Methods in

Molecular Biology 1770

Sarah Covshoff Editor

Photo-
synthesis
Methods and Protocols
METHODS IN MOLECULAR BIOLOGY

Series Editor
John M. Walker
School of Life and Medical Sciences
University of Hertfordshire
Hatfield, Hertfordshire, AL10 9AB, UK

For further volumes:

http://www.springer.com/series/7651
Photosynthesis

Methods and Protocols

Edited by

Sarah Covshoff
Niceville, FL, USA
Editor
Sarah Covshoff
Niceville, FL, USA

ISSN 1064-3745 ISSN 1940-6029 (electronic)

Methods in Molecular Biology
ISBN 978-1-4939-7785-7 ISBN 978-1-4939-7786-4 (eBook)
https://doi.org/10.1007/978-1-4939-7786-4
Library of Congress Control Number: 2018940274

© Springer Science+Business Media, LLC, part of Springer Nature 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply,
even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations
and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to
be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty,
express or implied, with respect to the material contained herein or for any errors or omissions that may have been made.
The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Humana Press imprint is published by the registered company Springer Science+Business Media, LLC part of
Springer Nature.
The registered company address is: 233 Spring Street, New York, NY 10013, U.S.A.
Preface

Photosynthesis is the foundation for life on Earth. In the last 10 years, interest in photosyn-
thesis has increased dramatically as a means to address two concerns related to global
population growth: increasing natural photosynthesis to ensure food security and harnessing
the power of photosynthesis for unlimited energy. A multidisciplinary cadre of scientists has
joined the photosynthetic research community with those aims in mind.
The goal of this volume entitled Photosynthesis: Methods and Protocols is to provide
fundamental protocols for the study of photosynthesis in a manner accessible to a broad
spectrum of researchers. A range of classical and modern techniques written in user-friendly,
accessible language is provided. The contents are broken into four parts covering the
measurement of physiological photosynthetic parameters, quantifying photosynthetic
enzyme abundance and catalytic activity, visualizing cellular and subcellular phenotypes,
and photosynthesis-inspired energy generation.
In the first part, researchers wishing to learn the fundamentals of physiological techni-
ques will find an introductory chapter surveying current tools for in vivo measurements. The
remaining chapters in the part provide in-depth protocols for the measurement of gas
exchange at small-through-large scales, light response curves, techniques utilizing chloro-
phyll fluorescence, carbon and oxygen stable isotopes, and a comparison of methods for
measurements of photosynthetic oxygen evolution.
The second and third parts present easily adoptable protocols to aid researchers in
identifying phenotypes of interest and elucidating their underlying biology. These include
the quantification of photosynthetic enzymes, investigation of the catalytic properties and
activity of Rubisco, visualizing cellular phenotypes by staining and molecular labeling, as
well as the quantification of thylakoid lipid classes.
The final part explores the rapidly expanding field of photosynthesis-inspired energy
generation. Protocols for two approaches—water-splitting photoelectrochemical cells and
biophotovoltaic devices—are presented. These nascent technologies aim to use the natural
world as a springboard to unlimited clean fuel.
As the photosynthetic research community diversifies, exciting developments in the
understanding and application of photosynthesis are sure to happen. I hope you are inspired
in your studies.

Niceville, FL, USA Sarah Covshoff

v
Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

PART I MEASURING PHOTOSYNTHETIC PARAMETERS

1 Survey of Tools for Measuring In Vivo Photosynthesis . . . . . . . . . . . . . . . . . . . . . . 3
Berkley J. Walker, Florian A. Busch, Steven M. Driever,
Johannes Kromdijk, and Tracy Lawson
2 Photosynthetic Gas Exchange in Land Plants at the Leaf Level . . . . . . . . . . . . . . . 25
Florian A. Busch
3 Design and Use of a Digitally Controlled Device for Accurate,
Multiplexed Gas Exchange Measurements of the Complete Foliar
Parts of Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Gavin M. George, Katharina Kölling, Roland Kuenzli,
Matthias Hirsch-Hoffmann, Patrick Flütsch, and Samuel C. Zeeman
4 Measuring Canopy Gas Exchange Using CAnopy Photosynthesis
and Transpiration Systems (CAPTS). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
Qingfeng Song and Xin-Guang Zhu
5 Light-Response Curves in Land Plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Robert A. Coe and HsiangChun Lin
6 Chlorophyll Fluorescence on the Fast Timescale . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Olubukola O. Ajigboye, Rumiana V. Ray, and Erik H. Murchie
7 Sub-saturating Multiphase Flash Irradiances to Estimate Maximum
Fluorescence Yield . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Thomas J. Avenson and Aaron J. Saathoff
8 Chlorophyll Fluorescence Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Tracy Lawson and Silvere Vialet-Chabrand
9 Measurement of O2 Uptake and Evolution in Leaves In Vivo
Using Stable Isotopes and Membrane Inlet Mass Spectrometry. . . . . . . . . . . . . . . 141
Steven M. Driever and Neil R. Baker
10 Using Stable Carbon Isotopes to Study C3 and C4
Photosynthesis: Models and Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
Nerea Ubierna, Meisha-Marika Holloway-Phillips,
and Graham D. Farquhar
11 Liquid-Phase Measurements of Photosynthetic Oxygen Evolution . . . . . . . . . . . . 197
Dmitriy Shevela, Wolfgang P. Schröder, and Johannes Messinger

vii
viii Contents

PART II MEASURING PHOTOSYNTHETIC ENZYME ABUNDANCE

AND CATALYTIC ACTIVITY

12 Quantification of Photosynthetic Enzymes in Leaf Extracts

by Immunoblotting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
J. Alejandro Perdomo, Cristina R. G. Sales, and Elizabete Carmo-Silva
13 Extraction of RuBisCO to Determine Catalytic Constants . . . . . . . . . . . . . . . . . . . 229
Douglas J. Orr and Elizabete Carmo-Silva
14 Spectrophotometric Determination of RuBisCO Activity
and Activation State in Leaf Extracts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Cristina R. G. Sales, Gustaf E. Degen, Anabela Bernardes da Silva,
and Elizabete Carmo-Silva

PART III DETERMINING CELLULAR AND SUB-CELLULAR PHENOTYPES

15 Creating Leaf Cell Suspensions for Characterization of Mesophyll

and Bundle Sheath Cellular Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
Roxana Khoshravesh and Tammy L. Sage
16 Determining the Subcellular Localization of Fluorescently
Tagged Proteins Using Protoplasts Extracted from Transiently
Transformed Nicotiana benthamiana Leaves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Vivien Rolland
17 3D Clearing and Molecular Labeling in Plant Tissues . . . . . . . . . . . . . . . . . . . . . . . 285
William M. Palmer, Jamie R. Flynn, Antony P. Martin,
Stephanie L. Reed, Christopher P. L. Grof, Rosemary G. White,
and Robert T. Furbank
18 Evaluation of Lipids for the Study of Photosynthetic Membranes . . . . . . . . . . . . . 305
Helmut Kirchhoff and Robert Yarbrough

PART IV PHOTOSYNTHESIS-INSPIRED ENERGY GENERATION

19 “Click” Methodology for the Functionalization of Water

Oxidation Catalyst Iridium Oxide Nanoparticles with Hydrophobic
Dyes for Artificial Photosynthetic Constructs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
Jackson D. Megiatto Jr. and Catia Ornelas
20 Biophotovoltaics: Design and Study of Bioelectrochemical Systems
for Biotechnological Applications and Metabolic Investigation . . . . . . . . . . . . . . . 335
Stephen J. L. Rowden, Paolo Bombelli, and Christopher J. Howe

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
Contributors

OLUBUKOLA O. AJIGBOYE  Division of Plant and Crop Sciences, School of Biosciences,

The University of Nottingham, Leicestershire, UK
THOMAS J. AVENSON  LI-COR Biosciences, Lincoln, NE, USA; Department of Biochemistry
and Molecular Biology, Michigan State University, East Lansing, MI, USA
NEIL R. BAKER  School of Biological Sciences, University of Essex, Colchester, Essex, UK
ANABELA BERNARDES DA SILVA  Faculdade de Ciências, Departamento de Biologia Vegetal
e Centro de Biodiversidade, Genómica Integrativa e Funcional (BioFIG), Universidade de
Lisboa, Lisbon, Portugal
PAOLO BOMBELLI  Department of Biochemistry, University of Cambridge, Cambridge, UK
FLORIAN A. BUSCH  Research School of Biology and ARC Centre of Excellence for
Translational Photosynthesis, The Australian National University, Acton, ACT, Australia
ELIZABETE CARMO-SILVA  Lancaster Environment Centre, Lancaster University, Lancaster,
UK
ROBERT A. COE  C4 Rice Centre, International Rice Research Institute (IRRI), Los Baños,
Philippines
GUSTAF E. DEGEN  Lancaster Environment Centre, Lancaster University, Lancaster, UK
STEVEN M. DRIEVER  Centre for Crop Systems Analysis, Wageningen University and
Research, Wageningen, The Netherlands
GRAHAM D. FARQUHAR  Research School of Biology, Australian National University,
Canberra, ACT, Australia
PATRICK FLÜTSCH  Department of Biology, Institute of Agricultural Sciences, ETH Zurich,
Zurich, Switzerland
JAMIE R. FLYNN  School of Biomedical Sciences and Pharmacy, University of Newcastle,
Callaghan, NSW, Australia
ROBERT T. FURBANK  ARC Centre of Excellence for Translational Photosynthesis, Australian
National University, Acton, ACT, Australia
GAVIN M. GEORGE  Department of Biology, Institute of Agricultural Sciences, ETH Zurich,
Zurich, Switzerland
CHRISTOPHER P. L. GROF  School of Environmental and Life Sciences, University of
Newcastle, Callaghan, NSW, Australia
MATTHIAS HIRSCH-HOFFMANN  Department of Biology, Institute of Agricultural Sciences,
ETH Zurich, Zurich, Switzerland
MEISHA-MARIKA HOLLOWAY-PHILLIPS  Research School of Biology, Australian National
University, Canberra, ACT, Australia
CHRISTOPHER J. HOWE  Department of Biochemistry, University of Cambridge, Cambridge,
UK
ROXANA KHOSHRAVESH  Department of Ecology and Evolutionary Biology, University of
Toronto, Toronto, ON, Canada
HELMUT KIRCHHOFF  Institute of Biological Chemistry, Washington State University,
Pullman, WA, USA
KATHARINA KÖLLING  Department of Biology, Institute of Agricultural Sciences, ETH
Zurich, Zurich, Switzerland

ix
x Contributors

JOHANNES KROMDIJK  Carl R. Woese Institute for Genomic Biology, University of Illinois,
Urbana, IL, USA
ROLAND KUENZLI  DMP Ltd, Fehraltorf, Switzerland
TRACY LAWSON  School of Biological Sciences, University of Essex, Colchester, Essex, UK
HSIANGCHUN LIN  C4 Rice Centre, International Rice Research Institute (IRRI), Los
Baños, Philippines
ANTONY P. MARTIN  School of Environmental and Life Sciences, University of Newcastle,
Callaghan, NSW, Australia
JACKSON D. MEGIATTO JR  Institute of Chemistry, University of Campinas (UNICAMP),
Campinas, São Paulo, Brazil
JOHANNES MESSINGER  Department of Chemistry, Umeå University, Umeå, Sweden;
Department of Chemistry—Ångström Laboratory, Uppsala University, Uppsala, Sweden
ERIK H. MURCHIE  Division of Plant and Crop Sciences, School of Biosciences, The University
of Nottingham, Leicestershire, UK
CATIA ORNELAS  Institute of Chemistry, University of Campinas (UNICAMP), Campinas,
São Paulo, Brazil
DOUGLAS J. ORR  Lancaster Environment Centre, Lancaster University, Lancaster, UK
WILLIAM M. PALMER  School of Environmental and Life Sciences, University of Newcastle,
Callaghan, NSW, Australia
J. ALEJANDRO PERDOMO  Plant Science Department, Rothamsted Research, Harpenden, UK
RUMIANA V. RAY  Division of Plant and Crop Sciences, School of Biosciences, The University of
Nottingham, Leicestershire, UK
STEPHANIE L. REED  School of Environmental and Life Sciences, University of Newcastle,
Callaghan, NSW, Australia
VIVIEN ROLLAND  Commonwealth Scientific and Industrial Research Organisation
(CSIRO), Agriculture & Food, Canberra, ACT, Australia
STEPHEN J. L. ROWDEN  Department of Biochemistry, University of Cambridge, Cambridge,
UK; Faculty of Engineering and Science, University of Greenwich, Chatham Maritime,
Kent, UK
AARON J. SAATHOFF  LI-COR Biosciences, Lincoln, NE, USA; School of Natural Resources,
University of Nebraska-Lincoln, Lincoln, NE, USA
TAMMY L. SAGE  Department of Ecology and Evolutionary Biology, University of Toronto,
Toronto, ON, Canada
CRISTINA R. G. SALES  Lancaster Environment Centre, Lancaster University, Lancaster, UK
WOLFGANG P. SCHRÖDER  Department of Chemistry, Umeå University, Umeå, Sweden
DMITRIY SHEVELA  Department of Chemistry, Umeå University, Umeå, Sweden
QINGFENG SONG  National Key Laboratory of Plant Molecular Genetics, CAS Center for
Excellence in Molecular Plant Sciences, Shanghai Institute of Plant Physiology and Ecology,
Chinese Academy of Sciences, Shanghai, China; State Key Laboratory of Hybrid Rice,
Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China
NEREA UBIERNA  School of Biological Sciences, Molecular Plant Sciences, Washington State
University, Pullman, WA, USA
SILVERE VIALET-CHABRAND  School of Biological Sciences, University of Essex, Colchester,
Essex, UK
BERKLEY J. WALKER  Biochemistry of Plants, Heinrich-Heine University, Düsseldorf,
Germany
ROSEMARY G. WHITE  CSIRO Agriculture, Black Mountain, ACT, Australia
 Contributors xi

ROBERT YARBROUGH  Institute of Biological Chemistry, Washington State University,

Pullman, WA, USA
SAMUEL C. ZEEMAN  Department of Biology, Institute of Agricultural Sciences, ETH Zurich,
Zurich, Switzerland
XIN-GUANG ZHU  National Key Laboratory of Plant Molecular Genetics, CAS Center for
Excellence in Molecular Plant Sciences, Shanghai Institute of Plant Physiology and Ecology,
Chinese Academy of Sciences, Shanghai, China
 Part I

Measuring Photosynthetic Parameters

 Chapter 1

Survey of Tools for Measuring In Vivo Photosynthesis

Berkley J. Walker, Florian A. Busch, Steven M. Driever,
Johannes Kromdijk, and Tracy Lawson

Abstract
Measurements of in vivo photosynthesis are powerful tools that probe the largest fluxes of carbon and
energy in an illuminated leaf, but often the specific techniques used are so varied and specialized that it is
difficult for researchers outside the field to select and perform the most useful assays for their research
questions. The goal of this chapter is to provide a broad overview of the current tools available for the study
of in vivo photosynthesis so as to provide a foundation for selecting appropriate techniques, many of which
are presented in detail in subsequent chapters. This chapter also organizes current methods into a
comparative framework and provides examples of how they have been applied to research questions of
broad agronomical, ecological, or biological importance. The chapter closes with an argument that the
future of in vivo measurements of photosynthesis lies in the ability to use multiple methods simultaneously
and discusses the benefits of this approach to currently open physiological questions. This chapter,
combined with the relevant methods chapters, could serve as a laboratory course in methods in photosyn-
thesis research or as part of a more comprehensive laboratory course in general plant physiology methods.

Key words Photosynthesis, CO2 exchange, O2 exchange, Chlorophyll fluorescence, Online mass
spectrometry

1 Principles of and Perspectives on Measuring In Vivo Photosynthetic Flux

The challenge of quantifying photosynthetic rates in vivo lies in the

unique substrates and products of carbon assimilation. Photosynthe-
sis involves a series of interconnected reactions that sequentially con-
vert light energy into chemical energy and then use this energy to
reduce carbon into usable sugars as shown non-stoichiometrically as
light
CO2 þ H2 O ! O2 þ glucose ð1Þ

Critical information concerning the mechanisms and biochem-

istry of photosynthesis can be determined in isolated or reconsti-
tuted enzymatic systems and can be probed in vivo to gain an
integrated understanding of how photosynthesis is regulated and
contributes to plant growth. Here, we focus on methods that

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_1, © Springer Science+Business Media, LLC, part of Springer Nature 2018

3
4 Berkley J. Walker et al.

measure an intact photosynthesizing system and how the informa-

tion gained can contribute to broader research questions. Since
photosynthetic reactions occur in the aqueous phase and glucose
is rapidly converted to other forms, many measurement techniques
examine gas exchange (assimilation of CO2 or production of O2) to
measure photosynthetic activity in vivo. Additionally, given the
unique properties of light energy capture and subsequent conver-
sion to short-term chemical storage as adenosine triphosphate
(ATP) and nicotinamide adenine dinucleotide phosphate
(NADPH), there exist an array of light-based biophysical probes
(i.e., chlorophyll fluorescence and leaf spectroscopy) that can also
be employed to monitor specific components of the light reactions
nondestructively and under natural conditions.
Today a wide array of instruments able to measure CO2 and O2
exchange and other light-based biophysical probes of the light
reactions in vivo are commercially available and can be further
modified to interface with additional analytical platforms (Fig. 1),
as is the case with online isotopic analysis. The variety of available

Fig. 1 Survey of tools used to measure photosynthetic fluxes discussed in this

chapter. Shown are three different measurement principles (O2 exchange, CO2
exchange, and light-based biophysical probes) plotted against the different
spatial scales measurements can be made on for each technique ranging from
algal cultures to growing regions. Techniques plotted include O2 electrode,
isotope analysis including dry matter discrimination and online mass
spectroscopy (MS), clamp-on infrared gas analysis (IRGA), eddy covariance,
enclosure and chamber-based analysis, pulse amplitude-modulated (PAM)
fluorometry (including guided approaches that use fiber-optic cables or light
sources in close proximity to the leaf surface and image-based approaches),
in vivo spectroscopy for monitoring absorption shifts of wavelengths of interest,
and remote sensing approaches using satellite or aerial imaging
 Tools for Measuring in vivo Photosynthesis 5

and user-friendly instrumentation means that photosynthesis is

easier to measure today than at any previous time, a benefit to not
only researchers who focus on photosynthesis, but also those from
other fields who want the insight that in vivo measurements of
photosynthetic fluxes can provide.
One such application extending the relevance of in vivo mea-
surements of photosynthesis to other fields is in understanding the
response of crop production to climate change. Models of canopy
crop production often incorporate biochemical sub-models which
describe the response of leaf carbon fixation using in vivo biochem-
ical parameters [1–3]. The biochemical parameters used to build
leaf-level models of photosynthesis in response to changing climate
are derived from in vivo measurements of carbon assimilation, and
these parameters can vary greatly among species [4–6]. While these
parameters are constant within some species like tomato [7], they
are significantly different when measured among rice cultivars [8]
and can even vary temporally throughout a growing season as seen
in wheat flag leaf [9]. Given the variability of these photosynthetic
parameters temporally and among species, it is apparent that in vivo
measurements of photosynthetic flux under field conditions are
needed to provide the parameters for higher order climate models.
Fortunately, current instrumentation provides the portability
needed to make these measurements possible.
Instruments and experimental setups for measuring photosyn-
thetic flux can be characterized broadly by what parts of Eq. (1)
they measure (viz. CO2 exchange, O2 exchange, or light-based
biophysical probes, Fig. 1). They can be further subdivided accord-
ing to what scale they measure, varying from chloroplast suspen-
sions to regional land surfaces. Proper selection and use of these
techniques depend on a correct understanding of the purposes,
principles, advantages, and disadvantages behind each
measurement.
The following section gives an overview of the major purposes,
principles, advantages, and disadvantages of each of the methods
discussed in later chapters. We also included additional techniques
for completeness. Given the breadth of techniques described
herein, our coverage will be necessarily brief with more comprehen-
sive discussions available in the referenced work. We begin with gas
exchange-based methods and then move to light-based probes of
the photosynthetic light reactions.

2 Measurements of Gas Exchange

The observation that more carbon fixation should drive greater

plant production and yield was the motivation for constructing
the first system for measuring CO2 exchange in an intact leaf
[10]. Measurements of CO2 were made possible with the
6 Berkley J. Walker et al.

development of commercially available infrared gas analyzers

(IRGA) which measure CO2 concentrations via the absorbance of
characteristic wavelengths in the infrared spectrum. Measurements
can be made in closed systems where the drawdown of CO2 is
measured in a closed cuvette or chamber as a function of time, or
via open systems where CO2 fluxes are resolved from the differences
in gas concentrations in an airstream measured essentially before
and after exposure to the plant material at a known flow rate. CO2
fluxes can be determined at multiple scales to better understand the
leaf-to-canopy interactions of the plant with the environment.
Examples of commercially and custom-built systems used to mea-
sure CO2 exchange at diverse scales reveal the wide variety of tools
available to researchers today (Fig. 2). Note that the systems shown
are only examples of commercially available instruments and nei-
ther intended to cover every manufacturer nor endorse a specific
manufacturer. The reader is encouraged to consult the literature
and contact a variety of manufacturers to determine which device
(s) can provide the features needed for a specific research
application.

Fig. 2 Example tools used to measure CO2 exchange across diverse scales. At
the canopy to regional scale shown is a sample eddy covariance flux station from
Campbell Scientific (a). Commercially available portable CO2 exchange systems
for measuring leaf photosynthesis such as the LI-6800 from LI-COR Biosciences
(b) and CIRAS-3 from PP Systems International, Inc. (c) can be used for leaf or
whole-plant measurements. For measurements of an enclosed canopy shown
are the EGAS-2, a custom-built, multiplexed whole-plant gas exchange system
(d) and the custom-built canopy enclosure of the CAPTS system (e). Instruments
from the various manufacturers are shown only for illustrative purposes and do
not imply any specific recommendation. Above images are reproduced with kind
permission from their respective copyright holders: Campbell Scientific (a),
LI-COR Biosciences (b), PP Systems International, Inc. (c), Gavin George (d ),
and Qingfeng Song (e)
 Tools for Measuring in vivo Photosynthesis 7

Modern gas exchange systems based on IRGA technology

combine CO2 and H2O measurements by making use of differ-
ences in the absorption spectra of CO2 and H2O. Estimating tran-
spiration fluxes from differences in H2O concentrations in the air is
necessary to account for the dilution effect of transpired H2O on
the CO2 concentration in the air and determine the intercellular
CO2 concentration. Gas exchange systems usually also measure
and/or control other environmental conditions the leaf or plant is
exposed to, such as temperature, light intensity, and air pressure. In
combination, these parameters are used to derive the net CO2
uptake or release of the measured plant.

2.1 Leaf CO2 Off-the-shelf gas exchange systems for measuring leaf gas exchange
Exchange are readily available and increasingly user friendly, which allows
researchers to collect physiological measurements with the push
of a few buttons. Leaf-level CO2 exchange is routinely used to
estimate instantaneous rates of net CO2 assimilation (Anet) under
ambient conditions (indicating the carbon uptake of the leaf at that
moment), or under light- or CO2-saturated conditions, allowing
direct comparison of the CO2 assimilation capacity between differ-
ent leaves. Such measurements can be taken in a timespan on the
order of minutes, allowing sampling of a large number of plants. A
more detailed assessment of the biochemistry underlying the pho-
tosynthetic processes of the studied leaf can be obtained by taking
repeated measurements on the same leaf under varying CO2 con-
centrations, light intensities, or temperatures. Measurements of
Anet in response to the CO2 concentration can be used in combi-
nation with a biochemical model of photosynthesis (such as the
Farquhar, von Caemmerer, Berry (FvCB) model; [11]) to estimate,
e.g., the maximum rate of RuBisCO carboxylation (Vcmax) [12],
maximum rate of electron transport (Jmax), or rate of phosphate
limitation (TPU limitation) [13]. Techniques to measure net gas
exchange to resolve biochemical parameters such as Vcmax, Jmax, and
TPU limitation are presented herein [14]. Similarly, assessing how
Anet changes with light intensity yields estimates of the maximum
rate of photosynthetic electron transport (Jmax; [15]). In addition,
measurements of transpiration can be used to quantify the diffusive
barriers for CO2 to enter the leaf via the stomata, termed stomatal
conductance (gs). These response measurements require more time
to perform than those under a single condition, often more than an
hour, but help elucidate whether differences between plants are due
to the environment the plant is exposed to and the plant’s transient
response to it, or whether they are caused by some more long-term
acclimation of its biochemistry or anatomy. In addition to these
plant-specific parameters, gas exchange has been used to resolve
RuBisCO kinetics in vivo [16–18]. This approach provides values
for some of the most important input parameters for the widely
used FvCB model.
8 Berkley J. Walker et al.

Leaf-level gas exchange can also be combined with other

techniques to gain further insight into plant physiology and metab-
olism. In combination with stable isotope measurements it can be
used to assess internal CO2 diffusion (see below), or to quantify
gross fluxes of CO2 into and out of the leaf [19]. Net gas exchange
fluxes can help constrain the labeling kinetics of metabolite pools to
map fluxes through central carbon metabolism [20]. A similar
approach allows for a detailed flux analysis of specific biochemical
pathways such as photorespiration, which has been achieved with
leaf-level gas exchange in combination with quantitative NMR
analysis [21].

2.2 Whole-Plant CO2 Leaf-level gas exchange, as described above, can yield valuable
Exchange insights into the physiology and biochemistry of the leaf. However,
in some cases these measurements cannot be obtained directly, e.g.,
when the leaves are too small or oddly shaped to be measured via a
clamp-on leaf chamber. In addition, for some purposes leaf-level
measurements are too specific, as they only include measurements
of the photosynthetic tissue of the leaf and neglect the effect of
other parts of the plant that are not contained within the chamber,
such as other leaves, stems, and roots. This is particularly an issue
when one wants to relate photosynthetic rate to plant growth or to
integrate photosynthesis across leaves of different ages on the same
plant. One way around these issues is to measure the CO2 uptake of
the enclosed shoots of an entire plant [22].
Whole-shoot measurements of plant CO2 exchange also pro-
vide different types of data as compared to leaf-level measurements.
For example, the CO2 uptake integrated over the entirety of the
plant and the whole growth period can be used to estimate growth
nondestructively in real time through carbon balance when it is
measured at regular enough intervals [23]. In addition, whole-
plant experiments can be combined with carbon isotope labeling
to obtain insights into carbohydrate metabolism [24]. The benefit
of a whole-plant approach in labeling approaches is that more than
one leaf of the plant is labeled, allowing inclusion of the effects of
both photosynthetic and non-photosynthetic tissue of the above-
ground parts of the plant in experimental analyses. Whole-plant
exchange can be measured in much the same way as leaf-level
exchange by means of closed, open, or semi-closed chambers
placed over or around the plant to be measured. Gas exchange is
measured by the drawdown of CO2 within the chamber for closed
and semi-closed systems, or the difference in CO2 and H2O con-
centration between the air entering and exiting the chamber in
open systems. Chamber-based measurements can be effectively
employed, but are biased by differences in temperature, light inten-
sity, turbulent mixing, and CO2 concentration between the inside
and outside of the chamber (see [25]). Many of these differences
can be minimized through construction of semi-closed systems
 Tools for Measuring in vivo Photosynthesis 9

with air-conditioners, mixing fans, dehumidifiers, and CO2

injection systems [26]. Chamber-based methods can be deployed
on the scale of small herbaceous plants [27] and recent develop-
ments in multichannel systems enable the measurement of intact
shoot material in many plants simultaneously as described herein
[22]. Chamber-based measurements have even been scaled to fully
grown plants such as in the Hawkesbury Forest experiment where
Eucalyptus saligna trees are reared from seedlings and grown within
chambers that allow for trees up to 9 m in height [28]. The Haw-
kesbury Forest chambers are revealing important physiology in
relation to climate change; for example, gas exchange data from
chambers with elevated temperatures revealed that in a Eucalyptus
species, respiration increases more with temperature than net pho-
tosynthesis does despite physiological adaptation to growth condi-
tions, which may exacerbate the impacts of climate change in these
species since their net CO2 uptake will decrease with
temperature [29].
Some things that work well on the leaf level cause problems on
the whole-plant level; whole-plant gas exchange measurements are
difficult when performing responses to environmental conditions,
such as temperature, gas exchange, and light intensity, since these
variables are not easily controlled uniformly across the whole can-
opy of the plant. An additional consideration is the increased
investment in equipment, reagents, and setup time, especially
since there is currently a lack of commercially available instruments
and most systems must be custom-built.

2.3 Canopy CO2 The next step along the continuum of scales after whole-plant
Exchange measurements is the measurement of CO2 exchange at the canopy
scale. Canopy-scale measurements of CO2 and H2O exchange are
most useful when research questions are focused on the interaction
between plants and the growth environment, for example to deter-
mine the net carbon balance of ecosystems in response to present
and future climates [30]. Canopy photosynthesis is measured in
much the same way as whole-plant measurements by enclosing a
portion of the canopy in a translucent chamber and measuring gas
fluxes either via an open, closed, or semi-closed path design. Can-
opy chambers are helping to resolve the importance of canopy
effects on in-field photosynthesis and recent advances in automa-
tion have made them more practical for larger scale studies as
discussed herein [31].
Eddy covariance has become a powerful method to determine
CO2 and H2O exchanges from canopies ranging in size from
hundreds to thousands of meters noninvasively and over long time-
scales [30, 32]. Eddy covariance measures flux into and from the
canopy by measuring trace gas (CO2 and H2O) concentrations in
tandem with wind speed and direction. These data are analyzed
using a statistical model that represents turbulent mixing to
10 Berkley J. Walker et al.

produce measurements of canopy CO2 and H2O exchange. This

technique has been employed in a myriad of sites around the world
and is more straightforward over “smooth” vegetation like most
agricultural systems, but more complicated over “rough” canopies
such as those found over forests [33, 34]. While not presented as a
method within this book, it is nevertheless an important tool for
scaling photosynthetic flux to canopy and regional scales and there-
fore mentioned for the sake of completeness.

2.4 Measurements The O2 produced during H2O splitting from photosystem II

of O2 Exchange (PSII) provides a direct assay of the activities of the light reactions.
O2 in living systems was first measured using manometric techni-
ques before a critical review of cardiovascular researcher Leland
Clark’s work with blood oxygenation led him to develop an elec-
trode for the measurement of dissolved O2 [35]. The Clark-type
electrode determines dissolved O2 concentration by monitoring
the reduction of O2 catalyzed via a platinum surface separated
from the liquid being assayed by a semipermeable membrane
[36]. O2 electrodes are routinely used to determine the photosyn-
thetic capacity of algae cultures and chloroplast suspensions and
have even been adapted for use with excised leaf disks. Rates are
determined by monitoring the increase of dissolved O2 as a func-
tion of time in a closed, illuminated reaction cuvette. O2 electrodes
can also be employed for gas-phase measurements, but for greater
accuracy online mass spectroscopy can be employed as discussed
below and within this book. Techniques and background on both
approaches are presented herein [37, 38]. Measurements of O2
exchange in algal cultures and isolated chloroplasts were instru-
mental in resolving the maximum quantum efficiency of photosyn-
thesis (discussed further below), which was critical for subsequent
work exploring its mechanisms [39]. Measurements of O2
exchange are still valuable alone, but are especially informative
when combined with isotopic methods as discussed below.

2.5 Using Isotopes Whereas measurements of the molecular fluxes of O2, CO2, and
to Resolve Net Fluxes H2O vapor provide a rich source of information on photosynthetic
of Gas and Physiology physiology of the measured sample, parallel analysis of stable iso-
topes performed on the same fluxes can be used both to expand and
better constrain the analysis as discussed herein [40]. Small predict-
able differences in reaction and diffusion rates between lighter and
heavier isotopologues of CO2 and H2O create small alterations in
the natural abundance of stable isotopes during photosynthetic gas
exchange. Slight changes in the relative abundance of 13CO2 can be
used to quantify internal conductance to CO2 in C3 species [41],
and establish the presence and expression of carbon concentrating
mechanisms [42] as well as the extent of CO2 leakage away from
the site of concentration [43]. Simultaneous determination of
changes in the relative natural abundance of 18O in CO2 and
 Tools for Measuring in vivo Photosynthesis 11

H2O vapor can be used to determine the internal conductance to

CO2 in C4 species and CO2 permeability of the chloroplast mem-
brane in C3 species [44]. These techniques make use of changes in
the natural abundance of stable isotopes. However, it is also possi-
ble to manipulate the stable isotope composition of air to directly
distinguish between in- and outgoing gas fluxes of leaves. For
example, fluxes of O2 evolution and O2 uptake in leaves can be
determined using air heavily enriched in 18O2 and H216O
[45, 46]. Similarly, when using air heavily enriched in 13CO2,
distinction can be made between, e.g., fluxes of respired (12CO2)
and assimilated 13CO2 [47]. These flux measurements can also
provide an estimation of the chloroplastic CO2 concentration,
and thereby can be used to quantify the internal conductance to
CO2 between the intercellular air space and the chloroplast
[19]. These isotope exchange techniques are often combined with
chlorophyll fluorescence measurements [48–50]. When using
isotopologues of CO2 with singly (C18O16O) and doubly labeled
O2 (C18O16O), it is possible to study specific enzyme activity
in vivo of carbonic anhydrase in leaves [51]. These types of analyses
allow relatively fast kinetic studies in vivo (seconds, minutes),
providing a more detailed dissection of processes underlying CO2
and O2 gas exchange of leaves. Moreover, when using air heavily
enriched with isotopologues for prolonged periods (minutes,
hours, days), isotopic labeling of metabolites, proteins, and struc-
tural components can be achieved in so-called pulse-chase experi-
ments. This longer term isotope labeling, especially when
combined with leaf gas exchange techniques, can provide valuable
insight into metabolic pathways, fluxes, and enzyme activities and
changes therein as discussed above.

3 Light-Based Probes of Photosynthesis

The unique photochemistry of the light reactions provides several

useful tools for understanding photosynthetic rates from the per-
spective of light energy utilization and subsequent electron trans-
port (Fig. 3). As discussed below, these light-based probes of
photosynthesis are noninvasive and can be applied remotely to
resolve the flux of electrons and protons in vivo through photosyn-
thetic systems. These techniques differ from gas exchange methods
in that flux is not measured directly via the uptake of substrates, but
rather via the interactions of molecules and proteins with photo-
chemically driven redox reactions or protonation states. These
signatures of photosynthesis are then detected as emitted photons,
as is the case with chlorophyll fluorescence, or as shifts in light
absorption at characteristic wavelengths. The emitted photons
and/or absorbance shifts are then used to derive either operational
efficiencies in the case of chlorophyll fluorescence and PSI redox
12 Berkley J. Walker et al.

Fig. 3 Tools used to assay light-based biophysical probes of the light reactions
across diverse scales. Shown are pulse amplitude-modulated (PAM) fluorometry
methods including “guided probe” approaches, which deliver measuring and
actinic light via fiber-optic cables or LED sources in close proximity to the leaf
surface, such as the Dual-PAM-100 from Heinz Walz GmbH (a), the MultispeQ
from PhotosynQ (b), and the FluorPen from Photon Systems Instruments (c) as
well as imaging-based platforms such as the Open FluorCam from Photon
Systems Instruments (d ). Remote sensing platforms include the use of drones
such as the Phantom series from Dà-Jiāng Innovations Science and Technology
Co (e). Monitoring of other spectral signatures can be accomplished using the
MultispeQ from PhotosynQ ( f ) and the JTS-10 spectrometer from BioLogic
Science Instruments (g). Instruments from the various manufacturers are
shown only for informational purposes and do not imply any specific
recommendation. Above images are reproduced with kind permission from
their respective copyright holders: Heinz Walz GmbH (a), PhotosynQ (b and f ),
Photon Systems Instruments spol. s r.o. (c and d), Clément Bucco-Lechat under
a Creative Commons Attribution-Share Alike 3.0 license (e), and BioLogic
Science Instruments (g)

state or relative flux units as is the case with the electrochromic shift
(ECS). As with gas exchange, these techniques can be applied from
the single leaf level to entire canopies, and to some extent moni-
tored remotely using drone or satellite imagery.

3.1 Chlorophyll Chlorophyll fluorescence is a powerful tool for probing the opera-
Fluorescence tion of photochemistry at the level of PSII, the enzyme complex
responsible for H2O splitting during photosynthesis and providing
electrons for downstream photochemical energy conversion from
absorbed light energy [52, 53]. The link between chlorophyll
fluorescence and photochemistry lies in the various fates of light
energy absorbed by a chlorophyll molecule. Once excited, there are
three main routes by which absorbed light energy is “quenched,”
or dissipated [54]: (1) photochemical quenching through passage
of the excitation energy to PSII, where it is used to transfer
 Tools for Measuring in vivo Photosynthesis 13

electrons from H2O to the mobile electron-carrier plastoquinone;

(2) non-photochemical quenching when the excitation energy is
dissipated as heat; and (3) chlorophyll fluorescence when the
energy is reemitted as a photon with a shifted wavelength. Since
chlorophyll fluorescence represents the balance of the three pro-
cesses and is readily measured, it can be monitored under various
conditions to understand relative rates of photochemical and
non-photochemical quenching [55]. Fluorescence can be
measured using “guided probe” systems, which orient the measur-
ing and excitation beams close to the leaf surface with an optic fiber
or via more remote tools such as imaging systems (as described
below) or more recently via solar-induced fluorescence measured
using satellites.
The deconvolution of quenching fates is accomplished by mea-
suring chlorophyll fluorescence emitted before and during a short
(<1 s) flash of light that saturates the capacity of photochemical
quenching resulting in a proportional increase in fluorescence
[56]. The transient rise in fluorescence can then be used to infer
the behavior of photochemical quenching during the condition
immediately before the saturating flash of light. One difficulty
with this approach is imposing a flash of light that fully saturates
PSII without oversaturating PSII and inducing alterative dissipative
fates for absorbed light energy. While this can be accomplished by
careful selection of saturating light intensities, the problem can also
be circumvented by exposing the leaf to a multiphase flash of
sub-saturating intensities and extrapolating fluorescence yields to
a saturating value [57]. The pre-flash condition can either be a
light- or dark-adapted state and each reveals details of leaf light
use in the presence or absence of non-photochemical quenching,
respectively. Chlorophyll fluorescence can be measured on “fast” or
“slow” timescales to reveal potentially complementary information
on the fate of absorbed light energy. “Fast” timescale measure-
ments (millisecond) attempt to deconvolute the transient induction
of the fluorescence signal in response to a saturating light flash into
various phases of PSII electron transfer of a leaf in the fully dark-
adapted state [58]. Techniques in making sensible “fast” fluores-
cence measurements are discussed herein [59].
Chlorophyll fluorescence is often used at the “slow” timescales
to resolve details of photochemical quenching and non-photo-
chemical quenching (NPQ). In this chapter, we refer to “slow”
timescales as looking at the maximum fluorescence yield during the
application of a saturating light pulse without resolving the kinetics
of that initial induction as mentioned in the previous paragraph.
These measurements can be used in both the light- and dark-
adapted state, to resolve actual rates of linear electron flux, non--
photochemical quenching, and maximum efficiency of PSII
[52, 60]. These data are also important for determining the induc-
tion or relaxation kinetics of photochemical and/or
14 Berkley J. Walker et al.

non-photochemical processes, particularly in combination with

changes in environmental variables such as temperature, light, or
CO2 [61]. Changes in these estimations and kinetics can function
as a noninvasive proxy for a plant’s stress responses because of light-
induced damage to the photosynthetic apparatus [62–64]. Recently
developed techniques in using sub-saturating flashes of light are
making these measurements even less invasive with techniques
discussed herein [65]. Leaf-level chlorophyll fluorescence can also
be combined with CO2 and O2 gas exchange to determine energy
partitioning between CO2 assimilation, photorespiration, respira-
tion, and other processes [49, 50].
Leaf-level chlorophyll fluorescence measurements have been
applied broadly to probe and predict many aspects of plant physiol-
ogy. For example, the CO2 fixation rate of C4 plants can be accu-
rately determined through a now-simple measurement of quantum
efficiency using chlorophyll fluorescence since the majority of
reductants in this photosynthetic type go towards carbon reduction
[66]. Furthermore, since the non-photochemical quantum yield is
related to other plant stresses, dark-adapted chlorophyll fluores-
cence measurements have been used to screen for cold hardiness in
barley [67] and other crops.

3.2 Chlorophyll Chlorophyll fluorescence imaging provides the opportunity to spa-

Fluorescence Imaging tially resolve chlorophyll fluorescence signals and parameters at a
range of different scales. High-resolution chlorophyll fluorescence
microscopes enable photosynthetic efficiency (and the associated
quenching parameter) to be determined at the cellular [68, 69] and
the subcellular [70] level, while the integration of chlorophyll
fluorescence into many large-scale industrial and commercial phe-
notyping platforms has provided the ability to screen large numbers
of plants. Such capabilities have been invaluable for screening for
perturbations in plant metabolism, for example improvements in
photosynthesis [71] and the impact of herbicides [72]. Many of the
commercial instruments can be used on a range of different photo-
synthetic materials—from screening algae in media in petri dishes
[73] to plants in 96-well plates [72], from individual leaves [74] to
whole plants [75]. Chlorophyll fluorescence imaging also allows
spatial heterogeneity to be assessed which is often present within or
between samples [76]. This can be a problem with traditional fiber-
optic approaches that only measure a small proportion of the
photosynthetic material defined by the users which can lead to
increased variation between measurements due to spatial heteroge-
neity. Chlorophyll fluorescence imaging can also be combined with
other physiological techniques to provide advanced screening
methodologies and powerful physiological measurement
approaches. For examples, combined chlorophyll fluorescence and
thermal imaging has been used to screen for intrinsic H2O use
efficiency [77], and chlorophyll fluorescence imaging in
 Tools for Measuring in vivo Photosynthesis 15

combination with infrared gas exchange has been used to produce

spatial images of internal CO2 concentrations within the leaf in
order to calculate lateral gas fluxes [78]. Carrying out chlorophyll
fluorescence imaging under known gas environments has also been
used as a screening tool for photorespiratory mutants [79].
The advantages of chlorophyll fluorescence imaging have been
outlined above and are explored in greater detail herein [80]. How-
ever, disadvantages include the size of the equipment, which often
means that such imaging systems are not portable and often rely on
mains power. There are exceptions to this, with portable field-based
systems commercially available, although the area for imaging is in
general small compared to many of the lab instruments. Another
challenge is that due to the large imaging area of many systems, it
can be difficult to obtain even actinic illumination over the entire
imaging area, which can result in the introduction of spatial varia-
tion into the images. Another problem with illumination can be the
intensity of the saturating pulse, which can be lower than
4000 μmol m2 s1 and is therefore unlikely to be saturating for
species grown in high-light environments or that perform C4
photosynthesis.

3.3 Whole-Leaf Characteristic absorbance shifts originating from components of

Spectroscopy electron transport downstream of Photosystem II (PSII) provide
valuable insight into the fate and results of light energy capture.
The two main spectral signatures downstream of PSII stem from
the reaction center chlorophyll pigments of photosystem I (PSI),
P700, and from the ECS of carotenoid proteins embedded in the
thylakoid membranes [81]. The redox status of P700 is estimated
by absorbance shifts in the 800–850 nm range as measured under
steady state and rapid-saturating light pulses, thus probing quan-
tum efficiency of PSI [58, 82]. Since the ECS responds to the
electric field across the thylakoid membrane, the relaxation of the
ECS in a dark interval is proportional to the proton motive force
maintained by proton pumping and its decay kinetics provide infor-
mation concerning the conductivity of ATP synthase [83, 84].
The absorbance shifts driven by P700 and the ECS can be
measured on an intact leaf and are helping to resolve important
aspects of how the light reactions flexibly provide reductant and
ATP to optimally provide for the dynamic demands of carbon
fixation, photorespiration, and other aspects of central metabolism
without resulting in the damaging consequences of over-reduction
of the electron transport chain [85, 86]. These spectral
signatures have been used to provide indications of cyclic electron
transport around PSI when measured in conjunction with PSII
chlorophyll fluorescence [87, 88], determine the light response of
PSI quantum efficiency as affected by environmental variables
[89, 90], and estimate in vivo steady-state proton fluxes, ATP
synthase activity, and components of the transthylakoid proton
motive force [88, 91, 92].
16 Berkley J. Walker et al.

3.4 Light Response Measurements for probing photosynthesis, such as gaseous

Curves exchange of CO2 or O2, chlorophyll fluorescence, or wavelength-
specific absorption changes, are often integrated into experiments
where environmental factors such as temperature, and light inten-
sity CO2 can be varied. In the case of light intensity, several impor-
tant properties can be extracted from such measurements as
discussed herein [93].
Saturated photosynthetic capacity can be derived as the asymp-
tote to which the response curve saturates at infinite light intensity.
Additionally, the initial slope of the response provides a measure for
the quantum efficiency, i.e., the maximum number of evolved O2
molecules or assimilated CO2 molecules per absorbed photon. This
parameter is typically designated by the symbol φ and was instru-
mental in investigating a long-running scientific controversy
concerning the photon requirements of water splitting (reviewed
by [39]), until the involvement of two photosystems in photosyn-
thesis was discovered [94] and development of the Z-Scheme [95]
provided firm theoretical underpinning for the minimum require-
ment of eight photons per evolved O2.
While in the 1950s photosynthetic light response measurements
were performed using extremely tedious and error-prone manome-
try, modern instruments utilize more user-friendly techniques to
measure the light response of photosynthesis via liquid- or
gas-phase gas exchange. It is however still important to realize that
the measurement conditions during the light response curve (tem-
perature, light spectrum, atmospheric gas composition, etc.) as well
as pretreatment of the sample can have important consequences for
the measured light response. Photoinhibition or non-photochemical
quenching (NPQ) can substantially reduce the quantum efficiency of
photosynthesis (φ, [96]) and induce substantial variation in
measured values, whereas φ measurements on non-stressed, fully
dark-adapted material typically lead to a very narrow range of values
[97]. Considering the lack of variation associated with dark-adapted
φ in C3 photosynthesis, differences in the photosynthetic light
response can reflect physiological adaptations of photosynthesis,
such as the presence or absence of a carbon-concentrating mecha-
nism (e.g., [98]). For instance, in species with C4 photosynthesis, the
additional two ATP required for the carbon-concentrating C4 acid
shuttle is reflected in φ and light saturation of photosynthesis also
typically occurs at considerably higher light levels compared to C3.
Light response curves are deceptively tricky to execute properly.
For example, in the case of leaf gas exchange, it can be challenging
to keep temperature, CO2 concentration, and vapor pressure deficit
constant with variations light intensity. In traditional light response
curves with sequential changes in light intensity on the same sam-
ple, it is also important to consider the order of changes in light
intensity. A measurement sequence from high to low light intensity
helps to fully induce enzyme activity and stomatal opening at the
 Tools for Measuring in vivo Photosynthesis 17

start, but will also generate substantial non-photochemical quench-

ing (NPQ) or even photoinhibition, leading to decreased φ. Alter-
natively, measurements from low to high light intensity can provide
a non-biased estimate for φ, but require much more time to allow
full enzyme activation and stomatal opening after each increase in
light intensity and rushed measurements would again lead to
underestimation of φ, especially at the intermediate light range.
Both examples emphasize the issue of performing measurements at
different light intensities sequentially on the same sample, which
carries the risk of generating a light history component throughout
the response curve. Subdividing the sample to measure light inten-
sities in parallel can be used to avoid this problem, as was recently
demonstrated by [99]. This method also holds promise to reduce
the time required per response curve from approximately 0.5–1 h
to only a few minutes; however its interpretation is limited to values
derived from chlorophyll fluorescence since it is not able to resolve
leaf gas exchange.

4 The Future of In Vivo Measurements of Photosynthetic Flux

The future of measurements of photosynthetic flux relies on

research that exploits the unprecedented accessibility of current
technology to multiplex and increase the throughput of measure-
ments at diverse scales. As discussed below, these developments will
provide novel insight into basic plant biology and help bring pho-
tosynthetic insights into -omic scale approaches.
The ability to multiplex, or perform simultaneous assays on the
same sample, is a hallmark of genomic- and metabolomic-based
approaches and these have helped to greatly advance these fields
and hold promise to help advance our understanding of photosyn-
thesis. Given the inherently lower throughput limitations of pho-
tosynthetic measurements that require physically clamping a
measuring device to each leaf, it makes sense to combine many
assays in the same device to get as much data from each measure-
ment as possible. Additionally, deeper physiological meaning can be
determined when certain measurements are combined, for example
the determination of the CO2 transfer conductance across the
mesophyll from combined measurements of gas exchange and
chlorophyll fluorescence [100]. Past efforts in multiplexing photo-
synthesis measurements have been limited by the physical size of
each component, but recent advances in LED technology and
market pressure to engineer more compact analytical equipment
have helped alleviate this problem. For example, a recently
launched effort (PhotosynQ) to provide a low-cost, open-source
instrument has produced a handheld device capable of measuring
over a dozen different signatures of photosynthesis during a 15-s
measurement [101]. The PhotosynQ platform additionally uploads
18 Berkley J. Walker et al.

timestamped and geotagged data into a freely accessible database,

allowing leaf-level measurements to be interpreted at multiple
scales. Another example in multiplexing is when measurements of
carbon assimilation have been combined with mass flux balance
approaches to constrain the photosynthetic uptake of carbon in
models of central metabolism [20] and for confirming the stoichi-
ometry of CO2 release from photorespiration [21].
In the past, the limited throughput of tools measuring photo-
synthetic flux has restricted their application from approaches tying
gene to function, such as through parallel quantitative trait locus
mapping and genome-wide association studies [102, 103]. Increas-
ing measurement throughput is also important to properly screen
transgenic plants carrying multigene constructs for increased pho-
tosynthesis in replicated field trials. Measuring in vivo photosyn-
thetic parameters in tandem with other techniques, e.g., including
those discussed herein such as quantifying photosynthesis-related
enzymes [104], purifying RuBisCO for determining catalytic con-
stants and [105] quantifying RuBisCO activity and activation state
[106], evaluating thylakoid lipid content [107], and determining
high-resolution ultrastructure [108] and chloroplast structure
[109], would be useful to more fully understand the physiological
and molecular effects of gene function or environmental changes,
as determined by the experiment. Fortunately, there are several
emergent approaches that may help increase throughput, in addi-
tion to the PhotosynQ platform mentioned previously. For exam-
ple, key photosynthetic parameters derived from measurements of
the response of carbon assimilation to CO2 (maximum rate of
RuBisCO carboxylation and electron transport) correlate strongly
with spectral regions of the fresh leaf hyperspectral reflectance
[110, 111]. Since hyperspectral leaf reflectance measurements
take <1 s as compared to 20–40 min for the standard measurement
by gas exchange, this approach holds great promise for application
for higher throughput phenotyping, yet the mechanism behind this
relationship remains unresolved. Validation of this approach, and of
determining photosynthetic gas exchange parameters in general
from gas exchange, will be aided by the development of non-
steady-state approaches for measuring the response of photosyn-
thetic assimilation to CO2 concentrations [112].

References
1. Wu A, Song Y, van Oosterom EJ, Hammer
GL (2016) Connecting biochemical photo-
synthesis models with crop models to
support crop improvement. Front Plant Sci
7:1518. https://doi.org/10.3389/fpls.
2016.01518
2. De Pury D, Farquhar GD (1997) Simple scal-
ing of photosynthesis from leaves to canopies
without the errors of big-leaf models. Plant
Cell Environ 20(5):537–557
3. von Caemmerer S (2000) Biochemical models
of leaf photosynthesis, Techniques in plant
sciences, vol 2. CSIRO, Collingwood
4. Galmés J, Hermida-Carrera C, Laanisto L,
Niinemets U € (2016) A compendium of tem-
perature responses of Rubisco kinetic traits:
 Tools for Measuring in vivo Photosynthesis
variability among and within photosynthetic
groups and impacts on photosynthesis mod-
eling. J Exp Bot 67(17):5067–5091. https://
doi.org/10.1093/jxb/erw267
5. Hermida-Carrera C, Kapralov MV, Galmés J
(2016) Rubisco catalytic properties and tem-
perature response in crops. Plant Phys 171
(4):2549–2561. https://doi.org/10.1104/
pp.16.01846
6. Orr D, Alcântara A, Kapralov MV,
Andralojc J, Carmo-Silva E, Parry MAJ
(2016) Surveying Rubisco diversity and tem-
perature response to improve crop photosyn-
thetic efficiency. Plant Phys 172(2):707–717.
https://doi.org/10.1104/pp.16.00750
7. Fullana-Pericàs M, Conesa MÀ, Soler S,
Ribas-Carbó M, Granell A, Galmés J (2016)
Variations of leaf morphology, photosynthetic
traits and water-use efficiency in Western-
Mediterranean tomato landraces. Photo-
synthetica 55(1):121–133. https://doi.org/
10.1007/s11099-016-0653-4
8. Gu J, Yin X, Stomph T-J, Wang H, Struik PC
(2012) Physiological basis of genetic variation
in leaf photosynthesis among rice (Oryza
sativa L.) introgression lines under drought
and well-watered conditions. J Exp Bot 63
(14):5137–5153. https://doi.org/10.
1093/jxb/ers170
9. Sun J, Sun J, Feng Z (2015) Modelling pho-
tosynthesis in flag leaves of winter wheat Tri-
ticum aestivum considering the variation in
photosynthesis parameters during develop-
ment. Funct Plant Biol 42(11):1036–1044.
https://doi.org/10.1071/FP15140
10. Gaastra P (1959) Photosynthesis of crop
plants as influenced by light, carbon dioxide,
temperature, and stomatal diffusion resis-
tance. Wageningen University, Wageningen:
Veenman
11. Farquhar GD, von Caemmerer S, Berry JA
(1980) A biochemical model of photosyn-
thetic CO2 assimilation in leaves of C3 species.
Planta 149(1):78–90. https://doi.org/10.
1007/bf00386231
12. von Caemmerer S, Farquhar GD (1981)
Some relationships between the biochemistry
of photosynthesis and the gas exchange of
leaves. Planta 153(4):376–387
13. Harley P, Sharkey T (1991) An improved
model of C3 photosynthesis at high CO2:
reversed O2 sensitivity explained by lack of
glycerate reentry into the chloroplast. Photo-
synth Res 27(3):169–178
14. Busch FA (2017) Photosynthetic gas
exchange in land plants at the leaf level. In:
19
Covshoff S (ed) Photosynthesis: methods and
protocols, vol 1. Springer Press, New York
15. Farquhar G, Wong S (1984) An empirical
model of stomatal conductance. Funct Plant
Biol 11(3):191–210. https://doi.org/10.
1071/PP9840191
16. von Caemmerer S, Evans J, Hudson G,
Andrews T (1994) The kinetics of ribulose-
1, 5-bisphosphate carboxylase/oxygenase
in vivo inferred from measurements of photo-
synthesis in leaves of transgenic tobacco.
Planta 195(1):88–97
17. Bernacchi CJ, Pimentel C, Long SP (2003) In
vivo temperature response functions of para-
meters required to model RuBP-limited pho-
tosynthesis. Plant Cell Environ 26
(9):1419–1430. https://doi.org/10.1046/j.
0016-8025.2003.01050.x
18. Walker BJ, Ariza LS, Kaines S, Badger MR,
Cousins AB (2013) Temperature response of
in vivo Rubisco kinetics and mesophyll con-
ductance in Arabidopsis thaliana: compari-
sons to Nicotiana tabacum. Plant Cell
Environ 36(12):2108–2119. https://doi.
org/10.1111/pce.12166
19. Busch FA, Sage TL, Cousins AB, Sage RF
(2013) C3 plants enhance rates of photosyn-
thesis by reassimilating photorespired and
respired CO2. Plant Cell Environ 36
(1):200–212. https://doi.org/10.1111/j.
1365-3040.2012.02567.x
20. Ma F, Jazmin LJ, Young JD, Allen DK (2014)
Isotopically nonstationary (13)C flux analysis
of changes in Arabidopsis thaliana leaf metab-
olism due to high light acclimation. Proc Natl
Acad Sci U S A 111(47):16967–16972.
https://doi.org/10.1073/pnas.1319485111
21. Abadie C, Boex-Fontvieille ERA, Carroll AJ,
Tcherkez G (2016) In vivo stoichiometry of
photorespiratory metabolism. Nat Plants
2:15220
22. George GM, Kölling K, Kuenzli R, Hirsch-
Hoffmann M, Flütsch P, Zeeman SC (2017)
Design and use of a digitally-controlled device
for accurate, multiplexed gas exchange mea-
surements of the complete foliar parts of
plants. In: Covshoff S (ed) Photosynthesis:
methods and protocols, vol 1. Springer
Press, New York
23. Dutton RG, Jiao J, Tsujita MJ, Grodzinski B
(1988) Whole plant CO2 exchange measure-
ments for nondestructive estimation of
growth. Plant Phys 86(2):355–358. https://
doi.org/10.1104/pp.86.2.355
24. Zeeman SC, Rees TA (1999) Changes in car-
bohydrate metabolism and assimilate export in
20 Berkley J. Walker et al.
starch-excess mutants of Arabidopsis. Plant Cell
Environ 22(11):1445–1453. https://doi.org/
10.1046/j.1365-3040.1999.00503.x
25. Baldocchi DD, Amthor JS (2001) Canopy
photosynthesis: history, measurements, and
models. In: Roy J, Mooney HA, Saugier B
(eds) Terrestrial global productivity. Aca-
demic Press, USA, pp 9–31
26. Hall DO, Scurlock J, Bolhar-Nordenkampf H,
Leegood RC, Long S (1993) Photosynthesis
and production in a changing environment: a
field and laboratory manual. Chapman &
Hall, London
27. Kölling K, George GM, Künzli R, Flütsch P,
Zeeman SC (2015) A whole-plant chamber
system for parallel gas exchange measure-
ments of Arabidopsis and other herbaceous
species. Plant Methods 11(1):48. https://
doi.org/10.1186/s13007-015-0089-z
28. Barton CVM, Ellsworth DS, Medlyn BE,
Duursma RA, Tissue DT, Adams MA,
Eamus D, Conroy JP, McMurtrie RE,
Parsby J, Linder S (2010) Whole-tree cham-
bers for elevated atmospheric CO2 experi-
mentation and tree scale flux measurements
in south-eastern Australia: The Hawkesbury
Forest Experiment. Agric For Meteorol 150
(7–8):941–951. https://doi.org/10.1016/j.
agrformet.2010.03.001
29. Drake JE, Tjoelker MG, Aspinwall MJ, Reich
PB, Barton CVM, Medlyn BE, Duursma RAC
(2016) Does physiological acclimation to cli-
mate warming stabilize the ratio of canopy
respiration to photosynthesis? New Phytol
211(3):850–863. https://doi.org/10.1111/
nph.13978
30. Baldocchi DD (2003) Assessing the eddy
covariance technique for evaluating carbon
dioxide exchange rates of ecosystems: past,
present and future. Glob Chang Biol 9
(4):479–492. https://doi.org/10.1046/j.
1365-2486.2003.00629.x
31. Song Q, Zhu X-G (2017) Measuring canopy
gas exchange using CAnopy Photosynthesis
and Transpiration Systems (CAPTS). In: Cov-
shoff S (ed) Photosynthesis: methods and
protocols. Springer Press, New York
32. Schmid HP (1994) Source areas for scalars
and scalar fluxes. Boundary-Layer Meteorol
67(3):293–318. https://doi.org/10.1007/
bf00713146
33. Raupach MR (1979) Anomalies in flux-
gradient relationships over forest. Boundary-
Layer Meteorol 16(3):467–486. https://doi.
org/10.1007/bf03335385
34. Simpson IJ, Thurtell GW, Neumann HH,
Den Hartog G, Edwards GC (1998) The
validity of similarity theory in the roughness
sublayer above forests. Boundary-Layer
Meteorol 87(1):69–99. https://doi.org/10.
1023/a:1000809902980
35. Severinghaus JW (2002) The invention and
development of blood gas analysis apparatus.
Anesthesiology 97(1):253–256
36. Clark LC, Lyons C (1962) Electrode systems
for continuous monitoring in cardiovascular
surgery. Ann N Y Acad Sci 102(1):29–45.
https://doi.org/10.1111/j.1749-6632.
1962.tb13623.x
37. Shevela D, Schröder WP, Messinger J (2017)
Liquid-phase measurements of photosyn-
thetic oxygen evolution. In: Covshoff S
(ed) Photosynthesis: methods and protocols.
Springer Press, New York
38. Driever SM, Baker NR (2017) Measurement
of O2 uptake and evolution in leaves in vivo
using stable isotopes and membrane inlet
mass spectrometry. In: Covshoff S
(ed) Photosynthesis: methods and protocols,
vol 1. Springer Press, New York
39. Hill JF, Govindjee (2014) The controversy
over the minimum quantum requirement for
oxygen evolution. Photosynth Res 122
(1):97–112. https://doi.org/10.1007/
s11120-014-0014-8
40. Ubierna N, Holloway-Phillips M-M, Farqu-
har GD (2017) Using carbon stable isotopes
to study C3 and C4 photosynthesis: models
and calculations. In: Covshoff S
(ed) Photosynthesis: methods and protocols.
Springer Press, New York
41. Evans JR, Kaldenhoff R, Genty B, Terashima I
(2009) Resistances along the CO2 diffusion
pathway inside leaves. J Exp Bot 60
(8):2235–2248. https://doi.org/10.1093/
jxb/erp117
42. Cernusak LA, Ubierna N, Winter K, Holtum
JAM, Marshall JD, Farquhar GD (2013)
Environmental and physiological determi-
nants of carbon isotope discrimination in ter-
restrial plants. New Phytol 200(4):950–965.
https://doi.org/10.1111/nph.12423
43. Raven JA, Beardall J (2016) The ins and outs
of CO2. J Exp Bot 67(1):1–13
44. Barbour MMC (2016) Understanding regu-
lation of leaf internal carbon and water trans-
port using online stable isotope techniques.
New Phytol 213(1):83–88. https://doi.org/
10.1111/nph.14171
45. Canvin DT, Berry JA, Badger MR, Fock H,
Osmond CB (1980) Oxygen exchange in
leaves in the light. Plant Phys 66
(2):302–307. https://doi.org/10.1104/pp.
66.2.302
 Tools for Measuring in vivo Photosynthesis
46. Badger MR (1985) Photosynthetic oxygen
exchange. Annu Rev Plant Physiol 36
(1):27–53. https://doi.org/10.1146/
annurev.pp.36.060185.000331
47. Haupt-Herting S, Klug K, Fock HP (2001) A
new approach to measure gross CO2 fluxes in
leaves. Gross CO2 sssimilation, photorespira-
tion, and mitochondrial respiration in the
light in tomato under drought stress. Plant
Phys 126(1):388–396. https://doi.org/10.
1104/pp.126.1.388
48. Maxwell K, Badger M, Osmond C (1998) A
comparison of CO2 and O2 exchange patterns
and the relationship with chlorophyll fluores-
cence during photosynthesis in C3 and CAM
plants. Aust J Plant Physiol 25(1):45–52
49. Ruuska SA, Badger MR, Andrews TJ, von
Caemmerer S (2000) Photosynthetic electron
sinks in transgenic tobacco with reduced
amounts of Rubisco: little evidence for signif-
icant Mehler reaction. J Exp Bot 51:357–368.
https://doi.org/10.1093/jexbot/51.suppl_
1.357
50. Driever SM, Baker NR (2011) The water–-
water cycle in leaves is not a major alternative
electron sink for dissipation of excess excita-
tion energy when CO2 assimilation is
restricted. Plant Cell Environ 34
(5):837–846. https://doi.org/10.1111/j.
1365-3040.2011.02288.x
51. Peltier G, Cournac L, Despax V, Dimon B,
Fina L, Genty B, Rumeau D (1995) Carbonic
anhydrase activity in leaves as measured in vivo
by 18O exchange between carbon dioxide and
water. Planta 196(4):732–739. https://doi.
org/10.1007/bf01106768
52. Baker NR (2008) Chlorophyll fluorescence: a
probe of photosynthesis in vivo. Annu Rev
Plant Biol 59(1):89–113. https://doi.org/
10.1146/annurev.arplant.59.032607.
092759
53. Maxwell K, Johnson G (2000) Chlorophyll
fluorescence-a practical guide. J Exp Bot 51
(345):659–668
54. Butler WL (1978) Energy distribution in the
photochemical apparatus of photosynthesis.
Annu Rev Plant Physiol 29(1):345–378
55. Genty B, Briantais J-M, Baker NR (1989) The
relationship between the quantum yield of
photosynthetic electron transport and
quenching of chlorophyll fluorescence. Bio-
chim Biophys Acta Gen Subj 990(1):87–92.
https://doi.org/10.1016/S0304-4165(89)
80016-9
56. Bradbury M, Baker NR (1981) Analysis of the
slow phases of the in vivo chlorophyll fluores-
cence induction curve. Changes in the redox
21
state of Photosystem II electron acceptors and
fluorescence emission from Photosystems I
and II. Biochim Biophys Acta Bioenerg 635
(3):542–551. https://doi.org/10.1016/
0005-2728(81)90113-4
57. Loriaux SD, Avenson TJ, Welles JM, McDer-
mitt DK, Eckles RD, Riensche B, Genty B
(2013) Closing in on maximum yield of chlo-
rophyll fluorescence using a single multiphase
flash of sub-saturating intensity. Plant Cell
Environ 36(10):1755–1770. https://doi.
org/10.1111/pce.12115
58. Schreiber U, Neubauer C (1987) The poly-
phasic rise of chlorophyll fluorescence upon
onset of strong continuous illumination:
II. Partial control by the Photosystem II
donor side and possible ways of interpreta-
tion. Zeitschrift für Naturforschung C 42
(11–12):255–1264. https://doi.org/10.
1515/znc-1987-11-1218
59. Ajigboye OO, Ray RV, Murchie EH (2017)
Chlorophyll fluorescence on the fast time-
scale. In: Covshoff S (ed) Photosynthesis:
methods and protocols. Springer Press,
New York
60. Murchie EH, Lawson T (2013) Chlorophyll
fluorescence analysis: a guide to good practice
and understanding some new applications. J
Exp Bot 64(13):3983–3998. https://doi.
org/10.1093/jxb/ert208
61. Baker NR, Rosenqvist E (2004) Applications
of chlorophyll fluorescence can improve crop
production strategies: an examination of
future possibilities. J Exp Bot 55
(403):1607–1621. https://doi.org/10.
1093/jxb/erh196
62. Fryer MJ, Oxborough K, Martin B, Ort DR,
Baker NR (1995) Factors associated with
depression of photosynthetic quantum effi-
ciency in maize at low growth temperature.
Plant Phys 108(2):761–767
63. Fracheboud Y, Leipner J (2003) The applica-
tion of chlorophyll fluorescence to study light,
temperature, and drought stress. In: DeEll
JR, Toivonen PMA (eds) Practical applica-
tions of chlorophyll fluorescence in plant biol-
ogy. Springer, Boston, MA, pp 125–150.
https://doi.org/10.1007/978-1-4615-
0415-3_4
64. Gray GR, Chauvin LP, Sarhan F, Huner N
(1997) Cold acclimation and freezing toler-
ance (A complex interaction of light and tem-
perature). Plant Phys 114(2):467–474.
https://doi.org/10.1104/pp.114.2.467
65. Avenson TJ, Saathoff AJ (2017)
Sub-saturating multiphase flash irradiances
to estimate maximum fluorescence yield. In:
22 Berkley J. Walker et al.
Covshoff S (ed) Photosynthesis: methods and
protocols. Springer Press, New York
66. Poulson ME, Edwards GE, Browse J (2002)
Photosynthesis is limited at high leaf to air
vapor pressure deficit in a mutant of Arabi-
dopsis thaliana that lacks trienoic fatty acids.
Photosynth Res 72(1):55–63. https://doi.
org/10.1023/a:1016054027464
67. Rizza F, Pagani D, Stanca AM, Cattivelli L
(2001) Use of chlorophyll fluorescence to
evaluate the cold acclimation and freezing tol-
erance of winter and spring oats. Plant Breed
120(5):389–396. https://doi.org/10.1046/
j.1439-0523.2001.00635.x
68. Lawson T, Oxborough K, Morison JIL, Baker
NR (2002) Responses of photosynthetic elec-
tron transport in stomatal guard cells and
mesophyll cells in intact leaves to light, CO2,
and humidity. Plant Phys 128(1):52–62.
https://doi.org/10.1104/pp.010317
69. Lawson T, Oxborough K, Morison JIL, Baker
NR (2003) The responses of guard and meso-
phyll cell photosynthesis to CO2, O2, light,
and water stress in a range of species are simi-
lar. J Exp Bot 54(388):1743–1752. https://
doi.org/10.1093/jxb/erg186
70. Baker NR, Oxborough K, Lawson T, Morison
JIL (2001) High resolution imaging of pho-
tosynthetic activities of tissues, cells and chlor-
oplasts in leaves. J Exp Bot 52(356):615–621.
https://doi.org/10.1093/jxb/52.356.615
71. Simkin AJ, McAusland L, Headland LR,
Lawson T, Raines CA (2015) Multigene
manipulation of photosynthetic carbon assim-
ilation increases CO2 fixation and biomass
yield in tobacco. J Exp Bot 66
(13):4075–4090. https://doi.org/10.1093/
jxb/erv204
72. Barbagallo RP, Oxborough K, Pallett KE,
Baker NR (2003) Rapid, noninvasive screen-
ing for perturbations of metabolism and plant
growth using chlorophyll fluorescence imag-
ing. Plant Phys 132(2):485–493. https://doi.
org/10.1104/pp.102.018093
73. Caspari O, Meyer M, Tolleter D, Wittkopp T,
Cunniffe N, Lawson T, Grossman A, Griffiths
H (2017) Pyrenoid loss in Chlamydomonas
reinhardtii causes limitations in CO2 supply,
but not thylakoid operating efficiency. J Exp
Bot 68(14):3903–3913
74. Oxborough K (2004) Imaging of chlorophyll
a fluorescence: theoretical and practical
aspects of an emerging technique for the
monitoring of photosynthetic performance. J
Exp Bot 55(400):1195–1205. https://doi.
org/10.1093/jxb/erh145
75. von Caemmerer S, Lawson T, Oxborough K,
Baker NR, Andrews TJ, Raines CA (2004)
Stomatal conductance does not correlate
with photosynthetic capacity in transgenic
tobacco with reduced amounts of Rubisco. J
Exp Bot 55(400):1157–1166. https://doi.
org/10.1093/jxb/erh128
76. Lawson T, Lefebvre S, Baker NR, Morison
JIL, Raines CA (2008) Reductions in meso-
phyll and guard cell photosynthesis impact on
the control of stomatal responses to light and
CO2. J Exp Bot 59(13):3609–3619. https://
doi.org/10.1093/jxb/ern211
77. McAusland L, Vialet-Chabrand SRM, Mat-
thews JSA, Lawson T (2015) Spatial and tem-
poral responses in stomatal behaviour,
photosynthesis and implications for water-
use efficiency. In: Mancuso S, Shabala S (eds)
Rhythms in plants: dynamic responses in a
dynamic environment. Springer International
Publishing, Cham, pp 97–119. https://doi.
org/10.1007/978-3-319-20517-5_5
78. Morison JIL, Gallouët E, Lawson T,
Cornic G, Herbin R, Baker NR (2005) Lat-
eral diffusion of CO2 in leaves is not sufficient
to support photosynthesis. Plant Phys 139
(1):254–266. https://doi.org/10.1104/pp.
105.062950
79. Badger MR, Fallahi H, Kaines S, Takahashi S
(2009) Chlorophyll fluorescence screening of
Arabidopsis thaliana for CO2 sensitive photo-
respiration and photoinhibition mutants.
Funct Plant Biol 36(11):867–873. https://
doi.org/10.1071/FP09199
80. Lawson T, Vialet-Chabrand S (2017) Chloro-
phyll fluorescence imaging. In: Covshoff S
(ed) Photosynthesis: Methods and Protocols,
vol 1. Springer Press, New York
81. Baker NR, Harbinson J, Kramer DM (2007)
Determining the limitations and regulation of
photosynthetic energy transduction in leaves.
Plant Cell Environ 30(9):1107–1125.
https://doi.org/10.1111/j.1365-3040.
2007.01680.x
82. Harbinson J, Woodward FI (1987) The use of
light-induced absorbance changes at 820 nm
to monitor the oxidation state of P-700 in
leaves. Plant Cell Environ 10(2):131–140.
https://doi.org/10.1111/1365-3040.
ep11602090
83. Sacksteder C, Kramer D (2000) Dark-interval
relaxation kinetics (DIRK) of absorbance
changes as a quantitative probe of steady-
state electron transfer. Photosynth Res 66
(1):145–158
84. Cruz JA, Sacksteder CA, Kanazawa A, Kramer
DM (2001) Contribution of electric field
(Δψ) to steady-state transthylakoid proton

motive force (pmf) in vitro and in vivo. Con-
trol of pmf parsing into Δψ and ΔpH by ionic
strength. Biochemistry 40(5):1226–1237.
https://doi.org/10.1021/bi0018741
85. Kramer DM, Evans JR (2011) The impor-
tance of energy balance in improving photo-
synthetic productivity. Plant Phys 155
(1):70–78. https://doi.org/10.1104/pp.
110.166652
86. Walker BJ, Strand DD, Kramer DM, Cousins
AB (2014) The response of cyclic electron
flow around photosystem I to changes in pho-
torespiration and nitrate assimilation. Plant
Phys 165(1):453–462. https://doi.org/10.
1104/pp.114.238238
87. Finazzi G, Rappaport F, Furia A,
Fleischmann M, Rochaix JD, Zito F, Forti G
(2002) Involvement of state transitions in the
switch between linear and cyclic electron flow
in Chlamydomonas reinhardtii. EMBO Rep 3
(3):280–285. https://doi.org/10.1093/
embo-reports/kvf047
88. Avenson TJ, Cruz JA, Kanazawa A, Kramer
DM (2005) Regulating the proton budget of
higher plant photosynthesis. Proc Natl Acad
Sci U S A 102(27):9709–9713. https://doi.
org/10.1073/pnas.0503952102
89. Laisk A, Oja V (1994) Range of photosyn-
thetic control of postillumination P700þ
reduction rate in sunflower leaves. Photo-
synth Res 39(1):39–50. https://doi.org/10.
1007/bf00027141
90. Harbinson J (1994) The responses of thyla-
koid electron transport and light utilisation
efficiency to sink limitation of electron trans-
port. In: Baker NR, Bowyer J (eds) Photoin-
hibition of photosynthesis. Bios Scientific
Publishers, Oxford, pp 273–295
91. Kramer DM, Avenson TJ, Edwards GE
(2004) Dynamic flexibility in the light reac-
tions of photosynthesis governed by both
electron and proton transfer reactions. Trends
Plant Sci 9(7):349–357
92. Cruz JA, Avenson TJ, Kanazawa A,
Takizawa K, Edwards GE, Kramer DM
(2005) Plasticity in light reactions of photo-
synthesis for energy production and photo-
protection. J Exp Bot 56(411):395–406.
https://doi.org/10.1093/jxb/eri022
93. Coe RA, Lin H (2017) Light-response curves
in land plants. In: Covshoff
(ed) Photosynthesis: methods and protocols.
Springer Press, New York
94. Emerson R, Chalmers RV (1958) Specula-
tions concerning the function and phyloge-
netic significance of the accessory pigments
of algae. Phycol Soc News Bull 11:51–56
Tools for Measuring in vivo Photosynthesis 23
95. Hill R, Bendall F (1960) Function of the two
cytochrome components in chloroplasts: A
working hypothesis. Nature 186
(4719):136–137
96. Kromdijk J, Głowacka K, Leonelli L, Gabilly
ST, Iwai M, Niyogi KK, Long SP (2016)
Improving photosynthesis and crop produc-
tivity by accelerating recovery from photopro-
tection. Science 354(6314):857–861.
https://doi.org/10.1126/science.aai8878
97. Singsaas E, Ort D, DeLucia E (2001) Varia-
tion in measured values of photosynthetic
quantum yield in ecophysiological studies.
Oecologia 128(1):15–23. https://doi.org/
10.1007/s004420000624
98. Skillman JB (2008) Quantum yield variation
across the three pathways of photosynthesis:
not yet out of the dark. J Exp Bot 59
(7):1647–1661. https://doi.org/10.1093/
jxb/ern029
99. Serôdio J, Ezequiel J, Frommlet J, Laviale M,
Lavaud J (2013) A method for the rapid gen-
eration of nonsequential light-response curves
of chlorophyll fluorescence. Plant Phys 163
(3):1089–1102. https://doi.org/10.1104/
pp.113.225243
100. Loreto F, Harley PC, Di Marco G, Sharkey
TD (1992) Estimation of mesophyll conduc-
tance to CO2 flux by three different methods.
Plant Phys 98(4):1437–1443. https://doi.
org/10.1104/pp.98.4.1437
101. Kuhlgert S, Austic G, Zegarac R, Osei-Bonsu-
I, Hoh D, Chilvers MI, Roth MG, Bi K,
TerAvest D, Weebadde P, Kramer DM
(2016) MultispeQ Beta: a tool for large-scale
plant phenotyping connected to the open
PhotosynQ network. R Soc Open Sci 3
(10):160592. https://doi.org/10.1098/
rsos.160592
102. Korte A, Farlow A (2013) The advantages
and limitations of trait analysis with GWAS: a
review. Plant Methods 9(1):29. https://doi.
org/10.1186/1746-4811-9-29
103. Koornneef M, Alonso-Blanco C, Vreugdenhil
D (2004) Naturally occurring genetic varia-
tion in Arabidopsis thaliana. Annu Rev Plant
Biol 55:141–172
104. Perdomo JA, Sales CRG, Carmo-Silva E
(2017) Quantification of photosynthetic
enzymes in leaf extracts by immunoblotting.
S In: Covshoff S (ed) Photosynthesis: methods
and protocols, vol 1. Springer Press,
New York
105. Orr DJ, Carmo-Silva E (2017) Extraction of
Rubisco to determine catalytic constants. In:
Covshoff S (ed) Photosynthesis: methods and
protocols, vol 1. Springer Press, New York
24 Berkley J. Walker et al.
106. Sales CRG, Degen GE, da Silva AB, Carmo-
Silva E (2017) Spectrophotometric determi-
nation of Rubisco activity and activation state
in leaf extracts. In: Covshoff S send PA (2011) Leaf optical properties reflect
(ed) Photosynthesis: methods and protocols,
vol 1. Springer Press, New York
107. Kirchhoff H, Yarbrough R (2017) Evaluation
of lipids for the study of photosynthetic mem-
branes. In: Covshoff S (ed) Photosynthesis:
methods and protocols, vol 1. Springer
Press, New York
108. Palmer WM, Flynn JR, Martin AP, Reed SL,
Grof CPL, White RG, Furbank RT (2017)
3D clearing and molecular labeling in plant
tissues. In: Covshoff S (ed) Photosynthesis:
methods and protocols, vol 1. Springer
Press, New York
109. Khoshravesh R, Sage TL (2017) Creating leaf
cell suspensions for characterization of meso-
phyll and bundle sheath cellular features. In:
Covshoff S (ed) Photosynthesis: methods and
protocols, vol 1. Springer Press, New York
110. Serbin SP, Dillaway DN, Kruger EL, Town-
variation in photosynthetic metabolism and
its sensitivity to temperature. J Exp Bot 63
(1):489–502
111. Yendrek C, Tomaz T, Montes CM, Cao Y,
Morse AM, Brown PJ, McIntyre L,
Leakey A, Ainsworth E (2016) High-
throughput phenotyping of maize leaf physi-
ology and biochemistry using hyperspectral
reflectance. Plant Phys. https://doi.org/10.
1104/pp.16.01447
112. Stinziano JR, Morgan PB, Lynch DJ, Saathoff
AJ, McDermitt DK, Hanson DT (2017) The
rapid A–Ci response: photosynthesis in the
phenomic era. Plant Cell Environ 40
(8):1256–1262. https://doi.org/10.1111/
pce.12911
 Chapter 2

Photosynthetic Gas Exchange in Land Plants

at the Leaf Level
Florian A. Busch

Abstract
Leaf-level gas exchange enables insights into the physiology and in vivo biochemical processes of plants.
Advances in infrared gas analysis have resulted in user-friendly off-the-shelf gas exchange systems that allow
researchers to collect physiological measurements with the push of a few buttons. Here, I describe how to
set up the gas exchange equipment and what to pay attention to while making measurements, and provide
some guidelines on how to analyze and interpret the data obtained.

Key words Gas exchange, Photosynthesis, Transpiration, Conductance, FvCB model, RuBisCO,
Mitochondrial respiration, Infrared gas analyzer

1 Introduction

Leaf-level gas exchange has been established as a reliable tool to

estimate the photosynthetic performance of plants. Modern equip-
ment has made it easy to measure gas exchange on leaves and collect
data “right out of the box.” While the user-friendly handling of the
instruments is encouraging more and more researchers to enter the
area of plant physiology, the principles of gas exchange, technical
details to consider during the measurements, and interpretation of
the obtained data remain complex. The protocol for setting up and
operating the gas exchange equipment will vary depending on the
machine used, and can be looked up in the manual provided by the
manufacturer [1–3]. However, some details are not often described
in the manuals, as they will depend on the questions asked by the
user and be affected by the experimental necessities, and are there-
fore not easily discussed comprehensively. Here, I provide some
general advice on the practical aspects of gas exchange that usually
come with experience, to help jump-start novice users in their
attempt to collect useful data. Additional information and guide-
lines for measuring photosynthetic gas exchange, such as how to

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_2, © Springer Science+Business Media, LLC, part of Springer Nature 2018

25
26 Florian A. Busch

Table 1
List of parameters and their units

Parameter Description Unit

A or Anet Net CO2 assimilation rate μmol m
2 s
1
Asat Light-saturated CO2 assimilation rate μmol m
2 s
1
Amax Maximum CO2 assimilation rate (at high light and μmol m
2 s
1
saturated CO2 concentration)
Ca Ambient CO2 concentration μmol mol
1
Ci Intercellular CO2 concentration μmol mol
1
E Rate of transpiration mmol m
2 s
1
gsc Stomatal conductance to CO2 mol m
2 s
1
gm Mesophyll conductance mol m
2 s
1 bar
1
J Potential electron transport rate μmol m
2 s
1
Jmax Maximum electron transport rate μmol m
2 s
1
Vcmax Maximum rate of carboxylation by RuBisCO μmol m
2 s
1
Vc Rate of carboxylation of RuBP by RuBisCO μmol m
2 s
1
Vo Rate of oxygenation of RuBP by RuBisCO μmol m
2 s
1
VPD Vapor pressure deficit kPa
Rd Rate of mitochondrial respiration in the light μmol m
2 s
1

Tleaf Leaf temperature C
Tp Rate of triose phosphate utilization μmol m
2 s
1
Γ CO2 compensation point μmol mol
1
Γ* CO2 compensation point in the absence of mitochondrial respiration μmol mol
1

assess photorespiration or mesophyll conductance, can be found for

example in [4–6].
Commercially available infrared gas analyzers (IRGA) measure
CO2 and H2O concentrations via the absorbance of characteristic
wavelengths in the infrared spectrum. The differences in gas con-
centrations between a sample cell containing a leaf and an empty
reference cell, together with the flow rate of the air through the
chambers, are then used by the gas exchange equipment to deter-
mine the rate of net photosynthetic CO2 uptake of leaves (Anet; a
list of parameters can be found in Table 1) concomitant with the
rate of water loss through transpiration (E). The gas exchange
equipment uses E to derive a value of stomatal conductance for
CO2 (gsc), which is a measure of the ease by which CO2 can diffuse
through the stomata from the outside of the leaf to the substomatal
 Photosynthetic Gas Exchange at the Leaf Level 27

cavity. Because stomata represent a resistance to CO2 diffusion into

the leaf, the CO2 concentration in the intercellular space (Ci) is
different from the CO2 concentration surrounding the leaf (Ca),
and can be calculated according to Fick’s law as
A net
Ci ¼ Ca
ð1Þ
g sc

Based on the parameters Anet and Ci determined from the

photosynthetic gas exchange measurements, photosynthetic mod-
els can be used to provide deep insights into the leaf biochemistry,
such as the maximum rate of carboxylation by RuBisCO (Vcmax).
The relation between photosynthetic biochemistry and Anet has
been described by the widely used Farquhar, von Caemmerer, and
Berry (FvCB) model, which centers on the biochemical properties
of the CO2-fixing enzyme ribulose-1,5-bisphosphate carboxylase/
oxygenase (RuBisCO) [7]. RuBisCO is a dual-functioning enzyme
that uses CO2 to carboxylate its substrate ribulose
1,5-bisphosphate (RuBP), and also performs a competing reaction
involving O2 that causes the oxygenation of RuBP. The rates of
carboxylation and oxygenation are denoted as Vc and Vo, respec-
tively. The oxygenation reaction of RuBP results in a subsequent
release of CO2 in a process called photorespiration. The FvCB
model [7] accounts for the uptake of CO2 determined by the
carboxylation reaction and the release of CO2 from photorespira-
tion and mitochondrial respiration (Rd) to derive Anet of the leaf as
A net ¼ V c
0:5V o
Rd ð2Þ

Here, the factor 0.5 assumes that 0.5 moles of CO2 are released
for every mole of oxygenation reactions. Vc and Vo can be related to
the CO2 concentration at which the rate of CO2 uptake equals the
rate of CO2 release from photorespiration (Γ*) by

V o 2Γ∗
¼ ð3Þ
Vc C
where C denotes the CO2 concentration at the site of carbox-
ylation. Anet can then be modeled by combining Eqs. (2 and 3) as
 
 Γ∗
A net ¼ min W c , W j , W p 1

Rd ð4Þ
C

where Wc, Wj, and Wp describe the potential rates of carboxyl-

ation that can be achieved under the limiting biochemical processes
restricting photosynthesis in the leaf: the rate of RuBP consump-
tion by RuBisCO (Wc) and the regeneration capacity of RuBP in
the Calvin-Benson cycle, which can be limited by the availability of
NADPH generated by the photosynthetic electron transport chain
(Wj) or ATP (Wp). The net CO2 assimilation rates that would be
28 Florian A. Busch

possible under these three limitations are denoted as Ac, Aj, and Ap,
respectively, and the actual CO2 assimilation rate Anet corresponds
to the minimum of these three rates. The potential rates of carbox-
ylation are given by

V cmax C
Wc ¼ ð5Þ
C þ K c ð1 þ O=K o Þ

depending on Vcmax, the Michaelis-Menten constants of

RuBisCO for CO2 (Kc) and O2 (Ko), and the O2 concentration
(O):

JC
Wj ¼ ð6Þ
4C þ 8Γ∗
where J is the rate of photosynthetic electron transport
and
3T p C
Wp ¼ ð7Þ
C
ð1 þ 3αÞΓ∗

which depends on the rate of triose phosphate utilization (Tp).

In this equation, α denotes the fraction of glycolate carbon that
leaves the photorespiratory pathway as amino acids and is not
returned to the chloroplast [8, 9]. In its simplest form, when the
photorespiratory pathway is fully closed, α ¼ 0.
Leaf CO2 uptake can be modeled under a wide range of envi-
ronmental conditions, if the parameters of the model are known.
Some of these parameters described above are usually measured
independently, such as Kc, Ko, and Γ*. Others, such as Vcmax, J, Tp,
Rd, and α, can be estimated with gas exchange from the response of
Anet to Ci (referred to as the A/Ci curve). This is why A/Ci curves
belong to the standard set of measurements used by plant physiol-
ogists. In general, increasing Ci values will cause the underlying
biochemical limitation of A to shift from Ac at low Ci values, via Aj,
to Ap at high Ci values, although any of these limitations may be
missing in a particular A/Ci response (see Fig. 1). An analysis like
this involves the simplifying assumption that C equals Ci. However,
just as stomata impose a resistance for CO2 diffusion, so does the
mesophyll [10]. A parameter called mesophyll conductance (gm)
can account for this difference in CO2 concentrations, but it is
difficult to estimate. Some fitting routines for analyzing A/Ci
data have gm as one of their outputs (e.g., [11]), but it is best to
have independent estimates of gm using, e.g., carbon isotope dis-
crimination techniques.
Below I describe how to set up the gas exchange equipment,
take measurements, and give some insights into how to fit the
photosynthetic model described above to the collected data.
 Photosynthetic Gas Exchange at the Leaf Level 29

40

30

A ( mol m-2 s-1)

20 Ac
Aj
Ap; =0
10
Ap; = 0.3
Anet
(Γ*, −Rd)
0

0 250 500 750 1000 1250

Ci ( mol mol-1)

Fig. 1 A/Ci response modeled with the equations outlined in the introduction. The colored lines refer to the
potential rates of photosynthesis that can be achieved under the three biochemical limitations (red: Ac; green:
Aj; orange: Ap). Ap is often insensitive to a change in CO2 concentration (α ¼ 0; see text), but sometimes a
decrease in Anet can be observed with increasing Ci (0 < α < 1). The actual rate of photosynthesis (Anet) at any
given Ci is equal to the minimum of Ac, Aj, and Ap. Ac and Aj intersect at a CO2 concentration corresponding to
Γ*, at which Anet represents -Rd (indicated by arrow)

2 Materials

1. Open path infrared gas analyzer (IRGA; see Note 1).

2. Light source (see Note 2).
3. CO2 mixer assembly (see Note 3).
4. Tripod to mount the IRGA head.
5. CO2 cartridges.
6. Soda lime (used as CO2 scrub).
7. Desiccant (such as anhydrous calcium sulfate or silica gel).
8. Power supply and/or batteries (see Note 4).
9. Computer for data download and processing.
Optional materials:
10. Humidifying granules (Stuttgarter Masse; for instruments that
are capable of humidifying the airstream in addition to drying
it).
11. Squirt bottle with distilled water (for wetting the soda lime, or
Stuttgarter Masse, if applicable).
12. CO2 tank with regulator and appropriate fittings to connect to
the machine (for measurements exceeding the time a CO2
cartridge lasts, or for extended stationary laboratory use).
30 Florian A. Busch

3 Methods

3.1 Machine Setup 1. Connect the IRGA head to the console according to the
and Calibration manufacturer’s instructions. It is best to set up the equipment
in an indoor location with little temperature fluctuations
(see Note 5).
2. Unless operating in the field for survey measurements, attach
IRGA head to a tripod (see Note 6).
3. Fill scrub bottles with fresh soda lime and desiccant (see Note 7),
and if applicable add water to Stuttgarter Masse.
4. Turn on the machine and perform all start-up checks recom-
mended by the manufacturer. These daily checks should
include some form of items 5–16 of this list.
5. Use the desiccant and soda lime columns to scrub water and
CO2 from the air by turning them on full scrub to prepare the
IRGA for a calibration check.
6. Check the temperature values, in particular Tleaf. Make sure
that Tleaf is accurate and responding to temperature changes
(see Note 8).
7. Check the proper functioning of the light source, flow control,
and sensors, including the light and pressure sensors.
8. With the chamber closed and no leaf inserted, check H2O
zero of the IRGA with the desiccant and soda lime on full
scrub (see Note 9).
9. If necessary, perform a H2O zero calibration. This should only
be done with fresh chemicals.
10. Bypass the desiccant to check the CO2 zero (see Note 10).
11. While soda lime is on full scrub, briefly blow into the air intake
of the IRGA. A spike in the CO2 reading indicates that the soda
lime is exhausted and needs to be replaced (or “regenerated”
by adding a bit of water, see Note 7).
12. If necessary, perform a CO2 zero calibration (see Note 11).
This should only be done if blowing into the air intake and
around the leaf chamber does not change the measured CO2
concentration.
13. If applicable, test the gas analyzer matching function.
14. Insert a CO2 cartridge into the CO2 mixer, or turn on CO2 if a
tank is in use.
15. Wait for a few minutes for the CO2 to replace any air in the
lines and perform a CO2 mixer calibration (optional; this is
useful when performing A/Ci measurements).
 Photosynthetic Gas Exchange at the Leaf Level 31

16. Before inserting the first leaf, set the desired chamber condi-
tions (CO2 concentration, light intensity, and temperature; see
Note 12; as well as humidity; see Note 13).

3.2 Making 1. On the IRGA console open a file where the data is to be stored.
Measurements 2. Place the leaf in the leaf chamber and close the chamber care-
fully. Make sure that the thermocouple touches the leaf firmly,
but gently enough so it does not damage the leaf or the
thermocouple. Ideally, the leaf covers the whole chamber (see
Note 14).
3. Monitor parameters for some time to make sure that the set
environmental conditions are appropriate for the measure-
ment. In particular, make sure that the humidity inside the
chamber stays below 85%; otherwise condensation may occur
(see Note 13).
4. Once the CO2 concentration has stabilized inside the chamber,
blow around the chamber to test for air leaks. If a spike of more
than about 1 ppm in the sample CO2 concentration is visible,
then a leak exists somewhere, likely at the interface of leaf and
gasket. In this case, try to locate the leak by blowing at individ-
ual spots around the leaf chamber with a straw (see Note 15).
5. When the parameters such as Anet, E, and stomatal conduc-
tance look stable, a measurement can be logged (see Note 16).
6. If only one individual measurement is needed, such as Anet
under ambient conditions, Asat, or Amax, then the chamber
can be opened and the leaf removed at this point. If the leaf
did not fill the whole chamber, then now is the time to measure
the leaf area by outlining the part of the leaf that was inside the
chamber with a permanent marker and subsequently measur-
ing the outlined leaf area (see Note 17).
7. Insert the next leaf to be measured.
Response curves require multiple measurements of the same
leaf under different environmental conditions. In general, all
environmental parameters, apart from the one to be varied,
need to be kept constant. The most common of these responses
is the CO2 response (or A/Ci) curve, which is described in
some detail below.
8. Insert leaf into the leaf chamber and check for air leaks as
outlined above.
9. Even if no leak is detected by blowing around the chamber,
small diffusion leaks may still cause trouble due to the large
(and nonrandom) CO2 gradient between the inside and the
outside of the chamber. Perform a leak correction as recom-
mended by the manufacturer even if no leak is detected (see
Note 18). An indication of potential leaks can be obtained by
32 Florian A. Busch

measuring a CO2 response curve without a leaf inside the

chamber. If no leak is present, then Anet should be zero,
independent of CO2 concentration. However, keep in mind
that inserting a leaf into the chamber may introduce additional
leaks.
10. Set the desired temperature, controlled on the basis of Tleaf (see
Note 19).
11. Set the CO2 concentration to ambient or close to ambient, i.e.,
400 μmol mol
1 (for an A/Ci curve it is best for the CO2 to be
controlled on the reference value, see Note 12).
12. Set the light intensity to a reasonably high value, as A/Ci
curves are typically measured under saturating light to confi-
dently estimate Vcmax and Tp. The intensity at which light
saturation is reached will depend on the plant and how it is
grown. This can be reached at 500 μmol photons m
2 s
1 or
below for nutrient deficient, senescing, or shade leaves, but
healthy leaves grown under full sunlight may not even saturate
at 2000 μmol photons m
2 s
1 (see Note 20).
13. Check whether the humidity is in a good range, considering
that E will vary throughout the measurement (see Note 19). A
good starting point is around 70% relative humidity; adjust if
the value is significantly different. This value either can be
adjusted automatically if the equipment is capable of doing so
or will have to be manually adjusted by directing more or less
air through the desiccant scrub bottle (see Note 21).
14. Let the leaf acclimate to these conditions for 5–10 min, until
parameters are stable and RuBisCO is fully activated. Ideally,
stomata are wide open at this point, with a Ci/Ca of >0.7 for
C3 species (see Note 22). For plants that come out of the
shade/dark this acclimation time can take much longer.
15. Start the sequence of measurements at ambient CO2 concen-
tration (400 μmol mol
1) and stepwise decrease the CO2
concentration to the lowest value while taking a measurement
at each concentration. After returning to 400 μmol mol
1,
stepwise increase the CO2 concentration until the highest
value is reached, again taking a measurement at each concen-
tration. Suggested here is the following sequence of CO2 con-
centrations: 400, 350, 300, 250, 200, 150, 100, 50, 400,
400 (see Note 23), 450, 500, 650, 800, 1000, 1250, 1500,
1750, 2000, and 400. This ensures enough measurements for
the fitting of Vcmax, Rd, as well as J and/or Tp (see Note 24).
Wait for a minimum of 90 s and a maximum of 150 s
between measurements (see Note 25).
16. Make sure to match the reference and sample analyzer at every
CO2 concentration (see Note 26).
 Photosynthetic Gas Exchange at the Leaf Level 33

17. If the equipment allows, set up the sequence of measurements

described in the previous points as an auto program. This will
make the measurements faster, easier, and more repeatable.
During the measurement, try not to breath at or around the
leaf chamber; this will increase the level of noise if air leaks are
present.
Some comments on light and temperature response curves:
18. Light response curves: Light response curves can be used to
estimate the maximum rate of electron transport, Jmax (one of
the parameters that cannot be measured with an A/Ci curve) .
To reach a stable CO2 supply, control CO2 on the sample CO2
value, as Anet (and therefore the CO2 gradient between sample
and reference) will vary with light intensity. Measure a
sequence of light intensities starting from high light to low
light to ensure full RuBisCO activation: (2000), 1500, 1000,
750, 500, 300, 200, 150, 100, 50, 25, 10, and 0 μmol photons
m
2 s
1. Wait for a minimum of 5 min and a maximum of
6–8 min between measurements (see Note 27).
19. Temperature response curves: Ideally, temperature responses are
measured in a climate-controlled cabinet, or a room in which
the temperature can be adjusted; the IRGA itself will have a
limited range of temperatures that can be achieved relative to
room temperature. The greater the temperature difference
between the chamber wall and the leaf is, the larger are poten-
tial inaccuracies in Tleaf measurements [12]. Special attention
has to be paid to the humidity inside the chamber. At low
temperatures, the relative humidity generally increases, which
easily results in condensation at temperatures below room
temperature (see Note 13). At high temperatures, the relative
humidity tends to decrease, resulting in stomatal closure and
large changes in Ci values. These effects necessitate a tight
control of the humidity of the ingoing air by scrubbing or
adding water, which is easier to do when the temperature
difference between the room and the chamber air is small.

3.3 Analyzing The previous sections were to ensure that the quality of the
and Interpreting obtained primary data is suitable. While collecting primary data
the Data requires some skill and practice, I consider analysis and interpreting
data the “art” of performing gas exchange. This section describes
how photosynthetic parameters can be calculated from collected
A/Ci data, and some of the potential pitfalls of doing so.
1. Download the data to a computer by the appropriate means (see
Note 28).
2. Plot the data as Anet against Ci. A sample data set collected with
an LI-6400XT according to the protocol described here is
shown in Fig. 2.
34 Florian A. Busch

a
30

A ( mol m-2 s-1)

20

10
measured data (1800)
'extra' measurements (1800)
measured data (250)
'extra' measurements (250)
0

0 500 1000 1500

Ci ( mol mol )-1

b
30
A ( mol m-2 s-1)

20

'good' measurements
10 fitted Ac
fitted Aj
fitted Ap
0

0 500 1000 1500

Ci ( mol mol-1)

Fig. 2 (a) Sample A/Ci curve measured on a Eucalyptus grandis leaf with an
LI-6400XT using the protocol described here. To highlight the impact of
saturating vs. non-saturating light intensity, the same leaf was measured at
1800 and 250 μmol photons m
2 s
1 (circles and squares, respectively). Open
symbols represent the “extra” measurements at a CO2 concentration of
400 μmol mol
1 that should be discarded for the fitting process. (b) The
data measured under high light shown in (a) fitted with the FvCB model. At
1800 μmol photons m
2 s
1Aj only occupies a small Ci range, or may be
missing altogether, while at lower light intensities (such as the data measured
at 250 μmol photons m
2 s
1 in (a)) Ap and/or Ac may be missing or hard to fit.
The fitted values are Vcmax ¼ 90 μmol m
2 s
1, J ¼ 129 μmol m
2 s
1,
Tp ¼ 8.6 μmol m
2 s
1, and Rd ¼ 0.8 μmol m
2 s
1
 Photosynthetic Gas Exchange at the Leaf Level 35

3. Assess the quality of the collected data. Some potential pro-

blems during data collection, their effects on A/Ci data, and
how they can be avoided are outlined in Fig. 3. Remove “out-
lier” data points, such as extra measurements at 400 μmol
mol
1 CO2, or points that are off the A/Ci trajectory for any
other valid reason (e.g., the machine did not match).
4. If the data is satisfactory, then photosynthetic model para-
meters can be estimated from the data with a fitting tool.
Fitting the FvCB model to the obtained data can be done
with a customized program, or with the use of one of the
many fitting tools that are available (their applications are
described for example in [6, 11, 13–15]). They all have their
advantages and disadvantages, and below I outline some gen-
eral issues to consider.
5. Up to three functions of the FvCB model can be fitted to the
data: Ac, Aj, and Ap. For the fitting process, individual data
points have to be assigned to one of these functions, either
explicitly by assigning them manually or implicitly via maximiz-
ing the goodness of the overall fit with the tool where manual
assignment is not applicable. Assigning the points manually
requires some idea of the underlying biochemical limitations.
The automated tools have the downside that not all of them
include the Ap limitation, and that any one of the limitations
may be missing, which can be misinterpreted by the fitting
routine. A careful check of the fitted results is necessary with
any fitting approach. In general, when using the protocol
described here, data points measured below ambient [CO2]
are likely limited by Ac, data points above ambient [CO2]
most often limited by Ap (see Note 29), with some remaining
data points around ambient [CO2] limited by Aj (see [16]
about the general pattern of limitations and how they change
with environmental parameters). Assigning the wrong limita-
tion to the data can have large consequences for the estimated
parameters (see [14] for a detailed discussion). When in doubt,
it is often best to not assign that particular measurement to any
limitation. The protocol described above purposefully contains
plenty of measurements to have the flexibility to leave some
measurements out from the fitting routine.
6. If the light intensity during the measurement was truly saturat-
ing, then only few (if any) data points can be assigned to Aj. In
this case, the estimation of J from the A/Ci curve will be error
prone. If J is a parameter of interest, then consider performing a
light response curve, or an A/Ci curve at a light intensity that
you can safely consider sub-saturating (e.g., at 200–500 μmol
photons m
2 s
1; see Fig. 2a). In the latter case, most data
points can be assigned to Aj, resulting in a good model fit. The
so-obtained value for J cannot be used as approximation for
36 Florian A. Busch

a b
30 30
A ( mol m-2 s-1)

A ( mol m-2 s-1)

20 20

10 10

0 0
0 500 1000 1500 0 500 1000 1500
Ci ( mol mol-1) Ci ( mol mol-1)
c d
30 30
A ( mol m-2 s-1)

A ( mol m-2 s-1)

20 20

10 10

0 0
0 500 1000 1500 0 500 1000 1500
Ci ( mol mol-1) Ci ( mol mol-1)

Fig. 3 Schematic A/Ci curves with the telltale signs of issues with the measurement. (a) Data is very noisy—it
will be difficult to accurately fit the biochemical model to very noisy data. Potential reasons for noisy data
include exhausted soda lime allowing room-air CO2 to make its way into the chamber, dirt or debris inside the
leaf chamber, or a reduced sensitivity of the IRGA. Noisy data can also be caused by excessive humidity inside
the chamber, which is strongly indicated if Ci values are negative. If the relative humidity is close to saturating,
condensation can occur, resulting in erroneous calculations of Ci values. Lastly, noisy data can arise when the
minimum wait time between measurements is not long enough to establish a stable CO2 concentration inside
the chamber. Increase the minimum wait time, or decrease the change in CO2 concentration between
measurement steps. A small amount of noise can be partially compensated for by having more than three
or four measurements assigned to each limitation, but this should only be the solution of choice if the source
of the noise cannot be tracked down. (b) The A/Ci curve does not flatten out even at high values of Ci (see
arrow). An increase of Anet with Ci is the expected response under sub-saturating light intensities, especially at
low light (see (a) at 250 μmol photons m
2 s
1). Was the light intensity high enough during the measurement?
If you are confident that it was, then this response indicates a diffusion air leak around the leaf chamber. Track
down leaks and mitigate them as described in the text. (c) No high Ci values can be reached, despite setting Ca
concentrations up to 2000 μmol mol
1. While nothing may be wrong per se with this measurement, it will be
hard to fit values for J and Tp. The likely reason for this effect is a plant with very low stomatal conductance. Is
the humidity inside the chamber around 70%? Did the plant have enough time to acclimate to the chamber
conditions and fully open the stomata? Has the plant been watered with the desired amount of water? (d) The
Ac limited data, when extrapolated to Γ*, does not agree with the expected Γ* value, i.e., it appears as if the
whole curve is shifted to the left or to the right (see arrows). This can indicate that the leaf temperature
measurement is off (Γ* is highly dependent on leaf temperature), or that there is a problem with the CO2 zero
calibration. Check all the calibrations using fresh chemicals
 Photosynthetic Gas Exchange at the Leaf Level 37

Jmax, but allows for a consistent comparison between measure-

ments of different plants.
7. Mesophyll conductance (gm) is another parameter that can be
estimated from A/Ci curves [11, 17], and some fitting tools
are set up to include gm as an output. However, as outlined by
Sharkey [6], estimating gm with an A/Ci curve fitting approach
is not ideal, and it is best to measure gm via some independent
means.

