Bacterial small regulatory RNAs: how to reconcile target specificity and pleiotropic action – SalsARN

Small non-coding RNAs have emerged as one of the most important classes of regulatory molecules in all living organisms. In bacteria, small RNAs (sRNA) exert their regulatory function by base-pairing with sequences near the translation initiation codon of target messenger RNAs. Most often, formation of the duplex obstructs ribosome access to the mRNA, resulting in translation inhibition and mRNA destabilization. Chaperon protein Hfq participates in the regulatory mechanism by stabilizing the sRNA-mRNA interaction. This stabilization is critical as the pairing region is generally short and often punctuated by mispairs and/or bulging bases. Lack of full complementarity might be important to allow a single sRNA to act pleiotropically on multiple non-identical targets. Compromising between regulatory efficacy and pleiotropic action is a key aspect of sRNA function in bacteria. The present proposal is aimed at dissecting the determinants of these properties in representative sRNAs. Our main strategy will involve the analysis of mutations affecting regulation efficacy or target specificity, or both, in three different sRNAs. MicA and RybB sRNAs accumulate during the sigmaE envelope stress response and down-regulate the lelves of several outer membrane proteins (OMPs). RybB has multiple mRNA targets including ompC, ompD, and ompW mRNAs. MicA was until recently thought to act solely on ompA. However, our work identified lamB (maltoporine) mRNA as an additional MicA target (Bossi & Figueroa-Bossi, submitted). A third sRNA, identified in our laboratory and named MicZ specifically down-regulates a poorly characterized OMP, which we named ompZ. Unlike MicA and RybBs, which are made under stress conditions, MicZ is synthesized constitutively and is responsible for the complete silencing of the ompZ mRNA during normal growth. Isolation of mutations in micA, rybB and micZ, and in some of their target genes (lamB, ompC/ompD and ompZ) will be made possible by a procedure developed in our laboratory that combines random PCR mutagenesis to l Red-mediated recombination. Mutants will be identified in the chromosome by a sensitive lac-based chromogenic screen. The procedure has proven successful in preliminary tests that have already produced interesting alleles of all three sRNA genes. In being developed entirely in the bacterial chromosome our study constitutes a novelty in the sRNA field, normally heavily tributary to the use with multi-copy plasmids. We plan to generate a large collection of mutants that will then be subjected to in-depth molecular characterization. This will include analyzing the effects of the mutations on the levels of sRNAs and mRNA targets in vivo, on the sRNA-mRNA pairing in vitro (by gel retardation and foothprinting assays), and, for some of the alleles, on Hfq binding. In sRNA with multiple targets, special attention will be given to alleles differentially affecting separate targets. Overall, this study is expected to produce a fine map of structure/function relations in the three sRNAs and provide insight on the determinants of target selectivity in MicA and RybB. In parallel with the above work, part the project will be devoted to the characterization of some Salmonella loci that show Hfq-dependent regulation (Figueroa-Bossi et al., Mol Microbiol 62: 838-852, 2006). One of these loci is the malB operon, which comprises the malK, lamB and malM genes. Our recent data show that, besides intervening in MicA-mediated down-regulation of lamB during the sigmaE response, Hfq acts in a mechanism that causes lamB and malM genes to be differentially expressed during normal growth. We postulate that this mechanism involves Hfq binding to the lamB-malM spacer, mediated by a yet unidentified small RNA. We will actively try to identify such presumptive sRNA. An Hfq involvement is also observed in the Salmonella ompD gene. Here too, the effects can be separated from the RybB-mediated down-regulation that occurs during the sigmaE response (also Hfq-dependent). Here too, the participation of an sRNA can be inferred and will be investigated.