Tracking putative <i>Microcystis</i> viruses and virus-host associations across distinct phases of a <i>Microcystis</i> -dominated bloom

Viruses significantly impact microbial community composition and function. Yet their role in the fate of freshwater cyanobacterial harmful algal blooms (cHABs), an increasing threat to freshwater systems, remains poorly understood. Here, we address this with a metagenomic analysis of viruses of bloom-forming <i>Microcystis aeruginosa</i> through a seasonal cHAB in the western basin of Lake Erie. We identified globally distributed <i>Microcystis</i> viruses in Lake Erie based on sequence homology to well-studied isolates. A machine-learning model was then used to predict associations between uncharacterized viral populations and the <i>Microcystis</i> and non-<i>Microcystis</i> hosts of the cHAB. Size fractionation of water samples allowed us to identify significant fraction-specific trends in <i>Microcystis</i> viral diversity that corresponded with <i>Microcystis</i> genetic diversity. Viral diversity was highest in the non-colony-associated fraction and lowest in the colony-associated fraction, suggesting that colony formation may lead to bottlenecks in viral diversity in cHABs. Significant turnover of predicted <i>Microcystis</i> virus populations was observed through time, but not between stations miles apart. The virus-host networks revealed extensive interconnectivity and the potential for virus-mediated cross-species genetic exchange. The networks predicted that Lake Erie <i>Microcystis</i> viruses infect hosts spanning phyla, in agreement with lab studies in other systems but challenging previous notions of "narrow" host-virus associations in this genus. Abundant <i>Microcystis</i> virus genes revealed a potential role in key metabolic pathways and host adaptation. These findings advance our understanding of <i>Microcystis</i> viruses and their potential influence on host metabolism, species interactions, and coevolution in <i>Microcystis</i>-dominated cHABs.IMPORTANCEUnderstanding associations between viruses, their hosts, and environmental factors is key for identifying the mechanisms behind the rise and fall of cyanobacterial harmful algal blooms. This study explores the diversity and host ranges of viruses predicted to infect <i>Microcystis</i>, reporting how these properties vary over time, across sample stations in western Lake Erie, and among different filter size fractions. We found that <i>Microcystis</i> virus diversity is highest in non-colony-associated fractions and the lowest in colony-associated fractions, suggesting a link between <i>Microcystis</i> colony formation and reduced viral diversity. We identify abundant genes belonging to predicted <i>Microcystis</i> viruses and their potential roles in key metabolic pathways and adaptation to environmental changes. These findings enhance our understanding of the interplay among viruses, <i>Microcystis</i>, and co-occurring bacteria in cHABs, offering insights into the mechanisms driving bloom dynamics, species interactions, and coevolutionary processes.