Environmental Microbiology (2010) 12(2), 467479
Resources drive trade-off between viral lifestyles
in the plankton: evidence from freshwater
microbial microcosms emi_2088
A. S. Pradeep Ram and T. Sime-Ngando*
Laboratoire Microorganismes: Gnome et
Environnement, UMR CNRS 6023, Universit Blaise
Pascal, Clermont-Ferrand II, 63177 Aubire Cedex,
France.
Summary
Viruses have evolved different life strategies for
coping with environmental challenges and this is a
key explanation for their omnipresence in aquatic
systems. However, factors that determine the balance
between lytic versus lysogenic decision within
natural virioplankton are poorly documented, prima-
rily in freshwaters. This study was designed to inves-
tigate the experimental short-term (24 h incubation)
effects of added organic and inorganic nutrients on
the two viral lifestyles in nutrient-depleted freshwater
microbial (i.e. < 0.8 mm fraction) microcosms, using
mitomycin C as prophage inductor agent. In the
absence of mitomycin, viral lytic production
increased as a functional response to the strong
stimulation of bacterial growth rates (0.70.8 day-1) by
the added nutrients, primarily the organic nutrients
which appeared scarcer than inorganic nutrients and
was related to the sampling period and the geomor-
phological peculiarities of Lake Pavin. In the presence
of mitomycin, temperate phage production (fre-
quency of lysogenically infected bacterial cells,
FLC = 1719% of total cells) significantly exceeded
lytic production (frequency of lytically infected bacte-
rial cells, FIC = 911%) in natural samples (i.e. without
nutrient additions) as a result of enhanced prophage
induction, which relatively increased with the
decreasing contact probability between viruses and
their potential hosts. In contrast, addition of nutrients
drastically reduced FLC (< 4%) and increased FIC
(> 22%). Both variables were antagonistically corre-
lated as was the correlation between FLC and bacte-
rial growth rates, supporting the idea that lysogeny
Received 8 June, 2009; accepted 10 September, 2009. *For corre-
spondence. E-mail telesphore.sime-ngando@univ-bpclermont.fr;
Tel. (+33) 4 73407836; Fax (+33) 4 73407670.
2009 Society for Applied Microbiology and Blackwell Publishing Ltd
doi:10.1111/j.1462-2920.2009.02088.
467..479
may represent a maintenance strategy for viruses in
harsh nutrient/host conditions which appeared as
major instigators of the trade-off between the two
viral lifestyles. Overall, at the community level, we
reject the hypothesis that nutrients but mitomicyn C
stimulate temperate phage induction, and retained
the hypotheses that nutrients rather (i) stimulate lytic
viruses via enhanced host growth and (ii) when limit-
ing, promote lysogenic conversion in natural waters.
Introduction
Heterotrophic prokaryotes (i.e. bacteria) are a prime bio-
logical component and integral part of any normally func-
tioning natural ecosystems, where they play key roles in
the transformation and mineralization of organic matter.
Factors controlling their abundance, activity and commu-
nity composition are central to the understanding of their
ecological and biogeochemical roles in the transfer of
energy and materials in aquatic systems. Among these
factors, top-down control through viral lysis has recently
received, perhaps, the most attention. Pelagic studies
conducted during the past one decade have revealed that
viruses can contribute to bulk of the bacterial mortality via
selective lysis of host cells, which at times can exceed
protistan bacterivory depending upon the environment
(Weinbauer and Hfle, 1998; Fischer and Velimirov, 2002;
Pradeep Ram et al., 2005). As a result, viruses can regu-
late carbon and nutrient fluxes, food web dynamics, and
microbial diversity and diversification in aquatic systems
(Fuhrman, 1999; Wilhelm and Suttle, 1999; Weinbauer,
2004; Weinbauer and Rassoulzadegan, 2004). It is now
well established that viruses, primarily bacteriophages,
are the most abundant biological entities in a wide variety
of aquatic environments (Weinbauer, 2004; Sime-Ngando
and Colombet, 2009). Basically, this is because viruses
are excellent survivors which have evolved different
strategies for coping with environmental challenges
(Suttle, 2007), although availability of specific hosts gen-
erally is a crucial constraint for viral proliferation.
One of the key explanations for the omnipresence of
viruses in natural ecosystems undoubtedly is through the
existence of several lifestyles, of which two major path-
ways, namely lysis and lysogeny, are prevalent in aquatic
468 A. S. Pradeep Ram and T. Sime-Ngando

systems (Weinbauer, 2004). Lytic infections are by far the infection and the factors regulating their exclusive- or
best studied of the virushost interactions, which result in co-occurrence in natural populations of pelagic bacteria
host cell death via lysis following the replication of viral are poorly understood in freshwater systems. Also it is not
particles. Lysogeny has been less studied in aquatic envi- quite clear as to what environmental factors influence the
ronments where the temperate phage can alternatively switch from lysogenic to lytic decision and vice versa in
integrate into the host genome as prophage. Lysogenic aquatic environments.
conversion has been described as a means of survival for To test the hypotheses that (i) labile nutrient additions
viral populations that are threatened by poor host cell stimulate prophage induction and lytic production while (ii)
abundance and therefore cannot sustain population nutrient-limiting conditions rather promote lysogenic
numbers through lytic infection alone (Stewart and Levin, conversion, microbial samples taken from an oligome-
1984). Prophages can be stable in their host for long sotrophic freshwater lake were treated with a variety of
periods of time (months to years, Paul, 2008) with low inorganic and organic nutrients, in combination with mito-
probability of bacteriophages being released by sponta- mycin C. Generally, the relative abundance of induced
neous lysis. Mathematical modelling has demonstrated lysogenic bacteria is calculated from burst size (BS) esti-
that there is a reciprocal relationship between bacterial mates (Jiang and Paul, 1996; Williamson et al., 2002). In
diversity and viral abundance, with lysogeny theorized to addition to this approach, we also adopted the frequency
act as the repository preventing viruses from disappearing of visibly infected cell (FVIC) approach using transmission
altogether during periods of low relative abundances of electron microscope, as an alternative to determine the
host populations (Thingstad, 2000). Recent studies have frequency of lysogenic cells in our experimental samples.
suggested that prophage induction events can change The present study is an attempt to simultaneously inves-
bacterial community structure by increasing the diversity tigate two viral lifestyles (lytic and lysogenic) at the natural
and richness of natural bacterial populations (Hewson bacterial community level, in order to assess the specific
and Fuhrman, 2007). It is thus worth investigating the role of viruses for microbial food-web processes under
factors that may lead to the induction of prophages in different nutrient regimes.
natural conditions. This is an important issue for assess-
ing the whole role of viruses in aquatic systems and the Results
related ecological potentials, through major processes
Initial conditions
such as biogeochemical catalyses from lysis products
(Fuhrman, 1999) and horizontal gene transfers, e.g. Initial physicochemical and biological characteristics of
transduction, transformation (Sime-Ngando and Colom- the natural and experimental samples are listed in
bet, 2009). Table 1. For all these variables, the within-sample variabil-
Examinations of natural bacterial communities induc- ity was judged satisfactory with coefficient of variation
ible with a mutagenic agent (e.g. mitomycin C) have sug-
gested that the fraction of lysogenic bacteria typically is
Table 1. Mean (SD for triplicates) surface water characteristics of
< 50% of the total abundance (Jiang and Paul, 1996; Lake Pavin for the two experiments (Expt.).
Weinbauer and Suttle, 1996; Cochran and Paul, 1998),
with most of the studies restricted to marine environments Expt. 1 Expt. 2
and few on soil (Williamson et al., 2007) and desert (19 September (17 October
Parameters 2007) 2007)
(Prigent et al., 2005) ecosystems. Several studies have
suggested that availability of nutrients may have an Temperature (C)a 14.7 0.1 14.0 0.1
Dissolved oxygen (O2 mg l-1)a 8.9 0.2 7.5 0.2
important influence upon lytic or lysogenic decision with Total nitrogen (mg l-1)a 0.17 0.005 0.18 0.01
respect to the availability of host cells (Bratbak et al., Soluble reactive phosphorus 0.02 0.002 0.01 0.005
1993; Wilson and Mann, 1997; Hewson et al., 2001). (mg l-1)a
Total phosphorus (mg l-1)a 0.03 0.001 0.04 0.006
Wilson and Mann (1997) proposed that lysogenic conver- Total organic carbon (mg l-1)a 1.5 0.2 1.6 0.3
sion is favoured when nutrient concentrations, especially Chlorophyll a (mg l-1)a 1.7 0.3 2.1 0.2
inorganic phosphates, are low and the virus-to-bacteria Bacterial abundance 2.9 0.3 2.2 0.2
( 109 cells l-1)b
abundance ratio is high. In addition to nutrients, tempera- Viral abundance ( 1010 l-1)b 1.8 0.3 1.8 0.2
ture has also been suggested as a possible constraint for Viruses-to-bacteria ratiob 6.8 1.3 8.6 1.7
viral lifestyles but with contrasting results. Cochran and Frequency of lytically infected 9.5 2.1 5.7 1.4
cells (%)b
Paul (1998) have detected lysogens only when the water Viral-induced bacterial mortality 11.3 2.5 6.3 1.6
temperature exceeded 19C, while in the same site Will- (%)b
iamson and colleagues (2002) have successfully induced Burst size (viruses cell-1)b 22 4.1 19 3.0

