Silencing of the Escherichia coli bgl operon by RpoS requires Crl

Karin Schnetz1

Institut für Genetik, Universität zu Köln, Weyertal 121, 50931 Köln, Germany1

Tel: +49 221 4703815. Fax: +49 221 4705975. e-mail: schnetz{at}uni-koeln.de


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Silencing of the Escherichia coli bgl operon is mediated by histone-like protein H-NS and affected by other pleiotropic regulators, including sigma factor RpoS. Silencing is relieved and the bgl operon is activated in hns mutants and by mutations that map in the vicinity of the bgl promoter. However, the expression level of activated bgl operon derivatives varies with the strain background. Here it is shown that the repression of the bgl operon by RpoS requires Crl. Crl is a protein that is necessary for the RpoS-dependent expression of the csgBA operon and that enhances the expression of other RpoS-dependent genes. In a Crl-negative strain RpoS had no effect on the bgl operon. The crl gene maps close to the proBA locus in the lac operon region and is deleted in many commonly used E. coli strains. Crl may therefore account for some of the observed strain-dependent variations of bgl operon expression levels and effects of pleiotropic regulators on bgl operon regulation.

Keywords: ß-glucoside, pleiotropic regulation, H-NS, lac operon region


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
The bgl operon, encoding the gene products necessary for the fermentation of aryl-ß-glucosides, is silent in wild-type Escherichia coli (Schaefler & Maas, 1967 ). The abundant histone-like nucleoid-structuring protein H-NS is essential for bgl silencing and in hns-null strains the operon is expressed at high levels (Defez & de Felice, 1981 ; Higgins et al., 1988 ). The activity of the CRP/cAMP-dependent bgl promoter is also enhanced and expression of the operon is activated by spontaneous mutations that disrupt a silencer sequence located upstream of the promoter, and by point mutations that improve the CRP-binding site (Reynolds et al., 1981 , 1986 ; Schnetz & Rak, 1992 ; Singh et al., 1995 ). Silencer sequences located downstream of the promoter contribute to silencing of the operon (Schnetz, 1995 ).

The expression level of the activated bgl operon derivatives and the effect of mutations that partially disrupt the hns gene can vary with the strain background (Higgins et al., 1988 ; Lopilato & Wright, 1990 ; Tsui et al., 1994 ; Free et al., 2001 ). In addition to H-NS and CRP, the bgl operon is affected by other pleiotropic regulators including the DNA-bending protein FIS, which competes with CRP for binding (Caramel & Schnetz, 2000 ), and the transcriptional-regulator-like proteins LeuO and BglJ (Giel et al., 1996 ; Ueguchi et al., 1998 ). RpoS, the key regulator in the stress response of E. coli and the H-NS homologue StpA are necessary for silencing of the wild-type bgl operon by a truncated H-NS protein lacking its C-terminal DNA-binding domain (Free et al., 1998 , 2001 ; Ohta et al., 1999 ). Furthermore, RpoS downregulates the expression of activated bgl operon alleles (Dole et al., 2002 ). Interestingly, the up to 50-fold repression of the bgl operon by RpoS is based on the amplification of a moderate (two- to threefold) repression of the transcription rate by RpoS via a second, post-transcriptional level of regulation involving the specific antiterminator protein BglG. BglG is encoded by the first gene of the operon. At low transcription rates BglG is limiting and transcription halts at rho-independent transcriptional terminators t1 in the leader and t2 within the operon. If the transcription rate increases above a threshold the basal synthesis of BglG is sufficient for antitermination and, as a result, bglG and the operon are expressed at high levels (Dole et al., 2002 ).

