SummaryIn bacteria, Hfq is a core RNA chaperone that catalyzes the interaction of mRNAs with regulatory small RNAs (sRNAs). To determine in vivo RNA sequence requirements for Hfq interactions, and to study riboregulation in a bacterial pathogen, Hfq was UV crosslinked to RNAs in enterohemorrhagic Escherichia coli (EHEC). Hfq bound repeated trinucleotide motifs of A-R-N (A-A/G-any nucleotide) often associated with the Shine-Dalgarno translation initiation sequence in mRNAs. These motifs overlapped or were adjacent to the mRNA sequences bound by sRNAs. In consequence, sRNA-mRNA duplex formation will displace Hfq, promoting recycling. Fifty-five sRNAs were identified within bacteriophage-derived regions of the EHEC genome, including some of the most abundant Hfq-interacting sRNAs. One of these (AgvB) antagonized the function of the core genome regulatory sRNA, GcvB, by mimicking its mRNA substrate sequence. This bacteriophage-encoded “anti-sRNA” provided EHEC with a growth advantage specifically in bovine rectal mucus recovered from its primary colonization site in cattle.

The can (previously yadF) gene of Escherichia coli encodes a ␤-class carbonic anhydrase (CA), an enzyme which interconverts CO 2 and bicarbonate.Various essential metabolic processes require either CO 2 or bicarbonate and, although carbon dioxide and bicarbonate spontaneously equilibrate in solution, the low concentration of CO 2 in air and its rapid diffusion from the cell mean that insufficient bicarbonate is spontaneously made in vivo to meet metabolic and biosynthetic needs. We calculate that demand for bicarbonate is 10 3 -to 10 4 -fold greater than would be provided by uncatalyzed intracellular hydration and that enzymatic conversion of CO 2 to bicarbonate is therefore necessary for growth. We find that can expression is ordinarily required for growth in air. It is dispensable if the atmospheric partial pressure of CO 2 is high or during anaerobic growth in a closed vessel at low pH, where copious CO 2 is generated endogenously. CynT, the single E. coli Can paralog, can, when induced with azide, replace Can; also, the ␥-CA from Methanosarcina thermophila can at least partially replace it. Expression studies showed that can transcription does not appear to respond to carbon dioxide concentration or to be autoregulated. However, can expression is influenced by growth rate and the growth cycle; it is expressed best in slow-growing cultures and at higher culture densities. Expression can vary over a 10-fold range during the growth cycle and is also elevated during starvation or heat stress.

RNA sequencing studies have identified hundreds of non‐coding RNAs in bacteria, including regulatory small RNA (sRNA). However, our understanding of sRNA function has lagged behind their identification due to a lack of tools for the high‐throughput analysis of RNA–RNA interactions in bacteria. Here we demonstrate that in vivo sRNA–mRNA duplexes can be recovered using UV‐crosslinking, ligation and sequencing of hybrids (CLASH). Many sRNAs recruit the endoribonuclease, RNase E, to facilitate processing of mRNAs. We were able to recover base‐paired sRNA–mRNA duplexes in association with RNase E, allowing proximity‐dependent ligation and sequencing of cognate sRNA–mRNA pairs as chimeric reads. We verified that this approach captures bona fide sRNA–mRNA interactions. Clustering analyses identified novel sRNA seed regions and sets of potentially co‐regulated target mRNAs. We identified multiple mRNA targets for the pathotype‐specific sRNA Esr41, which was shown to regulate colicin sensitivity and iron transport in E. coli. Numerous sRNA interactions were also identified with non‐coding RNAs, including sRNAs and tRNAs, demonstrating the high complexity of the sRNA interactome.