Lake Pavin History, geology, biogeochemistry, and sedimentology of a deep meromictic maar lake

(Chris Devlin) #1

236


observed. Infected prokaryotic cells were detected in the
whole sediment core, and their vertical distribution corre-
lated with both viral and prokaryotic abundances.
Pleomorphic ellipsoid VLPs were visible in fi lamentous
cells (Fig. 14.3 ) tentatively identifi ed as representatives of
the archaeal genus Methanosaeta , a major group of methane
producers on earth. We consider these empirical observa-
tions as preliminary spot fi ndings which may pave the way to
one of the last frontiers of our knowledge in environmental
microbiology: the connection between the viral world and
the prokaryotic species that consume or produce greenhouse
gases, such as methane. More details on methane cycle and
the related prokaryotic actors are provided in Chap. 16.
The above fi ndings which are typical of Lake Pavin cor-
roborate a study in the same lake which has unveiled a com-
plete shift in the composition of the prokaryotic assemblages
between the mixolimnion and the monimolimnion , with a
maximal Archaea / Bacteria ratio that is reached below mixo-
limnic layers (Lehours et al. 2005 ). It is thus likely that (i)
deep anoxic viruses in Lake Pavin include endemic, typical
populations, with different morphology and diversity charac-
teristics compared to surface water viruses, and (ii) the
potential viral infection of Archaea which are abundant in
the monimolimnion and in the sediments of Lake Pavin
could explain the depth-related changes in viral
morphology.


14.4.2 Molecular Genetic Diversity


Viral species are mostly known from their isolated hosts in
laboratory cultures, which, in the case of environmental sam-
ples, may not exceed 1 % of the total prokaryotes (Hugenholtz
et al. 1998 ). Molecular approaches are thus critical and offer
windows to the largest uncharacterized reservoir of diversity
on the earth. Polymerase chain reaction (PCR)-based meth-
ods are restricted to chosen viral groups as no gene is univer-
sally conserved among viruses, while metagenomics gives
access to an exhaustive view of uncultured viral diversity,
and has so far revealed an important unknown diversity and
an unexpected richness of viral communities (Edwards and
Rohwer 2005 ). In aquatic viral ecology, the most used
molecular fi ngerprinting approach is the pulse-fi eld gel elec-
trophoresis (PFGE), which separates PCR-generated
dsDNA products (Wommack et al. 1999 ). Application to a
series of spring samples collected in the mixolimnion o f
Lake Pavin and concentrated via the “pegylation” approach
yielded viral genomes in the size range from about 15 to 290
kbp, with a bulk around 35 kbp in the metalimnion. Visible
electrophoretic bands in our agarose gels, calculated both
from computer-scanning and densitometry profi ling, ranged
in number from 7 to 9 (Colombet et al. 2007 ). Above results


are within the ranges known from aquatic ecosystems (Sime-
Ngando and Colombet 2009 ).
In Lake Pavin, we have performed one of the fi rst studies
on the composition and diversity of freshwater viral com-
munities through metagenomics , based on 454 pyrose-
quencing derived virome of 649,290 reads with an average
length of 420 bp (Roux et al. 2012 ). In 0.2 μm fi ltered sam-
ples collected on June and July 2008, the proportion of reads
similar to protein sequences of the non-redundant NCBI
database was at 14 % (Fig. 14.5 ), which is among the highest
compared to published viromes (range 1–28 %, mean 6.3 %).
However, when our known fraction was determined using
reads randomly reduced to 100 bp as in previous studies, this
fraction dramatically dropped to 0.7 %, which is the lowest
among aquatic viromes. Among the known fraction, the
majority of reads (60 %) was most similar to non-viral
sequences, whereas our virome was not contaminated by
bacteria: the absence of 16S rRNA in our sample was
checked by PCR amplifi cation and BLAST search for 16S
rRNA sequences, and only a paucity of virome reads was
found to be partly similar to ribosomal protein. This result is
consistent with previous studies and is thought to be an indi-
cation of both the lack of viral gene annotation and the hori-
zontal gene transfers between viral and host genomes
(Edwards and Rohwer 2005 ).
Phylogenetic analyses performed on the major viral fami-
lies of our virome using different marker genes (Roux et al.
2011 ) highlight the dominance of unexpected single stranded
DNA viruses: Micro -, Nano - and Circoviridae , the later fam-
ily being the most abundant and represented 60 % of the total
sequences (Fig. 14.5 ). This sheds light on a hitherto unde-
scribed viral diversity for freshwaters. In addition, using 29
previously published viromes , the cluster richness in viral
communities was shown to vary between different environ-
ment types and appeared signifi cantly higher in marine eco-
systems than in other biomes. Furthermore, signifi cant
genetic similarity between viral communities of related envi-
ronments was highlighted as freshwater, marine and hyper-
saline environments were separated from each other despite
the vast geographical distances between sample locations
within each of these biomes (Roux et al. 2012 ). Overall, our
pioneer viral metagenomics in Lake Pavin spotlighted a
very broad diversity and previously unknown clades unde-
tectable by PCR and microscopic analyses. The Lake Pavin
virome appears closely related to other freshwater viromes ,
despite the signifi cant ecological differences between lakes.
Furthermore, freshwater viral communities appear geneti-
cally distinct from other aquatic ecosystems , demonstrating
the specifi city of freshwater viruses at a community scale.
This was the fi rst time that such results were reported. More
details on the metagenetics and metagenomics of prokary-
otes and viruses are provided in Chap. 15.

T. Sime-Ngando et al.
Free download pdf