Portal de Periódicos da CAPES

Crl, a Low Temperature-induced Protein in Escherichia coli That Binds Directly to the Stationary Phase σ Subunit of RNA Polymerase

2004; Elsevier BV; Volume: 279; Issue: 19 Linguagem: Inglês

10.1074/jbc.m314145200

1083-351X

Alexandre Bougdour, Cécile Lelong, Johannes Geiselmann,

Protein Structure and Dynamics

The alternative sigma factor σS (RpoS) of Escherichia coli RNA polymerase regulates the expression of stationary phase and stress-response genes. σS is also required for the transcription of the cryptic genes csgBA that encode the subunits of the curli proteins. The expression of the csgBA genes is regulated in response to a multitude of physiological signals. In stationary phase, these genes are transcribed by the σS factor, and expression of the operon is enhanced by the small protein Crl. It has been shown that Crl stimulates the activity of σS, leading to an increased transcription rate of a subset of genes of the rpoS regulon in stationary phase. However, the underlying molecular mechanism has remained elusive. We show here that Crl interacts directly with σS and that this interaction promotes binding of the σS holoenzyme (EσS) to the csgBA promoter. Expression of Crl is increased during the transition from growing to stationary phase. Crl accumulates in stationary phase cells at low temperature (30 °C) but not at 37 °C. We therefore propose that Crl is a second thermosensor, besides DsrA, controlling σS activity. The alternative sigma factor σS (RpoS) of Escherichia coli RNA polymerase regulates the expression of stationary phase and stress-response genes. σS is also required for the transcription of the cryptic genes csgBA that encode the subunits of the curli proteins. The expression of the csgBA genes is regulated in response to a multitude of physiological signals. In stationary phase, these genes are transcribed by the σS factor, and expression of the operon is enhanced by the small protein Crl. It has been shown that Crl stimulates the activity of σS, leading to an increased transcription rate of a subset of genes of the rpoS regulon in stationary phase. However, the underlying molecular mechanism has remained elusive. We show here that Crl interacts directly with σS and that this interaction promotes binding of the σS holoenzyme (EσS) to the csgBA promoter. Expression of Crl is increased during the transition from growing to stationary phase. Crl accumulates in stationary phase cells at low temperature (30 °C) but not at 37 °C. We therefore propose that Crl is a second thermosensor, besides DsrA, controlling σS activity. RNA polymerase of Escherichia coli is composed of a core enzyme (E) with subunit structure α2ββ′ which associates with one of seven different σ subunits to form the holoenzyme (Eσ). The core enzyme carries the RNA polymerization function, and the σ subunit is required for promoter recognition and binding (1Burgess R.R. Travers A.A. Dunn J.J. Bautz E.K. Nature. 1969; 221: 43-46Crossref PubMed Scopus (639) Google Scholar, 2Helmann J.D. Chamberlin M.J. Annu. Rev. Biochem. 1988; 57: 839-872Crossref PubMed Scopus (716) Google Scholar). Each σ subunit targets RNA polymerase to a different set of promoters, thereby profoundly modulating the gene expression pattern (3Gaal T. Ross W. Estrem S.T. Nguyen L.H. Burgess R.R. Gourse R.L. Mol. Microbiol. 2001; 42: 939-954Crossref PubMed Scopus (150) Google Scholar, 4Ishihama A. Annu. Rev. Microbiol. 2000; 54: 499-518Crossref PubMed Scopus (454) Google Scholar). RNA polymerase holoenzyme containing the σ70 subunit is responsible for the transcription of the majority of genes during exponential growth. Upon entry into stationary phase, σ38 also called σS and encoded by the rpoS gene, begins to accumulate in the cell, associates with the core enzyme, and directs the transcription of genes essential for stationary phase survival (5Lange R. Hengge-Aronis R. Mol. Microbiol. 