You are on page 1of 9

NMR IN BIOMEDICINE

NMR Biomed. 2008; 21: 129137


Published online 21 May 2007 in Wiley InterScience
(www.interscience.wiley.com) DOI:10.1002/nbm.1169

Fast screening of paramagnetic molecules in zebrafish


embryos by MRI

Laurence Canaple,1,3* Olivier Beuf,2 Mircea Armenean,1 Jens Hasserodt,4 Jacques Samarut 3
and Marc Janier1,5
1
Plateforme ANIMAGE, Batiment CERMEP, Bron, France
2
Universite de Lyon, Universite Lyon 1, CNRS, UMR 5012 and Laboratoire RMN-MIB ESCPE Lyon, Villeurbanne, France
3
Laboratory of Molecular Cell Biology, CNRS INRA, UMR 5161, Universite de Lyon, IFR 128 BioSciences Lyon-Gerland, Ecole Normale Superieure,
Lyon, France
4
CNRS UMR 5182, Universite de Lyon, IFR 128 BioSciences Lyon-Gerland, Laboratoire de Chimie, Ecole Normale Superieure, Lyon, France
5
Universite de Lyon, Universite Lyon 1, Faculte de medecine Lyon Nord, CREATIS, CNRS UMR 5515, INSERM U630, Bat. CERMEP and Hospices Civils
de Lyon, Service de Medecine Nucleaire, Hopital Edouard Herriot, Lyon, France

Received 14 December 2006; Revised 12 January 2007; Accepted 3 February 2007

ABSTRACT: Zebrafish embryo is a well-established model used in many fields of modern experimental biology. We
demonstrate that it provides a promising model platform for exploring fundamental MR aspects that can be used to screen and
study active MR molecules before progressing to more complex living systems. Setting up a dedicated MRI methodology, we
arrayed a large number of living embryos, which were microinjected at very early stages of development with different
contrast agents. We also showed that MRI signal intensity correlates with the gadolinium content of zebrafish embryos. This
allowed us to validate a new approach for MR compound screening. Using a specific surface coil of 5 mm inner diameter, we
obtained for the first time high-spatial-resolution images at 7 T of living zebrafish embryos with a 47 mm isotropic voxel size
with an acquisition time of 39 min. Finally, we discuss potential applications of this development: a viable in vivo assay for
screening small pharmacological compounds; assessment of and tracking the action of molecules over time. Exploring
in vivo biological activity, gene function analysis, and detailed characterization of disease processes in fish are natural
extensions of these preliminary studies. Copyright # 2007 John Wiley & Sons, Ltd.
KEYWORDS: MRI; zebrafish; contrast agent; screening; gene function

INTRODUCTION also used as a pharmacological tool in drug discovery and


ecotoxicity research (46). As it can live in as little as a
The zebrafish (Danio rerio) is a useful and relevant model few microlitres of fluid, only micrograms of compound
of vertebrate development and organogenesis (1,2). It per assay are needed for screening. This facilitates
exhibits high fecundity (about 200 eggs per week per screening of compounds with a different order of
female), with embryos developing rapidly in fresh water magnitude of synthetic chemistry resource, enabling
externally to the mother and therefore easily manipulated in vivo analysis of compound action at much earlier stages
ex utero, progressing from fertilized eggs to free and higher throughput than hitherto possible. Recent
swimming larvae in 60 h. Moreover, zebrafish embryos reports underline the fact that the zebrafish is also an
are transparent. Visualization of all stages of organ appropriate model for studying human diseases (7,8),
development is relatively easy in the first few days of life such as haematological disorders (9,10), kidney pathol-
when the fish is only a few millimetres in length, and this ogy (11) and cancers (1214). Clearly, the zebrafish
can be combined with manipulation of gene expression at model combines the relevance of a vertebrate with the
RNA and protein levels. Importantly, gene expression can scalability of an invertebrate and, in some studies, could
be induced using microinjection of specific vectors at provide an interesting intermediate vertebrate model to
very early stages and is generally highly effective for at laboratory small mammals.
least the first 2 days of development [http://zfin.org(3)]. Although in the literature there is a trend to pharma-
Because it can absorb chemicals from water, zebrafish is cological and genetic use of zebrafish to obtain a deeper
understanding of cellular mechanisms, advances have
*Correspondence to: L. Canaple, Plateforme ANIMAGE, Batiment also occurred in molecular imaging, with characterization
CERMEP, 59 Boulevard Pinel, 69677 Bron, France. and measurement of molecular events in living animals
E-mail: laurence.canaple@ens-lyon.fr with high sensitivity and spatial resolution. The progress
Contract/grant sponsor: ANR project; contract/grant number: ANR
05-EMPB-010-03. of molecular imaging has been accelerated by the
Copyright # 2007 John Wiley & Sons, Ltd. NMR Biomed. 2008; 21: 129137
130 L. CANAPLE ET AL.

