Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Occurrence and seasonal dynamics of RNA viral genotypes in three contrasting temperate lakes

  • Ian Hewson ,

    Contributed equally to this work with: Ian Hewson, Kalia S. I. Bistolas, Jason B. Button, Elliot W. Jackson

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    hewson@cornell.edu

    Affiliation Department of Microbiology, Cornell University, Ithaca, NY United States of America

  • Kalia S. I. Bistolas ,

    Contributed equally to this work with: Ian Hewson, Kalia S. I. Bistolas, Jason B. Button, Elliot W. Jackson

    Roles Data curation, Formal analysis, Writing – original draft, Writing – review & editing

    Affiliation Department of Microbiology, Cornell University, Ithaca, NY United States of America

  • Jason B. Button ,

    Contributed equally to this work with: Ian Hewson, Kalia S. I. Bistolas, Jason B. Button, Elliot W. Jackson

    Roles Data curation, Formal analysis, Writing – original draft, Writing – review & editing

    Current address: Illumina Corporation, San Diego, CA United States of America

    Affiliation Department of Microbiology, Cornell University, Ithaca, NY United States of America

  • Elliot W. Jackson

    Contributed equally to this work with: Ian Hewson, Kalia S. I. Bistolas, Jason B. Button, Elliot W. Jackson

    Roles Data curation, Formal analysis, Writing – original draft, Writing – review & editing

    Affiliation Department of Microbiology, Cornell University, Ithaca, NY United States of America

Occurrence and seasonal dynamics of RNA viral genotypes in three contrasting temperate lakes

  • Ian Hewson, 
  • Kalia S. I. Bistolas, 
  • Jason B. Button, 
  • Elliot W. Jackson
PLOS
x

Abstract

Decades of research have demonstrated the crucial importance of viruses in freshwater ecosystems. However, few studies have focused on the seasonal dynamics and potential hosts of RNA viruses. We surveyed microbial-sized (i.e. 5–0.2 μm) mixed community plankton transcriptomes for RNA viral genomes and investigated their distribution between microbial and macrobial plankton over a seasonal cycle across three temperate lakes by quantitative reverse transcriptase PCR (qRT-PCR). A total of 30 contigs bearing similarity to RNA viral genomes were recovered from a global assembly of 30 plankton RNA libraries. Of these, only 13 were found in >2 libraries and recruited >100 reads (of 9.13 x 107 total reads), representing several picornaviruses, two tobamoviruses and a reovirus. We quantified the abundance of four picornaviruses and the reovirus monthly from August 2014 to May 2015. Patterns of viral abundance in the >5 μm size fraction and representation in microbial-sized community RNA libraries over time suggest that one picornavirus genotype (TS24835) and the reovirus (TS148892) may infect small (<5 μm) eukaryotic microorganisms, while two other picornaviruses (TS24641 and TS4340) may infect larger (>5 μm) eukaryotic microorganisms or metazoa. Our data also suggest that picornavirus TS152062 may originate from an allochthonous host. All five viral genotypes were present in at least one size fraction across all 3 lakes during the year, suggesting that RNA viruses may easily disperse between adjacent aquatic habitats. Our data therefore demonstrate that RNA viruses are widespread in temperate lacustrine ecosystems, and may provide evidence of viral infection in larger eukaryotes (including metazoa) inhabiting the lakes.

Introduction

Nearly 30 years of research has highlighted the crucial role of viruses in aquatic ecosystems, where they cause significant mortality of microbial and metazoan hosts, facilitate gene exchange, and influence elemental cycling [15]. Viral communities in aquatic ecosystems are enormously diverse, comprising genotypes that infect bacteria, eukaryotic microorganisms, and multicellular organisms [613]. While aquatic viruses may bear either DNA or RNA genomes, most studies to date have focused on identification and characterization of DNA viruses [1420]. Understanding of RNA viral diversity has undergone considerable revision in the last 5 years with the application of shotgun-based sequencing (i.e. metagenomics) approaches to size-fractioned environmental samples [2127] and whole tissues of non-model metazoa [2832]. These studies have revealed a wide array of viruses that expand known taxonomy of RNA viruses [28]. Despite a growing appreciation for the potential significance of RNA viruses to aquatic food web processes and host biodiversity, there remains little information about the taxonomic and genetic composition, biogeography and seasonal recurrence of RNA viruses across many aquatic habitats.

Temperate lacustrine ecosystems represent interesting habitats in which to study the diversity and temporal dynamics of aquatic viruses for several reasons. Lakes receive constant and episodic allochthonous inputs from terrestrial environments (i.e. runoff). Lakes at high latitudes experience distinct seasonal shifts in biological phenomena, including seasonal (primarily spring) blooms, post-bloom clear-water phases, and ice cover in winter. They are also subject to greater variability in natural disturbance events (e.g. ice cover, storms, seiches, etc.) than lakes at lower latitudes. Lakes also experience various mixing regimes, from meromictic to polymictic which influence their productivity and microbial ecology. These distinct mixing regimes and productivity cycles provide ideal conditions to examine dynamics of viral groups by providing stark temporal changes in host abundance.

The composition of viral communities in aquatic ecosystems is a function of production of new virus particles through autochthonous infection and import from allochthonous sources and their decay [3338]. Runoff delivers viruses from surrounding non-aquatic habitats to lake waters. Lakes also receive inputs from airborne animals which facilitate dispersal of other organisms and presumably their associated microbiota and viruses [39]. Decay of aquatic viruses is mostly attributed to labile organic matter which presumably includes nucleases and proteases [40], and sunlight which degrades capsids and genomes [41]. Viral particles may also adsorb to surfaces of particles and sink into deeper waters and sediments [42]. Most studies of marine viral decay converge on a rapid 2–4% h-1 [1, 43], while in freshwater ecosystems study of individual genotypes indicated slower decay rates (0.1–1.5% h-1) [33, 44], and some types of viruses (e.g. nucleopolyhedroviruses) may persist for months-years without significant decay [45, 46]. Thus, the assemblage of viruses present in lake waters may represent viruses infecting autochthonous hosts, those infecting allochthonous hosts that are recently arrived, and viruses that accumulate in the lake and are resistant to decay.

