The curli biosynthesis regulator CsgD co-ordinates the expression of both positive and negative determinants for biofilm formation in Escherichia coli

Eva Brombacher1, Corinne Dorel2, Alexander J. B. Zehnder1 and Paolo Landini1

1 Swiss Federal Institute of Environmental Technology (EAWAG), Überlandstrasse 133, CH-8600 Dübendorf, Switzerland
2 Unité de Microbiologie et Génétique (CNRS UMR 5122), Institut National des Sciences Appliquées de Lyon, 10 rue Dubois, 69622 Villeurbanne Cedex, France

Correspondence
Paolo Landini
landini{at}eawag.ch


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Production of curli, extracellular structures important for biofilm formation, is positively regulated by OmpR, which constitutes with the EnvZ protein an osmolarity-sensing two-component regulatory system. The expression of curli is cryptic in most Escherichia coli laboratory strains such as MG1655, due to the lack of csgD expression. The csgD gene encodes a transcription activator of the curli-subunit-encoding csgBA operon. The ompR234 up-mutation can restore csgD expression, resulting in curli production and increased biofilm formation. In this report, it is shown that ompR234-dependent csgD expression, in addition to csgBA activation during stationary phase of growth, stimulates expression of the yaiC gene and negatively regulates at least two other genes, pepD and yagS. The promoter regions of these four genes share a conserved 11 bp sequence (CGGGKGAKNKA), necessary for csgBA and yaiC regulation by CsgD. While at both the csgBA and yaiC promoters the sequence is located upstream of the promoter elements, in both yagS and pepD it overlaps either the putative -10 sequence or the transcription start point, suggesting that CsgD can function as both an activator and a repressor. Adhesion experiments show that csgD-independent expression of both yagS and pepD from a multicopy plasmid negatively affects biofilm formation, which, in contrast, is stimulated by yaiC expression. Thus it is proposed that CsgD stimulates biofilm formation in E. coli by contemporary activation of adhesion positive determinants (the curli-encoding csg operons and the product of the yaiC gene) and repression of negative effectors such as yagS and pepD.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In natural environments, micro-organisms, either prokaryotic or eukaryotic, are often found as organized communities growing on surfaces, rather than as free-swimming organisms. In such communities, usually referred to as biofilms, microbial cells are enclosed by a matrix mainly composed of polysaccharides (Costerton et al., 1987, 1995). Biofilms can harbour different species of micro-organisms, able to establish mutual interactions through complex chemical signalling known as quorum sensing (Fuqua et al., 1994; Costerton et al., 1995). A mature biofilm can display complex structural and functional architecture, with cells located in different areas showing different biochemical and morphological properties (Lawrence et al., 1991; Costerton et al., 1994). Cells growing in biofilm are more resistant to biocides, antibiotics, desiccation and oxidative stress than are their planktonic counterparts (Hoyle & Costerton, 1991; Finlay & Falkow, 1997; Stewart, 2001). Finally, biofilms are more resistant to predation by protozoa and to attack by bacteriophages (Weiner et al., 1995). These observations strongly suggest that biofilms can be considered as a ‘resistance form’ of growth, able to withstand stress conditions more efficiently than planktonic cells. Biofilm-growing cells display different patterns of gene expression when compared with planktonic cells (Prigent-Combaret et al., 1999). Expression of determinants for initial adhesion is likely to take place already in planktonic cells in response to environmental conditions (Pratt & Kolter, 1998; Prigent-Combaret et al., 1999). The adhesion event itself stimulates the expression of biofilm-specific genes, such as the algC operon in Pseudomonas (Davies & Geesey, 1995). As cells start growing on the colonized surface, their local concentration increases, leading to activation of quorum-sensing-dependent genes, several of which encode proteins involved in the production of the exopolymeric substance typical of mature biofilm (Swift et al., 1998; Miller & Bassler, 2001).

