Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Jun 1.
Published in final edited form as: Mol Microbiol. 2010 May 4;76(5):1111–1126. doi: 10.1111/j.1365-2958.2010.07192.x

N-acetyllactosamine-induced retraction of bundle-forming pili regulates virulence-associated gene expression in enteropathogenic Escherichia coli

Romney M Humphries 1,6, Thomas P Griener 1, Stefanie L Vogt 2, George L Mulvey 1, Tracy Raivio 2, Michael S Donnenberg 4, Pavel I Kitov 5, Michael Surette 1,3, Glen D Armstrong 1
PMCID: PMC2900475  NIHMSID: NIHMS207650  PMID: 20487271

Summary

Enteropathogenic Escherichia coli (EPEC) are a major cause of infant morbidity and mortality due to diarrhoea in developing countries. The pathogenesis of EPEC is dependent on a coordinated multi-step process culminating in the intimate adherence of the organisms to the host's intestinal mucosa. During the initial stages of the EPEC colonization process, the fimbrial adhesin, bundle-forming pili (BFP), plays an integral role. We previously reported that the major BFP structural subunit, bundlin, displays lectin-like properties which enables BFP to initially tether EPEC to N-acetyllactosamine (LacNAc) glycan receptors on host cell surfaces. We also reported that incubating EPEC with synthetic LacNAc-bearing neoglycoconjugates not only inhibits their adherence to host cells, but also induces BFP retraction and subsequent degradation of the bundlin subunits. Herein, we demonstrate that the periplasmic serine protease, DegP, is required for degrading bundlin during this process. We also show that DegP appears to act as a bundlin chaperone during BFP assembly and that LacNAc-BSA-induced BFP retraction is followed by transcriptional up-regulation of the BFP operon and down-regulation of the locus of enterocyte effacement operons in EPEC.

Introduction

Enteropathogenic Escherichia coli (EPEC) cause infant diarrhoea, mainly in developing countries. Key to EPEC virulence is their expression of type IV bundle-forming pili (BFP) (Bieber et al., 1998) which promote the formation of EPEC microcolonies (Giron et al., 1991), as well as the initial adherence of the organisms to host enterocytes (Hyland et al., 2008). Together, these two BFP-mediated events are referred to as localized adherence (LA).

Following LA, EPEC inject effector proteins into host cells by way of a type III secretion system (T3SS). These effector proteins induce enterocyte cytoskeletal restructuring and formation of attaching and effacing (A/E) lesions. One such effector protein is the translocated intimin receptor (Tir) which is inserted into the host cell plasma membrane where it is phosphorylated by a host cell kinase and serves as a receptor for another EPEC adhesin called Intimin. The A/E phenotype is dependent on a 35 kbp chromosomal pathogenicity-associated island known as the locus of enterocyte effacement (LEE). LEE harbours 41 open reading frames which are organized into five polycistronic operons, LEE1 to LEE5. The LEE genes can be separated into three functional categories: a region encoding the T3SS components, another encoding the secreted effector proteins and their chaperones, and a third encoding Intimin (Garmendia et al., 2005).

BFP are expressed from a 14 gene operon present on a large virulence-associated plasmid known as pEAF (EPEC adherence factor plasmid (Donnenberg et al., 1992)). The first gene of this operon, bfpA, encodes bundlin, the major structural subunit of the BFP filaments. EPEC isolates may express one of two bundlin alleles, α and β. These are further subdivided into α alleles 1-3 and β alleles 1-7 (Blank et al., 2000, Fernandes et al., 2007, Blank et al., 2003). We previously demonstrated that EPEC LA to the host intestine is initiated by the interaction between α bundlin (Hyland et al., 2008) and an enterocyte receptor which contains, at a minimum, the N-acetyllactosamine (LacNAc) glycoside (Humphries et al., 2009a, Hyland et al., 2008, Vanmaele et al., 1995). LacNAc conjugated to bovine serum albumin (LacNAc-BSA) inhibits LA of α bundlin-expressing EPEC to Hep-2 cells (Vanmaele et al., 1999), but only during the first 45 minutes of the in vitro LA assay (Hyland et al., 2008). We call this “early LA” to differentiate it from the intimate adherence phase observed in the longer 3-hour assay (Donnenberg & Nataro, 1995). This intimate adherence phase, which is not inhibited by LacNAc-BSA, is mediated by other EPEC adhesins, including Intimin and EspA (Cleary et al., 2004). Early LA to human adult (Hyland et al., 2006a) and paediatric (Humphries et al., 2009c) intestinal biopsy specimens is also inhibited by LacNAc-BSA.

In addition to its LA inhibiting activity, LacNAc-BSA induces BFP retraction and the rapid degradation of bundlin from α bundlin-expressing EPEC strains (Vanmaele et al., 1999). Although the mechanism is unknown, it is dependent on the presence of BfpF (Hyland et al., 2006b), an ATPase thought to be required for energizing the BFP retraction process (Bieber et al., 1998, Anantha et al., 1998). BFP retraction may cause microcolony dispersal following the initial adherence phase of EPEC colonization (Knutton et al., 1999) and incubating EPEC with Caco-2 cell membrane preparations also results in BfpF-dependant BFP loss (Tobe & Sasakawa, 2001). Further, BFP retraction is apparently required for EPEC pathogenesis in vivo, since a bfpF mutant is attenuated in its ability to cause diarrhoea in adult volunteers (Bieber et al., 1998).

The ability of EPEC to retract their BFP and eliminate the resulting accumulated pool of bundlin subunits from the inner membrane (Ramboarina et al., 2005, Giron et al., 1991) within 30 minutes (Vanmaele et al., 1999) is remarkable. The investigations reported herein were undertaken to gain insight into how EPEC respond to the sudden influx of bundlin subunits upon LacNAc-BSA-induced BFP retraction. Specifically, we asked if BFP retraction and disposal of bundlin requires the E. coli two-component Cpx membrane stress-response system and if this alters the expression rate of BFP and LEE genes in EPEC.

Results

Bundlin degradation following LacNAc-BSA-induced BFP retraction requires DegP

Since the Cpx membrane stress-response system, which is activated by the inappropriate accumulation of excess or misfolded proteins in the bacterial cell envelope (Raivio, 2005, Buelow & Raivio, 2005, Raivio & Silhavy, 2001), is required for efficient BFP biogenesis (Nevesinjac & Raivio, 2005, Vogt et al., accompanying article), we investigated whether DegP, a Cpx-regulated periplasmic serine protease, might degrade internalized bundlin monomers following LacNAc-BSA-induced BFP retraction. To test this hypothesis, we introduced an insertional mutation into the degP gene of E2348/69, to produce the strain, ALN188 (Vogt et al., accompanying article). Figure 1A demonstrates that ALN188 expresses BFP, albeit less efficiently than the wild-type strain, taking an additional 60 minutes of incubation in DMEM to induce BFP expression, which we detected by both western immunoblot analysis for bundlin (Figure 1A) as well as by light microscopy for microcolony formation (data not shown). Therefore, we incubated ALN188 an additional 60 minutes in DMEM to ensure BFP expression. This alteration in protocol did not appear to affect the ability of ALN188 to bind HEp-2 cells in the early LA assay relative to the wild-type E2348/69 strain (p > 0.05, n = 3, Student's t-test, Figure 1B).