RpoS activity is controlled by multiple signals and at various levels. Predominant is the post-transcriptional control, for example its translation is regulated by the RNA-binding protein Hfq and small regulatory RNAs and its proteolysis is controlled by the response regulator RssB and the ClpXP protease (Hengge-Aronis, 2000 ). In addition, RpoS activity is affected by Crl, a 15 kDa protein that is conserved among the Enterobacteriaceae (for example, it is present in the complete genome sequences of Salmonella enterica, Salmonella typhimurium LT2, Yersinia pestis and Vibrio cholerae) but shares no homology with other proteins. Crl was discovered because it is required for the RpoS-dependent expression of the csgBA operon encoding the fibronectin-binding curli fimbria (Crl=curli) (Olsen et al., 1989 ; Arnquist et al., 1992 ). Genetic analyses revealed that Crl stimulates to various extents all RpoS-dependent transcription processes and that Crl acts either in concert with RpoS or upstream of it in the same pathway (Pratt & Silhavy, 1998 ).

Here the expression level of activated bgl operon derivatives and bgllacZ reporter constructs were tested in strain CSH50 which carries a large deletion in the lac region, {Delta}(gptlac), and derivatives that carry the smaller {Delta}(argFlac) deletion or the wild-type gpt–pro–lac ({Delta}lacZ) region. In strain CSH50, activated bgl operon alleles are expressed at high levels. However, their expression is low in the other two deletion derivatives. In these strains the expression is increased up to 50-fold when a rpoS mutation is introduced (see also Dole et al., 2002 ). RpoS has no effect in CSH50. A complementation experiment shows that Crl, which is encoded in the gptargF region, is required for the RpoS-dependent repression of the bgl operon and responsible for these strain-dependent variations in the expression level of the bgl operon.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Strains and plasmids.
Plates and liquid media were used as described by Dole et al. (2002) . The genotypes of the E. coli K-12 strains used in this study are listed in Table 1. All strains are derivatives of CSH50 (Miller, 1972 ). [The {Delta}(prolac)X111 deletion present in CSH strains was mapped and renamed as {Delta}(gptlac), Genetic Stock Center, http://cgsc.biology.yale.edu/] Transductions were performed with phage T4GT7 (Wilson et al., 1979 ). For transduction of rpoS359::Tn10, cells were selected on LB tetracycline (12 µg ml-1) plates and for transduction of {Delta}(argFlac) (Pro+) cells, minimal glucose B1 plates were used. Integrations of bgllacZ reporter constructs into the chromosomal phage lambda attachment site attB using plasmids pKESK11, pKESK12, pKESK13, pKESK14, pKESK20 or pKESD13 were performed and confirmed by PCR as described (Diederich et al., 1992 ; Dole et al., 2002 ). Plasmids were constructed according to standard techniques (Sambrook et al., 1989 ). The structure of plasmids pKESD13 and pKESD20 is similar to pKESK11 (Dole et al., 2002 ). Plasmid pKESD13 carries the lac operator promoter and the lacZ gene cloned into a pACYC backbone, including the kanamycin-resistance neo gene, attP, and the spectinomycin-resistance aadA gene. Plasmid pKESK20 carries the bgl promoter activated by integration of IS1 (allele bgl-R1243) (Schnetz & Rak, 1992 ), bgl terminator mutant t1-L, bglG and lacZ. Plasmid pKESK19 carries the crl gene cloned under control of the constitutive antisense promoter of the tetracycline resistance gene (Stüber & Bujard, 1981 ). For its construction the crl gene fragment was amplified by PCR and cloned into pBR322, replacing the bla gene. In addition, a kanamycin resistance gene cassette was inserted into the tet gene. PCR fragments were sequenced after cloning. Details of constructions and compiled sequences of the plasmids are available upon request.


View this table:
[in this window]
[in a new window]
 
Table 1. E. coli K-12 strains

 
Phospho-ß-glucosidase and ß-galactosidase assays.
These were performed as described by Miller (1972) , Schnetz & Rak (1988) and Dole et al. (2002) . Cells were grown in M9 medium containing 1% glycerol and 0·66% Casamino acids (Difco). For induction of the bgl operon, 0·5% (w/v) salicin was added in the phospho-ß-glucosidase assays. Cultures were inoculated from non-induced fresh overnight cultures grown in the same medium and cells were harvested after approximately 3 h growth at 37 °C at OD600 of 0·5. Enzyme activities were determined at least three times from at least two independent transformants or integration derivatives. Standard deviations were less than 10% in case of the ß-galactosidase assays and less than 20% in case of the phospho-ß-glucosidase assays.