Numerous studies in the last two decades have focused on surveying viral composition in aquatic ecosystems to understand patterns of diversity as they relate to environmental parameters and hosts [1427]. Most studies to date have approached viral diversity via shotgun sequencing of mixed-community viral genomes (i.e. viral metagenomics), where viruses are purified from tissues or directly from environmental samples, their nucleic acids extracted, and then sequenced [47, 48]. Viral metagenomics was first applied to marine plankton [13] and human fecal samples [49]. Since the advent of economical third-generation sequencing, the approach has become common in studies of viral diversity worldwide. RNA viral diversity in freshwater habitats was first determined using metagenomics in 2009 [50], and has since been applied elsewhere [5154]. Early study of community transcriptomics from environmental samples highlighted the unexpectedly large proportion of sequence reads associated with viruses [55]. More recently, shotgun sequencing of invertebrate transcriptomes (‘RNAseq’) has confirmed that transcript-based sequencing efforts are an effective means for retrieving viral genomes [29, 56].

The purpose of this study was to understand RNA viral composition in three temperate lakes with variable hydrology and allochthonous inputs. We first surveyed RNA viral genomes using a community transcriptomics approach targeting microbial-sized (0.2–5 μm) material. We identified RNA viral genomes within transcriptomes, then evaluated their seasonal distribution in the lakes on particles >5 μm (i.e. present in aggregates or particles, large microbial eukaryotes or metazoa). Our results revealed that several RNA viral genotypes are temporally transient and lake-specific, may relate to precipitation and runoff, and may follow seasonal patterns of productivity.

Methods

Description of Finger Lakes

The Finger Lakes (New York State, USA) are a series of 11 freshwater lakes oriented approximately north-south, ranging in length from 4.8 km to 64 km, and in maximum depth from 29 m to 188 m. Cayuga Lake, which is the longest of the Finger Lakes, is mesotrophic [57], having received inputs of fertilizer and runoff from agricultural lands in the region [58]. Seneca Lake, which is the second largest and deepest of the Finger Lakes, is also mesotrophic [59]. Owasco Lake, which is considerably smaller (18km long and 54m deep), has a large catchment (540 km2) relative to its size, and is surrounded by agricultural land [59]. As a consequence, it is more eutrophic than other lakes in the region [59].

Sample collection

Samples were collected monthly between September 2014 and August 2015 from three locations: Owasco Lake (42.754468oN, 76.470780oW; OL); Seneca Lake (42.618192oN, 76.878926oW; SL) and Cayuga Lake (42.537017oN, 76.550545oW; CL) (S1 Table). In February 2015 lake water was inaccessible due to ice. Water was collected from a pier at each location in approximately 1 m water using a sample-rinsed bucket and placed into duplicate sample-rinsed 20 L HDPE cubitainers. Temperature, conductivity and O2 were measured during each sampling expedition using a hand-held YSI probe. Samples were transported in coolers to maintain ambient water temperature to the lab at Cornell University. Water samples were subsampled within 2 h collection for bacterial abundance (~10 mL), chlorophyll a concentration (~ 5 L), and plankton RNA (~20 L). No specific permissions were required for these locations/activities, and this work did not involve endangered or protected species.

Chlorophyll a analyses

Duplicate samples for chlorophyll a analysis (0.1–1 L) were diafiltered through 47 mm diameter GF/F filters, which were subsequently encased in alumninium foil and frozen at -20°C prior to analysis. Samples for chlorophyll a were analysed by the acetone extraction-fluorometry approach [60]. Filters were immersed in 10 mL of 90% acetone in 13x100mm borosilicate glass tubes and incubated overnight in darkness at -20°C. Filters were subsequently removed and the chlorophyll a concentration determined without correction for phaeophytin in a hand-held Turner Designs Fluorometer.

Bacterioplankton abundance

Samples for bacterioplankton abundance (10 mL) were collected and fixed in 2% formalin and stored at 4°C prior to analysis. Preparation of slides for bacterial enumeration occurred 7–14 d after sample collection. Bacterial abundance was determined in fixed samples by DAPI epifluorescence microscopy [61]. Duplicate subsamples (1 mL) of water were stained in darkness with DAPI (20 μL mL [lake water]-1) for 2 min before being filtered through black 0.2 μm polycarbonate filters. The filters were then mounted on glass slides using PBS:Glycerol (1:1) as mountant. Slides were kept frozen prior to microscopy. Bacteria were counted on an Olympus BX-51 epifluorescence microscope under blue light excitation and at 1000X magnification. Ten random fields on each slide containing >200 cells were counted for each slide and abundance calculated from the average bacterial abundance per field.

Preparation of plankton RNA

Duplicate samples for plankton RNA (0.5–2 L, depending on sampling date; S1 Table) were serially filtered through 142 mm diameter 5 μm (Isopore) and 0.2 μm (Durapore) filters using positive air pressure. The filters were immediately frozen at -80°C prior to downstream analyses. Plankton RNA libraries were prepared from the 0.2 μm size fraction filters following the protocols of Hewson et al. [62]. RNA was extracted from the 142 mm filters in ZR RNA Buffer (Zymo Research) and subject to bead beating for 2 min. The homogenate was processed through the ZR RNA Mini Isolation Kit. After extraction, RNA was depleted of contaminating DNA using the DNA-free RNA Kit (Zymo Research), and treated with terminator exonuclease (Epicentre) to increase the ratio of mRNA relative to rRNA. Following treatment, samples were amplified using the TransPlex kit (Sigma Aldrich). Plankton RNA libraries were sequenced at the Cornell Biotechnology Resource Center following library preparation (Nextera). Each plankton RNA library was run on 1/16 of an Illumina MiSeq paired-end 250nt lane.

