RpoN‐dependent transcription of rpoH?

Sir, Kallipolitis and Valentin-Hansen (1998, Mol Microbiol29: 1091–1099) recently reported that transcription of rpoH (which encodes the heat shock sigma factor in Escherichia coli ) is controlled by the cAMP–CRP/CytR complex. This adds to the known complexity of the regulation of rpoH transcription, which involves at least four promoters (P1, P3, P4 and P5) and two sigma factors, sigma-E (RpoE) and sigma-70 (RpoD). However, it seems likely that the situation is yet more complex, as rpoH is probably also transcribed by the core enzyme in association with yet another sigma factor, sigma-54 (RpoN or sigma-N), a sigma factor structurally unrelated to the sigma-70 family (Merrick, 1993, Mol Microbiol10: 903–909). While searching for promoters dependent on RpoN, I discovered an excellent match to the RpoN promoter consensus sequence (TGGCAC-N5-TTGCa/t; Merrick, 1993, Mol Microbiol10: 903–909; Buck and Cannon, 1989, Nucleic Acids Res17: 2597–2612) upstream of the E. coli rpoH coding region (Fig. 1). The putative RpoN binding site begins 10 bp downstream of the P5 start site (the most downstream of the known promoters), and the RpoN-dependent transcriptional start site (P6) can be predicted to occur 31 bases upstream of the ATG start codon. The sequence and position of the putative RpoN binding site is well conserved in the regions upstream of rpoH in Citrobacter freundii and Enterobacter cloacae (Nakahigashi et al., 1995, Nucleic Acids Res23: 4383–4390), adding credence to its likely biological significance (Fig. 2). . The proximal part of the rpoH promoter region. The proven transcriptional start sites from promoters P3 P4 and P5 and the predicted start site P6 are marked by vertical lines. The RpoN binding site is underlined. Numbering is set with +1 as the first base of the rpoH coding region. The start codon is indicated in bold. Sequence from Kallipolitis and Valentin-Hansen, 1998, Mol Microbiol29: 1091–1099. . Putative RpoN binding sites upstream of the rpoH coding regions in related organisms. Consensus from Buck and Cannon, 1989, Nucleic Acids Res17: 2597–2612, cited by Merrick, 1993, Mol Microbiol10: 903–909. Sequences from Nakahigashi et al., 1995, Nucleic Acids Res23: 4383–4390. As the regulation of RpoN-dependent genes is governed by activator proteins, it is difficult to predict under what circumstances the RpoN-dependent promoter might be active (Merrick, 1993, Mol Microbiol10: 903–909). However, for those with the interest and experimental expertise, it should be relatively straightforward to show experimentally that RpoN binds to the predicted site. If these predictions are confirmed, then a previously unsuspected link between two important regulatory networks, the RpoH regulon and the RpoN regulon, will be established. I thank Martin Buck for helpful discussion, and Gordon Dougan and the Wellcome Trust for sponsoring my Research Leave Fellowship.