Figure 1.

Figure 1

Figure 1

Open in a new tab

Role of DegP in LacNAc-BSA-induced loss of bundlin. A) Temporal expression of Bundlin in EPEC E2348/69 (WT), ALN188 (degP::kn) and ALN188 complemented in trans with degP (pCAdegP). The amount of bundlin was determined by densitometric analysis of three independent immunoblots. The data represent average % bundlin, relative to the MBP band density. B) Early LA of E2348/69, ALN188 and ALN188 (pCAdegp) to HEp-2 cells. The bars represent the mean number of Hep-2 cells with LA organisms from three experiments, with error bars representing the standard deviation from the means. No significant difference was seen between the LA of the three strains (p > 0.05) C) Effect of 0.8 mg/mL LacNAc-BSA (L) relative to underivitized BSA (B) on the amount of bundlin present in E2348/69 (WT), ALN188 (degP::kn) and ALN188 complemented with degP in trans (pCAdegP) following 30 minutes of incubation. The blot is representative of three independent experiments. The graph represents the cumulative results from the three experiments, and represents mean bundlin immunoblot band density, relative to MBP band density for each lane of the blots. The error bars represent standard deviations from the means. D) Effect of 0.8 mg/mL LacNAc-BSA, relative to underivitized BSA, on the autoaggregation of E2348/69 (WT), ALN188, ALN188 (pCAdegP) UMD916 (ΔbfpF), or UMD916 (pMSD217). The bars represent the mean autoaggregation indexes for three independent experiments, with error bars representing the standard deviations from the means. **p < 0.001 relative to underivatized BSA. E) Enzymatic activity of recombinant DegP on purified bundlin and lysozyme in the presence or absence of 5mM DTT. Samples were visualized by Coomassie brilliant blue staining of SDS-PAGE gels.

In contrast to wild-type E2348/69, exposing ALN188 to LacNAc-BSA did not result in a reduction in the amount of bundlin observed on western immunoblots (Figure 1C), even after an extended (1.5 h) incubation time (data not shown). The ability of LacNAc-BSA to induce bundlin degradation was restored in ALN188 by introducing a functional degP gene in trans on the plasmid, pCAdegP (Figure 1C). However, this was only observed if we did not add IPTG to induce degP expression in this strain. Presumably, a low rate of degP expression, due to promoter leakiness in pCA24 (Kitagawa et al., 2005), was sufficient to restore BFP expression and the ability to adhere to HEp-2 cells (Figure 1B) to ALN188 (pCAdegP). Adding IPTG to cause overexpression of degP prevented ALN188 (pCAdegP) from producing sufficient bundlin to be detectable on immunoblots (data not shown). These observations suggest that overexpressing DegP in the periplasm prevents BFP expression by EPEC.

The observation that inactivating degP prevents bundlin degradation after exposing ALN188 to LacNAc-BSA could be attributed to either failure of the BFP retraction process, or failure to of the organisms to degrade bundlin. In order to exclude one of these two possibilities, we assessed the autoaggregation index of ALN188 in the presence of LacNAc-BSA. The autoaggregation index is a functional measure of EPEC microcolony formation in liquid culture. EPEC that do not produce BFP do not express the autoaggregation phenotype (Anantha et al., 1998). In the wild-type E2348/69 strain, the autoaggregation index was reduced from 9.71 ± 0.15% in the presence of underivatized BSA, to 1.16 ± 0.02% in the presence of LacNAc-BSA (Figure 1D), indicating that LacNAc-BSA induced BFP retraction and dispersal of microcolonies in this strain. In ALN188, the autoaggregation index was similarly reduced from 9.71 ± 0.18% to 2.91 ± 0.07% when LacNAc-BSA was added to the cultures. To confirm that the reduced autoaggregation index correlated with the loss of BFP on the E2348/69 cell surface we recorded the autoaggregation index of UMD916, an E2348/69 strain incapable of retracting its BFP due to an insertional mutation in its bfpF gene (Anantha et al., 1998). In this strain, the autoaggregation index was not significantly (p > 0.4, n = 4, Student's t-test) affected by LacNAc-BSA (Figure 1D). Therefore, it appears most likely that the degP mutant strain ALN188 is unable to degrade its bundlin monomers following LacNAc-BSA-induced BFP retraction.

To further investigate the role of DegP in eliminating retracted bundlin monomers, we tested the ability of purified recombinant DegP to degrade soluble, recombinant bundlin. As reported previously (Spiess et al., 1999), DegP completely digested lysozyme but only in the presence of dithiothreitol (DTT) (Figure 1E). When DegP was incubated with recombinant bundlin, we also observed a significant (16 ± 4% of control, p < 0.001, Student's t-test, n = 3) reduction in the amount of bundlin when DTT was present relative to when it was omitted from the reactions (89.9% of control, p > 0.05, Student's t-test, n=3, Figure 1E). No lysozyme or bundlin degradation was evident on analysis by SDS-PAGE when DegP was omitted from the reactions (Figure 1E).

DegP is required for efficient BFP expression

In view of the results presented in Figure 1A, we were interested in determining the basis for the reduced BFP expression rate in ALN188. We therefore investigated BFP expression at the transcriptional level by quantitative real time PCR (qPCR) of cDNA synthesized from ALN188 and E2348/69 which were inoculated into DMEM and incubated for 45 minutes to induce BFP expression. These experiments revealed no significant difference in ΔCt (Ct bfpA gene relative to 16S RNA gene) values between the two strains (p > 0.05, Student's t-test, n = 4, Figure 2A). Therefore, the transcription rate of the BFP genes is not affected by the degP mutation in ALN188.

Figure 2.

