A closed genome of Streptococcus agalactiae ST347/II reveals a composite integrative element coupling nisin P biosynthesis to erm(B) and tet(O) in Dominican group B Streptococcus

Abstract

Clindamycin is the alternative for intrapartum prophylaxis against group B Streptococcus in penicillin-allergic women, and whether it works depends on which lineages carry the resistance and on the mobile element that moves it. No Dominican group B Streptococcus genome has been closed. A Streptococcus agalactiae isolate from residual diagnostic material was sequenced twice on Oxford Nanopore, annotated, and screened against 83 genomes from a published Dominican maternal colonisation survey. The assembly is one circular chromosome of 2,066,859 bp with all seven rRNA operons resolved: ST347/CC19, serotype II, carrying erm(B), tet(O), mef(J) and msr(I). Two integrative conjugative elements were delimited: one of ~34 kb related to Tn5252 carrying erm(B) and tet(O), another of ~35 kb at a tRNA-Lys gene carrying mef(J) and msr(I). An eleven-gene class I lanthipeptide cluster lies 96 bp from the first element, on the same strand and with no annotated feature between them; its precursor encodes a 31-residue core identical to nisin P. The complete operon was found in 12 of the 83 genomes, 13 of 84 (15.5%) with the isolate. tet(O) was present in all 12 carriers and none of the 71 others (Fisher p = 1.03 x 10^-14), erm(B) in 12 of 12 against 3 of 71 (p = 4.70 x 10^-12). All eleven ST1 genomes lack both the operon and the nisin resistance protein; all 15 ST1010/IV genomes carry aac(6')-Ie/aph(2'')-Ia. Clindamycin resistance is therefore tightly associated with a lantibiotic-encoding element and contiguous with it in the one closed genome; in the drafts the relationship is statistical, and co-occurrence is not linkage. The ST1 absence predicts lineage-wide susceptibility, testable by deferred antagonism.