Soil conditions rather than long-term exposure to elevated CO2 affect soil microbial communities associated with N-cycling

Abstract

Continuously rising atmospheric CO2 concentrations may lead to an increased transfer of organic C from plants to the soil through rhizodeposition and may affect the interaction between the C- and N-cycle. For instance, fumigation of soils with elevated CO2 (eCO2) concentrations (20% higher compared to current atmospheric concentrations) at the Giessen Free-Air Carbon Dioxide Enrichment (GiFACE) sites resulted in a more than 2-fold increase of long-term N2O emissions and an increase in dissimilatory reduction of nitrate compared to ambient CO2 (aCO2). We hypothesized that the observed differences in soil functioning were based on differences in the abundance and composition of microbial communities in general and especially of those which are responsible for N-transformations in soil. We also expected eCO2 effects on soil parameters, such as on nitrate as previously reported. To explore the impact of long-term eCO2 on soil microbial communities, we applied a molecular approach (qPCR, T-RFLP, and 454 pyrosequencing). Microbial groups were analyzed in soil of three sets of two FACE plots (three replicate samples from each plot), which were fumigated with eCO2 and aCO2, respectively. N-fixers, denitrifiers, archaeal and bacterial ammonia oxidizers, and dissimilatory nitrate reducers producing ammonia were targeted by analysis of functional marker genes, and the overall archaeal community by 16S rRNA genes. Remarkably, soil parameters as well as the abundance and composition of microbial communities in the top soil under eCO2 differed only slightly from soil under aCO2. Wherever differences in microbial community abundance and composition were detected, they were not linked to CO2 level but rather determined by differences in soil parameters (e.g., soil moisture content) due to the localization of the GiFACE sets in the experimental field. We concluded that +20% eCO2 had little to no effect on the overall microbial community involved in N-cycling in the soil but that spatial heterogeneity over extended periods had shaped microbial communities at particular sites in the field. Hence, microbial community composition and abundance alone cannot explain the functional differences leading to higher N2O emissions under eCO2 and future studies should aim at exploring the active members of the soil microbial community. © 2017 Brenzinger, Kujala, Horn, Moser, Guillet, Kammann, Müller and Braker.