Capsule gene (cps) expression, which normally occurs at low levels in Escherichia coli lon ؉ cells, increased 38-fold in lon ؉ cells carrying a Tn10::⌬kan insertion mapping to 24 min on the E. coli chromosome. Null mutations in rcsA, rcsB, or rcsC abolished the effect of the Tn10::⌬kan insertion. Sequencing of both sides of the Tn10::⌬kan insertion localized the insertion to the previously reported mdoH gene, which encodes a protein involved in biosynthesis of membrane-derived oligosaccharides (MDOs). A model suggesting that the periplasmic levels of MDOs act to signal RcsC to activate cps expression is proposed.Regulation of colanic acid capsular polysaccharide gene (cps) expression in Escherichia coli is multilayered, as evidenced by the numerous direct and indirect regulatory mechanisms that have been identified to date (for reviews, see references 6 and 8). The current model proposes two pathways for activating cps expression (6). One pathway involves a twocomponent regulatory system (24). RcsC (regulator of capsule synthesis), which has been described as a membrane-bound sensor protein based on homology with the sensor component of two-component sensor regulator pairs (24), appears to be activated by environmental stimuli, such as desiccation (18) or osmotic shock (22). Presumably, activated RcsC either directly or indirectly modifies RcsB, the proposed effector of the twocomponent system, which in turn activates cps expression (6, 7). In the alternate pathway, the other positive effector of cps expression, RcsA, presumably forms a complex with RcsB, resulting in the activation of cps expression (2, 6, 25). RcsA is highly unstable and appears to be degraded in a Lon-dependent fashion (26; for a review, see reference 6). In lon ϩ cells, RcsA levels are low, and these cells produce little colanic acid (26; for a review, see reference 6). Conversely, in ⌬lon cells, RcsA levels are high, leading to increased colanic acid production and mucoid colonies (26; for a review, see reference 6). In this model, both pathways require RcsB for high-level expression of cps (6).Current evidence suggests that additional regulators of cps expression exist: mutations in hns (6 min) (21, 29), capS (22.5 min) (14), opsX (62 min) (30), or capT (unknown location) (14) lead to an increase in colanic acid production. Furthermore, mutations within the rfa locus (82 min) both alter lipopolysaccharide structure and synthesis and increase colanic acid production (19). Given the complexity of the cps system, identification of additional cps regulators seems probable. (SG20780 [2]) strains carrying the zce-23::⌬kan insertion were assayed for ␤-galactosidase activity (Table 1). A 38-fold increase in the level of cps expression was observed in either Luria-Bertani (LB) or minimal (M63 salts, 0.4% glucose, 0.1% Casamino Acids) medium with the introduction of the zce-23:: ⌬kan insertion into a lon ϩ strain. The zce-23::⌬kan insertion had no effect on cps expression in ⌬lon cells.The increase in cpsB10::lacZ expression is abolished in lon ؉ s...

Analysis of the nucleotide sequence downstream from the Xanthomonas oryzae pv. oryzae recA gene reveals two orfs designated orfX and recX. The former has the potential to code for a 5.6 kDa protein of unknown function while the latter encodes for a putative 14.6 kDa protein with homology to RecX from various bacteria. Northern blot analysis and RT-PCR results show that recA-orfX-recX are co-regulated and arranged in an operon. A recX mutant was constructed. The mutant has no obvious growth defects or stress response defects, except that it cannot support high-level expression of recA from an expression vector. Introduction of the plasmid containing recA into the recX mutant resulted in reduced transformation efficiency and all transformants tested had mutations with reduced RecA levels. Moreover, the recX mutant has reduced basal levels of RecA. This has not been observed in other bacteria. When inactivated recX was complemented in trans, both changes were reversed. recX mutation has no effect on the regulation of the recA promoter, suggesting that its effect on the RecA level could be post-transcriptional.

A Xoo recA insertion inactivation mutant was constructed. The mutant, lacking RecA, showed increased sensitivity towards mutagen killing. This phenotype could be complemented by a cloned, functional recA. Unlike other bacteria, both the recA mutant and the parental strain had similar level of resistance to H2O2 killing and peroxide-induced mutagenesis.