Collaboration with Nelson Hairston, Cornell E&EB

Over two decades of research has highlighted the crucial role of viruses in aquatic ecosystem function. They can be significant agents of mortality for microbial hosts, influence elemental cycles, impact the composition of host communities, and mediate gene flow through microbial communities.  Yet, we know remarkably little about the role of metazoan viruses in pelagic lake ecosystems, and whereas there are a number of different ways in which viruses could be ecologically important for the dynamics of higher trophic levels, among the most likely is as pathogens of grazing zooplankton.  Because viruses are essentially invisible to the traditional tools used to study lake plankton, it is quite possible that they are significant regulators of grazer abundance, and hence ecosystem processes, without generations of limnologists having found out – the tools simply were not readily available – until now through genomics and high throughput sequencing. This project is an exploratory investigation of whether viruses are present and prevalent in a population of the major grazing taxon, Daphnia, in a very well-studied system, Oneida Lake, New York.  We wish to find out if viruses are present in Oneida Lake Daphnia, and if they are address the questions:  How does prevalence change seasonally?  How does prevalence correspond to changes in the abundance, birth rate and death rate of the hosts?  This study combines ecological investigations of the life cycle of Daphnia with metagenomics and post-genomic quantification of viral genotypes using quantitative PCR.

Year 1 activities

During this first year of this early concepts project, we focused on viral discovery in Daphnia populations of Oneida Lake using viral metagenomics. Daphnia were collected by net tows from surface waters of Oneida Lake on two dates in midsummer 2010 and transported immediately to the lab at Cornell. Daphnia were initially picked from tow material into virus-free (30 kDa ultrafiltered) Oneida Lake water, then re-picked into sterile bottles containing virus-free water. The bottles were amended with ~ 100 mg of autoclaved bentonite to encourage gut passage, and incubated for 6 hours in a rotator. After this time, the Daphnia were filtered over a bleach-treated Nitex screen, concentrated, placed into sterile falcon tubes and flash-frozen on liquid N2.
Metaviromic libraries were prepared from Daphnia tissues following the protocols of {Thurber, 2009 #2400}. Daphnia were homogenized with a mortar and pestle in virus-free phosphate buffered saline. The homogenate was then filtered through a 0.2 µm Acrodisc filter to remove large particles, then subject to overnight polyethylglycol precipitation. The viral pellet was then resuspended in 3 mL PBS and subject to a CsCl2  step gradient centrifugation. Viruses at the interface between the 1.3 and 1.5 g ml-1 density layers were removed, treated with DNAse and RNAse to digest extraneous DNAs and RNAs (leaving viral DNA and RNA intact), and extracted using the DNEasy Tissue kit (Qiagen) for DNA viruses (half the sample), and RNEasy kit for RNA viruses (half the sample). Following extraction, the viral DNA was amplified using Φ29 strand displacement amplification (Genomiphi, GE Healthcare). The RNA was amplified using the Transplex II kit (Sigma). The amplified viral DNA and RNA was cleaned and pyrosequenced at Engencore (University of South Carolina), where each library comprised 1/8th of a picotiter plate.
We obtained a total of 132,968 sequence reads from June and August samples (Table 1). Reads were subject to de novo assembly using the CLC Genomics Workbench 4.0 using stringent parameters (0.2 of sequence read, 95% identity). This resulted in 6,263 contigs. Open reading frames on contigs were identified using the getorf algorith, and ORFs were blasted against the non-redundant database (nr) at NCBI. We identified 38 contigs which shared significant (e < 0.001) homology with viral proteins and focused on these for further analyses.