Bioinformatic analyses of environmental RNA libraries

Sequence libraries were first vetted for poor quality sequence (Ambiguous nucleotides ≤1), trimmed for adapters and length using the CLC Genomics Workbech 4.0. Following this, libraries were assembled using the program’s de novo assembly algorithm with default settings and with contig lengths ≥ 500 nt, minimum overlap 0.2 and minimum similarity 0.8. To identify contigs which may represent RNA virus genomes or genome fragments, all contigs were first subject to tBLASTx against an in-house database of complete RNA genomes collected from NCBI GenBank (keyword search “RNA Virus” with filter “complete genome” as of May 2016). Contigs matching the RNA viral database at an e-value < 10−10 were then subject to BLASTn against the non-redundant (nr) database at NCBI to confirm viral identity. RNA viral contigs (i.e. those matching RNA genomes via tBLASTx) were then subject to fragment recruitment using a minimum overlap of 0.1 and minimum similarity of 0.95 for all environmental RNA libraries. Only contigs recruiting ≥ 100 reads across ≥ 2 libraries were considered further. Contigs fulfilling these criteria were then subject to tBLASTx against the nr database to identify their phylogenetically closest relative. Sequence data is available at NCBI under BioProject accession PRJNA417956.

Determination of viral genotype abundance by qRT-PCR

Five candidate RNA viral phylotypes were selected for focus that were represented in ≥ 5 libraries (Table 1; NCBI accessions MG550036—MG550040). The abundance of RNA viral genotypes was determined by quantitative reverse transcriptase PCR (qRT-PCR). Total RNA was extracted from 5 μm filters using the Zymo RNA Mini Isolation Kit. RNA was then depleted of contaminating DNA by the Zymo DNA-free RNA kit, and subsequently converted to cDNA using Superscript III reverse transcriptase. Primers and probes were designed based on capsid proteins for the 5 genotypes using the program Primer3 (http://www.primer3.com) [63]. qRT-PCR was performed in duplicate reactions where each reaction contained 1X SSO Probes Universal Master Mix (Bio-Rad), 200pmol of each primer (S2 Table) and 5’-FAM labeled probe (3’-TAMRA as quencher), and 2 μl of template cDNA in 25μl reactions. Each run was performed with 4 no template controls and duplicate oligonucleotide standards over 8 orders of magnitude copy number. Quantity was determined by comparison of cycle threshold values against the standards and multiplied by 2 to account for single-stranded standards. The practical detection threshold of all 5 primer/probe pairs was 1 genotype copies reaction-1 based on linearity of standards and absence of amplification in no template controls; this equates to <0.001 genotype copies mL-1 based on the volumes of water filtered.

thumbnail
Table 1. Contigs matching known viruses by BLASTx against the non-redundant (nr) database at NCBI.

The closest cultured virus and uncultured virus relatives are indicated.

https://doi.org/10.1371/journal.pone.0194419.t001

Statistical analyses

All statistical analyses were performed in Microsoft Excel using the XLStat plugin (Addinsoft SARL). Pairwise correlation significance (Pearsons ρ) α values were corrected for Type II error by dividing 0.05 by the number of comparisons made.

Results and discussion

RNA viral communities inhabiting lacustrine ecosystems are a milieu of temporally and spatially variable agents which may arise from both aquatic infections and from infection of hosts in non-aquatic habitats. Our data provide evidence that RNA viruses observed in microbial-sized (0.2–5 μm) plankton RNA libraries may, in fact, represent viruses of larger (>5 μm) microbial eukaryotes, aggregates of eukaryotes, or metazoa. Some RNA viral genotypes had a seasonal trend which may be related to overall patterns of lake productivity (i.e. fall/spring phytoplankton blooms) and precipitation/runoff. Our data provide evidence that several RNA viral genotypes were dispersed between all three lakes, but their presence in different size fractions varied between lakes.

Microbial biomass and physicochemical conditions

All three lakes demonstrated a seasonal pattern of temperature consistent with mean air temperature (Fig 1). Lakes experienced ice cover for approximately 6 weeks from mid-December 2014 through mid-February 2015, followed by thaw in mid-February. The wettest month in the region occurred in mid-June 2015. The temperature at the sampling site in Cayuga Lake was higher in March 2015 than in other lakes, which may have been due to meltwater inputs from nearby creeks. Amongst the three lakes, Owasco had the greatest variation in water temperature from 24.4°C in mid-summer and 0.5°C in February, while Cayuga had the least variation from 23.1°C in mid-summer and 2°C in January. The overall change in temperature reflects polymictic conditions across all three lakes.

thumbnail
Fig 1. Physical and meteorological conditions at Cayuga, Owasco and Seneca Lakes over the sampling period.

Meteorological data were taken from the National Weather Service station at the Ithaca Tompkins Regional Airport (https://www.ncdc.noaa.gov/). Temperature was determined by YSI handheld probe. The shaded region indicates approximate ice covered conditions.

https://doi.org/10.1371/journal.pone.0194419.g001

Reductions in chlorophyll a concentration between August and December 2014 were concurrent with decreased water temperature across all three lakes. Elevated chlorophyll a from March to June 2015 suggested a spring bloom occurred (Fig 2). The timing of maximum phytoplankton concentration differed between lakes. Owasco experienced a spring bloom in April 2015, while both Seneca and Cayuga Lakes experienced blooms in May–June 2015. A pronounced ‘clear water phase’ after bloom exhaustion occurred in Owasco in May 2015. However, there was no clear water phase in Seneca and Cayuga over the sampling period. Bacterial abundance generally increased across all three lakes from Fall 2014 through Summer 2015. Bacterial abundance was greatest in Owasco Lake during the spring bloom in April 2015. Overall bacterial abundance did not demonstrate a clear seasonal pattern related to water temperature, with all three lakes maintaining high cell abundances during ice-covered months.

thumbnail
Fig 2.

Chlorophyll a (top) and bacterial abundance (bottom) in Cayuga, Owasco and Seneca lakes during the sampling period. Chlorophyll a concentration was determined by acetone-extracted fluorometry. Bacterioplankton abundance was determined by DAPI staining and epifluorescence microscopy. The shaded region indicates approximate ice covered conditions.

https://doi.org/10.1371/journal.pone.0194419.g002

Identification of RNA viral genotypes in plankton RNA libraries

An attempt was made to generate duplicate plankton RNA libraries from each lake monthly between September 2014 and May 2015. However, not all attempts were successful, due to poor nucleic acid recovery or sample preparation problems. A total of 30 environmental RNA libraries were successfully prepared, bearing a total of 9.13 x 107 reads (S1 Table). Of these, viruses represented only a small (0.6%) fraction, with the remainder representing transcripts of cellular organisms (S1 Fig).

