Long-term elevated precipitation promotes an acid metabolic preference in soil microbial communities in a Tibetan alpine grassland

Abstract

Alpine ecosystems store vast amounts of soil organic carbon but are highly sensitive to climate change. Despite this, the response of in situ soil microbial metabolic processes, especially carbon substrate utilization, to climatic shifts remains underexplored. Here, we assessed microbial activity by metatranscriptomics in a Tibetan alpine grassland after a decade of experimental warming (+2°C) and altered precipitation (+50% and -50% of ambient precipitation). The experiment revealed that altered precipitation, rather than warming, shaped the active microbial community. Altered precipitation and warming had significant interactions: warming combined with increased precipitation generally suppressed microbial carbohydrate metabolism and methane oxidation, while warming with decreased precipitation enhanced these processes. Notably, increased precipitation induced a shift in microbial communities towards acid metabolism over sugar metabolism, predominantly driven by taxa such as Betaproteobacteria. This metabolic shift corresponded with an increased emission ratio of methane (CH₄) to carbon dioxide (CO₂), a change primarily driven by CH₄, underscoring the critical role of microbial carbon metabolic preferences in regulating greenhouse gas emissions. Our findings highlight the necessity of integrating microbial carbon metabolic preferences and their interactions with climatic factors into models to accurately predict carbon-climate feedbacks.IMPORTANCEMicrobes have specific preferences for different carbon substrates, but their responses to climate change remain unclear. Our study, conducted through a long-term climate manipulation experiment in a Tibetan alpine grassland, reveals that increased precipitation leads soil microbial communities to favor acid metabolism over sugar metabolism. This shift significantly affects greenhouse gas emissions by increasing the CH₄/CO₂ ratio, which has important implications for global warming. These findings are crucial for accurately forecasting carbon-climate feedbacks and managing alpine ecosystems as climate change progresses.

Keywords: active soil microbes; alpine ecosystems; carbon metabolic preferences; climate change; greenhouse gas emissions; metatranscriptomics.

Fig 1

Effects of experimental treatments on soil variables and carbon fluxes. (A) Effects of altered precipitation and warming on environmental factors, microbial biomass carbon, and carbon fluxes. (B) Effects of altered precipitation and warming on CH₄/CO₂. The effect sizes were regression coefficients in the LMMs [LMMs: y ~ altered precipitation*warming + (1|block)]. The effect sizes of drought and wet were the effect sizes of altered precipitation multiplied by −0.5 and 0.5, respectively. Data were coefficients and standard errors of LMMs. Type II Wald χ tests were used to determine statistical significance: ***P < 0.001, **P < 0.01, *P < 0.05, ^# P < 0.1. Abbreviations in subplot: P, altered precipitation; W, warming; P*W, interaction between altered precipitation and warming. Data in the corner of plot B were regression coefficients (β) in the LMMs. Environmental factors: soil temperature, soil moisture, pH. Microbial biomass carbon: MBC. Ecosystem carbon fluxes: NEE (net ecosystem carbon exchange), ER (ecosystem respiration). Soil carbon fluxes: CH₄, Rs (soil respiration), Rh (heterotrophic respiration).

Fig 2

Effects of experimental treatments on active soil microbes. (A) Proportion of the relative abundance of phylum-level affiliations of bacterial transcripts and class-level affiliations of archaeal transcripts across all plots. Pseudominadota was divided into Alphaproteobacteria, Betaproteobacteria, and Gammaproteobacteria based on the class level. (B) Difference in active soil microbial composition. The figure showed the first and second axes of principal coordinate analysis (PCoA) based on the weighted Bray-Curtis distance of the relative abundance of species-level transcripts. The values in parentheses on the axes represented the relative eigenvalues of PCoA. The values at the top of the subplot indicated the statistical significance (P values) determined by PERMANOVA using Adonis. (C) Relative abundance of taxonomic affiliations of transcripts associated with bacteria. (D) Relative abundance of taxonomic affiliations of transcripts associated with methanotrophs and methanogens. The values in the corner of each subplot represented the effect sizes of treatments on them [LMMs: relative abundance ~altered precipitation*warming + (1|block)]. Type II Wald χ tests were used to determine statistical significance: ***P < 0.001, **P < 0.01, *P < 0.05, ^# P < 0.1. Abbreviations in subplots: P, altered precipitation; W, warming; P*W, interaction between altered precipitation and warming.

