RESEARCH ARTICLE

High and dierential viral infection rates within bacterial

morphopopulations in a shallow sand pit lake (Lac de Creteil,
France)
A.S. Pradeep Ram1,2, Bashir Arnous1,2, Michael Danger3,4, Jean-Francois Carrias1,2, Gerard
Lacroix3 &
1,2

Telesphore Sime-Ngando
1
Clermont Universite, Blaise Pascal, Clermont-Ferrand, France; 2Laboratoire Microorganismes: Genome
et Environnement, CNRS, UMR 6023, Aubie`re,
France; 3Laboratoire Bioemco, Biogeochimie
et Ecologie des Milieux Continentaux, UMR 7618 (Universite Paris 6, CNRS, INRA, ENS), Ecole Normale

Superieure, Paris, France; and 4Laboratoire des Interactions Ecotoxicologie, Biodiversite,
Ecosyste`mes LIEBE, Universite,
Paul Verlaine Metz, CNRS UMR
7146, Metz, France

Correspondence: Telesphore Sime-Ngando,
Blaise Pascal, BP 10448
Clermont Universite,
F-63000 Clermont-Ferrand, France. Tel.: 133
4 7340 7836; fax: 133 4 7340 7670; e-mail:
telesphore.sime-ngando@univ-bpclermont.fr
Received 3 December 2009; revised 18 May
2010; accepted 25 May 2010.
Final version published online 30 June 2010.
DOI:10.1111/j.1574-6941.2010.00920.x
Editor: Riks Laanbroek
MICROBIOLOGY ECOLOGY
Abstract
The ecology of viruses in shallow artificial freshwaters is poorly documented and
there is no reference for sand pit lakes. We examined the seasonal abundances and
infection rates of viruses in the sand pit Lake Cre teil (France). Bacteria were the
best predictor of viral abundance (4.07.8  1010 viruses L1), with an average
virus-to-bacteria ratio of 13.5  1.9. Virus-induced bacterial mortality (range
3786%, mean 65%) was higher than that in typical pelagic situations. This was
related to high specific contact rates between viruses and bacterial hosts and high
burst size (BS) estimates. Seasonal fluctuations in viruses and bacteria were rather
homeostatic, although temperature was a major driver of microbial activities.
Different bacterial morphotypes, i.e. morphopopulations, were analysed. Rod

Keywords
sand pit lakes; seasonal dynamics; viruses;
bacteria; lytic infection; microbial ecology.
cells dominated the total (90%) and infected (89%) communities. Elongated rods
were the most infected (45% of infected cells), whereas fat rods exhibited the
highest BS estimates (mean = 72 viruses per bacterium) due to a larger specific cell
volume. We conclude that the lytic activity of viruses is high and heterogeneous for
different bacterialhost phenotypes in the sand pit Lake Creteil. A theoretical
exercise shows that this can exert a strong influence on the processes occurring in
plankton food webs.

mer, 2004; Ram et al., 2009). Reports have suggested that

Introduction
Over the last two decades, progress in viral ecology has
considerably changed our conception of the structural and
functional organization of microbial food webs in aquatic
ecosystems. It is now well recognized and accepted that
viruses are the most conspicuous, abundant and dynamic
biological entities and form an integral component of the
microbial food web in the world aquatic environment. They
play crucial roles in the regulation of carbon and nutrient
fluxes, govern bacterial diversity and diversification, mediate
lateral gene transfer and even have direct implications for
the global climate (Wilhelm & Suttle, 1999; Weinbauer,
2004; Suttle, 2007). Studies examining larger datasets have
revealed that heterotrophic prokaryotes are the predomi-
nant hosts for viruses in pelagic systems (Peduzzi & Schie-
virally induced bacterial mortality can account for up to
60100% of the daily bacterial production, especially in
some freshwater systems (Fischer & Velimirov, 2002), and
can exceed the rates of grazing by bacterivorous protists
(Weinbauer & Hofle, 1998a; Ram et al., 2010). This is a
significant departure from the traditional view that preda-
tion and resource availability are the major factors control-
ling bacterial abundance (BA) and production in pelagic
systems.
Although viral abundances (VAs) and infection rates have
been estimated for a variety of aquatic environments,
including freshwater, seawater and hypersaline ponds, there
are no such investigations pertaining to shallow lakes of
artificial origin such as sand pit lakes. Studies in sand pit
lakes are important as their artificial origin makes them

FEMS Microbiol Ecol 74 (2010) 8392 

c 2010 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
84
study models both for understanding phenomena related to
eutrophication and for determining the relative importance
of different environmental factors in microbial metabolism.
A lack of opportunity to document viral ecology in such
systems has resulted in a paucity of knowledge about these
environments. Sand pit lakes usually originate from dred-
ging activities due to the increasing need for building
materials and are mainly located in the flood plains of large
rivers and streams. Although such lakes are oligotrophic in
nature, at least for a few decades, they can attain a eutrophic
status in the due course of time through high nutrient input
supply by river waters. Sand pit lakes are often more
productive than deeper lakes because the organic matter
formed does not sediment for a long time; instead, it is
broken down and its constituent nutrients are frequently
stirred back into the euphotic realm and quickly reused to
produce more organic matter. Shallow lakes are also fre-
quently warmer than deeper lakes, which indeed leads to
increased matter decomposition and production rates (Bur-
ghis & Morris, 1987).
We hypothesized that viral bacteriolysis should be much
higher in sand pit lakes compared with other systems, largely
because of higher host abundance and activity. In such a
case, viruses could have a strong and selective impact on
different morphotypes of bacterial populations. A study of
the above aspect is crucial as variations in bacterial biomass
(carbon pool) are strongly related to the morphotype
composition of bacterial populations (Sime-Ngando et al.,
1991). Morphological adaptation by bacteria serves as an
important biological function through cell surface-to-
volume ratios to respond to environmental cues, and to
gain a competitive advantage for coping with a changing
environment, especially in pelagic systems (Young, 2007).
Very little is known about the potential variations of infection
frequency and burst sizes (BSs) (number of phages released
during bacterial lysis) in different bacterial morphotypes.
Moreover, trophic status could be a possible driving force in
controlling the abundance of viruses through host activity, an
issue that has been less addressed in freshwater than in marine
systems. To enhance our understanding of aquatic microbial
food webs, information on forces that determine the relative
importance of viruses as a source of bacterial mortality
among bacterial morphotypes is of considerable interest.
In the present study, seasonal variations in VA and phage
infectivity in a shallow sand pit lake (Lake Cre teil, France)
were examined in relation to environmental parameters,
with the aim of improving our knowledge of the viral
dynamics in artificial aquatic basins. Variations in the viral
infection frequency and BS were evaluated for different
bacterial morphopopulations, in the context of total viral-
mediated bacterial mortality (VIBM) and viral production.
This is the first study on the seasonal VA and infectivity in
Lake Cre teil.

