23914.6 Activities of Viruses
14.6.1 Lytic Activity and Impact on Bacterial
Community Composition
and Metabolism
It is now well accepted that lysis is one of the main causes of
microbial mortality in aquatic systems. Based on our method
(described previously) for direct observation of infected
cells, the seasonal study conducted in Lake Pavin in 2000
yielded one of the fi rst estimates of the impact of lytic viruses
on bacterial communities (Bettarel et al. 2004 ). To estimate
the impact of viruses on bacterial mortality, as percent of
bacterial biomass production loss, a series of conversions
must be applied (Binder 1999 ; Weinbauer et al. 2002 ).
The levels and the seasonal fl uctuations in the basic lytic
parameter (i.e. FVIC) in Lake Pavin have been described
previously (see section 17.3). The translation into viral pro-
duction yields an average of 80 × 10^6 viruses liter −1 hour −1
produced in the mixolimnion , with the highest values
recorded in the metalimnion (Table 14.1 ). This corresponds
to a mean bacteria production loss due to viral lysis of 16 %,
which is lower than the grazing impact from fl agellates
(38 %), but higher than that from bacterivorous ciliates (3 %)
(Bettarel et al. 2004 ). Viral bacteriolysis levels in the mixo-
limnion of Lake Pavin appear conservative, compared to the
world marine and freshwaters where viral-mediated mortal-
ity averages 10–50 % of the production of heterotrophic pro-
karyotes, and is considered approximately equal to the
bacterivory from grazers (Fuhrman and Noble 1995 ; Pradeep
Ram et al. 2005 ). Based on the combined average mortality
related to viral lysis and protistan bacterivory, a signifi cant
fraction of bacterial production in the mixolimnion of Lake
Pavin is controlled by factors other than viral lysis and pro-
tistan predation, such as metazoan bacterivory. This is par-
ticularly credible during the period of thermal stratifi cation ,
which usually coincides with the maximum development of
metazoan zooplankton, which then obtain most of their
resources from intense bacterivory (Thouvenot et al. 1999 ).
During autumn, all of the bacterial production in Lake Pavin
is balanced by viral lysis and protistan grazing (Bettarel et al.
2004 ).
The impacts of viral lysis and ciliate grazing in Lake
Pavin were highest in the metalimnion, while that from fl ag-
ellates was highest in the hypolimnion (Table 14.1 ). From a
study conducted in 11 lakes located in the French Massif
Central (including Lake Pavin), we found that the magnitude
of prokaryotic mortality due to viruses and fl agellates varies
with sampled depth (Pradeep Ram et al. 2013 ). The average
ratio of viral mortality to fl agellate grazing of prokaryotes in
the euphotic depth was 0.5, with lowest ratio (0.1) observed
in Lake Pavin (Fig. 14.6C ). In the aphotic zone, the domi-
nance of viruses as a factor of prokaryotic mortality was
largely due to signifi cantly higher viral infection which was
accompanied by the signifi cantly lower grazing rates com-
pared to the euphotic zone. Lower levels of dissolved oxygen
at the aphotic compared to euphotic depth explained for low
fl agellate grazing rates. A high-resolution vertical sampling
in Lake Pavin (Colombet et al. 2006 ) has confi rmed the pat-
tern of high relative abundance and activity of viruses and
bacteria in the monimolimnion where protistan fl agellates
were almost absent (Fig. 14.7 ). Further seasonal investiga-
tions of depth-related gradients in microbial components in
Lake Pavin (Colombet et al. 2009 ; Colombet and Sime-
Ngando 2012a ) have provided evidence that the ecology of
the deep, cold, dark and permanently anoxic monimolimnic
waters may be driven by viral loop (dissolved organic mat-
ter–prokaryotes–viruses) processes, which can sequester
organic matter and nutrients for a long-lived turnover time.
In general, under anoxic conditions where most of the
eukaryotes are absent, the predominance of viruses vs graz-
ers as a factor of bacterial mortality in the aquatic environ-
ment is still poorly understood at the community level.
Indeed, it has been reported that high viral control of bacte-
rial production occurs either at reduced or high grazing
activities (Bettarel et al. 2004 ), while experimental studies
have provided evidence of synergistic interactions between
viral bacteriolysis and protistan bacterivory. In these interac-
tions, the presence of grazers appears as a stimulating factor
for prokaryotic growth and viral proliferation (Sime-NgandoTable 14.1 Mean seasonal contributions of viruses, phagotrophic fl agellates, and ciliates to bacterial mortality and of fl agellates to viral losses in
the three main layers of the mixolimnion of Lake Pavin
Sampled depth% (CV) of bacterial production removed by: % (CV) of viral production
Viral lysis Flagellate grazing Ciliate grazing removed by fl agellates
Epilimnion 13.6 (63.1) 25.5 (117.4) 1.6 (137.1) 4.8
Metalimnion 18.8 (53.2) 23.4 (112.2) 4.6 (122.0) 3.6
Hypolimnion 16.1 (124.0) 64.2 (88.8) 1.8 (93.6) 3.8
Mean 16.2 (13.1) 37.7 (49.8) 2.7 (50.3) 4.1 (12.3)
CV coeffi cient of variation
14 Ecology of Viruses in Lake Pavin