1991; 5: 49-59Crossref PubMed Scopus (591) Google Scholar, 6Tanaka K. Takayanagi Y. Fujita N. Ishihama A. Takahashi H. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 3511-3515Crossref PubMed Scopus (193) Google Scholar, 7Loewen P.C. Hengge-Aronis R. Annu. Rev. Microbiol. 1994; 48: 53-80Crossref PubMed Scopus (473) Google Scholar). The synthesis of σS is also induced in response to many other stress conditions such as high osmolarity, low pH, and high temperature. σS is therefore considered a general stress-response regulator (8Lange R. Hengge-Aronis R. Genes Dev. 1994; 8: 1600-1612Crossref PubMed Scopus (415) Google Scholar, 9Small P. Blankenhorn D. Welty D. Zinser E. Slonczewski J.L. J. Bacteriol. 1994; 176: 1729-1737Crossref PubMed Google Scholar, 10Hengge-Aronis R. Mol. Microbiol. 1996; 21: 887-893Crossref PubMed Scopus (293) Google Scholar). Regulation of σS occurs at transcriptional and post-transcriptional levels and involves numerous regulators. In brief, σS accumulates at the beginning of stationary phase because of many factors acting in concert: increased transcription of the rpoS gene, stabilization and activation of the rpoS mRNA, as well as increased protein stability (8Lange R. Hengge-Aronis R. Genes Dev. 1994; 8: 1600-1612Crossref PubMed Scopus (415) Google Scholar). A number of small molecules, RNAs, and proteins belong to the complex regulatory network that controls σS expression; ppGpp (the stringent control signal) (11Gentry D.R. Hernandez V.J. Nguyen L.H. Jensen D.B. Cashel M. J. Bacteriol. 1993; 175: 7982-7989Crossref PubMed Google Scholar), homoserine lactone (12Huisman G.W. Kolter R. Science. 1994; 265: 537-539Crossref PubMed Scopus (178) Google Scholar), inorganic phosphate (13Shiba T. Tsutsumi K. Yano H. Ihara Y. Kameda A. Tanaka K. Takahashi H. Munekata M. Rao N.N. Kornberg A. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 11210-11215Crossref PubMed Scopus (160) Google Scholar), UDP-glucose (14Bohringer J. Fischer D. Mosler G. Hengge-Aronis R. J. Bacteriol. 1995; 177: 413-422Crossref PubMed Google Scholar), and cAMP-CRP (8Lange R. Hengge-Aronis R. Genes Dev. 1994; 8: 1600-1612Crossref PubMed Scopus (415) Google Scholar) have been reported to influence the transcription of rpoS. The nucleoid-associated protein HU, the histone-like protein H-NS, and the small regulatory RNAs DsrA and OxyS in conjugation with the RNA chaperone Hfq control the translation of rpoS by modulating the secondary structure of the rpoS mRNA (15Balandina A. Claret L. Hengge-Aronis R. Rouviere-Yaniv J. Mol. Microbiol. 2001; 39: 1069-1079Crossref PubMed Scopus (104) Google Scholar, 16Repoila F. Gottesman S. J. Bacteriol. 2001; 183: 4012-4023Crossref PubMed Scopus (123) Google Scholar). σS proteolysis is controlled by the response regulator RssB (the σS recognition factor) and the ClpXP protease (recently reviewed in Ref. 17Hengge-Aronis R. Microbiol. Mol. Biol. Rev. 2002; 66: 373-395Crossref PubMed Scopus (764) Google Scholar). Despite this abundance of information about the regulation of σS expression, little is known about the regulation of σS activity. The total concentration of σS does not exceed one-third of the concentration of σ70 even at the onset of stationary phase when σS concentration is at its highest (18Jishage M. Ishihama A. J. Bacteriol. 1995; 177: 6832-6835Crossref PubMed Scopus (185) Google Scholar, 19Jishage M. Iwata A. Ueda S. Ishihama A. J. Bacteriol. 