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Expression levels of activated bgl operon derivatives are affected by deletions in the lac operon region
The expression levels of the wild-type and activated bgl operon alleles were tested in a set of isogenic CSH50 derivatives including a {Delta}(gptlac), a {Delta}(argFlac) and a {Delta}lacZ deletion (Fig. 1). The activated alleles of the bgl operon that were tested include a mutant carrying an improved CRP-binding site or an integration of IS1 (Fig. 2). The expression level of the bgl operon was determined in phospho-ß-glucosidase assays from exponential-phase cultures grown in minimal M9 glycerol medium containing salicin as inducer. Phospho-ß-glucosidase is encoded by the third gene of the bgl operon (bglB).



View larger version (7K):
[in this window]
[in a new window]
 
Fig. 1. Chromosomal map of the gpt–prolac operon region. Expression of the bgl operon was tested in strain background CSH50 (Miller, 1972 ) carrying a large deletion in the lac operon region {Delta}(gptlac), and CSH50-derived strains that carry a smaller deletion in the lac region, {Delta}(argFlac), or a deletion of the lacZ gene only (Dole et al., 2002 ). Map positions are given in kbp (gpt is at 256 kbp, crl is at 257·8 kbp, proBA is at 260·0 kbp, argF is at 289·5 and lac is at 365·5 kbp).

 


View larger version (18K):
[in this window]
[in a new window]
 
Fig. 2. RpoS and Crl are mutually required for repression of activated bgl operon alleles. Expression levels of (a) the wild-type bgl operon and alleles activated by (b) improvement of the CRP-binding site (CRP+, a C to T exchange at position -66 relative to the transcription start site), and (c) integration of IS1 (bgl::IS1-R1243) (Dole et al., 2002 ) are given in units of phospho-ß-glucosidase activity. Phospho-ß-glucosidase is encoded by BglB. BglF encodes the ß-glucoside specific permease enzyme IIBgl. Stem–loop structures t1 and t2 are rho-independent terminators at which BglG mediates antitermination. Expression levels were determined in CSH50 and derivatives, which differ in the size of the deletion in the lac operon region [{Delta}(gptlac), {Delta}(argFlac) and {Delta}lacZ]. Also tested were rpoS359::Tn10 (Lange & Hengge-Aronis, 1991 ) mutants of CSH50 and its derivatives (Table 1). +Crl indicates expression levels of transformants of the crl expressing plasmid pKESK19. Strains used were: (a) S524, S1071, S484, S1489, CSH50 and S1484; (b) S544, S1079, S1491, S1499, S432 and S1497; (c) S1245, S1250, S1493, S1501, R1243 and S1495.

 
As expected, the expression level of the silent wild-type bgl operon was low in all strains tested (Fig. 2a). However, the expression level of activated alleles varied significantly with the CSH50 strain and its deletion derivatives. These alleles are expressed at low levels in the {Delta}lacZ and {Delta}(argFlac) strains (Fig. 2). However, when the isogenic {Delta}(gptlac) strain was used, the activated alleles were expressed at high levels. Expression of the IS1-activated allele is higher than that of the allele carrying an improved CRP-binding site. Low expression of activated alleles in the {Delta}lacZ strain when cells are grown in medium containing glycerol was observed previously (Dole et al., 2002 ). Glycerol causes some catabolite repression of the operon (Lopilato & Wright, 1990 ). Irrespective of the background, all strains carrying an activated operon show a Bgl-positive phenotype on salicin indicator plates. In the {Delta}lacZ and {Delta}(argFlac) strain backgrounds, the introduction of a mutation in rpoS caused an up to 50-fold increase in the expression of the activated alleles (Fig. 2) (see also Dole et al., 2002 ), while RpoS had no further effect in the {Delta}(gptlac) strain (Fig. 2). This indicated that a gene in the gptargF region is required for repression of the bgl operon by RpoS (Fig. 1).

