Regulation of the phosphate starvation response in Corynebacterium glutamicum by the PhoRS two-component system

In bacteria, the recognition of environmental changes and the necessary adaptations to these changes is in many cases accomplished by two-component signal transduction systems. They are composed of a usually membrane-bound sensor kinase and a response regulator. In response to specific environmental signals, the sensor kinase controls the phosphorylation state of the response regulator and thus its activity. With few exceptions, response regulators function as transcriptional regulators. Within the 3.3 Mb-genome of the gram-positive bacterium Corynebacterium glutamicum genes for 13 two-component systems were identified. By directed mutagenesis, a set of 12 strains was constructed each lacking the genes for one sensor kinase and its response regulator. In one case, only the sensor kinase, but not the response regulator could be deleted. In this work it was tested, whether one of the two-component systems of C. glutamicum is involved in the phosphate starvation response. To this end, the mutants mentioned above were cultivated under phosphate excess and phosphate limitation. One of the mutants grew poorer than the wild type under phosphate limitation, but not under phosphate excess. The growth defect could be abolished by expression of the missing genes from a plasmid. These results indicate a function of the deleted two-component system in the phosphate starvation response and therefore the corresponding genes were named phoS (phosphate sensor kinase) and phoR (phosphate response regulator). With the aim to identify the target genes of response regulator PhoR, DNA microarray analyses were performed. In a first set of experiments, the transcriptom of the $\Delta$phoRS mutant before and after a shift from phosphate excess to phosphate starvation was compared. Genes known to be induced by phosphate starvation in the wild type were not induced in the mutant, explaining the impaired growth of the $\Delta$phoRS mutant under phosphate limitation. The pstSCAB genes encoding a high-affinity ABC transporter for phosphate were still partially induced in the $\Delta$phoRS mutant under phosphate limitation, indicating the existence of an additional, currently unknown regulation of the pst operon. In a second set of experiments, the transcriptom of the $\Delta$phoRS mutant was compared to that of the wild type. Under phosphate excess, 27 genes showed a lower and 15 genes a higher mRNA level in the mutant compared to the wild type. Under phosphate-limited conditions, all of the known P$_{i}$ starvation-inducible genes (pstSCAB, ugpAEBC, glpQ, phoH, nucH and ushA) showed lower mRNA levels in the deletion mutant, sustaining the previous DNA microarray data. Interestingly, the mRNA level of the pitA gene encoding a low-affinity P$_{i}$ transporter decreased after a shift from P$_{i}$ excess to P$_{i}$ limitation in the wild type, but not in the $\Delta$phoRS mutant. By primer extension analysis of pstS, ugpA and phoR, the DNA microarray data were confirmed and the transcriptional start sites of these genes were determined. In summary, the data indicated that the PhoS-PhoR two-component system is responsible for the activation of many phosphate starvation genes, but also for repression of the pitA gene. In order to demonstrate the phosphorylation reactions characteristic for two-component signal transduction systems, the sensor kinase PhoS and the response regulator PhoR were both modified with a carboxyterminal histidine tag, overproduced in Escherichia coli and subsequently purified by affinity chromatography. Solubilization and purification of the integral membrane protein PhoS were performed with the detergent N,Ndimethyldodecylamine-N-oxide (LDAO). Both the autokinase activity of solubilized PhoS and the phosphoryl transfer from phosphorylated PhoS to PhoR were demonstrated with the isolated proteins. The binding of the response regulator PhoR to the promoters of pstSCAB, ugpAEBC, phoRS and pitA was analysed by gel shift analyses. With the entire PhoR protein it was not possible to show binding, irrespective of the conditions applied. However, a derivative of PhoR in which the aminoterminal 125 amino acid residues were deleted and replaced by a histidine tag did bind to the four promoter regions mentioned above, but not to the negative control promoters clpP1P2 and clpC. Additionally, the PhoR binding site within the pst promoter was analysed by DNase I footprinting, which indicated a protected region localized 170 to 205 bps upstream of pstS transcriptional start site