We used fabXL::gusA fusions to measure the relative expression of the fabXL genes in different conditions. The fabXL genes were induced approximately two-fold in TY, or VMM with 1% hydrolyzed casein, compared with the expression in VMM with mannitol (data not shown). The induction of ropB expression in peptide-containing media requires the ChvG/ChvI two-component system (Foreman et al., 2010). We used a chvG mutant in R. leguminosarum VF39SM (we were unable to construct a chvG

mutant in 3841) to determine whether ChvG AZD9291 order is involved in regulating the fabXL genes. The putative promoter sequences used to construct the fabXL::gusA transcriptional fusions are identical to those found in VF39SM, and expression of the fusions in VF39SM was similar to the expression in 3841. VF39SM and the chvG mutant carrying the fabXL::gusA fusions were grown on solid VMM with calcium chloride or VMM with tryptone and calcium chloride, as described previously (Foreman Everolimus et al., 2010). A roughly fourfold induction in fabXL gene expression was observed in wild-type grown in the presence of tryptone. In contrast, there was no induction of the fabXL fusions in the chvG mutant, suggesting that ChvG is a positive regulator of the fabXL genes in R. leguminosarum (Table 4). As well, the fabXL fusions were generally expressed

at a lower level in the chvG mutant than in wild-type (Table 4), suggesting ChvG is also important for gene expression under non-inducing conditions. Given that the ChvG–ChvI two-component system is thought to be a global regulatory system (Chen et al., 2009), it is possible that the nonspecific decrease in transcription of the fabXL genes is an indirect effect resulting from a significantly affected cell physiology. Our results differ from those found for the chvG homolog, bvrS, in Brucella abortus, where there was no change in the transcript abundance for acpXL or lpxXL in a bvrS Sitaxentan mutant (Manterola et al., 2005). This observation suggests that although

this two component system (TCS) is a highly conserved global regulator, its role in regulation of cell envelope components may have species-specific complexities. In conclusion, our results demonstrate that the fabXL and ropB genes are part of a cell envelope network required to maintain outer membrane stability. These cell envelope components are strongly conserved among the order Rhizobiales; therefore, continued investigation into the mechanisms and proteins controlling the interactions between ropB and the fabXL genes will provide insight into the mechanisms that regulate gene expression during cell envelope development in free-living and host-associated environments. Lastly, the biological significance of the down-regulation of ropB expression following mutation of LPS structural genes remains an intriguing question meriting further investigation. We gratefully acknowledge Dr. R. W. Carlson for supplying us with the acpXL mutant strain, Rlv22, and Dr. K.