Figure 2

Open in a new tab

Chaperone activity of DegP is required for efficient BFP expression in E2348/69. A) bfpA expression by the strains, ALN188 and E2348/69 after 45 minutes was determined by qPCR of cDNA isolated from the cultures. ΔCt values were obtained by subtracting the Ct of the bfpA amplicon from that of the 16S RNA amplicon. The graph represents the mean of three independent experiments. Error bars indicate the standard deviations from the means. B) Temporal expression of bundlin by the EPEC strains, E2348/69, ALN188, and ALN188, complemented in trans with either DegPS210A (pCAdegPS210A) or wild-type DegP (pCAdegP), was monitored by immunoblot analysis of the bundlin and MBP proteins. The data presented in the graph represent % bundlin immunoblot band density, relative to MBP immunoblot density for each lane of the blot. Each point in the graph represents the mean result from three independent immunoblot experiments. The graph represents the cumulative results from the three experiments, and shows mean bundlin immunoblot band density, relative to MBP band density for each lane of the blots. The error bars represent standard deviations from the means.

DegP is a multi-functional protein possessing both protease and chaperone activities. Accordingly, we investigated whether delayed BFP expression by ALN188 (Figure 1A), relative to E2348/69, could be attributed to the loss of either the DegP chaperone or proteolytic activities. To accomplish this, we investigated the effect of a protease-defective S210A point mutation variant of DegP on bundlin expression. Serine at position 210 is one of three amino acids which comprise a catalytic triad of amino acids (histidine, serine, aspartate) that are required for DegP serine protease activity. We found that, whereas ALN188 displayed delayed bundlin expression, ALN188 (pCAdegPS210A), produced an equivalent amount of bundlin, relative to the wild-type E2348/69 strain or to ALN188 complemented with wild-type DegP (pCAdegP) (Figure 2B). These observations indicate that the chaperone, not the protease activity of DegP is required for efficient BFP expression. No significant difference was observed in the early LA of the strains, E2348/69, ALN188, ALN188 (pCAdegP) and ALN188 (pCAdegPS210A) all bound equally to Hep-2 cells (Figure 1B and data not shown).

LacNAc-BSA-induced BFP retraction causes a transient reduction in bundlin expression

To investigate the dynamics of LacNAc-BSA-induced loss of BFP, we determined if exposing E2348/69 to LacNAc-BSA altered the temporal expression of bundlin at the transcriptional level. The effect of LacNAc-BSA on bfpA promoter activity was measured by introducing the plasmid, pMS210A, which contains the luciferase operon under the control of the bfpA promoter (Table S1), into E2348/69 and monitoring light production by this strain when grown in DMEM in the presence of LacNAc-BSA. The resulting data (Figure 3A) revealed that the transcription rate of the BFP operon increases transiently (for 30 minutes), in the presence of LacNAc-BSA, and then declines thereafter to a level similar to that observed in organisms exposed to underivatized BSA. A similar profile was observed for bundlin protein expression as the transcriptional results presented in Figure 3A; albeit protein expression was delayed by 60 minutes relative to mRNA expression (Figure 3B). Thirty minutes after adding LacNAc-BSA to E2348/69, bundlin transcription peaked (Figure 3A), whereas bundlin translation peaked at 90 minutes after E2348/69 was exposed to LacNAc-BSA (p < 0.0001 relative to organisms exposed to underivatized BSA, Student's t-test, n = 3, Figure 3B). The relative differences in bundlin expression between LacNAc-BSA and BSA-treated organisms disappeared after 180 min of incubation, when little bundlin was evident in either sample, relative to bundlin expression at the 45 and 90 min intervals (p < 0.0001, Student's t-test, n = 3, Figure 3B).

Figure 3.

Figure 3

Figure 3

Open in a new tab

Effect of LacNAc-BSA on bfpA transcription andbundlin expression. A) bfpA transcription was monitored by the production of light in E2348/69 harbouring the plasmid pRMHluxA, which encodes the luciferase operon under the control of the bfpA promoter. The data are presented as luminescence (cps) relative to the OD600nm. B) Temporal expression of bundlin, monitored by immunoblot analysis, in E2348/69 in the presence (L) or absence (B) of 0.8 mg/mL LacNAc-BSA. The blot is representative of three independent experiments and the graph represents the bundlin band immunoblot densities relative to those of the corresponding MBP bands for each lane of the blots. **p < 0.001 relative to underivatized BSA. The graph presents the means of the three independent experiments and the error bars are standard deviations from the means. C) Effect of LacNAc-BSA on the expression of bfpA in E2348/69, UMD916 (ΔbfpF), UMD916 (pMSD217), ALN188 and ALN188 (pCAdegP) determined by qPCR of cDNA isolated from bacteria treated with LacNAc-BSA or underivatized BSA for 30 minutes. Fold increase in gene expression was determined by the 2−ΔΔCt calculation, as detailed in the Materials and Methods section. **p < 0.001 relative to underivatized BSA (n = 3).

These luminescence and western immunoblot results were confirmed by qPCR analysis of cDNA prepared from E2348/69 following 30 min incubation with LacNAc-BSA (Figure 3C). This analysis revealed a 6.84-fold increase in bfpA mRNA in the LacNAc-BSA-treated relative to underivatized BSA-treated organisms. When this experiment was performed using UMD916, no significant (p > 0.05, Student's t-test, n = 3) difference was observed in the amount of bfpA mRNA isolated from organisms incubated with LacNAc-BSA or underivatized BSA for 30 minutes (Figure 3C). These observations suggest that BFP retraction is required to alter the rate of bundlin expression in E2348/69 following exposure to LacNAc-BSA. Complementing UMD916 with an active bfpF gene in trans, on the plasmid pMSD217 returned this phenotype to that of the wild-type E2348/69 strain (Figure 3C).

The role of DegP in bundlin expression was also investigated by determining the effect of LacNAc-BSA on bfpA expression in the degP mutant strain, ALN188 (Figure 3C). In this experiment, ALN188 and the degP-complemented strain, ALN188 (pCAdegP) were grown in DMEM in the presence or absence of LacNAc-BSA for 30 minutes. While ALN188 (pCAdegP) demonstrated the wild-type phenotype, that is an increase in bfpA mRNA transcription, LacNAc-BSA had little effect on the expression of bfpA mRNA in ALN188 (p > 0.05, Student's t-test, n = 3). The expression of bfpA mRNA in ALN188 was also recorded following 45 and 60 minutes incubation under these conditions, and similarly, no change in bfpA mRNA expression was observed (data not shown).