Fig 3

Effects of experimental treatments on microbial carbohydrate and methane metabolism. (A) Relative abundance of transcripts associated with CAZy and its modules, i.e., glycoside hydrolases (GH) and glycosyltransferases (GT). (B) Effects of altered precipitation and warming on the relative abundance of transcripts associated with polysaccharide lyases (PL), auxiliary activities (AA), carbohydrate esterases (CE), and carbohydrate binding (CBM). (C) Relative abundance of transcripts associated with methane metabolism across drought, ambient, wet, drought and warming, warming, as well as wet and warming plots (left). Effects of altered precipitation and warming on the relative abundance of transcripts associated with methane metabolism (right). Data were coefficients and standard errors of LMMs. Methane metabolism included methanotrophy (four gene families) and methanogenesis (14 gene families), which was divided into methylotrophic methanogenesis (two gene families), aceticlastic methanogenesis (three gene families), and hydrogenotrophic methanogenesis (six gene families) based on the substrates and core methanogenesis (three gene families). (D) Relative abundance of transcripts associated with K10945 (pmoB-amoB). The values in the corner of each subplot represented the effect sizes of experimental treatments on them [LMMs: relative abundance ~altered precipitation*warming + (1|block)]. Type II Wald χ tests were used to determine statistical significance: **P < 0.01, *P < 0.05, ^#P < 0.1. Abbreviations in subplots: P, altered precipitation; W, warming; P*W, interaction between altered precipitation and warming.

Fig 4

Responses and taxonomical affiliations of sugar and acid transporters. (A-C) Responses of the relative abundance of transcripts associated with sugar transporters (A) and acid transporters (B) and SAR-Transporter (C) to experimental treatments. The values at the top of each subplot represented the effect sizes of treatments determined by LMMs [relative abundance/SAR ~ altered precipitation*warming + (1|block)]. Type II Wald χ tests were used to determine statistical significance: **P < 0.01, *P < 0.05. Abbreviations in subplots: P, altered precipitation; W, warming; P*W, interaction between altered precipitation and warming. (D) Linking microbial carbon metabolic preferences with active microbes. Proportion of the relative abundance of taxonomical affiliations of transcripts associated with sugar or acid transporters (left). Sankey diagram showing the contributions of active microbes to sugar and acid transporters (right).

Fig 5

Linking microbial carbon metabolic preferences to CH₄/CO₂. (A) Correlation between SAR-Transporter and CH₄/CO₂. r represented the correlation coefficient determined by LMMs [CH₄/CO₂ ~SAR-Transporter + (1|block)], and P represented the statistical significance of LMMs in the corner of subplot. (B) Effects of environmental variables and microbial carbon metabolism on CH₄/CO₂. Standardized regression coefficients of LMMs [CH₄/CO₂ ~soil moisture + soil temperature + SAR-Transporter + CAZy + (1|block)] represented the effect sizes of environmental variables and microbial carbon metabolism on CH₄/CO₂ based on rescaled predictors. Type II Wald χ tests were used to determine the statistical significance: *P < 0.05. (C) Linking microbial carbon metabolic preferences with soil carbon fluxes. The size of the circles represented the magnitude of similar indexes. In the sugar and acid transporters, the colors of the circles indicated transcript taxonomy (see Fig. 4). For CO₂, CH₄, CH₄/CO₂, and SAR, the colors of the circles represented the magnitude of the values, with darker colors indicating higher values. The data excluded transcripts without taxonomic annotation. Sugar, acid, and SAR represented sugar transporters, acid transporters, and SAR-Transporter, respectively. CO₂ represented the heterotrophic respiration flux.