941
J. Phycol. 35, 941±948 (1999)
FLOW CYTOMETRIC ANALYSES OF VIRAL INFECTION IN TWO MARINE
PHYTOPLANKTON SPECIES, MICROMONAS PUSILLA (PRASINOPHYCEAE) AND
PHAEOCYSTIS POUCHETII (PRYMNESIOPHYCEAE)1
Corina P. D. Brussaard,2 Runar Thyrhaug
Department of Microbiology, University of Bergen, N-5020 Bergen, Norway
Dominique Marie
Station Biologique, CNRS, INSU et Universite Pierre et Marie Curie, BP 74, 29682 Roscoff cx, France
and
Gunnar Bratbak
Department of Microbiology, University of Bergen, N-5020 Bergen, Norway
Cell characteristics of two axenic marine phyto-
plankton species, Micromonas pusilla (Butscher)
Manton et Parke and Phaeocystis pouchetii (Hariot)
Lagerheim, were followed during viral infection us-
ing ¯ow cytometry. Distinct differences between
noninfected and infected cultures were detected in
the forward scatter intensities for both algal species.
Changes in side scatter signals on viral infection
were found only for P. pouchetii. Chlorophyll red
¯uorescence intensity per cell decreased gradually
over time in the infected cultures. DNA analyses
were performed using the nucleic acid±speci®c ¯uo-
rescent dye SYBR Green I. Shortly after infection
the fraction of algal cells with more than one ge-
nome equivalent increased for both species because
of the replication of viral DNA in the infected cells.
Over time, a population of algal cells with low red
auto¯uorescence and low DNA ¯uorescence devel-
oped, likely representing algal cells just prior to viral
lysis. The present study provides insight into basic
virus±algal host cell interactions. It shows that ¯ow
cytometry can be a useful tool to discriminate be-
tween virus infected and noninfected phytoplankton
cells.
Key index words: algal viral infection; ¯ow cytome-
try; Micromonas pusilla; Phaeocystis pouchetii; phyto-
plankton cell characteristics
Abbreviations: FCM, ¯ow cytometry; FSC, forward
scatter; GFL, green ¯uorescence; LRFLÐP. pouche-
tii, population of P. pouchetii with low RFL; MpV,
Micromonas pusilla virus; NRFLÐP. pouchetii, popu-
lation of P. pouchetii with normal RFL; PpV, Phaeo-
cystis pouchetii virus; RFL, red ¯uorescence; SSC,
side scatter
The signi®cance of viruses in natural aquatic eco-
systems has received much attention recently. Cur-
1 Received 18 December 1998. Accepted 25 May 1999.
2 Author for reprint requests; e-mail corina.brussaard@im.uib.no.
rent working hypotheses suggest that viruses are im-
portant regulating factors in marine ecosystems. Lyt-
ic viruses directly control population dynamics by
viral lysis (Suttle et al. 1990, Bratbak et al. 1993, Na-
gasaki et al. 1994, Brussaard et al. 1996). By gene
transfer they may alter community composition
(Chirura 1997). As a result of the release of cell con-
tents due to viral lysis of host organisms, viruses in-
directly affect the carbon, nutrient, and sulfur ¯ux
in pelagic ecosystems (Fuhrman and Suttle 1993,
Bratbak et al. 1994, Gobler et al. 1997, Hill et al.
1998).
Although it is clear that viruses do play an impor-
tant role in phytoplankton ecology, most work in
this ®eld of research has concentrated on the de-
tection and enumeration of the algal viruses. De-
tailed information on the direct effect of a viral in-
fection on phytoplankton cells is still lacking. Waters
and Chan (1982) and Suttle et al. (1990) showed
that viral infection may lead to a reduction in chlo-
rophyll ¯uorescence and primary production of the
algal species, but these data do not provide any in-
formation about individual cells within the algal
population. The detection of viruslike particles with-
in algal cells using transmission electron microscopy
(TEM) provides only information about the ®nal
stages of the lytic cycle when mature viral particles
are visible (Fields et al. 1996). Furthermore, TEM
analyses are time consuming.