development of dedicated small-animal imaging equip- Preparation of zebrafish embryos


ment (1517). MRI is now widely used as a clinical
diagnostic tool because it offers a non-invasive means Zebrafish were housed in the zebrafish facility of the
of acquiring multiple orientations of two- or three- IFR128 at the ENS of Lyon. They were raised and
dimensional images, enabling high-spatial-resolution maintained under standard laboratory conditions at 288C
views of numerous tissues (18). MRI can also map on a 14 h light/10 h dark cycle, as described by
in vivo structure and function and could be a useful and Westerfield (3). Morphological features were used to
appropriate tool for the observation of biological define the developmental stages of the embryos. Embryos
processes in animals. Images are based on the NMR were incubated at 288C, and different developmental
signal from proton nuclei, and the signal intensity is a stages were determined according to the description in the
function of proton concentration and longitudinal and Zebrafish Book (3). Embryos were obtained by natural
transverse relaxation times, T1 and T2. Paramagnetic mating of adult zebrafish.
contrast agents are used to enhance the intrinsic contrast
between regions, tissues, and cells that are magnetically
similar but histologically distinct (19). There have been Microinjection of contrast agent
recent exciting achievements in MRI with the develop-
ment of innovative contrast agents, such as smart sensor Each contrast agent was adjusted, except when indicated
probes, that modify relaxation times by interaction with a otherwise, to a final concentration of 0.4 M in 1 
specific biological target or the biochemical activities of Danieau solution [5 mM Hepes, pH 7.6, 58 mM NaCl,
biological systems, allowing detection of precise mol- 0.7 mM KCl, 0.4 mM MgSO4, and 0.6 mM Ca(NO3)2]
ecular events (20,21). However, owing to their permanent containing 0.5% phenol red and injected into the zebrafish
paramagnetic nature, these gadolinium-based agents start embryo from one-cell to four-cell stages using the
their signal change from an off agent, which already Eppendorf FemtoJet microinjector system (about 1 nL
modifies T1 relaxation time to a significant extent, to solution is injected per embryo). As a control for viability
an on agent with a greater degree of T1 change. The of each batch, some embryos were injected with water
main challenge to the use of such MRI smart agents for only. After injection, embryos were kept at 288C, after
investigating biological questions is their delivery 6 h, the medium was changed. Normal development of the
through cell membranes (22). embryos at 24, 48, and 72 h was observed under a Nikon
The goal of our study was to investigate the true SMZ-U stereomicroscope.
potential of the MRI approach with the characteristics For signal intensity quantification based on MR
that have made zebrafish a powerful model for biologists images, several solutions of Dotarem at different con-
and pharmacologists. We report for the first time MRI of centrations were prepared and injected into the embryos
living zebrafish embryos. Using different contrast agents, either at the same pressure for the different concentrations
we were able to drive high-throughput MRI screening of or at different pressures for one specific concentration.
zebrafish embryos over time. We also show that MRI The injected volume is proportionally dependent on the
signal intensity can be reliably and accurately correlated injection pressure and injection time.
with the gadolinium content of zebrafish embryos.
Developing a dedicated MRI methodology around
zebrafish embryos, we have obtained high-spatial- MRI setup
resolution images with a 47 mm isotropic voxel size
and an acquisition time of 39 min. Finally, we discuss To image several zebrafish embryos simultaneously, a
potential applications of this development: a viable specific sample holder was designed: a sample well
in vivo assay for screening of small and pharmacological former (well size 4 mm  1 mm) was placed in a
compounds; assessment of and tracking the action of polystyrene cell culture plate (35 mm diameter) and
molecules over time. It also offers a powerful tool for melted aqueous agarose solution was poured in. When the
the study of in vivo biological activity, analysis of gene agarose gel was completely formed, the sample well
function, and detailed characterization of disease pro- former was removed, and the surface of the agarose wells
cesses in a whole organism. covered with 1  Danieau solution. For MRI examin-
ation, the zebrafish embryos were placed in the
MATERIALS AND METHODS submerged agarose wells. To minimize embryo move-
ment during imaging, tricain anaesthetic was added to the
Contrast agents Danieau medium at a final concentration of 0.01%. MR
images were obtained on a 7 T Biospec system (Bruker
Three different gadolinium complexes were used in this Biospin, Ettlingen, Germany) using a birdcage coil of
study: Gd-HP-DO3ATM (ProHanceTM) and Gd-DO3A 72 mm inner diameter for radio frequency transmitting
were provided by Bracco Imaging S.p.A. (Milan, Italy), and a surface coil of 15 mm diameter for receiving
and Dotarem1 by Guerbet (Aulnay sous Bois, France). (Bruker Biospin). A T2-weighted turbo spin-echo
Copyright # 2007 John Wiley & Sons, Ltd. NMR Biomed. 2008; 21: 129137
DOI: 10.1002/nbm
MRI SCREENING IN ZEBRAFISH EMBRYOS 131