lysogens during winter months at temperature < 5C. The a. Raw natural water.
functional importance of lysogeny as an alternative to lytic b. Filtered experimental samples (< 0.8 mm) at t0.

2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479
 Nutrient effects on lytic versus lysogenic viruses 469

generally < 20%. On both the sampling dates, the in situ 4.5 A
water temperature was at about 14C and oxygenic con- 4.0

Viral abundance ( x 10 7 ml-1)

ditions (dissolved oxygen content > 7 mg l-1) prevailed in 3.5
the euphotic layer of the sampled depth (i.e. 1 m). Among
3.0
inorganic nutrients, total nitrogen in raw natural waters
2.5
exhibited the highest concentrations, one order of magni-
2.0
tude higher than phosphorus concentrations (Table 1).
Total nitrogen concentration was about fourfold lower than 1.5

the typical annual mean concentration of 0.8 mg l-1 while 1.0

total phosphorus was similar to the annual mean of 0.02 0.5
mg l-1 in Lake Pavin (Rioual, 2002), but the N : P atomic 0.0
ratio was less than 10. In terms of biological environment, U C N P NP CNP
chlorophyll a pigment concentration in the raw natural
B
samples was at the lower end of annual range 7.0 Bacterial growth rate 1.0

Bacterial abundance ( x 10 6 cells ml-1)

(1.710.3 mg l-1) recorded for the current studied year (i.e.

Bacterial growth rate ( day -1)

2007) in the upper euphotic depths of Lake Pavin (S.
Rasconi, pers. comm.). Initial abundances (t0) of viruses
(VA) and of bacteria (BA) in the microbial microcosms
(< 0.8 mm experimental samples) were quite similar for
the two experiments, although the viral-to-bacteria ratios
(VBR) in the microcosms were relatively higher for the
second experiment, contrasting with the between-
experiment differences for the frequency of lytically
infected cells (FIC), the viral-induced bacterial mortality
(VIBM) and the maximum BS estimates. However, the two
experiments did not differ significantly (P > 0.05) for none
of the physicochemical and biological variables under
study.
Impact of nutrient additions on viral and
bacterial abundances
Without mitomycin C. Relative to the control treatment
(U) for the first experiment (Expt. 1), additions of organic
carbon and of inorganic nutrients, alone (C, N and P) or in
combination (NP and CNP), provoked an increase in the
abundance of viral particles in the microcosms over the
24 h incubation period, clearly as a response to the stimu-
lation of bacterial growth (Fig. 1). Significant effects
(P < 0.004) of nutrients were recorded in the two micro-
cosms with organic carbon added, i.e. treatments C and
CNP, where viral abundance increased by 153% and
184%, respectively, and bacterial abundance by 210%
and 218%, corresponding to bacterial-specific growth
rates (BGR) of 0.7 day-1 in both microcosms. The highest
BSs in Expt. 1 were also noted in C and CNP treatments
(Table 2). In the second experiment (Expt. 2) where dif-
ferent organic nutrients, namely bacterial lysate (treat-
ment BL), organic phosphate (OP) and amino acid (AA),
were added to separate microcosms, viral and bacterial
abundances, bacterial growth rates and BSs also
responded positively to these additions, compared with
the unamended control (U) (Table 2, Fig. 2). These
responses were significant (P < 0.002) in all cases, with a
6.0
0.8
5.0
0.6
4.0
3.0 0.4
2.0
0.2
1.0
0.0 0.0
U C N P NP CNP
Fig. 1. Experiment 1. Viral (abundance, A) and bacterial
(abundance and growth rate, B) responses to organic and inorganic
nutrients (light bars) and the effect of mitomycin C (grey bars),
following 24 h incubation of experimental samples. Viral and
bacterial abundances at t0 were at 1.8 0.3 107 ml-1 and
2.9 0.3 106 cells ml-1 respectively. Values are mean SD of
triplicate samples. U, unamended (i.e. no nutrient addition); C,
carbon; N, nitrogen; P, phosphate; NP, nitrogen + phosphate; CNP,
carbon + nitrogen + phosphate.
maximum BGR of 0.8 day-1 due to organic phosphate
addition. Besides, a viral reduction approach was adopted
during Expt. 2 to significantly lower (i.e. by 8- to 10-fold)
the contact probability between viruses and their prokary-
ote hosts in the related treatment [viral-reduced fraction
(VR)], compared with the initial abundances of both com-
munities in all microcosms at the start of the incubations.
For this treatment and after incubation, VA and BA also
increased relative to their initial levels, this being more
marked for VA (increase by about 50%) than for BA (15%)
(Fig. 2).
With mitomycin C. In Expt. 1, VA also increased in the
presence of mitomycin C but in all treatments, i.e. with
and without nutrient additions. For each treatment taken
separately, the differences due to mitomycin was signifi-
cant (P < 0.001) only for the microcosms without any
nutrient addition (i.e. comparison of the treatments U
versus U + mitomycin), although the addition of mitomycin
in the nutrient-enriched treatments also resulted in more

2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479
470 A. S. Pradeep Ram and T. Sime-Ngando
Table 2. Estimates of the frequencies of lysogenically (FLC) and of lytically (FIC) infected cells in experimental samples.