Crl is required for the RpoS-mediated repression of the bgl operon
The crl gene maps in the gptargF region and thus was an obvious candidate for the observed difference. To test whether Crl complements the {Delta}(gptlac) strain, plasmid pKESK19 carrying the crl gene under control of the constitutive antisense promoter of the tetracycline resistance gene (Stüber & Bujard, 1981 ) was constructed. In transformants of the {Delta}(gptlac) strain the repression of the activated bgl operon alleles by RpoS was restored (Fig. 2; +Crl). As a control, in the RpoS-negative {Delta}(gptlac) strain providing Crl in trans caused a less than twofold decrease [Fig. 2; {Delta}(gptcrllac) rpoS +Crl]. This demonstrates that Crl is required for the RpoS-mediated repression.

Crl is required for the RpoS-mediated repression of bgllacZ reporter constructs
Next a chromosomal bgllacZ reporter system was used in which expression is independent of the ß-glucoside specific regulation by BglG-mediated antitermination due to a mutation (t1-L) of terminator t1 (Fig. 3) (Dole et al., 2002 ). In this system the changes in the transcription rate are not amplified by BglG-mediated antitermination and thus correlate with changes in the levels of expressed ß-galactosidase (see Introduction). RpoS has an approximately threefold effect on the expression level of the bgllacZ reporter gene system (Dole et al., 2002 ). Analysed were bgllacZ fusions that carry the wild-type bgl promoter (Fig. 3a) or activated alleles, including the mutants that carry the improved CRP-binding site (Fig. 3b) or the integration of IS1 (Fig. 3c). In addition, mutants carrying an integration of IS5 (Fig. 3d) and a deletion of the upstream silencer were tested (Fig. 3e) (Dole et al., 2002 ). Expression levels are given in units of ß-galactosidase activity and thus the values are higher than in the phospho-ß-glucosidase assay shown in Fig. 2. Similar to the results obtained using the bgl operon, the bgllacZ constructs were repressed by RpoS when Crl was present ({Delta}lacZ) but not when Crl was missing ({Delta}gptcrllac) (Fig. 3). Repression of the activated alleles by RpoS is approximately two- to threefold (Fig. 3). Again, providing Crl in trans restored repression by RpoS in the {Delta}gptcrllac strain, but had no effect in the rpoS mutant of this strain (Fig. 3). As a control, the expression of a lacOPlacZ fragment integrated into attB varied only moderately in the various strain backgrounds (Fig. 3).



View larger version (31K):
[in this window]
[in a new window]
 
Fig. 3. Crl is required for the RpoS-mediated repression of BglG-independent bgllacZ reporter constructs. Chromosomally encoded bgllacZ reporter constructs (Dole et al., 2002 ) carrying the wild-type bgl promoter (a) or alleles activated by (b) improvement of the CRP-binding site (CRP+), (c) integration of IS1, (d) integration of IS5 and (e) a deletion of the upstream silencer ({Delta}2 in Table 1) were tested in the CSH50 {Delta}lacZ derivative and in CSH50, {Delta}(gptcrllac) [rpoS+], and the rpoS359::Tn10 mutant derivatives of these strains. Crl was provided in trans (+Crl) using plasmid pKESK19. All the bgllacZ constructs carry mutation t1-L of the first terminator, a 3 bp exchange in the left stem of the inverted repeat forming the terminator, which renders expression independent of BglG-mediated antitermination. As a control (f) expression of a lacOPlacZ construct integrated into attB (Diederich et al., 1992 ; Dole et al., 2002 ) was also tested. Strains used were: (a) S1142 ({Delta}lacZ), S1155 ({Delta}lacZ rpoS359::Tn10), S1163 ({Delta}gpt–lac), and S1171 ({Delta}(gptlac) rpoS359::Tn10); (b) S1144, S1157, S1165 and S1173; (c) S1503, S1507, S1505 and S1508; (d) S1146, S1159, S1167 and S1512; and (e) S1274, S1276, S308 and S1305.

