The evolution and diversity of noncanonical microbial nitrogen metabolisms

Abstract

Microbial nitrogen cycling underpins Earth’s most basic ecology. The environmental and commercial significance of nitrogen biogeochemistry has prompted substantial study of the modern nitrogen cycle, but much remains unknown about the evolution of nitrogen biochemistry through deep time. This thesis explores the evolution and nature of divergent or noncanonical nitrogen metabolisms. The first section of this work explores divergent enzymes of denitrification. I report a novel cytochrome-type nitrite reductase (nirS) structure within members of Phylum Chloroflexi, which includes a cytochrome superfamily domain in addition to functional domains conserved in Proteobacterial nirS. Phylogenetic domain mapping reveals that this gene resulted from a chimeric domain fusion in ancestral Chloroflexi. I also identify an underreported variant of nitric oxide reductase (eNOR) within Chloroflexi MAGs, and provide a detailed phylogeny of this enzyme variant, revealing much broader diversity than previously reported. Next, I explore the evolution of microbial cyanide metabolism. Cyanide-degrading enzymes have been extensively studied for bioremediation and biotechnology, but the extant diversity of such enzymes remains underexplored. Additionally, while cyanide is hypothesized to play a central role in prebiotic chemistry, few biological data constrain the age and emergence of cyanide metabolisms. I provide a comprehensive analysis of the distribution and evolution of the Class I nitrilases, a subfamily specialized for hydrogen cyanide (HCN) reduction. Gene trees reveal that cyanide-reducing nitrilases originated in bacteria and were transferred into eukaryotes, refuting earlier eukaryotic origin hypotheses. Molecular clock analyses indicate that this enzyme subfamily shares an ancestor that emerged 1-2 billion years ago in the Paleo- to Mesoproterozic. I also analyze the constrained prokaryotic distributions of other nonhomologous nitrile-reducing enzymes, thiocyanate hydrolases and nitrile hydratases. Finally, I detail the evolution of promiscuous nitrogen metabolism in nitrogenases. Though highly specialized for dinitrogen reduction, molybdenum and vanadium nitrogenases have been shown to reduce off-target nitrogenous substrates, including HCN. Using ancestral sequence reconstruction, I compare the sequence space of predicted ancestral and extant nitrogenase substrate channels. The results indicate that the predicted highest-likelihood ancestral states for key residues are not represented in extant sequence space, and the physicochemical types of these high-likelihood residue combinations are only rarely represented in divergent nitrogenases. These data suggest that ancestral nitrogenases likely had alternative substrate channel compositions, possibly reflecting selection in early Earth environments. Together, these data suggest that the diversity and age of microbial nitrogen metabolisms are currently underestimated, and furtherstudy of these pathways should shed light on large-scale patterns of microbial evolution and ecology.