Global assembly of all 30 plankton RNA libraries generated 191,574 contigs that were ≥ 500 nt in length. Thirty-five contigs bore similarity to RNA viral genomes by tBLASTx against complete RNA genomes. Seventeen of these recruited ≥ 100 reads each in ≥ 2 libraries. Of these, 13 contigs produced significant alignments to eukaryote-associated RNA viruses by BLASTn against the non-redundant (nr) library at NCBI (Table 1; S2 Fig), while the remaining 4 produced significant alignments to bacteria or eukaryotes. The 13 viral contigs detected were predominately picornaviruses (n = 10), however 2 contigs matched most closely the putative tobamovirus associated with the green alga Chara australis [64] and one matched the Orungo virus (Reovirus) associated with mammals. Picornaviruses have a wide host range including microbial eukaryotes, plants, invertebrates and vertebrates. Tobamoviruses typically infect terrestrial plants. The genome of Chara australis virus bears similarity to both tobamoviruses and benyviruses, and infects a freshwater macrophyte [64]. These data demonstrate that viruses of metazoa may be common in freshwater ecosystems.

We chose to focus on 5 viral genotypes (4 Picornaviruses and 1 Reovirus) that matched most closely metazoan RNA viruses at NCBI and were represented in ≥ 5 libraries. These were: 1) Contig TS152062, which was most similar to Wenling Crustacean Virus (APG78478.1, 44% amino acid ID, Picornaviridae) (Fig 3); 2) Contig TS4340, which was most similar to Bee Paralysis virus (YP_003622540.1, 24% amino acid ID, Iflaviridae) (Fig 3); 3) Contig TS24835, which was most similar to Behai picorna-virus, (APG78024.1, 44% amino acid ID, Picornavirales); 4) Contig TS148892, which was most similar to the Orungo Virus (AFX73387.1, 30% amino acid ID, Reoviridae); and 5) TS24641, which was most similar to Beihai Papia Shell Virus (APG78606.1, 45% amino acid ID, Picornavirales) (Fig 4). Phylogenetic analyses of these contigs indicated that all were most similar to metazoan-associated viruses, but were also homologous to viruses of unicellular eukaryotes (Figs 3 and 4). Picornaviruses are frequently encountered in aquatic plankton [6567], and may represent pathogens of eukaryotic microorganisms or metazoans. They typically contain a small (~7–9 kBp) positive sense single stranded RNA genome, a non-enveloped capsid, and one or only a few non-overlapping open reading frames (ORFs, i.e. polyprotein gene organization). Four of the genotypes (TS24835, TS24641, TS152062 and TS4340) selected in this study belong to the order Picornavirales. TS24835 and TS24641 do not contain sufficient genome architecture or phylogenetic information to assign them to a specific family. TS4340 has genome architecture and amino acid sequence consistent with Iflaviridae. TS152062 is phylogenetically most similar to Nora viruses, which are within the Picornaviridae, but remains unclassified to family. Finally, TS148892 is most similar to orbiviruses (Reoviridae), which typically have segmented double stranded RNA genomes and represent arboviruses infecting vertebrates and invertebrates.

thumbnail
Fig 3.

Contig architecture and pylogenetic representations of three viral genotypes observed in this study; Picornavirus TS24835 (top), Reovirus TS148892 (middle) and Picornavirus TS24641 (bottom). Phylogenetic dendrograms were generated based on an amino acid alignment (polyprotein for TS24835 and TS24635, and RNA polymerase VP1 for TS148892) using neighbor joining. Arrows indicate reading direction of ORFs on contigs.

https://doi.org/10.1371/journal.pone.0194419.g003

thumbnail
Fig 4.

Contig architecture and phylogenetic representations of two viral genotypes observed in this study; TS152062 (top) and iflavirus TS4340 (bottom). Phylogenetic dendrograms (polyprotein) were generated based on an amino acid alignment using neighbor joining. Arrows indicate reading direction of ORFs on contigs.

https://doi.org/10.1371/journal.pone.0194419.g004

Temporal quantification of RNA viral genotypes

Read recruitment of all 5 viral genotypes revealed that reovirus TS148892 had the greatest representation in plankton RNA libraries (i.e. 5–0.2 μm) (Fig 5). All five contigs were better represented in plankton libraries during Fall 2014 than Winter or Spring 2015, with the exception of TS152062 in Seneca Lake (Fig 5). The representation of picornaviruses TS24841, TS24835 and iflavirus TS4340 in libraries was lower than the reovirus TS148892 and picornavirus TS152062 throughout the sampling period across all three lakes.

thumbnail
Fig 5. Heat map representation of reads recruiting from plankton RNA libraries against five candidate RNA viral genotypes.

Read recruitment was performed in CLC Genomics Workbench 4.0 with minimum identity 0.95 and minimum overlap of 0.2. Brighter hues indicate a greater % of reads (standardized by column) than darker hues.

https://doi.org/10.1371/journal.pone.0194419.g005

Quantitative PCR targeting the 5 genotypes was applied to the >5 μm size fraction cDNA from the time series samples (Fig 6). The five viral genotypes demonstrated different temporal patterns with season and between lakes. Picornavirus TS24641 had the greatest abundance during winter months in Owasco Lake, but had a very large abundance in Seneca and Cayuga during the spring (Mar-April 2015). Picornavirus TS24835 was absent entirely from Seneca and Cayuga Lakes but was abundant in winter months in Owasco Lake. Reovirus TS148892 was abundant in all 3 lakes throughout the sampling period. Picornavirus TS152062 was present after the onset of winter and into Spring in Owasco and Cayuga Lakes, however was absent from Seneca Lake. Iflavirus TS4340, on the other hand, had large abundances in Seneca Lake in both early Fall and Spring, but the genotype was only detected in late fall in Owasco Lake and was absent entirely from Cayuga Lake.

thumbnail
Fig 6. Abundance of 5 RNA viral genotypes in >5 μm size fraction plankton in Cayuga, Seneca and Owasco Lakes during the sampling period.

