Marine prasinovirus genomes show low evolutionary divergence and acquisition of protein metabolism genes by horizontal gene transfer

Abstract

Although marine picophytoplankton are at the base of the global food chain, accounting for half of the planetary primary production, they are outnumbered 10 to 1 and are largely controlled by hugely diverse populations of viruses. Eukaryotic microalgae form a ubiquitous and particularly dynamic fraction of such plankton, with environmental clone libraries from coastal regions sometimes being dominated by one or more of the three genera Bathycoccus, Micromonas, and Ostreococcus (class Prasinophyceae). The complete sequences of two double-stranded (dsDNA) Bathycoccus, one dsDNA Micromonas, and one new dsDNA Ostreococcus virus genomes are described. Genome comparison of these giant viruses revealed a high degree of conservation, both for orthologous genes and for synteny, except for one 36-kb inversion in the Ostreococcus lucimarinus virus and two very large predicted proteins in Bathycoccus prasinos viruses. These viruses encode a gene repertoire of certain amino acid biosynthesis pathways never previously observed in viruses that are likely to have been acquired from lateral gene transfer from their host or from bacteria. Pairwise comparisons of whole genomes using all coding sequences with homologous counterparts, either between viruses or between their corresponding hosts, revealed that the evolutionary divergences between viruses are lower than those between their hosts, suggesting either multiple recent host transfers or lower viral evolution rates.

FIG. 1.

Schematic representation of colinearity of common prasinovirus genes. The six prasinovirus genomes are represented proportionally to their respective sizes, and each CDS is shown as a small open rectangle. Each rectangle has an identical size, unrelated to its real size, except for the two big extra CDSs found in BpV1 and BpV2. The genes common to the six viral genomes are shaded, similarly and joined to their orthologs in neighboring genomes by lines. At right are Venn diagrams showing the numbers of common and specific genes among prasinoviruses: the three Ostreococcus viruses OlV1, OtV1, OtV5 (A), the two Bathycoccus viruses (B), and the six prasinoviruses, showing counts of genes specific to each genus and those common to two or three genera (C).

FIG. 2.

Phylogeny of NCLDV, including diverse eukaryote and prokaryote sequences, based on Bayesian inference (BI) and maximum likelihood (ML) analyses. Black dots indicate posterior probabilities (pp; BI) of >0.9 and bootstrap proportions (bp; ML) of >70%, and white dots indicate pp of >0.9 or bp of >70%. The overall phylogeny of diverse lineages was accomplished using the full-length DNA polymerase B, predicted amino acid sequence, which is a gene conserved in all NCLDV and all prokaryotes and eukaryotes. To resolve the branches concerning prasinoviruses and their hosts more clearly (boxes), the amino acid sequences of five conserved genes common to both hosts and viruses (DNA polymerase B, proliferating cell nuclear antigen, ribonucleotide reductase large and small subunits, and thymidine synthase) were concatenated, permitting comparison of evolutionary distances. Numbers on branches are maximum likelihood/Bayesian inference support values.

FIG. 3.

Distribution of amino acid identities of three host proteomes versus their respective viruses. (A) O. tauri RCC745 versus O. lucimarinus CCMP2972 with OtV5 versus OlV1. (B) O. tauri RCC745 versus Micromonas (M) sp. RCC299 with OtV5 versus MpV1. (C) O. tauri RCC745 versus Bathycoccus sp. RCC1105 with OtV5 versus BpV1. n, number of orthologous genes analyzed; av, average amino acid identity. All 15 possible host-virus pairwise comparisons showed the same trends.

FIG. 4.

(A) Gene clusters encoding amino acid synthesis pathways. Representation of CDS positions located between bp 17000 and about bp 30000 for BpV1, MpV1, and OtV5. (B) Amino acid biosynthesis pathways encoded by Micromonas or Ostreococcus viruses: biosynthesis pathways for Leu, Ile, Val, Tyr, Trp, and Phe. Text written in black corresponds to reactions catalyzed by enzymes encoded by both Micromonas and Ostreococcus viruses, text in red represents reactions with enzymes found only in the Micromonas genome, and text in pink represents reactions specific to the Ostreococcus genomes. Note that none of these enzymes is present in the two Bathycoccus viral genomes.