 
Role of RpoS and Crl in bgl operon regulation
The bgl operon is regulated by various pleiotropic regulators including H-NS and CRP, as essential factors. In addition, rather moderate changes (approximately two- to threefold) in the transcription rate of the operon can be amplified post-transcriptionally into large changes in the expression level (Dole et al., 2002 ). Therefore, the expression level of the bgl operon and the effect of any non-essential factor (for example RpoS and Crl) involved in bgl operon regulation may vary significantly depending on the strain background and the growth conditions that are used. RpoS causes a moderate repression of the bgl promoter of activated bgl operon derivatives (Dole et al., 2002 ). The finding that the RpoS-mediated repression requires Crl supports earlier findings that Crl stimulates RpoS activity. How Crl stimulates RpoS is unknown. Interestingly, RpoS is required for silencing of a bgllacZ fusion by a truncated H-NS protein in a {Delta}(gptlac), i.e. Crl-negative, strain (Ohta et al., 1999 ). Likewise, the contribution of Crl to the RpoS-mediated transcription of the Curli (csgBA) operon varies with the strain background (Olsen et al., 1989 ; Provence & Curtiss, 1992 ). This indicates that the enhancement of RpoS-mediated regulation by Crl may depend on H-NS and other pleiotropic regulators.


   ACKNOWLEDGEMENTS
 
I thank Sandra Kühn for excellent technical assistance, and rotation students Bettina Pies and Monika Przybytek for contribution to the experiments. The work was funded by the Deutsche Forschungsgemeinschaft through SFB274.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Arnquist, A., Olsén, A., Pfeifer, J., Russell, D. G. & Normark, S. (1992). The Crl protein activates cryptic genes for curli formation and fibronectin binding in Escherichia coli HB101. Mol Microbiol 6, 2443-2452.[Medline]

Caramel, A. & Schnetz, K. (1998). Lac and Lambda repressor relieve silencing of the Escherichia coli bgl promoter: activation by alteration of a repressing nucleoprotein complex. J Mol Biol 284, 875-883.[Medline]

Caramel, A. & Schnetz, K. (2000). Antagonistic control of the E. coli bgl promoter by FIS and CAP in vitro. Mol Microbiol 36, 85-92.[Medline]

Defez, R. & de Felice, M. (1981). Cryptic operon for ß-glucoside metabolism in Escherichia coli K12: genetic evidence for a regulatory protein. Genetics 97, 11-25.[Abstract/Free Full Text]

Diederich, L., Rasmussen, L. J. & Messer, W. (1992). New cloning vectors for integration into the lambda attachment site attB of the Escherichia coli chromosome. Plasmid 28, 14-24.[Medline]

Dole, S., Kühn, S. & Schnetz, K. (2002). Post-transcriptional enhancement of Escherichia coli bgl operon silencing by limitation of BglG-mediated antitermination at low transcription rates. Mol Microbiol 43, 217-226.[Medline]

Free, A., Williams, R. M. & Dorman, C. J. (1998). The StpA protein functions as a molecular adapter to mediate repression of the bgl operon by truncated H-NS in Escherichia coli. J Bacteriol 180, 994-997.[Abstract/Free Full Text]

Free, A., Porter, M. E., Deighan, P. & Dorman, C. J. (2001). Requirement for the molecular adapter function of StpA at the Escherichia coli bgl promoter depends upon the level of truncated H-NS protein. Mol Microbiol 42, 903-918.[Medline]

Giel, M., Desnoyer, M. & Lopilato, J. (1996). A mutation in a new gene, bglJ, activates the bgl operon in Escherichia coli K-12. Genetics 143, 627-635.[Abstract/Free Full Text]

Hengge-Aronis, R. (2000). The general stress response in Escherichia coli. In Bacterial Stress Responses , pp. 161-178. Edited by G. Storz & R. Hengge-Aronis. Washington, DC:American Society for Microbiology.