FLC (%)
Maximum BS
Treatments (viruses bacterium-1) FIC (%) BS method FVIC method FIC : FLCa

Expt. 1
U 23 3.8 11.5 1.9 19.3 2.6 18.2 2.3 0.6
C 40 4.2 30.2 3.1 3.4 0.4 0.4 0.1 8.9
N 22 2.3 22.3 2.3 3.2 0.4 0.6 0.2 7.0
P 23 2.1 22.5 2.8 3.0 0.5 0.6 0.2 7.5
NP 25 2.7 22.9 2.5 2.8 0.4 0.5 0.1 8.2
CNP 42 4.8 33.0 3.5 3.2 0.5 0.4 0.1 10.3
Expt. 2
Viral reduced 17 2.3 8.7 2.2 17.5 2.2 20.3 2.7 0.4
U 28 1.8 9.5 1.9 17.2 2.8 16.2 2.1 0.6
BL 32 3.2 27.7 2.8 2.8 0.6 0.5 0.2 9.9
OP 30 3.5 25.4 2.2 3.0 0.5 0.6 0.3 8.5
AA 33 2.9 25.8 1.9 2.8 0.4 0.5 0.3 9.2

a. FLC estimated from burst size (BS) approach; for acronyms refer to the legends of Figs 1 and 2 or to the main text.

production of viral particles (Fig. 1A). In contrast, mitomy-

cin appeared inhibitory for bacterial communities in all
4.5 A
treatments. Addition of mitomycin C indeed maintained
Viral abundance ( x 10 7 ml-1)

BA to a relatively constant level around the initial natural 4.0

abundances (i.e. t0 level) in all experimental treatments 3.5

(Fig. 1B). For Expt. 2, quite similar results on the effects of 3.0

mitomycin versus nutrient additions were obtained com- 2.5

pared with those recorded in Expt. 1. Indeed, VA also 2.0

increased in the treatments U, BL, OP and AA with mito- 1.5

mycin added but the difference due to mitomycin was 1.0

significant only for the treatment without nutrient (i.e. U), 0.5

while mitomycin appeared inhibitory for bacteria in all 0.0

VR U BL OP AA
treatments (Fig. 2). Regarding the viral reduction treat-
ment in Expt. 2, addition of mitomycin significantly B
(P < 0.001) increased the level of VA compared with the 7.0 Bacterial growth rate 1.0
Bacterial abundance ( x 10 6 cells ml-1)

same treatment but without mitomycin addition (Fig. 2A).

Bacterial growth rate ( day -1)

6.0
This contrasts with the observation for bacterial commu- 0.8
nities in VR treatment, where mitomycin induced a weak 5.0
decrease in BA (Fig. 2B). 4.0 0.6

3.0
Interactions between nutrient and mitomycin effects for 0.4
VA variability. For both experiments, two-way analysis of 2.0
variance (ANOVA) showed strong variability (P < 0.001) in 0.2
1.0
VA in response to nutrient additions after 24 h incubation,
contrasting with the weak effects (P < 0.24) of mitomycin 0.0 0.0
C. There were near-significant but weak (P < 0.056) VR U BL OP AA

interactions between the two factors (i.e. nutrients and
mitomycin C).
Impact of nutrient additions on viral lifestyles
Burst size approach. Table 2 summarizes the frequency
of lytically (FIC) and of lysogenically (FLC) infected cells in
experimental samples. For both experiments, prophage
Fig. 2. Experiment 2. Viral (abundance, A) and bacterial
(abundance and growth rate, B) responses to organic nutrients
(light bars) and the effect of mitomycin C (grey bars), following 24 h
incubation of experimental samples. Viral and bacterial abundances
at t0 were at 0.4 0.1 107 ml-1 and 4.7 0.8 106 cells ml-1,
respectively, for VR, and at 1.8 0.2 107 ml-1 and 2.2 0.2 106
cells ml-1 for the other treatments. Values are mean SD of
triplicate samples. VR, viral-reduced fraction (see text for details);
U, unamended (i.e. no nutrient addition); BL, bacterial lysates; OP,
organic phosphate; AA, amino acid.

2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479

inductions calculated from maximum BS estimates were
significantly higher (P < 0.01) in natural experiment
microcosms (FLC > 17%), compared with treatments with
organic and/or inorganic nutrients added (FLC < 4%).
This contrasts with the differences in the FIC which were
< 12% in the absence of nutrients but > 20% in nutrient-
enriched microcosms. As a consequence, FIC : FLC ratio
was < 1 (range 0.40.6) in treatment with no nutrient and
was largely > 1 (range 710) due to nutrient additions
(Table 2).
Frequency of visibly infected cell (FVIC) approach. An
attempt was made to directly detect visibly infected
bacteria using transmission electron microscopy (TEM)
approach in experimental samples treated with mitomycin
C, from which alternative FLC calculations were per-
formed. The between-treatment comparisons for FVIC-
derived FLC versus FIC were exactly the same yielded by
those previously described from BS approach (r = 0.99,
Table 2). When comparing the two approaches, FLC esti-
mated from FVIC did not differ significantly (P > 0.05) with
those determined from BS in the microcosms without
nutrient additions, although FVIC-derived FLC in VR treat-
ment for Expt. 2 appeared relatively higher. In the nutrient-
enriched microcosms, FVIC-derived FLC (< 1%) were
significantly (P < 0.001) lower than those calculated from
BS (> 2.5%) approach (Table 2). Indeed, very rarely we
were able to detect any visibly infected bacterial cells in
nutrient-enriched samples with mitomycin C added.
Relationship between lysogeny and lytic viral infection
Overall, addition of nutrients to experimental samples did
not induce significant prophage induction and thereby
resulted in the decrease in the FLC, with concomitant
increase in the FIC, irrespective of the approaches used
in this study (Table 2). The FLC in the unamended
samples was on average about twofold higher than FIC in
both Expt. 1 and Expt. 2, whereas in nutrient-treated
samples, FIC was about one order of magnitude higher
(P < 0.001) than FLC (Table 2). For both experiments, FIC
and FLC were correlated with BA, VA and BGR, the rela-
tionships being positive for FIC and negative for FLC
(Table 3). As a consequence, FLC and FIC were nega-
tively correlated (r = -0.94, P < 0.001) to each other,
which could be best described by a polynomial function
(log FLC = 0.00018FIC2 - 0.1018FIC + 1.8865).
Discussion
Methodological and general considerations
Methodological constraints often limit an accurate deter-
mination of lysogens in aquatic systems. In the present
2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479
Nutrient effects on lytic versus lysogenic viruses 471
Table 3. Pearson productmoment correlation coefficient (r)
between different parameters during the experimental study, n = 30.
Parameters BA VA BGR FIC
VA 0.51***
BGR 0.74*** 0.58***
FIC 0.81*** 0.55*** 0.81***
FLCa -0.66*** -0.69*** -0.85*** -0.94***
a. Correlation analysis was conducted after log transformation to
reduce high variance for larger values.
Levels of significance: *P < 0.05, **P < 0.01, ***P < 0.001.
BA, bacterial abundance; VA, viral abundance; BGR, bacterial growth
rate; FIC, frequency of lytically infected cells; FLC, frequency of
lysogenically infected cells.
study, mitomycin C, a chemical-mediated prophage induc-
tion method, was used for estimating the FLC in natural
microbial microcosms. Mitomycin C is an antibiotic,
extremely toxic antineoplastic agent (Windholz et al.,
1976), and a potent DNA cross-linker. When activated it
acts as a DNA-alkylating agent leading to mispairing of
bases, DNA strand breakage, and cross-linking of com-
plimentary strands which prevents DNA synthesis. This
results in prohage induction by direct DNA damage (Iyer
and Syzbalski, 1963) and activation of the rec A repair
system. To date, mitomycin C is the most frequently and
currently used agent to estimate FLC in aquatic microbial
ecology (Jiang and Paul, 1996). Compared with other
agents, e.g. UV, hydrogen peroxide, mitomycin has been
shown to be the best prophage inductor for marine
bacterioplankton (Weinbauer and Suttle, 1996; 1999),
although some environmental pollutants (e.g. aroclor
1248, PCB mixture) (Cochran et al., 1998) and sunscreen
products (Danovaro and Corinaldesi, 2003) could also be
effective in inducing temperate phages.
However, the effectiveness of mitomycin as prophage-
inducing agent is largely unknown for natural freshwater
bacterial communities. In our experimental samples, FLC
was estimated from two different approaches: (i) the BS
method, which is based on the difference in viral abun-
dance between treated (with mitomycin C) and control
(without mitomycin C) divided by maximum BS, and (ii)
the FVIC method, which is based on the detection and
enumeration of visibly infected bacteria after induction of
lysogenic bacteria. The latter approach so far has not
been done in environmental samples. Both methods
yielded consistently similar result trends for the two
experiments conducted, although FLC levels for FVIC-
based method were generally lower compared with those
from BS-based approach, this being significant when
nutrients were added in microcosms (Table 2). The finding
that FVIC approach yielded varying BS levels (range
566 viruses bacterium-1, mean = 20 viruses bacterium-1,
Fig. 3) is consistent with the consideration that low BS
observed in mitomycin C-treated samples may be
472 A. S. Pradeep Ram and T. Sime-Ngando