1996; 178: 5447-5451Crossref PubMed Google Scholar). Moreover, among all σ factors, σS has the lowest affinity for the core enzyme (20Maeda H. Fujita N. Ishihama A. Nucleic Acids Res. 2000; 28: 3497-3503Crossref PubMed Google Scholar). σS thus appears to be in a difficult position in the competition for binding to core enzyme. Recently, it has been shown that ppGpp modulates this competition in favor of σS (21Jishage M. Kvint K. Shingler V. Nystrom T. Genes Dev. 2002; 16: 1260-1270Crossref PubMed Scopus (264) Google Scholar). Other factors, such as glutamate, trehalose, and inorganic polyphosphate, modulate the activity of σS holoenzyme at the steps of holoenzyme formation and/or holoenzyme binding to promoters (22Ding Q. Kusano S. Villarejo M. Ishihama A. Mol. Microbiol. 1995; 16: 649-656Crossref PubMed Scopus (68) Google Scholar, 23Kusano S. Ishihama A. J. Bacteriol. 1997; 179: 3649-3654Crossref PubMed Google Scholar, 24Kusano S. Ishihama A. Genes Cells. 1997; 2: 433-441Crossref PubMed Scopus (58) Google Scholar). In addition, recent studies have identified the crl gene product as a regulator of σS activity (25Pratt L.A. Silhavy T.J. Mol. Microbiol. 1998; 29: 1225-1236Crossref PubMed Scopus (92) Google Scholar, 26Schnetz K. Microbiology. 2002; 148: 2573-2578Crossref PubMed Scopus (15) Google Scholar). Crl is known for stimulating the transcription of csgBA, the operon encoding for the two curli subunits, in a σS-dependent manner (27Olsen A. Arnqvist A. Hammar M. Sukupolvi S. Normark S. Mol. Microbiol. 1993; 7: 523-536Crossref PubMed Scopus (232) Google Scholar, 28Arnqvist A. Olsen A. Normark S. Mol. Microbiol. 1994; 13: 1021-1032Crossref PubMed Scopus (142) Google Scholar, 29Hammar M. Arnqvist A. Bian Z. Olsen A. Normark S. Mol. Microbiol. 1995; 18: 661-670Crossref PubMed Scopus (385) Google Scholar). In a crl null strain, the transcriptional activity of the csgBA promoter is reduced ∼4-fold compared with the wild type strain (25Pratt L.A. Silhavy T.J. Mol. Microbiol. 1998; 29: 1225-1236Crossref PubMed Scopus (92) Google Scholar). Curli fibers are thin aggregative surface fimbriae, which are involved in cell-cell attachment (30Romling U. Bian Z. Hammar M. Sierralta W.D. Normark S. J. Bacteriol. 1998; 180: 722-731Crossref PubMed Google Scholar, 31Romling U. Sierralta W.D. Eriksson K. Normark S. Mol. Microbiol. 1998; 28: 249-264Crossref PubMed Scopus (346) Google Scholar) and adhesion to extracellular matrices (32Olsen A. Jonsson A. Normark S. Nature. 1989; 338: 652-655Crossref PubMed Scopus (455) Google Scholar, 33Hammar M. Bian Z. Normark S. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 6562-6566Crossref PubMed Scopus (229) Google Scholar, 34Sjobring U. Pohl G. Olsen A. Mol. Microbiol. 1994; 14: 443-452Crossref PubMed Scopus (125) Google Scholar). Curli are expressed under special environmental conditions, such as low temperature, low osmolarity, and in stationary phase (32Olsen A. Jonsson A. Normark S. Nature. 1989; 338: 652-655Crossref PubMed Scopus (455) Google Scholar, 35Olsen A. Arnqvist A. Hammar M. Normark S. Infect. Agents Dis. 1993; 2: 272-274PubMed Google Scholar). Control of curli production in Escherichia coli and Salmonella typhimurium involves a complex network of regulatory proteins. Global regulators such as H-NS (27Olsen A. Arnqvist A. Hammar M. Sukupolvi S. Normark S. Mol. Microbiol. 1993; 7: 523-536Crossref PubMed Scopus (232) Google Scholar, 28Arnqvist A. Olsen A. Normark S. Mol. Microbiol. 1994; 13: 1021-1032Crossref PubMed Scopus (142) Google Scholar), OmpR (30Romling U. Bian Z. Hammar M. Sierralta W.D. Normark S. J. Bacteriol. 1998; 180: 722-731Crossref PubMed Google Scholar), and IHF (36Gerstel U. Park C. Romling U. Mol. Microbiol. 2003; 49: 639-654Crossref PubMed Scopus (150) Google Scholar) control the expression of the curli. Osmolarity sensing occurs through the two-component regulatory systems EnvZ/OmpR and CpxA/CpxR (37Prigent-Combaret C. Vidal O. Dorel C. Lejeune P. J. Bacteriol. 