Rns is known to regulate one member, CS1, of fimbrial subclass 5b. In addition, the expression of the related CS19 fimbrial proteins is enhanced when CS19+ strains are transformed with a plasmid expressing CfaD [GenBank:P25393] which is 97% identical to Rns 1819. However, it is not known if the enhancement of CS19 expression results from positive regulation of the fimbrial promoter or an indirect effect involving non-fimbrial genes within the Rns/CfaD regulon 2021. Currently, there is no evidence to suggest that the other subclass 5b members, CS17 and PCFO71, are regulated by Rns. However, the available data does not exclude this possibility. To determine if other subclass 5b fimbria are regulated by Rns, we cloned the promoters of CS17, CS19, and PCFO71 into a promoterless Lac reporter plasmid that was then integrated into the chromosomal attB_HK022 site of K-12 strain MC4100 by site specific recombination. Quantitative enzymatic assays revealed that the reporter strains expressed 8 to 10 times more β-galactosidase when they were transformed with a Rns expression plasmid, pGPMRns, than when they were transformed with the rns::kan negative control plasmid pGPMRns<Tn>2 (Table 1). Thus, when these results are combined with previous analyses of the CS1 fimbrial promoter 15, they reveal that all known subclass 5b fimbrial promoters are activated by Rns.

The CS17, CS19, and PCFO71 promoters have not been previously characterized; therefore, the transcription start site of each promoter was mapped by primer extension. We found that the Rns-dependent transcription start site is 16 nucleotides upstream of csbB (CS17), 17 nucleotides upstream of csdB (CS19), and 20 nucleotides upstream of cosB (PCFO71) (Figure 1). We did not observe primer extension products in the absence of Rns although our β-galactosidase assays indicated that each promoter has at least a low level of Rns-independent activity (Table 1). This difference is most likely a consequence of the relatively long half-life of β-galactosidase compared to the short half-life of most mRNA's in E. coli. Thus, β-galactosidase assays reflect the accumulative activity of the promoter over an extended period of time while mRNA levels more accurately reflect the promoter's activity at a specific time. Nevertheless, the two different assays, one qualitative the other quantitative, both demonstrate that the CS17, CS19, and PCFO71 fimbrial promoters are positively regulated by Rns.

In vitro DNase I footprinting was used to determine Rns binding site locations at the CS17, CS19, and PCFO71 promoters. Like other AraC/XylS family members that have been characterized, Rns is too insoluble for the characterization of protein-DNA complexes in vitro. However, it has been previously shown that the fusion of maltose binding protein (MBP) to the amino-terminus of Rns substantially increases its solubility without affecting its activity in vivo 17. With purified soluble fusion protein we found that Rns binds to a site adjacent to each promoter's -35 hexamer, which were predicted based on primer extension results and sequence analyses (Figure 2). Each promoter also carries a second binding site centered, relative to the Rns-dependent transcription start site, at -109.5 for CS17p and PCF071p or -108.5 for CS19p. DNase I footprints of MBP-Rns typically cover 33 to 36 nucleotides. Due to steric occlusion, these footprints overestimate the actual site of protein-nucleotide contacts 17. However, each footprint encompasses a central core of 12 nucleotides with at least 67% identity to the Rns binding site consensus sequence TATTTTTTTATC 21. Since Rns binding sites are not symmetrical, the conserved sequences are either on the coding or noncoding strand of each promoter (Figure 2).

Since DNase I footprinting identified two Rns binding sites at each fimbrial promoter, each site was subjected to oligonucleotide directed mutagenesis to determine its function with regards to promoter activation. Prior to in vivo analyses, we used gel mobility assays to determine if the point mutations reduced or abolished Rns binding in vitro. We found that the ability of MBP-Rns to alter the mobility of DNA fragments carrying mutagenized sites was substantially reduced compared to wild-type binding sites (Figure 3). This was expected because three or four nucleotides within the conserved core of each site were changed. In addition to large nonspecific protein-DNA complexes trapped in the wells of most gels at high concentrations of MBP-Rns, we also observed a low mobility complex with the wild-type csbB2o binding site. This suggests the presence of an additional low affinity binding site that was not observed by DNase I footprinting. This site is likely a pseudo binding site and is probably not relevant in vivo because high concentrations of protein were required for binding.

