A single gram of soil contains between 10 million and 1 billion viruses. Most of those viruses do not infect plants, animals or people - but they do target bacteria and other microbes.
Because of their influence on microbial communities, viruses can affect nutrient cycling and soil health. Understanding how they behave is therefore crucial for supporting agriculture, food production and water quality.
A new study published in mSystems by researchers at Lawrence Livermore National Laboratory (LLNL) shows that of the many viruses present in soil, only some are active participants at any given time. Over three quarters of the viruses in the study were dormant lurkers, persisting in the soil and biding their time for suitable conditions to emerge.
"Traditionally, viruses were often viewed as particles that either quickly found a host or were rapidly degraded in the environment," said LLNL scientist and author Gary Trubl. "Instead, our results suggest that many soil viruses form a viral 'seed bank' that can remain in the environment and potentially reactivate when conditions become favorable."
To reveal these dormant viruses, the team used stable isotope probing, or SIP. The technique has historically been labor-intensive and expensive, but LLNL's semi-automated, high-throughput pipeline substantially reduces cost and increases output.
All microbes need the oxygen from water to synthesize DNA, and when viruses infect a host, they rewire this process to synthesize their own viral DNA. The SIP approach works by adding water with a natural but rare and heavy form of oxygen to, in this case, a seasonally dry grassland soil. Any virus that replicates during the experiment will contain DNA essentially tagged with that heavy oxygen.
For the first time, SIP was paired with a viral-targeted metagenome protocol optimized for soils. That process separated the active viruses from the soil, extracted their DNA for sequencing and characterized them computationally.
"In Mediterranean-like grasslands, soils can remain dry for months and then experience rapid rewetting after the first rains," said Trubl. "Previous work from our team showed that viruses become active after wet-up and contribute to microbial mortality, but it was unclear how much of the viral community was newly produced versus persisting from the dry season."
During the week-long experiment, only 22 percent of viruses showed the oxygen tags that indicate replication. The rest - 78 percent - remained inactive.
This work moves beyond a survey of which viruses are present in a sample. Instead, it provides an understanding of what viruses are actually doing in nature. The result will allow researchers to better incorporate viruses into models of ecosystems and nutrient cycling.
Going forward, the team is exploring how long the dormant viruses can remain stable in soil, as well as which viral or host traits promote long-term survival. They also aim to expand SIP analysis beyond grassland soil to study agricultural systems, forests, aquatic environments and the human microbiome.
This research was supported by an LLNL Laboratory Directed Research & Development grant and by the U.S. Department of Energy Office of Biological and Environmental Research Genomic Science Program.