sequence and a T1-weighted spin-echo sequence were inner diameter. The embryo was centred on the loop coil
systematically performed at the same location in both for maximum sensitivity. A three-dimensional fast
coronal and transverse orientation. The T2-weighted low-angle shot (FLASH) sequence was used with the
images were acquired with TR/TE 3000/68 ms, a turbo following parameters: 908 flip angle, TR/TE 32/3.5 ms,
factor of 8, and number of experiments (NEX) of 2. The 25 kHz receiver bandwidth, and 8 NEX. A cubic field of
T1-weighted images were acquired with TR/TE 100/14 ms view of 6 mm was acquired with a matrix size of 96 
and 6 NEX. For all scans, the field of view was 35  96  128. The acquisition volume was reconstructed to a
35 mm2, the slice thickness was 1.5 mm, and the matrix 128  128  128 matrix leading to a 47 mm isotropic
size was 256  256. voxel size. The scan time was 39 min. Three-dimensional
MRI acquisition allows multi-planar reconstruction and
maximum intensity projection visualization.
Data analysis and quantification of MRI
signal intensity
RESULTS
MRI detection was carried out 24 h after the adminis-
tration of Dotarem at different concentrations to Dedicated MRI method for zebrafish
anaesthetized zebrafish embryos using a 7 T Bruker embryos
system coupled to a 15-mm inner-diameter surface coil.
T1-weighted images of at least three living embryos were For the screening protocol, zebrafish embryos were
used to quantify the signal intensity in the embryos, which placed in agarose wells immersed in 1% Danieau solution
was corrected for experimental noise background and supplemented with the anaesthetic agent, tricain (0.01%),
normalized to a reference signal measured in the and placed above a surface coil of 15 mm inner diameter
anaesthetic medium surrounding the embryos. Results as shown in Fig. 1A. A protocol consisting of sequences
are expressed as mean  SEM from three independent with various orientations and contrast was designed. A
measurements, each performed with embryos injected good compromise was reached between image contrast,
with a different concentration of contrast agent. spatial resolution, and scan duration to screen zebrafish
Significant differences between groups were determined embryos between 24 and 72 h.
using Students t test. For high-resolution MRI, an imaging single-loop coil
of 5 mm inner diameter was designed and built in our
laboratory (Fig. 1B). The coil was interfaced with the
High-spatial-resolution MRI Bruker decoupling box to allow tuning and matching of
the coil as well as active uncoupling of the receiver coil
For high-spatial-resolution MRI, gadolinium-injected during radio frequency transmission. The hole in the
zebrafish embryos were placed in agarose-coated NMR centre of the loop allowed one thin NMR tube containing
sample tubes containing the anaesthetic medium. MRI zebrafish embryos to be positioned exactly in the area of
acquisition was carried out at 7 T 56 h after injection maximum coil sensitivity in order to obtain the best
using a home-made dedicated single-loop coil of 5 mm signal-to-noise ratio.

Figure 1. Digital photograph of the experimental setup and position-


ing. (A) Sample holder containing the zebrafish embryos placed above
the 15-mm surface coil. (B) Dedicated home-made surface coil with
5-mm inner-diameter single loop. The zebrafish embryos are placed in a
thin tube and located at the coil loop centre. The tuning and matching
circuit is driven by a Bruker decoupling box.
Copyright # 2007 John Wiley & Sons, Ltd. NMR Biomed. 2008; 21: 129137
DOI: 10.1002/nbm
132 L. CANAPLE ET AL.