After gel mobility assays, we evaluated the function of each wild-type and mutagenized binding site in vivo by β-galactosidase assays. Overall levels of enzymatic activity were higher in stationary phase (Figure 4) than late log phase (Table 1) because as explained above, β-galactosidase is a stable enzyme that accumulates over time. For each promoter, we observed that mutations within either binding site reduced Rns-dependent expression of the enzyme (Figure 4). However, the most dramatic reductions were observed when both binding sites were simultaneously mutagenized. Taken together, these results demonstrate that the CS17, CS19, and PCFO71 fimbrial promoters each carry two functional binding sites that are required for full Rns-dependent activation.

Bacterial strains are described in Table 2. The CS17 fimbrial promoter was amplified from strain WS6788A with primers SN419-Bam and SN552-Eco. Primer sequences are listed in Table 3. The CS19 promoter was amplified from strain WS0115A with primers SN421-Bam and SN552-Eco. The PCFO71 promoter was amplified from strain WS2173A with primers SN530-Bam and SN553-Eco. The 0.7 kb PCR products were digested with BamHI and EcoRI, then ligated into the same sites of pHKLac1 to construct pCS17Lac2, pCS19Lac2, and pCFO71Lac2. Plasmid pHKLac1 carries lacZYA flanked by transcriptional terminators, aadA, attP_HK022 and the pir-dependent γ origin of replication from R6K 20. It is a derivative of pAH144 (GenBank:AY048731) with the addition of a 5.5 kb BamHI-MfeI fragment carrying lacZYA from pRS550 3132.

For the purpose of mutagenesis, each reporter plasmid was digested with PvuII then ligated to excise 5 kb of lacZYA. The resulting plasmids (pCS17c, pCS19c, and pCFO71c) were then subjected to oligonucleotide directed mutagenesis to create point mutations in Rns binding sites. The point mutations were designed so that they also generate unique restriction sites. Plasmid pCS17c was subjected to inverse PCR with primers SN562-AgeI and SN563-AgeI or SN568-KpnI and SN569-KpnI. Plasmid pCS19c was subjected to inverse PCR with primers SN572-SpeI and SN573-SpeI or SN574-KpnI and SN575-KpnI. Plasmid pCFO71c was subjected to inverse PCR with primers SN560-AgeI and SN561-AgeI or SN570-KpnI and SN571-KpnI. The 3.6 kb PCR products were then digested with AgeI, KpnI, or SpeI then ligated to yield pCS17cAge, pCS17cKpn, pCS19cSpe, pCS19cKpn, pCFO71cAge, and pCFO71cKpn. The mutagenized promoter fragments were then cloned into pHKLac1 as 0.7 kb BamHI-EcoRI fragments. Plasmids containing double mutations, where both Rns-binding sites are mutagenized, were generated by subjecting plasmids pCS17Lac3, pCS19Lac4, and pCFO71Lac3 to an inverse PCR reaction with primers SN568-KpnI and SN569-KpnI, SN574-KpnI and SN575-KpnI, and SN570-KpnI and SN571-KpnI, respectively. The PCR products were then digested with KpnI and recircularized.

Plasmid pCS17Lac3 carries CS17p with the unique AgeI restriction site in the Rns promoter proximal binding site csbB1o. pCS17Lac4 carries CS17p with the unique KpnI site in the Rns promoter distal binding site csbB2o. pCS17Lac5 carries both unique restriction sites. Plasmids pCS19Lac3 and pCS19Lac4 carry CS19p with a unique KpnI site in the promoter distal Rns binding site csdB2o or a SpeI site in the promoter proximal binding site csdB1o, respectively. pCS19Lac5 carries both mutations. Plasmids pCFO71Lac3 and pCFO71Lac4 carry CFO71p with a unique AgeI site in the promoter proximal Rns binding site cosB1o or a KpnI site in the promoter distal binding site cosB2o, respectively. pCFO71Lac5 carries both mutations. Reporter plasmids were integrated into the chromosomal attB_HK022 site of MC4100 by Int_HK022 dependent, site-specific recombination 31. Strains with single plasmid integrants were identified by colony PCR 31 and used for enzymatic assays.

Plasmid pRARE2 (EMD Biosciences) provides seven rare tRNAs to supplement the rare codon usage of rns. Plasmid pMBPRns1 expresses MBP-Rns from an isopropyl-β-D-1-thiogalactopyranoside (IPTG)-inducible tac promoter 20. Plasmid pGPMRns is a derivative of pNEB193 (New England Biolabs) and expresses Rns from lacp 20. Plasmid pGPMRns<Tn>2 carries rns::kan and was produced by transposon mutagenesis of pGPMRns.

