J. Phycol.
47, 692–702 (2011)
 2011 Phycological Society of America
DOI: 10.1111/j.1529-8817.2011.00991.x
FLOW CYTOMETRIC APPLICABILITY OF FLUORESCENT VITALITY PROBES ON
PHYTOPLANKTON 1
Louis Peperzak and Corina P. D. Brussaard2
Department of Biological Oceanography, Royal Netherlands Institute for Sea Research ⁄ NIOZ, P.O. Box 59,
NL-1790 AB Den Burg, the Netherlands
The applicability of six fluorescent probes (four
esterase probes: acetoxymethyl ester of Calcein
[Calcein-AM], 5-chloromethylfluorescein diacetate
[CMFDA], fluorescein diacetate [FDA], and 2¢,7¢-di-
chlorofluorescein diacetate [H2DCFDA]; and two
membrane probes: bis-(1,3-dibutylbarbituric acid)
trimethine oxonol [DiBAC4(3)] and SYTOX-Green)
as vitality stains was tested on live and killed cells of
40 phytoplankton strains in exponential and station-
ary growth phases, belonging to 12 classes and
consisting of four cold-water, 26 temperate, and
four warm-water species. The combined live ⁄ dead
ratios of all six probes indicated significant differ-
ences between the 12 plankton classes (P < 0.01)
and between individual species (P < 0.05). No spe-
cific differences were observed among strains of
one species, among species or strains from different
origin, nor between cells in exponential and station-
ary growth phase except for FDA. FDA showed a sig-
nificant (P < 0.05) drop of <20% in fluorescence
intensity in stationary cells. Of the four esterase
probes, the live ⁄ dead ratios of FDA and CMFDA
were not significantly different from each other,
and both performed better than Calcein-AM and
H2DCFDA (P < 0.001). Of the two membrane
probes, DIBAC4(3) stained rhodophytes and eugle-
nophytes much better than SYTOX-Green. The 13
algal strains best stainable (high live ⁄ dead ratios)
among all six probes belonged to nine genera from
six classes of phytoplankton. In conclusion, FDA,
CMFDA, DIBAC4(3), and SYTOX-Green represent a
wide choice of vitality probes in the study of phyto-
plankton ecology, applicable in many species from
different algal classes, originating from different
regions and at different stages of growth.
Key index words: Calcein-AM; CMFDA (CellTracker
Green); DiBAC 4 (3); FDA; flow cytometry;
H 2 DCFDA; live ⁄ dead assay; phytoplankton;
SYTOX-GREEN; vitality
Abbreviations: Calcein-AM, acetoxymethyl ester of
Calcein; CMFDA or CellTracker Green, 5-chlorom-
ethylfluorescein diacetate; DiBAC 4 (3), bis-(1,3-
dibutylbarbituric acid) trimethine oxonol; FDA,
1
Received 26 May 2010. Accepted 15 November 2010.
2
Author for correspondence: e-mail corina.brussaard@nioz.nl.
692
fluorescein diacetate; FL1, fluorescence 1 (green
probe fluorescence); FL4, fluorescence 4 (phyto-
plankton autofluorescence); FLS, forward light
scatter (cell size); H 2 DCFDA, 2¢,7¢-dichlorofluores-
cein diacetate
Phytoplankton forms the base of most aquatic
food chains. Therefore, changes in the processes
of phytoplankton growth and mortality may have
wide-ranging ecological consequences. The factors
controlling phytoplankton growth and physiology
during growth and division have been relatively well
studied, in contrast to the factors involved in mortal-
ity (Franklin et al. 2006). Mortality of phytoplankton
in nature includes three major processes: sedimenta-
tion, herbivore grazing, and cell lysis (Cloern 1996).
Cell lysis can be the outcome of pathogens (e.g.,
viral infection), physiological stress, or automortality
or programmed cell death (Berges and Falkowski
1998, Brussaard et al. 1998, Peperzak et al. 2000,
Brussaard 2004, Franklin et al. 2006). Mortality by
lysis is fundamental in understanding phytoplankton
ecology, urging for proper and fast methodologies
to determine this process on a cellular basis. Differ-
ent fluorescent probes are available that have poten-
tial applicability in measuring phytoplankton vitality.
Often these probes are used in combination with
epifluorescence microscopy (Garvey et al. 2007).
Alternatively, flow cytometry has been used (Dorsey
et al. 1989, Humphreys et al. 1994, Veldhuis et al.
1997, Brussaard et al. 2001, Franklin et al. 2005).
Flow cytometry has the major advantage of rapidly
analyzing single-cell abundance and optical proper-
ties such as scatter and fluorescence. The fast, multi-
variable examination of individual cells has made
flow cytometry an invaluable tool for both qualita-
tive and quantitative analyses.
The mode of action of one class of vitality probes
is based on intracellular esterase activity. The most
common esterase probe is the cell membrane-
permeable FDA, which is transformed in the green
fluorescent fluorescein after cleavage of the acetates
by intracellular esterases (Dorsey et al. 1989, Gilbert
et al. 1992, Jochem 1999, Franklin et al. 2001, 2005,
Lage et al. 2001, Garvey et al. 2007). Cells that are
metabolically active are considered alive, whereas
 V I T A L I T Y P R O B E S F O R P H Y T O PL A N K T O N
those that are not active and thus remain unstained
are defined as dead. Because the fluorescein may
rapidly leak out of a cell, and can lead to an appar-
ent loss of vitality, other esterase probes such as Cal-
cein-AM and CMFDA or CellTracker Green have
been developed. After the acetoxymethyl (AM) ester
of Calcein-AM is hydrolyzed, the green fluorescent
and charged, membrane-impermeable Calcein is
formed. Calcein-AM has been used to assess the
vitality of virus-infected cells of the phytoplankton
species Phaeocystis pouchetii and Micromonas pusilla
(Brussaard et al. 2001). The chloromethyl group of
CMFDA reacts covalently inside the cell with thiols
such as glutathione and becomes fluorescently
green after cleavage of the acetates by intracellular
esterases. Because the reaction product is cell
impermeable, the green fluorescence should be
retained much better than that of FDA. CMFDA has