4 Notes

1. The materials described here refer most closely to measuring

gas exchange with the frequently used LI-6400 and
LI-6400XT gas analyzers (Li-Cor Biosciences, Lincoln, NE,
USA). Other commercial instruments (e.g., GFS-3000 (Walz,
Effeltrich, Germany), CIRAS-3 (PP Systems International,
Amesbury, MA, USA), LCpro-SD (ADC BioScientific Ltd.,
Hoddesdon, UK), or LI-6800 (Li-Cor, Lincoln, NE, USA))
or custom-built devices of similar capability may differ slightly
in their requirements and details in measurement outputs, but
the general principle of operation and data analysis/interpreta-
tion still applies.
2. The simplest option is the LED light source made specifically
for the leaf chamber in use, designed by the IRGA manufac-
turer. Depending on experimental requirements other light
sources may be required, e.g., when a custom-built or clear
top chamber is used, or when a certain light quality is required.
Natural daylight may also be an option.
3. This is a separate part for the LI-6400 and LI-6400XT
machines, but is built into the GFS-3000, LCpro-SD, and
LI-6800 machines. It is strongly recommended not to use
ambient air without having the machine control the CO2
concentration, since the air CO2 concentration can vary greatly,
which can add a large amount of unnecessary noise to the data.
4. While it is recommended to use the batteries supplied by the
manufacturer, it may be possible to connect a standard car
battery. This can be beneficial for long measurement days in
the field where it is impractical to recharge the batteries
regularly.
5. If the ambient air temperature fluctuates, the IRGA may have
trouble keeping the leaf temperature constant. This can
increase the noise in the collected data and is easily avoided
by keeping the equipment at a controlled temperature. For
similar reasons, avoid exposing the equipment to direct
sunlight.
38 Florian A. Busch

6. The tripod ensures that the leaf chamber is fixed in place in a

convenient location, which allows an easy placement of the leaf
and minimizes the chances of damaging the leaf or the
equipment.
7. Depending on the humidity of the air and/or the moisture
coming from the soda lime, expect the desiccant to need
replacement once a day or more often. The CO2absorption
capacity of the soda lime will typically last significantly longer.
However, soda lime absorbs CO2 in exchange for water, so
over time it will dry out and lose its CO2-scrubbing ability
(unless the humidity of the ambient air is very high). In this
case, the soda lime can be “regenerated” once or twice before
needing replacement by adding a small amount of water
(around 5 mL from a squirt bottle) to the scrub column.
Shake well, and wait for several minutes to let the water soak
deep into the soda lime. DO NOT use when standing water is
visible; this will damage the IRGA! Refer to your manual for
further details.
8. Tleaf is used to calculate several of the recorded parameters,
including stomatal conductance (gsc) and Ci. An accurate Tleaf
is critical for obtaining accurate parameter values, which
becomes particularly important when performing A/Ci
measurements.
9. Water tends to “stick” to the tubing and chamber walls. It may
take up to 30 min to fully scrub all the water out of the
measured air.
10. Desiccant releases small amounts of trapped CO2. Bypassing
the desiccant for the CO2 zero ensures that the calibration gas
is CO2 free.
11. A zero offset of a ppm or two of CO2 may not affect measure-
ments of Amax or Asat very much, but will introduce an error of
the same magnitude as the offset into measurements of Γ or Γ*.
It will also affect the fit of an A/Ci curve.
12. Depending on the nature of the measurement, one needs to
consider how to control these parameters. The CO2 concen-
tration can be controlled on the basis of the reference or the
sample CO2 value. While the sample CO2 value corresponds to
the CO2 concentration that the leaf is exposed to, it is often
useful to set and control for the reference CO2 value. Especially
when measuring an A/Ci curve, CO2 concentrations need to
be changed rather quickly, which is better handled by
controlling the fast-responding reference CO2 value. Other
measurements may require the sample CO2 to be constant,
e.g., light response curves. Similarly, for response curves the
temperature should generally be controlled as a fixed Tleaf, but
spot measurements may require the control of the less fluctu-
ating block temperature.
 Photosynthetic Gas Exchange at the Leaf Level 39

13. Keep in mind that the leaf chamber contains a fan that is not
only intended to mix the air, but also to reduce the boundary
layer. This means that at a comparable air humidity the loss of
water from the leaf will be higher inside the chamber than in
the plant’s natural environment, which can be counterbalanced
by measuring the leaf under relatively humid conditions. A
good range for relative humidity inside the chamber is in the
range of 60–75%, or at a VPD in the range of 1–1.5 kPa, to
prevent the leaf from drying out on the one hand and avoid
condensation of water on the chamber walls on the other hand.
If condensation occurs, then the measured parameters will
become inaccurate and the system has to be dried out
completely before further use. Negative Ci values are often an
indication of condensation, but condensation can occur even
when the reported Ci values are positive.
14. In the interest of a good signal-to-noise ratio, covering as much
as possible of the chamber with the leaf is preferable. Smaller
leaves may be used, but the change in leaf area has to be
accounted for in the software/when calculating the measured
parameters. Multiple straight-edged leaves, such as grass leaves,
can also be combined to fill the whole chamber area (DO NOT
overlap the leaves!). Small grass leaves or conifer needles can be
arranged on and fixed in place with a piece of tape, before
placing them inside the chamber (Fig. 4). Make sure that the
parts of the leaves inside the chamber are not covered with
tape. In any of these cases, special care has to be taken to make
sure that the leaf thermocouple is in close contact with a leaf.
15. Large veins increase the chances of diffusion air leaks through
the gasket of the chamber, as they may introduce a gap between
leaf and gasket. If the leaf shape allows, try to avoid placing the
midvein or any other large veins inside the chamber. Leaks can
be minimized by carefully applying some vacuum grease or
adhesive plastic sealant on the leaf/gasket interface from the
outside. In addition, filling the whole chamber area with a leaf
minimizes the relative contribution of a possible leak to the
overall rate of CO2 uptake.
16. A quick way to decide whether the parameters are stable is to
assess them from their graphs. What to consider “stable” will
depend on the type of measurement and your machine; consult
your manual for further details.
17. The area of the leaf inside the chamber can be estimated at the
end of the measurement with a leaf area meter (if available), or
can be estimated from a picture that includes a scale, such as a
ruler, as a reference (see Fig. 4). The green pixel area of an
image can be calculated semiautomatically, e.g., with the freely
40 Florian A. Busch

Fig. 4 Measuring small leaves. If the measured leaf is not big enough to fill the entire chamber, then the leaf
area inside the chamber has to be quantified and the gas exchange data adjusted with the actual measured
leaf area. Multiple straight-edged leaves, such as grass leaves, can be combined to fill a larger part of the
chamber (make sure that the leaves do not overlap). Small grass leaves or conifer needles can be arranged on
and fixed in place with a piece of tape, before placing them inside the chamber. A photograph like this that
shows the measured leaves (with the chamber extent marked using a marker pen, here outlined as a red
rectangle) on a white background and including a ruler can be used to semiautomatically quantify the leaf area
with software such as ImageJ

available software ImageJ (National Institutes of Health, USA.

Website: https://imagej.nih.gov/ij/).
18. Several ways of correcting for air leaks have been put forward in
the past [1, 18, 19]. Figure 3b shows the telltale signs of a
diffusion leak through the gasket.
19. Stomata are fairly sensitive to changes in CO2 concentration.
At low concentrations gsc increases, while at higher concentra-
tions gsc decreases. This affects the rate of transpiration (E)
and, as a consequence, leaf temperature. The biochemical para-
meters estimated from an A/Ci curve are highly temperature
sensitive. Therefore leaf temperature has to be kept constant
throughout the measurement. To ensure this, it is advisable to
control temperature on the basis of Tleaf.
20. While photoinhibition due to excessive light should be
avoided, it is generally better to err on the side of too much
light. For most plants, this means a light intensity of
1500 μmol photons m
2 s
1 or above is a good choice. Photo-
inhibition is minimized by increasing the light intensity up to
its final value in multiple steps, waiting at each light intensity
 Photosynthetic Gas Exchange at the Leaf Level 41

for 2 or 3 min, or until Anet has stabilized. The light intensity

used for the A/Ci measurements can be adjusted, if an initial
response curve indicates that Amax is fairly low despite high
light. Then PAR can be decreased, as long as this does not
affect the values of the A/Ci response. Conversely, if Anet is
increasing with increasing Ci even at high CO2 concentrations
(such as the low light example in Fig. 2a), then increase PAR
(if this is the case, then Anet around ambient CO2 concentra-
tions should increase with higher PAR).
21. Adjusting the humidity during the A/Ci measurement, in
particular when using an auto program, may result in incorrect
estimates of gsc due to prolonged times to reach steady state.
22. An A/Ci curve removes the effect of stomatal opening for the
analysis of the biochemical parameters. Open stomata ensure a
wide range of Ci values that can be achieved, and thereby
improve the estimation of these parameters. Ideally, measured
Ci values should exceed 1500 μmol mol
1 for a good estima-
tion of Tp.
23. Often the first measured value upon returning to 400 μmol
mol
1 from low CO2 has a lower Anet than the initial measure-
ment. This may be due to a slow response time for adjusting
the CO2 concentration inside the leaf chamber, but can also
indicate responses of the plant. These responses include photo-
inhibition, which can occur at low CO2 concentrations, or
photorespiration CO2 being released from the metabolite
pools that have accumulated during the exposure to low CO2
concentrations. Sometimes it may be necessary to include three
or four measurements at 400 μmol mol
1 after returning from
low CO2 concentrations to make sure that Anet reaches a value
close to the starting value. Only the starting value and the last
of the values after returning from low CO2 should be kept for
analysis. All other values measured at 400 μmol mol
1 should
be discarded from the A/Ci analysis.
24. Any of the three biochemical limitations can be missing from a
measured A/Ci curve or not be present in the measured plant.
In general, you can expect to comfortably estimate Vcmax and
Rd from the A/Ci data. If the light intensity chosen was
saturating, then several data points at high CO2 concentrations
can be attributed to the TPU limitation, whereas only few data
points (if any) fall into the RuBP-regeneration limitation at
intermediate Ci values. If the light intensity used is not fully
saturating, then most (sometimes all) values above ambient will
have to be attributed to the RuBP-regeneration limitation. In
this case, quantifying Tp will be difficult. The number and
spread of measurements suggested here maximize the chance
that at least two or three values can be allocated to each of the
42 Florian A. Busch

three limitations. The number of measurements can be

reduced if time is an issue (at the cost of the quality of the
model fit), or restricted to a certain range, if e.g., only Vcmax
needs to be estimated.
25. The reasoning for the minimum time is twofold: (1) Sufficient
time has to be given so the IRGA can reach a stable concentra-
tion of the new CO2 value inside the chamber and (2) photo-
respiration CO2 is released from photorespiratory metabolites
some time after the oxygenation reaction (up to several min-
utes, depending on the pool size) and the plant must be given
time to reach steady state in that regard, so the photorespira-
tory CO2 release matches the rate of oxygenation by RuBisCO.
During the measurement, the leaf should not be given suffi-
cient time to acclimate to the changes in CO2 concentration
(e.g., the activation state of RuBisCO should not change dur-
ing the A/Ci measurement). For this reason, and to keep the
overall measurement time as short as possible, a measurement
should be logged as soon as the assimilation rate becomes
stable, which generally happens within 2 min. The actual mini-
mum and maximum wait times may have to be adjusted some-
what, depending on the behavior of the leaf. Faster methods
for measuring A/Ci curves are starting to become available as
the IRGA technology is improving [20].
26. The infrared gas analyzers are somewhat CO2 dependent and
also drift apart with time. Therefore, they need to be matched
after each change in CO2 concentration and after some time
even without changing the CO2 concentration. Forgetting to
match can result in unusable data that cannot be corrected for
later!
27. The highest light intensity of 2000 μmol photons m
2 s
1 is
only needed if Anet does not saturate at lower values. The
number of measurements and their values can be adjusted to
suit the measured leaf and the experimental requirements. The
minimum wait time between measurements should be longer
than for A/Ci measurements, so that the stomata are given
time to acclimate to the new light intensity. This reduces the
variation in Ci that invariably will occur when changing the
light intensity. Measuring from high light to low light has the
advantage that stomata will close throughout the light
response measurement. Since stomata tend to close faster
than they open, this minimizes the time required for each
measurement.
28. Depending on the equipment used, this may be done via a
USB, RS-232, or network cable, or with a SD or CF
memory card.
 Photosynthetic Gas Exchange at the Leaf Level 43

29. Under an Ap limitation, Anet is usually insensitive to changes in

CO2 concentration. Frequently, however, a decrease in Anet
with increasing Ci can be observed and has been attributed to
glycolate carbon leaving the photorespiratory pathway as
amino acids and is not returned to the chloroplast [9] (see
Fig. 1 for α > 0). Some fitting tools provide a function to
account for this effect (e.g., [6]).

Acknowledgments

I thank Ross Deans for helpful comments on this chapter.

References

1. Li-Cor (2012) Using the LI-6400 / Peguero-Pina JJ, Pou A, Ribas-Carbó M,

LI-6400XT portable photosynthesis system,
version 6. Li-Cor Biosciences, Lincoln, NE
2. Walz (2013) Portable gas exchange fluores-
cence system GFS-3000, 7th edn. Heinz Walz
GmbH, Effeltrich
3. Li-Cor (2016) Using the LI-6800 portable
photosynthesis system, version 1. Li-Cor Bios-
ciences, Lincoln, NE
4. Long SP, Bernacchi CJ (2003) Gas exchange
measurements, what can they tell us about the
underlying limitations to photosynthesis? Pro-
cedures and sources of error. J Exp Bot 54
(392):2393–2401
5. Evans JR, Santiago LS (2014) PrometheusWiki
Gold Leaf Protocol: gas exchange using
LI-COR 6400. Funct Plant Biol 41
(3):223–226. https://doi.org/10.1071/
FP10900
6. Sharkey TD (2016) What gas exchange data
can tell us about photosynthesis. Plant Cell
Environ 39(6):1161–1163. https://doi.org/
10.1111/pce.12641
7. Farquhar GD, von Caemmerer S, Berry JA
(1980) A biochemical model of photosynthetic
CO2 assimilation in leaves of C3 species. Planta
149(1):78–90
8. von Caemmerer S (2000) Biochemical models
of leaf photosynthesis. CSIRO Publishing,
Collingwood
9. Harley PC, Sharkey TD (1991) An improved
model of C3 photosynthesis at high CO2:
Reversed O2 sensitivity explained by lack of
glycerate reentry into the chloroplast. Photo-
synth Res 27(3):169–178
10. Flexas J, Barbour MM, Brendel O, Cabrera
HM, Carriquı́ M, Dı́az-Espejo A, Douthe C,
Dreyer E, Ferrio JP, Gago J, Gallé A, Galmés J,
Kodama N, Medrano H, Niinemets U, €
Tomás M, Tosens T, Warren CR (2012) Meso-
phyll diffusion conductance to CO2: An unap-
preciated central player in photosynthesis.
Plant Sci 193–194:70–84. https://doi.org/
10.1016/j.plantsci.2012.05.009
11. Sharkey TD, Bernacchi CJ, Farquhar GD,
Singsaas EL (2007) Fitting photosynthetic car-
bon dioxide response curves for C3 leaves.
Plant Cell Environ 30(9):1035–1040.
https://doi.org/10.1111/j.1365-3040.2007.
01710.x
12. Mott KA, Peak D (2011) Alternative perspec-
tive on the control of transpiration by radia-
tion. Proc Natl Acad Sci U S A 108
(49):19820–19823. https://doi.org/10.
1073/pnas.1113878108
13. Duursma RA (2015) Plantecophys - An R
package for analysing and modelling leaf gas
exchange data. PLoS One 10(11):e0143346.
https://doi.org/10.1371/journal.pone.
0143346
14. Dubois JJB, Fiscus EL, Booker FL, Flowers
MD, Reid CD (2007) Optimizing the statisti-
cal estimation of the parameters of the
Farquhar-von Caemmerer-Berry model of
photosynthesis. New Phytol 176:402–414.
https://doi.org/10.1111/j.1469-8137.2007.
02182.x
15. Gu LH, Pallardy SG, Tu K, Law BE, Wulls-
chleger SD (2010) Reliable estimation of bio-
chemical parameters from C3 leaf
photosynthesis-intercellular carbon dioxide
response curves. Plant Cell Environ 33
(11):1852–1874. https://doi.org/10.1111/j.
1365-3040.2010.02192.x
16. Busch FA, Sage RF (2017) The sensitivity of
photosynthesis to O2 and CO2 concentration
identifies strong Rubisco control above the
44 Florian A. Busch
thermal optimum. New Phytol 213
(3):1036–1051. https://doi.org/10.1111/
nph.14258
17. Ethier GJ, Livingston NJ (2004) On the need
to incorporate sensitivity to CO2 transfer con-
ductance into the Farquhar-von Caemmerer-
Berry leaf photosynthesis model. Plant Cell
Environ 27(2):137–153
18. Flexas J, Diaz-Espejo A, Berry JA, Cifre J,
Galmes J, Kaidenhoff R, Medrano H, Ribas-
Carbo M (2007) Analysis of leakage in IRGA’s
leaf chambers of open gas exchange systems:
quantification and its effects in photosynthesis
parameterization. J Exp Bot 58
(6):1533–1543. https://doi.org/10.1093/
jxb/erm027
19. Rodeghiero M, Niinemets U, Cescatti A
(2007) Major diffusion leaks of clamp-on leaf
cuvettes still unaccounted: how erroneous are
the estimates of Farquhar et al. model para-
meters? Plant Cell Environ 30(8):1006–1022.
https://doi.org/10.1111/j.1365-3040.2007.
001689.x
20. Stinziano JR, Morgan PB, Lynch DJ, Saathoff
AJ, McDermitt DK, Hanson DT (2017) The
rapid A–Ci response: photosynthesis in the
phenomic era. Plant Cell Environ 40
(8):1256–1262. https://doi.org/10.1111/
pce.12911
 Chapter 3

Design and Use of a Digitally Controlled Device for Accurate,

Multiplexed Gas Exchange Measurements of the Complete
Foliar Parts of Plants
Gavin M. George, Katharina Kölling, Roland Kuenzli,
Matthias Hirsch-Hoffmann, Patrick Flütsch, and Samuel C. Zeeman

Abstract
Performing accurate measurements of photosynthetic and respiration rates is vital to a large proportion of
plant-based studies. While several commercial systems exist to perform such measurements, few are ideal for
whole-plant measurements of small herbaceous plants such as Arabidopsis and none offer the capacity for
simultaneous analysis of multiple plants. We, therefore, designed a multi-chamber, computer-controlled,
infrared gas analyzer-coupled system for the continuous measurement of gas exchange in whole-plant
shoots or rosettes. This system was called ETH Gas Exchange System-1 (EGES-1). We have subsequently
expanded the device to accommodate a wider variety of species while providing precise control over
environmental parameters. Critically, we have (1) increased the flow rates through each of the eight
chambers, (2) introduced a computer-controlled feedback loop for the precise introduction of CO2, and
(3) added an additional feedback loop for the introduction and control of humidity. The advantages of this
new system (EGES-2) are illustrated here in the context of a variety of physiological experiments.

Key words Gas exchange, Photosynthesis, Respiration, Transpiration, Computer-controlled, Multi-

chamber

1 Introduction

Photosynthesis is a process whereby atmospheric carbon dioxide is

assimilated into organic compounds using light energy. In this way,
the light’s electromagnetic energy is converted to reduced organic
molecules. The resulting organic compounds are metabolized into
the structural and functional components of the plant or respired at
night, or in non-photosynthetic cells, to produce ATP and reducing
power, releasing CO2 [1]. Essentially all the carbon found in a plant

Electronic supplementary material: The online version of this article (https://doi.org/10.1007/978-1-4939-

7786-4_3) contains supplementary material, which is available to authorized users.

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_3, © Springer Science+Business Media, LLC, part of Springer Nature 2018

45
46 Gavin M. George et al.

is derived from the atmosphere and so its assimilation by photosyn-

thesis is one of the fundamental factors which determine plant
productivity.
Measuring gas exchange of plants offers an accurate and direct
means by which photosynthetic assimilation of CO2 can be deter-
mined. In principle, a plant or leaf can be sealed in a chamber which
is either transparent or internally lit. Two tubes allow air to pass in
and out of this chamber [2]. The concentrations of CO2 and H2O
are measured in the airstream both before and after passing through
this chamber. Carbon assimilation and transpiration, or water loss,
can be calculated from the difference (Δ) between the input and
output air. Commercial systems for measuring gas exchange typi-
cally utilize a leaf-clip cuvette which isolates a small area on a leaf of
2–10 cm2. The air supplied to and received from this leaf is typically
measured by an infrared gas analyzer (IRGA). These provide accu-
rate measurements of assimilation rates and are often portable,
making them ideal for environmental and field studies. They are
limited, however, by only providing information on the particular
leaf area being measured. Photosynthetic assimilation rates vary
greatly over a single plant, depending on which leaf is measured
and the developmental stage [3]. In addition, stomatal density and,
therefore, assimilation and transpiration rates can also vary across a
single leaf lamina [4, 5]. Mutations affecting assimilation and
metabolism of carbon often yield pleiotropic alterations in leaf
architecture leading to changes in the leaf area (LA) to dry mass
ratio (DM). This, in turn, may result in differences in assimilation
rates between two genotypes being either missed or overrepre-
sented when ΔCO2 is normalized only to LA. Leaf-clip cuvettes
are also often bulky and may shade parts of a small plant that are not
being measured, which may introduce errors in downstream ana-
lyses. Whole-plant chambers for measuring gas exchange are also
available; however, these generally measure the entire plant in a pot.
This results in a dilution of the assimilation rate by CO2 respired
from the roots. Furthermore, these systems usually measure only a
single plant at a time and are not always suitable for use over the
course of several days.
In order to overcome the limitations of the commercial sys-
tems, we designed and built a computer-controlled, gas switching
system allowing for the automated measurement of up to eight
attached custom-made plant chambers. We recently described this
ETH Zurich gas exchange system 1 (EGES-1, [6]) and have subse-
quently upgraded it to greatly increase its flexibility and stability
(EGES-2). The upgraded system uses a molecular scrubber to strip
incoming air of CO2. Water is also removed from the air in the
process; however, compressed air supplies usually contain low con-
centrations of water to begin with. The CO2 and H2O are reintro-
duced by a bottled supply and a humidification column,
respectively. Both are controlled by a digital feedback loop, which
 Gas Exchange Measurements of Complete Foliar Parts of Plants 47

constantly measures the concentrations of both in a reference air-

stream, and adjusts the supply accordingly. The controller for these
feedback loops is integrated into environmental control software,
based on the LabVIEW platform [7], adapted specifically for the
EGES systems. This allows for near-perfect control of both CO2
and H2O for several days or more. It also allows both to be set from
very low to high concentrations as required by the experiment. In
addition, the maximal airflow rates that the system can provide have
been increased in order to accommodate plants larger than Arabi-
dopsis, such as the common pea (Pisum sativum).
EGES-2 expands on a number of advantages of the previous
iteration, EGES-1, and several commercial devices. It allows for the
nondestructive measurement of the entire aerial portion of a plant
while isolating the roots and the contaminating respired CO2 that
they produce. As multiple chambers can be measured, it allows for
replicated comparisons between plant eco- or genotypes as well as
experimental treatments. It also allows for the introduction of
plants grown either in soil or in hydroponics-based systems,
which is useful for quantitative root treatments. Furthermore, the
system allows for supplied air CO2 or H2O concentrations to be
changed at any point during an experiment inducing carbon limi-
tation/surplus or leaf-specific drought. Finally, EGES-2 can be run
continuously for days or weeks allowing for the distinction between
short- and long-term adaptive processes.

2 Materials

2.1 Multi-chamber, 1. Compressed air source (root air supply).

IRGA-Linked, Gas 2. Gas switching unit (built in-house, see schematic in Fig. 1)
Switching Unit consisting of ten mass flow controllers (MFCs).
3. Humidification column: Comprised of two zones, air bubbles
through water and then 2 kg of defined ceramic stones (Stutt-
garter Masse) which when wet allow for secondary humidifica-
tion of passing air in a single-pressure vessel (built in-house, see
schematic in Fig. 2).
4. Plant chambers (built in-house, see schematic in Fig. 3).
5. Molecular scrubber: An inline regenerative dual-tower adsorp-
tion system which provides CO2-free and water-free air.
6. 400 mL Compressed air accumulator: Employing a throttle
check valve, this allows for compensation of pressure fluctua-
tions from the root air supply and the molecular scrubber.
7. Compressed air reservoir (7 L).
8. Five gas regulators with built-in pressure gauges: 1–10 PSI.
9. Compressed 10–100% CO2 gas supply.
48 Gavin M. George et al.

Fig. 1 Schematic showing the arrangement of the EGES-2. The incoming air is passed through a molecular
scrubber (Scrub.), a pressure reservoir (PR), and three regulators before humidification. A digital bypass (PWM)
around the hydration column allows for humidity control of the incoming air. CO2 is added before the air is split
between a reference gas line and the plant chambers. All airflow rates are precisely set by mass flow
controllers (MFCs). Air returning from the plant chambers moves to a gas switching unit that controls which
chamber is measured by the IRGA (LI-7000 in this system). The data acquisition and environmental controls
are managed by the attached PC and related software

10. A digital proportional valve: This will constitute the dry-air

bypass and provides unidirectional airflow at a rate determined
by a pulse width-modulated (PWM) signal.
11. IRGA (LI-COR 7000 as used here or similar).
12. Tubing of outer diameter (OD) 6 mm and OD 4 mm (poly-
tetrafluoroethylene [PTFE]; see Note 1).
13. Tubing interconnects: KQ2 series, one-touch connectors
(SMC Pneumatics) as used here or similar.
14. Plant growth chamber.
15. Light sensor: Quantum LI-190 as used here or similar.
16. Temperature probe: PT-100 as used here or similar.
17. Computer control: A customized package run on the Lab-
VIEW software platform provides digital control of the envi-
ronmental parameters while recording output data (available
on request).
 Gas Exchange Measurements of Complete Foliar Parts of Plants 49

Fig. 2 The humidification column was designed to have the capacity to hydrate air at both low and high flow
rates. A central spire (i) attached to the lid (ii) carries input (iii) air to well below the water (iv) level. The lid is
fastened by six bolts which compress a gasket to create an airtight seal. An air diffuser can be attached to the
bottom of the spire to create small bubbles in order to increase the total air-water surface area. Three open
mesh platforms (v) and an open-base container (vi) are attached to the central spire. The mesh impedes
bubble movement allowing for longer contact with the water. The container houses crushed, wet ceramic
through which the air must pass before exiting (viii) the column at the lid. The “S” denotes the cross section
displayed in “S–S”
50 Gavin M. George et al.

Fig. 3 Plant chambers (a) are constructed with plastic supporting structures (i) and an aluminum stage (ii). The
base of the chamber is constructed in two halves which close around the plant stem at the foam seal (iii),
along the stage union with a rubber lip (iv), and are held firmly in place by two tightening bolts (v). The lids
(vi) are fastened by a thread which compresses a foam gasket (vii). Air is transferred in (viii) and out (ix) of the
 Gas Exchange Measurements of Complete Foliar Parts of Plants 51

18. Data analysis software: The frequently large datasets generated

by EGES-2 are suitable for automatic processing for which we
have a custom-written application (Supplemental File 2).

2.2 Plant Growth Various substrates can be used for plant growth and will depend on
Substrates the physiological experiment (see Note 2).
1. Soil.
2. Hydroponics (Cramer’s) solution: 1.5 mM Ca(NO3)2,
1.25 mM KNO3, 0.75 mM Mg(SO4), 0.5 mM KH2PO4,
1 mM (NH4)2SO4, 72 μM C10H12FeN2NaO8, 100 μM
Na2O3Si, 50 μM KCl, 10 μM MnSO4, 1.5 μM CuSO4, 2 μM
ZnSO4, 50 μM H3BO3, and 0.075 μM (NH4)6Mo7O24
adjusted to pH 6 with HNO3 (based on [8]), or similar.
3. Conical microfuge tubes (0.5 mL) prefilled with sterile agar
(0.5% m/v).

2.3 Seed Sterilization 1. 70% (v/v) ethanol containing 0.015% (v/v) Triton X-100.
for Hydroponics 2. 100% Ethanol.
and Defined Substrate 3. Sterile H2O.
Growth (See Note 3)

2.4 Stress Induction 1. Aluminum foil: For premature or extended nights.

(See Note 2) 2. Salt stress solution: Hydroponics solution supplemented with
200 mM NaCl.
3. Polyethylene glycol (PEG) hypotonic stress solution: Hydro-
ponics solution supplemented with 160 g.L
1 (m/v)
PEG-6000.

3 Methods

3.1 Construction
of the Multi-chamber,
IRGA-Linked, Gas
Switching Unit
1. High-pressure gas supply: Before assembly close the root air
supply valve. For all connecting tubing in the high-pressure
part of the device, use OD 6 mm PTFE tubing (see Note 1).
Attach a regulator to the compressed atmospheric air supply
and set it to provide a pressure of 400 kPa. Connect the re-
gulated air sequentially to the molecular scrubber (see Note 4),
the air accumulator, and then an air reservoir (Fig. 1). From
this point the air must be passed through three further regula-
tors which step down the gas pressure to 200-, to 100-, and

Fig. 3 (continued) plant chamber through passages integrated into the stage. The base of the chambers (b)
can accommodate a range of pots or hydroponics tanks. The “S” marks the plane of the cross section shown.
A range of lids (c) allow for various plant species to be measured. From left to right, 113 cm3, 196 cm3, and
530 cm3 are shown
52 Gavin M. George et al.

then to 50 kPa. We discovered empirically that these pressure-

step regulators, along with the upstream air accumulator and
compressed air reservoir, were sufficient to stabilize any detect-
able pressure fluctuations. It is important to ensure that no
air leaks exist in the high-pressure portion of the device
(see Note 5). The gas that is now supplied will be at the
operating pressure for the remainder of the system and will be
effectively free of CO2 and H2O.
2. The CO2- and H2O-free air supply must be first connected to
the humidification system consisting of a hydration column (see
Note 6) and a digital proportional valve, dry-air bypass. The
column itself (Fig. 2) should contain water in the lower portion
and wet, crushed ceramic in an upper basket. The dry air is first
bubbled through the water and then pushed through the wet
ceramic to achieve a high level of humidity. The digital bypass
can then reduce the humidity to the required level by allowing
dry air to reenter the gas line (see Note 7).
3. Attach a bottle of pressurized 10–100% CO2, regulated to
50 kPa, to a digitally modulated MFC in line but after the
humidification unit. Both the digital bypass of the humidifica-
tion system and the MFC on the CO2 supply can be actively
controlled by the computer software on a feedback loop from
the IRGA to produce air which contains precise and adjustable
concentrations of CO2 and H2O.
4. Attach tubes so that the air is split to travel to the reference gas
cell (Cell A) on the IRGA, and also to the eight plant chambers
(Fig. 1). Each of the channels supplying air to the plant cham-
bers must be independently regulated by a MFC (see Note 8).
The returning air from each channel must be attached to the
gas switching unit (see Note 9). This device sequentially directs
returning air from a chamber to the measurement cell (Cell B)
of the IRGA. After a period of time, which can be set in the
software, the next chamber is switched to supply the IRGA
while the air from the previously measured chamber is vented.
In this way, all plant chambers are constantly supplied with
defined air but only one is measured at any time. The ΔCO2
and ΔH2O concentrations measured in Cell A and Cell B are
later used for the calculation of assimilation or transpiration by
the plant. This system allows for up to eight plant chambers to
be measured periodically or one continuously for periods of
minutes to days or more.
5. The plant chambers should be constructed so as to limit the
possibility of air leaks once they are closed (see Note 10). Sche-
matics for our design are provided in Fig. 3a–c (see Note 11).
6. Temperature and light sensors must be attached to the system
and placed near the plant chambers during an experiment in
 Gas Exchange Measurements of Complete Foliar Parts of Plants 53

order to accurately track these parameters. All of these data are

collected by the software package and recorded in a text file for
later processing.

3.2 Introduction Plants can be grown in either soil or hydroponics systems.

of Plants
1. For soil: Soon after germination, transfer individual plants to a
for Measurement pot which will fit into the base of the gas-exchange plant
chamber. The soil must fill these pots so that the stem of the
plant will be above the rim of the pot allowing for alignment
with the foam seal of the plant chamber (see Note 11). Alter-
natively, soft plastic pots can be used and cut down to soil level
just before mounting.
2. For hydroponics: Fill conical tubes (0.5 mL) with sterile agar
(0.5% m/v) and allow to cool before removing the bottom
cone of the tube as well as the lid with a sharp blade (see Note
3). Place a single sterilized seed on top of the agar on the lid
side of the tube. Place the tube in the rack of a tip box (for
200 μL tips) which has been filled with hydroponics solution so
that the opening at the bottom of the tube is immersed in the
liquid.
3. Grow plants for 3–4 weeks in controlled conditions until ready
for gas exchange measurements. For Arabidopsis we recom-
mend a 12- to 16-h photoperiod, 150 μmol quanta m
2 s
1 at
20  C, and 60% relative humidity (RH).
4. Select plants which are at a similar developmental stage for
measurement (see Note 12).
5. Open the plant chambers and place a small amount of soft
paraffin on the edges of the foam gasket which will make
contact with each other and the plant’s stem, creating an
airtight seal when closed (see Note 10).
6. Place the pot in the open plant chamber base so that the stem
makes contact with the attached foam gasket (see Note 11).
7. Close the base of the plant chamber by sliding the second half
of the stage toward the first while gently lifting leaves out of the
way. Be careful to clear any dirt or soil debris that may interfere
with the union of the stage plate upon closing the chamber base
since these may create air leaks.
8. Use the tightening bolts to secure the two halves of the base
together.
9. Repeat until all required chambers are occupied. If less than
eight chambers are used, unoccupied channels can be inacti-
vated (see Note 9).
10. Place the chambers, without the lids, in a growth cabinet to
acclimate the plants as required (see Note 13).
11. Attach plant chamber air supply and return tubes.
54 Gavin M. George et al.

3.3 Preparation 1. Open the root air supply valve and the supply valve of the CO2
and Calibration bottle.
of the System 2. Confirm that the pressures of the root air (400 kPa), the three
regulators (200, 100, and 50 kPa) after the molecular scrubber,
as well as that of the CO2 (50 kPa) supply are correct by
looking at the built-in gauges.
3. Power on the molecular scrubber, gas switching unit, IRGA,
and computer.
4. Start the environmental control (LabVIEW) software.
5. Open the plant chamber air bypass so that the reference air
measured in Cell A is immediately routed to Cell B in
the IRGA.
6. In the software, set the concentration of CO2 and H2O to zero
and then up to 400 μmol mol
1 CO2 and 16 mmol mol
1
H2O (see Note 14). If the system responds to both of these
commands, then it will confirm that the root air supply, molec-
ular scrubber, CO2 supply, and dry-air bypass are functioning
normally.
7. Allow the system to stabilize and then run for 30 min during
which time the supplied and ΔCO2 and ΔH2O concentrations
will be plotted in the appropriate real-time windows of the Lab-
VIEW software (Fig. 4a). All of these values should be stable to
1 μmol mol
1. If this is not the case, then check for air leaks in
the high-pressure portion of the device (see Note 5).
8. Calibrate the IRGA so that the value measured by Cell A
matches that of Cell B for CO2 and H2O concentrations;
ΔCO2 and ΔH2O will now be zero (Fig. 4b).
9. Close the bypass valve so that air from the gas switching unit is
directed to Cell B of the IRGA.

3.4 Defining 1. Set the required concentration of CO2 and H2O for the
Experimental intended experiment. For Arabidopsis, we use 380 μmol
Parameters mol
1 CO2 and ca. 15 mmol mol
1 H2O, which will yield
and Starting 60% RH at 20  C [9], as our standard conditions.
Measurements 2. Set the airflow rate which will pass through the plant chambers.
A typical experiment with Arabidopsis requires a flow rate of
300 mL min
1 (see Note 15).
3. Set the waiting time to 90 s. Upon switching the measured
chamber, data is not recorded in the data file during the waiting
time. This is required for the air returning from the newly
measured chamber to fully displace that of the previously
measured one in the tubing between the gas switches and
Cell B in the IRGA (Fig. 1; see Note 9). A waiting time of
90 s is conservative for a 300 mL min
1 flow rate; however,
lower airflow rates may require longer waiting times and vice
versa.
 Gas Exchange Measurements of Complete Foliar Parts of Plants 55

Fig. 4 The control and real-time measurement interface of the LabVIEW software. From here all environmental
parameters of the EGES-2 can be set and adjusted as required. The real-time graphical output of the device
when (a) running with all chambers containing plants and (b) correctly calibrated and run on bypass mode. (c)
CO2 and H2O control panes allowing the user to define the set points (SP) which will be targeted manually or by
the feedback loop control
56 Gavin M. George et al.

4. Set the measuring time. We routinely measure for 5 min; how-

ever, this can be adjusted based on the requirements of the
experiment to a value between 1 and 60 min. Shorter measur-
ing times are not recommended, particularly if the waiting time
is also short (see Note 9).
5. Set the interval between measured data points. This value is the
rate at which data points are generated in the output file (see
Note 16).
6. Close the lids of the plant chambers (see Note 17).
7. Check for pressure loss on the plant chamber return air bubble
meter for each channel. The returning air should be the same as
the user-defined flow rate. A lower reading indicates an air leak
in the chamber (see Note 10).
8. Create a file in the LabVIEW software to which data will be
saved.
9. Allow the system to automatically measure gas exchange of the
plants for hours or days, as required. The environmental con-
ditions, such as CO2 and humidity, supplied to all of the
chambers can be changed at any time during an experiment.
Additionally, some environmental parameters can be altered for
single or a subset of chambers. The intensity of the light or
temperature supplied can be changed by moving a chamber to
a different plant growth cabinet. Light can also be selectively or
completely excluded by covering chambers with a filter or
aluminum foil. Conversely, specific wavelengths could be sup-
plied to the plants by additional lighting. Furthermore, the
airflow rate can also be adjusted on a per-chamber basis so
that accurate measurements can still be acquired under the
changed conditions.

3.5 Concluding 1. Copy data file generated during the experiment to a new loca-
an Experiment tion and delete the file path in the LabVIEW software.
2. Open the plant chamber bypass and record the ΔCO2 values in
order to confirm the stability of the calibration between Cell A
and Cell B. Any drift should be compensated for in the analysis
of the experimental data (see Note 18).
3. Open the plant chamber lids and unplug the air supply and
return tubes.
4. Use scissors to harvest the entire portion of the plant which was
contained above the foam gasket, i.e., inside the chamber, and
record the fresh mass (FM) using a balance (see Note 19).
5. Photograph each plant on white paper, avoiding overlapping
leaves, with a ruler in frame for the calculation of leaf area (LA;
see Note 19).
 Gas Exchange Measurements of Complete Foliar Parts of Plants 57

6. Place individually bagged plants in an oven (65  C for 3 days)

before reweighing on a precise balance to determine the dry
mass (DM; see Note 19).

3.6 Data Analysis Data generated during an experiment is stored in a user-defined

text file. Each entry records the date, time, measured channel,
airflow rate, supplied and ΔCO2 and ΔH2O as well as temperature
and light intensity. The ΔCO2 and ΔH2O gas exchange values can
be converted to assimilation (A) and transpiration (E) rates using
previously described methods [10] (see Note 20). Due to the large
amount of data that can be produced by EGES-2, we developed a
database and processing application which is detailed in Note 20
(Supplemental File 2).

3.7 Example Photosynthesis is dependent on light, CO2, and water. Using

Physiological EGES-2 and a standard plant growth cabinet, these environmental
Experiments Using parameters, as well as humidity and temperature, can be set before
the EGES-2 or altered during an experiment. In addition, the sensitivity of this
system allows for accurate measurements of respiration rates at
night. This allows for a range of possible experiments that other
systems may not be capable of (see Note 21). A number of standard
experiments have already been described [6]. The following three
examples, however, have been chosen to highlight the potential
application of EGES-2 or a similar device.

3.7.1 Decoupling Light
and CO2-Driven
Assimilation in Arabidopsis
thaliana
Conditional assimilation profiles reveal when CO2 or light is limit-
ing (Fig. 5). This information can be used to alter assimilation rates
to a specific level through the supply of CO2 or light. Similarly,
assimilation rates can be maintained at a set rate through opposing
but complementary adjustments. It is true that measurements such
as these could be achieved, perhaps more rapidly, with a leaf-clip
cuvette. However, our system has the distinct advantage of being
able to subject the entire aerial part of the plant to the varied
environments allowing for downstream analyses such as metabolite
quantification or transcriptional profiling (see Note 21).
1. Mount 3- to 4-week-old Arabidopsis plants in the plant cham-
bers. In this example, use only four wild-type, Col-0, plants.
2. In the LabVIEW software, set only the occupied chambers to
be measured (Fig. 4a).
3. Initially set the environmental parameters to light intensity of
150 μmol m
2 s
1, RH of 60%, CO2 concentration to
400 μmol mol
1, and airflow rate to 300 mL min
1.
4. Measure the four chambers for four complete cycles of the gas
switching unit.
58 Gavin M. George et al.

25

20
Assimilation rate
(nmol gFM⁻⁻¹ s⁻¹)

15

10

200
5 150
NM NM NM 100
50
NM NM NM NM NM 30
0
100 200 300 400 500 600 700 800
Supplied CO2 concentration (µmol mol-1)

Fig. 5 Photosynthetic assimilation rates are dependent on supplied CO2 and light concentrations. Similar rates
of assimilation can be reached by decreasing one and increasing the other, allowing for decoupling of the two
variables. NM refers to conditions not measured

5. Step the supplied CO2 concentration from 100 to 200, 300,

600, and then to 800 μmol mol
1 to acquire measurements of
all four chambers for a minimum of three cycles each.
6. Repeat steps 2–5 at a range of supplied CO2 concentrations
while stepping the light intensity from 30 to 200 μmol
m
2 s
1.
7. Process the resulting data to acquire the assimilation rates at the
varied conditions.

3.7.2 Comparison of Two
Nicotiana sylvestris
Genotypes’ Assimilation
and Respiration Rates
in a Normal Day/Night
Cycle Or with a Premature
Night
Gas exchange can be used to investigate carbon starvation in plants.
Here we employ two different modes of induction of starvation: the
first is genetic and the second in a physical truncation of the day.
Plastidial phosphoglucomutase (pgm) mutants are unable to pro-
duce transitory leaf starch [11]. Instead the products of assimilation
are partitioned toward sugars, which in turn lead to an increased
and decreased rate of respiration at the beginning and end of the
night, respectively. This can be seen by the negative assimilation
rates (i.e., CO2 release) during the night in Fig. 6a. Introduction of
a premature night on day 3 results in a similar profile; however,
respiration in the wild type also decreases at the end of the night as
its carbohydrate reserves begin to be depleted.
1. For the comparison of two genotypes, mount four of each in
the plant chambers, place in a growth cabinet, and connect to
the gas switching unit. In this case plastidial pgm knockout
mutants were compared to wild type.
 Gas Exchange Measurements of Complete Foliar Parts of Plants 59

2. Set the airflow rate to 300 mL min
1, humidity to 60% RH,

CO2 to 400 μmol mol
1, and light intensity to 150 μmol
m
2 s
1.
3. Collect measurements in complete day/night cycles, i.e., from
beginning of day to the next beginning of day, for as long as
required.

Fig. 6 Gas exchange profiles of (a) Nicotiana sylvestris wild-type (purple dots) and pgm (green dots) and (b)
hydroponics-grown Arabidopsis thaliana plants subjected to treatment with either PEG-6000 (16% m/v; blue
dots) or sodium chloride (200 mM; red dots)
60 Gavin M. George et al.
3.7.3 Quantification
of Differences in Carbon
Assimilation Rates Affected
by Osmotic Stresses
Applied to the Roots
of Arabidopsis thaliana
in a Hydroponics System
Over Time
4. To introduce a premature night, cover the chambers with
aluminum foil and continue measurements through the course
of the night. Extending the night can be achieved in a
similar way.
5. Record LA, FM, and DM for rate calculations as described in
Subheading 3.5.
6. Calculate and plot the assimilation and respiration rates
(Fig. 6a).
Many stresses which a plant may encounter, such as nitrogen or
water limitation, temperature, and light quality or quantity, will
have an effect on the photosynthetic rate in the short or long term.
An experiment may, therefore, call for different genotypes to be
treated with the same or varied stress/es to measure both short-
and long-term responses. In this example, wild-type plants were
treated with either 200 mM NaCl or 16% (w/v) PEG-6000
(Fig. 6b). The salt treatment induced a rapid response in the rate
of assimilation likely due to “drought”-induced stomatal closure.
Both treatments, in addition, elicited a long-term response over the
following days. The ability to continuously measure these long-
term adaptive responses highlights a key advantage of the EGES-2.
1. Sterilize seeds by treating them with sequential washes of 5 min
each with 70% (v/v) ethanol containing 0.015% (v/v) Triton
X-100, 100% (v/v) ethanol, and sterile water. Sow individual
seeds on top of the sterile 0.65% agar preset in a 0.5 mL conical
tube with the bottom tip removed (see Note 3). Place these, in
turn, in a standard tip box (for 200 μL tips) filled with hydro-
ponics solution.
2. During 2 days of stratification, cover the tip box with its lid,
and then remove before transfer to a growth cabinet.
3. Grow plants for 3 to 4 weeks exchanging the hydroponics
solution every 3 to 4 days.
4. Transfer the plants to the gas-exchange chambers with the tip
box containing hydroponics medium in the base. Fill the tip
box with medium so the roots are completely submerged (see
Note 11).
5. For measurement, set the airflow rate to 300 mL min
1,
humidity to 60% RH, CO2 to 400 μmol mol
1, and light
intensity to 150 μmol m
2 s
1.
6. Take measurements before any treatment is applied (see Note
22).
7. Treat the plants with a stress solution by siphoning off the
hydroponics solution contained in the tip box and replacing it
with either new hydroponics solution containing 16%
 Gas Exchange Measurements of Complete Foliar Parts of Plants 61

PEG-6000 or the more severe osmoticum, 200 mM NaCl (see

Notes 11 and 22).
8. Continue measurements for several days (see Note 23).
9. Finally, conclude the experiment by recording LA and FM and
DM, and process results (see Note 19).

4 Notes

1. Plastics and other materials may have capacity for CO2 diffu-
sion as well as adsorption and subsequent desorption which
may lead to imprecise measurements. We, therefore, use poly-
tetrafluoroethylene (PTFE) tubing and ethylene-propylene-
diene foam where required.
2. EGES-2 can accommodate plants potted in soil or grown in
hydroponics. Soil is a natural substrate for plants; however, it
may be advantageous to measure hydroponically grown plants.
For example, in Subheading 3.7.3, we apply a precise osmotic
stress to the roots of a plant. Media can be adapted for experi-
ments where phosphate, nitrogen, or indeed any nutrient can
be made replete, limiting, or absent. Stresses can also be applied
to the foliar portion of the plant since light and atmospheric
conditions can be altered as required.
3. While plants grown in hydroponics solutions are not sterile, we
take care to sterilize seeds and the agar before germination.
Once the conical tube is immersed in the hydroponics solution,
nutrients diffuse into the agar which will supply the plant but
will also support bacterial and fungal growth. Failure to steril-
ize often leads to microbial contamination around the seed
during germination which may kill the plant at this early
stage. After germination, this is no longer a problem presum-
ably due to the plant’s developed defense mechanisms.
4. The dual-tower CO2 adsorber used in EGES-2 (MCA-6, Pure-
gas, USA) functions by switching the gas flow between col-
umns every 30 s allowing for alternating sequestration and
regeneration. This allows for continuous use but also produces
pressure fluctuations. Infrared gas analyzers, such as the
LI-COR 7000 used here, require a constant pressure in order
to gather accurate measurements, and so these fluctuations
must be dampened. This is achieved here by, first, performing
the CO2 and H2O scrubbing at a higher pressure of 400 kPa.
Then, this air is passed sequentially through an air accumulator,
a large air reservoir, and then three regulators. Each of these
regulators step down the air pressure to 200, 100, and lastly
50 kPa. Should a different molecular scrubber be used when
62 Gavin M. George et al.

building the system, the downstream pressure effects would

need to be empirically considered and regulated accordingly.
5. Air leaks in the over-pressurized portion of the system result in
instability which cannot necessarily be overcome by the digital
feedback controls. To check for leaks, dish soap foam can be
placed around connection points. Bubbling will indicate the
presence of an air leak.
6. Humidification of the air must be performed at the pressure at
which the system is run. Subsequent decreases in pressure will
result in condensation which may damage the instrument and
make measurements impossible.
7. The temperature probe is not required for the modulation of
the concentration of H2O but it is needed to be able to deter-
mine the resultant % RH in the supplied air.
8. Airflow can be set from 50 to 1000 mL min
1. Ideally the flow
rate should be set to a rate per minute that is at least threefold
larger than the volume of the chamber being supplied. Lower
flow rates can be tolerated but may yield more momentary
variation due to poor gas mixing. In such cases, each plant
chamber should be measured for no less than 5 min allowing
for accurate averages. Alternately, miniature electric mixing
fans could be installed in the lids; however, we have not
found this to be necessary under our tested conditions.
9. Each channel terminates at a gas switch which either vents the
returning air or directs it toward Cell B of the IRGA for
measurement (Fig. 1). The gas switches and the IRGA are
connected by a common tube which carries the air of the
channel currently being measured. As such, upon switching
from one channel to the next this tube will still contain the
air of the previously measured channel. For this reason, a wait-
ing time is introduced where readings are not logged in the
data file for a period of time that allows air from the newly
measured chamber to displace that of the previously measured
one. The LabVIEW software controls which channel is open to
the IRGA and switches through the measured channels
sequentially at a user-defined interval. As such, the number of
chambers used for an experiment can also be set using the
toggles on the left of the control panel in the LabVIEW soft-
ware (Fig. 4a).
10. Small air leaks around the stem of the plant can be tolerated by
the system but may affect the waiting time required between
channel switching before a stable reading is obtained
[6]. Larger leaks, however, will affect the concentrations of
CO2 and H2O measured by the IRGA and should be avoided.
This is especially important when supplying the chambers with
Gas Exchange Measurements of Complete Foliar Parts of Plants 63

very high or low CO2 and H2O concentrations compared to

the outside environment.
11. The plant chambers were designed in two halves that meet at
the stem of the plant. Two bolts compress a rubber lip which
seals the interface of the two parts of the stage while soft foam
and a small amount of paraffin wax create an airtight seal. Once
the lid is on, air can only enter and leave the chamber via the
respective passages. The base can accommodate up to a stan-
dard 6 cm height pot. We routinely use smaller pots placed on a
layer of water-conducting cloth, inside a petri dish allowing for
watering of the plants in long-term experiments. Hydroponics
containers can also be accommodated (see Note 2). Before
mounting hydroponically grown plants, the conical tube in
which they were germinated should be carefully removed leav-
ing just a small amount of agar around the root. This is done by
holding the plant stem and sliding the tube down past the
roots. The plant can then be mounted in the chamber with
the roots submerged in hydroponics medium. Mounted in this
way, the plant is completely supported by the foam gasket
around the stem and the tip rack is no longer required. As
the vessel containing the medium is not fixed in position, it can
be slid to one side while the roots are still submerged. This
allows access for medium exchange as required.
12. We have found that the most consistent results are obtained
when plants, within and between genotypes, are as similar in
size and developmental stage as possible. If a mutation causes a
significant difference in growth rate, then we recommend
staggering germination so that the control plants are a similar
size to the mutants at the time of measurement.
13. When an experiment is run for less than a day, we recommend
setting plants in the chambers at least a day before to allow time
for acclimation and recovery from the mechanical stresses that
result from handling and introduction of the plant to the
chambers. We have observed that these mechanical stresses
appear to have an effect on photosynthetic and growth rates
but they recover within 24 h. For this reason, we have con-
structed a duplicate set of eight chambers so that plants can be
acclimated in advance avoiding measurement at the time of
handling and allowing for continuous use of EGES-2.
14. The concentration of supplied CO2 (and H2O) can either be
set manually or by the digital feedback loop (Fig. 4c). The
feedback loop provides the precise concentrations of CO2 set
by the user by measuring the concentration of CO2 in the
reference air (Cell A) and dynamically adjusting the supply
(Fig. 1). On the other hand, the manual mode introduces
CO2 at a calculated rate based on the total airflow. This mode
64 Gavin M. George et al.

is generally accurate to 10% but can be further adjusted in the

control software to match the target concentration of CO2.
Manual supply mode is useful when alternate isotopes of CO2
are provided such as 13CO2, which would be invisible to the
IRGA, and could not be regulated by the feedback loop.
15. The airflow rate used in an experiment should be determined
empirically so that the measured ΔCO2 is between 5 and
50 μmol mol
1. This will ensure that the measurements are
well within the detectable and linear ranges of the IRGA. We
routinely target a ΔCO2 value of 20 μmol mol
1. Lower flow
rates will result in greater Δ values measured. Flow rates can be
set for each chamber individually which may be useful when
comparing mutants with severely impaired CO2 assimilation or
respiration rates with wild-type plants. Higher flow rates may
be required when larger plants, such as Pisum sativum, are
being measured.
16. Short measuring intervals should be avoided if the data will be
processed manually due to the size of the resulting dataset. A
1-s interval over the course of a single day will yield a file with
86,400 records of all measured parameters, which is difficult to
process in typical spreadsheet applications. If processing is to be
done in this way a 10–30-s interval should be used. We further
recommend that the measuring time be a minimum of five
times the interval. However, if data processing will be per-
formed automatically, then intervals of 1 s could be used to
yield high resolution of any fluctuations in assimilation or
transpiration.
17. We have four types of lids with differing dimensions and
volumes to accommodate a variety of plant species. The smal-
lest lid has a volume of 38 cm3, is wide in diameter (7 cm) but
low in height (1 cm), and is suitable for prostrate species such
as Arabidopsis. In addition, this lid has accessory, transparent
inserts which can reduce the diameter to 5 cm and, thereby, the
chamber volume to 20 cm3 for the accurate measurement of
small plants at low flow rates. The three further lid volumes are
113 cm3, 196 cm3, and 530 cm3 and are suitable for species
such as Mesembryanthemum crystallinum, Lotus japonicus, and
Pisum sativum, respectively (Fig. 3c).
18. After the conclusion of an experiment, it is important to deter-
mine whether any calibration drift has occurred between the
two cells of the IRGA. This often occurs in experiments which
span several days. The drift is usually less than 1–2 μmol mol
1
for CO2; however, this may be critical to compensate for when
measuring mutants with low assimilation or respiration rates.
We do not measure drift during the experiments; however, if it
is observed at the conclusion of an experiment, then we assume
Gas Exchange Measurements of Complete Foliar Parts of Plants 65

that it occurred at a linear rate and compensate for it accord-

ingly. Each data point is adjusted by a proportional addition of
the drift value divided by the total number of data points
measured. This adjustment can be performed automatically in
the data analysis software (Supplemental File 2).
19. In order to calculate photosynthetic rates from the ΔCO2 and
ΔH2O provided by EGES-2 a normalization parameter is
required. This can be either leaf area (LA), fresh mass (FM),
or dry mass (DM). For prostrate plant species such as Arabi-
dopsis, LA can be obtained by photographing the plant in the
chamber with a ruler in the frame. Care should be taken to
capture the photograph with the lens directly above the plant
to avoid errors in perspective which may affect later analysis.
Some plant species may require leaves to be carefully separated
to avoid overlap and underestimation of the LA. Should
destructive harvesting be performed after an experiment,
then we suggest taking a photograph of the plant, on a white
paper background, with the leaves carefully removed and
arranged flat in the photograph frame. This will aid in accurate
measurements of LA, particularly in species with long stems
such as Lotus japonicus, and Pisum sativum. The area can then
be calculated readily using freely available software such as
ImageJ [12]. To obtain FM measurements, we harvest the
plants and immediately weigh them on an accurate balance.
These plants can then be placed in a non-stick paper bag, dried
at 65  C for 3 days, and weighed again to obtain DM measure-
ments. Obtaining all three measurements is recommended
since these parameters change in plants with different ecophy-
siologies and between genotypes within a species.
20. We have provided a spreadsheet which includes the required
calculation in order to process ΔCO2 and ΔH2O measurements
to assimilation and transpiration rates (Supplemental File 1).
Spreadsheet analysis of the data generated by EGES-2 is time
consuming due to the high degree of variability of the data
structure in the output file. This can change based on the
number of chambers being read, measurement interval, waiting
time, measurement time, or read interruptions. For this reason,
we developed a data processing and graphing application to
handle large datasets (Supplemental File 2). It functions by
storing experiment-related information on the plants in each
chamber including their genotype, measured masses, and/or
leaf areas, as well as the day length and any calibration drift in a
MySQL database. To this, the EGES-2-generated output file is
uploaded. An SQL query of the data is then executed by the
user through a PHP-generated web interface. The interface
allows for selection of data for analysis by a time range and the
exclusion of specific chambers while specifying the
66 Gavin M. George et al.

normalization parameter (FM, DM, or LA) to be used for the

analysis. The query generates a summary of the data which
includes the processed assimilation and transpiration rates.
These can be aggregated by the initially defined genotypes in
each chamber as averages with standard deviations or repre-
sented as individuals. The summary can then be downloaded
as a text output file or used to generate a graphical output (D3:
Data-Driven Documents JavaScript library) such as those in
Fig. 6.
21. Leaf-clip cuvettes offer a convenient method for measuring the
gas exchange of a single leaf, and also allow the environment of
that leaf to be altered in much the same way as EGES-2 can.
However, there are several key advantages to EGES-2 which
may make it more appropriate for certain experiments. Firstly,
mechanical stress on the plant is limited to the time when the
plant is introduced into the chamber. A recovery period can be
introduced so that little or no handling of the plant occurs on
the day of the experiment (see Note 13). Secondly, unlike a leaf
cuvette, the entire rosette or foliar part of the plant is subjected
to the same environmental conditions. This allows for treat-
ment of the entire foliar parts of the plant and then subsequent
sampling from individual leaves or organs, all treated iden-
tically. This is particularly important in small herbaceous spe-
cies such as Arabidopsis, where the bulk of a leaf-clip cuvette
will cause significant shading of the unmeasured parts of the
plant. Experiments where photosynthesis may be increased
through supplied CO2 using a clip-on system may be compli-
cated by increased sugar export to untreated leaves, or vice
versa. This could conceivably introduce systematic errors in
downstream metabolite or transcriptional analyses; EGES-2-
avoids this. A further advantage of this system is that treat-
ments can be extended for days on replicated samples that are
measured in an automated way, which is impractical with a
single leaf clip. With a leaf clip, a similar result might be
achieved with repeated measurements of the same plants over
the course of the experiment. However, as a leaf clip requires an
airtight seal around the leaf, repeated measurements of the
same area on fragile plants, such as Mesembryanthemum crystal-
linum, will quickly lead to damage. This is also avoided by
EGES-2.
22. Osmotic stress can be applied with, but not limited to,
PEG-6000 or salts such as sodium chloride. The high molecu-
lar weight (limiting root absorption), solubility, and low toxic-
ity of PEG allow for a broad range of osmotic pressures to be
applied and provide a suitable analogue for soil dehydration in
hydroponics systems [13]. Sodium chloride treatment, on the
other hand, causes Naþ ions to accumulate in the leaf leading
 Gas Exchange Measurements of Complete Foliar Parts of Plants 67

to direct osmotic effects on the cells of the shoot [14]. The

osmotic pressure generated by both PEG-6000 and NaCl can
be varied greatly as required by the experiment [15, 16]. We
recommend that when performing similar experiments, irre-
spective of what the treatment may be, the plants be measured
before the treatment is applied. Ideally, if a single genotype is to
be analyzed by application of two different treatments, then the
initial measurements should be similar and will act as control
measurements. In this regard, we suggest that all chambers are
measured for a minimum of three cycles.
23. In an experiment where, for instance, a mild stress is applied it
may take days or weeks before a notable change in assimilation
or transpiration can be detected. In such cases, the data file can
be copied and preliminarily processed by normalizing to the
leaf area (see Notes 19 and 20). In this way it can be established
whether the experiment can be stopped or extended.

Acknowledgments

Thank you to Daniel Carrera and Elisabeth Truernit for their advice
and assistance.

References

1. Stitt M, Zeeman S (2012) Starch turnover: 7. Elliott C, Vijayakumar V, Zink W, Hansen R

pathways, regulation and role in growth. Curr
Opin Plant Biol 15:282–292
2. LI-COR Biosciences (2016) LI-6400XT sys-
tem. www.licor.com/env/products/photosyn
thesis. Accessed 5 May 2017
3. Duan Z, Homma A, Kobayashi M, Nagata N,
Kaneko Y, Fujiki Y, Nishida I (2015) Photo-
assimilation, assimilate translocation and plas-
modesmal biogenesis in the source leaves of
Arabidopsis thaliana grown under an increased
atmospheric CO2 concentration. Plant Cell
Physiol 55:358–369
4. Vanhaeren H, Gonzalez N, Inzé D (2015) A
journey through a leaf: phenomics analysis of
leaf growth in Arabidopsis thaliana. Arabidop-
sis Book 13:e0181
5. Tanaka Y, Sugano SS, Shimada T, Hara-
Nishimura I (2013) Enhancement of leaf pho-
tosynthetic capacity through increased stoma-
tal density in Arabidopsis. New Phytol
198:757–764
6. Kölling K, George GM, Künzli R, Flütsch P,
Zeeman SC (2015) A whole-plant chamber
system for parallel gas exchange measurements
of Arabidopsis and other herbaceous species.
Plant Methods 16:11–48
(2007) National instruments LabVIEW: a pro-
gramming environment for laboratory auto-
mation and measurement. JALA 12:17–24
8. Gibeaut DM, Hulett J, Cramer GR, Seemann
JR (1997) Maximal biomass of Arabidopsis
thaliana using a simple, low-maintenance
hydroponic method and favorable environ-
mental conditions. Plant Physiol 115:317–319
9. Buck AL (1981) New equations for computing
vapor pressure and enhancement factor. J Appl
Meteorol 20:1527–1532
10. von Caemmerer S, Farquhar GD (1981) Some
relationships between the biochemistry of pho-
tosynthesis and the gas exchange of leaves.
Planta 153:376–387
11. Huber SC, Hanson KR (1992) Carbon parti-
tioning and growth of a starchless mutant of
Nicotiana sylvestris. Plant Physiol
99:1449–1454
12. Schneider CA, Rasband WS, Eliceiri KW
(2012) NIH Image to ImageJ: 25 years of
image analysis. Nat Methods 9:671–675
13. George GM, Ruckle ME, Lloyd JR (2015)
Virus-induced gene silencing as a scalable tool
68 Gavin M. George et al.
to study drought tolerance in plants. Methods
Mol Biol 1287:243–253
14. Berthomieu P, Conéjéro G, Nublat A et al
(2003) Functional analysis of AtHKT1 in Ara-
bidopsis shows that Na(þ) recirculation by the
phloem is crucial for salt tolerance. EMBO J
22:2004–2014
15. Coons JM, Kuehl RO, Simon NR (1990) Tol-
erance of ten lettuce cultivars to high tempera-
ture combined with NaCl during germination.
J Am Soc Hortic Sci 115:1004–1007
16. Michel BE, Kaufmann MR (1973) The
osmotic potential of polyethylene glycol
6000. Plant Physiol 51:914–991
 Chapter 4

Measuring Canopy Gas Exchange Using CAnopy

Photosynthesis and Transpiration Systems (CAPTS)
Qingfeng Song and Xin-Guang Zhu

Abstract
Canopy photosynthesis (Ac), rather than leaf photosynthesis, is critical to gaining higher biomass produc-
tion in the field because the daily or seasonal integrals of Ac correlate with the daily or seasonal integrals of
biomass production. The canopy photosynthesis and transpiration measurement system (CAPTS) was
developed to enable measurement of canopy photosynthetic CO2 uptake, transpiration, and respiration
rates. CAPTS continuously records the CO2 concentration, water vapor concentration, air temperature, air
pressure, air relative humidity, and photosynthetic photon flux density (PPFD) inside the chamber, which
can be used to derive CO2 and H2O fluxes of a canopy covered by the chamber. Here we describe the
protocol of using CAPTS to perform experiments on rice (Oryza sativa L.) in paddy field, wheat (Triticum
aestivum L.) in upland field, and tobacco (Nicotiana tabacum L.) in pots.

Key words Automatic canopy chamber, CAPTS, Gas exchange, Canopy photosynthesis, Canopy
transpiration, Canopy respiration, Rice, Wheat, Tobacco

1 Introduction

Canopy photosynthesis (Ac) is the total photosynthesis of an entire

canopy. The daily or seasonal integrals of canopy photosynthesis
correlate with the daily or seasonal integrals of biomass production.
Although measurement of leaf-level gas exchange is widely used to
study leaf photosynthetic properties, the measurement of canopy-
level gas exchange properties at the plot scale is relatively less
common due to technical challenges, such as the logistics difficulty
of moving the canopy chamber in the field and mixing gases in the
canopy chamber. Canopy gas exchange represents the total CO2
uptake (release) rate during the day (night) time of the entire
canopy and is influenced by the heterogeneity of microclimates
inside the canopy. Variations in photosynthetic photon flux density
(PPFD) at different layers of the canopy [1] can cause up to 25%
variation in total Ac. Similarly humidity and CO2 concentration
differ at different canopy layers. In addition, there are substantial

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_4, © Springer Science+Business Media, LLC, part of Springer Nature 2018

69
70 Qingfeng Song and Xin-Guang Zhu

variations in physiological properties, such as leaf photosynthesis,

for leaves at different layers of a canopy [2]. These heterogeneities
can be influenced by canopy architecture, chlorophyll content,
planting densities, planting directions, etc. These heterogeneities
make it difficult to calculate canopy photosynthesis using measure-
ments at the leaf scale.
Ac can be measured by micrometeorological approaches
including the Bowen ratio/energy balance (BREB) [3] and the
eddy correlation [4], which are suitable for large-scale (hectares)
gas exchange measurements, and by canopy chambers [5] including
open-chamber systems [6–10] and closed-chamber systems
[11–14], which are suitable for plot-scale (meter squares) gas
exchange measurement [15–19]. Canopy chambers are widely
used in plant physiology and ecology-related studies; for instance,
they have been used to study the influence of plant age on photo-
synthesis [20], to trace 13C in soil respiration [21], and to quantify
the effects of elevated CO2 concentration on photosynthesis and
transpiration [22]. The open canopy chamber system [6–10] mea-
sures stable CO2 concentration ([CO2]) or H2O concentration
([H2O]) at both inlet and outlet of a chamber and flow rate
under a controlled environment [6]. An air-conditioner for
controlling air temperature is usually needed as well as a fan with
a large flow rate to exchange gas in the chamber with the outside.
The measurement needs [CO2] and [H2O] to be stable, which
takes several minutes or longer time depending on the stability of
the environment. In contrast, the closed canopy chamber system is
more portable and has a lower cost than the open-chamber system
as it does not control air temperature and does not need to
exchange gas in the chamber with the outside. Closed-chamber
systems measure gas exchange rate by continuously recording the
changes of [CO2] and [H2O] in the chamber [11, 12] and then
calculating gas exchange-related parameters based on the slope of
gas concentration changes at the time of chamber closure. Linear
regression, modified rate regression, and quadratic regression
methods are used to calculate the slope of gas concentration
changes at the time of chamber closure [12, 23, 24]. Measurements
performed in closed-chamber systems take a short time (around
1 min) [24] and it is not necessary to control the environment, i.e.,
air temperature, [CO2], and [H2O] in the chamber [11]. Further-
more, the closed-chamber system is suitable for multi-chamber
measurement, i.e., it can simultaneously measure a number of
plant canopies [24]. This chapter describes a protocol for using
an automatic closed canopy photosynthesis and transpiration mea-
surement system (CAPTS) [13, 24] for rice (Oryza sativa L.),
wheat (Triticum aestivum L.), and tobacco (Nicotiana tabacum
L.) canopies. These examples are given to demonstrate how CAPTS
can be used on plants growing in paddy, upland, and potted soil
conditions, respectively.
 Measuring Canopy Gas Exchange with CAPTS 71

2 Materials

To measure canopy photosynthesis, we need CAPTS to cover plant

canopies or soil. The canopy or soil area being measured needs to fit
into the size of the chamber. In this section, we describe the plant
canopies of rice, wheat, and tobacco; the soil; CAPTS chamber; and
CAPTS controller needed for measurements.
1. Rice canopy: Sow rice seeds in paddy soil. Transplant the seed-
lings into a paddy field, planting seven rows with 20 cm row
distance and seven columns with 20 cm column distance using
a density of 25 plants/m2 (see Note 1). The chamber (with size
of 100 cm
100 cm) covers 25 plants in the center of the plot.
The border-protecting plants of the plot are not used for
measurements (see Note 2; Fig. 1a).
2. Wheat canopy: Sow wheat directly to soil in seven rows with
20 cm row distance, and in each row the average distance
between plants is 5 cm (see Note 1). Use the chamber to
cover the plants in the center of the plot (100 cm
100 cm).
Border-protecting plants surrounding the chamber are not
measured (see Note 2; Fig. 1b).
3. Tobacco canopy: Sow seeds in four pots, with one plant in each,
for measurements (Fig. 1c). Align the four pots to form a 2
2
matrix with a row distance of 50 cm and a column distance of
50 cm. Border-protecting plants are not required when leaves

Fig. 1 For CAPTS with a chamber size of 100 cm
100 cm, (a) a rice canopy (shown with filled green circles)
with a row distance of 20 cm and plant distance of 20 cm and (b) wheat canopy (shown with filled green bars)
grown with a row distance of 20 cm and distance between plants of 5 cm are used, respectively. Border-
protecting plants around the chamber (shown with empty green circles and bars) are used for rice and wheat
experiments in the field. (c) Four tobacco plants grown in pots form a tobacco canopy for CAPTS measure-
ments. No border-protecting plants are used when measuring plants in pots when the distance between plants
is large enough and there is no shading between plants
72 Qingfeng Song and Xin-Guang Zhu

Fig. 2 The basic procedure for using a CAPTS chamber for automatic Ac measurements. (a) Assemble the
chamber sides with connector and screws, (b) apply the chamber cover, (c) cover the plants by moving the
chamber to the field or putting plants in the chamber, and (d) chamber is periodically closed and opened by the
controller for measurements

of different plants do not shade each other. Conversely, border-

protecting plants are required when leaves of different plants
do shade each other (see Note 2; Fig. 1c).
4. Soil: Soil with no plants is used for the measurement of soil
respiration, which is needed during the calculation of net can-
opy photosynthesis or respiration (see Note 3). For rice grown
in a paddy field, the canopy chamber covers the paddy soil, with
the bottom of the chamber sealed with standing water above
the paddy. Remove weeds from the area covered by the cham-
ber to avoid contamination of measurements (see Note 4).
5. CAPTS chamber: A chamber with a size of 100 cm
100 cm is
used for typical Ac measurements (Fig. 2). The height of the
chamber should be higher than the plant (see Note 5). The
chamber is built with a metal frame and polycarbonate film,
which can be built by users or commercially bought (see
Note 6).
6. CAPTS controller: CAPTS controller is a machine with single-
chip microcomputer, infrared gas analyzer (IRGA) unit, a
pump, and a gas switching unit. To measure multiple canopies
simultaneously, the CAPTS controller can control multiple
chambers (see Note 7, Fig. 3). A pump in the controller
pumps gas from a chamber to the infrared gas analyzer in the
controller to measure [CO2]. The CAPTS controller automat-
ically records the [CO2], time, PPFD, air temperature, air
pressure, and relative humidity. The data are saved on a secure
digital (SD) memory card. All the sensors for measuring envi-
ronmental parameters, including PPFD, air temperature, air
humidity, and air pressure, are installed inside the chamber
and connected to the controller by cables.
 Measuring Canopy Gas Exchange with CAPTS 73

Fig. 3 The CAPTS controller can control multiple chambers. Fans connected with
tubes are installed inside the chamber to mix the gas in the chamber

Table 1
Parts list for CAPTS

Item Quantity Function

Chamber(s) with automatic
top covers
Controller
Temperature and relative
humidity sensors
Photosynthetic photon flux
density (PPFD) sensor
Air pressure sensor
Fans with tubes
Sampling tubes
Power supply
CO2 scrub tube
1–10 Covering a canopy or soil
1 Control chamber opening/closure, pump gas from chamber
to controller for analysis and data recording
1 Measure air temperature and relative humidity in a chamber
1 Measure PPFD in a chamber
1 Measure air pressure in a chamber
4–40 Mix air in a chamber
2–20 Sample gas from chamber and pump back to chamber after
analysis
1 Provide 24VDC for controller, automatic top cover, and fans
1 Zero the IRGA in a controller

7. Fans for mixing the gas in the chamber: Four fans connected
with 70-cm-long (depending on chamber height) tubes are
used to mix the gas in the chamber. The size of the fan is
80 mm
80 mm
38 mm, the rated voltage is 24VDC, the
rated current is 1.12A, and the airflow is 133 CFM. The
protection degree of the fan for dust and water is IP55.
8. Parts list for the CAPTS: A CAPTS has a minimum of one
chamber and a maximum of ten chambers. The parts list for
each CAPTS is shown in Table 1.
74 Qingfeng Song and Xin-Guang Zhu

3 Methods

3.1 Chamber The CAPTS chamber is assembled before its setup in the field. One
Assembly chamber consists of four sides and one automatic top cover as
shown in Fig. 2. The four sides are identical (see Note 6).
1. Assemble any two chamber side plates with a special mechanical
connector and fix them with two screws (Fig. 2a).
2. Take another plate and assemble it with the two connected side
plates and fix in place with two screws.
3. Take the fourth plate and connect to the three assembled side
plates to complete the chamber walls.
4. Install fans in the four corners of chamber and plug the fans’
cables into the controller (Fig. 3).
5. Check the gaskets at the bottom surface of the top cover (see
Note 8).
6. Put the top cover onto the chamber (Fig. 3).
7. Connect the cable of linear electronic actuator with the
controller.

3.2 Chamber A leakage test should be done before using the CAPTS to ensure
Leakage Test that the chamber is well assembled and all the gaskets are well
pasted (see Note 9).
1. Connect to the chamber with gas sampling tubes.
2. Blow some gas by mouth into the chamber.
3. Close the top cover of the chamber.
4. The controller automatically measures the [CO2] inside the
chamber (Cinside).
5. Calculate the change of C (ΔC) during time interval T.
6. Measure the C outside of the chamber (Coutside) and Cinside.
7. Calculate the leakage L as

ΔC
L¼ ð1Þ
T ðC inside  C outside Þ

3.3 Measurement 1. Move the assembled, leak-tested chamber to the field and cover
a canopy (see Note 10).
2. Seal the bottom of the chamber with soil. If the chamber is set
up in an upland field, then the soil should be 5 cm high from
the bottom of the chamber. If the chamber is set up in a paddy
field, then seal the bottom of the chamber with water. The
water level should be kept higher than the bottom of the
chamber (see Note 11).
 Measuring Canopy Gas Exchange with CAPTS 75

3. Start the fans to mix the air in the chamber.

4. Start the controller and set the parameters for the automatic
measurement program. Set the chamber close time to 45 s; set
the chamber open time to 3 min; set the chamber log wait time
to 10 s; set the chamber number to be the number of chambers
used during measurements (see Note 12).
5. Start the automatic measurement program and wait until the
experiment finishes (see Note 13).
6. Export data from controller as text files to a PC by a SD card
(see Note 14).

3.4 Data Analysis CAPTS Suite is a software for analyzing the raw data exported from
Using the CAPTS Suite the CAPTS after measurement. CAPTS Suite software can be used
Software on a PC machine with a Windows operating system. The following
describes the procedure used to run the CAPTS Suite software to
calculate parameters related to canopy photosynthesis.
1. Import the measured data to the CAPTS suite software by
selecting the data folder from the CAPTS suite graphic user
interface (GUI) (Fig. 4) (see Note 15).
2. In the CAPTS Suite software, the parameters need to be
entered in the GUI before the software is used for data analysis.
These parameters include controller version, controller ID,
chamber volume, chamber ground area, air pressure, and air
temperature (see Note 16).
3. The CAPTS Suite software can be used to filter the low-quality
data by setting the log duration cutoff, pre-deletion time,
R-square cutoff (for CO2 linear fit), and standard error cutoff
(for PPFD) in the GUI (see Note 17).
4. Select the output file format (either .txt or .csv). The data in
.txt files are separated by “tab” and data in .csv files are sepa-
rated by “,”.
5. Click the run button for analysis.
6. Directly measured parameters can now be analyzed. These
include [CO2], [H2O], air temperature T, air pressure P, rela-
tive humidity (RH), and PPFD. These parameters are recorded
every second after the log wait time until the chamber opens.
7. Derived parameters can now be calculated with the CAPTS
Suite software. These include canopy photosynthetic CO2
uptake rate (Ac), canopy respiration rate (Rc), and canopy
transpiration rate (Ec). They are calculated with Eqs. 2, 3,
and 4, where the dC/dT (unit: μmol mol1 s1) is the slope
of [CO2] change with time, V (unit: m3) is the volume of gas in
the chamber, P (unit: kPa) is the air pressure in the chamber,
S (unit: m2) is the ground area that the canopy occupied, R is
the universal gas constant (8.3
103 m3 kPa mol1 K1), and
76 Qingfeng Song and Xin-Guang Zhu

Fig. 4 GUI of the CAPTS Suite software. CAPTS data directory: gives the path to the CAPTS raw data folder.
CAPTS version: select the version of the controller used. Controller ID: enter the ID of the controller used (for
home-built system, see Note 18). Chamber volume: enter the volume of the chamber. Chamber ground area:
enter the ground area covered by a chamber. Air pressure (default): the default air pressure. Air temperature
(default): the default air temperature. Use default checkbox: determines whether to apply (check) or not apply
(uncheck) the default values. Log duration cutoff: enter a filter threshold that indicates the time for recording or
logging during a measurement, which is usually set as the chamber closure time minus one (unit is second).
Pre-deletion time: the time deleted at the beginning of one measurement, when the CO2 concentration still
fluctuates. R-square cutoff (CO2 linear fit): sets a quality threshold of the linear regression for [CO2] with time.
Standard error cutoff (PPFD): monitors fluctuation of PPFD during measurements. This cutoff filters out those
measurements obtained under highly fluctuating light. Output file format: data in .txt files are separated by
“tab” and data in .csv files are separated by “,”. Processed data: the path and filename of the filtered raw data
of all chambers used for calculation of derived parameters. Result: the path and filename of the result file of
derived parameters as well as the averaged values of directly measured parameters

T (unit: K) is the air temperature in the chamber. dW/dT is the

slope of [H2O] change with time:

dC V  P
Ac ¼   ð2Þ
dT S  R  T
dC V  P
Rc ¼  ð3Þ
dT S  R  T
dW V P
Ec ¼  ð4Þ
dT S  R  T

8. A results file containing Ac, Rc, and Ec with the averaged values
of PPFD, air temperature, air RH, air pressure, [CO2], [H2O]
and time, and a processed data file which contains the filtered
data used for the calculations will be exported.
 Measuring Canopy Gas Exchange with CAPTS 77

4 Notes

1. The row and column distance can be changed to form a user-

defined planting density.
2. The border-protecting plants are important when there are
strong plant-plant interactions, such as leaf shading between
plants. The border-protecting plants help ensure that the light
environments inside CAPTS are comparable to those in the
center of a field instead of at the border of a field. In a crop
canopy, the microclimate, especially the light environment, of
the border-protecting plants, is dramatically different from that
of plants not at the border. So, in general, border-protecting
plants are needed when canopy photosynthesis is measured.
Only when measuring a canopy where distance between plants
is sufficiently large, i.e., there is no shading between plants,
border-protecting plants surrounding the chamber are not
needed.
3. When net Ac is calculated, the gas exchange rate for both the
plant canopy and soil needs to be measured. The soil measured
should be the same as that used for growing the plant. In a
paddy field, the water level for the soil should be the same as
that for the plant canopy because microbes in the water also
influence the measurement of respiration.
4. It is important to keep no plants in a canopy chamber during
the measurement of soil respiration, because plant photosyn-
thetic CO2 uptake flux per leaf area is usually much higher than
soil respiratory CO2 release rate per ground area under light.
5. The CAPTS chamber should be taller than the plants
measured, and the leaf and spike of a plant should not touch
the top cover of the chamber. Otherwise changes in canopy
structure will affect light distribution inside the canopy and
alter the measurements.
6. User-designed chambers have been used in earlier studies
[13, 14, 25]. When building in-house, care should be taken
to prevent chamber leakage and use fans with high flow rate in
the chamber to mix air. Commercial infrared gas analyzer
(IRGA) units and data loggers are available for the user to
incorporate into the system. Commercial options for the can-
opy chamber and the whole automatic measurement system
described in this chapter are also available. CAPTS chambers
can be built by end users, or commercial products are available.
The frames should be made with aluminum or stainless steel.
The transparent film used for the sides and cover should be
made from a polycarbonate that allows equally good transmit-
tance of different light wavelengths. Fans should be used for
78 Qingfeng Song and Xin-Guang Zhu

mixing the gas in the chamber. The transparent film needs to be

replaced when its transmittance is less than 70% or broken.
7. A controller includes several units: (a) the single-chip micro-
computer, which runs a program that sets time for each cham-
ber’s closing and opening. The single-chip microcomputer
generates signals to drive linear electric actuators, which move
the top cover of a chamber to close or open. The main board
also collects signals from sensors and IRGA and stores them on
a SD card, (b) an infrared gas analyzer (IRGA) unit, which
measures [CO2] and [H2O], and (c) a pump and gas switching
unit, which pumps gas from different chambers to the IRGA
for measurement.
8. To prevent leakage at the junction between a top cover and a
chamber, gaskets are used. The gaskets need to be checked
every time before using. Replace gaskets when they are broken.
9. The leakage test is important because the [CO2] in the cham-
ber is lower than ambient [CO2] during the measurement of Ac
or higher than ambient [CO2] during the measurement of
respiration. Based on the principle of the closed-chamber sys-
tem, the amount of CO2 (unit: μmol) uptake needs to be
calculated based on the [CO2] (unit: ppm), air pressure (unit:
kPa), volume of gas in chamber (unit: m3), and temperature
(unit: K). If a chamber has leakage or exchange of gas between
the inside and outside of a chamber, then the [CO2] will be
influenced by parameters unrelated to photosynthesis or respi-
ration, and the measurement will be inaccurate.
10. Put the chamber over the plant canopy slowly and carefully to
avoid destroying the plants during chamber setup. It is better
to have four people work together during setup. Two people
move the chamber slowly and two people move the plants’
leaves to prevent destruction by the chamber.
11. The bottom of the chamber is open and 5 cm soil or above-
bottom-level water is enough to seal the chamber bottom
because there is no air pressure difference between inside and
outside.
12. Chamber close time is the time from the start of the chamber
closing to the start of the chamber opening. Chamber open
time is the time from the start of the chamber opening to the
start of the chamber closing. Chamber log wait time is the time
from the start of the chamber closing to the start of data
recording. Chamber close time is determined by the gas
exchange rate. Shorter chamber close times (30 s–1 min) will
increase the accuracy of canopy photosynthesis and transpira-
tion measurement as the impacts of temperature and RH
change will be smaller. Longer chamber close times
(1–3 min) will increase precision of canopy photosynthesis
 Measuring Canopy Gas Exchange with CAPTS 79

and transpiration measurement because more data points will

increase the R-square of curve fitting of [CO2] or [H2O]
against time. A typical chamber close time is 45 s and should
be longer when respiration is measured. The default chamber
open time is 3 min. When multiple chambers are used for
measurement, the chamber open time is the max of the cham-
ber open time and the (n-1) times chamber close time (n is the
total chamber number).
13. One experiment usually lasts for 1–2 days, during which the
diurnal canopy gas exchange is measured. When ten chambers
are used, we can compare Ac measurements for three groups
(with three replicates in each group) of canopies with nine
chambers and have one chamber to measure soil respiration.
The nine canopies should be randomly placed in a 3
3 plot
matrix. The controller is set near the chamber at the center of
the plot matrix.
14. The file name of each data file is an 8-digit number where the
first 2 digits indicate the chamber ID (from 01, 02 to 10) and
the 3rd–8th digits indicate the date of measurement (with
YYMMDD format). For example, a filename of “03160702.txt”
represents the data in the file measured by chamber No. 3 on
2016-July-02.
15. The CAPTS software is available upon request to the author.
The data folder contains only the whole raw data files exported
from the CAPTS controller.
16. Controller version should be consistent with the controller
used. Controller ID can be defined by users to label data
measured with different controllers. Chamber volume is the
volume of the gas in the chamber, which equals to the volume
of the chamber when the volume that plant stem and leaf
occupied is relatively small and ignored. When plants in pots
are measured, the volume of the pot should be excluded from
the volume of the chamber. Chamber ground area is the area of
the ground covered by the chamber. Air pressure and air tem-
perature are normally measured. However, if these data are not
measured or there are errors during the measurement, air
temperature and pressure can be input into the GUI and used
during the calculation of canopy photosynthesis and
transpiration.
17. Log duration cutoff is a threshold to determine whether the
logged data during a close-open cycle can be used. Any mea-
surements during a close-open cycle shorter than the log dura-
tion cutoff will not be used. The log duration cutoff is typically
set as the chamber closure time minus one (unit is second).
Pre-deletion time is the time threshold before which the
recorded data are not used during one measurement cycle
80 Qingfeng Song and Xin-Guang Zhu

due to the high level of fluctuations in [CO2]. This value

depends on the curve of [CO2] change with time during
measurements and is usually set as 10–15 s. R-square cutoff
of [CO2] is another measure used to determine whether a set
of recordings during a measurement cycle is of good quality. If
the R2 for the linear regression of [CO2] with time is less than
this threshold, this set of recordings will not be used. Typically,
this cutoff is set as 0.5 by default. Standard error cutoff (for
PPFD) is used to monitor fluctuation of light during measure-
ments. This cutoff filters out measurements made under highly
fluctuating PPFD. A typical value for this cutoff is 15%, which
means the coefficient of variation of PPFD should be less than
15% to keep a recording.
18. If a user needs to use the CAPTS Suite of software to analyze
data from a home-built system following designs described in
this chapter, the data format of raw data from the home-built
system need to match (or convert to) the data format of
CAPTS raw data. An example CAPTS raw data file is available
upon request to the authors.

Acknowledgments

The authors acknowledge funding from the National Natural Sci-

ence Foundation of China young scientist grant (grant #
31501240) to QS and open funding from State Key Laboratory
of Hybrid Rice (grant #2016KF06) to QS, the Chinese Academy of
Sciences strategic leading project (XDA08020301) to XZ, and the
CAS-CSIRO collaboration grant (GJHZ1501) to XZ.

References
1. Song Q, Zhang G, Zhu X-G (2013) Optimal
crop canopy architecture to maximise canopy
photosynthetic CO2 uptake under elevated
CO2- a theoretical study using a mechanistic
model of canopy photosynthesis. Funct Plant
Biol 40:109–124. https://doi.org/10.1071/
FP12056
2. Song Q, Chu C, MAJ P, Zhu X-G (2016)
Genetics-based dynamic systems model of can-
opy photosynthesis: the key to improve light
and resource use efficiencies for crops. Food
Energy Secur 5:18–25. https://doi.org/10.
1002/fes3.74
3. Held AA, Steduto P, Orgaz F et al (1990)
Bowen ratio/energy balance technique for esti-
mating crop net CO2 assimilation, and compar-
ison with a canopy chamber. Theor Appl
Climatol 42:203–213. https://doi.org/
10.1007/BF00865980
4. McMillen RT (1988) An eddy correlation tech-
nique with extended applicability to
non-simple terrain. Boundary-Layer Meteorol
43:231–245
5. Bugbee B (1992) Steady-state canopy gas
exchange: system design and operation.
Hortscience 27:770–776
6. Long SP, Farage PK, Garcia RL (1996) Mea-
surement of leaf and canopy photosynthetic
CO2 exchange in the field. J Exp Bot
47:1629–1642
7. Dragoni D, Lakso AN, Piccioni RM (2005)
Transpiration of apple trees in a humid climate
using heat pulse sap flow gauges calibrated with
whole-canopy gas exchange chambers. Agric
For Meteorol 130:85–94. https://doi.org/
10.1016/j.agrformet.2005.02.003
 Measuring Canopy Gas Exchange with CAPTS
8. Graydon JA, St Louis VL, Lindberg SE et al
(2006) Investigation of mercury exchange
between forest canopy vegetation and the
atmosphere using a new dynamic chamber.
Environ Sci Technol 40:4680–4688
9. Burkart S, Manderscheid R, Weigel H-J (2007)
Design and performance of a portable gas
exchange chamber system for CO2- and H2O-
flux measurements in crop canopies. Environ
Exp Bot 61:25–34. https://doi.org/
10.1016/j.envexpbot.2007.02.007
10. Muller J, Eschenroder A, Diepenbrock W
(2009) Through-flow chamber CO2/H2O
canopy gas exchange system—Construction,
microclimate, errors, and measurements in a
barley (Hordeum vulgare L.) field. Agric For
Meteorol 149:214–229. https://doi.org/
10.1016/j.agrformet.2008.08.007
11. Reicosky DC (1990) Canopy gas exchange in
the field: Closed chambers. Remote Sens Rev
5:163–177
12. Wagner SW, Reicosky DC (1992) Closed-
chamber effects on leaf temperature, canopy
photosynthesis, and evapotranspiration.
Agron J 84:731–738
13. Steduto P, Cetinkoku O, Albrizio R, Kanber R
(2002) Automated closed-system canopy-
chamber for continuous field-crop monitoring
of CO2 and H2O fluxes. Agric For
111:171–186
14. Pérez-Priego O, Testi L, Orgaz F, Villalobos FJ
(2010) A large closed canopy chamber for
measuring CO2 and water vapour exchange of
whole trees. Environ Exp Bot 68:131–138.
https://doi.org/10.1016/j.envexpbot.2009.
10.009
15. Dugas WA, Fritschen LJ, Gay LW et al (1991)
Bowen ratio, eddy correlation, and portable
chamber measurements of sensible and latent
heat flux over irrigated spring wheat. Agric For
Meteorol 56:1–20
16. Dugas WA, Reicosky DC, Kiniry JR (1997)
Chamber and micrometeorological measure-
ments of CO2 and H2O fluxes for three C4
grasses. Agric For Meteorol 83:113–133.
81
https://doi.org/10.1016/S0168-1923(96)
02346-5
17. Dugas WA (1993) Micrometeorological and
chamber measurements of CO2 flux from bare
soil. Agric For Meteorol 67:115–128. https://
doi.org/10.1016/0168-1923(93)90053-K
18. Angell RF, Svejcar T, Bates J et al (2001)
Bowen ratio and closed chamber carbon diox-
ide flux measurements over sagebrush steppe
vegetation. Agric For Meteorol 108:153–161.
https://doi.org/10.1016/S0168-1923(01)
00227-1
19. Johnson DA, Saliendra NZ, Walker JW, Hen-
drickson JR (2003) Bowen ratio versus canopy
chamber CO2 fluxes on sagebrush rangeland. J
Range Manage 56:517–523
20. Peng S, Krieg DR (1991) Single leaf and can-
opy photosynthesis response to plant age in
cotton. Agron J 83:704–708
21. Barthel M, Sturm P, Gentsch L, Knohl A
(2010) Technical note: a combined soil/can-
opy chamber system for tracing δ 13C in soil
respiration after a 13CO2 canopy pulse label-
ling. Biogeosci Discuss 7:1603–1631
22. Hileman DR, Huluka G, Kenjige PK et al
(1994) Canopy photosynthesis and transpira-
tion of field-grown cotton exposed to free-air
CO2: enrichment (FACE) and differential irri-
gation. Agric For Meteorol 70:189–207
23. Wagner SW, Reicosky DC, Alessi RS (1997)
Regression models for calculating gas fluxes
measured with a closed chamber. Agron J
89:279–284
24. Song Q, Xiao H, Xiao X, Zhu X-G (2016) A
new canopy photosynthesis and transpiration
measurement system (CAPTS) for canopy gas
exchange research. Agric For Meteorol
217:101–107. https://doi.org/10.1016/j.
agrformet.2015.11.020
25. Baker JT, Gitz DC, Lascano RJ (2014) Field
evaluation of open system chambers for mea-
suring whole canopy gas exchanges. Agron J
106:537–544. https://doi.org/10.2134/
agronj2013.0449
 Chapter 5

Light-Response Curves in Land Plants

Robert A. Coe and HsiangChun Lin

Abstract
Light-response curves reveal the photosynthetic properties of plants. Depending upon the methodology
selected they can be used to characterize CO2 assimilation, photochemistry, photoacclimation, photoinhi-
bition, and kinetics of photoprotective mechanisms in response to changing light conditions. They are
widely used to describe the ontogeny and range in physiological plasticity of plants. Here we describe
methods for acquiring light-response curves using CO2 gas exchange and chlorophyll a fluorescence
measurements that are applicable to a wide range of land plants.

Key words Light-response curves, Gas exchange, Chlorophyll a fluorescence

1 Introduction

Growth of autotrophic land plants is intrinsically linked to light, the

driving force for photosynthesis. There is a well-defined relation-
ship between the amount of irradiance absorbed by a leaf, the
electron transport rate (ETR), and the rate of RuBisCO (ribulose-
1,5-bisphosphate carboxylase/oxygenase) carboxylation or oxyge-
nation reactions [1]. As the quantity (quantum flux density) and
quality (spectral composition) of light are dynamic properties, it is
desirable to gain insights into the characteristics and adaptation of
photosynthesis to the light environment, and/or probe the physi-
ological properties of different species. This can be achieved by
measuring light-response curves.
Photosynthetic light-response curves (P-I) describe the net
CO2 assimilation (P) as a function of irradiance (I). While various
mathematical models have been used to describe the relationship,
physiologists frequently use a non-rectangular hyperbola equation
[2–5]
θP 2
ðφI þ P max Þ P
φI P max ¼ 0

where P is the photosynthetic rate, I is the absorbed irradiance, φ is

the maximum quantum yield for CO2 uptake, θ is the convexity of

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_5, © Springer Science+Business Media, LLC, part of Springer Nature 2018

83
84 Robert A. Coe and HsiangChun Lin

Rate of CO2 as similation, P

P max

φ
I comp

Irradiance, I

Fig. 1 Idealized light-response curve of photosynthesis

the curve, and Pmax is the light-saturated rate of photosynthesis.

The initial portion of the curve is characterized by the light com-
pensation point (Icomp), the value of I at which CO2 assimilated by
photosynthesis is equal to the CO2 produced by respiration
(Fig. 1). Beyond this point there is a supposed linear response of
P to I up to approximately 200 μmol m
2 s
1, which is used to
calculate φ. This is followed by a region of nonlinear die-off
described by θ before reaching a semi-plateau (Pmax) where further
increases in I do not lead to an increase in P. Occasionally, after
reaching Pmax there is a decline in P resulting from photoinhibition.
Interpreting the mechanistic basis of the P-I response is not
straightforward. It is difficult to pinpoint experimentally the transi-
tion between RuBP-regenerated and RuBisCO-limited photosyn-
thesis, two major rate limitations of photosynthesis which are
important for understanding the component processes controlling
the response of photosynthesis to light. The model does not
account for all the physiological or biochemical processes of the
leaf [6], nor do all parameters derived from it, such as θ, have an
obvious physiological meaning [7–9]. Despite this, P-I response
curves provide a means with which to characterize CO2 assimilation
in response to light.
More recently, measurements of chlorophyll a fluorescence
have been established as a useful and informative indicator charac-
terizing the light reactions of photosynthesis. Light energy
absorbed by chlorophylls associated with photosystem II (PSII)
can be used to drive photochemistry in which an electron (e
) is
transferred from the reaction center chlorophyll of PSII (P680) to
the primary quinine acceptor of PSII (QA). Alternatively, absorbed
light energy can be lost from PSII as chlorophyll fluorescence or as
heat. These three processes are in direct competition with each
other; if the rate of one process increases, then the rates of the
other two will decrease. Thus, the yield of chlorophyll fluorescence
 Light Response Curves 85

emission gives us valuable information about the quantum effi-

ciency of photochemistry and heat dissipation [10].
Combining gas exchange with chlorophyll a fluorescence
means it is possible to simultaneously characterize the dark and
light reactions of photosynthesis and gain greater insights into the
photosynthetic processes. In this chapter we present methods for
measuring light-response curves using gas exchange and chloro-
phyll a fluorescence. These are the most commonly employed
techniques and offer the greatest flexibility in order to investigate
photosynthetic performance in response to light.

2 Materials

A variety of commercial infrared gas exchange analyzers (IRGAs)

and chlorophyll fluorometers capable of pulse-amplitude-modu-
lated (PAM) measurements [11–13] are available. Discussion, com-
parison, and detailed operation instructions are beyond the scope of
this chapter and so the reader is referred to the user manuals and
wider literature.

3 Methods

3.1 Light-Response There are several strategies for acquiring light-response curves each
Curve Strategy with its own biological interpretation. The two most common are
the following (see Note 1):
1. Steady-state light-response curves (LRCs): These are designed
to characterize the potential response of steady-state photosyn-
thesis under a range of light conditions (photoacclimation)
(see Note 2).
2. Rapid light-response curves (RLCs): These are designed to
characterize the dynamic response of the photosynthetic appa-
ratus in a rapidly changing light environment. They are used to
detect short-term adjustments in the functioning of the photo-
synthetic apparatus including the activation and deactivation of
carbon metabolism, operation of photoprotective mechanisms,
and photoinhibition (see Note 3).

3.2 Photosynthetic 1. Set up the gas exchange system according to the manufac-
Light-Response turer’s instructions.
Curves 2. A CO2 concentration of 400 μmol mol
1 and 21% O2 is
representative of ambient conditions (see Note 4).
3.2.1 Operational
3. A constant leaf temperature should be used. For most experi-
Considerations
ments a temperature of 25  C is recommended (see Note 5).
86 Robert A. Coe and HsiangChun Lin

4. Where possible, a relative humidity (RH) similar to the growth

environment should be used in order to avoid stomatal effects
(see Note 6).
5. A constant airflow should be used; 500 μmol m
2 s
1 is sug-
gested (see Note 7).
6. A standard protocol for C3 plants is a photosynthetic photon
flux density (PPFD) of 1500, 1000, 500, 250, 120, 60, 40,
20, 10, and 0 μmol m
2 s
1 (see Notes 8 and 9). A maximum
light intensity of 2000 μmol m
2 s
1 is suggested for C4 plants
or 500 μmol m
2 s
1 for C3 or C4 plants grown in the shade
(see Note 10).
7. If plants were in the dark prior to starting the curves, then they
should be equilibrated for 15–20 min at a light intensity repre-
sentative of the natural growth environment (see Note 11).
8. For steady-state light-response curves (LRCs) leaves should be
equilibrated at each light intensity for 3–5 min (see Note 12).
For rapid light-response curves (RLCs) the leaves should be
equilibrated at high light for a minimum of 5 min (see Note 13)
and then the light intensity should be decreased in intervals of
200 μmol m
2 s
1 every 1–3 min (see Note 14).
9. Typically, during gas exchange measurements, a CO2 zero
point is obtained every hour or after changing the CO2 con-
centration. When acquiring light-response curves this is only
important at the start of each curve and does not need to be
repeated.

3.2.2 Methods 1. Set starting light intensity, CO2 concentration, and flow.
2. Insert leaf (see Note 15).
3. Set leaf temperature and humidity.
4. Program and start the desired light-response curve protocol.
5. When the protocol has finished remove the leaf. If measuring
other leaves, go back to step 1.
6. Download data, graph, and analyze data (see Notes 16 and 17).
7. Icomp is calculated using least square regression analysis of the
intercept of the initial part of the curve (PPFD
<200 μmol m
2 s
1).
8. φ is calculated using least square regression analysis to the initial
slope (PPFD <200 μmol m
2 s
1).
9. The convexity of the response (θ) curve can be calculated using
curve fitting routines (see Note 16), although frequently curve
shapes are simply compared.
10. Pmax can be extracted from the dataset if light saturation of
photosynthesis is reached.
11. Rd, respiration rate in the dark, can be estimated from rates of
P at a PPFD of 0 μmol m
2 s
1 (see Note 18).
 Light Response Curves 87

3.3 Chlorophyll 1. Set up the chlorophyll fluorescence system according to the

a Fluorescence Light- manufacturer’s instructions.
Response Curves 2. The distance between the leaf and sensor should be optimized
to ensure that a strong fluorescence signal is obtained (see
3.3.1 Operational Note 19).
Considerations
3. A constant leaf temperature should be used. For most experi-
ments a temperature between 25 and 30  C is recommended.
4. Where possible, measurements should be made at a RH similar
to the growth environment in order to avoid stomatal effects.
5. The user needs to decide whether a period of dark adaption is
necessary before making the first measurement (see Note 20) as
this may affect the experimental setup.
6. A standard protocol for C3 plants is a PPFD of 0, 200,
400, 600, 800, 1000, 1200, 1400, and 1500 μmol m
2 s
1
(see Note 21), acclimating the leaf at each actinic light intensity
for between ~20 s (RLCs) (see Note 22) and 20 min (LRCs).
Consideration of the type of light-curve strategy is discussed
above (see Subheading 3.1). A maximum light intensity of
2000 μmol m
2 s
1 is recommended for C4 plants or
500 μmol m
2 s
1 for C3 or C4 plants grown in the shade (see
Note 10).

