Salmonella typhimurium encounters a variety of acid stress situations during growth in host and nonhost environments. The organism can survive potentially lethal acid conditions (pH <4) if it is first able to adapt to mild or more moderate acid levels. The molecular events that occur during this adaptive process are collectively referred to as the acid tolerance response and vary depending on whether the cells are in log- or stationary- phase growth. The acid tolerance response of logarithmically growing cells includes the participation of an alternate sigma factor, σ(S) (RpoS), commonly associated with stationary-phase physiology. Of 51 acid shock proteins (ASPs) induced during shifts to pH 4.4, 8 are clearly dependent on σ(S) for production (I. S. Lee, J. Lin, H. K. Hall, B. Bearson, and J. W. Foster, Mol. Microbiol. 17:155-167, 1995). The acid shock induction of these proteins appears to be the result of an acid shock-induced increase in the level of σ(S) itself. We have discovered that one component of a potential signal transduction system responsible for inducing rpoS expression is the product of the mouse virulence gene mviA+. MviA exhibits extensive homology to the regulatory components of certain two-component signal transduction systems (W. H. Benjamin, Jr., and P. D. Hall, abstr. B-67, p. 38, in Abstracts of the 93rd General Meeting of the American Society for Microbiology 1993, 1993). Mutations in mviA (mviA::Km) caused the overproduction of σ(S) and σ(S)-dependent ASPs in logarithmically growing cells, as well as increases in tolerances to acid, heat, osmolarity, and oxidative stresses and significant decreases in growth rate and colony size. Mutations in rpoS suppressed the mviA::Km-associated defects in growth rate, colony size, ASP production, and stress tolerance, suggesting that the effects of MviA on cell physiology occur via its control of σ(S) levels. Western blot (immunoblot) analyses of σ(S) produced from natural or arabinose-regulated promoters revealed that acid shock and MviA posttranscriptionally regulate σ(S) levels. Turnover experiments suggest that MviA regulates the stability of σ(S) protein rather than the translation of rpoS message. We propose a model in which MviA or its unknown signal transduction partner senses some consequence of acid shock, and probably other stresses, and signals the release of σ(S) from proteolysis. The increased concentration of σ(S) drives the elevated expression of the σ(S)-dependent ASPs, resulting in an increase in stress tolerance. The avirulent nature of mviA insertion mutants, therefore, appears to result from inappropriate σ(S)-dependent gene expression during pathogenesis.