been used to track cells in grazing experiments of
3–5 h duration (Stoecker et al. 2000). A fourth
esterase probe is H2DCFDA. This membrane-perme-
able probe produces a green fluorescent dye in the
cell after the acetate groups have been removed by
intracellular esterases and oxidation has occurred
by reactive oxygen species (ROS). Oxidative stress,
an imbalance between oxidants and antioxidants in
favor of the oxidants (Sies 1997), may occur in auto-
trophic plankton under several kinds of environ-
mental stress, such as ultraviolet-B (UVB) radiation
and CO2 limitation (Vardi et al. 1999, He and
Häder 2002).
Another mode of action of vitality probes is based
on membrane potential and integrity. The anionic
oxonol DiBAC4(3) is conventionally used to gauge
the cells’ membrane potential. This probe increases
in fluorescence intensity upon entry into the mem-
brane after depolarization. A second probe based
on membrane integrity is SYTOX-Green, a com-
pound that can only enter cells with a compromised
cell membrane, where it binds to nucleic acids to
form a green fluorescent dye. The vitality probe
SYTOX-Green has been tested on phytoplankton
species from 12 taxonomic classes, but so far not on
Rhodophyceae and Euglenophyceae (Veldhuis et al.
1997).
In the present paper, the six fluorescent probes
were tested for their applicability in measuring phy-
toplankton vitality. To avoid confusion, we define
vitality as ‘‘manifesting life,’’ and viability as ‘‘the
capability to grow.’’ Hence, cells can be (i) vital and
viable, (ii) vital with reduced viability (e.g., as an
effect of a toxic substance), (iii) vital but with no
viability (e.g., after a nutrient has been depleted),
and (iv) nonvital and nonviable (dead).
In total, 34 taxonomically different species, and
multiple strains from two species, were examined to
test the probes in their competence as vital stains.
Species from polar, temperate, and warm-water
regions were included in the experiments to investi-
gate if origin and growth temperature influenced
693
probe performance. Both live (vital) and killed
(nonvital) cells in exponential (viable) and in sta-
tionary growth phase (nonviable) were tested to
examine if the probes were able to distinguish
between growing, nongrowing but live, and dead
cells.
MATERIALS AND METHODS
Phytoplankton. A list of the phytoplankton (including
cyanobacteria) species and strains is given in Table 1. The
prymnesiophytes contained four temperate strains of Phaeocystis
globosa, and in the culture of one (Pg6-1), the nonflagellates
transformed into flagellates during the experiments. Both the
warm-water P. globosa (Ph627) and the Phaeocystis antarctica
culture contained nonflagellate cells. The prasinophytes con-
tained four strains of M. pusilla, two from temperate and two
from warm-water regions (Table 1).
Culturing. All strains were cultured in a 1:1 mix of f ⁄ 2
(Guillard 1975) and enriched seawater, artificial water (Harri-
son et al. 1980) with Tris-HCl and selenium added (Cottrell
and Suttle 1991). Specifically for the diatoms, the medium was
also enriched with 150 lM silicate. Temperate strains were
cultured at 15C; cold-water strains, at 4C; and warm-water
strains, at 22C. Cultures were incubated at a photon flux of
17–26 lmol Æ m)2 Æ s)1 with a 16:8 light:dark (L:D) cycle for the
temperate and cold-water species and a 13:11 L:D cycle for the
warm-water species. Growth was monitored at least three times
per week by measuring in vivo chl autofluorescence and cell
abundance using a Hitachi F2000 spectrofluorimeter (Peper-
zak et al. 2000) and a benchtop flow cytometer, respectively
(see below).
Flow cytometry. Stained cells were analyzed using a standard
Coulter Epics XL-MCL benchtop flow cytometer (Beckman
Coulter Inc., Miami, FL, USA), equipped with an Argon laser
(excitation wavelength 488 nm). Green probe fluorescence
intensity (FL1) was measured at 515 ± 20 nm. Red autofluo-
rescence intensity (far red, FL4) was measured at >610 nm and
was set as the trigger. Median values of green probe fluores-
cence intensity were calculated from cell clusters selected in
FL1*FL4 plots. Living cells and killed controls were measured
at least twice from duplicate cultures.
Forward light scatter (FLS) was measured as a proxy of cell
size (Cunningham and Buonnacorsi 1992). In the case of
large cells, a neutral filter was placed in the FLS light path
that effectively reduced the FLS signal by a factor 9. Fluores-
cent calibration beads (Flow Set and Flow Check, Coulter
Diagnostics, Brea, CA, USA) were used to standardize the
green fluorescence of well-stained FDA-probed strains in
exponential growth phase to correlate FDA fluorescence
intensity with cell size as FLS.
Probe protocols. Vitality probes (all from Invitrogen Inc.,
Carlsbad, CA, USA) were tested on phytoplankton cells in both
exponential and stationary growth phase. Controls consisted of
cells from both growth phases, killed by adding 50 lL 18%
formaldehyde to 1 mL sample and stored at 4C for at least 1 h.
For each probe, a number of initial tests were run on a subset
of species to determine probe concentrations and incubation
specifics (light or dark, temperature, and incubation time). All
tests were performed during fixed intervals, 2–9 h into the light
period. The final protocols that stained the test species best
were used throughout the study and are summarized in
Table 2.
Data analysis. The response (live ⁄ dead ratio) of each strain
to a specific probe was calculated by dividing the green
fluorescence intensity of the living cells by the green fluores-
cence intensity of killed cells in the case of esterase probes and
vice versa for the membrane probes, because the intensity of
694 L O U I S PE P E R Z A K AN D C O R I N A P . D . B R U S S A A R D