Reporter strains were transformed with pGPMRns (rns+ bla) or pGPMRns<Tn>2 (rns::kan bla) and grown aerobically at 37°C in Luria-Bertani (LB) medium with 100 μg/ml ampicillin. Cells were harvested, lysed and assayed for β-galactosidase activity as previously described 33. Stationary phase cells were harvested from overnight cultures with an optical absorbance between 0.7 and 0.8 at 580 nm. Late log phase cells were harvested 3 to 4 hours after inoculation when the optical absorbance of the culture reached 0.4 to 0.6 at 580 nm.

Strain KS1000/pRare2/pMBPRns1 was grown aerobically at 37°C in LB medium containing 0.2% wt/vol glucose, 30 μg/ml chloramphenicol, and 100 μg/ml ampicillin. During mid-log phase the incubation temperature was lowered to 30°C then IPTG was added to a final concentration of 300 μM to induce the expression of MBP-Rns. After several hours of induction, cells were harvested at 4°C and concentrated ≥ 100 fold in ice-cold lysis buffer (10 mM TrisCl [pH 7.6], 200 mM NaCl, 1 mM EDTA, 0.5 mM CaCl₂, 10 mM β-mercaptoethanol, 100 μg/ml DNase I). A French press was used to mechanically lyse the cells. The lysate was clarified by high-speed centrifugation to pellet insoluble material. The supernatant was passed through an amylose column that was then washed with several column volumes of buffer A (10 mM TrisCl [pH 7.6], 200 mM NaCl, 1 mM EDTA, 15% vol/vol glycerol, and 10 mM β-mercaptoethanol). The fusion protein was eluted from the column with buffer B (buffer A with 10 mM maltose). The eluted protein was stored at -80°C in 10 mM TrisCl [pH 7.6], 50 mM KCl, 1 mM β-mercaptoethanol, 30% vol/vol glycerol after buffer exchange. The concentration of MBP-Rns was determined by the Bradford method relative to a standard curve of bovine serum albumin 34.

MBP-Rns was equilibrated with ³²P-end-labeled promoter DNA at 37°C in 10 mM TrisCl [pH 7.6], 50 mM KCl, 1 mM DTT, 0.4 mM MgCl₂, 0.2 mM CaCl₂, 2 ng/μl polydI-dC, 10 μg/ml bovine serum albumin. After equilibration, DNase I was added to a final concentration of 100 ng/μl for 1 min. at 37°C. The enzymatic reaction was quenched by addition of 10 volumes of 570 mM NH₄OAc, 50 μg/ml tRNA, 80% vol/vol ethanol. The solution was vortexed then placed on dry ice to precipitate DNA. The DNA was pelleted, washed with 70% vol/vol ethanol, dried, then resuspended in 4 μl 80% vol/vol formamide, 50 mM Tris-Borate [pH 8.3], 1 mM EDTA, 0.1% wt/vol xylene cyanol and bromophenol blue. The samples were heat denatured and separated on sequencing gels. Dried gels were visualized by exposure to phosphorimager plates. The Maxam-Gilbert method was used to generate sequence ladders 35.

MBP-Rns was equilibrated with ³²P-labeled CS17, CS19, or PCFO71 promoter fragments in 10 mM TrisCl (pH 7.6), 50 mM KCl, 1 mM DTT, 2 ng/ul poly (dI-dC), 0.1 mg/ml bovine serum albumin, and 6% (vol/vol) glycerol at 37°C for 20 minutes. After equilibration, the reactions were loaded onto non-denaturing 5% acrylamide gels with TAE (40 mM Tris-acetate, 1 mM EDTA [pH 8.5]) as the gel and running buffer. The gels were run at 150 volts for 2 hours, dried, and visualized by exposure to phosphorimager plates.

Cells were harvested from cultures after they reached an optical absorbance of 1.0 at 595 nm. Total RNA was isolated from reporter strains GPM1133 (CS17p::lacZYA), GPM1134 (CS19p::lacZYA), and GPM1131 (PCFO71p::lacZYA) transformed with pGPMRns (rns+ bla) or pGPMRns<Tn>2 (rns::kan bla) as previously described 20. Subsequently, 2.4 picomoles of a ³²P-end-labeled oligonucleotide (SN420-Eco for CS17p and CS19p; SN531-Eco for PCFO71p) was combined with 54–169 μg of total RNA and 0.8 mM dNTPs. The solution was heated to 65°C for 5 minutes then cooled on ice for 2 minutes to anneal the primer, which was then extended with SuperScript™ III Reverse Transcriptase according to the supplier's protocol (Invitrogen). Aliquots were separated on DNA sequencing gels alongside Maxam-Gilbert sequencing ladders after heat denaturation.