Genomic and metagenomic survey of microbial carbonic anhydrase genes reveals novel clades, high diversity, and biome specificity

Carbonic anhydrase (CA) enzymes catalyze the interconversion of carbon dioxide and bicarbonate with an efficiency exceeded only by superoxide dismutase. CA enzymes have evolved convergently in phylogenetically distant organisms, forming eight structurally unrelated classes that share physiological functions involved in photosynthesis, respiration, pH homeostasis, CO2 transport, and carbonyl sulfide hydrolysis that play central roles in medicine and the environment. Here, we leverage the recent surge in publicly available genomes and metagenomes to re-examine our understanding of the abundance, diversity, and phylogenetic relationships of the three major CA classes in Bacteria/Archaea and microbial Eukaryotes (Fungi, algae). We recovered a total of 57 218 α-, β-, and γ-CA sequences from 24 184 metagenomes and genomes, including the first putative α-CA from an archaeal species. CA sequences formed 3859 protein clusters (1188 with three or more sequences). Sequences within a cluster were typically taxonomically conserved only at higher levels (i.e. Superkingdom, Phylum). When viewed within a phylogenetic framework, the majority of subclades for each CA class contained CAs representing multiple Superkingdoms, although numerous novel β-CA clades appear unique to Fungi. Queries of CA Hidden Markov models against all public metagenome and metatranscriptome datasets revealed that CA is a ubiquitous enzyme present in virtually all sampled environments. However, CA clusters that were taxonomically conserved also appeared more environment-specific, which may explain high CA diversity. This work represents an important contribution to our understanding of the evolution, diversity, and environmental distribution of an enzyme that is key to life and has broad environmental and industrial applications.