Table 1: Viral Metagenomic library characteristics

Month	Nucl. Acid Type	Total Reads	Total Contigs	Total Contig Matches to nr database at e < 0.001	Putative Viral Contigs
June	DNA	35,900	3,007	690	10
	RNA	32,372	374	359	7
August	DNA	34,490	2,299	1,024	19
	RNA	30,206	583	250	2

The majority of viral hits were to bacteriophage in DNA and RNA libraries as both reads and contigs (Fig. 1). The largest proportion of DNA read matches to nr across both sampling dates were to Circoviruses (a larger % of viral matches to retroviruses in June likely represents spurious matches since retroviruses are ssRNA viruses, and our viromes were targeting ss and ds DNA). Contigs not matching viruses had strongest matches to bacterial proteins, but only a small number (14) of contigs matched Daphnia proteins. Bacteria and associated bacteriophage were likely contaminants that were not removed from Daphnia tissues prior to sequencing by our washing technique. Contigs matching eukaryotic viral proteins (15 total) were all in the DNA viral libraries, with the exception of a contig containing a Geminivirus RdRp gene from the June RNA viral library. Since geminiviruses are ssDNA viruses, and the match was weak (e = 0.0001) we believe this to be a spurious hit.

Fig 2: Percentages of total nr matches to viral genomes of sequence reads in DNA libraries prepared from Daphnia tissues.

Amongst DNA contig viral matches, 6 were to circoviruses. This group of relatively unstudied viruses causes diseases in birds and pigs, however they have recently been observed as major contituents of antarctic lake plankton {Lopez-Bueno, 2009 #2981}, have been detected in the Chesapeake Bay  and Sargasso Sea {Rosario, 2009 #2986}. As part of a concurrent project examining the presence of viruses in marine copepods (NSF OCE 1049670, EAGER: Discovery of viruses infecting marine copepoda, M. Breitbart, K. Daly and I. Hewson, co-PIs), we have detected circoviruses actively replicating at high prevalence and viral load in two species of marine copepod, Labidocera aestiva and Acartia tonsa (unpub data). Hence, focused our efforts on circoviruses detected in Daphnia populations in Oneida Lake.

Year 2 Activities

Candidate Daphnia circoviruses were the targets of Quantitative PCR (qPCR) primer and probe design for application in summer of 2011. qPCR primers and probes were developed around the two contigs using Primer3, and validated against environmental extracts of bacterioplankton (negative controls) and against Daphnia collected from Oneida Lake to ensure positive amplification.

In summer 2011, we engaged a REU student, Gabriel Ng, to conduct routine sampling of Daphnia for comparison of life history traits to viral parameters. Samples of Daphnia were collected weekly, and comprised 50 picked individual cladocerans, that were frozen at the Cornell Biological Field Station, Shackleton Point, Oneida Lake.  The samples were transported to the lab at Cornell for processing by undergraduate researcher, WenFang Li, and high school student, James Eaglesham. DNA was extracted from whole animals using the Zymo Insect & Tissue DNA kit, and assayed by qPCR for two circoviruses across 20 – 50 individuals from each collection time. qPCR reactions were condudcted in duplicate reactions per individual animal. The prevalence of circoviruses was calculated as the total number of individuals in which > 1 copy of virus was detected per animal in both duplicates. Viral load was calculated as the average number of viral genotype copies per individual.

Both viral load and prevalence of Daphnia circoviruses increased during the summer before experiencing decreases in late summer (Fig. 3). The two circoviruses demonstrated opposing dynamics, with higher circovirus 2 prevalence corresponding to an early season Daphnia population crash, and higher circovirus 1 prevalence corresponding to higher mortality rates in late summer.

Fig 3: Dynamics of circovirus genotypes over time in relation to Daphnia abundance and birth/death rates.

Work is currently underway to assess the environmental reservoirs of Daphnia circoviruses in the lake environment, and we hope to conduct TEM analysis of Daphnia tissues to locate sites of virus particle formation.

Work published on this project:

Hewson I, Ng G, Li W, LaBarre BA, Aguirre I, Barbosa JG, Breitbart M, Greco AW, Kearns CM, Looi A, Schaffner LR, Thompson PD, Hairston NG (2013) “Metagenomic identification, seasonal dynamics and potential transmission mechanisms of a Daphnia-associated putative RNA-DNA hybrid virus in two temperate lakes” Limnology and Oceanography 58: 1605-1620