TABLE 1. List of phytoplankton classes (n = 12), species (n = 34), strains (n = 40), and culture temperature (C) used to
screen six fluorescent vitality probes.

Class Number Species Strain C

Bacillariophyceae 1 Chaetoceros calcitrans CCMP 1315 15

Bacillariophyceae 2 Ditylum brightwellii Bigelow 358 15
Bacillariophyceae 3 Phaeodactylum tricornutum U. Wollenzien, UTCC162 15
Bacillariophyceae 4 Pseudo-nitzschia pungens G. Casteleyn, UG Belgium 15
Bacillariophyceae 5 Pseudo-nitzschia sp. B. Bontes, NIOZ 4
Bacillariophyceae 6 Skeletonema costatum CCMP781 15
Bacillariophyceae 7 Thalassiosira antarctica B. Bontes, NIOZ 4
Bacillariophyceae 8 Thalassiosira oceanica CCMP1006 22
Chlorophyceae 9 Dunaliella sp. CCMP1320 15
Chlorophyceae 10 Nannochloris sp. CCAP 251 ⁄ 2 15
Cryptophyceae 11 Rhodomonas baltica Hagmeier, Helgoland 15
Cryptophyceae 12 Rhodomonas salina CCMP 1319 15
Cyanophyceae 13 Synechococcus sp. 1 G. Nieuwland, NIOZ 15
Cyanophyceae 14 Synechococcus sp. 2 CCMP1334 15
Dinophyceae 15 Alexandrium minutum CCMP 113 15
Dinophyceae 16 Amphidinium carterae CCMP1314 22
Dinophyceae 17 Gymnodinium simplex A. Noordeloos, NIOZ 15
Dinophyceae 18 Prorocentrum minimum Biol. Hel. Damm E66 15
Dinophyceae 19 Scrippsiella sp. A. Noordeloos, NIOZ 15
Euglenophyceae 20 Eutreptiella cf. gymnastica CCMP1594 15
Euglenophyceae 21 Eutreptiella marina CCMP390 15
Eustigmatophyceae 22 Nannochloropsis salina CCAP 849 ⁄ 4 15
Pelagophyceae 23 Pelagococcus subviridis CCMP 1429 15
Prasinophyceae 24 Micromonas pusilla Mp38, Roscoff 15
Prasinophyceae 25 M. pusilla Mp1545, Roscoff 15
Prasinophyceae 26 M. pusilla Mp449, Roscoff 22
Prasinophyceae 27 M. pusilla Mp450, Roscoff 22
Prasinophyceae 28 Pyramimonas sp. A. Buma, NIOZ 4
Prasinophyceae 29 Tetraselmis suecica CCMP883 15
Prymnesiophyceae 30 Chrysochromulina polylepis Oldenburg 15
Prymnesiophyceae 31 Emiliania huxleyi T. Ietswaart, NIOZ 15
Prymnesiophyceae 32 Isochrysis galbana CCMP 1323 15
Prymnesiophyceae 33 Phaeocystis antarctica (nonflags) B. Bontes, NIOZ 4
Prymnesiophyceae 34 Phaeocystis globosa (nonflags) CCMP627 22
Prymnesiophyceae 35 P. globosa (flags) RIKZ Ph91mf, NIOZ Pg2 15
Prymnesiophyceae 36 P. globosa (flags) NIOZ Pg5 15
Prymnesiophyceae 37 P. globosa (nonflags, flags) Pg6-1, L. Peperzak, NIOZ 15
Raphidophyceae 38 Heterosigma akashiwo K. de Boer, RUG 15
Rhodophyceae 39 Dixoniella grisea U. Wollenzien, CEME 15
Rhodophyceae 40 Porphiridium cruentum U. Wollenzien, CEME 15
The 40 strains belong to four cold-water, four warm-water, and 26 temperate species. The strain numbers (1–40) appear in the
results of statistical analyses.

TABLE 2. Final probe incubation protocols.

Probe Working stock solvent Probe added (lL Æ mL)1) Final (lM) Incubation T Incubation time (min)

Calcein-AM Acetone 10 10 L CT 60 + ice

CMFDA Acetone 10 10 L CT 60 + ice
DiBAC4(3) EtOH 10 2 D RT <30
FDA Acetone 10 10 L CT 60 + ice
H2DCFDA Acetone 10 10 L CT 60 + ice
SYTOX-Green Milli-Q 10 0.5 D RT 5
Hydrophobic probes were dissolved in acetone or in ethanol. Of each probe, 10 lL was added to 1 mL sample, yielding final
concentrations ranging from 0.5 to 10 lM. Incubations were performed in the light (L) or dark (D) at culture (CT) or room tem-
perature (RT). After 60 min of incubation, the samples stained with esterase probes were stored on ice before measurement.
Calcein-AM, acetoxymethyl ester of Calcein; CMFDA or CellTracker Green, 5-chloromethylfluorescein diacetate; DiBAC4(3), bis-
(1,3-dibutylbarbituric acid) trimethine oxonol; FDA, fluorescein diacetate; H2DCFDA, 2¢,7¢-dichlorofluorescein diacetate.

their green fluorescence increases in dead cells (Fig. 1). The weak signal only; ++ meant an intermediate result, while +++
median value of the live ⁄ dead ratios (n ‡ 2) was coded as: 0 indicated a more than 10-fold difference in signal between
(ratio <2), + (ratio £5), ++ (ratio £10), and +++ (ratio >10). dead and live cells. To identify the strains with the highest
Zero and + meant that the probe did not work or produced a live ⁄ dead ratios, a simple individual ranking, for both
 V I T A L I T Y P R O B E S F O R P H Y T O PL A N K T O N 695

FIG. 1. Examples of flow cyto-

metry dot plots of three vitality
probes: esterase probe fluorescein
diacetate (FDA) (a, b), membrane
potential probe bis-(1,3-dibutyl-
barbituric acid) trimethine oxonol
[DiBAC4(3)] (c, d), and mem-
brane integrity probe SYTOX-
Green (e, f). Samples in the left
column (a, c, e) were alive; sam-
ples in the right column had been
fixed with formaldehyde (b, d, f).
Samples consisted of a prymnesio-
phyte (Phaeocystis globosa: a, b), a
dinoflagellate (Alexandrium minu-
tum: c, d), and a prasinophyte
(Micromonas pusilla: e, f). FL4, flu-
orescence 4 (phytoplankton auto-
fluorescence); FL1, fluorescence 1
(green probe fluorescence).

exponential and stationary growth phase cells, was done after
adding up all coded values (+, ++, or +++) for the six probes
(maximum sum = 18).
The effect of taxonomic differences and of differences in
strain origin (polar, temperate, and warm water) on the
live ⁄ dead ratios among the 12 classes of phytoplankton was
analyzed using the coded relative green fluorescence (0, +, ++,
+++) in the multivariate statistical program PRIMERTM version
6 (PRIMER-E Ltd., Lutton, UK). The first step of the analysis,
using the results of all six probes and both growth phases, was
the calculation of a Euclidean distance matrix, a two-dimen-
sional table containing the differences between all the strains
tested. The distances in this matrix express that strains that
have high live ⁄ dead ratios in both growth phases for all probes
are less distant to each other than strains that differ between
growth phases or that have low live ⁄ dead ratios. Secondly, the
Euclidean distance matrix was used in a cluster analysis with a
simultaneous SIMPROF (similarity profile) permutation test
(n = 999). The SIMPROF test calculates if individual strains are
significantly (P < 0.05) different from each other. The third
step consisted in visualizing the Euclidean distance matrix in a
nonmetric multidimensional scaling (NMS) diagram. Basically,
696 L O U I S PE P E R Z A K AN D C O R I N A P . D . B R U S S A A R D

the rank differences between the strains in the matrix were
used to determine their position in a two-dimensional ordina-
tion, such that dissimilarities between strains result in further
distances apart. Plotting cluster boundaries, based on the
cluster analysis in the second step, emphasized the similarities
between strains in the NMS diagram. The Euclidean distance
matrix was also used to investigate the null hypotheses that
there are no differences between groups (class or origin) using
a one-way analysis of similarities (ANOSIM). The significance
of the ANOSIM test statistic R was computed by a permutation
(n = 999) test.
To examine the staining characteristics between cells in
exponential and in stationary growth phase, the differences in
live ⁄ dead ratios between both growth phases of all species were
investigated for each probe separately using a Kruskal–Wallis
test. Furthermore, the differences between the six probes were
tested simultaneously with a Kolmogorov–Smirnov test. Both
nonparametric tests, as well as linear regression analyses of the
FDA probe were performed in SYSTATTM version 12 (SYSTAT
Software, Chicago, IL, USA). The linear regression of FDA
fluorescence intensity and live ⁄ dead ratios of exponentially
growing cells was performed to test the possibility that both
may not only be dependent on taxonomy and origin but also
on cell size.
RESULTS
Species differences. The live ⁄ dead fluorescence
ratios of the 40 plankton strains tested varied widely,
depending on species and probe (Table 3). Exam-
ples of flow cytometer dot plots are provided in
Figure 1. For some species considerable different

TABLE 3. List of phytoplankton species and response to six different probes in either exponential (exp) or stationary (stat)
growth phase.

