An Upstream Sequence Element Required for NucC-Dependent Expression of the Serratia marcescens Extracellular Nuclease

Abstract

The Serratia marcescens extracellular nuclease gene, nucA, is positively regulated by the product of the nucC gene. In this study, the upstream region required for NucC-dependent nuclease expression was defined by using fusions to the gene encoding chloramphenicol acetyltransferase (cat). This sequence includes an element of hyphenated dyad symmetry identified previously as the binding site for the P2 Ogr family of activators. Footprint analysis confirmed that members of this family of activator proteins bind to this site, protecting a region between −76 and −59 relative to the start of transcription. The activator binding site in the nucA promoter lies one turn of the helix upstream from the corresponding sites in the P2 and P4 late promoters. The effects of deletions between the downstream end of the activator binding site and the putative −35 region are consistent with a strict helical phasing requirement for activation.

Expression of the Serratia marcescens extracellular nuclease, NucA, exhibits both growth phase dependence and an SOS-mediated response (3, 4). Previous studies have demonstrated that maximal expression of nucA also requires the product of the nucC gene (15). NucC is homologous to a family of small zinc-binding transcription factors encoded by the P2-related phages and their satellites. The nucC gene, although chromosomally located, lies in a transcription unit that appears to be part of a cryptic prophage found in S. marcescens (10, 15).

All members of the P2 Ogr family of proteins tested to date can substitute for each other to activate transcription at the same positively regulated promoters; this includes not only Ogr but 186 B, PSP3 Pag, P4 Delta, and φR73 Delta as well as NucC (13, 15, 17). The P2 and P4 late promoters share a conserved sequence element of interrupted dyad symmetry, TGT-N₁₂-ACA, centered at −57 upstream of the transcription start (6, 8, 9). Deletion and mutational analyses of the P2 P_F and P4 P_sid promoters (1, 7, 12, 22) indicated a key role for this sequence element in positive control of transcription. Activator binding to a region spanning these nucleotides was confirmed by DNase I footprinting analysis (16, 17). Inspection of sequences upstream of the nucA gene revealed several potential upstream binding sites that resembled this conserved sequence element, all of which were located farther upstream than the sites in the P2 and P4 late promoters. In order to localize the site of NucC action, we constructed a series of promoter derivatives fused to the cat gene and assayed them for NucC-dependent transcription in Escherichia coli.

Previous studies had identified the nucA transcription start site, as well as the presence of a LexA binding site overlapping +1 (5). Inclusion of 120 bp of sequence upstream of the transcription start site was sufficient to confer enhanced levels of nuclease expression, indicating that the NucC binding site was within this region (15). A 140-bp nucA promoter fragment, from −125 to +16, was obtained by PCR from plasmid pNuc2-LacZ (3), using the synthetic oligonucleotides Pnuc5 (5′ATCGATGTCTGTTGGACCCGT3′) and Pnuc3 (5′GTCGACATTTACAGTGAATTAAACT3′). This introduced the underlined ClaI and SalI sites at the 5′ and 3′ ends, respectively, and destroyed the LexA-binding site. This fragment was ligated into the TA vector pCR2.1 (Invitrogen) to create pTG198, and the sequence was verified. The fragment was then excised by digestion with ClaI and SalI and ligated with pSL100 (19), a derivative of the promoterless chloramphenicol acetyltransferase (CAT) vector pKK232-8 (4), which had been cleaved with the same two enzymes. The resulting plasmid, pJC1, exhibited activator-dependent expression from the nucA promoter (Fig. 1). Basal (activator-independent) promoter activity from this construct was below the limits of detection (<0.005 U/mg) of the CAT assay conditions used. Deletion constructs were created by taking advantage of conveniently located restriction sites to generate new 5′ ends at −87 (AflIII) and −72 (RsaI). In the case of AflIII, the DNA was treated with Klenow fragment after restriction to generate a blunt end. The truncated promoter fragments were then excised by digestion with ClaI and ligated with pKK232-8 that had been digested with SmaI and ClaI. As shown in Fig. 1, constructs pJC1 and pJC3 retain full activity, while removal of sequences upstream from the RsaI site abolishes all activation by NucC, P2 Ogr, and P4 δ. This endpoint corresponds directly to the upstream arm of a sequence element resembling those in the P2 and P4 late promoters.

FIG. 1.

Mapping the activator binding site on P_nucA. Each of the fragments tested is indicated schematically below a map of the promoter region. These fragments were used to create derivatives of the cat expression vector pKK232-8 (4). Compatible plasmids encoding P2 Ogr (pGC57) and P4 Delta (pGVB1) under λ p_L control have been described previously (12, 22). An equivalent plasmid encoding NucC was constructed by amplifying a 310-bp fragment containing the nucC gene from pSE380TacNucC (15) by using Vent DNA polymerase (New England Biolabs) with primers SM1 (5′-CGGAATTCTATGATGCACTGTCCACT-3′) (the translation initiation codon for NucC is underlined) and SM2 (5′-CACGTTGCATTTGCGAG-3′). Following digestion with EcoRI, this fragment was ligated with pBB105 (2) to yield pTG136. Cultures of E. coli C-1a carrying each activator plasmid and promoter construct were assayed 30 min after synthesis of the activator was induced by a shift to 42°C, as described by Grambow et al. (12). CAT activity is expressed as micromoles of chloramphenicol acetylated per minute at 37°C per milligram of protein in the cell extract. Each value represents the average of two determinations from each of at least two independent cultures.

