Associations between Bacillus subtilis B regulators in cell extracts

Abstract

The general stress regulon of Bacillus subtilis is induced by the activation of the sigma(B) transcription factor. Activation of sigma(B) occurs as a consequence of the dephosphorylation of its positive regulator RsbV by one of two phosphatases that respond to either physical or nutritional stress. The physical stress phosphatase (RsbU) requires a second protein (RsbT) for activity. Stress is thought to initiate a process that triggers the release of RsbT from a large inhibitory complex composed of multiple copies of two protein species, RsbR (and/or its paralogues) and RsbS. The stress-derived signal driving RsbT release is unknown, but it fails to develop in B. subtilis lacking either ribosome protein L11 or the ribosome-associated protein Obg. RsbR, RsbS, RsbT, Obg and ribosomes elute in common high-molecular-mass fractions during gel-filtration chromatography of crude B. subtilis extracts. This paper reports the investigation of the basis of this coelution by the examining of associations between these proteins in extracts prepared from wild-type and mutant B. subtilis, and Escherichia coli engineered to express RsbR, RsbS and RsbT. Large RsbR/RsbS complexes, distinct from ribosomes, were detected in extracts of both B. subtilis and E. coli. In E. coli, high-molecular-mass forms of RsbS were less abundant when RsbR was absent, but in B. subtilis, only when both RsbR and its principal paralogues were missing from the extract was this form less abundant. This finding is consistent with the notion that the RsbR paralogues, present in B. subtilis but not E. coli, can substitute for RsbR in such complexes. RsbT was not bound to RsbR/RsbS in any extract that was examined, including one prepared from a B. subtilis strain with an RsbS variant (RsbS59SA) that is believed to continuously associate with RsbT. The high-molecular-mass forms of RsbT were found to be Triton-sensitive and independent of any other B. subtilis protein for their formation. These probably represent RsbT aggregates. The data suggest that the contribution of ribosomes/Obg to sigma(B) activation does not involve formation of a stable association between these proteins and the Rsb complex. In addition, the binding of RsbT to RsbS/RsbR appears to be more labile than the binding between the previously analysed Rsb proteins which form inhibitory complexes. This, and the apparent proclivity of RsbT to aggregate, suggests an inherent instability in RsbT which may play a role in its regulation.