Methylmercury (MeHg) is a bioaccumulative neurotoxin that is produced by certain anaerobic microorganisms, but the abundance and importance of different methylating populations in the environment is not well understood. We combined mercury geochemistry, hgcA gene cloning, rRNA methods, and metagenomics to compare microbial communities associated with MeHg production in two sulfate-impacted lakes on Minnesota’s Mesabi Iron Range. The two lakes represent regional endmembers among sulfate-impacted sites in terms of their dissolved sulfide concentrations and MeHg production potential. rRNA amplicon sequencing indicates that sediments and anoxic bottom waters from both lakes contained diverse communities with multiple clades of sulfate reducing Deltaproteobacteria and Clostridia. In hgcA gene clone libraries, however, hgcA sequences were from taxa associated with methanogenesis and iron reduction in addition to sulfate reduction, and the most abundant clones were from unknown groups. We therefore applied metagenomics to identify the unknown populations in the lakes with the capability to methylate mercury, and reconstructed 27 genomic bins with hgcA. Some of the most abundant potential methylating populations were from phyla that are not typically associated with MeHg production, including a relative of the Aminicenantes (formerly candidate phylum OP8) and members of the Kiritimatiellaeota (PVC superphylum) and Spirochaetes that, together, were more than 50% of the potential methylators in some samples. These populations do not have genes for sulfate reduction, and likely degrade organic compounds by fermentation or other anaerobic processes. Our results indicate that previously unrecognized populations with hgcAB are abundant and may be important for MeHg production in some freshwater ecosystems.
Bibliographical noteFunding Information:
Acknowledgements We thank D. Engstrom, M. Kelly, and M. Berndt for insightful discussion and assistance with sample collection. Thanks to C. Nguyen, L. Thomas, the University of Minnesota College of Science and Engineering Systems staff, and the Minnesota Supercomputing Institute for computing support and the use of facilities. Thanks to H. Huang for providing mercury analytical support. Sequencing was performed at the University of Minnesota Genomics Center (UMGC). We thank the University of Minnesota Undergraduate Research Fellowship Program (UROP) for supporting G.M. W. This work was supported by University of Minnesota MnDRIVE initiative and the University of Minnesota BioTechnology Institute. The metagenomic sequencing was funded by a grant from the University of Minnesota Duluth’s Natural Resources Research Institute Permanent University Trust Fund and by an MnDRIVE-supported research grant from the University of Minnesota Duluth’s Swenson College of Science and Engineering.