Several extracellular structures including lipopolysaccharide (Jucker et al., 1998; Williams & Fletcher, 1996), flagella and pili (Pratt & Kolter, 1998) and curli (Olsen et al., 1989; Vidal et al., 1998) are involved in initial adhesion of bacterial cells to a solid surface and/or in subsequent steps of biofilm formation. Curli are fibrils made of protein present in Escherichia coli and Salmonella spp., where they are known as thin aggregative fimbriae (Romling et al., 1998b). In E. coli, curli promote both initial adhesion and cell–cell interaction (Prigent-Combaret et al., 2000). Genes involved in curli production are clustered in two divergent operons: the csgBA operon, encoding the structural components of curli; and the csgDEFG operon, encoding genes necessary for export of the curli subunit and stabilization of the fibres (the csgE–G genes; Chapman et al., 2002) and csgD, a transcription factor belonging to the luxR family necessary for csgBA expression (Hammar et al., 1995). Despite this gene organization being extremely conserved among E. coli and Salmonella enterica strains (Romling et al., 1998a), the curli-encoding genes are not expressed in many laboratory strains of E. coli, due to silencing of the csgD promoter (Hammar et al., 1995). Expression of the csgD promoter is affected, either positively or negatively, by several transcriptional regulators, including rpoS, crl, hns and ompR (Romling et al., 1998b; Prigent-Combaret et al., 2001). Mutations in these regulatory genes, such as the ompR234 mutation, which results in more efficient ompR-dependent activation of the csgD promoter, can stimulate curli production and biofilm formation in laboratory strains (Vidal et al., 1998; Prigent-Combaret et al., 2001). In this report, we investigated the effects of the ompR234 mutation on global transcription regulation in the E. coli MG1655 laboratory strain and on the ability of this strain to attach to solid surfaces. We show that ompR234-dependent activation of csgD results in activation of the csgBA and yaiC operons in the stationary phase of growth. An 11 bp sequence conserved in the two promoters appears to be necessary for csgD activation. This putative binding site is conserved in other promoters, two of which, pepD and yagS, are negatively regulated by csgD. Thus we conclude that ompR-dependent activation of csgD results in a co-ordinated response leading to regulation of at least four independent genes or operons.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Strains and plasmids.
The bacterial strains and plasmids used in this report are listed in Table 1. Unless otherwise stated, the bacteria were grown at 28 °C in M9/Glucose medium supplemented with 5 % L broth (Miller et al., 1972). To obtain the different luxAB reporter plasmids, promoter regions of interest were amplified by PCR, using the following primers (sequences given in Table 2): for csgD, primers d1 and d2; for csgB, b1 and b2; for yaiC, y1 and y2. The PCR products were purified and cloned into the pGEM-T Easy plasmid. The insert carrying the promoter regions was then subcloned into pJAMA8 (Jaspers et al., 2000) using the SphI and XbaI restriction sites (underlined in the primer sequences given in Table 2). Mutagenesis of the putative CsgD-binding site was achieved by amplification of the csgB and yaiC promoter regions using primers in which the putative binding site was substituted with a random sequence. The primers used for mutagenesis were mutcsgB1, mutcsgB2, mutyaiC1 and mutyaiC2 (see Table 2). Purified PCR products were cloned into the pGEM-T Easy plasmid. The inserts carrying the mutagenized promoter regions were subcloned as SphI–XbaI fragments into the pJAMA8 reporter plasmid. Wild-type and mutagenized promoters were sequenced and luciferase assays were performed as described below.


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

 

View this table:
[in this window]
[in a new window]
 
Table 2. Primers used in the study

 
Low-level, csgD-independent expression of adhesion-related genes was achieved by PCR-amplification of the various ORFs (csgB, csgD, csgG, pepD, yagS, yaiC) followed by direct cloning into the pGEM-T Easy plasmid. Primers used were csgB1, csgB2, csgD1, csgD2, csgG1, csgG2, yaiC1, yaiC2, yagS1, yagS2, pepD1 and pepD2 (see Table 2). Expression of the cloned genes relies on low-level expression from promoter-like sequences in the vector plasmid, upstream of the cloning site.

Biofilm and adhesion assays.
Determination of biofilm formation in microtitre plates was carried out as in Dorel et al. (1999). The ratio between surface-attached and unattached bacteria was estimated by measuring the OD600. At least three independent assays were performed and means were calculated. To determine initial attachment to a solid surface, we used the sand column assay described in Simoni et al. (1998). Bacteria were grown to either mid-exponential or stationary phase in M9/Glucose supplemented with 5 % L broth, harvested, washed and resuspended in PBS to an OD280=0·8 (initial absorbance, A0). The bacterial suspension was loaded onto a column filled with fine sea sand grains (9 g sand) and fractions were collected at the outlet. The ratio between the number of bacteria in the flow through and in the initial suspension (A/A0, where A is the absorbance of the bacterial suspension at the output of the column) was determined spectrophotometrically. The percentage of cells adhering to the column is calculated as (1-A/A0)x100. Microscopical analysis of the column sand grains shows that bacteria attach as single cells in the conditions used in our experiments (data not shown).

Global transcription experiments.
The Panorama gene array system (Sigma) was used to compare global gene expression in the MG1655 (wild-type) and the PHL628 (ompR234) strains. Both strains were grown for 14 h in M9/Glucose supplemented with 5 % L broth at 28 °C. Total RNA was isolated as described by Sambrook et al. (1989) and 2 µg was subjected to RT-PCR according to the manufacturer's instructions in the presence of 20 µCi (740 kBq) per reaction of [{alpha}-33P]ATP in a final volume of 50 µl. The mixture was loaded onto Sephadex G-50 to remove the unincorporated nucleotides; the products of the RT-PCR were eluted in 200 µl TE buffer (10 mM Tris/HCl, pH 8·0; 1 mM EDTA) and directly used for hybridization. The gene arrays were exposed to a phosphorimager (Molecular Dynamics) and analysed as described in Tao et al. (1999). The intensity of each spot was quantified and the local background was subtracted using the Array Vision software (Research Imaging). The duplicate spots were averaged and expressed as a percentage of the total of intensities of all the spots on the DNA array. This value was used to calculate the ratio of mRNA levels of PHL628 and MG1655. The correlation coefficients of the percentage intensities determined individually for the duplicate spots on a single blot ranged from 0·953 to 0·990. Two independent experiments were performed; we considered as significant differences in expression between MG1655 and PHL628 higher than 2·5-fold. Sequence analysis and searches were carried out using the Colibri Web Server (http://genolist.pasteur.fr/Colibri/genome.cgi).