Flow cytometry (FCM) allows a rapid and precise
analysis of the characteristics of individual cells and
is a tool with great potential for the study of changes
in cellular parameters of virus infected and nonin-
fected algal populations. Because of de novo synthesis
of viral DNA and possible digestion of host DNA,
the cellular DNA content of virus infected and non-
infected phytoplankton cells may be expected to be
different. The use of FCM in combination with nu-
cleic acid speci®c ¯uorescent dyes allows quanti®-
cation of cellular DNA. Recently, FCM has also been
successfully applied to enumerate algal viruses (Ma-

942 CORINA P. D. BRUSSAARD ET AL.
rie et al. 1999), making FCM a potentially useful
tool for studies on basic virus±host cell interactions.
The purpose of the present study was to investi-
gate how infection with a lytic virus alters the host
cellular parameters and cell cycle. Furthermore, we
searched for differences in cell characteristics be-
tween virus-infected and noninfected algal popula-
tions to allow differentiation between the two pop-
ulations. Forward-angle and side-angle light scatter,
as well as red auto¯uorescence (which is directly
linked to chlorophyll ¯uorescence per cell) and the
green ¯uorescence from DNA staining of individual
algal cells, were measured to differentiate between
virus-infected and noninfected algal cells.
We used two axenic host-virus model systems: the
¯agellated free-living life stage of Phaeocystis pouchetii
(Prymnesiophyceae, 4±6 mm diameter) with its viral
pathogen PpV-01 (130±160 nm diameter) and the
naked pico¯agellate Micromonas pusilla (Prasinophy-
ceae, 1.5±3.0 mm diameter) with its virus MpVUF10-
38 (100±140 nm diameter). Both species are geo-
graphically widespread and abundant (e.g. Cottrell
and Suttle 1991, Baumann et al. 1994, Vaulot et al.
1994), making them and their viruses interesting
host-virus model systems.
MATERIALS AND METHODS
Strains and culture conditions. The axenic algal-host-virus systems
used in this study were Micromonas pusilla strain LAC38 with the
virus MpVUF10-38 (MpV) and Phaeocystis pouchetii strain AJ-01
with the virus PpV-01 (PpV). M. pusilla was obtained from the
culture collection of the Marine Research Centre of Goteborg
University and P. pouchetii AJ-01 from the culture collection at the
University of Bergen, Norway. The virus MpV was isolated from
Central Kattegat (E. Sahlsten, pers. comm.) and PpV from Rau-
nefjorden, Western Norway (Jacobsen et al. 1996).
Algal cultures became axenic after treatment (1±10 days) with
a mixture of the antibiotics carbenicillin (Sigma Chemical Co.,
St. Louis, Missouri) and cefotaxime (Sigma) and yeast extract
(Difco Laboratories). Successful ®nal concentrations of the anti-
biotic mixture used were between 0.02 and 0.1 g´L21. Purity of
the algal cultures was tested in three different ways: by epi¯u-
orescence microscopy (Zeiss) using DAPI (adding 1:1 v/v of 10
mg´mL21 stock solution ®ltered through 0.2-mm pores; Sigma), by
checking for bacterial contamination after addition of peptone
and yeast extract (0.5 and 0.3 mg mL21 ®nal concentration, re-
spectively), and by transmission electron microscopy (Jeol 100S)
using the method of Bratbak and Heldal (1993). The virus cul-
tures became axenic after double ®ltration through Supor 200
®lters (0.2 mm pore size; Gelman Sciences).
P. pouchetii was grown in f/2 medium (Guillard 1975) with the
following modi®cations. Concentrations of added KH2PO4 and
NaNO3 were 5 and 80 mM, respectively. The concentrations of
the vitamins were doubled. All stock solutions of nutrients and
vitamins were added to the basic seawater before autoclaving (bot-
tles of 1 L containing 500 mL of medium were autoclaved with
closed lids). The pH of the medium after autoclaving was 7.9 to
8.0. M. pusilla was growing in modi®ed ESAW medium (Harrison
et al. 1980) as described by Cottrell and Suttle (1991). Stock so-
lutions of KBr, NaF, KCl, NaHCO3, H3BO3, and Tris-HCl were
autoclaved separately and added afterward. The pH of the me-
dium was 7.7. Temperatures were maintained at 8.0 6 0.58 C for
P. pouchetii and 15 6 0.58 C for M. pusilla. Light was supplied at
a 16:8 h LD (light:dark) cycle of at a photon ¯ux density of 100
6 20 mmol´m22´s21 (Philips TLD 33 and Sylvania T8/CW) for P.
pouchetii and M. pusilla, respectively.
