Heterogeneity of rock-hosted microbial communities in a serpentinizing aquifer of the Coast Range Ophiolite

Abstract

The movement of groundwater through fractured bedrock provides favorable conditions for subsurface microbial life, characterized by constrained flow pathways and distinctive local environmental conditions. In this study, we examined a subsurface microbial ecosystem associated with serpentinized rocks recovered from the Coast Range Ophiolite in northern California, USA. The distribution and diversity of microbial communities at various depths within two separate cores reaching up to 45 m below the land surface were investigated with microbiological and geochemical approaches. Core samples contained low total organic carbon content, low DNA yields, and low copy numbers of 16S rRNA genes, yet some samples still yielded amplifiable DNA sequences. The microbial community composition of rock cores was distinct from groundwater, and source tracking of DNA sequences indicated that groundwater is not a significant source of DNA into basement rocks. In contrast, the microbial community of some rock core samples shared similarities with overlying soil samples, which could indicate potential contamination, weathering of shallow serpentinites, or a combination of both. Individual DNA sequences of archaea and bacteria predicted to be endemic to the basement rocks were identified by differential abundance analyses. Core-enriched sequences were distinct from those in groundwater or in the overlying soils and included OTUs related to Serpentinimonas as well as putatively anaerobic, deep subsurface-associated taxa such as methanogens and Bathyarchaeia. Stable isotope analyses of organic and inorganic carbon did not reveal a chemoautotrophic signal and were instead consistent with a primarily surface vegetation source of organic carbon into the basement. This census of archaeal and bacterial DNA sequences associated with altered ultramafic rocks provides a useful resource for further research into the potential for deep subsurface microbial activity fueled by geochemical reactions associated with serpentinization.

Keywords: 16S rRNA; low biomass; rock hosted; serpentinization; subsurface.

Figure 1

Figure 2

Statistical methods were used to identify endemic core microbes. SourceTracker2 was used to identify the percent of each core community that likely came from either soils (brown) or fluids (blue) for bacteria (A) and archaea (D). Samples with greater than 50% contribution of OTUs from soils and fluids (QV7R, QV18R, and QV30R for bacteria; QV30R for archaea) were removed from the “core” classification for downstream differential expression analysis in EdgeR, used to identify core-enriched sequences (red). Comparisons were performed for bacterial data between core and soil samples (B) and core and fluid samples (C) and archaea between core and soil samples (E).

Figure 3

Relative abundance of core-enriched bacterial taxa. The sizes of the bars represent the percent of the bacterial community composed of core-enriched bacterial taxa, with white space beyond the bar representing the portion of the microbial community that could not be attributed to core-enriched taxa (as defined by the differential abundance metrics). The core-enriched taxa are identified at the phylum level, except for the Proteobacteria, which are represented at the class level (Alphaproteobacteria and Gammaproteobacteria).

Figure 4

Relative abundance of core-enriched archaeal taxa. The sizes of the bars represent the percent of the archaeal community composed of core-enriched taxa, with white space beyond the bar representing the portion of the archaeal community that could not be attributed to core-enriched taxa (as defined by the differential abundance metrics). The core-enriched taxa are identified at the class level. Archaea were only able to be sequenced from five core samples. All the archaeal taxa identified in QV30R were identical to soil-enriched taxa; therefore, there are no core-enriched taxa in this sample to be represented here.

Figure 5

Environment of best hit to core-enriched sequences. Representative sequences from abundant core-enriched OTUs (>1% in any sample) were BLASTed to identify the best hit. The environmental origin of each best hit was identified from the literature and classified into an environmental category. The cumulative relative abundances of core-enriched OTUs are plotted according to the environmental category of their best hit. Note this is only the percentage of the core-enriched OTUs, which constitute a subset of the whole community.