c 2010 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
A.S. Pradeep Ram et al.
Materials and methods
Study site and sampling
Lake Cre teil is a mesoeutrophic, turbid, small (surface area:
42 ha), artificial shallow (mean depth: 4 m, maximum
depth: 6 m) sand pit lake situated 15 km southeast of Paris,
France. The lake originates from an excavation of alluvial
sediments near the confluence of the Seine and Marne
Rivers. The lake is mainly supplied with anoxic phreatic
waters circulating through alluvial deposits and diverse
filling materials and also by a rainwater sewer from the 0.5-
km2 Mont Mesly urbanized zone. The water residence time
of the lake was estimated at 612 months. The high total
salinity (1.4 g L1) of lake water originates from the compo-
sition of the supplying ground water (Garnier et al., 1992).
Well exposed to winds, Lake Cre teil weakly stratifies for
short unpredictable periods in summer, and the water
column is well oxygenated throughout the year. Water
samples were collected every month at 1 m depth from
October 2005 through August 2006 at a designated point of
the lake (48146 0 3700 N, 02126 0 4700 E). Collected samples were
poured into clean 10-L recipient vessels washed previously
with lake water.
Physicochemical analysis
Water temperature and dissolved oxygen concentrations
were followed as vertical profiles using an oxygen probe
(O2-meter DO200; VWR International, Strasbourg, France).
Dissolved inorganic (DIN, i.e. NO3-N and NH4-N) and
organic (DON) nitrogen were measured after persulphate
oxidation, whereas dissolved inorganic and organic phos-
phorus were measured after persulphate digestion (AFNOR,
Association Francaise de Normalisation, 1990). Dissolved
organic (DOC) and inorganic (DIC) carbon were measured
by nondispersive infrared gas analysis using a Shimadzu
TOC-5000 analyzer. Total dissolved carbon (TC) was ex-
tracted by total combustion at 680 1C, whereas DIC was
recovered by acidification with H3PO4. DOC was calculated
by subtracting DIC from TC concentrations (APHA, 1998).
Phytoplankton biomass was estimated in situ as mg equiva-
lent chlorophyll-a (Chla) L1 using a fluorescence photo-
meter (Fluoroprobe; BBE-Moldaenke, Kiel, Germany). The
device was calibrated for the system studied (see Danger
et al., 2008) so as to minimize any likely errors associated
with its measurements.
Bacteria and VAs
For the enumeration of viruses (VA) and bacteria (BA), 100-mL
water samples were fixed with 0.02 mm filtered buffered alkaline
formalin (final concentration 2% v/v, from a 37% w/v solution
of commercial formaldehyde). Subsamples (12 mL  three
FEMS Microbiol Ecol 74 (2010) 8392
Seasonal viruses in shallow sand pit lake
replicates) were then filtered (o 15 kPa vacuum) through 0.02-
mm pore-size Anodisc filters (Whatman) using cellulose acetate
backing filters (1.2 mm pore size). Following SYBR Green I
staining (Molecular Probes, Carlsbad, CA; final dilution,
2.5  103 fold) as described by Noble & Fuhrman (1998), the
filters were air dried on an absorbent paper and mounted
between slides and glass coverslips with an antifading mountant
glycerol/phosphate-buffered saline solution (i.e. Citifluor,
London, UK) amended with a special antifading solution,
c. 20% v/v of Vecta Shield (Vector Laboratories). This amend-
ment significantly reduced fading of the fluorochrome and
yielded highly stable fluorescence (Ram et al., 2009). The slides
were stored at  20 1C and counted within a week using a
model 300F epifluorescent microscope (Leica DC). Bacteria
were distinguished from virus-like particles on the basis of their
relative size and brightness. A blank was routinely examined to
control for contamination of the equipments and reagents.
Lytic phage infection
Bacterial cells present in 8 mL of formalin-fixed (final
concentration 2% v/v) water samples were collected on
triplicate electron microscope grids (400-mesh, carbon-
coated Formvar film) by ultracentrifugation (Optima LE-
80K, Beckman Coulter SW40 Ti Swing-Out-Rotor at
70 000 g for 20 min at 4 1C) according to Sime-Ngando
et al. (1996). Each grid was stained at room temperature (c.
20 1C) for 30 s with uranyl acetate (2%, pH = 4), rinsed twice
with 0.02 mm filtered distilled water to remove excess stain
and then dried on a filter paper. Grids were examined using
a JEOL 1200Ex transmission electron microscope (TEM)
operated at 80 kV and a magnification of 20 00060 000 to
distinguish between bacterial cells with and without intra-
cellular viruses. A bacterium was considered infected when
at least five viruses, identified by shape and size, were clearly
visible inside the host cell. At least 600800 bacterial cells
were inspected per grid to determine the frequency of visibly
infected bacterial cells (FVIC). The viral mean BS (viruses
per bacterium) was estimated for every sample as the
average number of viral particles in infected cells (at least
15 infected cells were examined per sample). Because mature
phages are visible only late in the infection cycle, FVIC
counts were converted to the frequency of infected cells
(FIC) using the equation FIC = 9.524FVIC  3.256 (Wein-
bauer et al., 2002). Assuming a steady state of infected and
uninfected bacteria and a latent period that equalled the
bacterial generation time, FIC was converted to VIBM (as a
percentage per generation) using the following equation:
VIBM = (FIC10.6FIC2)/(1  1.2FIC) (Binder, 1999).
Viral and bacterial size class analyses
The diameter of intracellular viruses was examined via TEM
for the relative abundances of four arbitrarily divided size
FEMS Microbiol Ecol 74 (2010) 8392
85
classes based on capsid diameter: 4 30, 3060, 6090 and
4 90 nm. Similarly, different bacterial morphotypes were
subjectively recorded as elongated thin rod, short rod, fat
rod, filamentous or cocci (i.e. bacterial morphopopulations)
based on TEM observations. The above classification can,
however, result in some overlaps of groups. To minimize
this, cocci were defined as having a length : width ratio
between 1 and 2, fat rods were defined as having a
length : width ratio between 2 and 5, or having a width
4 200 nm, and thin rods were defined as having a
length : width ratio 4 5 and a width o 200 nm (Brum
et al., 2005). The mean cell dimensions of both infected
and noninfected bacterial cells were determined from at least
200 cells per sample, using TEM photographs made at a
magnification of c. 4000. Cell volumes (V) were calculated
assuming that the true shapes of cells could be approximated
as a sphere or as a cylinder with hemispherical ends (Sime-
Ngando et al., 1991; Bratbak, 1993): V = (p/4)  W2 
[L  (W/3)], where L = cell length and W = cell width. For
each bacterial morphopopulation, BS was estimated from
the number of viruses in those visibly infected cells that were
totally filled with phages, i.e. the maximum BS (BSmax).
Contact rates
The rate of contact (R) between viruses and bacteria was
calculated using the following formulae (Murray & Jackson,
1992): R = (Sh2pwDv)VP, where Sh is the Sherwood number
(1.06 for a bacterial community with 10% motile cells)
(Wilhelm et al., 1998), w is the cell diameter (calculated
from the mean bacterial cell volume assuming that the cells
are spheres), V and P are the abundance of viruses and the
abundance of undamaged cells, respectively, and Dv is the
diffusivity of viruses, Dv = kT/3pmdv, where k is the Boltz-
mann constant (1.38  1023 J K1), T is the in situ tempera-
ture (in K), m is the viscosity of water (in P s1; m was
calculated from the values given by Schworbel (1987) for
temperatures in the range from 4 to 15 1C) and dv is the
mean diameter of the viral capsid (42  16 nm for Lake
Creteil, present study). The contact rate was corrected for
BA to estimate the number of contacts per cell on a daily
basis.
Heterotrophic nanoflagellate (HNF) abundance
and grazing potential
Samples for the measurements of HNF abundance were
fixed immediately after sampling with 1% glutaraldehyde
(final concentration). Primulin-stained HNF collected on
0.8-mm polycarbonate black filters (25 mm diameter) were
counted under UV excitation under a Leica epifluorescent
microscope (Caron, 1983). A total of 200400 nanoflagel-
lates from each slide were counted on several transects
(SD o 10%). All solutions were also filter sterilized and a