A DNase I footprint extending from −70 to −43 on the late promoters of bacteriophage P2 and P4 had previously been demonstrated using P4 Delta protein that was copurified with a maltose-binding protein-Delta fusion protein (16). DNase I protection analysis was carried out on the NucA promoter in the same manner. The region of protection, extending from −81 to −51, also corresponds to the dyad element implicated by the deletion analysis (Fig. 2). As has been seen for the P2 and P4 late promoters (17), two other activators from the Ogr-like family, the Delta protein of retronphage φR73 and the Pag protein from phage PSP3, protected the same upstream region of P_nucA (data not shown). This strongly suggests that binding to the dyad element is similar among the entire family of activators, including NucC.

FIG. 2.

DNase I footprint analysis of the nucA promoter. A promoter fragment generated by cleavage of pTG198 was labeled at a SalI or XhoI site to generate probes for the top strand and bottom strand, respectively. Footprinting with a mixture of P4 Delta and maltose-binding protein–Delta was carried out as described by Julien and Calendar (16). The G and G+A lanes indicate the Maxam-Gilbert sequencing reactions, and a minus sign indicates reaction mixtures to which no protein was added. The numbers beside the connected arrows give the location of the protected region relative to the transcription start site. The region of protection for each strand is also indicated by connected arrows on the sequence shown below the footprints. The partial dyad sequences are indicated by inverted arrows, and the transcription start site is shown by the bent arrow above the sequence.

The activator binding site identified in this study contains the consensus TGT-N₁₂-ACA motif found in the P2 and P4 late promoters but is centered at −67, which corresponds to one full turn of the helix upstream from the binding sites in the P2 and P4 late promoters. This provided us with an opportunity to elucidate the spacing requirements involved in activation from these promoters. An NdeI restriction site located between the downstream half of the dyad and the −35 region (Fig. 1) facilitated the deletion of multiple bases in this region. PCR using the synthetic oligonucleotide Pnuc5 with either NdeΔ4 (5′CATATGTTGTGAATGTGTCAGTCAT3′) or NdeΔ9 (5′CATATGAATGTGTCAGTCATGGTAC3′) resulted in fragments that moved the NdeI site upstream either four or nine bases. These products were initially ligated into the TA vector pCR2.1 (Invitrogen), and their sequences were verified. Mutant fragments were then excised by digestion with ClaI and NdeI and ligated into pTG198, replacing the corresponding wild-type promoter fragment. Additional spacing variants of these mutant promoters were created by NdeI digestion followed by treatment with either mung bean nuclease or Klenow fragment. The resulting blunt-end plasmid DNAs were religated, and the sequences of the deletions were determined. Mutant promoter fragments were then subcloned into pSL100 as described above and assayed for NucC-dependent CAT activity. Only the deletion mutant pNucA-Δ11 showed activity similar to wild-type levels (Fig. 3). This shift in position, in fact, appears to promote slightly higher levels of activation. This represents one complete rotation along the DNA helix, consistent with a requirement for these activators to bind to a specific face of the DNA relative to RNA polymerase. Genetic evidence supports a model for transcription activation by the Ogr-like proteins that involves a direct interaction between the activator and the alpha subunit(s) of RNA polymerase (2, 14, 18, 20, 23). The strict helical phasing requirement for activation of the nuclease promoter is consistent with such a model, in which NucC bound to the upstream sequence makes a specific contact with RNA polymerase. A similar phase dependence has been demonstrated for activation of certain promoters by the cyclic AMP receptor protein (11, 21), which also interacts directly with RNA polymerase via one of the alpha subunits.

FIG. 3.

Deletion analysis of P_nucA. The NdeI restriction site used to facilitate deletion of multiple bases in this region is underlined in the wild-type sequence. The highly conserved TGT and ACA sequences in the activator binding site are in boldface. Mutant promoter fragments were assayed for NucC-dependent CAT activity as described in the legend to Fig. 1 in E. coli C-4595 (12) carrying the NucC plasmid pTG136.

The S. marcescens bss gene, encoding bacteriocin 28b, has also been shown to be positively regulated by NucC (10). While the transcription start site for the bss gene has not been reported, Ferrer et al. (10) identified putative −10 and −35 regions in the sequence upstream of the bacteriocin 28b coding region. There is a consensus dyad element, TGT-N₁₂-ACA, located upstream of this presumptive promoter between −47 and −64. This is the same location as the binding sites in the P2 and P4 late promoters, and we predict that this dyad element is the NucC binding site for the bss promoter. The aberrant location of the NucC binding site in P_nucA may reflect the way in which the nuclease promoter evolved to allow regulation by a transcription factor that was originally phage encoded.