AUTHOR=Zhang Qian , Tang Jie , Angel Roey , Wang Dong , Hu Xingyi , Gao Shenghua , Zhang Lei , Tang Yuxi , Zhang Xudong , Koide Roger T. , Yang Haishui , Sun Qixiang TITLE=Soil Properties Interacting With Microbial Metagenome in Decreasing CH4 Emission From Seasonally Flooded Marshland Following Different Stages of Afforestation JOURNAL=Frontiers in Microbiology VOLUME=13 YEAR=2022 URL=https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2022.830019 DOI=10.3389/fmicb.2022.830019 ISSN=1664-302X ABSTRACT=

Wetlands are the largest natural source of terrestrial CH4 emissions. Afforestation can enhance soil CH4 oxidation and decrease methanogenesis, yet the driving mechanisms leading to these effects remain unclear. We analyzed the structures of communities of methanogenic and methanotrophic microbes, quantification of mcrA and pmoA genes, the soil microbial metagenome, soil properties and CH4 fluxes in afforested and non-afforested areas in the marshland of the Yangtze River. Compared to the non-afforested land use types, net CH4 emission decreased from bare land, natural vegetation and 5-year forest plantation and transitioned to net CH4 sinks in the 10- and 20-year forest plantations. Both abundances of mcrA and pmoA genes decreased significantly with increasing plantation age. By combining random forest analysis and structural equation modeling, our results provide evidence for an important role of the abundance of functional genes related to methane production in explaining the net CH4 flux in this ecosystem. The structures of methanogenic and methanotrophic microbial communities were of lower importance as explanatory factors than functional genes in terms of in situ CH4 flux. We also found a substantial interaction between functional genes and soil properties in the control of CH4 flux, particularly soil particle size. Our study provides empirical evidence that microbial community function has more explanatory power than taxonomic microbial community structure with respect to in situ CH4 fluxes. This suggests that focusing on gene abundances obtained, e.g., through metagenomics or quantitative/digital PCR could be more effective than community profiling in predicting CH4 fluxes, and such data should be considered for ecosystem modeling.