3.3.2 Methods 1. Dark-adapt the leaf for a minimum of 5–30 min (see Note 23).
This can be achieved using the manufacturer’s leaf clips or the
user’s own protocol.
2. Switch off the measuring beam and apply the fluorescence
detector to ensure that the reading is zero.
3. Switch on the measuring beam. Exposing the leaf to a weak
modulated measuring beam (PPFD of 0.1 μmol m
2 s
1)
results in the minimal level of fluorescence (Fo). Applying a
pulse of weak far-red light prior to the measurements of Fo is
recommended (see Note 24).
4. Expose leaves to a saturating light pulse (>4000 μmol m
2 s
1)
for a duration of 0.6–0.8 s (see Note 25); QA will be maximally
reduced and the maximal fluorescence level (Fm) will be
observed. The difference between Fm and Fo is defined as the
variable fluorescence (Fv). The ratio of Fv/Fm is used to
estimate the maximum quantum yield of QA reduction (see
Note 26).
5. Switch on the actinic light.
6. When steady state is reached, measure fluorescence
emission (F0 ).
7. Apply a saturating pulse and attain maximum fluorescence in
the light (Fm0 ).
88 Robert A. Coe and HsiangChun Lin

8. If dark-adapted measurement are not required, then you can

directly apply the light-response curve protocol after inserting
the leaf. Start from step 2 above (see Note 27).
9. Remove leaf.
10. Download data, graph, and analyze data (see Table 1 for a list of
formulae).
11. The maximum quantum yield of PSII photochemistry in the
dark (Fv/Fm) is calculated as (Fm
Fo)/Fm.

Table 1
Chlorophyll fluorescence parameters and equations frequently used in studies of photosystem II
photochemistry

Parameter Definition Formula Description

Fv/Fm Maximum quantum (Fm
Fo)/Fm Maximum efficiency at which
efficiency of PSII light absorbed by PSII is used
photochemistry for reduction of QA
ΔF/Fm0 PSII operating efficiency (Fm0
F0 )/Fm0 Quantum efficiency of PSII
electron transport in the light.
Also known as φII or Fq0 /Fm0
Fv0 /Fm0 PSII maximum efficiency (Fm0
Fo0 )/Fm0 Maximum efficiency of PSII
photochemistry in the light,
which is the PSII operating
efficiency if all the PSII centers
were “open” (QA oxidized)
ETR Photosynthetic electron PPFD  0.84  0.5  (ΔF/ Electron transport rate through
transport rate of PSII Fm 0 ) photosystem II
NPQ Nonphotochemical (Fm/Fm0 )
1 Estimates the apparent rate
quenching constant for heat loss from PSII
qp PSII efficiency factor (Fm0
F0 )/(Fm0
Fo) Relates PSII maximum efficiency
to operating efficiency.
Nonlinearly related to the
proportion of PSII centers that
are “open” (QA oxidized). See
qL
qL Fraction of PSII centers (ΔF/Fv0 )(Fo0 /F0 ) ¼ [(Fm0
Estimates the fraction of “open”
that are “open” F0 )/(Fm0
Fo0 )]  (Fo0 / PSII centers (with QA oxidized)
F0 ) on the basis of a lake model
[11] for the PSII
photosynthetic apparatus
qN Coefficient of (Fm
Fm0 )/(Fm
Fo) Estimated from the proportion of
nonphotochemical quenching of the variable
quenching fluorescence Fv, (Fm
Fo). On
a scale of 0–1
 Light Response Curves 89

12. The effective quantum yield of PSII, ΔF/Fm0 , is calculated as

(Fm0
F0 )/Fm0 , where F0 is fluorescence yield of the light-
adapted sample and Fm0 is the maximum light-adapted fluores-
cence yield (see Note 28).
13. The photosynthetic electron transport rate (ETR) of PSII is
calculated as PPFD  0.84  0.5  (ΔF/Fm0 ) (see Note 29).
14. Nonphotochemical quenching (NPQ) is calculated as
(Fm/Fm0 )
1 (see Note 30).
15. The PSII efficiency factor (qP) is calculated as (Fm0
F0 )/
(Fm0
Fo) (see Note 31).
16. The fraction of PSII centers that are “open” (qL) is obtained as
(ΔF/Fv0 )(Fo0 /F0 ) ¼ [(Fm0
F0 )/(Fm0
Fo0 )]  (Fo0 /F0 ) (see
Note 32).

3.4 Combined Gas Combining chlorophyll fluorescence with gas exchange enables
Exchange correlation of PSII photosynthetic efficiency directly with CO2
and Chlorophyll assimilation. This is achieved by installing a fluorescence measure-
a Fluorescence Light- ment head to the infrared gas analyzer (see Note 33).
Response Curves

4 Notes

1. When selecting a light curve strategy careful consideration

needs to be given to the response you want to measure; typi-
cally this will be either the short-term responses of photosyn-
thesis to rapid changes in light conditions or the response of
photosynthesis to steady-state light conditions.
2. LRCs measure steady-state conditions which are not typically
found in the natural environment. They are used to investigate
plasticity and inherent properties of photosynthesis. They can
be slow to acquire, and so, to save time, light curves can be
performed on multiple leaves equilibrated at different light
intensities. Caution is required in order to avoid introducing
variability into measurements by using leaves with different
ages or adapted to different conditions due to their position
and orientation within the canopy. Leaves of similar physiolog-
ical development should be selected from a single plant or
multiple representative plants.
3. RLCs are relatively fast to acquire and can be used to mimic
natural light conditions. They are typically used to investigate
short-term responses to rapid changes in the light environ-
ment. To overcome light dosage effects, non-sequential light-
response curves can be acquired, where curves are based upon
90 Robert A. Coe and HsiangChun Lin

independent measurements at each light level on different

leaves taken from a population in the same physiological state.
4. It is important that a constant CO2 and O2 concentration is
maintained to avoid confounding the effects of CO2 with light
and/or O2.
5. Photosynthesis is strongly affected by temperature and exhibits
a thermal optimum of ~25  C. Thermal acclimation changes
this thermal optimum and so the temperature of the growth
environment needs to be considered before measurements are
made. The mechanisms controlling temperature response in C3
and C4 plants are numerous and varied [14] and so care should
be taken when deciphering temperature responses.
6. Excessive humidity (>85%) will lead to condensation and erro-
neous measurements in some instruments. Humidity will
change alongside conductance and transpiration and so very
high (>85%) or low (<15%) values should be avoided in order
to accommodate these changes.
7. In some gas-exchange systems, flow rate will affect the speed of
detection: the higher the flow rate the faster the gas mixtures
reach the infrared gas analyzers. Flow rate is also important for
partially controlling humidity; at high flow rates, air inside the
cuvette is replenished quickly with dry air and humidity will
remain low, even if the transpiration rate is high. Flow rates
should be decreased when CO2 assimilation or stomatal con-
ductance is low as the difference between the CO2 concentra-
tion in the incoming and outgoing air will be very small and
high flow rates under these conditions may lead to measure-
ment error.
8. LED actinic light source with peak wavelengths of
~655  15 nm and a smaller proportion of light at 465  5 nm
are used in commercial systems. It is possible to perform a
light-response curve without a light source by using neutral
density filters.
9. LRCs can be performed by increasing light intensity, but the
responses of stomatal conductance (gs) and intercellular CO2
concentration (Ci) will be different to those under decreasing
light and A may be limited by gs as stomata can be relatively
slow to respond to changes in I.
10. Ideally, light-saturated photosynthesis (Pmax) or maximum
quantum efficiency (Fv/Fm or Fv0 /Fm0 ) should be determined
before any measurements are made. Exposing plants to light
intensities above those experienced in the normal growth envi-
ronment will lead to photoinhibition and photodamage; this
will adversely affect subsequent measurements of light
response. C4 species have a higher light use efficiency than C3
 Light Response Curves 91

species and so high irradiances can be used for the light curve.
Shade-adapted species tend to have lower dark respiration
rates, lower compensation points, and lower maximum photo-
synthetic rates than sun-adapted leaves.
11. Many photosynthetic enzymes are light regulated and require
an activation period before reaching steady state.
12. Sufficient time is required to allow stomatal conductance to
adjust and equilibrate to each light intensity. This can be deter-
mined by monitoring Ci (intercellular CO2 concentration) and
waiting for it to stabilize (typically this takes 3–5 min at each
light intensity). On some machines, this wait time is defined by
the user at the start of the program, and so care must be taken
to ensure that it is long enough for Ci to stabilize. Otherwise
the curve must be repeated. If the resulting P-I curve is not
smooth, then this is an indication that the wait time was not
long enough.
13. High light can lead to the onset of photoinhibition and
decreases in A and φ; recovery can take minutes to hours
depending upon the mechanism involved [15].
14. Stomata will not be able to adjust and equilibrate to each new
light intensity because the acclimation time at each intensity is
too short. Pre-adaptation to a high light intensity will ensure
that stomata are fully open and do not limit CO2 assimilation.
This means that as the light intensity is decreased, gs and Ci will
be higher than under increasing light. While values will be
higher, response between leaves can still be compared.
15. CO2 assimilation is typically expressed on a leaf area basis
(μmol m
2 s
1) so a known leaf area is required.
16. Curve fitting routines for light-response curves are not as
straightforward as those for photosynthetic CO2 response
curves. Microsoft Excel-based tools have been developed
[16, 17] but use of these requires the user to possess sound
statistical and mathematical knowledge as well as the biological
understanding to interpret the results. Using the published
tools on your own datasets and taking time to understand
how they work is the best way to acquire this knowledge.
17. Typical values: φ ¼ 0.425–0.5 mol e
(mol absorbed PAR
photon)
1; θ ¼ 0.7–0.96; ETRmax ¼ 50–200 μmol e
m
2 s
1;
Rd ¼ 0.1–3.0 μmol CO2 m
2 s
1; Icomp 0–50 μbar; and
Pmax ¼ <50 μmol CO2 m
2 s
1.
18. In the absence of photorespiration, Rd reflects rates of mito-
chondrial respiration; measurements are best made predawn to
ensure that there is no photorespiratory CO2 release. Measure-
ments should be made and averaged for a period of 30 min.
The values will be negative as they represent CO2 loss;
92 Robert A. Coe and HsiangChun Lin

however, these are typically expressed as positive values in final

data presentation.
19. If the leaf is positioned too far from the fluorometer this may
result in a weak fluorescence signal. The leaf should be held
between 90 and 45 horizontal to the light source in order to
provide homogeneous illumination and must not be changed
during measurements. If using an open system, then measure-
ments may be affected by strong fluctuations in ambient
irradiance.
20. The decision whether to dark-adapt a leaf or not depends on
the objectives of the experiment and the time and equipment
available. Dark adaptation is essential for studying the kinetics
of nonphotochemical quenching (NPQ) and for quenching
analysis. Dark adaptation is not necessary if you are only inter-
ested in chlorophyll kinetics in the light.
21. For chlorophyll a fluorescence light-response curves, light
intensity is always increased; it is not possible to decrease
light intensity because high irradiance leads to the onset of
photoinhibitory mechanisms which can take minutes to hours
to relax [15].
22. Always check whether there is photoinhibition resulting from
the repeated application of saturating pulses over short time
periods by monitoring Fm. When photoinhibition occurs, Fm
declines following sequential pulses and may not recover.
23. If the dark-adaptation period is not long enough, then QA may
not become maximally oxidized. To establish the length of
dark adaptation, non-stressed leaves acclimated in the light at
steady-state conditions should be dark-adapted and Fv/Fm
measured every 5–10 min until a value of ca. 0.83 is reached.
24. A pulse of weak far-red light should be applied prior to the
measurement of Fo to excite photosystem I (PSI) and remove
electrons from QA. If the chlorophyll fluorescence analyzer
does not feature an intrinsic far-red light source, then QA may
not maximally oxidize leading to overestimation of Fo; thus
care should be taken in interpreting the data.
25. Saturating pulses of ~2000 μmol m
2 s
1 are sufficient to
obtain dark-adapted Fm in most situations, but may be too
low to produce a true maximum fluorescence (Fm0 ) under
high actinic illumination. Most commercial instruments pro-
vide saturating light pulses with an intensity of between 4000
and 8000 μmol m
2 s
1. C4 plants will require a higher satur-
ating pulse than C3 plants and the intensity required will vary
according to the light environment in which the plants have
been grown. Users must monitor Fm to ensure that the inten-
sity of the pulse is strong enough to fully saturate PSII but does
not lead to quenching as standard parameters cannot be given.
 Light Response Curves 93

26. Fv/Fm values for non-stressed leaves are remarkably consistent

among all species at ca. 0.83 [10].
27. If dark adaptation is not required, then the functional status of
PSII will reflect photoacclimation and photoinhibition to the
light environment directly prior to the start of measurements.
28. ΔF/Fm0 provides an estimate of the PSII operation efficiency
under different light and different environmental conditions.
Note that ΔF/Fm0 will be lower than the dark-adapted Fv/Fm
value due to quenching of PSII chlorophyll fluorescence by
photochemistry and NPQ.
29. Photosynthetic electron transport rate of PSII (ETR) is calcu-
lated as ETR ¼ I  Aleaf  fractionPSII  (ΔF/Fm0 ). The ratio
of light absorbed by the leaf, Aleaf, is frequently assumed to be
0.84 (84% of incident PPFD is assumed to be absorbed by
leaves), although large deviations do occur [10], and so
where possible Aleaf should be measured with an integrated
sphere. The fractionPSII is frequently assumed to be 0.5 (equal
excitation of both PSII and PSI).
30. Nonphotochemical quenching (NPQ) estimates the apparent
rate constant for heat loss from PSII [10].
31. PSII efficiency factor (qp) estimates the redox state of QA.
32. Fraction of PSII centers that are “open,” (qL) estimates the
fraction of “open” PSII centers (with QA oxidized) on the basis
of a lake model [11] for the PSII photosynthetic apparatus. Fo0
value can be obtained from a measurement or calculation. The
measurement of Fo0 : following the saturating pulse, the actinic
light is switched off and after a few seconds Fo0 measured. A
pulse of weak far-red light should be applied prior to the
measurements of Fo0 . The calculation of Fo0 : Fo0 ¼ Fo/[(Fv/
Fm) + (Fo/Fm0 )] [18].
33. Follow the method for P-I response curves in Subheading
3.2.1; in step 6 the light intensity should be increased not
decreased (see Note 21).

References
1. von Caemmerer S (2000) Biochemical models
of leaf photosynthesis. CSIRO Publishing,
Collingwood, VIC
2. Farquhar GD, von Caemmerer S, Berry JA
(1980) A biochemical model of photosynthetic
CO2 assimilation in leaves of C3 species. Planta
149:78–90
3. Marshall B, Biscoe PV (1980) A model for C3
leaves describing the dependence of net photo-
synthesis on irradiance. J Exp Bot 31(120):29–39
4. Prioul JL, Chartier P (1977) Partitioning of
transfer and carboxylation components of
intracellular resistance of photosynthetic CO2
fixation: a critical analysis of the methods used.
Ann Bot 41:789–800
5. Leverenz JW, Jarvis PG (1979) Photosynthesis
in Sitka pruce VIII. The effects of light flux
density and direction on the rate of net photo-
synthesis and stomatal conductance of needles.
J Appl Ecol 16:919–932
94 Robert A. Coe and HsiangChun Lin
6. Ye ZP (2007) A new model for relationship
between irradiance and the rate of photosyn-
thesis in Oryza sativa. Photosynthetica 45
(4):637–640
7. Ögren E (1993) Convexity of the photosyn-
thetic light-response curve in relation to inten-
sity and direction of light during growth. Plant
Physiol 101:1013–1019
8. Ögren E, Evans JR (1993) Photosynthetic
light-response curves. I. The influence of CO2
partial pressure and leaf inversion. Planta
189:182–190
9. Evans JR, Jakobsen I, Ögren E (1993) Photo-
synthetic light-response curves. 2. Gradients of
light absorptions and photosynthetic capacity.
Planta 189:191–200
10. Baker NR (2008) Chlorophyll fluorescence: a
probe of photosynthesis in vivo. Annu Rev
Plant Biol 59:89–113
11. Bradbury M, Baker NR (1984) A quantitative
determination of photochemical and non- pho-
tochemical quenching during the slow phase of
the chlorophyll fluorescence induction curve of
bean leaves. Biochim Biophys Acta
765:275–281
12. Schreiber U, Schliwa U, Bilger W (1986) Con-
tinuous recording of photochemical and
nonphotochemical fluorescence quenching
with a new type of modulation fluorometer.
Photosynth Res 10:51–62
13. Schreiber U, Bilger W (1993) Progress in chlo-
rophyll fluorescence research: major develop-
ments during the last years in retrospect.
Progr Bot 54:151–173
14. Sage RF, Kubien DS (2007) The temperature
response of C3 and C4 photosynthesis. Plant
Cell Environ 30:1086–1106
15. Quick WP, Stitt M (1989) An examination of
factors contributing to non-photochemical
quenching of chlorophyll fluorescence in barley
leaves. Biochim Biophys Acta 977:287–296
16. Lobo FA, de Barros MP, Dalmagro HJ et al
(2013) Fitting net photosynthetic light-response
curves with Microsoft Excel - a critical look at the
models. Photosynthetica 51(3):445–456
17. Bellasio C, Beerling DJ, Griffiths H (2016) An
Excel tool for deriving key photosynthetic
parameters from combined gas exchange and
chlorophyll fluorescence: theory and practice.
Plant Cell Environ 39:1180–1197
18. Oxborough K, Baker NR (1997) Resolving
chlorophyll a fluorescence images of photosyn-
thetic efficiency into photochemical and
non-photochemical components – calculation
of qP and Fv0 /Fm0 without measuring Fo0 .
Photosynth Res 54:135–142
 Chapter 6

Chlorophyll Fluorescence on the Fast Timescale

Olubukola O. Ajigboye, Rumiana V. Ray, and Erik H. Murchie

Abstract
Chlorophyll fluorescence is a rapid and non-invasive tool used for probing the activity of photosynthesis
that can be used in vivo and in the field. It is highly relevant to the demands of high-throughput crop
phenotyping and can be automated or manually applied. Here we describe protocols and advice for making
fast timescale fluorescence measurements using handheld equipment in the laboratory or in the field. While
interpretation of some measured parameters requires caution, we demonstrate that this technique is
appropriate for some applications where convenience, rapidity, and sensitivity are required.

Key words Chlorophyll fluorescence, OJIP, Fluorescence transients, Fluorometer, Leaf, Phenotype

1 Introduction

Chlorophyll fluorescence is a highly powerful tool, widely used to

probe the reactions and responses of photosynthesis. The tech-
nique is sensitive and non-invasive, able to provide rapid and
detailed information about the structure and function of the pho-
tosystem II (PSII) and how photosynthesis responds to environ-
mental and intrinsic stimuli. It has also gained relevance for the new
field of plant and crop phenotyping which has required an expan-
sion of previous methodologies, e.g., see [1]. Chlorophyll fluores-
cence analysis can be achieved in two ways: (a) (pulse-amplitude)
modulated or PAM [2] (b) non-modulated [3]. In most chloro-
phyll fluorescence-related studies, the modulated method tends to
be chosen, for several reasons. One of these is that measurements
may be made during the exposure of leaves to ambient light (and
other environmental stimuli) allowing a greater physiological rele-
vance [4]. PAM fluorescence allows long-term measurements of
photosynthetic functioning in the field and can be achieved simul-
taneously with other techniques for measuring gas exchange and
photosystem I activity. The non-modulated technique requires
measurement in the absence of any actinic light source (i.e., dark-
ness) and often consists of analysis of the fast kinetics (<1 s) of the

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_6, © Springer Science+Business Media, LLC, part of Springer Nature 2018

95
96 Olubukola O. Ajigboye et al.

Fig. 1 A typical fast chlorophyll fluorescence transient rise (OJIP) in an intact healthy wheat leaf. The O level
represents the minimum that rises in less than 1 s to a maximum P, with two intermediate steps, J and I. The
x-axis is expressed as a log since the transients frequently occur within 100 ms

fluorescence rise in PSII following application of an excitation light.

When this is done a set of fluorescence transients are observed, each
of which is assumed to arise from a specific step in excitation and
electron transport. Measurement of fast fluorescence rise is also
possible using PAM equipment, but there are important differences
in terms of the type of data derived and how these data are inter-
preted [5]. Fast induction fluorescence rise has recently gained
support for use in crop phenotyping using simple parameters [1],
based on analyses of non-modulated fast fluorescence transients
induced by the application of a 1 s saturating light pulse from a
state of darkness [6]. These transients, termed O, J, I, and P (OJIP
or JIP), usually represent the amplitude of fluorescence during
specific phases of induction. The O level (equivalent to F0 in
PAM) represents the origin of the fast fluorescence that rises in
less than 1 s to a maximum peak P (equivalent to FM in PAM). J and
I are two inflection points between O and P (Fig. 1). This is
followed by a long-term decline (less than 1 min) to S, a steady-
state level [6].
Using fairly inexpensive handheld equipment, these transients
can be analyzed [7] with the JIP test, which uses the major inflec-
tion points in the fast fluorescence transient [6] to derive a number
of parameters that in principle can quantify the energy usage and
reaction dynamics within PSII. These can be useful in experiments
that seek to understand how PSII functions in response to long-
term abiotic and biotic stimuli [8–10]. The mechanism and inter-
pretation of each phase are beyond the scope of this chapter;
 Fast Timescale Chlorophyll Fluorescence in Leaves 97

however we show the common parameters that are used by avail-

able software and we point to reviews that describe underlying
biophysical processes and some empirical applications [5, 11].
While based on sound theory and some experimentation, inter-
pretation of some of these OJIP test-derived parameters may be
subject to change and should be considered carefully within their
proper physiological context. PAM fluorescence analysis now has an
enormous body of literature within diverse ecological and physio-
logical contexts and its interpretation is solid, especially when run in
parallel with other techniques such as gas exchange and spectros-
copy [3]. Although advances have now been made with non-modu-
lated techniques, we still recommend careful consideration of
whether fast-rise approaches are appropriate, depending on the
experimental context and level of interpretation required. In
terms of plant analysis and phenotyping, the fast fluorescence rise
may prove particularly useful when high numbers of genotypes are
screened rapidly in the field, coupled with the ease of handling of
the device and the volume of data output. The OJIP method can
take a matter of seconds to generate a large amount of data that can
be used to distinguish genotypes and we note that OJIP has shown
its usefulness in certain applications such as genotypic discrimina-
tion to biotic stress [9]. Other advantages include the fact that
measurements are less likely to be affected by chloroplast move-
ment due to the brevity of measurement. PAM analysis can be just
as fast but generates less data, even though it may have a firmer
theoretical footing and has been used for genotype screening (for
reviews see [4]). PAM analysis with dark adaptation through to
induction and steady state following actinic light application may
take substantially longer [4].
Here we describe a procedure for fast-rise fluorescence analysis
that is appropriate for either field or laboratory context. It includes
a period of dark adaptation of the leaf. Due to the context of this
chapter we describe procedures and equipment appropriate for field
measurements using “off-the-shelf” devices.

2 Materials

1. Fluorometer: We use a commercially available inexpensive

handheld fluorometer. The method we describe should be
applicable to most devices of this type but some details of
operation may vary (check the manufacturer’s instructions).
Dedicated continuous measurement fluorometers may be con-
structed or purchased for laboratory use. Commercially avail-
able instruments have varying levels of complexity and
accessories for dark adaptation. The fluorometer we used has
dimensions of 120
50
30 mm, and can be gripped and
operated by a single hand, and data is stored directly on the
98 Olubukola O. Ajigboye et al.

device. Power is provided by 4
AAA alkaline or rechargeable

batteries (see Note 1). A computer and specialized software are
required for data download (see the manufacturer’s instruc-
tions). A built-in clip with black foam gaskets allows insertion
of a leaf and exclusion of ambient light. The measuring window
is less than 1 cm in diameter.
2. Leaf clips or aluminum foil to dark-adapt the plants.
3. Plants for measurement, i.e., your experiment.

3 Methods

The following is a general guide and should be adapted according

to experimental objectives and equipment. Here, the fluorometer
described above is used to carry out the fast chlorophyll fluorescent
transient (OJIP) measurements on plants in the glasshouse (Fig. 1).
1. Choose a time of day. The time of day for measurement should
be selected according to the experimental objectives. As a
general guide, maximum PSII activity is often achieved around
midday, 11 am–3 pm.
2. Choose a suitable plant and leaf. The method can be applied to
most plants as long as a portion of leaf attached to the plant can
be comfortably inserted into the clip so that light is excluded.
In the case of cereal plants, it is common to measure the most
recently emerged fully expanded leaf on the main tiller, deter-
mined by length and stem diameter. In cereals such as wheat
and rice always measure the flag leaf once fully emerged (see
Note 2). Avoid leaves lower in the canopy unless they are part
of the experimental hypothesis. This is partly to avoid the self-
shading that occurs in older leaves because shading can affect
photosynthesis phenotype. In rosette plants such as Arabidop-
sis it may be less important to measure the most recently
expanded leaf but it is still necessary to avoid shaded and
senescent leaves unless this is part of your experimental hypoth-
esis. If the sequence of leaf expansion is required in Arabidop-
sis, then it is possible to track leaf number via photography of
the plants during development or labeling/tagging leaves dur-
ing development for example with non-toxic paint.
3. Dark-adapt the leaf by wrapping in aluminum foil (or use leaf
clips that might be available for your instrument) for
15–20 min [4] keeping the time of dark adaptation the same
for each leaf. Avoid damage to the leaf while handling. Wrap a
large section of the entire leaf as much as possible to avoid light
entering the center portion of the foil where the measurement
will be made. Fold the foil around the leaf to block all light
from the leaf surface. This ensures that immediate areas around
 Fast Timescale Chlorophyll Fluorescence in Leaves 99

the point of fluorescence measurements are also dark adapted

(see Notes 3 and 4).
4. Switch on the fluorometer and allow it to equilibrate with
ambient temperature for 5–10 min (see Note 5).
5. Set up the fluorometer. The saturating light can be adjusted
from 0 to 100% (up to 3000 μmol photon m2 s1). Check
that the saturating light is set to the desired intensity. Here, we
have used the default setting of 70% saturating light intensity
under controlled conditions (see Note 6). It is strongly advis-
able to use the same light intensity for a given experiment or set
of experiments.
6. Check the device memory. Make sure that you have enough
capacity for your data. It is best to have 100% memory capacity
for field data since the JIP test generates a lot of large files (see
Note 7).
7. Insert the dark-adapted leaf into the clip. Press and hold open
the leaf clip on the fluorometer. With the foil still on the leaf,
place the sensor head over a central section of the foil. Make
sure that the adaxial surface of the leaf will completely cover the
aperture of the sensor window. Close the leaf clip.
8. Remove the foil. The most important point about this proce-
dure is that no light, or as little light as possible, reaches the leaf
while the foil is removed and where the leaf clip is positioned.
Ideally this process is carried out in a dark environment. In the
field it may be difficult to achieve perfect dark adaptation and it
is advisable to try to carry this out in shade (see Note 3). With
the same hand used to hold the device, gently press the leaf clip
to open it slightly, and with your free hand slide the foil off the
leaf (see Note 8). Avoid damage to the leaf. Do not pull on the
leaf; this can cause the leaf to detach from the plant.
9. Operate the device for the appropriate measurement. This may
have different terms depending on the device manufacturer. The
measurement should only be taken once on the same leaf area
within the leaf clip. If you need to repeat the measurement, then
change the leaf section and preferably the leaf itself (see Note 9).
10. Extract the data. On the fluorometer we use here, the software
on the device extracts data along the fluorescence transient
(Fig. 1) including fluorescence intensity at F0 (50 μs), fluores-
cence intensities at J-step (2 ms) and I-step (60 ms), and peak
of the transient FP  FM. The automated parameter calcula-
tions are normally made using log time and the original trace
can be exported for further calculation of parameters if
required. We recommend on a periodic basis to manually
check that the parameters calculated by the device software
utilize the correct positions on the original trace data. Biophys-
ical parameters involving energy fluxes per active reaction
100 Olubukola O. Ajigboye et al.

centers are automatically computed from the extracted data by

the software on the device (see the manufacturer’s instructions
in each case) [6] (Table 1). For example, the device we use
automatically computes parameters from the OJIP transient
using the JIP test method (see Note 10, Table 2). FV/FM is
computed as [(FM  F0)/FM].
11. Measured data and calculated parameters are sequentially
stored in the internal device memory. Transfer data and calcu-
lated parameters to a computer via USB or Bluetooth commu-
nication as per the manufacturer’s instructions.
12. Ensure that your dataset has appropriate replication for statis-
tical analyses. The choice of sample size per treatment should
be sufficient for meaningful statistical analysis with sufficient
power within the experimental design to draw an accurate
conclusion. As an example in our glasshouse experiment, the
sample size for each genotype studied was four, with one
measurement per plant, i.e., four biological replicates (see
Note 11). The data obtained were analyzed using analysis of
variance (ANOVA) from which conclusions were drawn
(Table 2).

4 Notes

1. Ensure that batteries are fully charged. Always keep with you a
spare set because the repeated use of the high-intensity pulse
can rapidly drain power.
2. The flag leaf contributes the largest proportion of photosyn-
thate to grain filling in wheat. It is best to measure fully
emerged leaves because the rate of photosynthesis in such
leaves is almost or at maximum after leaf emergence and then
plateaus before declining [13].
3. The short-term dark adaptation provided by the manufacturer’s
standard leaf clip may be sufficient for rapid screening of plants
in the field and a replacement for other methods such as alumi-
num foil, particularly when sampling large numbers of plants.
Great care must be taken with dark adaptation in the field and it
may be necessary to use material to further darken the leaf clip
(and scientist!) while the leaf is being moved between the dark-
adaptation position and the device cuvette. For example
growth-room lights could be switched off or a portable card-
board box can be used for small plants. In the field dark cloth
can be draped over the user’s head and if large enough it will
partially darken the plant being operated on below. If you
suspect that the leaf may have experienced a small amount of
light, then it can be worthwhile to wait 30 s before making the
measurement, in order to allow high-energy-state
 Fast Timescale Chlorophyll Fluorescence in Leaves 101

Table 1
Common derivations and definitions of JIP parameters directly obtained from the recorded OJIP
fluorescence transients

Technical fluorescence parameters Definition

F0 Minimal fluorescence at 50 μs, used as initial value of
fluorescence
F300 μs Fluorescence value at 300 μs
FJ  F2 ms Fluorescence value at 2 ms (J level)
FI  F60 ms Fluorescence value at 60 ms (I level)
FM ¼ FP Maximal fluorescence under saturating light
FV ¼ FM  F0 Maximum variable Chl fluorescence
VJ ¼ (F2 ms  F0)/(FM  F0) Relative variable fluorescence at 2 ms
VI ¼ (F60 ms  F0)/(FM  F0) Relative variable fluorescence at 60 ms
M0 ¼ (TR0/RC)  (ET0/ Initial slope of relative variable Chl fluorescence curve
RC) ¼ 4 ms1 (F300 μs  F0)/ (approximate value)
(FM  F0)
Energy fluxes Definition
ABS Total photon flux absorbed by the PSII antenna pigments
TR0 Trapped exciton flux by PSII reaction centers
DI0 Dissipated energy flux in PSII, e.g., as heat and fluorescence
ET0 Electron transport flux from QA to QB
Quantum yield Definition
FV/FM ¼ φPo ¼ TR0/ABS ¼ Maximum quantum yield of primary PSII photochemistry at
(FM  F0)/FM time 0
φEo ¼ φPo · (1  VJ) Quantum yield of electron transport at time 0
ψ0 ¼ (1  VJ) Probability at time 0 that a trapped exciton moves an electron
further into the electron transport chain beyond QA
Specific energy fluxes (per active PSII Definition
reaction centers)
ABS/RC ¼ (M0/VJ) · (1/φPo) Average absorbed photon flux per PSII reaction center
(apparent antenna size of an active PSII)
TR0/RC ¼ M0/VJ Maximum trapped exciton flux per PSII reaction center
DI0/RC ¼ (ABS/RC)  (TR0/RC) Dissipated energy flux per PSII reaction center
ET0/RC ¼ (M0/VJ) · (1  VJ) Electron transport flux from QA to QB per PSII reaction center
Reproduced from ref. [9] with permission from CSIRO Publishing. Formulas derived from [12]. Chl chlorophyll
 102

Table 2
Resistance rating, efficiency of PSII photochemistry (FV0 /FM0 ), dissipated energy flux (DI0/RC), electron transport flux (ET0/RC), absorbed photon flux
(ABS/RC), trapped energy flux (TR0/RC), probability that a trapped exciton moves an electron further than QA (ψ0), and quantum yield of electron
transport (φEo) of wheat varieties 7 and 14 days after inoculation (dai) with powdery mildew (Blumeria graminis)

Fv0 /Fm0 DI0/RC ET0/RC ABS/RC TR0/RC ψ0 φEo

Variety Resistance
7 dai 14 dai 7 dai 14 dai 7 dai 14 dai 7 dai 14 dai 7 dai 14 dai 7 dai 14 dai 7 dai 14 dai
Olubukola O. Ajigboye et al.

Duxford 5 0.788 0.782 0.56 0.62 1.01 1.10 2.61 2.80 2.05 2.18 0.50 0.51 0.39 0.40
Xi19 6 0.801 0.792 0.49 0.55 1.10 1.09 2.48 2.63 1.98 2.08 0.56 0.53 0.45 0.42
Claire 4 0.803 0.787 0.50 0.59 1.09 1.11 2.51 2.76 2.01 2.17 0.54 0.52 0.44 0.41
Mascot 7 0.795 0.790 0.53 0.56 1.14 1.14 2.57 2.69 2.04 2.12 0.56 0.54 0.45 0.43
Panorama 7 0.796 0.795 0.54 0.53 1.17 1.06 2.59 2.61 2.05 2.07 0.57 0.51 0.45 0.41
Battalion 8 0.795 0.799 0.54 0.52 1.08 1.11 2.60 2.60 2.06 2.07 0.53 0.54 0.42 0.43
Glasgow 5 0.796 0.785 0.53 0.60 1.12 1.11 2.59 2.77 2.06 2.17 0.55 0.52 0.43 0.41
P 0.166 0.042 0.285 0.023 0.004 0.608 0.381 0.023 0.452 0.033 <0.001 0.626 0.001 0.526
LSD 0.011 0.011 0.062 0.061 0.076 0.076 0.139 0.147 0.083 0.090 0.035 0.041 0.030 0.036
On 1–9 scale: 1 ¼ susceptible, 9 ¼ resistant
Chlorophyll fluorescence measurements taken on asymptomatic leaves (youngest fully expanded leaves) on the main tiller (reproduced from ref. [9] with permission from CSIRO
Publishing). ANOVA significant at P < 0.05, followed by LSD for pairwise comparisons between means. ANOVA analysis of variance; LSD least significant difference
 Fast Timescale Chlorophyll Fluorescence in Leaves 103

non-photochemical quenching (NPQ) to relax as much as pos-

sible. Better still, start the dark adaptation again on a
different leaf.
4. We recommend dark adaptation for many applications. Dark
adaptation, and the time allocated to this process, allows high-
energy-state quenching (qE) to relax and this can affect key
parameters in the JIP test calculation (Table 1). This is impor-
tant because the variability in environmental parameters (espe-
cially light) experienced by leaves in the field is large and this
can lead to a high level of variation in NPQ. Such data will be of
relevance where long-term abiotic and biotic stress is applied.
5. Allow the fluorometer to equilibrate to room or ambient tem-
perature. This is important for consistent readings because leaf
physiological processes and device electronic stability are tem-
perature sensitive. Variations between temperature of the fluo-
rometer and leaf sample may cause errors in readings.
Therefore try to avoid variations in temperature of the
instrument.
6. A high saturating intensity is recommended to separate the
O-J, J-I, and I-P phases (3000–10,000 μmol photon m2 s1
according to [4]). Options for this are provided with each
device. Follow the manufacturer’s instructions.
7. Captured data stored in the internal memory of the device are
allocated index numbers sequentially. When large numbers of
plants are sampled, such as in the field, measurements can be
tracked by recording at specific intervals the index number of
logged measurements and the corresponding plant number/
treatment or plot. An alternative is to make use of accurate GPS
positioning technology if available.
8. Sliding the foil off in this way may only be applicable to fluo-
rometers with attached leaf clip.
9. Light capture, excitation energy trapping, and transfer occur
on a very fast timescale. Therefore in case you need to repeat a
measurement, change the leaf section or preferably the leaf
itself to ensure that the reaction centers of PSII are fully
relaxed.
10. The JIP test, based on a number of assumptions, provides
information on the activity of the donor and acceptor sides of
PSII and is in turn dependent on the interpretation of the
different phases of the fast fluorescence transient curve. It is
important that parameters derived through the JIP test are
considered collectively and not individually in order to distin-
guish genotypes while assessing plant response [9, 13, 14] such
as biotic stress (see Table 2).
11. For field-grown plants such as wheat, a sample size of 7–10
plants per 20 m2 plot area may be sufficient.
104 Olubukola O. Ajigboye et al.
References
1. Pask AJD, Pietragalla J, Mullan D, Reynolds
MP (eds) (2012) Physiological breeding II: a
field guide to wheat phenotyping. Mexico,
International Wheat and Maize Improvement
Center (CIMMYT)
2. Schreiber U, Schliwa U, Bilger W (1986) Con-
tinuous recording of photochemical and
non-photochemical chlorophyll fluorescence
quenching with a new type of modulated fluo-
rometer. Photosynth Res 10:51–62
3. Strasser RJ, Govindjee (1991) The F0 and the
O-J–I–P fluorescence rise in higher plants and
algae. In: Argyroudi-Akoyunoglou JH
(ed) Regulation of chloroplast biogenesis. Ple-
num Press, New York, pp 423–426
4. Murchie EH, Lawson T (2013) Chlorophyll
fluorescence analysis: a guide to good practice
and understanding some new applications. J
Exp Bot 64:3983–3998
5. Lazár D (2006) The polyphasic chlorophyll a
fluorescence rise measured under high intensity
of exciting light. Funct Plant Biol 33:9–30
6. Strasser RJ, Tsimilli-Michael M, Srivastava A
(2004) Analysis of the chlorophyll a fluores-
cence transient. In: Papageorgiou GC, Govind-
jee (eds) Chlorophyll a fluorescence: a
signature of photosynthesis. Kluwer Academic
Publishers, Dordrecht, pp 321–362
7. Strasser RJ, Srivastava A, Govindjee (1995)
Polyphasic chlorophyll a fluorescence transi-
ents in plants and cyanobacteria. Photochem
Photobiol 61:32–42
8. Ajigboye OO, Murchie EH, Ray RV (2014)
Foliar application of isopyrazam and epoxico-
nazole improves photosystem II efficiency,
biomass and yield in winter wheat. Pestic Bio-
chem Physiol 114:52–60
9. Ajigboye OO, Bousquet L, Murchie EH, Ray
RV (2016) Chlorophyll fluorescence para-
meters allow the rapid detection and differenti-
ation of plant responses in three different wheat
pathosystems. Funct Plant Biol 43(4):356–369
10. Ajigboye OO, Lu C, Murchie EH, Schlatter C,
Swart G, Ray RV (2017) Altered gene expres-
sion by sedaxane increases PSII efficiency, pho-
tosynthesis and growth and improves tolerance
to drought in wheat seedlings. Pestic Biochem
Physiol 137:49–61
11. Stirbet A, Govindjee (2011) On the relation
between the Kautsky effect (chlorophyll
a fluorescence induction) and photosystem II:
basics and applications of the OJIP fluores-
cence transient. J Photochem Photobiol B
104:236–257
12. Strasser RJ, Srivastava A, Tsimilli-Michael M
(2000) The fluorescence transient as a tool to
characterize and screen photosynthetic sam-
ples. In: Yunus M, Pathre U, Mohanty P (eds)
Probing photosynthesis: mechanism, regula-
tion and adaptation, vol 25. Taylor and Francis,
London, pp 445–483
13. Suzuki S, Nakamoto H, Ku MSB, Edwards GE
(1987) Influence of leaf age on photosynthesis,
enzyme activity, and metabolite levels in wheat.
Plant Physiol 84:1244–1248
14. Ripoll J, Bertin N, Bidel LPR, Urban L (2016)
A user’s view of the parameters derived from
the induction curves of maximal chlorophyll
a fluorescence: perspectives for analysing stress.
Front Plant Sci 7:1679
 Chapter 7

Sub-saturating Multiphase Flash Irradiances to Estimate

Maximum Fluorescence Yield
Thomas J. Avenson and Aaron J. Saathoff

Abstract
Many intricacies of leaf-level photosynthesis can be probed by combining infrared gas analysis with pulse-
amplitude-modulation chlorophyll a fluorometry. A key fluorescence yield (ΦF) parameter required for
estimating many of the phenomena associated with the light reactions of photosynthesis is referred to as the
maximum ΦF, which is termed Fm0 when measured on a light-adapted leaf. While ubiquitously used to
assess many aspects of photosynthesis, Fm0 is problematic because it is prone to being underestimated. This
error can be propagated to parameters and phenomena that are based on estimation of Fm0 . Theoretical and
experimental observations have shown that ΦF increases hyperbolically in response to increasing irradiance,
asymptotically approaching the maximum ΦF, or Fm0 , at extreme irradiances. Importantly, depending upon
the convexity of the hyperbolic response, ΦF exhibits a linear and inverse relationship with the reciprocal of
irradiance, a relationship previously referred to as a reciprocal plot. Given the negative slope of the reciprocal
plot, estimates of ΦF at infinite irradiance can be obtained, even over sub-saturating irradiances, by linear
regression and extrapolation of the resultant reciprocal plot to the y-intercept. Here, we show how to obtain
data from a dynamic multiphase flash of sub-saturating irradiance, occurring within the time span of ~1 s, to
generate a reciprocal plot that subsequently provides an accurate estimate of ΦF at infinite irradiance, or Fm0 .

Key words Maximum fluorescence yield, Multiphase flash, Photosynthesis, Photosystem II, Pulse-
amplitude-modulation chlorophyll a fluorescence, Saturation flash

1 Introduction

In a leaf engaged in steady-state photosynthesis, pulse-amplitude-

modulation (PAM) chlorophyll a fluorometry can be used, in
conjunction with the traditional saturation flash (Q0 ) method [1],
to assess the steady-state minimum (Fs) and maximum (Fm0 ) fluo-
rescence yield (ΦF). Fm0 is explicitly assumed to represent the true
maximum ΦF (see Eqs. (1) and (2)), as opposed to the apparent
maximum ΦF (AFm0 ) [2–4]. PAM chlorophyll a fluorometry takes
advantage of a low intensity-modulated (i.e., pulsed) light source,
with the pulse duration typically taking place on a μsec timescale, to
elicit measurements of Fs and, in conjunction with application of a
Q0 , Fm0 . Application of a Q0 , typically severalfold higher in intensity

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_7, © Springer Science+Business Media, LLC, part of Springer Nature 2018

105
106 Thomas J. Avenson and Aaron J. Saathoff

than full sunlight and 0.5–1 s in duration, concomitant with the

modulated light, is a methodological means of manipulating the
redox dynamics of the photosystem (PS) II reaction center [5] for
the purpose of obtaining an estimate of Fm0 (but see the following
discussion). Estimation of Fm0 is thus a transitory measurement
involving a Q0 -induced transition from Fs ! Fm0 .
There are several, frequently unspoken assumptions underlying
accurate determination of Fm0 that can be helpful to explicitly
articulate in the present context. It is useful to define the mathe-
matical descriptions of Fs and Fm0 . Based on a lake model of PSII
photo-physics [6], Fs can be defined as
kF
F s ¼ ΦF ¼ ð1Þ
ΣðkF þ kd þ kISC þ kNPQ þ kpi  qLÞ

where kF, kd, kISC, kNPQ, and kpi correspond to rate constants for
fluorescence, basal non-radiative decay, intersystem crossing of sin-
glet excited chlorophyll into the triplet state, non-photochemical
quenching (NPQ) [7], and intrinsic capture of excitation energy by
reaction centers, respectively. qL represents the proportion of oxi-
QA
dized, or “open,” reaction centers (i.e., qL ¼ ΣðQ þQ  [6], where
Þ
A A

QA and QA correspond to the oxidized and reduced states, respec-

tively, of the primary electron acceptor within the PSII reaction
center). The expression for Fm0 can be described as
kF
F m 0 ¼ ΦF ¼
ð2Þ
Σ kF þ kd þ kISC þ kNPQ

According to comparison of Eqs. (1) and (2), technically accu-

rate estimation of Fm0 is based on the assumptions that the Q0
(1) does not cause a change in the rate constants for non-radiative
decay, intersystem crossing, and NPQ, processes that are in compe-
tition with fluorescence for dissipating chlorophyll excited states;
(2) does not “induce” any of several auxiliary reactions (i.e., those
capable of being activated by the Q0 and that can compete with
fluorescence for dissipating chlorophyll excited states [8]); and
(3) causes QA to be completely converted to its reduced state, or
QA (i.e., qL ! 0); thus kpi  qL does not appear in the denomi-
nator of Eq. (2). It is important to note that the expression
“qL ! 0” is meant to convey the assumptions that both qL ¼ 0
and dqL/dt ¼ 0 during a Q0 , which is to say that qL is assumed to
decrease to zero during the Q0 and remain so throughout the
duration of the saturation flash. As qL ! 0, formation of QA in
the PSII reaction center slows the rate of radical pair formation
(P680+-Pheo) [5], thereby increasing the probability that absorbed
light will be dissipated via fluorescence, i.e., Fm0 will be achieved.
Measurement of an assortment of photosynthetic phenomena
is dependent on accurate estimation of Fm0 . This includes the
 Fm0 from a Single Flash 107

quantum yield of PSII-mediated electron transfer (ΦPSII) (i.e.,

ΦPSII ¼ [Fm0  Fs]/Fm0 ) [9], which can be used to assess coupling
between the light-dependent and carboxylative reactions of photo-
synthesis. ΦPSII is a measure of the efficiency with which absorbed
light energy is used to promote linear electron transfer (J)
(J ¼ ΦPSII  PPFD  α  fII; α and fII represent absorbance and
fraction of absorbed light partitioned to PSII, respectively [9, 10]).
J transduces light energy into the chemical energy that is used for
assimilating CO2 and powering other processes in the chloroplast
[11]. Comparison of J with gross CO2 assimilation, independently
measured by infrared gas analyses (IRGA), provides an estimate of
the electron requirement of CO2 assimilation [10]. NPQ is a
composite of photoprotective mechanisms that is critical for plant
survival [12] and is calculated from Fm0 and the maximum ΦF (Fm)
obtained after prolonged (i.e., overnight) dark adaptation
(NPQ ¼ [Fm  Fm0 ]/Fm0 ) [7].
While estimates of Fm0 are widely used to estimate many differ-
ent photosynthetic phenomena, there is a problem: Fm0 is prone to
underestimation by the traditional saturation flash methodology
and consequently error can be propagated to estimation of
Fm0 -derived phenomena [2–4]. An extensive model of PSII exciton
dynamics was previously developed to quantitatively assess various
aspects of the irradiance dynamics of ΦF and, in part, explore the
underlying causative mechanisms responsible for the propensity of
Fm0 to be underestimated [3]. The model was developed to explic-
itly assess the response of ΦF solely to irradiance-induced changes in
the redox state (i.e., QA ! QA, etc.) of the PSII reaction center,
e.g., the model was constructed so that the rate constants for
competitive processes remained constant and rate constants for
auxiliary reactions were excluded from the model infrastructure.
The model revealed, consistent with the seminal findings of Mark-
graf and Berry [4], that ΦF increases hyperbolically in response to
increasing irradiance, implying that the reason for Fm0 being prone
to underestimation is that ΦF asymptotically approaches its maxi-
mum value, potentially requiring extreme irradiances (i.e., upwards
of 60,000 μmol m2 s1 under certain simulated adaptive states) to
reach the asymptotic maximum value. The modeling results
revealed that the causative reasons underlying the hyperbolic
response of ΦF to increasing irradiance involve, in part, a convolu-
tion of QA being a hyperbolic function of Q0 and ΦF itself being a
hyperbolic function of QA, the latter phenomenon of which being
due to excitons reversibly exchanging between PSII units (i.e., PSII
connectivity) [13].
These theoretical considerations have important, practical
implications for experimental determination of Fm0 . On the one
hand, achieving the high irradiances necessary to accurately assess
maximum Fm0 is problematic from an instrumental point of
view, especially if large leaf areas (2 cm2–9 cm2) or sample volumes
108 Thomas J. Avenson and Aaron J. Saathoff

(i.e., algal suspensions) are being assessed, because of limitations

imposed by light-emitting diode (LED) capacity. On the other
hand, even if an instrument could provide such extreme intensities,
it is likely that such intensities would induce auxiliary reactions,
e.g., formation of cation species within the PSII reaction center that
are capable of quenching fluorescence [8], and thereby cause viola-
tion of a key assumption of the traditional saturation flash method-
ology (i.e., the saturation flash intensity does not induce such
reactions).
A fascinating feature of the hyperbolicity of ΦF as a function of
increasing irradiance is that the same ΦF response is an inverse
function of the reciprocal of flash irradiance (Q0 1) [2–4], a rela-
tionship previously referred to as a reciprocal plot [3]. Notably, the
inverse relationship exhibits linearity, with a negative slope, the
closer ΦF is to the asymptotic maximum. Thus, an estimate of ΦF
at “infinite irradiance” (i.e., as Q0 1 ! 0), a value previously
referred to as extrapolated Fm0 (EFm0 ) and that has been shown to
be an accurate approximation of Fm0 [2, 3], can be obtained via
linear regression and extrapolation of the reciprocal plot to the y-
intercept [2–4]. These combined results indicate that while Fm0 is
prone to underestimation using extreme flash intensities, its value
can nonetheless be closely approximated using sub-saturating irra-
diances that take advantage of the inverse, linear relationship
between ΦF and Q0 1.
A method employing a sub-saturating, multiphase flash (MPF)
of variable irradiance, involving three contiguous, millisecond
phases of flash irradiance, all occurring within the time span of
~1 s, has been developed that is capable of providing ΦF dynamics
from which a reciprocal plot and a resultant estimate of EFm0 can be
obtained [3]. During a MPF, a maximum irradiance referred to as
phase 1 is followed by partial, linear attenuation of the phase
1 irradiance during phase 2, after which the irradiance returns to
the phase 1 intensity during phase 3. Accurate determination of Fm0
by MPF methodology is predicated on a subtly different combina-
tion of assumptions than those associated with traditional satura-
tion flash methodology. As assumed by traditional saturation flash
methodology, technically accurate determination of Fm0 by MPF
methodology is also predicated on the assumptions that (1) the rate
constants for unambiguous reactions (i.e., kNPQ) remain constant
during a MPF, and (2) the MPF does not induce ambiguous
reactions (i.e., long-lived cationic states that quench fluorescence
[8]). In contrast to the third assumption of the traditional satura-
tion flash method, with respect to qL, the MPF method is predi-
cated on the nuanced assumption that while dqL/dt ¼ 0 during
phases 1 and 3, qL 6¼ 0, which is to say that qL need not necessarily
decrease to zero during phases 1 and 3, but rather that qL does not
change during phases 1 and 3. Furthermore, ramp rates
>0.01 mol m2 s2 ensure that changes in qL during phase
 Fm0 from a Single Flash 109

2 occur via quasi-steady-state redox dynamics [3]. Therefore,

steady-state redox dynamics during phases 1 and 3, combined
with quasi-steady-state phase 2 redox dynamics, ensure accurate
determination of Fm0 by the MPF approach [3].
Herein, an approach for discerning MPF irradiation dynamics
from which to obtain an accurate approximation of Fm0 will be
described, including a discussion of example data and their analyses.
In addition, given the novelty of the MPF approach [3], in com-
parison to the traditional saturation flash method [1], the two
approaches will be conceptually juxtaposed to gain insight into
their validity, thereby providing guidance to researchers as to their
proper use.

2 Materials

1. Nicotiana tabacum: Sow seeds in soil and, if needed, grow

resultant plants in a greenhouse with supplemental LED
lighting. Use the most recent, fully expanded leaf for physio-
logic analyses (see Note 1).
2. Use a photosynthesis system that is capable of combined IRGA
and PAM chlorophyll a fluorescence analyses (see Note 2). The
gas exchange system must be capable of extensive environmen-
tal control of the leaf chamber and the PAM chlorophyll
a fluorometer must possess MPF functionality (see Note 3).
The LEDs producing the photosynthetically active radiation
(PAR) (see Note 4) and modulated light sources of the instru-
ment should overlap spectrally, where the actinic and MPF
irradiance dynamics are produced using the same LEDs. A
nuance, however, is that the MPF should be produced only
by the red LEDs that spectrally overlap with the red modulated
LEDs. MPF irradiance profiles are created and precisely timed
by a combination of onboard software and digital/analog elec-
tronic circuitry within the fluorometer. The modulated LEDs
are functionally separate and dynamic, so as to be capable of
being adjusted to optimize the ΦF signal-to-noise ratio while
simultaneously limiting the extent to which the integrated
modulation intensity contributes to PAR (see Note 5). Fs is
elicited by the modulated light [1] and is taken as the average
value of ΦF over several seconds timescale just prior to initia-
tion of a MPF.

3 Methods

1. Configure the gas exchange functionality [14] of the photo-

synthesis system by controlling the leaf chamber environment
110 Thomas J. Avenson and Aaron J. Saathoff

at (see Notes 3, 6, and 7) (1) a flow rate of 500 μmol air s1,
(2) a mixing-fan speed of, e.g., 10,000 rpm, (3) a leaf vapor
pressure deficit (VPDLeaf) of 1.5 kPa, (4) a leaf temperature
(TLeaf) of 25  C, (5) a PAR of 1000 μmol m2 s1 (90% red and
10% blue), and (6) a chamber [CO2] of 400 μmol CO2 mol
air1.
2. Configure the modulation light source of the PAM chlorophyll
a fluorometer (see Note 2). Set the modulation LED frequen-
cies during steady-state PAR and a MPF to 50 KHz and
250 KHz, respectively, and the modulation pulse amplitude
to 100 μmol m2 s1, resulting in integrated modulation
intensities, for example, of ~5 μmol m2 s1 and
25 μmol m2 s1, respectively (see Note 5).
3. Configure the maximum irradiances of the MPFs for the PAM
chlorophyll a fluorometer (see Notes 2 and 8). Control the
maximum irradiances of phases 1 and 3 by (1) varying their
intensities (see Subheading 3, steps 6–8) and (2) configuring
their durations to 300 ms.
4. Configure the phase 2 irradiance dynamics of the MPFs for the
PAM chlorophyll a fluorometer (see Notes 2 and 8). Set the
phase 2 “ramp depth” and duration to 25% and 300 ms,
respectively. The phase 2 ramp depth and duration define the
“ramp rate,” which should not exceed 0.01 mol m2 s2 [3].
5. Clamp the chosen leaf into the measuring chamber and estab-
lish steady-state conditions (see Subheading 3, step 1, and
Note 7). Allow enough time, e.g., approximately 45–60 min,
for net CO2 assimilation (ANet) and Fs to reach a steady state, as
determined by dAnet/dt ¼ 0 and dFs/dt ¼ 0.
6. Apply a MPF to the leaf engaged in steady-state photosynthesis
and analyze the irradiance and ΦF dynamics. Download the file
(see Notes 2 and 9) containing the time, irradiance, and ΦF
dynamics that were acquired during the MPF and explore
how the irradiation and ΦF dynamics occur during a MPF (see
Notes 10–13).
7. Determine the ΦF at infinite irradiance, i.e., EFm0 , using the
data acquired in Subheading 3, step 6. Plot and analyze the
phase 2 ΦF versus the reciprocal of the phase 2 irradiance (see
Note 14).
8. Determine the phase 1 intensity dependence of AFm0 and EFm0 .
Randomly apply a series of MPFs, separated in time by 2 min
(see Note 15) and ranging in phase 1 intensity, e.g., between
3000 μmol m2 s1 and 15,000 μmol m2 s1, to a selected
leaf engaged in steady-state photosynthesis (see Subheading 3,
steps 1–7, and Note 7). Each of the MPFs provides
two approximations of Fm0 : AFm0 (see Note 10) and EFm0 (see
Note 14). Plot the respective estimates of AFm0 and EFm0 as a
 Fm0 from a Single Flash 111

function of phase 1 intensity and analyze (see Notes 16

and 17).
9. Determine the technical accuracy of the MPFs (see Notes
18–22). Independently assess the percent
 differences (%D)
for ΦF and Q0 ( %D ¼ ðP3P1Þ
P3  100 ) during phases 1 and
3 for the various MPFs (see Subheading 3, steps 6–8). P1 and
P3 correspond to the respective values of either ΦF or Q0
averaged over the last 60 ms of phases 1 and 3, respectively.
10. Use the data obtained from Subheading 3, steps 6–8, to
carefully discern MPF irradiation dynamics from which to
obtain an accurate approximation of Fm0 (see Notes 18–24).

4 Notes

1. Leaves from different plants, no matter their growth environ-

ment (i.e., field-grown, greenhouse-grown) or developmental
stage, can be used for experimentation [3]. The main point is
that it is best if the leaf area of leaves being studied fills the
entire chamber of the combined gas exchange/PAM chloro-
phyll a fluorometer (see Subheading 2, item 2, and Note 2);
signal-to-noise ratios of the various signals will thus be
optimized.
2. Photosynthesis systems capable of performing the combined
gas exchange and PAM chlorophyll a fluorescence experiments
described herein have user-friendly software interfaces that
allow easy control of the pertinent parameters needed for gas
exchange and fluorometric functionality. For gas exchange
functionality, parameters such as the sample chamber [CO2]
and flow rate (see Note 7) are readily programmable. With
respect to PAM chlorophyll a fluorescence functionality,
instrumental user interfaces enable control of the (1) various
light sources (i.e., modulation, actinic, and saturation flash)
and (2) durations, and changes in intensities, of phases 1–3 of
a MPF. When duration and amplitude of phase 2 are selected,
the instrument automatically, and linearly, attenuates the phase
1 intensity by the chosen amplitude. Raw data, e.g., changes in
ΦF and irradiance during a MPF, from such photosynthesis
systems can be readily accessed through Ethernet exchange,
enabling post-analyses of raw data.
3. Not all commercial gas exchange/PAM chlorophyll
a fluorometric systems are capable of strict environmental con-
trol and MPF functionality. As of the writing of this chapter,
two companies engineer such systems: PP Systems and
LI-COR Biosciences. PP Systems engineers the CIRAS-3 pho-
tosynthesis system and LI-COR Biosciences engineers the
112 Thomas J. Avenson and Aaron J. Saathoff

LI-6400, equipped with the 6400-40 fluorometer chamber,

and the recently introduced LI-6800 photosynthesis systems.
All three instruments are “open” gas exchange systems and are
thus predicated on steady-state (see Note 7) mathematical
principles [15].
4. PAR is typically, although not exclusively, produced by differ-
ent sets of red (i.e., λpeak ¼ 632 nm) and blue (i.e.,
λpeak ¼ 450 nm) LEDs. Spectral compositions of PAR are
typically 90% red and 10% blue light. Photosynthesis is
“driven” by both red and blue light, whereas stomatal regula-
tion specifically requires blue light. The spectrum of the LEDs
of the MPF and the modulated light should spectrally overlap
so that similar chloroplast populations are probed (i.e., various
colors of light differentially penetrate into a leaf [16]).
5. The integrated intensity (II) of the modulated light can be
quantified as

I I ¼ P w  νml  P I ð3Þ

where Pw, νml, and PI correspond to the modulated “pulse

width” (i.e., μsec), the frequency of the modulated pulses
(pulses per sec), and the peak intensity (μmol m2 s1) of the
modulated pulse, respectively. Pw and PI are typically fixed in
the instrument hardware and software, respectively, at, for
example, ~2 * 105 s and ~100 μmol m2 s1. Thus, the
integrated intensity of the modulated light can be adjusted by
varying the modulation frequency. This flexibility allows for
signal averaging, ultimately facilitating optimization of the ΦF
signal-to-noise ratio.
6. Commercial instruments that perform gas exchange and PAM
chlorophyll a fluorescence, and which possess MPF functional-
ity (see Notes 2 and 3), are also capable of automatically
controlling the environmental conditions of the leaf chamber.
This environmental control is important to maintain steady-
state conditions during experiments (see Note 7).
7. Take advantage of steady-state conditions to avoid complexities
and ambiguities associated with non-steady-state dynamics.
The physiologic state required for accurate determination of
Fm0 by the approach described herein involves taking advantage
of steady-state, leaf-level ANet [14, 15], as achieved by exposing
a leaf to the incident PAR of interest and near-ambient levels of
CO2 (400 μmol CO2 mol air1) (see Subheading 3, steps 1 and
5). Leaf-level ANet, as measured by an open gas exchange
system [14, 15], is roughly quantified as
 Fm0 from a Single Flash 113

μðC R  C S Þ
A Net ¼ þ C SE ð4Þ
s
where μ, CR, CS, s, and E represent flow rate (μmol air s1),
reference [CO2], sample chamber [CO2], leaf area, and tran-
spiration, respectively. An important chamber parameter not
explicitly included in Eq. (4) concerns the extensive mixing of
the chamber gas, which is accomplished using high-speed fans
(see Subheading 3, step 1). Eq. (4) quantitatively expresses that
ANet is estimated as the product of flow rate multiplied by the
difference between the reference and sample [CO2], which is
then normalized to the leaf area that is enclosed in the cham-
ber. The final term in Eq. (4) (i.e., CSE) represents a H2O
correction term associated with leaf transpiration and is essen-
tial for accurate measurements of ANet. Importantly, Eq. (4) is
predicated on steady-state assumptions [15, 17].
8. While different aspects of a MPF can be varied, e.g., the dura-
tions of any one of the three phases, the amplitude of the phase
2 ramp, etc. use phase 1–3 durations, as well as a phase 2 ramp
depth, as defined above (see Subheading 3, steps 3–7); these
settings have been previously shown to be optimal for accu-
rately estimating Fm0 using the MPF method [3]. Thus, for the
purpose of determining appropriate MPF dynamics using a
given leaf, assess, as previously suggested (see Fig. 10 in [3]),
variability in phase 1 intensity.
9. The photosynthesis systems that perform gas exchange and
MPF functionality possess onboard analyses of the raw changes
in irradiance and ΦF during a MPF, providing as output the
respective values of AFm0 and EFm0 . It is nonetheless useful,
for didactic purposes, to manually analyze the raw irradiance
and ΦF dynamics that occur during a MPF. This can be
achieved by removing the spreadsheet files from the instrument
(see Note 2) that contain the data and post-analyzing the MPF
irradiation and ΦF dynamics.
10. Analyze the time dependence of the irradiance and ΦF of a
MPF. Figure 1a depicts what the irradiance and ΦF dynamics of
a MPF should look like when obtained from a leaf adapted to
the steady-state conditions (see Subheading 3, steps 1–6). The
irradiance initially, and near instantaneously, increases from the
steady state of 1000 μmol m2 s1 to a phase 1 maximum (i.e.,
effectively a traditional saturation flash) of 8000 μmol m2 s1,
during which AFm0 is attained. The phase 1 irradiance remains
constant for 300 ms (see Subheading 3, step 3); as such, any
observed changes in ΦF during phase 1 can be ascribed to leaf-
level, biophysical phenomena and not to changes in phase
114 Thomas J. Avenson and Aaron J. Saathoff

Fig. 1 Example of multiphase flash irradiation and ΦF dynamics. (a) Contiguous

changes in phase 1–3 irradiances (300 ms each) (closed squares) and ΦF (open
squares) that were obtained by applying a MPF to a leaf undergoing steady-state
photosynthesis at a PAR of 1000 μmol m2 s1. The phase 1 intensity was
8000 μmol m2 s1 and the phase 2 ramp (i.e., amplitude of linear attenuation
of phase 1 intensity) and duration were 25% and 300 ms, respectively, resulting
in a ramp rate of 0.01 mol photons m2 s2. The phase 1 apparent maximum ΦF
(AFm0 ) is indicated by the arrow. (b) The phase 2 ΦF is shown plotted versus the
reciprocal of the phase 2 irradiance (Q0 1). Linear regression and extrapolation
of the reciprocal plot to the y-intercept provide an estimate of ΦF at infinite
irradiance (i.e., EFm0 ). Note that as Q0 ! 1, Q0 1 ! 0 (i.e., the x-axis)

1 irradiance (see Note 11). Following phase 1, the irradiance

decreases linearly by 25% and for a duration of 300 ms during
phase 2, leading to a hyperbolic decrease in phase 2 ΦF. The
resultant phase 2 ramp rate is determined from both the phase
2 duration and amplitude of change in irradiance [3]; the ramp
rate was ~0.01 mol photons m2 s2 (see Note 12). Following
phase 2, the irradiance subsequently returns to the phase 1 irra-
diance of 8000 μmol m2 s1 for 300 ms during phase 3 (see
Notes 11 and 13), after which the irradiance near instanta-
neously (i.e., tens of ns timescale) decreases back to the steady-
state illumination of 1000 μmol m2 s1.
 Fm0 from a Single Flash 115

11. Maintaining a constant irradiance at extreme irradiances is

nontrivial. LEDs heat up upon illumination and the spectrum
of the dye in the LED can shift in a way that the LEDs become
less efficient, leading to a decrease in output. Fluorometric
systems possess electronic feedback control for maintaining con-
stant irradiances when necessary (i.e., during phases 1 and 3).
12. The phase 2 ramp rate is important because if the duration is
too short, or the amplitude too large, either scenario of which
could result in a ramp rate greater than 0.01 mol photons
m2 s2 (i.e., a rate faster than the composite rates of the
photo-physical processes controlling the ΦF (see Subheading
1)), then the phase 2 ΦF may not change enough, a circum-
stance that can result in non-optimized estimates of EFm0 (see
Figs. 4, 5, and 6 in [3]).
13. The return of the phase 1 irradiance during phase 3 allows for
determination of violations in assumptions about changes in
ΦF during extreme irradiances, i.e., the ΦF should be identical
during phases 1 and 3; otherwise assumption(s) concerning
proper use of the technique are violated (see Subheading 1 and
Notes 18–22).
14. Manually determine the value of EFm0 . Shown in Fig. 1b is a
plot of the phase 2 ΦF (see Note 10) plotted versus the recip-
rocal of the phase 2 irradiance (Q0 1) (see Subheading 3, steps
3–7), a relationship previously referred to as a reciprocal plot
[3]. Estimate EFm0 by fitting a line to the data using least square
regression and extrapolation to the y-intercept (open circle
along the y-intercept). The estimate of EFm0 was ~8% higher
than that of AFm0 (see Note 10), and indicates that Fm0 is not
attained during the phase 1 irradiance of 8000 μmol m2 s1,
whereas a closer approximation of Fm0 is achieved via MPF
methodology (see Notes 18–22). It is important to emphasize
that as Q0 increases, Q0 1 ! 1, which serves as the basis for
approximating ΦF at infinite irradiance (i.e., EFm0 ) via linear
regression and extrapolation to the y-axis.
15. The time interval allowed to elapse between applications of the
variably intense MPFs is critical for assessment of proper MPF
irradiance dynamics. When testing for proper MPF intensities
on the same leaf, it is critical that no feedback effect (hysteresis)
exists from prior MPFs that could confound determination of
proper MPF irradiation dynamics. In order to avoid this out-
come, a time interval of 2 min should be allowed between each
flash. However, since leaves can behave differently under the
many conditions that may be of interest, the proper interval can
be determined by monitoring Fs over time, i.e., using onboard
graphics, after a flash and ensuring that the chosen time interval
116 Thomas J. Avenson and Aaron J. Saathoff

Fig. 2 Experimental determination of multiphase flash irradiation dynamics from

which to obtain an accurate estimate of Fm0 . (a) Estimates of AFm0 based on both
phase 1 (closed squares) and phase 3 (open squares), as well as the
corresponding estimates of EFm0 (closed circles), are shown as a function of
increasing phase 1 irradiance. Note that phase 1 and 3 intensities of a MPF are
the same (i.e., as shown in Fig. 1a). Estimates of steady-state Fs (i.e., prior to
initiation of the various MPFs) are also shown (open triangles). (b) The maximum
ΦF and irradiances (Q0 ) of the variably intense MPFs are shown. The values of ΦF
and Q0 over the last 60 ms of phases 1 (P1) and 3 (P3) were averaged. The
percent differences (see equation in Subheading 3, step 9) for the ΦF (closed
squares) and Q0 intensities (open squares) are shown on the y-axis as a function
of increasing phase 1 irradiance

enables Fs to return to a steady-state condition prior to appli-

cation of an ensuing MPF.
16. Make a plot of the phase 1 intensity dependence of AFm0 and
E
Fm0 (see Subheading 3, step 8). Figure 2a shows a series of
values of AFm0 , as well as the corresponding estimates of EFm0 ,
based on application of a series (i.e., applied in a randomized
manner) of MPFs ranging in phase 1 intensities between
3000 μmol m2 s1 and 15,000 μmol m2 s1 (see Note 17).
In between phase 1 intensities of 3000 μmol m2 s1 and
 Fm0 from a Single Flash 117

15,000 μmol m2 s1, values of AFm0 that were measured

during phases 1 and 3 hyperbolically increased by 18% and
17%, respectively, but were nonetheless between 22% and 5%
lower in comparison to the estimate of EFm0 based on a phase
1 intensity of 8000 μmol m2 s1. These results indicate that
none of the values of AFm0 accurately approximate Fm0 (see
Notes 20–22 and [3]). In contrast, the respective values of
E
Fm0 only varied by ~1% in between phase 1 intensities of
6000 μmol m2 s1 and 15,000 μmol m2 s1, decreasing
slightly below 6000 μmol m2 s1. These combined results
have important implications as to the technical accuracy with
which Fm0 can be estimated (see Notes 18–22).
17. The values of AFm0 , EFm0 , and Fs have been normalized to the
estimate of EFm0 based on the phase 1 intensity of
8000 μmol m2 s1, allowing fractional differences between
values of AFm0 and EFm0 to be easily discerned.
18. The irradiances and ΦF associated with phases 1 and 3 of a MPF
are comparable to traditional saturation flash dynamics;
assumptions of both methods (see Subheading 1) can thus be
assessed from MPF dynamics. Confidence in accurate determi-
nation of Fm0 by the traditional saturation flash and MPF
approaches is imperative when quantifying Fm0 -derived para-
meters (see Subheading 1). Therefore, as a means of determin-
ing which method to use, assess the validity of the assumptions
of both by scrutinizing the maximum ΦF and irradiance
dynamics (i.e., using raw data; see Subheading 3, step 6) asso-
ciated with the series of variably intense MPFs (see Subheading
3, steps 6–9). For example, the percent differences between
the latter portions (see Subheading 3, step 9) of the phases
1 and 3 ΦF (see Subheading 3, step 8) quasi-exponentially
decreased from 1.7 to 0.92 in between phase 1 intensities
of 3000 μmol m2 s1 and 15,000 μmol m2 s1, passing
through ~0 at a phase 1 intensity of 8000 μmol m2 s1 (see
Subheading 3, step 9, and Fig. 2b, closed squares). These data
indicate between phase 1 intensities of 3000 μmol m2 s1 and
7000 μmol m2 s1 that dΦF/dt (i.e., during phases 1 and 3)
was positive (i.e., the ΦF continually increased during these
phases), implying that assumption 3 (i.e., dqL/dt ¼ 0) of both
methods was invalid at low flash irradiances (see Subheading 1).
The data show between phase 1 intensities of
9000 μmol m2 s1 and 15,000 μmol m2 s1 that dΦF/dt
was negative (i.e., the ΦF continually decreased during phases
1 and 3), implying that assumption(s) 1 and/or 2 of both
methods (see Note 19) were violated over these flash irra-
diances (note: the ΦF dynamics obtained using a phase 1 inten-
sity of 8000 μmol m2 s1 are quite unique and are discussed
below).
118 Thomas J. Avenson and Aaron J. Saathoff

19. While it is possible for irradiation dynamics to change during

such extreme intensities (see Note 11), which could account
for the observed differences in phases 1 and 3 ΦF over the
lower and higher flash irradiances, electronic circuitry of the
fluorometric systems used for these experiments were designed
to maintain the maximum irradiance constant during phases
1 and 3 (see Note 11). The percent differences in phases 1 and
3 maximum irradiances decreased from 0.03 to 0.28 (see
Subheading 3, step 9, and Fig. 2b, open squares). Such data
are consistent with relatively constant irradiances during phases
1 and 3. These combined results have important implications
with regard to whether the two methods were, under the
described conditions, capable of providing technically accurate
estimates of Fm0 (see Notes 20–22).
20. Based on the restricted conditions described herein (see Sub-
heading 3, steps 1–9), technically accurate estimates of Fm0 (see
Subheading 1) are achievable only when using sub-saturating
flash irradiances in conjunction with MPF methodological
dynamics. Assumption 3 of both methods was violated using
phase 1 intensities between 3000 μmol m2 s1 and
7000 μmol m2 s1, whereas assumptions 1 and/or 2 of
both methods were violated using phase 1 intensities between
9000 μmol m2 s1 and 15,000 μmol m2 s1 (see Note 18
and Fig. 2b, closed squares). While the difference between
phases 1 and 3 ΦF was ~0 using a MPF comprised of a phase
1 intensity of 8000 μmol m2 s1 (Fig. 2b), consistent with the
redox condition dqL/dt ¼ 0 being achieved, the corresponding
value of AFm0 was ~2% lower than that based on a phase 1 inten-
sity of 15,000 μmol m2 s1 (Fig. 2a). These results imply that
the redox condition qL ¼ 0 was not achieved using a flash
irradiance of 8000 μmol m2 s1, which technically violates
only the traditional saturation flash method (see Subheading 1),
suggesting that, under these conditions, estimates of Fm0 ought
to be assessed by the MPF methodology (see Note 21).
21. While none of the assumptions of the traditional saturation
flash method were technically valid over the entire range of flash
irradiances (see Note 20), which can ultimately lead to errors in
estimation of Fm0 -derived parameters (see Subheading 1),
all assumptions of the MPF method (i.e., dqL/dt ¼ 0 and
qL ¼6 0) were valid using a phase 1 intensity of 8000 μmol m2 s1
(see Note 18 and Figs. 1 and 2). In addition, based on MPF
methodological dynamics, the resultant estimate of EFm0 at
8000 μmol m2 s1 fell within a range of estimates of EFm0 (i.e.,
between 6000 μmol m2 s1 and 15,000 μmol m2 s1) that
were within a single percent of one another (Fig. 2a); such data
indicate that all values within such a constant range closely
approximate Fm0 (see Fig. 10 in [3]). The combined results
 Fm0 from a Single Flash 119

indicate that technically accurate estimates of Fm0 can only be

achieved when using a sub-saturating flash of irradiance (but see
Note 22), which in and of itself (i.e., 8000 μmol m2 s1) under-
estimated Fm0 by 8% (i.e., compare AFm0 and EFm0 at a phase
1 irradiance of 8000 μmol m2 s1) (see Fig. 2a), in conjunction
with MPF methodological dynamics.
22. While from a technical perspective, the only flash irradiance
under which all MPF assumptions were valid was at a phase
1 intensity of 8000 μmol m2 s1, the fact of the matter is that
estimates of EFm0 were within 1% of one another between phase
1 irradiances of 6000 μmol m2 s1 and 15,000 μmol m2 s1.
Such results are consistent with the notion that relatively accu-
rate estimates of Fm0 can be achieved over a broad range of
sub-saturating MPF irradiances (i.e., there is no need to use
extreme flash irradiances), a circumstance that is very likely
applicable to other conditions (see Note 23).
23. The approach described herein for determination of a suitable
MPF can be broadly applicable. On the one hand, the experi-
ments described herein are a sort of academic exercise aimed at
demonstrating how to properly use the MPF method. On the
other hand, armed with such an understanding of the MPF
approach, one could then apply it to other adaptive states (but
see Note 24), conditions, experiments, etc. For example, one
may have an interest in determining Fm0 -derived parameters
(Subheading 1) in leaves at the tops (i.e., exposed to full
sunlight) of transgenic plants of a crop species during a mid-
summer’s morning when the PAR is ~1000 μmol m2 s1 and
leaf-level photosynthesis is in a steady state. If so, the approach
described herein could be used to properly assess MPF irradia-
tion dynamics from which to obtain accurate estimates of Fm0
and, consequently, Fm0 -derived parameters.
24. Maximum ΦF in dark-adapted leaves, e.g., Fm (see Subheading 1),
is readily achieved using traditional saturation flash methodol-
ogy; use of the MPF method on dark-adapted leaves can cause
errors in estimation of Fm.

References
1. Schreiber U (2004) Pulsed-amplitude-modu-
lation (PAM) fluorometry and saturation pulse
method: an overview. In: Papageorgiou
(ed) Chlorophyll a fluorescence: a signature of
photosynthesis. Springer, Dordrecht
2. Earl HJ, Ennahli S (2004) Estimating photo-
synthetic electron transport via chlorophyll
fluorometry without Photosystem II light sat-
uration. Photosynth Res 82:177–186
3. Loriaux SD et al (2013) Closing in on maxi-
mum yield of chlorophyll fluorescence using a
single multiphase flash of sub-saturating inten-
sity. Plant Cell Environ 36:1755–1770
4. Markgraf T, Berry JA (1990) Measurement of
photochemical and non-photochemical
quenching: correction for trunover of PSII
during steady-state photosynthesis. In:
Baltscheffski M (ed) Current research in pho-
tosynthesis. Kluwer Academic Publishers,
Dordrecht
5. Schatz GH, Brock H, Holzwarth AR (1988)
Kinetic and energetic model for the primary
120 Thomas J. Avenson and Aaron J. Saathoff
processes in photosystem II. Biophys J
54:397–405
6. Kramer DM, Johnson G, Kiirats O, Edwards GE
(2004) New fluorescence parameters for the
determination of QA redox state and excitation
energy fluxes. Photosynth Res 79:209–218
7. Muller P, Li XP, Niyogi KK (2001)
Non-photochemical quenching. A response to
excess light energy. Plant Physiol 125:1558–1566
8. Kramer DM, Crofts AR (1996) Control and
measurement of photosynthetic electron trans-
port in vivo. In: Baker NR (ed) Photosynthesis
and the environment. Kluwer Academic Pub-
lishers, Dordrecht
9. Genty BE, Briantas JM, Baker NR (1989) The
relationship between the quantum yield of
photosynthetic electron transport and photo-
chemical quenching of chlorophyll fluores-
cence. Biochim Biophys Acta 990:87–92
10. Edwards GE, Baker NR (1993) Can CO2
assimilation in maize leaves be predicted accu-
rately from chlorophyll fluorescence analysis?
Photosynth Res 37:89–102
11. Maxwell K, Johnson GN (2000) Chlorophyll
fluorescence-a practical guide. J Exp Bot
51:659–668
12. Kulheim C, Ågren J, Jansson S (2002) Rapid
regulation of light harvesting and plant fitness
in the field. Science 297:91–93
13. Lavergne J, Trissl H-W (1995) Theory of
fluorescence induction in photosystem II:
derivation of analytical expressions in a
model including exciton-radical-pair equilib-
rium and restricted energy transfer between
photosynthetic units. Biophys J
68:2474–2492
14. Long SP, Bernacchi CJ (2003) Gas exchange
measurements, what can they tell us about the
underlying limitations to photosynthesis? Pro-
cedures and sources of error. J Exp Bot
54:2393–2401
15. Long SP, Garcia RL (1996) Measurement of
leaf and canopy photosynthetic CO2 exchange
in the field. J Exp Bot 47:1629–1642
16. Nishio JN, Sun J, Vogelmann TC (1993) Car-
bon fixation gradients across spinach leaves do
not follow internal light gradients. Plant Cell
5:953–961
17. Long SP, Zhu XG, Naidi SL, Ort DR
(2006) Can improvement in photosynthesis
improve crop yields? Plant Cell Environ
29:315–330
 Chapter 8

Chlorophyll Fluorescence Imaging

Tracy Lawson and Silvere Vialet-Chabrand

Abstract
Chlorophyll fluorescence imaging provides a noninvasive rapid screen to assess the physiological status of a
number of leaves or plants simultaneously. Although there are no standard protocols for chlorophyll
fluorescence imaging, here we provide an example of routines for some of the typical measurements.

Key words Chlorophyll fluorescence, Imaging, Photosynthetic efficiency, Phenotypic screening,

Plant stress

1 Introduction

Chlorophyll fluorescence is a useful noninvasive tool to explore

plant performance and, with the development of imaging systems
[1], can provide a high-throughput diagnostic screening tool for
evaluating the physiological status of whole plants or individual
leaves. Imaging also allows users to utilize the pixel area/number
to measure plant/leaf area (see Note 1). However, the main advan-
tage of imaging platforms is the ability to measure multiple plant
materials simultaneously and to assess the spatial heterogeneity
within plants or leaves (Fig. 1). Chlorophyll fluorescence imaging
systems are available at a range of resolutions providing imaging
capabilities for screening trays of plants [2, 3] to microscope sys-
tems that enable measurements to be made in individual cells [4, 5]
or even single chloroplasts [6].
Chlorophyll fluorescence provides a measure of how well plants
use the light energy absorbed at photosystem II (PSII) for electron
transport and is therefore a key measure of photosynthetic effi-
ciency and performance. However, because photosynthesis is linked
to many plant processes, alterations in metabolic pathways not
directly involved in photosynthesis will also influence the chloro-
phyll fluorescence signal [2, 7]. This sensitivity of chlorophyll
fluorescence to perturbations in metabolism means that measuring
fluorescence provides a rapid, nondestructive, and noninvasive

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_8, © Springer Science+Business Media, LLC, part of Springer Nature 2018

121
122 Tracy Lawson and Silvere Vialet-Chabrand

Fig. 1 Model of the possible fate of light energy absorbed by photosystem II (PSII) and the fluorescence
parameters associated with assessing these different fates. Light energy absorbed by chlorophylls associated
with PSII can be used to drive photochemistry in which an electron (e
) is transferred from the reaction center
chlorophyll of PSII (P680) to the primary quinone acceptor QA, or can be lost from PSII as chlorophyll
fluorescence and heat. The processes of photochemistry, chlorophyll fluorescence, and heat loss are in direct
competition for excitation energy and therefore changing the rate constant for one process will impact the
rates of the other possible fates (following [7]). Fo, Fm: minimum and maximum chlorophyll fluorescence under
dark conditions; F0 , Fm0 : minimum and maximum chlorophyll fluorescence under light conditions; Fv/Fm:
maximum quantum efficiency of PSII photochemistry; Fq0 /Fm0 : operating efficiency of PSII photochemistry

screening tool for plant performance and/or the ability to cope

with stress. How chlorophyll fluorescence can be used to evaluate
the extent and mechanisms of inhibition of metabolic processes
depends on the protocol employed and the user’s overall aim.
There are many examples of how chlorophyll fluorescence imaging
has been used to determine the effect of different stresses or treat-
ments including early detection of herbicide application [2, 8],
nutrient deficiency [9], drought stress [10], photorespiratory
mutants [3], improved photosynthesis [7, 11–14], freezing toler-
ance [15], insect herbivory [16], leaf fungal infection [17–19], and
ozone damage [20, 21]. In this chapter, we introduce the reader to
the most widely used chlorophyll fluorescence imaging parameters
and how these can be used to assess photosynthetic performance,
along with methodological protocols and possible pitfalls for cap-
turing these images. For an in-depth review on the principles of
chlorophyll fluorescence we refer readers to the following reviews
[7, 22–26].
 Chlorophyll Fluorescence 123

1.1 Basic Principles Light energy absorbed by chlorophyll within a leaf has one of the
of Chlorophyll three fates. It is (1) used to drive photosynthesis (photochemistry);
Fluorescence (2) reemitted and lost as heat (non-photochemical quenching
(NPQ)); or (3) reemitted at a slightly longer light wavelength
(chlorophyll fluorescence). These three processes compete with
each other and consequently a change in the rate constant of one
will be reflected in the other two (Fig. 1). Therefore by measuring
the yield of chlorophyll fluorescence, information on the quantum
efficiency of photochemistry and heat dissipation can be obtained,
providing a noninvasive indirect measure of plant photosynthetic
performance [24]. Changes in chlorophyll fluorescence reflect
changes in the redox state of the electron carriers and, as this can
be influenced by both photochemical and non-photochemical
quenching processes, chlorophyll fluorescence measurements are
carried out in a “known state.” To explain the principles behind
chlorophyll fluorescence we will first describe changes in chloro-
phyll fluorescence following a dark-light transition initially
observed and described in [27] (see Fig. 2). When a dark-adapted
leaf is transferred into light, the yield of chlorophyll fluorescence

Fig. 2 Example of a typical fluorescence trace recorded during an induction and relaxation protocol (see
Subheading 3.5). Each dot represents an average chlorophyll fluorescence value over the area of interest
represented by a whole rosette of an Arabidopsis thaliana plant. ML measuring light, SP saturating pulse, AL
actinic light. The black bars represent the light intensity in arbitrary units for the different light sources (ML,
SP, AL) and describe the different light regimes used to monitor chlorophyll fluorescence
124 Tracy Lawson and Silvere Vialet-Chabrand

increases for about 1 s. This rise in fluorescence is due to the

reduction of electron acceptors in the photosynthetic electron
transport system downstream of PSII. Light energy captured by
the antennae is funneled to the PSII reaction center chlorophyll
(P680) resulting in the transfer of an electron to the primary
quinone acceptor of PSII (QA). QA is not able to accept another
electron until it has passed this first electron onto the next carrier
(QB) and at this point the center is termed “closed.” An “open”
center is one in which QA is fully oxidized and capable of capturing
an electron. Following the initial rise in fluorescence after transfer
to light, the chlorophyll fluorescence signal decreases over a few
minutes, and this is termed “quenching” of fluorescence (Fig. 2)
[22]. The quenching of the fluorescence signal is driven by a
number of processes, including the light activation of photosynthe-
sis, which includes full activation of Calvin cycle enzymes (for
example RuBisCO) [28], which can take several minutes, as well
as stomatal opening to increase CO2 availability for photosynthesis
[29]. All of these processes increase the availability of sinks for
electrons derived from light-driven electron transport in the thyla-
koid membrane and contribute to photosynthesis and are therefore
termed photochemical quenching. Another process that aids in the
quenching of fluorescence is an increase in the dissipation of exci-
tation energy as heat, which is termed non-photochemical quench-
ing (NPQ). Excess excitation energy is dissipated as heat within the
chlorophyll antennae and is a protective mechanism to reduce the
possibility of the formation of damaging free radicals which can
damage the photosystems referred to as photoinhibition
[30]. NPQ is a complex mechanism and is driven by a number of
processes including the development of a ΔpH driven by the acidi-
fication of the thylakoid lumen during linear and cyclic electron
flow [31–33] and the conversion of the carotenoid violaxanthin to
zeaxanthin as part of the xanthophyll cycle [34–36]. The formation
of zeaxanthin and the regulatory protein PsbS result in conforma-
tional changes in PSII antennae which causes the quenching of
excitation energy in the antennae of PSII [33]. For further details
on NPQ and the components that make up this process we refer
readers to [37].

1.2 Measuring Here we briefly describe how to measure several of the most com-
Chlorophyll monly used chlorophyll fluorescence parameters and follow the
Fluorescence: Most nomenclature and abbreviations defined in [7]. We first describe
Commonly Used measurements made on dark-adapted material, followed by mea-
Parameters surements carried out on plant material in the light-adapted state.
The maximum quantum efficiency of PSII photochemistry Fv/Fm
provides a measure of the maximum efficiency at which light
absorbed by PSII is used for the reduction of QA. This value in
healthy non-stressed dark-adapted leaves is conserved across plant
species with values of ~0.84 [38]. A value less than this provides a
 Chlorophyll Fluorescence 125

good indication of damage to the photosynthetic apparatus or

photoinhibition and is routinely used as a measure of plant health.
The measurement of Fv/Fm is determined on a dark-adapted leaf
(see below for details on dark adaptation) to ensure that any NPQ
previously present in the leaf has relaxed and QA is maximally
oxidized and ready to accept electrons (“open center”). Minimal
fluorescence (Fo) can be determined by a weak measuring light
beam (typically <1 μmol m
2 s
1). Following the measurements
of Fo, if the sample is exposed to a “saturating flash” of light of
several thousand (>4000–10,000) μmol m
2 s
1 for several
100 ms (between 600 and 1000 ms depend on the species), then
QA is transiently fully reduced and the centers termed “closed”
resulting in maximum fluorescence (Fm) (Fig. 2). The measure-
ments of Fm and Fo can be used to determine variable fluorescence
(Fv) by the following equation Fv ¼ Fm
Fo which can subse-
quently be used to calculate the maximum quantum efficiency of
PSII, Fv/Fm.
In the light-adapted state a number of reaction centers will be
in the closed state, the proportion of which depends on the inten-
sity of actinic illumination (defined as absorbed light that will drive
photosynthesis; [7]). Steady-state fluorescence in the light is
termed F0 and following a saturating pulse of light as described
above maximum fluorescence in the light (Fm0 ) is determined. Fm0 is
usually lower than the dark-adapted Fm due to NPQ present in the
light-adapted plant material. The difference between Fm0 and F0 is
termed Fq0 and is used to calculate the most useful fluorescence
parameter: the operating efficiency of PSII photochemistry (Fq0 /
Fm0 ; [39]). This parameter is also termed ϕPSII or ΔF/Fm0
[7, 24]. It provides a measure of the proportion of absorbed light
that is actually used in PSII photochemistry [40] and can therefore
be used to estimate the rate of electron transport (ETR) through
PSII using Eq. (1), provided that the following information is
available: light intensity (PPFD), light absorption by the leaf
(Aleaf), and proportion of the light energy that is absorbed by
PSII (fractionPSII):
ETR ðμmol m
2 s
1 Þ ¼ F q 0 =F m 0  A leaf  f ractionPSII  PPFD ð1Þ

Leaf absorbance is often assumed to be ca. 0.85 but can be

influenced by a number of factors (see Note 2) and the proportion
of light energy used by PSII can be assumed to be 50%. The fact
that Fq0 /Fm0 provides a measure of photosynthetic efficiency directly
related to ETR and that a dark adaption measurement is not
required for the calculation make it an attractive, rapid, and popular
measurement (see Notes 3 and 4). Fq0 /Fm0 can be broken down into
two products (Fig. 1), the level of photochemical quenching of
PSII (Fq0 /Fv0 ), which provides a measure of the proportion of PSII
centers that are “open” (QA oxidized) and is mathematically
126 Tracy Lawson and Silvere Vialet-Chabrand

identical to the coefficient of photochemical quenching (qP; [7]),

and the maximum efficiency of PSII in the light (Fv0 /Fm0 ), which
provides an estimate of the maximum efficiency of PSII photo-
chemistry at the given light intensity if all the PSII centers were
“open,” with any decreases in Fv0 /Fm0 reflecting an increase in NPQ
[1, 7]:

F q 0 =F m 0 ¼ F q 0 =F v 0  F v 0 =F m 0 ð2Þ

However, in order to determine these quenching parameters, a

measure of minimal fluorescence in the light (Fo0 ) is required. Fo0 in
conventional fiber optic-based fluorimetry is measured prior to the
saturating pulse used to determine Fm0 and following the removal
of actinic light and the application of a far-red pulse (see Fig. 2).
Far-red light is preferentially absorbed by photosystem I (PSI)
relative to PSII, which draws electrons through the electron trans-
port chain away from QA, resulting in oxidized QA and open PSII
centers. The application of a far-red light pulse is particularly chal-
lenging in imaging systems due to the complication of including
different LEDs into the lighting arrays and the requirement for
homogeneity in these light wavelengths (710–850 nm). Therefore
imaging systems often employ a mathematical calculation of Fo0
that was developed by [41] for the purpose of chlorophyll fluores-
cence imaging:

Fo
F o0 ¼    0 ð3Þ
Fv
Fm þ FFmo 0

However as you will note from the above equation a dark-

adapted measurement of Fo is required for this calculation along
with the various assumptions that go with the calculation of this
parameter (see Note 5). Additionally, the process of taking images
on dark-adapted material has other challenges for imaging systems
(see Subheading 2 and Note 6).

2 Materials

To carry out chlorophyll fluorescence imaging requires the pur-

chase and setup (per the manufacturer’s instructions) of one of the
off-the-shelf imaging fluorimeters commercially available.
Although it is possible to build a custom system this requires
extensive knowledge and experience in electronics, computer pro-
gramming, and an in-depth understanding of the biophysics
behind the chlorophyll fluorescence signal. A key requirement of
any chlorophyll fluorescence imaging system is homogeneous
actinic light intensity over the entire imaging area as well as
the ability to provide a saturating pulse over the entire area.
 Chlorophyll Fluorescence 127

A heterogeneous light intensity over the measuring area will result

in uncontrolled spatial variations of the fluorescence signals
measured and parameters determined from these signals. The
intensity of the saturating pulse does not necessarily have to be
uniform but it does have to be saturating over the entire area being
imaged, both of which can be difficult to achieve (see Note 7).
Due to minimal overlap in spectrum with chlorophyll fluores-
cence but similar ability to drive photosynthesis as red light, blue
light LEDs (λ of ca. 470 nm) are generally used for the measuring
beam and the actinic light, and changes in chlorophyll fluorescence
are collected by a black-and-white camera equipped with a specific
band-pass or long-pass filter (e.g., λ > 660 nm). One key point to
note about imaging systems is that different fluorescence filters will
alter the amount and depth from which the chlorophyll fluores-
cence will be detected. Long-wavelength long-pass filters will gen-
erally collect fluorescence from deeper in the leaf compared with
narrow shorter wavelength band-pass filters (e.g., 680  10 nm)
which are generally preferred when the user requires leaf surface
measurements (see [5]).
Historically there have been two main types of instruments
used for chlorophyll fluorescence imaging, pulse-amplitude modu-
lation (PAM), which employs modulated illumination, and contin-
uous illumination systems.
Both produce similar results but the PAM method does not
require complex calibration and can be less sensitive to small differ-
ences of chlorophyll fluorescence. Here we describe the typically
used PAM imaging.
The PAM method requires a high frequency of image capture,
which is inversely proportional to the image resolution and integra-
tion time, and was more difficult to achieve with earlier camera
models. For example, [42] used two different light sources to
provide (measuring light) ML and (actinic light) AL. With this
dual-light source, building a fluorescence image requires a picture
to be captured under actinic light and subtracted from an image
captured during the measuring light, lowering the frequency to
produce fluorescence images and at the same time reducing the
dynamic range of these outputs. Currently, typical PAM fluorimetry
can employ up to four light sources [42–44], each of which enables
measurement of a different physiological state:
(a) A measuring light (ML), provided by light-emitting diodes
(LED), uses short pulses (1–3 μs) of a red light with a very low
integral photon flux density (PFD < 0.1 μmol m
2 s
1). The
intensity, frequency, and duration of the pulses influence the
integral photon flux density, and generally there is a trade-off
between intensity of the fluorescence signal and noise of
the signal due to the sampling rate (frequency at which
pictures/measurements are captured). This trade-off needs
128 Tracy Lawson and Silvere Vialet-Chabrand

to be set to achieve the highest signal quality without driving

photosynthesis (see Note 8). The red radiation with a peak
wavelength λ of 650 nm is passed through a short-pass filter
(λ < 670 nm) and is absorbed by the leaf. The emitted chlo-
rophyll fluorescence is detected by a camera protected by long-
pass (λ > 700 nm) and heat filters. The measuring light
enables measurement of steady-state fluorescence (F0 ) in the
light and minimal fluorescence in the dark (Fo).
(b) Actinic light (AL) generates a continuous photon flux of up to
hundreds of μmol m
2 s
1 that triggers primary photosyn-
thetic processes. Changes in the chlorophyll fluorescence sig-
nal related to the variation of PSII efficiency can be detected as
the amplitude modulation of a pulsing signal induced by the
ML (Fig. 2).
(c) A saturating pulse (SP) is achieved by a short square flash
(0.3–1.2 s) of saturating light intensity (>7000 μmol m
2 s
1)
that temporarily converts all reaction centers of PSII to the
reduced (or closed) state (see Note 9). SP also fully reduces all
electron carriers in the plastoquinone (PQ) pool. During SP,
the frequency of ML is automatically increased in order to
achieve a better signal-to-noise ratio and time resolution
(Fig. 2).
(d) A far-red pulse (FR) is a flash of far-red light (5–8 s) with λ of
735 nm and PFD of approx. 10 μmol m
2 s
1 used to pro-
mote PSI activity resulting in a rapid reoxidation of the PQ
pool. It is used to determine minimal fluorescence in the light-
(Fo0 ) and dark-adapted (Fo) state. Recently the differences
between Fo0 and Fo have been used to infer the level of photo-
inhibition of the photosystems [33] (see Note 5).

3 Methods

Although numerous custom-built systems have been described in

the literature (reviewed by [1, 42]), only a few companies sell
systems off the shelf and ready to use, and each provides solutions
with different specifications such as absorbance measurements and
single turnover pulse. Unlike many other plant physiological tech-
niques there are no standard or formal protocols for chlorophyll
fluorescence imaging and the procedure very much depends on the
user. Below we provide some examples for some of the commonly
used approaches; however the timing and light intensities depend
on the questions being addressed as well as species, growth condi-
tions, and treatments. Many of the protocols described below
have been used for determining stress as well as alterations in
photosynthesis. Additionally, it is worth bearing in mind that com-
bining these measurements with other techniques such as infrared
 Chlorophyll Fluorescence 129

gas exchange analysis can provide further information and the

possibility for mechanistic approaches. For example, combined
chlorophyll fluorescence and infrared gas exchange approaches
have been utilized to examine lateral gas fluxes in leaves [45, 46]
and guard cell photosynthesis [4, 5]. As mentioned above chloro-
phyll fluorescence protocols are user specific and generally devel-
oped depending on the questions being addressed, as well as
consideration of the number of samples to be measured/screened
and the timeframe available to conduct the measurements. Below
we provide a few examples of protocols used to evaluate plant
photosynthetic performance; however, it is important to keep in
mind that the environmental conditions and variations in light
intensity will influence the leaf temperature and stomatal conduc-
tance, which will ultimately influence the chlorophyll fluorescence
signal (see Notes 10 and 11). Before we describe specific protocols,
we will briefly outline the typical procedures for setting up a chlo-
rophyll fluorescence imaging system. In particular we will outline
the procedures based on the chlorophyll fluorescence (CF) imager
supplied by Technologica Ltd. (Essex), which is used in our labora-
tory, but the methodology can be generally applied.

3.1 Setting Up 1. Place plant material in the imaging system (with the appropri-
a Chlorophyll ate dark or light adaption depending on the user measurement
Fluorescence Imaging protocol) at the correct height (see Note 12) or the specific
System distance from the camera (as outlined by the manufacturer).
Ensure that the actinic light intensity has been calibrated (see
Note 13) and that there is even illumination over the imaging
area (see Note 7).
2. Focus the camera to ensure that the image is as sharp as possible
and make appropriate aperture adjustments as outlined by the
manufacturer (see Note 14). The amount and how aperture
adjustments are made to optimize the fluorescence signal differ
depending on the instrumentation.
3. After setting up the camera, a map image can be recorded and
used to subtract the background or regions not of interest to
the user. Several systems enable the map image or region of
interest to be defined post-image capture, while others have the
ability to define this when the plant material is initially placed in
the imaging system (see Notes 15 and 16).
4. Wait for the fluorescence signal to stabilize before initiating the
protocol (see Note 17).

3.2 Dark-Adapted These measurements are useful for evaluating treatments or

Measurements environmental effects on the maximum quantum efficiency,
of Fv/Fm providing an indication of photoinhibition and plant stress.
1. Dark-adapt plant material for a minimum of 20 min by placing
the plant in the dark or measuring the plant prior to
130 Tracy Lawson and Silvere Vialet-Chabrand

illumination. Place the plant/leaf in the chlorophyll fluores-

cence imaging system at the correct height, maintaining the
leaf in darkness for the entire measurement of Fv/Fm. Follow
the setup protocol described in Subheading 3.1.
2. Switch on the measuring beam and wait until Fo is stable
(usually only 1–2 min) and measure Fo.
3. Apply saturating pulse, typically 0.8 s, at an intensity of at
least 4000 μmol m
2 s
1 to obtain an image of Fm and an
image of Fv/Fm.
4. The value of Fv/Fm should be between 0.8 and 0.84 (see Note
18). It should be noted however that values obtained from
imaging systems tend to be a little lower than those obtained
from typical PAM instruments that use a fiber optic to capture
the fluorescence signal. This is due to spectral differences in the
actinic light and measuring beam as well as the selection of
emission wavelengths.

3.3 Procedure A light–response curve is a commonly used approach to distinguish

for Conducting a Light differences in photosynthetic capacity that can be further examined
Intensity–Response for differences in photochemical and non-photochemical processes.
Curve Along with information on light absorption and partitioning of
energy to both PSII and PSI, these data can be used to produce
ETR as a function of light, which directly relates to the rate of
carbon assimilation (see Notes 3). An example of a light–response
curve is shown in Fig. 3.
1. After setting up the fluorescence imager, place dark-adapted
plant material in the imager and wait for the fluorescence
signal to stabilize. Follow the setup protocol described in
Subheading 3.1.
2. Initiate a saturating pulse to record Fv/Fm.
3. Following the measurement of Fv/Fm (see Subheading 3.2)
subject the leaf/plant to a light intensity similar to that under
which the plant was grown and leave for a period of time in
order to activate photochemical enzymes and open stomata. To
determine how long is required for this, apply a saturating
pulse every 3–4 min to determine Fq0 /Fm0 and determine the
time it takes for this value to become stable. The time required
for this depends on species and growth conditions but can take
up to 45 min.
4. After reaching a plateau in Fq0 /Fm0 at growth light intensity the
light–response protocol can be initiated. Change the actinic
light intensity using the following series: 50, 100, 150, 200,
300, 400, 500, 700, 900, 1100, 1300, and 400. At each light
level allow 2–3 min before applying a saturating pulse and
record Fq0 /Fm0 .
 Chlorophyll Fluorescence 131

Fig. 3 The effect of cold on chlorophyll fluorescence induction and relaxation

curve and light–response curve—an example of a combined protocol. The first
grey shading illustrates measurements taken for a light induction following a
step change in light from darkness to 600 μmol m
2 s
1 and the second shaded
area illustrates the capture of relaxation from 2000 μmol m
2 s
1 to darkness.
In the period represented by the white area, a light–response curve was
measured from 0 to 2000 μmol m
2 s
1. Measurements from control plants
are represented by black squares while cold-treated plants are represented by
white circles

3.4 Rapid Rapid measurements of photosynthetic efficiency (Fq0 /Fm0 ) provide

Measurements a high-throughput method for screening differences in plant
of Fq0 /Fm0 metabolism. These measurements are useful to explore differences
in photosynthetic efficiency and have been used to screen trans-
genic plants with differences in photosynthetic rates (e.g.,
[14, 47]). Protocols often employ measurements at two light
intensities, usually growth light intensity and a higher intensity.
The reason for taking a measurement at the higher light intensity
is to ensure that photosynthesis is pushed toward higher levels, in
case differences in capacity can only be distinguished when photo-
synthesis is maximized (see Note 19).
1. Adapt plant material to the light level that the initial measure-
ments will be taken at, which is often the growth light level.
2. Set up the fluorescence imaging system, as described above.
3. Place the plant material into the imager directly from where it
has been acclimated (e.g., from the growth room) and set the
light intensity in the imager to the first light intensity and wait
for the fluorescence signal to stabilize which is usually less than
1 min.
4. Initiate a saturating pulse, which in most systems will record
both F0 and Fm0 and produce a calculated image of Fq0 /Fm0 .
132 Tracy Lawson and Silvere Vialet-Chabrand

5. Increase the light intensity to a high value. This should typically

be the intensity that is saturating for photosynthesis but not
inhibitory or a sub-saturating value but one which will greatly
increase the rate of photosynthesis relative to the first intensity
(see Note 20). Leave this to stabilize for 2–5 min and apply a
saturating pulse to determine Fq0 /Fm0 at the higher light level.
The values obtained at the two different light levels can be used
to directly assess differences between species or treatments by
determining any significant differences or the difference
between the two measurements assessed.
6. Evaluation of the Fq0 /Fm0 values recorded at the two different
light intensities can be used to determine differences between
treatments or mutant/wild type or can be compared within the
treatment to assess mechanistic differences between the differ-
ent plant materials.

