| dc.description.abstract | Global terrestrial ecosystems are subjected to various climate change factors, including the concurrent elevated CO2 (eCO2) and nitrogen deposition (eN). Despite the increasing appreciation that eCO2 and eN can interactively affect aboveground plants, how they will affect soil microbial communities and associated ecoprocesses remain understudied. This dissertation addresses this gap by examining the response of soil microbial communities associated with two plant functional groups (C3 grass and legumes) to eCO2 and eN in a long-term (12-year) field experiment (BioCON).
In the beginning of this study, we investigated soil bacterial and archaeal communities subjected to CO2 (ambient, 368 µmol mol−1, versus elevated, 560 µmol mol−1) and N (ambient, 0 g m-2 yr-1, versus elevated, 4 g m-2 yr-1) treatments using Illumina MiSeq sequencing of 16S rRNA gene amplicons. Over 2.3 million passing sequences were obtained from a total 24 samples, corresponding to 38 known phyla, 96 classes, and 565 genera. Elevated CO2 significantly altered the diversity and structure of microbial communities, but these changes vary greatly depending on soil N conditions and plant functional groups. In C3 grass plots, community diversity increased with eCO2. A positive eN effect on community richness was also observed. These shifts in community structure and composition may be driven by differential responses of microbial taxonomic groups to eCO2 and/or eN. For example, Actinobacteria abundance decreased with the main effect of eCO2, accounting for about 20.3% of the total population in the C3 grass. Chlamydiae increased with eCO2 but only under eN condition. The abundance of Woesearchaeota increased with eN, but no effect of eCO2 on its abundance was observed. Whereas in legume plots, community richness increased with eCO2. The abundance of Actinobacteria, Chloroflexi, Armatimonadetes, Saccharibactiera, and Euryarchaeota, accounting for about 21.2% of the total population in legume plots, decreased with eCO2, eN or both. Only Nitrospirae and Latescibacteria increased with eCO2 in their abundance. Changes in community diversity and composition were significantly related to plant and soil properties including plant biomass, biomass N content and C/N ratio, soil ammonium and nitrate, pH, moisture, temperature, and soil C and N contents by Mantel analysis. In addition, our results suggested that copiotrophic-like bacteria appear to be more abundant in the legume than in the C3 grass plots, whereas oligotrophic-like bacteria appear to be more abundant in the C3 grass than in the legume plots. Collectively, these results revealed different impacts of eCO2 and eN on soil microbial community diversity and composition with few common trends observed across plant functional groups, providing new information for our understanding of the feedback response of soil microbial communities to global change factors.
In the following, we used high-throughput microbial functional gene microarray (GeoChip), stable isotope-based microbial C-sequestration and N-fixation measurements to detect and identify the impacts of eCO2 and eN on soil microbial functional communities. We found that long-term changes in CO2 and N availability dramatically altered the diversity and structure of C3 grass-associated soil microbial functional genes via several mechanisms, such as altering plant fine root production, exudation and soil moisture. There was an antagonistic relationship between eCO2 and eN that affected a large number of microbial functional genes, with eCO2 generally increasing the abundance of these genes at aN, but either decreasing or increasing abundance to a minimal degree at eN. These results imply that eCO2 may accelerate C and nutrient cycling in the C3 grass system, but the magnitude of effect is strongly dependent on the relative availability of N. Particularly, microbial activities associated with chemically recalcitrant soil organic matter (SOM) turnover significantly increased with eCO2 in low fertility condition (unfertilized C3 grass plots of this study). These changes in C degradation genes suggested enhancement of microbial N mining under long-term eCO2, an effect may limit soil C storage and stability. Meanwhile, eCO2 and eN had surprisingly minor effects in the legume-associated soil microbial functional community with generally lower gene abundances at eCaN condition and higher gene abundances at eCeN condition, suggesting that the impacts of eCO2 and eN are plant-functional-group-specific. This study provides new insights into our understanding of microbial functional processes in response to multiple global change factors.
The potential impact of global change factors (e.g., eCO2, eN) on microbial activities, such as decomposition of various organic compounds, remains largely inferred from metagenomic analysis targeting the 16S, 28S rRNA, ITS or functional genes. However, the actual activity of microorganism can't be directly measured by such DNA-based technologies. Thus, the following study focused on assessing the influence of long-term eCO2 on belowground microbial metabolic potential on different C sources provided on Biolog EcoPlate and determining whether the effect of eCO2 was regulated by soil N conditions in the two plant functional groups (C3 grass and legumes). By cultivating soil samples on Biolog EcoPlate containing 31 low molecular weight C substrates, we constructed sole C source utilization profiles of microbial communities in soil samples mentioned above. We found that community composition of soil microbes based on metabolic potential (utilization rates of 31 C sources) in the C3 grass plot soils was significantly different from those in the legume plot soils by both DCA and nonparametric dissimilarity tests. Microbial communities in legume plots had significantly higher metabolic potential than in C3 grass plots for decomposing organic substrates. Specifically, compared to the C3 grass plots, microbes in legume plots can use a larger number of C sources provided on EcoPlate and with greater decomposition rates during the measured time. Elevated CO2 and eN didn't significantly alter metabolic potential of microorganisms in the C3 grass plots. In contrast, overall microbial metabolic activities significantly increased with eCO2 by 20.6% in the fertilized legume plots, while there was no evidence for a CO2 effect in the non-fertilized legume soils. Total soil N content, root ingrowth biomass, aboveground biomass, and root biomass N content as environmental attributes were closely correlated with microbial C utilization patterns as suggested by the Mantel test. In addition, PLFA analysis showed both total microbial and bacterial biomass were significantly lower in legume than in C3 grass plots, showing an opposite trend to the microbial metabolic potential in these plots. Collectively, these results demonstrated that eCO2 effects on active microbial metabolic activities are contingent on N conditions, and such effect differs between plant functional groups. Differences in microbial metabolic potential among treatments and between plant functional groups were not attributed to population size (biomass) but likely attributed to changes in community structure and/or enzymatic activities of belowground microbes.
We further analyzed the impacts of long-term eCO2 and eN on AOA communities by using 454 pyrosequencing of archaeal-amoA gene amplicons. A total of 87 amoA OTUs (95% identity cutoff) were generated from 26,211 qualified reads. The diversity of AOA communities measured by OTU richness (Chao1), Pielou evenness, Shannon and phylogenetic hill number were significantly reduced by eCO2 but the CO2 effects were confined within ambient N condition. PCoA, βMNTD and non-parametric dissimilarity tests revealed significant CO2 effects on the community structure of AOA regardless of N deposition, but no effect of N was observed. We also detected significant changes of several AOA taxa in their relative abundance, which were significantly correlated with plant root biomass, proportional soil moisture, and pH. In addition, significant positive correlations between AOA taxa and soil nitrification rate were observed, indicating AOA may be actively involved in the nitrification process in grassland soil. Interestingly, eCO2 and eN alone and combined did not significantly alter the abundance of AOA. These results are important in furthering the understanding of the global change impacts on AOA community in the long term.
All studies included in this work provided novel insights into the long-term eCO2 effects on belowground microbial communities. Our results demonstrated that eCO2 effects are contingent on soil N conditions and plant functional groups, underscoring the difficulty toward predictive modeling of soil ecosystem under future climate scenarios and necessitating more detailed studies. | en_US |