Other in vivo assays.
For luciferase assays, bacterial strains MG1655 and PHL628 containing the different reporter plasmids were grown overnight. These cells were either directly assessed for reporter gene activity, or diluted 1 : 200 in fresh medium for the time-course experiments, where samples were taken at different time points, starting at OD600=0·1. The samples were adjusted to an OD600 of 0·05–0·1 in PBS buffer. Twenty microlitres of this solution was tested for luciferase activity by adding 200 µl PBS containing n-decanal to a final concentration of 2 nM. Measurement of relative light units (RLU) was conducted by a 2 s pre-measurement delay followed by a 3 s measurement after addition of the substrate in a MicroLumat LB 96 P luminometer (Berthold Technologies). Results are expressed as RLU per OD600 of the tested bacterial samples.

Primer extension was carried out as described in Prigent-Combaret et al. (2001). RNA from stationary phase cultures of either MG1655 or PHL628 (wherever expression of the gene of interest was higher) transformed with pJPcsgB, pJPyaiC, pJPpepD or pJPyagS was used for transcript analysis. We used the 5'-GATAAGTGAGAAGGAAGTTTC-3' primer, which anneals to the coding strand between 147 and 167 nucleotides downstream of the luxAB gene transcription start. The primer was labelled at the 5' end with fluorescent dye IRD-800 (MWG Biotech). Twenty micrograms of total RNA (extracted as for the gene array experiments) was used for each assay. The start site was determined using a sequencing ladder of the corresponding gene as molecular mass marker.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Effects of the ompR234 mutation on adhesion properties
In a previous report, we showed that an up-mutation in the ompR gene, ompR234, increases biofilm formation and attachment to solid surfaces through the activation of the csgD gene, which, in turn, up-regulates the curli biosynthetic operons (Prigent-Combaret et al., 2001). To verify if the ompR234-dependent ability to adhere to solid surfaces is mediated solely by increased curli expression, we inactivated csgA, encoding the main curli subunit, in both MG1655 (wild-type) and PHL628 (ompR234), and tested the ability of the csgA derivatives to attach to sand grains in a column assay. The sand column assay mimics dissemination of micro-organisms in soil, an important factor in contamination of drinking water; this method allows rapid and precise determination of initial attachment to a solid surface by non-growing cells (Simoni et al., 1998; Prigent-Combaret et al., 2001; Landini & Zehnder, 2002). While the ompR234 mutation has little or no effect on adhesion in cells harvested during exponential phase of growth, it strongly stimulates adhesion in cells that reached stationary phase (Fig. 1), confirming our previous results (Prigent-Combaret et al., 2001). As expected, inactivation of the csgA gene results in a decrease of the ability to attach to sand by the ompR234 strain. However, the effects of the ompR234 mutation are not completely reverted by disruption of csgA, strongly suggesting that the increased adhesion properties of the ompR234 strain do not depend solely on curli production. Additional determinants for adhesion might be either regulated directly by the ompR gene or, as for the csgBA curli biosynthetic operon, controlled by the csgD regulatory gene. Indeed, a recent paper showed that agfD, the Salmonella homologue of csgD, controls at least two operons encoding genes involved in the multicellular phenotype: the agfBA operon, homologue of csgBA, and adrA, which controls cellulose biosynthesis (Romling et al., 2000; Zogaj et al., 2001). Thus we tested expression of the yaiC gene, the adrA homologue in E. coli, in either the MG1655 (wild-type) or the PHL628 (ompR234) strain, and compared it with expression of the csgBA promoter (Fig. 2). The results of this experiment suggest that yaiC expression also depends on csgD in E. coli. While csgD up-regulation by ompR234 occurs in mid-exponential phase (Prigent-Combaret et al., 2001), full activation of both csgBA and yaiC can only be detected in stationary phase (Fig. 2a). These observations show that ompR234-dependent expression of csgD and activation of the csgD-dependent operons take place at very different stages in cell growth, strongly suggesting that the CsgD protein might be activated post-transcriptionally in stationary phase. Both the csgBA and the yaiC promoters share a similar 11 bp sequence, respectively CGGGTGAGTTA in csgBA and CGGGTGAGCTA in yaiC (Fig. 3b). Interestingly, in the yaiC promoter, an almost perfect inverted repeat of the CGGGTGAGCTA sequence is located 6 bp downstream. Conservation of this 11 bp sequence in both csgB and yaiC points to a possible regulatory role; indeed, substitutions in this sequence [to either ATTCCATTAGG (PmutcsgB1) or ATTCCATTTTA (PmutcsgB2) in the csgB and to ATTCCATTAGG (PmutyaiC1) in the yaiC promoters] result in total loss of activation by CsgD (Fig. 2b). Substitution of the downstream arm of the inverted repeat (TGGCTCACCCG) in the yaiC promoter to ATTGCATTAGG also results in loss of yaiC activation in the ompR234 mutant strain (Fig. 2b). The results of the luciferase assays were confirmed by quantification of mRNAs using quantitative PCR for both wild-type and mutant promoters (data not shown). ompR234-dependent activation of either promoter was completely abolished in the ompR234 csgD double mutant PHL1087 strain (data not shown), confirming that the expression of both csgBA and yaiC is strictly csgD-dependent. In the PHL1087 strain, mutations in the conserved 11 bp sequence did not affect expression of the csgB promoter (data not shown). To confirm that the sequence we identified indeed serves as a CsgD-binding site, we attempted to purify the CsgD protein for in vitro studies. Unfortunately, we have failed to obtain any active CsgD protein thus far.