3.5 Procedure for an To assess mechanistic differences between plants, differences in the
Induction/Relaxation light activation of biochemical processes driving photosynthesis
of the Photochemistry (and photochemical quenching) and NPQ are often assessed. Vari-
Using Chlorophyll ation of chlorophyll fluorescence parameters provides the user with
Fluorescence information about rate of induction of photosynthesis and NPQ as
well as the steady-state values of each of these two parameters. This
protocol can be a useful tool to assess plant material with alterations
in photosynthesis, enzyme kinetics, stomatal conductance, or NPQ
that would influence the rate of induction of photosynthesis. Addi-
tionally the rate of relaxation of NPQ following a period of illumi-
nation can be a useful tool to assess differences in this process
contributing to quenching in different plant material as well as
distinguishing quenching related to heat dissipation in the antenna
and that due to photodamage.
1. After setting up the fluorescence imager, place a dark-adapted
plant in the imager and wait for the fluorescence signal to
stabilize. Follow the setup protocol described in Subheading
3.1. Initiate a saturating pulse to record Fv/Fm.
2. Following the measurement of Fv/Fm subject the plant to the
desired light intensity. This is often similar to plant growth light
intensity or higher (see Note 21) and apply a saturating pulse
every 2–3 min to determine Fq0 /Fm0 . Repeat this until Fq0 /Fm0
becomes stable or for a set period of time (see Note 22).
3. The above induction curve can be followed by a relaxation
curve where the light is switched off and the plant is returned
to the maximum quantum efficiency measured in dark-adapted
material (Fig. 3). This allows the contribution of NPQ to
photosynthetic efficiency to be assessed in different plant mate-
rial (see Note 23). A saturating pulse is applied every 2–3 min
until Fv/Fm has returned to the initial value recorded from
dark-adapted material.
 Chlorophyll Fluorescence 133

3.6 Combined 1. Above we have described some of the typical protocols that
Protocols users often employ as common chlorophyll fluorescence tools
to assess plant performance; however, it is worth pointing out
that there are no standard protocols and users can combine any
of the individual measurement or protocols described above in
any order or repetition required. Figure 3 provides an example
of a protocol that combines light induction and relaxation
kinetics as well as photosynthetic efficiency as a function of
light (all described as individual protocols above) to assess the
effect of low temperature on plant physiological status). The
initial part of the curve represents the light induction, and the
middle section the light–response curve followed by the relax-
ation kinetics.
2. It is clear from these data that the cold treatment reduced Fq0 /
Fm0 at all light intensities as well as the rate of light induction;
however no effect on dark relaxation was observed.
3. From this protocol we can examine images of Fq0 /Fm0 and the
corresponding parameters that make up this efficiency, Fv0 /Fm0
and Fq0 /Fv0, from the light induction kinetics (Fig. 4).
4. These images provide a visual illustration of the increase in Fq0 /
Fm0 with time following the application of light, but clearly
illustrate that, although the efficiency of PSII photochemistry
is generally uniform over the entire plant during the induction,
this is driven by spatial heterogeneity in photochemical (Fq0 /
Fv0 ) and non-photochemical (Fv0 /Fm0 ) quenching processes.
5. Higher non-photochemical quenching values are initially
observed in the older leaves, while the younger leaves had
higher photochemical efficiency after 8 min in the light (see
Note 24).
6. This illustrates the power of imaging for not only quantifying
heterogeneity within the plant but also the processes that
determine this. The same approach can be applied to multiple
plants for screening processes.

4 Notes

1. The pixel number of an image can be used to determine the

area of the image (plant or leaf area) using the appropriate scale.
Images with the greatest pixel intensity (e.g., Fm) ensure that
no blank pixels artificially reduce the measurement of the area.
2. Chloroplast movements in the mesophyll cells introduce varia-
tion in leaf absorbance that influences the values of Fo, Fo0 Fm,
and Fm0 between dark and light conditions, which could result
in inaccurate parameters that are determined using both light
and dark measurements (e.g., NPQ).
134 Tracy Lawson and Silvere Vialet-Chabrand

Fig. 4 Chlorophyll fluorescence images of Fq0 /Fm0 , Fv0 /Fm0 , and Fq0 /Fv0 at 2, 8, and 20 min of a light induction
curve (see Fig. 3). Leaves were dark-adapted for 20 min and Fv/Fm recorded, after which light was increased
to 600 μmol m
2 s
1 and Fq0 /Fm0 measured every 2 min for 30 min from which Fv0 /Fm0 and Fq0 /Fv0 were
determined. The color bar represents the range of values for each parameter. All images were adjusted to fit
the same scale. The false color scale represents the variation in each chlorophyll fluorescence parameter over
the image. To evaluate differences between images there are a number of options: (1) compare whole image
average values between treatment/plants, (2) compare selected areas of interest within images (e.g., young
and old leaves), and (3) use and evaluate histogram distribution of the range of values from the entire image
taking into account the entire spatial variation of the image

3. The operating efficiency of PSII photochemistry and electron

transport closely correlates with photosynthetic CO2 assimila-
tion; however, this relationship is only linear when photorespi-
ration is removed by performing measurements under low
[O2]. This is because the end products of electron transport
can be used in photorespiration to fix O2 which will act as a sink
for electrons and maintain the efficiency of PSII.
4. As photorespiration acts as a sink for the end products of
electron transport, the influence of stomatal behavior on pho-
tosynthetic processes (including those often observed when
water availability is reduced) may not be apparent when using
chlorophyll fluorescence as a measure of photosynthesis.
 Chlorophyll Fluorescence 135

A decrease in stomatal conductance can often limit assimilation

rate (as measured by gas exchange) but has little effect on
photosynthetic efficiency because photorespiration maintains
photochemical quenching. It is therefore important that the
user understands the effect that a treatment or condition may
have on the photosynthetic processes before designing the
protocol. Chlorophyll fluorescence imaging can be a useful
tool to detect patchy stomatal behavior, but this can only be
visualized under lower O2 preventing photorespiration acting
as an alternative sink for the products of electron transport.
5. The calculation of Fo using Eq. (3) is only valid based on the
following assumptions: (1) all PSII centers are open when Fo is
measured; (2) there is no reversal in downregulation between
the measurement of Fo and Fm; and (3) there is no reversal of
photoinhibition between the measurement of Fm0 and Fm (see
[1, 4] for further information). Based on the assumption that
the calculated Fo is affected by the previously cited processes,
0
Fo and Fo can be used to assess photoinhibition [33].
6. Fo images are generally much noisier due to the low light
intensity and long integration time required to collect the
chlorophyll fluorescence signal, which will influence the quality
of the picture generated for parameters such as NPQ. Ensuring
the highest possible camera aperture and highest sampling
frequency (without driving photosynthesis; see Note 8) aids
in reducing the noise.
7. To ensure homogenous illumination and a saturating pulse
over the entire imaging area a set of orientable LED bricks
can be used to focus illumination on a measuring area
providing homogenous light and reaching a light intensity
high enough for a saturating pulse.
8. In order to determine the frequency of sampling chlorophyll
fluorescence, the user usually uses a dark-adapted plant and
increases the measuring beam Hz until there is no change in
the fluorescence signal. Any change in the Fo signal implies that
the measuring beam is actinic.
9. Saturating pulse light intensities of 6000–8000 μmol m
2 s
1
may be insufficient for plants grown under high-light condi-
tions such as those encountered in the field. Sub-saturating
pulses produce a lower value of Fm or Fm0 making comparison
between treatments or species under these conditions difficult.
Conventional PAM systems have recently introduced a new
feature allowing the estimation of Fm from sub-saturating
pulses [48] that could be adapted to imaging systems. The
main challenge to adapting this approach to imaging systems
is the high frequency of image capture required by the tech-
nique and the increased cost of image processing.
136 Tracy Lawson and Silvere Vialet-Chabrand

10. Measurements should be conducted at a constant room tem-

perature, as any variations will influence the biochemistry and
ultimately the chlorophyll fluorescence signal.
11. Humidity in the room should be around 50–60% at 22  C
(or similar to growing conditions) to maintain open stomata
and not stress the plant. If stomata are closed, then photo-
chemistry will be rapidly limited by CO2 under an increasing
light intensity and differences in chlorophyll fluorescence sig-
nals could be due to the difference of stomatal opening
between plants.
12. If the sample is not positioned at the same height at which the
illumination was calibrated, then the light intensity received
can be under- or overestimated, which in the case of a light
intensity–response curve will shift all the points on the x-axis
resulting in incorrect measurements. If the height is different
between each measurement (typically plants with different
canopy heights), then a small error in the height (1–2 cm)
will not be important as long as all the samples are measured
in the same conditions.
13. LEDs can lose efficiency with age and it is important to recali-
brate the light intensity at the measuring distance every month
to account for this effect. To check whether the actinic light has
been calibrated correctly, place a light sensor at the same height
a sample would be placed at and measure the illumination at a
range of selected light levels (including both high and lower
levels). The user-defined actinic light intensity should be iden-
tical to that measured on the light sensor.
14. The camera focus determines the spatial precision of chloro-
phyll fluorescence measurements and should be as sharp as
possible. The camera aperture determines the dynamic range
of the signal maintaining a high enough chlorophyll fluores-
cence signal under low light intensity and at the same time
preventing signal saturation under high light intensity. This is
generally achieved by following the calibration procedure
described in the supplied software.
15. When imaging, it is important to remove the background and
maintain a region or regions of interest. Measurements taken
over a long period of time can be influenced by leaf movements
which can interfere with chlorophyll fluorescence measure-
ments if the mask used to calculate chlorophyll fluorescence
parameter pixel by pixel is not updated regularly. The calcu-
lated chlorophyll fluorescence parameter can be artificially
decreased if the images from which it had been derived
included background elements.
16. Image software and the ease with which it can be implemented
and used to isolate regions of interest and extract data as well as
 Chlorophyll Fluorescence 137

having the ability to input user-defined protocols should be a

significant consideration before purchasing an imaging
platform.
17. Stabilization should only take 2–5 min but will depend on the
species and pretreatment. Stabilization is achieved when the
signal does not change for 30 s or so.
18. A value of Fv/Fm lower than 0.8 could be a sign that the
intensity of the measuring light (ML) is too high and is driving
photosynthesis, or a plant that has not been fully dark-adapted
or has been stressed and damaged. Under a non-actinic ML
and using a healthy plant, Fo should rapidly reach a plateau
(in <1 min) and remain stable until the conditions are
changed. Driving photosynthesis by the ML results in inaccu-
rate measurements of photosynthetic efficiency.
19. Difference in photosynthetic efficiency between species, treat-
ment, and transgenic material is often not detected unless
photosynthesis and electron flow through the photosystems
are “pushed” toward a maximum. This is because performance
can often be maintained at lower rates (e.g., light levels) and it
is only when measurements are performed at saturating light
and the system pushed toward a maximum that deviation in
photosynthetic efficiency is realized. An example could be
reduced stomatal conductance due to water stress—photosyn-
thesis and therefore photosynthetic efficiency could be main-
tained at lower light levels and the effect of reduced stomatal
conductance only apparent at saturating light and when limited
diffusion of CO2 into the plant interior lowers the assimilation
rate and photosynthetic efficiency.
20. The light intensity used to achieve a saturating photosynthetic
rate and push the system toward maximum photosynthesis
must be determined by the user. This could be achieved from
a prior light–response curve or from knowledge of the species
under investigation. Alternatively an educated guess can be
used with knowledge of the growth conditions and plant
type. For example, Arabidopsis is often grown under con-
trolled environmental conditions at a light intensity of
ca. 150–200 μmol m
2 s
1 and will often be saturated with
light intensities between 500 and 1000 μmol m
2 s
1, while
wheat in the field will often experience light levels of 1000–-
1200 μmol m
2 s
1 and therefore will need 2000 μmol m
2 s
1
to saturate photosynthesis. Light of too high an intensity can
result in damage and photoinhibition that will reduce the
measure of Fq0 /Fm0 .
21. The light intensity used for induction curves is defined by the
user. Two commonly used intensities are the average growth
light intensity or the light intensity at which photosynthetic
138 Tracy Lawson and Silvere Vialet-Chabrand

rate is saturated. Information from a light–response curve can

often be useful in deciding on this light level.
22. The time for stabilization depends on the species, pretreat-
ment, and light intensity used for the induction. However,
this is often in the period of 20–45 min.
23. Relaxation is usually complete within a maximum period of
45 min, but will depend on the species, pretreatment, and
whether there has been any damage to the leaf.
24. Many imaging systems produce an automatically scaled image
that uses the minimum and maximum fluorescence values to
determine the scale and maps the individual pixels to this scale.
However, to use false color to visually compare images, images
must be adjusted to share the same scale (see Fig. 4).

References

1. Oxborough K (2004) Imaging of chlorophyll a
fluorescence: theoretical and practical aspects
of an emerging technique for the monitoring
of photosynthetic performance. J Exp Bot
55:1195–1205. https://doi.org/10.1093/
jxb/erh145
2. Barbagallo RP, Oxborough K, Pallett KE,
Baker NR (2003) Rapid, noninvasive screening
for perturbations of metabolism and plant
growth using chlorophyll fluorescence imag-
ing. Plant Physiol 132:485–493. https://doi.
org/10.1104/pp.102.018093
3. Badger MR, Fallahi H, Kaines S, Takahashi S
(2009) Chlorophyll fluorescence screening of
Arabidopsis thaliana for CO2 sensitive photo-
respiration and photoinhibition mutants.
Funct Plant Biol 36:867. https://doi.org/10.
1071/FP09199
4. Lawson T, Oxborough K, Morison J, Baker N
(2001) Evaluating guard cell photosynthesis in
intact green leaves using chlorophyll fluores-
cence imaging. Sci Access 3
5. Lawson T, Oxborough K, Morison JIL, Baker
NR (2002) Responses of photosynthetic elec-
tron transport in stomatal guard cells and
mesophyll cells in intact leaves to light, CO2,
and humidity. Plant Physiol 128:52–62
6. Baker NR, Oxborough K, Lawson T, Morison
JI (2001) High resolution imaging of photosyn-
thetic activities of tissues, cells and chloroplasts
in leaves. J Exp Bot 52:615–621. https://doi.
org/10.1093/jexbot/52.356.615
7. Baker NR (2008) Chlorophyll fluorescence: a
probe of photosynthesis in vivo. Annu Rev
Plant Biol 59:89–113. https://doi.org/10.
1146/annurev.arplant.59.032607.092759
8. Jin ZL, Tian T, Naeem MS, Jilani G, Zhang F,
Zhou WJ (2011) Chlorophyll fluorescence
responses to application of new herbicide
ZJ0273 in winter oilseed rape species. Int J
Agric Biol 13:43–50
9. Mauromicale G, Ierna A, Marchese M (2006)
Chlorophyll fluorescence and chlorophyll con-
tent in field-grown potato as affected by nitro-
gen supply, genotype, and plant age.
Photosynthetica 44:76–82. https://doi.org/
10.1007/s11099-005-0161-4
10. Rahbarian R, Khavari-Nejad R, Ganjeali A,
Bagheri A, Najafi F (2011) Drought stress
effects on photosynthesis, chlorophyll fluores-
cence and water relations in tolerant and sus-
ceptible Chickpea (Cicer Arietinum L.)
genotypes. Acta Biol Cracoviensia Ser Bot 53
(1):47–56. https://doi.org/10.2478/
v10182-011-0007-2
11. Baker NR (2004) Applications of chlorophyll
fluorescence can improve crop production stra-
tegies: an examination of future possibilities. J
Exp Bot 55:1607–1621. https://doi.org/10.
1093/jxb/erh196
12. Chaerle L, Leinonen I, Jones HG, Van Der
Straeten D (2006) Monitoring and screening
plant populations with combined thermal and
chlorophyll fluorescence imaging. J Exp Bot
58:773–784. https://doi.org/10.1093/jxb/
erl257
13. Harbinson J, Prinzenberg AE, Kruijer W, Aarts
MG (2012) High throughput screening with
chlorophyll fluorescence imaging and its use in
crop improvement. Curr Opin Biotechnol
23:221–226. https://doi.org/10.1016/j.
copbio.2011.10.006

14. Simkin AJ, McAusland L, Headland LR,
Lawson T, Raines CA (2015) Multigene
manipulation of photosynthetic carbon assimi-
lation increases CO2 fixation and biomass yield
in tobacco. J Exp Bot 66:4075–4090. https://
doi.org/10.1093/jxb/erv204
15. Ehlert B, Hincha DK (2008) Chlorophyll fluo-
rescence imaging accurately quantifies freezing
damage and cold acclimation responses in Ara-
bidopsis leaves. Plant Methods 4:12. https://
doi.org/10.1186/1746-4811-4-12
16. Tang JY (2006) The differential effects of her-
bivory by first and fourth instars of Trichoplusia
ni (Lepidoptera: Noctuidae) on photosynthesis
in Arabidopsis thaliana. J Exp Bot
57:527–536. https://doi.org/10.1093/jxb/
erj032
17. Scholes JD, Rolfe SA (2009) Chlorophyll fluo-
rescence imaging as tool for understanding the
impact of fungal diseases on plant perfor-
mance: a phenomics perspective. Funct Plant
Biol 36:880. https://doi.org/10.1071/
FP09145
18. Mutka AM, Bart RS (2015) Image-based phe-
notyping of plant disease symptoms. Front
Plant Sci 5:734. https://doi.org/10.3389/
fpls.2014.00734
19. McElrone AJ, Hamilton JG, Krafnick AJ,
Aldea M, Knepp RG, DeLucia EH (2010)
Combined effects of elevated CO2 and natural
climatic variation on leaf spot diseases of red-
bud and sweetgum trees. Environ Pollut
158:108–114. https://doi.org/10.1016/j.
envpol.2009.07.029
20. Leipner J (2001) Primary sites of ozone-
induced perturbations of photosynthesis in
leaves: identification and characterization in
Phaseolus vulgaris using high resolution chlo-
rophyll fluorescence imaging. J Exp Bot
52:1689–1696. https://doi.org/10.1093/
jexbot/52.361.1689
21. Aldea M, Frank TD, DeLucia EH (2007) A
method for quantitative analysis of spatially
variable physiological processes across leaf sur-
faces. Photosynth Res 90:161–172. https://
doi.org/10.1007/s11120-006-9119-z
22. Krause GH, Weis E (1991) Chlorophyll fluo-
rescence and photosynthesis: the basics. Annu
Rev Plant Physiol Plant Mol Biol 42:313–349.
https://doi.org/10.1146/annurev.pp.42.
060191.001525
23. Logan BA, Adams WW, Demmig-Adams B
(2007) Viewpoint: avoiding common pitfalls
of chlorophyll fluorescence analysis under field
conditions. Funct Plant Biol 34:853. https://
doi.org/10.1071/FP07113
Chlorophyll Fluorescence 139
24. Maxwell K, Johnson GN (2000) Chlorophyll
fluorescence—a practical guide. J Exp Bot
51:659–668. https://doi.org/10.1093/
jexbot/51.345.659
25. Murchie EH, Lawson T (2013) Chlorophyll
fluorescence analysis: a guide to good practice
and understanding some new applications. J
Exp Bot 64:3983–3998. https://doi.org/10.
1093/jxb/ert208
26. Adams WW, Demmig-Adams B (2004) Chlo-
rophyll fluorescence as a tool to monitor plant
response to the environment. In: Chlorophyll a
fluorescence. Springer, Dordrecht, pp
583–604
27. Kautsky H, Appel W, Amann H (1960) Chlo-
rophyll fluorescence and carbon assimilation.
Part XIII. The fluorescence and the photo-
chemistry of plants. Biochem Z 332:277–292
28. Buchanan BB, Balmer Y (2005) Redox regula-
tion: a broadening horizon. Annu Rev Plant
Biol 56:187–220. https://doi.org/10.1146/
annurev.arplant.56.032604.144246
29. Lawson T, Kramer DM, Raines CA (2012)
Improving yield by exploiting mechanisms
underlying natural variation of photosynthesis.
Curr Opin Biotechnol 23:215–220. https://
doi.org/10.1016/j.copbio.2011.12.012
30. Demmig-Adams B, Adams WW (2006) Photo-
protection in an ecological context: the
remarkable complexity of thermal energy dissi-
pation. New Phytol 172:11–21. https://doi.
org/10.1111/j.1469-8137.2006.01835.x
31. Horton P, Ruban AV, Walters RG (1996) Reg-
ulation of light harvesting in green plants.
Annu Rev Plant Physiol Plant Mol Biol
47:655–684. https://doi.org/10.1146/
annurev.arplant.47.1.655
32. Horton P, Johnson MP, Perez-Bueno ML, Kiss
AZ, Ruban AV (2008) Photosynthetic acclima-
tion: does the dynamic structure and macro-
organisation of photosystem II in higher plant
grana membranes regulate light harvesting
states? FEBS J 275:1069–1079. https://doi.
org/10.1111/j.1742-4658.2008.06263.x
33. Ruban AV, Murchie EH (2012) Assessing the
photoprotective effectiveness of
non-photochemical chlorophyll fluorescence
quenching: a new approach. Biochim Biophys
Acta Bioenerg 1817:977–982. https://doi.
org/10.1016/j.bbabio.2012.03.026
34. Li X-P, Muller-Moule P, Gilmore AM, Niyogi
KK (2002) PsbS-dependent enhancement of
feedback de-excitation protects photosystem
II from photoinhibition. Proc Natl Acad Sci
99:15222–15227. https://doi.org/10.1073/
pnas.232447699
140 Tracy Lawson and Silvere Vialet-Chabrand
35. Kiss AZ, Ruban AV, Horton P (2008) The
PsbS protein controls the organization of the
photosystem II antenna in higher plant thyla-
koid membranes. J Biol Chem
283:3972–3978. https://doi.org/10.1074/
jbc.M707410200
36. Murchie EH, Niyogi KK (2011) Manipulation
of photoprotection to improve plant photosyn-
thesis. Plant Physiol 155:86–92. https://doi.
org/10.1104/pp.110.168831
37. Ruban AV (2016) Nonphotochemical chloro-
phyll fluorescence quenching: mechanism and
effectiveness in protecting plants from photo-
damage. Plant Physiol 170:1903–1916.
https://doi.org/10.1104/pp.15.01935
38. Bjorkman O, Demmig B (1987) Photon yield
of O2 evolution and chlorophyll fluorescence
characteristics at 77 K among vascular plants of
diverse origins. Planta 170:489–504. https://
doi.org/10.1007/BF00402983
39. Genty B, Briantais J-M, Baker NR (1989) The
relationship between the quantum yield of
photosynthetic electron transport and quench-
ing of chlorophyll fluorescence. Biochim Bio-
phys Acta Gen Subj 990:87–92. https://doi.
org/10.1016/S0304-4165(89)80016-9
40. Genty B, Goulas Y, Dimon B, Peltier G, Moya I
(1992) Modulation of efficiency of primary
conversion in leaves, mechanisms involved at
PSII. Res Photosynth 4:603–610
41. Oxborough K, Baker NR (1997) Resolving
chlorophyll a fluorescence images of photosyn-
thetic efficiency into photochemical and
non-photochemical components - calculation
of qP and Fv0 /Fm0 without measuring Fo0 .
Photosynth Res 54:135–142. https://doi.
org/10.1023/A:1005936823310
42. Nedbal L, Soukupová J, Kaftan D,
Whitmarsh J, Trtı́lek M (2000) Kinetic imag-
ing of chlorophyll fluorescence using modu-
lated light. Photosynth Res 66:3–12. https://
doi.org/10.1023/A:1010729821876
43. Roháček K, Soukupová J, Barták M (2008)
Chlorophyll fluorescence: a wonderful tool to
study plant physiology and plant stress. Plant
Cell Compart Top Res Signpost, Kerala, India
41–104
44. Zangerl AR, Hamilton JG, Miller TJ, Crofts
AR, Oxborough K, Berenbaum MR, de Lucia
EH (2002) Impact of folivory on photosynthe-
sis is greater than the sum of its holes. Proc Natl
Acad Sci 99:1088–1091. https://doi.org/10.
1073/pnas.022647099
45. Morison JIL (2005) Lateral diffusion of CO2
in leaves is not sufficient to support photosyn-
thesis. Plant Physiol 139:254–266. https://
doi.org/10.1104/pp.105.062950
46. Lawson T, Morison J (2006) Visualising pat-
terns of CO2 diffusion in leaves. New Phytol
169:641–643. https://doi.org/10.1111/j.
1469-8137.2006.01655.x
47. Lawson T, Lefebvre S, Baker NR, Morison JIL,
Raines CA (2008) Reductions in mesophyll
and guard cell photosynthesis impact on the
control of stomatal responses to light and
CO2. J Exp Bot 59:3609–3619. https://doi.
org/10.1093/jxb/ern211
48. Loriaux SD, Avenson TJ, Welles JM, Mcder-
mitt DK, Eckles RD, Riensche B, Genty B
(2013) Closing in on maximum yield of chlo-
rophyll fluorescence using a single multiphase
flash of sub-saturating intensity. Plant Cell
Environ 36:1755–1770. https://doi.org/10.
1111/pce.12115
 Chapter 9

Measurement of O2 Uptake and Evolution in Leaves In Vivo

Using Stable Isotopes and Membrane Inlet Mass
Spectrometry
Steven M. Driever and Neil R. Baker

Abstract
Oxygen is both product and substrate of photosynthesis and metabolism in plants, by oxygen evolution
through water splitting and uptake by photorespiration and respiration. It is important to investigate these
processes simultaneously in leaves, especially in response to environmental variables, such as light and
temperature. To distinguish between processes that evolve or take up O2 in leaves in the light, in vivo gas
exchange of stable isotopes of oxygen and membrane inlet mass spectrometry is used. A closed-cuvette
system for gas exchange of leaf disks is described, using the stable isotopes 16O2 and 18O2, with a
semipermeable membrane gas inlet and isotope mass separation and detection by mass spectrometry.
Measurement of evolution and uptake, as well as CO2 uptake, at a range of light levels allows composition
of a light–response curve, here described for French bean and maize leaf disks.

Key words Oxygen evolution, Oxygen uptake, Stable isotopes, Membrane inlet mass spectrometry,
MIMS

1 Introduction

Oxygen in plants is both produced and consumed in several pro-

cesses and tissues throughout the plant, such as photosynthetic
water splitting, photorespiratory oxygenation, and metabolic respi-
ration. As previously reviewed [1], the study of net O2 exchange of
photosynthesis and respiration was generally made using a Clark-
type O2 electrode. This classic and relatively inexpensive method
allows for examination of changes in the net exchange of O2 of leaf
disks in a closed cuvette [2], in combination with simultaneous
measurement of chlorophyll fluorescence [3]. Other, more recent
techniques for measuring net oxygen exchange include photoa-
coustic spectrometry and electron paramagnetic resonance oxime-
try, as reviewed by van Gorkom, Gast [4], and electrochemical fuel
cell-based differential oxygen analyzers [5, 6].

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_9, © Springer Science+Business Media, LLC, part of Springer Nature 2018

141
142 Steven M. Driever and Neil R. Baker

Although determining net O2 exchange is valuable, it does not

allow distinction between out- and ingoing fluxes of O2 under the
same conditions, e.g., in the light. To allow the simultaneous study
of O2-consuming and -producing processes, it is necessary to dif-
ferentiate between O2 uptake and evolution. This can be achieved
by the use of stable isotope forms of oxygen (isotopologues) with
different masses: 16O and 18O. In brief, in an atmosphere of 18O2,
and leaf water predominantly containing 16O (as H216O), an
increase in atmospheric 16O2 in the presence of light will originate
from water splitting due to photochemistry, while decreases in 18O2
will be primarily due to O2-consuming processes such as photores-
piration and the Mehler reaction. A mass spectrometer is required
for this type of measurement, to separate and distinguish the con-
centration changes of 16O2 and 18O2. However, both 18O2 gas and
mass spectrometers are relatively expensive. Still, this is a powerful
technique, particularly when combined with simultaneous mea-
surement of CO2 gas exchange and chlorophyll fluorescence. It
allows assessment of several underlying processes, such as photo-
respiration [7, 8], respiration [9], and oxygen-reducing processes
such as the Mehler reaction [10–14], which are studied in C3 as
well as CAM and C4 plants [15–17].
In this chapter a description is given of the use of 18O2 and
membrane inlet mass spectrometry (MIMS) for the determination
of O2 uptake and evolution in leaf disks of French bean and maize.

2 Materials

2.1 Plant Material Grow plants of French bean (Phaseolus vulgaris) and maize (Zea
mays) from seed in controlled-environment cabinets with a 12-h
day length at 20
C day and 18
C night temperature and light
intensity of 300 μmol m2 s1 photosynthetically active photon
flux density (PPFD). Keep air humidity at an air vapor pressure
deficit of ca. 0.75 kPa. Grow French bean seedlings in peat-based
compost and maize in soil-based compost, in 1 L pots, and water
regularly with Hoagland’s nutrient solution [18]. Only take fully
expanded leaves for experiments described here.

2.2 Cuvette The purpose-built, homemade measurement cuvette is designed to

accommodate the measurement of O2 exchange of a leaf disk in a
small closed volume, based on a previously described design [15]
with modifications. As shown in Fig. 1, the cuvette consists of a
brass base that is temperature controlled via a circulating water
bath. The brass base, that supports a gas-permeable membrane, is
held in place by a steel plate with a central hole and nitrile O-rings
and a large compression ring, to prevent movement of the mem-
brane. A small porous Teflon support beneath the membrane
allows gas diffusion through the membrane into the vacuum line
 Leaf O2 Exchange Using Membrane inlet Mass Spectrometry 143

Light
PAM source
Fluorometer
Insert
Optic fibre
(Chl fluorescence)
Optic fibre
(Actinic light)
Gas outlet Gas inlet
(with septum) (with septum)

Compression Compression Nitrile o-ring

Leaf disk on ring ring
metal mesh Coated glass

Temperature controlled
brass block
Permeable
membrane

Porous membrane
Mass spectrometer support
(vacuum)

Fig. 1 Schematic drawing of the cuvette used for leaf disk O2 exchange measurements using 18O2 and online
mass spectrometry. The cuvette consists of a brass temperature-controlled base, which supports a
gas-permeable membrane. This membrane is pressed onto the base by a metal plate (light grey) with two
nitrile O-rings. The metal plate is pressed onto the brass base and a large plastic compression ring. The metal
plate also forms the bottom half of the leaf chamber, with a metal mesh supporting a leaf disk. On top of the
metal plate, the metal insert is screwed on, forming the top half of the closed leaf cuvette. The insert allows for
introduction of gas via a gas inlet that is covered with a septum. To allow gas flow through the chamber, a gas
outlet is located on the opposite side of the insert. The leaf disk in the chamber is illuminated via an optic fiber
and an anti-reflectance-coated glass window in the insert. Through this glass window, chlorophyll fluores-
cence can be measured during gas exchange via an optic fiber

of the mass spectrometer. The steel plate allows positioning of a leaf

disk on a metal mesh over the central hole. The cuvette is closed by
a removable steel insert with a glass window, which is screwed onto
the metal plate, where nitrile O-rings assure a gastight seal. Actinic
light is introduced by an optic fiber, which is placed in the central
hole of the insert over the glass window. At an angle, an optic fiber
can be positioned to take measurements of chlorophyll fluorescence
during gas exchange. For injection of gases, the insert has a septum-
sealed inlet and outlet, which allow rapid flushing, injection, and
correction of pressure changes when introducing the desired gases
into the closed cuvette.

2.3 Vacuum Line, Gases diffuse from the cuvette into the vacuum of a mass spectrom-
Water Trap, and Mass eter via a stainless steel tube and pass through a water trap packed
Spectrometer with dry ice and ethanol (see Note 1), allowing passage of both O2
144 Steven M. Driever and Neil R. Baker

Leaf disk cuvette Mass

with membrane spectrometer
Tap Ion
source
Vacuum line +
+

Faraday
Quadrupole
cup
rod electrodes
detector

Dry ice and

ethanol

Water Pre- High-

trap vacuum vacuum
pump pump Control and
acquisition

Fig. 2 Schematic drawing of the leaf cuvette, vacuum line, water trap, and mass spectrometer setup. The
vacuum line tubing passes from the cuvette through a water trap, containing dry ice and ethanol, to prevent
water from entering the mass spectrometer. The tubing is first vacuumed with the pre-vacuum pump. Once
the tubing is under vacuum, the tap to the mass spectrometer is opened. Gases from the leaf cuvette pass via
the membrane into the vacuum line and are introduced into the mass spectrometer. Ionization of gases is done
by the ion source and mass separation by quadrupole rod electrodes. Ions of different, separated masses are
detected with a Faraday cup detector. The mass spectrometer is controlled and data is acquired with a control
unit and computer system

and CO2 but trapping water (Fig. 2). Gases are detected with a
quadrupole mass spectrometer system with a cross-beam ion source
fitted with a tungsten filament to ionize the diffused gases. The
mass spectrometer separates ions by mass-to-charge ratio via rod
electrodes alternating in polarity controlled by a high-frequency
generator. Separated ions are then collected with a Faraday cup
and signals are amplified by an electronic preamplifier before signals
are recorded every 3 s via a computer. Several specified masses
should be detected continuously (e.g., peak jump mode) and
dwell time and resolution for optimum performance with air (N2,
CO2, 16O2, Ar) mixed with 18O2 should be optimized. During
experiments m/e 28 (N2), 32 (16O2), 36 (18O2), 40 (Ar), and
44 (CO2) are recorded. Additionally, m/e 5.5 is recorded as a
measure for signal stability and data are rejected if excessive changes
in m/e 5.5 are recorded. To reduce signal noise m/e 32, 36 and
44 are also recorded as a ratio to m/e 40.

3 Methods

3.1 Measurement 1. Pre-illuminate a leaf (while attached to the plant) and take a
Protocol Overview leaf disk.
2. For measurement of O2 and CO2 gas exchange, insert the leaf
disk into the measurement cuvette with an 18O2 atmosphere
and high CO2 concentration.
 Leaf O2 Exchange Using Membrane inlet Mass Spectrometry 145

3. Sample the changes of gases in the cuvette via a membrane inlet

and monitor gases with a mass spectrometer. During an initial
dark period of 10 min, determine oxygen uptake (Uo), then
expose the leaf disk to a specific PPFD, and determine CO2
assimilation rate (A), Uo, and oxygen evolution (Eo) in the
light until CO2 is depleted. Repeat this procedure with a mini-
mum of three leaf disks (biological replicates) per PPFD and a
minimum of six points per light–response curve.
4. Measure responses of leaf disks at 25
C, over a range of PPFDs
(ca. 0–1000 μmol m2 s1), to obtain a light–response curve.
To predetermine the PPFDs required (minimum of 6), a
light–response curve with an open gas exchange system can
be used (as described in Chapter 5 [19]) at the CO2 compen-
sation point or in an atmospheric CO2 concentration of
ca. 6000 μmol mol1, which is saturating for photosynthesis
in a non-mixed, closed cuvette.

3.2 Determining Leaf To ensure measurements at different PPFDs take a similar amount
Disk Size Per PPFD of time, determine the size of leaf disk needed using pilot experi-
ments following the procedure above but without 18O2. Vary the
leaf disk size per PPFD (0.25–0.9 cm2), to ensure a similar mea-
surement period of approximately 15 min, following CO2 uptake
from saturating CO2 to the CO2 compensation point. Note that
the diameter of the leaf disk must always be a smaller diameter than
the metal mesh of the leaf cuvette.

3.3 Membrane 1. Cut the membrane to the size of the base and apply a light coat
Installation, of high-vacuum grease to the outer perimeter on the brass base,
Calibration, Effective without contaminating the porous Teflon membrane support
Cuvette Volume, in the center of the base (see Note 2).
and Drawdown 2. Apply the membrane to the base and smooth out any wrinkles.
3. Install the steel plate and screw on the large compression ring.
4. Check that there are no leaks and the vacuum is stable by
creating a vacuum in the vacuum line tubing using the
pre-vacuum pump. When pre-vacuum has reached values near
to that of the mass spectrometer (as indicated by the vacuum
gauge and specification of the mass spectrometer manufacturer),
allow the membrane to settle onto the porous Teflon support for
at least 3 h (or overnight) before starting measurements.
5. After the membrane has settled and pre-vacuum is reached,
open the vacuum line to the mass spectrometer, close the line
to the pre-vacuum pump and leave the vacuum of the whole
line to return to operational vacuum (according to mass spec-
trometer manufacturer’s specification).
6. Using different gases (N2, CO2, Ar, 16O2, and 18O2), tune the
peaks for the masses of interest (according to the manufac-
turer’s instruction for the mass spectrometer used). Make
146 Steven M. Driever and Neil R. Baker

sure that the sensitivities for 16O2 (m/e 32) and 18O2 (m/e 36)
are equal, to allow a calibration of both masses to 16O2. For
handling 18O2 gas, see Note 3.
7. To calibrate signals of masses of interest to known concentra-
tions in an empty clean cuvette, it is recommended to use two
or more premixed gases with CO2 and O2 concentrations in
the range of measurement (0–2% for CO2 and 0–21% for O2)
and pure N2 gas for zero values. Gas mixture concentrations
may vary, but should include concentrations around
400, 10,000, and 20,000 ppm for CO2 and 2, 10, and 21%
for O2. With these known premixed gases and zero, determine
a linear calibration curve for each gas concentration against its
corresponding mass peak intensity signal. With all calibrations,
ensure that a filter paper disk (mimicking a leaf disk) is present
in the cuvette, to correct for volume effects of the leaf disk.
8. Determine the effective cuvette volume by injecting a
measured volume of argon (Ar) gas into the cuvette containing
air, via the inlet port, with a gastight syringe. The change in the
signal of the injected Ar can then be used to calculate cuvette
volume, as this depends on the known amount of gas in the
syringe. The cuvette volume (V) can be described as

vinj
V ¼ 0:0093b ð1Þ
a  0:0093

where vinj is the injected volume of Ar (ml), a is the concentra-

tion or signal of Ar in air, b is the concentration or signal of Ar
after injection, and 0.0093 is the mole fraction of Ar. Effective
cuvette volume is best done after flushing the cuvette with
compressed air and should contain a filter paper or leaf disk
(in the dark). Mass peak intensity signals are determined when
the cuvette is closed. To determine effective cuvette volume
when using different sized leaf disks (as determined per light
intensity), this procedure should be repeated with a
corresponding-size filter paper disk.
9. Correct for drawdown of gas by the system. During an experi-
ment, a small amount of gas is drawn through the membrane
into the mass spectrometer vacuum. To correct for this, the
drawdown is determined for a range of CO2, 16O2, and 18O2
concentrations during a period of ~20 min, described by a
linear relationship of drawdown rate (signal or concentration
min1) for each gas based upon their respective starting signal/
concentration. The methodology for correction is given in the
experiment described in Subheading 3.4.

3.4 Measurement 1. Take a leaf disk of a pre-illuminated, fully expanded leaf and
directly cover the outer edges of the disk by placing it between
two wet filter paper rings (see Note 4). Place the leaf disk on the
 Leaf O2 Exchange Using Membrane inlet Mass Spectrometry 147

metal mesh support in the cuvette and close the cuvette with
the insert.
2. Flush the cuvette with N2 gas for several minutes until 16O2
and CO2 are depleted. Then, inject 18O2 (95% isotopic enrich-
ment), Ar, and CO2 to create the desired gas mixture (21% O2
and >1% CO2 and Ar in N2). To allow instantaneous pressure
correction, each injection is done into the gas inlet, with the gas
outlet connected to a tube immersed in water. This ensures a
rapid, reproducible, and accurate way to create a gas mixture in
the closed cuvette. The outlet tube in water ensures that no gas
retro-diffuses intro the cuvette after injection. Inject a second
pulse of CO2 in the same way to assure a saturating CO2
concentration (>1%) after several minutes into the measure-
ment, only if the CO2 level rapidly drops below saturating level
(indicated by arrow, Fig. 3a).
3. Close the gas inlet and outlet and measure gas exchange in the
dark for about 5 min.
4. Turn on the actinic light; measure over a period of approxi-
mately 15–20 min the changes in CO2, 18O2, and 16O2; and
record their corresponding masses. Typical exchanges of CO2,
18
O2, and 16O2 for French bean and maize are shown in Fig. 3.
5. After CO2 depletion to the compensation point (see Fig. 3,
indicated by *), continue the measurement for another
5 min, stop, and empty the cuvette by flushing with N2 gas.
6. To correct for leakage and drawdown of gases, repeat the
procedure with a wet filter paper disk of the same size as the
leaf disk with gas mixtures corresponding to those used at
saturating CO2 concentration and the CO2 compensation
point during experiments.

3.5 Analysis The calculation of the oxygen evolution and oxygen uptake by leaf
disks is based on a simple model that describes the consumption of
oxygen from the atmosphere and the production of oxygen from
water. The natural abundance of 18O2 in water is known to be very
low (0.2%) and therefore the production of oxygen from water is
mainly observed as 16O2. The consumption of oxygen from the
atmosphere is the sum of three different oxygen species: 16O2,
16 18
O O, and 18O2. When the available oxygen in the atmosphere
is replaced by 18O2, as used in this method, the sum of the three
species is predominantly determined by 18O2. In this way, a distinc-
tion can be made between oxygen evolution producing 16O2 and
oxygen uptake consuming 18O2. The contribution of 16O18O in
this model is considered small as the natural abundance of this
isotopologue is negligible in both water and 18O2 gas used.
148 Steven M. Driever and Neil R. Baker

a 16O
2
18O
2 CO2
16
14 French bean

12
10
8

%
6
4
2
*
0
0 5 10 15 20
Time (min)
b 16O
2
18O
2 CO2
16
14 maize

12
10
8
%

6
4
2
*
0
0 5 10 15 20
Time (min)

Fig. 3 Typical changes of 16O2, 18O2, and CO2 during gas exchange in the light of
leaf disks in a closed-cuvette system using membrane inlet mass spectrometry.
The exchange of a French bean leaf disk (a) from saturating CO2 concentration,
which reaches the CO2 compensation point after 14 min (as indicated by *). The
arrow indicates an extra injection of CO2 into the cuvette. The exchange of a
maize leaf disk (bottom) from saturating CO2 concentration reaches the CO2
compensation point after 10 min (*). The evolution of O2 is shown as an increase
in 16O2 and the uptake of O2 as a decrease in 18O2 in both graphs

In order to adopt this model for a closed-cuvette system, both

the concentration changes of 16O2 and 18O2 over time as a result of
both evolution and uptake have to be taken into account. Note that
the concentration changes mentioned are corrected for drawdown,
leaks, and effective cuvette size.
1. Calculation of oxygen uptake and evolution: oxygen uptake
(Uo) and evolution (Eo) in a closed system are calculated
according to Radmer and Kok [20]:

  18
16
O2 ∂ O2
U o ¼  1 þ 18 ð2Þ
O2 ∂t
 Leaf O2 Exchange Using Membrane inlet Mass Spectrometry 149

16 16  18
∂ O2 O2 ∂ O2
Eo ¼  ð3Þ
∂t 18 O
2 ∂t

where 16O2 and 18O2 represent the concentrations of the

respective oxygen species at a given time point, corrected for
drawdown, leaks, and effective cuvette size. ∂16O2/∂t and
∂18O2/∂t are the first derivatives of their respective species over
time. Uo describes the uptake of 18O2 by the leaf disk over time,
where the changes in 18O2, ∂18O2/∂t, are corrected for the
18 16
relative abundance of O2 compared to O2
(1 + 16
O2/ O2). Eo is calculated from the changes in 16O2
18

over time, ∂16O2/∂t, minus the changes in 18O2, ∂18O2/∂t,

again corrected for its relative concentration to 16O2.
2. Calculation of CO2 assimilation: Assimilation of CO2 (A) in a
closed system is calculated as
 
∂CO 2
A¼ ð4Þ
∂t

where CO2 represents the concentration of CO2 at a given time

point, corrected for drawdown, leaks, and effective cuvette
size. ∂ CO2/∂t is the first derivative of CO2 over time.
3. Composing a light–response curve: Although the closed-
cuvette system described here only allows measurement of a
single PPFD, different leaf disks can be measured at a range of
different PPFDs. Such measurements can be used to compose a
light–response curve for saturating CO2 or at the compensa-
tion point, as these are the two levels at which the leaf disk is at
steady state (see Note 5). Examples of light–response curves of
Eo, Uo, and A at saturating CO2 concentration and CO2
compensation point for French bean (Fig. 4) and maize
(Fig. 5) are given.

4 Notes

1. To prepare a water trap with dry ice and ethanol, place the trap
line in a dewar and fill it ¾ full with dry ice. Then carefully add
ethanol (>95% pure) until all dry ice is submerged. Ethanol
and dry ice will evaporate, so top up with more ethanol when
dry ice starts to emerge over time (approximately after 3 h).
Wear safety goggles and insulated gloves when preparing the
water trap and adding the ethanol.
2. Apply silicone-based vacuum grease with a fingertip to the
outside perimeter of the brass base, just enough so the
150 Steven M. Driever and Neil R. Baker

A
35

O2 or CO2 m-2 s-1)

Eo Uo A
30
25
20
umol/m2/s

15
A, Eo, Uo (µmol

10
5
0
-5
0 200 400 600 800 1000 1200
PPFD (umol/m2/s)
B
14
O2 or CO2 m-2 s-1)

Eo Uo A
12
10
8
umol/m2/s

6
A, Eo, Uo (µmol

4
2
0
-2
0 200 400 600 800 1000 1200
PPFD
PPFD (umol/m2/s)
(µmol m-2 s-1)

Fig. 4 Response curves of O2 evolution (Eo), O2 uptake (Uo), and CO2 assimilation (A) to photosynthetically
active photon flux density (PPFD) for French bean leaf disks at saturating CO2 concentration (ca. 7000 ppm,
(A)) and at the CO2 compensation point (B). Each symbol represents a measurement on a single leaf disk at a
single PPFD. With increasing PPFD, both A and Eo increase until saturating PPFD at saturating CO2
concentration (A). At the compensation point, A remains around 0 and Eo and Uo increase significantly with
increasing PPFD, due to photorespiration (B). Data from Driever and Baker [11]

membrane sticks to the base. Don’t contaminate the porous

Teflon membrane support. If there is contamination, then
replace the Teflon membrane support with a new one.
3. Pure grade 18O2 gas (>98%) is costly and therefore best used in
low quantities in a small-volume cuvette (as described in this
chapter). To use gas sparingly and avoid dilution, a cylinder
fitted with a headspace unit is used. This headspace unit con-
sists of a four-way cross fitting, connecting the cylinder to a
tube with septum, a pressure gauge, and two valves (at sides
with and without septum).
 Leaf O2 Exchange Using Membrane inlet Mass Spectrometry 151

a
35

A, Eo, Uo (µmol O2 or CO 2 m-2 s-1)

Eo Uo A
30

25

20

15

10

-5
0 200 400 600 800 1000

b PPFD (umol/m2/s)

14
A, Eo, Uo (µmol O2 or CO2 m-2 s-1)

Eo Uo A
12

10

-2
0 200 400 600 800 1000
PPFD
PPFD (umol/m2/s)
(µmol m-2 s-1)

Fig. 5 Response curves of O2 evolution (Eo), O2 uptake (Uo), and CO2 assimilation (A) to photosynthetically
active photon flux density (PPFD) for maize leaf disks at saturating CO2 concentration (ca. 7000 ppm, (a)) and
at the CO2 compensation point (b). Each symbol represents a measurement on a single leaf disk at a single
PPFD. With increasing PPFD, both A and Eo increase, but do not reach a saturating PPFD at saturating CO2
concentration (a). At the compensation point, A remains around 0 and Eo and Uo only increase slightly with
increasing PPFD (b). Data from Driever and Baker [11]

18
To remove air from the headspace and obtain O2 gas to
inject:
l Check pressure gauge: when reading (close to) 0, there is air
in the headspace. Proceed to remove the air with a
vacuum pump.
l Connect side without septum to the pre-vacuum pump.
l Open both headspace valves.
l Turn on pump and drawdown headspace volume to vacuum.
152 Steven M. Driever and Neil R. Baker

l Close both headspace valves tight.

l Turn off pump.
l Open main cylinder valve and increase pressure just above
atmospheric (~1 bar).
l Close main cylinder valve.
l Put needle of gastight syringe (Luer-lock type) through
septum, open headspace valve at septum side, and watch
pressure (on gauge) drop when purging gas.
l Close the headspace valve, close syringe lock, and remove
gas syringe needle from septum.
To save 18O2 gas from the headspace back into the main
cylinder (wear safety glasses and insulated gloves):
l Plunge cylinder into liquid nitrogen, and wait until fully
cooled (liquid nitrogen stops fizzing).
l Open main cylinder valve.
l Watch pressure drop on gauge (the leftover 18O2 gas in the
headspace shoots back into the cylinder).
l Close main valve and let the cylinder warm up to room
temperature.
4. Leaf disks are placed between two wet filter paper rings that
cover the outer edges of the disk and prevent those edges from
drying out. This is done with filter paper rings which are of
similar diameter as the outer edge of the leaf disk. Rings are
used instead of a wet filter paper disk (as is common in O2
electrode studies) to allow for free diffusion of gases from both
sides of the disk. The leaf disk edge is sandwiched between the
two wet filter paper rings, providing ample hydration of the
entire leaf disk edge for the period of measurement.
5. The design of the non-mixed closed gas-exchange cuvette
creates a high boundary layer resistance. Therefore, high con-
centrations of CO2 are needed to saturate CO2 assimilation
(>6000 ppm), compared to mixed, open gas-exchange sys-
tems. Furthermore, water vapor is not measured and therefore
transpiration, stomatal conductance, and intercellular CO2
concentration are unknown. It is therefore important that a
steady state is reached, to derive a true rate of O2 and CO2
exchange. This can only be achieved at two states: at saturating
CO2 level and at the level of the CO2 compensation point. In
the measurement, the compensation point can be determined
when the CO2 level has reached a minimum. It is important to
 Leaf O2 Exchange Using Membrane inlet Mass Spectrometry 153

note that the described closed-cuvette setup does not allow for
a CO2–response curve, similar to that measured in frequently
used open gas-exchange systems.

Acknowledgments

This work was initiated and done under the supervision of Prof.
Neil R. Baker (University of Essex, UK) and supported by a
research studentship from the Department of Biological Sciences
at the University of Essex to Dr. Steven M. Driever. We thank Prof.
Suzanne von Caemmerer (Australian National University) for
providing details on the leaf cuvette as used by Maxwell et al.
[15] and Ruuska et al. [12].

References

1. Hunt S (2003) Measurements of photosynthe- photorespiratory CO2 release. Plant Physiol

sis and respiration in plants. Physiol Plantarum
117(3):314–325. https://doi.org/10.1034/j.
1399-3054.2003.00055.x
2. Delieu T, Walker DA (1981) Polarographic
measurement of photosynthetic oxygen evolu-
tion by leaf-disks. New Phytol 89(2):165–178.
https://doi.org/10.1111/j.1469-8137.1981.
tb07480.x
3. Delieu TJ, Walker DA (1983) Simultaneous
measurement of oxygen evolution and chloro-
phyll fluorescence from leaf pieces. Plant
Physiol 73(3):534–541. https://doi.org/10.
1104/Pp.73.3.534
4. van Gorkom HJ, Gast P (1996) Measurement
of photosynthetic oxygen evolution. In: Bio-
physical techniques in photosynthesis.
Springer, Dordrecht, pp 391–405
5. Davey PA, Hunt S, Hymus GJ, DeLucia EH,
Drake BG, Karnosky DF, Long SP (2004)
Respiratory oxygen uptake is not decreased by
an instantaneous elevation of [CO2], but is
increased with long-term growth in the field
at elevated [CO2]. Plant Physiol 134
(1):520–527. https://doi.org/10.1104/pp.
103.030569
6. Willms JR, Dowling AN, Dong ZM, Hunt S,
Shelp BJ, Layzell DB (1997) The simultaneous
measurement of low rates of CO2 and O2
exchange in biological systems. Anal Biochem
254(2):272–282. https://doi.org/10.1006/
abio.1997.2416
7. Cousins AB, Pracharoenwattana I, Zhou WX,
Smith SM, Badger MR (2008) Peroxisomal
malate dehydrogenase is not essential for pho-
torespiration in arabidopsis but its absence
causes an increase in the stoichiometry of
148(2):786–795. https://doi.org/10.1104/
pp.108.122622
8. Cousins AB, Walker BJ, Pracharoenwattana I,
Smith SM, Badger MR (2011) Peroxisomal
hydroxypyruvate reductase is not essential for
photorespiration in arabidopsis but its absence
causes an increase in the stoichiometry of
photorespiratory CO2 release. Photosynth Res
108(2–3):91–100. https://doi.org/10.1007/
s11120-011-9651-3
9. Walker BJ, Cousins AB (2013) Influence of
temperature on measurements of the CO2
compensation point: differences between the
Laisk and O2-exchange methods. J Exp Bot
64(7):1893–1905. https://doi.org/10.1093/
jxb/ert058
10. Biehler K, Haupt S, Beckmann J, Fock H,
Becker TW (1997) Simultaneous CO2- and
16
O2/18O2-gas exchange and fluorescence
measurements indicate differences in light
energy dissipation between the wild type and
the phytochrome-deficient aurea mutant of
tomato during water stress. J Exp Bot 48
(312):1439–1449. https://doi.org/10.1093/
jxb/48.7.1439
11. Driever SM, Baker NR (2011) The water-water
cycle in leaves is not a major alternative electron
sink for dissipation of excess excitation energy
when CO2 assimilation is restricted. Plant Cell
Environ 34(5):837–846. https://doi.org/10.
1111/j.1365-3040.2011.02288.x
12. Ruuska SA, Badger MR, Andrews TJ, von
Caemmerer S (2000) Photosynthetic electron
sinks in transgenic tobacco with reduced
amounts of Rubisco: little evidence for signifi-
cant Mehler reaction. J Exp Bot 51:357–368.
154 Steven M. Driever and Neil R. Baker
https://doi.org/10.1093/jexbot/51.suppl_
1.357
13. Haupt-Herting S, Fock HP (2000) Exchange
of oxygen and its role in energy dissipation
during drought stress in tomato plants. Physiol
Plant 110(4):489–495. https://doi.org/10.
1111/j.1399-3054.2000.1100410.x
14. Haupt-Herting S, Fock HP (2002) Oxygen
exchange in relation to carbon assimilation in
water-stressed leaves during photosynthesis.
Ann Bot 89:851–859. https://doi.org/10.
1093/aob/mcf023
15. Maxwell K, Badger MR, Osmond CB (1998) A
comparison of CO2 and O2 exchange patterns
and the relationship with chlorophyll fluores-
cence during photosynthesis in C3 and CAM
plants. Aust J Plant Physiol 25(1):45–52
16. Shirao M, Kuroki S, Kaneko K, Kinjo Y,
Tsuyama M, Forster B, Takahashi S, Badger
MR (2013) Gymnosperms have increased
capacity for electron leakage to oxygen (Mehler
and PTOX reactions) in photosynthesis com-
pared with angiosperms. Plant Cell Physiol 54
(7):1152–1163. https://doi.org/10.1093/
pcp/pct066
17. Siebke K, Ghannoum O, Conroy JP, Badger
MR, von Caemmerer S (2003) Photosynthetic
oxygen exchange in C4 grasses: The role of
oxygen as electron acceptor. Plant Cell Environ
26(12):1963–1972. https://doi.org/10.
1046/j.1365-3040.2003.01112.x
18. Arnon D, Hoagland D (1940) Crop produc-
tion in artificial culture solutions and in soils
with special reference to factors influencing
yields and absorption of inorganic nutrients.
Soil Sci 50:463–485
19. Coe RA, Lin H (2018) Light-response curves
in land plants. In: Covshoff S (ed) Photosyn-
thesis: methods and protocols, Methods in
molecular biology, vol 1770. Springer,
New York
20. Radmer RJ, Kok B (1976) Photoreduction of
O2 primes and replaces CO2 assimilation. Plant
Physiol 58(3):336–340. https://doi.org/10.
1104/Pp.58.3.336
 Chapter 10

Using Stable Carbon Isotopes to Study C3 and C4

Photosynthesis: Models and Calculations
Nerea Ubierna, Meisha-Marika Holloway-Phillips,
and Graham D. Farquhar

Abstract
Stable carbon isotopes are a powerful tool to study photosynthesis. Initial applications consisted of
determining isotope ratios of plant biomass using mass spectrometry. Subsequently, theoretical models
relating C-isotope values to gas exchange characteristics were introduced and tested against instantaneous
online measurements of 13C photosynthetic discrimination. Beginning in the twenty-first century, tunable
diode laser spectroscopes with sufficient precision for determining isotope mixing ratios became commer-
cially available. This has allowed collection of large data sets, at low cost and with unprecedented temporal
resolution. With more data and accompanying knowledge, it has become apparent that there is a need for
increased complexity in models and calculations. This chapter describes instantaneous online measurements
of 13C photosynthetic discrimination, provides recommendations for experimental setup, and presents a
thorough compilation of equations needed for different applications.

Key words Tunable diode laser absorption spectroscope, C3, C4, Carbon isotope discrimination,
Laser, Leakiness, Photosynthesis, Mesophyll conductance, Online, δ13C

1 Introduction

Isotopes are atoms having the same atomic number (number of

protons) but different atomic masses because of different numbers
of neutrons. Stable isotopes are isotopes that do not undergo
radioactive decay over time. There are two naturally occurring
stable isotopes of carbon (C): the most abundant (98.9%) 12C
and 13C (1.1%). Fractionation (also called isotope effect) occurs
when the relative abundance of the heavy isotope in a product
differs from that of the reactant (¼ substrate) from which it origi-
nated. Isotope fractionation is defined as (see Note 1 and Table 1
for symbol definitions):

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_10, © Springer Science+Business Media, LLC, part of Springer Nature 2018

155
156 Nerea Ubierna et al.

Table 1
List of symbols used

Variable (units) Definition

2 1
A (μmol m s ) Photosynthetic rate
a (‰) Weighted 12C/13C fractionation for diffusion across the boundary layer and
stomata in series, Eq. (10)
ab(‰) 12
C/13C fractionation for CO2 diffusion in the boundary layer, ab ¼ 2.9‰
al (‰) 12
C/13C fractionation for diffusion of CO2 through water, al ¼ 0.7‰
am (‰) Summed 12C/13C fractionations during liquid-phase diffusion (al) and
dissolution of CO2 (es), am ¼ 1.8‰
as (‰) 12
C/13C fractionation for CO2 diffusion in air, as ¼ 4.4‰
bp (‰) 12
C/13C fractionation during carboxylation by PEPC, bp ¼ 2.2‰
brub (‰) 12
C/13C fractionation of RuBisCO alone, brub ¼ 29 – 30‰, but variable
depending on the report
12
b3 (‰) C/13C fractionation during carboxylation by RuBisCO including
respiration and photorespiration fractionations, Eq. (19)
b 03 (‰) Overall in vivo 12C/13C fractionation during carboxylation by RuBisCO and
PEPC, Eq. (25)
b3-sim (‰) Eq. (59)

b 3 (‰) Eq. (62)

b4 (‰) Combined 12C/13C fractionation by CO2 dissolution, hydration, PEP

carboxylation, and respiration, Eq. (20)
b 04 (‰) Net 12C/13C fractionation that occurs as CO2 is converted to HCO‐3 ,
including fractionations by CO2 dissolution, hydration, and PEPC activity,
Eqs. (26) and (27)
b4-sim (‰) Eq. (60)

b 4 (‰) Eq. (63)

Ca (Pa) CO2 partial pressure in the ambient air

Cbs (Pa) CO2 partial pressure in the bundle-sheath cells, Eq. (50)
Cc (Pa) CO2 partial pressure in the chloroplast
CCA (Pa) CO2 partial pressure at the sites of CA
Ci (Pa) CO2 partial pressure inside the leaf
Cm (Pa) CO2 partial pressure in the mesophyll cytosol
Cs (Pa) CO2 partial pressure at the leaf surface
CA Carbonic anhydrase
d (‰) 12
C/13C fractionation during dehydration of HCO‐3 , d ¼ 10.1‰
E (mol m2 s1) Transpiration rate

(continued)
 On-line Measurements of 13C Discrimination 157

Table 1
(continued)

Variable (units) Definition

12
e (‰) C/13C fractionation during decarboxylation
e0 (‰) 12
C/13C fractionation during decarboxylation including the effect of a
respiratory substrate isotopically distinct from recent photosynthate,
Eq. (28)
e* (‰) Modulation factor for e0 to account for shift in respiratory substrate, Eq. (29)
12
eb (‰) C/13C fractionation associated with the catalyzed hydration of
CO2 þ H2 O $ HCO‐3 , eb ¼  9‰, but changing with temperature
es (‰) 12
C/13C fractionation as CO2 dissolves, es ¼ 1.1‰
F Photorespiratory rate
12
f C/13C fractionation during photorespiration
gac (mol m2 s1) Conductance to diffusion of CO2 in air
2 1
gbl (mol m s ) Boundary layer conductance to CO2
gbs (μmol m2 s1 Pa1) Bundle-sheath conductance to CO2
gc (μmol m2 s1 Pa1) Chloroplast conductance to CO2
gm (μmol m2 s1 Pa1) Mesophyll conductance to CO2
h (‰) Catalyzed 12C/13C fractionation during CO2 hydration, h ¼ 1.1‰
KC (Pa of CO2) Michaelis-Menten constant of RuBisCO for CO2
KO (Pa of O2) Michaelis-Menten constant of RuBisCO for O2
2 1
L (μmol m s ) Leak rate of CO2 out of the bundle sheath
Om (Pa) O2 partial pressure in the mesophyll cells
Os (Pa) O2 partial pressure in the bundle-sheath cells
pCO2 (Pa) CO2 partial pressure
pO2 (Pa) O2 partial pressure
PEPC Phosphoenolpyruvate carboxylase
ℛd (μmol m2 s1) Non-photorespiratory CO2 released in the dark
2 1
ℛm (μmol m s ) Mesophyll mitochondrial respiration rate
1
rc (m s Pa μmol
2
) Chloroplast resistance to CO2 diffusion
rm (m2 s Pa μmol1) Mesophyll resistance to CO2 diffusion
1
r w (m s Pa μmol
2
) Wall resistance to CO2 diffusion
12
s (‰) C/13C fractionation during leakage of CO2 out of the bundle-sheath cells,
s ¼ 1.8‰, when there is no HCO‐3 leakage out of the bundle-sheath cells
t Ternary correction coefficient, Eq. (9)
2 1
Vc (μmol m s ) RuBisCO carboxylation rate

(continued)
158 Nerea Ubierna et al.

Table 1
(continued)

Variable (units) Definition

Vh (μmol m2 s1) Hydration rate
2 1
Vo (μmol m s ) RuBisCO oxygenation rate
2 1
Vp (μmol m s ) PEP carboxylation rate
Vpmax (μmol m2 s1) Maximal PEP carboxylation rate
2 1
Vcmax (μmol m s ) Maximal RuBisCO carboxylation rate
α Isotope fractionation, Eqs. (1) and (2)
αac αac ¼ 1 þ a

αam αam ¼ 1 + am
αb αb ¼ 1 + b03
αe αe ¼ 1 + e
αf αf ¼ 1 + f
β Fraction of leaf carbon fixed by PEPC (Eq. 25)
βp Coefficient to partition the total mesophyll resistance into wall and chloroplast
resistances (Eq. 45)
δa (‰) δ13C of the CO2 in the air in the leaf cuvette
δ13C (‰) C isotopic composition, Eq. (4)
Δb (‰) Discrimination associated with RuBisCO, Eq. (36)
Δe (‰) In Eq. (33): Most of the discrimination associated with respiration
In Eq. (39): Discrimination associated with respiration
Δf (‰) Discrimination associated with photorespiration, Eq. (34)
Δgm (‰) In Eq. (32): Discrimination associated with the diffusion of CO2 from the
intercellular airspaces to the chloroplast
In Eq. (38): Most of the discrimination associated with the diffusion of CO2
from the intercellular airspaces to the chloroplast
Δgs (‰) Discrimination associated with diffusion of CO2 through the boundary layer
and stomata, Eq. (37)
Δi (‰) In Eq. (31): Discrimination that would occur if Cc ¼ Ci in the absence of any
respiratory fractionation
In Eq. (41): Discrimination that would occur if Cc ¼ Ci
Δobs (‰) Observed 13C photosynthetic discrimination, Eq. (5)
Δ13C (‰) 12
C/13C photosynthetic discrimination, Eq. (3)
Δ3‐com (‰) Comprehensive model for 13C photosynthetic discrimination in C3 species,
Eqs. (8), (14), (15), (17), (18a), and (21a)

(continued)
 On-line Measurements of 13C Discrimination 159

Table 1
(continued)

Variable (units) Definition

Δ4‐com (‰) Comprehensive model for 13C photosynthetic discrimination in C4 species,
Eqs. (18b) and (21b)
Δ3‐sim (‰) Simplified model for 13C photosynthetic discrimination in C3species,
Eq. (22a)
Δ4‐sim (‰) Simplified model for 13C photosynthetic discrimination in C4 species,
Eq. (22b)
ϕ (unitless) Leakiness, calculated with Eq. (49) (ϕ1, no simplifications); Eq. (52) (ϕ2,
when Cbs is much larger than Cm); Eq. (53) (ϕ3, when gm is infinite);
Eq. (54) (ϕ4, when Cbs is much larger than Cm and gm is infinite)
γ ∗ (unitless) Half of the reciprocal of RuBisCO specificity
∗
Γ (Pa) CO2 compensation point in the absence of mitochondrial respiration
ζ (unitless) Ratio of the 12CO2 mole fraction in the dry air coming into the gas-exchange
cuvette over the difference in 12CO2 mole fractions of air in and out of the
cuvette, Eq. (6)
The equation numbers refer to equations in the main text

Rr
α¼ ð1Þ
Rp

where R is the 13C/12C ratio of the reactant (r) or product (p). The
12
C reacts faster than 13C (i.e. k12 > k13 where k is the reaction
constant), meaning that the 12C/13C ratio is larger in the product
that in the reactant.
Fractionation (α) is often used in chemistry, geology, and
oceanography, but in the biological literature we use discrimination
(Δ) because of its numerical convenience [1]. Δ and α relate as
Δ ¼ ðα  1Þ ð2Þ

In the case of 13C photosynthetic discrimination (Δ13C), the

substrate is ambient air and the product is plant carbon
(or photosynthate). Thus Δ13C describes the change in 13C com-
position induced by the plant during photosynthesis and it is calcu-
lated as [1]
R a  R p δa  δp
Δ13 C ¼ ¼ ð3Þ
Rp 1 þ δp

where R and δ are ratios of isotope abundance and δ13C values,

respectively, of CO2 in air (a) and plant carbon (p). The δ13C is the
13
C isotopic composition and represents the 13C/12C ratio in a
given sample with respect to the internationally accepted standard,
160 Nerea Ubierna et al.

which is the Vienna Pee Dee Belemnite (VPDB). The VPDB was
adopted after the original Pee Dee Belemnite standard (PDB,
RPDB ¼ 0.0112372 [2]) was exhausted. The 13C/12C ratio in
VPDB is RVPDB ¼ 0.0111797 [3], with a reported uncertainty of
28  106. Therefore, the value should be rounded to
RVPDB ¼ 0.01118 [4]. Nevertheless, some authors retain the
extra decimal points, given the measurement precision of mass
spectrometers and tunable diode laser absorption spectroscopes
(e.g. [5]). The δ13C values are calculated as

Rsample  RVPDB
δ13 C ¼ ð4Þ
RVPDB

Both Δ13C and δ13C values are conveniently expressed in parts

per thousand (‰), but care must be taken in recognizing that 1‰
means 1  103. For example, δ13C ¼ 26‰ indicates that the
13
C/12C ratio of the sample is less than the standard by 26‰ or
2.6%. When comparing δ13C values in two samples (A and B), and if
δ13CA < δ13CB, we can state that A is depleted in 13C or “lighter”
than B. In contrast, A is enriched or “heavier” than B if
δ13CA > δ13CB.
Photosynthetic discrimination can be measured instanta-
neously by combining measurements of leaf gas and 13C isotope
exchange [6]:
ζðδout  δin Þ
Δobs ¼ ð5Þ
1 þ δout  ζðδout  δin Þ
C in
ζ¼ ð6Þ
C in  C out

where C is the 12CO2 mole fraction and δ is the δ13C of the CO2, as
measured in dry air coming in (reference airstream) and out (sample
airstream) of the gas-exchange cuvette (see Notes 2 and 3).
Photosynthetic discrimination against 13C can also be modeled
and the most applied models are those of Farquhar and colleagues
[7–9]. When observations (Eq. 5) are matched to the theoretical
models, the equation can be solved for an unknown variable, such
as mesophyll conductance in C3 species (gm) or leakiness (ϕ) in C4.
Stable C isotopes are a powerful tool to investigate photosyn-
thetic processes and their response to environmental gradients.
Initial applications consisted of determining the δ13C values of
plant biomass using isotope-ratio mass spectrometry (IRMS).
These analyses showed that plants are depleted in 13C compared
with ambient air and that δ13C of plant tissue varies with photosyn-
thetic pathway [10, 11] and in response to environmental factors
(i.e. [12, 13]). Park and Epstein [14] proposed a two-stage process
to explain the observed 13C depletion in plants. First, the diffusion
 On-line Measurements of 13C Discrimination 161

of CO2 from the atmosphere into the leaf favors 12CO2. Secondly,
the largest fractionation is via the photosynthetic enzyme ribulose-
1,5-bisphosphate carboxylase/oxygenase (RuBisCO), which
discriminates against 13CO2 because of the intrinsically lower reac-
tivity of 13C [15]. Subsequently theoretical models relating δ13C
values to gas exchange characteristics were introduced [7, 8, 16]
and tested against measurements of Δobs [6].
Initial measurements of Δobs required cryogenically trapping
the CO2 in the airflow out of a gas-exchange cuvette for posterior
analysis with dual-inlet IRMS [6, 17]. This technique had high
instrument precision but it was time consuming and could only
be completed in a laboratory setting. Subsequently, the introduc-
tion of continuous-flow IRMS (CF-IRMS) allowed for faster sam-
ple throughput yet retained high precision [18]. Gas-exchange
systems and chambers were then interfaced to CF-IRMS for simul-
taneous measurements of leaf gas and C-isotope exchange
[19–21]. Beginning in the twenty-first century, tunable diode
laser absorption spectroscopes (TDLAS) became commercially
available and permitted measurements of δ13C of CO2 in air at
high temporal resolution and with similar precision (0.2‰) to
IRMS [22–24]. In the last 10 years, TDLAS have become fre-
quently used in the study of plant isotope exchange from the leaf
to the ecosystem level [22, 25–34].
This chapter is dedicated to describing isotopic applications
that use instantaneous online measurements of Δ13C to study
photosynthesis and plant relationships with their environment.
We provide recommendations for the experimental setup and pres-
ent a thorough compilation of the equations derived from theoret-
ical Δ13C models that are needed for different applications. In
particular we describe (1) theoretical models for Δ13C in C3 and
C4 species, (2) apportioning C3–Δ13C into its component fractio-
nations, (3) calculation of gmin C3 species, and (4) calculation of ϕ
in C4 species. We don’t discuss applications based on the analysis of
δ13C in plant biomass. These types of analyses have been used in
plant biology for over 50 years and their fundamentals have been
extensively reviewed [35–45].

2 Materials

2.1 Tunable Diode TDLAS rely on absorption at different wavelengths by different

Laser Absorption isotopologues (i.e. 13CO2 and 12CO2) and use Beer’s absorbance
Spectroscope (TDLAS) law to relate measured spectroscopic absorbance of a molecule to its
abundance. A simplified synopsis of the operation is as follows:
(1) the light source (laser) wave number is selected to match an
individual absorption line of a target molecule, (2) the source
radiation is collimated and passed through an absorption cell con-
taining the sample gas before focusing in a detector, and (3) the
162 Nerea Ubierna et al.

absorbance (ratio of incident to transmitted light) recorded by the

detector is proportional to molecular abundance.
Lasers need to meet two specific criteria: (1) ability to distin-
guish between two different isotopic species (selectivity) and
(2) high absorption strength (sensitivity). Increasing temperature
and pressure results in line broadening and reduced instrument
selectivity. Therefore TDLAS are operated at subatmospheric pres-
sures. When measuring isotopes it is desirable that lines chosen for
each isotopologue have similar effective strengths. For example,
12
CO2 is about 100 more abundant than 13CO2; thus the
13
CO2 line should be ideally 100 stronger to produce similar
absorbance [32]. However, different strengths for the main and
minor isotopologues can result in larger instrumental temperature
sensitivity. Therefore temperature should be carefully controlled.
Commercially available TDLAS share these basic principles
with some differences in the implementation. Lead alloy tunable
diode lasers (TDLs, [32]) and cavity ring-down spectroscopes
(CRDS, [46]) use pulsed laser operation, while quantum cascade
lasers (QCLs, [23, 47]) are continuous wave. Some TDLs use
liquid nitrogen to cool the laser, while this is not necessary for the
other two instruments. TDLs and QCLs generally have a dual-cell
arrangement, where the source radiation is split into two multi-pass
cells (sample/reference). In CRDS the multi-pass cell is replaced by
a stable optical resonator, called the ring-down cavity [48].

2.2 Coupling of a The coupling of a TDLAS with a cuvette or gas-exchange system

TDLAS with a Gas- varies depending on the type of TDLAS used and the particular
Exchange System application. It is not a turnkey operation, but rather the user needs
to carefully build the setup. Overall all the settings have some
common features such as (Fig. 1) the following: (1) the gas lines
in and out of the leaf cuvette are split and part of the flow is diverted
to the TDLAS; (2) needle valves, critical flow orifices, or other
controls are used to regulate the flow of sample into the TDLAS;
(3) the sample gas is dried before entering the TDLAS; and (4) a
manifold, either manual or automatic, is used to switch between
lines (i.e. reference, sample, or calibration gases) sent to the ana-
lyzer. Additional recommendations for the setup are provided in
Note 4. A typical measuring cycle is described in Note 5.

2.3 Calibration TDLAS raw data are mole fractions (mixing ratio) for the major and
of TDLAS Raw Data minor isotopologues (μmol CO2 mol dry air1). TDLAS require
frequent calibration to achieve the accuracy required for isotope
ratio measurements. Several calibration methods have been used,
for example (see Supporting Information 1 in [28]): (1) one-point
calibration, (2) two-point calibration, and (3) concentration series.
One-point calibration uses one standard gas of known isotopolo-
gues mixing ratio to calculate a gain factor by dividing the true by
the measured molar fractions of the isotopologue of interest. Data
 On-line Measurements of 13C Discrimination 163

a. Synthetic air mix b. Measurement environment c. Tunable diode laser absorption

spectroscope

2 Leaf
* cuvette
*
IRGA
console 8b
7
3 L ‘outlet gas’ TDLAS

Manifold
P ‘inlet gas’

calibration gas
1 Parent synthetic air mix
*
MFC

MFC

2 Excess air exit points *

Calibration
1 4 system
3 Humidifier

4 Pure CO2

5 Inlet (reference) line

Outlet (sample) line

CO2

6
O2
N2

7 Matching valve loop

8 8b Drying system

Fig. 1 Generalized schematic for TDLAS-gas exchange coupling. In the following description the numbers in
brackets correspond to numbers in the schematic. (a) The synthetic air mix system consists of a parent N2/
O2 mix (1) that supplies air for the gas-exchange system with a range of O2%. Mass flow controls (MFC) are
used to regulate the flow rate, which should be greater than the flow required by the gas exchange and TDLAS,
to ensure that the lines downstream are not pulling ambient air. Excess flow is evacuated through T-valves
(2) to avoid over-pressurizing the line. The air mix flows through a humidifier (3). Pressurized CO2 is supplied
directly to the IRGA console (4). (b) The measurement environment consists of controlling the conditions
(mainly temperature) outside the leaf cuvette so they are similar to those experienced by the leaf inside the
cuvette. Large temperature gradients between inside and outside the leaf cuvette should be avoided because
they can result in condensation, which could result in unwanted isotopic fractionations as CO2 dissolves in
water and can also damage the gas-exchange system and give inaccurate estimates of parameters such as Ci.
The inlet gas line (reference) (5) to the leaf cuvette is split and part of the flow is diverted to the TDLAS. The
outlet gas (sample) (6) is often collected from the leaf cuvette matching tube (7). A three-way valve or
equivalent setting is used to allow for TDLAS to sample either outlet gas or, during matching mode, room air.
(c) The TDLAS requires a manifold to sample between inlet, outlet, and calibration gas streams. Before
reaching the TDLAS, the vapor is dried with a water trap which can be located in the inlet and outline lines
(8) or right before the TDLAS (8b). Additional recommendations for this setup are provided in Note 4
164 Nerea Ubierna et al.

are then corrected by multiplying measured mole fractions by the

corresponding gain. Two-point calibration expands this scheme by
using two standard gases to calculate a gain and an offset for each
isotopologue [5, 32]. This requires that the working standards span
the range of sample mixing ratios for each isotopologue, not the
isotope ratio per se [32]. The concentration series [25] requires two
calibration gases and one of them is analyzed at different CO2
concentrations. The first calibration gas is used to calculate a
12
CO2gain to correct 12CO2 raw data as in the one-point calibration
scheme. This 12CO2 gain gas is not required for obtaining accurate
δ13C values per se, but to enable the comparison of [CO2] between
TDLAS and gas-exchange machine. The second calibration gas can
be either pure CO2 or partly diluted, e.g. 10% CO2 is often used,
which is diluted further by mixing it with variable amounts of CO2-
free air. This generates a series of calibration standards with differ-
ent CO2 mole fractions but identical δ13C. The raw 13CO2 mole
fractions in the series of calibration standards are regressed (qua-
dratic fit) against their theoretical 13CO2 content, which is calcu-
lated from the known δ13C and 12CO2 values of the different
dilutions. Subsequently, the quadratic fit is applied to measured
raw sample 13CO2 values to obtain a corrected sample value.
Finally, δ13C values are calculated using the 12CO2 (from the gain
correction) and 13CO2 (from the quadratic fit). A fourth calibration
scheme (method 4) that also uses the same composition gas
measured at different concentrations is what we refer to as the offset
calibration method. In this case, a series of offsets between the true
and the measured δ13C values are calculated from measuring several
CO2 concentrations of the same gas. These data are used to gener-
ate the “offset calibration line” by regressing CO2 concentrations
against offsets. Finally the corrected δ13C value of a sample is
calculated by applying the offset calibration line to the raw sample
δ13C value derived from the uncorrected molar fractions for each
isotopologue.
In the calibration Methods 1–3, after calibration of raw data,
δ13C values are calculated with Eq. (4) and total CO2 concentration
with [5]

12 
CO2 þ13 CO2
½CO2  ¼ ð7Þ
1  0:00474

where 0.00474 is the fraction of CO2 containing all isotopologues

other than 12C16O16O and 13C16O16O.
The choice of calibration method will depend on laser perfor-
mance and experiment design. Method 1 requires only one calibra-
tion standard but might introduce errors because it does not
account for offsets. Methods 2 and 3 require at least two calibration
standards. Method 3 is often used to correct for any CO2 concen-
tration dependency of the instrument, which is a common feature
 On-line Measurements of 13C Discrimination 165

of many lasers. Method 4 also corrects for CO2 concentration

dependency using only one calibration gas. However, it could
introduce bias if the isotopic composition of the calibration gas is
very different to the gas supplying the gas-exchange machine due to
the fact that the laser may have very different concentration depen-
dencies for 12CO2 and 13CO2. Users should carefully test laser
performance to select an appropriate calibration scheme. Further-
more, it is important to have one or more independent tanks of
known composition that are used to quality check how well the
calibration system is performing. Practically, this could be the pure
CO2 tank supplying the gas-exchange machine.

2.4 Obtaining The error associated with Δobs measurements results from combin-
a Tolerable ing ζ (Eq. 6) values and instrument precision (see Fig. 2). The Δobs
“Precision” Error (Eq. 5) is approximately ζ multiplied by (δout  δin). If the instru-
in Δobs ment precision is X and the measurement errors for each δout and
δin are uncorrelated, then the
pffiffiffierror associated with the calculation
of Δobs is approximately 2  ζ  X. For example, if instrument
precision is 0.2‰ and ζ ¼ 10, the error associated with Δobs
determination is 2.8‰. When measurements are performed at
ambient air CO2 concentration (i.e. Cout ¼ 400 ppm), ζ ¼ 10 can
be achieved with a drawdown in CO2 by the photosynthetic leaf of
44 ppm (444/(444–400) ¼ 10). A typical value for Δobs in C3
plants is 20‰, and therefore an error of 2.8‰ represents 14% of the
total value. During C4 photosynthesis, Δobs is approximately 4‰,
so a 14% error would be only 0.6‰. Achieving 0.6‰ error with an
instrument precision of 0.2‰ when measurements are performed
at ambient air CO2 concentration would require a drawdown in
CO2 by the leaf of 357 ppm. This is a very large drawdown that
would require large leaf areas. For example, Ubierna et al. [28] used
a large leaf area of 45 cm2 to measure maize leaves with large
photosynthetic rates and only achieved a drawdown of 238 ppm.
Often, TDLAS instrument precision error is smaller than 0.2‰.
In the previous maize example, a TDL was used and the standard
deviation of 50 measurements of a reference gas over a 3-h period
was 0.06‰. With this machine precision, achieving an error of
0.6‰ only requires ζ ¼ 7, which translates to a CO2 drawdown
of 67 ppm.
Instrument imprecisions occur due to the finite signal-to-noise
ratio of the detection process (random shorter term data fluctua-
tions) and drift associated with temperature and pressure changes
(longer term data fluctuations). At short-time integrations the
variance is dominated by random noise whereas at longer time
integrations machine drift dominates. Allan variance is used to
determine the time integration that minimizes the mean squared
deviation between adjacent sample averages. Sampling a calibration
gas for the likely time length of measurements (e.g. 1 h) and
plotting the Allan variance against time integration on a log-log
166 Nerea Ubierna et al.

Fig. 2 Error associated with measurements of Δobs (Eq. 5) for different photosynthetic rates (A). Panels (a and
b) show the response of ζ (Eq. 6) to leaf area and flow rate, respectively, for A ¼ 30 (solid line), ¼15 (dotted
line), or ¼10 μmol m2 s1 (dashed line). In (a) calculations were done assuming a constant flow of
400 μmol s1, and in (b) a constant leaf area of 6 cm2 was used. Panels (c and d) show the error pffiffiffi in Δobs
when instrument precision (X) is 0.2‰ and ζ are the values displayed in (a and b). The error is  2  ζ  X ;
thus it can be reduced by either decreasing ζ or X. In this example X was kept constant to illustrate the impact
of varying ζ. Increasing leaf area or reducing the flow rate reduces ζ. When X ¼ 0.2‰, values for ζ  10
result in an error in Δobs  2.8‰, which for typical C3 photosynthesis is 14% of Δobs. The grey arrows
indicate multiple combinations of leaf areas, flow rates, and photosynthetic rates that result in ζ ¼ 10 and
therefore error of 2.8‰. Lowering X would allow for higher ζ for similar errors

scale can aid in identifying the best sampling procedure for the
measurements.
For each experiment, the user should assess what is a tolerable
error in Δobs and which values for ζ are needed as a function of the
instrument precision. Different ζ can be obtained by manipulating
the CO2 partial pressure ( pCO2), leaf area, and airflow through the
 On-line Measurements of 13C Discrimination 167

chamber (Fig. 2). Care should be taken that manipulations of leaf

area and airflow do not result in relative humidity inside the leaf
chamber detrimental to plant performance or the potential of
condensation occurring at very high humidity.

3 Methods

3.1 Theoretical Farquhar and coworkers have developed comprehensive theoretical

Model models to describe 13C photosynthetic discrimination both for C3
for Photosynthetic and C4 species [7–9]. In the literature, these models are presented
Discrimination Against with a plethora of different equations, which sometimes are equiv-
13
C alent to the original formulations, but frequently are presented in
simplified versions. This situation can lead to confusion regarding
the assumptions and implications of each formulation. Here we
describe the comprehensive model for 13C discrimination and
then detail the different modifications that can be applied to it in
order to derive either equivalent or simplified versions.

3.1.1 Mathematical The current comprehensive model for photosynthetic discrimina-

Formulations tion against 13C in C3 plants is [7, 9]
 
1 Ca  Ci 1þt
Δ3‐com ¼

a þ
1t Ca 1t
  ð8Þ
Ci  Cc 0 Cc α b ℛd C c α b Γ ∗
am þ b3  e  f
Ca C a αe V c C a αf C a

where the variables are defined in Table 1, and t and a
are the
ternary correction factor and the weighted fractionation for diffu-
sion across the boundary layer and stomata in series, respectively,
calculated as
αac E
t¼ ð9Þ
2g ac

ab ðC a  C s Þ þ a s ðC s  C i Þ
a
¼ ð10Þ
Ca  Ci

where E is the transpiration rate, gac is the combined boundary layer

and stomatal conductance to CO2, and αac ¼ 1 þ a
(see Note 6).
Equation (8) can be written in several equivalent formulations,
which are used in different applications, that we detail below:
(a) The rates of RuBisCO carboxylation (Vc) and photorespiration
(F) can be calculated as [49]

V c ¼ A þ ℛd þ F ð11Þ
168 Nerea Ubierna et al.

Γ∗ ðA þ ℛd Þ
F ¼ ð12Þ
C c  Γ∗
Combining Eqs. (11) and (12) and solving for Vc result in
ðA þ ℛd ÞC c
Vc ¼ ð13Þ
C c  Γ∗
Substituting Vcgiven by Eq. (13) in Eq. (8) results in
2 3
Ci  Cc 0 Cc
  am þ b3
1 Ca  Ci 1þt6 6 Ca Ca 7
7
Δ3‐com ¼ a
þ 6 7 ð14Þ
1t Ca 1  t 4 αb ℛd C c  Γ ∗
αb Γ ∗5
 e  f
α e A þ ℛd C a αf C a

The term CAþℛ ∗ ¼ C is sometimes given as k, carboxylation

d Vc
c Γ c
efficiency [7]. Equation (14) uses A instead of Vc, which is
convenient because this is what is actually measured during
gas exchange.
(b) Equation (14) can be rearranged to [50]
2   3
0 0 a
Ca  Ci 0
Ci  Cc
b3  b3   b 3  am
1þt66 1þt Ca Ca 7 7
Δ3‐com ¼ 4 ∗ ∗ 5 ð15Þ
1t αb ℛd C c  Γ αb Γ
 e  f
αe A þ ℛd C a αf C a

Equation (15) is useful to investigate the relative contribu-

tions to discrimination by RuBisCO, stomatal conductance,
mesophyll conductance, respiration, and photorespiration (see
Eq. (40)).
(c) An expression as a function of mesophyll conductance (gm) can
be obtained substituting the pCO2 in the chloroplast (Cc) from
Eq. (14) by

A
Cc ¼ Ci  ð16Þ
gm

and rearranging terms [25, 26]:

a
1
 Ci
Δ3‐com ¼ þ ð1 þ t Þb 03  a

1t 1t Ca
  
1þt αb ℛd A αb ℛd C i  Γ∗ αb Γ∗
 b 03  a m  e þ e þ f
1t α e A þ ℛd g m C a α e A þ ℛd C a αf C a
ð17Þ
 On-line Measurements of 13C Discrimination 169

Equation (17) is useful to derive an expression to calculate

gm (see Eqs. (35) and (44)).
∗
(d) In Eq. (8) the term ααbf f ΓC a can be replaced by ααbf f 0:5VV oc C
C a (this is
c

Γ∗
because V c ¼ 0:5C c , [49]). This is convenient because it pro-
Vo

duces an expression of Δ3‐com with a similar structure to expres-

sions used for C4species [8, 9]. With this modification the
comprehensive Δ13C models in C3 and C4 species are
8
    9
> 1 Ca  Ci 1þt Ci  Cc Cc >
>
> a C ! Δ ¼ a
þ a þ b >
>
>
>
3 3‐com
1t 1t
m 3 >
3>
> Ca Ca Ca   >
< 2 =
b 3 C bs
Δcom ¼   b4 þ ϕ s
>
>
1 Ca  Ci 1þt6 6a m C i  C m þ C bs  C m C m77>
>
>
5>
> b C4 ! Δ4‐com ¼ 1  t a C a
>
>
þ
1  t 4 C ϕC C >
>
>
: a
1þ
m a ;
C bs  C m
ð18Þ

The b3 refers to the combined effects of RuBisCO  fractionation


0
αb eℛd
b 3 , and fractionations
 associated
 with respiration αe V c and
αb 0:5fV o
photorespiration αf V c [8]:
αb eℛd αb 0:5fV o eℛd 0:5fV o
b 3 ¼ b 03   ffi b 03   ð19Þ
αe V c αf V c Vc Vc
The terms αb,αe, and αf are defined as 1 + b 30 , 1 + e and 1 + f,
respectively, and can be approximated to 1 when multiplying terms
that are already in per mil. The b4is the combined fractionation
during PEP carboxylation, hydration, and respiration [8, 19]:
 
Vp V p eℛm
b 4 ¼ b 04 1  þ ðe s þ h Þ  ð20Þ
Vh Vh Vp

Substituting Ccin Eq. (18a) and Cm( pCO2 in the mesophyll) in

Eq. (18b) with Ci  A/gm results in the following expression for
Δcomas a function of gm:
8 2 3 9
>
>   Ci 
A
  >
>
>
>
6 7 >
>
>
>

a 1 þ t 6

a A g

a 7 >
>
>
> a C3 ! Δ3‐com ¼ þ 4 am  þ m
b3  5 >
>
>
> 1  t 1  t 1 þ t g C C 1 þ t >
>
>
< m a a >
=
Δcom ¼ 2 0   13
>
>   Ci 
A
C bs ðb 4 þ ϕðb 3  s ÞÞ þ C i 
A
ðϕs  b 4 Þ >
>
>

> C7 >>
>
> a
1þt6 6 a
A gm B
B gm C 7 >
>
>
> b C4 ! Δ4‐com ¼ þ 4 am  þ @   A 5 >
>
>
> 1  t 1  t 1 þ t g C C A >
>
>
: m a a
C bs þ C i  ð ϕ  1Þ >
;
gm

ð21Þ

Values for fractionation factors ab, as, am, e, f, s, es, h, b 03 , and b 04

and simplifications that can be applied to b3 and b4 are discussed
later.
170 Nerea Ubierna et al.

Equations (8), (14), (15), (17), (18a), and (21a) are equivalent
formulations of Δ13C in C3species. Several simplifications (see
Fig. 3) can be applied to these equations resulting in the simplified
form [7, 8]:
8
0
Ci 9
>
< a C3 ! Δ3‐sim ¼ as þ b 3  a s >
=
Δsim ¼ C a
>



: b C4 ! Δ4‐sim ¼ as þ b 0 þ ϕ b 0  s  as C i > ;
4 3
Ca
ð22Þ

Equation (22) omits effects other than diffusion through sto-

mata and carboxylation (C3 and C4) and leakage out of the bundle-
sheath cells (C4). If the objective is to derive gm or other parameters
from measurements and models of discrimination, the detailed
equation is required. For ecosystem or global scale applications
such as isotope-constrained C-budgets or forecasting Δ13C with
reduced data sets, a simplified model accounting for mesophyll
contribution by lowering b 30 might suffice [50]. Alternatively, if
gmis known, Eq. (22) could be used with Cc instead of Ciand
adjusting as. When diurnal patterns in Δ13C need to be resolved,
it might be necessary to account for respiratory substrate shifts
(e.g. [51]) and therefore respiratory fractionations should be
included in the calculation of Δ13C. For crude applications, such
as using Δ13 C to correct 14C data [52], it would be pointless going
beyond the simplest model.
To this point we have described the different modifications that
can be applied to the comprehensive Farquhar et al. [7, 9] discrimi-
nation model in order to derive either equivalent or simplified
versions. These formulations for Δ3‐com conveniently place respired
(ℛd) and photorespired CO2 (F) inside the chloroplast. This is
correct for leaf geometries where chloroplasts are tightly adpressed
against the plasmalemma and cell walls adjacent to intercellular air
spaces. Mitochondria are located deeper inside the mesophyll cells.
However, when the arrangement of chloroplasts is sparse, the mito-
chondrial CO2 released in the cytosol (F + ℛd) may diffuse out
through the plasmalemma and cell wall without having to first cross
the chloroplast [53–55]. In this case Δ3‐com is modified to [26]
a
1
 ci
Δ3‐com ¼ þ ð1 þ t Þb 03  a

1t 1t ca
2 3
∗ eℛd  
1 þ t 6 0
rwA f Γ þ
 4 b 3  am þ k þ b 0  a
þ fF þ eℛd r c V c 7
5
m
1t ca ca 3
Vc ca

ð23Þ
Fig. 3 Box A: Simplifications applied to the “comprehensive” model of 13C discrimination (Δ13C) for C3 and C4
photosynthesis. Several theoretical models (Δtheo) can be derived applying a series of simplifications to the
comprehensive model (Δcom) resulting in the simplified expression (Δsim). The ternary correction factor is
t ¼ α2gacacE and the weighted fractionation for diffusion across the stomata and boundary layer in series is
a
¼ a b ðC a CCs aÞþa s ðC s C i Þ
C i . Other variables are defined in Table 1. Box B: Values for the fractionation factors used
in the equations of Box A. The values included here are the most commonly used. A detailed discussion about
fractionation factor values and their ranges can be found in Subheading 3.1, step 2. We recommend
undertaking a sensitivity analysis due to the uncertainty in some of these values
172 Nerea Ubierna et al.

where r w ¼ C i C
A
m
and r c ¼ C mVC
c
c
are the wall and chloroplast
resistances, respectively. An expression equivalent to Eq. (23), but
formulated in terms of Cc and including the second-order terms
αam, αb, αe, and αf, is [56]
 
a
C a  C i 1 þ t Ci  Cc C c αb Γ∗ αb ℛd C c  Γ∗
Δ3‐com ¼ þ am þ b 03  f  e
1  t Ca 1t Ca C a αf C a αe A þ ℛd C a
 ∗ 
1 þ t αam A þ ℛd Γ αam ℛd
f   þ e
1  t αf C c  Γ∗ C a αe Ca

gc
ð24Þ
where gc ¼ 1/rc.
The utility of Eqs. (23) and (24) is compromised because they
both need values for Cm, which cannot be measured directly.
Recently the “Δ18O method” [57–59] has been used to derive
Cmand rc in C3 species [60]. However, there are three assumptions
in this method [complete isotopic equilibrium, CCA ¼ Cm, where
CCA is pCO2 at the sites of carbonic anhydrase (CA), and no
isotopic gradients between the sites of CA and evaporation]
which are still debated. It has also been suggested that rc and r w
could be simultaneously estimated combining Δ13C and gas
exchange equations in a nonlinear optimization program [56].
Equations (23) and (24) only consider the extreme case where
all the mitochondrial CO2 production goes into the cytosol. In
reality a proportion will diffuse out of the cell only via the cytosol,
while the rest will first enter the chloroplast; the predominance of
the diffusion pathway will depend on how densely packed around
the cell wall the chloroplasts are arranged and the relative resis-
tances of wall and chloroplast membrane. This modification has
been recently incorporated into C3photosynthetic equations [61]
and we have incorporated it in the Δ13C model though this model
is yet to be published.

3.1.2 Values The ab and as are the 12C/13C fractionations for CO2 diffusion in
for the Fractionation the boundary layer (2.9‰) and in air (4.4‰), respectively
Factors [6, 62]. The am ¼ 1.8‰ (see Note 6) combines the fractionations
for dissolution of CO2 (es ¼ 1.1‰) [63] and diffusion through
water (al ¼ 0.7‰) [64]. The temperature dependencies of es
(0.08‰ for 20 C increase in temperature) [63] and al [65] are
very small and can be ignored. The fractionation during leakage of
CO2 out of the bundle-sheath cells is termed s and has a value of
1.8‰ when there is no HCO‐3 leakage out of the bundle-sheath
cells. It can vary between 1.8 and 6.3‰ depending on the
amount of CA and equilibration between CO2 and HCO‐3 in the
bundle sheath [8, 66, 67].
 On-line Measurements of 13C Discrimination 173

Fractionation of RuBisCO alone determined in vitro, brub, is

often taken as 29–30‰ [68–71]. Notice that some reports
(i.e. [71]) express their results in terms of fractionation with respect
to dissolved CO2, but to apply them in a gas-phase context, as
discussed here, ~1‰ should be added to the value. In vivo some
proportion of C fixation happens through phosphoenolpyruvate
(PEP) carboxylation,

and then the overall fractionation during
carboxylation b 03 is given by [1] (see Note 7)
b 03 ¼ βb 04 þ ð1  βÞb rub ð25Þ

where b 04 is the fractionation factor for phosphoenolpyruvate car-

boxylase (PEPC, 5.7‰ at 25 C, discussed below in detail), and β
is the fraction of leaf carbon fixed by PEPC. Suggested values for β
range from 0.02 to 0.055 [7, 72], but there are reported values as
high as β ¼ 0.15 [73, 74]. The β might also vary with factors such as
N metabolism [75]. Combining extreme values for brub and β
results in a range for b 03 from 23 to 30‰ [76]. However, the
more reasonable range β ¼ 0.02–0.04 only lowers brub by
0.7–1.4‰ [72], which is in agreement with the most commonly
used values for b 03 ranging from 28 to 30‰ [71, 72, 77]. Experi-
mental data show b 03 to be independent of temperature [78] but
theory suggests a small decrease of 0.04‰ C1 with increasing
temperature [79]. It is unknown whether b 03 varies between C3 and
C4 species; the few data available for C4 (27.8
0.8 to
33.7
6.6‰) suggest that values are not significantly different
from those in C3 plants [80, 81]. When the simplified model for
Δ13C is used (Eq. 22), b 03 is often substituted by a lower value,
ranging from 25 to 27‰ [31, 40, 82, 83], in order to compensate
for the effects ignored in that equation [50]. As a rule of thumb, b 03
can be taken as 29 or 30‰ when the comprehensive model is used,
though other values are possible as discussed above.
The b 04 combines a series of fractionations that occur as CO2 is
converted to HCO‐3 , and it can be calculated as [8, 84]
b 04 ¼ e s þ e b þ b p ð26Þ

Firstly, the dissolution of CO2 into water concentrates 13C in

the gas phase by 1.1‰ (¼es). Subsequently, there is a strong isotope
effect (eb) associated with the catalyzed hydration of
CO2 þ H2 O $ HCO‐3 , which concentrates the heavy isotope in
HCO‐3 . The eb ¼ h – d ¼ 9‰ where h ¼ 1.1‰ and d ¼ 10.1‰ are
the fractionations during hydration of CO2 and dehydration of
HCO‐3 . The net equilibrium fractionation during dissolution and
hydration of CO2 is 7.9‰ at 25 C but changes with temperature
[63]. This enriched HCO‐3 pool is the substrate for PEPC, which
discriminates against the heavy isotope only by bp ¼ 2.2‰
174 Nerea Ubierna et al.

[16]. Accordingly, b 04 ¼ 5:7‰ at 25 C but is variable with

temperature due to the temperature dependency of eb [63, 66]:
 
0 9:483  1000
b4 ¼ þ 23:89 þ 2:2 ‰ ð27Þ
273 þ T ¤ C

There is uncertainty about the values for the fractionation

during photorespiration ( f ) and respiration (e). Estimates for
f range from 8 to 16‰ [26, 85–88], with a theoretical value of
11‰ [89], which is close to the experimental value of 11.6‰
[86]. Estimations of f are dependent on the values assumed for
b 03 . The 11.6‰ [86] value was derived with b 03 ¼ 26‰ but when b 03
¼ 29‰ it resulted in f ¼ 16‰ [26]. The f was independent of
temperature from 20 to 35 C and slightly greater at 15 C
[26]. The values of f are positive, indicating that photorespiration
favors the lighter isotope.
On the other hand, respired CO2 is enriched in 13C. Reported
values for e range from 0 to 5‰ [22, 90–95]. Because of the
uncertainty in values for f and e, users should undertake a sensitivity
analysis of their data using different parameter values to determine
how the interpretation of measurements shifts. A much larger
apparent respiratory fractionation can occur when the substrate
for respiration is isotopically distinct from recent photosynthate.
This typically occurs when the air used for discrimination measure-
ments has a different 13C composition than air during plant growth
(e.g. when a very depleted tank is used as reference gas). In field
conditions, apparent fractionation has been observed at dusk or
dawn when there is a shift in respiratory substrates [33, 51, 96]. To
account for apparent respiratory fractionation, e is replaced by
e0 [40, 51]:
e0 ¼ e þ e∗ ð28Þ

where e is the fractionation during day respiration as defined above

(0 to 5‰) and e* is
e ∗ ¼ δa  Δobs  δsubstrate ð29Þ

with δa and δsubstrate being the δ13C values of the air in the leaf
cuvette and of the likely respiratory substrate, respectively. If the
substrate for respiration is recent assimilates formed during the
discrimination measurements, then δsubstrate ffi δa  Δobs, e* ¼ 0.
In this case there is no apparent fractionation and the fractionation
associated with respiration is e. If the substrate for respiration is old
photosynthate then δsubstrate ffi δgrowth  Δgrowth, where δgrowth and
Δgrowth are the δ13C of the air in which the plants grew and the
discrimination at which C was fixed during growth, respectively. If
growing and measurement conditions are similar (i.e. similar
 On-line Measurements of 13C Discrimination 175

irradiance and CO2) and Δobs  Δgrowth, Eq. (29) can be simplified
to [97]
e ∗ ¼ δa  δgrowth ð30Þ

3.2 Splitting Δ3‐com The Δ3‐com can be rewritten in a form that highlights the relative
into its Components contributions of the major fractionations that ambient CO2 under-
goes until its fixation in recent photosynthate. These are fractiona-
tions associated with diffusion through the stomata and mesophyll,
by RuBisCO, respiration, and photorespiration. Two different
approaches have been used:
1. Evans and von Caemmerer [26] divided Δ3‐com into the follow-
ing components:
(a) Δi: Discrimination that would occur if Ci ¼ Cc in the
absence of any respiratory fractionation (e ¼ 0):
a
1
C i
Δi ¼ þ ð1 þ t Þb 03  a
ð31Þ
1t 1t Ca
(b) Δgm: Discrimination associated with the diffusion of CO2
from the intercellular airspaces into the chloroplast:
 
1þt 0 α b ℛd A
Δgm ¼ b 3  am  e ð32Þ
1t α e A þ ℛd g m C a

(c) Δe: Most of the discrimination associated with respiration:

1 þ t αb ℛd C i  Γ ∗
Δe ¼ e ð33Þ
1  t α e A þ ℛd Ca
(d) Δf: Discrimination associated with photorespiration:

1 þ t αb Γ ∗
Δf ¼ f ð34Þ
1  t αf C a
Therefore Δ3‐com given in Eq. (17) can be written as
Δ3‐com ¼ Δi  Δgm  Δe  Δf ð35Þ

2. Ubierna and Farquhar [50] divided Δ3‐com into the following

components:
(a) Δb: Discrimination associated with RuBisCO:

1þt 0
Δb ¼ b ð36Þ
1t 3
176 Nerea Ubierna et al.