FDA CMFDA H2DCFDA Calcein-AM DiBAC4(3) SYTOX

Species Exp Stat Exp Stat Exp Stat Exp Stat Exp Stat Exp Stat

Chaetoceros calcitrans +++ ++ +++ +++ + ++ ++ 0 + +++ +++ +++

Ditylum brightwellii +++ +++ +++ +++ +++ +++ +++ +++ 0 0 0 +++
Phaeodactylum tricornutum +++ 0 +++ 0 0 0 0 0 ++ +++ 0 +++
Pseudo-nitzschia pungens +++ +++ +++ +++ ++ + +++ +++ + ++ +++ +++
Pseudonitzschia sp. +++ +++ +++ +++ ++ + +++ +++ 0 + + ++
Skeletonema costatum +++ ++ +++ +++ + 0 +++ +++ + ++ ++ +++
Thalassiosira antarctica +++ +++ +++ +++ 0 0 + + + ++ +++ +++
Thalassiosira oceanica +++ +++ +++ +++ +++ ++ +++ +++ + 0 ++ +
Dunaliella sp. +++ +++ +++ +++ 0 0 0 0 +++ +++ +++ +++
Nannochloris sp. 0 0 0 0 0 0 + + 0 0 0 0
Rhodomonas baltica +++ +++ +++ +++ ++ +++ 0 0 +++ +++ +++ +++
Rhodomonas salina +++ +++ +++ +++ + 0 0 0 +++ +++ +++ +++
Synechococcus sp. 1 0 0 0 0 0 0 + + 0 0 0 0
Synechococcus sp. 2 0 0 0 0 0 0 0 0 0 0 0 0
Alexandrium minutum +++ +++ +++ +++ +++ +++ 0 0 ++ + ++ +++
Amphidinium carterae +++ +++ +++ ++ 0 0 0 + +++ +++ + +++
Gymnodinium simplex +++ +++ 0 + 0 0 + ++ +++ +++ ++ ++
Prorocentrum minimum +++ +++ +++ +++ +++ +++ 0 0 +++ +++ +++ +++
Scrippsiella sp. +++ +++ +++ +++ +++ + 0 0 + + +++ +++
Eutreptiella cf. gymnastica +++ +++ +++ +++ 0 0 0 0 +++ +++ + 0
Eutreptiella marina +++ +++ +++ +++ ++ ++ 0 0 +++ +++ + 0
Nannochloropsis salina + 0 0 ++ 0 0 0 + 0 0 0 0
Pelagococcus subviridis +++ +++ +++ +++ 0 + + 0 +++ +++ +++ +++
Micromonas pusilla Mp38 +++ + +++ +++ ++ +++ + 0 +++ +++ +++ +++
M. pusilla Mp1545 +++ + +++ +++ ++ + 0 + +++ +++ +++ +++
M. pusilla Mp449 +++ ++ +++ + 0 ++ + ++ +++ ++ +++ +++
M. pusilla Mp450 +++ + +++ + ++ ++ + + ++ ++ +++ +++
Pyramimonas sp. +++ +++ +++ +++ 0 0 0 0 +++ +++ 0 +
Tetraselmis suecica +++ +++ +++ +++ 0 0 0 0 0 + +++ ++
Chrysochromulina polylepis + + 0 0 0 0 0 0 +++ +++ +++ +++
Emiliania huxleyi +++ +++ +++ +++ 0 0 0 0 ++ 0 +++ +++
Isochrysis galbana +++ +++ +++ +++ + + +++ 0 ++ 0 ++ +++
Phaeocystis antarctica +++ +++ +++ +++ 0 ++ + + +++ + +++ +++
Phaeocystis globosa Ph627 +++ +++ +++ +++ ++ + ++ +++ +++ +++ +++ +++
P. globosa Ph91mf +++ +++ +++ +++ +++ ++ ++ + ++ + +++ +++
P. globosa Pg5 +++ + +++ ++ ++ + +++ + ++ 0 +++ +++
P. globosa Pg6 fl +++ +++ +++ +++ 0 ++ 0 0 +++ ++ +++ +++
P. globosa Pg6 nfl – – +++ – ++ – – – +++ 0 +++ +++
Heterosigma akashiwo +++ + +++ +++ +++ +++ 0 0 +++ + +++ ++
Dixoniella grisea +++ +++ +++ +++ 0 0 0 0 +++ +++ + +
Porphiridium cruentum +++ + +++ ++ 0 0 0 0 +++ +++ + 0
The response is the ratio of green fluorescence of live cells compared to dead controls.
+++, ratio >10; ++, ratio 5 £ 10; +, ratio 2 £ 5; 0, ratio <2; –, not measured.
Calcein-AM, acetoxymethyl ester of Calcein; CMFDA or CellTracker Green, 5-chloromethylfluorescein diacetate; DiBAC4(3),
bis-(1,3-dibutylbarbituric acid) trimethine oxonol; FDA, fluorescein diacetate; H2DCFDA, 2¢,7¢-dichlorofluorescein diacetate.
 V I T A L I T Y P R O B E S F O R P H Y T O PL A N K T O N 697

live ⁄ dead ratios were observed between the esterase
probes. For example, all growth phases of Gymnodi-
nium simplex showed better staining using FDA than
CMFDA, whereas stationary cells of M. pusilla (both
strain Mp38 and Mp1545) showed higher live ⁄ dead
ratios with CMFDA than with FDA.
The overall response of all phytoplankton strains
to all six probes resulted in four significantly (SIM-
PROF, P < 0.05) different clusters in the NMS ordi-
nation (Fig. 2a). Synechococcus spp., Nannochloris sp.,
and Nannochloropsis salina, which had no or low
probe responses appeared in cluster I (Fig. 2a).
Cluster II contained three species from three classes
(classes given in Table 1), Phaeodactylum tricornutum,
Chrysochromulina polylepis, and G. simplex, which did
not stain with H2DCFDA and stained hardly when