View larger version (26K):
[in this window]
[in a new window]
 
Fig. 1. Adhesion to sand column by MG1655 (wild-type), PHL628 (ompR234), PHL856 (csgA) and PHL857 (ompR234 csgA) during either exponential phase (left-hand side, OD600=0·25) or stationary phase (right-hand side, OD600>1·1). Results are a mean of four independent experiments. Standard errors are shown.

 


View larger version (18K):
[in this window]
[in a new window]
 
Fig. 2. In vivo transcription of csgD-dependent genes in MG1655 (wild-type) and PHL628 (ompR234) strains. (a) In vivo transcription of csgB (left block) and yaiC (right block) measured in either the exponential (OD600=0·25) or the stationary (OD600>1·0) phase of growth. White bar, expression in MG1655 during exponential phase; light-grey bar, expression in PHL628 during exponential phase; hatched bar, expression in MG1655 during stationary phase; dark-grey bar, expression in PHL628 during stationary phase. At least three independent experiments were performed with very similar results. Data shown are from a typical experiment: light measurement was repeated three times, and the mean and standard deviation of the values are indicated. (b) In vivo transcription of wild-type and mutant csgB and yaiC promoters in either the MG1655 (wild-type; white bars) or the PHL628 (ompR234; grey bars) strain. Luciferase activity was measured in stationary phase (OD600>1·0). Mutant promoters are described in the text. At least three independent experiments were performed with very similar results. Data shown are from a typical experiment: light measurement was repeated three times, and the mean and standard deviation of the values are indicated. RLU, relative light units.

 


View larger version (31K):
[in this window]
[in a new window]
 
Fig. 3. (a) Determination of the transcription start points for the csgB, yaiC and pepD promoters. The start site was determined using a sequencing ladder of the gene of interest as a molecular mass marker. (b) Location of the putative csgD-binding sites relative to transcription starts and/or putative promoter elements in csgD-dependent promoters. The transcription start points are shown in bold. The putative promoter -10 and -35 elements are shown in italics and underlined. The black boxes indicate the putative CsgD-binding site (CGGGKGAKNKA).

 
Effects of the ompR234 allele and of csgD on global gene expression
To identify possible additional adhesion determinants regulated by either the ompR234 allele or the csgD gene, we compared global transcription levels in MG1655 (wild-type) and in PHL628 (ompR234) using a gene array system. Since adhesion properties are only affected by ompR234 during stationary phase (Fig. 1; Prigent-Combaret et al., 2001), we performed gene array experiments on overnight cultures grown at 28 °C. The results of two independent gene array experiments are shown in Table 3. We considered as significant a difference in gene expression between wild-type and ompR234 strains higher than 2·5-fold (Tao et al., 1999). The ompR234 mutation results in the altered expression of 10 genes in stationary phase of growth. Of the six up-regulated genes (mean induction levels 3·6–6·1-fold; Table 3), three are genes of unknown function (yhiE, yjbR and ydjC); recT encodes a single-stranded DNA-binding protein able to promote renaturation of homologous DNA (Hall et al., 1993). Both the csgB and the csgA genes were expected to be differentially regulated since the csgBA operon is activated by csgD during stationary phase (Fig. 2; Prigent-Combaret et al., 2001). Lack of differential expression of the csgD gene is consistent with the results showing that ompR234-dependent stimulation of csgD only takes place during the exponential phase of growth (Prigent-Combaret et al., 2001). In contrast, yaiC, which in luciferase assays showed higher expression in the ompR234 strain, consistent with its being part of the csgD regulon, did not show any differential expression in gene array experiments. One explanation for this discrepancy might be possible lack of stability of yaiC RNA.


View this table:
[in this window]
[in a new window]
 
Table 3. Global transcription experiments

Genes differentially expressed in PHL628 (ompR234) compared to MG1655.

 
Four genes appear to be down-regulated (mean down-regulation levels 3–4·8-fold) in the PHL628 (ompR234) strain. Two of these genes encode proteins involved in nucleotide metabolism: thyA (thymidine synthetase) and yagS (putative xanthine dehydrogenase). The other down-regulated genes are glnS, encoding glutaminyl-tRNA synthetase, and pepD, encoding carnitine dipeptidase. The pepD and the yagS genes showed the highest degree of down-regulation (4·6- and 4·8-fold on average, respectively) in the ompR234 mutant. Interestingly, both genes possess 11 bp sequences (CGGGTGATCGA for pepD and CGGGGGAGATA for yagS) very similar to the regulatory sequences found in the csgB and yaiC promoters (Fig. 3b). The presence of the conserved sequence suggests that both pepD and yagS might be directly regulated by the CsgD protein. To confirm that the genes carrying the putative binding site for CsgD are indeed regulated by this protein, and not by OmpR234, we performed quantitative PCR experiments in strains expressing the CsgD protein independently of the ompR234 allele. The MG1655 strain (unable to express the csg operons) was transformed either with the pT7 plasmid or with the same plasmid carrying the csgD ORF. ompR234-independent CsgD expression resulted in activation of the csgB promoter, and in repression of pepD and yagS expression (data not shown), strongly suggesting that these genes are indeed directly controlled by CsgD.