c2010 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
86
blank was examined routinely to control for contamination
of the equipment and reagents as well. To estimate the loss of
bacteria to HNF grazing, i.e. the potential rate of bacterivory
by HNF, an approach using an average flagellate clearance
rate (1.9 nL ind1 h1) obtained from published reports in
freshwater lakes (see Lymer et al., 2008) was used for calcu-
lations. Potential HNF grazing rates (FLG, cells mL1 h1)
were estimated from in situ BA, total HNF abundance and
the mean clearance rate.
Statistical analysis
Differences in the physicochemical and biological variables
between seasons (Autumn, OctoberDecember; Winter,
JanuaryMarch; Spring, AprilMay; Summer, JuneAugust)
were tested using one-way ANOVA. The potential relation-
ships among variables were tested by linear pair-wise
correlations (i.e. Pearsons correlation analysis) and stepwise
multiple regressions. Data were log-transformed to satisfy
the requirements of normality and homogeneity of variance
necessary for parametric statistics. All statistical analyses
were performed using MINITAB software for Windows (Re-
lease 12, MINITAB).
Results
Physicochemical environment
Water column temperature showed strong seasonal changes
(ANOVA, P o 0.001), which are typical of temperate lakes.
The values increased from 3.3 1C (January) to 24.6 1C (July),
with a mean seasonal average of 13.5  7.9 1C. The water
column of Lake Cre teil was generally well oxygenated
(mean  SD = 9.0  1.7 mg L1) during the entire study per-
iod, with an average saturation of 89%. In most cases,
nutrients did not vary significantly with the season. Among
nutrients, nitrate exhibited the highest values, with a mean
concentration of 1.1  0.9 mg L1. The concentrations of
potentially limiting nutrients such as total nitrogen and
phosphorous were well above the threshold levels known to
induce any kind of limitation for the growth of planktonic
organisms (Table 1). The mean atomic N : P ratio was
20.0  8.3. Like inorganic nutrients, DOC did not exhibit
seasonality and was relatively constant, with a seasonal
average of 6.0  0.4 mg L1. Chla ranged from 2.4 to
25.4 mg L1 (mean  SD = 13.3  8.2 mg L1), with the max-
imum value observed in April.
Standing stocks
VA showed several peaks, with the maximum value observed
on 7 August (7.8  1010 L1), coinciding with the peak in BA
(6.3  109 cells L1). These maximum VA and BA were
twofold higher than the lowest values obtained in February

