AbstractOstreococcus tauri is a ubiquitous marine pico-eukaryote that is susceptible to lysis upon infection by its species specific Ostreococcus tauri viruses (OtVs). In natural populations of O. tauri, costs of resistance are usually invoked to explain the persistence or reappearance of susceptible individuals in resistant populations. Given the low costs of resistance measured in laboratory experiments with the O. tauri/OtV system to date, the question remains of why susceptible individuals persist in the wild at all. Epidemiological models of host and pathogen population dynamics are one useful approach to understand the conditions that can allow the coexistence of susceptible and resistant hosts. We used a SIR (Susceptible-Infected-Resistant) model to investigate epidemiological dynamics under different laboratory culturing regimes that are commonly used in the O.tauri/OtV system. When taking into account serial transfer (i.e. batchcycle lengths) and dilution rates as well as different resistance costs, our model predicts that no susceptible cells should be detected under any of the simulated conditions – this is consistent with laboratory findings. We thus considered an alternative model that is not used in laboratory experiments, but which incorporates one key process in natural populations: host populations are periodically re-seeded with new infective viruses. In this model, susceptible individuals re-occurred in the population, despite low costs of resistance. This suggests that periodic attack by new viruses, rather than (or in addition to) costs of resistance, may explain the high proportion of susceptible hosts in natural populations, and underlie the discrepancy between laboratory studies and observations of fresh isolates.ImportanceIn natural samples of Ostreococcus sp. and its associated viruses, susceptible hosts are common. However, in laboratory experiments, fully resistant host populations readily and irreversibly evolve. Laboratory experiments are powerful methods for studying process because they offer a stripped-down simplification of a complex system, but this simplification may be an oversimplification for some questions. For example, laboratory and field systems of marine microbes and their viruses differ in population sizes and dynamics, mixing or migration rates, and species diversity, all of which can dramatically alter process outcomes. We demonstrate the utility of using epidemiological models to explore experimental design and to understand mechanisms underlying host-virus population dynamics. We highlight that such models can be used to form strong, testable hypotheses about which key elements of natural systems need to be included in laboratory systems to make them simplified, rather than oversimplified, versions of the processes we use them to study.
Boom-bust population dynamics can increase diversity in evolving competitive communities
