<p>  </p> <p>PhoQP is a two-component system that regulates the transcription of ~5% of the genes of <em>Salmonella enterica</em>. The membrane-bound PhoQ protein is phosphorylated in response to low extracellular Mg²⁺ concentration, acid pH, and a number of antimicrobial peptides. The inorganic phosphate bound to PhoQ is transferred to PhoP, which according to the classical model, acts as a typical transcriptional activator of its target genes. However, Will et al. (doi.org/10.1038/ncomms6270) proposed an alternate “counter-silencing” model, according to which genes in the PhoP regulon that were acquired by <em>Salmonella</em> via horizontal transfer are repressed by the generalized DNA-binding protein H-NS at high [Mg²⁺] and are induced at low [Mg²⁺] because the phosphorylated PhoP displaces the H-NS from the promoters and lifts repression. We evaluated this model by examining the transcriptional regulation of the <em>mgtCBRU-cigR </em>operon, which encodes the virulence protein MgtC and the Mg²⁺ transport protein MgtB and is in the SPI-3 pathogenesis island that has been acquired by <em>Salmonella</em> via horizontal transfer. Our main finding was that in the non-pathogenic strain of <em>S</em>. Typhimurium (LT2), induction of the <em>mgtCBRU-cigR</em> operon by Mg²⁺ limitation requires a functional PhoP protein, regardless of the presence or absence of H-NS. Interestingly, the pathogenic strain of <em>S</em>. Typhimurium (ATCC 14028s) revealed PhoP-independent transcription in the absence of H-NS, but only under inducing conditions. Thus, our results do not support the counter-silencing model and are consistent with the canonical view that PhoP is needed as a transcriptional activator of genes in the PhoP regulon.</p>

No description available.

Greater understanding of how we can use protein simulations and statistical characteristics of biomolecular interfaces as proxies for biological function will make manifest major advances in protein engineering. Here we show how to use calculated change in binding affinity and coevolutionary scores to predict the functional effect of mutations in the interface between a Histidine Kinase and a Response Regulator. These proteins participate in the Two-Component Regulatory system, a system for intracellular signalling found in bacteria. We find that both scores work as proxies for functional mutants and demonstrate a ~30 fold improvement in initial positive predictive value compared with choosing randomly from a sequence space of 160 000 variants in the top 20 mutants. We also demonstrate qualitative differences in the predictions of the two scores, primarily a tendency for the coevolutionary score to miss out on one class of functional mutants with enriched frequency of the amino acid threonine in one position.