LacNAc-BSA also alters LEE gene expression

To determine if LacNAc-induced BFP retraction affects subsequent steps in the EPEC colonization strategy, we tested its effect on LEE operon promoter activity. In these experiments, the strains E2348/69, UMD916 and UMD916 (pMSD217) were transformed with plasmids pRMHlux1 to pRMHlux5 which encode the luciferase genes under the control of the LEE1 to LEE5 promoters, respectively. A promoter-less luciferase gene (pRMHlux) construct served as a negative control. These experiments revealed that exposing E2348/69 to LacNac-BSA induced a significant (p < 0.001, Student's t-test, n = 3) reduction in promoter activity in LEE operons 1, 2, 4, and 5, and a slight increase in promoter activity in the LEE3 operon (p > 0.05, Student's t-test, n = 3, Figure 4A). Similar to the results presented in Figure 3A, the decrease in LEE operon promoter activity was most prominent 30 minutes after adding LacNAc-BSA to the cultures (Figure 4), and diminished over time until, after 90 minutes, no difference in LEE operon promoter activity was observed between organisms treated with LacNAc-BSA or underivatized BSA (data not shown). The LacNAc-BSA-induced alteration of LEE promoter activity was BfpF-dependent since we detected no significant difference in LEE promoter activity in the strain UMD916 exposed to LacNAc-BSA (Figure 4B), a phenotype that was restored to that of the wild-type E2348/69 strain by adding back a functional bfpF gene in trans (Figure 4C).

Figure 4.

Figure 4

Open in a new tab

Effect of LacNAc-BSA on transcription of LEE operons in A) E2348/69, B), UMD916 C) and UMD916 (pMSD217). Transcription was monitored by the production of light in the three strains, which harboured the plasmids pRMHlux1 to pRMHlux5, and the data represent results obtained following 30 minutes of incubation in the presence of LacNAc-BSA. Luminescence (cps/OD600nm) data are presented here relative to bacteria exposed to underivitized BSA. **p < 0.001, *p < 0.01 relative to underivatized BSA. Each bar represents the mean of three independent experiments and the error bars are standard deviations from the means.

To investigate if DegP activity was also required for BFP-retraction-mediated signalling, we introduced the pRMHlux1 to pRMHlux5 plasmids into ALN188, and monitored light production in the presence and absence of LacNAc-BSA. In these experiments, E2348/69 and ALN188 responded in a similar fashion to the presence of LacNAc-BSA, suggesting that bundlin degradation is not required for the LacNAc-BSA-induced alteration of LEE operon transcription rates (data not shown).

LacNAc-BSA represses expression of the Cpx and RpoE systems

MacRitchie and colleagues recently demonstrated that, similar to the effect of LacNAc-BSA-induced BFP retraction, activating the Cpx membrane stress-response system causes down-regulation of LEE1, 2, 4, and 5 promoter activity, and slightly increased LEE3 promoter activity (Macritchie et al., 2008). Since the Cpx system is induced in response to misfolded or over-expressed proteins accumulating in the bacterial cell envelope (Raivio, 2005, Buelow & Raivio, 2005, Raivio & Silhavy, 2001) we investigated the effect of exposing E2348/69 to LacNAc-BSA on the promoter activity of two Cpx system genes, cpxP and degP. LacNAc-BSA significantly repressed both cpxP (p < 0.001, Student's t-test, n = 3) and degP (p < 0.01, Student's t-test, n = 3) promoter activities (Figure 5A and 5B). Therefore, it appears that both the Cpx system and DegP are repressed when E2348/69 is exposed to LacNAc-BSA.

Figure 5.

Figure 5

Open in a new tab

Effect of LacNAc-BSA on the temporal transcription of A) degP, B) cpxP, C) and rpoE in E2348/69. Transcription was monitored by the production of light in E2348/69, which harboured the plasmids pRMHlux6, pRMHlux7 and pRMHluxrpoE, respectively. Data are representative of three independent experiments.

DegP expression is also regulated by the σE (RpoE) regulon (Raivio & Silhavy, 2001, Raivio & Silhavy, 1999) which is also induced by the accumulation of misfolded protein in the periplasm and outer membrane of Gram negative organisms. σE is positively autoregulated at the transcriptional level and therefore we tested the effect of LacNAc-BSA on rpoE expression. Accordingly, the plasmid, pRMHluxrpoE, was introduced into E2348/69, and luminescence was monitored over 1.5 h in the presence of LacNAc-BSA. In these experiments, we observed reduced luminescence in E2348/69 treated with LacNAc-BSA (Figure 5C). The activity of pRMHluxrpoE was confirmed by the observation that heat shocking E2348/69 (pRMHluxrpoE) at 43°C increased luminescence (rpoE expression) output by 130 ± 10% (p < 0.0001, Student's t-test, n = 6, data not shown).

LacNAc conjugated to polyacrylamide inhibits EPEC adherence to tissue culture cells but does not induce BFP retraction, nor alter BFP or LEE operon gene transcription

We recently described a novel LacNAc glycoconjugate, EP-II-189 (Humphries et al., 2009b), which consists of LacNAc dimers conjugated to a linear flexible polyacrylamide scaffolding (Figure 6A). Like LacNAc-BSA and LacNAc coupled to gold nanoparticles (LacNAc-Au), this molecule inhibits E2348/69 early LA to Hep-2 cells (Figure 6B). The EP-II-189-mediated inhibition of E2348/69 early LA was significant, relative to the early LA of organisms exposed to underivatized polyacrylamide (PAA, p < 0.0001, Student's t-test, n = 3, Figure 6B) or underivatized BSA (p < 0.0001, data not shown). Further, EP-II-189 inhibited early LA better (p < 0.001, Student's t-test, n = 3) than LacNAc-BSA, when these compounds were used at equivalent (90 μM) concentrations of LacNAc (Figure 6B). This difference was less apparent when 40 μM LacNAc equivalents were used (p > 0.05, Student's t-test, n = 3, Figure 6B). However, unlike LacNAc-BSA and LacNAc-Au (Vanmaele et al., 1999, Hyland et al., 2006b), exposing E2348/69 to EP-II-189 did not induce BFP retraction (Figure 7A) or bundlin degradation (Figure 7B). We therefore utilized this compound to confirm the observation that BFP retraction is necessary for LacNAc-induced alteration of gene expression in E2348/69. No difference (p > 0.05, Student's t-test, n = 3) was observed in bfpA mRNA transcription rates in organisms exposed to EP-II-189 for 30 min and 1 h relative to bfpA mRNA transcription in organisms exposed to underivatized PAA (Figure 7C). Also, EP-II-189 did not affect the expression of the LEE1 to LEE5 operons (Figure 7D).

Figure 6.

Figure 6

Open in a new tab

A) Structure of EP-II-189. B) Effect of EP-II-189 and LacNAc-BSA on E2348/69 early LA to Hep-2 cells at equimolar concentrations of LacNAc. Data are presented relative to organisms exposed to underivatized BSA (for LacNAc-BSA) or PAA (for EP-II-189). **p < 0.0001, *p < 0.01 relative to equimolar LacNAc-BSA concentration. For both early LA experiments, bars represent the mean of three experiments, with error bars representing the standard deviations from the means.

Figure 7.

