Cyanophage Dynamics at the San Pedro Ocean Time Series: Generalists, Specialists, and One-shot-wonders
Cyanophage Dynamics at the San Pedro Ocean Time Series: Generalists, Specialists, and One-shot-wonders
Abstract:
Viruses exert important controls on the abundance, productivity and strain-level diversity of microbial communities. Long term time series show that natural phage communities exhibit predictable seasonal patterns while models, such as Kill-the-Winner, suggests continual overturn of virus and host strains. Important, basic knowledge is still lacking about the dynamics and population structure of what phage strains are most successful in controlling host populations. Here we analyze multi-year stain level dynamics of the cyanobacteria genera Prochlorococcus and Synechococcus and the T4 cyanophage that infect them at the San Pedro Ocean Time Series. By utilizing 5 years worth of cellular and viral metagenomes along with host marker gene data, we demonstrate that different phage strains within the cyanophage community exhibit different long term dynamics. The cyanophage community consists broadly of three groups. 1) Generalist phage remain at low, constant relative abundance throughout the time series and exhibit no discernible seasonal pattern or significant correlation with host ecotypes as is consistent with them having a broad host range. 2) Seasonal specialist phage occur repeatedly at high relative abundance and are strongly correlated with host ecotypes that exhibit similar seasonal patterns. 3) “One-shot-wonder” phage occur sporadically at high relative abundance and correlate with hosts that also occur briefly within the time series. These data suggest that cyanobacterial lysis and productivity are impacted mostly by a combination of both seasonally re-occurring and sporadically successful phage. These results indicate that the complex co-evolutionary arms race of viruses and hosts is influenced by both recurrent and stochastic processes. This in turn may have important implications for informing models of virus-host dynamics.