Metagenomic Revelations

A review of: Kristensen DM, Mushegain AR, Dolja VV, Koonin EV (2009) New dimensions of the virus world discovered through metagenomics. Trends in Microbiology, 18 (1): 11-19.

The study of marine viruses using electron and fluorescent microscopy has revealed viruses as the most abundant microbes in the sea, with approximately 10⁶ to 10⁹ virus particles per millilitre of sea water.

Viral metagenomics is conducted by sequencing samples enriched with virus particles and through detection of known viral-specific sequences in available databases. Recent viral metagenomic analysis of double-stranded DNA (dsDNA) sequences have shown marine viromes as the most genetically diverse biological communities with an estimated >10³⁰ unique viral genotypes. The putative viral sequences were incongruent to previously determined sequences, and sequences with detectable homologues seem to be of bacterial origin.

Using computational methods, a database containing protein coding genes for known viruses and prokaryotes, virus enriched samples from the Sargasso Sea, the Arctic, the Gulf of Mexico and the coast of British Columbia; the authors were able to elucidate viral sequences statistically different to that of known bacterial and viral genomes. This implies that the viral genes identified reflect the true diversity of viruses, or the genes originated from cellular organisms and represent poorly conserved regions of the cellular pangenome, thus accounting for its low homology with known database sequences.

It is also possible that the diversity of cellular organisms is still poorly characterized, or the sequences are a mixture of bacterial and viral genes from gene transfer agents (GTAs) which originate from defective prophages and cannot replicate like “true viruses”. The prevalence of specialized vehicles like GTAs provides new insights to current ideas on the structure and dynamics of the viral and prokaryotic world, as well as the movement and exchange of genes in the marine environment.

The diversity and host range of nucleocytoplasmic large DNA viruses (NCLDVs) which infect eukaryotes, has also been made apparent from analysis of metagenomic sequences from the Sargasso Sea, Atlantic and Pacific Oceans. BLAST searches and comparisons showed that NCLDVs are ubiquitous in the oceans, deemed to be dominated by prokaryotic organisms. Phycodnaviruses appear to be the most abundant, followed by mimiviruses and a much lower occurrence of apparent relatives of iridoviruses and asfarviruses. The source samples were enriched for prokaryotic cells therefore NCLDV sequences may have been derived from picoeukaryotes, which are small enough to pass through bacterial filters, or from NCLDVs not assimilated into hosts. It is also possible that NCLDVs can infect prokaryotes.

Marine RNA viruses are poorly characterized due to their size and fragility, however metagenomics has expounded on the current understanding of virus evolution and global ecology. In contrast to dsDNA viromes, approximately 40% of RNA viral sequences taken from coastal waters of British Columbia and Hawaii exhibit significant similarity to known RNA viromes. A variety of RNA viruses infect plants and animals, but metagenomics indicates that marine RNA viruses consist of a single evolutionary lineage from the picornavirus-like superfamily. The reason for the pervasiveness of picorna-like viruses in the marine environment remains unclear.

Kristensen et al (2009) repeatedly present the possibility of unknown biases in sampling techniques which could explain the vast diversity of putative dsDNA virus sequences and the dominance of picorna-like viruses in RNA metagenomics. For example, the filtration of samples may selectively collect unicellular eukaryotes which are the primary hosts of the aforementioned RNA and dsDNA viruses. If this was the case then the analysis highlights the limitations of current metagenomic protocols.

Viral metagenomics is still in its infancy but has already revealed unexpected findings. Further sampling of the oceans and improvements on enrichment protocols and sequencing will enhance our current understanding of viral evolution and diversity, and provide further insights to their roles in marine ecology and geochemistry.