Abundance of viral genotypes was determined by quantitative reverse transcriptase PCR (qRT-PCR). The practical detection threshold of all 5 primer/probe pairs was <0.001 genotype copies mL-1; abundances indicated at 0.0001 indicates that the genotype was not detected. The shaded region indicates approximate ice covered condition.

https://doi.org/10.1371/journal.pone.0194419.g006

These data provide evidence that the host of picornavirus TS24641 may be a large microscopic eukaryote or metazoan, since it was more abundant than other viral genotypes in the >5 μm size fraction than its representation in microbial size-fractioned plankton RNA libraries. Given that the virus was present year-round in all three lakes and demonstrated no temporal pattern with season, it is unlikely to infect phytoplankton, which exhibit decreased abundance during winter months. Furthermore, the lower abundance of this virus during spring in Owasco during a large bloom of phytoplankton, suggests it may target a heterotrophic organism (e.g. heterotrophic protozoa or metazoa). TS24641 abundance correlated with water temperature (R2 = 0.37) in Owasco Lake, with bacterial abundance (R2 = 0.81) in Seneca Lake, and weakly with chlorophyll a concentration (R2 = 0.34) in Cayuga Lake (Fig 7). These correlations reflect changes in overall productivity, suggesting that the host of TS24641 may respond to resource availability (e.g. nutrient conditions) and be controlled by overall temperature.

thumbnail
Fig 7. Correlation matrices between viral genotype abundance and environmental parameters in all lakes together; and in Cayuga Lake (red), Owasco Lake (green), and Seneca Lake (blue) separately.

Numbers in bold represent significant correlations (Pearson’s ρ, p< 0.05).

https://doi.org/10.1371/journal.pone.0194419.g007

In contrast, picornavirus TS24835 was only detected in Owasco Lake in late fall and winter, which suggests that it may infect dying phytoplankton, or possibly a heterotrophic eukaryote consuming decaying phytoplankton. The higher relative contribution of TS24835 to microbial-sized plankton RNA libraries year-round, which contrasts with its pattern of abundance in the >5 μm size fraction, confirms its association with small a eukaryotic microorganism.

The seasonal abundance of reovirus TS148892 was greater than other viral genotypes in both the >5 μm size fraction throughout the year and microbial (5–0.2 μm) size fraction in the fall of 2014. Reoviruses infect a wide range of hosts, including fungi, plants, invertebrates and vertebrates. Since TS148892 occurred in both microbial and >5 μm size fractions and was especially enriched in fall samples, we speculate that it may infect phytoplankton or a eukaryote that consumes decaying primary production.

In contrast to reovirus TS148892, iflavirus TS4340 was only detected in two lakes (Owasco and Seneca), only during fall 2014 and spring 2015, and did not contribute substantially to microbial-sized libraries of plankton RNA (Fig 3). Our observation of TS4340 mainly in larger plankton is consistent with a metazoan host. The concordance of iflavirus TS4340 with elevated chlorophyll a in Seneca Lake in fall and spring suggests that it may infect a host influenced by enhanced resource availability. Zooplankton in the lake ecosystem exhibit seasonal abundance shifts according to resource availability, with peak abundances during spring and fall blooms, followed by declines in summer and fall [68, 69]. Iflavirus TS4340 correlated with Picornavirus TS24641 across all three lakes (R2 = 0.67). The correspondence between these viral genotypes may further indicate the presence of a potential pathogen of phytoplankton (TS24641) and a potential pathogen of its grazer (TS4340).

TS152062 was sporadically present in two lakes (Seneca and Owasco) and correlated with chlorophyll a concentration in Owasco lake (R2 = 0.59) but did not correlate with phytoplankton biomass in Seneca Lake. The absence of TS152062 from the >5 μm size fraction in early Fall 2014 and sporadic presence in Spring 2015 suggests that it may infect a rare host that was seldom captured in our survey. The larger contribution of this virus to microbial-size fraction plankton RNA libraries during months that it was absent in the >5 um size fraction by qPCR suggests that it either infects a smaller eukaryotic microorganism that form aggregates during enhanced phytoplankton production, or that it may be present in cell debris during peak phytoplankton abundance. TS152062 also correlated with picornavirus TS24641 in Cayuga Lake (R2 = 0.44), but did not significantly correlate with any genotype in Seneca Lake. A key question in our study was whether there is evidence for allocthonous viruses in lake plankton. Such viruses may be present on soil particles, animal carcasses or cell debris, or fecal matter that is washed into the lake by freshwater inputs or runoff [45, 70]. Of all viruses examined, picornavirus TS152062 had a temporal pattern consistent with enhanced rainfall or runoff. The abundance of picornavirus TS152062 was greatest after both rainfall peaks in September-October 2014 and after ice melt in March-April 2015. Given that picornavirus TS152062 was also abundant in the microbial-size fraction (5–0.2 μm) during rainfall events in early Fall 2014 and was present throughout the year at lower abundances, we hypothesize that TS152062 may infect an allochthonous host in nature. The similarity of TS152062 to a known virus of freshwater mollusk (Biomphalaria sp.) and several terrestrial arthropods (Drosophila sp., Apis sp. and Spodoptera sp.) is consistent with either a terrestrial or shallow-water invertebrate host.

Our data also provide evidence that some viruses were widespread between adjacent lakes, while others were specific to one or two lakes. For example, picornavirus TS24641 and reovirus TS148892 were present in all three lakes in >5 μm plankton, while iflavirus TS4340 and picornavirus TS152062 were present in larger plankton only two lakes, and picornavirus TS24835 detected in only one lake. All five viral genotypes were, however, observed in microbial size fraction (5–0.2 μm) libraries in all three lakes at least for some part of the sampling period. These data suggest viruses of larger unicellular eukaryotes or metazoans may persist in cell debris for some time after host lysis. Alternatively, our data could indicate that small eukaryotic hosts aggregate at certain times during the year and therefore are detectable in the larger size fraction. However, our data indicate that all viruses detected in our study are widely distributed across all three lakes.

