New publication from Mohn Lab shows diversity and significance of steroid-degrading bacteria is largely underestimated

Steroids in the environment accumulate from both natural and anthropogenic sources. Cholesterol, for example, is an essential part of cellular membranes and a natural source of steroids in the environment. Anthropogenic sources include steroid hormones associated with birth control pills. Regardless of where they originate, however, steroids have been found to accumulate in soil, wastewater treatment plants, and aquatic environments, where even at low concentrations they have negative impacts on animals—including humans. So far, only a few types of bacteria are known to degrade steroids in the environment and these species will play a big role in regulating steroidal pollution and its impacts.

To better understand the distribution and ecological significance of these steroid degraders, researchers from the Canadian universities of British Columbia and Waterloo, along with collaborators from Georgetown University in Washington, set out to apply a metagenomics approach to studying these bacteria. This approach uses DNA sequencing to find genes from the 9, 10-seco pathway responsible for steroid degradation in environmental samples and not the bacteria itself. The team then builds phylogenies to find the bacteria according to the phyla in which these genes occur.

The results of this paper supported earlier work showing that bacteria using the 9, 10-seco pathway belong to the Actinobacteria and Proteobacteria phyla. Members of both phyla coexist in wastewater, while species of Actinobacteria alone are found in soil and rhizospheres. While the complete set of genes used in this pathway were not assigned to any other phylum, evidence for steroid degradation ability was found for the first time in the alphaproteobacterial lineages Hyphomonadaceae, Rhizobiales, and Rhodobacteraceae, as well as the gammaproteobacterial lineages Spongiibacteraceae and Halieaceae. Actniobacterial degraders were found in the deep ocean samples while alpha- and gammaproteobacterial degraders were found in other marine samples, including sponges. Furthermore, the authors confirmed that the steroid-degrading bacteria from sponges, Spongiibacteraceae and Halieaceae, catabolize steroids.

The metagenomics approach is a useful one because many bacterial species cannot be cultured and identified directly. However, the techniques involved in DNA extraction and sequencing have inherent biases that cannot be avoided. It is therefore important to note that the absence of steroid degradation proteins from a sample does not definitely mean that the bacteria are not present. Despite this potential underestimation, this study is, according to researchers, “the first analysis of aerobic steroid degradation in diverse natural, engineered, and host-associated environments via bioinformatic analysis of an extensive metagenome data set.” Not only does this confirm the usefulness of the technique; it also demonstrates that the ecological significance and taxonomic and biochemical diversity of these bacteria have been largely underestimated.

Holert J, Cardenas E, Bergstrand LH, et al. Metagenomes reveal global distribution of bacterial steroid catabolism in natural, engineered, and host environments. MBio. 2018; 9: e02345-17.

About Microbiome Insights

Microbiome Insights, Inc. is a global leader providing end-to-end microbiome sequencing and comprehensive bioinformatic analysis. The company is headquartered in Vancouver, Canada where samples from around the world are processed in its College of American Pathologist (CAP) accredited laboratory. Working with clients from pharma, biotech, nutrition, cosmetic and agriculture companies as well as with world leading academic and government research institutions, Microbiome Insights has supported over 999 microbiome studies from basic research to commercial R&D and clinical trials. The company's team of expert bioinformaticians and data scientists deliver industry leading insights including biomarker discovery, machine-learning based modelling and customized bioinformatics analysis.