c 2010 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
A.S. Pradeep Ram et al.
Table 1. Mean  SD (range) physicochemical characteristics, Chla con-
centrations, bacterial and viral parameters in the surface waters (1 m) of
Lake Creteil
Parameters (units) Values
Temperature ( 1C) 13.5  7.9 (3.324.6)
Dissolved oxygen (mg L1) 9.0  1.7 (6.412.0)
Total nitrogen (mg L1) 1.25  0.92 (0.673.77)
Total phosphorous (mg L1) 0.06  0.07 (0.0350.25)
Dissolved organic carbon (mg L1) 6.0  0.4 (5.56.8)
Chlorophyll a (mg L1) 13.3  8.2 (2.425.4)
Viral abundance (1010 L1) 5.7  1.5 (4.07.8)
Bacterial abundance (109 per cell) 4.3  1.0 (2.86.3)
Virus-to-bacteria ratio 13.5  1.9 (10.116.6)
Frequency of infected cells (%) 32.0  5.9 (23.238.0)
Virally induced bacterial mortality (%) 64.4  20.1 (36.785.7)
Burst size (viruses per cell L1) 44.0  16.0 (6350)
Viral capsid diameter (nm) 42  16 (2577)
Specific contact rate (contacts per cell day1) 549  141 (382746)
Heterotrophic nanoflagellates (106 cells L1) 1.4  0.8 (0.22.6)
Flagellate grazing (106 bacteria L1 h1) 11.5  8.0 (1.225.3)
(VA = 4.0  1010 L1; BA = 2.8  109 cells L1) (Fig. 1a).
Despite the variability observed in VA and BA, the virus
bacterium ratio (VBR) did not differ significantly
(P 4 0.05) with season and ranged from 10.1 to 16.6, with
an average of 13.5  1.9. Among abiotic variables, tempera-
ture correlated significantly with BA and VA. VA and BA
were significantly correlated to each other (Table 2).
Phage infection, BS and contact rates
The FIC ranged from 23.2% to 38.0%, with a mean value of
32.0  5.9% corresponding to 64.6  20.1% of VIBM. Mul-
tiple peaks in FIC were observed, among which the max-
imum value was observed on 21 November (2005), 11 April,
10 July and 7 August (2006); the latter peak coincided with a
peak in BA and VA (Fig. 1a and b) and corresponds to a
VIBM level of 85.7%. FIC was more strongly related to VA
than to BA (Table 2). The BS of individual cells varied over
two orders of magnitude and ranged from 6 to 350 viruses
per bacterium (mean = 44  16 viruses per bacterium) and,
for the entire bacterial community, was significantly corre-
lated with water temperature and FIC, but negatively
correlated with BA (Table 2). Theoretical contact rates,
which are necessary to estimate the rate of successful
infection, were calculated according to the model of Murray
& Jackson (1992). In Lake Cre teil, the specific contact rate
(SCR), i.e. the number of viruses encountering a single
bacterial cell per time, ranged between 382 and 746,
averaging 549  141 contacts per cell day1 (Table 1), and
peaked in August with VA, BA and FIC. All these variables
were correlated to each other and to the water temperature
as well (Table 2).
FEMS Microbiol Ecol 74 (2010) 8392
Seasonal viruses in shallow sand pit lake 87

VA
10 (a) 8
BA

Bacteria (109 cells L1)

8

Viruses (1010 L1)

6

6
4
4

2
2

0 0

40
Frequency of infected cells (%) (b)
35
30
25
20
15
10
5
0

3.5 (c) 30.0

HNF

FLG ( 106 bacteria L1 h1)

3.0 FLG 25.0
HNF (106 cells L1)

2.5
20.0
2.0
15.0
1.5
Fig. 1. Seasonal variations in the abundance of 10.0
1.0
viruses and bacteria (a), frequency of infected
bacterial cells (b) and abundance of HNFs and 0.5 5.0
grazing potential (see the main text for more

details) in the surface waters (1 m) of Lake Creteil, 0.0 0.0
October 2005 to August 2006. Error bars indicate O N J F M A M J J A
SE (n = 3). 2005 2006

Table 2. Pearsons correlation coefficient (r) between different variables in the surface waters (1 m) of Lake Creteil (n = 20)
Bacterial Viral Frequency of Heterotrophic
Parameters Temperature abundance abundance infected cells nanoflagellates
Bacterial abundance 0.83
Viral abundance 0.62 0.92
Frequency of infected cells 0.83 0.58 0.93
Heterotrophic nanoflagellates 0.79 NS NS NS
Flagellate grazing 0.80 0.48 NS 0.50 0.89
Specific contact rates 0.62 0.81 0.94 0.69 NS
Burst size 0.62  0.82 NS 0.70 NS

Levels of significance: P o 0.05, P o 0.01, P o 0.001.

NS, not significant.

FEMS Microbiol Ecol 74 (2010) 8392 

c2010 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
88 A.S. Pradeep Ram et al.

Size class analyses of infective viruses and
infected bacteria
The diameter of intracellular infective viruses ranged from
20 to 110 nm, with an average of 42  16 nm (n = 210).
Among the size range, 4 80% of the head size diameter of
free-living (data not shown) and intracellular phages was
 60 nm (Fig. 2a), suggesting that they were typical bacter-
iophages. Among bacterial morphopopulations, rods were
the most dominant, representing about 90% of the entire
bacterial community (data not shown) and also the largest
(a) Intracellular viral size classification
2%
15%
21%
<30 nm
3060 nm
6090 nm
>90 nm
fraction (89%) of infected cells, whereas cocci and filamen-
tous forms contributed to 10% and 1% of infected cells,
respectively (Fig. 2b). Among rods, the elongated rods were
the most infected (45%), followed by fat rods (23%) and
short rods (21%). The bacterial morphotypes investigated
thus exhibited strong and significant (P o 0.001) differences
in terms of the frequency of cells containing mature phages
and the BSmax (cf. Fig. 3). Among rods, a considerable
difference in BSmax was observed (Fig. 2c). Indeed, the
number of infection cells in fat rods was lower than that in
elongated thin rods, contrasting with a BSmax that was
significantly higher (P o 0.01) due to the larger cell volume
(0.81.2 mm3 per cell) of fat rods. The specific cell volumes
of infected cells ranged from 0.03 to 1.2, with an average of
0.08  0.02 mm3 per cell, and did not differ signifi-
cantly from the cell volumes of noninfected cells
(mean = 0.07  0.02 mm3 per cell). The cell volumes of both
types of cells were quite stable over seasons, and were sig-
nificantly correlated to BSmax: r = 0.85 and 0.76 (P o 0.001
for both) for infected and noninfected cells, respectively.