1999; 181: 5993-6002Crossref PubMed Google Scholar, 38Dorel C. Vidal O. Prigent-Combaret C. Vallet I. Lejeune P. FEMS Microbiol. Lett. 1999; 178: 169-175Crossref PubMed Google Scholar), but temperature-sensing of curli expression is not understood. In the present study we have investigated, at the molecular and the physiological levels, the mechanism by which Crl influences σS activity. We report that Crl interacts directly with the σS subunit in vitro. Electrophoretic mobility shift assays (EMSA) 1The abbreviations used are: EMSA, electrophoretic mobility-shift assays; Cm, chloramphenicol; DTT, dithiothreitol; Ni-NTA, nickel-nitrilotriacetic acid; Kn, kanamycin; wt, wild type. indicate that Crl associates with the transcriptional complex formed by the EσS (σS holoenzyme) and the csgBA promoter (pcsgBA). Our work provides evidence that Crl positively controls EσS recruitment to pcsgBA. Immunoblot analysis of Crl demonstrates that Crl expression is growth phase- and temperature-dependent. Strains and Plasmids—The genotypes of E. coli K12 strains and plasmids used in this study are listed in Table I. Media used were Luria-Bertani broth with 5 g/liter NaCl (LB) and LB containing 12 g/liter agar (LA) (39Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual,2nd Ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1989Google Scholar). Media were supplemented with 15 μg/ml chloramphenicol (Cm), 25 μg/ml kanamycin (Kn), 100 μg/ml ampicillin, 40 μg/ml 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-gal), or 1 mm isopropyl-1-thio-β-d-galactopyranoside as required. Strains were grown at 37 or 30 °C with shaking (200 rpm).Table IBacterial strains and plasmidsStrain or plasmidGenotypeRef. or sourceE. coli strainsMC4100F- araD139 Δ(argF-lac) U169 rpsL150 relA1 150flbB5301 deoC1 ptsF25 rhsRSilhavy et al. (56Silhavy T.J. Berman M.L. Enquist L.W. Experiments with Gene Fusions. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1984Google Scholar)W3110λ- IN(rrnD-rrnE)1. rphlBachmann et al. (57Bachmann B.J. Low K.B. Taylor A.L. Bacteriol. Rev. 1976; 40: 116-167Crossref PubMed Google Scholar)ZK1000W3110 tna-2 ΔlacU169 rpoS::KnrBohanonn et al. (42Bohannon D.E. Connell N. Keener J. Tormo A. Espinosa-Urgel M. Zambrano M.M. Kolter R. J. Bacteriol. 1991; 173: 4482-4492Crossref PubMed Google Scholar)BL001W3110 crl::CmrFaure et al. (41Faure D. Lelong C. Portier P. Blot M. Diversification, Adaptation et Conservation de Populations Bactériennes d'Origine Clonale. Bureau des Ressources Génétiques, Paris2004Google Scholar)BL002W3110 rpoS::KnrThis studyBL003W3110 crl::Cmr rpoS::KnrThis studyFI1202MC4100 rpoS::Tn10 lacIqBecker et al. (43Becker G. Klauck E. Hengge-Aronis R. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6439-6444Crossref PubMed Scopus (158) Google Scholar)LP895MC4100 crl::CmrPratt et al. (25Pratt L.A. Silhavy T.J. Mol. Microbiol. 1998; 29: 1225-1236Crossref PubMed Scopus (92) Google Scholar)M15F-Lac-Thi-Mtl-QiagenPlasmidspCR® II-TOPO®Ampr KnrInvitrogenpQE-30AmprQiagenpTOPO-crlpCR®II-TOPO/crlThis studypHis6-CrlpQE-30/his6-crlThis studypHis6-σSpQE-30/his6-rpoSBecker et al. (43Becker G. Klauck E. Hengge-Aronis R. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6439-6444Crossref PubMed Scopus (158) Google Scholar)pTOPO-pcsgBpCR®II-TOPO/pcsgBThis studypREP4Knr/lacIqQiagenpDEB2Multicopy plasmid/rpoS; AmprA. R. Hengge Open table in a new tab The crl mutation was transferred from E. coli LP895 (25Pratt L.A. Silhavy T.J. Mol. Microbiol. 1998; 29: 1225-1236Crossref PubMed Scopus (92) Google Scholar) to W3110 using P1(vir)-mediated transduction (40Miller J.H. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1972: 201-205Google Scholar). Transduction of the crl::Cmr allele was obtained by selection on Cm plates (41Faure D. Lelong C. Portier P. Blot M. Diversification, Adaptation et Conservation de Populations Bactériennes d'Origine Clonale. Bureau des Ressources Génétiques, Paris2004Google Scholar). The rpoS mutation was transferred from the strain ZK1000 (42Bohannon D.E. Connell N. Keener J. Tormo A. Espinosa-Urgel M. Zambrano M.M. Kolter R. J. Bacteriol. 1991; 173: 4482-4492Crossref PubMed Google Scholar) to W3110. Transduction of the rpoS::Knr allele was obtained by selection on Kn plates. We verified by Southern blot that the mutations were introduced at the correct locus. All clones were stored at –80 °C in a 20% glycerol solution. Overproduction and Purification of His6-σs and His6-Crl Proteins— The coding sequence of the rpoS gene was cloned on a His6 tag PQE-30 vector (Qiagen), yielding the pHis6-σS plasmid as described (43Becker G. Klauck E. Hengge-Aronis R. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6439-6444Crossref PubMed Scopus (158) Google Scholar). To produce the His6-Crl fusion protein, we PCR-amplified the coding region of the crl gene using chromosomal DNA from the W3110 strain as template and the primers C5 (5′-ACGTTACCGAGTGGACAC-3′) and C6 (5′-CGCCGTTAACTTCACCGG-3′). The PCR fragment was cloned into pQE-30 UA vector (Qiagen), yielding the pHis6-Crl plasmid. The E. coli strains FI1202 and M15[pREP4], transformed with pHis6-σS and pHis6-Crl, respectively, were grown in LB medium containing selective antibiotics until an A600 of 0.6. We then added isopropyl-β-d-thiogalactopyranoside to 1 mm. After 5–6 h of induction, cells were harvested and stored at –80 °C until use. Cell lysates were prepared under native conditions. Cell pellets were resuspended in lysis buffer (50 mm NaH2PO4, 300 mm NaCl, 10 mm imidazole, pH 8.0) before sonication. The lysates were centrifuged at 10,000 × g for 30 min at 4 °C, and the soluble fractions were applied to a Ni2+-NTA column (Qiagen) equilibrated with lysis buffer. The column was washed with 10× column volumes of buffer A (50 mm NaH2PO4, 300 mm NaCl, 20 mm imidazole, pH 8.0), followed by elution with buffer E (50 mm NaH2PO4, 300 mm NaCl, 250 mm imidazole, pH 8.0). In control experiments we used a similar protocol except that the sample was incubated with either 600 units of micrococcal nuclease or 300 μg/ml ethidium bromide for 30 min at room temperature before the washes. When needed, the eluates were purified by gel filtration using an Amersham Biosciences PD-10 desalting column equilibrated with storage buffer (50 mm Tris-HCl, pH 7.5, EDTA 0.1 mm, 150 mm KCl, 5 mm CaCl2, 5 mm MgCl2, 0.1 mm DTT). Glycerol was added to a final concentration of 10% (v/v), and the samples were stored at –20 °C until use. His6-σ70 was a generous gift of A. Kolb. Protein concentrations were determined using the Bradford protein assay kit (Bio-Rad) and bovine serum albumin as a standard. Electrophoresis and Immunoblot Analysis of Proteins—Crude cell extracts under native conditions were prepared as follows. Cells were grown in LB medium at 30 or 37 °C for the time indicated. Cell pellets were resuspended in buffer B (20 mm HEPES, pH 7.4, 1 mm EDTA, 1 mm DTT, 10% glycerol) before sonication. The lysates were centrifuged at 10,000 × g for 30 min at 4 °C, and the soluble fractions were stored at –80 °C. The protein concentrations were determined using the Bradford protein assay kit (Bio-Rad) and bovine serum albumin as a standard. Protein samples were analyzed by SDS-PAGE (39Sambrook J. Fritsch E.F. Maniatis T. Molecular Cloning: A Laboratory Manual,2nd Ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY1989Google Scholar). Amersham Biosciences prestained protein standards were used for molecular weight estimation. Antibodies against Crl were