It has been hypothesized that the dinR gene product of Bacillus subtilis acts as a repressor of the SOS regulon by binding to DNA sequences located upstream of SOS genes, including dinR and recA. Following activation as a result of DNA damage, RecA is believed to catalyse DinR-autocleavage, thus derepressing the SOS regulon. The present results support this hypothesis: a dinR insertion mutation caused a high, constitutive expression of both dinR and recA, which could not be further elevated by SOS-induction. In addition, gel-retardation assays demonstrated a direct interaction between the dinR gene product and the recA and dinR promoter regions. Epistatic interactions and gel-retardation assays demonstrated that the previously reported competence-specific expression of recA directly depended upon the gene product of comK, the competence transcription factor. These data demonstrate the existence of a direct regulatory link between the competence signal-transduction pathway and the SOS reguion.

In Bacillus subtilis, exposure to DNA damage and the development of natural competence lead to the induction of the SOS regulon. It has been hypothesized that the DinR protein is the cellular repressor of the B. subtilis SOS system due to its homology to the Escherichia coli LexA transcriptional repressor.

Recently, the DinR protein was established as the cellular repressor of the SOS response in the bacterium Bacillus subtilis. It is believed that DinR functions as the repressor by binding to a consensus sequence located in the promoter region of each SOS gene. The binding site for DinR is believed to be synonymous with the formerly identified Cheo box, a region of 12 bp displaying dyad symmetry (GAAC-N4-GTTC). Electrophoretic mobility shift assays revealed that highly purified DinR does bind to such sites located upstream of the dinA, dinB,dinC, and dinR genes. Furthermore, detailed mutational analysis of the B. subtilis recA operator indicates that some nucleotides are more important than others for maintaining efficient DinR binding. For example, nucleotide substitutions immediately 5′ and 3′ of the Cheo box as well as those in the N4 region appear to affect DinR binding. This data, combined with computational analyses of potential binding sites in other gram-positive organisms, yields a new consensus (DinR box) of 5′-CGAACRNRYGTTYC-3′. DNA footprint analysis of the B. subtilis dinR and recA DinR boxes revealed that the DinR box is centrally located within a DNA region of 31 bp that is protected from hydroxyl radical cleavage in the presence of DinR. Furthermore, while DinR is predominantly monomeric in solution, it apparently binds to the DinR box in a dimeric state.