HNFs
62%
The seasonal pattern in the abundance of HNF was marked
(b) Viral infected bacterial cell morphotypes by two similar peaks, which were recorded in May and July,
10%
reaching about 2.6  106 cells L1 (Fig. 1c). Unlike abun-
dance, the HNF potential grazing rate (FG) showed large
1% 21%
and significant variability (P o 0.001) with seasons and
Short rods
ranged from 1.2 to 25.3  106 bacteria L1 h1 (mean-
Elongated rods 11.5  8.0  106 bacteria L1 h1), with a peak in August that
23% Fat rods coincided with a peak in FIC and VIBM. FG was signifi-
Filamentous cantly (P o 0.05) correlated to BA and to FIC and to the
Cocci water temperature as well (Table 2).

Regression analyses
45% Forward stepwise multiple regression analysis was con-
(c) Viral burst size for different cell morphotypes ducted to select the variables than can account for variations
20
in VA and FIC. The results indicated that BA together with
36
Chla and temperature accounted for 72% of the variation
(VA = 1.53  0.0380Temp10.792BA10.107Chla, r2 = 0.72,
Short rods
n = 20). The other variables listed in Table 2 did not add
Elongated rods
significantly to the ability of the equation to predict VA. For
Fat rods
FIC, temperature and BA were selected as significant
48
Cocci
(P o 0.05) predictor variables, explaining 50% of its varia-
bility (FIC = 35.110.476Temp  1.74BA, r2 = 0.50, n = 20).

72 Discussion
The present investigation documents the standing stock of
Fig. 2. Size class analyses of infective viruses and infected bacteria in the

surface waters (1 m) of Lake Creteil. (a) Percentage of different intracel-
viruses and phage infection in relation to environmental
lular viral size classes (capsid diameter), (b) FICs (% total infected cells) parameters in the pelagic realm of the sand pit Lake Cre teil.
and (c) max BS estimates (viruses per bacterium) for different bacterial The virioplankton abundances in the surface waters of the
morphotypes. lake were within the range reported in other lakes


c 2010 Federation of European Microbiological Societies FEMS Microbiol Ecol 74 (2010) 8392
Published by Blackwell Publishing Ltd. All rights reserved
Seasonal viruses in shallow sand pit lake
(a)
(e)
Fig. 3. TEMs of typical infected bacterial cells
representing different morphotypes (see the
main text for more details) in the surface waters
(1 m) of Lake Creteil, namely elongated thin rods
(a, e), fat rods (b, c), short rods (d, g) and cocci (f).
Scale bar = 100 nm.
(Weinbauer, 2004), and similar to those from temperate
eutrophic lakes such as Lake Grangent, France (Ram et al.,
2009), Lake Plusee, Germany (Weinbauer & Hofle, 1998a),
and Lake Loosdrecht, the Netherlands (Tijdens et al., 2008).
The seasonal abundances of both viruses and bacteria in Lake
Creteil were rather homeostatic as they did not vary by
4 twofold (Fig. 1a), which is similar to those reported by
Ram et al. (2009) in Lake Grangent (France), Mathias et al.
(1995) in a backwater system of River Danube (Austria) and
Hennes & Simon (1995) in the mesotrophic Lake Constance
(Germany). We consider that this was not related to a
methodological problem because, as in most of the literature
studies and in our proper previous studies, we have used the
routine protocol proposed by Noble & Fuhrman (1998) for
counting viruses and bacteria. In addition, among the abiotic
variables, water temperature largely explained significant varia-
tions in the measured microbial parameters, including those
from the viral community (Table 2), which highlights the
seasonal forcing of these parameters. This was likely enhanced
during the present study because of the weakened stratification
and the related low stability and thermal inertia of shallow
water column lakes such as the sand pit Lake Creteil. Under
such conditions, the physical environment, primarily tempera-
ture, has been shown to be a strong forcing factor in setting
conditions for optimal metabolic activities of microbial com-
munities (cf. Ram et al., 2010). Our data also reveal interesting
potential trophic links, primarily between viral and bacterial
components in Lake Cre teil. A large fraction of the variation in
VA was explained by BA, similar to what is generally known
from temperate lakes (Ram et al., 2009, 2010). This usually
suggests that the majority of viruses were coupled to prokar-
yotes, and that the number and role of typical algal viruses
FEMS Microbiol Ecol 74 (2010) 8392
89
(b) (c) (d)
(g)
(f)
could be lower compared with bacteriophages in Lake Cre teil.
In this lake, the average VBR value was 14, which is similar to
those observed in temperate eutrophic lakes such as Alte
Donau, Austria (Fischer & Velimirov, 2002), and Rimov
Reservoir, Czech Republic (Simek et al., 2001), but slightly
higher than the typical values of 310 for pelagic freshwater
systems in general (Wilhelm & Matteson, 2008). VBR did not
fluctuate considerably over the study period, supporting the
tight coupling between bacteria and viruses, with a relatively
constant level of viral production and losses (Ram et al., 2009).
Studies have suggested that viral infection may contribute
significantly to bacterial mortality in aquatic systems (cf.
Weinbauer, 2004). We have used TEM (i.e. whole-cell
examination) for the determination of FVIC (expressed as
% of the total bacterial cells) and BS estimates. FVIC
estimates not only provide direct evidence of phage infec-
tion and facilitate the calculation of virus-induced bacterial
mortality rates but also allow a relatively easy comparison of
viral infection among aquatic systems. However, the meth-
odological problems inherent to the FVIC approach should
be commented on. The determination of VIBM from FVIC
estimates in our study could be biased due to the use of
empirical models of Weinbauer et al. (2002) for converting
FVIC to FIC, and of Binder (1999) for converting FIC to
VIBM. Although the TEM-based derivation of FVIC, com-
bined with the above models, has been used increasingly in
recent studies conducted in freshwater lakes (Ram et al.,
2009, 2010), it should be noted that there are several critical
assumptions underlying the derivation of the models. While
the empirical model of Weinbauer et al. (2002) (obtained
from a viral dilution approach) is an improvement on a
previous existing model (Procter et al., 1993), the model