(b) Δgs: Discrimination associated with diffusion of CO2

through the boundary layer and stomata:
 
1þt 0 a
C a  C i
Δgs ¼ b3  ð37Þ
1t 1þt Ca

(c) Δgm: Most of the discrimination associated with diffusion of

CO2 from the intercellular airspaces to the chloroplast:

1þt 0
Ci  Cc
Δgm ¼ b  am ð38Þ
1t 3 Ca
(d) Δe: Discrimination associated with respiration:

1 þ t αb eℛd C c  Γ∗
Δe ¼ ð39Þ
1  t α e A þ ℛd Ca
(e) Δf: Discrimination associated with photorespiration, calcu-
lated with Eq. (34).
Therefore Δ3‐com given in Eq. (15) can be written as

Δ3‐com ¼ Δb  Δgs  Δgm  Δe  Δf ð40Þ

Equation (40) illustrates how RuBisCO fractionation is low-

ered by the contributions of Δgs, Δgm, Δe, and Δf, with Δgs and Δgm
being the largest effects.
These two approaches to apportioning component discrimina-
tions differ in terms of the following: (1) Evans and von Caem-
merer [26] lump together into Δi the terms Δb and Δgs of Ubierna
and Farquhar [50], and (2) the definition of Δe in Ubierna and
Farquhar [50] includes all the discrimination associated with respi-
ration. This differs from that of Evans and von Caemmerer [26]
where Δe includes “most” of the respiratory discrimination, the
1 þ t αb eℛd A
omitted term being . By separating Δb and
1  t α e A þ ℛd g m C a
Δgs, the approach of Ubierna and Farquhar [50] allows direct
comparison of the magnitude of stomatal and mesophyll contribu-
tions to Δ3‐com. Therefore, this method is more convenient when
the aim is to investigate the relative contributions to Δ3‐com by the
different components, for example, in field studies testing the
predictive ability of different discrimination models [31, 33,
51]. Alternatively, if the objective is to derive gm, then the Evans
and von Caemmerer [26] approach should be used, because Ubi-
erna and Farquhar [50] include a small contribution of gm in the
respiratory term.
 On-line Measurements of 13C Discrimination 177

3.3 Calculation of gm Mesophyll conductance (gm) describes the movement of CO2 from
in C3 Species the sub-stomatal cavity across the intercellular spaces to the sites of
first carboxylation, which in C3 plants is RuBisCO located in the
chloroplast stroma [98]. Thus gm determines the drawdown in
pCO2 from the intercellular airspaces (Ci) to the chloroplast stroma
(Cc). Reported values for gm in C3 species and at ambient pCO2
range from 0.8 to 4.5 μmol m2 s1 Pa1 [99–101]. Mesophyll
conductance is a more dynamic trait than initially thought; varia-
tion in gm has been described in response to many factors, both
long-term acclimation and short-term reversible responses
(i.e. [99, 102–105]). Despite the plethora of studies focusing on
gm, a complete mechanistic understanding of gm responses is still
lacking.
There are several methods available to estimate gm (for reviews
see [106, 107]) but here we focus only on those using online 13C
discrimination, which are (1) the single-point method [6, 108,
109] and (2) the slope method [6, 72, 111].
Both methods rely on comparing measurements and observa-
tions of photosynthetic discrimination against 13C. They differ in
that the single-point method calculates a value of gm for an individual
measurement, while the slope method calculates an average gm value
from a series of measurements along a range of A-Ca values.
The most widespread approach is the single-point method
because it allows for the study of responses of gm to short-term
variations in environmental factors. However, it has been suggested
that some of the results with this method might be artefactual,
generated by including the factor of interest (Ci) as input variable
for the calculation of gm [56]. Subsequently, uncertainties in other
input parameters needed for gm calculation could result in unreal
estimates of gm. An argument suggesting that observed trends in gm
are not simply an artefact is that similar conclusions have been
reported with other non-isotope methods, which rely on different
assumptions [108, 111]. The largest source of error in gm calcula-
tions is uncertainty in the value for b 03. For example, Pons et al.
[106] estimated that 3‰ change in b 03 varied gm by 20%. Allowing
b 03 to vary by 3–4‰ across environmental gradients can cancel some
of the short-term observed trends in gm [25, 76]. However, this
variation in b 03 would imply almost a 200% increase of the relative
contribution of PEPC to total carbon fixation [76]. Even if theo-
retically possible, it seems unlikely.
The slope method cannot be used to describe fast changes in
gm, but rather it produces an average gm value. It does not require
Cias input variable. However, it assumes that gm is independent of
changes in environmental factors, such as pCO2 or irradiance, but
experimental data indicate otherwise [25, 99, 102, 108, 112,
113]. Here we present the calculations needed for both methods.
Additional studies are necessary to better understand the potential
limitations of each.
178 Nerea Ubierna et al.

3.3.1 Single-Point Farquhar and Cernusak [9] defined Δias the predicted Δ3‐com when
Method gm is infinite and Cc ¼ Ci:
   
1 Ca  Ci 1 þ t 0 C i αb ℛd C i  Γ∗ αb Γ∗
Δi ¼ a
þ b3  e  f
1t Ca 1t C a αe A þ ℛd C a αf C a
ð41Þ

This definition differs from that given in Eq. (31) [26, 72] in
that Eq. (31) considers fractionations associated with respiration
and photorespiration zero. In Farquhar and Cernusak [9] Δi is
equivalent to Δi  Δe  Δf (Eqs. 31, 33, and 34) using Evans and
von Caemmerer [26] notation.
Using Farquhar and Cernusak [9] notation, the “single-point
method” derives gmby subtracting Δi(Eq. 41) from the Δ3‐com
(Eq. 14), which results in
 
1þt 0 α b ℛd Ci  Cc
Δi  Δ3‐com ¼ b 3  am  e ð42Þ
1t αe A þ ℛd Ca

In Eq. (42), Δ3‐com can be substituted by Δobs (Eq. 5) and

(Ci  Cc) by A/gm. Then the gm can be solved as (see Note 8)
0 α ℛd
1 þ t b 3  a m  αbe e Aþℛ A
gm ¼ d
ð43Þ
1t Δi  Δobs Ca
One can arrive at an equivalent expression using Evans and von
Caemmerer [26] notation by substituting in Eq. (35) the Δgmby its
expression given in Eq. (32) and solving for gm:
0 α ℛd
1 þ t b 3  a m  αbe e Aþℛ A
gm ¼ d
ð44Þ
1  t Δi  Δe  Δf  Δobs C a
where Δi, Δe, Δf, and Δobs are given by Eqs. (31), (33), (34), and
(5), respectively.
Pons et al. [106] estimated that for ζ ¼ 4–33, the % error in gm
calculation increased from 1 to 12%, 6 to 94%, and 11 to 263% for
instrument precisions of 0.02, 0.10, and 0.20‰, respectively. Obvi-
ously, large ζ should be avoided by adjusting the leaf area and
airflow through the gas-exchange cuvette.
In the past (i.e. before 2012), several simplifications were
applied to the calculation of gmwith Eq. (43) or (44). Firstly, no
study included ternary effects, but now we know that ternary
corrections can change values for gm, especially at high vapor pres-
sure deficits [9, 26]. Secondly, fractionations during respiration and
photorespiration have occasionally been considered negligible.
Values for e and f are still under debate, but they are unlikely to
be zero. Uncertainty in e has the largest impact at low assimilation
fluxes, whereas the impact of f increases with temperature and pO2.
Experimentally, uncertainty in f has been avoided by measuring at
 On-line Measurements of 13C Discrimination 179

low pO2 (e.g. 2%); however, not all plants perform well under
prolonged exposure to low oxygen concentrations [77]. For best
gm estimations we recommend using the complete equation with-
out simplifications, especially when studying short-time responses
to environmental gradients.
If photorespiratory and mitochondrial CO2 release occurs in
the cytosol separated from RuBisCO in the chloroplast, then gmis
calculated from Eq. (23) as [26]
 
0
ð1βp ÞA  0 
fF þeℛd βp V c
1þt
1t b 3  a m Ca þ b 3  a m þ Vc Ca
gm ¼ ∗ eℛ
ð45Þ
1þt f Γ þ k
d
Δi  Δobs  1t  Ca

where r w ¼ (1  βp)rm, rc ¼ βprm, and Δi is given by Eq. (31). In

the original formulation of this equation βp was referred as β but we
modified it here so it is not confused with the β previously used in
Eq. (25). The βp is a coefficient to partition the total mesophyll
resistance (rm ¼ r w + rc) into wall (r w) and chloroplast (rc) resis-
tances. The value for βp depends on the spatial arrangement of
mitochondria relative to chloroplast and therefore it is expected
to vary across species. However, at present there are no published
data for βp. To give an idea of its impact on the calculation of gm,
using βp ¼ 0.5 increased gm by 10% at 21% O2 over values estimated
when βp ¼ 0 [26].
A modification of the single-point method is the photosynthate
method, which also uses Eq. (43) to calculate gmbut instead of
measuring Δobs it estimates it with Eq. (3) from the δ13C values
of recent photosynthate given by leaf-soluble sugars [114–116] or
phloem contents [83]. This method can be useful for ecophysio-
logical applications in the field, where online measurements of Δobs
are impractical. Additionally it might be used when large sample
sizes are required, such as breeding programs or selection of geno-
types, where collecting and analyzing many leaves is faster and
easier than online discrimination measurements.

3.3.2 The Slope Method Traditionally the slope method has used the definition of Δi given in
Eq. (31) that assumes negligible respiratory and photorespiratory
fractionations [26, 72]. Then substituting in Eq. (35) the Δgm by its
expression given in Eq. (32) and solving for Δi  Δ3‐com result in
  
1þt 0 αb ℛd A
Δi  Δ3‐com ¼ b 3  am  e þ Δe þ Δf
1t αe A þ ℛd g m C a
ð46Þ

where Δi, Δe, and Δf are given in Eqs. (31), (33), and (34),
respectively. Equation (46) shows that Δi  Δ3‐com depends linearly
on A/Ca with the slope proportional to 1/gm(¼rm), and the
180 Nerea Ubierna et al.

intercept reflecting the respiratory and photorespiratory terms (see

Note 9).
The Farquhar and Cernusak [9] definition of Δi(Eq. 41)
doesn’t assume that respiratory and photorespiratory contributions
are negligible. In this case the Δi  Δ3‐com is
 
1þt 0 αb ℛd A
Δi  Δ3‐com ¼ b 3  am  e ð47Þ
1t α e A þ ℛd g m C a

Equation (47) shows that Δi  Δ3‐com depends linearly on A/

Ca with the slope proportional to 1/gm(¼rm), and the
intercept zero.
Comparing Eqs. (46) and (47) it can be seen that the slope is
identical, and they just differ in the intercept. Both approaches
result in the same calculation of gm, but we favor Eq. (47) where
the difference Δi  Δ3‐com is solely due to gm. Nevertheless the
alternative approach (Eq. 46) provides, besides gm, an estimation of
“most” of the respiratory and photorespiratory contribution.
In practice, with both approaches, gas and 13C isotope
exchange is measured under varying conditions resulting in a
range of A/Ca values (i.e. different irradiances or pCO2). Then
Δi  Δobs (Δobs ¼ Δ3‐com), where Δi can be calculated either with
Eq. (31) or (41), is plotted against A/Ca values, resulting in a line
with slope m. Subsequently the gm is calculated:
 
0 αb ℛd
1þt
1t b 3  a m  e
αe Aþℛd
gm ¼ ð48Þ
m

3.4 Calculation of ϕ Leakiness (ϕ) describes the proportion of carbon fixed by PEP
in C4 Species carboxylation that subsequently leaks out of the bundle-sheath
cells. Reported values for ϕ vary as much as 0.04–0.9 depending
on the method used for calculation and measurement conditions
(for a compilation of values and review of the methods see [117]). All
methods overlap with ϕ ¼ 0.2–0.3 for most species under regular
ambient conditions, and most report fluctuations in response to
short-term changes in irradiance [19, 28, 97, 118, 119]. Here we
focus on derivations of ϕ with C-isotope discrimination, which is the
method commonly used to calculate this variable.
Comparing Δobswith Δ4‐com (Eq. 18b), ϕ can be solved as
8 9
>
C bs  C m Δobs ð1  t ÞC a  a
ðC a  C i Þ  am ðC i  C m Þð1 þ t Þ  b 4 ð1 þ t ÞC m >
>
> a  >
>
< Cm ð1 þ t Þ½b 3 C bs  s ðC bs  C m Þ þ am ðC i  C m Þ þ a
ðC a  C i Þ  C a Δobs ð1  t Þ =
ϕ1 ¼
>

> b m bs
g ð C  C Þ þ A g ½ Δ ð 1  t ÞC  a
ð C  C Þ  b ð 1 þ t ÞC  þ ð 1 þ t ÞA ðb  a Þ >
>
>
:
i

m obs a

a i 4 i
 4 m >
;
g mC i  A g m ð1 þ t Þ b 3 C bs þ s C i  C bs þ a
ðC a  C i Þ  C a Δobs ð1  t Þ þ ð1 þ t ÞA ðam  s Þ
ð49Þ

where a) and b) are equivalent, but expressed either as a function of

Cm (a) or gm (b).
 On-line Measurements of 13C Discrimination 181

To date there is no formal analysis evaluating the effect of

ignoring ternary corrections (t) in the calculation of ϕ. Minor
effects of ignoring t when calculating ϕ were found across a range
of irradiances [28] and temperatures [84]. However, t could have
larger impacts under situations where the leaf to air vapor pressure
difference is large or variable. Therefore, and because the calcula-
tion of t is simple, we recommend including it in ϕ derivations.
Calculating ϕ with Eq. (49) requires variables that are difficult
to quantify for C4species. Because of the complexity and uncertain-
ties in Δ4‐com, it is tempting to use a simplified form (Fig. 3) and
then derive ϕ (Fig. 4) from that simplified equation. However, any
error introduced by the simplification will be absorbed in the
calculation of ϕ (for a discussion of these errors see [28, 117,
120]). Below, we discuss the simplifications that are applied to the
calculation of Δ4‐com to ease derivations of ϕ, namely (1) assuming
that the pCO2 in the bundle-sheath cells (Cbs) is large compared to
the pCO2 in the mesophyll (Cm), (2) assuming that mesophyll
conductance is infinite, and (3) modifications in the calculation
of b3and b4 that negate the need for additional modeling
(see Note 10).

3.4.1 The pCO2 With the enzyme limited model the Cbsis calculated as [49]
in the Bundle-Sheath Cells
V p  A  Rm
(Cbs) Is Large C bs ¼ C m þ
g bs
  
L γ ∗ O s þ K c 1 þ KOos Aþℛ
V cmax
d

¼ Cm þ ¼ ð50Þ
g bs 1  Aþℛ
V cmax
d

where variables’ definitions are in Table 1.

The Cbs is often tenfold higher than air concentrations [121],
and therefore much larger than the CO2 in the mesophyll (Cm).
Under these conditions Cbs/(Cbs  Cm)  1 and Cm/
(Cbs  Cm)  0 and Δ4‐com (Eq. 18b) can be simplified to
8     9
>
1 Ca  Ci 1þt Ci  Cm Cm >
>
> a a
þ am þ ðb 4 þ ϕ ðb 3  s ÞÞ >
>
>
> 1t 1t >
3>
> Ca Ca Ca >
< 2
A =
Δ4‐sim1 ¼     Ci 
>
>
a
1þt6 6 a
A a
g m7
7>>
>
> b þ 4 a  þ b þ ϕ ð b  s Þ  5>>
>
> 1  t 1  t
m
1 þ t g C
4 3
1 þ t C >
>
: m a a ;

ð51Þ

where a) and b) are equivalent, the only difference being that a) is

written in terms of Cmand b) in terms of gm. Note that Eq. (51b) is
182 Nerea Ubierna et al.

Fig. 4 Flowchart for selecting an expression to calculate leakiness (ϕ) depending on the measurement
conditions and simplifications applied in models of 13C discrimination during C4 photosynthesis. Sim.
1 ¼ simplification 1, which is Cbs/(Cbs  Cm)  1 and Cm/(Cbs  Cm)  0; Sim. 2 ¼ simplification
2, which is Vp/Vh  0; Sim. 3 ¼ simplification 3, which is gm is infinite and Ci ¼ Cm; Subs. 1 ¼ substitution
1, which consists of substituting b3 and b4 with b 3 and b 4 . The equations for the different expressions of ϕ are
ϕ1 (Eq. 49), ϕ2 (Eq. 52), ϕ3 (Eq. 53), ϕ4 and (Eq. 54). The ϕ2b and ϕ4b are calculated with Eqs. (52) and (54),
respectively, but substituting b3 and b4 with b 3 and b 4 . Calculations of b3, b4, b 3 , and b 4 are presented in the
main text in Eqs. (19), (20), (62), and (63), respectively

similar to that presented in von Caemmerer et al. [84] except that

the term (1 + t) dividing a
was omitted in that work.
Expressions for ϕ as a function of Cm (Eq. 52a) or gm (Eq. 52b)
for the situation when Cbs is large and Cbs/(Cbs  Cm)  1 and
Cm/(Cbs  Cm)  0 can be derived from Eq. (51):
 On-line Measurements of 13C Discrimination 183
8 9
>
> Δobs ð1  t ÞC a  a
ðC a  C i Þ >
>
>
>
 a C þ C ða  b 4 >
Þ >
>
> 1 þ t
m i m m
>
>
>
>a >
>
>
< C m ðb 3  s Þ   >
=
ϕ2 ¼ 1  t a
A

a C i
> Δobs   ða m  b 4 Þ  b4 
>
>
1þt
> 1þt g mC a 1 þ t Ca > >
>
>
>
> b   >
>
>
> Ci A >
>
>
:  ðb 3  s Þ >
;
C a g mC a
ð52Þ

There are two situations when this simplification is not recom-

mended because of the low Cbs predicted by the C4 photosynthetic
model: (1) low light intensities and (2) large bundle-sheath con-
ductance (gbs).
Under low irradiances, the fluxes through the C4 and C3 cycles
are small and therefore Cbsis low. In this case, it cannot be assumed
that Cbs is much larger than Cm and ϕ is overestimated when
calculated with the simplified Eq. (52) [28, 120].
The gbs cannot be directly measured but most estimated values
range between 0.01 and 0.2 μmol m2 s1 Pa1 [122, 123]. Meth-
ods to estimate gbs have been reviewed [117]. One of them calcu-
lates gbs as the value that minimizes the differences between
predicted and observed Δ13C [28, 97, 120]. This approach requires
assumptions about some of the variables included in the Δ13C
model. However, alternative methods to derive gbs are either diffi-
cult to implement or carry significant assumptions. The gbs has been
shown to be smaller in plants grown at low light [97, 118] and to
increase with temperature [124], but more studies are needed to
understand gbs variation. The difficulty in estimating gbs makes
accounting for gbs in ϕ calculations challenging.

3.4.2 Mesophyll If gm is assumed infinite (Ci ¼ Cm), then ϕ is calculated [28]:

Conductance (gm) Is Infinite
C bs  C i Δobs ð1  t ÞC a  a
ðC a  C i Þ  b 4 ð1 þ t ÞC i
ϕ3 ¼
Ci ð1 þ t Þ½b 3 C bs  s ðC bs  C i Þ þ a
ðC a  C i Þ  C a Δobs ð1  t Þ
ð53Þ

If gmis infinite, and additionally Cbsis much larger than Cm,

then ϕ calculation is [28]
1 Δobs ð1  t ÞC a  a
ðC a  C i Þ  ð1 þ t ÞC i b 4
ϕ4 ¼ ð54Þ
Ci ð1 þ t Þðb 3  s Þ

Values and sources of variation of C4-gm are still poorly under-

stood and data are limited because methods to estimate gm in C3
species [106] are not applicable for C4photosynthesis
(see Note 11). Therefore, almost all derivations to date of ϕ
184 Nerea Ubierna et al.

from Δ13C have assumed gm to be infinite [28, 118, 120] or

large [97].
Evaluating the impact of gmon Δ13C calculations in C4 species
is complicated due to the uncertainty in both gm and ϕ. Finite gm
compared with infinite results in either increased or decreased Δ13C
depending on whether ϕ is low (0.3) or high [66, 84]. Addition-
ally, the change in discrimination produced by gm is much lower in
C4than in C3 plants. For example, in C3 plants, when assimilation
flux is high, low gm can reduce Δ13C by 6–14‰ [50, 72, 84]. In
contrast, in C4 species low gm can change Δ13C by
3‰ for
extreme values of ϕ, but only 1‰ when ϕ ¼ 0.2 [84]. From
Cm ¼ Ci  A/gm, it is evident that the impact of including gm
should be minimal at low light intensities, when A is low and
Cm  Ci [120]. In a modeling exercise, assuming high irradiance,
ambient pCO2, and 25 C, ϕ calculated with Eq. (53) was mostly
insensitive to values of gm > 5 μmol m2 s1 Pa1, but sharply
declined for lower gm [84]. A recent study showed that ϕ estimated
with gm ¼ 2–24 μmol m2 s1 Pa1 was ~0.2 larger than ϕ
estimated assuming gm infinite [125]. The sensitivity of ϕ to gm
might be different under other assumptions, but currently there is
no formal study describing this variation. Given the potentially low
gm values and variation with environmental factors, we recommend
including gm in ϕ calculations whenever this information is avail-
able, though it is challenging at present due to the scarcity of data
on C4-gm.

3.4.3 Modifications of b3 In all the previous expressions for ϕ (Eqs. 49, 52, 53, and 54), b3
and b4 and b4are calculated with Eqs. (19) and (20), respectively. This
requires values for RuBisCO carboxylation (Vc), PEPC carboxyla-
tion (Vp), RuBisCO oxygenation (Vo), and hydration (Vh) rates,
which can be modeled with C4 photosynthesis equations [49]. In
certain conditions the following simplifications and substitutions
can be applied to avoid modelling these parameters:
1. If Vh is large, the ratio Vp/Vh  0 and b4 is simplified to

eℛm
b 4 ¼ b 04  ð55Þ
Vp

Large Vh is a fair assumption at ambient air pCO2, because the

conversion of CO2 into bicarbonate happens very fast when
catalyzed by carbonic anhydrase (CA) [126]. However, CA
could limit CO2 assimilation at low pCO2. Additional cases
where Vp/Vh might not be approximated to zero include trans-
genic plants with reduced amounts of CA (e.g. [19]) and some
C4 species with low CA content, such as C4 monocots [127].
2. C4 photosynthetic rate can be calculated both in terms of meso-
phyll (Eq. 56a) and bundle-sheath reactions (Eq. 56b) as [8, 49]
 On-line Measurements of 13C Discrimination 185


a
V c  0:5V o  ℛd
A¼ ð56Þ
b V p  L  ℛm ¼ V p ð1  ϕÞ  0:5ℛd

If assimilation flux is large and Vo is small compared with Vc,

Eq. (56a) can be solved for Vc as [118]
V c  A þ ℛd ð57Þ

From Eq. (56b), the Vp is [118]

A þ 0:5ℛd
Vp ¼ ð58Þ
1ϕ
Substituting in the equations for b3 (Eq. 19) and b4 (Eq. 20) the
Vc and Vp for the expressions given in Eqs. (57) and (58),
respectively, assuming Vp/Vh  0 and rewriting Vo/Vc as
2Γ∗/Cbs [49] results in formulations that do not include Vc,
Vp, Vo, or Vh:
eℛd f Γ∗
b 3‐sim ¼ b 03   ð59Þ
A þ ℛd C bs
e0:5ℛd
b 4‐sim ¼ b 04  ð1  ϕÞ ð60Þ
A þ 0:5ℛd
Equations (59) and (60) are valid when both hydration and
assimilation fluxes are large. A priori Eq. (60) can seem inconve-
nient, since it includes ϕ, which is the variable we aim to
calculate. However, introducing b3-sim and b4-sim into Δ13C
equations (for example into Eq. (51b)) and regrouping terms
result in
   
a
1þt a
A e0:5ℛd eℛd
Δ4‐sim1 ¼ þ am  þ b 04  þ ϕ b 03  þ
1t 1t 1 þ t g mC a A þ 0:5ℛd A þ ℛd
3
A
  Ci  ð61Þ
e0:5ℛd fΓ ∗
a
g m7
7
 s 
A þ 0:5ℛd C bs 1þt Ca 5

and this allows us to define b 3 and b 4 [84]:

eℛd e0:5ℛd f Γ∗
b 3 ¼ b 03  þ  ð62Þ
A þ ℛd A þ 0:5ℛd C bs
e0:5ℛd
b 4 ¼ b 04  ð63Þ
A þ 0:5ℛd

If Vo can be assumed 0, for example when measurements are

performed under 2% O2, then Eq. (62) can be further simplified
186 Nerea Ubierna et al.

∗
by assuming fCΓbs  0 [118]. Otherwise Cbs needs to be calculated
with Eq. (50). The b 3 and b 4 do not follow the original Farquhar
[8] mechanistic definition, but rather are convenient mathemat-
ical derivations that are valid when hydration and assimilation
rates are large. Substituting b 3 and b 4 into Eq. (61):
"   C  A#
a
1þt a
A
a
i gm
Δ4‐sim1 ¼ þ am  þ b4 þ ϕ b3  s 
1t 1t 1 þ t g mC a 1þt Ca
ð64Þ

which is the same as Eq. (51b) with the exception that b3 and b4
have been replaced by b 3 and b 4 . This is convenient because
solving ϕ from Eq. (64) results in the same expression as
Eq. (52) (ϕ2) but uses b 3 and b 4 instead of b3 and b4. The
same reasoning can be applied to Eq. (54) (ϕ4).
3. If fractionations during respiration (e) and photorespiration ( f )
are zero, or the ratios ℛd/Vc, ℛm/Vp, and Vo/Vc are very small,
then b3 and b4 simplify to b 03 ¼ 29–30‰ and b 04 ¼ 5.7‰ at
25 C but varies with temperature according to Eq. (27). The
values for e and f, discussed previously, are unlikely to be zero.
The e is weighted by ℛd/Vc or ℛm/Vp (Eqs. 19 and 20) and
these ratios are small when assimilation rate is large (and Vc and
Vp are large) as occurs under high irradiance (e.g. saturating
levels) or ambient pCO2. ℛd is the mitochondrial CO2 release
during illumination not associated with photorespiration. The
ℛd is often assumed to approximate measured rates of CO2
release in darkness (e.g. [28, 97]). ℛm is the proportion of ℛd
originating from mesophyll cells, calculated as 0.5 ℛd [49]. This
convenient approach could misrepresent actual ℛd [128] and
more complex techniques have been used to derive ℛd, such as
combinations of chlorophyll fluorescence and gas exchange
[119, 129, 130] or 13C labeling [131]. Nevertheless, it was
recently recommended approximating ℛd with dark respiration
until our knowledge of ℛd improves [61]. The ℛd increases
with temperature and this variation has been shown to result in
1‰ and 0.1 change in b3 and ϕ, respectively, in the range of
20–45 C [84]. Therefore, whenever assimilation flux is low or
temperature gradients are expected, ignoring respiratory frac-
tionation can result in errors in ϕ, especially if the value for e is
artificially enlarged by apparent respiratory fractionations
[28, 120].
The effect of f is to reduce b3 by the magnitude 0.5Vo/
Vc(¼Γ∗/Cbs). When the assimilation flux is large, Cbs is large
and Vo/Vc is small. Then the contribution to Δ13C by photores-
piration, expressed by the term f  Γ∗/Cbs, is small, and ignoring
it results in negligible changes in Δ13C (<0.2‰) [66, 84, 120].
However, as the flux decreases, f  Γ∗/Cbs increases, more so
 On-line Measurements of 13C Discrimination 187

when the pO2 around RuBisCO (Os) is large because Γ∗linearly

increases with pO2. When 21% O2 air is used for measurements,
Os can exceed ambient values in species with significant amounts
of photosystem II in the bundle-sheath cells because bundle-
sheath conductance to O2 is much less than to CO2 [49, 132,
133]. Photorespiratory contribution can be reduced by
performing measurements under 2% O2 atmosphere. In a mod-
eling exercise [84], at very low assimilation rates (Cbs ¼ 50 Pa),
values for f  Γ∗/Cbs ranged from 1.2 to 3.1‰ as Os increased
from 20,000 to 50,000 Pa. In these situations, ignoring the
contribution of photorespiratory fractionation will underesti-
mate ϕ.

3.4.4 How to Choose There is no clear-cut choice regarding how to select an expression
a Formulation for ϕ for ϕ. The user should evaluate the conditions of the experiment
pondering the error introduced by a simplification, which will be
expressed in the estimation of ϕ. Figure 4 is a flowchart for choos-
ing different formulations of ϕ according to the measurement
conditions and the simplifications that can be applied.
The equation with no simplifications (ϕ1, Eq. (49)) can be
applied in any situation, but requires extra modeling and calcula-
tions. When large assimilation flux can be expected, such as moder-
ately large light intensities and ambient pCO2 or larger, the
calculation can be simplified to ϕ2 (Eq. 52). When mesophyll
resistance is negligible (large gm), the expression to use is ϕ3
(Eq. 53), and if additionally there is large assimilation flux then
leakiness can be calculated with ϕ4 (Eq. 54). If both assimilation
and hydration rates are large, ϕ2 and ϕ4 can be further simplified by
replacing b3 and b4 with b 3 and b 4 . Finally, if fractionations
associated with respiration and photorespiration are negligible, for
example when assimilation flux is high, measurements are per-
formed at 2% O2, at constant temperature and pCO2, and there
are no apparent respiratory fractionations, then b 3  b 03 and
b 4  b 04 .

4 Notes

1. The α should have been defined as the 12C/13C ratio of the

product divided by that of the reactant. However, for historical
reasons it has been thought more convenient to think in
13
C/12C terms when measuring composition. Therefore to
obtain the desired α as defined in Eq. (1), one needs the
13
C/12C ratio of the reactant divided by that of the product.
2. The calculation of ζ uses 12CO2 mole fractions expressed by
mole of dry air (μmol 12CO2 mol dry air1). In any other
equation in this chapter, CO2 concentrations (i.e. Ca, Ci, Cm,
188 Nerea Ubierna et al.

Cc, Cbs) can be used either in units of partial pressures (Pa) or if

expressed in mole fractions they are per mol of wet air.
3. When calculating Δobs, some confusion is possible when
interpreting the per mil units. We provide an example here.
If ζ ¼ 10, δin ¼ 8.0‰, and δout ¼ 6.0‰, then
ð6ð8ÞÞ‰
Δobs ¼ 1þð0:00610Þ10ð0:006ð0:008ÞÞ ¼ 0:9940:02 ¼ 20:5‰.
20‰

4. Additional recommendations for TDLAS-gas exchange cou-

pling (Fig. 1): The TDLAS-gas exchange system is not pur-
chased as a kit, but rather it is a setup that the user needs to
carefully assemble. Here we provide information based on our
experience coupling a gas-exchange system (LI-6400, Li-Cor,
Lincoln, NE, USA) with a TDLAS (TGA 200A, Campbell
Scientific, Inc., Logan, UT, USA, or CO2 TILDAS-QCL
instrument, Aerodyne Research Inc., Billerica, MA, USA). We
include brand names for certain items in the setup that we have
tested and found optimal. But, these are not the only options
commercially available. Alternative brands may also be utilized,
but the users should test their performance. Recommendations
for the setup: (1) when air supply to the gas-exchange system is
from a pressurized tank, humidity needs to be added. This can be
achieved by circulating the air through a bubbler with vapor
saturation set by the water bath temperature, or else a commer-
cial dew point generator can be used; (2) the gas-exchange
system can use CO2 from small cartridges or a free-standing
tank, both of which are available with a range of δ13C values. It
is recommended to use a CO2 supply with similar δ13C values to
those in the air where the plant grew in order to avoid large
apparent respiratory fractionation. Unfortunately, the δ13C
values of small cartridges or ordinary tanks are often unknown
a priori, unless a calibrated tank is used. As a general rule, the
CO2 derived from industrial processes is often highly depleted in
13
C, whereas laser- or food-grade CO2 tends to have a composi-
tion closer to atmospheric. Nevertheless, the CO2 supply should
be analyzed for δ13C; (3) if leaf temperature is varied during the
experiment, condensation in the lines coming out of the leaf
cuvette should be avoided. The user should check the dew point
temperature at which water vapor will condense. For example, if
working with a LiCor-6400, a quick way to determine the dew
point temperature at which water vapor will condense is to look
at “Td_S_ C”. The room or controlled environment chamber
should always be above this temperature. Gas lines out of the
controlled environment chamber that are not under suction
should be heated, for example with electrical heating tape;
(4) if using a TDL (TGA 200A, Campbell Scientific, Inc.,
Logan, UT, USA), sufficient flow from the gas exchange lines
should be diverted to the analyzer because the instrument relies
on having excess flow to regulate pressure. However, care
 On-line Measurements of 13C Discrimination 189

should be taken not to disturb pressure within the leaf chamber;

(5) leaks can be avoided by sealing leaf chamber gaskets with
insulating material, for example with Terostat putty (Terostat IX,
Henkel Technologies, Düsseldorf, Germany). If other putty is
used, it should be checked that it effectively prevents CO2
diffusion. Leak proofing the leaf chamber and pipe connections
can be done by blowing through a small length of pipe around
the gaskets and joints and checking whether that results in a CO2
spike. Best results are obtained when the chamber is supplied
with CO2-free air. Then, the leaf chamber should have no irra-
diance during the leak testing so as not to stress the leaf. Pipe
joints can also be bubble-tested with detergent products such as
Snoop, but do not use this in the gas-exchange system because
water will damage the instrument; (6) tubing for connections
should have low permeability to CO2. Often, ultralow-porosity
Type 1300 Synflex with brass Swagelok compression fittings is
used (Saint Gobain Performance Plastics, Northboro, MA,
USA); (7) it is not recommended to use anhydrous calcium
sulfate to dry the air before entering the TDLAS; a low-flow
Nafion counterflow water trap (i.e. PD625 dual configurations,
Campbell Scientific), dry-ice traps, or a series or combination of
drying loops can be used; (8) to generate a series of reference
gases with the same isotope mixing ratio but different CO2
concentrations, the TDL uses a capillary gas mixing system,
but custom-built settings are also feasible.
5. A typical measuring cycle can be as follows (see Fig. 1 in [25]): a
N2/O2 gas (zero), calibration gas #1 measured at six different
concentrations, followed by a calibration gas #2, and then gas
exchange reference (inlet) and sample (outlet) lines. Each gas
can be measured for anywhere from 20 to 40 s (Allan variance
used to optimize sampling duration for instrument noise and
drift—refer to Subheading 2, item 4). For example if 20 s
intervals are used then the entire cycle will last 200 s (¼10
lines 20 s per line). For calculations, only the data collected
during the last 10 s are used to ensure complete turnover of the
gas inside the optical cell after switching lines.
6. For the calculation of t, the units for E and gac are mol m2 s1.
To calculate αac ¼ 1 þ a,
for example if a
¼ 4:4 ‰, then
αac ¼ 1 + 0.0044 ¼ 1.0044. The same applies to the calculation
of αb, αe, αf, and αam. Equivalent terms that have been used to
refer to am are ai and aw.
7. In models of C3 Δ13C, the term b3 has been used for RuBisCO
fractionation. However, in models of C4 Δ13C, b3 often refers
to the combined effects of RuBisCO fractionation and fractio-
nations associated with respiration and photorespiration. Here
we chose to use b 03 for the fractionation during carboxylation
(Eq. 25) and b3 for the combined effects of RuBisCO,
190 Nerea Ubierna et al.

respiration, and photorespiration. With this approach b3 and b 03

are the same in the models of C3 and C4 Δ13C.
8. In the calculation of gmwith Eqs. (43) or (44), the Ca is
expressed in units of partial pressures. The gm should be
derived from a drawdown in partial pressures so it is indepen-
dent of the temperature sensitivity of the solubility of CO2.
Using SI units, the Ca is in Pa, and gm in μmol m2 s1 Pa1. If
Ca is used in mole fractions (μmol mol1) then the entire
equation needs to be multiplied by 1/P, where P is atmospheric
pressure. Many studies report gm in units of mol m2 s1 bar1.
The conversion from mol m2 s1 bar1 to μmol m2 s1 Pa1
is   10.
9. The intercept of Eq. (46) (Δe + Δf) includes “most” of the
respiratory contribution, with the missing part being
1þt αb eℛd A
1t αe Aþℛd g C a .
m

10. There is one more simplification that could be applied to all the
expressions for Δ4‐com, and that is to assume that boundary
layer conductance (gbl) is infinite. Then a
can be replaced by as
(4.4‰). Any leaf cuvette should have large gbl. Occasionally, gbl
values might be compromised when custom-built leaf cham-
bers are used. Including gbl in calculations is so simple that
there is really not much of an advantage accrued by ignoring it.
11. In C4 species, gm cannot be derived from 13C discrimination
because Δ4‐com includes two unknowns, gm and ϕ. Recently,
C4-gm values have been derived with two methods: 18O-dis-
crimination [57–59, 134] and in vitro Vpmax [59, 135]. Values
of gm derived with both methods (1) were similar across tem-
peratures 10–40 C [59], (2) corresponded to the high end of
gm values reported for C3 species [58, 59, 134], and (3) varied
with leaf age [58] and temperature [59] in a similar fashion to
C3 observations.

References
1. Farquhar GD, Richards RA (1984) Isotopic
composition of plant carbon correlates with
water-use efficiency of wheat genotypes. Aust
J Plant Physiol 11(6):539–552
2. Craig H (1957) Isotopic standards for carbon
and oxygen and correction factors for mass-
spectrometric analysis of carbon dioxide.
Geochim Cosmochim Acta 12:133–149
3. Zhang QL, Li WJ (1990) A calibrated mea-
surement of the atomic weight of carbon.
Chin Sci Bull 35:290–296
4. Tans PP, Crotwell AM, Thoning KW (2017)
Abundances of isotopologues and calibration
of CO2 greenhouse gas measurements. Atmos
Meas Tech 10:2669–2685
5. Griffis TJ, Baker JM, Sargent SD, Tanner BD,
Zhang J (2004) Measuring field-scale isotopic
CO2 fluxes with tunable diode laser absorp-
tion spectroscopy and micrometeorological
techniques. Agric For Meteorol 124
(1–2):15–29
6. Evans JR, Sharkey TD, Berry JA, Farquhar
GD (1986) Carbon isotope discrimination
measured concurrently with gas-exchange to
investigate CO2 diffusion in leaves of higher-
plants. Aust J Plant Physiol 13(2):281–292
 On-line Measurements of 13C Discrimination
7. Farquhar GD, O’Leary MH, Berry JA (1982)
On the relationship between carbon isotope
discrimination and the intercellular carbon
dioxide concentration in leaves. Aust J Plant
Physiol 9(2):121–137
8. Farquhar GD (1983) On the nature of carbon
isotope discrimination in C4 species. Aust J
Plant Physiol 10(2):205–226
9. Farquhar GD, Cernusak LA (2012) Ternary
effects on the gas exchange of isotopologues
of carbon dioxide. Plant Cell Environ 35
(7):1221–1231
10. Bender MM (1968) Mass spectrometric stud-
ies of 13C variations in corn and other grasses.
Radiocarbon 10(2):468–472
11. Troughton JH (1979) δ13C as an inidcator of
carboxylation reactions. In: Latzko GM
(ed) Encyclopedia of plant physiologhy, vol
6. Springer, Berlin/New York, pp 140–147
12. Winter K (1981) CO2 and water vapour
exchange, malate content and δ13C value in
Cicer arietinum grown under two water
regimes. Zeitschrift für Pflanzenphysiologie
101:421–430
13. Guy RD, Reid DM, Krouse HR (1980) Shifts
in carbon isotope ratios of two C3 halophytes
under natural and artificial conditions. Oeco-
logia 44:241–247
14. Park R, Epstein S (1961) Metabolic fraction-
ation of C13 and C12 in plants. Plant Physiol
36(2):133–138
15. Melander L, Saunders WH (1979) Reaction
rates of isotopic molecules. John Wiley and
Sons, New York
16. O’Leary MH (1981) Carbon isotope frac-
tionation in plants. Phytochemistry 20
(4):553–567
17. Cernusak LA, Farquhar GD, Wong SC,
Stuart-Williams H (2004) Measurement and
interpretation of the oxygen isotope composi-
tion of carbon dioxide respired by leaves in the
dark. Plant Physiol 136(2):3350–3363
18. Ribas-Carbó M, Still C, Berry J (2002) Auto-
mated system for simultaneous analysis of
δ13C, δ18O and CO2 concentrations in small
air samples. Rapid Commun Mass Spectrom
16:339–345
19. Cousins AB, Badger MR, von Caemmerer S
(2006) Carbonic anhydrase and its influence
on carbon isotope discrimination during C4
photosynthesis. Insights from antisense RNA
in Flaveria bidentis. Plant Physiol 141
(1):232–242
20. Schnyder H, Sch€aufele R, Lötscher M, Geb-
bing T (2003) Disentangling CO2 fluxes:
direct measurements of mesocosm-scale natu-
ral abundance 13CO2/12CO2 gas exchange,
191
13
C discrimination, and labelling of CO2
exchange flux components in controlled
environments. Plant Cell Environ 26
(11):1863–1874
21. Klumpp K, Sch€aufele R, Lötscher M, Lattanzi
FA, Feneis W, Schnyder H (2005) C-isotope
composition of CO2 respired by shoots and
roots: fractionation during dark respiration?
Plant Cell Environ 28(2):241–250
22. Barbour MM, McDowell NG, Tcherkez G,
Bickford CP, Hanson DT (2007) A new mea-
surement technique reveals rapid post-
illumination changes in the carbon isotope
composition of leaf-respired CO2. Plant Cell
Environ 30(4):469–482
23. Tuzson B, Mohn J, Zeeman MJ, Werner RA,
Eugste W, Zahniser MS, Nelson DD, McMa-
nus JB, Emmenegger L (2008) High preci-
sion and continuous field measurements of
δ13C and δ18O in carbon dioxide with a
cryogen-free QCLAS. Appl Phys B
92:451–458
24. Berryman EM, Marshall JD, Rahn T, Cook
SP, Litvak M (2011) Adaptation of
continuous-flow cavity ring-down spectros-
copy for batch analysis of δ13C of CO2 and
comparison with isotope ratio mass spectrom-
etry. Rapid Commun Mass Spectrom
25:2355–2360
25. Tazoe Y, von Caemmerer S, Estavillo GM,
Evans JR (2011) Using tunable diode laser
spectroscopy to measure carbon isotope dis-
crimination and mesophyll conductance to
CO2 diffusion dynamically at different CO2
concentrations. Plant Cell Environ
34:580–591
26. Evans JR, von Caemmerer S (2013) Temper-
ature response of carbon isotope discrimina-
tion and mesophyll conductance in tobacco.
Plant Cell Environ 36:745–756
27. Walker B, Ariza LS, Kaines S, Badger MR,
Cousins AB (2013) Temperature response of
in vivo Rubisco kinetics and mesophyll con-
ductance in Arabidopsis thaliana: compari-
sons to Nicotiana tabacum. Plant Cell
Environ 36:2108–2119
28. Ubierna N, Sun W, Kramer DM, Cousins AB
(2013) The efficiency of C4 photosynthesis
under low light conditions in Zea mays, Mis-
canthus x giganteus and Flaveria bidentis.
Plant Cell Environ 36:365–381
29. Lehmann MM, Wegener F, Barthel M, Maur-
ino VG, Siegwolf RTW, Buchmann N,
Werner C, Werner RA (2016) Metabolic fate
of the carboxyl groups of malate and pyruvate
and their influence on δ13C of leaf-respired
CO2 during light enhanced dark respiration.
Front Plant Sci 7:739
192 Nerea Ubierna et al.
30. Hanson DT, Collins AM, Howland DTJ,
Roesgen J, López-Nieves S, Timlin JA
(2014) On-line stable isotope gas exchange
reveals an inducible but leaky carbon concen-
trating mechanism in Nannochloropsis salina.
Photosynth Res 121:311–322
31. Bickford CP, McDowell NG, Erhardt EB,
Hanson DT (2009) High-frequency field
measurements of diurnal carbon isotope dis-
crimination and internal conductance in a
semi-arid species, Juniperus monosperma.
Plant Cell Environ 32(7):796–810
32. Bowling DR, Sargent SD, Tanner BD, Ehler-
inger JR (2003) Tunable diode laser absorp-
tion spectroscopy for stable isotope studies of
ecosystem-atmosphere CO2 exchange. Agric
For Meteorol 118(1–2):1–19
33. Gentsch L, Hammerle A, Sturm P, Ogée J,
Wingate L, Siegwolf R, Plüss P, Baur T,
Buchmann N, Knohl A (2014) Carbon iso-
tope discrimination during branch photosyn-
thesis of Fagus sylvatica: a Bayesian modelling
approach. Plant Cell Environ 37
(7):1516–1535
34. Wehr R, Munger JW, McManus JB, Nelson
DD, Zahniser MS, Davidson EA, Wofsy SC,
Saleska SR (2016) Seasonality of temperate
forest photosynthesis and daytime respiration.
Nature 534:680–683
35. Dawson TE, Mambelli S, Plamboeck AH,
Templer PH, Tu KP (2002) Stable isotopes
in plant ecology. Annu Rev Ecol Syst
33:507–559
36. Marshall JD, Brooks JR, Lajtha K (2007)
Sources of variation in the stable isotopic
composition of plants. In: Michener R, Lajtha
K (eds) Stable isotopes in ecology and envi-
ronmental science. Wiley-Blackwell, Malden,
MA, pp 22–60
37. Farquhar GD, Ehleringer JR, Hubick KT
(1989) Carbon isotope discrimination and
photosynthesis. Annu Rev Plant Physiol
Plant Mol Biol 40:503–537
38. Farquhar GD, Hubick KT, Condon AG,
Richards RA (1989) Carbon isotope discrimi-
nation and water-use efficiency. In: Rundel
PW, Ehleringer JR, Nagy KA (eds) Stable iso-
topes in ecological research. Springer-Verlag,
New York, pp 21–46
39. Fry B (2006) Stable isotope ecology.
Springer-Verlag
40. Cernusak LA, Ubierna N, Winter K, Holtum
JAM, Marshall JD, Farquhar GD (2013)
Environmental and physiological determi-
nants of carbon isotope discrimination in ter-
restrial plants. New Phytol 200:950–965
41. Ehleringer JR, Rundel PW (1989) Stable Iso-
topes: history, units, and instrumentation. In:
Rundel PW, Ehleringer JR, Nagy KA (eds)
Stable isotopes in ecological research. Ecolog-
ical studies (analysis and synthesis), vol 68.
Springer, New York, NY
42. Tieszen LL, Boutton TW (1989) Stable car-
bon isotopes in terrestrial ecosystem research.
In: Rundel PW, Ehleringer JR, Nagy KA (eds)
Stable isotopes in ecological research, Ecolog-
ical studies (analysis and synthesis), vol 68.
Springer, New York, NY, pp 167–195
43. Griffiths H (1998) Stable isotopes: integra-
tion of biological, ecological and geochemical
processes, Environmental plant biology series.
BIOS Scientific Publishers, Oxford
44. McCarroll D, Loader NJ (2004) Stable iso-
topes in tree rings. Quat Sci Rev 23
(7–8):771–801
45. Ellsworth PZ, Cousins AB (2016) Carbon
isotopes and water use efficiency in C4 plants.
Curr Opin Plant Biol 31:155–161
46. Wahl EH, Fidric B, Rella CW, Koulokov S,
Kharlamov B, Taz S, Kachanov AA, Richman
BA, Crosson ER, Paldus BA, Kalaskar S,
Bowling DR (2006) Applications of cavity
ring-down spectroscopy to high precision iso-
tope ratio measurement of 13C/12C in carbon
dioxide. Isotopes Environ Health Stud
42:21–35
47. McManus JB, Nelson DD, Shorter JH,
Jimenez R, Herdon S, Saleska SR, Zahniser
MS (2005) A high precision pulsed quantum
cascade laser spectrometer for measurements
of stable isotopes of carbon dioxide. J Mod
Opt 52(16):2309–2321
48. Berden G, Peeters R, Meijer G (2000) Cavity
ring-down spectroscopy: experimental
schemes and applications. Int Rev Phys
Chem 19:565–607
49. von Caemmerer S (2000) Biochemical models
of leaf photosynthesis. CSIRO Publishing,
Collingwood
50. Ubierna N, Farquhar GD (2014) Advances in
measurements and models of photosynthetic
carbon isotope discrimination in C3 plants.
Plant Cell Environ 37:1494–1498
51. Wingate L, Seibt U, Moncrieff JB, Jarvis PG,
Lloyd J (2007) Variations in 13C discrimina-
tion during CO2 exchange by Picea sitchensis
branches in the field. Plant Cell Environ 30
(5):600–616
52. Drake BL (2014) Using models of carbon
isotope fractionation during photosynthesis
to understand the natural fractionation ratio.
Radiocarbon 56:29–38
 On-line Measurements of 13C Discrimination
53. Tholen D, Ethier G, Genty B, Pepin S, Zhu
XG (2012) Variable mesophyll conductance
revisited. Theoretical background and experi-
mental implications. Plant Cell Environ
35:2087–2103
54. Tholen D, Zhu XG (2011) The mechanistic
basis of internal conductance: a theoretical
analysis of mesophyll cell photosynthesis and
CO2 diffusion. Plant Physiol 156:90–105
55. Busch FA, Sage TL, Cousins AB, Sage RF
(2013) C3 plants enhance rates of photosyn-
thesis by reassimilating photorespired and
respired CO2. Plant Cell Environ 36:200–212
56. Gu L, Sun Y (2014) Artefactual responses of
mesophyll conductance to CO2 and irradiance
estimated with the variable J and online iso-
tope discrimination methods. Plant Cell Envi-
ron 37:1231–1249
57. Gillon JS, Yakir D (2000) Internal conduc-
tance to CO2 diffusion and C18OO discrimi-
nation in C3 leaves. Plant Physiol 123
(1):201–213
58. Barbour MM, Evans JR, Simonin KA, von
Caemmerer S (2016) Online CO2 and H2O
oxygen isotope fractionation allows estima-
tion of mesophyll conductance in C4 plants,
and reveals that mesophyll conductance
decreases as leaves age in both C4 and C3
plants. New Phytol 210(3):875–889
59. Ubierna N, Gandin A, Boyd RA, Cousins AB
(2017) Temperature response of mesophyll
conductance in three C4 species calculated
with two methods: 18O discrimination and
in-vitro Vpmax. Corrigendum. New Phytol
217:956–959
60. Loucos KE, Simonin KA, Barbour MM
(2017) Leaf hydraulic conductance and meso-
phyll conductance are not closely related
within a single species. Plant Cell Environ
40:203–215
61. Farquhar GD, Busch FA (2017) Changes in
the chloroplastic CO2 concentration explain
much of the observed Kok effect: a model.
New Phytol 214:570–584
62. Craig H (1953) The geochemistry of the sta-
ble carbon isotopes. Geochim Cosmochim
Acta 3:53–92
63. Mook WG, Bommerson JC, Staverman WH
(1974) Carbon isotope fractionations
between dissolved bicarbonate and gaseous
carbon dioxide. Earth Planet Sci Lett
22:169–176
64. O’Leary MH (1984) Measurement of the iso-
tope fractionation associated with diffusion of
carbon dioxide in aqueous solution. J Phys
Chem 88(4):823–825
193
65. Zeebe RE (2011) On the molecular diffusion
coefficients of dissolved CO2, HCO3, and
CO32 and their dependence on isotopic
mass. Geochim Cosmochim Acta
75:2483–2498
66. Henderson SA, von Caemmerer S, Farquhar
GD (1992) Short-term measurements of car-
bon isotope discrimination in several C4 spe-
cies. Aust J Plant Physiol 19(3):263–285
67. von Caemmerer S, Ludwig M, Millgate A,
Farquhar GD, Price GD, Badger M, Furbank
RT (1997) Carbon isotope discrimination
during C4 photosynthesis: insights from
transgenic plants. Aust J Plant Physiol 24
(4):487–494
68. Roeske CA, O’Leary MH (1984) Carbon iso-
tope effects on the enzyme-catalyzed carbox-
ylation of ribulose bisphosphate.
Biochemistry 23(25):6275–6284
69. Guy RD, Fogel ML, Berry JA (1993) Photo-
synthetic fractionation of the stable isotopes
of oxygen and carbon. Plant Physiol 101
(1):37–47
70. Guy RD, Fogel MF, Berry JA, Hoering TC
(1987) Isotope fractionation during oxygen
production and consumption by plants. In:
Biggins J (ed) Progress in photosynthesis
research III. Martinus Nijhoff, Dordrecht,
pp 597–600
71. McNevin DB, Badger MR, Whitney SM, von
Caemmerer S, Tcherkez G, Farquhar GD
(2007) Differences in carbon isotope discrim-
ination of three variants of d-ribulose-1,5-
bisphosphate carboxylase/oxygenase reflect
differences in their catalytic mechanisms. J
Biol Chem 282:36068–36076
72. von Caemmerer S, Evans JR (1991) Determi-
nation of the average partial-pressure of CO2
in chloroplasts from leaves of several C3
plants. Aust J Plant Physiol 18(3):287–305
73. Vu J, Yelenosky G, Bausher MG (1985) Pho-
tosynthetic activity in the flower buds of
‘Valencia’ orange (Citrus sinensis [L.]
Osbeck). Plant Physiol 78:420–423
74. Saurer M, Maurer S, Matyssek R, Landolt W,
Günthardt-Goerg MS, Siegenthaler U (1995)
The influence of ozone and nutrition on δ13C
in Betula pendula. Oecologia 103:397–406
75. Raven JA, Farquhar GD (1990) The influence
of N metabolism and organic acid synthesis on
the natural abundance of isotopes of carbon in
plants. New Phytol 116:505–529
76. Douthe C, Dreyer E, Brendel O, Warren CR
(2012) Is mesophyll conductance to CO2 in
leaves of three Eucalyptus species sensitive to
short-term changes of irradiance under
194 Nerea Ubierna et al.
ambient as well as low O2? Funct Plant Biol
38:434–447
77. Evans JR, von Caemmerer S, Setchell BA,
Hudson GS (1994) The relationship between
CO2 transfer conductance and leaf anatomy in
transgenic tobacco with a reduced content of
Rubisco. Aust J Plant Physiol 21(4):475–495
78. O’Leary MH, Madhavan S, Paneth P (1992)
Physical and chemical basis of carbon isotope
fractionation in plants. Plant Cell Environ 15
(9):1099–1104
79. Tcherkez G, Farquhar GD (2005) Carbon
isotope effect predictions for enzymes
involved in the primary carbon metabolism
of plant leaves. Funct Plant Biol 32
(4):277–291
80. Whelan T, Sackett WM, Benedict CR (1973)
Enzymatic fractionation of carbon isotopes by
phosphoenolpyruvate carboxylase from
C4-plants. Plant Physiol 51:1051–1054
81. von Caemmerer S, Tazoe Y, Evans JR, Whit-
ney SM (2014) Exploiting transplastomically
modified Rubisco to rapidly measure natural
diversity in its carbon isotope discrimination
using tuneable diode laser spectroscopy. J Exp
Bot 65(13):3759–3767
82. Farquhar G, Ball MC, von Caemmerer S,
Roksandic Z (1982) Effect of salinity and
humidity on δ13C value of halophytes-
evidence for diffusional isotope fractionation
determined by the ratio of intercellular/
atmospheric partial pressure of CO2 under
different environmental conditions. Oecolo-
gia 52:121–124
83. Ubierna N, Marshall JD (2011) Estimation of
canopy average mesophyll conductance using
δ13C of phloem contents. Plant Cell Environ
34:1521–1535
84. von Caemmerer S, Ghannoum O, Pengelly
JJL, Cousins AB (2014) Carbon isotope dis-
crimination as a tool to explore C4 photosyn-
thesis. J Exp Bot 65(13):3459–3470
85. Gillon JS, Griffiths H (1997) The influence of
(photo)respiration on carbon isotope discrim-
ination in plants. Plant Cell Environ 20
(10):1217–1230
86. Lanigan GJ, Betson N, Griffiths H, Seibt U
(2008) Carbon isotope fractionation during
photorespiration and carboxylation in Senecio.
Plant Physiol 148(4):2013–2020
87. Rooney MA (1988) Short-term carbon iso-
tope fractionation by plants. PhD thesis, Uni-
versity of Wisconsin, Madison, WI
88. Igamberdiev AU, Mikkelsen TN, Ambus P,
Bauwe H, Lea PJ, Gardestrom P (2004) Pho-
torespiration contributes to stomatal regula-
tion and carbon isotope fractionation: a study
with barley, potato and Arabidopsis plants
deficient in glycine decarboxylase. Photosynth
Res 81:139–152
89. Tcherkez G (2006) How large is the carbon
isotope fractionation of the photorespiratory
enzyme glycine decarboxylase? Funct Plant
Biol 33(10):911–920
90. Tcherkez G, Schaufele R, Nogués S, Piel C,
Boom A, Lanigan G, Barbaroux C, Mata C,
Elhani S, Hemming D (2010) On the
13
C/12C isotopic signal of day and night res-
piration at the mesocosm level. Plant Cell
Environ 33:900–913
91. Tcherkez G, Farquhar G, Badeck F, Ghash-
ghaie J (2004) Theoretical considerations
about carbon isotope distribution in glucose
of C3 plants. Funct Plant Biol 31(9):857–877
92. Tcherkez G, Mauve C, Lamothe M, Le
Bras C, Grapin A (2011) The 13C/12C isoto-
pic signal of day-respired CO2 in variegated
leaves of Pelargonium  hortorum. Plant Cell
Environ 34:270–283
93. Ghashghaie J, Duranceau M, Badeck FW,
Cornic G, Adeline MT, Deleens E (2001)
δ13C of CO2 respired in the dark in relation
to δ13C of leaf metabolites: comparison
between Nicotiana sylvestris and Helianthus
annuus under drought. Plant Cell Environ
24(5):505–515
94. Bathellier C, Badeck FW, Couzi P, Harscoet S,
Mauve C, Ghashghaie J (2008) Divergence in
δ13C of dark respired CO2 and bulk organic
matter occurs during the transition between
heterotrophy and autotrophy in Phaseolus vul-
garis plants. New Phytol 177(2):406–418
95. Sun W, Resco V, Williams DG (2010) Noc-
turnal and seasonal patterns of carbon isotope
composition of leaf dark-respired carbon
dioxide differ among dominant species in a
semiarid savanna. Oecologia 164(2):297–310
96. Stutz S, Edwards GE, Cousins AB (2014)
Single-cell C4 photosynthesis: efficiency and
acclimation of Bienertia sinuspersici to growth
under low light. Plant Physiol 202:220–234
97. Kromdijk J, Griffiths H, Schepers HE (2010)
Can the progressive increasse of C4 bundle-
sheath leakiness at low PFD be explained by
incomplete supression of photorespiration?
Plant Cell Environ 33:1935–1948
98. Evans JR, von Caemmerer S (1996) Carbon
dioxide diffusion inside leaves. Plant Physiol
110(2):339–346
99. Flexas J, Ribas-Carbó M, Dı́az-Espejo A,
Galmés J, Medrano H (2008) Mesophyll con-
ductance to CO2: current knowledge and
future prospects. Plant Cell Environ 31
(5):602–621

100. Ethier GJ, Livingston NJ (2004) On the need
to incorporate sensitivity to CO2 transfer con-
ductance into the Farquhar-von Caemmerer-
Berry leaf photosynthesis model. Plant Cell
Environ 27(2):137–153
101. Niinemets U, € Dı́az-Espejo A, Flexas J,
Galmés J, Warren CR (2009) Role of meso-
phyll diffusion conductance in constraining
potential photosynthetic productivity in the
field. J Exp Bot 60(8):2249–2270
102. Piel C, Frak E, Le Roux X, Genty B (2002)
Effect of local irradiance on CO2 transfer con-
ductance of mesophyll in walnut. J Exp Bot
53(379):2423–2430
103. Pérez-Martı́n A, Flexas J, Ribas-Carbó M,
Bota J, Tomas M, Infante JM, Dı́az-Espejo
A (2009) Interactive effects of soil water defi-
cit and air vapour pressure deficit on meso-
phyll conductance to CO2 in Vitis vinifera
and Olea europaea. J Exp Bot 60:2391–2405
104. Loreto F, Centritto M, Chartzoulakis K
(2003) Photosynthetic limitations in olive
cultivars with different sensitivity to salt stress.
Plant Cell Environ 26(4):595–601
105. von Caemmerer S, Evans JR (2015) Temper-
ature responses of mesophyll conductance dif-
fer greatly between species. Plant Cell
Environ 38:629–637
106. Pons TL, Flexas J, von Caemmerer S, Evans
JR, Genty B, Ribas-Carbó M, Brugnoli E
(2009) Estimating mesophyll conductance
to CO2: methodology, potential errors, and
recommendations. J Exp Bot
(8):2217–2234
107. Warren CR (2008) Stand aside stomata,
another actor deserves centre stage: the for-
gotten role of the internal conductance to
CO2 transfer. J Exp Bot 59:1475–1148
108. Loreto F, Harley PC, Dimarco G, Sharkey TD
(1992) Estimation of mesophyll conductance
to CO2 flux by three different methods. Plant
Physiol 98(4):1437–1443
109. Sharkey TD, Vassey TL, Vanderveer PJ, Vier-
stra RD (1991) Carbon metabolism enzymes
and photosynthesis in transgenic tobacco
(Nicotiana tabaccum L.) having excess phy-
tochrome. Planta 185:287–296
110. Lloyd J, Syvertsen JP, Kriedemann PE, Farqu-
har GD (1992) Low conductances for CO2
diffusion from stomata to the sites of carbox-
ylation in leaves of woody species. Plant Cell
Environ 15(8):873–899
111. Flexas J, Ribas-Carbó M, Hanson DT, Bota J,
Otto B, Cifre J, McDowell N, Medrano H,
Kaldenhoff R (2006) Tobacco aquaporin
NtAQP1 is involved in mesophyll conduc-
tance to CO2 in vivo. Plant J 48(3):427–439
On-line Measurements of 13C Discrimination 195
112. Hassiotou F, Ludwig M, Renton M, Venek-
laas EJ, Evans JR (2009) Influence of leaf dry
mass per area, CO2, and irradiance on meso-
phyll conductance in sclerophylls. J Exp Bot
60(8):2303–2314
113. Warren CR, Low M, Matyssek R, Tausz M
(2007) Internal conductance to CO2 transfer
of adult Fagus sylvatica: variation between sun
and shade leaves and due to free-air ozone
fumigation. Environ Exp Bot 59(2):130–138
114. Brugnoli E, Lauteri M, Guido MC (1994)
Carbon isotope discrimination and photosyn-
thesis: response and adaptation to environ-
mental stress. In: de Kouchkovsky Y, Larher
F (eds) Plant sciences, second general collo-
quium on plant sciences. Renners, SFPV,
Universite de Renners, pp 269–272
115. Scartazza A, Lauteri M, Guido MC, Brugnoli
E (1998) Carbon isotope discrimination in
leaf and stem sugars, water-use efficiency and
mesophyll conductance during different
developmental stages in rice subjected to
drought. Aust J Plant Physiol 25:489–498
116. Monti A, Brugnoli E, Scartazza A, Amaducci
MT (2006) The effect of transient and con-
tinuous drought on yield, photosynthesis and
carbon isotope discrimination in sugar beet
(Beta vulgaris L). J Exp Bot 57
(6):1253–1262
117. Kromdijk J, Ubierna N, Cousins AB, Griffiths
H (2014) Bundle-sheath leakiness in C4 pho-
tosynthesis: a careful balancing act between
60 CO2 concentration and assimilation. J Exp
Bot 65(13):3443–3457
118. Pengelly JJL, Sirault XRR, Tazoe Y, Evans JR,
Furbank RT, von Caemmerer S (2010)
Growth of the C4 dicot Flaveria bidentis:
photosynthetic acclimation to low light
through shifts in leaf anatomy and biochemis-
try. J Exp Bot 61(14):4109–4122
119. Bellasio C, Griffiths H (2014) Acclimation to
low light by C4 maize: implications for bundle
sheath leakiness. Plant Cell Environ 37
(5):1046–1058
120. Ubierna N, Sun W, Cousins AB (2011) The
efficiency of C4 photosynthesis under low
light conditions: assumptions and calculations
with CO2 isotope discrimination. J Exp Bot
62(9):3119–3134
121. Furbank RT, Hatch MD (1987) Mechanism
of C4 photosynthesis - the size and composi-
tion of the inorganic carbon pool in bundle-
sheath cells. Plant Physiol 85(4):958–964
122. He DX, Edwards GE (1996) Estimation of
diffusive resistance of bundle-sheath cells to
CO2 from modeling of C4 photosynthesis.
Photosynth Res 49(3):195–208
196 Nerea Ubierna et al.
123. von Caemmerer S, Furbank RT (2003) The
C4 pathway: an efficient CO2 pump. Photo-
synth Res 77(2–3):191–207
124. Yin X, van der Putten PEL, Driever SM,
Struik PC (2016) Temperature response of
bundle-sheath conductance in maize leaves. J
Exp Bot 67(9):2699–2714
125. Gong XY, Sch€aufele R, Schnyder H (2016)
Bundle-sheath leakiness and intrinsic water
use efficiency of a perennial C4 grass are
increased at high vapour pressure deficit dur-
ing growth. J Exp Bot 68(2):321–333
126. Stryer L (1988) Biochemistry. W. H. Free-
man, New York
127. Cousins AB, Badger MR, von Caemmerer S
(2008) C4 photosynthetic isotope exchange
in NAD-ME- and NADP-ME-type grasses. J
Exp Bot 59(7):1695–1703
128. Tcherkez BR, Gout E, Mahe A, Hodges M,
Cornic G (2008) Respiratory metabolism of
illuminated leaves depends on CO2 and O2
conditions. Proc Natl Acad Sci 105:797–802
129. Yin X, Sun Z, Struik PC, Gu J (2011) Evalu-
ating a new method to estimate the rate of leaf
respiration in the light by analysis of com-
bined gas exchange and chlorophyll fluores-
cence measurements. J Exp Bot
62:3489–3499
130. Bellasio C, Griffiths H (2014) Acclimation of
C4 metabolism to low light in mature maize
leaves could limit energetic losses during
progressive shading in a crop canopy. J Exp
Bot 65(13):3725–3736
131. Gong XY, Sch€aufele R, Feneis W, Schnyder H
(2015) 13CO2/12CO2 exchange fluxes in a
clamp-on leaf cuvette: disentangling artefacts
and flux components. Plant Cell Environ
38:2417–2432
132. Ghannoum O, Evans JR, Chow WS, Andrews
TJ, Conroy JP, von Caemmerer S (2005) Fas-
ter rubisco is the key to superior nitrogen-use
efficiency in NADP-malic enzyme relative to
NAD-malic enzyme C4 grasses. Plant Physiol
137:638–650
133. Berry JA, Farquhar GD (1978) The CO2
concentrating function of C4 photosynthesis:
a biochemical model. In: Hall D, Coombs J,
Goodwin T (eds) Proc. 4th Int. Congr. Pho-
tosynthesis, Reading, England 1977 Bio-
chem., Soc., London. pp 119–131
134. Osborn HL, Alonso-Cantabrana H, Shar-
wood RE, Covshoff S, Evans JR, Furbank
RT, von Caemmerer S (2017) Effects of
reduced carbonic anhydrase activity on CO2
assimilation rates in Setaria viridis: a trans-
genic analysis. J Exp Bot 68(2):299–310
135. Ubierna N, Gandin A, Cousins AB (2018)
The response of mesophyll conductance to
short-term variation in CO2 in the C4 plants
Setaria viridis and Zea mays. J Exp Bot,
69(5), 1159–1170
 Chapter 11

Liquid-Phase Measurements of Photosynthetic Oxygen

Evolution
Dmitriy Shevela, Wolfgang P. Schröder, and Johannes Messinger

Abstract
This chapter compares two different techniques for monitoring photosynthetic O2 production: the wide-
spread Clark-type O2 electrode and the more sophisticated membrane inlet mass spectrometry (MIMS)
technique. We describe how a simple membrane inlet for MIMS can be made out of a commercial Clark-
type cell, and outline the advantages and drawbacks of the two techniques to guide researchers in deciding
which method to use. Protocols and examples are given for measuring O2 evolution rates and for
determining the number of chlorophyll molecules per active photosystem II reaction center.

Key words Photosynthetic water oxidation, O2 evolution, Photosystem II, Clark-type electrode,
Membrane inlet mass spectrometry

1 Introduction

While there are many techniques available for measuring dissolved

O2, only a few are suitable for online monitoring of fast-changing,
light-induced photosynthetic O2 production by algae, cyanobac-
teria, chloroplast, and various types of photosystem II (PSII)-
containing preparations [1]. The predominant methods for such
applications are based on electrochemistry [1, 2], luminescence
[3, 4], or mass spectrometry [5–8]. There are two principal types
of electrochemical methods used in photosynthesis research: the
bare-platinum “Joliot”-type electrode and the membrane-covered
Clark-type O2 electrode (CTOE). The Joliot-type electrode is used
“only” for obtaining flash-induced oxygen evolution patterns and is
not further covered here, but see [1]. For the most common appli-
cation, the measurement of oxygen evolution rates, the CTOE has
become the lab standard due to its simple handling, durability, low
cost, and its commercial availability [9, 10]. More recently, also, the
luminescence method is used in a similar way to the CTOE
[4]. Compared to time-resolved membrane inlet mass spectrometry
(MIMS), the CTOE electrode has a significantly lower sensitivity.

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_11, © Springer Science+Business Media, LLC, part of Springer Nature 2018

197
198 Dmitriy Shevela et al.

Perhaps the most important difference is that the CTOE can mea-
sure only the net change in oxygen concentration in the sample.
This means that if photosynthetic O2 evolution is overlaid by a
simultaneous oxygen consumption reaction such as respiration or
photorespiration, then the photosynthetic oxygen evolution rates
will be underestimated or even overlooked, while they can be
separated by MIMS using stable isotope labeling [7, 8]. In addi-
tion, MIMS can, in contrast to the CTOE, detect also other pho-
tosynthetically relevant gases such as CO2, N2, and H2. Despite
such obvious advantages of MIMS, this technique is presently used
in only a few laboratories. The two main reasons are that firstly
there is no complete setup commercially available and thus the users
must be experienced, and secondly isotope ratio mass spectro-
meters are considerably more expensive than the CTOE. Below,
we compare the CTOE and MIMS in more detail, provide brief
outlines of their fundamental principles and key elements, and
describe how a commercial CTOE cell can be simply converted
into a MIMS inlet. This comparison is summarized in Table 1.
Although both techniques can be used for liquid- as well as for
gas-phase measurements of O2 evolution, this chapter is restricted
to liquid-phase O2 measurements. For gas-phase O2 assays in pho-
tosynthesis research (e.g., in whole leaves) see [8, 11, 12].

1.1 Basic Principle The CTOE method was developed in the 1950s by Leland Clark
and Key Elements and co-workers for monitoring blood O2 tensions during cardiac
of the CTOE Technique surgery [13]. The basic principle of this amperometric technique is
the electrochemical reduction of O2 at a Pt cathode that has a
constant potential of about
700 mV versus an Ag/AgCl or
Ag/AgOH anode. The two electrodes are connected by electrolyte
solution (in most cases a saturated KCl) and separated from the
sample by an O2-permeable membrane (Fig. 1). The electric cur-
rent generated by the reduction of O2 is proportional to the con-
centration of O2 at the surface of the Pt cathode, and thus
continuous stirring of the sample is required to prevent the forma-
tion of a diffusion gradient caused by the O2 consumption of the
electrode. The gas-permeable hydrophobic (Teflon) membrane is a
key constituent of the CTOE; it permits the permeation of gaseous
O2 from the reaction chamber into the cathode/anode/electrolyte
system while preventing interactions of other redox-active sub-
stances in the sample suspension with the electrode and the leakage
of the electrolyte into the sample. Since the degree of O2 solubility
in solution depends on the temperature, the electrode chamber
must be thermostated. Typically, this is performed by means of a
water jacket around the reaction vessel coupled to a thermoregu-
lated circulating water bath. Before performing an O2 assay, the
CTOE must be calibrated so that the electric currents can be
converted into O2 concentrations. Finally, a light source, which
 Measurements of O2 Evolution in Photosynthesis 199

Table 1
A comparison of the CTOE and MIMS techniques

The CTOE technique The MIMS technique

Instrument costs Relatively low High to very high
Availability The entire setup is commercially MS commercially available, but requires
available from several in-house sample inlet
companies
Membrane Typically PTFE (Teflon); Depending on the tasks, either silicone
supplied with electrodes or Teflon membranes can be used;
purchase separately
Sensitivitya 1 μM O2b 0.1 μM (0.1 nM) O2c
Separate O2 evolution and Not possible Possible in combination with 18O2 or
O2 consumption H218O labeling (see [6, 8] and
references therein)
Required time between 1 min or less (normally stable >10 min (thorough degasation is
loading the samples and background is reached almost required for low drift of background
measurement immediately) signal)
Possibility for liquid- and Both possible Both possible
gas-phase (whole leaf)
measurements
a
It is difficult to define absolute sensitivity of the CTOE and MIMS instruments, since the sensitivity is highly dependent
on many factors, including sample inlet system, membrane type and thickness, and working volume of the cell. In our
case, an estimate of sensitivity was performed under similar physical conditions using a commercial CTOE chamber
(adjusted to the same volume) as the MIMS sample inlet and using a PTFE membrane in both cases
b
Estimate was done with the standard electrode disk as provided by the supplier. It should be noted that the CTOE setups
with larger cathode surface area may have higher sensitivity
c
Estimate was performed with an isotope ratio mass spectrometer (equipped with magnetic sector ion optics and array
detection) as detector instrument monitoring non-labeled 16O2 at m/z ¼ 32 (with cup amplification of 1  109) via a
Teflon membrane. We note that the sensitivity value of 0.1 μM for MIMS obtained under these constraints largely
underestimates the real potential of this technique. Significantly higher (up to 1000 times) sensitivity can be reached by
using a silicon membrane and by labeling the samples with H218O for monitoring the 18O-labeled oxygen isotopologues
(16O18O or 18O2) on a very low background level (that can be achieved with a small reaction vessel volume of
150–200 μL) and with the most sensitive Faraday cup of the mass spectrometer

today normally is a LED light panel to minimize heat generation, is

required for sample illumination. For detailed overviews on the
CTOE method see [1, 9–11].

1.2 Fundamental MIMS involves the detection of gaseous analytes such as O2 based
Principle and Key on their mass-to-charge ratio (m/z) after they have been ionized in
Elements of the MIMS the ion source of a mass spectrometer and separated from other gas
Method ions or isotopologues in the mass analyzer. To avoid disturbances of
the ions’ trajectories by collisions with other molecules, all mass
spectrometers must operate under high vacuum. Therefore, O2
measurement on liquid suspension of photosynthetic samples is a
challenging task. In 1963, George Hoch and Bessel Kok solved this
technical challenge by separating the liquid sample from the high
vacuum by a semipermeable membrane [14]. The principle is
200 Dmitriy Shevela et al.

Fig. 1 Simplified scheme of a Clark-type electrode and its components. For the
sake of simplicity, a fine paper wick, which is normally applied between the
membrane and the electrode disk (to create an electrolyte bridge between the
cathode and the anode), is not shown. Dissolved O2 is involved in the following
reactions at the Pt cathode: O2 + H2O + 2 e
! H2O2 + 2 OH
;
H2O2 + 2 e
! 2 OH
. Reactions at the Ag/AgCl anode are
4 Ag ! 4 Ag+ + 4 e
; 4 Ag+ + 4 Cl
! 4 AgCl

similar to that described for the CTOE, only that the thin mem-
brane must be stabilized by a porous support to avoid rupture by
the vacuum. The membrane allows the gaseous analytes dissolved
in the sample to enter the mass spectrometer while keeping the
water and the PSII sample out of the detection system. This tech-
nical solution allows continuous online O2 assays with a resolution
of a few seconds.
Different types of mass spectrometers can be employed as a
basis for the MIMS instrumentation [15]. Here, we describe the
MIMS system comprising a stable isotope ratio mass spectrometer
(Fig. 2a). In this type of mass spectrometer, gaseous analytes are
ionized by electron impact and are then separated according to
their m/z ratios by a magnetic sector field. Such a mass analyzer
system allows simultaneous online collection of ion currents by an
array of individual detectors (seven Faraday cups are shown in
Fig. 2a).
The key part of the MIMS setup is the membrane inlet system
that is integrated within a MIMS chamber (Fig. 2b). The transmis-
sion of gaseous analytes through the membrane occurs via perva-
poration [16], a complex process which includes absorption and
permeation of analytes through the membrane and desorption into
the vacuum on the other side of the membrane. Upon desorption,
the gaseous analytes enter the high-vacuum line that delivers the
gas to the ion source of the mass spectrometer. On its way, the
analytes pass a cryogenic trap, which freezes out trace amounts of
water vapor that inadvertently enter the system (Fig. 2a) and also
protects the mass spectrometer against sample entry in case of
membrane damage. Similar to the CTOE method, the samples
need to be stirred vigorously to avoid the formation of concentra-
tion gradients.
 Measurements of O2 Evolution in Photosynthesis 201

Fig. 2 The MIMS setup and its membrane inlet. (a) Schematic representation of a time-resolved MIMS setup
based on a magnetic sector mass spectrometer. (b) MIMS cell assembled on the basis of a commonly
available DW1 O2 electrode chamber (Hansatech Instruments Ltd., UK) and an in-house MIMS inlet; (c)
in-house MIMS inlet shown from various perspectives

1.3 Comparison Although the fundamental principles of O2 detection are different

of the CTOE and MIMS in the CTOE and MIMS techniques, both techniques are identical
Methods in using a semipermeable membrane as a key functional element of
their sample inlet systems. The membrane is not permeable for
liquids and ions, but for gases; as such it prevents direct contact
of the aqueous sample and dissolved chemicals with the detection
system, which yields stability of operation and reliable results. The
membranes most commonly used are made either of Teflon or
silicon. The former is preferentially used in the CTOE electrode,
but our test showed that nearly identical results are obtained with a
silicon membrane (data not shown). In contrast, for MIMS they are
not interchangeable and only the silicon-based membrane gave a
fast response. The measuring capabilities, some technical para-
meters, as well as advantages and drawbacks of both methods are
summarized in Table 1.
202 Dmitriy Shevela et al.

2 Materials

2.1 Biological/ Intact algae and cyanobacteria, as well as chloroplasts, thylakoid

Photosynthetic membranes, PSII-enriched membranes (BBY membranes), or PSII
Material core complexes isolated from higher plants, algae, or cyanobacteria;
for detailed protocols describing isolation of these various prepara-
tions see relevant chapters in [17]. The O2 measurements shown in
this chapter were performed with BBY membranes from spinach,
referred to as PSII samples here after.

2.2 Buffers The optimal assay buffer solutions depend on the sample prepara-
and Solutions tions and the aim of the experiments. Below a buffer composition is
given that works well for BBY and broken thylakoids. Use deio-
nized (>1 MΩ  cm at 25  C) and filtered (0.22 μm filter) water
and analytical grade reagents for making buffered solutions. Follow
all waste disposal regulations when disposing waste materials.
1. Buffer (assay buffer): 40 mM MES-NaOH (pH 6.0–6.5),
15 mM NaCl, 5 mM MgCl2, 5 mM CaCl2.
2. Stock solution from electron acceptor PPBQ: 50–100 mM
2-phenyl-p-benzoquinone dissolved in dimethyl sulfoxide
(DMSO).
3. Stock solution for electron acceptor FeCy: 50–100 mM potas-
sium ferricyanide (III) dissolved in water (see Note 1).

2.3 CTOE Setup Complete (ready-to-use) CTOE setups are commercially available
through several companies. For a full list of the required compo-
nents of the CTOE setup, see operating manuals and references
[10, 11]. Here, only some key parts of the CTOE are described.
1. Clark-type electrode chamber: The electrode chamber (that
can also serve as a MIMS chamber as shown in Fig. 2b) with
accompanying magnetic follower and adjustable plunger.
2. Clark-type O2 electrode: The O2 electrode disk with control
unit connected via an electrode disk connection cable. The
magnetic stirrer is often integrated into the electrode control
unit, but sometimes supplied also as separate unit. Before use,
the O2 electrode must be prepared to form an electrolyte bridge
between the anode and cathode and covered by the membrane
according to the procedure described in detail in the manufac-
turer’s operating manuals or in [11]. We also refer readers to
these references to learn about other practical issues, including
maintenance, troubleshooting, and cleaning of the electrode.
3. Membrane: Typically, the electrode system is supplied with the
PTFE (Teflon) membrane (see Note 2) and a paper spacer
(such as a cigarette paper), which is placed beneath the Teflon
membrane to provide a uniform layer of electrolyte between
anode and cathode.
 Measurements of O2 Evolution in Photosynthesis 203

4. Circulating water bath: A thermoregulated circulating water

bath is required to be attached to the water jacket of the O2
electrode chamber to keep constant temperature, which is
critical for the calibration and measurements.
5. CW illumination source: Saturating CW illumination is
required for monitoring the O2 evolution rates by photosyn-
thetic samples. Any strong light can be employed for illumina-
tion. For the measurements shown in Figs. 2a and 3a, we used a
high-intensity illumination source. Using a light meter, adjust
the distance between the O2 electrode chamber and the light
source so that the light intensity measured inside of the sample
vessel lies between 1000 and 1500 μmol photons m
2 s
1.
6. Flash lamp: For excitation of photosynthetic samples with
repetitive-saturating single-turnover flashes, Xe flash lamps or
LED lamps can be used. For the assays depicted in Figs. 2b and
3b, we employed a Xe flash lamp with a flash duration of 5 μs
(full width at half maximum) and an intensity of 500 mJ per
flash.
7. Liquid-phase calibration: For liquid-phase O2 measurements,
the output is calibrated through comparison between
air-equilibrated/saturated water content at a specific tempera-
ture and atmospheric pressure, and oxygen-free water. To
obtain an oxygen-free solution, two principal methods are
used: bubbling with nitrogen to displace all dissolved oxygen
in the air-saturating water, or the addition of the reducing
agent sodium dithionite, which reacts with oxygen in the solu-
tion (see Note 3). Table values of oxygen concentrations for
various temperature, air pressure, and salinity can be obtained
at the following webpage: http://www.colby.edu/chemistry/
CH331/O2%20Solubility.html.

2.4 MIMS Setup Unlike the CTOE device, the entire MIMS setup is not commer-
cially available and its assembly requires additional engineering
work for the design of the MIMS chamber. However, for standard
experiments, such efforts can be significantly minimized by
employing an O2 electrode chamber as a basis for the MIMS cell.
1. Isotope ratio mass spectrometer: Any isotope ratio mass spec-
trometer for the mass range m/z  2–100 is suitable for setting
up the MIMS device. Alternatively, a cheaper quadrupole mass
spectrometer with suitable mass range can be used. The latter
comes with a small loss in sensitivity and stability and less
accurate isotope ratios. If several masses shall be recorded in
parallel, e.g., for obtaining isotope ratios, then peak jumping is
required. On the plus side, quadrupole mass spectrometers
generally give a higher flexibility, since one is not fixed to a
specific Faraday cup arrangement. MIMS measurements pre-
sented in this chapter were carried out with an isotope ratio
204 Dmitriy Shevela et al.

Fig. 3 CTOE measurements of O2 evolution in spinach PSII samples at 20  C and

pH 6.5. (a) Calibrated polarographic O2 signal induced by 1 min illumination of
the samples with saturating continuous light (1500 μmol photons m
2 s
1). The
final Chl concentration inside of the O2 electrode chamber (with sample volume
adjusted to 1 mL) was 0.01 mg mL
1. The measurements were performed in the
presence of 300 μM PPBQ and 600 μM FeCy as artificial electron acceptors. The
rate of O2 evolution of the PSII samples was determined to be ~480 μmol
O2 (mg Chl)
1 h-1 (see Eq. 1). Note that illumination of the measuring buffer in
the absence of the O2-evolving PSII samples may cause a small O2 consumption
(grey curve). (b) Calibrated polarographic O2 signal obtained in the PSII samples
in response to 120 saturating Xe flashes given at a frequency of 2 Hz. The
volume of the O2 electrode chamber was adjusted to 0.35 mL. The measure-
ments were performed at [Chl] ¼ 0.05 mg ml
1 in the presence of 500 μM PPBQ
and 1000 μM FeCy as artificial electron acceptors. Together with the [Chl], the
O2 signal obtained can be employed for calculating the number of Chl molecules
per active PSII reaction center (see Eq. 2)

mass spectrometer equipped with magnetic sector ion optics

and array detection.
2. MIMS chamber: The design of the MIMS chamber greatly varies
depending on the measuring purpose [5, 6, 8]. The easiest way is
to use an O2 electrode chamber as the basis for assembling a
MIMS cell (Fig. 2b). The Hansatech DW1 chamber was used to
optimize this protocol. Other devices may also be suitable but
testing and configutation as appropriate are required.
3. Sample inlet for MIMS chamber: A simple stainless steel or
brass sample inlet that fits to the electrode chamber (Fig. 2b, c)
needs to be made in a workshop. This is then fitted with a disk
of a porous plastic support, which on the one hand prevents
rupture of the thin membrane and, on the other hand, does not
impose a significant diffusion barrier for the traversing analytes.
For highest sensitivity, the diameter of the plastic support
should be the same as the internal diameter of the reaction
vessel of the O2 electrode chamber, i.e., commonly about
 Measurements of O2 Evolution in Photosynthesis 205

10 mm; however, if lower consumption rates are desirable, then

a smaller inlet hole is advantageous. A membrane needs to be
placed over the plastic support (see Note 4). For this cut a
2 cm  2 cm piece of membrane. Then apply a thin layer of
high-vacuum grease to the well-polished metal surface around
the plastic support; place the membrane without wrinkles. This
is best done by connecting the inlet to a rotary vacuum pump
(Line V1 in Fig. 2a).
4. Thermoregulated circulating water bath: Since the gas solubility
and permeability through the membrane are temperature depen-
dent, the temperature must be kept constant within 1  C.
5. Stirring system: In order to avoid the measurements of just the
boundary layers of the liquid sample above the membrane,
constant stirring of the sample is required. For this, the
MIMS chamber can be stirred using a magnetic stirrer (inbuilt
in the O2 electrode control unit or external) and a small PTFE-
coated magnetic stir bar inside of the reaction vessel. A constant
and efficient stirring speed is essential for achieving good S/N
ratios in the measurements of liquid samples. The surface of the
stirrer and of the plastic support underneath the membrane
need to be smooth to avoid formation of pin holes or destruc-
tion of the membrane.
6. Cryogenic trap and vacuum lines of the MIMS setup (Fig. 2a):
The cryogenic trap is normally made by forming a loop in the
vacuum line that passes through a dewar filled with either dry
ice in ethanol (~200 K) or liquid nitrogen (~77 K). In standby
mode, the vacuum Line V2 leading to the mass spectrometer is
closed, while the Line V1 is open. The Line V1 connects the
MIMS chamber to a rotary backing pump that creates a
pre-vacuum of about 2  10
3 mbar. This is needed for
removal of water or other solvents that were frozen in the
cooling loop from previous measurements. Prior to starting
new measurements, the dewar with coolant is placed around
the cooling loop. Then, Line V1 is closed and Line V2 is slowly
opened so that the MIMS chamber is connected to the mass
spectrometer. Opening and closing of the lines are performed
with gastight valves (Fig. 2a; for more details, see [6]).
7. CW and flash illumination sources: The illumination sources
are identical to those described in Subheading 2.3.

3 Methods

All manipulations with the PSII samples before the O2 assays must
be performed in the dark or under a dim green light. Thaw the PSII
sample aliquots in the dark on ice shortly before the measurements.
206 Dmitriy Shevela et al.
3.1 O2 Measure-
ments by CTOE
3.1.1 Determination
of O2-Evolving Activity
(O2 Evolution Rate)
1. Calibrate the O2 signal as described in the operating manual by
using the signal difference between air-saturated and air-free
water (see Note 5). At 20  C, air-saturated, salt-free water
contains 0.284 μmol O2 mL
1 at 1013.25 mbar air pressure.
2. To simplify calculations, adjust the sample volume to 1 mL
using the plunger of the O2 electrode chamber (Fig. 2b). It is
practical to make a double-concentrated measuring buffer, and
then add 500 μL of this 2 buffer into the sample chamber,
followed by the artificial electron acceptors (PPBQ and FeCy;
see Note 6), PSII sample, and remaining water. Typically, the
final chlorophyll (Chl) concentration in the chamber should be
between 0.01 and 0.02 mg (Chl) mL
1 to assure light satura-
tion. Concentration of the artificial electron acceptors and
other additions (e.g., uncouplers such as gramicidin or ammo-
nia in the case of intact thylakoids) are made such that a final
volume of 1 mL can be obtained. Close the cell with the
plunger and start stirring.
3. Check for a stable baseline of the amperometric O2
signal  1 min.
4. Turn on measuring light and register oxygen evolution. After
1–2 min, turn off light and record signal decay for about 1 min.
5. Evaluate the registered traces (Fig. 3). Normally, the initial
rate, i.e., oxygen production during the first 10–15 s after a
short lag, is the maximum oxygen evolution rate of the sample.
After that, the curve usually bends due to acceptor limitation
and/or photoinhibition. To evaluate the latter effect, one
alternative manner to evaluate the traces is to illuminate for
1 min, and obtain initial rates (first 15 s), final rates (last 15 s),
and the average between start and end points (the whole min-
ute). In general, the O2 evolution rate R(O2) can be calculated
according to the following equation:
½O2 sat 1
RðO2 Þ ¼ S net    3600 ð1Þ
½Chl Sc
where Snet is the net slope of the linear signal rise (in mV s
1;
taking background drifts into account if needed), [O2]sat is the
concentration of oxygen in air-saturated water at a given tem-
perature and air pressure (in μmol; see step 1 in Subheading 3.1
and item 7 in Subheading 2.3), Sc is the amplitude of the
calibration (in mV), and 3600 is a factor for converting s into-
h. As a result, the R(O2) is expressed in units of μmol
(mg Chl)
1 h
1. In our case, the O2 evolution trace shown in
Fig. 3a corresponds to ~480 μmol (mg Chl)
1 h
1.
 Measurements of O2 Evolution in Photosynthesis 207

6. Depending on additions made (e.g., type of electron acceptors,

uncouplers, inhibitors), it is important to rinse the O2 elec-
trode chamber thoroughly with water between measurements,
especially if additions are hydrophobic and could stick to the
plastic or glass in the reaction vessel.

3.1.2 Determination 1. Calibrate as described in Subheading 3.1.1.

of the Number 2. Add an aliquot of the PSII sample to make final [Chl] ¼ 0.05 μg
of Chlorophylls per PSII mL
1 to a sample vessel that contains the assay buffer, which
Reaction Center (RC) includes 300 μM PPBQ and 600 μM FeCy as artificial electron
acceptors. The total volume of the sample with buffer and
acceptors was 0.35 mL in our experiments (volume optimized
to ensure light saturation while maintaining an acceptable
signal-to-noise ratio).
3. Instead of using the regular plunger, close the sample chamber
with a Plexiglas rod (built in-house) that has a slope on the
sample end and a small grove on the shorter side for allowing
bubble-free insertion (see Note 7).
4. Connect flash lamp to the other end of the Plexiglas rod. A lab
stand and plastic adapter can be used for this. Typically, the rod
should be placed right against the window of the flash lamp.
5. Allow the dark-adapted PSII sample to equilibrate for at least
1–2 min in the dark to reach the desired temperature for the
experiment (normally, the optimal temperature range is
20–25  C).
6. Start recording the O2 signal for at least 1 min before switching
on the flash lamp.
7. Illuminate PSII sample with saturating single-turnover Xe
flashes given at a frequency of 1–2 Hz. Continue to record
the amount of O2 evolved by the sample as a function of time
(Fig. 3b). You may vary all specified parameters (flash number
and flash frequency) so as to get the best possible O2 yield per
flash with your sample type and flash lamp. Another concern is
to have a good signal-to-noise ratio.
8. Calculate the concentration of O2 generated by the flashes.
Together with the known [Chl] in the sample chamber, calcu-
late the number of chlorophyll molecules per active PSII reac-
tion center, Z(O2), according to

½Chl
Z ðO2 Þ ¼ Δ½O2 n
ð2Þ
4 n

where Δ[O2]n is the change of the [O2] (in μM) induced by a

number (n) of given flashes (n ¼ 120 in our example). A lower
estimate for the O2 yield (~6 μM in Fig. 3b) is obtained by
208 Dmitriy Shevela et al.

measuring the amplitude at the maximum of the signal with

respect to the forward extrapolated baseline. This results in an
upper estimate of Z(O2) ¼ 250 Chl/RC. Typical numbers for
active BBY membranes may vary from 200 to 250 Chl/RC. In
active PSII core complexes these numbers normally vary from
35 to 40 Chl/RC.

3.2 MIMS Except for the calibration procedure, which is done by injections of
Measurements of O2 known volumes of air-saturated, salt-free water into the degassed
Evolution buffer inside the MIMS cell, all other measuring steps are identical
and Determination to those described under Subheading 3.1. It is important that
of the Number during the calibration, both the injected air-saturated water and
of Chlorophylls per the thermostated MIMS cell are adjusted to the same temperature
PSII RC (in our case 20  C). In this way, a linear correlation (calibration
curve) between the concentrations of dissolved O2 (μM) and the
signal amplitudes (mV) is obtained that allows for quantification of
the light-induced [O2] signals of PSII ([O2]light) in terms of
μmol s
1. Accordingly, the R(O2) can be estimated as follows:
½O2 light
RðO2 Þ ¼  3600 ð3Þ
½Chl
Figure 4 represents typical m/z ¼ 32 MIMS signals obtained
after exposure of PSII samples to continuous (panel a) and flashing
(panel b) light. From these signals one can determine the R(O2)
and Z(O2) values by employing Eq. (3) and Eq. (2), respectively.

Fig. 4 MIMS assays of O2 evolution in spinach PSII samples at 20  C and pH 6.5. (a) Calibrated MIMS O2 signal
(m/z ¼ 32) induced by 1-min illumination of the PSII samples with saturating continuous light (1500 μmol
photons m
2 s
1). Measuring conditions: [Chl] ¼ 0.01 μg mL
1; [PPBQ] ¼ 300 μM PPBQ; [FeCy] ¼ 600 μM;
sample volume ¼ 1 mL. (b) Calibrated MIMS O2 signal (m/z ¼ 32) obtained from PSII samples in response to
120 saturating Xe flashes given at a frequency of 2 Hz. Measuring conditions: [Chl] ¼ 0.05 μg mL
1;
[PPBQ] ¼ 600 μM PPBQ; [FeCy] ¼ 1000 μM; sample volume ¼ 0.35 mL. All other measuring conditions are
identical to those in Fig. 3
 Measurements of O2 Evolution in Photosynthesis 209

4 Notes

1. We find that it is practical to freeze the stock solutions of

artificial electron acceptors as small aliquots (30–50 μL each)
in capped plastic tubes (0.5–1.5 mL) and store them at
80  C
until used. Shortly before the measurements, thaw one fresh ali-
quot of the acceptors in the dark at room temperature.
2. Any other type of membrane can be used to increase sensitivity
and time response of the O2 electrode. However, we find that it
is best to use a standard PTFE membrane. Our test measure-
ments show that usage of a silicone membrane (as used for
MIMS inlet system) does not improve the sensitivity of the
electrode (data not shown).
3. In case O2-free nitrogen gas is not available, one may use
sodium dithionite instead. Normally, a few crystals of sodium
dithionite are sufficient to establish the “zero oxygen line”
from the air-saturated water. However, we find that bubbling
of air-saturated water with nitrogen gas inside the reaction
vessel is a more convenient method of calibration. Usage of
sodium dithionite for calibration causes the risk of contaminat-
ing the photosynthetic sample during the subsequent O2 mea-
surements. Another disadvantage is that sodium dithionite is
toxic and an irritant and thus needs to be handled with care;
especially, it should not be mixed with acids to avoid toxic gas
production.
4. The choice of membrane is an important factor in the MIMS
protocol. The “best” membrane depends on the experimental
purpose. The thickness and permeability of the membrane
determine both the overall sensitivity of detection and the
response time of the MIMS setup. High-permeability mem-
branes combined with a large inlet area and small sample vol-
ume provide the highest sensitivity and the fastest response
time, but come with a fast gas consumption rate, which is not
always desirable [6]. The following membranes can be used for
MIMS applications: Teflon films, silicone membranes (e.g.,
silicone elastomer films or polydimethylsiloxane membranes),
silicone rubber, O2 electrode membranes (e.g., standard PTFE
membrane or high-sensitivity (0.5 mil stretch) membranes), or
some other plastic films from various sources [15]. Silicon
membranes with an embedded metal grid can be used for
measurements under elevated gas pressure (e.g., up to 20 bar)
[18]. Note that many organic solvents (methanol, ethanol,
acetonitrile, etc.) easily penetrate through the membranes
into the mass spectrometer and should be avoided or used in
minimal amounts only.
210 Dmitriy Shevela et al.

5. Air-equilibrated water is best obtained by periodic shaking and

gentle stirring of a small volume of water in an open vessel.
This must be done at the same temperature as used for the
experiments.
6. Usually, the O2 evolution measurements are performed in
combination with two artificial electron acceptors: FeCy, at
concentrations from 0.5 to 1 mM, and a quinone (PPBQ or
DCBQ), at concentrations from 200 to 300 μM. We recom-
mend using PPBQ, since it is known to support higher PSII
turnover efficiencies than DCBQ at high flash frequencies
(2 Hz) and accepts electrons from QA
with similar efficiency
as plastoquinone in intact cells (see [19] and references therein).
7. The diameter of the rod equals the inner diameter of the sample
chamber, while the length is selected for convenient attach-
ment of the flash lamp. Both ends of the rod are polished for
optimal light transmission. A small ring with screw can be used
to fix the rod in place.

Acknowledgments

We thank Dr. Mun Hon Cheah for careful reading of the manu-
script and valuable suggestions. The Swedish Energy Agency (Ener-
gimyndigheten), Swedish Science Foundation (Vetenskapsrådet),
and the K & A Wallenberg Foundation are acknowledged for
financial support.

References

1. Renger G, Hanssum B (2009) Oxygen detec- 6. Konermann L, Messinger J, Hillier W (2008)

tion in biological systems. Photosynth Res
102:487–498
2. Van Gorkom HJ, Gast P (1996) Measurement
of photosynthetic oxygen evolution. In:
Amesz J, Hoff AJ (eds) Biophysical techniques
in photosynthesis, Advances in photosynthesis,
vol 3. Kluwer Academic Publishers, Dordrecht,
pp 391–405
3. Lakowicz JR (1999) Principles of fluorescence
spectroscopy, 2nd edn. Kluwer Academic/Ple-
num Publishers, New York
4. Chodavarapu VP, Shubin DO, Bukowski RM
et al (2007) CMOS-based phase fluorometric
oxygen sensor system. IEEE Trans Circ Syst
54:111–118
5. Shevela D, Messinger J (2013) Studying the
oxidation of water to molecular oxygen in pho-
tosynthetic and artificial systems by time-
resolved membrane-inlet mass spectrometry.
Front Plant Sci 4:473. https://doi.org/10.
3389/fpls.2013.00473
Mass spectrometry based methods for studying
kinetics and dynamics in biological systems. In:
Amesz J, Hoff AJ (eds) Biophysical techniques
in photosynthesis, Series advances in photosyn-
thesis and respiration, vol 26. Springer, Dor-
drecht, pp 167–190
7. Cheah MH, Millar AH, Myers RC et al (2014)
Online oxygen kinetic isotope effects using
membrane inlet mass spectrometry can differ-
entiate between oxidases for mechanistic stud-
ies and calculation of their contributions to
oxygen consumption in whole tissues. Anal
Chem 86:5171–5178
8. Beckmann K, Messinger J, Badger MR et al
(2009) On-line mass spectrometry: membrane
inlet sampling. Photosynth Res 102:511–522
9. Delieu T, Walker DA (1972) An improved
cathode for the measurement of photosyn-
thetic oxygen evolution by isolated chloro-
plasts. New Phytol 71:201–255
 Measurements of O2 Evolution in Photosynthesis
10. Gonzalez L, Bolano C, Pellissier F (2001) Use
of oxygen electrode in measurements of pho-
tosynthesis and respiration. In: Reiggosa Roger
MJ (ed) Handbook of plant ecophysiology
techniques. Kluwer Academic Publishers, Dor-
drecht, pp 141–153
11. Walker D (1987) The use of the oxygen elec-
trode and fluorescence probes in simple mea-
surements of photosynthesis. Oxygraphics
Limited, Sheffield
12. Canvin DT, Berry JA, Badger MR et al (1980)
Oxygen exchange in leaves in the light. Plant
Physiol 66:302–307
13. Clark LC, Wolf R, Granger D et al (1953)
Continuous recording of blood oxygen ten-
sions by polarography. J Appl Physiol
6:189–193
14. Hoch G, Kok B (1963) A mass spectrometer
inlet system for sampling gases dissolved in
liquid phases. Arch Biochem Biophys
101:160–170
211
15. Johnson RC, Cooks RG, Allen TM et al (2000)
Membrane introduction mass spectrometry:
trends and applications. Mass Spectrom Rev
19:1–37
16. Silva ACB, Augusti R, Dalmazio I et al (1999)
MIMS evaluation of pervaporation processes.
Phys Chem Chem Phys 1:2501–2504
17. Carpentier R (ed) (2011) Photosynthesis
research protocols, Methods in molecular biol-
ogy, vol 684, 2nd edn. Springer, Clifton
18. Shevela D, Beckmann K, Clausen J et al (2011)
Membrane-inlet mass spectrometry reveals a
high driving force for oxygen production by
photosystem II. Proc Natl Acad Sci U S A
108:3602–3607
19. Shevela D, Messinger J (2012) Probing the
turnover efficiency of photosystem II
membrane fragments with different electron
acceptors. Biochim Biophys Acta
1817:1208–1212
 Part II

Measuring Photosynthetic Enzyme Abundance

and Catalytic Activity
 Chapter 12

Quantification of Photosynthetic Enzymes in Leaf Extracts

by Immunoblotting
J. Alejandro Perdomo, Cristina R. G. Sales, and Elizabete Carmo-Silva

Abstract
In this chapter, we describe a method to extract and quantify photosynthetic enzymes using sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting. The method is particularly
suitable for characterizing altered protein amounts in leaves of plants produced from genetic engineering or
gene-editing approaches. We focus on RuBisCO and RuBisCO activase, a molecular chaperone required to
sustain the activity of RuBisCO and CO2 fixation, yet the method can be easily adapted to investigate other
leaf proteins of interest.

