Assessing the biogeography of marine giant viruses in four oceanic transects

Abstract

Viruses of the phylum Nucleocytoviricota are ubiquitous in ocean waters and play important roles in shaping the dynamics of marine ecosystems. In this study, we leveraged the bioGEOTRACES metagenomic dataset collected across the Atlantic and Pacific Oceans to investigate the biogeography of these viruses in marine environments. We identified 330 viral genomes, including 212 in the order Imitervirales and 54 in the order Algavirales. We found that most viruses appeared to be prevalent in shallow waters (<150 m), and that viruses of the Mesomimiviridae (Imitervirales) and Prasinoviridae (Algavirales) are by far the most abundant and diverse groups in our survey. Five mesomimiviruses and one prasinovirus are particularly widespread in oligotrophic waters; annotation of these genomes revealed common stress response systems, photosynthesis-associated genes, and oxidative stress modulation genes that may be key to their broad distribution in the pelagic ocean. We identified a latitudinal pattern in viral diversity in one cruise that traversed the North and South Atlantic Ocean, with viral diversity peaking at high latitudes of the northern hemisphere. Community analyses revealed three distinct Nucleocytoviricota communities across latitudes, categorized by latitudinal distance towards the equator. Our results contribute to the understanding of the biogeography of these viruses in marine systems.

Fig. 1. Global map of sampling locations, colored by transect.

Blue dots indicate the start of cruise tracks.

Fig. 2. Unique genomes and genomes shared across different transects and depth strata.

The bar graphs show the intersections of genome distribution between the four transects (A) and water depth layers (B). Horizontal bars (right) indicate the total number of genomes found in each transect; black dots indicate the presence in one or multiple transects; the corresponding vertical bars indicate the number of genomes with the presence described by the dots.

Fig. 3. Distribution of giant viruses in each transect.

Giant virus distribution along the transects (A) GA02 (B) GA03 (C) GA10 (D) GP13. Each column represents a sampling location. The y-axis shows the number of different viral genomes that were recovered at a given location, separated into three depth ranges (2–80 m, 80–150 m, and 150–5500 m). Locations are arranged in increasing distance from left to right on the x-axis, based on their distance from the starting location and follow the indicated orientation (N to S for GA02 and GA03, W to E for GA10 and GP13). There are no samples collected at >150 m in the GA10 transect.

Fig. 4. General mapping statistics of viruses found in surface waters <150 m.

The y-axis shows the average abundance of a given individual genome (in RPKM), and the x-axis shows the number of samples from which the virus was recovered. Dots are colored by the viral order and dot sizes represent the length of the genomes.

Fig. 5. Distribution of viruses throughout the water column along the transects.

Graphs show the viral abundance (calculated in RPKM, natural log transformed) of (A) total giant viruses present in each of the transect, (B) viruses of the Imitervirales order, and (C) viruses of the Algavirales order. Samples were ordered based on the distance along transects, beginning from the first sampling location of cruise tracks (0 km). White dots denote the sampling location along the transect of each sample.

Fig. 6. The latitudinal pattern of giant virus diversity across the transect GA02.

Diversity was calculated using (A) Shannon’s H index and (B) Genome richness. Shannon index calculations were based on the viral abundances (in RPKM) produced from 10 M subsampled reads. Stars showing significant difference between two latitudinal groups (Wilcox test, p values < 0.05) (* <0.05, ** <0.01, *** <0.001, **** <0.0001) Panels left: Total virus community; center: Imitervirales communities; right: Algavirales communities. EQ Equator.

Fig. 7. Giant virus communities across latitudes.

A Relative abundance of giant virus communities along the GA02 transect. Pie charts indicate the community composition in each sampling location, with sizes corresponding to the total abundance of the Nucleocytoviricota community. Community composition between latitudinal locations NMDS ordination based on Bray-Curtis distance matrices of viral communities collected in the (B) GA02 transect, stress = 0.3 (C) All four bioGEOTRACES, stress = 0.3. Latitudinal groups are color-coded by sample locations at higher than 40°N/S, from 20° to 40°N/S, and below 20°N/S (equatorial). Ellipses represent 95% confidence intervals. Viral communities are significantly different between groups (Permanova p < 0.001).

Fig. 8. Functions encoded in the genomes of the six viruses that are widespread in oligotrophic waters.

On the x-axis, genomes are ranked from left to right in order of decreasing number of samples in which the viruses were detected; the horizontal colored bar shows the taxonomic family of the genome (purple: Imitervirales family 1, green: Algavirales family 1). The y-axis denotes the functional annotation found in genomes; putative genes are color-coded by functional categories. Gene function abbreviations: PPDK Pyruvate phosphate dikinase, GAPDH Glyceraldehyde 3-P dehydrogenase, SDH Succinate dehydrogenase, LHCB Chlorophyll a/b binding protein, ACAD Acyl-CoA dehydrogenase, ACBP Acyl-CoA binding protein, GMD GDP-mannose dehydrogenase, GMDH GDP-mannose 4,6 dehydratase, GlcNAc epimerase UDP-N-acetylglucosamine 2-epimerase, GNAT Glucosamine-6-phosphate N-acetyltransferase, PCNA Proliferating cell nuclear antigen, PI3K phosphatidylinositol 3-kinases, PDXK PD-(D/E)XK nuclease superfamily.