2D Stress: 0.12 P<0.05

a 2 d
c
b
a
Distance
8 5.3
5 IV

I 6
36
4
10 + 13
10
13 32
22 3435
14
29 7 27 19
1 15
31 26
33
38
II 17
2337 25 24
11 18
16 912
39
30
40 28 21 III
3 20

2D Stress: 0.12 Class

b 2 Bacillariophyceae
Chlorophyceae
Cryptophyceae
Cyanophyceae
Dinophyceae
8 Euglenophyceae
5 IV Eustigmatophyceae
Pelagophyceae
Prasinophyceae
Prymnesiophyceae
Raphidophyceae
6 4
I 36 Rhodophyceae

1013 Distance
10
+ 13 32 5.3
22 3435
14
29 7 27 19
1 15
31 26
33
38
II 17
2337 25 24
11 18
16 912

30 39
40 28 21 III
3 20

FIG. 2. Differences among phytoplankton strains and classes: nonmetrical dimensional scaling (NMS) ordination of the live ⁄ dead fluo-
rescence ratios of 40 autotrophic plankton strains stained with six different vitality probes. Symbols depict significant (SIMPROF, P < 0.05)
differences between either phytoplankton species or classes. Numbers represent the phytoplankton species according to Table 1. (a) The
four clusters (I–IV) encompass the significant (P < 0.05) differences between strains (a, b, c, d) based on live ⁄ dead ratios measured with
six different probes. Cluster I contains species that were not stained; cluster IV contains species that were very successfully stained. (b)
The same ordination as in (a), but now the symbols depict the 12 phytoplankton classes of this study. Cluster IV is the least diverse clus-
ter, containing only bacillariophytes (diatoms) and prymnesiophytes.
698 L O U I S PE P E R Z A K AN D C O R I N A P . D . B R U S S A A R D

using CMFDA (Table 3). The majority of the species strains did not respond differently from each other
fell in clusters III and IV, staining well with CMFDA (R = )0.1, n.s.). Individual nonparametric tests
and FDA. Cluster III is the most diverse; it con- revealed that there were no significant differences
tained euglenophytes and rhodophytes that had in staining between cells in exponential and station-
higher live ⁄ dead ratios with DiBAC4(3) than with ary growth phase, with the exception of FDA (Krus-
SYTOX-Green. Cluster IV contains a number of dia- kal–Wallis, P < 0.05). However, the difference
toms and prymnesiophytes with high esterase live ⁄ between exponential and stationary phase cells
dead ratios, including Calcein-AM (Table 3). stained with FDA was <20% (Fig. 3).
The taxonomic differences in probe success in To identify the species and strains that had the
Figure 2b were confirmed by an ANOSIM, showing highest live ⁄ dead ratios, the results (1+, 2+, 3+) of
a significant difference between the 12 plankton all six probes were summed and ranked. The top
classes (R = 0.31, P < 0.01). In contrast, the three 33% of strains investigated scored ‡14, which is
groups of cold-water, temperate, and warm-water 88% of the maximum sum of 18. These 13 top-rank-
ing strains with high live-dead ratio strains belonged
to six classes of phytoplankton (Table 4). Three P.
1.0
globosa strains, one of warm-water and two of tem-
exponential perate origin, achieved the highest ranks. The high-
stationary est-ranking cold-water species, not belonging to the
0.8 top 33%, were another Phaeocystis strain, P. antarc-
tica, and the diatom Pseudo-nitzschia sp. (sum = 13
Relative probe intensity

0.6
0.4
0.2
0.0
FDA CMFDA H2DCFDA Calcein-AM
Probe
FIG. 3. Comparison of the relative fluorescence intensity for
six vitality probes tested on 40 strains of autotrophic phytoplank-
ton, in exponential and stationary growth phase, respectively. The
relative probe fluorescence (rpf) per growth phase was calculated
as the sum of the species ratios (0 to 3+) in Table 3 divided by
the total maximum fluorescence (number of species tested times
3+), yielding 0 £ rpf £ 1. Calcein-AM, acetoxymethyl ester of Cal-
cein; CMFDA, 5-chloromethylfluorescein diacetate; DiBAC4(3),
bis-(1,3-dibutylbarbituric acid) trimethine oxonol; FDA, fluores-
cein diacetate; H2DCFDA, 2¢,7¢-dichlorofluorescein diacetate.
and 12, respectively). The lowest ranking species,
hardly reacting to any probe (sum £3), were both
strains of the cyanophyte Synechococcus, the chloro-
phyte Nannochloris sp., and the eustigmatophyte
N. salina, confirming their presence in cluster I of
Figure 2.
Probe differences. Comparing of the six stains
showed that the fluorescence intensities were rela-
tively high for the esterase probes FDA and CMFDA
DiBAC SYTOX
and the membrane probes DIBAC4(3) and SYTOX-
Green, and relatively low for the esterase probes
H2DCFDA and Calcein-AM (Fig. 3). Generally, SY-
TOX-Green and DiBAC4(3) stained most of the
tested phytoplankton species (Table 3). SYTOX-
Green performed better than DiBAC4(3) in six of
eight prymnesiophyte strains (including four strains
of Phaeocystis). Conversely, DiBAC4(3) stained the
prasinophyte Pyramimonas sp., both euglenophytes
(Eutreptiella cf. gymnastica and E. marina) and
both rhodophytes (Dixoniella grisea and Porphiridium

TABLE 4. Strains with the highest live ⁄ dead ratios measured with six vitality probes.

Class Species Origin Growth phase Score

Prymnesiophyceae Phaeocystis globosa Ph627 w Exp + stat 16 + 16

Prymnesiophyceae P. globosa Pg2 t Exp 16
Prymnesiophyceae P. globosa Pg5 t Exp 16
Bacillariophyceae Ditylum brightwellii t Stat 15
Raphidophyceae Heterosigma akashiwo t Exp 15
Prasinophyceae Micromonas pusilla Mp38 t Exp 15
Dinophyceae Prorocentrum minimum t Exp + stat 15 + 15
Bacillariophyceae Pseudo-nitzschia pungens t Exp + stat 15 + 15
Cryptophyceae Rhodomonas baltica t Exp + stat 14 + 15
Bacillariophyceae Thalassiosira oceanica w Exp 15
Prymnesiophyceae Isochrysis galbana t Exp 14
Prasinophyceae M. pusilla Mp1545 t Exp 14
Prasinophyceae M. pusilla Mp450 w Exp 14
The cells were either in exponential (exp) or stationary (stat) growth phase. The maximum score for each growth phase sepa-
rately is 18 (6 · 3), that is, six probes times three, which is the maximum live ⁄ dead ratio from Table 3. Four species scored high
in both exponential and stationary growth phases. All species listed are from temperate (t) or warm (w) origin and were cultured
at 15C or 22C.
 V I T A L I T Y P R O B E S F O R P H Y T O PL A N K T O N 699

Table 5. Differences between the six probes tested using the live ⁄ dead ratios (0, 1+, 2+, 3+) of all 40 strains.