At the start of each experiment, the algal culture was split into
two subcultures of 300 to 400 mL each (in sterile 1-L Erlenmeyer
¯asks) and diluted to cell concentrations around 4 3 105 mL21
for P. pouchetii and 1.8 3 107 mL21 for M. pusilla. One of the
subcultures was infected with 40 mL of fresh virus lysate, whereas
the other subculture served as a control. The ratio of virus to
algal cell was 10 for M. pusilla and 30 for P. pouchetii. Thus, even
when infectivity of the viruses added is low (Cottrell and Suttle
1995, Bratbak et al. 1998), we intended to have a multiplicity of
infection (m.o.i.; ratio of number of infectious viruses in the in-
oculum added to the number of algal cells in the culture) of at
least one. For both algal species the experiments were performed
in duplicate.
Samples for algal cell and virus counts were taken at regular
intervals of generally 2 to 6 h. Subsamples were ®xed with 0.5%
glutaraldehyde (25%, grade I, Sigma) for 15 min, quick frozen
in liquid nitrogen, and then stored at 2808 C (Vaulot et al. 1989,
Marie et al. 1999).
Analyses. All analyses were performed with a FACSort ¯ow cy-
tometer (Becton Dickinson) equipped with an air-cooled laser
providing 15 mW at 488 nm and with the standard ®lter setup.
All stains were purchased from Molecular Probes Inc.
For enumeration of phytoplankton cells, fresh samples were
diluted up to ®ve-fold for P. pouchetii and 100-fold for M. pusilla
with 0.22-mm-®ltered culture medium. The trigger was set on the
red ¯uorescence (RFL), and samples were analyzed on the ¯ow
cytometer for 2 min at a delivery rate of 50 mL´min21. Fluorescent
microspheres with a diameter of 0.95 mm were added as internal
reference.
For cell cycle analysis, ®xed frozen samples (Vaulot et al. 1989)
were quickly thawed at 378 C and then diluted ®ve-fold for P.
pouchetii and 100-fold for M. pusilla in 0.22-mm-®ltered seawater.
Diluted samples were incubated for 30 min at 378 C in presence
of 1% of RNase A (Sigma R-4875) and then for 15 min at room
temperature in the dark and in presence of SYBR Green I at a
concentration of 1024 of the commercial dilution. The discrimi-
nator was set on the red ¯uorescence, and samples were analyzed
by ¯ow cytometry for 4 min at a delivery rate of 50 mL´min21.
Virus enumeration was performed on ®xed frozen samples. Di-
lutions in TE buffer (Tris 10 mM, EDTA 1 mM, pH 7.5) ranged
from 1:50 to 1:2000. Dilutions were incubated for 15 min in the
dark in the presence of SYBR-I at a ®nal concentration of 0.5 3
1024 (Marie et al. 1999). The discriminator was set on the green
¯uorescence (GFL), that is, proportional to the nucleic acids±
SYBR-I complex, and samples were analyzed by ¯ow cytometry for
1 to 4 min at a delivery rate of 50 mL´min21.
Data were collected on logarithmic scale for all experiments
except for cell cycle analysis, for which both linear and logarith-
mic DNA ¯uorescence were collected. Data were computed with
the custom-designed software CYTOWIN (Vaulot, unpubl.; soft-
ware is available freely through anonymous FTP on the ftp.sb-
roscoff.fr server in the /pub/cyto directory), which discriminates
cell populations by using the combination of all parameters re-
corded. Cell cycle data were analyzed using MultiCycle (Phoenix
Flow Systems Inc., San Diego, California). Graphs were drawn
with the WinMDI freeware (Joseph Trotter).
The single-parameter DNA histograms of the control cultures
were analyzed by assuming that cells with one genome equivalent
corresponds to G1 (postmitosis), cells with two complements of
DNA to G2 (cells have replicated their DNA), and those cells in
the region that separates the two peaks G1 and G2 are in the
process of DNA replication (the S-phase). Infected cells possess
not only genomic DNA but also viral DNA. Thus, for the purpose
of discussion, we refer for the virus infected cultures to the DNA
peak corresponding to G1 in the control cultures as G1-like, the
population with twice the DNA content of G1 as G2-like, and to
the intermediate population as S-like.
RESULTS
Micromonas pusilla
The population of Micromonas pusilla could be de-
tected on the dual-parameter dot plot of RFL versus