RESPONSES OF TOTAL AND ACTIVE SOIL MICROBIAL COMMUNITIES TO CLIMATE WARMING

Abstract

As illustrated by accumulating scientific evidence, unconscionable anthropogenic activities since industrialization such as intensive land utilization and accumulation of various greenhouse gases due to fossil fuel combustion have caused global climate warming, which has in turn caused instability of the earth’s ecosystems and impacts on human society. Granted that huge efforts through scientific research have been devoted to address the interactions between the biosphere and the warmer climate, there are still numerous understudied scientific areas and questions of this topic due to the complicacy of both the biosphere and the climate system. Microbial communities are the most abundant, diverse and complex assemblages in the biosphere, and play crucial roles in geochemical processes closely related to climate warming. However, due to the difficulties in observing and cultivating the microorganisms, responses and feedbacks of microbial communities to climate warming are difficult to observe and predict, in terms of microbial taxonomic, functional and interactional patterns under warming. High-throughput genomic technologies have revolutionized microbial ecology. Such technologies are capable to provide detailed characterization and thus great insight into studies of complex and uncultivated microbial communities and the microbially-mediated mechanisms governing the carbon balance under a warmer climate. Using several such high-throughput genomic technologies, this dissertation attempts to assess responses of soil microbial community to warming, based on field experiments and laboratory incubations. The high-latitude permafrost region (tundra) could be a “hot spot” in global carbon balance and the changing climate because it possesses the largest carbon reservoir globally. This dissertation focuses on the tundra regions and the residing microbial communities, while tall-grass prairie (temperate grassland), an understudied but important ecosystem type among the terrestrial ecosystems, should also be studied and compared to the tundra ecosystem to assess the sensitivities of different types of ecosystems to warming. As short-term warming has been reported as altering microbial functional potentials instead of taxonomic composition, we first attempt to illustrate the impact of long-term (5 years) experimental warming on the responses of the total soil microbial community in the Alaska tundra soils of 0–15 cm depth and their correlations with environmental factors and ecosystem C balance. We applied an amplicon sequencing approach of both bacterial/archaeal 16S rRNA gene and fungal internal transcribed spacer (ITS) to assess microbial taxonomic profiles, and applied the GeoChip 5.0 microarray to assess microbial functional profiles. We observed that longer-term experimental warming altered the structure of tundra bacterial communities (p < 0.040 as revealed by Adonis test) but not fungal communities. Thaw depth was the strongest environmental factor correlating with microbial community assembly and interaction networks, suggesting that warming-accelerated tundra thaw fundamentally restructured the microbial communities. Both carbon decomposition and methanogenesis genes increased in relative abundance under warming, and the functional structures strongly correlated (R2 > 0.725, p < 0.001) with ecosystem respiration and CH4 flux, respectively. These results demonstrate that microbial responses associated with carbon cycling could lead to positive feedbacks that accelerate soil organic carbon decomposition in tundra regions, which is alarming because soil organic carbon loss due to tundra thaw is unlikely to subside owing to changes in microbial community composition. Soil microbial nitrogen fixation serves as a crucial factor in ecosystem feedbacks to climate warming since it largely determines plant growth and plant carbon fixation in tundra regions. Therefore, it is crucial to examine the responses of diazotrophic communities to warming across the depths of Alaska tundra soils. We assessed the dynamics of soil diazotrophic communities spanning both the organic and mineral layers under long-term (5 years) experimental warming and their correlations with ecosystem carbon balance through amplicon sequencing of nifH genes, the α-subunit of bacterial nitrogenase. As observed, warming significantly (p<0.050) enhanced diazotrophic absolute abundance by 86.3% and aboveground plant biomass by 25.2%. Diazotrophic composition in the organic soil layers was markedly altered with an increase of α-diversity. Changes in diazotrophic abundance and composition significantly correlated to soil thaw duration, soil moisture and plant biomass, as shown by structural equation modeling analyses, indicating similar environmental drivers for total microbial and diazotrophic communities. We conclude that more abundant diazotrophic communities induced by warming may serve as an important mechanism of negative feedback to warming by supplementing biologically available nitrogen in the tundra ecosystem. After assessing soil total microbial community in the tundra soil, it is necessary to further assess microbial community that is active in degrading soil carbon compounds, which is more responsible and representative in the microbial interactions with ecosystem carbon balance. Compared to biologically labile carbon compounds, studying the dynamics of biologically recalcitrant carbon compounds under warming could be crucial as they are the dominant components of tundra carbon storage. Lignin is a major component among the biologically recalcitrant carbon compounds in tundra soil. After depleting soil labile C through a 975-day laboratory incubation, the identity of microbial decomposers of lignin and their responses to warming were characterized by applying stable isotope probing to the active layer of Arctic tundra soils. Warming considerably increased both total abundance and functional capacities of all potential lignin decomposers. A β-Proteobacteria genus, Burkholderia, accounted for 95.1% of total abundance of potential lignin decomposers. Consistently, Burkholderia strains isolated from our tundra soils could grow with lignin as the sole carbon source. In addition to Burkholderia, α-Proteobacteria species capable of lignin decomposition (e.g. Bradyrhizobium and Methylobacterium genera) was stimulated by 82-fold, indicating that α-Proteobacteria species are more stimulated by warming than Burkholderia species. Those community changes collectively doubled the priming effect, i.e., decomposition of existing carbon after fresh carbon input to soil. Consequently, warming would cause a higher rate of soil carbon decomposition in the long-term as verified by microbially-enabled climate-carbon modeling. Our alarming findings demonstrate that accelerated carbon decomposition under warming conditions will make tundra soils a larger biospheric carbon source than anticipated. To allow the comparison of microbial feedbacks to warming in different ecosystems, we further assessed the active bacterial community of carbon degradation and its response to warming in grassland ecosystems. Combining metagenomic technologies with a stable isotope probing incubation using isotopically labelled straw to simulate grass litter, we examined the dynamics of active bacterial communities and carbon dioxide emissions of grassland soil in response to a long-term (7 years) experimental warming treatment. Our study unveiled a comprehensive stimulation of warming on active bacterial communities in terms of abundance and carbon-degrading potentials, which collectively increased the carbon degradation rates of both straw input and old soil organic matter, indicative of an increased positive feedback to climate warming. This stronger positive feedback could be permanent as warming compositionally changed the active bacterial communities, turning the majority of phylum Firmicutes into active (18.5% of total bacterial abundance), which was reported as efficient in degradations of both biologically labile and recalcitrant carbon compounds. Moreover, divergent successions (larger internal dissimilarities) of active bacterial communities were observed under warming, which may cause less predictable dynamics of active bacterial communities. The response of microbial carbon-degrading activities to warming displayed a phylogenetic clustering pattern (Firmicutes), indicative of a phylogenetically conserved ecological strategy, which could be related to the alarming positive feedback to the warmer climate. Overall, this dissertation provided valuable observations based on field experiments and laboratory incubations on responses of microbial communities to climate warming, revealed that both total and active microbial communities could be sensitive to warming, and captured warming-induced environmental drivers of microbial communities such as tundra soil thaw and aboveground plant biomass. These findings collectively accumulate valuable experience for microbially-mediated mechanisms underlying carbon cycle ecosystem models in a warmer world.