Fig. 3. Transmission electron micrographs showing bacterial intracellular viruses from mitomycin C-treated experimental samples, with varying
burst sizes, i.e. the number of intracellular viruses per infected cell. Each bar denotes 100 nm.

characteristic of lysogeny trait (Weinbauer et al., 2007).
However, because FLC levels, primarily in nutrient-
amended microcosms, were close to or even under the
detection limit of the TEM used for detecting the infected
cells, therefore we feel that our FVIC-based approach
needs more testing and time-course calibration for an
accurate determination of FLC in natural waters.
In general, the differences in viral abundances when
comparing experimental microcosms with and without
mitomycin were low, primarily in the nutrient-enriched
microcosms. Viral production due to nutrient amendments
could be largely explained by the growth of bacterial hosts
in the absence of mitomycin, contrasting with the situation
in nutrient- and mitomycin-treated microscosms where
bacteria were apparently under inhibition, based on their
total abundances (cf. Figs 1 and 2). In fact, few numbers
of bacteria were needed for the observed viral production.
For the microscosms with nutrients and mitomycin, the
numbers of bacteria needed for the observed viral pro-
duction after 24 h incubation can be calculated by dividing
these productions by the maximum BSs in the different
microcosms, i.e. (VAt24h - VAt0h)/BS. These calculations
suggested that the numbers of lysed bacteria ranged from
2 to 6 105 cells ml-1, which are within the standard
deviations around the total bacterial mean abundances in
the microcosms of interest, and are close to the detection
upper limit of our epifluorescence microscopy counts, i.e.
c. 104-105 cells ml-1. Clearly, these considerations imply
that viral production in the presence of nutrients and mito-
mycin could be explained by the ambient relatively con-
stant abundances of bacteria, when considering the
within-replicate variability and the precision of the bacte-
rial counting method. It has been shown that the action of
mitomycin is effective after a minimum of 1115 h in the

2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479

western Gulf of Mexico (Weinbauer and Suttle, 1996;
1999). In addition, the physiology of some bacterial
lysogens can escape from the action of mitomycin (Ack-
ermann and Dubow, 1987), indicating (i) that we might
underestimate the relative abundance of lysogens in
natural waters and (ii) that, for our case study and for the
mitomycin-treated microcosms, bacterial assemblages
during the first hours of incubations and at least a fraction
of these assemblages during the whole incubation time
were able to respond to nutrient additions and produce
lytic viruses also, in spite of the relatively constant level of
total cells.
The lysogenic potential of the experimental samples
Lysogeny in aquatic samples has largely been determined
by BS approach and in our case study, error sources were
minimized by using maximum BS estimates for the calcu-
lation of FLC (Weinbauer and Suttle, 1996). This is in
contrast with previous studies where theoretical average
BSs (i.e. known from the literature) have been used and
could result in rough estimates of FLC (Jiang and Paul,
1996; Tapper and Hicks, 1998; Williamson et al., 2002;
Swstrm et al., 2006). In aquatic systems, lysogeny has
largely been determined in either natural (Tapper and
Hicks, 1998; Colombet et al., 2006) or viral-reduced water
samples (Weinbauer and Suttle, 1996; Williamson et al.,
2002). In our study, comparison of the two types of
samples did not yield any discrepancy in FLC estimates,
suggesting that our results were accurate. In the Tampa
Bay, USA, the relative abundance of lysogens was found
higher for the viral reduced (mean FLC = 36%) than for
the natural (20%) experimental samples (Williamson
et al., 2002). In our natural samples without nutrient
amendments, estimates of FLC (1719%) using BS
approach was little higher but closer to the previous find-
ings (0.116%) reported in spring in Lake Pavin (Colom-
bet et al., 2006), than to more lower values known from
the few studies conducted in productive temperate
(0.17.4%, Tapper and Hicks, 1998) and tropical
(0.17.1%, Bettarel et al., 2006) freshwater systems,
using the same methodological approach. This is prob-
ably not due to our relatively low sample temperature
(14C) that was maintained during incubations, as FLC as
high as > 50% has been reported at temperatures < 15C
in the Baltic and deep Mediterranean Seas (Weinbauer
et al., 2003). It is likely that FLC levels in Lake Pavin may
be higher than in more productive lakes, and we believe
that nutrient environment is a significant factor that may
help explain these differences.
Our experiments were conducted in the late seasonal
clear-water phase (i.e. early autumn), when chlorophyll
pigment concentration and especially inorganic nutrients
(N, P) typically were at their lowest annual concentrations,
2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479
Nutrient effects on lytic versus lysogenic viruses 473
e.g. fourfold lower than the annual mean for nitrogen (S.
Rasconi, pers. comm.). In general, this is characteristic of
nearby regional lakes in the French Massif Central area in
late summer towards autumn, where microbial metabo-
lisms suffer from low ambient nutrient concentrations. In
Lake Pavin, this could be due to its peculiarities such a
small surface area and a low catchment-to-lake area ratio,
and imbalance between sublacustrine spring-water inputs
and the surface water outputs which leads to an apparent
deficit of 20ls-1, with a residence time in the lake of nearly
100 years (Aeschbach-Hertig et al., 2002). Moreover, the
dominantly rocky (basaltic, trachyandesitic, granitic and
gneissic) geological substratum of the Lake Pavin area
(Assayag et al., 2008) and the absence of riverine inputs
and of agricultural activities in the vicinity of lake which
prevent eutrophication also add in driving Lake Pavin to
organic and inorganic nutrient limitation, primarily in its
late seasonal cycle with thermally stratified mixolimnion.
Such environmental conditions are well known to promote
temperate phage proliferation and lysogenic conversion
within bacterioplankton genomes (Wilson and Mann,
1997).
Nutrient-commanded trade-off between lytic versus
lysogenic decisions
The concentration of nutrients which often limits bacterial
growth and activity has been proposed as one of the
environmental factors that influence the lysogenic deci-
sion (Levin and Lenski, 1983), by modulating the meta-
bolic activity of host populations. In our experimental
study, the addition of nutrients did not interfere nor interact
significantly with mitomycin C (as inferred from two-way
ANOVA), which strongly suggests that prophage induction
was stimulated under low natural nutrient concentration,
i.e. for both U and VR samples. In contrasts, addition of
organic and inorganic nutrients in experimental micro-
cosms failed to significantly induce prophages. For both
experiments, lytic viruses were proliferating with manually
added nutrient, clearly because of the stimulation of bac-
terial growth potential. In contrast, lysogenic decision was
enhanced in natural short-nutrient conditions. Compari-
son of natural microbial microcosm (i.e. U) with the
so-called viral-reduced microcosms (VR) known from
Expt. 2 implies that the ambient nutrient levels, together
with low virus-to-host ratios and the related contact rate
probability, are prevalent in the lysogeny conversion deci-
sion. The FLC levels in U microscosms were relatively
lower than in the VR samples (Table 2), where the com-
petition for nutrient was enhanced among concentrated
bacterial populations exposed to weak viral hit. Most of
the grazers (> 98%, data not shown) whose activities are
known as a major source of organic and inorganic nutri-
ents production in the plankton (Nagata, 2000) were
474 A. S. Pradeep Ram and T. Sime-Ngando

removed from our microcosms, in order to maintain intact y = -110.55x2 + 131.2x - 10.231
R2 = 0.65
the ambient nutrients and the related competitiveness 35.0
within bacterial communities. In previous freshwater
30.0
microcosm experiments, we have indeed shown that
nutrient regeneration from grazers can have a strong 25.0
influence on the interactions between viruses and their

