UC Berkeley

Microbial communities play important roles in natural, engineered, and anthropogenically altered systems. Specifically, microorganisms can metabolize contaminants and contribute to nutrient cycling. These processes can be achieved by one species or by the combined effects of multiple species. Hence, communities can possess emergent properties that are not always obvious from work on isolates or taxonomic profiling. Furthermore, the majority of microorganisms have not been cultivated in the laboratory and even for cultivated species, the full metabolic capacity may not be known. Examining genomes, the blueprints for life, and proteomes, the parts assembled from these blueprints, can provide insight into microbial physiologies and their roles in systems of interest. A more detailed understanding of which organisms are responsible for key processes could improve monitoring of in situ bioremediation, allow better targeted biostimulation, and even direct the manipulation of applied microbial systems to be more efficient.

Genome-resolved metagenomics and metaproteomics (“meta-omics”) techniques are approaches that sample biomolecules (DNA and protein) from an intact microbial community, sequence or identify these molecules, and assign the sequences to specific populations. Analysis of the resulting data can yield a species-resolved view of the metabolic potential present in a community. These methods were used to investigate the structure and functioning of microbial communities in mining-contaminated systems. Bioinformatic analyses across three different systems sought to elucidate ecological roles for members of novel bacterial phyla and identify organisms that contributed to contaminant transformation. Contaminant studies focused on removal of common mining-related compounds including thiocyanate, cyanide, and reduced sulfur species. Metagenomes taken in series were used to examine the stability of consortia over time and increased thiocyanate loading while metagenomes taken across a mining landscape were used to assess the diversity, metabolic potential, and seasonal variation of microbial communities.

Most microbial communities include bacteria from major branches of the tree of life with no cultivated representatives. These lineages are referred to as Candidate Phyla (CP), and in the absence of complete genomes or cultivated representatives, many aspects of their biology and ecological roles remained unclear. Extensive characterization of sediment- and groundwater-associated microbial communities in Rifle, Colorado, USA, provided some of the first genomic observations of these CP. The site of a former uranium and vanadium mill in Rifle has been the subject of in situ biostimulation experiments, most notably, acetate addition to increase uranium reduction by the native microbial community. A series of metagenomes from acetate-amended aquifer sediment yielded three complete and one near-complete bacterial genomes from CP, some of the first ever reported. Subsequent exploration of microbial communities involved in thiocyanate remediation and acid mine drainage also recovered genomes from the CP. Metabolic analyses based on the four Rifle genomes revealed the lack of an electron transport chain and pointed to energy generation based on fermentation of organic substrates including sugars, organic acids, amino acids, and DNA. A significant portion of genes in the unusually small genomes were involved in attachment, motility, and cell surface modification. Perhaps most importantly, none of the four genomes contained genes required for the complete biosynthesis of nucleic acids and amino acids. Taken together, all evidence suggests an obligately symbiotic or parasitic lifestyle for all four organisms.

Thiocyanate (SCN-) is a common industrial contaminant produced at high quantities in gold mining. Chemical degradation of this compound is expensive and can produce other toxic byproducts, whereas biological treatment produces sulfate, ammonium, and carbon dioxide. Thiocyanate bioremediation has been successful at the pilot and industrial scale, but the biological underpinnings of the process were not well understood. In order to identify key pathways and organisms involved in thiocyanate degradation, microbial communities of two laboratory-scale continuous flow bioreactors were studied. The first reactor was a long-running system fed at high thiocyanate loadings whereas the second, inoculated with mixed culture from the first, was fed both thiocyanate and cyanide. Metagenomic sequencing and analysis of the two reactor communities resulted in a total of 93 bacterial and two eukaryotic genome bins. Based on coverage, the most abundant organisms in both reactors belonged to the genus Thiobacillus. Importantly, the genomes for these organisms encoded the enzyme thiocyanate hydrolase, located in an operon with cyanase and a predicted thiocyanate transporter. Other organisms in the reactor were predicted to oxidize sulfur or ammonium produced during thiocyanate degradation, and some possessed genes encoding denitrification. Whereas prior culture-based approaches had suggested heterotrophic organisms were responsible for thiocyanate degradation and that this process requires oxygen, the Thiobacillus spp. genomes encoded autotrophic metabolism and anaerobic respiration using nitrate. Altogether, a qualitative model of the community suggested that further optimization of the bioreactor system to encourage the growth of autotrophs could improve thiocyanate removal and that some ammonium removal may be possible via conversion to nitrogen gas.