Figure 7

Figure 7

Open in a new tab

Effect of 90 μM LacNac equivalents of EP-II-189 on BFP retraction. A) Autoaggregation index of E2348/69 following 30 minutes incubation in the presence of 90 μM LacNAc equivalents of LacNAc-BSA, EP-II-189 or 0.8 mg/mL underivatized BSA. Data are presented at two time points, 30 min and 3 h, and represent the mean of three experiments. **p < 0.0001 relative to organisms exposed to underivatized BSA. B) Bundlin expression, determined by western immunoblot analysis, in E2348/69 in the presence of 90 μM relative LacNAc concentration of EP-II-189, LacNAc-BSA or equivalent concentrations of underivatized polyacrylamide (PAA) or BSA. Graph bars represent the bundlin band density relative to that of the MBP density, for each lane of the blot. Data were normalized using the underivatized BSA negative control, and represent the means of three experiments. **p < 0.0001 relative to organisms exposed to underivatized BSA. C) Effect of EP-II-189 on bfpA expression, as assessed by real-time PCR of bfpA and 16s RNA from cDNA isolated from the bacteria following a 30 min or 1 h incubation with or without EP-II-189. Fold gene up-regulation was determined by calculating 2−ΔΔCt, as described in the materials and methods section. **p < 0.0001, *p < 0.01 relative to organisms exposed to underivatized BSA. D) Effect of a 30 minute incubation with 90 μM LacNAc equivalent of EP-II-189 on light production by E2348/69 harbouring the plasmids pRMHlux1-5 (LEE 1-5 promoters) and pRMHlux (a promoterless plasmid). Bars represent the mean of three experiments. For all panels, error bars are the standard deviations from the means.

Discussion

BFP represent important virulence determinants for classical EPEC strains in humans (Bieber et al., 1998). BFP, like other Type IV pili (Tfp), can be extended and retracted by the bacterium (Gauthier, 2009). Further, when Tfp are bound to the host cell membrane, their retraction may influence epithelial cell signalling and subsequent response to infection (Merz & So, 2000). We now demonstrate that BFP retraction also alters virulence-associated gene expression in EPEC.

Previously, we reported that the initial interaction between E2348/69 and cultured epithelial cells in vitro as well as human intestinal biopsy specimens is mediated by the binding of α1-bundlin to a carbohydrate receptor containing the LacNAc glycan sequence (Hyland et al., 2008). Synthetic versions of this receptor, when conjugated to BSA or gold nanoparticles, inhibit this early adherence phase of the EPEC intestinal colonization strategy. These synthetic neoglycoconjugates also appear to induce BFP retraction and rapid degradation of bundlin (Hyland et al., 2006b, Vanmaele et al., 1999).

Once BFP are retracted, the results presented in Figure 1 now demonstrate that the periplasmic serine protease, DegP, rapidly degrades the bundlin monomers that accumulate in the bacterial cell envelope. We demonstrate herein that DegP is also required for the proper assembly of BFP, since a degP mutant strain expresses BFP more slowly than a wild-type strain (Figure 1A). This attenuation of BFP expression is not due to changes at the gene transcription level, since no differences were observed in the rate of bfpA gene transcription between the wild-type and degP insertional mutant (Figure 2A). Rather, it would appear that the difference in BFP expression is due to the loss of DegP chaperone activity in the ALN188 strain, a conclusion which is supported by the data presented in Figure 2B. This experiment demonstrates that complementing ALN188 with a mutant DegP that is protease-deficient, but maintains its chaperone activity, restores efficient BFP expression.

The functional switch between DegP protease and chaperone activities is apparently mediated by both temperature (Spiess et al., 1999), which induces conformational changes in the protein, and by substrate recognition, which is controlled by the DegP PDZ1 domain (Iwanczyk et al., 2007). It has therefore been proposed that DegP might recognize substrate molecules targeted for degradation in these two different ways. How bundlin is protected from DegP degradation during BFP assembly remains to be determined, but this may be due to alternate physical presentations of “outgoing” and “incoming” bundlin monomers during the BFP assembly and retraction processes, respectively. This hypothesis is supported by the observation that a mutated bundlin, in which the single C129-C179 disulfide bond which links the two C-terminal α helices of the protein cannot be formed (Ramboarina et al., 2005) does not assemble into BFP and is rapidly degraded by the bacterium (Zhang & Donnenberg, 1996). The data presented in Figure 1E suggest that reducing the disulfide bond with DTT may also expose bundlin's normally inaccessible C-terminal amino acids (Ramboarina et al., 2005) to the DegP PDZ1 domain (Iwanczyk et al., 2007, Krojer et al., 2008) thereby leading to its degradation.

Bundlin is also processed by DsbA, an oxidoreductase found in the E. coli periplasm (Zhang & Donnenberg, 1996). DsbA catalyzes the formation of the C129-C179 disulfide bond in bundlin. It is possible that the interaction between DsbA and bundlin during BFP assembly protects the protein from degradation by masking its C-terminal end thereby preventing PDZ recognition by DegP; however, this hypothesis remains to be tested

Activating the Cpx system in E2348/69 increases DegP expression, and also induces the loss of bundlin in EPEC in a manner reminiscent of LacNAc-BSA-induced BFP retraction and bundlin degradation (Vogt et al., accompanying article). However, in our experiments, we detected no up-regulation of degP, cpxP or rpoE transcription when E2348/69 was exposed to LacNAc-BSA. It is possible, therefore that a basal amount of DegP in the EPEC periplasm is sufficient to cope with the influx of bundlin monomers (Ramboarina et al., 2005, Giron et al., 1991) upon exposing the organisms to LacNAc-BSA. Perhaps eliminating retracted bundlin from the periplasm, by DegP (Figure 1), prevents envelope stress such that the Cpx and σE regulons are not induced (Figure 5). In the accompanying article, Vogt et al. demonstrate that activating the Cpx response represses BFP expression at the transcriptional level. In our experiments, we observed a second phase of bundlin over-expression following the initial LacNAc-BSA-induced loss of BFP and degradation of bundlin (Figure 3). Suppressing the Cpx system would allow this over-expression of bundlin to occur. Therefore, our observations that Cpx and σE are repressed in E2348/69, subsequent to exposing the organisms to LacNAc-BSA, are consistent with the model proposed by Vogt and colleagues (Vogt et al., accompanying article) whereby inducing the Cpx system represses the expression of cell-surface structures during times of envelope stress.