Screening of contrast agents by MRI in three times less for Gd-DO3A than for Dotarem and
zebrafish embryos ProHance. Moreover, at the concentration we used, no
teratogenic effect on the surviving embryos was observed,
As little is known about the toxicity, distribution, and as they developed perfectly with no visible defect.
release of injected MR agents in zebrafish embryos, we To use zebrafish embryos in a MRI-based in vivo
first evaluated embryo survival 24 h after the injection of high-throughput molecule-screening assay, we performed
different gadolinium complexes. The choice of three MRI acquisition on anaesthetized embryos that had been
available MRI non-specific agents, Dotarem and Pro- injected with Dotarem (Fig. 2A), ProHance or Gd-DO3A
Hance on the one hand and Gd-DO3A on the other, was (Fig. 2B), or nothing in the case of the control embryos.
based on the desire to measure the differences in T1 During the acquisition time, they were randomly placed
relaxation time in a genuine in vivo situation. In fact, in submerged agarose wells above the 15 mm surface coil.
anionic Dotarem and neutral ProHance allow one water With this setup, we were able to screen at least 12
molecule into their first coordination sphere, and neutral embryos by acquisition. We systematically performed a
Gd-DO3A is known to be coordinated by two water turbo spin-echo T2-weighted contrast image to verify the
molecules (19). Hence, Gd-DO3A shows a considerable presence of all the embryos, which appear in hyposignal
increase in T1 relaxivity in vitro, an effect that may be compared with the surrounding medium, as shown in
diminished in vivo because of the presence of endogenous Fig. 2A right, followed by a spin-echo T1-weighted
bidentate molecules. It is therefore important to detect in contrast image. MRI detection was as expected, with the
our context their in vivo MRI signal activity. Dotarem and signal contrast of the injected embryos being enhanced
ProHance appear to be well tolerated by zebrafish and the injected embryos appearing in hypersignal
embryos. The centile viability of embryos injected with intensity. On T1-weighted contrast images, the control
Dotarem did not differ from that of the non-injected con- embryos are not visible, whereas the signal intensity of
trols and was 8595%. Survival was 68% for ProHance the embryos injected with the three contrast agents show
and only 11% for Gd-DO3A. The high mortality observed large contrast enhancement because of the shortening of
for Gd-DO3A can be explained by the fact that the the longitudinal relaxation time, T1, of the environmental
working concentration is close to the lethal dose, which is protons (Fig. 2A,B).

Figure 2. MR molecule screening in injected zebrafish embryos. MRI was


carried out at 7 T, 24 or 48 h after the administration of the contrast agent
using a 15-mm inner-diameter surface coil. (A) Left, A global view of one
plate containing agarose wells with injected or non-injected living zebrafish
embryos. Right, top, representative T2-weighted contrast image of living
zebrafish embryos (hyposignal); bottom, representative T1-weighted con-
trast image of living zebrafish embryos. Only the Dotarem-injected embryos
appear in hypersignal. C, Non-injected control embryo; I, Dotarem-injected
embryo. (B) A representative T1-weighted contrast image of Dotarem-,
ProHance (Gd-HPDO3A)- or Gd-DO3A-injected living embryos. The high-
intensity regions of the embryos indicate the presence of contrast agent. C,
Non-injected control embryo; DOTA, Dotarem-injected embryo; PH, Pro-
Hance-injected embryo; DO3A, Gd-DO3A-injected embryo.
Copyright # 2007 John Wiley & Sons, Ltd. NMR Biomed. 2008; 21: 129137
DOI: 10.1002/nbm
MRI SCREENING IN ZEBRAFISH EMBRYOS 133

Monitoring of Dotarem-injected zebrafish pronephric kidney and the observation of increasing


embryos over time using MRI background signal from the medium surrounding the
embryo if it is not changed during the experiment. As all
Zebrafish embryos develop rapidly, with embryos the injections were into the cell and not the yolk, only the
progressing from fertilized eggs to free swimming larvae embryo displays more or less uniform increase in signal
in 2.5 days. In this short time window, a functional intensity, suggesting that the distribution of the contrast
pronephric kidney is formed allowing blood filtration, agent follows mosaic cell division.
tubular resorption, and fluid excretion, and the embryo
increases its volume by a factor 5. This leads to great
dilution and secretion of microinjected MR products, MRI signal intensity as a function of the
which will be a major limitation of MRI use for screening concentration of Dotarem injected into
MR molecules in zebrafish embryos. zebrafish embryos
In keeping with this, we studied the MRI signal from
Dotarem-injected zebrafish embryos 24, 48 and 72 h after To determine the effect on the MRI signal intensity of the
injection using a 15-mm-diameter imaging probe quantity of contrast agent injected into the zebrafish
(Fig. 3A). We detected a MRI contrast-enhanced signal embryos, we used injections of either different gadoli-
up to 72 h after injection. The time course of MRI signal nium concentrations at the same pressure or one
intensity in the embryos injected in similar conditions is concentration at different pressures. The injection volume
shown in Fig. 3B. The signal decreases over time and is depends on injection pressure and injection time. The
significantly lower at 72 h than at 24 h, essentially because mean signal intensity was calculated and corrected for
of dilution of the Dotarem in the whole embryo and its image background (standard deviation of noise) as well as
secretion into the environmental aquatic medium. This is the reference signal intensity in the surrounding medium
in agreement with the developing functionality of the as shown in Fig. 4. There was a significant increase in

