Mutational analysis of DsbA and DsbB, two disulfide catalysts, to identify critical residues involved in their functions.

Abstract

Proper folding of many secreted proteins is reliant on the formation of disulfide bonds. Although the catalytic CXXC motifs of disulfide bond forming enzymes such as DsbA or DsbB have been extensively studied, we do not know what other critical residues are involved in their functions. I used a genetic approach to analyze the contribution of such residues to DsbA and DsbB functions. A M153R substitution of DsbA appears to affect the interactions of DsbA with DsbB, and Leu43, as well as His91 of DsbB, appear to be important for the catalytic activity of DsbB. The R133C and L146R substitutions affect the overexpression of <italic>dsbB</italic>, suggesting that such substitutions negatively affect the structural stability of DsbB. In addition to point mutants in DsbA and DsbB, I also recovered other types of mutations in the upstream and downstream regions of these genes. Insertion sequence elements integrated upstream and downstream of <italic>dsbA</italic> cause a defect in the expression of <italic> dsbA</italic>. Mutations in the untranslated leader sequence of <italic>dsbB </italic> might possibly affect <italic>dsbB</italic> expression. I also recovered a mutant that does not seem to be DsbA- or DsbB-linked, raising the exciting possibility of an unknown player in the Dsb system. In a separate project, I analyzed the role of RrmJ, a 23S rRNA methyltransferase, by overproducing proteins that would suppress the growth defect of an RrmJ null mutant. Surprisingly, I found that two GTPases, YhbZ and EngA, were able to rescue the growth and ribosome defects of the RrmJ null mutant. The overproduction of a GTPase Era, a close family member of EngA, fails to rescue the growth defect of an RrmJ null mutant, suggesting that the rescue is not due to a simple alteration of guanine nucleotide pools in the cell. The ability of the 50S ribosomal subunits isolated from an RrmJ null mutant overexpressing <italic> yhbZ</italic> or <italic>engA</italic> to be methylated by RrmJ in an <italic> in vitro</italic> methylation assay, suggests that such an overexpression could allow for the GTPases to bind and stabilize nbosomes, replacing the need for methylation at U2552. My results, together with observations by other groups working on these small GTPases, seem to suggest that these small GTPases are functionally linked to the ribosome.