Key words Photosynthesis, CO2 assimilation, Calvin-Benson-Bassham cycle, RuBisCO, RuBisCO

activase, Quantitative immunoblotting, Protein extraction, Immunoblotting, Genetic engineering,
Transgenic and gene-edited plants

1 Introduction

Improving photosynthetic efficiency is a major target for improving

plant biomass production, crop yields, and agricultural resource use
efficiency [1]. A number of photosynthetic enzymes have been
identified as targets for bioengineering plants to improve carbon
assimilation [2, 3]. These targets include enzymes of the Calvin-
Benson-Bassham cycle, such as those involved in the carboxylation
and regeneration of RuBP (ribulose 1,5-bisphosphate) [4, 5].
RuBisCO catalyzes the carboxylation of RuBP, enabling CO2
assimilation in photosynthesis. It is an essential, yet inefficient
enzyme, characterized by a number of severe limitations. For
instance, RuBisCO is a relatively slow catalyst, by comparison
with other plant enzymes, and acts both as a carboxylase and an
oxygenase, with CO2 and O2 competing for the reaction with
RuBP, which decreases the efficiency of photosynthesis. Conse-
quently, RuBisCO accounts for up to 50% of the total soluble
protein in the leaves of C3 plants, such as wheat [6]. RuBisCO is
also prone to inhibition by the binding of sugar phosphates that

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_12, © Springer Science+Business Media, LLC, part of Springer Nature 2018

215
216 J. Alejandro Perdomo et al.

lock the enzyme’s catalytic sites in a closed conformation. RuBisCO

activase (Rca) is a catalytic chaperone that remodels the conforma-
tion of RuBisCO, facilitating the release of inhibitors and restoring
catalytic competence to the CO2-fixing enzyme [7]. Rca is present
in all plant species and green algae [8]. The leaves of most plant
species contain two Rca isoforms, a short β-isoform (41–43 kDa)
and a longer α-isoform (45–46 kDa) that differs from the shorter
form by the presence of a redox-sensitive C-terminal extension
[7, 8]. Immunoblotting analyses have shown the presence of both
Rca isoforms in numerous plant species [8]. However, like many
other enzymes, the number of Rca isoforms and their relative
abundance vary from species to species.
Accurate protein quantification of the target enzymes for pho-
tosynthetic improvement, such as RuBisCO and Rca, is key to
evaluate the impact of altering their abundance on carbon assimila-
tion and plant biomass production. As Rca affects RuBisCO func-
tion, it is important to evaluate whether altered Rca amounts have
an additional and indirect impact on the amounts of RuBisCO [1];
thus we routinely quantify both enzymes simultaneously. This
chapter describes (1) extraction of soluble proteins from frozen
leaves, including RuBisCO and Rca; (2) separation of the extracted
proteins by SDS-PAGE and Coomassie staining of protein bands;
(3) immunoblotting of the separated proteins to a nitrocellulose
membrane and subsequent immunodetection of Rca; and (4) rela-
tive protein quantification by reference to a dilution series.

2 Materials

All solutions are prepared with deionized pure laboratory water.

Deionized pure water is also used for rinsing gels, membranes, and
soaking filter paper to be used in protein transfer.

2.1 Extraction 1. Frozen leaf tissue, e.g., 2.5 cm2 samples (see Note 1), stored at
of Total Soluble
80  C until transferred to a liquid nitrogen container prior
Proteins from Leaves to use.
2. Ice-cold leaf extraction buffer: 50 mM Tricine-NaOH pH 8.0,
10 mM EDTA, 1% PVP40, 20 mM 2-mercaptoethanol, 1 mM
phenylmethylsulfonyl fluoride (PMSF), 10 μM leupeptin
(see Note 2).
3. Ice-cold mortar and pestle or tissue homogenizer.
4. Refrigerated microcentrifuge with fixed-angle rotor set to 4  C.
5. SDS loading buffer: 3.75% Sodium dodecyl sulfate (SDS),
22.5% sucrose, 0.5% bromophenol blue. Store frozen in ali-
quots at
20  C. Combine with protein extracts on a 2:2.5
ratio, e.g., 20 μL loading buffer to 25 μL sample extract.
 Quantification of Photosynthetic Enzymes by Immunoblotting 217

6. SDS blank solution: Mix SDS loading buffer above with extrac-
tion buffer in the same proportion as used for the samples
(2:2.5 ratio), e.g., add 20 μL loading buffer to 25 μL extraction
buffer.
7. Heating block to incubate samples at 95  C.

2.2 SDS-PAGE 1. Running gel unit.

2. Polyacrylamide gel for electrophoresis (e.g., 12% acrylamide/
bis-acrylamide 37.5:1, 15 well, 1.5 mm thick; home-made or
commercial).
3. Pre-stained protein standards (10–250 kDa, including the size
range of the target proteins).
4. Bradford reagent and bovine serum albumin (BSA).
5. Running buffers: The following upper and lower gel-running
reservoir buffers provide for good protein separation and well-
defined bands. Upper buffer 1 final concentrations: 40 mM
Trizma-base, 40 mM borate, 0.1% SDS, pH 8.4 adjusted with
borate (see Note 3). Lower buffer 1 final concentrations:
0.4 M Trizma-base pH 9.18 adjusted with 0.03 M HCl
(see Note 3).
6. Rocking platform shaker.

2.3 Protein Transfer 1. Dry blotting system device.

2. Transfer stack including membrane compatible with fluores-
cence signal detection, such as nitrocellulose or low-fluores-
cence PVDF. Commercially available transfer stacks for dry
blotting contain a copper electrode and appropriate cathode
and anode buffers in the gel matrix to allow fast and reliable
transfer of proteins.
3. Tris-buffer-saline (TBS): 20 mM Trizma-base, 150 mM NaCl,
pH 7.5 adjusted with HCl (see Note 4).
4. Tris-buffer-saline-Tween (TBST): 20 mM Trizma-base,
150 mM NaCl, pH 7.5 adjusted with HCl, 0.05% (v/v)
Tween-20 (see Note 5).
5. 4% Blocking solution (Blotto): 4% Skim milk powder in TBS.
6. Primary antibody: RuBisCO activase antibody produced in
rabbit against cotton Rca (see Note 6), diluted 1:10000 in
TBS, 0.5% skim milk powder, and 0.04% NaN3 (the diluted
antibody can be used at least five times).
7. Secondary antibody: Anti-rabbit IgG suitable for fluorescence
detection diluted in TBS, 0.5% skim milk powder, and 0.04%
NaN3 (see Note 7).
8. Black opaque immunoblot incubation box.
9. Filter paper, pore size 11 μm.
218 J. Alejandro Perdomo et al.

3 Methods

3.1 Extraction
of Total Soluble
Protein from Leaves
1. Homogenize leaf disks (2.5 cm2) in 500 μL of ice-cold extrac-
tion buffer in an ice-cold mortar and pestle or tissue
homogenizer.
2. Centrifuge the homogenate at 14,000  g for 3 min at 4  C.
Collect the supernatant and transfer 50 μL to a tube containing
40 μL SDS-loading buffer, and then heat at 95  C for 4 min
(a minimum of 3 min is recommended to ensure protein
denaturation).
3. In order to estimate the content of RuBisCO and Rca in the
leaves of the various plants analyzed relative to a control plant
(e.g., wild-type), prepare an SDS standard using the leaf extract
from one of the control plants (see Note 8) as follows: add
250 μL of leaf extract supernatant prepared from a control
plant to 200 μL of SDS-loading buffer, and then heat at
95  C for 4 min.
4. Determine the total soluble protein (TSP) concentration in
each sample leaf extract supernatant by reaction with Bradford
reagent [9]. Based on the TSP concentration of the 90 μL
SDS-ready sample (step 2), estimate the required dilution
(final volume made up with SDS blank solution) in order to
load a sensible and identical amount of each sample per gel lane
(e.g., dilute samples to 0.6 μg μL
1 to load 3 μg ¼ 5 μL per
lane; Table 1; see Note 9).
5. Prepare a dilution series (0.25, 0.5, 1, and 2 TSP) using an
SDS standard prepared from a control sample (step 3) and
diluted with SDS blank solution.

Table 1
Example of SDS sample preparation for final concentration of 0.6 μg TSP μL
1

Vol to
[TSP] in SN [TSP] in SDS SDS sample 0.6 μg Vol of blank to Final [TSP]
Sample (μg μL
1) sample (μg μL
1) Vol (μL) μL
1 add (μL) (μg μL
1)
1 2.1 1.17 90 175.5 85.5 0.6
2 2.2 1.22 90 183 92 0.6
3 2.0 1.11 90 166.5 76.5 0.6
4 2.1 1.17 90 175.5 85.5 0.6
The TSP concentration in the SDS sample prepared during extraction (Subheading 3.1) is 55.5% of the TSP concentra-
tion in the supernatant (SN)
 Quantification of Photosynthetic Enzymes by Immunoblotting 219

3.2 SDS-PAGE These instructions assume the use of a vertical gel running system,
most of which are compatible with a wide range of commercially
available gels for protein separations, making them highly adaptable
for different experimental needs. The protocol described here can
also be adapted to home-made gels [e.g., 10].
1. Carefully remove the storage comb from the top and any tape
or insert from the bottom of the gel cassette. If necessary,
straighten the sides of the wells gently with a fine pipette tip.
2. Place the gel(s) into the electrode assembly with the short gel
cassette plate(s) facing inward. Two cassettes are required to
create a functioning assembly; if running only one gel, the extra
solid plastic cassette provided functions as a “buffer dam” that
completes the upper reservoir assembly. Clamping the gel
(s) onto the electrode assembly should form a seal by joining
each gel cassette to the notch in the gasket. Fill the inner
compartment with upper buffer so that the buffer level is
higher than the inner gel plate and just below the outer gel
plate. Check that there are no leaks to the outer lower reservoir.
Wash wells with the upper buffer using a syringe or a fine
pipette tip. Fill the outer compartment with the lower reservoir
buffer.
3. Load 2 μL of the pre-stained molecular weight marker, 5 μL of
the SDS standard dilution series, and 5 μL of each sample
(0.6 μg μL
1 TSP, i.e., 3 μg TSP) into separate wells with a
fine pipette tip, layering gently from the bottom (Fig. 1; see
Note 10).
4. Place the lid on the running chamber and set the powerpack to
150 V constant. The running time for the gel is ca. 70 min.
Progress can be observed via the blue dye front migrating
down the gel. Once the dye front has moved just beyond the
bottom of the gel, turn off the power supply and remove the
gel from the running unit. Separate the gel plates using a
spatula or equivalent tool and rinse the gel with pure water.
The gel can subsequently be stained with Coomassie blue (e.g.,
for RuBisCO quantification, Fig. 2a) or used for immunoblot-
ting (e.g., for Rca quantification; Fig. 3a).
5. For staining the gel, transfer to a box with a smooth surface,
add water to wash the gel for three times of 5 min each while

Fig. 1 Typical gel layout, with protein molecular marker (M), followed by the SDS standards for the dilution
series (containing 0.25, 0.5, 1, and 2 the TSP amount used in the samples) and the samples. We
typically load SDS blank solution (B) in the very last well, which tends to generate less well-defined protein
bands when used for samples
220 J. Alejandro Perdomo et al.

Fig. 2 RuBisCO detection and quantification. (a) Proteins were separated by

SDS-PAGE and visualized by staining with Coomassie blue. From left to right:
protein molecular marker (M), sequence of Total soluble protein (TSP) extraction/
SDS standard preparation prepared from a control sample and representing a
TSP dilution series (corresponding to 0.25, 0.5, 1, and 2 the TSP amount
used in the samples) and samples of interest (1 TSP). RuBisCO large subunit
corresponds to the most abundant protein (56 kDa). (b) Calibration curve to
quantify RuBisCO amounts in leaf samples relative to a control sample. Squares
represent RuBisCO in the SDS standards (Std RuBisCO). The amount of RuBisCO
large subunit in the samples (crosses) is estimated by reference to the equation
of the calibration curve using the band intensity signal. The values for the
samples fall within the calibration curve prepared for RuBisCO
 Quantification of Photosynthetic Enzymes by Immunoblotting 221

Fig. 3 Rca detection by near-infrared fluorescence and quantification. (a) Immu-

noblot of Rca after separation by SDS-PAGE. From left to right: protein molecular
marker (M), sequence of SDS standards prepared from a control sample and
representing a TSP dilution series (corresponding to 0.25, 0.5, 1, and
2 the TSP amount used in the samples) and samples of interest (1 TSP). Both
the Rca α-isoform (46 kDa) and Rca β-isoform (42 kDa) are visible. (b) Calibra-
tion curve to quantify Rca amounts in leaf samples relative to a control sample.
Triangles and circles represent Rca α-isoform and Rca β-isoform, respectively,
in the SDS standards (Std Rca). The amount of Rca in the samples (crosses) is
estimated by reference to the equation of the calibration curve using the band
intensity signal. The values for the samples fall within the calibration curve
prepared for each Rca isoform

shaking on a rocking platform. After the three washes, replace

the water with a Coomassie blue-based stain solution and
incubate with shaking for up to 2 h. Replace the stain solution
(which should be disposed of according to your institution
222 J. Alejandro Perdomo et al.

guidelines and level of toxicity of the solution used) with water

and incubate with shaking overnight to de-stain the back-
ground. Replacing the water frequently will enable faster
de-staining of the background. The gel is then ready to photo-
graph in a suitable imaging system.

3.3 Immunoblotting Proteins that have been separated by SDS-PAGE are transferred to
a membrane electrophoretically and then detected using specific
antibodies to allow quantification. These directions assume the use
of a dry blotting system with nitrocellulose membranes (see Note
11). The procedure we use requires 2 days to be completed, but can
be adjusted to fit within a single day, if required.

3.3.1 Day 1 1. After SDS-PAGE, remove the gel from the cassette, rinse with
deionized pure laboratory water, and ensure that the gel is
clean and free of any small gel pieces prior to blotting. Remove
the top part of the gel containing the wells using a clean and
sharp razor blade and discard.
2. Place the bottom layer of the transfer stack with the nitrocellu-
lose membrane into the dry blotting transfer system. Carefully
position the gel over the membrane, adjusting as needed to
ensure that there are no bubbles between the gel and the
membrane (see Note 12). We typically transfer two mini-gels
(8.6  6.7 cm) side by side into one nitrocellulose membrane
(13  8.3 cm). Cover the gel(s) with filter paper pre-wet with
deionized pure laboratory water and the top layer of the trans-
fer stack. Use a blotting roller to ensure that there are no
bubbles in the stack. Close the dry blotting system and select
a suitable transfer method (we commonly use 20 V and 7 min; a
lower voltage or transfer time may be required when using
thinner gels).
3. Once the transfer is complete, open the lid of the dry blotting
transfer system and discard the top stack and filter paper.
Inspect the gel and membrane visually: the pre-stained protein
molecular marker should have transferred completely from the
gel to the membrane. Discard the gel. If needed, carefully use a
clean and sharp razor blade to cut off unwanted areas of the
membrane. We typically cut the membrane section
corresponding to each mini-gel and cut the top right corner
of each section to indicate the orientation of the gel/mem-
brane. Place the membrane section corresponding to each
mini-gel in a small immunoblot incubation box containing
20–25 mL TBS (enough to cover the membrane completely)
and incubate with shaking for 10 min.
4. Replace the TBS buffer with 20–25 mL 4% Blotto and incubate
with shaking for 2 h at room temperature.
 Quantification of Photosynthetic Enzymes by Immunoblotting 223

5. Rinse membrane with ~0.5% Blotto. This is done by diluting a

small volume of 4% Blotto used in the previous step in approxi-
mately 7 volumes of TBS.
6. Replace the 0.5% Blotto with 20–25 mL primary antibody per
mini-gel and incubate with shaking overnight at room temper-
ature. If needed, this incubation time can be decreased to 2 h.

3.3.2 Day 2 1. Recover the primary antibody into the tube for reusing. Rinse
membrane with pure water. Wash six times with 20–25 mL
TBST, 15 min each, with shaking.
2. Incubate the membrane for 2 h with 20–25 mL secondary
antibody with shaking (see Note 13).
3. Wash four times with TBST buffer, 15 min each, with shaking.
4. Rinse with deionized pure laboratory water. Keep the mem-
brane in pure water or dry on a filter paper in the dark until
detection.

3.4 Detection RuBisCO and Rca bands on the gel and membrane, respectively, are
and Quantification visualized by imaging with a near-infrared fluorescence detector.
1. For imaging the Coomassie-stained gel, wearing gloves, hold
the gel gently to transfer the gel from the box with water to the
imaging system tray. Select the 700 nm channel and acquisition
time. We typically use 2 min, which provides a good signal
corresponding to the protein bands with limited background
interference. Increasing the acquisition time may be useful to
increase the signal-to-background ratio for detection of faint
bands associated with low abundance or less efficiently stained
proteins.
2. For imaging the membrane, using fine-tipped or flat-head
tweezers, hold the membrane gently from one of the corners,
avoid touching the membrane in the area containing the bands
of interest, and transfer to the imaging system tray. Select the
channels and acquisition time to detect the fluorescence signal.
We typically use channels 700 and 800 nm to visualize the
pre-stained marker and the Rca bands, respectively, and a
2-min exposure time for image acquisition in each channel (for
low-abundance proteins it may be necessary to increase the
acquisition time to increase the signal-to-background ratio).
3. After capturing the image, the software accompanying the
imaging system is used to quantify band intensities for each
protein of interest. This typically involves drawing or adding a
rectangle to each band and selecting the area to be considered
as the background (e.g., area around the protein band). The
software should then output a signal that corresponds to the
relative intensity of each band. The band intensity signal
224 J. Alejandro Perdomo et al.

corresponding to the RuBisCO large subunit (56 kDa, Fig. 2a)

is used to estimate the amount of RuBisCO holoenzyme in the
samples relative to the control. RuBisCO is the most abundant
protein in leaves [1, 6] and the band observed at 56 kDa will by
and large correspond to the RuBisCO large subunit.
4. Once the signal intensity of the bands has been determined, the
amounts of RuBisCO (in the Coomassie-stained gel; Fig. 2b)
and Rca (in the membrane; Fig. 3b) in the leaf samples relative
to the control plants are determined by reference to a calibra-
tion curve prepared using the dilution series of the SDS stan-
dard (see Notes 14 and 15).

4 Notes

1. For glasshouse- and field-grown wheat plants, we typically

homogenize ca. 2.5 cm2 leaf samples (corresponding to
ca. 40 mg of fresh weight) in 500 μL extraction buffer (i.e., a
5:1 ratio of cm2 leaf area to mL extraction buffer) to obtain a
total soluble protein concentration in the leaf extracts suitable
for our needs. Different leaf areas and extraction buffer
volumes can be used, according to leaf material availability,
and taking into account the protein content in leaves from
various plant species or plants grown under varying environ-
mental conditions.
2. Keeping the extraction at low temperature (4  C) and adding
fresh reducing agents and protease inhibitors are critical to
prevent proteolysis of photosynthetic proteins.
2-Mercaptoethanol, PMSF, and leupeptin must be added
fresh just before extraction.
3. We typically prepare concentrated (10 or 5) stock solutions
of the SDS-PAGE running buffers. Upper buffer 10 stock
solution: 0.4 M Trizma-base, 0.4 M borate, 1% SDS, final pH
adjusted to 8.4 by adding small amounts of borate powder as
needed. Filtered through a 0.22 μm nitrocellulose membrane,
stored at 4  C, and diluted 1:10 with deionized pure water to
have 1 solution. Lower buffer 5 stock solution: 4 M Trizma-
base adjusted to pH 9.18 with concentrated HCl. Filtered
through a 0.22 μm nitrocellulose membrane, stored at 4  C,
and diluted 1:5 with deionized pure water to have 1 solution.
4. We typically prepare a concentrated (10) stock solutions of
the Tris-buffer-saline (TBS) buffer: 0.2 M Tris, 1.5 M NaCl,
pH 7.5 adjusted with HCl. Filtered through a 0.22 μm nitro-
cellulose membrane, stored at 4  C, and diluted 1:10 with
deionized pure water to have 1 solution.
 Quantification of Photosynthetic Enzymes by Immunoblotting 225

5. Tween-20 is a viscous solution. Trimming the end of the tip

and slowly drawing up the liquid allow for easier and more
accurate pipetting. To make Protein transfer/ tris-buffer-
saline-tween (TBST), we use the 10 concentrated TBS
described in Note 4, dilute it 1:10 with deionized pure water,
and then add the 0.05% (v/v) Tween-20 (e.g., 0.5 mL Tween-
20 in 1 L of solution).
6. Anti-Rca antibody was a gift of Dr. Mike Salvucci. An equiva-
lent antibody is available commercially.
7. Different secondary antibodies and corresponding detection
methods can be used, such as chemiluminescence and colori-
metric methods. Whichever method is used it is important to
ensure compatibility between primary and secondary antibo-
dies. Fluorescence imaging in the near-infrared region provides
a wide linear dynamic range of detection, providing high sensi-
tivity for weak bands and avoiding saturation of signal for
strong bands.
8. Absolute quantification requires the use of purified proteins
(e.g., see purification methods in [11–13] for RuBisCO and
Rca). It is preferable to use proteins purified from the same
species as the plants being analyzed to avoid any confounding
differences in antibody reactivity. When purified proteins are
not available, it is possible to quantify the amount of a target
protein relative to a control sample.
9. The volume of sample loaded in each lane can be adjusted
depending on the samples’ TSP concentration and the gel
properties. The well volume will depend on the gel thickness
and number of wells per gel. A volume of 5 μL of sample is
conveniently loaded into each well in 15-well, 1.5 mm thick
gels; and 3 μg of TSP loaded per lane results in clear and well-
defined protein bands after separation by SDS-PAGE.
10. Loading SDS blank solution in unused wells in the gel, includ-
ing the well at the edge of the gel, promotes a homogeneous
run of all the samples in all the lanes and ensures that all bands
are horizontal and produce an even picture.
11. There are several types of membranes for blotting, including
nitrocellulose, nylon, and polyvinylidene difluoride (PVDF). It
is important to ensure that the membrane type used does not
fluoresce at the wavelength selected for detection (e.g., some
PVDF membranes fluoresce at 700 nm and this can obscure
faint bands).
12. Air bubbles between the gel and the membrane will prevent
protein transfer in the area of the bubble. Adding a small
amount (ca. 5 mL, by eye) of deionized pure laboratory
water on the membrane before carefully placing the gel on
226 J. Alejandro Perdomo et al.

top of the membrane helps smoothen this task and prevent air
bubbles.
13. Incubation of the membrane with a fluorescent secondary
antibody should be carried out in the absence of light. For
this purpose, we typically carry out all membrane incubations
and washes in a black opaque immunoblot incubation box. The
fluorophore is stable for several years and the membrane can be
stored in darkness and imaged repeatedly.
14. To quantify the abundance of RuBisCO in the gel and Rca in
the immunoblot a calibration curve is prepared using the band
intensity signal and the relative TSP concentrations of the SDS
standard dilution series (Figs. 2b and 3b). The amounts of
RuBisCO and Rca are determined by applying the equation
of the calibration curve to the band intensity signal for
RuBisCO (in the gel) or Rca (in the immunoblot) for each
sample.
15. An alternative to using a standard curve prepared with a SDS
standard dilution series would be to use a loading control (e.g.,
housekeeping protein or total protein stain). Some imaging
systems enable the detection of more than one protein on the
same immunoblot, e.g., by using two antigens derived from
different host species and corresponding secondary antibodies
labeled with spectrally distinct fluorescent dyes.

Acknowledgments

We thank Dr. Mike Salvucci (USDA-ARS) for the generous gift of

Rca antibody and Dr. Doug Orr for helpful comments on the
manuscript. JAP was funded through a Rothamsted Research Fel-
lowship awarded to ECS. Rothamsted Research receives grant-
aided support from the Biotechnology and Biological Sciences
Research Council (BBSRC) 20:20 Wheat Institute Strategic
Programme. CRGS and ECS acknowledge funding from the Inter-
national Wheat Yield Partnership (IWYP64). ECS also acknowl-
edges support from a sub-contract to the Bill & Melinda Gates
Foundation award RIPE.

References

1. Carmo-Silva E, Scales JC, Madgwick P, Parry
MAJ (2015) Optimising Rubisco and its regu-
lation for greater resource use efficiency. Plant
Cell Environ 38:1817–1832
2. Raines CA (2011) Increasing photosynthetic
carbon assimilation in C3 plants to improve
crop yield: current and future strategies. Plant
Physiol 155:36–42
3. Parry MAJ, Andralojc P, Scales J, Salvucci M,
Carmo-Silva A, Alonso H, Whitney S (2013)
Rubisco activity and regulation as targets for
crop improvement. J Exp Bot 64:717–730
4. Lefebvre S, Lawson T, Fryer M, Zakhleniuk
OV, Lloyd JC, Raines CA (2005) Increased
sedoheptulose-1,7-bisphosphatase activity in
transgenic tobacco plants stimulates
 Quantification of Photosynthetic Enzymes by Immunoblotting
photosynthesis and growth from an early stage
in development. Plant Physiol 138:451–460
5. Simkin AJ, McAusland L, Headland LR,
Lawson T, Raines CA (2015) Multigene
manipulation of photosynthetic carbon assimi-
lation increases CO2 fixation and biomass yield
in tobacco. J Exp Bot 66:4075–4090
6. Carmo-Silva E, Andralojc P, Scales J, Driever S,
Mead A, Lawson T, Raines C, Parry MAJ
(2017) Phenotyping of field-grown wheat in
the UK highlights contribution of light
response of photosynthesis and flag leaf longev-
ity to grain yield. J Exp Bot 68(13):3473–3486.
https://doi.org/10.1093/jxb/erx169
7. Portis AR (2003) Rubisco activase – Rubisco’s
catalytic chaperone. Photosynth Res 75:11–27
8. Salvucci ME, Werneke J, Ogren W, Portis AJ
(1987) Purification and species distribution of
Rubisco activase. Plant Physiol 84:930–936
9. Bradford MM (1976) A rapid and sensitive
method for the quantitation of microgram quan-
tities of protein utilizing the principle of protein-
dye binding. Anal Biochem 72:248–254
227
10. Kurien BT, Scofield RH (2015) Multiple
immunoblots by passive diffusion of proteins
from a single SDS-PAGE gel. In: Kurien BT,
Scofield RH (eds) Western blotting: methods
and protocols, Methods in molecular biology.
Springer, New York, pp 77–86
11. Carmo-Silva E, Barta C, Salvucci ME (2011)
Isolation of ribulose-1,5-bisphosphate carbox-
ylase/oxygenase from leaves. In: Carpentier R
(ed) Photosynthesis research protocols, Meth-
ods in molecular biology, vol 684, 2nd edn.
Springer, New York, pp 339–347
12. Barta C, Carmo-Silva E, Salvucci ME (2011)
Purification of Rubisco activase from leaves or
after expression in Escherichia coli. In: Carpen-
tier R (ed) Photosynthesis research protocols,
Methods in molecular biology, vol 684, 2nd
edn. Springer, New York, pp 363–374
13. Orr DJ, Carmo-Silva E (2018) Extraction of
Rubisco to determine catalytic constants. In:
Covshoff S (ed) Photosynthesis: methods and
protocols, Methods in molecular biology, vol
1770. Springer, New York
 Chapter 13

Extraction of RuBisCO to Determine Catalytic Constants

Douglas J. Orr and Elizabete Carmo-Silva

Abstract
RuBisCO enables net carbon fixation through the carboxylation of RuBP during photosynthesis. Its
complex biochemistry and catalytic diversity found among different plants make characterization of
RuBisCO properties useful for investigations aimed at improving photosynthetic performance. This chap-
ter reports methods for rapid extraction of soluble proteins to examine RuBisCO catalytic properties, and
for large-scale purification of RuBisCO from leaves to measure the specificity of the enzyme toward its
gaseous substrates.

Key words RuBisCO, CO2 fixation, Calvin-Benson-Bassham cycle, Protein extraction, Enzyme
catalysis

1 Introduction

Ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) is

the chloroplast protein responsible for the assimilation of carbon
into the biosphere. RuBisCO is noted for its seemingly inefficient
nature compared to other carboxylases [1, 2]. The enzyme’s
bifunctional nature as a carboxylase/oxygenase results from poor
specificity for CO2, as RuBisCO frequently combines ribulose
1,5-bisphosphate (RuBP) with O2 instead of CO2. RuBP oxygena-
tion initiates the photorespiratory cycle, with additional energy
requirements and loss of previously fixed CO2 and NH4 [3, 4]. In
addition, RuBisCO is prone to inhibition by sugar phosphates,
including “misfire” products generated from incomplete or incor-
rect attempts at either carboxylation or oxygenation of RuBP. The
removal of inhibitors requires the action of the molecular chaper-
one RuBisCO activase (Rca) to restore catalytic competency [5].
These RuBisCO inefficiencies are consistent across higher
plants, despite variation in catalysis among species [6–8], and
mean that plants typically invest significant resources into produc-
ing sufficient RuBisCO to maintain adequate rates of photosynthe-
sis. For example, in wheat leaves, RuBisCO can constitute 50% of

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_13, © Springer Science+Business Media, LLC, part of Springer Nature 2018

229
230 Douglas J. Orr and Elizabete Carmo-Silva

the total leaf-soluble protein [9]. Improving the efficiency of the

enzyme is therefore a major target for improvement in crops in
order to increase photosynthetic and resource-use efficiency, bio-
mass, and yield [5, 10].
It is often desirable to isolate RuBisCO from other leaf proteins
for closer scrutiny of its biochemistry under defined conditions. For
analyses of the maximum carboxylation activity and affinity of the
RuBisCO enzyme for CO2, a rapid extraction is typically preferable
in order to maintain the highest measurable catalytic activity and
minimize potential effects of proteolysis on the ca. 560 kDa hex-
adecameric structure [11]. Measurement of RuBisCO specificity
[12] requires separation from any leaf proteins that may consume
CO2 or O2. The large size of the enzyme and its abundance mean
that ion-exchange chromatography can be used for purifying larger
quantities of the protein, suitable for specificity measurements as
well as other techniques such as immunoblotting [13] and investi-
gation of the interaction with Rca [e.g., 14].
This chapter details (1) procedures that have been developed
and optimized for the rapid extraction of highly active native
RuBisCO from leaves, and (2) a technique for large-scale purifica-
tion of RuBisCO. Both methods are also suitable for extracting
RuBisCO from transgenic plants expressing a modified or foreign
RuBisCO.

2 Materials

2.1 Rapid Isolation 1. Frozen leaf tissue, approximately 20–30 cm2, stored at
80  C.
of RuBisCO from 2. Leaf extraction buffer: 100 mM Bicine-NaOH, pH 7.9, 5 mM
Leaves MgCl2, 1 mM EDTA, 2 mM benzamidine, 5 mM aminoca-
proic acid, 50 mM 2-mercaptoethanol, 5% PEG 4000, 1 mM
phenylmethylsulfonate (PMSF), 10 mM dithiothreitol (DTT),
plant protease inhibitor cocktail (PI), 10 mM NaHCO3, and
1% (w/v) polyvinylpolypyrrolidone (PVPP) (see Note 1). Add
PEG, PMSF, DTT, PI, NaHCO3, and PVPP fresh just before
extracting.
3. Mortar (~60 mm) and pestle.
4. Fine-grade sand.
5. Refrigerated high-speed centrifuge with fixed-angle rotors.
6. Plastic microcentrifuge tubes, 1.5 and 2 mL capacity.
7. Sephadex G-25 desalting columns of approximately 8 mL bed
volume, 5 cm long (see Note 2).
8. Desalt buffer: 100 mM Bicine-NaOH, pH 8.0, 10 mM MgCl2,
1 mM EDTA, 1 mM benzamidine, 1 mM aminocaproic acid,
10 mM DTT, 10 mM NaHCO3, 2% PEG 4000. Add DTT,
 Rubisco Extraction From Leaves 231

NaHCO3 and PEG fresh on the day of extraction (see Notes 1

and 3).
9. Compressed N2 gas tank.
10. Liquid nitrogen and dewar.
11. Retort stand and clamps to hold desalting columns.

2.2 Large-Scale 1. Frozen leaf tissue, approximately 100 g, stored at
80  C. For
Purification most species, this corresponds to approximately two 50 mL
of RuBisCO from tubes full of leaves.
Leaves 2. Leaf extraction buffer: 40 mM Triethanolamine (TEA)-
NaOH, pH 8.0, 10 mM MgCl2, 1 mM EDTA, 2 mM
KH2PO4, 2 mM benzamidine, 5 mM aminocaproic acid,
50 mM 2-mercaptoethanol, 5% PEG 4000, 1 mM phenyl-
methylsulfonate (PMSF), 10 mM DTT, 10 mM NaHCO3,
and 1% (w/v) polyvinylpolypyrrolidone (PVPP) (see Note 1).
Add PMSF, DTT, NaHCO3, and PVPP fresh just before
extracting.
3. Anion-exchange buffer A (Buffer A): 25 mM Triethanolamine
(TEA)-NaOH, pH 7.8, 5 mM MgCl2, 1 mM EDTA, 2 mM
benzamidine, 5 mM aminocaproic acid, 12.5% (v/v) glycerol,
10 mM DTT, 10 mM NaHCO3. Both buffer A and B can be
filtered through a 0.2 μM filter and stored at 4  C for up to
6 months. Add DTT and NaHCO3 fresh immediately before
use (see Notes 1 and 4).
4. Anion-exchange buffer B (Buffer B): As described for buffer A,
with addition of 1 M NaCl. Add DTT and NaHCO3 fresh
immediately before use (see Notes 1 and 4).
5. 60% (w/v) PEG 4000 solution (see Note 3).
6. Plant protease inhibitors (see Note 1).
7. Blender with ~1 L capacity.
8. Muslin cloth.
9. Large plastic or glass funnel.
10. 250 mL Centrifuge bottles.
11. 10 mL Ultracentrifuge tubes.
12. Anion-exchange columns, 5 mL sepharose bed volume (6%
highly cross-linked agarose, 90 μm particle size, strong anion
ligand such as quaternary amine), suitable for use with chro-
matographic systems.
13. Low-, medium-, or high-pressure chromatographic system
with UV detector and fraction collector.
14. Refrigerated high-speed centrifuge with fixed-angle rotors to
accommodate 250 mL bottles.
232 Douglas J. Orr and Elizabete Carmo-Silva

15. Refrigerated ultra-speed centrifuge with fixed-angle rotors to

accommodate 10 mL tubes.
16. 0.45 μM Nitrocellulose membrane filters, compatible with
syringes.
17. 10 mL Disposable plastic syringe.
18. Plastic microcentrifuge tubes, 1.5 and 2 mL capacity.
19. Glass measuring cylinders (100 mL, 250 mL).
20. 10 mL Pyrex homogenizer (“Wheaton” type or similar).

3 Methods

This section describes procedures for isolation of RuBisCO from

leaves. Unless stated otherwise, all steps are performed at 4  C and
as quickly as possible to minimize proteolysis and subsequent loss
of RuBisCO activity.

3.1 Rapid Isolation 1. Precool a mortar and pestle either in a refrigerator or on ice.
of RuBisCO from Precool the centrifuge to 4  C. Transfer tube or foil packet
Leaves containing previously
80  C stored tissue to liquid nitrogen
container.
2. Sparge the desalting buffer by bubbling with N2 (~5 min/
100 mL) before adding the DTT, NaHCO3, and PEG (see
Note 5).
3. Pre-equilibrate a Sephadex G-25 column with 20 mL of freshly
prepared desalting buffer. Have the column set up on a retort
stand at 4  C.
4. With the mortar sitting on ice, add 3.5 mL of ice-cold extrac-
tion buffer to the mortar, and then add the additional compo-
nents (PMSF, DTT, NaHCO3, PEG 4000, PI, PVPP), and
100 mg of sand.
5. Remove leaf tissue from liquid nitrogen, add to the mortar, and
grind on ice. The tissue should homogenize in the buffer (see
Note 6). Avoid grinding for excessive time (1 min max) to
prevent the buffer from warming due to friction.
6. Transfer the homogenate to two 1.5 mL tubes and centrifuge
for 2 min at 14,000  g to pellet insoluble cell debris.
Remove at least 2 mL of supernatant and transfer to ice.
7. Load the supernatant onto the Sephadex G-25 column and
allow it to run through the column. Add 1.5 mL of desalt
buffer and allow to run through.
8. Place a 2 mL tube underneath the column, add 1.5 mL of
desalt buffer to the column, and collect the flow through.
This will contain the majority of the active RuBisCO. Add
 Rubisco Extraction From Leaves 233

20 μL of 1 M MgCl2 and 20 μL of protease inhibitor cocktail to

the collected flow through, and mix well.
9. The extract can now be allowed to activate before use to
determine RuBisCO catalytic constants by 14CO2 consump-
tion. An aliquot should be set aside for quantification of
RuBisCO in the extract, either by SDS-PAGE/
immunoblotting [13] or by 14C–CABP binding [15, 16].

3.2 Large-Scale For some in vitro analyses, large amounts of RuBisCO protein are
Purification required. This procedure is designed to extract a large (>50 mg)
of RuBisCO from quantity of RuBisCO from leaves and can be applied to diverse
Leaves species (see Table 1 for examples, and see Note 7). Unless indicated
otherwise, all steps are performed at 4  C. Ideally, equipment such
as the HPLC and stirring plates should be in a refrigerator or cold
room to ensure that samples are kept cold throughout the entire
process. Buffers and stock solutions are prepared as described in
Subheading 2 and accompanying notes.
1. Precool all buffers and equipment including centrifuges and
blender. Pre-soak a piece of muslin (ca. 20 cm2) in deionized
pure laboratory H2O at 4  C.
2. Add 195 mL of ice-cold extraction buffer to the blender, and
then add: 0.975 mL 0.5 M NaHCO3 (to 1 mM final concen-
tration); 0.975 mL 1 M DTT (to 5 mM); 1.95 mL 0.1 M
PMSF (to 1 mM); and 1.95 g PVPP. Mix for a few seconds in
the blender.

Table 1
Protein yields of RuBisCO purified from the indicated species using the
protocol described herein

Species mg RuBisCO
Amaranthus caudatus 56
Coriandrum sativum 118
Hordeum vulgare 58
Nicotiana tabacum 81
Phaseolus acutifolius 63
Phaseolus vulgaris 61
Puccinellia distans 54
Solanum tuberosum 110
Triticum aestivum 160
Vigna unguiculata 46
In all cases, approximately 100 g fresh weight of leaf material was used (two 50 mL tubes
full of leaves)
234 Douglas J. Orr and Elizabete Carmo-Silva

3. Transfer frozen leaf material from liquid N2 to the blender, and

blend into a homogenate, usually 45–60 s (see Notes 6 and 8).
4. Remove excess water from the muslin and layer it into the
funnel with a 250 mL centrifuge bottle underneath. Filter the
supernatant by pouring through the muslin.
5. Clarify the homogenate by centrifugation for 12 min at
22,000  g and 4  C.
6. Pour the supernatant into a cold 250 mL cylinder and deter-
mine the volume. Transfer to a cold 600 mL beaker and add
60% PEG to a final concentration of 20% using the following
formula:

Vol:60%PEG ¼ Vol:supernatant  0:52

And add additional MgCl2 to return the concentration to

20 mM via the formula

Vol:supernatant þ Vol:60%PEG
Vol:1 M MgCl2 ¼
83

7. Mix on a stirring plate for 30 min at 4  C to precipitate the

RuBisCO.
8. Transfer to clean cold 250 mL centrifuge bottles and pellet the
PEG-precipitated RuBisCO via centrifugation for 20 min at
22,000  g and 4  C. Use a slow deceleration setting on the
centrifuge (see Note 9).
9. Discard the supernatant and use a cold glass rod to resuspend
the pellet in 10 mL of ice-cold, freshly prepared Buffer
A. Transfer the homogenate and any lumps of pellet into a
cold glass homogenizer and add 100 μL protease inhibitor
cocktail. Homogenize the remaining pellet (1–2 min).
10. Transfer the homogenate to 10 mL ultracentrifuge tubes and
clarify at 200,000  g and 4  C for 20 min, with slow
deceleration.
11. Use a small flexible piece of plastic tubing and a 10 mL syringe
to remove the supernatant, taking care to avoid disturbing the
pellet. Filter through a 0.45 μM syringe filter (see Note 10).
12. Using a chromatographic system equipped with a 5 mL anion-
exchange column, load the supernatant onto the column pre-
viously equilibrated with Buffers A and B, as described in the
manufacturer’s instructions.
13. Apply a 200 mL linear gradient of 0–1 M NaCl over the
column at 2 mL min
1, mixing Buffer A (no NaCl) and Buffer
B (1 M NaCl) to produce the gradient. Fractionate the eluant.
RuBisCO elutes from the column at ca. 250 mM NaCl. The
use of a chromatography system with a UV detector for
 Rubisco Extraction From Leaves 235

Fig. 1 Example trace from purification of RuBisCO from Triticum aestivum using a 5 mL HiTrap Q Sepharose
column and an ÄKTA protein purification system. Colored lines indicate monitoring of eluant absorbance at
280 nm (blue), eluant conductivity (red), and Buffer B percentage being loaded (yellow). Red markers along the
x-axis indicate fractionation. Elution of RuBisCO from the column is highlighted in green

Fig. 2 SDS-PAGE gel of purified RuBisCO from Triticum aestivum (wheat) and
Gossypium hirsutum (cotton). Bands for RuBisCO large (LSU, ~55 kDa) and small
(SSU, ~13 kDa) subunits are indicated

monitoring absorbance at 280 nm allows easy identification of

the primary protein peak (see Fig. 1 for an example).
14. Pool the fractions containing the RuBisCO peak. RuBisCO
purity can be assessed by SDS-PAGE (Fig. 2) and concentra-
tion can be determined via absorbance at 280 nm [17, 18],
using a RuBisCO-specific quantification method [15, 16], or
immunoblotting [13]. If a more concentrated sample is
236 Douglas J. Orr and Elizabete Carmo-Silva

desired, then the pooled fractions can be added to a

pre-washed, ice-cold centrifugal concentrating filter unit and
centrifuged at 1500  g and 4  C for 20–30 min (see Note 11).
15. The purified protein can then be diluted to a desired concen-
tration (using Buffer A), and aliquots snap-frozen for storage at

80  C.

4 Notes

1. For extracting numerous samples, we prepare a number of

solutions in advance and add a small aliquot of these to the
leaf extraction buffer immediately prior to grinding. We typi-
cally prepare DTT as a 1 M stock, and NaHCO3 as a 0.5 M
stock, and store these at 4  C. We then add 35 and 70 μL of
each, respectively, to 3.5 mL of extraction buffer. We add
320 μL of 60% (w/v) PEG 4000 to obtain a 5% final concen-
tration in the extraction buffer (see Note 3). We prepare a
0.1 M PMSF stock in ethanol and store this at
20  C, adding
35 μL to the extraction buffer before use. In addition, we use a
plant protease inhibitor cocktail that comes as a solution in
DMSO and add 10 μL cocktail mL
1 extraction buffer. We
recommend users include, either as a cocktail or individually,
suitable protease inhibitors that inhibit as wide a range of
proteases as possible (e.g., trypsin, metalloproteases, serine
proteases, acid proteases, and cysteine proteases). Storage of
the complete leaf extraction buffer is not recommended.
2. On the day of use, Sephadex G-25 columns should be equili-
brated with 20 mL of freshly prepared desalt buffer. After
elution of RuBisCO, they must be washed with at least
40 mL deionized pure laboratory H2O. We use each Sephadex
G-25 column no more than three times.
3. To prepare 60% (w/v) PEG 4000 solution, add 600 g PEG
4000 to 520 mL H2O and mix on a heated stirring plate at
approximately 50  C until PEG dissolves.
4. All solutions used for chromatographic separation are filtered
through a 0.22 μm nitrocellulose membrane filter upon
preparation.
5. Where the extract is intended to be used in assaying RuBisCO
activity through 14CO2 consumption, it is critical to have
known CO2 and O2 conditions in the assay. Thus, the desalt
buffer is bubbled with N2 to remove dissolved CO2 and O2
prior to the addition of the final buffer components on the day
of use.
6. When adapting this method for different plant species, the
amount of leaf tissue is an important consideration. Excessive
 Rubisco Extraction From Leaves 237

tissue amounts will make grinding difficult and prevent recovery

of the buffer containing the homogenized material. Ideally all of
the leaf material should be homogenized to an extent that remov-
ing the liquid leaves very little leaf residue in the mortar. For the
larger scale extraction, we have found ca. 100 g fresh weight in
200 mL leaf extraction buffer to be a good starting point. To
avoid overloading the blender, add half the material (ca. 50 g),
blend for ca. 20 s, and if the homogenate is still very liquid then
add the remaining ca. 50 g of leaf material (see also Note 8).
7. Alternative equivalent techniques are available, both for
RuBisCO and related proteins, such as RuBisCO activase
[e.g., 19, 20]; selection of a particular technique will largely
depend on available equipment and materials in the laboratory,
such as sucrose gradients.
8. If possible, the blending step should be done at 4  C. Alterna-
tively, minimize the time the blender is at room temperature.
Excessive amounts of time blending will heat the sample and
increase the likelihood of RuBisCO proteolysis. When extract-
ing RuBisCO from grasses, it is recommended to cut the leaf
blades into short (ca. 5 cm long) sections before snap-freezing.
This will prevent long pieces of grass from fouling the blades of
the blender and help to produce a consistent homogenate.
9. A slow deceleration setting will help to prevent the PEG pre-
cipitate (which contains the RuBisCO) from being dislodged,
and retain most of the precipitate in the pellet.
10. After ultracentrifugation, there may be a small amount of
insoluble material in the supernatant. Avoid transferring this to
the syringe. The filtration step will ensure removal of small
amounts that may get into the syringe. For dense extracts, it
may be necessary to use a number of filters to efficiently clean the
supernatant in a reasonable time and avoid warming the extract.
11. When selecting a centrifugal concentrating filter, be aware of
the molecular weight cutoff of the filter (typically quoted as
MWDA, molecular weight in Daltons). Using a filter with a
MWDA of 100 kDa will allow smaller proteins to go through
the membrane while maintaining the large RuBisCO complex
in the reservoir. These filters can also be used to facilitate buffer
exchange if required.

Acknowledgments

The authors thank André Alcântara and Dr. Karen Harper (Lan-
caster University) for technical assistance that led to method
improvements, and Dr. Cristina Sales (Lancaster University) for
providing the gel image. The authors acknowledge funding
through a sub-contract from the University of Illinois as part of
238 Douglas J. Orr and Elizabete Carmo-Silva
the Bill & Melinda Gates Foundation award RIPE: Realizing
Increased Photosynthetic Efficiency.
References
1. Tcherkez G (2013) Modelling the reaction
mechanism of ribulose-1,5-bisphosphate car-
boxylase/oxygenase and consequences for
kinetic parameters. Plant Cell Environ
36:1586–1596
2. Tcherkez GG, Farquhar GD, Andrews TJ
(2006) Despite slow catalysis and confused
substrate specificity, all ribulose bisphosphate
carboxylases may be nearly perfectly optimized.
Proc Natl Acad Sci U S A 103:7246–7251
3. Raines CA (2003) The Calvin cycle revisited.
Photosynth Res 75:1–10
4. Andersson I (2008) Catalysis and regulation in
Rubisco. J Exp Bot 59:1555–1568
5. Carmo-Silva E, Scales JC, Madgwick PJ, Parry
MAJ (2015) Optimizing Rubisco and its regu-
lation for greater resource use efficiency. Plant
Cell Environ 38:1817–1832
6. Galmés J, Kapralov MV, Andralojc PJ et al
(2014) Expanding knowledge of the Rubisco
kinetics variability in plant species: environ-
mental and evolutionary trends. Plant Cell
Environ 37:1989–2001
7. Orr DJ, Alcântara A, Kapralov MV et al (2016)
Surveying Rubisco diversity and temperature
response to improve crop photosynthetic effi-
ciency. Plant Physiol 172:707–717
8. Sharwood RE, Ghannoum O, Kapralov MV
et al (2016) Temperature responses of Rubisco
from Paniceae grasses provide opportunities
for improving C3 photosynthesis. Nat Plants
2:16186
9. Carmo-Silva E, Andralojc PJ, Scales JC et al
(2017) Phenotyping of field-grown wheat in
the UK highlights contribution of light
response of photosynthesis and flag leaf lon-
gevity to grain yield. J Exp Bot 68:3473–3486
10. Parry MAJ, Andralojc PJ, Scales JC et al (2013)
Rubisco activity and regulation as targets for
crop improvement. J Exp Bot 64:717–730
11. Sharwood RE, Sonawane BV, Ghannoum O,
Whitney SM (2016) Improved analysis of C4
and C3 photosynthesis via refined in vitro assays
of their carbon fixation biochemistry. J Exp Bot
67:3137–3148
12. Prins A, Orr DJ, Andralojc PJ et al (2016)
Rubisco catalytic properties of wild and domes-
ticated relatives provide scope for improving
wheat photosynthesis. J Exp Bot
67:1827–1838
13. Perdomo JA, Sales CRG, Carmo-Silva E
(2018) Quantification of photosynthetic
enzymes in leaf extracts by immunoblotting.
In: Covshoff S (ed) Photosynthesis: methods
and protocols, Methods in molecular biology,
vol 1770. Springer, New York
14. Carmo-Silva E, Salvucci ME (2013) The regu-
latory properties of Rubisco activase differ
among species and affect photosynthetic
induction during light transitions. Plant
Physiol 161:1645–1655
15. Parry MAJ, Andralojc PJ, Parmar S et al (1997)
Regulation of Rubisco by inhibitors in the
light. Plant Cell Environ 20:528–534
16. Whitney SM, Von Caemmerer S, Hudson GS,
Andrews TJ (1999) Directed mutation of the
Rubisco large subunit of tobacco influences
photorespiration and growth. Plant Physiol
121:579–588
17. Wishnick M, Lane MD (1971) Ribulose
diphosphate carboxylase from spinach leaves.
Methods Enzymol 23:570–577
18. McCurry SD, Gee R, Tolbert NE (1982) Ribu-
lose-1,5-bisphosphate carboxylase/oxygenase
from spinach, tobacco, or tobacco leaves.
Methods Enzymol 90:515–521
19. Barta C, Carmo-Silva E, Salvucci ME (2011)
Purification of Rubisco activase from leaves or
after expression in Escherichia coli. In: Carpen-
tier R (ed) Methods in molecular biology, Pho-
tosynthesis research protocols, vol 684, 2nd
edn. Humana Press, New York, pp 363–374
20. Carmo-Silva E, Barta C, Salvucci ME (2011)
Isolation of ribulose-1,5-bisphosphate carbox-
ylase/oxygenase from leaves. In: Carpentier R
(ed) Methods in molecular biology, Photosyn-
thesis research protocols, vol 684, 2nd edn.
Humana Press, New York, pp 339–347
 Chapter 14

Spectrophotometric Determination of RuBisCO Activity

and Activation State in Leaf Extracts
Cristina R. G. Sales, Gustaf E. Degen, Anabela Bernardes da Silva,
and Elizabete Carmo-Silva

Abstract
RuBisCO plays a central role in photosynthesis and, due to its catalytic inefficiencies, frequently limits CO2
assimilation in fully illuminated leaves at the top of unstressed crop canopies. The CO2-fixing enzyme is
heavily regulated and not all the enzyme present in the leaf is active at any given moment. In this chapter, a
spectrophotometric assay is described for measuring RuBisCO activity and activation state in leaf extracts.
Most of the assay components are available commercially and others can be produced by established
protocols, making adoption of the assay achievable by most plant biochemistry laboratories. Its relative
high-throughput capacity enables large-scale experiments aimed at screening germplasm for improved
RuBisCO function.

Key words Carboxylation, Enzyme activity, Activation state, Spectrophotometry, Microplate reader,
NADH, PK-LDH, 14CO2

1 Introduction

Ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO)

plays a central role in all photosynthetic organisms, enabling CO2
fixation via the carboxylation of ribulose 1,5-bisphosphate (RuBP)
[1]. However, RuBisCO is catalytically inefficient and is commonly
the limiting step of CO2 assimilation [1, 2]. Increasing the rate of
carbon assimilation in plants can contribute significantly toward
meeting future crop yield demands in order to provide food secu-
rity [3]. Catalytic diversity in RuBisCO is found in a range of crop
cultivars, landraces, and wild relatives, and provides a resource that
can be used in breeding programs to improve carbon fixation [e.g.,
4, 5]. In order to exploit this diversity, an accurate and reasonably
high-throughput method for studying RuBisCO activity is
required.
A great proportion of previous research measuring RuBisCO
activity has used assays with 14CO2 to monitor production of the

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_14, © Springer Science+Business Media, LLC, part of Springer Nature 2018

239
240 Cristina R. G. Sales et al.

acid-stable product 3-phosphoglycerate (3-PGA) radiometrically.

While highly accurate and specific, this method depends on hazard-
ous and expensive substances, rigorous safety rules, as well as highly
trained and qualified researchers, which limits its application. This
chapter provides a detailed description of a spectrophotometric
assay previously proposed by Scales, Parry, and Salvucci [6]. By
using a microplate assay system, it is possible to analyze a large
number of plants in a cost-effective way, without requiring
specialized facilities associated with the 14CO2-based assay.
The spectrophotometric assay described here uses five reactions
to couple RuBP carboxylation and 3-PGA formation to NADH
oxidation (Fig. 1). 3-PGA is converted to 2-phosphoglycerate
(2-PGA) by 2,3-dPGA-dependent phosphoglycerate mutase

Fig. 1 Scheme for the spectrophotometric assay method to determine RuBisCO

activity in leaf extracts. Five reactions couple RuBP carboxylation to NADH
oxidation. Red indicates components added to the assay, while the products of
each reaction are indicated in black. In reaction 1, RuBisCO in the leaf extract
catalyzes the carboxylation of RuBP (ribulose 1–5, bisphosphate) which yields
two 3-PGA (3-phosphoglycerate). In the second reaction, dPGM
(phosphoglycerate mutase) catalyzes the transfer of phosphate from C-3 to
C-2 via the intermediate 2,3diPGA (2,3-bisphosphoglycerate), resulting in
2-PGA (2-phosphoglycerate). 2-PGA is converted into PEP
(phosphoenolpyruvate) in the third reaction, catalyzed by enolase. In the fourth
step, PK (pyruvate kinase) converts PEP to pyruvate by transferring a phosphate
group onto ADP, resulting in the formation of ATP. In the fifth and final reaction,
pyruvate is reduced to lactate by LDH (lactate dehydrogenase), which also
oxidizes NADH to NAD+ (highlighted by the green box). The decrease in NADH
concentration is monitored by the change in absorbance at 340 nm
 Spectrophotometric Determination of Rubisco Activity 241

(dPGM). Enolase then converts 2-PGA to phosphoenolpyruvate

(PEP), which is a substrate for pyruvate kinase (PK), leading to the
formation of pyruvate with use of ADP. In the last step, pyruvate is
reduced to lactate by lactate dehydrogenase (LDH), and this final
reaction is coupled to the oxidation of NADH to NAD+. The
resulting decrease in NADH concentration can be monitored by
measuring the absorbance of the assay solution at 340 nm by a
microplate reader using UV-transparent 96-well plates.

2 Materials

Ultrapure laboratory water and high-grade reagents are recom-

mended. The chemicals can be obtained from commercial sources,
except dPGM, which can be produced according to [7]. RuBP is
available commercially or can alternatively be synthesized enzymat-
ically [8] (see Note 1). Stock solutions can be prepared and frozen
in aliquots that are stable for several months. Using aliquots of the
same solution preparation maximizes repeatability of results across
measurements. This protocol can be used for analysis of purified
RuBisCO [e.g., 9, 10], or RuBisCO extracted from leaves as
described herein. It is applicable to a diverse number of species.

2.1 Equipment 1. Frozen leaf tissue, e.g., 2.5 cm2 samples (see Notes 2 and 3;
and Materials Fig. 2), stored at
80  C. Transferred to a container with
liquid nitrogen prior to use.
2. Ice-cold mortar and pestle or tissue homogenizer.
3. Refrigerated microcentrifuge set to 4  C with fixed-angle rotor.
4. Microplate reader capable of measuring absorbance in the
ultraviolet-visible (UV-Vis) region in kinetics mode to monitor
the change in absorbance at 340 nm.
5. 96-Well polystyrene microplate with clear flat bottom.

2.2 Solutions 1. Leaf extraction buffer: 50 mM Bicine pH 8.2 (adjusted with

to Prepare Beforehand NaOH), 20 mM MgCl2∙6H2O, 1 mM EDTA, 2 mM benza-
(See Notes 4 and 5) midine, 5 mM aminocaproic acid, 50 mM 2-mercaptoethanol.
Prepared CO2-free by purging with N2 prior to adjusting the
pH. This basic leaf extraction buffer mix can be stored at

20  C. Prior to starting the extraction, add 10 mM DL-
dithiothreitol (DTT), 1 mM phenylmethylsulfonyl fluoride
(PMSF), and 1% plant protease inhibitor cocktail (see Note 6).
2. CO2-free H2O: Purge ultrapure H2O with N2 (ca. 5 min per
100 mL) to release any CO2 in equilibrium with pure water.
3. 1 M Bicine pH 8.2 (adjusted with NaOH): Filter and store at
4  C.
4. 1 M MgCl2.6H2O: Filter and store at 4  C.
242 Cristina R. G. Sales et al.

Fig. 2 Leaf sampling and extraction prior to the photometric assay for determining RuBisCO activity. (a) Cutting
ca. 2.5 cm2 leaf sample with a razor blade. (b) Snap freezing leaf material in liquid nitrogen. The sample is
rapidly dropped from the razors into the liquid nitrogen, and then transferred into a frozen tube and stored at

80  C until extraction. (c) Grinding of the leaf sample. Mortar and microtubes are kept on ice during the
extraction and homogenization is completed within 45 s to avoid proteolysis

5. 0.5 M NaHCO3: Filter and store at 4  C.

6. 3 M KCl: Filter and store at 4  C.
7. 1 M DTT: Store in 0.5 mL aliquots at
20  C.
8. 1.5 KU mL
1 2,3-diPGA-dependent phosphoglycerate mutase
(d-PGM): Recombinant d-PGM containing a C-terminal
StrepTactin (S-Tag) is conveniently expressed in E. coli [7],
purified by affinity chromatography [6], and stored at
80  C.
9. Approx. 1 KU mL
1 pyruvate kinase/lactate dehydrogenase
(PK-LDH) enzyme from rabbit muscles (buffered aqueous
glycerol solution); Store at
20  C.
10. 35 mM RuBP: High-purity (99%) RuBP is available commer-
cially or can be produced enzymatically from AMP-50
 Spectrophotometric Determination of Rubisco Activity 243

monohydrate and ATP disodium salt [8] (see Note 1). Store at

80  C.
The following solutions can be stored in small aliquots
(~50 μL) at
80  C (see Notes 4 and 7):
11. 0.1 M 2,3-Disphospho-D-glyceric acid pentasodium salt
(2,3-diPGA).
12. 0.5 M β-Nicotinamide adenine dinucleotide reduced diso-
dium salt hydrate (NADH).
13. 0.5 M Adenosine 50 -diphosphate sodium salt (ADP; see Note
8).
14. 5 KU mL
1 enolase (from S. cerevisiae).

2.3 Assay Mix The basic assay mix (Table 1) should be prepared prior to starting
the assays and kept in a tube wrapped in aluminum foil on ice
during the measurements (see Note 9).

Table 1
Stock and final concentrations of each component in the assay mix for measuring RuBisCO activity in
microplates

Final concentration Volume per 200 μL

Component Stock concentration in the assay assay (μL)
Bicine-NaOH pH 8.2 1M 100 mM 20
MgCl2 1M 20 mM 4
NaHCO3 0.5 M 10 mM 4
KCl 3M 20 mM 1.3
DTT 1M 5 mM 1
2,3-diPGA 0.1 M 0.2 mM 0.4

1
1
Enolase 5 KU mL 5 U mL 0.2
d-PGM 1.5 KU mL
1 3.75 U mL
1 0.5
NADH 0.5 M 1 mM 0.4

1
1
PK-LDH ~1 KU mL ~12.5 U mL 2.5
ADP 0.5 M 2 mM 0.8
Total volume 35.1
Volumes per 200 μL assay
244 Cristina R. G. Sales et al.

3 Methods

3.1 RuBisCO Protein leaf extracts are prepared as described by Carmo-Silva et al.
Extraction from Leaf [11]:
Samples
1. Add ice-cold extraction buffer to an ice-cold mortar, and then
add frozen leaf sample.
2. Grind for 45 s with an ice-cold pestle (Fig. 2).
3. Collect the homogenate into an ice-cold tube and centrifuge at
4  C for 1 min at 14,000  g.
4. Transfer the supernatant into a second ice-cold tube and use
immediately for the RuBisCO activity assays (see Note 10).

3.2 RuBisCO Activity 1. Adjust all microplate reader settings prior to commencing
Assay extractions. Conducting measurements at 30  C provides fast
rates and reliable slopes, but the temperature can be adjusted
according to the experimental aims and plant species used.
Select wavelength 340 nm, shaking prior to reading (we use
500 rpm for 5 s). Most modern microplate readers allow for
selection of typical microplates (size, model, and brand).
Choose the option for path length correction according to
the final volume per well or calculate the path length correction
manually (see Note 11).
It is important to ensure that air bubbles are not introduced in
the wells during the following steps, as these will interfere with the
absorbance measurements. Table 2 provides a template for prepa-
ration of the various assays in the microplate: blank, initial, and total
activity (see Note 12). We recommend the use of three biological
and three technical replicates per assay.
2. Pipet 159.9 μL CO2-free H2O for the blank and 153.9 μL for
the samples into each well, followed by 35.1 μL assay mix
(avoiding exposure to light, i.e., keep light above the micro-
plate preparation low). Gently mix components by pipetting up
and down five times while stirring. Add 6 μL RuBP to the wells
for measuring initial RuBisCO activity (Table 2).
3. Add 5 μL of sample supernatant (leaf extract; see Note 13) to
the wells for total activity first, followed by those for initial
activity, mixing well by pipetting up and down ten times
while stirring. Place microplate in the reader and start moni-
toring the change in absorbance at 340 nm immediately (initial
activity) while incubating RuBisCO with CO2 and MgCl2 in
the absence of RuBP (total activity) for 5 min at 30  C to
enable carbamylation of the enzyme. The absorbance value
should start decreasing in the wells for the initial activity assay
(containing RuBP).
 Spectrophotometric Determination of Rubisco Activity 245

Table 2
Pipetting scheme for the microplate-based RuBisCO activity assay indicating volumes of each
component (in μL)

Volumes to add (μL)

Initial Total
Solution (in pipetting order) Blank activity activity
CO2-free H2O 159.9 153.9 153.9
Assay mix (from Table 1) 35.1 35.1 35.1
20 mM RuBP (diluted from original 35 mM 0 6 0
stock for easier pipetting)
Leaf extract 5 5 5
Start measuring absorbance at 340 nm while incubating the plate at the desired temperature (e.g., 30  C).
Pause reading after 5 min to start reaction for total activity.
20 mM RuBP (diluted from original 35 mM stock for easier 0 0 6
pipetting)
Continue measuring absorbance at 340 nm until a plateau is reached.
A blank is prepared without RuBP to account for any changes in absorbance not related to RuBisCO activity. The reaction
is started by adding leaf extract to all the components (initial activity assay) or by adding RuBP after the RuBisCO
contained in the leaf extract is incubated with all remaining components to allow carbamylation (total activity assay)

4. Pause the reading in order to add 6 μL RuBP to the wells for

measurement of total RuBisCO activity 5 min after addition of
sample supernatant, ideally maintaining the activation time
constant between samples of the same experiment (see Note
12). Place the microplate in the reader and continue monitor-
ing the change in absorbance.
5. The reading can be stopped once the reaction reaches a plateau.

3.3 Calculations 1. The activity of RuBisCO is inferred from the consumption of

3.3.1 RuBP Consumption
and RuBisCO Activity
RuBP (μmol s
1) measured by absorbance change per second
at 340 nm due to NADH oxidation, using an extinction coeffi-
cient of 6220 M
1 cm
1 or 6.22 μmol
1 mL cm
1 (if the
microplate reader used does not correct automatically for the
path length in the microplate wells, this needs to be accounted
for here, i.e., multiply 6.22 by the path length in cm; see Note
11). The carboxylation of one molecule of RuBP results in two
molecules of 3-PGA, thus requiring two NADH in the final
step. The rate of RuBP consumption (μmol s
1) in the assay
volume is therefore calculated by

Slope  Assay vol

RuBP consumption ¼ ð1Þ
6:22  2  Path length
246 Cristina R. G. Sales et al.

where the Slope represents the change in absorbance per second

in the linear part of the absorbance trace change, Assay vol is the
final volume per well in mL, 6.22 is the extinction coefficient of
NADH in μmol
1 mL cm
1, and the factor 2 is used to account
for the two molecules of NADH which are oxidized per mole-
cule of RuBP. The Path length of the assay mix contained in
each well is measured in cm.
2. RuBisCO initial (Vi) and total (Vt) activity expressed on a leaf
area basis (μmol m
2 s
1) is then calculated by

RuBP consumption  Extraction vol

V i or V t ¼ ð2Þ
LA  Aliquot vol
where the Extraction vol is the volume of buffer in mL used for
leaf extraction, LA is the sample leaf area in m2, and Aliquot vol is
the volume of leaf extract supernatant used in the assay in mL.
3. RuBisCO activity can also be expressed on a RuBisCO content
basis (μmol min
1 mg
1; see Note 14):
RuBP consumption  60
V i or V t ¼ ð3Þ
Rubisco content  Aliquot vol
where 60 is to convert seconds to minutes, Rubisco content is in
mg RuBisCO mL
1 leaf extract supernatant, and Aliquot vol is
the volume of leaf extract supernatant used in the assay in mL.

3.3.2 RuBisCO Activation From the RuBisCO activity calculations above for both the initial
State (Vi) and total activity (Vt), the Rubisco activation state (%) can be
calculated:
Vi
Rubisco activation state ¼ 100  ð4Þ
Vt
An example is shown in Fig. 3 for RuBisCO initial and total
activities (on a RuBisCO content basis, μmol min
1 mg
1), and
RuBisCO activation state in wheat leaves acclimated for 60 min at
two different light levels: normal light (500 μmol m
2 s
1) and low
light (15 μmol m
2 s
1). The low-light condition illustrates the
lower RuBisCO initial activity and activation state typically found in
shaded leaves [12].

4 Notes

1. High-purity RuBP (99%), available commercially or pro-

duced enzymatically from AMP-50 monohydrate and ATP dis-
odium salt [8], is required to avoid interference in measurable
 Spectrophotometric Determination of Rubisco Activity 247

Fig. 3 (a) RuBisCO initial and total specific activity, and (b) RuBisCO activation state given in percentage of
active RuBisCO, calculated according to Eq. (3). Values are means  SE (n ¼ 3). Wheat leaves were sampled
from plants acclimated to a photosynthetic photon flux density (PPFD) of 400–500 μmol m
2 s
1 or
15–50 μmol m
2 s
1 for 1 h. The latter, low-light treatment was performed to verify the accuracy of the
photometric assay, as it is well established that RuBisCO initial activity and activation state decrease under
low-light conditions [e.g., 6]. RuBisCO content was determined as described in [17]

activity due to the presence of RuBP analogs that inhibit

carboxylation [13].
2. For glasshouse- and field-grown wheat plants, we typically
homogenize ca. 2.5 cm2 leaf samples (corresponding to
ca. 40 mg of fresh weight) in 500 μL extraction buffer (i.e., a
5:1 ratio of cm2 leaf area to mL extraction buffer) to obtain a
protein concentration in the leaf extracts suitable for our needs
(~1 mg RuBisCO mL
1). Different leaf areas and extraction
buffer volumes can be used, according to leaf material availabil-
ity, and taking into account the protein content in leaves from
various plant species or plants grown under varying environ-
mental conditions.
3. In order to accurately estimate RuBisCO activation states, it is
critical to rapidly collect leaf samples and freeze immediately in
liquid nitrogen. It is recommended to take note of the photo-
synthetic photon flux density (PPFD) at the leaf level and the
sample leaf area. We frequently use a cork borer of known
diameter or a homemade device (Fig. 2) consisting of two
sharp razor blades oriented in parallel and 2 cm apart. In the
latter case, we take note of the leaf width at each cut end with a
caliper to enable estimation of the sample leaf area.
4. If chemicals are stored as a powder at
20  C or
80  C, then
let them warm up to room temperature in an enclosed box with
desiccant prior to opening.
248 Cristina R. G. Sales et al.

5. Buffers and solutions will last longer if filtered through a

0.22 μm nitrocellulose membrane.
6. Keeping extracts at low temperature (<4  C) and adding fresh
reducing agents and protease inhibitors are critical to prevent
proteolysis of RuBisCO. DTT, PMSF, and plant protease
inhibitor cocktail must be added fresh before extraction.
7. An efficient way to prepare 2,3-dPGA, NADH, ADP, and
enolase stock solutions is to calculate the volume in which the
amount provided by the supplier should be dissolved to obtain
the desired stock concentration and add pure water directly to
the original container, and then dispense into aliquots for
freezing.
8. ADP is very acidic and it is advisable to adjust the stock solution
pH in order not to affect the pH of the assay solution. To do so,
when preparing the stock solution, add 50% of the water vol-
ume to dissolve the powder in the original bottle, adjust the pH
to 6–8 with 4 M NaOH, then add the remaining water to
complete the final volume, and dispense in aliquots.
9. The basic assay mix (Table 1) can be prepared the day before
the assays and kept in the fridge, or prepared in advance (e.g., at
the start of an experiment), snap-frozen in aliquots and kept at

80  C. Each assay mix aliquot should only be thawed once, as
repeated freeze thawing can result in degradation of the cou-
pling enzymes. Stock solutions and assay mix should thaw on
ice. The assay mix should be kept in a tube wrapped in alumi-
num foil on ice during the assay, as NADH is light sensitive.
10. Delaying the start of the assay may result in a decrease in the
measurable RuBisCO activity.
11. Modern microplate readers frequently include a path length
correction option. This feature allows for the measured absor-
bance values to be normalized to a 1 cm path length, which
would be found in a typical cuvette [14]. Measurements are
corrected using Lambert-Beer’s law and consider both the
volume in each well and the specific well dimensions for each
type of microplate. Since the absorption coefficient corre-
sponds to a path length of 1 cm, if the equipment does not
have a path length correction option, then a path length cor-
rection factor can be empirically determined. To do this, com-
pare the absorbance of a known NADH concentration in the
assay mix, using the same volume as used in the microplate
assay, with the absorbance of the same solution in a standard
cuvette (path length ¼ 1 cm). The calculated path length is
then used in the denominator of Eq. (1). It is important to use
the assay mix (and not for example water) in determining the
path length correction factor as the meniscus will affect the
path length and absorbance reading in the microplate.
 Spectrophotometric Determination of Rubisco Activity 249

12. The initial activity is measured immediately upon extraction, by

adding the leaf extract to the assay mix containing all compo-
nents, while the total activity is measured by first incubating the
leaf extract with all components except RuBP to allow carba-
mylation of all available RuBisCO catalytic sites in the presence
of saturating concentrations of CO2 and Mg2+ for 3–5 min
prior to initiating the assay by addition of RuBP [e.g., 1,
15]. The optimal time of RuBisCO activation for determina-
tion of total activity should ideally be tested for in leaves
collected from different plants grown under specific growth
conditions and subjected to the specific extraction procedure.
In our hands, both leaves of glasshouse- and field-grown wheat
and brassica plants presented highest values of measurable total
RuBisCO activity after 3–5-min activation. We typically collect
leaf samples 3–4 h after the start of the photoperiod, when
leaves are most photosynthetically active.
13. For maximum accuracy in the measurements, the final
RuBisCO concentration in the assay should be between
~12.5 and 37.5 μg mL
1. To achieve this, we use leaf extracts
(supernatants) containing approximately 0.5–1.5 mg
RuBisCO mL
1 (equivalent to 1–3 mg total soluble protein
mL
1, [16]), and add 5 μL per 200 μL assay. This range of
RuBisCO concentrations corresponds to rates of RuBP con-
sumption of 0.04–0.08 μmol s
1, which are not limited by any
of the assay components.
14. RuBisCO activity expressed on a leaf area basis (μmol m
2 s
1)
is easily comparable to rates of CO2 assimilation in leaves, as
measured in vivo by infrared gas analyses. RuBisCO-specific
activity (expressed on a RuBisCO content basis) may be useful
for example to compare the speed of RuBisCO extracted from
different species. RuBisCO content in the leaf extracts can be
determined by immunoblotting [18].

Acknowledgments

We thank Dr. Mike Salvucci (previously USDA-ARS) for useful

discussions, Prof. Rebekka Wachter and Mr. Matthew Hilton (Ari-
zona State University) for advice on implementing this method in
our laboratory, and Dr. Doug Orr for helpful comments on the
manuscript. CRGS and ECS acknowledge funding from the Inter-
national Wheat Yield Partnership (IWYP64). ECS also acknowl-
edges support from a sub-contract to the Bill & Melinda Gates
Foundation award RIPE, Realizing Increased Photosynthetic
Efficiency.
250 Cristina R. G. Sales et al.
References
1. Parry MAJ, Andralojc PJ, Scales JC et al (2013)
Rubisco activity and regulation as targets for
crop improvement. J Exp Bot 64:717–730
2. Whitney SM, Baldet P, Hudson GS, Andrews TJ
(2001) Form I Rubiscos from non-green algae
are expressed abundantly but not assembled in
tobacco chloroplasts. Plant J 26:535–547
3. Ray DK, Mueller ND, West PC, Foley JA
(2013) Yield trends are insufficient to double
global crop production by 2050. PLoS One 8:
e66428
4. Orr DJ, Alcantara A, Kapralov MV et al (2016)
Surveying Rubisco diversity and temperature
response to improve crop photosynthetic effi-
ciency. Plant Physiol 172:707–717
5. Prins A, Orr DJ, Andralojc PJ et al (2016)
Rubisco catalytic properties of wild and domes-
ticated relatives provide scope for improving
wheat photosynthesis. J Exp Bot 67:1827–1838
6. Scales JC, Parry MAJ, Salvucci ME (2014) A
non-radioactive method for measuring
Rubisco activase activity in the presence of var-
iable ATP: ADP ratios, including modifications
for measuring the activity and activation state
of Rubisco. Photosynth Res 119:355–365
7. van de Loo FJ, Salvucci ME (1996) Activation
of ribulose-1,5-bisphosphate carboxylase/oxy-
genase (Rubisco) involves Rubisco activase
Trp16. Biochemistry 35:8143–8148
8. Wong C-H (1980) Practical enzymatic synth-
eses of ribulose 1,5-bisphosphate and ribose
5-phosphate. J Am Chem Soc 102:7938–7939
9. Carmo-Silva E, Barta C, Salvucci ME (2011)
Isolation of ribulose-1,5-bisphosphate carbox-
ylase/oxygenase from leaves. In: Carpentier R
(ed) Methods in molecular biology, Photo-
synth research protocols, vol 684, 2nd edn.
Humana Press, New York, pp 339–347
10. Orr DJ, Carmo-Silva E (2018) Extraction of
RuBisCO to determine catalytic constants.
In: Covshoff S (ed) Photosynthesis: methods
and protocols, Methods in molecular biology,
vol 1770. Springer, New York
11. Carmo-Silva E, Andralojc P, Scales J et al
(2017) Phenotyping of field-grown wheat in
the UK highlights contribution of light
response of photosynthesis and flag leaf longev-
ity to grain yield. J Exp Bot 68
(13):3473–3486. https://doi.org/10.1093/
jxb/erx169
12. Carmo-Silva E, Salvucci ME (2013) The regu-
latory properties of Rubisco activase differ
among species and affect photosynthetic
induction during light transitions. Plant
Physiol 161:1645–1655
13. Kane HJ, Wilkin J-M, Portis AR, Andrews TJ
(1998) Potent inhibition of ribulose-
bisphosphate carboxylase by an oxidized impu-
rity in ribulose-1,5-bisphosphate. Plant Physiol
117:1059–1069
14. Burnett RW (1972) Accurate measurement of
molar absorptivities. J Res Natl Bur Stand Sect
A Phys Chem 76A:483–489
15. Parry MAJ, Andralojc PJ, Parmar S et al (1997)
Regulation of Rubisco by inhibitors in the
light. Plant Cell Environ 20:528–534
16. Carmo-Silva E, Keys AJ, Andralojc PJ et al
(2010) Rubisco activities, properties and regu-
lation in three different C4 grasses under
drought. J Exp Bot 61:2355–2366
17. Whitney SM, von Caemmerer S, Hudson GS
et al (1999) Directed mutation of the Rubisco
large subunit of tobacco influences photorespi-
ration and growth. Plant Physiol 121:579–588
18. Perdomo JA, Sales CRG, Carmo-Silva E
(2018) Quantification of photosynthetic
enzymes in leaf extracts by immunoblotting.
In: Covshoff S (ed) Photosynthesis: methods
and protocols, Methods in molecular biology,
vol 1770. Springer, New York
 Part III

Determining Cellular and Sub-cellular Phenotypes

 Chapter 15

Creating Leaf Cell Suspensions for Characterization

of Mesophyll and Bundle Sheath Cellular Features
Roxana Khoshravesh and Tammy L. Sage

Abstract
Imaging of mesophyll cell suspensions prepared from Arabidopsis has been pivotal for forming our current
understanding of the molecular control of chloroplast division over the past 25 years. In this chapter, we
provide a method for the preparation of leaf cell suspensions that improves upon a previous method by
optimizing cellular preservation and cell separation. This technique is accessible to all researchers and
amenable for use with all plant species. The leaf suspensions can be used for imaging chloroplast features
within a cell that are important for photosynthesis such as size, number, and distribution. However, we also
provide examples to illustrate how the cells in the suspensions can be easily stained to image other features,
for example pit fields where plasmodesmata are located and organelles such as mitochondria, to improve our
understanding of traits that are important for photosynthetic physiology.

Key words Leaf cell suspensions, Light microscopy, Chloroplasts, Cellular architecture, Plasmodes-
mata, Callose, Live-cell fluorescence probes

1 Introduction

Microscopy-based imaging of cell architecture has been an essential

tool used to unravel the relationship between plant cellular form,
physiology, and evolution since the initial observations of plant cells
by Robert Hooke published in Micrographia in 1665. Current
techniques employed for characterization of plant cell structure
are quite sophisticated in comparison to those used during Hooke’s
time with, for example, the incorporation of confocal microscopy
for live-cell imaging to detect proteins labeled with fluorescent
probes such as GFP and YFP [1, 2]. The integrated use of confocal
microscopy with one or more techniques such as light, scanning,
and transmission electron microscopy and immunohistochemistry
[3] has contributed substantially to our understanding of photo-
synthesis. As examples, these aforementioned techniques, used
either in combination or alone, have been indispensable for asses-
sing pyrenoid formation [4], thylakoid assembly [5, 6], formation

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_15, © Springer Science+Business Media, LLC, part of Springer Nature 2018

253
254 Roxana Khoshravesh and Tammy L. Sage

of dimorphic chloroplasts [7], plasmodesmata distribution between

the two photosynthetic cell types in C4 species [8], and control of
plastid volume per cell [numbers and size; 9]. Many of the studies
characterizing plastid volume per cell have utilized a very easy,
rapid, and informative method to prepare mesophyll cell suspen-
sions that allows imaging of chloroplasts in intact cells using a
compound light microscope (Fig. 1a–c; [10]). This technique has
proven to be an essential tool for identification of plastid division

Fig. 1 Images of leaf cells in leaf cell suspension. (a) Spongy mesophyll cell from Arabidopsis (DIC). (b)
Palisade parenchyma cell from Heliotropium calcicola, a C3 photosynthetic species (bright field). (c) Arabi-
dopsis mesophyll cells with 1–2 large plastids isolated from the mutant line pdv1pdv2 generated by
Miyagishima et al. [13]. DIC image by Dr. Siddhartha Dutta. (d) Bundle sheath cells of the C4 photosynthetic
species Atriplex rosea stained with aniline blue illustrating callose at pit fields on a lateral wall between two
bundle-sheath cells (arrows). Imaged with fluorescence microscope. (e) Bundle-sheath cells of Oryza sativa
stained with IKI (brown). Imaged with DIC. Bars, 10 μm. bs bundle sheath; c chloroplast; m mesophyll; s stoma
 Leaf Cell Suspensions 255

mutants, such as those with one or two chloroplasts per cell

(Fig. 1c), thereby aiding in transforming our understanding of
the molecular control of plastid division over the past 25 years
[10–14]. More recently the leaf cell suspension method has been
employed to study shifts in chloroplast numbers, size, and position-
ing in mesophyll and bundle-sheath cells during the evolution from
C3 to C4 photosynthesis in over 12 lineages [15, 16] and identifi-
cation of candidate genes involved in the evolutionary transition of
chloroplast traits between those two cell types [16]. The latter two
studies [15, 16] underscore the applicability of this technique for
use in a broad number of plant species.
Here, we describe the method for preparation of cell suspen-
sions from leaves for imaging of chloroplasts and other cellular
features that are important for photosynthetic physiology. We pro-
vide slight modifications to the original technique to enhance the
ease of cell separation and quality of cell preservation through the
use of a lower concentration of tissue fixative and a pectinase
treatment. The pectinase treatment enables leaf cell suspensions
to be made from a wide range of species that have tightly adherent
cells. Enhanced cell preservation provides a more accurate assess-
ment of planar chloroplast size and chloroplast distribution within
the cell, a trait important for understanding light perception, CO2
diffusion, and refixation of photorespired CO2 [15–17]. Chloro-
plasts in the isolated cells are easily viewed in the absence of staining
with regular bright-field microscopy (Fig. 1b) as well as by using
differential interference contrast imaging (DIC; Fig. 1a, c, e). Thus,
this technique is accessible to all researchers and amenable for use
with all plant species.
Although this quick and easy procedure can be used for rapid
screening of plastid traits in all plant species, the leaf cell suspen-
sions are also valuable for characterizing other cellular features
important for photosynthetic physiology. Plasmodesmata reside in
pit fields and C4 species have been demonstrated to have a greater
pit field area at the mesophyll-bundle sheath interface than C3
species [8]. Leaf cell suspensions can be utilized to image pit fields
using the fluorochrome aniline blue, which stains plasmodesmata
associated β-1,3 glucan (callose, Fig. 1d; [18]). This use of the leaf
cell suspensions provides either an alternative to time- and labor-
intensive techniques for immunolocalization of callose on cleared
leaf tissue [8] or an initial rapid screening for pit field/plasmodes-
mata phenotypes prior to the use of more time- and labor-intensive
techniques. As a second example, leaf suspensions can be employed
to image mitochondria and peroxisomes to test hypotheses addres-
sing the role(s) of these organelles in C4 evolution; unfixed leaf
sections can be used for live-cell labeling with fixable fluorescent
probes that stain mitochondria and peroxisomes. These labeled
cells can subsequently be prepared for leaf cell suspensions. The
combined use of leaf cell suspensions and fixable fluorescent probes
256 Roxana Khoshravesh and Tammy L. Sage

to detect mitochondria, peroxisomes, as well as other cellular fea-

tures is tractable for use with plants that are not easily transformed
with organelle-specific probes such as GFP or YFP. Finally, Pyke
and Leech [10] used potassium iodide (IKI), which stains starch, to
enhance chloroplast detection as illustrated in Fig. 1e. IKI staining
can also be used in combination with leaf cell suspensions when an
investigator wishes to follow the course of starch accumulation in
specific leaf cell types during a given time period. Once cells have
been imaged, cellular parameters can be quantified to test hypoth-
eses of interest as previously described [3, 10–18].

2 Materials

2.1 Tissue Fixation 1. EM-grade glutaraldehyde aqueous solution, 25%.

2. EM-grade paraformaldehyde aqueous solution, 20%.
3. 0.1 M Sodium cacodylate buffer, titrate to the pH 6.9 with
2 M HCl.
4. Fixative: 0.5% Glutaraldehyde in 0.1 M sodium cacodylate
buffer, or 1% paraformaldehyde in sodium cacodylate buffer
(see Note 1)
5. Double-edged razor blades (see Note 2).
6. Dissecting microscope.
7. Borosilicate glass vials, 1 dram volume.
8. High precision fine tip tweezers.
9. Transfer pipettes, 1 mL.

2.2 Preparation 1. NaOH pellets.

of Leaf Cell 2. 0.2 M Disodium EDTA, pH 9.0: Add 37.22 g disodium EDTA
Suspensions and bring to approximately 450 mL volume. Stir to produce a
milky solution. Add NaOH pellets slowly, while stirring, to
clear the solution. Bring to a total volume of 500 mL by adding
water, pH to 9.0.
3. Digestion buffer, pH 5.3: 0.15 M Sodium hydrogen phos-
phate, 0.04 M citric acid.
4. Pectinase from Aspergillus niger (polygalacturonase, 1 U/mg).
5. Water bath.
6. Sonicating bath.
7. Glass microscope slides, 25  75  1 mm.
8. Glass coverslips, No. 1., 22  50 mm.
9. Wooden applicator stick.
 Leaf Cell Suspensions 257

10. Light microscope adjusted for imaging with bright field

and/or differential interference contrast microscopy (DIC)
with digital image capture capability or confocal microscope.
11. Fluorescence microscope with appropriate filters for aniline
blue excitation (excitation filter 390 nm; dichroic mirror
420 nm; emission filter 460 nm).

2.3 Detection
of Callose, Cellular
Feature of Interest
with Fluorescent Probe
or Starch
1. 0.01% Aniline blue in 150 mM K2HPO4 (see Note 3) for
callose detection.
2. Appropriate live-cell labeling fixable fluorescent probe for
imaging cell structure of interest (see Note 4).
3. 1% IKI solution in dH2O for starch detection.

3 Methods

3.1 Tissue Fixation 1. Prepare all materials required for fixation.

2. Place a drop of fixative on the dissecting scope-stage plate and
immerse the leaf tissue into the drop (Fig. 2a). Add more
fixative if required to cover the leaf tissue.
3. Use the single-edged razor to cut the tissue into 1–2 mm2 pieces.
Use tweezers to hold the leaf sample (Fig. 2a; see Note 5).
4. Fill the vial with fixative.
5. Using tweezers or a 1 mL plastic transfer pipette with the tip
cut off, transfer the pieces of tissue into the vial. Overfill the vial
with fixative and quickly place the lid on the vial ensuring that
no air is introduced into the vial (Fig. 2b). Most tissue pieces
will immediately sink to the bottom of the vial (Fig. 2c; see
Note 6).
6. Leave fixed tissue at room temperature in the dark for 1 h.

Fig. 2 The process of tissue fixation. (a) A razor blade cutting immersed tissue into 1–2 mm wide strips to
enhance fixative penetration. Note the presence of tape on the razor to protect fingers. (b) Tissue pieces
floating on fixative in an overfilled vial. (c) Tissue pieces should immediately sink to the bottom of the vial
following rapid placement of the lid onto the overfilled vial. Samples sinking immediately after placement of lid
on the vial indicates that fixative is penetrating the tissue
258 Roxana Khoshravesh and Tammy L. Sage

3.2 Live-Cell 1. Place a drop of fluorescent probe solution on the dissecting

Labeling with Fixable scope-stage plate.
Fluorescent Probe 2. Place live tissue directly in the fluorescent probe solution and
cut the tissue into 1–2 mm2 pieces with a razor as described in
Subheading 3.1, step 3.
3. Transfer the tissue into a vial filled with the fluorescent probe
solution. Overfill the vial with the probe solution and quickly
place the lid on the vial ensuring that no air is introduced into
the vial as described above.
4. Incubate in the dark at room temperature for time recom-
mended by probe supplier.
5. Remove the fluorescent probe solution from the vial with a
transfer pipette and replace with phosphate-buffered saline
(PBS) solution (1X PBS, 3 x 10 min).
6. Remove the PBS and fix in 1% paraformaldehyde in sodium
cacodylate buffer.
7. Proceed as described in Subheading 3.3.

3.3 Preparation 1. Remove the fixative from the vial using a transfer pipette and
of Leaf Cell add 2 mL of 0.2 M disodium EDTA to the vial.
Suspensions 2. Heat the vials in a water bath for 2–3 h at 55  C on a floating
rack in a water bath. Either store at 4 C (see Note 7) and
proceed as described in Subheading 3.4 or if cells do not separate
during squashing for imaging, proceed to the next step.
3. Remove EDTA from the vials and rinse the samples with diges-
tion buffer for 15–20 min at room temperature. Add enough
buffer to cover the tissue.
4. Incubate in 2% pectinase dissolved in digestion buffer for 1 h at
45  C in a water bath.
5. Rinse in digestion buffer (2  30 min) to remove pectinase at
room temperature.
6. Store samples at 4  C until ready for viewing (see Note 8).
7. If using the samples for aniline blue detection of callose,
remove the digestion buffer, rinse in tap water (2  30 min),
and replace with 1 mL of aniline blue solution, room tempera-
ture (see Note 9).

3.4 Microscopic 1. Use a pipette to place a drop of digestion buffer onto a glass
Observation microscope slide.
2. Use tweezers to transfer a piece of digested leaf tissue from the
3.4.1 Imaging Unstained vial directly into the digestion buffer on the microscope slide.
Tissue
3. Cover tissue with a coverslip and gently squash the tissue with a
wooden applicator stick until the cells are separated as visua-
lized under the dissecting scope (see Note 10).
 Leaf Cell Suspensions 259

4. View the cell suspension with bright-field microscopy or DIC.

3.4.2 Imaging Tissue 1. Mount and squash digested leaf tissue on a microscope slide as
Stained with Fluorescent directed in Subheading 3.4.1.
Probe 2. View the cell suspension with a conventional fluorescence
microscope or confocal microscope as recommended by fluo-
rescent probe manufacturer.

3.4.3 Imaging Tissue 1. Mount digested leaf tissue on microscope slide as directed in
Stained with IKI Subheading 3.4.1.
2. Use a pipette or dropper to place a drop of IKI directly on the
tissue.
3. Allow the stain to penetrate for 5 min.
4. Cover tissue with a coverslip and gently squash as directed in
Subheading 3.4.1.
5. View the cell suspension with bright-field microscopy or DIC.

4 Notes

1. Paraformaldehyde is recommended for confocal microscopy

and if fluorescent probes are being employed to detect cellular
features of interest.
2. Break double-edged razor blade in half before removing the
packaging paper and then clean with 70% EtOH to remove oil
and packaging residue. Cover broken edge of the razor blade
with tape to protect fingers (Fig. 2a).
3. To make aniline blue solution, add aniline blue to K3PO4,
cover with foil, and place in room temperature overnight.
The initially blue solution will turn a clear light yellow and is
ready to use. Store indefinitely at 4  C.
4. Choose a fluorescent probe that is retained after fixation. Zhu
et al. [19] describes the use of fluorescent probes for sensing
and imaging organelles.
5. Very small tissue size allows the digestion solutions to penetrate
into the cell walls and facilitates the process of digestion. If
fixing grass leaves, longitudinal thin sections parallel to the
veins increase the efficiency of the middle lamella digestion.
6. This method of tissue fixation eliminates the need for vacuum
infiltration of tissues that can lead to cellular artifact. If tissue
does not sink, shake the vial. Use only tissue samples that sink
260 Roxana Khoshravesh and Tammy L. Sage

to the bottom. If no tissue sinks, open the lid, add more

fixative, dry the lid, and try again.
7. Tissue can be stored in EDTA for up to 6 months at 4  C.
8. Tissue can be preserved in the digestion buffer for 2–4 weeks
at 4  C.
9. Tissue can be preserved in the aniline blue for 4 months at 4  C.
10. If you have trouble separating the cells, then place the vial
containing the digested leaf tissue in a sonicating bath for
1–2 min to facilitate cell separation.

References

1. Hanson MR, Köhler RH (2001) GFP imaging:
methodology and application to investigate
cellular compartmentation in plants. J Exp
Bot 52:529–539
2. Sankaranarayanan S, Samuel MA (2015) Guid-
ing principles for live cell imaging of plants
using confocal microscopy. In: Yeung ECT,
Stasolla C, Sumner MJ, Huang BQ (eds)
Plant microtechniques and protocols. Springer,
New York, pp 213–224
3. Khoshravesh R, Lundsgaard-Nielsen V,
Sultmanis S, Sage TL (2017) Light micros-
copy, transmission electron microscopy and
immunohistochemistry protocols for study of
the role of photorespiration. In: Fernie A,
Bauwe H, Weber APM (eds) Photorespiration:
methods and protocols. Springer, New York,
pp 243–270
4. Mackinder LCM, Meyer MT, Mettler-Altmann
T et al (2016) A repeat protein links Rubisco to
form the eukaryotic carbon-concentrating
organelle. PNAS 113:5958–5963
5. Gao H, Sage TL, Osteryoung KW (2006)
FZL, an FZO-like protein in plants, is a deter-
minant of thylakoid and chloroplast morphol-
ogy. PNAS 103:6759–6764
6. Armbruster U, Labs M, Pribil M et al (2013)
Arabidopsis CURVATURE THYLAKOID1
proteins modify thylakoid architecture by
inducing membrane curvature. Plant Cell
25:2661–2678
7. Rosnow J, Yerramsetty P, Berry JO et al (2014)
Exploring mechanisms linked to differentiation
and function of dimorphic chloroplasts in the
single cell C4 species Bienertia sinuspersici.
BMC Plant Biol 14:34
8. Danila F, Quick WP, White RG et al (2016)
The metabolite pathway between bundle
sheath and mesophyll: quantification of
plasmodesmata in leaves of C3 and C4 mono-
cots. Plant Cell 28:1461–1471
9. Larkin RM, Stefano G, Ruckle ME et al (2016)
REDUCED CHLOROPLAST COVERAGE
genes from Arabidopsis thaliana help to estab-
lish the size of the chloroplast compartment.
PNAS 113:E1116–E1125
10. Pyke KA, Leech RM (1991) Rapid image anal-
ysis screening procedure for identifying chloro-
plast number mutants in mesophyll mells of
Arabidopsis thaliana (L.) Heynh. Plant Physiol
96:1193–1195
11. Osteryoung KW, Stokes KD, Rutherford SM
et al (1998) Chloroplast division in higher
plants requires members of two functionally
divergent gene families with homology to bac-
terial ftsZ. Plant Cell 10:1991–2004
12. Colletti KS, Tattersall EA, Pyke KA et al (2000)
A homologue of the bacterial cell division site-
determining factor MinD mediates placement
of the chloroplast division apparatus. Curr Biol
10:507–516
13. Miyagishima SY, Froehlich JE, Osteryoung
KW (2006) PDV1 and PDV2 mediate recruit-
ment of the dynamin-related protein ARC5 to
the plastid division site. Plant Cell
18:2517–2530
14. Glynn JM, Yang Y, Vitha S et al (2009)
PARC6, a novel chloroplast division factor,
influences FtsZ assembly and is required for
recruitment of PDV1 during chloroplast divi-
sion in Arabidopsis. Plant J 59:700–711
15. Stata M, Sage TL, Rennie TD, Khoshravesh R,
Sultmanis S, Khaikin Y, Ludwig M, Sage RF
(2014) Mesophyll cells of C4 plants have fewer
chloroplasts than those of closely related C3
plants. Plant Cell Environ 37:2587–2600

16. Stata M, Sage TL, Hoffman N et al (2016)
Mesophyll chloroplast investment in C3, C4
and C2 species of the genus Flaveria. Plant
Cell Physiol 57:904–918
17. Busch FA, Sage TL, Cousins AB, Sage RF
(2013) C3 plants enhance rates of photosyn-
thesis by reassimilating photorespired and
respired CO2. Plant Cell Environ 36:200–212
18. Zavaliev R, Epel BL (2015) Imaging callose at
plasmodesmata using aniline blue: quantitative
Leaf Cell Suspensions 261
confocal microscopy. In: Heinlein M
(ed) Plasmodesmata: methods and protocols.
Springer New York, New York, NY, pp
105–119
19. Zhu H, Fan J, Du J, Peng X (2016) Fluores-
cent probes for sensing and imaging within
specific cellular organelles. Acc Chem Res
49:2115–2126
 Chapter 16

Determining the Subcellular Localization of Fluorescently

Tagged Proteins Using Protoplasts Extracted from
Transiently Transformed Nicotiana benthamiana Leaves
Vivien Rolland

Abstract
In plants, stable expression is arguably the method of choice to test transgene function but it is a slow and
labor-intensive process. This bottleneck generally limits the number of transgenes that can be tested, and as
such hinders construct optimization. In the face of this challenge, transient expression in Nicotiana
benthamiana leaves has emerged as a powerful screening platform to test gene expression, as well as
subcellular distribution and function of many proteins within a week. This system relies on the infiltration
of Agrobacterium tumefaciens (Agrobacterium) carrying DNA of interest into the leaf air spaces of
N. benthamiana plants. Agrobacterium rapidly transforms the plant cells and the leaves can be analyzed
within a few days. Investigating the subcellular localization of a protein of interest often relies on its fusion
to a fluorescent tag. While the amount of accumulation of such fusion proteins can often be gauged by
observing the fluorescence of the tag at the whole-leaf level, subcellular protein distribution is best
determined in protoplasts extracted from transformed leaves. Here I present a simple and effective method
to transform N. benthamiana leaves with Agrobacterium and to prepare protoplasts from these leaves to
characterize the subcellular localization of proteins of interest.

Key words Protoplasts, Transient expression, Nicotiana benthamiana, Fluorescent proteins,

Agrobacterium tumefaciens, Protein targeting, Subcellular localization, Agroinfiltration, GFP

1 Introduction

Transgenesis, the action of transferring genetic material (e.g., a

transgene) from one species to another, is often used in genetic,
developmental, and cellular studies as well as in plant biotechnol-
ogy strategies. It can be used, for example, to disrupt the genomic
sequence of the target species to assess gene function (mutagene-
sis), or to express foreign DNA in the form of RNA or protein
(transgene expression).

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_16, © Springer Science+Business Media, LLC, part of Springer Nature 2018

263
264 Vivien Rolland

For a transgenic protein to be effective, a functional level of

expression must be attained in the right tissue at the right time.
Equally important, the resulting protein must be targeted to the
correct location (i.e., in the cytosol, in a particular organelle, or
secreted), where it can act on its own or interact with its partner/
target proteins. While published literature and bioinformatics can
guide transgene construct design, successful plant engineering
requires that each construct be individually tested and optimized.
In plants, transgenesis can be achieved in a stable or a transient
manner. In the case of stable transformation, successful recombi-
nant cells are selected (most often using antibiotics) and used to
generate plants in which all cells are transformed. This type of
transformation is typically achieved using Agrobacterium tumefa-
ciens (Agrobacterium) or by bombarding tissue with DNA-coated
particles, depending on the species and on whether the nuclear or
the chloroplastic genome is being targeted [1]. While stable trans-
formation can be achieved in an ever-increasing number of species,
it is a slow and labor-intensive approach which limits the number of
constructs that can be tested [1]. To circumvent this bottleneck, a
number of faster, transient expression systems have been developed
in which the foreign DNA is expressed by the transformed plant
tissue without being transmitted to the next generation.
Agrobacterium-mediated transient transformation of Nicoti-
ana benthamiana leaves has emerged as the method of choice to
reach high levels of in planta transgene expression within a week.
In the system presented here, DNA of interest is inserted into the
T-DNA region of a binary vector which is then introduced into the
pathogenic soil bacterium Agrobacterium. Agrobacterium cultures
are infiltrated into the air spaces of Nicotiana benthamiana leaves
(termed “agroinfiltration”), where they transform plant cells which
in turn express the transgene within a few days. Post-transcriptional
gene silencing (PTGS) then rapidly takes place, limiting and even-
tually shutting down transgene expression within 2–3 days
[2–4]. To bypass PTGS, Agrobacterium carrying DNA of interest
is co-infiltrated with Agrobacterium carrying a construct encoding
the tomato bushy stunt virus P19 protein, allowing prolonged
transgene expression [4].
This assay enables the testing of multiple constructs within a
week, making it a powerful screening platform to test protein
expression, subcellular distribution, or function. Here I describe
how to use this method to express fluorescently-tagged proteins of
interest and how to prepare protoplasts from transiently expressing
leaves in order to study the subcellular distribution of proteins of
interest.
 Subcellular Localization of Transiently-Expressed Proteins in N. benthamiana Protoplasts 265

2 Materials

2.1 Growing 1. Wild-type N. benthamiana seeds (see Note 1).

Nicotiana 2. A growth cabinet with controlled environment. The cabinet
benthamiana Plants settings should be 16-h/8-h light/dark cycle at 25
C/20
C
with ~60% humidity (see Note 2).
3. Seed-raising mix.
4. Pierced plastic punnets of ~10 cm  10 cm.
5. Plastic pots of ~12–15 cm in diameter and ~15 cm in height.
6. A plastic tray, ideally able to contain 6–8 pots.
7. A pair of fine forceps such as the Dumont #5.
8. Wash bottle filled with tap water.

2.2 Transformation 1. Aliquots of electrocompetent Agrobacterium, strain GV3101

and Culture (pMP90) (see Note 3).
of Agrobacterium 2. Purified binary vector containing a gene of interest fused in
frame to a fluorescent protein. In most cases, 1–2 μL
(~200–500 ng) of plasmid purified with a commercial kit is
sufficient (see Notes 4 and 5).
3. Purified binary vector containing the coding sequence of the
tomato bushy stunt virus P19 protein. In most cases, 1–2 μL
(~200–500 ng) of plasmid purified with a commercial kit is
sufficient (see Note 6).
4. LB: 1% (w/v) Bacto-tryptone, 1% (w/v) NaCl, 0.5% (w/v)
yeast extract. Autoclave and keep at room temperature.
5. LB-RK: 1% (w/v) Bacto-tryptone, 1% (w/v) NaCl, 0.5% (w/v)
yeast extract, 50 μg/mL rifampicin, and 30 μg/mL kanamycin.
Autoclave before adding antibiotics and store at 4
C for up to
3 weeks once antibiotics have been added (see Notes 7 and 8).
6. LB-agar-RK plates: 1% (w/v) Bacto-tryptone, 1% (w/v) NaCl,
0.5% (w/v) yeast extract, 1.5% (w/v) agar, 50 μg/mL rifampi-
cin, 30 μg/mL kanamycin. Prepare LB-agar without antibio-
tics and autoclave. LB-agar can be kept at room temperature.
Melt the mixture in a microwave and allow it to cool down up
to the point where it is still liquid but the bottle can be
handheld. Add antibiotics, mix well by swirling, and pour
into standard 90 mm petri dishes. When solidified, plates can
be stored in a sterile manner at 4
C for up to 3 weeks.
7. 30% Glycerol: 30% (v/v) Glycerol, 70% (v/v) distilled water.
Autoclave solution before use.
8. An electroporation apparatus.
9. Sterile 10 mL culture tubes.
10. Electroporation cuvettes.
266 Vivien Rolland

11. A spectrophotometer.
12. Liquid nitrogen (LN2).
13. Sterile cryo-tubes with screw-on lids.

2.3 Preparation 1. 100 mM MES (2-(N-morpholino)ethanesulfonic acid),

of Agrobacterium pH 5.6: Store at room temperature.
for Infiltration 2. 100 mM MgCl2: Store at room temperature.
3. 150 mM Acetosyringone (30 ,50 -dimethoxy-40 -hydroxya-
cetophenone): Prepare aliquots in DMSO and store at 20
C
(see Note 9).
4. Infiltration solution: 10 mM MES pH 5.6, 10 mM MgCl2,
150 μM acetosyringone. This solution must be prepared fresh
each time.