Location of the putative CsgD-binding site at different promoters
The results of the previous experiments suggest that CsgD might function as either an activator (at the csgB and yaiC promoters) or a repressor (at the pepD and yagS promoters). Since the effect of a regulatory protein is determined by the location of its binding site relative to the promoter elements (Lloyd et al., 2001), we determined the position of the conserved 11 bp sequence necessary for CsgD-dependent regulation (Fig. 2b). The transcription start points for the various promoters were identified using primer extension; the transcription start points for the csgB, yaiC and pepD promoters are shown in Fig. 3. The yagS is the second gene in an operon controlled by the yagT promoter (Colibri Web Server, http://genolist.pasteur.fr/Colibri/). Despite the presence of possible promoter elements in the DNA region immediately upstream of the yagS ORF, we were not able to find any specific transcription start for the yagS gene (Fig. 3b). Our experiments confirmed the already reported transcription start site for csgB (Arnqvist et al., 1994); the conserved sequence necessary for activation by CsgD overlaps the -35 region (from -42 to -32; Fig. 3b), consistent with the location for an activator binding site (Busby & Ebright, 1994). The yaiC promoter possesses a -10 sequence perfectly matching the consensus for RNA polymerase, although in an unusual location relative to the transcription start point. We could identify no -35 sequence for yaiC, consistent with its role of positively controlled promoter (Fig. 3b). Unlike at the csgB promoter, the 11 bp sequence necessary for CsgD-dependent regulation is positioned between -70 and -60 relative to the transcription start site. This location would still be consistent with the possible role of CsgD as an activator (Busby & Ebright, 1994).

According to a previous study, pepD transcription in the exponential phase of growth is directed by two distinct promoters, called pepD1 and pepD2 (Henrich et al., 1990). Although we could confirm the presence of two different transcription start sites in the stationary phase of growth (Fig. 3), only the start site corresponding to pepD1 matched the one previously reported. In contrast, we find a second transcription start 71 bp downstream of pepD1 (39 bp downstream of the previously detected pepD2 start site). Thus, according to our results, the start point for pepD2 would be located within the conserved 11 bp sequence, consistent with a possible role of CsgD as a repressor at the pepD promoter.

Effects of csgD-regulated genes on biofilm formation
The csgD gene controls the expression of factors involved in biofilm formation, such as the curli operon and the adrA gene in Salmonella. Thus we investigated the possible involvement in biofilm formation by the other csgD-dependent genes identified in this study. Either pepD or yagS was cloned into the multicopy pGEM-T Easy plasmid to obtain low-level expression, independent of either CsgD or OmpR234. We transformed the MG1655 strain, which does not express csgD, with the plasmids carrying either pepD or yagS and tested their effects on biofilm formation (Fig. 4). As a control, we also expressed in a csgD-independent fashion known determinants for biofilm formation (csgB, csgG, yaiC and csgD itself). Expression of none of these genes resulted in significant effects on MG1655 growth rate (data not shown). As expected, expression of CsgD from pGEM-T Easy resulted in increased formation of biofilm by MG1655, while csgD-independent expression of either csgB or csgG had little or no effect on biofilm formation. This result was also expected, since production of curli requires concomitant expression of both the csgBA and the csgDEFG operons. In contrast, low-level csgD-independent expression of yaiC resulted in a significant increase of attached cells (from 30 to 52 % of total cells). Expression of either the yagS or the pepD genes resulted in the opposite effect, with a significant reduction in the number of attached cells (down to 19 and 17 %, respectively).



View larger version (12K):
[in this window]
[in a new window]
 
Fig. 4. Biofilm formation by MG1655 (wild-type) harbouring pGEM-T Easy derivatives carrying either csgD or csgD-dependent genes. MG1655 carrying the vector plasmid was used as control in this experiment (Co). The black line indicates the adhesion value obtained for the control. Data are the mean of four independent experiments; standard errors are shown.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Determinants for biofilm formation and for bacterial adhesion to solid surfaces are subject to complex regulation, involving several global regulatory genes. The OmpR regulator is directly involved in activation of curli expression, in concert with several other regulatory proteins such as RpoS, H-NS, CpxR, and other yet unknown factors (Romling et al., 1998b; Prigent-Combaret et al., 2001). The OmpR protein stimulates curli production through a regulatory cascade pathway, i.e. via activation of the csgD regulatory gene, which in turn up-regulates the csgBA operon, encoding the curli subunits. In several laboratory strains of E. coli, such as the MG1655 strain, expression of curli is cryptic, but the ompR234 up-mutation can restore curli production and stimulate adhesion to solid surfaces and biofilm formation (Prigent-Combaret et al., 2001). In this report, we showed that csgD-dependent curli activation is only partially responsible for the adhesion properties shown by the ompR234 mutant strain of MG1655 (Fig. 1), suggesting that ompR234-dependent genes other than curli contribute to biofilm formation. Indeed, we determined through gene array experiments (Table 3) that at least 10 genes are differentially expressed in the ompR234 mutant PHL628 compared to its MG1655 parental strain. These genes might be regulated by the OmpR234 protein directly, or through activation of csgD. Indirect evidence that csgD not only controls the csgBA operon, but also other genes, comes from the observations that csgD overexpression induces activation of the glyA gene (Chirwa & Herrington, 2003) and that mutations in the csgD gene affect nutritional requirements and ability to grow on different carbon sources in environmental isolates of E. coli (Uhlich et al., 2001). Romling et al. (2000) showed that the S. enterica serovar Typhimurium csgD homologue, the agfD gene, controls transcription of the adrA gene, involved in cellulose biosynthesis, in addition to curli expression. The adrA gene also plays a role in biofilm formation and in the development of the so-called multicellular phenotype in S. enterica serovar Typhimurium. In a csgA adrA double mutant, biofilm formation is completely abolished (Romling et al., 2000; Zogaj et al., 2001).