CMFDA Calcein-AM DiBAC4(3) FDA H2DCFDA SYTOX

CMFDA –
Calcein-AM 0.62** –
DiBAC4(3) 0.30* 0.42** –
FDA 0.04 0.59** 0.27* –
H2DCFDA 0.62** 0.16 0.32* 0.59** –
SYTOX 0.17 0.50** 0.13 0.14 0.46** –
Exponential and stationary growth phase data were pooled. Significance of differences: *, P < 0.01; **, P < 0.001 (Kolmogorov–
Smirnov test).
Calcein-AM, acetoxymethyl ester of Calcein; CMFDA or CellTracker Green, 5-chloromethylfluorescein diacetate; DiBAC4(3), bis-
(1,3-dibutylbarbituric acid) trimethine oxonol; FDA, fluorescein diacetate; H2DCFDA, 2¢,7¢-dichlorofluorescein diacetate.

2D Stress: 0.13 Class

Ph627 Prasinophyceae
Prymnesiophyceae
Distance
3
Mp449 Phant 3.5
Pg2
Pg6fl
Pg5 Mp450

Mp1545

Mp38

FIG. 4. Comparison of five strains of the genus Phaeocystis (Prymnesiophyceae) and four strains of the species Micromonas pusilla (Prasin-
ophyceae) stained with six vitality probes in one nonmetrical dimensional scaling (NMS) ordination. Pg = Phaeocystis globosa strains Pg2,
Pg5, and Pg6-flagellates; Ph627 = P. globosa strain CCMP627; Phant = Phaeocystis antarctica; Mp = M. pusilla (see Table 1) of origins: cold
water (bold), temperate (italic), and warm water (underlined).

cruentum) better than SYTOX-Green, regardless of
their growth phases.
A nonparametric test was performed to investi-
gate live ⁄ dead ratio differences between the various
probes. Of the four esterase probes, Calcein-AM
and H2DCFDA were not significantly different from
each other, and both probes performed less than
FDA and CMFDA (not significantly different from
each other; Table 5). The two membrane probes,
DIBAC4(3) and SYTOX-Green, were not signifi-
cantly different from each other (Table 5), although
species-specific differences did occur (Table 3).
DIBAC4(3) did differ significantly from all esterase
probes, whereas SYTOX-Green differed only signifi-
cantly from Calcein-AM and H2DCFDA.
Intraspecific differences. An intraspecific compari-
son of live ⁄ dead ratios from six probes was made
for the different species and strains of Phaeocystis
and M. pusilla. This comparison was made in one
ordination (Fig. 4) because both Phaeocystis and Mi-
cromonas had relatively high live ⁄ dead ratios
(Tables 3 and 4), showing how the intraspecific
differences compared to interspecific differences.
Phaeocystis consisted of cold-water, temperate, and
warm-water strains, M. pusilla of temperate and
warm-water strains (Table 1). P. antarctica (cold-
water) and P. globosa Pg6 (temperate) showed a
more similar response to all six probes compared to
the two temperate (Pg2 and Pg5) and the warm-
water strain (Ph627) (Fig. 4). There were only small
differences in live ⁄ dead ratios between the temper-
ate and warm-water strains of M. pusilla, so their
positions in the NMS diagram were very close
(Fig. 4). Overall, the intraspecific differences
between strains or species were comparable to the
differences between Phaeocystis and Micromonas.
Kruskal–Wallis and SIMPROF tests confirmed that
the intraspecific differences for Phaeocystis as well as
for Micromonas were not significant.
Size. To test the possibility that small cells, with
a lower esterase content than large cells, have a
relatively low green FDA-fluorescence and hence a
low live ⁄ dead ratio, both fluorescence intensity and
live ⁄ dead ratios were correlated with cell size. Indeed,
the green fluorescence intensities of FDA-probed
cells were significantly (P < 0.001) correlated to cell
700 L O U I S PE P E R Z A K AN D C O R I N A P . D . B R U S S A A R D

3.6 6.0 10.0 19.6

DISCUSSION
3 This study provides a screening of the applicabil-
a
2 ity of six fluorescent vitality stains for phytoplank-
15
FDA fluorescence (log FL1)

2 ton. Although most of these stains have been used

18 in algal studies, they were tested simultaneously on
19
21
38 a large range of algal species, originating from dif-
392820
1 11
17
12
ferent locations and at different growth stages. The
8 results show significant differences between classes
40 and species for the combined live ⁄ dead ratios of all
32 29 1
31 5
0 37 34 six probes, but not between strains of one species.
23 9
33 6 16 4 Of the four esterase probes, FDA and CMFDA per-
35
3 formed best. The two membrane probes DIBAC4(3)
-1 7 r 2 = 0.72
and SYTOX-Green performed well for many species,
36 while showing complementary performance for rho-
25 P < 0.001
27 24 dophytes and euglenophytes.
-2 Applicability of the probes. Ideally, the best applica-
26 ble vitality probe should be capable of staining all
species regardless of class, origin, or growth rate.
-3 The data show that the six probes had varying
success in measuring vitality among the 12 phyto-
4 plankton classes and 34 species tested. A multivari-
b ate analysis made clear that probe applicability was
21 significantly related to phytoplankton class. A few
31
23 34 28 species, belonging to three classes, Synechococcus
Log FDA live/dead ratio

