News Releases

Dyeing to Learn More About Marine Viruses

Tagging strategy allows for population surveys

The sheer volume of cyanobacteria in the oceans makes them major players in the global carbon cycle and responsible for as much as a third of the carbon fixed. These photosynthetic microbes, which include Prochlorococcus and Synechococcus, are tiny – as many as 100 million cells can be found in a single liter of water – and yet they are not the most abundant entities on Earth. That distinction goes to viruses, up to 100 million of which can be found per 1 mL of seawater. However, researchers know very little about the viruses in the water, other than that there are three kinds of viruses, and that they are capable of drastically decreasing cyanobacterial populations, affecting the global regulation of biogeochemical cycles.

viral tagging schematic by C. Schirmer

This schematic highlights the steps of the viral tagging method – a high-throughput method to investigate viruses that infect a specific microbial host to understand viral-host interactions and natural host-specific viral diversity and evolution. Briefly, natural viral communities are fluorescently labeled with nucleic acid stain and incubated with a cultivated microbial host that was treated with 15N-ammonium chloride to make the DNA isotopically heavy. The sample is then sorted with a flow-cytometer to obtain the viral-tagged cells based on differential fluorescence. DNA is extracted from the tagged population, and viral DNA is separated from microbial DNA on a cesium chloride density gradient. Metagenomic sequencing of the viral DNA then enables genomic examination of viruses that infect the particular host.
(Christine Schirmer, University of Arizona)

To help resolve this conspicuous lack of knowledge and learn more about viral diversity, a team led by Matt Sullivan, a professor at the University of Arizona and a collaborator with the U.S. Department of Energy Joint Genome Institute (DOE JGI), a DOE Office of Science User Facility, conducted a population-scale survey using a game-changing new technique. Their results, he said, suggest that there is an ecology of viruses and it can be studied by harnessing more traditional approaches that have been applied to larger organisms. The work was published online July 13, 2014 in Nature.

“I often joke that viruses are only as interesting as their microbial hosts,” Sullivan said, “which makes cyanophages pretty important. Not only do they affect marine photosynthesis through mortality of cyanobacteria, but these viruses also encode photosynthesis genes – decade-old finding made in collaboration with JGI – that means cyanophages help drive global biogeochemical cycles that are crucial for running all kinds of energy conversions on the planet. The challenge for us, if we wanted to develop predictive capacity, was to develop a method that allowed us to simultaneously examine thousands (or more) of wild cyanobacterial viruses from the millions of non-cyanobacterial viruses in seawater – this would get us beyond learning about them one at a time.”

Sullivan’s team focused on the cyanophages isolated from a single sample of water collected in Monterey Bay, Calif. To resolve the challenge of figuring out “who infects whom” among marine viruses and cyanobacteria, they used a technique known as viral tagging, in which viruses and so-called “host bait” are stained with a fluorescent dye in order to find out with which hosts the phages associate. Sullivan credits the original idea for the project to study co-author Phil Hugenholtz, formerly a DOE JGI researcher and now Director of the Australian Centre for Ecogenomics at the University of Queensland.

“We subsequently found out that the basic idea of “labeling” bacteria with fluorescent viruses wasn’t novel, it had been proposed back in 1985, but the novelty of viral tagging lies in combining this with flow sorting and sequencing,” said Hugenholtz. “Viral tagging combined with flow sorting and sequencing provides an unbiased view of host-phage range.”

TEM of phage isolate collected off Monterey Bay, CA

Transmission Electron Microscopy image of the coastal phage isolate S-MbCM25 collected on the R/V Western Flyer research cruise off Monterey Bay, CA. (Li Deng, University of Arizona)

After screening for cyanophages that were tagged as being associated with a single Synechococcus strain (SynWH803), samples of the viral community metagenomes were sent to the DOE JGI for sequencing as part of a Community Science Program project proposed by Sullivan. The results indicated that there were “at least” 26 viral populations associated with the cyanobacterial strain, and many of them had been found and identified in culture. Additionally, the researchers wrote, “viral tagging also provided evidence… for 42 new uncultured viruses specific to SynWH7803…. [A]n unprecedented diversity of specific viruses were recovered for this single host despite two decades of isolation studies.”

The study also benefited from collaboration with Joshua Weitz, an Associate Professor and theoretical ecologist from the Georgia Institute of Technology, who was on sabbatical in the Sullivan Lab at UA. “This new method provides incredibly novel sequence data on viruses linked to a particular host,” Weitz explained. “The work is foundational for developing a means to count genome-based populations that serve as starting material for more rigorous predictive models of how viruses interact with their host microbes. Instead of counting ‘dots’ we can now map viral populations with their genomes, providing information about who they are and what they do.”

The team made several key findings, perhaps chief among them was that the fluorescently-tagged cyanophages sequenced existed in discrete populations when plotted in ‘genome sequence space’ (an abstract method the researchers used to visualize the relatedness of many viruses at once). This means that the conventional knowledge of viral genomes evolving by “rampant mosaicism” – i.e., recombining segments or modules from different cyanophages – might be wrong. Instead, these cyanophage genome “clusters” suggested that there was discrete population structure in the wild.

“The novel finding here isn’t the number of viruses, but rather the structured nature of the populations,” Sullivan said. “With these discrete populations in a complex natural community and the genome sequence information linked to each population, we are generating hypotheses on what might be driving particular population-host interactions and the abundances of particular populations – that’s viral ecology. And you can track how one population changes over time at a genetic level – that’s viral evolution,” he said. “The thinking before was that the viral genome sequence space would be one big blur, but this suggests there are units that we can count and study. That represents a whole new ballgame and opens up viral ecology to utilize decades of theory and practice from the study of more traditional study of larger organisms. Additionally, our method of viral tagging should be generalizable to many other virus-host studies so it should transform the way viruses in nature are studied moving forward.”

Matt Sullivan spoke about marine viruses at the 2012 DOE JGI Genomics of Energy & Environment Meeting. Watch the video at The Sullivan lab maintains publicly available protocols and informatics tools at

The U.S. Department of Energy Joint Genome Institute, User Facility of Lawrence Berkeley National Laboratory supported by the DOE Office of Science, is committed to advancing genomics in support of DOE missions related to clean energy generation and environmental characterization and cleanup. DOE JGI, headquartered in Walnut Creek, Calif., provides integrated high-throughput sequencing and computational analysis that enable systems-based scientific approaches to these challenges. Follow @doe_jgi on Twitter.

DOE’s Office of Science is the largest supporter of basic research in the physical sciences in the United States, and is working to address some of the most pressing challenges of our time. For more information, please visit