Termite gas emissions select for hydrogenotrophic microbial communities in termite mounds

Abstract

Organoheterotrophs are the dominant bacteria in most soils worldwide. While many of these bacteria can subsist on atmospheric hydrogen (H ₂), levels of this gas are generally insufficient to sustain hydrogenotrophic growth. In contrast, bacteria residing within soil-derived termite mounds are exposed to high fluxes of H ₂ due to fermentative production within termite guts. Here, we show through community, metagenomic, and biogeochemical profiling that termite emissions select for a community dominated by diverse hydrogenotrophic Actinobacteriota and Dormibacterota. Based on metagenomic short reads and derived genomes, uptake hydrogenase and chemosynthetic RuBisCO genes were significantly enriched in mounds compared to surrounding soils. In situ and ex situ measurements confirmed that high- and low-affinity H ₂-oxidizing bacteria were highly active in the mounds, such that they efficiently consumed all termite-derived H ₂ emissions and served as net sinks of atmospheric H ₂. Concordant findings were observed across the mounds of three different Australian termite species, with termite activity strongly predicting H ₂ oxidation rates ( R ² = 0.82). Cell-specific power calculations confirmed the potential for hydrogenotrophic growth in the mounds with most termite activity. In contrast, while methane is produced at similar rates to H ₂ by termites, mounds contained few methanotrophs and were net sources of methane. Altogether, these findings provide further evidence of a highly responsive terrestrial sink for H ₂ but not methane and suggest H ₂ availability shapes composition and activity of microbial communities. They also reveal a unique arthropod–bacteria interaction dependent on H ₂ transfer between host-associated and free-living microbial communities.