Two-component systems, composed of a sensor kinase and a response regulator, are a major tool used by microorganisms to adapt to environmental conditions. Typically, phosphorylation of the response regulator by the sensor kinase results in gene activation. In the case of B. pertussis, the response regulator BvgA is phosphorylated by the sensor kinase BvgS to yield a form (BvgA-P) that binds to the promoter regions of the virulence genes, activating their transcription during infection. These genes include adhesins, needed to adhere to the ciliated epithelial cells within the upper respiratory tract, and toxins, which cause major symptoms of the whooping cough disease. Adhesins and toxins are particularly important because they are typical components of the acellular pertussis vaccine. In B. pertussis the serologically distinct fimbriae 2 and fimbriae 3 are composed primarily of the major subunit proteins Fim2 and Fim3. Previously its been shown that the promoters for these genes behave differently from each other both in vivo and in vitro. In vivo Pfim2 is significantly stronger than Pfim3, even though predictions based on the DNA sequences of BvgA binding motifs and core promoter elements would indicate the opposite. In vitro Pfim3 demonstrates robust BvgAP-dependent transcriptional activation, while none is seen with Pfim2. With our collaborators Drs. Scott Stibitz and Qing Chen (CBER, FDA), we have investigated this apparent contradiction further. By swapping sequence elements, a number of hybrid promoters were created and assayed for their strength in vivo. These hybrid promoters demonstrated that while Pfim3 promoter elements upstream of the +1 transcriptional start site do indeed direct Bvg-activated transcription more efficiently than those of Pfim2, the overall promoter strength of Pfim3 in vivo is reduced due to sequences downstream of +1 that inhibit transcription more than 250-fold. This element, the DRE (downstream repressive element), was mapped to the 15 bp immediately downstream of the Pfim3 +1. Placing the DRE in different promoter contexts indicated that its activity was not specific to fim promoters, or even to Bvg-regulated promoters. We also found that with Pfim3 and nonphosphorylated BvgA, the presence of the DRE in vitro results in the formation of novel complexes, which are not able to produce full length RNA. Further investigation of DRE function in vivo and in vitro is expected to elucidate both the complex regulation of fimbrial gene expression and a novel mechanism of regulation during promoter clearance.