2.4 Agroinfiltration 1. Sterile 1 mL syringes.

of N. benthamiana 2. Plastic paraffin film strips.
Leaves
3. Low-lint soft wipes.

2.5 Isolation 1. Blue LED light source (~470 nm) and glasses with orange
of Protoplasts filters (see Note 10).
2. Imaging solution: 0.4 M Mannitol, 20 mM KCl, 20 mM MES
pH 5.6, 10 mM CaCl2, 0.1% (w/v) BSA. Store at 4
C for up to
1 month.
3. Digestion solution: Make fresh every time by adding 1.5%
(w/v) Cellulase R-10 (Yakult Pharmaceutical Industry) and
0.4% (w/v) Macerozyme R-10 (Yakult Pharmaceutical Indus-
try) to imaging solution (see Note 11).
4. 70% EtOH: 70% (v/v) EtOH, 30% (v/v) water.
5. Scalpel blades.
6. Standard 90 mm petri dishes.
7. Clean 5 mL syringes.

2.6 Protoplast 1. Fluorescence microscope or confocal laser scanning

Imaging microscope.
2. 1 mL Syringes filled with petroleum jelly (see Note 12).
3. Standard microscopy glass slides.
4. Glass coverslips (see Note 13).
 Subcellular Localization of Transiently-Expressed Proteins in N. benthamiana Protoplasts 267

3 Methods

3.1 Growing 1. Place seed-raising mix in a punnet, water abundantly, and allow
N. benthamiana Plants excess water to drain (see Note 14).
2. Add seeds on top of the wet seed-raising mix. Close the punnet
and place it in a plastic tray containing ~1 cm deep water.
3. Leave in growth cabinet for 2 weeks. During this time, keep the
soil humid by adding water to the tray as needed (see Note 15).
4. Fill pots with fresh seed-raising mix, water abundantly from the
top, and allow excess water to drain. Create a thin and shallow
trench in each pot and use thin forceps to gently transfer a
single seedling into each trench, laying its roots horizontally
(see Note 16).
5. Use a wash bottle filled with water to gently push the surround-
ing seed-raising mix over the roots. Once the roots are covered,
continue watering abundantly with the wash bottle to eliminate
air pockets in the soil. Let excess water drain and place the pots
in a plastic tray containing ~2 cm deep water (see Note 17).
6. Place the tray with the pots in the growth cabinet for ~2 weeks.
During this time, make sure that the plants never run out of
water to prevent stress. Always water plants from the bottom
(see Note 15).
7. When grown in these conditions, plants aged ~3.5–4.5 weeks
can be used for infiltration (Fig. 1a, c).

3.2 Transformation 1. On ice, defrost an aliquot of electrocompetent Agrobacterium

and Culture cells.
of Agrobacterium 2. Add 200–500 ng of DNA to the cells and gently mix by gently
flicking the tube and keep on ice until ready to proceed to the
next step (see Note 18).
3. Use a pipette to place the mixture of cells and DNA at the
bottom of a clean electroporation cuvette (see Note 19).
4. Fit the cuvette in the shocking chamber of the electroporation
machine and ensure that it is in firm contact with the
electrodes.
5. Adjust the current delivered by the machine according to the
manufacturer’s instructions (for example: here a Bio-Rad E. coli
Pulser™ was used and adjusted to 1.8 kV). Charge the capaci-
tor and deliver a current pulse (see Note 20).
6. Remove cuvette from the shocking chamber and immediately
add 800 μL LB (see Note 21).
7. Gently mix the cells and the medium using a 1 mL pipette and a
sterile pipette tip. Transfer the mixture to a 2 mL microfuge
268 Vivien Rolland

tube. At this point the cells can wait at room temperature until
all transformations have been done (see Note 22).
8. When all transformations have been done, place all tubes in a
28
C incubator shaking at ~190 rpm for 1.5–2 h.
9. Remove all tubes from the shaking incubator and plate each
culture on separate LB-agar-RK plates in a sterile manner (see
Note 23).
10. Seal all plates with paraffin film strips and place upside-down in
a 28
C incubator for 2–3 days (see Notes 24 and 25).
11. In a sterile manner, for each transformed construct use a
200 μL pipette tip to pick a single Agrobacterium clone and
drop the tip into a 10 mL culture tube containing 2 mL
LB-RK. Place culture tubes in a 28
C incubator, shaking at
~190 rpm for ~24 h (see Note 26).
12. After this incubation, place 500 μL of each culture in a sterile
cryo-tube and add 500 μL 30% glycerol. Mix thoroughly by
inverting the tube a few times. Snap-freeze in LN2 and store at
80
C (see Note 27).

3.3 Preparation 1. Thaw relevant Agrobacterium glycerol stocks on ice. The min-
of Agrobacterium imum number of stocks needed is two: one containing P19 and
for Infiltration the other containing your gene of interest (GOI) fused to a
fluorescent protein (see Note 28).
2. For each stock, place 10 mL of LB-RK in a sterile 50 mL
conical tube and add 10 μL of thawed glycerol stock (see
Note 29).
3. Place all cultures in a 28
C incubator, shaking at
~190–220 rpm, for ~24 h (see Note 30).
4. For each culture, prepare a 1:1 culture dilution by mixing
0.5 mL of culture and 0.5 mL of LB. Use a 1 mL pipette to
homogenize the mixture.
5. Measure the optical density (OD) at 600 nm (OD600) of each
dilution. Because these are 1:1 dilutions, multiply the reading
value by two to obtain the OD600 of each undiluted culture.
The OD600 of each undiluted culture should be between ~1
and ~3 (see Notes 31 and 32).
6. Calculate the volume of P19 (V19) and of GOI (VGOI) culture
required using the following equations:
VP19 ¼ V0.3ODP19  Vleaf  nleaf and VGOI ¼ V0.5OD-
GOI  Vleaf  nleaf (see Note 33).

7. For the P19 control infiltration, place VP19 in a 50 mL sterile

conical tube. For the GOI infiltration, mix VP19 and VGOI in a
50 mL sterile conical tube (see Note 34).
 Subcellular Localization of Transiently-Expressed Proteins in N. benthamiana Protoplasts 269

8. Centrifuge all mixtures for 8 min at 2150  g prior to discard-

ing the supernatant.
9. Resuspend each pellet in nleaf  Vleaf mL infiltration solution
(i.e., the same values as used when calculating VP19 and VGOI).
10. Leave on bench for 2 h at room temperature (see Note 35).

3.4 Agroinfiltration 1. Carefully choose which leaves to infiltrate (Fig. 1a, c)

of N. benthamiana (see Note 36).
Leaves 2. Inject the infiltration solution into the leaf air spaces using a
sterile 1 mL syringe (Fig. 1b, d, e). To do so, place the syringe
on the abaxial side of the leaf (underside) where you wish to
inject the solution. Place your opposing index finger on the
adaxial side of the leaf (upper side) and press it against the
syringe tip. While maintaining gentle pressure with your finger,
slowly press on the syringe plunger until the solution enters the

Fig. 1 N. benthamiana plants, 26 days (~4 weeks, a and b) or 32 days after sowing (~4.5 weeks, c), either
before (a and c) or after (b) agroinfiltration (infiltrated areas are marked by black dashed lines). In (a) and (c),
red stars indicate leaves that are ideal for agroinfiltration, while leaves marked by black stars are too old and
those marked by white stars are too young and difficult to infiltrate. The leaf marked with a yellow star in c can
be used if not too difficult to infiltrate. (d and e) Close-up view of the infiltration process in which a syringe is
pressed on the abaxial side of the leaf and the index of the other hand is used to create a pressure point which
allows the solution to enter the leaf air spaces via open stomata
270 Vivien Rolland

leaf, resulting in a darkening of the leaf (Fig. 1d, e). To stop

infiltrating, slowly release the pressure on the plunger, and then
slowly pull your index finger away from the leaf surface (see
Note 37).
3. Remove excess solution from the surface of the leaf by gently
patting both sides of the infiltration spot with a clean low-lint
soft wipe. Never reuse the wipes to avoid contaminations (see
Note 38).
4. Repeat as many times as needed, on the same leaf or on a
different leaf. Depending on the purpose of the experiment,
you might want to infiltrate several construct combinations on
the same leaf (Fig. 2a) or most of a leaf with a single construct
combination (Fig. 2e) (see Notes 39 and 40).
5. Return plants to growth chamber.

3.5 Isolation 1. It is recommended to check protein expression 2–5 days post-

of Protoplasts infiltration (dpi). For GFP-tagged constructs, use a blue light
source and orange glasses to determine the optimal expression
time point for protoplast extraction (Fig. 2b, d) (see Notes
41–43).
2. Once the optimal expression time point has been determined,
cut out ~6–8 cm2 of infiltrated tissue from which to extract
protoplasts. Depending on the size of each infiltration spot this
corresponds to an entire spot or to only part of it (Fig. 2a, c, e).
3. Place the leaf piece(s) on the lid of a petri dish and use a clean
razor blade to cut the tissue into small squares of ~4–-
5 mm  ~4–5 mm (see Note 44).
4. Remove the plunger of a clean 5 mL syringe and place the leaf
pieces in the syringe barrel. Place the plunger back into the
barrel and press the air out of the barrel without squashing the
leaf tissue.
5. Add 2 mL of digestion solution via the tip of the syringe using a
1 mL pipette. To facilitate entry of the solution into the
syringe, gently pull on the plunger as the solution is pipetted
into the barrel.
6. Orientate the syringe with the tip up and remove air
bubbles from the barrel by gently pushing on the plunger
(see Note 45).
7. Put a small piece of paraffin film onto the tip of the syringe and
press down on the film using your index finger to block the
opening of the syringe. Pull on the plunger until resistance is
met, to force the solution to enter the leaf air spaces, and let the
plunger return to its initial position. As soon as the solution
enters the leaf pieces, the tissue turns dark green and it is time
to stop the procedure (see Note 46).
Subcellular Localization of Transiently-Expressed Proteins in N. benthamiana Protoplasts 271

Fig. 2 Examples of N. benthamiana leaves infiltrated with Agrobacterium con-

taining P19 and P19 þ GFP (infiltration areas marked by black dashed lines) at
the time of infiltration (a, c, e) or several days post-infiltration (dpi) (2 dpi in b and
5 dpi in d). When Agrobacterium containing P19 or P19 þ GFP were infiltrated
sufficiently far apart from one another (a), GFP was detected where it was
infiltrated with P19 and not where P19 was infiltrated alone (b, which is the
same leaf as in a). If the two infiltration areas were too close together (c), then
cross-contamination was observed and GFP was detected where P19 was
infiltrated alone (white arrowhead in d; this leaf is the same one as in c).
Black arrowheads in b and d indicate the scars caused by the tip of the syringe.
When a single construct combination is infiltrated in a leaf (e), the zone high-
lighted by white rectangles provides enough leaf material for protoplast extrac-
tion. In the examples shown in this figure, GFP was fused to the end of
GmRBCScTP [5]
272 Vivien Rolland

Fig. 3 Examples of GFP-tagged proteins localizing to the endoplasmic reticulum (ER, a–d), chloroplast stroma
(e–h), chloroplast envelopes (i–l), Golgi apparatus (m–p), or mitochondria (q–t) in protoplasts isolated from
N. benthamiana leaves, as described in this protocol. For details on AtPLGG1-GFP, GmPsRBCS79-GFP,
 Subcellular Localization of Transiently-Expressed Proteins in N. benthamiana Protoplasts 273

8. Push the solution through the syringe tip into a 2 mL micro-

fuge tube, without squashing the leaf pieces. Remove the
plunger and place leaf pieces into the same tube using a spatula
(see Note 47).
9. Repeat for each sample, using fresh syringes, paraffin film, and
clean spatula, to avoid cross-contamination.
10. Incubate for 1 h at room temperature, gently inverting each
tube a few times halfway through the incubation process (see
Note 48).
11. Remove leaf fragments with a thin pair of forceps and let
protoplasts settle for a few minutes (see Note 49).
12. Once the protoplasts have sedimented at the bottom of the
tube, discard the supernatant and replace with 2 mL of imaging
solution. Gently invert the tube a few times, allow the proto-
plasts to sediment, and replace solution once more. At this
point, the protoplasts are ready for imaging.

3.6 Protoplast 1. Image protoplasts within 2–3 h of completing the

Imaging previous step.
2. Specific subcellular compartments can be stained at this point.
For examples of staining protocols see [5, 6]. For examples of
images of protoplasts expressing fluorescent proteins and
stained with different subcellular markers see Fig. 3.
3. Using a 1 mL syringe, deposit four very small drops of petro-
leum jelly onto a glass slide (see Note 50).
4. Cut the end off a 1 mL pipette tip with scissors and use the tip
to place a drop of protoplast-containing imaging solution
between the four petroleum jelly drops.
5. Place a coverslip over the solution and the jelly drops and
gently press down with a pair of fine forceps. The samples are
now ready for imaging (see Note 51).
6. The specifics of how to image protoplasts largely depend on the
number and the properties of the signals to record (i.e., fluo-
rescent proteins, stains, chloroplast autofluorescence) as well as
the equipment available. For general information about fluo-
rescence and confocal microscopy, please refer to https://www.
microscopyu.com. For specific information about how to
image samples using locally available instruments, seek advice
from your local imaging expert (see Notes 52–54).
ä

Fig. 3 (continued) AtHP59-GFP, and PCC7942 SbtA-GFP refer to [5]. For details on AtpFAγ-GFP refer to
[6]. The ER was marked using the signal peptide of AtWAK2 fused to mCherry-HDEL [11], while the Golgi
apparatus was marked by GmMAN149 fused to mCherry [12]. Mitochondria were stained with MitoTracker®
Red CMXRos (for a staining protocol refer to [5]). CHLO: chloroplast autofluorescence (chlorophyll). MITO:
Mitochondria. Scale bars: 10 μm
274 Vivien Rolland

4 Notes

1. Store the seeds at 4
C in a dry container.

2. Slightly different conditions may be used, but this protocol is
optimized for the conditions mentioned here. Regardless of
the growth condition used, it is essential that they be kept
stable. These must include light and temperature control, but
best results are achieved when humidity levels are controlled as
well. It is not recommended to use a glasshouse because light,
temperature, and humidity conditions are too variable.
3. It is possible to use other strains of Agrobacterium such as
AGL1, but this protocol is optimized for GV3101(pMP90)
and I recommend using this strain. GV3101(pMP90) is resis-
tant to gentamicin and rifampicin and sensitive to kanamycin,
ampicillin, and chloramphenicol. Agrobacterium is slow grow-
ing, so when preparing competent cells use gentamicin and
rifampicin to prevent contamination by faster growing organ-
isms. Use rifampicin combined with either kanamycin, ampicil-
lin, or chloramphenicol (depending on the plasmid used) to
select cells containing your plasmid of interest.
4. There are a large number of plasmids which are compatible
with this protocol. Important features of such plasmids are
(1) a set of left and right borders to allow insertion in the
plant nuclear genome, (2) a strong constitutive promoter
such as a single- or a double-cauliflower mosaic virus promoter
(CaMV 35S) to ensure rapid and strong expression of the
transgene, (3) a fluorescent tag to detect the protein of interest,
and (4) a gene conferring antibiotic resistance (preferably
against kanamycin or ampicillin). To illustrate this protocol,
here I have used vectors from the pMDC series, which are
readily available from the ABRC website (https://abrc.osu.
edu) [7]. pMDC83, for example, enables the fusion of GFP
at the C-terminus of the protein of interest and provides resis-
tance to kanamycin. When growing Agrobacterium trans-
formed with a gene of interest in pMDC83, use rifampicin
and kanamycin as selection markers. If using a different plas-
mid, then replace kanamycin with the appropriate antibiotic.
Many other vectors from many different sources are available to
suit different experiments.
5. Locally available imaging instruments may limit which fluores-
cent tag can be detected and/or distinguished from autofluor-
escence or from other fluorescent proteins/dyes.
Consequently, carefully selecting the right fluorescent tag is
key to a successful experimental design. As a general guideline,
(1) ensure that the fluorescent tag is distinguishable from leaf
tissue/protoplast autofluorescence (e.g., chlorophyll),
Subcellular Localization of Transiently-Expressed Proteins in N. benthamiana Protoplasts 275

potential vital stains of interest, and other fluorescent proteins

when several tagged constructs are to be co-expressed;
(2) ensure that the fluorescent protein is compatible with the
pH of the organelle where it is expected to localize; (3) select a
monomeric fluorescent protein with stable properties, such as
mCherry (Red), mTurquoise2 (Cyan), or mGFP6 (Green) as
appropriate [8–10]; and (4) check whether or not a retention
signal is present in the fluorescent protein(s). Most fluorescent
proteins, including the three mentioned above, possess a
MDEL motif two amino acids before the end of their sequence.
This motif resembles the KDEL and HDEL ER retention
signals and as such may lead to unwanted ER retention of
protein fusions. When designing and assembling a new plasmid
it is recommended to add an additional motif, such as four
glycines, just before the stop of the fluorescent protein of
interest. When using existing plasmids, it is strongly recom-
mended to check the protein sequence before cloning. In
pMDC83, for example, mGFP6 contains an additional six
histidines at its very C-terminus which makes this vector safe
to use.
6. Here, I have used pMDC32 (kanamycin resistance, [7]), which
is similar to pMDC83 but allows expression of coding
sequences such as P19 without a tag [5].
7. Rifampicin does not dissolve in water so the stock solution
must be prepared with DMSO.
8. The antibiotics must be added when the LB solution is cool
enough that the bottle can be handheld (typically <50
C).
Adding antibiotics when the solution is too hot will result in
their degradation so that they are nonfunctional.
9. Acetosyringone is a phenolic compound involved in plant-
pathogen recognition. It is used to increase the efficiency of
plant cell transformation by Agrobacterium. Do not freeze-
thaw aliquots. Instead use a new aliquot each time.
10. This light/filter combination works well to detect GFP and
chloroplast autofluorescence at the whole-leaf level (see Fig. 2
for examples). This step is most useful when establishing this
protocol and should be performed with GFP because compati-
ble light sources are fairly easy to source. If this protocol is
tested with a different fluorescent protein, the light source and
filter would have to be adapted to the tag being used. For
example, the maximum excitation and emission wavelengths
for mTurquoise2 are 434 nm and 474 nm, respectively, and the
maximum excitation and emission wavelengths for mCherry
are 587 nm and 610 nm, respectively.
276 Vivien Rolland

11. While it is possible to buy enzymes from other companies, this

protocol has been optimized for Cellulase R-10 and Macero-
zyme R-10 sourced from Yakult Pharmaceutical Industry. Pro-
ducts sourced from other companies would need to be
empirically tested to ensure suitability.
12. To prepare these, melt petroleum jelly using a water bath or a
microwave and use molten jelly to fill the syringes. The syringes
can be used as soon as the jelly has cooled down. They can be
stored at room temperature for years.
13. The dimensions of the coverslips do not matter so long as they
are smaller than the slide; however, their thickness is important
as many objectives on a confocal microscope have a thickness
correction collar which needs to be adjusted to match that of
the coverslips. The standard coverslips are 17 μm thick.
14. Punnets should not be reused. Pots and trays must be cleaned
with detergent and thoroughly rinsed with water prior to being
used or reused to prevent/limit algal growth.
15. Make sure that the plants never run out of water to prevent
stress. Do not water seedlings from the top to prevent covering
them with soil or disturbing their young roots. Instead, water
them from the tray.
16. It is critical to ensure that the seedlings, and in particular their
roots, are not damaged during the transfer. To avoid physical
damage, do not firmly hold onto seedlings. Instead, use the
forceps to lift the seedlings (a little bit like a forklift), and try to
keep a bit of soil around the roots to protect them.
17. Using a black tray, rather than a white one, can help prevent/
limit the development of algae in the water.
18. It is not advised to add a large volume of DNA-containing
solution as it can decrease transformation efficiency. For exam-
ple, it is recommended to add ~1 μL of a 200–500 ng/μL
DNA plasmid prep to 25–50 μL of cells.
19. Make sure to remove possible air bubbles as they can negatively
impact transformation efficiency.
20. Arcing, the formation of an electric arc, can occur if air bubbles
are present or if the salt concentration in the mixture is too
high. Bubbles can be eliminated by tapping the cuvette before
electroporating. High salts can result from poorly washed cells
or poorly purified plasmid DNA.
21. It is essential to add the recovery medium as soon as possible,
and failing to do so can reduce transformation efficiency.
22. Here, a sterile 200 μL pipette tip can be used as a substitute for
a 1 mL pipette tip. While a wider tip ensures a quicker transfer
of the liquid, the main advantage of using a smaller tip is that it
Subcellular Localization of Transiently-Expressed Proteins in N. benthamiana Protoplasts 277

provides better access to the bottom of the electroporation

cuvette. However, to prevent damaging the cells, cutting the
end of the 200 μL tip with a razor blade or a pair of scissors, in a
sterile fashion, is recommended.
23. To avoid overcrowding the plates, it is recommended to split
each culture onto two plates: one with 200 μL of culture and
one with the rest of the culture.
24. Placing plates upside-down prevents water condensing on the
top lid from dripping back onto the agar.
25. A. tumefaciens grows more slowly than bacteria such as E. coli.
Large colonies appearing after 24 h are contaminants.
26. Plates can be resealed and stored upside-down at 4
C for up to
2 weeks in case the liquid cultures do not work.
27. Such glycerol stocks can be stable for years, provided that they
do not go through too many freeze-thaw cycles (see Note 29).
28. Each Agrobacterium stock must be grown separately. Only mix
cultures when indicated.
29. Using 10 μL of thawed glycerol stock, rather than scraping the
surface of a frozen glycerol stock, leads to more consistent
culture growth rates. Each stock can go through ~10–15
freeze-thaw cycles without a noticeable effect on growth. Nev-
ertheless, repeated freeze-thaw cycles may decrease long-term
shelf life of glycerol stocks. To prevent losing an Agrobacter-
ium stock, it is recommended to regularly prepare fresh glyc-
erol stocks by following steps 2 and 3 of Subheading 3.3,
“Preparation of Agrobacterium for infiltration”, followed by
step 12 of Subheading 3.2, “Transformation and culture of
Agrobacterium”.
30. It is recommended to inoculate the cultures around midmorn-
ing to ensure that infiltrations can be done around midday the
next day.
31. After ~24 h, the red coloration of the LB-RK, caused by
rifampicin, will have disappeared. The spectrophotometer can
be calibrated with LB or LB-RK, as the impact of media color-
ation on OD600 measurements is negligible.
32. If the OD600 value is too low, do not use the cells as the
resulting transformation efficiency will be too low. A low
OD600 value can indicate that the glycerol stock is too old or
that the construct is toxic to the cells. If the OD600 is too high,
then set up a new culture and adjust the ratio of glycerol stock:
volume of LB-RK to avoid excess cell death of bacteria.
33. V0.3ODP19 represents the volume of undiluted P19 culture
corresponding to an OD600 of 0.3. So, V0.3ODP19 ¼ 0.3/
OD600(P19), where OD600(P19) is the OD600 of the
278 Vivien Rolland

undiluted P19 culture. V0.5ODGOI represents the volume of

undiluted GOI culture corresponding to an OD600 of 0.5.
So, V0.5ODGOI ¼ 0.5/OD600(GOI), where OD600(GOI) is
the OD600 of the undiluted GOI culture. Vleaf represents the
volume of infiltration solution available per leaf. I recommend
using Vleaf ¼ 3 mL to make sure that the final volume is not
limiting the success of the experiment. Nleaf represents the
number of leaves independently infiltrated with the same con-
struct combination. I recommend using nleaf  2. So, if each
construct needs to be infiltrated in two leaves, with a volume
of 3 mL per leaf, if OD600(P19) is 1, and if OD600(GOI) is
2, then VP19 and VGOI can be calculated as follows:
VP19 ¼ (0.3/1)  3  2 ¼ 1.8 mL and VGOI ¼ (0.5/
2)  3  2 ¼ 1.5 mL.
34. Depending on the tags used and the imaging setup available, it
is possible to combine two or three fluorescently tagged con-
structs in one infiltration spot. To ensure that enough proto-
plasts from all construct combinations can be imaged within
2–3 h on the final day, preparing a maximum of eight separate
construct combinations, including the P19 control, is
recommended.
35. This step increases the ability of Agrobacterium to transform
plant cells.
36. Leaf selection is a critical step to achieve high transformation
efficiency. Ideally, selected leaves should be able to (1) let the
infiltration solution enter the air spaces, and (2) allow for high
transgene expression. In Fig. 1a, c, the leaves marked with a red
star are ideal for infiltration as they combine both characteris-
tics. Older leaves (black star) are smaller, rather thick and hairy,
and while they can take up solution without difficulty, they are
easily damaged by the syringe. Younger leaves (white stars) are
also small, rather thin and less hairy, feel slightly elastic, and will
not take up solution easily.
37. Safety glasses and gloves must be worn for this step as the
solution can squirt back into your face or leak out of the leaf.
It is also very important to minimize the number of infiltration
spots required for a single construct as leaf damage can stress
the plant and trigger a wound response. If the plants are grown
in suitable conditions and if appropriate leaves are selected,
then it is normally possible to infiltrate most of a leaf from a
single spot. If the solution does not enter the air spaces and
squirts back, then it can indicate that (1) the leaf is too young,
(2) the plants are stressed (i.e., plants were grown in subopti-
mal conditions), or (3) the leaves have closed their stomata
(i.e., experiment done too late during the day). If the solution
squirts back after having entered the air spaces, then it can
Subcellular Localization of Transiently-Expressed Proteins in N. benthamiana Protoplasts 279

indicate that (1) the syringe is not held straight or (2) the
syringe was removed before the pressure of the plunger was
released. If the solution stops progressing into the air spaces
and starts leaking through stomata, then stop infiltrating this
area and infiltrate next to it, if needed.
38. Do not press too hard on the leaf surface as this will cause leaf
damage. Avoid touching other parts of the leaf with the used
low-lint soft wipe to avoid contamination.
39. Use a fresh pair of gloves each time you infiltrate a new con-
struct or a new leaf to avoid cross-contamination. For the same
reason, when infiltrating several construct combinations in a
single leaf, allow enough space between infiltration areas or
cross-contamination may occur (Fig. 2a–d).
40. It is possible to use a wide-tip marker to highlight the edge of
each infiltration patch. If doing so, then avoid damaging the
leaf with the marker. Note that if the patch is still wet, then it
can be hard to label leaves. If the constructs are tagged with a
detectable marker (e.g., GFP) or if the whole leaf has been
infiltrated, then the need to use a marker may be superfluous.
41. Determining the optimal expression time for a given construct
or experiment depends on the vector used, the type of protein
tested (membrane proteins are less bright than cytosolic pro-
teins for example), and the subcellular compartment it localizes
in (e.g., a stable cytosolic protein is likely to be detected earlier
than a protein localizing to mitochondria). To determine in
which subcellular compartment a protein localizes, it is recom-
mended to look after 2–3 days, rather than later, to avoid
overexpression artifacts. To achieve maximum protein expres-
sion or to observe the formation of a phenotype, it may be
useful to wait a few more days. It is generally not recommended
to exceed 5 dpi, as necrosis may start to appear beyond this
point.
42. Use untransformed leaves and P19-infiltrated patches as nega-
tive controls. Untransformed leaves provide information about
leaf autofluorescence, mainly caused by chloroplasts (red,
Fig. 2b, d). P19-infiltrated patches provide information about
infiltration-induced leaf fluorescence (Fig. 2b, d).
43. Visualizing the fluorescence of whole leaves can also be used to
determine whether or not expression levels of a given construct
are comparable across leaves and/or plants, or compare the
expression levels of two constructs co-infiltrated into the same
leaf. Nevertheless, it is important to keep in mind that an
absolute quantification is not possible because levels of trans-
formation can vary from one infiltration patch to another.
280 Vivien Rolland

44. Cutting leaf tissue into pieces of this size increases the surface
area of tissue in contact with the degrading enzymes, and hence
increases digestion efficiency. In addition, such pieces are big
enough that at the end of the digestion step, remaining pieces
of undigested tissue can be discarded easily. If leaf tissue is cut
into smaller pieces than indicated above, then it is harder to
remove undigested bits of tissue, resulting in an increased
proportion of debris in the final solution.
45. Flicking the syringe can help detach small bubbles from the
surface of the leaf pieces. It is essential to remove all air bubbles
at this stage. Failing to do so prevents the liquid from entering
the leaf air spaces.
46. Do not pull the plunger too far, or too many times, to prevent
tissue damage. If all air bubbles were removed in step 6 (Sub-
heading 3.5), then some resistance will be felt when pulling on
the plunger. Pulling the plunger once or twice until strong
resistance is met is usually sufficient for the solution to enter all
leaf pieces. As a general rule, stop as soon as the leaf pieces turn
dark green. If resistance is met but the leaf pieces do not turn
dark green, then pull on the plunger a little further. If no
resistance is felt when pulling on the plunger, then expanding
air bubbles will be observed during this process, indicating that
not all air bubbles were eliminated. If this is the case, then
unblock the syringe tip and start over from Subheading 3.5,
step 6, to remove all air bubbles.
47. Invert the tube a few times to make sure that all the leaf pieces
can move freely. If this is not the case, then discard a few leaf
pieces. Adjust digestion solution final volume to 2 mL, if
needed.
48. If all pieces are immersed in the solution, then there is no need
for the samples to be rotating or shaking during the incubation
process.
49. Rinse the pair of forceps with water and ethanol and carefully
dry with fresh low-lint soft wipes between each sample to avoid
cross-contamination.
50. Organize the drops as a square or a rectangle, depending on
the size of the coverslip, such that each corner of the coverslip is
supported by a drop of jelly.
51. Apply gentle pressure over the petroleum jelly drops rather
than to the middle of the coverslip to avoid damaging the
protoplasts or breaking the coverslip. It is recommended to
image a slide within 30 min–1 h to ensure that cells remain
fresh.
52. It is recommended to use a confocal laser scanning microscope,
primarily because it offers greater flexibility and accuracy when
separating multiple signals. Another advantage of confocal
 Subcellular Localization of Transiently-Expressed Proteins in N. benthamiana Protoplasts 281

microscopy is that it blocks the out-of-focus light, which allows

for sharper and more accurate images. However, a standard
fluorescence microscope can also be used, provided that it is
equipped with suitable filters.
53. When using a confocal laser scanning microscope, it is recom-
mended to image multiple planes through roughly half of each
selected protoplast—whichever half is closer to the coverslip:
the top half on an upright microscope or the bottom half on an
inverted microscope. These different planes can then be
(1) shown as individual planes (Fig. 4a–c), which is often the
most accurate way to show where a protein localizes within a

Fig. 4 Example of three different ways of presenting data from the same protoplast expressing the inner
envelope membrane protein AtTIC20-mTurquoise2 (green in a, c, d, f, g, and i) and exhibiting chloroplast
autofluorescence (magenta in b, c, e, f, h, and i). (a–c) show a single plane, while (d–f) and (g–i) show a
maximum projection and 3D rendering, respectively. The maximum projection and 3D rendering were
generated from 15 single planes taken at evenly spaced depths into the protoplast. All images were collected
using a confocal laser scanning microscope. For details on AtTIC20-mTurquoise2, refer to [5]. CHLO:
chloroplast autofluorescence (chlorophyll). Scale bars: 10 μm
282 Vivien Rolland

cell; (2) used to generate a maximum projection, where multi-

ple single planes are combined into a single plane (Fig. 4d–f);
and/or (3) used to reconstitute a 3D rendering of the volume
imaged (Fig. 4g–i). When using a maximum projection, keep
in mind that two structures which are present at the same x, y
position but in different planes (i.e., different z positions) will
have the same z position after projection onto a single plane
and could consequently look as though they co-localize when
they do not.
54. When co-expressing several constructs, make sure to also pre-
pare protoplasts from leaf patches infiltrated with each fluores-
cently tagged protein alone, to check for autofluorescence and
bleed-through (e.g., signal detected in the YFP channel, which
originates from excitation of GFP). When staining protoplasts
expressing fluorescently tagged proteins, make sure to also
image unstained protoplasts from the same infiltrated leaf
patch (i.e., expressing the same constructs) for the same
reasons.

Acknowledgments

I would like to thank Rosemary White, Rob Allen and Sarah

Covshoff for useful comments on the manuscript. This research
was supported by Bill and Melinda Gates Foundation grant
OPP1060461, titled “RIPE—Realizing increased photosynthetic
efficiency for sustainable increases in crop yield”, sub-objective on
“CCMs into chloroplasts” held by MR Badger and GD Price.

References

1. Leonor A, Gómez-lim M, Fernández F, Loske 5. Rolland V, Badger MR, Price GD (2016)

AM (2012) Physical methods for genetic plant
transformation. Phys Life Rev 9:308–345.
https://doi.org/10.1016/j.plrev.2012.06.
002
2. Jorgensen RA, Cluster PD, English J et al
(1996) Chalcone synthase cosuppression phe-
notypes in petunia flowers: comparison of
sense vs. antisense constructs and single-copy-
vs. complex T-DNA sequences. Plant Mol Biol
31:957–973. https://doi.org/10.1007/
BF00040715
3. Que Q, Jorgensen RA (1998) Homology-
based control of gene expression patterns in
transgenic petunia flowers. Dev Genet
109:100–109
4. Roth BM, Pruss GJ, Vance VB (2004) Plant
viral suppressors of RNA silencing. Virus Res
102:97–108. https://doi.org/10.1016/j.
virusres.2004.01.020
Redirecting the cyanobacterial bicarbonate
transporters BicA and SbtA to the chloroplast
envelope: soluble and membrane cargos need
different chloroplast targeting signals in plants.
Front Plant Sci 7:1–19. https://doi.org/10.
3389/fpls.2016.00185
6. Allen RS, Tilbrook K, Warden AC et al (2017)
Expression of 16 nitrogenase proteins within
the plant mitochondrial matrix. Front Plant Sci
8:1–14. https://doi.org/10.3389/fpls.2017.
00287
7. Curtis MD, Grossniklaus U (2003) A gateway
cloning vector set for high-throughput func-
tional analysis of genes in planta. Plant Physiol
133:462–469. https://doi.org/10.1104/pp.
103.027979
8. Goedhart J, von Stetten D, Noirclerc-Savoye
M et al (2012) Structure-guided evolution of
cyan fluorescent proteins towards a quantum
 Subcellular Localization of Transiently-Expressed Proteins in N. benthamiana Protoplasts
yield of 93. Nat Commun 3:751. https://doi.
org/10.1038/ncomms1738
9. Haseloff J (1999) GFP variants for multispec-
tral imaging of living cells. Methods Cell Biol
58:139–151
10. Shaner NC, Campbell RE, Steinbach PA et al
(2004) Improved monomeric red, orange and
yellow fluorescent proteins derived from Disco-
soma sp. red fluorescent protein. Nat Biotech-
nol 22:1567–1572. https://doi.org/10.
1038/nbt103710.1038/nbt1037
283
11. Nelson BK, Cai X, Nebenführ A (2007) A mul-
ticolored set of in vivo organelle markers for
co-localization studies in Arabidopsis and
other plants. Plant J 51:1126–1136. https://
doi.org/10.1111/j.1365-313X.2007.03212.x
12. Saint-jore-dupas C, Nebenfu A, Plasson C et al
(2006) Plant N -glycan processing enzymes
employ different targeting mechanisms for
their spatial arrangement along the secretory
pathway. Plant Cell 18:3182–3200. https://
doi.org/10.1105/tpc.105.036400
 Chapter 17

3D Clearing and Molecular Labeling in Plant Tissues

William M. Palmer, Jamie R. Flynn, Antony P. Martin, Stephanie L. Reed,
Christopher P. L. Grof, Rosemary G. White, and Robert T. Furbank

Abstract
Plant histology and imaging traditionally involve the transformation of tissues into thin sections to
minimize light scatter in opaque material, allowing optical clarity and high-resolution microscopy. Recently,
new techniques in 3D tissue clearing, including PEA-CLARITY, have been developed to minimize light
scatter within intact, whole samples. These techniques can achieve equivalent microscopic resolution to that
of thin section imaging with the added benefit of maintaining the original 3D structure and position of
biomolecules of interest. Furthermore, PEA-CLARITY is compatible with standard stains and immuno-
histochemistry, allowing molecular interrogation of intact, 3D tissues. This chapter outlines the current
methods available for 3D histology in plants and details the materials, equipment, reagents, and procedure
for the PEA-CLARITY technique.

Key words Three-dimensional (3D) imaging, Molecular labeling, Immunohistochemistry,

Photosynthesis

1 Introduction

Microscopy techniques are fundamental to the study of plant biol-

ogy. Plants, like all eukaryotic organisms, consist of a specialized
organization of differentiated cells. For example, photosynthesis in
leaves of C3 and C4 plants occurs in numerous cell types such as
palisade and spongy mesophyll and bundle-sheath cells. In the case
of C4 photosynthesis, photosynthetic mesophyll and bundle-sheath
cells are specialized to carry out the C4 CO2 concentrating mecha-
nism [1]. However, when using microscopy techniques to study
photosynthesis and in particular, the leaf organ, limitations in terms
of viewing depth are usually met at ~50 μm due to differences in
refractive index (RI), which cause light scattering. In plant tissues,
RI can range from ~1.3 in the cell cytosol to 1.4–1.5 in plasma
membranes and cellulose. In addition, most leaves have cuticle
surfaces and light-gathering pigments such as chlorophyll that
further decrease optical clarity. Overcoming these limitations has

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_17, © Springer Science+Business Media, LLC, part of Springer Nature 2018

285
286 William M. Palmer et al.

traditionally relied on physically sectioning the material to a thick-

ness of less than 50 μm, which only encompasses 2–3 cell layers.
Other techniques include removing the differing RI components,
e.g., the cytosol; however, this also removes molecules of interest
including DNA, RNA, and proteins, limiting studies to visualizing
cellular organization without the ability to perform molecular
interrogations such as immunohistochemistry and in situ
hybridization.
In recent years, optical clearing techniques have advanced to
make three-dimensional (3D) whole-mount organs such as leaves
optically transparent while maintaining structure and biological
molecules of interest [2]. Achieving optical transparency of intact,
3D biological tissue has a number of benefits. Apart from eliminat-
ing the need for labor-intensive cutting of thin “2D” sections,
optical clearing allows the complete 3D structure of biological
tissue to remain intact for analysis, providing the following advan-
tages: (1) an immediate understanding of spatial relationships
between structures or molecules of interest; (2) holistic and quan-
tifiable cell counting, cell tracing, and volumetric data; and
(3) global visualization of branching networks such as vasculature
and cell-to-cell communication. As an example application, plas-
modesmatal connections between mesophyll and bundle-sheath
cells of C4 species are typically difficult to visualize and quantify in
2D. Intact, 3D imaging of this system would make this type of
investigation feasible. In addition, synthetic biology has enabled
stable genetic transformation of multiple genes targeted to differ-
ent cell types within a leaf in a single transformation. 3D visualiza-
tion of expression patterns of multiple ectopic proteins in the same
leaf sample is essential for analysis of such transgenic plants. When
tagged with molecular markers such as GFP, the spatial interaction
and location of these ectopic proteins can be ascertained without
loss of data associated with histological sectioning.
Techniques to optically clear 3D tissue aim to homogenize the
RI throughout the sample, allowing light to pass through without
scattering, allowing undistorted, high-resolution imaging deep
within the tissue. Unfortunately, early techniques that use harsh
reagents such as ethanol, benzyl alcohol, and benzyl benzoate
(BABB) or chloral hydrate severely quench fluorescent probes
such as GFP, limiting the ability to interrogate cleared samples
with conventional molecular techniques [3, 4]. Within the last
5 years however, a range of new clearing techniques and reagents
including Scale [4, 5], ClearT [6], 3DISCO/iDISCO [7–9],
SeeDB [10, 11], CUBIC [12, 13], “active” CLARITY [14–19],
and “passive” CLARITY (PACT/PARS; [2, 15, 18, 20] were
developed to circumvent this and many other issues (see Table 1
for comparison). The focus of this chapter is the application of
“passive” plant enzyme-assisted (PEA)-CLARITY, which involves
enzymatic cell wall degradation allowing passage of antibodies into
 3D Clearing and Labeling 287

whole leaves, with useful insights from recent publications and our
own experience [2].
In brief, CLARITY involves the fixation of molecules of interest
(proteins and nucleic acids) within a solid polymer network using
formaldehyde and hydrogel monomers (acrylamide/bisacryla-
mide). This is followed by the removal of light scattering lipids
with the strong ionic detergent sodium dodecyl sulfate (SDS) and
then enzymatic degradation of the cell wall. The tissue is then
amenable to molecular interrogation using techniques such as
immunohistochemistry or chemical staining. After molecular label-
ing, the sample is submerged in a RI-matched fluid for imaging
(Fig. 1).
PEA-CLARITY has the following useful attributes: (1) com-
plete 3D structure of the plant tissue/organ analyzed, (2) can be
applied to any plant organ or tissue, (3) does not quench signal
from heterologous expression of fluorescent proteins (e.g., GFP),
(4) amenable to multiple rounds of immunolabeling and chemical
staining, (5) can be applied to historical/previously fixed samples,
and (6) once cleared, tissue can be stored for long periods (several
years).
“Passive” CLARITY takes longer than “active” CLARITY;
however, it does not require a complicated electrophoresis/filtra-
tion setup and strikes a balance between effective clearing and
simplicity. Thus, it provides a good introduction to 3D tissue
clearing techniques. While the focus of this chapter is on the passive
PEA-CLARITY technique, it is important that researchers looking
to employ a 3D tissue clearing protocol carefully consider the range
of techniques available to properly address the biological question
of interest (Table 1). For example, if you want to clear multiple
tissue types, maintain signal of heterologous fluorescence proteins,
attempt multiple rounds of immunolabeling/chemical staining,
and maintain good tissue integrity, then PEA-CLARITY is a good
option. However, for examination of simple heterologous fluores-
cence in a tissue sample, protocols such as ClearSee [23] and
TOMEI [24] are much faster (hours/days compared to days/
weeks). For further reading, recently published reviews by Richard-
son and Lichtman [26] and Timmers [27] provide an excellent
description of current 3D clearing techniques for mammalian and
plant tissues, respectively.

2 Materials

2.1 Hydrogel 1. Ensure that all reagents and mixing vessels/containers are
Solution chilled prior to use and kept on ice to stop premature polymer-
ization of the hydrogel during preparation.
Table 1
Comparison of 3D tissue clearing by mammalian and plant techniques

Technique Preservation of
(chemical signal from
variation) heterologous Advantages and/or
[original Validated fluorescence Staining techniques Effect on tissue disadvantages of the
reference(s)] Mechanism of action species and tissues expression tested integrity Protocol complexity Time investment technique

3DISCO Dehydration with Mouse and rat No IHC confirmed. Tissue shrinkage Incubation in Several days for Rapidly quenches
(EtOH/ EtOH followed CNS and Must be multiple whole-mouse signal from
BABB) [21] by RI embryos. performed solutions over brain or heterologous
homogenization Drosophila before clearing variable periods embryos. 1 day expression of
with BABB melanogaster of time for small tissues fluorescence
embryos genes. Not
effective in
heavily
myelinated
tissue. Only
minimal
validation for
IHC

3DISCO Dehydration with Mouse and rat Yes. However IHC confirmed. Tissue shrinkage. Incubation in Overnight for Less background
(THF/DBE) THF followed CNS, embryos, signal will Must be Tissues become multiple whole-mouse fluorescence and
[3] by RI kidney, heart, quench within performed solid solutions over brain or faster/better
homogenization muscle, 1–2 days before clearing variable periods embryos. A few clearing than
with DBE vasculature of time hours for small EtOH/BABB.
tissues Extensive
validation for
IHC

ScaleA2/U2 RI Mouse brain and Yes. However IHC confirmed in Significant swelling Incubation in a Weeks to months Not effective in
[4] homogenization embryos signal undergoes 30 μm sections and fragility. single solution heavily
via aqueous urea some quenching Clearing is myelinated tissue
and glycerol reversible
mixture. Results
in lipid removal
 ScaleS [5] RI Mouse and human Yes IHC, lipophilic Excellent. No Incubation in Several days Major
Plant variant: homogenization brain. dyes and significant multiple (or several hours improvements
Warner et al. via aqueous Nicotiana chemical stains change in size. solutions over using elevated over ScaleA2/
[22] urea, sorbitol benthamiana, confirmed. Compatible with variable periods temperature) U2. Heavily
and glycerol Medicago IHC confirmed with EM. of time myelinated
mixture truncatula, Zea use of ETC Loss of structural regions still cause
mays, Pisum integrity after problems
sativum leaves ETC
and roots

ClearT/T2 [6] RI Mouse and rat Yes (ClearT2 only) IHC confirmed No significant Incubation in Overnight for Only brain sections
Plant variant: homogenization brain, embryos, with ClearT2 to a change in size multiple whole-mouse up to 1 mm
ClearSee [23] using formamide intestine, depth of for ClearT. solutions over brain or thickness tested.
(ClearT) or muscle. 120 μm. Slight swelling in variable periods embryos. Very high
formamide/ Arabidopsis IHC possible with ET ClearT2. of time Minutes/hours primary antibody
PEG mixture thaliana leaves Clearing is for small tissue concentration
(ClearT2) and roots reversible. sections required for
Loss of structural IHC.
integrity after Compatible with
ET lipophilic dyes

SeeDB [10] RI Mouse and rat Yes IHC confirmed No significant Incubation in Several days Compatible with
homogenization brain and change in size. multiple lipophilic dyes.
with aqueous embryos Clearing is solutions over Optical
fructose solution reversible variable periods transparency is
of time limited
compared to
other techniques

CUBIC [12] RI Mouse brain, lung, Yes IHC confirmed, No significant Incubation in Approx. 2 weeks Whole-body
homogenization kidney, liver, but must be change in size multiple for whole- clearing is
via urea and spleen, heart, performed solutions over mouse brain possible via
aminoalcohol- pancreas, during clearing variable periods transcardial
based chemical muscle, skin, process of time perfusion with
cocktails. Results intestine CUBIC
in lipid removal. reagents. Elution
Unbinds heme, of the
increasing endogenous
transparency in chromophore
tissue heme results in

(continued)
Table 1
(continued)

Technique Preservation of
(chemical signal from
variation) heterologous Advantages and/or
[original Validated fluorescence Staining techniques Effect on tissue disadvantages of the
reference(s)] Mechanism of action species and tissues expression tested integrity Protocol complexity Time investment technique

excellent
transparency

CLARITY Formation of Mouse brain, spinal Yes IHC and in situ Some tissue Polymerization Approx. 2 weeks Requires
(active) [14] tissue-hydrogel cord, spleen, hybridization swelling after followed by for whole- complicated and
hybrid followed pancreas, confirmed. clearing. complex mouse brain or expensive ETC
by lipid removal intestine, kidney, Performed after Reduction in electrophoresis 3 days with setup. Variable
via lung, testis, clearing process size when step and stochastic results with
electrophoresis muscle incubated in mounting in electrotransport. standard ETC.
and SDS. RI FocusClear imaging Several days for Marked
matching in medium small tissues improvements
imaging described with
medium stochastic
electrotransport.
Variability
reported with
IHC

CLARITY Formation of Mouse and rat Yes IHC and in situ Some tissue Polymerization Several weeks for No ETC chamber
(passive) aka tissue-hydrogel brain, spinal hybridization swelling after followed by whole-mouse required. Simple
PACT/PARS hybrid followed cord, intestine, confirmed. clearing. clearing in a brain. Several “low-
[15, 20] by lipid removal kidney, lung, Performed after Reduction in single solution days for small maintenance”
Plant variant: with SDS. RI liver, pancreas, clearing process size when and mounting in tissues. Weeks procedure
PEA-CLARITY matching in human brain, incubated in imaging for plant tissue
[2] imaging zebrafish, whole. FocusClear medium
medium N. tabacum
expanded leaf
disks.
A. thaliana
expanded leaves.
Setaria viridis
stems
 TOMEI-I [24] RI A. thaliana Yes Not compatible Maintains cell wall Incubation in a 4 days to 4 weeks, Good protocol for
homogenization seedlings with IHC structure single solution tissue dependant samples with
with TDE heterologous
solution fluorescence only

mPS-PI [25] RI A. thaliana No Not compatible Removal of Incubation in Several days Maintains cell wall
homogenization seedlings with IHC chlorophyll, multiple structure, but
with chloral heterologous solutions over removes a
hydrate fluorescence and variable periods significant
cell contents. of time amount of cell
Maintains cell contents and
wall structure pigments
BABB Benzyl alcohol benzyl benzoate, CNS central nervous system, DBE dibenzyl ether, EM electron microscopy, ET enzyme treatment, ETC electrotransport chamber, EtOH ethanol, RI
refractive index, IHC immunohistochemistry, PEG polyethylene glycol, SDS sodium dodecyl sulfate, TDE 2,20 -thiodiethanol, THF tetrahydrofuran.
292 William M. Palmer et al.

2. Prepare 200 mL of hydrogel solution: 20 mL of 40% (wt/vol)

acrylamide (4% (wt/vol) final concentration), 2.5 mL of 2%
(wt/vol) bisacrylamide (0.025% (wt/vol) final concentration),
20 mL of 10 phosphate-buffered saline (PBS), 157.5 mL of
distilled water, and 0.5 g of VA-044 thermal initiator (0.25%
(wt/vol) final concentration). Use serological pipettes to trans-
fer solutions efficiently (see Notes 1 and 2).
3. Divide hydrogel solution into 10–15 mL aliquots in 15 mL
conical tubes and store them at 4  C. For long-term storage,
hydrogel solution can be stored at 20  C.

2.2 Clearing Solution 1. Use a 6 L beaker and magnetic stirrer to prepare 5 L of clearing
solution.
2. Clearing solution: 400 g of SDS (8% wt/vol) in 3 L 1 PBS.
Make up to 5 L with dH20 and pH to 8.5 with NaOH solution
(1–5 M) (see Notes 3 and 4)
3. Use a large funnel to pour clearing solution into 1 or 2 L
reagent bottles.

2.3 Enzyme 1. Make complete enzyme treatment solution fresh each time.
Treatment Solution 2. Enzyme treatment solution: In a 15 mL tube, prepare 1.1 mg
calcium chloride in 9.94 mL of 1 PBS [1 mM final concen-
tration], with 0.005% sodium azide (NaN3). If necessary adjust
pH to 7.4 with sodium hydroxide (NaOH) or hydrochloric
acid (HCl) (see Notes 5 and 6)
3. Add 10 μL α-amylase [10 U], 10 μL α-L-arabinofuranosidase
[5 U], 10 μL β-mannanase [50 U], 10 μL cellulase [14 U],
10 μL pectate lyase [6.6 U], 10 μL xyloglucanase [10 U] to
enzyme treatment solution.
4. Mix complete enzyme solution by inverting tube.

3 Methods

3.1 Hydrogel Unique to the CLARITY procedure is the formation of a hydrogel

Embedding polymer network throughout the tissue, providing a semirigid
framework to support structures and biomolecules.
Formaldehyde-fixed tissue is infused with cold acrylamide/bisacry-
lamide monomers in combination with a thermally triggered poly-
merization initiator. Formaldehyde binds to both amine-
containing molecules and acrylamide/bisacrylamide polymers,
linking the hydrogel network with fixed biomolecules of interest.
This includes proteins and nucleic acids, but not cellular membrane
phospholipids or cell wall polymers. The following steps describe
how to embed the tissue with hydrogel solution. Please note that
 3D Clearing and Labeling 293

tissue must be fixed with 4% paraformaldehyde prior to hydrogel

embedding (see Notes 7 and 8).
1. Fix tissue in 4% paraformaldehyde and 1 PBS, pH 7.4, over-
night at 4  C.
2. Place all solutions, containers, and fixed tissue(s) on ice. This is
to ensure that the hydrogel does not polymerize due to high
temperatures.
3. If hydrogel solution is frozen, then thaw the frozen vial on ice
or in a refrigerator. Gently mix the thawed monomer solution
by inverting (see Note 9).
4. Submerge fixed tissue(s) in the hydrogel solution. Ensure that
there is enough volume of hydrogel solution so that it does not
become appreciably diluted by cellular contents, i.e., PBS from
initial fixation step that is still retained within tissue sample. For
small tissues (e.g., a 4 mm diameter leaf disk), a 1.5 mL micro-
fuge tube is suitable. Scale tubes accordingly with tissue size.
5. Incubate the sample in hydrogel solution at 4  C for a mini-
mum of 1 day (large tissues will require several days). Tissue
that has sunk and is uniformly embedded indicates complete
infiltration. Protect the sample from light when fluorophores
are present.
6. Intermittent use of approx. 100 kPa vacuum for 5 min at the
beginning of hydrogel embedding can help facilitate hydrogel
infiltration. Be sure to keep on ice during vacuum treatment.

3.2 Hydrogel Once the tissue is thoroughly infused with hydrogel solution,
Polymerization polymerization of acrylamide/bisacrylamide monomers (which
are now covalently bound to formaldehyde-fixed proteins and
nucleic acids) creates a cross-linked network within the tissue.
This polymer network creates a gel within the tissue that maintains
the 3D structure and position of native biomolecules within the
cells throughout the subsequent clearing procedures (Fig. 1).
1. After incubation in hydrogel solution at 4  C (1–2 day), place
sample into a small container (e.g., a microfuge tube) and
completely fill it with hydrogel solution. There must be no air
bubbles present in the container because oxygen will inhibit
hydrogel polymerization. Seal closed lid with paraffin film to
ensure that no air can enter (see Notes 7, 8, 10, and 11).
2. Transfer the closed, airtight container to a water bath or incu-
bator shaker set to 37  C for 3 h. This will allow the hydrogel to
set. Make sure that the hydrogel has fully solidified before
proceeding to the next step. Note that lower acrylamide or
bisacrylamide concentrations (Table 2) may not solidify fully
upon polymerization. In these cases, the alternative
294 William M. Palmer et al.

Fig. 1 PEA-CLARITY protocol overview. (a) The tissue sample (e.g., an intact Arabidopsis leaf) is perfused with
a hydrogel solution that contains a cocktail of acrylamide, bisacrylamide, and a thermal polymerization
initiator. Formaldehyde links amine groups of biomolecules to the acrylamide/bisacrylamide monomers.
Hydrogel polymerization is initiated by incubating the tissue at 37  C for 3 h, resulting in a network of fibers
that maintain the position of biomolecules in 3D space. (b) Lipid membranes are removed by passive clearing
in sodium dodecyl sulfate (SDS) solution at 37  C. (c) Specific cell wall-degrading enzymes remove
polysaccharide polymers to allow passage of antibodies throughout the tissue. (d) The resulting intact
tissue-hydrogel hybrid can undergo multiple rounds of molecular and structural interrogation using stains or
immunohistochemistry. (e) Sample is incubated in refractive index (RI)-matched solution before being imaged
using confocal or light-sheet microscopy
Table 2
Modifications to hydrogel solution suggested for protocol optimization on tissues/organs of interest

Standard þ Standard Standard

saponin variation 1 A4P0 variation 2 Hydrogel “A” Hydrogel “B”
Standard [14] [14] [15] [20, 28] [17] [18] [18]

Acrylamide 4 4 0.5–4 4 3–4 4 4

(% wt/vol)

Bisacrylamide 0.05 0.05 0.0125–0.05  0.025–0.05  

(% wt/vol)

Paraformaldehyde 4 4 4  3–4 4 2
(% wt/vol)

VA-044 (% 0.25 0.25 0.25 0.25 0.25 0.25 0.25

wt/vol)

Saponin  0.05     

Acrolein (%       2
vol/vol)

Notes: Original hydrogel Saponin enhances Lower Significantly increases Lower monomer Same as standard Includes acrolein to
solution. We penetration of concentrations speed of clearing. and solution but enhance tissue
suggest this as hydrogel into of monomers Leads to greater tissue formaldehyde without rigidity and signal-
a good starting the sample. May enhance speed of swelling than concentrations bisacrylamide to-noise ratio for
point for be useful for lipid clearing “standard” hydrogel. enhance speed immunolabeling.
optimization passive infusion Requires tissue to be of lipid clearing Acrolein imparts
of historical/ fixed with and tissue some tissue
prefixed samples paraformaldehyde or transparency. discoloration but
formalin prior to Lower does not impede
hydrogel infusion due concentrations microscopy
to exclusion of also cause some
paraformaldehyde in tissue swelling
hydrogel solution
All hydrogel variations are made up with 1 PBS
3D Clearing and Labeling
295
296 William M. Palmer et al.

“degassing” step (see Note 10) may be crucial to ensure that

absolutely no oxygen is present.
3. Completely remove the gel from the tube by placing lint-free
paper into the end of the tube and slowly pulling the gel out
while attached to the lint-free paper.
4. Carefully remove excess hydrogel from the surface of the tissue
with a lint-free paper by attaching the gel plug to another piece
of clean lint-free paper and pulling apart. Repeat this until just
the leaf disk remains without excess gel surrounding it.
5. Thoroughly wash the tissue with at least two changes of 50 mL
SDS clearing solution over 2 h at room temperature to dialyze
the remaining unpolymerized hydrogel.

3.3 Passive Clearing After polymerization, lipids are removed with an ionic detergent-
(Lipid Removal) based clearing solution containing SDS (see Note 3). This provides
passage for photons that would otherwise be irregularly scattered
by high-refractive-index cell wall lipids. The hydrogel polymer
network maintains the structure and position of molecules of inter-
est such as proteins and nucleic acids (Fig. 1).
1. Place the sample in a 50 mL conical tube and fill with clearing
solution.
2. Gently shake the sample in a shaker incubator at 37  C. Replace
the solution at least every 2 days to maximize the diffusion
gradient for lipids out of the sample.
3. Continue until tissue is transparent (i.e., easy visualization of
high-contrast signals such as a black-and-white grid or printed
text through the tissue; Fig. 2) and clearing is homogenous
(i.e., even distribution of transparency across the tissue). A
typical sample such as a 5 mm diameter leaf disk takes 2–3
weeks. Each tissue type is different and will require consider-
ation of clearing time and possible variations of the hydrogel
and clearing solution (see Notes 11–16).
4. After clearing, wash the sample for at least 48 h in 1 PBS (see
Note 12). Change 1 PBS three times daily. It is critical to
ensure that all the SDS is washed out of the tissue for the
following enzymatic step to be successful (see Note 17).

3.4 Enzymatic Plants differ fundamentally from mammalian tissue in cellular orga-
Treatment nization due to the surrounding cell wall. The main component of
the cell wall, cellulose, does not inhibit optical clarity, as the refrac-
tive index is ~1.43, close to that of the final refractive index match-
ing solution. The cell wall however creates a physical barrier to the
movement of labeling molecules such as IgG antibodies used in
immunohistochemistry that are ~150 kDa; cell wall porosity is
limited to ~60 kDa. To overcome this limitation a mixture of
cellulase, hemicellulase, pectinase, and amylase is used to degrade
 3D Clearing and Labeling 297

Fig. 2 PEA-CLARITY clearing of whole-mount leaf tissue. (a) Fresh leaf disk from
a fully expanded Nicotiana tabacum leaf (left) and a whole fully expanded
Arabidopsis thaliana leaf (right). (b) Fixed, hydrogel embedded, passively
cleared leaves. (c) Cleared cell wall enzyme-treated leaves for
immunohistochemistry and imaging. Scale bar: 1 mm (reproduced from
Palmer et al. [2] under CC license)

cell wall components (and starch) to allow permeability of relatively

large IgG antibodies into the tissue (Fig. 1c; see Note 18). If you do
not wish to perform immunohistochemistry or tissue staining, then
proceed directly to Subheading 3.6. Instructions for generic cell
wall-degrading enzymes are given here; however specific tailoring
may be required for different plant organs [29].
1. Thoroughly wash all SDS from prior clearing steps
(see Note 17).
2. Place leaf disk into 5 mL tube containing 1 mL enzyme treat-
ment solution (see Subheading 2.3).
3. Keep sample in a dark shaker incubator at 37  C with gentle
agitation (30–50 rpm) for 5 days with one change of enzyme
treatment solution on the third day.
4. Intermittent use of vacuum can help facilitate enzymatic deg-
radation (see Note 19).
5. After enzyme incubation, wash the sample for at least 24 h in
PBST (1 PBS with 0.1% Triton X), changing the wash at least
three times. Proceed to labeling step within 48 h.

3.5 Molecular Standard stains (e.g., DAPI and propidium iodide) and immuno-
Labeling histochemistry can be applied to CLARITY-cleared 3D tissue
(Fig. 3). Importantly, CLARITY tissue can undergo multiple
rounds of staining and immunolabeling. If a sample contains fluor-
ophores (e.g., GFP, tdTomato), then fluorescence is generally
298 William M. Palmer et al.

Fig. 3 Representative confocal laser scanning microscopy 3D projection of a passively cleared Nicotiana
tabacum leaf showing nuclei stained with propidium iodide (purple) and cell walls stained with calcofluor-
white (yellow). Enzymatic digestion of the cell wall was not performed as only small-molecule dyes were used

conserved throughout the CLARITY procedure (Fig. 1d). Follow

this section for small-molecule staining or immunostaining-cleared
tissue. If you do not wish to perform staining or immunohisto-
chemistry, then proceed directly to Subheading 3.6.
1. For a small tissue block, using a sealable container, incubate
sample in 1–2 mL of primary antibody/PBST solution for
2 days at 37  C. For larger tissues, such as a mature Arabidopsis
leaf, incubate in 5 mL primary antibody/PBST solution for up
to 1 week, with additional antibody periodically added to
account for sequestering of antibody by the large 3D tissue
volume. As with any staining procedure, it is important to
systematically optimize staining conditions (detergent, temper-
ature, concentrations, tissue size, etc.). When testing a primary
antibody for the first time, a 1:50 dilution is a good starting
point for a 5 mm leaf disk, replacing the antibody solution
every 2 days over 1–2 weeks (see Notes 20–29).
2. Wash off the primary antibody with PBST solution at 37  C for
1–3 days depending on the size of the tissue (refresh PBST
every 4–6 h, with overnight steps when required).
3. Incubate the sample with the desired secondary antibody
(1:50–1:500) in PBST for 2 days to 1 week at 37  C (see
Notes 20–29). Other stains, such as DAPI (1 μg/mL), can
also be added at this step.
 3D Clearing and Labeling 299

4. Wash off the secondary antibody with PBST at 37  C for 1–3

days depending on the size of the tissue. Refresh PBST every
4–6 h with overnight steps as necessary.

3.6 Refractive Index Refractive index (RI) matching is required to homogenize RI

Matching and Sample throughout the tissue with the surrounding imaging medium.
Mounting for Imaging Matching the RI limits inconsistent scattering of photons at the
tissue borders and throughout the tissue. This step is critical for
optimal imaging quality and depth. Typically, an RI matching
solution will have a similar RI to the cleared tissue (e.g., 1.45 in
the case of CLARITY-cleared mammalian tissue). 3D imaging is
typically performed with confocal or light-sheet microscopy
(Fig. 1e). FocusClear, glycerol (80–87% vol/vol), RIMS (88%
Histodenz wt/vol in 0.02 M phosphate buffer), sRIMS (70% sor-
bitol wt/vol in 0.02 M phosphate buffer), and 2,20 -thiodiethanol
(60% vol/vol) have been tested in several studies [14, 15, 17, 18,
20, 24, 30]. Of these, only 2,20 -thiodiethanol has been reported as
an RI matching solution in studies of cleared plant tissue.
1. Transfer the PBST-washed tissue into RI matching medium
(e.g., FocusClear). Periodically check the visual clarity of the
sample over the next few hours. One or two millimeter thick
tissue blocks will typically complete the RI homogenization
process in a few hours, but can be left overnight (see Note 30).
2. Place the sample in fresh/unused RI matching medium several
hours before imaging. This is to ensure that the final prepara-
tion does not contain any diluted PBST from the first incuba-
tion, ensuring that the intended RI is achieved.
3. Once the tissue is completely see-through (Fig. 2), take a clean
glass slide and place it on a dust-free surface.
4. Take a small piece of removable adhesive putty and roll a
constant-diameter cylinder shape using gloved hands. Make it
approximately the same thickness as the sample.
5. Make a circular/donut-shaped tissue well in the middle of the
slide using the rolled putty.
6. With a small spatula, apply petroleum jelly or silicone grease
around the base of the putty to ensure that it is fluid-tight and
will not leak.
7. Carefully take the sample with a spatula and place it within the
putty tissue well.
8. Use a 5 mL plastic transfer pipette to completely fill the tissue
well with RI matching medium. Fill so that the fluid surface is
convex over the putty walls.
9. Carefully place a glass coverslip over the tissue, starting at one
side of the tissue well and gently laying it across to the other
300 William M. Palmer et al.

side. Ensure that there are no bubbles between the sample and
coverslip.
10. The sample is now ready for imaging using confocal micros-
copy (see Note 31).

4 Notes

1. !CAUTION! Acrylamide, bisacrylamide, and VA-044 are

toxic. Perform all procedures in a fume hood and wear appro-
priate personal protective equipment (PPE) including a lab
coat, gloves, safety goggles, face mask, and closed-toe shoes.
2. The percent of acrylamide, percent or presence of bisacryla-
mide, and percent or presence of paraformaldehyde can be
altered to modify the pore size of the polymerized hydrogel
for faster clearing and immunohistochemistry (Table 2). Our
recipe is similar to A4P0 used by Yang et al. [20], but with
added bisacrylamide (0.025%). Note that tissue samples must
be initially fixed with 4% paraformaldehyde when the fixative is
not included in the hydrogel solution [20].
3. !CAUTION! SDS is a toxic irritant. Perform all procedures in
a fume hood and wear appropriate PPE including a lab coat,
gloves, safety goggles, face mask, and closed-toe shoes.
4. A 0.2 M borate buffer can be used instead of PBS to reduce
background staining [14, 15]. 0.1% Triton-X can be added to
improve tissue penetration of clearing solution.
5. !CAUTION! Sodium azide is toxic. Perform all procedures in
a fume hood and wear appropriate PPE including a lab coat,
gloves, safety goggles, face mask, and closed-toe shoes.
6. !CAUTION! The hydrogel solution is toxic. Perform all pro-
cedures in a fume hood and wear appropriate PPE including a
lab coat, gloves, safety goggles, face mask, and closed-toe
shoes.
7. !CAUTION! DO NOT PUT HYDROGEL SOLUTION
DOWN THE SINK. The hydrogel will polymerize at room
temperature and block drains. Collect paraformaldehyde and
hydrogel in a designated waste container for proper disposal.
8. Make sure that there is no precipitation floating in the hydrogel
solution; precipitation is an indicator of spontaneous polymer-
ization of monomers in the stored hydrogel solution. If pres-
ent, then discard and make fresh hydrogel solution.
9. A 4–5-mm-thick layer of mineral oil can be gently laid over the
top of the hydrogel solution before/during the 4  C incuba-
tion in step 1 (Subheading 3.2). This will prevent oxygen
contact with the hydrogel solution. With this method, the
 3D Clearing and Labeling 301

sample container can simply be taken directly from the fridge

and transferred to the water bath in step 2 (Subheading 3.2).
10. A “degassing” procedure which involves the vacuum removal
of air from the sample container and purging with nitrogen gas
can be used. This adds a significant layer of complexity to the
process, but may lead to enhanced polymerization within the
tissue [14, 15, 17].
11. Do not incubate sample in clearing solution (3.3) for more
than 2 months as the tissue loses antigenicity.
12. Sample will become less “see-through” when placed in PBS
(3.3.4). This is because the refractive index of PBS is quite
different from the cleared tissue. It will only become
completely transparent when incubated in a refractive index-
matched imaging solution.
13. Always ensure that the pH of the clearing solution remains at
8.5 and that no microbial growth occurs within the stock
clearing or staining solutions. 0.01% Sodium azide can be
added to inhibit microbial growth for long incubations.
14. The clearing process can be sped up by using lower concentra-
tions of acrylamide/bisacrylamide in the hydrogel solution (see
Table 2) or increasing the SDS concentration in the clearing
solution [20].
15. Higher temperatures, up to 60  C, can be used for faster
clearing. However this may quench fluorescence signal from
heterologous fluorescence proteins (e.g., GFP). In our experi-
ence, yellowing of the sample can also occur at higher tem-
peratures. This is a result of the Maillard reaction and can be
inhibited by the addition of 0.5% α-thioglycerol to the clearing
solution [11].
16. “Stochastic electrotransport” (a modified version of the elec-
trotransport chamber (ETC) originally described for “active”
CLARITY) can be used to rapidly and consistently clear and
label thick tissues without damage. However, this requires
specialized equipment [31].
17. SDS will precipitate and form an opaque mass if it is not
thoroughly washed out before it is placed into the
RI-matched imaging medium. SDS is also a strong inhibitor
of enzymatic activity. Therefore, when enzymatic degradation
of the cell wall is necessary for immunohistochemistry, take
care to completely wash all SDS clearing solution from the
tissue.
18. Be sure to use quality-controlled, pure enzymes as some com-
monly used cell wall-degrading enzymes (such as those used in
protoplast digestion) also contain DNA, RNA, and proteases
that will degrade other molecules of interest.
302 William M. Palmer et al.

19. Enzymatic treatment can be sped up via application of an

intermittent vacuum to assist penetration of enzymes into the
tissue sample. This is applied in 5-min bursts at 100 kPa, 2–3
times daily [2].
20. High antibody concentrations (1:20–1:100) are usually
required for immunostaining 3D tissue. 1:1000 dilutions typi-
cally lead to inadequate tissue penetration.
21. A major factor for successful immunostaining is complete
removal of lipids during the clearing step.
22. Immunolabeling can be sped up via application of an intermit-
tent vacuum to assist penetration of antibodies into the tissue
sample as with enzyme treatment (see Note 19) [2].
23. “Stochastic electrotransport” can be used to rapidly and con-
sistently label thick tissues with antibodies or dyes. However,
this requires specialized equipment [31]. A more simple
method is described in [32].
24. 0.2 M Sodium borate buffer with 0.1% (vol/vol) Triton X-100
(pH 8.5) can be used in place of PBST for antibody incubations
to reduce background staining [16, 17].
25. Smaller secondary antibody formats such as immunoglobulin F
(ab) (55 kDa) and F(ab’)2 (110 kDa) fragments, or camelid
nanobodies (15 kDa), are suggested for faster diffusion
throughout the hydrogel matrix [28].
26. It is possible to elute (remove) labels by incubating in clearing
solution at 60  C overnight in preparation for multiple rounds
of labeling after imaging has been performed. Once labels have
been removed, repeat the immunolabeling procedure with the
desired antibodies.
27. Make sure that there is no microbial growth during long
incubations at 37  C. 0.01% (wt/vol) sodium azide can be
added to the primary and secondary antibody solutions to
prevent this [2].
28. To make immunolabeling more cost effective, recycle the pri-
mary antibody solutions by adding 0.01% sodium azide and
storing at 4  C after use. It is, however, preferable to use fresh
antibody solution.

29. Treatment at 37 C will accelerate the immunolabeling
process.
30. Before or after RI homogenization, the tissue may be stored
indefinitely in PBST (with 0.01% (wt/vol) sodium azide) at
4  C or room temperature. Extended incubation of tissue in
FocusClear is not advised as it may form precipitates that affect
the RI of the tissue sample.

31. Light-sheet microscopy is a faster alternative to confocal
microscopy for imaging cleared 3D tissue while decreasing
phototoxicity [26].
References
1. Furbank RT (2016) Walking the C4 pathway:
past, present, and future. J Exp Bot 67
(14):4057–4066. https://doi.org/10.1093/
jxb/erw161
2. Palmer WM, Martin AP, Flynn JR, Reed SL,
White RG, Furbank RT, Grof CPL (2015)
PEA-CLARITY: 3D molecular imaging of
whole plant organs. Sci Rep 5:13492.
https://doi.org/10.1038/srep13492
3. Becker K, Jahrling N, Saghafi S, Weiler R, Dodt
HU (2012) Chemical clearing and dehydration
of GFP expressing mouse brains. PLoS One 7
(3):e33916. https://doi.org/10.1371/jour
nal.pone.0033916
4. Hama H, Kurokawa H, Kawano H, Ando R,
Shimogori T, Noda H, Fukami K, Sakaue-
Sawano A, Miyawaki A (2011) Scale: a chemi-
cal approach for fluorescence imaging and
reconstruction of transparent mouse brain.
Nat Neurosci 14(11):1481–1488. https://
doi.org/10.1038/nn.2928
5. Hama H, Hioki H, Namiki K, Hoshida T,
Kurokawa H, Ishidate F, Kaneko T, Akagi T,
Saito T, Saido T, Miyawaki A (2015) ScaleS: an
optical clearing palette for biological imaging.
Nat Neurosci 18(10):1518–1529. https://doi.
org/10.1038/nn.4107
6. Kuwajima T, Sitko AA, Bhansali P, Jurgens C,
Guido W, Mason C (2013) ClearT: a deter-
gent- and solvent-free clearing method for
neuronal and non-neuronal tissue. Develop-
ment 140(6):1364–1368. https://doi.org/
10.1242/dev.091844
7. Erturk A, Becker K, Jahrling N, Mauch CP,
Hojer CD, Egen JG, Hellal F, Bradke F,
Sheng M, Dodt HU (2012) Three-
dimensional imaging of solvent-cleared organs
using 3DISCO. Nat Protoc 7(11):1983–1995.
https://doi.org/10.1038/nprot.2012.119
8. Erturk A, Bradke F (2013) High-resolution
imaging of entire organs by 3-dimensional
imaging of solvent cleared organs (3DISCO).
Exp Neurol 242:57–64. https://doi.org/10.
1016/j.expneurol.2012.10.018
9. Renier N, Wu Z, Simon DJ, Yang J, Ariel P,
Tessier-Lavigne M (2014) iDISCO: a simple,
rapid method to immunolabel large tissue sam-
ples for volume imaging. Cell 159
(4):896–910. https://doi.org/10.1016/j.cell.
2014.10.010
3D Clearing and Labeling 303
10. Ke MT, Fujimoto S, Imai T (2013) SeeDB: a
simple and morphology-preserving optical
clearing agent for neuronal circuit reconstruc-
tion. Nat Neurosci 16(8):1154–1161. https://
doi.org/10.1038/nn.3447
11. Ke MT, Imai T (2014) Optical clearing of fixed
brain samples using SeeDB. Curr Protoc Neu-
rosci 66:Unit 2.22. https://doi.org/10.1002/
0471142301.ns0222s66
12. Susaki EA, Tainaka K, Perrin D, Kishino F,
Tawara T, Watanabe TM, Yokoyama C,
Onoe H, Eguchi M, Yamaguchi S, Abe T,
Kiyonari H, Shimizu Y, Miyawaki A,
Yokota H, Ueda HR (2014) Whole-brain
imaging with single-cell resolution using
chemical cocktails and computational analysis.
Cell 157(3):726–739. https://doi.org/10.
1016/j.cell.2014.03.042
13. Susaki EA, Tainaka K, Perrin D, Yukinaga H,
Kuno A, Ueda HR (2015) Advanced CUBIC
protocols for whole-brain and whole-body
clearing and imaging. Nat Protoc 10
(11):1709–1727. https://doi.org/10.1038/
nprot.2015.085
14. Chung K, Wallace J, Kim SY,
Kalyanasundaram S, Andalman AS, Davidson
TJ, Mirzabekov JJ, Zalocusky KA, Mattis J,
Denisin AK, Pak S, Bernstein H,
Ramakrishnan C, Grosenick L, Gradinaru V,
Deisseroth K (2013) Structural and molecular
interrogation of intact biological systems.
Nature 497(7449):332–337. https://doi.
org/10.1038/nature12107
15. Tomer R, Ye L, Hsueh B, Deisseroth K (2014)
Advanced CLARITY for rapid and high-
resolution imaging of intact tissues. Nat Protoc
9(7):1682–1697. https://doi.org/10.1038/
nprot.2014.123
16. Lee H, Park JH, Seo I, Park SH, Kim S (2014)
Improved application of the electrophoretic
tissue clearing technology, CLARITY, to intact
solid organs including brain, pancreas, liver,
kidney, lung, and intestine. BMC Dev Biol
14:48. https://doi.org/10.1186/s12861-
014-0048-3
17. Epp JR, Niibori Y, (Liz) Hsiang H-L,
Mercaldo V, Deisseroth K, Josselyn SA, Frank-
land PW (2015) Optimization of CLARITY
for clearing whole-brain and other intact
304 William M. Palmer et al.
organs. eNeuro 2:3. https://doi.org/10.
1523/eneuro.0022-15.2015
18. Zheng H, Rinaman L (2015) Simplified
CLARITY for visualizing immunofluorescence
labeling in the developing rat brain. Brain
Struct Funct 221(4):2375–2383. https://doi.
org/10.1007/s00429-015-1020-0
19. Kim SY, Chung K, Deisseroth K (2013) Light
microscopy mapping of connections in the
intact brain. Trends Cogn Sci 17
(12):596–599. https://doi.org/10.1016/j.
tics.2013.10.005
20. Yang B, Treweek JB, Kulkarni RP, Deverman
BE, Chen CK, Lubeck E, Shah S, Cai L, Gra-
dinaru V (2014) Single-cell phenotyping
within transparent intact tissue through
whole-body clearing. Cell 158(4):945–958.
https://doi.org/10.1016/j.cell.2014.07.017
21. Dodt HU, Leischner U, Schierloh A,
Jahrling N, Mauch CP, Deininger K, Deussing
JM, Eder M, Zieglgansberger W, Becker K
(2007) Ultramicroscopy: three-dimensional
visualization of neuronal networks in the
whole mouse brain. Nat Meth 4(4):331–336.
https://doi.org/10.1038/nmeth1036
22. Warner CA, Biedrzycki M, Jacobs SS, Wisser
RJ, Caplan JL, Sherrier DJ (2014) An optical
clearing technique for plant tissues allowing
deep imaging and compatible with fluores-
cence microscopy. Plant Physiol 166
(4):1684–1687. https://doi.org/10.1104/
pp.114.244673
23. Kurihara D, Mizuta Y, Sato Y, Higashiyama T
(2015) ClearSee: a rapid optical clearing
reagent for whole-plant fluorescence imaging.
Development 142(23):4168–4179. https://
doi.org/10.1242/dev.127613
24. Hasegawa J, Sakamoto Y, Nakagami S, Aida M,
Sawa S, Matsunaga S (2016) Three-
dimensional imaging of plant organs using a
simple and rapid transparency technique.
Plant Cell Physiol 57(3):462–472. https://
doi.org/10.1093/pcp/pcw027
25. Truernit E, Bauby H, Dubreucq B,
Grandjean O, Runions J, Barthélémy J, Palau-
qui J-C (2008) High-resolution whole-mount
imaging of three-dimensional tissue organiza-
tion and gene expression enables the study of
phloem development and structure in Arabi-
dopsis. Plant Cell 20(6):1494–1503. https://
doi.org/10.1105/tpc.107.056069
26. Richardson DS, Lichtman JW (2015) Clarify-
ing tissue clearing. Cell 162(2):246–257.
https://doi.org/10.1016/j.cell.2015.06.067
27. Timmers AC (2016) Light microscopy of
whole plant organs. J Microsc 263
(2):165–170. https://doi.org/10.1111/jmi.
12394
28. Treweek JB, Chan KY, Flytzanis NC, Yang B,
Deverman BE, Greenbaum A, Lignell A,
Xiao C, Cai L, Ladinsky MS, Bjorkman PJ,
Fowlkes CC, Gradinaru V (2015) Whole-
body tissue stabilization and selective extrac-
tions via tissue-hydrogel hybrids for high-
resolution intact circuit mapping and pheno-
typing. Nat Protoc 10(11):1860–1896.
https://doi.org/10.1038/nprot.2015.122
29. Danila FR, Quick WP, White RG, Furbank RT,
von Caemmerer S (2016) The metabolite path-
way between bundle sheath and mesophyll:
quantification of plasmodesmata in leaves of
C3 and C4 monocots. Plant Cell 28
(6):1461–1471. https://doi.org/10.1105/
tpc.16.00155
30. Liu AK, Hurry ME, Ng OT, DeFelice J, Lai
HM, Pearce RK, Wong GT, Chang RC, Gen-
tleman SM (2015) Bringing CLARITY to the
human brain: visualisation of Lewy pathology
in three-dimensions. Neuropathol Appl Neu-
robiol 42(6):573–587. https://doi.org/10.
1111/nan.12293
31. Kim SY, Cho JH, Murray E, Bakh N, Choi H,
Ohn K, Ruelas L, Hubbert A, McCue M, Vas-
sallo SL, Keller PJ, Chung K (2015) Stochastic
electrotransport selectively enhances the trans-
port of highly electromobile molecules. Proc
Natl Acad Sci USA 112(46):E6274–E6283.
https://doi.org/10.1073/pnas.1510133112
32. Li J, Czajkowsky DM, Li X, Shao Z (2015) Fast
immuno-labeling by electrophoretically driven
infiltration for intact tissue imaging. Sci Rep
5:10640. https://doi.org/10.1038/srep10640
 Chapter 18

Evaluation of Lipids for the Study of Photosynthetic

Membranes
Helmut Kirchhoff and Robert Yarbrough

Abstract
The biological role of lipids goes far beyond the formation of a structural membrane bilayer platform for
membrane proteins and controlling fluxes across the membranes. For example, in photosynthetic thylakoid
membranes, lipids occupy well-defined binding niches within protein complexes and determine the struc-
tural organization of membrane proteins and their function by controlling generic physicochemical
membrane properties. In this chapter, two-dimensional thin-layer chromatography (2D TLC) and gas
chromatography (GC) techniques are presented for quantitative analysis of lipid classes and fatty acids in
thylakoid membranes. In addition, lipid extraction methods from isolated thylakoid membranes and leaves
are described together with a procedure for derivatization of fatty acids to fatty acid methyl esters (FAME)
that is required for GC analysis.

Key words Thylakoid membrane, Thin-layer chromatography, Gas chromatography, Monogalactosyl-

diacylglycerol, Digalactosyldiacylglycerol, Sulfoquinovosyldiacylglycerol, Phosphatidyldiacylglycerol

1 Introduction

In oxygen-evolving photosynthetic organisms, the energy-

converting machinery is harbored in highly specialized thylakoid
membranes inside the chloroplasts. The lipid composition of thyla-
koid membranes of different photosynthetic organisms is highly
conserved from cyanobacteria to higher plants, containing four
main lipid classes [1]. Thylakoid membrane lipids are dominated
by uncharged galactoglycerolipids that are exclusively found in pho-
tosynthetic membranes. In higher plants, about 70–80% of all lipids
in thylakoid membranes are galactolipids [2]: monogalactosyldiacyl-
glycerol (MGDG) and digalactosyldiacylglycerol (DGDG). The
remaining lipids are charged phosphoglycerolipids: sulfoquinovosyl-
diacylglycerol (SQDG) and phosphatidylglycerol (PG). The four
main thylakoid membrane lipids often have highly unsaturated fatty
acids, [1, 3] typically 18:3 or 16:3 acyl chains, and in addition
contain 16:0 and 16:1(3 t) fatty acids, leading to a great variety of

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_18, © Springer Science+Business Media, LLC, part of Springer Nature 2018

305
306 Helmut Kirchhoff and Robert Yarbrough

different lipid species. In general, two lipid localizations can be

differentiated for thylakoid membranes [1, 4]. One class of lipids is
strictly bound to the photosynthetic membrane protein complexes,
supercomplex, and part of the overall protein structure whereas the
other class of bulk lipids form the membrane lipid bilayer. Both
protein-bound and bulk lipids have specific functional and structural
roles [1, 4], i.e., thylakoid membrane lipids do not just form an inert
lipophilic matrix for embedding membrane proteins but determine
vital functions of photosynthetic energy conversion by specific inter-
actions with proteins. Due to the importance of lipids for specific
structure-function relationships in photosynthetic membranes, the
quantification of both thylakoid lipid classes and their fatty acids is of
high scientific interest.
This chapter describes two main complementary techniques for
quantification of lipid classes and fatty acids in thylakoid mem-
branes of higher plants. Both techniques, however, can be applied
to other photosynthetic organisms as well. The chapter describes
first the extraction of lipids from leaves or isolated thylakoid mem-
branes. Then the procedure for quantification of thylakoid lipid
classes (MGDG, DGDG, SQDG, PG) by two-dimensional thin-
layer chromatography (2D TLC) is presented. Separation of thyla-
koid lipids on silica gel TLC plates is a widely used technique in
lipid research [for example 5–7]. It is an easy-to-establish technique
and provides reliable data on lipid composition. TLC is based on
differential interaction of the different lipid classes that run in the
mobile phase with the charged stationary phase of the plate (typi-
cally silica). The mobile phase that contains the lipid mixture is
drawn up the plate by capillary forces. Typically, one-dimensional
TLC is used [6, 7] but 2D TLC produces better separation in
particular for more complex lipid mixtures of leaf extracts [5]. In
2D TLC, the lipid–stationary interactions are different in the two
different solvents of the mobile phase since the solvent modifies
these interactions (for example by screening charges on the plate
surface). Therefore, by using different mobile phases in the two
dimensions, a further and better separation of lipid classes can be
achieved. We quantify lipid classes in 2D TLC by using reference
standard lipids (with known mol numbers) that run in parallel with
the unknown lipids. In the procedure presented in this chapter,
lipids are quantified by visualization with a copper sulfate/phos-
phoric acid solution followed by computer scanning and densito-
metric image analysis. Other staining methods exist and might be
used for different purposes [5, 7]. The second main method pre-
sented here is gas chromatography (GC) for quantification of thy-
lakoid fatty acids. GC is in most cases the first choice for fatty acid
analytics [8]. A prerequisite of GC application for fatty acid analysis
is the conversion into fatty acid methyl esters (FAME) to increase
their volatility and decrease their polarity. A high polarity of fatty
acids impairs the analysis by GC. Typically, fatty acids are analyzed
 Lipids in Thylakoid Membranes 307

by flame ionization detection [8], which provides high accuracy and

sensibility. The assignment of individual fatty acids in gas chroma-
tograms of lipid extracts is done by comparing profiles from known
samples, like plant seeds. The amount of each fatty acid is calculated
by comparing the area under the chromatogram peak with the area
of a known fatty acid with a specific amount (in mol). This “calibra-
tion” fatty acid is added to the lipid extract before FAME produc-
tion and therefore acts as an internal standard. A further advantage
of using flame ionization detection GC is that it works with very
small sampled amounts.

2 Materials

2.1 Thin-Layer 1. LE1 (lipid extraction 1, thylakoids): Chloroform þ methanol

Chromatography (1:1), 100 mL.
2. LE2 (lipid extraction 2, thylakoids): Chloroform þ metha-
nol þ dH2O (5:5:1), 100 mL.
3. LE3 (lipid extraction 3, leaves): Chloroform þ methanol (2:1),
100 mL.
4. KCl, 1 mol L1 (thylakoids), 100 mL.
5. KCl, 0.83 % (w/v) (leaves), 100 mL.
6. D1 (dimension 1 solvent): Chloroform þ methanol þ dH2O
(75:25:2.5), 200 mL.
7. D2 (dimension 2 solvent): Chloroform þ methanol þ acetic
acid þ dH2O (80:9:12:2), 200 mL.
8. Lipid reference standards: MGDG, 10 mg mL1; DGDG,
10 mg mL1; SQDG, 5 mg mL1; PG, 5 mg mL1. All in
toluene (see Note 1). Store at 20  C. Lipids are commercially
available.
9. Copper dye: 50 mL 85% H3PO4 þ 37.5 g CuSO4 þ 450 mL
dH2O.
10. 80% Acetone (acetone:water 80:20).
11. Chloroform (HPLC grade) for rinse.
12. Liquid N2.
13. Nitrogen gas.
14. TLC silica gel 60 (10  10 cm) plates.
15. Convection oven.
16. Table centrifuge (fixed-angle rotor).
308 Helmut Kirchhoff and Robert Yarbrough

2.2 Gas 1. TAG (15:0): 1,2,3-Tripentadecanoyl-sn-glycerol (commer-

Chromatography cially available) as internal fatty acid standard at 2 μg μL1 in
toluene. 2 μg μL1 TAG (15:0) translates into
2.613 nmol μL1.
2. FAME 1: 2.5% H2SO4 in methanol.
3. Hexanes.
4. Seed extract: Take about 30 Arabidopsis seeds and make a
FAME as detailed in Subheading 3.5.

3 Methods (See Note 2)

3.1 Lipid Extracts 1. Take 50–100 μL of isolated thylakoid membranes, equivalent

from Isolated to 30 μg of chlorophyll, in a glass test tube that was rinsed with
Thylakoid Membranes chloroform (fat free). Add first 400 μL LE1 and then 200 μL
LE2. Thoroughly mix by vortexing (10 s).
2. Add 600 μL 1 M KCl solution and vortex for 10 s.
3. Centrifuge for 5 min at 580  g. After centrifugation, the
upper aqueous phase is clear and the lower organic phase is
green.
4. Carefully harvest the lower green organic phase with a long-
neck glass Pasteur pipette and transfer it to a glass tube that has
been rinsed with chloroform.
5. Perform chlorophyll determination [9]. Add 50 μL sample to
1 mL 80% acetone in a microcentrifuge tube (1.5 mL tube size)
and mix by vortexing. Spin sample at 10,000  g for 4 min.
Collect the supernatant. Blank spectrophotometer at 750 nm
with 80% acetone and measure absorption (abs) of the 80%
acetone extract at 750 nm, 663.6 nm, and 646.6 nm. Calculate
the total chlorophyll concentration of the lipid extract
(ChlLiEx) with

ChlLiEx ¼ ½19:54∙ðabs646:6  abs750 Þ þ 8:29∙ðabs663:6  abs750 Þ∙1050=50

ChlLiEx is given in nmol Chl mL1 and will be used later for
calculation of the lipid-to-chlorophyll ratios.

6. If not used directly for TLC or GC, then the lipid extract can be
stored at 20  C (see Note 3).

3.2 Lipid Extracts 1. Pre-chill mortar and pestle with liquid N2. Place about 5 g
from Leaves fresh-cut leaves into the mortar and add enough liquid N2 to
cover the leaves.
2. After evaporation of liquid N2, use a pestle to crush the leaves
to coarse granules. If required add more liquid N2 and repeat.
 Lipids in Thylakoid Membranes 309

3. Add enough LE3 to completely submerge granules. Let sit for

10 min.
4. By using a Pasteur pipette, remove all solvent from mortar and
place in a chloroform-rinsed glass vial (15 mL size, with Teflon
lid). Add about 20% volume of 0.83 % (w/v) KCl and vortex
thoroughly (10 s).
5. Centrifuge for 5 min at 580  g and carefully harvest the lower
green organic phase with a long-neck glass Pasteur pipette and
transfer it to a glass tube that has been rinsed with chloroform.
6. If not used directly for TLC or GC, then the lipid extract can be
stored at 20  C (see Note 3).

3.3 Two- 1. Have two TLC chambers with a sealed lid ready, one with
Dimensional Thin- solvent D1 and the other with solvent D2, respectively. TLC
Layer Chromatography chambers with a latch lid (Fig. 1) are highly recommended
(2D TLC) since they allow for excellent solvent vapor equilibration in
the chamber, which is required for good lipid separation (see
Note 4). The tanks should be equilibrated for at least 1 h
(to allow the gas phase to be saturated with the respective
solvent vapor). The solvents in the chambers should not be
higher than 1 cm.
2. Activate the TLC silica gel 60 (10  10 cm) plate for 30 min at
70  C in a convection oven (see Note 5). Be careful not to
touch the plate with your fingers; use gloves, and hold the
plates at the rim.
3. With a soft pencil, draw two lines on the TLC plate as shown in
Fig. 2a: 1 cm from the bottom and 1 cm from the left to mark

Fig. 1 TLC chamber with a latch lid system. The latch lid system allows optimal
sealing required for efficient equilibration of solvent vapor within the chamber
310 Helmut Kirchhoff and Robert Yarbrough

a b Leaf extract c Thylakoid extract

2nd
R3 R1 R2 R1 R2 R1
References R2
R4 R3 Pigments R3 R3
R4 R4 R4

References
MGDG
1 R1 1 R1 R1
R1
2 R2 2 R2 DGDG R2
Pencil lines 5
6 SQDG PG
4 3 R3 4 R3
3 R3 R4
R4 R4
8 7
Sample Background
R1 R2 1st

Fig. 2 2D TLC analysis. (a) Schematic drawing of a TLC plate with indicated directions for solvent 1 (first
dimension) and solvent 2 (second dimension). (b) TLC plate lipid profile from an Arabidopsis leaf extract.
1, MGDG; 2, DGDG; 3, PG; 4, SQDG; 5, phosphatidylethanolamine; 6, DPG, diphosphatidylglycerol; 7, phosphati-
dylcholine; 8, phosphatidylserine and phosphatidylinositol (R, stands for reference lipids). (c) TLC plate lipid
profile from an isolated thylakoid membrane extract. The plate on the very right shows a converted negative
image version of the TLC plate for the thylakoid extract. The area of interest (AOI) for MGDG, DGDG, SQDG, and
PG and their references are highlighted by dashed circles. The grey intensity values (I) and areas (A) for each spot
are used for lipid quantification. R1 to R4 indicate reference standard lipids as specified in the text

the height levels of the solvents in the first and second dimen-
sions in the TLC chambers. The lipid spots should not be in
direct contact with the solvents.
4. Lipid reference standard preparation (see Note 6): Mix four
lipid reference standards (R1–R4 in Fig. 2a) in a 1 dram vial
(rinsed with chloroform), respectively, as follows:
R1: 1 μL MGDG [12.85 nmol/μL ¼ 10 μg/μL] þ 1 μL PG
[6.45 nmol/μL ¼ 5 μg/μL].
R2: 1 μL SQDG [5.67 nmol/μL ¼ 5 μg/μL] þ 1 μL DGDG
[10.64 nmol/μL ¼ 10 μg/μL].
R3: 1 μL MGDG [12.85 nmol/μL ¼ 10 μg/μL] þ 1 μL
DGDG [10.64 nmol/l ¼ 10 μg/μL].
R4: 1 μL SQDG [5.67 nmol/μL ¼ 5 μg/μL] þ 1 μL PG
[6.45 nmol/μL ¼ 5 μg/μL].
Store reference standards on ice until use.
5. Remove the entire chloroform phase of the lipid extract (either
from isolated thylakoids or from leaves) under a stream of N2
to get a completely dry lipid film. Take care that you do not
splash the lipid extract at the sides of the vial due to a high flow
rate of N2. Try to concentrate the lipid film to the bottom of
the vial. Dissolve the dry lipid film with 40 μL chloroform and
store on ice.
6. By using a glass capillary column, transfer the lipid extract to
the start point of the TLC plate (see sample circle in Fig. 2a).
The lipid extract will be drawn into the plate by capillary forces.
Before you add the next drop wait until the previous one has
 Lipids in Thylakoid Membranes 311

dried by gently blowing N2 gas over the sample spot. The aim is
to produce a well-defined and small sample spot. Be careful not
to damage the silica surface of the plate with the capillary
column. Repeat until the entire lipid extract is transferred
onto the TLC plate at the same spot. Rinse the lipid extract
vial with 20 μL chloroform and add the rinse onto the plate at
the same spot to ensure that the lipid extract is quantitatively
transferred to the plate. Repeat until no traces of chlorophyll
remain in the vial.
7. Add 20 μL chloroform to the lipid reference standards R1 and
R2. Add them quantitatively to the TLC plate in the same way
as for the lipid extract at the positions shown in Fig. 2a. Use
new capillary columns for each reference standard.
8. Place the TLC plate in the chamber with solvent D1 (first
dimension). Close the lid and run the plate until solvent
reaches about ¾ of the plate height. Remove the TLC plate
from the chamber and dry under N2 gas.
9. Turn the plate 90 counterclockwise (Fig. 2a).
10. Add 20 μL chloroform to the lipid reference standards R3 and
R4 and transfer as for R1 and R2 at the positions indicated in
Fig. 2a.
11. Place the TLC plate in the chamber with solvent D2 (second
dimension). Close the lid and run the plate until solvent
reaches about ¾ of the plate height. Remove the TLC plate
from the chamber and dry under N2 gas.
12. Pour copper dye solution into a flat glass dish that can hold the
10  10 cm TLC plate. Submerge the plate for 3 s in the dye.
Remove the plate and hold at an angle to let excess copper dye
drain away.
13. Incubate the plate at 80  C in a convection oven for 10–20 min
until the lipid spots are clearly visible (Fig. 2b, c, left). Do not
“overstain.” If the spots are too dark, then just place the plate
at room temperature and wait until spot intensity is adequate
(the spots fade away in time) (see Note 7).
14. The four thylakoid membrane lipids that should be seen are
indicated by numbers (1–4) in Fig. 2b, c.
15. Scan the plate immediately with a standard image scanner.

3.4 Densitometric 2D 1. Open scanned TLC plate with an image analysis software.
TLC Analysis Convert the image into an 8-bit grey image and invert the
and Quantification grey scale, i.e., the lipid spots appear whitish instead of grayish
of Lipids (Fig. 2c, right). Adjust the grey scale (brightness) so that all
lipid spots are clearly visible.
2. Draw an area of interest (AOI) around each reference standard
lipid spot and all unknown lipid spots (dashed circles in con-
verted image on the right) (see Note 8). Also take an AOI from
312 Helmut Kirchhoff and Robert Yarbrough

the plate background. From the AOIs determine the areas (A)
and mean grey intensities (I) of the reference (Aref, Iref),
unknown (Aunk, Iunk), and background spots (Iback).
3. Calculate the stain intensities (Int) for the reference (Intref) and
unknown (Intunk) spots by
Intref ¼ ðI ref  I back Þ∙A ref and Intunk ¼ ðI unk  I back Þ∙A unk

The quantity of each lipid class is given by

mollipid x ¼ ðmolref x ∙Intunk x Þ=Intref x

where the subscript “x” stands for each of the four lipids
MGDG, DGDG, SQDG, and PG, respectively.
4. A convenient and informative way to quantify thylakoid lipids is
to give the molar lipid-to-chlorophyll ratio. The amount of
lipid (mollipid x) for each lipid class “x” is divided by the amount
of chlorophyll that was on the plate. The latter is calculated in
Subheading 3.1 as ChlLiEx multiplied by the volume of the lipid
extract that was transferred to the TLC plate (see Note 9).

3.5 Fatty Acid Methyl 1. Place 20–100 μL of lipid extract (either from isolated thylakoid
Ester (FAME) Analysis membranes or from leaves) in a glass vial (any threaded small-
with Leaves bore vial will do) (see Note 10).
and Isolated Thylakoid 2. For each vial, add an internal fatty acid standard of 20 μL of
Membranes 2 μg μL1 TAG (15:0 fatty acid) (see Note 11).
3. Add 1 mL of FAME1 solution and seal the vial with Teflon seal
screw caps. Incubate the vial for 1 h at 80  C in a water bath (see
Note 12). This treatment cleaves the lipid ester bonds releasing
free fatty acid methyl esters by transesterification.
4. Remove from water bath and let cool. Open the vial and add
200 μL hexanes and 1.5 mL dH2O. Mix thoroughly by manu-
ally shaking the vial rack for 1 min.
5. Spin at 580 x g for 2 min. This generates a two-phase system
with the upper green organic phase and a lower clear aqueous
phase.
6. Transfer 100 μL of the very small upper organic phase to a
micro reaction GC vial (cone-shaped borosilicate glass for easy
removal of small sample volumes with PTFE-faced rubber
septum and cap) (see Note 13). Seal the vial with a manual or
an automatic crimper.
7. The FAME can be used directly for GC or stored at 20  C.

3.6 Gas 1. Perform GC analysis with the FAMEs (see Note 14).
Chromatography 2. Obtain a chromatogram. A typical GC chromatogram from a
and Analysis thylakoid extract and from plant seeds (see Note 15) is shown
in Fig. 3. The seed chromatogram is used as a reference to
 Lipids in Thylakoid Membranes 313

Thylakoid extract
18:3
15:0 (standard)

16:0 16:3

16:1(3t) 18:0 18:2

18:1

Seed extract

Fig. 3 Example of a gas chromatogram. One microliter of FAME was injected into
the system. The lower profile is from a seed extract that is used for peak
assignment on the thylakoid chromatogram (dashed lines). The 15:0 TAG
internal standard is indicated and is used for quantification of the fatty acids

assign the different peaks to fatty acid types based on their

retention times. Commercial GC systems usually make these
assignments automatically after the seed profile is established.
3. Calculate the area under each peak relative to an internal stan-
dard. With the area for the TAG standard and the information
that 1 μL TAG (15:0) contains 2.613 nmol fatty acid, the
amount of the unknown fatty acids can be calculated. Before
moles of fatty acids can be calculated by comparison to the
TAG signal, the peak areas must be corrected for differences in
molecular weight for the different fatty acids [10] (see Note
16). Standard molecular weight correction factors (MWF) for
the GC peaks shown in Fig. 3 are (C16:0 was arbitrarily set to
one) 15:0 (TAG)—1.055; 16:0—1.000; 16:1—1.007;
16:13t—1.007; 16:2—1.015; 16:3—1.023; 18:0—0.906;
18:1—0.912; 18:2—0.918; and 18:3—0.925 [10].
4. Calculate the mol amount for a fatty acid by

Areafatty acid ∙MWFfatty acid

nmolLipid ¼ 2:613 nmol∙
AreaTAG ∙MWFTAG
314 Helmut Kirchhoff and Robert Yarbrough

4 Notes

1. Store in 1 mL or smaller brown glass vials with a Teflon-sealed

lid at 20  C to reduce evaporation of the toluene. For TLC
use, keep lipid standards cooled by placing them on ice. Reduce
opening of the vial to a minimum.
2. Hazards: Gloves and lab jacket are to be worn due to carcino-
genic, toxic, and corrosive nature of solvents and acids. All
work is only to be conducted in a fume hood except for the
chlorophyll determination. Most solvents are flammable and
should be kept away from open flames.
3. Since chloroform is volatile, it is important to use sealed lids for
storing the lipid extract. Teflon sealing is preferable.
4. If no latch lid TLC chambers are available, then a simple glass
lid can be used instead. In this case, the back of the chamber
should be covered with filter paper to allow better solvent
saturation of the chamber. The glass lid can be sealed with
grease and weighted down to allow good sealing of the
chamber.
5. The heating removes all the water from the plate.
6. The reference lipids are pre-quantified based upon their manu-
facturer or you can quantify them yourself via GC. The refer-
ence spot for MGDG should be equivalent to 12.85 nmol of
lipid, DGDG should be equivalent to 10.64 nmol of lipid,
SQDG should be equivalent to 5.67 nmol of lipid, and PG
should be equivalent to 6.45 nmol of lipid.
7. If the plate shows some “streaking,” i.e., lipid classes do not
appear as well-defined spots but more like a band, then this is
most likely caused by not having the solvent vapor equilibrated
in the TLC chambers. In that case, use fresh LE1 and LE2
solvents and let the chambers equilibrate for at least 1 h. If the
problem persists, then use filter paper at the back of the cham-
bers that stands in the solvents and reaches the top of chamber
to help solvent vapor equilibration.
8. Draw AOI circle generously around the spot to enclose the
whole stain intensity.
9. Typical numbers determined by this method for Arabidopsis
leaves are 56 mol% MGDG, 24 mol% DGDG, 13 mol% PG,
7 mol% SQDG, and a ratio of 2.8 mol total lipids/mol
chlorophyll.
10. If the fatty acid content is unknown, then it is advisable to
prepare vials with different volumes to identify the best volume
for GC analysis.
 Lipids in Thylakoid Membranes 315

11. TAG (15:0) is an adequate standard since this is not found in

leaves and can therefore be easily identified in GC
chromatographs.
12. Check often as samples have a tendency to boil if caps are not
tightly sealed.
13. Be extremely careful not to contaminate with the aqueous
phase since this could clog the GC system.
14. A typical GC column is a long, very thin gel capillary column
(e.g., 30 m  0.53 mm) with the inner wall fused to a 1.0 μm
thick layer of silica. The carrier gas is often inert helium. For
our measurements, we use an Agilent 6890 series GC system.
Methods for GC vary depending upon the machine, column
type, gases, etc. With our setup it takes around 8 min to detect
lipids through 20: fatty acids at operation temperatures of
190–250  C for proper flame ionization.
15. Plant seeds are used because it is straightforward to produce a
FAME from seeds and they have a well-defined fatty acid
composition.
16. The GC chromatogram signal is proportional to the amount of
the fatty acid and its molecular weight.