Conclusions

RNA viruses are constituents of all aquatic viral communities. However, their predicted hosts, seasonal dynamics and biogeography have not been extensively investigated in temperate lakes [50, 71]. Our data demonstrate that RNA viruses are widespread within regions, but may be temporally transient and associated with particles in different size fractions throughout the year. Most viruses observed in this study were likely associated with eukaryotic microorganisms, and we found evidence that these viruses may originate from metazoa within lakes or from the surrounding catchment. The presence of putative RNA viruses affiliated with eukaryotic microorganisms or metazoa piques interest into the role of viruses in host ecology, and more widely into their impact on freshwater ecosystem function.

Supporting information

S1 Table. Library and assembly characteristics for plankton RNA obtained from the 0.2–5 μm size fraction in three temperate lakes.

https://doi.org/10.1371/journal.pone.0194419.s001

(DOCX)

S2 Table. Primers, hybridization probes and olignucleotide standards used to determine abundance of viral genotypes.

https://doi.org/10.1371/journal.pone.0194419.s002

(DOCX)

S1 Fig. Percentage of viral reads amongst all reads annotated in community RNA libraries.

Phylogenetic annotation occurred via the MG-RAST server ([72] as at May 2016) using an e-value cut-off of 0.001 (as at May 2016).

https://doi.org/10.1371/journal.pone.0194419.s003

(TIF)

S2 Fig. Heat map representation of total reads recruited to each of 13 contigs that matched most strongly RNA viruses at NCBI.

Time Series library dates and locations can be found with reference to S1 Table.

https://doi.org/10.1371/journal.pone.0194419.s004

(TIF)

Acknowledgments

We are grateful to M. Johnson, J. Flanzenbaum and B. Gudenkauf for assistance with field sampling and sample processing.