Figure 3. (A) A representative T1-weighted contrast image of living embryos over time. MRI was
carried out 24, 48 or 72 h of development after administration of the Dotarem contrast agent in
zebrafish embryos at stages 14. MRI was performed at 7 T with a 15-mm inner-diameter surface coil.
Only the injected embryos (I) appear in hypersignal. The non-injected control embryos (C) appear in
hyposignal. On the right, photographs of the zebrafish embryos at 24, 48 and 72 h of development
with the corresponding embryo length (EL, embryos longest linear dimension) are shown. (B) MRI
signal as a function of Dotarem concentration injected into zebrafish embryos over time. Results are
expressed as mean  SEM from at least three independent measurements. Significant difference
between groups at 24 and 72 h is shown: P < 0.001.
Copyright # 2007 John Wiley & Sons, Ltd. NMR Biomed. 2008; 21: 129137
DOI: 10.1002/nbm
134 L. CANAPLE ET AL.

Figure 4. MRI signal as a function of Dotarem concentration injected into zebrafish embryos. MRI
detection was carried out 24 h after injection of the Dotarem contrast agent at different concentrations
into the embryos. Representative T1-weighted contrast images of living embryos (3 or 4 as indicated)
were used for the quantification. Relative MRI signal enhancement is displayed after normalization to
the experimental background. Results are expressed as mean  SEM from three independent measure-
ments, each carried out with embryos injected with different concentrations of contrast agent.
Significant difference between groups is indicated by asterisks: P < 0.05; P < 0.01.

MRI signal intensity, which was directly proportional to DISCUSSION


the quantity of gadolinium injected up to 200 pmol/
embryo. However, for higher concentrations (200 pmol/ In this study, we demonstrate that the zebrafish embryo can
embryo), the signal intensity did not appear to change be used as a benchmark for screening, assessment, and
significantly, suggesting that the limit of signal saturation tracking of contrast agents over time using MRI at 7 T.
was reached with the sequence parameters used. The zebrafish has a number of advantageous features as a
Subsequently, within the limits of detection, significant model organism for the study of vertebrate development,
variations in MRI signal intensity correlated with the disease, and biological pathways, which have been
gadolinium content of the embryos, and this was taken validated further by the demonstration that many zebrafish
into account in a screening assays. genes show a high degree of structural and functional
similarity to their human homologues. For a wide spectrum
of biological questions, the combined use of traditional
High-spatial-resolution MRI of microscopy, which is restricted to thick specimens, and
Dotarem-injected zebrafish embryos more in-depth studies at organ and tissue level using
immunohistochemistry and in situ hybridization methods
To answer a number of biological questions in living allows zebrafish investigators to describe, at cellular
embryos, we proposed a high-resolution three- resolution, biological processes and their mechanisms.
dimensional MRI approach, with intrinsically high spatial As these invasive methods do not allow longitudinal studies
resolution and multiple source of contrast, to provide, in a with the same specimen, imaging techniques that allow
non-destructive way, sufficient anatomical detail. Opti- non-invasive physiological and molecular monitoring
mizing anatomical detail requires a compromise between in vivo are more suitable for assessing dynamics in vivo
contrast, spatial resolution, experimental time, anaes- and gene expression in any cell type.
thetic effects, and survival of the embryo. Using the 5-mm
inner-diameter dedicated single-loop coil, we designed
and maximized an MRI procedure at 7 T as previously MRI on living zebrafish embryos for
described for gadolinium-injected embryos placed in molecular screening or molecular tracking
agarose-coated NMR tubes. MRI performed on Dotar- over time
em-injected embryos allows three-dimensional images to
be obtained with either a resolution of 31  31  62 mm Traditionally, large-scale molecular screening is per-
voxel size for an acquisition time of 45 min (Fig. 5A) or formed using cell lines or in vitro protein binding assays,
an isotropic 47 mm voxel size for an acquisition time of which are limited by the lack of spatial and temporal
39 min (Fig. 5B). The high signal intensity depicted in the context with regard to tissue type and stage of
embryos indicates the presence of the contrast agent. A development, and do not represent normal physiological
maximum 908 flip angle was used to maximize contrast status in a multicellular organism (23,24). Zebrafish is a
between injected embryos and surrounding solution. well-characterized model used for in vivo selection of
Copyright # 2007 John Wiley & Sons, Ltd. NMR Biomed. 2008; 21: 129137
DOI: 10.1002/nbm
MRI SCREENING IN ZEBRAFISH EMBRYOS 135