c2010 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
90
could not be applied for a wide variety of freshwater systems,
as the parameters of the model vary depending on varying
environmental conditions. The extent and cause of this
variability remain unclear. Moreover, the model of Binder
(1999) assumes that infected and uninfected bacteria are
grazed at the same rate, which may not be true in a wide
variety of environments (Weinbauer & Peduzzi, 1995).
Because the applicability of these parameters in Lake Cre teil
has yet to be validated, the absolute estimates of virally
induced mortality that we report in this paper should be
interpreted cautiously, but we still consider them to be
comparable to systems where similar approaches were used.
One striking feature observed in Lake Cre teil was the
incidence of high viral infection rates, at times reaching up to
86% of bacterial production. This is among the highest VIBM
levels reported in euphotic oxic zones of freshwater ecosystems.
Viral lytic control of 4 100% of bacterial production has so
far only been reported in the oxic euphotic zone of Lake Alte
Donau, Austria (Fischer & Velimirov, 2002). Compared with
other temperate lakes, the average value of VIBM in Lake
Cre teil (about 65%) is much higher than those reported in the
oxic zones of eutrophic systems such as Lake Plusee,
Germany (Weinbauer & Hofle, 1998a), Lake Aydat, France
(Bettarel et al., 2004), and Lake Grangent, France (Ram et al.,
2009). In Lake Cre teil, the increase in viral infection with
increasing VA and BA suggests that the lytic mode of infection
is important, which is supported by the high SCRs (mean
SCR = 549 per cell day1) between viruses and their prokaryo-
tic hosts. Such high encounter rates, resulting in high mortality
rates, suggest that the bacterioplankton community could be
dominated by a few species, favouring specific adsorption of
viruses to their co-occurring hosts. This also suggests that the
maintenance of a high number of viruses is dependent on the
most active fraction of bacterioplankton (Ram et al., 2010). In
some reports where SCR were on the order of 100250 per
cell day1, viruses are considered a major source of bacterial
mortality (Weinbauer & Hofle, 1998a; Fischer & Velimirov,
2002). This is likely also the case for the shallow, feebly
stratified sand pit Lake Cre teil, where microbial activities are
further enhanced by seasonal temperature, although data on
bacterial community structure and activity and on the lyso-
genic lifestyle in the viral community are lacking.
BS is a crucial parameter for the estimation of viral
production, depending on the number of lysed cells and on
the number of viruses set free per lysed cell. The mean BS (44
viruses per bacterium) reported for Lake Cre teil in this study
(i.e. from those cells containing at least five mature viruses) is
higher than those of mesoeutrophic lakes (27 viruses per
bacterium) such as Rimov Reservoir, Czech Republic (Simek
et al., 2001), but was similar to those reported in eutrophic
systems (46 viruses per bacterium) (Weinbauer & Hofle,
1998a; Fischer & Velimirov, 2002). Examples of typical
infected cells and intracellular viruses from Lake Cre teil are