Higgins, C. F., Dorman, C. J., Stirling, D. A., Waddell, L., Booth, I. R., May, G. & Bremer, E. (1988). A physiological role for DNA supercoiling in the osmotic regulation of gene expression in S. typhimurium and E. coli. Cell 52, 569-584.[Medline]

Lange, R. & Hengge-Aronis, R. (1991). Identification of a central regulator of stationary-phase gene expression in Escherichia coli. Mol Microbiol 5, 49-59.[Medline]

Lopilato, J. & Wright, A. (1990). Mechanisms of activation of the cryptic bgl operon of Escherichia coli K-12. In The Bacterial Chromosome , pp. 435-444. Edited by K. Drlica & M. Riley. Washington, DC:American Society for Microbiology.

Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory.

Ohta, T., Ueguchi, C. & Mizuno, T. (1999). rpoS function is essential for bgl silencing caused by C-terminally truncated H-NS in Escherichia coli. J Bacteriol 181, 6278-6283.[Abstract/Free Full Text]

Olsen, A., Jonsson, A. & Normark, S. (1989). Fibronectin binding mediated by a novel class of surface organelles on Escherichia coli. Nature 338, 652-655.[Medline]

Pratt, L. A. & Silhavy, T. J. (1998). Crl stimulates RpoS activity during stationary phase. Mol Microbiol 29, 1225-1236.[Medline]

Provence, D. L. & Curtiss, R.III (1992). Role of crl in avian pathogenic Escherichia coli: a knockout mutation of crl does not affect hemagglutination activity, fibronectin binding, or Curli production. Infect Immun 60, 4460-4467.[Abstract]

Reynolds, A. E., Felton, J. & Wright, A. (1981). Insertion of DNA activates the cryptic bgl operon of E. coli K12. Nature 293, 625-629.[Medline]

Reynolds, A. E., Mahadevan, S., LeGrice, S. F. J. & Wright, A. (1986). Enhancement of bacterial gene expression by insertion elements or by mutation in a CAP-cAMP binding site. J Mol Biol 191, 85-95.[Medline]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Schaefler, S. & Maas, W. K. (1967). Inducible system for the utilization of ß-glucosides in Escherichia coli. II. Description of mutant types and genetic analysis. J Bacteriol 93, 264-272.[Medline]

Schnetz, K. (1995). Silencing of Escherichia coli bgl promoter by flanking sequence elements. EMBO J 14, 2545-2550.[Abstract]

Schnetz, K. & Rak, B. (1988). Regulation of the bgl operon of Escherichia coli by transcriptional antitermination. EMBO J 7, 3271-3277.[Abstract]

Schnetz, K. & Rak, B. (1992). IS5: a mobile enhancer of transcription in Escherichia coli. Proc Natl Acad Sci USA 89, 1244-1248.[Abstract]

Singh, J., Mukerji, M. & Mahadevan, S. (1995). Transcriptional activation of the bgl operon of E. coli: negative regulation by DNA structural elements near the promoter. Mol Microbiol 17, 1085-1092.[Medline]

Stüber, D. & Bujard, H. (1981). Organization of transcriptional signals in plasmids pBR322 and pACYC184. Proc Natl Acad Sci USA 78, 167-171.[Abstract]

Tsui, H.-C. T., Leung, H.-C. E. & Winkler, M. E. (1994). Characterization of broadly pleiotropic phenotypes caused by an hfq insertion mutation in Escherichia coli K-12. Mol Microbiol 13, 35-49.[Medline]

Ueguchi, C., Ohta, T., Seto, C., Suzuki, T. & Mizuno, T. (1998). The leuO gene product has a latent ability to relieve the bgl silencing in Escherichia coli. J Bacteriol 180, 190-193.[Abstract/Free Full Text]

Wilson, G. G., Young, K. Y. K., Edlin, G. J. & Konigsberg, W. (1979). High-frequency generalised transduction by bacteriophage T4. Nature 280, 80-82.

Received 25 March 2002; revised 3 May 2002; accepted 7 May 2002.