Figure 5. High-spatial-resolution MR T1-weighted contrast image of Dotar-


em-injected zebrafish embryo obtained at 7 T. (A) MRI performed on embryos
56 h after injection using the 5-mm inner-diameter dedicated single-loop coil.
The high-intensity signal depicted in the embryo indicates the presence of
contrast agent. (B) Reformatted dorsal (a), ventral (b) and lateral views (c, d) of a
single zebrafish embryo, 48 h after Dotarem injection from a three-dimensional
acquisition with an isotropic 47 mm voxel size and an acquisition time of 39 min.
A photograph of an embryo at the same developmental stage is also
presented (e). This figure is available in colour online at www.interscience.
wiley.com/journal/nbm

bioactive molecules via injection or natural absorption molecular processes occurring in healthy living tissues
from the surrounding medium (4,25), and there is at and during disease development.
present a surge of interest in new technologies to achieve As genes can readily be overexpressed in zebrafish,
fast and successful screening in real time in living either by microinjection or electroporation of expression
zebrafish embryos (26,27). The development of a vectors (28,29) or by transgenic technology (30), the
dedicated coil and a specific MRI protocol was mandatory zebrafish is currently used to express specific reporter or
to be able to image zebrafish embryos of miniature size bioactive gene products in a controlled manner. Reporter
(24 mm in length) with an experimental setup adapted to genes expressed in specific tissues can thus be used for
the aquatic lifestyle of the zebrafish. Such a development real time imaging to elucidate cell migration patterns and
allows us to demonstrate that MRI can be used with tissue differentiation. Existing transgenic fish carrying a
zebrafish embryos as a screening tool of MR molecules reporter gene could also be analyzed by MRI for fine
over time. However, this approach is not limited to mapping of expression in living embryos or adult fish and
embryos, as techniques are already well established for for detailed fate mapping of lineage tracers. Imaging of
introducing compounds into adult zebrafish by intraper- reporter gene expression could in time become a valuable
itoneal or intracardiac injection. method for optimizing molecular therapies.
Interestingly, a large collection of existing mutants and Such an approach using MRI requires screening of
transgenic lines of zebrafish is also available (ZFIN bioactive compounds that display high affinity and
database, http://zfin.org). The use of high-spatial- specificity of action. The bioactive compound can be
resolution MRI should help in the screening of com- described as an imaging molecular probe which is
pounds that can rescue a molecularly defined defect. directed specifically against a molecular target, then
Further screening of mutants may also reveal important processed or activated in a living organism to generate a
clinical clues to diagnosis, treatment and prevention by specific MRI signal (off/on). This constitutes a consider-
identifying new active agents that affect cellular able technological challenge. Several MR contrast agents
processes in various dysfunctions. MRI of zebrafish have recently been successfully developed as specific
embryos could also be a biological research tool in toxic reporters of physiological status (calcium, zinc, pH or
assays and pharmacodynamic studies. pO2 activated) or biochemical activities [b-galactosidase-
activated EgadMe, chemical exchange saturation transfer
(CEST) agents] of biological systems (20,21,31,32).
Monitoring gene expression Another technological hurdle is delivery of these smart
agents to the cell, as they generally have poor ability to
One particularly exciting application is to exploit the penetrate the cell membrane (22,33). As well as providing
relevance of zebrafish in biology and MRI resolution to an accurate measure of gene expression in vivo, the
analyze gene function and further visualize cellular and reporter system must be temporally responsive, with a
Copyright # 2007 John Wiley & Sons, Ltd. NMR Biomed. 2008; 21: 129137
DOI: 10.1002/nbm
136 L. CANAPLE ET AL.

quick response in gene expression. It is important to be and disease in fish is a natural extension of these early
able to quantify the specificity of active products, which is studies.
directly proportional to expression of the targeted gene.
As recently reported (34) and clearly shown in this paper,
Acknowledgements
the correlation between the injected dose of contrast agent
and the MR signal intensity should allow accurate
We thank Rhone-Alpes Genopole and Fondation Rho-
quantification of gene expression by MRI. However,
ne-Alpes Futur funded by the Reseau National des Geno-
rational use of the zebrafish model with MRI techniques
poles. This research is funded by the ANR project (ANR
will require optimization of data analysis. Indeed, once
05-EMPB-010-03). We also thank Laure Bernard, Boua-
images are acquired, the greatest challenge is to extract
lem Azazi and Jean-Baptiste Langlois for technical assist-
the large amount of biological information and turn it
ance.
with ease into reliable quantitative information.

