Abstract
Microbes usually provide important functions as the basal trophic level of ecosystems. Their high reproduction and dispersal capabilities can result in their ubiquity. The “everything is everywhere” hypothesis of bacterial biogeography has been challenged recently by an indication that abiotic factors determine bacterial communities. Vernal pools are insulated environments, which can function similar to islands in regard to the movement of organisms. An alternative hypothesis is that vernal pools in close proximity will have increased movement of organisms from one to the other, while pools that are further away would have less flow of organisms resulting in closer pools supporting similar bacterial communities. California vernal pool communities (primarily plants) have been researched at the local and landscape level but little to no characterization of the microbial community has been done. For this environment it is hypothesized that there will be a difference in the bacterial communities in different California vernal pool ecosystems (CVPE) and that the difference will be driven by location or abiotic factors. Six vernal pools from three different sites were used to evaluate whether location or abiotic factors drive bacterial community structure. Two vernal pools were sampled from each of these three locations; Mather Field Air Force Base, Travis Air Force Base and Beale Air Force Base. Turbidity, pH, phosphates and conductivity were used as abiotic factors. All of the bacterial analysis was carried out using community analysis techniques including; community level physiological profiles, terminal restriction fragment length polymorphism and next generation sequencing. Bacterial cells in each sample were enumerated using fluorescent staining and flow cytometry. Biolog EcoPlates™ were used to assess differences in metabolism for the communities. Terminal restriction fragment length polymorphism (t-RFLP) and next generation sequencing (NGS) were used to evaluate bacterial community diversity. Water samples contained between 4x107 and 6x107 cells per ml of water. Clustering analysis using the Biolog EcoPlates™ data showed two distinct groups: one contained Travis2, and both Beale pools and the other containing Travis1 and both Mather pools. Principal component analysis (PCA) was conducted showing the same grouping. Axis 1 showed a moderate negative correlation with phosphates (r = -0.473, r2 = 0.223, tau = -0.200) and axis two showed strong negative correlations with phosphates (r= -0.760, r2 = 0.577, tau= -0.333) and pH (r = -0795, r2= 0.632, tau= -0.733). t-RFLP showed differences dependent on the enzyme used, with all enzymes showing low values of the Bray-Curtis index of similarity. The highest value produced was from using the restriction enzyme RsaI between Travis1 and Travis2 (0.37). NGS showed six main classes of bacteria found in each pool, although relative abundance of each differed between pools: Spartobacteria, Verrucomicrobiae, Gammaproteobacteria, Betaproteobacteria, Alphaproteobacteria, and Sphingobacteria. All are common freshwater bacteria, which generally function in degradation of organic matter. The Bray-Curtis Index of similarity was higher using NGS data, with the highest value being 0.89 between Travis1 and Beale2. A PCA was conducted using NGS data also showing correlations with pH, phosphates and turbidity. UniFrac Unweighted Pair Group Method with Arithmetic Mean (UPGMA), resulted in two strongly supported groups one containing the Travis pools and the other containing the Beale and Mather pools. This indicates that an evolutionary significant branch was found in the Travis pools, which was not located in the rest of the pools. All of this data together supports the idea that there are differences in the bacterial communities in CVPE and that the data supports that abiotic factors are driving those differences. Further research will be needed to understand any biogeographical patterns, since only a few pool exhibited biogeographical influences.