University of British Columbia. Oceanography

Since the 'discovery' of the high abundance of viruses in aquatic environments, it has been generally assumed that viruses in lakes are similar to those in oceans. I directly compared these two systems using a large, robust data set. Viral abundance was significantly different among the surveyed environments. The relationship between viral and bacterial abundance indicated a fundamental difference between lakes and oceans, and suggested that viruses infecting phytoplankton may be more important in lakes. Molecular techniques (PCR & DGGE) were used to document spatial and temporal variations in the richness of viruses that infect eukaryotic phytoplankton (Phycodnaviridae) in lakes at the Experimental Lakes Area (ELA). Phycodnavirus richness was highest in the eutrophic lake, and during the spring/early summer in all the lakes. Viral richness was closely associated with phytoplankton abundance and composition. As a result, richness was influenced by trophic status, while patterns of richness were affected by regional climatic conditions. Phylogenetic analysis of environmental Phycodnavirus DNA polymerase (pol) sequences indicated that freshwater Phycodnaviruses are genetically different from cultured isolates and marine environmental sequences. A genetic distance analysis indicated that pol sequences > 7 % different infected different host species. Therefore, the 20 different freshwater sequences likely infected nine different hosts. Multivariate statistics identified seven possible phytoplankton hosts, including chlorophytes, chrysophytes, diatoms and dinoflagellates. Finally, the modified dilution experiment was evaluated as an approach for estimating viral-mediated phytoplankton mortality in two lakes at the ELA. Experiments resulted in non-significant apparent growth rate regressions. While a model analysis, indicated that the method was sensitive to poorly constrained parameters such as burst size and length of the lytic cycle, making it unsuitable for estimating mortality rates in these lakes. These studies indicate that Phycodnaviridae are a genetically rich and dynamic component of lakes. Their richness is influenced by both the chemical and physical components of their environment. Although the presence of these viruses indicates that they are a source of phytoplankton mortality, the magnitude of their impact on structuring phytoplankton communities awaits methodological advances. Nonetheless, these findings support the view that viruses infecting phytoplankton are ecologically important components of lake ecosystems.