produced in rabbits by injecting highly purified His6-Crl protein (Eurogentec). Antibodies against σS were a gift from R. Hengge-Aronis (8Lange R. Hengge-Aronis R. Genes Dev. 1994; 8: 1600-1612Crossref PubMed Scopus (415) Google Scholar). No cross-reaction was observed for the antibodies used in this study. The immunoblot analysis of proteins electrotransferred (Bio-Rad system) onto nitrocellulose or polyvinylidene difluoride membranes (Amersham Biosciences) was performed with polyclonal antibodies raised against Crl and σS as described previously. The blots were developed either with ECL (Amersham Biosciences) or nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (Sigma) systems, and staining intensity was quantified with the ImageGauge (Fujifilm) software. Gel Filtration—Gel filtration was performed using the AKTA system (Amersham Biosciences). Samples of 500 μl containing 167 μg of either His6-Crl (13.5 nmol) or His6-σS (5 nmol) or both proteins in a 20 mm Tris-HCl, pH 8, buffer (containing EDTA 0.2 mm, 500 mm KCl, 0.5 mm DTT, 10% glycerol, and 0.1% Nonidet P-40) were incubated for 20 min at room temperature. The mixture was loaded onto a Superdex 200 HR 10/30 column (Amersham Biosciences) equilibrated with the same buffer. Filtration was performed at 4 °C at a flow rate of 0.250 ml/min. Fifty fractions of 500 μl each were collected and analyzed by immunoblot, as indicated previously, to determine the protein content of each fraction. The column was calibrated using molecular weight standards from Amersham Biosciences. EMSA—The E. coli RNA polymerase core enzyme was purchased from Epicenter Technologies. DNA for binding assay was generated by PCR from pTOPO-pcsgBA plasmid and by using the primers C1 (5′-ATACTTTGGTATGAACTAAAAAAGAA-3′) and C2 (5′-CTGGTCGTACATTTAAGAAATT-3′). The PCR product (158 bp) from –81 to +77 of csgBA promoter region (named pcsgBA) was purified (Qiagen) and labeled with [γ-32P]ATP using T4 polynucleotide kinase. Binding assays with purified proteins were conducted in 20-μl reaction mixtures containing 20 mm HEPES, pH 7.4, 50 mm KCl, 1 mm DTT, 10% glycerol, 3 mm MgCl2, 0,1 μg/μl bovine serum albumin, 6 ng/μl herring sperm DNA as competitor, 1 nm labeled DNA, 20 nm RNA polymerase core enzyme, 60 nm His6-σS protein or 60 nm His6-σ70, and 120 nm His6-Crl protein. The purified proteins were diluted into the reaction buffer (see above) before use. The σ subunits were pre-incubated with His6-Crl for 10 min at room temperature before reconstitution of holoenzymes (see below). The core enzyme and the σ subunits (with or without His6-Crl) were mixed at a ratio of 1:3 in reaction buffer and incubated at 30 °C for 15 min. Binding of EσS or Eσ70 to the csgBA promoter region was allowed to proceed for 20 min at 30 °C in order to allow open complex formation. The reaction mixtures were loaded on a 5% non-denaturing polyacrylamide gel and then run with HEPES buffer (50 mm HEPES, pH 7.4) at 3 watts for 30–45 min at room temperature. Binding assays with crude extracts (1 μg/μl) were conducted in 20-μl reaction mixtures containing 20 mm HEPES, pH 7.4, 50 mm KCl, 1 mm DTT, 10% glycerol, 1 mm EDTA, 10 nm labeled DNA, and 6 ng/μl herring sperm DNA as competitor. Binding was allowed to proceed for 15 min at room temperature. The reaction mixture was loaded on a 4–10% non-denaturing polyacrylamide gradient gel and then run with TBE-1× buffer at 2 watts for 72 min with a cooling system. Gels were dried before being scanned and quantified using a FLA-8000 PhosphorImager (Fujifilm). Intensity profiles of the retarded bands were fitted to a Lorentzian curve shape as shown in Equation 1, (y=A1+[(x−xmean)⋅2σ]2)(Eq. 1) in order to separate and quantify accurately the overlapping peaks. Footprinting Experiment—A 0.175-kb