References

  1. 1. Weinbauer MG. Ecology of prokaryotic viruses. FEMS Microbiology Reviews. 2004;28(2):127–81. pmid:15109783
  2. 2. Wommack KE, Colwell RR. Virioplankton: Viruses in aquatic ecosystems. Microbiology and Molecular Biolology Reviews. 2000;64(1):69–114.
  3. 3. Suttle CA. Viruses in the sea. Nature. 2005;437:356–61. pmid:16163346
  4. 4. Fuhrman JA. Marine viruses and their biogeochemical and ecological effects. Nature. 1999;399(6736):541–8. pmid:10376593
  5. 5. Breitbart M. Marine viruses: Truth or dare. Annual Reviews in Marine Science. 2012;4:425–48.
  6. 6. Chow CET, Suttle CA. Biogeography of viruses in the Sea. Annual Reviews in Virology. 2015;2:41–66.
  7. 7. Comeau AM, Hatfull GF, Krisch HM, Lindell D, Mann NH, Prangishvili D. Exploring the prokaryotic virosphere. Research in Microbiology. 2008;159(5):306–13. pmid:18639443
  8. 8. Nishimura Y, Watai H, Honda T, Mihara T, Omae K, Roux S, et al. Environmental viral genomes shed new light on virus-host interactions in the ocean. Msphere. 2017;2(2).
  9. 9. Sunagawa S, Coelho LP, Chaffron S, Kultima JR, Labadie K, Salazar G, et al. Structure and function of the global ocean microbiome. Science. 2015;348(6237).
  10. 10. Brum JR, Ignacio-Espinoza JC, Roux S, Doulcier G, Acinas SG, Alberti A, et al. Patterns and ecological drivers of ocean viral communities. Science. 2015;348(6237).
  11. 11. Angly FE, Felts B, Breitbart M, Salamon P, Edwards RA, Carlson C, et al. The marine viromes of four oceanic regions. PLoS Biology. 2006;4(11):2121–31.
  12. 12. Breitbart M, Felts B, Kelley S, Mahaffy JM, Nulton J, Salamon P, et al. Diversity and population structure of a near-shore marine-sediment viral community. Proceedings of the Royal Society B-Biological Sciences. 2004;271(1539):565–74.
  13. 13. Breitbart M, Salamon P, Andresen B, Mahaffy JM, Segall AM, Mead D, et al. Genomic analysis of uncultured marine viral communities. Proceedings of the National Academy of Sciences of the United States of America. 2002;99(22):14250–5. pmid:12384570
  14. 14. Rosario K, Schenck RO, Harbeitner RC, Lawler SN, Breitbart M. Novel circular single-stranded DNA viruses identified in marine invertebrates reveal high sequence diversity and consistent predicted intrinsic disorder patterns within putative structural proteins. Frontiers in Microbiology. 2015;6.
  15. 15. Hopkins M, Kailasan S, Cohen A, Roux S, Tucker KP, Shevenell A, et al. Diversity of environmental single-stranded DNA phages revealed by PCR amplification of the partial major capsid protein. ISME Journal. 2014;8(10):2093–103. pmid:24694711
  16. 16. Martinez-Hernandez F, Fornas O, Gomez ML, Bolduc B, de la Cruz Pena MJ, Martinez JM, et al. Single-virus genomics reveals hidden cosmopolitan and abundant viruses. Nature Communications. 2017;8.
  17. 17. Green JC, Rahman F, Saxton MA, Williamson KE. Metagenomic assessment of viral diversity in Lake Matoaka, a temperate, eutrophic freshwater lake in southeastern Virginia, USA. Aquatic Microbial Ecology. 2015;75(2):117–28.
  18. 18. Martinez JM, Swan BK, Wilson WH. Marine viruses, a genetic reservoir revealed by targeted viromics. ISME Journal. 2014;8(5):1079–88. pmid:24304671
  19. 19. Steward GF, Preston CM. Analysis of a viral metagenomic library from 200m depth in Monterey Bay, California constructed by direct shotgun cloning. Virology Journal. 2011;8.
  20. 20. Bench SR, Hanson TE, Williamson KE, Ghosh D, Radosovich M, Wang K, et al. Metagenomic characterization of Chesapeake bay virioplankton. Applied and Environmental Microbiology. 2007;73(23):7629–41. pmid:17921274
  21. 21. Miranda JA, Culley AI, Schvarcz CR, Steward GF. RNA viruses as major contributors to Antarctic virioplankton. Environmental Microbiology. 2016;18(11):3714–27. pmid:26950773
  22. 22. Culley AI, Mueller JA, Belcaid M, Wood-Charlson EM, Poisson G, Steward GF. The characterization of RNA viruses in tropical seawater using targeted PCR and metagenomics. Mbio. 2014;5(3).
  23. 23. Steward GF, Culley AI, Mueller JA, Wood-Charlson EM, Belcaid M, Poisson G. Are we missing half of the viruses in the ocean? ISME Journal. 2013;7(3):672–9. pmid:23151645
  24. 24. Lang AS, Rise ML, Culley AI, Steward GF. RNA viruses in the sea. FEMS Microbiology Reviews. 2009;33(2):295–323. pmid:19243445
  25. 25. Culley AI, Steward GF. New genera of RNA viruses in subtropical seawater, inferred from polymerase gene sequences. Applied and Environmental Microbiology. 2007;73(18):5937–44. pmid:17644642
  26. 26. Culley AI, Lang AS, Suttle CA. The complete genomes of three viruses assembled from shotgun libraries of marine RNA virus communities. Virology Journal. 2007;4.
  27. 27. Culley AI, Lang AS, Suttle CA. Metagenomic analysis of coastal RNA virus communities. Science. 2006;312(5781):1795–8. pmid:16794078
  28. 28. Shi M, Lin XD, Tian JH, Chen LJ, Chen X, Li CX, et al. Redefining the invertebrate RNA virosphere. Nature. 2016;540(7634):539–+.
  29. 29. Holmes EC. The expanding virosphere. Cell Host Microbe. 2016;20(3):279–80. pmid:27631697
  30. 30. Wood-Charlson EM, Weynberg KD, Suttle CA, Roux S, van Oppen MJH. Metagenomic characterization of viral communities in corals: mining biological signal from methodological noise. Environmental Microbiology. 2015;17(10):3440–9. pmid:25708646
  31. 31. Weynberg KD, Wood-Charlson EM, Suttle CA, van Oppen MJH. Generating viral rnetagenomes from the coral holobiont. Frontiers in Microbiology. 2014;5.
  32. 32. Hewson I, Brown JM, Burge CA, Couch CS, LaBarre BA, Mouchka ME, et al. Description of viral assemblages associated with the Gorgonia ventalina holobiont. Coral Reefs. 2012;31(2):487–91.
  33. 33. Hewson I, Barbosa JG, Brown JM, Donelan RP, Eaglesham JB, Eggleston EM, et al. Temporal dynamics and decay of putatively allochthonous and autochthonous viral genotypes in contrasting freshwater lakes. Applied and Environmental Microbiology. 2012;78(18):6583–91. pmid:22773646
  34. 34. Kostner N, Scharnreitner L, Jurgens K, Labrenz M, Herndl GJ, Winter C. High viral abundance as a consequence of low viral decay in the Baltic Sea redoxcline. PLoS One. 2017;12(6).
  35. 35. Corinaldesi C, Dell'Anno A, Magagnini M, Danovaro R. Viral decay and viral production rates in continental-shelf and deep-sea sediments of the Mediterranean Sea. FEMS Microbiology Ecology. 2010;72(2):208–18. pmid:20337705
  36. 36. Parada V, Sintes E, van Aken HM, Weinbauer MG, Herndl GJ. Viral abundance, decay, and diversity in the meso- and bathypelagic waters of the North Atlantic. Applied and Environmental Microbiology. 2007;73(14):4429–38. pmid:17496133
  37. 37. Bongiorni L, Magagnini M, Armeni M, Noble R, Danovaro R. Viral production, decay rates, and life strategies along a trophic gradient in the north Adriatic sea. Applied and Environmental Microbiology. 2005;71(11):6644–50. pmid:16269692
