Abstract:

Aims: The purpose of this study is to systematically examine soil viral community composition and assess their carbon metabolic functional potential across urban wetland ecosystems.  

Methods: We took five habitat types of sample plots, including trees, shrubs and grasslands, shrub grasslands, reed marshes, shoals and ponds, distributed widely in Hangzhou Xixi National Wetland Park, Zhejiang Province (Xixi Wetland) as the research objects. This study investigated soil viral diversity patterns and carbon metabolic gene composition across these five habitat types using virus metagenomic sequencing coupled with bioinformatics analysis.

Results: There were significant differences in viral richness and diversity among the five habitat types (P < 0.05), with hierarchical rankings as follows: shrub grasslands > trees, shrubs and grasslands > reed marshes > shoals > ponds; The structural equation modeling analysis showed that this spatial pattern was predominantly mainly driven by soil physicochemical properties, with the order of effect size was pH > soil moisture > SOC ≈ TN > soil temperature. The contribution of comprehensive effect size of viral host and viral diversity was in the order of TN > SOC > soil temperature, and the direct effect of viral host on viral diversity was the strongest (effect size = 0.87); A total of 158 unique carbohydrate transport and metabolism (G) genes were identified, including 13 distinct carbohydrates active enzyme (CAZyme) families. Among these glycosyltransferases accounted for the highest proportion (65.8%), indicating that soil viruses may play an important role in wetland carbon cycling through glycosylation-mediated processes.

Key words: wetlands, virus metagenomic sequencing, soil organic carbon sequestration, CAZymes genes, ecological function