Similar to its Salmonella adrA homologue, the yaiC gene is regulated by csgD, and could be the main determinant for adhesion in the PHL857 (csgA ompR234 double mutant) strain. Both csgB and yaiC require an 11 bp conserved sequence for activation by CsgD (Fig. 2). This sequence was also found in two other genes negatively regulated in the ompR234 mutant strain, pepD and yagS (Table 3, Fig. 3). The conserved sequence (CGGGKGAKNKA) is totally unrelated to any so far reported OmpR-binding sites, and we propose that it might be the target sequence for CsgD. Thus CsgD might be able to act as both a positive (at the csgBA and yaiC promoters) and a negative (at the pepD and yagS promoters) transcription regulator. Interestingly, the locations of the putative CsgD-binding site differ in the csgB and yaiC promoter regions, suggesting that CsgD might activate transcription at these two promoters with different mechanisms. At csgB, the CsgD-binding site overlaps the -35 sequence, typical of an activator that contacts the {sigma} subunit of RNA polymerase (Busby & Ebright, 1994), while at yaiC its target sequence is located at -70. The binding site is present as an inverted repeat in the yaiC promoter region, possibly suggesting that CsgD might bind this promoter as a dimer. The different location and the presence of the inverted repeat strongly suggest that CsgD might activate transcription at the csgB and at the yaiC promoter regions with different mechanisms. We could not identify any other sites for known regulators common to the csgD-dependent promoters.

Since CsgD appears to regulate genes involved in biofilm formation, we tested the possibility that the newly identified pepD and yagS genes might also play a role in this process. Low-level, csgD-independent expression of either pepD or yagS negatively affects biofilm formation (Fig. 4). Thus our results strongly suggest that in addition to the positive regulation of the csgBA and yaiC promoters, CsgD can repress the expression of negative determinants for biofilm formation such as pepD and yagS (Table 3). While the functions of the csgBA operon (encoding curli) and of the yaiC gene (regulator of cellulose biosynthesis) are strictly related to production of extracellular polymers and to biofilm formation, neither pepD nor yagS appears to encode extracellular proteins or to be directly involved in the biosynthesis of adhesion determinants. The product of the pepD gene, dipeptidase D, cleaves the unusual dipeptide carnosine, and is induced by phosphate starvation (Klein et al., 1986; Henrich et al., 1992). Interestingly, pepD is up-regulated in luxS-deficient mutants of the enterohaemorrhagic E. coli O157 : H7, suggesting that quorum sensing negatively controls dipeptidase D expression (Sperandio et al., 2001). Quorum sensing is necessary for efficient biofilm formation in several Gram-negative species (Davies et al., 1998; Miller & Bassler, 2001). This suggests that carnosine might act as a signal molecule and that repression of the pepD gene might allow its accumulation as part of a switch to biofilm growth.

The yagS gene appears to be controlled by the yagT promoter, and we could not map any yagS-specific transcription start. However, yagS was down-regulated in the PHL628 strain, as determined by global gene expression (Table 3) and RT-PCR assays (data not shown). It is possible that CsgD might repress yagS transcription by preventing elongation by RNA polymerase, rather than controlling the expression of a specific yagS promoter. Low-level csgD-independent expression of yagS negatively affects biofilm formation (Fig. 4), suggesting a role for this gene as a negative determinant for biofilm formation. The yagS gene encodes a putative FAD-binding subunit of xanthine dehydrogenase, an enzyme involved in purine catabolism; thus yagS, as pepD, might be involved in the synthesis or in the degradation of a signal molecule important for biofilm formation.

In addition to the genes that show differential expression in the ompR234 mutant, a search for the putative CsgD-binding sequence (CGGGKGAKNKA) performed using the Colibri Web Server (http://genolist.pasteur.fr/Colibri/genome.cgi) reveals that this sequence is present in only one more gene in the E. coli chromosome, the putative aldehyde dehydrogenase gene (aldH). Despite the presence of the putative CsgD-binding site, aldH is expressed at similar levels in both the MG1655 and PHL628 strains in the growth conditions we tested, suggesting that CsgD might only regulate aldH in response to specific growth or environmental conditions, possibly in concert with additional transcriptional regulators. Thus our results show that csgD regulates a limited set of genes, suggesting that the central role of this regulator is the establishment of the biofilm phenotype in E. coli.