REFERENCES
An innovative cancer approach
1. Detrich HW, Westerfield M, Zon LI. Overview of the zebrafish
Many of the major genes involved in the mammalian cell system. Methods Cell Biol. 1999; 59: 310.
cycle, cell survival, cell death and tumour suppression are 2. Golling G, Amsterdam A, Sun Z, Antonelli M, Maldonado E, Chen
highly conserved in zebrafish [myc, ras, p53, bcl-2 W, Burgess S, Haldi M, Artzt K, Farrington S, Lin SY, Nissen RM,
Hopkins N. Insertional mutagenesis in zebrafish rapidly identifies
(13,14,35,36)]. As a wide variety of neoplasms, which are genes essential for early vertebrate development. Nat. Genet. 2002;
histopathologically strikingly similar to human cancers, 31(2): 135140.
are observed in zebrafish after the administration of 3. Westerfield M. The Zebrafish Book. A Guide for the Laboratory
Use of Zebrafish (Danio rerio). University of Oregon Press:
carcinogens or spontaneously arising through the Eugene, OR, 1995.
generation of transgenic fish, the zebrafish has recently 4. Zon LI, Peterson RT. In vivo drug discovery in the zebrafish. Nat.
been suggested to be a well-suited whole organism for the Rev. Drug Discov. 2005; 4(1): 3544.
5. Goldsmith P. Zebrafish as a pharmacological tool: the how, why
study of different types of cancer [cutaneous papilloma, and when. Current Opinion in Pharmacology 2004; 4(5): 504512.
pancreatic adenocarcinoma, hepatocarcinoma, osteosar- 6. Ankley GT, Johnson RD. Small fish models for identifying and
coma, leukaemia, neuroblastoma (12,13,37)]. MRI assessing the effects of endocrine-disrupting chemicals. ILAR J.
2004; 45(4): 469483.
screens based on zebrafish cancer models have significant 7. Ackermann GE, Paw BH. Zebrafish: a genetic model for vertebrate
advantages in identifying chemicals and therapeutic organogenesis and human disorders. Front Biosci. 2003; 8:
drugs for suppression, shrinking, and regression of a d12271253.
8. Dooley K, Zon LI. Zebrafish: a model system for the study of
cancer and in the prevention of a specific malignant human disease. Curr. Opin. Genet. Dev. 2000; 10(3): 252256.
phenotype. This could lead to identification of promising 9. de Jong JLO, Zon LI. Use of the zebrafish system to study primitive
chemotherapeutic and chemopreventive agents and could and definitive hematopoiesis. Annu. Rev. Genet. 2005; 39(1):
481501.
be an ideal entry point for the development of new drugs 10. Shafizadeh E, Paw BH. Zebrafish as a model of human hemato-
for the treatment of human cancers. Similarly, MRI based logic disorders. Curr. Opin. Hematol. 2004; 11(4): 255261.
on zebrafish cancer models could be used to monitor the 11. Drummond IA. Kidney development and disease in the zebrafish.
J. Am. Soc. Nephrol. 2005; 16(2): 299304.
delivery and magnitude of gene, cell, or drug transfer 12. Berghmans S, Jette C, Langenau D, Hsu K, Stewart R, Look T,
under pathological conditions or to follow up gene Kanki JP. Making waves in cancer research: new models in the
expression in cancer using specific activatable contrast zebrafish. Biotechniques. 2005; 39(2): 227237.
13. Lam SH, Gong Z. Modeling liver cancer using zebrafish: a
agents and genes known to suppress cancer-related comparative oncogenomics approach. Cell Cycle. 2006; 5(6):
phenotypes as a molecular target. The ability to follow 573577.
gene expression in vivo has huge applications for 14. Stern HM, Zon LI. Cancer genetics and drug discovery in the
zebrafish. Nat. Rev. Cancer 2003; 3(7): 533539.
biomedical research and is crucial in translational 15. Paulus MJ, Gleason SS, Easterly ME, Foltz CJ. A review of high-
medicine and gene therapy to improve our understanding resolution X-ray computed tomography and other imaging modal-
of the regulation in vivo of key molecular pathways. To ities for small animal research. Lab Anim (NY). 2001; 30(3): 3645.
16. Lewis JS, Achilefu S, Garbow JR, Laforest R, Welch MJ. Small
fully understand the complex gene regulation of many animal imaging: current technology and perspectives for oncolo-
biological processes, spatial and temporal information on gical imaging. Eur. J. Cancer 2002; 38(16): 21732188.
gene expression in real time in living organisms is 17. Koo V, Hamilton PW, Williamson K. Non-invasive in vivo imaging
in small animal research. Cell Oncol. 2006; 28(4): 127139.
required. 18. Delikatny EJ, Poptani H. MR techniques for in vivo molecular and
In summary, by combining the relevance of the cellular imaging. Radiol. Clin. North. Am. 2005; 43(1): 205220.
zebrafish model and the ability of MRI, we have 19. Merbach AE, Toth E. The Chemistry of Contrast Agents in Medical
Magnetic Resonance Imaging. Wiley: New York, 2001.
produced a powerful protocol for MR molecule screen- 20. Louie AY, Huber MM, Ahrens ET, Rothbacher U, Moats R, Jacobs
ing. The work described here also contributes a new way RE, Fraser SE, Meade TJ. In vivo visualization of gene expression
to visualize and decipher gene expression in living using magnetic resonance imaging. Nat. Biotechnol. 2000; 18(3):
321325.
embryos, identify new drugs, and select innovative 21. Lowe MP. Activated MR contrast agents. Curr. Pharm. Biotechnol.
imaging approaches for cancer. Exploring development 2004; 5(6): 519528.