fragment containing the csgBA promoter was produced by PCR using primers P1(5′-ATCGGTCGACCTTTGGTATGAACTAAAAAAGAA-3′) and P2(5′-CGATCTCGAGCTGGTCGTACATTTAAGAAATT-3′) containing SalI and XhoI restriction sites, respectively. The 175-bp DNA products were labeled with [γ-32P]ATP, using T4 polynucleotide kinase prior to being digested with either SalI or XhoI in order to produce a DNA fragment with only one labeled end. The reaction mixture (50 μl) containing end-labeled DNA (1 nm) and reconstituted RNA polymerase (20 nm core enzyme and 60 nm His6-σS), with or without His6-Crl (20–200 nm) in reaction buffer (20 mm HEPES, pH 7.4, 50 mm NaCl, 1 mm DTT, 10% glycerol, 3 mm MgCl2, 0,1 μg/μl bovine serum albumin), was incubated for 30 min at room temperature. The σS subunit was pre-incubated with His6-Crl for 10 min at room temperature before reconstitution of the holoenzyme. DNase I (3 ng) was added, and the mixture was incubated for 5 min at 22 °C. The reaction was stopped by adding 25 μl of stop solution (4 m ammonium acetate and 0,2 μg/ml herring sperm DNA). DNA was precipitated from the reaction mixture with LiCl and ethanol. The reaction products were analyzed by electrophoresis in a 8% polyacrylamide sequencing gel containing 8 m urea. Gels were dried before being autoradiographed and finally quantified using a FLA-8000 PhosphorImager (Fujifilm). Unmodified bands were used for normalization of the results. Crl interacts directly with the σS subunit—The crl gene product stimulates the expression of curli fimbriae (44Arnqvist A. Olsen A. Pfeifer J. Russell D.G. Normark S. Mol. Microbiol. 1992; 6: 2443-2452Crossref PubMed Scopus (119) Google Scholar), which are involved in cell-cell aggregation and adhesion to extracellular matrices (32Olsen A. Jonsson A. Normark S. Nature. 1989; 338: 652-655Crossref PubMed Scopus (455) Google Scholar). Transcription of csgBA, coding for the two subunits of the curli, is activated by Crl in a σS-dependent way (27Olsen A. Arnqvist A. Hammar M. Sukupolvi S. Normark S. Mol. Microbiol. 1993; 7: 523-536Crossref PubMed Scopus (232) Google Scholar, 44Arnqvist A. Olsen A. Pfeifer J. Russell D.G. Normark S. Mol. Microbiol. 1992; 6: 2443-2452Crossref PubMed Scopus (119) Google Scholar). Furthermore, Crl has a stimulatory effect on the expression of many other σS-activated genes (25Pratt L.A. Silhavy T.J. Mol. Microbiol. 1998; 29: 1225-1236Crossref PubMed Scopus (92) Google Scholar), and it participates in the negative effects of σS on the expression of ompF (45Pratt L.A. Silhavy T.J. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 2488-2492Crossref PubMed Scopus (208) Google Scholar) and the bgl operon (26Schnetz K. Microbiology. 2002; 148: 2573-2578Crossref PubMed Scopus (15) Google Scholar). Crl therefore either (i) increases σS expression or (ii) somehow stimulates σS activity. We eliminated the first possibility because neither the expression of a transcriptional rpoS::lacZ fusion (data not shown) nor the σS levels are reduced in a crl mutant strain (Fig. 6, A and B, lanes 1 and 2). These data agree with the results published by Pratt and Silhavy (25Pratt L.A. Silhavy T.J. Mol. Microbiol. 1998; 29: 1225-1236Crossref PubMed Scopus (92) Google Scholar), who observed a greater amounts of σS in a crl null strain than in wild type cells. The most straightforward way to investigate the second possibility would be to test the effect of Crl on σS-dependent transcription in vitro. Unfortunately, Crl did not significantly modify the activity of σS as measured by this assay (data not shown). We therefore used a less direct approach to investigate this second hypothesis. We used His6-tagged protein fusions and Ni-NTA-agarose as an affinity matrix for His6-Crl in an in vitro binding assay. This method was chosen in order to get rid of possible aggregation of His6-Crl proteins when purified. Aggregation is prevented by immobilizing the protein on the gel matrix, thus avoiding excessively high local concentrations (46Reischl S. Wiegert T. Schumann W. J. Biol. Chem. 2002; 277: 32659-32667Abstract Full Text Full Text PDF PubMed Scopus (43) Google Scholar). Considering that the carboxyl-terminal end of Crl is responsible for its ability to stimulate csgBA transcription (44Arnqvist A. Olsen A. Pfeifer J. Russell D.G. Normark S. Mol. Microbiol. 