  38. 38. Heldal M, Bratbak G. Production and decay of viruses in aquatic environments. Marine Ecology Progress Series. 1991;72(3):205–12.
  39. 39. Simonis JL, Ellis JC. Bathing birds bias beta-diversity: Frequent dispersal by gulls homogenizes fauna in a rock-pool metacommunity. Ecology. 2014;95(6):1545–55. pmid:25039219
  40. 40. Noble RT, Fuhrman JA. Virus decay and its causes in coastal waters. Applied and Environmental Microbiology. 1997;63(1):77–83. pmid:16535501
  41. 41. Bettarel Y, Bouvier T, Bouvy M. Viral persistence in water as evaluated from a tropical/temperate cross-incubation. Journal of Plankton Research. 2009;31(8):909–16.
  42. 42. Hewson I, Fuhrman JA. Viriobenthos production and virioplankton sorptive scavenging by suspended sediment particles in coastal and pelagic waters. Microbial Ecology. 2003;46:337–47. pmid:14502409
  43. 43. Heldal M, Bratbak G. Production and decay of viruses in aquatic environments. Marine Ecology Progress Series. 1991;72:205–12.
  44. 44. M Long A, Short S. Seasonal determinations of algal virus decay rates reveal overwintering in a temperate freshwater pond. ISME Journal. 2016. 10:1602–12 pmid:26943625
  45. 45. Hewson I, Brown JM, Gitlin SA, Doud DF. Nucleopolyhedrovirus detection and distribution in terrestrial, freshwater, and marine habitats of Appledore Island, Gulf of Maine. Microbial Ecology. 2011;62(1):48–57. pmid:21509607
  46. 46. Holmes SB, Fick WE, Kreutzweiser DP, Ebling PM, England LS, Trevors JT. Persistence of naturally occurring and genetically modified Choristoneura fumiferana nucleopolyhedroviruses in outdoor aquatic microcosms. Pest Management Science. 2008;64(10):1015–23. pmid:18470960
  47. 47. Breitbart M, Salamon P, Andresen B, Mahaffy J, Segall A, Mead D, et al. Genomic analysis of uncultured marine viral communities. Proceedings of the National Academy of Sciences of the United States of America. 2002;99(22):14250–5. pmid:12384570
  48. 48. Thurber RV, Haynes M, Breitbart M, Wegley L, Rohwer F. Laboratory procedures to generate viral metagenomes. Nature Protocols. 2009;4(4):470–83. pmid:19300441
  49. 49. Breitbart M, Hewson I, Felts B, Mahaffy JM, Nulton J, Salamon P, et al. Metagenomic analyses of an uncultured viral community from human feces. Journal of Bacteriology. 2003;185(20):6220–3. pmid:14526037
  50. 50. Djikeng A, Kuzmickas R, Anderson NG, Spiro DJ. Metagenomic analysis of RNA viruses in a fresh water lake. PLoS One. 2009;4:e7264. pmid:19787045
  51. 51. Boujelben I, Yarza P, Almansa C, Villamor J, Maalej S, Anton J, et al. Virioplankton community structure in Tunisian solar salterns. Applied and Environmental Microbiology. 2012;78(20):7429–37. pmid:22904045
  52. 52. Bruder K, Malki K, Cooper A, Sible E, Shapiro JW, Watkins SC, et al. Freshwater metaviromics and bacteriophages: A current assessment of the state of the art in relation to bioinformatic challenges. Evolutionary Bioinformatics. 2016;12:25–33.
  53. 53. Lopez-Bueno A, Rastrojo A, Peiro R, Arenas M, Alcami A. Ecological connectivity shapes quasispecies structure of RNA viruses in an Antarctic lake. Molecular Ecology. 2015;24(19):4812–25. pmid:26198078
  54. 54. Skvortsov T, de Leeuwe C, Quinn JP, McGrath JW, Allen CCR, McElarney Y, et al. Metagenomic characterisation of the viral community of Lough Neagh, the largest freshwater lake in Ireland. PLoS One. 2016;11(2):19.
  55. 55. Gilbert JA, Field D, Huang Y, Edwards R, Li WZ, Gilna P, et al. Detection of large numbers of novel sequences in the metatranscriptomes of complex marine microbial communities. PLoS One. 2008;3(8).
  56. 56. Shi M, Lin X-D, Tian J-H, Chen L-J, Chen X, Li C-X, et al. Redefining the invertebrate RNA virosphere. Nature. 2016;540:539–43.
  57. 57. Oglesby R.T. The limnology of Cayuga Lake. In: B A., editor. Lakes of New York State I, Ecology of the Finger Lakes. New York: Academic Press; 1978. p. 2–121.
  58. 58. Godfrey PJ. The eutrophication of Cayuga Lake—a historical analysis of the phytoplanktons response to phosphate detergents. Freshwater Biology. 1982;12(2):149–66.
  59. 59. Schaffner WR, Oglesby RT. Limnology of eight finger lakes: Hemlock, Canadice, Honeoye, Keuka, Seneca, Owasco, Skaneateles, and Otisco. In: Bloomfield JA, editor. Lakes of New York State. 1. Ecology of the Finger Lakes. Albany, NY: Academic Press; 1978. p. 313–470.
  60. 60. Parsons TR, Maita Y, Lalli CM. A manual of chemical and biological methods for seawater analysis. Oxford: Pergamon Press; 1985. 172 p.
  61. 61. Porter KG, Feig YS. The Use of Dapi for Identifying and Counting Aquatic Microflora. Limnology and Oceanography. 1980;25(5):943–8.
  62. 62. Hewson I, Poretsky R, Beinart R, White AE, Shi T, Bench SR, et al. In situ transcriptomic analysis of the globally important keystone N2-fixing taxon Crocosphaera watsonii. The ISME Journal. 2009: pmid:19225552
  63. 63. Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. In: Krawertz S, Misener S, editors. Bioinformatics Methods and Protocols. Methods in Molecular Biology. Totowa, NJ: Humana Press; 2000. p. 365–86.
  64. 64. Gibbs AJ, Torronen M, Mackenzie AM, Wood JT, Armstrong JS, Kondo H, et al. The enigmatic genome of Chara australis virus. Journal of General Virology. 2011;92(11):2679–90.
  65. 65. Culley A, Lang AS, Suttle CA. High diversity of unknown picorna-like viruses in the sea. Nature. 2003;424:1054–7. pmid:12944967
  66. 66. Culley AI, Lang AS, Suttle CA. Metagenomic analysis of coastal RNA virus communities. Science. 2006;312:1795–8. pmid:16794078
  67. 67. Culley AI, Lang AS, Suttle CA. The complete genomes of three viruses assembled from shotgun libraries of marine RNA virus communities. Virology Journal. 2007;4:-.
  68. 68. Schindler DW, Noven B. Vertical distribution and seasonal abundance of zooplankton in 2 shallow lakes of experimental lakes area, northwestern Ontario. Journal of Fisheries Researd Board Canada. 1971;28(2):245–+.
  69. 69. Watson NHF. Seasonal distribution and abundance of crustacean zooplankton in Lake Erie. 1970. Journal of Fisheries Research Board Canada. 1976;33(3):612–21.
  70. 70. Hewson I, Barbosa JG, Brown JM, Donelan RP, Eaglesham JB, Eggleston EM, et al. Temporal dynamics and decay of putatively allochthonous and autochthonous viral genotypes in contrasting freshwater lakes. Applied and Environmental Microbiology. 2012; pmid:22773646
  71. 71. Mohiuddin M, Schellhorn H. Spatial and temporal dynamics of virus occurrence in two freshwater lakes captured through metagenomic analysis. Frontiers in Microbiology. 2015;6(960).
  72. 72. Meyer F, Paarmann D, D'Souza M, Olson R, Glass EM, Kubal M, Paczian T, Rodriguez A, Stevens R, Wilke A, Wilkening J, Edwards RA. The metagenomics RAST server—a public resource for the automatic phylogenetic and functional analysis of metagenomes. BMC Bioinformatics 2008; 9 (Artn 386)