FIC (%)
20.0
hosts (Sime-Ngando and Pradeep Ram, 2005; Pradeep
15.0
Ram and Sime-Ngando, 2008). It is thus likely that energy
or nutrient limitations which are typical of most natural 10.0
aquatic systems (Pradeep Ram and Sime-Ngando, 2008) 5.0
are major instigators of viruses to shift their infection strat-
0.0
egy and increase the lysogeny maintenance strategy over
0.0 0.2 0.4 0.6 0.8 1.0
the lytic lifestyle (and vice versa). The mechanisms
behind such trade-off behaviour are not clear but we
y = 4.929x2 - 5.8077x + 2.0494
believe that (i) decrease in the viral to host contact rates 1.4 R2 = 0.76
and/or (ii) weak bacterial growth rates in response to 1.2
harsh nutrient conditions are important explicative factors.
1.0
Furthermore, our case study also suggests that the type

of the limiting nutrient is an important issue for the lysog-
eny versus lytic decisions. Our experiments have shown
that organic carbon stimulated bacterial growth more than
conservative N and P, leading to higher BS, FIC and
FIC : FLC ratio. Although we did not measure the natural
concentration of organic carbon during our study, this
suggests that the bioavailability of labile carbon was
prevalent over that of inorganic N and P. Indeed, N : P
atomic ratio (< 10) at the start of experiments was under
the limiting Redfield threshold. In marine systems such as
the Gulf of Mexico where the severely limiting nutrient is
inorganic P, studies have shown that this nutrient rather
than others is in the centre of the trade-off between lytic
versus lysogenic back-and-forth in the context of virus
host interactions in aquatic microbial ecology (Williamson
et al., 2002; Williamson and Paul, 2004).
Ecological inferences
Our findings of higher lytic over temperate phage produc-
tion (by one order of magnitude) and the significant
inverse relation of FLC with bacterial abundance and
growth rates were suggestive of alteration of physiological
and metabolic status of natural bacterial cells (Del Giorgio
et al., 1997), which directly depend on the protein
repressors (Ptashne, 1992). The above relation thereby
supports lysogeny as a survival strategy for viruses
(Weinbauer et al., 2003; Colombet et al., 2006), which
can in this way maintain both their genomic and pheno-
typic traits in the environment at low host density and
growth rate (Williamson et al., 2002; Weinbauer et al.,
2003). Negative relationship between FLC and FIC in our
experimental samples could thus be viewed as antago-
nistic interaction between lysogenic and lytic activities in
relation to the resource availability, as previously indicated
in freshwater (Colombet et al., 2006) and marine systems
log FLC
0.8
0.6
0.4
0.2
0.0
0.0 0.2 0.4 0.6 0.8 1.0
Bacterial growth rate ( day -1 )
Fig. 4. Relationships between the frequencies of lytically (FIC) and
lysogenically (FLC) infected bacterial cells with bacterial growth
rates in the experimental samples.
(Weinbauer et al., 2003). Our experimental study inferred
that lysogeny tend to occur during periods or at depths of
low nutrient concentrations or when bacteria are exposed
to low viral strikes.
Changes in the FLC and FIC with respect to bacterial
growth rate were best explained by a second-order poly-
nomial regression model (Fig. 4), indicating that albeit
nutrient addition led the temperate phage genome to
enter into lytic pathway, there was still a small fraction of
lysogens which were not affected. Similarly, most of the
bacteria in our samples were not concerned by viral activi-
ties of interest, probably including phage-resistant bacte-
ria in spite of the general large competitive cost
associated with resistance, primarily in oligotrophic con-
ditions. In such conditions, sensitive cells may even out-
compete resistant ones as bacteria might develop decoy
receptors to invite infection, in order to assess nutrient-
riched phage protein and DNA (Fuhrman, 1999; Wein-
bauer, 2004). Maintaining the capacity to respond to
nutrient influxes and installing a gradient of sensitivity to
ambient phages, i.e. from resistance, abortive phage
infection, cryptic phage infection (i.e. lysogeny,
pseudolysogeny) to lytic infection, is a matter of trade-off
within virushost communities relative to resources. This