A second study of thiocyanate degradation examined the stability and proteome of the microbial community across increasing loadings of thiocyanate. Biomass from the same long-running thiocyanate bioreactor was used to inoculate two new laboratory-scale reactors. One was fed thiocyanate at increasing loadings, while the other was fed ammonium sulfate at parallel nitrogen loadings to mimic the breakdown products of thiocyanate. Metagenomes from the ammonium sulfate reactor showed enrichment of community members involved in nitrogen cycling and heterotrophic metabolism compared to the inoculum. Meanwhile the thiocyanate reactor showed an increase in the relative abundance of three thiocyanate-degrading Thiobacillus species. Two of these species were composed of substrains whose differential abundances shifted across the time series, suggestive of within-species competition or niche specialization. Proteomic data collected at the final time point, when thiocyanate loading was highest, showed that the thiocyanate hydrolase enzyme was highly expressed by all three Thiobacillus spp. in whose genomes it was found. Proteins involved in sulfide oxidation, ammonium oxidation, nitrite oxidation, and carbon fixation were also detected, consistent with the model for thiocyanate degradation proposed in the prior study. Supporting the prediction of anaerobic zones in biofilm, proteins involved in denitrification were also detected in the proteomics. The biofilm and planktonic community compositions were similar, but more biomass appeared to be present in the biofilm. The importance of biofilm for uncoupling hydraulic retention time from bacterial growth rates and improving denitrification suggests consideration of biofilm-based rather than sludge-based bioreactor designs in the future.

While laboratory-scale reactors allow detailed investigations of simplified communities, microbial communities in the environment can be much more complex. Mining waste sites exhibit a range of geochemical conditions that can shape the local microbial community, and in turn this impacted community may contribute to in situ remediation. Metagenomics was applied to four high-sulfur mining wastewaters and to a reservoir used to treat those wastes in order to construct a picture of the biological processes in play. Elemental sulfur was the most abundant form of sulfur at all sites, and specific clades of sulfur oxidizing organisms were enriched at each. Acidic waste rock sites and wastewater piped to the reservoir were enriched in acidophilic iron- and sulfur-oxidizing organisms. In contrast, drainage from a tailings-dewatering site high in organic carbon contained mainly methylotrophic bacteria, suggesting less sulfur oxidation occurs there. Metagenomes from the reservoir showed that it contained common freshwater bacteria as would be expected due to inputs from higher in the watershed, but it also contained numerous sulfur oxidizing bacteria. Some of these bacteria were likely capable of sulfur oxidation coupled to nitrate reduction under anoxic summer conditions. Importantly, sulfur oxidizers were present in late summer and early winter at multiple depths, despite a seasonal shift in the reservoir community. Mining operations rely on unmanaged in situ remediation in the reservoir to convert all sulfur compounds to sulfate, preventing acidification of receiving water bodies. This improved understanding of microbial metabolism present at various sites could lead to the development of active management strategies to achieve more complete, reliable remediation.

Metagenomic approaches provide foundational information about microbial communities and can be used to monitor these communities during remediation. Further biochemical validation and controlled experiments are needed to enable quantitative modeling of reaction rates required for engineering new remediation solutions. Mining represents perhaps the largest anthropogenic manipulation of the surface and subsurface of the planet, but it is also tied to the economic development of many countries. Ultimately, deeper knowledge of mining waste microbiology may lead to safer, more effective handling of mining wastes, benefiting environmental and public health.