   ACKNOWLEDGEMENTS
 
We would like to thank Teresa Colangelo for carrying out the sand column experiments. This work was supported by the research grant no. 3100-058871 from the Swiss National Science Foundation.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Arnqvist, A., Olsen, A. & Normark, S. (1994). Sigma S-dependent growth-phase induction of the csgBA promoter in Escherichia coli can be achieved in vivo by sigma 70 in the absence of the nucleoid-associated protein H-NS. Mol Microbiol 13, 1021–1032.[Medline]

Busby, S. & Ebright, R. H. (1994). Promoter structure, promoter recognition and transcription activation in prokaryotes. Cell 79, 743–746.[Medline]

Chapman, M. R., Robinson, L. S., Pinkner, J. S., Roth, R., Heuser, J., Hammar, M., Normark, S. & Hultgren, S. J. (2002). Role of Escherichia coli curli operons in directing amyloid fiber formation. Science 295, 851–855.[Abstract/Free Full Text]

Chirwa, N. T. & Herrington, M. B. (2003). CsgD, a regulator of curli and cellulose synthesis, also regulates serine hydroxymethyltransferase synthesis in Escherichia coli K-12. Microbiology 149, 525–535.[Abstract/Free Full Text]

Costerton, J. W., Cheng, K. J., Geesey, G. G., Ladd, T. I., Nickel, J. C., Dasgupta, M. & Marrie, T. J. (1987). Bacterial biofilms in nature and disease. Annu Rev Microbiol 41, 435–464.[CrossRef][Medline]

Costerton, J. W., Ellis, B., Lam, K., Johnson, F. & Khoury, A. E. (1994). Mechanism of electrical enhancement of efficacy of antibiotics in killing biofilm bacteria. Antimicrob Agents Chemother 38, 2803–2809.[Abstract]

Costerton, J. W., Lewandowski, Z., Caldwell, D. E., Korber, D. R. & Lappin-Scott, H. M. (1995). Microbial biofilms. Annu Rev Microbiol 49, 711–745.[CrossRef][Medline]

Davies, D. G. & Geesey, G. G. (1995). Regulation of the alginate biosynthesis gene algC in Pseudomonas aeruginosa during biofilm development in continuous culture. Appl Environ Microbiol 61, 860–867.[Abstract]

Davies, D. G., Parsek, M. R., Pearson, J. P., Iglewski, B. H., Costerton, J. W. & Greenberg, E. P. (1998). The involvement of cell-to-cell signals in the development of a bacterial biofilm. Science 280, 295–298.[Abstract/Free Full Text]

Dorel, C., Vidal, O., Prigent-Combaret, C., Vallet, I. & Lejeune, P. (1999). Involvement of the Cpx signal transduction pathway of E. coli in biofilm formation. FEMS Microbiol Lett 178, 169–175.[CrossRef][Medline]

Finlay, B. B. & Falkow, S. (1997). Common themes in microbial pathogenicity revisited. Microbiol Mol Biol Rev 61, 136–169.[Abstract]

Fuqua, W. C., Winans, S. C. & Greenberg, E. P. (1994). Quorum sensing in bacteria: the LuxR-LuxI family of cell density-responsive transcriptional regulators. Annu Rev Microbiol 50, 727–751.[CrossRef]

Hall, S. D., Kane, M. F. & Kolodner, M. D. (1993). Identification and characterization of the Escherichia coli RecT protein, a protein encoded by the recE region that promotes renaturation of homologous single-stranded DNA. J Bacteriol 175, 277–287.[Abstract]

Hammar, M., Arnqvist, A., Bian, Z., Olsen, A. & Normark, S. (1995). Expression of two csg operons is required for production of fibronectin- and congo red-binding curli polymers in Escherichia coli K-12. Mol Microbiol 18, 661–670.[Medline]

Henrich, B., Monnerjahn, U. & Plapp, R. (1990). Peptidase D gene (pepD) of Escherichia coli K-12: nucleotide sequence, transcript mapping, and comparison with other peptidase genes. J Bacteriol 172, 4641–4651.[Medline]

Henrich, B., Backes, H., Klein, J. R. & Plapp, R. (1992). The promoter region of the Escherichia coli pepD gene: deletion analysis and control by phosphate concentration. Mol Gen Genet 232, 117–125.[Medline]

Hoyle, B. D. & Costerton, W. J. (1991). Bacterial resistance to antibiotics: the role of biofilms. Prog Drug Res 37, 91–105.[Medline]

Jaspers, M. C., Suske, W. A., Schmid, A., Goslings, D. A., Kohler, H. P. & van der Meer, J. R. (2000). HbpR, a new member of the XylR/DmpR subclass within the NtrC family of bacterial transcriptional activators, regulates expression of 2-hydroxybiphenyl metabolism in Pseudomonas azelaica HBP1. J Bacteriol 182, 405–417.[Abstract/Free Full Text]

Jucker, B. A., Harms, H. & Zehnder, A. J. B. (1998). Polymer interaction between five gram-negative bacteria and glass investigated using LPS micelles and vesicles as model system. Colloid Surf B Biointerfaces 11, 33–45.[CrossRef]

Klein, J., Henrich, B. & Plapp, R. (1986). Cloning and expression of the pepD gene of Escherichia coli. J Gen Microbiol 132, 2337–2342.[Medline]

Landini, P. & Zehnder, A. J. (2002). The global regulatory hns gene negatively affects adhesion to solid surfaces by anaerobically grown Escherichia coli by modulating the expression of lipopolysaccharide and flagellar genes. J Bacteriol 184, 1522–1529.[Abstract/Free Full Text]

