My Research

UPDATE: My dissertation is available, open-access here:×916033. Additional publications to follow.

My dissertation research uses tools for describing genes, genomes, and microbial communities to learn more about the metabolism of specific chlorine compounds.

Perhaps no other element sounds more unnatural than “chlorine.” Indeed, chlorine deserves its reputation. Chlorine gas, chlorine dioxide, and hypochlorous acid (bleach) are used to kill bacteria and viruses, perchlorate is used in explosives, organic chlorine is dangerous as a solvent and inert as a plastic, and, most dramatically, volatile chlorine eats away at the ozone layer protecting all of life on Earth. Accidental and intentional release of these chemicals into the environment has caused innumerable damage.

Chlorine compounds used in industry. (Chlorite is a byproduct of chlorine dioxide application).

However, many of these chlorine compounds are produced in low amounts as part of continual natural processes, or cycles, that have existed for hundreds of millions of years. That does not mean that chlorine compounds are any less toxic. But by studying the tools that life uses to change one form of chlorine to another, we can create biotechnology to better use or dispose or chlorine compounds. New methods, such as being able to sequence and analyze DNA directly from the environment (metagenomics), have rapidly expanded the information available to understand the chlorine cycle.

Studying communities using genomics

Perchlorate (ClO4-) can be respired in the absence of oxygen by perchlorate-reducing bacteria, which are the essential agents in remediating contaminated sites. Because many bacteria cannot be grown in the laboratory, it was suspected that wild communities degrading perchlorate had more diverse genes, organisms, and interactions than had been previously found.

In a 2018 paper [1], I examined communities of microorganisms from the San Francisco Bay that degraded perchlorate. I assembled short reads from Illumina sequencing into fragments of genomes (contigs) using the Berkeley Research Computing high performance computer cluster and sorted the contigs together into high quality genomes using manual and automated tools. I also used assembly graphs to link horizontally transferred genes to specific genomes, which was a new addition to the field. As expected, I found additional diversity of organisms, and a panel of comparative genomics, protein alignments, and phylogenetics provided support for a new type of enzyme necessary for perchlorate reduction, which I am now validating. Finally, I summarized the metabolisms of all genomes present to identify interactions that may influence perchlorate degradation.

A summary of genes (A), abundance (B), and an estimate of growth rate (C) for each genome from different communities. Figure reproduced from Barnum et. al 2018.

[1] Barnum TP, Figueroa IA, Carlström CI, Lucas LN, Engelbrektson AL, Coates JD. (2018). Genome-resolved metagenomics identifies genetic mobility, metabolic interactions, and unexpected diversity in perchlorate-reducing communities. ISME J 12: 1568–1581. Open-access PDF.

Studying microbe-microbe metabolic interactions

Chlorate (ClO3-), chlorite (ClO2-), and oxygen (O2) are the intermediates in the biochemical pathway for reducing perchlorate to chloride and water. Each of those compounds – but not perchlorate – can be used by microbes called chlorate-reducing bacteria. A chlorate-reducing bacterium had been identified in perchlorate-reducing communities [1], raising the intriguing possibility that these metabolisms interact but without direct confirmation.

While isolating perchlorate-reducing bacteria, a process that should result in one single organism in culture, I discovered that cultures were regularly contaminated by chlorate-reducing bacteria that were altering the rate of perchlorate reduction. Beyond simply persisting in the culture, genomic sequencing showed that these chlorate-reducing bacteria dominated their small community. To understand the mechanism of the interaction, I used genetic deletion of pathway genes to measure phenotypes in co-cultures, modeled co-culture growth kinetics, and identified specific inhibitors of each metabolism [2]. This discovery greatly improves of our understanding of how perchlorate and chlorate are recycled.

Confirmation of the interaction using qPCR of different strain combinations and amplicon sequence variants from perchlorate-reducing communities. Mock figure from a paper in preparation.

[2] Barnum TP, Cheng Y, Hill KA, Lucas LN, Carlson HK, Coates JD. Identification of a parasitic symbiosis between respiratory metabolisms in the biogeochemical chlorine cycle. ISME J 14, 1194–1206 (2020).

Gene discovery using genomes and metagenomes

Chlorite (ClO2-) and hypochlorous acid (HOCl) are strong oxidants that react with various biological molecules. For example, our immune system generates hypochlorous acid for its efficiency in killing bacterial pathogens, and residual chlorite in wastewater treatment plants can affect bacteria and humans alike. Several bacterial genes have been found that respond to reactive chlorine stress, including a chlorite-degrading enzyme called chlorite dismutase (Cld). Unlike hypochlorous acid, however, chlorite is not known to be common, and the prevalence of this specific enzyme in hundreds of bacteria has left researchers perplexed.

I am approaching the problem from a comparative genomics perspective. What features of genomes are predictive of containing the Cld gene? What genes found near the Cld gene are involved in reactive chlorine stress production and consumption? In addition, I am mining metagenome databases to expand data scope and environmental context. The results will be the most comprehensive analysis of genes involved in reactive chlorine stress yet performed.

Phylogeny of Cld genes from metagenomes and genomes. Relative contribution of environments and taxa (class level) to each phylogenetic group is shown. Work in progress.


The most exciting new discoveries in microbiology will involve both experimentation and mining large sets of genomes and metagenomes. Using this approach, my research has identified new enzymes and interactions that participate in converting oxidized chlorine to chloride. These findings provide a clearer picture of how microorganisms drive the chlorine cycle and will support research in fundamental and applied environmental microbiology.


Data availability