2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479

could well be understood by the fact that trade-offs do
occur along a continuum from resistance to highly com-
petitive ability in natural aquatic systems, which allows
potentially competing bacterial populations to coexist
(Bohannan et al., 2002). The concept of evolutionary
trade-off often assumes that resource scarcity and abun-
dance are major determining factors for lifestyles. Based
on the above concept, lysogenic phages can be classified
as k-strategists or equilibrium species (Suttle, 2007;
Sime-Ngando and Colombet, 2009). This concept can
also be extended to resource-independent survival
mechanisms such as resistance to sunlight-induced
nucleic acid damage, genetic repair capacity, temperature
or salinity tolerance, which can strongly influence viral
lifestyles (Weinbauer et al., 2007). The magnitude of
trade-offs can thus have both direct and indirect effects on
a community, and understanding their nature would allow
a better prediction of the response of a community struc-
ture and dynamics to environmental changes.
Conclusions
Overall, this study presents original experiments on the
interactions between resources, bacteria and viral lif-
estyles (lytic versus lysogenic) in a site that offers a
unique geological and catchment peculiarities, with low
allochthonous inputs primarily during the target nutrient-
depleted study period. The experimental design adopted
avoids internal nutrient regeneration (i.e. from grazers)
but favours manually added nutrients, with an incubation
time (24 h) that was realistic with intrinsic metabolic rates
and generation time of the main biological components
under study. In conjunction with the volume of micro-
cosms (5 l), we consider that the bottle effect due to
confinement was largely weakened during the short-term
incubations. This experimental design has established
added nutrients, primarily organic nutrients in the case of
our study site, as one of the major driving forces which
could dictate the balance between the two viral lifestyles
explored, by enhancing bacterial growth and, as a conse-
quence, the virushost contact probability and the adsorp-
tion, replication and proliferation of lytic phages. In poor
nutrient conditions, such as in our natural environment,
lytic and temperate phages coexist to a large extent but
with higher trade-off advantage towards the lysogenic
lifestyle. From our microcosm experiments, we were able
to prove that nutrients stimulated lytic viral production via
enhanced host growth and, when limiting, promote
lysogenic conversion. Our experiments followed a
community-level approach and the results represent
average responses by bacteria and viruses. Both compo-
nents are not homogenous assemblages of organisms,
and there could be species-specific differences in nutrient
limitations as well as differences resulting from patchiness
2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479
Nutrient effects on lytic versus lysogenic viruses 475
(metabolic hot spots). This could result in more complex
interactions between resources, viruses and bacteria in
natural conditions.
Experimental procedures
Study site and sampling
Experimental samples originated from Lake Pavin (altitude,
1197 m above sea level), an oligomesotrophic, meromictic
and dimictic lake (volume, 22.98 106 m3) with partial over-
turns, located in the French Massif Central (4529N, 256E).
It is a typical crater mountain lake characterized by a
maximum depth of 92 m, with a small surface area (44 ha)
about equal to the catchments area (50 ha). There is there-
fore a very low inflow of organic inputs from the drainage
basin, and the watershed essentially consists of beech
forests. This site offers a unique environment with low human
influences and high annual reproducibility of seasonal
dynamics in the water column. The annual pigment mean
concentration is at 3 mg chlorophyll a l-1, while the typical
annual mean values for total nitrogen and total phosphorus
are at 57 and 5 mM respectively (Rioual, 2002).
Two microcosm experiments were performed on different
dates, Expt. 1 (19 September 2007) and Expt. 2 (17 October
2007), with water samples collected from a fixed surface
depth (1 m) in the deepest central point of the lake. All experi-
mental samples were collected in triplicates with a Van Dorn
bottle and filled in 40 ( 3 replicates) l capacity-insulated
carboys, which had been previously cleaned with 1.2 N HCl
and rinsed three times with Milli-Q and lake water. Water
samples were maintained at in situ water temperature during
transport to the laboratory. The sampling periods corre-
sponded to the clear water phase, with a euphotic layer of
about 25 m (estimated from Secchi depth).
Experimental design
For the two experiments, natural water samples were pre-
filtered through a succession of nylon fibre sieves (150, 63
and 25 mm porosities) and were sequentially filtered through
two membrane barriers (3- and 0.8-mm-pore-size cellulose
acetate filters, Sartorius) under attenuated light conditions
and low differential pressure (< 50 mmHg), to yield grazer-
free fractions which were confirmed by microscopic exami-
nation. Upon completion of filtration, these experimental
microbial samples were acclimatized for 2 h in the dark at in
situ conditions and then equally distributed in 5 l capacity
polycarbonate bottles previously washed with 1.2 N HCl and
rinsed three times with Milli-Q water and the experimental
water of interest (Fig. 5). For Expt. 1, each set of bottles
received organic and inorganic nutrients, individually and in
combinations, including organic carbon (glucose 100 mM C),
nitrate (NaNO3 25 mM N) and phosphate (NaH2PO4 5 mM P)
in five different treatments, namely C, N, P, NP and CNP.
Nutrients were added to reach up to their maximum concen-
trations known from the annual range recorded for the current
studied year (i.e. 2007) in the euphotic layer of Lake Pavin (S.
Rasconi, pers. comm.). One set of bottles that received no
nutrient addition served as unamended control (U) (Fig. 5A).
476 A. S. Pradeep Ram and T. Sime-Ngando
Fig. 5. Diagrammatic representation of two experiments (A = Expt.
1; B = Expt. 2) designed to investigate the short-term (24 h) effects
of added organic and inorganic nutrients on two viral lifestyles (lytic
versus mitomycin-induced lysogenic infections) in microbial
microcosms (< 0.8 mm fraction). Experimental samples were
collected in the surface waters of Lake Pavin, France.
For Expt. 2, each set of bottles received different sources
of organic matter: crude bacterial lysate (BL, dissolved
organic carbon 200 mM C, see below), organic phosphate
(OP, glycerophosphate 10 mm P final concentration) and
2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479
amino acid (AA, glutamic acid 100 mM final concentration).
One set of bottles that received no organic matter addition
served as unamended control (U). In addition, a set of bottles
containing VR of water samples which received no organic
nutrient addition was also included in the experimental design
(Fig. 5B). For obtaining the VR, the method of Weinbauer and
Suttle (1996) was employed to reduce the ambient back-
ground levels of viruses in the initial experimental samples.
Briefly, these samples were filtered through GF/C filters
(Whatman, England) and the bacteria in the filtrate was con-
centrated 10-fold by ultrafiltration using a high-performance
concentration/diafiltration system (Model DC 10LA, Amicon)
equipped with a reusable hollow fibre cartridge of 0.2 mm
cut-off and surface area of 0.45 m2 (Amicron, Epernon,
France), entry pressure 0.9 bar. An aliquot (50 ml) of the
bacterial concentrate was made up to 250 ml with virus-free
ultra filtered lake water. This procedure increased the bacte-
rial numbers by about twofold while decreasing the density of
viruses by 8- to 10-fold, compared with the concentrations of
both communities in the natural samples. Reducing the abun-
dances of viruses decreased contact rates between viruses
and bacteria and, we assumed, the lytic production of new
viruses. Bacterial lysate inoculum was obtained from VR
assays, corresponding to approximately two times the in situ
concentration of natural bacterial which were disrupted by
ultrasonication (5 2 min using the Bandelin Sonoplus model
HD 2070 sonicator, Bandelin electronic, Berlin) prior to
experimental set-up (Fig. 5B). The amount of organic carbon
obtained from BL and added to experimental bottles was
measured using Apollo 9000 TOC analyser.
All organic (C, OP and AA) and inorganic nutrients (N and
P) used for the experiments were purchased from Sigma
Chemical Company, USA. All experimental treatments were
prepared from the triplicate samples collected initially (i.e.
6 3 replicates for Expt. 1 and 5 3 replicates for Expt. 2
respectively) and incubated in the dark at in situ temperature
(14 1C) conditions for 24 h. Subsamples were taken at t0
in the initial experimental samples (i.e. before treatments
were set up) and at t24h in all treatments, for the assessment
of bacterial and viral variables of interest (see below). Pre-
cautionary measures were taken to avoid contamination of
the experimental samples. All containers, filtering devices,
glassware and tubing coming in contact with the experimental
samples were acid-washed (10% hydrochloric acid) and thor-
oughly rinsed with de-ionized water prior to use.
Enumeration of viruses and bacteria, BGR
Viruses (VA) and bacteria (BA) in formaldehyde (2% final
concentration) preserved samples were filtered (12 ml,
< 15 kPa vacuum) onto 0.02-mm-pore-size Anodisc mem-
branes (Whatman, Maidstone, UK), stained with SYBR
Green I (Molecular Probes Europe, Leiden, the Netherlands),
and mounted between microscope slides and glass cover-
slips using a mixture of 80% AF1 Citifluor (Citifluor, London,
England) and 20% Vectashield (Vector Laboratories, Burlin-
game, CA, USA) as antifading mounting medium. Counts
were performed under a Leica DC 300F epifluorescence
microscope as described by Noble and Fuhrman (1998).
When not analysed immediately, slides were stored at -20C
and counted within a week. Heterotrophic prokaryotes were

