BACKGROUND : The Antarctic continent is considered the coldest and driest place on earth with simple ecosystems,
devoid of higher plants. Soils in the ice-free regions of Antarctica are known to harbor a wide range of
microorganisms from primary producers to grazers, yet their ecology and particularly the role of viruses is
poorly understood. In this study, we examined the virus community structures of 14 soil samples from the
Mackay Glacier region.
METHODS : Viral communities were extracted from soil and the dsDNA was extracted, amplified using single-primer
amplification, and sequenced using the Ion Torrent Proton platform. Metadata on soil physico-chemistry was collected
from all sites. Both read and contig datasets were analyzed with reference-independent and reference-dependent
methods to assess viral community structures and the influence of environmental parameters on their distribution.
RESULTS : We observed a high heterogeneity in virus signatures, independent of geographical proximity. Tailed
bacteriophages were dominant in all samples, but the incidences of the affiliated families Siphoviridae and
Myoviridae were inversely correlated, suggesting direct competition for hosts. Viruses of the families Phycodnaviridae
and Mimiviridae were present at significant levels in high-diversity soil samples and were found to co-occur, implying
little competition between them. Combinations of soil factors, including pH, calcium content, and site altitude, were
found to be the main drivers of viral community structure.
CONCLUSIONS : The pattern of viral community structure with higher levels of diversity at lower altitude and pH,
and co-occurring viral families, suggests that these cold desert soil viruses interact with each other, the host,
and the environment in an intricate manner, playing a potentially crucial role in maintaining host diversity and
functioning of the microbial ecosystem in the extreme environments of Antarctic soil.
Additional file 1: Table S1. Sample overview. Coordinates of sampling
sites and sequencing metadata of viromes. Cleaned reads = reads after
removal of contaminants; reads for assembly = reads considered by the
assembly and mapping algorithm of CLC Genomics Workbench at the
default parameters. Table S2. Taxonomic diversity of the assembled
viromes. Diversity was assessed by MetaVir and abundant virus families
defined as representing more than 1% of significant hits. All values are
given as relative abundances in percent. Values for sample MGM, which
needed to be split in two parts for upload to MetaVir, represent both
parts (a and b; individual values for the separate parts are shown at the
bottom). Sign. hits = significant reference database hits of the predicted
genes at a maximum e value of 10−5. Families belonging to the order
Caudovirales are marked with an asterisk. Table S3. Environmental
parameters measured for all soil samples. ex cat = exchangeable cations,
variables designated with an asterisk were used in the RDA. Table S4.
Microbial diversity in samples as determined by metagenomic sequencing,
as percentage of the total reads assigned. Figure S1. Accumulation plots of
the different taxa present in the 14 Antarctic virome read datasets. A) family
level, B) genus level, C) species level.