1992; 6: 2443-2452Crossref PubMed Scopus (119) Google Scholar), we decided to add the His6 tag to the amino-terminal end of Crl. For σS, the His6 derivative has already been constructed by Becker et al. (43Becker G. Klauck E. Hengge-Aronis R. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 6439-6444Crossref PubMed Scopus (158) Google Scholar). The His6-Crl and His6-σS fusion proteins were purified, using Ni2+-NTA technology, from crude extracts of E. coli strains M15 and FI1202, respectively. The major His6 protein fusion peaks were recovered by elution of the native proteins with 250 mm imidazole (Fig. 1A). As expected, His6-Crl (Mr, 15,5137) migrates with a relative mobility of ∼15 kDa in an SDS-PAGE gel, as does wild type Crl (data not shown). Despite the presence of minor contaminating proteins, His6-Crl and His6-σS represented about 95% of the eluted protein, as judged by Fig. 1A. The yield of His6 fusion proteins was 0.2 mg from 100 ml of induced culture. Immunoblot analysis of the eluted fraction containing His6-Crl with anti-σS polyclonal antibodies allowed us to demonstrate that σS coeluted with His6-Crl from the Ni2+ column (Fig. 1B). The interaction specificity was verified using salt gradients (data not shown). As a control, a cleared extract from the host strain that does not express the fusion protein was applied to the Ni2+-NTA-agarose matrix. Immunoblot analysis of the eluted fraction with imidazole did not revealed the presence of proteins, thus suggesting that the interaction between Crl and σS is specific. To establish that the association of these two proteins was not mediated through DNA, we repeated the purification protocol in the presence of either ethidium bromide (300 μg/ml) or micrococcal nuclease (600 units) as described (47Lai J.S. Herr W. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 6958-6962Crossref PubMed Scopus (398) Google Scholar, 48Hakimi M.A. Bochar D.A. Schmiesing J.A. Dong Y. Barak O.G. Speicher D.W. Yokomori K. Shiekhattar R. Nature. 2002; 418: 994-998Crossref PubMed Scopus (234) Google Scholar). The association of σS with His6-Crl was unaffected by either treatment (Fig. 1C). In order to delineate which part of σS is responsible for the interaction with Crl, we repeated the His6-Crl purification in strain MNC10, which expresses relatively high amounts of a σS mutant called σS(Δ7–35) (49Studemann A. Noirclerc-Savoye M. Klauck E. Becker G. Schneider D. Hengge R. EMBO J. 2003; 22: 4111-4120Crossref PubMed Scopus (80) Google Scholar). The smaller protein σS(Δ7–35) still co-purified with His6-Crl, thus indicating that the amino-terminal region of σS is not essential for Crl binding in vitro (data not shown). The inverse protocol of using His6-σS to co-purify Crl was not possible because Crl appeared to bind non-specifically to the column. Therefore, the interaction between Crl and σS was further assessed using gel filtration. Purified proteins were applied to a Superdex 200 HR 10/30 column (Amersham Biosciences), and column samples were collected and analyzed by immunoblotting. Purified His6-σS alone, a 38-kDa protein, eluted earlier than expected (mainly in fractions 31–34), at the position predicted for a globular protein