c 2010 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
A.S. Pradeep Ram et al.
provided in Fig. 3. In the present study, the mean BS was
positively correlated with the FIC and negatively correlated
with BA, suggesting that high viral release should have
occurred under high bacterial productivity. In addition, BS
was the sole variable correlated to the specific cell volume of
bacteria, as was also found in the Adriatic Sea (Weinbauer &
Peduzzi, 1995). In contrast to reports suggesting that the cell
volume of infected cells is larger than that of uninfected cells
(Weinbauer & Hofle, 1998b), we did not observe such a trend
in this study, where the cell volumes of infected cells
(0.031.2 mm3) were similar to those from uninfected cells.
The current observation that a target bacterial cell mor-
photype, namely rod-shaped cells, largely dominated the
bacterial community in Lake Cre teil during this study is in
contrast to a previous observation made two decades ago in
the same lake, where small coccoids (mean cell volu-
me = 0.03 mm3) dominated and constituted 74% of the total
BA (Garnier & Benest, 1990). This strongly suggests com-
munity shifts in bacterioplankton structure over a period of
time, which could be attributed to the changing trophic
conditions in Lake Cre teil, from oligotrophic towards eu-
trophic. This argument is supported by the higher levels of
nutrient concentrations and the increasing chlorophyll
values measured in this study, compared with reports from
previous studies conducted two decades ago (Garnier et al.,
1992). In addition, the age of the lake can also have an
impact, as such a sand pit lake becomes shallower with time
due to sediment deposition, and the size and shape of the
lake can influence the stirring and mixing patterns of the
water column (Kattner et al., 2000).
Studies on the potential effects of viral lysis on different
bacterial morphotypes (i.e. the so-called bacterial morpho-
populations) and cell size distribution are limited, although
these variables directly determine bacterial biomass in
aquatic systems (Sime-Ngando et al., 1991). Morphological
changes in natural bacterioplankton with higher percentages
of elongated cell vs. coccal forms have been shown to
contribute to the increase of the bacterial carbon pool in
the Northern Adriatic Sea (Ferla & Leonardi, 2005). Bacteria
can modify their morphology in response to environmental
cues, and selective forces such as nutrient uptake capabilities
and predation have been shown to have a strong impact on
bacterioplankton (Young, 2007). In Lake Creteil, the strong
differences in the frequencies of bacterial cells filled with
mature phages (i.e. BSmax) and of viral infection among the
investigated bacterial morphopopulations (cf. Fig. 3) might
be due, at least to some extent, to the physiological dif-
ferences associated with different bacterial phenotypes. For
example, it was shown that rod-shaped cells are favoured in
nutrient-rich environments, whereas small coccal-shaped
bacteria are often dominant in low nutrient waters (Ferla &
Leonardi, 2005). Studies from Lake Plusee, Germany, have
revealed that bacterial mortality due to viral lysis was
FEMS Microbiol Ecol 74 (2010) 8392
Seasonal viruses in shallow sand pit lake
significant in the bacterial size class 0.91.2 mm, which
represented 2955% of the total abundance (Weinbauer &
Hofle, 1998b). The impact of phages on bacteria thus strongly
depends on the metabolic activity of host cells (Ram & Sime-
Ngando, 2008), which may be directly linked to the size of
these cells (Ferla & Leonardi, 2005). In the present study, the
abundances of the dominant and most infected rod bacteria
were well above the threshold level of 2  105 cells mL1
necessary for the occurrence of detectable phage infection in
the plankton (Weinbauer & Peduzzi, 1994). As phages target
the most active and dominant fraction of the bacterial com-
munity, we consider that selective lysis of cells belonging to a
particular morphotype, i.e. rod cells in our case study (cf. Fig.
3), may induce substantial changes in the functional roles of
the natural bacterial community and the related biogeochem-
ical processes. This is further supported by the observation of
different infection rates and BSmax in different categories of the
dominant rod cells in Lake Cre teil (Fig. 2b and c).
Finally, no correlation was found between bacteria and
HNFs in Lake Creteil, suggesting that bacterial biomass does
not seem to be a limiting factor for the growth of HNFs. The
relatively high flagellate grazing rates in Lake Cre teil (mean-
11.5  106 bacteria L1 h1) compared with other eutrophic
systems (Bettarel et al., 2004), along with high virally induced
mortality, suggest that both mortality factors could contribute
to the bulk of bacterial mortality, with viral lysis exceeding
grazing at certain periods of the year. As bacterial production
was not measured in our study, we could not directly estimate
and compare the two sources of mortality, as carried out in
previous studies (Bettarel et al., 2004; Lymer et al., 2008).
Conclusions
Overall, this first study on viral ecology in Lake Cre teil
provides evidence for high, but differential infection rates
within various bacterial phenotypes, partly as a result of
enhanced seasonal forcing in the shallow and weakly strati-
fied water column of the lake. The high viral infection likely
affected the most active fraction of the bacterial host
community, while the different infection rates for various
morphopopulations and submorphopopulations suggest
that viruses can have a significant impact on the bacterial
community composition. This could result in the disruption
of carbon flow to higher trophic levels, due to the virally
mediated release of substantial amounts of organic particles
and biogenic elements into the dissolved phase (Weinbauer,
2004; Suttle, 2007). During our study period, we estimated
that viruses were able to lyse 0.72.4  109 bacterial cells in
1 L of water (average = 1.5  109 cells L1), which corre-
sponds to dissolved carbon release rates of 1448 mg C L1
(average = 28 mg C L1), assuming a mean bacterial carbon
content of 20 fg C per cell (Norland, 1993). These potential
amounts of organic carbon released into the water column
FEMS Microbiol Ecol 74 (2010) 8392
91
as a result of viral lysis during the present study represented
658% (average = 30%) of the primary production, based
on the values measured in 1985 and 1986 in the water
column of Lake Creteil (Garnier & Benest, 1990). We
conclude that viruses can exert a strong influence on the
processes occurring in the plankton food web of Lake Cre teil
and the related carbon and energy flows. This study also
highlights the importance of a lack of data on bacterial
community composition, production and metabolic activ-
ity, and on the lysogenic lifestyle in the viral community,
which undoubtedly will improve our overall understanding
of the functional significance of viruses in Lake Creteil.
Acknowledgements
A.S.P.R. was supported by a postdoctoral fellowship from the
French ANR (Agence Nationale de la Recherche). We are
grateful to C. Oumarou, D. Benest and D. Dalger for their
technical help. This is a contribution to the research programs
PNBC (ACI-ECCO) and ANR Biodiversite AQUAPHAGE.
References
AFNOR, Association Francaise de Normalisation (1990) Eaux,
methodes dessais: recueil de normes francaises, Union Impr.
Paragraphic, Paris.
APHA (1998) Standard Methods for the Examination of Water and
Wastewater, 20th edn. American Public Health Association,
Washington, DC.
Bettarel Y, Sime-Ngando T, Amblard C & Dolan J (2004) Viral
activity in two contrasting lake ecosystems. Appl Environ
Microb 70: 29412951.
Binder B (1999) Reconsidering the relationship between viral
induced bacterial mortality and frequency of infected cells.
Aquat Microb Ecol 18: 207215.
Bratbak G (1993) Microscopic methods for measuring bacterial
biovolume: epifluorescence microscopy, scanning electron
microscopy and transmission electron microscopy.
Handbook of Methods in Aquatic Microbial Ecology (Kemp PF,
Sherr B, Sherr E & Cole JJ, eds), pp. 309317. Lewis Publishers,
Boca Raton, FL.
Brum JR, Steward GF, Jiang SC & Jellison R (2005) Spatial and
temporal variability of prokaryotes, viruses, and viral
infections of prokaryotes in an alkaline, hypersaline lake.
Aquat Microb Ecol 41: 247260.
Burghis MJ & Morris R (1987) The Natural History of Lakes.
Cambridge University Press, Cambridge.
Caron DA (1983) Techniques for enumeration of heterotrophic
and phototrophic nanoplankton using epifluorescent
microscopy, and comparison with other procedures. Appl
Environ Microb 46: 19221928.
Danger M, Lacroix G, Oumarou C, Benest D & Meriguet J (2008)
Effects of foodweb structure on periphyton stoichiometry in
eutrophic lakes: a mesocosm study. Freshwater Biol 53:
20892100.

