Taken the above observations a complex regulation of the operon, with multiple promoters and transcripts containing different sets of genes, cannot be ruled out. Since we were particularly interested in rnr and smpB we have searched for promoters in the vicinity that could regulate the expression of this particular set of genes. Even though bioinformatics analysis indicated a putative promoter immediately upstream selleck kinase inhibitor of rnr, we could not detect any active promoter, either by primer extension analysis or by 5’ RACE mapping (data not shown). Upstream of rnr lays a small ORF that encodes a protein with homology to SecG, an auxiliary protein in the Sec-dependent protein

export pathway. A transcript containing secG and rnr was detected and was also mainly expressed under cold shock (Figure 2b). In fact, a putative promoter upstream this ORF was identified in silico, which could also drive rnr transcription (see Figure 2a). Therefore, primer extension

and RACE experiments were conducted to check this possibility. A single fragment was extended from a primer that hybridizes with the 5’-end of the secG mRNA (rnm014) Wnt inhibitor as shown in Figure 3a. The size of this fragment, as determined by comparison with the M13 phage sequence, shows that its 5’-end matches the transcription start site (+1) of the in silico predicted promoter (see Figure 3c). To confirm this result the 5’-end of the transcript was mapped by 5’ Non-specific serine/threonine protein kinase RACE following a protocol that makes use of the tobacco acid selleck inhibitor pyrophosphatase (TAP) enzyme [32]. This method allows distinguishing between 5’-ends of primary transcripts from those generated by cleavage/processing. A 5’ RACE product that was only obtained from the TAP-treated samples (Figure 3b, lane T+) indicates that it carries a 5’-triphosphate group characteristic of primary transcripts. Cloning and sequencing of this RACE product allowed us to

identify the +1 site at the same position as that identified by primer extension. These results clearly show that this promoter is active and drives the expression of secG. Considering the lack of a promoter upstream rnr and since a transcription terminator could neither be identified in this region, we believe that the secG promoter may also contribute to the rnr expression. Since our data indicate that rnr and smpB are co-transcribed, this promoter most likely directs smpB transcription as well. Nonetheless, we searched for alternative promoters of smpB. We started by analysing the 5’-end of the smpB transcript by primer extension using a primer specific for the smpB 5’-end region (rnm002 – see Figure 2a). As shown in Figure 4a, two different fragments were extended from this primer (fragment a and fragment b). Analysis of the sequence revealed that the 5’-ends of both fragments are located right before the overlapping region between rnr and smpB (Figure 4c).