Viral community assembly during cyanobacterial blooms reveals ecosystem-specific regulation and nutrient cycling in freshwater lakes.

Cyanobacterial blooms pose escalating threats to freshwater ecosystems and public health, yet the ecological roles of viral communities in bloom development, succession and toxin dynamics remain poorly understood. We characterized bacterial and viral community dynamics during bloom seasons in two chronically affected lakes with contrasting ecological properties: Lake Erie, a large Great Lake, and Buckeye Lake, a shallow inland reservoir. Buckeye Lake exhibited substantially higher bloom severity, with fivefold higher microcystin concentrations (mean 24.40 vs. 5.30 μg L-¹), all samples exceeding drinking water advisory thresholds. Cyanobacterial composition differed markedly, with Microcystis dominating Lake Erie, whereas Buckeye Lake was co-dominated by Planktothrix and Microcystis. Buckeye Lake exhibited dramatic succession from > 90% cyanobacterial abundance (June-July) to increased abundances of Proteobacteria, Planctomycetes, and Verrucomicrobia (August-September), reflecting post-bloom transitions. Viral communities clustered by season rather than lake type, but exhibited lake-specific host associations and functional potential. Myoviridae abundance was strongly correlated with toxin-producing cyanobacteria in Lake Erie (ρ = 0.786-0.808), indicating selective host coupling, whereas viral-host associations in Buckeye Lake were broader and linked to post-bloom heterotrophic bacteria. Auxiliary metabolic genes related to amino acid metabolism and peptidoglycan degradation were abundant in Buckeye Lake, suggesting virus-mediated nutrient recycling during bloom decline. These findings indicate that viral communities play ecosystem-specific roles in shaping microbial succession, nutrient cycling, and bloom dynamics. Importantly, smaller inland lakes may be more vulnerable to severe bloom impacts than large lakes and therefore require direct toxin monitoring and taxa-specific nutrient management strategies. Our findings provide critical insights into virus-host interactions in bloom ecosystems and highlight the importance of incorporating viral ecology into bloom monitoring and management strategies.