c2010 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
92
Ferla RL & Leonardi M (2005) Ecological implications of biomass
and morphotype variations of bacterioplankton: an example
in a coastal zone of the Northern Adriatic Sea
(Mediterranean). Mar Ecol-Evol Persp 26: 8288.
Fischer UR & Velimirov B (2002) High control of bacterial
production by viruses in a eutrophic oxbow lake. Aquat Microb
Ecol 27: 112.
Garnier J & Benest D (1990) Seasonal coupling between phyto-
and bacterioplankton in a sand pit lake (Creteil Lake, France).
Hydrobiologia 207: 7177.
Garnier J, Chesterikoff A, Testard P & Garban B (1992)
Oligotrophication after a nutrient reduction in a shallow sand-
pit lake (Creteil Lake, Paris surburbs, France): a case of rapid
restoration. Ann Limnol-Int J Lim 28: 253262.
Hennes KP & Simon M (1995) Significance of bacteriophages for
controlling bacterioplankton growth in a mesotrophic lake.
Appl Environ Microb 61: 333340.
Kattner E, Schwarz D & Maier G (2000) Eutrophication of gravel
pit lakes which are situated in close vicinity to the River Donau:
water and nutrient transport. Limnologica 30: 261270.
Lymer D, Lindstrom ES & Vrede K (2008) Variable importance
of viral-induced bacterial mortality along gradients of
trophic status and humic content in lakes. Freshwater Biol 53:
11011113.
Mathias CB, Kirschner KT & Velimirov B (1995) Seasonal
variations of virus abundance and virus control of the bacterial
population in backwater system of the Danube River. Appl
Environ Microb 61: 37343740.
Murray AG & Jackson GA (1992) Viral dynamics: a model of the
effects of size, shape, motion and abundance of single-celled
planktonic organisms and other particles. Mar Ecol-Prog Ser
89: 103116.
Noble RT & Fuhrman JA (1998) Use of SYBR green I for rapid
epifluorescence counts of marine viruses and bacteria. Aquat
Microb Ecol 14: 113118.
Norland S (1993) The relationship between biomass and volume
of bacteria. Handbook of Methods in Aquatic Microbial Ecology
(Kemp PF, Sherr BF, Sherr EB & Cole JJ, eds), pp. 303307.
Lewis Publishers, Boca Raton, FL.
Peduzzi P & Schiemer F (2004) Bacteria and viruses in the water
column of tropical freshwater reservoirs. Environ Microbiol 6:
707715.
Proctor LM, Okubo A & Fuhrman JA (1993) Calibrating
estimates of phage-induced mortality in marine bacteria:
ultrastructural studies of marine bacteriophage development
from one-step growth experiments. Microb Ecol 25:
161182.
Ram ASP & Sime-Ngando T (2008) Functional responses of
prokaryotes and viruses to grazer effects and nutrient
additions in freshwater microcosms. ISME J 2: 498509.
Ram ASP, Sabart M, Latour D & Sime-Ngando T (2009) Low
effect of viruses on bacteria in deep anoxic water and sediment
of a productive freshwater reservoir. Aquat Microb Ecol 55:
255265.

c 2010 Federation of European Microbiological Societies
Published by Blackwell Publishing Ltd. All rights reserved
A.S. Pradeep Ram et al.
Ram ASP, Nishimura Y, Tomaru Y, Nagasaki K & Nagata T (2010)
Seasonal variation in viral induced-mortality of
bacterioplankton in the water column of a large mesotrophic
lake (Lake Biwa, Japan). Aquat Microb Ecol 58: 249259.
Schworbel J (1987) Handbook of Limnology. Ellis Horwood, New
York.
Simek K, Pernthaler J, Weinbauer MG, Hornak K, Dolan J,
Nedoma J, Masin M & Amann R (2001) Changes in bacterial
community composition and dynamics and viral mortality
rates associated with enhanced flagellate grazing in
mesotrophic reservoir. Appl Environ Microb 67: 27232733.
Sime-Ngando T, Bourdier G, Amblard C & Pinel-Alloul B (1991)
Short-term variations in specific biovolumes of
different bacterial forms in aquatic systems. Microb Ecol 21:
211226.
Sime-Ngando T, Mignot JP, Amblard C, Bourdier G, Devilettes C
& Quilblier-Lloberas C (1996) Characterization of planktonic
virus-like particles in a French mountain lake: methodological
aspects and preliminary results. Ann Limnol-Int J Lim 32: 15.
Suttle CA (2007) Marine viruses major players in the global
ecosystem. Nat Rev Microbiol 5: 801812.
Tijdens M, Hoogveld HL, Kamst-van Agterveld MP, Simis SGH,
Baudoux AC, Laanbroek HJ & Gons HJ (2008) Population
dynamics and diversity of viruses, bacteria and phytoplankton
in a shallow eutrophic lake. Microb Ecol 56: 2942.
Weinbauer MG (2004) Ecology of prokaryotic viruses. FEMS
Microbiol Rev 28: 127181.
Weinbauer MG & Hofle MG (1998a) Significance of viral lysis
and flagellate grazing as factors controlling bacterioplankton
production in a eutrophic lake. Appl Environ Microb 64:
431438.
Weinbauer MG & Hofle MG (1998b) Size-specific mortality of
lake bacterioplankton by natural virus communities. Aquat
Microb Ecol 15: 103113.
Weinbauer MG & Peduzzi P (1994) Frequency, size and
distribution of bacteriophages in different marine bacterial
morphotypes. Mar Ecol Prog Ser 108: 1120.
Weinbauer MG & Peduzzi P (1995) Significance of viruses versus
heterotrophic nanoflagellates for controlling bacterial
abundance in the northern Adriatic Sea. J Plankton Res 17:
18511856.
Weinbauer MG, Winter C & Hofle MG (2002) Reconsidering
transmission electron microscopy based estimates of viral
infection of bacterioplankton using conversion factors derived
from natural communities. Aquat Microb Ecol 27: 103110.
Wilhelm SW & Matteson AR (2008) Freshwater and marine
virioplankton: a brief overview of commonalities and
differences. Freshwater Biol 53: 10761089.
Wilhelm SW & Suttle CA (1999) Viruses and nutrient cycles in
the sea. Bioscience 49: 781788.
Wilhelm SW, Weinbauer MG, Suttle CA & Jeffrey WH (1998) The
role of sunlight in the removal and repair of viruses in the sea.
Limnol Oceanogr 43: 586592.
Young KD (2007) Bacterial morphology: why have different
shapes? Curr Opin Microbiol 10: 596600.
FEMS Microbiol Ecol 74 (2010) 8392