The transient up-regulation of bundlin expression observed 90 minutes following exposure to LacNAc-BSA (Figure 3) also appears to occur when EPEC engages host receptors (Leverton & Kaper, 2005). In our experiments, the change in bundlin expression was dependant on the presence of BfpF (Figure 3C), suggesting that BFP retraction itself is required for altering bundlin gene expression in EPEC. This is supported by the observation that EP-II-189 did not induce BFP retraction nor alter bundlin expression. The failure of EP-II-189 to induce BFP retraction in E2348/69 (Figure 7) may be due to its linear presentation of divalent LacNAc glycosides. That is, the LacNAc groups on EP-II-189 might be presented in such a way that, not only will these occupy the carbohydrate binding domain of bundlin (Humphries et al., 2009a) thereby preventing BFP binding to host cell receptors, but they may also effectively crosslink BFP filaments on adjacent bacteria, physically preventing their retraction. This activity may be functionally equivalent to that described by Helaine and colleagues for the minor pilin, PilX, that is intermittently incorporated into the Tfp of Neisseria meningitidis (Helaine et al., 2007). These investigators demonstrated that interactions between the protruding D-regions of the PilX subunits on pili expressed by adjacent bacteria counteracted the activity of PilT, the N. meningitidis BfpF equivalent, and antagonized the anti-parallel retraction of the pili.

Recently, Dietrich and colleagues demonstrated that PilT, is required for global regulation of gene expression in N. meningitidis. These authors reported that eliminating Tfp retraction, by deleting pilT, caused increased expression of the Neisserial Tfp pilin, PilE (Dietrich et al., 2009). We also observed increased bundlin expression in the bfpF mutant strain, UMD916, both at the translation (Hyland et al., 2006b) and transcriptional levels (unpublished observations). Perhaps entry of Tfp pilins into the bacterial inner membrane represents a signal for down-regulating pilin gene expression. In contrast, the absence of pilin in the inner membrane may signal for the up-regulation of pilin gene expression, in a feed-back-type regulation system. The absence of pilins from the inner membrane of the bacterial cell could be caused not only by the failure of Tfp retraction, but also by DegP-mediated pilin clearance from the inner membrane following BFP retraction. This hypothesis is supported by the observation that bfpA gene expression increased when bundlin was absent from the cell, caused by both the loss of BFP retraction in the UMD916 strain (unpublished observations) but also following bundlin clearance from the inner membrane by DegP (Figure 3B). When DegP was mutated, this effect was lost as bundlin expression was reduced in the presence of LacNAc-BSA in ALN188 (Figure 3C). These data suggest the continued presence of bundlin in the inner membrane after exposing ALN188 to LacNAc-BSA (Figure 1B) suppresses bundlin expression.

We further found that exposing E2348/69 to LacNAc-BSA caused down-regulation of the Cpx and RpoE systems (Figure 5), suggesting that membrane stress due to BFP retraction is relieved by removing the accumulated bundlin from the membrane. It follows, therefore, that expressing BFP induces membrane stress, which must be relieved by the activity of the Cpx and RpoE systems for efficient BFP assembly, as reported by Vogt et al., in the accompanying article, and by Nevesinjac and Raivio (2005).

LacNAc-BSA not only triggers BFP retraction and bundlin degradation, but also alters LEE gene expression. We found that BFP retraction is coupled to down-regulation of effector protein transcription, including the T3SS, intimin and Tir operons (Figure 5). The role of BFP retraction in the signalling process was demonstrated by the failure of both the bfpF retraction-deficient strain (Figure 4B), and the compound EP-II-189 to alter bundlin and LEE gene expression (Figure 7D).

The role of BFP retraction and bundlin degradation following EPEC engagement of host cell receptors during infection remains to be clarified. It has been reported, however, that EPEC disperse from their LA microcolonies after 6 hours of contacting HEp-2 cells (Cleary et al., 2004, Knutton et al., 1999), leaving only bacteria that have become intimately attached to the monolayers. In addition, Tobe and Sasakawa demonstrated that, similar to LacNAc-BSA, exposing EPEC to Caco-2 cell membrane preparations causes the disappearance of BFP (Tobe & Sasakawa, 2001). These investigators also studied the temporal expression of BFP during EPEC LA to Caco-2 cells. In this report, bundlin was detected for at least 2 hours. After 2 hours of co-incubating EPEC with Caco-2 cells, the amount of bundlin expressed by the organisms declined to sub-detectable amounts. These observations are consistent with the results presented in Figure 3, which demonstrate down-regulation of bundlin expression after 180 minutes of EPEC exposure to LacNAc-BSA (Figure 3B). Leverton and Kaper (2005) also monitored the temporal expression of bundlin and LEE genes in the presence and absence of HEp-2 cells. In this study, they observed enhanced bundlin expression after 60 minute of co-incubating EPEC with HEp-2 cells, similar to our results in Figure 3A. However, unlike our results, these investigators found increased expression of LEE 3, 4 and 5 and decreased expression of LEE 1 and 2 after 3 hours in the presence of HEp-2 cells, whereas we only observed an insignificant increase in LEE3 expression following exposure to LacNAc-BSA (Figure 4A). This suggests additional host cell factors may be involved in LEE gene modulation in vitro. However, in their investigations, Leverton and Kaper (2005) did not monitor LEE and bundlin mRNA levels during the same time period for which we recorded the effects of LacNAc-BSA-mediated BFP retraction.

After they have served their purpose in the initial attachment step, the presence of BFP may hinder the intimate adherence phase of the EPEC colonization strategy. Their retraction at this time might be necessary to facilitate intimate adherence. Perhaps, after initially binding to host cells, BFP retraction also allows organisms that have not established intimate contact to be released from their microcolonies. These freed organisms may then re-express their BFP to further colonize additional sites of the host's intestine thereby furthering the infection process. However, organisms exposed to LacNAc-BSA cannot intimately adhere to a host cell surface after retracting their BFP. In this special instance, all of the bacteria may be programmed to re-express their BFP upon dispersal of the microcolonies. This might explain the surge in bundlin expression (Figure 3A and B) following LacNAc-BSA-induced retraction of BFP in E2348/69. This might also explain why we observed LEE gene suppression upon exposing E2348/69 to LacNAc-BSA, since the factors encoded by the LEE operons are involved in the second, intimate stage, of adherence. EPEC that are not in contact with a host cell surface do not need to express the LEE operons, and avoiding an unnecessary expenditure of energy to do so may be of benefit to the organism's survival in the intestinal tract. However, other groups observed no loss of BFP after extended incubation with host cells (Saldana et al., 2009, Leverton & Kaper, 2005) and so further study is needed to determine how LacNAc-BSA-induced BFP retraction relates to the in vivo infection process.