Lawrence, J. R., Korber, D. R., Hoyle, B. D., Costerton, J. W. & Caldwell, D. E. (1991). Optical sectioning of microbial biofilms. J Bacteriol 173, 6558–6567.[Medline]

Lloyd, G. S., Landini, P. & Busby, S. J. W. (2001). Activation and repression of transcription initiation in bacteria. Essays Biochem 37, 17–31.[Medline]

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

Miller, M. B. & Bassler, B. L. (2001). Quorum sensing in bacteria. Annu Rev Microbiol 55, 165–199.[CrossRef][Medline]

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

Pratt, L. A. & Kolter, R. (1998). Genetic analysis of Escherichia coli biofilm formation: roles of flagella, motility, chemotaxis and type I pili. Mol Microbiol 30, 285–293.[CrossRef][Medline]

Prigent-Combaret, C., Vidal, O., Dorel, C. & Lejeune, P. (1999). Abiotic surface sensing and biofilm-dependent regulation of gene expression in Escherichia coli. J Bacteriol 181, 5993–6002.[Abstract/Free Full Text]

Prigent-Combaret, C., Prensier, G., Le Thi, T. T., Vidal, O., Lejeune, P. & Dorel, C. (2000). Developmental pathway for biofilm formation in curli producing Escherichia coli strains: role of flagella, curli and colanic acid. Environ Microbiol 2, 450–464.[CrossRef][Medline]

Prigent-Combaret, C., Brombacher, E., Vidal, O., Lejeune, P., Ambert, A., Landini, P. & Dorel, C. (2001). A complex regulatory network controls initial adhesion and biofilm formation in Escherichia coli via regulation of the csgB gene. J Bacteriol 183, 7213–7223.[Abstract/Free Full Text]

Romling, U., Bian, Z., Hammar, M., Sierralta, W. D. & Normark, S. (1998a). Curli fibers are highly conserved between Salmonella typhimurium and Escherichia coli with respect to operon structure and regulation. J Bacteriol 180, 722–731.[Abstract/Free Full Text]

Romling, U., Sierralta, W. D., Eriksson, K. & Normark, S. (1998b). Multicellular and aggregative behaviour of Salmonella typhimurium strains is controlled by mutations in the agfD promoter. Mol Microbiol 28, 249–264.[CrossRef][Medline]

Romling, U., Rohde, M., Olsen, A., Normark, S. & Reinkoster, J. (2000). AgfD, the checkpoint of multicellular and aggregative behaviour in Salmonella typhimurium regulates at least two independent pathways. Mol Microbiol 36, 10–23.[CrossRef][Medline]

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

Simoni, S., Harms, H., Bosma, T. N. P. & Zehnder, A. J. B. (1998). Population heterogeneity affects transport of bacteria through sand column at low flow rates. Environ Sci Technol 32, 2100–2105.[CrossRef]

Sperandio, V., Torres, A. G., Giron, J. A. & Kaper, J. B. (2001). Quorum sensing is a global regulatory mechanism in enterohemorrhagic Escherichia coli O157 : H7. J Bacteriol 183, 5187–5197.[Abstract/Free Full Text]

Stewart, P. S. (2001). Multicellular resistance: biofilms. Trends Microbiol 9, 34–39.[CrossRef][Medline]

Swift, S., Throup, J., Bycroft, B., Williams, P. & Stewart, G. (1998). Quorum sensing: bacterial cell-cell signalling from bioluminescence to pathogenicity. In Molecular Microbiology, pp. 185–208. Edited by S. J. W. Busby, C. M. Thomas & N. L. Brown. Berlin: Springer.

Tao, H., Bausch, C., Richmond, C., Blattner, F. R. & Conway, T. (1999). Functional genomics: expression analysis of Escherichia coli growing on minimal and rich media. J Bacteriol 181, 6425–6440.[Abstract/Free Full Text]

Uhlich, G. A., Keen, J. E. & Elder, R. O. (2001). Mutations in the csgD promoter associated with variations in curli expression in certain strains of Escherichia coli O157 : H7. Appl Environ Microbiol 67, 2367–2370.[Abstract/Free Full Text]

Vidal, O., Longin, R., Prigent-Combaret, C., Dorel, C., Hooreman, M. & Lejeune, P. (1998). Isolation of an Escherichia coli K-12 mutant strain able to form biofilms on inert surfaces: involvement of a new ompR allele that increases curli expression. J Bacteriol 180, 2442–2449.[Abstract/Free Full Text]

Weiner, R., Langille, S. & Quintero, E. (1995). Structure, function and immunochemistry of bacterial exopolysaccharides. J Ind Microbiol 15, 339–346.[Medline]

Williams, V. & Fletcher, M. (1996). Pseudomonas fluorescens adhesion and transport through porous media are affected by lipopolysaccharide composition. Appl Environ Microbiol 62, 100–104.[Abstract]

Zogaj, X., Nimtz, M., Rohde, M., Bokranz, W. & Romling, U. (2001). The multicellular morphotypes of Salmonella typhimurium and Escherichia coli produce cellulose as the second component of the extracellular matrix. Mol Microbiol 39, 1452–1463.[CrossRef][Medline]

Received 18 February 2003; revised 26 May 2003; accepted 28 May 2003.