32 8
spp., Nannochloris sp., and N. salina, had very poor
3 12
staining characteristics and belonged to one multi-
11 20
18 15
35 variate cluster. On the other hand, most species
average could be grouped in two clusters with good staining
24 37 7 33 17 1 2 characteristics, bringing the extent of probe applica-
6
29 38 5
2 27
26 3
9
16 19 bility to nine ecologically relevant classes of phyto-
40
25 plankton.
36 4 Variation in the staining characteristics between
different phytoplankton species has been reported
39
1 (Veldhuis et al. 1997, Brussaard et al. 2001, Zhang
et al. 2007). We found, for example, comparable
staining difficulties for Nannochloris sp. and N. salina.
Difficulties in staining Synechococcus were not
restricted to SYTOX-Green (see also Veldhuis et al.
0 1997) but included also the other probes tested. A
0 1 2 3 4 solution to low-staining intensities could be adapta-
Size (log FLS) tion of the standard protocols used here. For
instance, Zhang et al. (2007) were able to probe the
FIG. 5. (a) Correlations between green fluorescence intensity cyanophyte Microcystis aeruginosa with FDA but only
of fluorescein diacetate (FDA)–probed cells and cell size (n = 35) at a much higher final concentration (25 mM), that
and (b) the live ⁄ dead ratios of FDA-probed cells and cell size.
FLS, forward light scatter; and FL1, green fluorescence intensity. is, at a probe concentration a factor 1,000 higher
The smallest strains belong to Micromonas pusilla (24–27), while than used in the present investigation. In natural
the largest species investigated was Ditylum brightwellii. On top of samples with mixed species assemblages, these spe-
(a), the FLS of four latex reference beads with their diameter in cies-specific differences in probe reactivity must be
lm is indicated with arrows. The line in (a) is the linear regres-
sion line (r2 = 0.72, P < 0.001). No correlation was found in (b);
recognized when using general probe protocols to
instead the average live ⁄ dead ratio was plotted. Numbers are prevent false results, that is, declare a species dead
strains in Table 1. Five strains with no or only a weak signal were when it is not (well) stainable.
excluded (Table 3). For all stains tested, except FDA, there was no sig-
nificant difference in live ⁄ dead ratios between expo-
sizes measured as FLS (Fig. 5a). However, there was nential and stationary cells. For the FDA-stained
no correlation between cell size and the live ⁄ dead cells, this difference was <20%. Therefore in most
ratio (Fig. 5b). The smallest cells, the M. pusilla species, the probes are well suited to gauge the dif-
strains, had high live ⁄ dead ratios (102) that were ference between live and dead cells. Our results
comparable to the live ⁄ dead ratio of the largest indicate that the selected probes in combination
species tested: Ditylum brightwellii (Fig. 5b). with flow cytometry will most likely be applicable for
 V I T A L I T Y P R O B E S F O R P H Y T O PL A N K T O N
determining phytoplankton vitality in natural sam-
ples from a wide range of marine pelagic ecosystems
and environmental stressors.
Probe comparisons. Gauging phytoplankton vitality
does not need six probes but rather the application
of only one or two that meet species- or class-specific,
as well as experimental, requirements. The highest
live ⁄ dead ratios were obtained with two esterase
probes that were not significantly different from each
other: FDA and CMFDA. Two esterase probes, Calce-
in-AM and H2DCFDA, can be used satisfactorily when
studying diatoms and prymnesiophytes. Calcein-AM
works equally well in certain diatoms, as do FDA and
CMFDA. However, in most classes, the live ⁄ dead
ratios obtained with Calcein-AM are lower than FDA
and CMFDA, which might tentatively be related to
the higher molecular weight of Calcein-AM
(MW = 995), hence its permeability, as compared to
that of FDA and CMFDA (MW = 416 and 465, respec-
tively). The application of PluronicTM as suggested
by Invitrogen Inc., a surfactant that promotes the sol-
ubilization of Calcein-AM, did not significantly
enhance live ⁄ dead ratios (L. Peperzak and C. P. D.
Brussaard, unpublished data). Fluorescein from
FDA, however, may leak rapidly out of cells. In exper-
iments where cell vitality cannot be measured imme-
diately, Calcein-AM and CMFDA may, therefore, be
considered useful alternatives to FDA.
H2DCFDA has a low molecular weight compara-
ble to FDA and CMFDA (MW = 487), but this probe
needs oxidation by ROS. ROS can be produced
under stress, for example, due to low CO2 concen-
trations, UVB radiation, and toxic metals (Vardi
et al. 1999, He and Häder 2002, Knauert and
Knauer 2008). In the present study, H2DCFDA
yielded high ratios of fluorescently stained: non-
stained cells, even in exponentially growing cells, of
certain species of prasinophytes, dinoflagellates
(e.g., Alexandrium minutum) as well as the raphido-
phyte Heterosigma akashiwo. Interestingly, A. minutum
and H. akashiwo have been shown to produce high
amounts of the ROS species superoxide under nor-
mal circumstances, which suggests a defense mecha-
nism against bivalve grazing (Marshall et al. 2005).
On the contrary, typical aquaculture bivalve feeds
such as Dunaliella, Tetraselmis, and Nannochloropsis,
showed extremely low live ⁄ dead ratios with
H2DCFDA (this study), corresponding to negligible
production of superoxide (Marshall et al. 2005).
Therefore, H2DCFDA in conjunction with flow
cytometry may be used as a cell-specific alternative
to bulk ROS measurements on luminometers (Mar-
shall et al. 2005), providing added value to studying
ROS production.
Instead of esterase activity, DIBAC4(3) and
SYTOX-Green measure membrane polarization and
integrity, and one of them may therefore be chosen
as a second independent vitality probe to gauge cell
vitality. Interestingly, DiBAC4(3) stained the eugle-
nophytes and rhodophytes very well, in contrast to
701
SYTOX-Green. In addition, SYTOX-Green appeared
a better vitality probe for diatoms and prymnesio-
phytes than DiBAC4(3). The underlying reason for
these differing results is unknown. As with the ester-
ase probes, Synechococcus spp., Nannochloris sp., and
N. salina were not stained with DIBAC4(3) and
SYTOX-Green. Possible solutions to this problem
might be the addition of 0.05% Triton X-100
(Veldhuis et al. 1997) in the case of SYTOX-Green
or a change in the general protocols, that is, by
increasing the probe concentration as employed in
the present study.
Flow cytometry. Flow cytometers are capable of
handling a wide range in cell sizes and fluorescence
intensities in a fast and reproducible manner. The
species investigated spanned nearly three orders of
magnitude in size and four orders in green FDA flu-
orescence. The significant correlation between cell
size and green fluorescence (r2 = 0.72, P < 0.001)
indicates that the activity of the nonspecific intracel-
lular esterases that hydrolyze FDA to green fluores-
cent fluorescein is related to cell biomass. Because