Communication between prokaryotic and eukaryotic cells during the infection process occurs at multiple levels. It would appear that engaging intestinal cell receptors not only tethers the bacteria to the host, but also serves as a recognition signal which may trigger altered expression of both bacterial virulence-associated genes and host response genes. Therefore, it may be possible to develop carbohydrate-based cell receptor analogs that not only prevent bacterial adherence to host tissues but may also antagonize bacterial virulence-associated gene expression. By virtue of their specificity, carbohydrate-based drugs could be targeted to only interfere with the virulence of specific pathogens thereby, unlike conventional antibiotics, sparing the normal flora as well as their benefits to the host from eradication. Compounds such as EP-II-189 may prove valuable in interrupting this important aspect of EPEC pathogenesis, not only by masking BFP host-cell receptor binding domains, but also by preventing the appropriate signalling for regulating virulence-associated gene expression in these organisms.

Materials and Methods

Bacterial strains, plasmids and recombinant proteins

The bacterial strains and plasmids used in this study are listed in Table S1. Bacteria were routinely cultivated on conventional tryptic soy agar (TSA) plates at 37°C to obtain individual colonies. Ampicillin (100 ug/ml), kanamycin (50 ug/ml), chloramphenicol (25 ug/ml) and trimethoprim (100 ug/ml) were added, as necessary.

The plasmid, pCAdegP, which harbours degP under control of an IPTG-inducible promoter, was modified using the primer degPS210AF (Table S2) and its reverse compliment by the Quickchange protocol (Stratagene), as described in our previous article (Humphries et al., 2009a). The resulting mutation encoded S210A in DegP. This was confirmed by sequencing the pCAdegPS210A degP gene using the primer degPseq (Table S2) obtained from the University of Calgary Core DNA Sequencing Laboratory.

Recombinant DegP with a hexahistidine C-terminal tag was produced as described elsewhere (Spiess et al., 1999) from the plasmid, pCS20 (Table S1). Recombinant soluble bundlin was produced, as described in our previous articles, from the plasmid pPF401 (Humphries et al., 2009a, Hyland et al., 2008).

Construction of luciferase reporter plasmids

Luciferase reporter plasmids (Table S1) were constructed by introducing the dihydrofolate reductase gene from p34Tp (DeShazer & Woods, 1996), which was amplified with the primers dhrf1 and dhrf2 (Table S2), into the PstI site of pJW17 (Macritchie et al., 2008), to produce the plasmid, pRMHlux1 (Table S2). This plasmid was then digested with EcoRI and BamHI (NEB), blunt-ended by Taq (NEB) and ligated using T4 ligase (Invitrogen) to produce a promoter-less negative control plasmid, pRMHlux. Plasmids pRMHlux2 to pRMHlux7 were produced by ligating the EcoRI/BamHI fragments from pJW18, pJW19, pJW20, ptir-lux, pNLP27 and pJW25, respectively, into an EcoRI/BamHI digested pRMHlux1. The plasmid, pRMHluxA (TableS2) was produced by PCR-amplifying the BFP promoter from E2348/69 using the primers bfp01 and bfp03 (Table S2). This fragment was BamHI-digested and inserted into a BamHI-digested pRMHlux1. pRMHluxrpoE was produced by amplifying the rpoE promoter from E2348/69 using the primers, PrpoE1 and PrpoE2 (Table S2) and inserting the BamHI- and EcoRI-digested product into pRMHlux1. All plasmids were screened for inserts by PCR using the primers, pNLP10F and pNLP10R (Table S2).

Preparation of LacNAc glycoconjugates

LacNAc-BSA, which consists of a chemically synthesized 8-methoxycarbonyloctyl (MCO) glycoside (Lemieux, 1977) covalently coupled to BSA, was provided by Dr. Om Srivastava, Optimer Pharmaceuticals, San Diego, CA. Underivitized BSA was obtained from Sigma Aldrich.

Western immunoblotting analysis

EPEC strains were grown statically overnight in tryptic soy broth (TSB) without glucose, at 37°C. Cultures were inoculated 1:100 into DMEM which had been pre-equilibrated overnight in a CO2 incubator. BFP expression was induced by further incubating the cultures in the CO2 incubator for 45 minutes. 0.8 mg/ml LacNAc-BSA or underivitized BSA were then added and the cultures were incubated for an additional 30 min in the CO2 incubator. Finally, the bacteria were harvested by centrifugation and the resulting bacterial cell pellets were suspended in SDS-PAGE sample buffer containing 50 mM dithiothreitol. Bacterial proteins were separated by SDS-PAGE (12.5% polyacrylamide) and electrophoretically transferred to Immobilon-P membranes (Millipore, Bedford, Mass). The membranes were cut into three sections using the pre-stained molecular size standards as a guide. Nonspecific binding sites on the membranes were blocked with 5% (wt/vol) skim milk in PBS containing 0.05% Tween 20 (PBST) and the sections were incubated with either anti-MBP (Sigma Aldrich), or anti-bundlin antibodies(Hyland et al., 2008). The immunoblots were thoroughly washed in PBST and incubated with the appropriate peroxidase-conjugated antibody (Sigma Aldrich). The membranes were then developed using the enhanced chemiluminescence (ECL) color development reagents according to the manufacturer's (GE Healthcare, Oakville, ON) instructions and exposed to Kodak X-ray film (GE Healthcare, Oakville, ON). Bands corresponding to bundlin and MBP were analyzed using the Kodak Image Station 2000MM and associated Kodak ID Image Analysis Software (version 3.6.5). Sample loading inconsistencies were corrected for by normalizing the densitometry intensities of each of the bundlin bands in each gel lane to the intensity of the accompanying constitutively expressed E. coli MBP band.

Autoaggregation Assays

Overnight TSB cultures were diluted 1:100 in DMEM and incubated for 4 h at 37°C in a CO2 incubator. The A600 of each culture was then recorded and, after each measurement, the culture was shaken vigorously for 30 sec using a Vortex mixer and the A600 was re-recorded. The percent increase in A600 after disrupting the microcolonies was recorded as a quantitative autoaggregation index. Each experiment was repeated three times.

DegP Enzyme Activity Assay

The proteolytic activity of DegP was determined by incubating 5μg DegP with either 25 μg Lysozyme (Invitrogen) or recombinant purified α1-bundlin, in the presence or absence of 5 mM DTT overnight at 37°C. The reaction was stopped and proteins were precipitated by adding 6% (w/v) tricholoracetic acid. Following centrifugation (5 minutes, 20,800 × g, room temperature), the proteins were solubilised in SDS sample buffer and separated by SDS-PAGE (12.5% acrylamide). Gels were then stained with Coomassie Blue (Invitrogen).