distinguished from viruses on the basis of their relative size
and brightness. A blank was routinely examined to control for
contamination of the equipments and reagents. Bacterial-
specific growth rates (BGR, m) during the incubation time
interval were calculated assuming exponential growth:
m = (ln N2 - ln N1)/T, where N1 and N2 are the bacterial abun-
dances at t0 and t24h respectively.
Lytic infection
Viral lytic infection was inferred from the FVIC according to
Sime-Ngando and colleagues (1996). Bacteria contained in
8 ml of formalin-fixed subsamples were collected on electron
microscope grids (400-mesh, carbon-coated Formvar film)
using ultracentrifugation, positively contrasted with Uranyl
acetate (2% final concentration at room temperature), rinsed
twice with 0.02 mm filtered distilled water, and dried on a
filter paper. Grids were examined using a JEOL 1200Ex
transmission electron microscope operated at 80 kV at a
magnification of 20 00060 000, to distinguish between
prokaryotic cells with and without intracellular viruses. At
least 500 prokaryote cells were inspected per grid to deter-
mine FVIC. Burst size (BS, viruses bacterium-1) was esti-
mated for every single sample as average number of viral
particles in all visibly infected bacteria totally filled with
phages, i.e. the maximum BS. Because mature phages are
visible only late in the infection cycle, FVIC measurements
were converted to the FIC using the equation FIC = 9.524
FVIC - 3.256 (Weinbauer et al., 2002). Assuming that, in
steady state, the infected and uninfected cells were grazed
at the same rate, and that the latent period equalled
the prokaryotic generation time, FIC was converted to
bacterial mortality (VIBM, as percentage of prokaryotic
production) using the equation: VIBM = (FIC + 0.6FIC2)/
(1 - 1.2FIC) (Binder, 1999).
Prophage induction assays for lysogenic infection
For both experiments, the approach of Jiang and Paul
(1996; Paul and Jiang, 2001) was used to induce prophages
in bacterioplankton. At t0, 25 ml of triplicate experimental
samples were treated with the mutagen mitomycin C
(1 mg ml-1 final concentration, Sigma) or left untreated as
mitomycin C controls. These samples were then incubated
statically at in situ conditions in the dark for 24 h (i.e. in
parallel to the incubation of nutrient-amended treatments),
and fixed with 0.02 mm filtered formalin (2% final concentra-
tion). Slides were immediately prepared and stored frozen
(-20C) and counted within a week as described previously.
The FLC within bacterial communities was determined from
two different methods. For the first approach termed the BS
method, FLC was calculated from viral abundances in mito-
mycin C-treated (VAMC) and control (VAC) essays after incu-
bations, and from the bacterial abundance (BAt0) and
maximum burst size (BSt0) in original microbial samples:
FLC = [(VAMC - VAC)/(BSt0 BAt0)]100 (Weinbauer et al.,
2003). For the second approach termed FVIC method, FLC
was inferred from the occurrence of visibly infected bacterial
cells due to the presence of mitomycin C (i.e. as compared
with control) at the end of the incubation period.
2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479
Nutrient effects on lytic versus lysogenic viruses 477
Physicochemical analysis
Water temperature and dissolved oxygen concentration were
measured in situ with a WTW OXI 197 multiparametric probe.
Samples for initial nutrient concentrations, namely total nitro-
gen (TN), soluble reactive phosphorus (SRP) and total phos-
phorus (TP), were analysed spectrophotometrically (Tartari
and Mosello, 1997). Total organic carbon concentrations
were determined by high temperature combustion in an
Apollo 9000 TOC analyser set at 700C and calibrated with
standard additions of potassium hydrogen phthalate (KHP);
precision = 0.2 mM (APHA, 1998). The initial chlorophyll a
concentration was determined spectrophotometrically from
samples (500 ml) collected on Whatman GF/F filters. Pig-
ments were extracted in 90% acetone overnight in the dark at
4C, and concentrations were calculated from SCOR-
UNESCO (1966) equations. Nutrients and chlorophyll a con-
centrations were analysed in the triplicate samples collected.
Statistical analysis
The response of bacterial and viral parameters to different
nutrient (inorganic and organic) additions, with or without
mitomycin C, was tested by two-way ANOVA. A significant
response was considered to be at P < 0.05. Potential rela-
tionships among various measured variables were tested by
Pearsons correlation analysis. All statistical analyses were
performed using Minitabs software for Windows (Release
12, Minitab).
Acknowledgements
A.S.P.R. was supported by a Research Associate fellowship
from the French ANR Biodiversit. This is a contribution to the
Research Program ANR Biodiversit AQUAPHAGE. We
thank S. Rasconi, M. Jobard, L. Jouve and C. Portelli for their
logistic and field assistance.
References
Ackermann, H.W., and Dubow, M.S. (1987) Viruses of
prokaryotes. In General Properties of Bacteriophages, Vol.
1. Ackermann, H.W., and Dubow, M.S. (eds). Boca Raton,
CA, USA: CRC Press, p. 202.
Aeschbach-Hertig, W., Hofer, M., Schmid, M., Kipfer, R., and
Imboden, D.M. (2002) The physical structure and dynam-
ics of a deep, meromictic crater lake (Lac Pavin, France).
Hydrobiologia 487: 111136.
APHA (1998) Standard Methods for the Examination of Water
and Wastewater, 20th edn. Washington, DC, USA:
American Public Health Association.
Assayag, N., Jzquel, D., Ader, M., Viollier, E., Michard, G.,
Prvot, F., and Agrinier, P. (2008) Hydrological budget,
carbon sources and biogeochemical processes in Lac
Pavin (France): constraints from d18O of water and d13C of
dissolved inorganic carbon. Appl Geochem 23: 2800
2816.
Bettarel, Y., Bouvy, M., Dumont, C., and Sime-Ngando, T.
(2006) Virusbacterium interactions in water and sediment
478 A. S. Pradeep Ram and T. Sime-Ngando
of West African Inland aquatic systems. Appl Environ
Microbiol 72: 52745282.
Binder, B. (1999) Reconsidering the relationship between
viral induced bacterial mortality and frequency of infected
cells. Aquat Microb Ecol 18: 207215.
Bohannan, B.J.M., Kerr, B., Jessup, C.M., Hughes, J.B., and
Sandvik, G. (2002) Trade-offs and coexistence in microbial
microcosms. Antonie Van Leeuwenhoek 81: 107115.
Bratbak, G., Egge, J.K., and Heldal, M. (1993) Viral mortality
of the marine alga Emiliania huxleyi (Haptophyceae)
and termination of algal blooms. Mar Ecol Prog Ser 93:
3948.
Cochran, P.K., and Paul, J.H. (1998) Seasonal abundance of
lysogenic bacteria in a subtropical estuary. Appl Environ
Microbiol 64: 23082312.
Cochran, P.K., Kellogg, C.A., and Paul, J.H. (1998) Prophage
induction of indigenous marine lysogenic bacteria by envi-
ronmental pollutants. Mar Ecol Prog Ser 164: 125133.
Colombet, J., Sime-Ngando, T., Cauchie, H.M., Fonty, G.,
Hoffmann, L., and Demeure, G. (2006) Depth-related gra-
dients of viral activity in Lake Pavin. Appl Environ Microbiol
72: 44404445.
Danovaro, R., and Corinaldesi, C. (2003) Sunscreen prod-
ucts increase virus production through prophage induction
in marine bacterioplankton. Microb Ecol 45: 109118.
Del Giorgio, P.A., Prairie, Y.T., and Bird, D.F. (1997) Cou-
pling between rates of bacterial production and abundance
of metabolically active bacteria in lakes, enumerated using
CTC reduction and flow cytometry. Microb Ecol 34: 144
154.
Fischer, U.R., and Velimirov, B. (2002) High control of bac-
terial production by viruses in a eutrophic oxbow lake.
Aquat Microb Ecol 27: 112.
Fuhrman, J.A. (1999) Marine viruses and their biogeochemi-
cal and ecological effects. Nature 399: 541548.
Hewson, I., and Fuhrman, J.A. (2007) Characterization of
lysogens in bacterioplankton assemblages of the southern
California borderland. Microb Ecol 53: 631638.
Hewson, I., ONeil, J.M., Fuhrman, J.A., and Dennison, W.C.
(2001) Virus-like particle distribution and abundance in
sediments and overlying waters along eutrophication gra-
dient in two subtropical estuaries. Limnol Oceanogr 46:
17341746.
Iyer, V.N., and Syzbalski, W. (1963) A molecular mechanism
of Mitomycin action: linking of complementary DNA
strands. Proc Natl Acad Sci USA 50: 355362.
Jiang, S.C., and Paul, J.H. (1996) Occurrence of lysogenic
bacteria in marine microbial communities as determined by
prophage induction. Mar Ecol Prog Ser 142: 2738.
Levin, B.R., and Lenski, R.E. (1983) Coevolution in bacteria
and their viruses and plasmids. In Coevolution. Futuyma,
D.J., and Slatkin, M. (eds.). Sunderland, MA, USA: Sinauer
Associates, pp. 99127.
Nagata, T. (2000) Production mechanisms of dissolved
organic matter. In Microbial Ecology of the Oceans.
Kirchman, D.L. (ed.). New York, USA: Wiley-Liss, pp. 121
152.
Noble, R.T., and Fuhrman, J.A. (1998) Use of SYBR Green I
for rapid epifluorescence counts of marine viruses and
bacteria. Aquat Microb Ecol 14: 113118.
Paul, J.H. (2008) Prophages in marine bacteria: dangerous
2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479
molecular time bombs or the key to survival in the seas?
ISME J 2: 579589.
Paul, J.H., and Jiang, S. (2001) Lysogeny and transduction.
In Methods in Microbiology, Vol. 30. Paul, J.H. (ed.). San
Diego, CA, USA: Academic Press, pp. 105125.
Pradeep Ram, A.S., Boucher, D., Sime-Ngando, T., Debroas,
D., and Romagoux, J.C. (2005) Phage bacteriolysis, pro-
tistan bacterivory potential, and bacterial production in a
freshwater reservoir: coupling with temperature. Microb
Ecol 50: 6472.
Pradeep Ram, A.S., and Sime-Ngando, T. (2008) Functional
responses of prokaryotes and viruses to grazer effects and
nutrient additions in freshwater microcosms. ISME J 2:
498509.
Prigent, M., Leroy, M., Confalonieri, F., Dutertre, M., and
DuBow, M.S. (2005) A diversity of bacteriophage forms
and genomes can be isolated from the surface sands of the
Sahara Desert. Extremophiles 9: 289296.
Ptashne, M. (1992) A Genetic Switch, 2nd edn. Cambridge,
Oxford, UK: Cell Press and Blackwell Scientific
Publications.
Rioual, P. (2002) Limnological characteristics of 25 lakes of
the French Massif Central. Ann Limnol 38: 311327.
Swstrm, C., Anesio, M.A., Granli, W., and Laybourn-
Parry, J. (2006) Seasonal viral loop dynamics in two large
ultraoligotrophic Antarctic freshwater lakes. Microb Ecol
53: 111.
SCOR-UNESCO (1966) Determination of Photosynthetic
Pigments in Sea Water. Paris, France: UNESCO.
Sime-Ngando, T., and Colombet, J. (2009) Viruses and
prophages in the dynamics of aquatic ecosystems (in
French). Can J Microbiol 55: 95109.
Sime-Ngando, T., and Pradeep Ram, A.S. (2005) Grazer
effects on prokaryotes and viruses in a freshwater micro-
cosm experiment. Aquat Microb Ecol 41: 115124.
Sime-Ngando, T., Mignot, J.P., Amblard, C., Bourdier, G.,
Devilettes, C., and Quilblier-Lloberas, C. (1996) Character-
ization of planktonic virus-like particles in a French moun-
tain lake: methodological aspects and preliminary results.
Ann Limnol 32: 15.
Stewart, F.M., and Levin, B.R. (1984) The population biology
of bacterial viruses: why be temperate? Theor Popul Biol
26: 93117.
Suttle, C. (2007) Marine viruses major players in the global
ecosystem. Nat Rev Microbiol 5: 801812.
Tapper, M.A., and Hicks, R.E. (1998) Temperate viruses
and lysogeny in Lake Superior bacterioplankton. Limnol
Oceanogr 43: 95103.
Tartari, G.A., and Mosello, R. (1997) Metodologie Analitiche e
Controlli di Qualit nel Laboratorio Chimico dellIstituto
Italiano di Idrobiologia del Consiglio Nazionale delle
Ricerche. Documenta Ist Ital Idrobiol, Vol. 60. Pallanza,
Italy.
Thingstad, T.F. (2000) Elements of a theory for the mecha-
nisms controlling abundance, diversity, and biogeochemi-
cal role of lytic bacterial viruses in aquatic systems. Limnol
Oceanogr 45: 13201328.
Weinbauer, M.G. (2004) Ecology of prokaryotic viruses.
FEMS Microbiol Rev 28: 127181.
Weinbauer, M.G., and Hfle, M.G. (1998) Significance of viral
lysis and flagellate grazing as factors controlling bacteri-