Copyright # 2007 John Wiley & Sons, Ltd. NMR Biomed. 2008; 21: 129137
DOI: 10.1002/nbm
MRI SCREENING IN ZEBRAFISH EMBRYOS 137

22. Allen MJ, MacRenaris KW, Venkatasubramanian PN, Meade TJ. 31. Meade TJ, Taylor AK, Bull SR. New magnetic resonance contrast
Cellular delivery of MRI contrast agents. Chem. Biol. 2004; 11(3): agents as biochemical reporters. Curr. Opin. Neurobiol. 2003;
301307. 13(5): 597602.
23. Hudelist G, Singer CF, Kubista E, Czerwenka K. Use of high- 32. Yoo B, Pagel MD. A PARACEST MRI contrast agent to detect
throughput arrays for profiling differentially expressed proteins in enzyme activity. J. Am. Chem. Soc. 2006; 128(43): 1403214033.
normal and malignant tissues. Anticancer Drugs 2005; 16(7): 33. Allen MJ, Meade TJ. Synthesis and visualization of a membra-
683689. ne-permeable MRI contrast agent. J. Biol. Inorg. Chem. 2003; 8(7):
24. Yeung ES. High-throughput single molecule screening of DNA 746750.
and proteins. Chem. Rec. 2001; 1(2): 123139. 34. Shapiro MG, Atanasijevic T, Faas H, Westmeyer GG, Jasanoff A.
25. Parng C. In vivo zebrafish assays for toxicity testing. Curr. Opin. Dynamic imaging with MRI contrast agents: quantitative con-
Drug Discov. Dev. 2005; 8(1): 100106. siderations. Magn. Reson. Imaging 2006; 24(4): 449462.
26. Pichler FB, Laurenson S, Williams LC, Dodd A, Copp BR, Love 35. Langheinrich U, Hennen E, Stott G, Vacun G. Zebrafish as a model
DR. Chemical discovery and global gene expression analysis in organism for the identification and characterization of drugs and
zebrafish. Nat. Biotechnol. 2003; 21(8): 879883. genes affecting p53 signaling. Curr. Biol. 2002; 12(23):
27. Eggert US, Mitchison TJ. Small molecule screening by imaging. 20232028.
Curr. Opin. Chem. Biol. 2006; 10(3): 232237. 36. Lam SH, Wu YL, Vega VB, Miller LD, Spitsbergen J, Tong Y,
28. Rambabu KM, Rao SH, Rao NM. Efficient expression of transgenes Zhan H, Govindarajan KR, Lee S, Mathavan S, Murthy KR, Buhler
in adult zebrafish by electroporation. BMC Biotechnol. 2005; 5: 29. DR, Liu ET, Gong Z. Conservation of gene expression signatures
29. Xu Q. Microinjection into zebrafish embryos. Methods Mol. Biol. between zebrafish and human liver tumors and tumor progression.
1999; 127: 125132. Nat. Biotechnol. 2006; 24(1): 7375.
30. Amsterdam A, Becker TS. Transgenes as screening tools to probe and 37. Grabher C, Look AT. Fishing for cancer models. Nat. Biotechnol.
manipulate the zebrafish genome. Dev. Dyn. 2005; 234(2): 255268. 2006; 24(1): 4546.

Copyright # 2007 John Wiley & Sons, Ltd. NMR Biomed. 2008; 21: 129137
DOI: 10.1002/nbm