Eight Unexpected Selenoprotein Families in ABC transport, in Organometallic Biochemistry in Clostridium difficile and other anaerobes, and in Methylmercury Biosynthesis

Abstract

A novel protein family related to mercury resistance protein MerB, which cleaves Hg-C bonds of organomercurial compounds, is a newly recognized selenoprotein, typically seen truncated in sequence databases at CU (cysteine-selenocysteine) dipeptide sites fifty residues before the true C-terminus. Inspection shows this protein occurs in a nine-gene neighborhood conserved in more than fifty bacterial species, taxonomically diverse but exclusively anaerobic, including spirochetes, deltaproteobacteria, and Gram-positive spore-formers Clostridium difficile and C. botulinum. Three included families are novel selenoproteins in most instances, including two ABC transporter subunits, one a substrate-binding protein with another CU motif, the other a permease subunit with selenocysteine at the substrate-gating position. Phylogenetic profiling shows a strong pattern of co-occurrence with Stickland metabolism selenoproteins, but an even closer link to a group of 8Fe-9S cofactor-type double-cubane proteins. These 8Fe-9S enzymes vary in count and in genome location but frequently sit next to the nine-gene locus. We have named the locus SAO, because of the Selenocysteine-Assisted Organometallic (SAO) biochemistry implied by an uptake ABC transporter with apparent metal-binding selenocysteines, complementary metal efflux pump SaoE, the MerB-like cytosolic enzyme now called SaoL, and comparative genomics signatures suggesting energy metabolism rather than metal resistance. Hypothesizing cycles of formation and dismutation of organometallic compounds involved in fermentative metabolism, we examined methylmercury formation proteins, and discovered most HgcA proteins are selenoproteins as well, with a CU motif N-terminal to the previously predicted start. Seeking additional rare and overlooked selenoproteins, tricky because of their rarity, could help reveal more candidate cryptic biochemical processes.