the red chl autofluorescence also increased with cell
size, green fluorescein-fluorescence (FL1) and red
autofluorescence (FL4) were correlated to each
other as log FL1 = 0.77 + 1.08*log FL4, r2 = 0.61,
P < 0.001 (L. Peperzak and C. P. D. Brussaard,
unpublished data). This correlation explains why in
fluorescence microscopy, red autofluorescence in
both small (P. tricornutum and P. globosa) and large
(Coscinodiscus granii) cells may be too bright to
observe the fluorescein fluorescence (Garvey et al.
2007). In microscopy, the problem of red autofluo-
rescence overpowering green fluorescein fluores-
cence will be exaggerated when cells in stationary
growth phase are studied or when less intense green
fluorescent probes are used. On the other hand,
FDA live ⁄ dead ratios were independent of cell size
meaning that flow cytometry is a technique capable
of rapid, single-cell vitality analysis in a wide range
(2–20 lm) of cell dimensions. Using flow cytometry,
well-stainable model species such as P. globosa and
M. pusilla, and probes such as FDA or SYTOX-
Green, a rapid method emerges to measure the
effect of natural and anthropogenic toxicants on
phytoplankton vitality.
CONCLUSION
The vitality of the majority of phytoplankton spe-
cies can easily be probed flow cytometrically by the
esterase probes FDA or CMFDA and by the mem-
brane probes DIBAC4(3) and SYTOX-Green. The
live ⁄ dead ratio of the green probe fluorescence is
generally not influenced by origin (polar, temper-
ate, or warm water) or growth phase. Because live ⁄
dead ratios are independent of cell size, the use of
fluorescent probes combined with flow cytometry
provides a powerful tool in measuring vitality in
even the smallest phytoplankton species.
702 L O U I S PE P E R Z A K AN D C O R I N A P . D . B R U S S A A R D
We gratefully acknowledge the following colleagues for donat-
ing phytoplankton strains: A. A. M. Noordeloos and B. Bontes
(NIOZ, Texel), U. I. A. Wollenzien (CEME, Yerseke), M. K.
de Boer (RuG, Groningen), and G. Casteleyn (UG, Gent).
This work was supported by a grant from the FP6 European
Framework program (STREP project FACEiT, contract num-
ber 018391).
Berges, J. A. & Falkowski, P. G. 1998. Physiological stress and cell
death in marine phytoplankton: induction of proteases in re-
sponse to nitrogen or light limitation. Limnol. Oceanogr.
43:129–35.
Brussaard, C. P. D. 2004. Viral control of phytoplankton popula-
tions – a review. J. Eukaryot. Microbiol. 51:125–38.
Brussaard, C. P. D., Brookes, R., Noordeloos, A. A. M. & Riegman,
R. 1998. Recovery of nitrogen-starved cultures of the diatom
Dytilum brightwellii (Bacillariophyceae) upon nitrogen resup-
ply. J. Exp. Biol. Ecol. 227:237–50.
Brussaard, C. P. D., Marie, D., Thyrhaug, R. & Bratbak, G. 2001.
Flow cytometric analysis of phytoplankton viability following
viral infection. Aquat. Microb. Ecol. 26:157–66.
Cloern, J. E. 1996. Phytoplankton bloom dynamics in coastal eco-
systems: a review with some general lessons from sustained
investigation of San Francisco Bay, California. Rev. Geophys.
34:127–68.
Cottrell, M. T. & Suttle, C. A. 1991. Wide-spread occurrence and
clonal variation in viruses which cause lysis of a cosmopolitan,
eukaryotic marine phytoplankter Micromonas pusilla. Mar. Ecol.
Prog. Ser. 78:1–9.
Cunningham, A. & Buonnacorsi, G. A. 1992. Narrow-angle forward
light scattering from individual algal cells: implications for size
and shape discrimination in flow cytometry. J. Plankton Res.
14:223–34.
Dorsey, J., Yentsch, C. M., Mayo, S. & McKenna, C. 1989. Rapid
analytical technique for the assessment of cell metabolic
activity in marine microalgae. Cytometry 10:622–8.
Franklin, D. J., Brussaard, C. P. D. & Berges, J. A. 2006. What is the
role and nature of programmed cell death in phytoplankton
ecology? Eur. J. Phycol. 20:160–70.
Franklin, N. M., Stauber, J. L. & Adams, S. A. 2005. Improved
methods of conducting microalgal bioassays using flow
cytometry. In Ostrander, G. K. [Ed.] Techniques in Aquatic
Toxicology. CRC Publishers, Boca Raton, Florida, pp. 735–55.
Franklin, N. M., Stauber, J. L. & Lim, R. P. 2001. Development of
flow cytometry-based algal bioassays for assessing toxicity of
copper in natural waters. Environ. Toxicol. Chem. 20:160–70.
Garvey, M., Moriceau, B. & Passow, U. 2007. Appicability of the FDA
assay to determine the viability of marine phytoplankton un-
der different environmental conditions. Mar. Ecol. Prog. Ser.
352:17–26.
Gilbert, F., Galgani, F. & Cadiou, Y. 1992. Rapid assessment of
metabolic activity in marine microalgae: application in eco-
toxicological tests and evaluation of water quality. Mar. Biol.
112:199–205.
Guillard, R. R. L. 1975. Culture of phytoplankton for feeding
marine invertebrates. In Smith, W. L. & Chanley, M. H. [Eds.]
Culture of Marine Invertebrate Animals. Plenum Pub., New York,
pp. 29–60.
Harrison, P. J., Waters, R. E. & Taylor, F. J. R. 1980. A broad
spectrum artificial sea water medium for coastal and open
ocean phytoplankton. J. Phycol. 16:28–35.
He, Y.-Y. & Häder, D.-P. 2002. Reactive oxygen species and
UV-B: effect on cyanobacteria. Photochem. Photobiol. Sci. 1:729–
36.
Humphreys, M. J., Allman, R. & Lloyd, D. 1994. Determination of
the viability of Trichomonas vaginalis using flow cytometry.
Cytometry 15:343–8.
Jochem, F. J. 1999. Dark survival strategies in marine phytoplankton
assessed by cytometric measurement of metabolic activity with
fluorescein diacetate. Mar. Biol. 135:721–8.
Knauert, S. & Knauer, K. 2008. The role of reactive oxygen species
in copper toxicity to two freshwater green algae. J. Phycol.
44:311–9.
Lage, O. M., Sansonetty, F., O’Connor, J.-E. & Parente, A. M. 2001.
Flow cytometric analysis of chronic and acute toxicity of cop-
per(II) on the marine dinoflagellate Amphidinium carterae.
Cytometry 44:226–35.
Marshall, J.-A., de Salas, M., Oda, T. & Hallegraeff, G. 2005.
Superoxide production by marine microalgae I. Survey of 37
species from 6 classes. Mar. Biol. 147:533–40.
Peperzak, L., Duin, R. N. M., Colijn, F. & Gieskes, W. W. C. 2000.
Growth and mortality of flagellates and non-flagellate cells of
Phaeocystis globosa (Prymnesiophyceae). J. Plankton Res. 22:107–
19.
Sies, H. 1997. Oxidative stress: oxidants and antioxidants. Exp.
Physiol. 82:291–5.
Stoecker, D. K., Stevens, K. & Gustafson, D. E. 2000. Grazing on
Pfiesteria piscicida by microzooplankton. Aquat. Microb. Ecol.
22:261–70.
Vardi, A., Berman-Frank, I., Rozenberg, T., Hadas, O., Kaplan, A. &
Levine, A. 1999. Programmed cell death of the dinoflagellate
Peridinium gatunense is mediated by CO2 limitation and oxi-
dative stress. Curr. Biol. 9:1061–4.
Veldhuis, M. J. W., Cucci, T. L. & Sieracki, M. E. 1997. Cellular DNA
content of marine phytoplankton using two new fluoro-
chromes: taxonomic and ecological implications. J. Phycol.
33:527–41.
Zhang, M., Kong, F., Xing, P. & Tan, X. 2007. Effects of Interspe-
cific Interactions between Microcystis aeruginosa and Chlorella
pyrenoidosa on their growth and physiology. Int. Rev. Hydrobiol.
92:281–90.