Abstract
Photoferrotrophy, the light-induced oxidation of ferrous iron, is thought to have contributed to primary production within Earth’s early anoxic oceans yet is presumed to be of little modern environmental relevance. Here we use genome-resolved metagenomics and enrichment cultivation to explore the potential for photoferrotrophy in the anoxic water columns of globally abundant Boreal Shield lakes. We recovered four high-completeness and low-contamination draft genome bins assigned to the class Chlorobia (formerly phylum Chlorobi) from environmental metagenome data and enriched two novel sulfide-oxidizing species, also from the Chlorobia. The sequenced genomes of both enriched species, including the novel “Candidatus Chlorobium canadense”, encoded the cyc2 candidate gene marker for iron oxidation, suggesting the potential for photoferrotrophic growth. Surprisingly, one of the environmental genome bins encoded cyc2 and lacked sulfur oxidation gene pathways altogether. Despite the presence of cyc2 in the corresponding draft genome, we were unable to induce photoferrotrophy in “Ca. Chlorobium canadense”, suggesting that yet-unexplored mechanisms regulate expression of sulfide and ferrous iron oxidation gene systems, or that previously unrecognized functions for this outer membrane cytochrome exist. Doubling the known diversity of Chlorobia-associated cyc2 genes, metagenome data showed that putative photoferrotrophic populations occurred in one lake but that only sulfide-oxidizing populations were present in a neighboring lake, implying that strong ecological or geochemical controls govern the favourability of photoferrotrophy in aquatic environments. These results indicate that anoxygenic photoautotrophs in Boreal Shield lakes could have unexplored metabolic diversity that is controlled by ecological and biogeochemical drivers pertinent to understanding Earth’s early microbial communities.
Footnotes
The authors declare no competing financial interests.
One of the enriched members of the Chlorobium genus was provisionally named "Candidatus Chlorobium canadense". Despite encoding cyc2 (along with sulfide oxidation genes) in its genome, "Candidatus Chlorobium canadense" was not able to oxidize iron photoferrotrophically in lab testing. This implies that cyc2 might be tightly regulated this organism, which is partially supported by our metagenomic data showing the variable presence/absence of Chlorobia-affiliated cyc2 in neighbouring lakes. Alternative functions of cyc2 might also exist. The manuscript has been substantially revised to reflect these new data.