Abstract:Soil microbes are a critical component for maintaining stability and function of terrestrial ecosystems. Viruses, the most abundant biological entities on the earth, can regulate the dynamics of microbial communities in several types of habitats. Therefore, understanding the interactions between viruses and microorganisms is of great significance for further exploring the ecosystem processes including carbon cycle. Here we report a microcosm experiment that addresses the possible link between top-down regulation of bacteria by viruses and soil heterotrophic respiration. Soil microcosms were set up in the laboratory and incubated under three water-content treatments (constant low-water, constant high-water and fluctuating water content). The changes of viral and bacterial abundances responded to above-mentioned soil water content, and their correlation with soil heterotrophic respiration were investigated. Viruses and bacteria were extracted from soil using extraction buffer and enumerated with epifluorescence microscopy direct counting method, and the soil respiration rate was measured by CO2 concentration analyzer. The results showed that bacterial abundances were higher than viral abundances under the condition of three water treatments, regardless of sampling time. The viral abundances (P<0.001) and virus-to-bacteria ratio (P=0.0026) were significantly increased in the high-water treatment, while only viral abundances (P<0.001) were significantly increased in the fluctuating water treatment compared with the low-water treatment. Signal of oscillation in viral and bacterial abundances was observed in the treatment of high-water content. The bacterial and viral abundances showed a trend of increase-decrease-increase and increase-decrease over time, respectively, but the change of viruses lagged behind bacteria. These are consistent with an expectation that water content increase is akin to resource enrichment to our soil bacteria-viruses system. Soil heterotrophic respiration rate was positively correlated with soil water content (P<0.001), bacterial abundance (P=0.0045) and viral abundance (P<0.001). These results indicated that top-down control by viruses could be a crucial force to regulate soil bacterial abundance particularly under water-replete conditions. Our results imply an alternative explanation for the positive effect of water content on soil respiration:predation of bacteria by viruses that can accelerate the recruitment of the bacterial community was exacerbated by high water treatment. Therefore, the interactions between viruses and bacteria may be an important determinant of the soil carbon cycle. Overall, the present work contributes to expanding the current knowledge about the relationship between soil viruses and bacteria, and provides some empirical evidence supporting related researches on the effect of viruses on microbial community structure in soil.