Acknowledgments

The authors acknowledge support from the US National Science

Foundation (MCB-1616982), the US Department of Energy
(DE-SC0017160), the US Department of Agriculture (ARC
grant WNP00775), and the US National Institute of Food and
Agriculture (NIFA, 2011-68005-30416) and USDA National
Institute of Food and Agriculture, Hatch project 0119.

References
1. Boudière L, Michaud M, Petroutsos D,
Rébeillé F, Falconet D, Bastien O, Roy S,
Finazzi G, Rolland N, Jouhet J, Block MA,
Maréchal E (2014) Glycerolipids in photosyn-
thesis: composition, synthesis and trafficking.
Biochim Biophys Acta 1837:470–480
2. Block MA, Dorne AJ, Joyard J, Douce R
(1983) Preparation and characterization of
membrane fractions enriched in outer and
inner envelope membranes from spinach chlor-
oplasts. 2. Biochemical characterization. J Biol
Chem 258:3281–3286
3. Murata N, Siegenthalter PA (1998) Chapter 1:
Lipids in photosynthesis: an overview. In:
Murata N, Siegenthaler PA (eds) Lipids in
photosynthesis: structure, function and genet-
ics. Kluwer Academic Publishers, Dordrecht,
pp 1–20
4. Garab G, Ughy B, Goss R (2016) Role of
MGDG and non-bilayer lipid phases in the
structure and dynamics of chloroplast thyla-
koid membranes. Subcell Biochem
86:127–157
5. Christie WW, Dobson G (1999) Thin-layer
chromatography re-visited. Lipid Technol
11:64–66
6. Wever V, Doermann P, Hoelz G (2013)
Chapter 8: Analysis and quantification of
plant membrane lipids by thin-layer chroma-
tography and gas chromatography. In:
316 Helmut Kirchhoff and Robert Yarbrough
Munnik T, Heilmann I (eds) Plant lipid signal-
ing protocols. Springer Science and Business
Media, p 69–78
7. Wang Z, Benning C (2011) Arabidopsis thali-
ana polar glycerolipid profiling by thin layer
chromatography (TLC) coupled with
gas-liquid chromatography (GLC). J Vis Exp
49:2518
8. Wu Z, Zhang Q, Li N, Pu Y, Wang B, Zhang T
(2017) Comparison of critical methods devel-
oped for fatty acid analysis: a review. J Sep Sci
40:288–298
9. Porra RJ, Thompson WA, Kriedermann PE
(1989) Determination of accurate extinction
coefficient and simultaneous equations for
assaying chlorophylls a and b extracted with
four different solvents: verification of the con-
centration of chlorophyll standards by atomic
absorption spectroscopy. Biochim Biophys
Acta 975:384–394
10. Christie WW (1989) Chapter 5: Gas chro-
matographic analysis of fatty acid derivatives.
In: Christi WW (ed) Gas chromatography and
lipids. Oily Press Ltd, Dundee, pp 73–75
 Part IV

Photosynthesis-Inspired Energy Generation

 Chapter 19

“Click” Methodology for the Functionalization of Water

Oxidation Catalyst Iridium Oxide Nanoparticles
with Hydrophobic Dyes for Artificial Photosynthetic
Constructs
Jackson D. Megiatto Jr. and Catia Ornelas

Abstract
The unusually high tolerance toward chemical functional groups of the copper(I)-catalyzed Huisgen-
Sharpless-Meldal 1,3-dipolar cycloaddition of azides and alkynes protocol (the CuAAC or “click” reaction)
associated with its mild conditions and high yields has been explored in the present methodology to
successfully prepare water oxidation catalyst iridium oxide nanoparticles decorated with organic dyes.
The “click reaction” has proven to be an excellent synthetic tool to overcome the incompatible solubility
of the hydrophilic iridium oxide nanoparticles and the hydrophobic dyes. A complex artificial photosyn-
thetic model designed to mimic the photoinduced redox processes occurring in photosystem II is used as a
hydrophobic dye to highlight the efficiency and selectiveness of the method.

Key words Artificial photosynthesis, Water oxidation catalysts, “Click chemistry”, Photoelectro-
chemical cell, Photocatalyst, Iridium oxide nanoparticles

1 Introduction

Sunlight is an abundant and widely distributed energy source on

Earth. Conservative calculations estimate that the Sun delivers solar
energy to our planet on an annual basis at a rate of approximately
120,000 TW (1 TW ¼ 1012 W), which dwarfs both the 13.5 TW
current global annual rate of human energy consumption and the
43.0 TW projected annual rate of consumption by the end of this
century. Although abundant, solar energy in the form of sunlight is
an essentially useless form of energy. Sunlight is diffuse, with a
yearly average power rate of about 170 TW per square meter on
Earth’s surface. Therefore, solar energy utilization requires sun-
light capture, conversion, and storage [1–3].
Several technologies, which have been collectively called artifi-
cial photosynthesis, have been devised to efficiently harvest,

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_19, © Springer Science+Business Media, LLC, part of Springer Nature 2018

319
320 Jackson D. Megiatto Jr. and Catia Ornelas

convert, and store sunlight for societal use. Photovoltaics systems

are the pioneer technology to harness and convert sunlight into
electricity. However, modern society cannot rely on photovoltaic
solar electricity as a primary source of energy because of the diurnal
variation of insolation. Although solar electricity could be in prin-
ciple stored in batteries, current battery technologies are too expen-
sive to make photovoltaic solar electricity economically viable
[3–5].
Another approach to store sunlight is borrowed from nature:
storage of sunlight in chemical bonds, namely fuels. The natural
photosynthetic apparatus is composed of two photoactive enzymes
linked to each other in series, called photosystem II and I. Upon
irradiation, the chlorophyll-based pigment P680 in photosystem II
is activated and triggers a series of electron and proton transfer
reactions that produce a “wireless electrical current.” The anodic
component of the “wireless electrical current” is used to activate
the oxygen-evolving complex (OEC), the catalytic inorganic cluster
that oxidizes two water molecules after four light flashes into
molecular oxygen and protons. The cathodic component of the
“wireless electrical current” is captured by photosystem I, which
upon activation by sunlight reduces the protons released by photo-
system II to reduced hydrogen equivalents that are stored through
the conversion of NADP to NADPH [3, 6].
The machinery of natural photosynthesis is too complex to be
replicated in laboratory. However, the chemical and physical prin-
ciples underlying such complex systems can be used as a valuable
guide to design simpler artificial photosynthetic devices [7–11]. In
this context, water-splitting photoelectrochemical cells, which are
composed of photoanodes able to oxidize water molecules
connected in series with photocathodes capable of reducing pro-
tons to hydrogen gas (fuel), have gained increased attention [12].
One approach to prepare the photoelectrodes of such artificial
photosynthetic systems is based on those developed for dye-sensi-
tized solar cells (DSSC) [13, 14]. The key difference between
standard DSSC photoelectrodes and those needed for water-
splitting photoelectrochemical cells resides in sensitizing the semi-
conductor material with dyes bearing catalysts able to promote
water oxidation and proton reduction reactions. Figure 1 provides
a schematic diagram of a water-splitting photoelectrochemical cell
composed of a DSSC-based photoanode and a platinum “dark”
cathode (see Note 1) designed to produce hydrogen gas from water
and sunlight (see Note 2).
One of the biggest challenges in the development of water-
splitting photoelectrochemical cells is the proper assembling of the
photoanode component. In collaboration with Mallouk’s and
Gust-Moore-Moore’s groups at Penn State and Arizona State Uni-
versities, respectively, we have developed water-splitting photoa-
nodes composed of catalytically active iridium oxide nanoparticles
 Preparation of Photocatalysts for Water Splitting 321

Fig. 1 Water-splitting photoelectrochemical cell schematic diagram for production of hydrogen gas. For this
type of photoelectrochemical cell to achieve overall water splitting (decomposition of water into oxygen (O2)
and hydrogen (H2) gases), an external voltage of about 200 mV must be applied on the platinum “dark”
electrode (see Notes 1 and 2). WOC water oxidation catalyst

(IrOx-NPs) decorated with organic dyes that are thermodynami-

cally and kinetically designed to inject electrons upon irradiation
into semiconductor materials such as titanium and tin dioxides
(TiO2 and SnO2, respectively) while removing electrons from the
IrOx-NPs that in turn promote water oxidation reactions [4, 7,
10]. However, the lack of an efficient method to functionalize the
water-soluble IrOx-NP catalyst with hydrophobic dyes has limited
the successful assembly of such photoanodes. The tendency of
organic dyes to aggregate in water, associated with their chemical
instability to the alkaline conditions and high temperatures
required by the current IrOx-NPs protocols [15–18], has forced
us to develop innovative methods to prepare IrOx-based photo-
anode systems.
Here we describe an efficient and versatile methodology that
allows the functionalization of IrOx-NPs with any hydrophobic dye
in high yields and under mild conditions using the Cu(I)-catalyzed
Huisgen-Sharpless-Meldal-1,3-dipolar cycloaddition of azides and
alkynes (the CuAAC or “click” reaction) [19–21] to create photo-
catalysts that can be used to sensitize semiconductor materials and
produce photoanodes for water-splitting photoelectrochemical
cells.
322 Jackson D. Megiatto Jr. and Catia Ornelas

Fig. 2 Synthetic strategy used to prepare IrOx-NPs functionalized with iodomethyl (IrOx-CH2I-NPs 1) and
azidomethyl (IrOx-CH2N3-NPs 2) peripheral groups (see Note 3). IrOx-CH2N3-NPs 2 are able to react with
alkynyl-functionalized organic dyes under “click” conditions

Fig. 3 Synthesis of photosynthetic model 8. The synthetic work begins with preparation of porphyrin 5, which
relies on an approach based on the selective acid-catalyzed condensation reaction of one of the formyl groups
of compound 3 with 5-(pentafluorophenyl)dipyrromethane 4 and commercially available 4-iodobenzaldehyde.
This selectivity is grounded in the reduced chemical reactivity of one of the equivalent formyl groups, which is
involved in the intramolecular O—H---O¼C hydrogen bond. This H-bonding formation efficiently discourages
its activation by acid catalysts. The benzimidazole moiety is then formed upon cyclization of the remaining
formyl group in 5 with orthophenylenediamine to yield 6, which is in turn hydrolyzed under acid conditions to
yield 8. Target photosynthetic model 8 is obtained by amide-coupling reaction between 8 and propargyl amine
using 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDCI) as coupling agent and dimethylaminopyridine as
catalyst (see Notes 5 and 6)

In this method, IrOx-NPs with an average 2 nm diameter and

with peripheral iodomethyl functional groups IrOx-CH2I-NPs
1 (Fig. 2) are prepared from commercially available iridium salts,
using readily available iodoacetic acid as the stabilizing agent in
alkaline aqueous medium (see Note 3). Treatment of IrOx-CH2I-
NPs 1 with sodium azide in neutral aqueous medium creates IrOx-
CH2N3-NPs 2, whose surface is decorated with azide functional-
ities (see Note 4). The IrOx-CH2N3-NPs 2 dispersed in water
readily react under “click” conditions with terminal alkynyl-
functionalized hydrophobic dyes previously dissolved in
tetrahydrofuran.
 Preparation of Photocatalysts for Water Splitting 323

For the present purpose, this approach is illustrated using the

biomimetic artificial photosynthetic model 8 (Fig. 3) (see Notes 5
and 6) [8, 9, 22] as organic dye, which is able to undergo photoin-
duced electron transfer (ET) and proton-coupled electron transfer
(PCET) reactions once anchored to semiconductor materials
[9]. Those events mimic the photo-redox processes occurring
between the primary electron donor chlorophyll complex (P680)
and the tyrosine-histidine pair (Tyrz-D1His190) in PSII [6, 8, 9,
22]. We chose this complex photosynthetic model to demonstrate
the chemical selectiveness of the present methodology, although
the method can be extended to other dyes, provided that they bear
terminal alkynyl functionalities.
CuAAC or “click” reaction between IrOx-CH2N3-NPs 2 and
model 8 yields final photocatalyst 9 (Fig. 4). However, a major
challenge to overcome when using the CuAAC or “click” reaction
is the possibility of residual copper contamination of the final
products and/or damage of sensitive substrates. This is particularly
critical when working with porphyrin pigments, which are known
to readily coordinate copper species in solution to form Cu(II)
porphyrinates [23]. To circumvent this synthetic issue, we propose
the use of sulfonated bathophenanthroline 10 (Fig. 4) as an auxil-
iary ligand in the preparation of the “click” catalyst. Ligand 10 is
known to enhance the “click” cycloaddition reaction kinetics by
stabilizing the Cu(I) catalytic species, while sequestering it from the
reaction medium, thereby avoiding copper contamination of sensi-
tive substrates such as porphyrin dyes [24–28]. The present
method also describes the experimental details to prepare the
“click” catalyst from copper iodide and ligand 10.

2 Materials

2.1 Reagents 1. Iodoacetic acid (>98%).

2. Potassium hexachloroiridate (IV) (>99.99%).
3. Sodium hydroxide solution 25% (w/w) in ultrapure water.
4. Sodium azide (>99.5%).
5. Copper(I) iodide (>99.5%).
6. Bathophenanthrolinedisulfonic acid disodium salt hydrate,
98% (compound 10 in Fig. 4): This chemical compound is
also known as sulfonated bathophenanthroline.
7. (þ)-Sodium L-ascorbate (crystalline >98%).
Fig. 4 Synthetic strategy used to prepare target photocatalyst 9 from IrOx-CH2N3-NPs 2 and model 8 using the
CuAAC or “click” reaction. Inset: sulfonated bathophenanthroline auxiliary ligand 10 reacting with copper
iodide to yield the “click” catalyst that promotes the anchoring process of model 8 to the surface of the IrOx-
NPs’ surfaces through formation of triazole rings
 Preparation of Photocatalysts for Water Splitting 325

8. Photosynthetic model 8, which is prepared following the syn-

thetic strategy depicted in Fig. 3. Any other dye can be used
instead of 8. However, the dye must bear a terminal alkynyl
functionality for the method to work.

2.2 Solvents 1. Ultrapure water.

2. Tetrahydrofuran (>99%).
3. Methanol (spectroscopic grade).
4. Ethanol (>95%).

2.3 Gases 1. Argon gas.

3 Methods

3.1 Preparation 1. Use ultrapure water to synthesize the IrOx-NPs and in the
of the Iridium Oxide dialysis procedure.
Nanoparticles 2. Cut 10 cm of dialysis membrane. Here we use Spectra/Por®
6 MWCO, molecular weight cutoff 2000 with 38 mm flat
3.1.1 Dialysis width. A similar membrane may be used but must first be
empirically tested.
3. Place the aqueous solution containing the IrOx-NPs into the
dialysis bag.
4. Seal the dialysis membrane using resealable closures.
5. Place the sealed dialysis membrane in a beaker containing 1 L of
ultrapure water with gentle stirring for 5 h at room
temperature.
6. Replace the ultrapure water in the beaker six times every 5 h,
totalizing 30 h of dialysis purification for each batch of IrOx-
NPs.

3.1.2 Synthesis of IrOx- 1. Add 17 mL of ultrapure water into a 20 mL glass vial equipped
CH2I-NPs 1 with a magnetic stir bar.
2. Add iodoacetic acid (0.046 g, 0.25 mmol) and K2IrCl6
(0.010 g, 0.021 mmol) as solids to the glass vial (see Note 7).
3. Adjust the pH of the solution to 10 using NaOH aqueous
solution (about 70 μL of the NaOH solution 25% in water
(w/w)). Use universal pH paper to check the pH of the reac-
tion solution (see Note 8).
4. Close the vial with a plastic cap leaving the cap slightly loose,
and place the vial in an oil bath. Use a clamp to hold the vial.
5. Heat the solution at 90  C for 2 h under magnetic stirring
(500 rpm) (see Note 9).
326 Jackson D. Megiatto Jr. and Catia Ornelas

6. Remove the glass vial from the oil bath and let the resulting
blue solution cool to room temperature.
7. Purify the crude IrOx-CH2I-NPs 1 by dialysis against ultrapure
water (MWCO 2000) following the procedure described in
Subheading 3.1.1.

3.1.3 Synthesis of IrOx- 1. Add sodium azide (0.033 g, 0.50 mmol) (see Note 10) to the
CH2N3-NPs 2 glass vial containing the IrOx-CH2I-NPs 1 solution.
2. Close the vial with a plastic cap leaving the cap slightly loose,
and place the vial in an oil bath. Use a clamp to hold the vial.
3. Heat the mixture at 65  C in an oil bath for 2 h under magnetic
stirring (500 rpm).
4. Remove the glass vial from the oil bath and let the resulting
blue solution cool to room temperature.
5. Purify the crude IrOx-CH2N3-NPs 2 by dialysis against ultra-
pure water (MWCO 2000) following the procedure described
in Subheading 3.1.1 (see Note 11).

3.2 Characterization 1. Place 2 mL of the ultrapure water in a 1 cm quartz cuvette.

of the IrOx-NPs 1 and 2 2. Place the cuvette in the UV-Vis spectrometer and acquire the
Using UV-Visible blank.
Spectroscopy
3. Place 2 mL of the blue aqueous solution of IrOx-NPs 1 or
2 obtained from dialysis in a clean 1 cm quartz cuvette.

Fig. 5 Absorption spectra of the IrOx-CH2I-NPs 1 (blue line) and IrOx-CH2N3-NPs

2 (red line) in ultrapure water. The broad absorption with maximum at about
584 nm corresponds to the expected electronic transition for IrOx-NPs with
average diameter of about 2 nm
 Preparation of Photocatalysts for Water Splitting 327

Fig. 6 TEM images of IrOx-CH2I-NPs 1 (a) and IrOx-CH2N3-NPs 2 (b) after

purification by dialysis. The average diameter of the IrOx-particles is about 2 nm

4. Place the cuvette in the spectrometer and acquire the absorp-

tion spectrum.
5. Plot the data for IrOx-NPs 1 and 2 into graphical charts of
absorbance versus wavelength (in nanometers) to yield Fig. 5.

3.3 Characterization 1. Immerse the TEM gold grid in the aqueous solution contain-
of the IrOx-NPs 1 and 2 ing the IrOx-NPs for 1 s.
by Transmission 2. Let the gold grid dry at room temperature for 10 min.
Electron Microscopy
3. Place the gold grid in the sample holder of the TEM piece of
equipment and acquire the TEM images (Fig. 6) (see Note 12).

3.4 Preparation 1. In a 20 mL glass vial, place a magnetic stir bar and add 10 mL of
of the “Click” Catalyst a solvent mixture composed of ethanol and ultrapure water
(1:1, v/v).
2. Bubble the solution with argon gas for 5 min.
3. Add copper iodide (0.008 g, 0.042 mmol, 1 equiv.), sodium
ascorbate (0.033 g, 0.17 mmol, 4 equiv.), and sulfonated bath-
ophenanthroline 10 (0.045 g, 0.084 mmol, 2 equiv.) to the
glass vial.
4. Close the glass vial tightly with a Teflon-coated plastic cap.
5. Heat the resulting pink suspension at reflux for 2 min using a
heat gun under gentle magnetic stirring (200 rpm).
6. Allow the resulting dark red catalytic solution to cool to room
temperature.
7. This catalytic solution should be immediately (see Note 13)
used to promote the “click reactions” between the IrOx-
CH2N3-NPs 2 and the alkynyl-functionalized dye.
328 Jackson D. Megiatto Jr. and Catia Ornelas

3.5 “Click Reaction” 1. In a 20 mL glass vial, place a magnetic stir bar and add 10 mL of
Between the Artificial the IrOx-CH2N3-NPs 2 aqueous solution obtained after
Photosynthetic dialysis.
Model 8 and the IrOx- 2. Cap the vial with a rubber septum and bubble the solution with
NPs-N3 2 argon gas for 10 min.
3. Place 4.0 mg (3.53 μmol) of model 8 into a 1 mL micro
volumetric flask and complete the flask volume with tetrahy-
drofuran (THF).
4. Add 0.55 mL of the THF solution of model 8 (1.94 μmol) to
the reaction glass vial containing IrOx-CH2N3-NPs 2.
5. Add 0.9 mL of the freshly prepared “click” catalyst solution to
the reaction glass vial.
6. Cap the reaction glass vial with a rubber septum and bubble the
solution with argon gas for 1 min.
7. Place the reaction glass vial over a magnetic stirrer and stir the
reaction mixture (200 rpm) at room temperature for 24 h.
8. Place the resulting photocatalyst 9 in a separatory funnel and
wash the aqueous phase containing photocatalyst 9 with
dichloromethane (three times, 5 mL each time) (see Note 14).
9. Place the aqueous solution containing photocatalyst 9 in a clean
glass vial and bubble argon gas throughout the aqueous solu-
tion to remove any residual dichloromethane (see Note 15).
10. Purify photocatalyst 9 by dialysis against ultrapure water fol-
lowing the procedure described in Subheading 3.1.1.

3.6 Characterization 1. Place 2 mL of ultrapure water in a 1 cm quartz cuvette and add

of Photocatalyst 9 by 0.2 mL of methanol (see Note 16).
UV-Visible 2. Place the cuvette in the spectrometer and acquire the blank.
Spectroscopy
3. Place 2 mL of the aqueous solution of photocatalyst 9 in a 1 cm
quartz cuvette and add 0.2 mL of methanol.
4. Place the cuvette in the spectrometer and acquire the absorp-
tion spectrum.
5. Solubilize the necessary amount of compound 8 in 0.2 mL of
methanol and add 1.8 mL of ultrapure water to make a solution
with a concentration whose absorbance is within the range of
that measured for 9 for comparison purposes (see Note 17).
6. Place 2 mL of the solution of dye 8 in ultrapure water/metha-
nol (9:1, v/v) in a 1 cm quartz cuvette.
7. Place the cuvette in the spectrometer and acquire the spectra.
8. Plot the data as graphical charts of absorbance versus wave-
length (in nm) to yield Fig. 7.
 Preparation of Photocatalysts for Water Splitting 329

Fig. 7 Absorption spectra of photocatalyst 9 (pink line) and artificial photosyn-

thetic model 8 (green line) in ultrapure water/methanol solvent mixture (9:1, v/v).
Inset: zoom in for the spectrum region with wavelength higher than 450 nm to
show the typical relatively low-intense absorption Q-bands of the porphyrin core

Fig. 8 TEM image of photocatalyst 9 showing particle size of around 2 nm

3.7 Characterization 1. Immerse the TEM gold grid in the aqueous solution contain-
of Photocatalyst 9 by ing photocatalyst 9 for 1 s.
Transmission Electron 2. Let the gold grid dry at room temperature for 10 min.
Microscopy
3. Place the gold grid in the sample holder of the TEM piece
of equipment and acquire the TEM images (Fig. 8; see
Note 12).
330 Jackson D. Megiatto Jr. and Catia Ornelas

4 Notes

1. We have also designed photoelectrochemical cells in which

both anode and cathode are photoelectrodes, i.e., they need
light to function, that operate in a more similar manner to that
of the natural PSII and PSI systems. As the present method
describes the preparation of a photocatalyst that has been used
to prepare the anode electrode of our photoelectrochemical
cells, we decided to include a simpler photoelectrochemical
cell design with a platinum “dark” electrode as cathode in the
schematic diagram (Fig. 1) for instruction purposes.
2. The photoelectrochemical cell depicted in Fig. 1 is composed
of a DSSC-based photoanode wired in series with a platinum
“dark” cathode for sunlight-driven hydrogen gas production
from water. The anode and cathode compartments of the cell
are separated by a proton-permeable membrane that allows
proton movement between the compartments while avoiding
diffusion and mixture of other photo-generated products. The
photoanode contains a transparent conductive glass coated
with a thin layer of a semiconductor material sensitized with
an organic dye (porphyrin in this example), which is chemically
attached to a water oxidation catalyst (WOC). Upon irradia-
tion, the organic dye absorbs light and injects electrons into the
semiconductor, from whence they travel to the platinum cath-
ode and reduce protons (H+) to hydrogen gas (H2). The
oxidized dye in turn removes electrons from the WOC, thereby
activating it to split water molecules into oxygen gas (O2) and
protons. This is an ideal scenario. However, recent studies in
our group have revealed a kinetic mismatch in the photo-redox
processes occurring in this type of photoelectrochemical cell.
Briefly, the photo-injected electron in the semiconductor
recombines with the oxidized dye before the oxidize dye
removes an electron from the WOC. This recombination pro-
cess negatively affects the efficiency of the photoelectrochem-
ical cell. To overcome this kinetic hindrance and achieve overall
water oxidation and proton reduction, application of an exter-
nal voltage of about 200 mV has been necessary. For further
details, see our previous works [11, 12].
3. Some of the iodomethyl functionalities of the IrOx-NPs’ sur-
faces might undergo nucleophilic substitution reactions with
the excess of hydroxyl ions present in the reaction medium.
Therefore, the IrOx-NPs’ surfaces are partially and randomly
decorated with hydroxymethyl groups. Hydroxymethyl groups
on the IrOx-NPs’ surfaces are unreactive toward sodium azide
under the conditions described. Accordingly, this side reaction
deactivates some of the iodomethyl groups on the surface of
 Preparation of Photocatalysts for Water Splitting 331

the IrOx-NPs, thereby precluding full (100% yield) decoration

of the IrOxNPs’ surfaces with azide functionalities. We have
estimated the degree of iodomethyl substitution reactions indi-
rectly by preparing the IrOx-CH2I-NPs 1 following the proce-
dure described in Subheading 3.1 but using D2O instead of
ultrapure water. 1H NMR analyses of the reaction mixture
suggested that about 30% of the iodomethyl groups underwent
nucleophilic substitution reactions with hydroxyl ions after 2 h
under the conditions of the procedure. This measurement was
made by comparing the relative intensities of the signals at 3.65
and 4.03 ppm (500 MHz, D2O, 25  C), which correspond to
the resonances of the methylene groups in —CH2—I and
—CH2—OH moieties, respectively.
4. 1H NMR analyses of the crude IrOx-CH2N3-NPs 2 reaction
mixture suggested that about 80% of the remaining iodomethyl
groups underwent nucleophilic substitution reactions with the
azide ions after 2 h under the conditions of the procedure. This
measurement was made by comparing the relative intensities of
the signals at 3.65 and 3.83 ppm (500 MHz, D2O, 25  C),
which correspond to the resonances of the methylene groups in
—CH2—I and —CH2—N3 moieties, respectively. Therefore,
the surface of the isolated IrOx-CH2N3-NPs 2 is randomly
decorated with about 50% —CH2—N3, 20% —CH2—I and
30% —CH2—OH functionalities. As the hydroxymethyl and
iodomethyl groups are unreactive under the “click” conditions
reported within, the theoretical yield for the attachment of
model 8 or any other dye to the IrOx-NPs is about 50%.
5. Experimental conditions for the synthesis of compound
8 (Fig. 3): (i) 5-(pentafluorophenyl)dipyrromethane 4, com-
mercially available 4-iodobenzaldehyde, compound 3, boron-
trifluoride etherate, chloroform, 2,3-dichloro-5,6-dicyano-
1,4-benzoquinone, room temperature, 24 h, 9% yield;
(ii) orthophenylenediamine, nitrobenzene, argon atmosphere,
reflux, 8 h, 85% yield; (iii) hydrochloric acid, trifluoroacetic
acid, reflux, 24 h, 98% yield; and (iv) propargylamine,
1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDCI),
dimethylaminopyridine (DMAP), dichloromethane, room
temperature, 20 h, 80% yield. Further experimental details
about the syntheses of the building blocks and photosynthetic
model 8 can be found in our previous works [8, 9, 22].
6. The 4-iodophenylgroup at position 10 of the porphyrin core in
artificial photosynthetic model 8 is a useful chemical function-
ality for further introduction of phosphonate groups that
might be used to anchor photocatalyst 9 to semiconductor
materials in order to produce the final photoanode. For details,
see our previous work [2]
332 Jackson D. Megiatto Jr. and Catia Ornelas

7. When iodoacetic acid is dissolved in ultrapure water, the result-

ing solution should be colorless with pH of about 3. Upon
addition of K2IrCl6, the solution turns brown and the pH
drops to about 2.
8. At pH 10, the reaction solution should change color from
brown to light yellow.
9. Upon heating, the original light yellow solution should turn
colorless within the first 10 min. After about 30 min, the
reaction solution should become deep blue. If the blue color
is not observed by eye after 30 min, then IrOx-NPs with a
diameter larger than about 2 nm is forming. Large particles
have low or no catalytic activity in water oxidation reactions.
Therefore, if the deep blue color does not appear after about
30 min reaction time, the whole batch must be discarded and
the synthesis of IrOx-CH2I-NPs 1 should start from the
beginning.
10. Sodium azide is highly toxic and dangerous to the environ-
ment. Use appropriate personal protective equipment to
manipulate it and dispose of any azide waste properly.
11. IrOx-CH2N3-NPs 2 are stable under the conditions described.
They should be stored in closed glass vials in the dark at room
temperature. However, sedimentation of the IrOx-CH2-N3-
NPs might be observed after a few days under storage condi-
tions. We have found that simple shaking by hand of the
storage glass vial promotes the redispersion of the nanoparti-
cles with no size changes and loss of chemical reactivity and/or
catalytic properties.
12. In TEM images, one should look for black round particles with
about 2 nm diameter. Sometimes, large aggregates composed
of several correctly sized IrOx-NPs might be observed on the
TEM images due to water evaporation during sample prepara-
tion for TEM analyses. As long as individual particles with
about 2 nm diameter are still observed within the aggregates,
TEM images confirm formation of active IrOx-NPs.
13. The “click” catalyst is highly sensitive to oxygen in the air.
Oxidation of the catalytic active Cu(I) species leads to inactive
Cu(II) ions. Therefore, use of freshly prepared catalysts is
highly recommended. However, if stored under argon atmo-
sphere, in the dark and room temperature, the deep red “click”
catalyst solution is still active after 1 week maximum. After this
period, a new batch must be prepared.
14. An indication of the success of the synthesis of photocatalyst 9
is the color of the aqueous and dichloromethane phases. The
aqueous phase should have the same color of that observed
from the pristine organic dye dissolved in organic solvents. In
the case of 8, the aqueous phase should be purple to reddish.
 Preparation of Photocatalysts for Water Splitting 333

Usually, the dichloromethane phase resulting from the first

washing is slightly yellow, indicating that the reaction between
8 and IrOx-CH2N3-NPs 2 was quantitative. If the dichloro-
methane phase maintains the dye color, keep washing the
aqueous phase until the dichloromethane phase is colorless.
15. The dialysis membrane is not chemically compatible with
dichloromethane. We have found that residual dichloro-
methane disrupts the membrane.
16. The absorption spectra of dye 8 and photocatalyst 9 were
obtained in a water/methanol (9:1, v/v) solvent mixture due
to the insolubility of 8 in pristine water. Therefore, the same
amount of methanol is added to the aqueous solution of 9 for
comparison purposes. If the dye is not soluble in methanol,
then tetrahydrofuran might be used instead. If the dye is water
soluble, there is no need for a co-solvent and the UV-Vis
analyses should be carried out in ultrapure water.
17. The absorption spectrum of photocatalyst 9 should be very
similar to that of compound 8; otherwise aggregation of the
dye on the surface of the IrOx-NPs probably occurred. Aggre-
gation of the organic dye on the NPs’ surfaces might alter the
physical and chemical properties of the final photocatalyst. If
the absorption spectra of model 8 and photocatalyst 9 are
different, then the batch should be discarded and the synthesis
of 9 started from the beginning.

Acknowledgments

This work was supported by FAPESP (The State of São Paulo

Research Foundation, Brazil) under Award Numbers 2013/
22160-0 and 2015/23761-2.

References

1. Blankenship RE et al (2011) Comparing pho- 5. Barber J (2008) Photosynthetic energy conver-

tosynthetic and photovoltaic efficiencies and
recognizing the potential for improvement.
Science 332:805–809
2. Hoffert MI et al (2002) Advanced technology
paths to global climate stability: energy for a
greenhouse planet. Science 298:981–987
3. Lewis NS, Nocera DG (2006) Powering the
planet: chemical challenges in solar energy uti-
lization. Proc Natl Acad Sci U S A
103:15729–15735
4. Gust D, Moore TA, Moore LA (2011) Realiz-
ing artificial photosynthesis. Faraday Discuss
155:9–26
sion: natural and artificial. Chem Soc Rev
38:185–196
6. Blankenship RE (2002) Molecular mechanisms
of photosynthesis. Blackwell Science, Oxford
7. Gust D, Moore TA, Moore LA (2009) Solar
fuels via artificial photosynthesis. Acc Chem
Res 42:1890–1898
8. Megiatto JD Jr et al (2012) Mimicking the
electron transfer chain in photosystem II with
a molecular triad thermodynamically capable of
water oxidation. Proc Natl Acad Sci U S A
109:15578–15583
9. Megiatto JD Jr et al (2014) A bioinspired
redox relay that mimics radical interactions of
334 Jackson D. Megiatto Jr. and Catia Ornelas
the Tyr–His pairs of photosystem II. Nat Chem
6:423–428
10. Zhao Y et al (2012) Improving the efficiency of
water splitting in dye-sensitized solar cells by
using a biomimetic electron transfer mediator.
Proc Natl Acad Sci U S A 109:15612–15616
11. Sherman BD et al (2014) Evolution of reaction
center mimics to systems capable of generating
solar fuel. Photosynth Res 120:59–70
12. Sherman BD et al (2016) A tandem
dye-sensitized photoelectrochemical cell for
light driven hydrogen production. Energy
Environ Sci 9:1812–1817
13. O’Regan B, Gr€a tzel M (1991) A low-cost,
high-efficiency solar cell based on
dye-sensitized colloidal TiO2 films. Nature
353:737–740
14. Gr€a tzel M (2001) Photoelectrochemical cells.
Nature 414:338–344
15. Youngblood WJ et al (2009) Photoassisted
overall water splitting in a visible light absorb-
ing dye-sensitized photoelectrochemical cell. J
Am Chem Soc 131:926–927
16. Hara M, Waraksa CC, Lean JT, Lewis BA,
Mallouk TE (2000) Photocatalytic water oxi-
dation in a buffered tris(2,20 -bipyridyl)ruthe-
nium complex-colloidal IrO2 system. J Phys
Chem A 104:5275–5280
17. Hoertz PG, Kim YI, Youngblood WJ, Mallouk
TE (2007) Bidentate dicarboxylate capping
groups and photosensitizers control the size
of IrO2 nanoparticle catalysts for water oxida-
tion. J Phys Chem B 111:6845–6856
18. Harriman A, Thomas JM, Millward GR (1987)
Catalytic and structural-properties of iridium-
iridium dioxide colloids. New J Chem
11:757–762
19. Huisgen R (1968) Cycloadditions – definition,
classification, and characterization. Angew
Chem Int Ed Engl 7:321–328
20. Kolb HC, Finn MG, Sharpless KB (2001) Click
chemistry: diverse chemical function from a
few good reactions. Angew Chem Int Ed
40:2004–2021
21. Tornøe CW, Christensen C, Meldal M (2002)
Peptidotriazoles on solid phase: [1,2,3]-
triazoles by regiospecific copper(I)-catalyzed
1,3-dipolar cycloadditions of terminal alkynes
to azides. J Org Chem 67:3057–3064
22. Megiatto JD Jr et al (2012) Intramolecular
hydrogen bonding as a synthetic tool to induce
chemical selectivity in acid catalyzed porphyrin
synthesis. Chem Comm 48:4558–4560
23. Wagner RW, Ciringh Y, Clausen C, Lindsey JS
(1999) Investigation and refinement of
palladium-coupling conditions for the synthe-
sis of ciarylethyne-linked multiporphyrin
arrays. Chem Mater 11:2974–2983
24. Lewis WG, Magallon FG, Fokin VV, Finn MG
(2004) Discovery and characterization of cata-
lysts for azidealkyne cycloaddition by fluores-
cence quenching. J Am Chem Soc
126:9152–9153
25. Megiatto JD Jr, Schuster DI (2008) General
method for synthesis of functionalized macro-
cycles and catenanes utilizing “click” chemis-
try. J Am Chem Soc 130:12872–12873
26. Megiatto JD Jr et al (2010) [2]catenanes deco-
rated with porphyrin and [60]fullerene groups:
design, convergent synthesis, and photoin-
duced processes. J Am Chem Soc
132:3847–3861
27. Kimer SV et al (2015) Synthesis and photophy-
sical properties of new catenated electron
donor-acceptor materials with magnesium and
free base porphyrins as donors and C60 as the
acceptor. Nanoscale 7:1145–1160
28. Megiatto JD Jr et al (2011) Optimizing reac-
tion conditions for synthesis of electron donor-
[60]fullerene interlocked multiring systems. J
Chem Mat 21:1544–1550
 Chapter 20

Biophotovoltaics: Design and Study of Bioelectrochemical

Systems for Biotechnological Applications and Metabolic
Investigation
Stephen J. L. Rowden, Paolo Bombelli, and Christopher J. Howe

Abstract
Biophotovoltaic methods rely on the fact that photosynthetic microorganisms, like many others, can export
small amounts of electric current. For photosynthetic organisms, this current usually increases on illumina-
tion. This “exoelectrogenic” property may be of biotechnological interest, and may also provide useful
experimental insights into the physiological status of the cell. We describe how to construct biophotovoltaic
devices, and the kinds of measurements that are typically made.

Key words Biophotovoltaics, Cyanobacteria, Photosynthesis, Energy, Biofilm

1 Introduction

Biophotovoltaic (BPV) devices are an emerging renewable electro-

chemical technology that utilizes photosynthetic organisms to con-
vert light and stored organic matter into electricity or pure
hydrogen. These BPV devices typically use exoelectrogenically
active (i.e., able to transfer electrons to the outside of the cell)
photosynthetic microbes such as cyanobacteria and eukaryotic
algae that can form biofilms directly on electrode surfaces, and
promote electron transfer to the electrode without the requirement
for catalysts or artificial redox mediators. However, because of low
power outputs, electron mediators such as potassium ferricyanide
are sometimes added to enhance electron transfer to the anode
allowing cells to be grown in suspension (planktonic mode). BPV
technology is a modification of the microbial fuel cell (MFC),
which typically uses exoelectrogenically active non-photosynthetic
microbes, such as Shewanella oneidensis and Geobacter sulfurredu-
cens, growing heterotrophically [1]. BPVs are also referred to as
photosynthetic microbial fuel cells (PMFCs) [2, 3] in the literature.
There are also systems that use isolated photosystem preparations

Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4_20, © Springer Science+Business Media, LLC, part of Springer Nature 2018

335
336 Stephen J. L. Rowden et al.

rather than intact cells [e.g., 4] but they will not be

considered here.
Both BPVs and MFCs can be regarded as a biological form of
the chemical fuel cells (FC) that can be used to provide electricity
from sources such as hydrogen, for example for powering vehicles.
BPVs use photosynthetic organisms as a biocatalyst instead of, for
example, the platinum used in FCs. Similar to FCs, BPVs produce
electricity by spatially separating reduction and oxidation half-
reactions. BPVs consist of anodic and cathodic chambers, typically
separated by a proton exchange membrane (PEM). However, if
cells are grown as a biofilm on the anode, without access to the
cathode, separation of the two compartments by a PEM may be
unnecessary. A fraction of the electrons that are usually transported
down the photosynthetic electron chain are accepted by the anode
in the device [2, 3, 5, 6]. Protons travel to the cathodic chamber
through the PEM. Electrons can be transferred from the biocatalyst
to the anode through soluble electron shuttles added to the system
[6], and potentially by long-range electron transport mediated by
conductive bacterial nanowires (electrically conductive appen-
dages) [7, 8]. Electrons then travel to the cathode electrode
through an external circuit. At the cathode, protons, electrons,
and molecular oxygen recombine to form water. The potential
difference between the two half-reactions drives a current from
anode to cathode through the external circuit. Exclusion of oxygen
from the cathode can be used to generate hydrogen in that com-
partment, depending on the configuration of the device [9, 10].
Light-dependent power generation in both artificial and ambient
light has been demonstrated [2, 3, 5, 6]. The maximum power
density reported so far was above 100 mW m
2, which was
achieved in cyanobacterial BPV devices [11, 12]. Single strains
and mixed populations of cyanobacteria and eukaryotic algae have
been utilized in BPVs for lab-scale energy production [13–17].
Although illumination leads to increased power generation,
significant amounts of power are also produced in the dark. In
this case, the devices are probably functioning like a conventional
microbial fuel cell, with the oxidation of organic compounds—in
this case stored metabolites—being the source of the electrons. The
devices are of interest for power generation, especially in situations
where there is no local power supply, small amounts of power are
required, and bringing in batteries or silicon photovoltaic devices is
considered undesirable in view of potential environmental impacts.
However, detailed analysis of the output from devices may also
provide insights into the biology of the underlying electron transfer
reactions [6, 18].
BPV technology can be used for both basic and applied
research. Exoelectrogenically active photosynthetic microorgan-
isms have many possible biotechnology applications due to their
ability to transfer electrons outside of the cell to insoluble electron
 Biophotovoltaics: Methods and Protocols 337

acceptors such as metal oxides and generate small amounts of

power, for example to run biosensors. Algal and cyanobacterial
strains differ widely in their exoelectrogenic activity [5, 19, 20]
and specific mutations can also affect activity [18, 21]. Screening
different strains or mutants may therefore provide insights into the
environmental or physiological role of exoelectrogenic activity or
its mechanism. Alternatively, artificially modulating the loss of elec-
trons (by directly holding the device at a particular potential, or by
altering the external resistance) can be used to study the effects on
the cell of electron loss.

2 Materials

2.1 Materials for BPV The materials used for constructing BPV are outlined with the
Construction description of construction below.

2.2 Stock Solutions The BPV can in principle be operated with any cyanobacterial or
Required for BG-11 microalgal strain. Many studies use the model cyanobacterium
Culture Medium Synechocystis sp. PCC 6803, which can be readily cultured in
BG-11 medium [22].
1. 100 BG-11: 1.76 M NaNO3, 0.030 M MgSO4.7H2O,
0.0324 M CaCl2, 0.0031 M citric acid, and 1.12 mL 0.25 M
Na2EDTA solution (adjust to pH 8.0 using sodium hydroxide
before mixing with other components of 100 BG-11).
2. Trace elements: 0.046 M H3BO3, 0.0091 M MnCl2.4H2O,
0.000765 M ZnSO4.7H2O, 0.00161 M Na2MoO4.2H2O,
0.0003204 M CuSO4.5H2O, 0.000172 M Co(NO3)2.6H2O.
3. Iron stock: 0.002449 M Ferric citrate or 0.002290 M ferric
ammonium citrate.
4. Phosphate stock: 0.1751 M K2HPO4 (filter sterilize).
5. 0.1887 M Na2CO3 (filter sterilize).
6. Glucose stock: 0.5 M Glucose.
7. TES buffer: 22 g TES in 100 mL, adjust to pH 8.2 using
sodium hydroxide.
8. 1 M NaHCO3 (autoclave).

2.3 BG-11 Culture 1. For 1 L of BG-11 add: 10 mL 100 BG-11, 1 mL trace

Medium elements, 1 mL iron stock, 10 mL TES buffer and fill to
988 mL with distilled water and then autoclave.
2. Under sterile conditions, add 1 mL phosphate stock, 1 mL
Na2CO3 stock, and 10 mL NaHCO3 stock (all previously
autoclaved separately) to the solution described in item 1.
These stocks are autoclaved separately and added after cooling
to minimize precipitation in the final solution.
338 Stephen J. L. Rowden et al.

2.4 BG-11 Agar 1. Prepare 2 concentrated BG-11 þ TES and 3 g sodium thio-
Plates and Slopes sulfate, mix together, and autoclave.
2. Prepare appropriate volume of agar (3% w/v in distilled water),
autoclave, and then cool to 50  C.
3. Mix equal volume of 2 BG-11 with agar.
4. Pour into plates (about 30 mL per 90 mm Petri dish).
5. Pour into autoclaved boiling tubes for slopes.
6. Dry plates and slopes overnight at 30  C, and then store sealed
with paraffin film
7. Puncture three air holes in paraffin film using tweezers when
growing cells.

2.5 Long-Term 1. Spin down 40 mL dense cell culture in a sterile centrifuge tube
Storage (1610  g, 5 min in a standard benchtop centrifuge).
of Cyanobacteria 2. Resuspend pellet in 2 mL sterile BG-11 and add 0.5 mL sterile
glycerol and mix.
3. Pipette 200 μL aliquots into sterile microfuge tubes, and then
plunge into liquid nitrogen to freeze and store at
80  C.
4. To grow cells again, thaw out one tube, pour onto agar plate or
slope, spread, and leave to grow at 10 μmol photons m
2 s
1.

3 Methods

3.1 Design of an We describe here (Fig. 1) the construction of a typical single-

Example of a chamber device, using a biofilm of cyanobacterial cells on the
Mediator-Less, Single- anode (see Note 1). This is based on the devices described by
Chamber McCormick et al. [5]. Of course, any number of variations on the
Biophotovoltaic Device size of the device is possible.
1. Coat a 50  50 mm cathode with gold. Deposit the layer of
gold on the cathode surface by vacuum deposition coating.
When this process is not available in-house, commercial facil-
ities are available.
2. Coat the gold-coated cathode with platinum.
3. Place the cathode parallel with a 50  50 mm anode (10 mm
apart) (see Note 2) composed of indium tin oxide-coated poly-
ethylene terephthalate (ITO-PET) (see Notes 3 and 4) in a
clear acrylic glass chamber, sealed with polydimethylsiloxane
(PDMS), and then fill with BG-11 medium.
4. Connect the anode to the external circuit through a
10  20 mm strip of hydrophobic carbon material.
 Biophotovoltaics: Methods and Protocols 339

Fig. 1 Construction of a biophotovoltaic device used to extract current from a biofilm culture attached to an
ITO-coated PET anode. The anodic biofilm was connected to a Pt-coated cathode in a sandwich-like horizontal
design filled with BG-11 medium solution. Schematic view of the BPV device (a), a photograph of the actual
BPV device (b), and the components forming the BPV device (c) (figure reproduced by permission of The Royal
Society of Chemistry [5])

5. Solder a copper wire to the gold-platinum glass cathode that

protrudes outside the BPV device, to allow connection to the
external circuit (see Note 5).
6. Construct the base and cover of the device using two
50  50 mm clear acrylic glass blocks.

3.2 Culturing 1. Streak out glycerol stocks, that have been kept in
80  C, of
of Cyanobacteria Prior cyanobacteria onto BG-11 agar Petri dishes (15 g/L), and
to Inoculation place in a 30  C incubator and expose the cells to constant
on Biophotovoltaic light of 10 μmol photons m
2 s
1.
Device 2. Leave these Petri dishes in the incubator for around 1 week,
then take out, and leave on lab bench indefinitely.
3. Cyanobacteria should be streaked onto new Petri dishes every
2–4 weeks. If there is any sign of contamination, dispose of
Petri dish and start from previous stock (glycerol or Petri dish)
(see Note 6).
4. Set up 50–100 mL cultures using a small amount of cyanobac-
terial cells from the Petri dish and BG-11 medium [22] supple-
mented with NaHCO3 to a final concentration of 10 mM, and
place in a shaking incubator (160 rpm) at 30  C and expose to
340 Stephen J. L. Rowden et al.

constant light with an irradiance of 50–70 μmol photons

m
2 s
1 (see Note 7).
5. Grow the cyanobacterial cells in the incubator until they
reach mid-log phase as measured by absorption spectroscopy.
These cells are ready to be inoculated into the BPV device
(see Note 8).

3.3 Characterization 1. Concentrate a planktonic culture containing 150 nmol chloro-

of Biofilm Growth phyll, as measured by absorption spectroscopy [23], of an
on Anode Material appropriate microorganism such as the cyanobacterium Syne-
chocystis sp. PCC 6803 (cultured as described above) using
centrifugation (906  g, 5 min in a benchtop centrifuge) and
then resuspend in 5 mL of fresh medium, for example BG-11.
2. To avoid contamination, autoclave the anode material.
3. In a flow hood, using sterile technique, pipette cells onto
50  50 mm of anode material and leave to settle for 30 min
before submerging in fresh medium as carefully as possible in
order to minimize disturbance to the newly forming biofilm.
4. These biofilms should then be grown for at least a week and can
then be inserted into the BPV. For ideal growth, grow in the
same culturing conditions given above (see Subheading 3.2).
5. If it is necessary to measure the amount of cellular material in
the biofilm at the end of the experiment, then the cells can be
scraped from the electrode and resuspended in a known volume
of fresh medium (e.g., 5 mL).
6. Cell density can then be determined by spectroscopic determi-
nation of chlorophyll content [23] or by direct cell counting.

3.4 Operation of a 1. Autoclave the BPV first, when it is desirable to run with as little
Mediator-Less, Single- contamination as possible.
Chamber BPV 2. Use alligator clamps and copper wire to connect to the anode
and cathode and complete the external circuit as shown in
Fig. 3.
3. Fill the chambers with fresh BG-11 medium and check to
ensure that the temperature (ideally 30  C) (see Note 9)
remains constant by either monitoring inside the BPV using a
sensor probe or outside the device using a thermometer.
4. The anode with biofilm from Subheading 3.3 should be care-
fully submerged into the medium in the chamber.
5. The BPV can be kept either with the anode horizontal and lit
from the top or vertical and lit from the side using LEDs of
known wavelength (see Note 10). The brightness can vary
significantly based on the experiment (see Note 11).
 Biophotovoltaics: Methods and Protocols 341

6. To allow for different light intensities to be used in a single

experiment vary the brightness of the LEDs in the experiment
by, for example, varying the current going into the LED (being
careful not to blow the LED), using dimmable LEDs, or
varying the distance between the biofilm and the LED.
7. To allow for different wavelengths of light, use multicolor
LEDs, switch the color of the LED bulb, or simply use optical
filters.
8. Use a digital multimeter to measure the current from the
operational BPV.

3.5 Techniques One of the most frequently used analytical methods is the polariza-
of Investigation: tion curve. This is a plot of current density against electrode poten-
Polarization and Power tial (i.e., the potential difference between anode and cathode), and
Curves it allows the maximum power output to be determined (Fig. 2).
The curve can be constructed by altering the resistance of the
external circuit in a number of steps by sequentially adding resistors
to the circuit. Increasing the resistance will cause a decrease in the
current, according to Ohm’s law V ¼ IR. Each time the resistance is
altered, the potential should be allowed to reach a steady state and
is then recorded. The “open circuit potential,” when there is no
current flowing, should also be measured. Power (W) can be
derived from the potential (V) and current (I) according to
W ¼ I  V. In this way power curves can then be derived by plotting
the power output against the current density. Power curves are
typically an inverted U-shape, and the peak power output can
thus be determined (Fig. 2). The peak power is usually observed
with an external resistance equal to the internal resistance of the
device.

Fig. 2 Power outputs of BPV devices with the cyanobacterium Synechococcus sp. WH 5701. Polarization
curves (a) and power curves (b) quantified under dark (filled diamonds) and 10 W m
2 light (open squares)
conditions are shown. Filled triangles show data without a biofilm under light condition (figure reproduced by
permission of The Royal Society of Chemistry [5])
342 Stephen J. L. Rowden et al.

3.6 Techniques These are measurements in which the external current (chronoam-
of Investigation: perometry) or potential (chronovoltammetry) is monitored after a
Chronoamperometry change, for example in illumination, as a function of time. Most
and Chrono- commonly chronoamperometry or chronovoltammetry is per-
voltammetry formed with a two-electrode system (anode/cathode). The
two-electrode system can be operated by a conventional multi-
meter. Chronoamperometry and chronovoltammetry measure-
ments may provide insights into the mechanism of electron
export, and might also be useful as an experimental tool to monitor
the state of the electron transfer chain of organisms under study (see
Notes 12 and 13). The protocol to perform a two-electrode system
(anode/cathode) chronovoltammetry is shown in Fig. 3 and
includes four steps.
1. Prepare the BPV as described in Subheading 3.1.
2. Use alligator clamps and copper wire to connect to the anode
and cathode of the BPV device with the red and black terminals
of a voltmeter.
3. Clamp an external resistor between the terminals of the alliga-
tor clamps.
4. Select mV (millivolt) on the voltmeter to measure the potential
from the operational BPV and record the data over time.
5. Plot the recorded potential difference between anode and
cathode against time to create a graph as shown in Fig. 4.

Fig. 3 Material (a) and diagram (b) required for performing a chronovoltammetry measurement with a
two-electrode system (anode and cathode only)
 Biophotovoltaics: Methods and Protocols 343

Fig. 4 Example of chronovoltammetry performed with BPV over ca. 1000 s. The
graph shows four cycles of dark/light. The grey and white backgrounds denote
the dark and light phases, respectively

4 Notes

1. In general, mediator-less systems are considered more desirable

than those requiring addition of an exogenous mediator such
as potassium ferricyanide, a compound that can facilitate elec-
tron transport from the organism to the anode. The key part of
designing a mediator-less BPV is to make a chamber for the
cyanobacterial cells whereby they are in continuous contact
with the anode and can form a biofilm. The easiest way to
achieve this is to have the anode at the bottom of the chamber,
or grow the biofilm on the anode prior to insertion into the
BPV. Concentrated planktonic cells should be pipetted directly
onto the anode in a small quantity of medium.
2. Ideally, all the cyanobacterial cells in the biofilm should experi-
ence the same electrochemical conditions, especially for experi-
ments studying environmental effects on electron output. The
geometry of the chamber should therefore be such that the
anode and cathode are parallel to each other. The distance
between anode and cathode affects the internal resistance of
the device, and in turn the output. Ideally the device should be
designed so that the distance between anode and cathode can
be adjusted, for example to minimize the internal resistance.
3. In addition to the usual requirements of the anode, the material
of the anode may affect the ability of the cyanobacterial cells to
adhere to the anode, which affects both the structure of the
biofilm and the electrical output [24].
4. ITO-PET is widely used as an anode material as it is conductive
and largely transparent, but other electrode materials are
possible [24].
344 Stephen J. L. Rowden et al.

5. As an alternative to the two-electrode system (i.e., anode and

cathode), some studies have also employed a reference elec-
trode in the anode compartment. A reference electrode permits
determination of the absolute potential of the anodic half-cell.
Without a reference electrode, the potential of the anode is
determined versus the potential of the cathode. This absolute
measurement of anodic potential allows information to be
gained about the redox species operating in the anodic cham-
ber [25] and allows variation in the anode potential to be
determined independently of any change in the cathode.
6. Cells will not survive being repeatedly frozen and thawed. It is
advisable to make multiple aliquots of each strain, and to keep
the strains alive on plates, to avoid repeated freeze–thaw cycles.
7. It is a good idea to set up cultures from the Petri dish rather
than diluting a previously grown culture of cyanobacteria,
because in our experience it leads to less contamination and
increased reproducibility.
8. If subsequent isolation of RNA or DNA is to be carried out, the
best results are usually obtained by inoculating with cells at
mid-log phase.
9. BPV systems can suffer from evaporation or condensation.
Increased temperature can increase these problems. Therefore
it is often desirable to grow the biofilm in the device at room
temperature. If evaporation does occur, top up to previous level
with autoclaved water. Do not use BG-11, as this would
increase the concentration of the components of the medium.
10. The lighting system is required to provide the biofilm with
light. Temperature, light intensity, and wavelength all affect
the electrical output of the device. It is therefore important to
ensure that the light system does not increase the temperature
of the device, and that light intensity and wavelength are
known and controllable.
11. To mimic the solar irradiation reaching the earth’s surface at
ground level, the BPV devices can be exposed to light photon
fluxes within the range 0–2000 μmol photons m
2 s
1
[26]. The brightness can vary significantly based on the cyano-
bacterial strain and cell density. Experimental tests are required
to define the optimal condition.
12. An environmental change may cause an initial fluctuation in the
output from a BPV. Although the profile of the change may be
informative (see, e.g., chronoamperometry above), for most
purposes it is best to use measurements of current or power
output once a steady state has been achieved.
 Biophotovoltaics: Methods and Protocols 345

13. Because the currents and potential differences measured are

typically very small, it is important to avoid external electrical
interference as much as possible.

Acknowledgments

CJH and PB thank the Leverhulme Trust for financial support.

SJLR thanks the European Commission (EU KBBE.2013.3.2-02
programme, D-Factory: 368 613870) for financial support. We
thank Mr. Pavel Artemov for designing Fig. 3.

References
1. Logan BE, Hamelers B, Rozendal R,
Schroeder U, Keller J, Freguia S, Aelterman P,
Verstraete W, Rabaey K (2006) Microbial fuel
cells: methodology and technology. Environ
Sci Technol 40:5181–5192
2. Zou Y, Pisciotta J, Billmyre RB, Baskakov IV
(2009) Photosynthetic microbial fuel cells with
positive light response. Biotechnol Bioeng
104:939–946
3. Pisciotta JM, Zou Y, Baskakov IV (2011) Role
of the photosynthetic electron transfer chain in
electrogenic activity of cyanobacteria. Appl
Microbiol Biotechnol 91:377–385
4. Sokol KP, Mersch D, Hartmann V, Zhang JZ,
Nowaczyk MM, Rogner M, Ruff A,
Schuhmann W, Plumere N, Reisner E (2016)
Rational wiring of photosystem II hierarchical
indium tin oxide electrodes using redox poly-
mers. Energy Environ Sci 9:3698–3709
5. McCormick AJ, Bombelli P, Scott AM, Philips
AJ, Smith AG, Fisher AC, Howe CJ (2011)
Photosynthetic biofilms in pure culture harness
solar energy in a mediatorless bio-photovoltaic
cell (BPV) system. Energy Environ Sci
4:4699–4709
6. Bombelli P, Bradley RW, Scott AM, Philips AJ,
McCormick AJ, Cruz SM, Anderson A,
Yunus K, Bendall DS, Cameron PJ, Davies
JM, Smith AG, Howe CJ, Fisher AC (2011)
Quantitative analysis of the factors limiting
solar power transduction by Synechocystis
sp. PCC 6803 in biological photovoltaic
devices. Energy Environ Sci 4:4690–4698
7. Gorby YA, Yanina S, McLean JS, Rosso KM,
Moyles D, Dohnalkova A, Beveridge TJ,
Chang I, Hong Kim B, Shik Kim K, Culley
DE, Reed SB, Romine MF, Saffarini DA, Hill
EA, Shi L, Elias DA, Kennedy DW, Pinchuk G,
Watanabe K, Ishii S, Logan BE, Nealson KH,
Fredrickson JK (2006) Electrically conductive
bacterial nanowires produced by Shewanella
oneidensis strain MR-1 and other microorgan-
isms. PNAS 103(30):11358–11363
8. Sure S, Ackland ML, Torriero AAJ,
Adholeya A, Kochar M (2016) Microbial nano-
wires: an electrifying tale. Microbiology
162:2017–2028
9. McCormick AJ, Bombelli P, Lea-Smith DJ,
Bradley RW, Scott AM, Fisher AC, Smith AG,
Howe CJ (2013) Hydrogen production
through oxygenic photosynthesis using the
cyanobacterium Synechocystis sp. PCC 6803 in
a bio-photoelectrolysis cell (BPE) system.
Energy Environ Sci 6:2682–2690
10. Pinhassi RI, Kallmann D, Saper G, Dotan H,
Linkov A, Kay A, Liveanu V, Schuster G,
Adir N, Rothschild A (2016) Hybrid bio--
photo-electro-chemical cells for solar water
splitting. Nat Commun 7:12552
11. Bombelli P, Muller T, Herling TW, Howe CJ,
Knowles TPJ (2015) A high power-density,
mediator-free, microfluidic biophotovoltaic
device for cyanobacterial cells. Adv Energy
Mater 5:1410229
12. McCormick AJ, Bombelli P, Bradley RW,
Thorne R, Wenzel T, Howe CJ (2015) Bio-
photovoltaics: oxygenic photosynthetic organ-
isms in the world of bioelectrochemical
systems. Energy Environ Sci 8:1092–1109
13. Anderson A, Laohavisit A, Blaby IK, Howe CJ,
Merchant SS, Davies JM, Smith AG (2015)
Exploiting algal NADPH oxidase for biopho-
tovoltaic energy. Plant Biotechnol J 14:22–28
14. Laohavisit A, Anderson A, Bombelli P,
Jacobs M, Howe CJ, Davies JM, Smith AG
(2015) Enhancing plasma membrane
NADPH oxidase activity increases current out-
put by diatoms in biophotovoltaic devices.
Algal Res 12:91–98
15. Darus L, Ledezma P, Keller J, Freguia S (2016)
Marine phototrophic consortia transfer
346 Stephen J. L. Rowden et al.
electrons to electrodes in response to reductive
stress. Photosynth Res 127:347–354
16. Wei X, Lee H, Choi S (2016) Biopower gener-
ation in a microfluidic bio-solar panel. Sensors
Actuators B 228:151–155
17. Wu Y, Xiao Y, Wang Z, Zhao F (2016) Perfor-
mance of bioelectrochemical systems inocu-
lated with Desmodesmus sp. A8 under different
light sources. Biorem J 20:233–239
18. Bradley RW, Bombelli P, Lea-Smith DJ, Howe
CJ (2013) Terminal oxidase mutants of the
cyanobacterium Synechocystis sp. PCC 6803
show increased electrogenic activity in
biological photo-voltaic systems. Phys Chem
Chem Phys 15:13611–13618
19. Ng FL, Phang SM, Periasamy V, Yunus K,
Fisher AC (2014) Evaluation of algal biofilms
on Indium Tin Oxide (ITO) for use in biopho-
tovoltaic platforms based on photosynthetic
performance. PLoS One 9:e97643
20. Hasan K, Grippo V, Sperling E, Packer MA,
Leech D, Gorton L (2017) Evaluation of pho-
tocurrent generation from different photosyn-
thetic organisms. ChemElectroChem
4:412–417
21. Cereda A, Hitchcock A, Symes MD, Cronin L,
Bibby TS, Jones AK (2014) A bioelectrochem-
ical approach to characterize extracellular
electron transfer by Synechocystis sp PCC6803.
PLoS One 9:e91484
22. Rippka R, Deruelles J, Waterbury JB,
Herdman M, Stanier RY (1979) Generic assign-
ments, strain histories and properties of pure cul-
tures of cyanobacteria. J Gen Microbiol 111:1–61
23. Porra RJ, Thompson WA, Kriedemann PE
(1989) Determination of accurate extinction
coefficients and simultaneous equations for
assaying chlorophylls a and b extracted with
four different solvents: verification of the con-
centration of chlorophyll standards by atomic
absorption spectroscopy. Biochim Biophys
Acta 975:384–394
24. Bombelli P, Zarrouati M, Thorne RJ,
Schneider K, Rowden SJL, Ali A, Yunus K,
Cameron PJ, Fisher AC, Wilson DI, Howe
CJ, McCormick AJ (2012) Surface morphol-
ogy and surface energy of anode materials
influence power outputs in a multi-channel
mediatorless bio-photovoltaic (BPV) system.
Phys Chem Chem Phys 14:12221–12229
25. Bard AJ, Faulkner LR (2000) Electrochemical
methods: fundamentals and applications, 2nd
edn. Wiley, ISBN: 0-471-04372-9
26. Hall DO, Rao KK (1999) Photosynthesis, 6th
edn. Cambridge University Press, Cambridge.
ISBN: 0-521-64497-6
 INDEX
A Ap ...........................................................28, 29, 34, 35, 43
Absorbance ..........................................241, 244, 245, 248
Absorption..............................................4, 7, 11, 16, 125,
130, 161, 162, 200, 308, 326–329, 333, 340
Ac ....................................... 29, 34–36, 69, 70, 72, 75–79
Acclimation..................................... 7, 32, 63, 90, 91, 176
Acetosyringone..................................................... 266, 275
Actinic illumination.........................................15, 92, 125,
126, 128–130, 136, 143
Adenine dinucleotide phosphate (ADP).....................240,
241, 243, 248
Agrobacterium-mediated transient
transformation ..........264, 265, 268, 271, 278
arcing ....................................................................... 276
binary vector.......................................... 264, 265, 274
calculations
volume of GOI (VGOI) culture ............... 268, 278
volume of P19 (V19) culture ........................... 264,
265, 268, 271, 277, 278
electroporation ..............................265, 267, 276, 277
electroporation cuvette ......................... 265, 267, 277
glycerol stocks ....................................... 265, 268, 277
GV3101(pMP90) .......................................... 265, 274
infiltration ...................................... 268–271, 277–279
infiltration solution ............................... 266, 269, 278
LB ......................................... 265, 267, 268, 275, 277
LB-agar-RK plates .......................................... 265, 268
LB-RK.................................................... 265, 268, 277
leaf selection ................................................... 269, 278
negative controls ............................................ 279, 282
nleaf ......................................................... 268, 269, 278
OD600 ............................................................. 268, 277
optimal expression time ................................. 270, 279
V0.3ODP19 ........................................................ 268, 277
V0.5ODGOI ....................................................... 268, 278
Vleaf ......................................................... 268, 269, 278
Agroinfiltration.................. 264, 266, 269–271, 278–280
Aj ......................................................................... 29, 34, 35
Algae ................................................................10, 13, 108,
197, 202, 216, 276, 335–337
Allan variance ....................................................... 164, 189
Amax ............................................................. 26, 31, 38, 41
Anet ....................................................................... 7, 26–29,
31–33, 36, 41–43, 110, 112, 113
Aniline blue .........................................254, 255, 257–260
Sarah Covshoff (ed.), Photosynthesis: Methods and Protocols, Methods in Molecular Biology, vol. 1770,
https://doi.org/10.1007/978-1-4939-7786-4, © Springer Science+Business Media, LLC, part of Springer Nature 2018
347
Apportioning........................................................ 161, 176
Argon gas..................................................... 325, 327, 328
Artificial photosynthesis.............................. 319, 320, 328
Asat ...................................................................... 26, 31, 38
Autofluorescence........................ 273–275, 279, 281, 282
Azide functionalities............................................. 322, 331
B
Bathophenanthrolinedisulfonic acid ............................ 323
BBY membranes ................................................... 202, 208
Beer’s absorbance law ................................................... 161
Biophotovoltaic (BPV) device
BG-11
agar plates ................................................. 338, 339
culture medium .........................................337–340
slopes.................................................................. 338
biofilm.......................... 335, 336, 338–341, 343, 344
characterization of biofilm growth......................... 340
chronoamperometry ...................................... 342, 344
chronovoltammetry ....................................... 342, 343
culturing of cyanobacteria .................... 339, 340, 344
description .....................................335, 336, 339, 343
design....................................................................... 343
operation................................................ 340, 341, 344
polarization curve.................................................... 341
power curve ............................................................. 341
Bundle-sheath cells ............................................. 156, 157,
168, 170, 180, 181, 183, 184, 187, 254, 255,
285, 286
C
C3 ................................................................ 10, 11, 16, 32,
86, 87, 90, 92, 142, 158–161, 164–166,
168–173, 179, 183–185, 189, 190, 215, 254,
255, 285
C4 ....................................................................... 11, 14–16,
86, 87, 90, 92, 142, 159–161, 164, 166,
168–171, 173, 179, 180, 182–185, 189, 190,
254, 255, 285, 286
Calibration.........................................................30, 36, 38,
56, 64, 127, 136, 145, 146, 162–165, 189,
203, 206, 208, 209, 220, 221, 224, 226
Callose .................................................254, 255, 257, 258
Calvin-Benson-Bassham cycle ...................................... 215
 PHOTOSYNTHESIS: METHODS AND PROTOCOLS
348 Index
Carbonic anhydrase (CA) ...................156, 168, 172, 184
Carboxylation efficiency (k).......................................... 168
Cavity-ring down spectroscope (CRDS) ..................... 162
Cellulase R-10 ...................................................... 266, 276
Chlorophyll fluorescence .................................... 4, 11–13,
18, 85, 95–103, 105, 108–118, 121, 123, 125,
127–135, 137, 161, 257
A
Fm0 .........................................................110, 113–119
dark-light transition ................................................ 123
E
Fm0 ................................................ 108, 110, 113–119
F0 ........................................................ 87, 89, 125, 128
fluorescence yield (ΦF).................. 105, 107–118, 123
Fm ......................................87, 92, 107, 119, 125, 130
Fm0 ...............................................87, 89, 92, 105–109,
111–113, 115–119, 125, 126, 131, 135
Fo ................................................................87, 92, 125,
126, 128, 130, 135, 137
Fo0 ..............................................................93, 126, 128
Fq0 .................................................................... 125, 131
Fq0 /Fm (see ϕPSII or ΔF/Fm0 )
Fq0 /Fv0 ...................................................................... 125
fractionPSII ........................................................ 93, 125
Fs .................................. 105, 106, 109, 110, 116, 117
Fv ....................................................................... 87, 125
imaging .................................................................... 121
filters .................................................127, 128, 257
induction curve ................................132, 134, 137
light sources.............................................. 123, 127
map image ................................................ 129, 138
phenotying............................................ 13, 18, 121
relaxation curve ........................................ 131, 132
spatial heterogeneity ..........................13, 121, 133
maximum quantum efficiency of PSII,
Fv/Fm ........................................................... 125
measurement type
non-modulated.............................................. 95–97
pulse amplitude modulated (PAM)..................... 4,
12, 85, 95–97, 105, 109–112, 127, 130, 135
photosynthetic efficiency (Fq0 /Fm0 ) ..... 130, 131, 137
quenching ............................................. 12, 13, 17, 92,
93, 103, 106, 108, 123, 124, 126
relaxation kinetics............................................. 13, 133
saturation flash (Q0 ) ...................................... 105–108,
111, 113, 117–119
sub-saturating multiphase flash (MPF)
irradiances .................................................... 108
hysteresis ............................................................ 115
integrated intensity (II) of the modulated
light .............................................................. 112
light sources.............................................. 109, 111
MPF irradiance dynamics ....... 109–111, 113–118
percent differences (%D)................................... 111
phase 1 ............................ 108, 110, 111, 113–118
phase 2 ............................ 108, 110, 111, 113–115
phase 3 ..............................................108, 114–116
ramp rate...........................................110, 114, 115
reciprocal of flash irradiance (Q0 –1)......... 114, 115
(see also Reciprocal plot)
red modulated LEDs ........................................ 109
time interval....................................................... 115
timescale
fast .......................................................... 11, 13, 95,
96, 98, 99, 101
slow ...................................................................... 13
traditional saturation flash methodology..... 107–109,
113, 117–119
Chloroplast ................................................... 5, 10, 11, 18,
28, 43, 97, 107, 112, 121, 133, 156–158, 168,
170, 172, 173, 176, 179, 197, 202, 229,
254–256, 272, 279, 305
Clark-type O2 electrode (CTOE) ...............................141,
197–200, 202, 204, 206
“Click” reaction ......................... 321–324, 327, 328, 332
catalyst preparation ........................................ 323, 327
Closed gas exchange system ................................ 152, 153
CO2 compensation point ..................................... 26, 145,
147–152, 159
CO2 compensation point in the absence of
mitochondrial respiration, (Γ*).................... 26,
159, 187
CO2 concentration in the intercellular
space (Ci) ...................................26–29, 32–34,
36, 38, 39, 41–43, 91, 163, 177
CO2 concentration surrounding the
leaf (Ca) ............................................. 26, 27, 36
Conductance
boundary layer conductance to CO2 (gbl).... 157, 190
bundle-sheath conductance to CO2 (gbs)..... 157, 183
chloroplast conductance to CO2 (gc) .................... 157
conductance to diffusion of CO2
in air (gac) ............................................ 157, 166
mesophyll conductance to CO2 (gm)...................... 26,
28, 37, 157, 168–170, 176–181,
183, 184, 190
Coomassie............................................216, 219–221, 223
CO2 partial pressure
in the ambient air (Ca) ............................................ 156
in the bundle-sheath cells (Cbs) ................... 156, 181,
183, 186
in the chloroplast (Cc) .................................. 107, 156,
168–170, 172, 176, 177, 179
in the leaf intercellular spaces (Ci) .......................... 27,
90, 91, 163, 170, 177
at the leaf surface (Cs)............................................. 156
in the mesophyll (Cm)................................... 156, 169,
172, 180, 181, 183
at the sites of CA (CCA).......................................... 156
Copper(I)iodide .......................................... 323, 324, 327

Cuvette (leaf chamber) ......................................... 6, 8, 30,
31, 33, 34, 36, 38, 39, 46, 57, 66, 90, 100,
109, 112, 141–145, 147–150, 158, 163, 189
Cyanobacteria ...................................................... 197, 202,
305, 335–341, 343, 344
Cycloaddition of azides and alkynes (CuAAC/“click”)
reaction
catalyst preparation ................................................. 332
reaction between the artificial photosynthetic
model 8 and the IrOx-NPs-N3 2 ............... 332
Cytosol................................................................. 156, 170,
179, 264, 285, 286
D
Dark-adapted.....................................................13, 14, 16,
87, 88, 92, 93, 97–100, 103, 107, 119,
123–126, 128–130, 132, 134, 135, 137, 207
3D clearing ..........................................286–288, 297, 298
Dialysis ..........................................................325–328, 333
Discrimination
carbon isotope discrimination (Δ13C) ......... 158–161,
170, 172, 184, 186
comprehensive model for 13C photosynthetic
discrimination in C3 species (Δ3-com)........ 158,
167–170, 175, 176, 178
comprehensive model for 13C photosynthetic
discrimination in C3 species
(Δ4-com) ...................................... 180, 181, 190
discrimination associated with diffusion
of CO2 through the boundary layer
and stomata (Δgs) ............................... 158, 176
discrimination associated with photorespiration
(Δf) ............................................. 158, 175, 176
discrimination associated with respiration
(Δe)............................................. 158, 175, 176
discrimination associated with Rubisco
(Δb) ..................................................... 158, 175
discrimination associated with the diffusion
of CO2 from the intercellular airspaces to
the chloroplast (Δgm) ................................. 158,
175, 176, 178, 179
discrimination that would occur if Cc ¼ Ci in
the absence of any respiratory fractionation
(Δi) ..................................... 158, 176, 178–180
observed 13C photosynthetic discrimination
(Δobs)......................................... 158, 160, 161,
165, 166, 178, 180, 188
oxygen isotope discrimination (Δ18O) ..............4, 172
simplified model for 13C photosynthetic
discrimination in C3 species
(Δ3-sim) ....................................... 159, 170, 171
simplified model for 13C photosynthetic
discrimination in C4 species
(Δ4-sim) ....................................... 159, 170, 171
PHOTOSYNTHESIS: METHODS AND PROTOCOLS
Index 349
2,3-dPGA-dependent phosphoglycerate mutase
(dPGM)............................................... 240–243
Drawdown correction .......................................... 146, 147
Dye-sensitized solar cells (DSSC) ....................... 320, 330
E
Eddy correlation.............................................................. 70
Eddy covariance ........................................................4, 6, 9
Electrochromic shift (ECS) ......................................12, 15
Electron transport rate (ETR)................................ 26, 83,
88, 89, 93, 125, 130
Enolase................................................................. 240, 241,
243, 248
F
F0 .................................................................................... 101
Farquhar, von Caemmerer, Berry (FvCB)
model .........................................................7, 27
Fatty acid methyl esters (FAME) ....................... 306, 308,
312, 313, 315
Ferricyanide ................................................................... 335
Flag leaf............................................................... 5, 98, 100
Fluorescence transients ........................... 96, 99, 101, 103
Fluorescent probes ............................................... 298, 301
green fluorescent probe (GFP) ............................. 253,
256, 270–272, 275, 282, 286
live-cell labeling fixable fluorescent probe ............ 255,
257–259
mCherry (Red)............................................... 273, 275
mGFP6 (Green) ...................................................... 275
mTurquoise2 (Cyan) ..................................... 275, 281
selecting the right fluorescent tag .......................... 274
yellow fluorescent probe (YFP)............ 253, 256, 282
Fluorometer............................................................. 85, 92,
97–99, 103, 109–112
Fractionation
catalyzed 12C/13C fractionation during CO2
hydration (h) ................................................ 157
12
C/13C fractionation as CO2 dissolves
(es) .............................................. 157, 172, 173
12
C/13C fractionation associated with the
catalyzed hydration of CO2 þ H2 O $ HCO 3
(eb).............................................. 157, 173, 174
12
C/13C fractionation during carboxylation
by PEPC (bp) ............................................... 156
12
C/13C fractionation during carboxylation by
Rubisco including respiration and
photorespiration fractionations
(b3) ............................................ 156, 169, 181,
184, 185, 189
12
C/13C fractionation during
decarboxylation (e) ............................ 157, 174,
178, 186
 PHOTOSYNTHESIS: METHODS AND PROTOCOLS
350 Index
Fractionation (cont.)
12
C/13C fractionation during decarboxylation
including the effect of a respiratory substrate
isotopically distinct from recent
photosynthate (e´) .............................. 157, 174
12
C/13C fractionation during dehydration
of HCO 3 (d )............................................... 156
12 13
C/ C fractionation during leakage of CO2 out
of the bundle-sheath cells (s) ............. 157, 172
12
C/13C fractionation during liquid phase
diffusion and dissolution of
CO2 (am) ............................................ 156, 172
12
C/13C fractionation during photorespiration
( f )......................................157, 174, 178, 186
12
C/13C fractionation for CO2 diffusion
in air (as) .................................... 156, 170, 172
12
C/13C fractionation for diffusion of CO2
in the boundary layer (ab) ................. 156, 172
12
C/13C fractionation for diffusion of CO2
through water (al) .............................. 156, 172
12
C/13C fractionation of Rubisco alone
(brub).................................................... 156, 173
combined 12C/13C fractionation by CO2
dissolution, hydration, PEP carboxylation
and respiration (b4) ........................... 156, 169,
181, 184, 185
modulation factor for e´ to account for shift in
respiratory substrate, apparent respiratory
fractionation (e*) ................................ 157, 174
net 12C/13C fractionation that occurs as CO2
is converted to HCO 3 , including
fractionations by CO2 dissolution,
hydration and PEPC activity (b0 4)............. 156,
173, 174
overall in-vivo12C/13C fractionation during
carboxylation by Rubisco and PEPC
(b30 ) .................. 156, 170, 173, 174, 177, 189
weighted 12C/13C fractionation for diffusion
across the boundary layer and stomata in
series (ā).............................................. 156, 171
G desiccant ................................................. 30, 32, 38
Gas exchange .............................................. 6, 8, 9, 29, 30,
36, 42, 46–48, 52–54, 56, 57, 59, 61–66,
69–80, 92, 145, 148, 152, 153
A/Ci curve....................................... 28, 31–38, 40–42
air leaks ................................................. 31, 33, 36, 39,
40, 52, 54, 56, 62, 74, 77, 78
air pressure..................................................... 7, 61, 72,
73, 75, 76, 78, 79
canopy photosynthesis and transpiration systems
(CAPTS)
border protecting plants ...............................71, 77
CAPTS suite software ......................75, 76, 79, 80
chamber .................................................. 70–77, 79
controller ..........................................71–76, 78, 79
fans .......................................................... 73, 74, 77
log duration cutoff................................. 75, 76, 79
plant growth ..................................................70, 71
soil respiration ..................................70, 72, 77, 79
weeds ................................................................... 72
condensation ............................................... 31, 33, 36,
39, 62, 90, 163, 188
ETH gas exchange system-2 (EGES-2)
air flow rate............................................ 48, 54, 56,
57, 59, 60, 62, 64, 86
calibration ......................................................54, 65
cell A .................................................52, 54, 56, 63
cell B .................................................52, 54, 56, 62
construction ..................................................48, 53
data processing and graphing application.......... 65
dual tower CO2 adsorber ................................... 61
ethylene-propylene-diene foam .......................... 61
gas switching unit ............................................... 48
humidification column.....................46–49, 52, 62
LabVIEW platform ................... 47, 48, 53–57, 62
lids ........................................................................ 64
mass flow controllers (MFC)...........47, 48, 52, 61
measuring time intervals ........................ 56, 64, 65
molecular scrubber...........................46–48, 54, 61
multichamber ........................................ 46, 47, 52,
53, 56–58, 62, 65, 72, 79
normalization parameters .............................65, 66
physiological experiments ................................... 57
plant growth ..................................................51, 60
polytetrafluoroethylene [PTFE]
tubing................................................ 48, 51, 61
pressure-step regulators ...................................... 52
seed sterilization ..................................... 48, 60, 61
Fick’s law ................................................................... 27
humidity .................................................................... 57
infra-red gas analyzer (IRGA) .................................. 29
calibration ............................................................ 30
CO2 mixer .....................................................29, 30
CO2 zero ............................................................. 30
drift ................................................................42, 64
H2O zero............................................................. 30
humidifying granules/ceramic stones................ 47
(see also Stuttgarter Masse)
matching ...........................................30, 42, 54, 64
soda lime...........................................29, 30, 36, 38
leaf area ........................................................ 31, 39, 40,
46, 56, 65, 67, 77, 91, 111, 113
leaf clip cuvette....................................................46, 66
light intensity............................................. 7–9, 34–36,
40–42, 57–59
mixing fans ....................................9, 62, 73, 110, 113

scale of measurement
canopy................................................... 6, 9, 69–80
whole shoot measurement.............................. 8, 46
Stuttgarter Masse ......................................... 29, 30, 47
system types
closed .......................................6, 8, 9, 70, 78, 148
open ....................................................... 6, 8, 9, 70,
92, 145, 152, 153
semi-closed ........................................................ 8, 9
temperature ............................................... 7–9, 33, 34,
36, 38, 40, 52, 56, 57, 70, 72, 75, 76, 78, 79
wait time .............................................................36, 42,
54, 62, 75, 78, 91
Glutaraldehyde aqueous solution................................. 256
H
1
H NMR analyses.......................................................... 331
Hydration rate (Vh) ..................................... 158, 184, 185
Hydroponics (Cramer’s) solution .......................... 48, 51,
52, 59, 61
Hyperspectral leaf reflectance ......................................... 18
I
ImageJ........................................................................40, 65
Immunoblotting ................................................. 216, 217,
219, 221, 222, 225, 230, 249
Immunodetection ....................................... 216, 223, 225
quantification........................ 220, 221, 223, 225, 226
Immunohistochemistry ...................................... 253, 286,
287, 291, 294, 296–298, 300, 301
In situ hybridization ............................................ 286, 290
Intrinsic H2O use efficiency ........................................... 13
Ion-exchange chromatography .................................... 230
Iridium oxide nanoparticles (IrOx-NPs) ........... 320, 322,
324–329, 331–333
iodoacetic acid ...............................322, 323, 325, 332
IrOx-CH2I-NPs 1.......................................... 321–322
characterization ................................326, 327, 330
synthesis ...................................322, 325, 326, 332
IrOx-CH2N3-NPs 2
characterization ........................................ 326, 327
synthesis .................................................... 322, 326
photocatalyst 9
characterization ................................328, 329, 333
color ................................................................... 332
synthesis ...................................323, 324, 328, 332
photosynthetic model 8................................. 322, 323
characterization ................................................. 333
description ................................................ 328, 331
experimental conditions.................................... 331
synthesis ............................................322, 331, 332
Isotope
13
C depleted ............................................................ 160
13
C enriched ................................................... 160, 174
PHOTOSYNTHESIS: METHODS AND PROTOCOLS
Index 351
C isotopic composition (δ13C)..................... 158–161,
164, 174, 179, 188
δ13C of the CO2 in the air in the leaf cuvette ....... 158
isotope fractionation (a) ................................ 155, 158
isotopic equilibrium ................................................ 172
isotopologues ............................................10, 11, 142,
147, 161, 162, 164, 199
stable carbon isotopes ........................... 8, 28, 64, 155
stable isotope measurements ........................... v, 8, 10,
141, 142, 148
stable isotope ratio mass spectrometer ........ 198–200,
203
Isotopologue ................................................................. 199
J
JIP test ..................................................... 96, 99, 100, 103
L
Lactate .................................................................. 240, 241
Lactate dehydrogenase (LDH) ........................... 240, 241
Laser.....................................................161, 162, 164, 165
Lead alloy tunable diode laser (TDL).........................162,
164, 188, 189
Leaf ......................................................254–258, 260, 300
cell suspensions .............................................. 254, 255
digestion buffer ................................256, 258–260
fixative ...............................................255–258, 260
imaging .............................................254, 255, 257
preparation ................................................255–258
discs....................................... 141, 142, 145–152, 218
extracts .................................. 240, 242, 245, 249, 306
spectroscopy ..........................................................4, 15
temperature (TLeaf) ............................................26, 30,
32, 33, 38, 40, 110
vapor pressure deficit (VPDLeaf)............................. 110
Leakiness (ϕ) ....................................................... 159–161,
180–185, 187
Ligand 10 ............................................................. 323, 324
Light-adapted .................................................89, 124, 125
Light response curves (P-I) ......................................16, 17
chlorophyll a fluorescence light-response
curves ................................................ 84, 87, 92
combined gas exchange and chlorophyll a
fluorescence light-response curves ............... 88
effective quantum yield of PSII, ΔF/Fm0 ................ 89
flow rate ..................................................................... 90
light compensation point (Icomp) ................ 84, 86, 91
light saturated rate of photosynthesis
(Pmax) ......................................... 84, 86, 90, 91
maximum quantum yield of PSII photochemistry
in the dark (Fv/Fm) ....................................... 88
measurement sequence
high to low .......................................................... 16
low to high .......................................................... 17
 PHOTOSYNTHESIS: METHODS AND PROTOCOLS
352 Index
Light response curves (P-I) (cont.)
non-rectangular hyperbola equation........................ 83
photosynthetic light response curves ....................... 83
pulse-amplitude modulated (PAM)
measurements ................................................ 85
rapid light-response curves (RLCs) ................... 85–88
steady-state light-response curves
(LRCs) .............................................. 85–88, 90
temperature .................................................. 85–87, 90
using a fluorescence imager ........................... 130, 132
wait time .................................................................... 91
Lipids .................................................................... 305–315
classes
digalactosyldiacylglycerol (DGDG) ........305–307,
310, 312, 314
monogalactosyldiacylglycerol
(MGDG).............................................305–307,
310, 314
phosphatidylglycerol (PG).......................305–307,
310, 312, 314
Sulfoquinovosyldiacylglycerol
(SQDG) .................... 305–307, 310, 312, 314
extraction
chlorophyll concentration of the lipid
extract (ChlLiEx) ................................. 308, 312
leaves ......................................................... 308, 309
thylakoid membranes .......................306, 308, 310
gas chromatography (GC)...................................... 306
analysis ...................................................... 312, 313
FAME 1 ............................................................. 308
method .............................................................. 308
nmolLipid ................................................... 308, 314
seed extract ............................................... 308, 313
standard molecular weight correction
factors (MWF) ............................................. 313
TAG ..................................................308, 313, 315
two-dimensional thin layer chromatography
(2D-TLC) .................................................... 306
copper dye solution.................................. 307, 311
densitometric analysis .............................. 310, 312
dimension 1 solvent (D1)................307, 309, 311
dimension 2 solvent (D2)................307, 309, 311
KCl ............................................................ 307, 309
lipid extraction 1 (LE 1) ................................... 307
lipid extraction 2 (LE 2) ................................... 307
lipid extraction 3 (LE 3) ................................... 307
lipid reference standards .......................... 306, 307
method ......................................................309–311
mobile phase...................................................... 306
molar lipid to chlorophyll ratio ........................ 312
streaking............................................................. 314
Liquid-phase O2 measurements .......................... 198, 203
Live-cell imaging ........................................................... 253
M
Macerozyme R-10................................................ 266, 276
Mass spectrometer .............................................. 142–145,
160, 198–201, 205
Mass-to-charge ratio (m/z) ................................ 144, 199,
200, 203, 208
Maximum power density .............................................. 336
Maximum rate of electron transport
(Jmax) .............................................7, 26, 33, 37
Mehler reaction ............................................................. 142
Membrane inlet mass spectrometry
(MIMS) ............................................... 197–199
cuvette............................................................ 141–143,
145–150, 152
cuvette volume (V) .......................145, 146, 148, 149
determining leaf disc size per PPFD ...................... 145
draw down correction.................................... 148, 149
Faraday cup....................................144, 199, 200, 203
light response curve ....................................... 145, 149
linear calibration curves .......................................... 146
membrane installation.................................... 145, 150
vacuum line .......................... 142, 144–146, 200, 205
water trap.............................................. 143, 144, 148,
163, 200, 205
MES ...................................................................... 202, 266
Mesophyll ............................................................ 157, 158,
176, 178, 179, 184, 255, 285, 286
cell suspensions .............................................. 254, 255
conductance (gm) .................................. 160, 161, 169
single point method ..................................176–179
slope method ....................................176, 177, 179
Metaboliten quantification/transcription
profiling.......................................................... 57
MgCl2 ......................................................... 202, 230, 231,
233, 234, 243, 266
Microbial fuel cell (MFC).................................... 335, 336
Microplate reader ................................241, 244, 245, 248
Microscopy
bright field .....................................254, 255, 257, 259
confocal................................................. 253, 257, 259,
266, 273, 276, 280, 281, 294, 298, 300, 303
differential interference contrast imaging
(DIC) .................................254, 255, 257, 259
fluorescence ........................................... 266, 273, 281
light................................................253, 254, 257, 294
light sheet ................................................................ 294
scanning ................................................................... 253
transmission electron ....................253, 327, 329, 332
Mitochondria....................................................... 170, 179,
255, 272, 273, 279
Multiplexed measurements..................................... 46, 47,
51–54, 56, 57, 60–67

N 2-phosphoglycerate (2-PGA) ....................................... 240
+
NAD .................................................................... 240, 241
NADH ................................................................. 240, 241,
243, 245, 246, 248
Nicotinamide adenine dinucleotide phosphate
(NADPH) ......................................... 4, 27, 320
Non-photochemical quenching (NPQ) ................ 13, 16,
17, 88, 89, 92, 93, 103, 106, 107, 123–126,
132, 133, 135
Nucleophilic substitution reactions .................... 330, 331
O
O2 evolution (Eo)..........................................11, 148–151,
198, 199, 203, 204, 207, 208
O2 measurements ..........................................10, 197–202,
204, 206–209
Clark type O2 electrode (CTOE)........................... 201
background.......................................................... 10
measurements ...................................197, 198, 204
set up.................................................198, 200–202
membrane inlet mass spectrometry
(MIMS)........................................................ 201
background............................................... 199, 209
measurements .................................................... 208
set up........................................200, 201, 203, 209
Z(O2) calculation .................................................... 207
assay buffer ........................................................ 202
FeCy stock solution ................................. 202, 209
measurements ........................................... 207, 210
PPBQ stock solution................................ 202, 209
OJIP......................................................... 96–98, 100, 101
Open gas exchange system ........................................... 112
Optical clearing techniques .......................................... 286
O2 uptake (Uo) ..................................... 11, 142–151, 199
Oxygen evolving complex (OEC)................................ 320
Oxygen partial pressure
in the bundle sheath cells (Os) ...................... 157, 187
in the mesophyll cells (Om) .................................... 157
P degassing.................................................................. 301
Paraformaldehyde aqueous solution .......... 256, 258, 259
Pectinase ..............................................255, 256, 258, 296
Peroxisomes.......................................................... 255, 256
2-phenyl-p-benzoquinone (PPBQ) ................... 202, 204,
206–208, 210
Phi (φ)
maximum quantum yield for CO2 uptake............... 83
quantum efficiency of photosynthesis...................... 16
Phosphoenolpyruvate (PEP)................................ 240, 241
fraction of leaf carbon fixed by PEPC........... 158, 173
maximal PEP carboxylation rate (Vpmax) ............... 158
PEP carboxylase (PEPC) ..............157, 169, 173, 177
PEP carboxylation rate (Vp) ................. 158, 184, 185
PHOTOSYNTHESIS: METHODS AND PROTOCOLS
Index 353
3-phosphoglycerate (3-PGA) .............................. 240, 245
Photoanodes........................................320, 321, 330, 331
Photocathodes............................................................... 320
Photochemical quenching ..............................12, 13, 124,
125, 132, 135
Photoelectrochemical cells ......................v, 320, 321, 330
Photoelectrodes.................................................... 320, 330
Photoinhibition .................................................16, 17, 40,
41, 84, 85, 90–93, 124, 125, 128, 129, 135,
137, 206
Photorespiration.................................................. 8, 14, 15,
18, 26, 27, 41, 42, 91, 134, 135, 142, 150,
156–158, 166, 169, 173, 174, 176, 178, 186,
198, 229
PhotosynQ.......................................................... 12, 17, 18
Photosynthate ................... 100, 157, 159, 173, 174, 179
Photosynthesis, photosynthetic rate (A).....................3, 8,
11, 59, 65, 91, 138, 156, 166, 184
Photosynthetic photon flux density (PPFD)................ 69,
72, 73, 75, 76, 80, 86–89, 93, 107, 125, 142,
145, 149–151, 247
Photosynthetically active radiation (PAR) ............. 41, 91,
109, 110, 112, 114, 119
Photosystem I (PSI)................................................ 11, 15,
47, 92, 126, 128, 320, 330
Photosystem II (PSII) ............................................ 13, 15,
84, 95, 96, 98, 103, 106, 107, 122, 124, 197,
202, 208, 320, 330
fraction of PSII centers that are ‘open’ (qL)........... 88,
89, 93, 125
PSII efficiency factor (qP) ............................ 88, 89, 93
PSII-mediated electron transfer (ΦPSII)................. 107
Photovoltaic systems ..................................................... 320
Plant developmental stage ..................46, 52, 63, 89, 111
Plant enzyme assisted (PEA)-CLARITY ........... 286, 287,
292–294, 296, 298, 302
clearing
lipid removal ............................287, 296, 297, 302
solution ..................................................... 292, 301
enzyme treatment
procedure........................................................... 296
solution .............................................................. 292
hydrogel
embedding procedure....................................... 292
polymerization procedure........................ 293, 296
solution .............................................287, 292, 300
molecular labeling .......................................... 287, 297
immunostaining ....................................... 298, 302
small molecule staining ..................................... 298
protocol optimization ............................298, 300–302
refractive index matching .............296, 298, 299, 301
Plasmodesmata ............................................ 254, 255, 286
 PHOTOSYNTHESIS: METHODS AND PROTOCOLS
354 Index
Post-transcriptional gene silencing (PTGS) ................ 264
Potassium ferricyanide (III) (FeCy)............................202,
204, 206–208, 210, 343
Potassium hexachloroiridate (IV) ................................ 323
Potassium iodide (IKI) .......................254, 256, 257, 259
Propidium iodide (PI) ......................................... 291, 298
Protein quantification ................................................... 216
Protein transfer.............................................................. 222
blocking solution (Blotto).................... 217, 222, 223
filter paper....................................................... 217, 222
low fluorescence PVDF membrane............... 217, 225
nitrocellulose membrane ........................................ 222
primary antibody ............................................ 217, 223
secondary antibody .......................217, 223, 225, 226
tris-buffer-saline (TBS) ..........................217, 222–225
tris-buffer-saline-Tween (TBST).......... 217, 223, 225
Proton exchange membrane (PEM) ............................ 336
Protoplasts ..................................................................... 264
digestion solution.................................. 266, 270, 280
imaging ................................................. 266, 273, 274,
278, 280, 281
imaging solution ............................................ 266, 273
isolation ......................................... 266, 270–273, 280
ϕPSII or ΔF/Fm0 ........................................................... 125
Pyruvate ................................................................ 240, 241
Pyruvate kinase (PK)............................................ 240–242
Q (Vcmax).....................7, 26, 27, 32, 41, 42, 158
QA ................................................................. 102, 106, 107
qE................................................................................... 103
qL .......................................................................... 106, 118
Quantum cascade laser (QCLs) ................................... 162
R desalt buffer .............................................. 230, 232
Rate of triose phosphate utilization (Tp) ....................... 28
Rates of carboxylation
Wc ............................................................................... 27
Wj ............................................................................... 27
Wp .............................................................................. 27
ζ Ratio of the 12CO2 mole fraction in the dry air
coming in the gas exchange cuvette over
the difference in 12CO2 mole fractions of
air in and out of the cuvette ......................159,
160, 165, 166, 178, 187, 188
Reciprocal plot ............................................ 108, 114, 115
Refractive index (RI)........................................... 285–287,
291, 294, 296, 298
Resistance
chlorophyll resistance to CO2 diffusion
(rc) .............................................. 157, 172, 179
mesophyll resistance to CO2 diffusion
(rm)...................................................... 157, 179
wall resistance to CO2 diffusion (r w)... 157, 172, 179
Respiration
mesophyll mitochondrial respiration rate
(ℛm).................................................... 157, 186
non-photorespiratory CO2 released in the
dark (ℛd) ................................... 157, 170, 186
Ribulose 1,5-bisphosphate (RuBP) ....................... 26, 27,
41, 215, 229, 239–242, 244–246, 249
Ribulose-1,5-bisphosphate carboxylase/oxygenase
(Rubisco) .................................. 161, 215, 223,
229–234, 237, 239, 241, 244, 248
activation ................................................................. 249
half of the reciprocal of Rubisco
specificity, γ∗ ................................................ 159
large scale purification............................................. 230
60% (w/v) PEG 4000 final concentration
calculation .................................................... 234
anion exchange buffer A (Buffer A)........ 231, 234
anion exchange buffer B (Buffer B)................. 231
anion-exchange columns ......................... 231, 234
chromatographic system .......................... 231, 234
leaf extraction buffer ................................ 231, 233
MgCl2 calculation ............................................. 234
nitrocellulose membrane filters ........................ 232
protease inhibitor cocktail ....................... 231, 234
protocol ............................................................. 233
pyrex homogenizer ........................................... 232
maximum rate of Rubisco carboxylation
Michaelis-Menten constant of Rubisco
for CO2 (Kc)................................................ 157
Michaelis-Menten constant of Rubisco
for O2 (Ko) .................................................. 157
rapid extraction ....................................................... 230
leaf extraction buffer ......................................... 230
protease inhibitor cocktail ...............230, 232, 233
protocol ............................................................. 232
sephadex G-25 desalting columns .......... 230, 232
rate of RuBP consumption by Rubisco
(Wc) ................................................................ 27
Rubisco carboxylation rate (Vc) ................... 157, 167,
168, 184, 185
Rubisco oxygenation rate (Vo) ............. 158, 184, 185
Rubisco activase (Rca) ........................................ 216, 217,
219, 221, 223, 229, 230, 237
Rubisco activity assay ...................................239–241, 244
assay mix ........................................ 243–246, 248, 249
initial Rubisco activity (Vi) ............................ 244, 246
measuring initial Rubisco activity......... 244, 245, 249
measuring total Rubisco activity .......... 244, 245, 249
microplate preparation................................... 244, 245
path length correction .......................... 244, 245, 248
protein extraction.................................. 217, 219, 244
rate of RuBP consumption ..................................... 245

Rubisco activation state calculation ....................... 246
solutions ......................................................... 241, 248
total Rubisco activity (Vt) .............................. 245, 246
S Tomato bushy stunt virus P19 protein264, 265, 275, 279
SDS-PAGE ..........................................216, 217, 220, 221
Bradford reagent ..................................................... 217
BSA .......................................................................... 217
polyacrylamide gel.......................................... 217, 219
pre-stained protein standards ................217, 219–222
running buffers ..................................... 217, 219, 224
Seed sterilization ............................................................. 48
Silica gel TLC plates ................................... 306, 307, 309
Sodium azide .............................................. 292, 300–302,
322, 323, 326, 330, 332
Sodium cacodylate buffer .................................... 256, 258
(+)-Sodium L-ascorbate................................................ 323
Spectrophotometer ..................................... 266, 277, 308
Starch .....................................................58, 256, 257, 297
Stochastic electrotransport ......................... 290, 301, 302
Stomatal .........................40, 46, 60, 86, 87, 91, 112, 158
conductance (gs) .....................................7, 26, 31, 36,
38, 90, 91, 129, 132, 135, 137, 152, 166
opening .............................................. 16, 41, 124, 136
Stress .................................. 58–61, 66, 67, 103, 122, 129
Stress induction ............................................................... 48
Stuttgarter Masse ......................................................29, 30
Sulfonated bathophenanthroline................ 323, 324, 327
See also Bathophenanthrolinedisulfonic acid
Synechocystis sp. PCC 6803 .................................. 337, 340
T offset ........................................................................ 164
Temperature response curves ......................................... 33
PHOTOSYNTHESIS: METHODS AND PROTOCOLS
Index 355
Ternary correction factor (t) .............................. 157, 166,
171, 181, 189
Tetrahydrofuran (THF) .............291, 322, 325, 328, 333
Thermal imaging ............................................................. 13
Total soluble protein (TSP) extraction ........................ 216
leaf extraction buffer ...................................... 216–219
SDS blank solution ................................217–219, 225
SDS loading buffer......................................... 216, 218
SDS standard preparation.............................. 218–221
TPU limitation ............................................................7, 41
Transgene expression ..........................263, 264, 274, 278
Transgenesis
stable ............................................................... 263, 264
transient .......................................................... 263, 264
Transpiration rate (E) .......................................46, 65, 66,
75, 90, 156, 166
Tris-buffer-saline-Tween (TBST)................................. 223
Tunable diode laser spectroscopes (TDLAS) .... 160–162,
164, 165, 174, 188, 189
coupling with gas exchange system...... 162, 163, 188
detector........................................................... 161, 162
gain ................................................................. 162, 164
gas
calibration (standard)........................................ 164
reference line-inlet............................162, 163, 189
sample line-outlet.............................162, 163, 189
tank ........................................................... 165, 188
instrument precision .....................161, 164, 166, 178
leaks.......................................................................... 189
mixing ratio (Mole fractions) .......162, 164, 185, 190
precision error ................................................ 165, 166
selectivity/sensitivity ............................................... 162