oplankton production in a eutrophic lake. Appl Environ
Microbiol 64: 431438.
Weinbauer, M.G., and Rassoulzadegan, F. (2004) Are
viruses driving microbial diversification and diversity?
Environ Microbiol 6: 111.
Weinbauer, M.G., and Suttle, C.A. (1996) Potential signifi-
cance of lysogeny to bacteriophage production and bacte-
rial mortality in coastal waters of the Gulf of Mexico. Appl
Environ Microbiol 62: 43744380.
Weinbauer, M.G., and Suttle, C.A. (1999) Lysogeny and
prophage induction in coastal and offshore bacterial com-
munities. Aquat Microb Ecol 18: 217225.
Weinbauer, M.G., Winter, C., and Hfle, M.G. (2002) Recon-
sidering transmission electron microscopy based esti-
mates of viral infection of bacterioplankton using
conversion factors derived from natural communities.
Aquat Microb Ecol 27: 103110.
Weinbauer, M.G., Brettar, I., and Hfle, M.G. (2003) Lysog-
eny and virus-induced mortality of bacterioplankton in
surface, deep and anoxic marine waters. Limnol Oceanogr
48: 14571465.
Weinbauer, M.G., Agis, M., Bonilla-Findji, O., Malits, A., and
Winter, C. (2007) Bacteriophage in the environment. In
2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 467479
Nutrient effects on lytic versus lysogenic viruses 479
Bacteriophage: Genetics and Molecular Biology. McGrath,
S., and Van Sinderen, D. (eds). London, UK: Caister Aca-
demic Press, pp. 6191.
Wilhelm, S.W., and Suttle, C.A. (1999) Viruses and nutrient
cycles in the sea. Bioscience 49: 781788.
Williamson, S.J., and Paul, J.H. (2004) Nutrient stimulation of
lytic phage production in bacterial populations of the Gulf of
Mexico. Aquat Microb Ecol 36: 917.
Williamson, S.J., Houchin, L.A., McDaniel, L., and Paul, J.H.
(2002) Seasonal variation in lysogeny as depicted by
prophage induction in Tampa Bay, Florida. Appl Environ
Microbiol 68: 43074314.
Williamson, S.J., Radosevich, M., Smith, D.W., and
Wommack, K.E. (2007) Incidence of lysogeny within tem-
perate and extreme soil environments. Environ Microbiol 9:
25632574.
Wilson, W.H., and Mann, N.H. (1997) Lysogenic and lytic
production in marine microbial communities. Aquat Microb
Ecol 13: 95100.
Windholz, M., Budavari, S., Stroumtsos, L.Y., and Fertig,
M.N. (1976) The Merck Index. Rahway, NJ, USA: Merck. &
Co.