Bioluminescence assay

Single colonies harbouring pRMHlux plasmids were inoculated into 5 ml of TSB plus antibiotics as required and grown at 37°C or 30°C statically overnight. The cultures were inoculated 1:100 in 1 ml of DMEM and grown at 37°C in a CO2 incubator for 30 minutes. Glycoconjugates at a concentration of 0.8 mg/mL were then added to some of the cultures, and 200 μl of these were transferred to a 96-well black-sided microtiter plate (Gibco) and covered with 10μl sterile mineral oil (Invitrogen) to prevent evaporation. The absorbance (A600) and bioluminescence (counts per second [cps]) were recorded at programmed intervals using a Wallac 420 multilabel automatic plate reader (Perkin-Elmer) at a temperature of 37°C over a period of 4 hours. The data represent values collected after 0.5, 1, or 3 hours of growth, and represent bioluminescence normalized to absorbance to correct for bacterial cell growth between experiments (cps/A600). Each experiment was repeated a minimum of three times on separate days. No bioluminescence was observed from EPEC harbouring the promoterless lux pRMHlux (Table S1) plasmid.

mRNA extraction, cDNA synthesis and, real-time quantitative PCR assays

BFP were induced in E2348/69 cultures using DMEM and these were subsequently treated with glycoconjugates as described above. The cultures were then diluted 1:1 in the Qiagen Bacteria Protect reagent and incubated at room temperature for 5 minutes. Bacteria were subsequently harvested by centrifugation and total RNA was isolated using the RNeasy Protect Bacteria mini kit (Qiagen, Mississauga, ON) following the manufacturer's instructions. First-strand cDNA was synthesized using the Invitrogen Superscript III First Strand Synthesis system, again following the manufacturer's instructions, and by using oligod(T) primers. Real-time quantitative PCR was performed using the Fermentas Maxima SYBR green qPCR master mix (Fermentas, Burlington, ON) and either the E. coli-specific 16s RNA primers Ec16sF and Ec16sR (Table S2), or bfpA-specific primers bfpAF and bfpAR (Table S2, (Leverton & Kaper, 2005)). Fold change in gene expression was determined by the comparative Ct method, and represent 2−ΔΔCt. Here, ΔΔCt = [(Ct 16sRNA amplicon − CtbfpA amplicon) BSA-treated samples] − [(Ct 16sRNA amplicon − CtbfpA amplicon) LacNAc-BSA treated samples].

Supplementary Material

Supp Table s1-s2

Acknowledgements

This work was supported by the Alberta Ingenuity Centre for Carbohydrate Science (GDA), the Alberta Heritage Foundation for Medical Research (TLR and MS), and the Canadian Institutes of Health Research (TLR and MS). RMH and TPG are recipients of NSERC Canada Graduate Scholarships.

Literature Cited

  1. Anantha RP, Stone KD, Donnenberg MS. Role of BfpF, a member of the PilT family of putative nucleotide-binding proteins, in type IV pilus biogenesis and in interactions between enteropathogenic Escherichia coli and host cells. Infect Immun. 1998;66:122–131. doi: 10.1128/iai.66.1.122-131.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bieber D, Ramer SW, Wu CY, Murray WJ, Tobe T, Fernandez R, Schoolnik GK. Type IV pili, transient bacterial aggregates, and virulence of enteropathogenic Escherichia coli. Science. 1998;280:2114–2118. doi: 10.1126/science.280.5372.2114. [DOI] [PubMed] [Google Scholar]
  3. Blank TE, Lacher DW, Scaletsky IC, Zhong H, Whittam TS, Donnenberg MS. Enteropathogenic Escherichia coli O157 strains from Brazil. Emerg Infect Dis. 2003;9:113–115. doi: 10.3201/eid0901.020072. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Blank TE, Zhong H, Bell AL, Whittam TS, Donnenberg MS. Molecular variation among type IV pilin (bfpA) genes from diverse enteropathogenic Escherichia coli strains. Infect Immun. 2000;68:7028–7038. doi: 10.1128/iai.68.12.7028-7038.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Buelow DR, Raivio TL. Cpx signal transduction is influenced by a conserved N-terminal domain in the novel inhibitor CpxP and the periplasmic protease DegP. J Bacteriol. 2005;187:6622–6630. doi: 10.1128/JB.187.19.6622-6630.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Cleary J, Lai LC, Shaw RK, Straatman-Iwanowska A, Donnenberg MS, Frankel G, Knutton S. Enteropathogenic Escherichia coli (EPEC) adhesion to intestinal epithelial cells: role of bundle-forming pili (BFP), EspA filaments and intimin. Microbiology. 2004;150:527–538. doi: 10.1099/mic.0.26740-0. [DOI] [PubMed] [Google Scholar]
  7. DeShazer D, Woods DE. Broad-host-range cloning and cassette vectors based on the R388 trimethoprim resistance gene. Biotechniques. 1996;20:762–764. doi: 10.2144/96205bm05. [DOI] [PubMed] [Google Scholar]
  8. Dietrich M, Mollenkopf H, So M, Friedrich A. Pilin regulation in the pilT mutant of Neisseria gonorrhoeae strain MS11. FEMS Microbiol Lett. 2009 doi: 10.1111/j.1574-6968.2009.01647.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Donnenberg MS, Giron JA, Nataro JP, Kaper JB. A plasmid-encoded type IV fimbrial gene of enteropathogenic Escherichia coli associated with localized adherence. Mol Microbiol. 1992;6:3427–3437. doi: 10.1111/j.1365-2958.1992.tb02210.x. [DOI] [PubMed] [Google Scholar]
  10. Donnenberg MS, Nataro JP. Methods for studying adhesion of diarrheagenic Escherichia coli. Methods Enzymol. 1995;253:324–336. doi: 10.1016/s0076-6879(95)53028-2. [DOI] [PubMed] [Google Scholar]
  11. Fernandes PJ, Guo Q, Donnenberg MS. Functional consequences of sequence variation in bundlin, the enteropathogenic Escherichia coli type IV pilin protein. Infect Immun. 2007;75:4687–4696. doi: 10.1128/IAI.00009-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Garmendia J, Frankel G, Crepin VF. Enteropathogenic and enterohemorrhagic Escherichia coli infections: translocation, translocation, translocation. Infect Immun. 2005;73:2573–2585. doi: 10.1128/IAI.73.5.2573-2585.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gauthier NC. Bundle-forming pili from enteropathogenic Escherichia Coli generate moderate forces. Biophysical Journal. 2009;96:519a–519a. [Google Scholar]
  14. Giron JA, Ho AS, Schoolnik GK. An inducible bundle-forming pilus of enteropathogenic Escherichia coli. Science. 1991;254:710–713. doi: 10.1126/science.1683004. [DOI] [PubMed] [Google Scholar]
  15. Helaine S, Dyer DH, Nassif X, Pelicic V, Forest KT. 3D structure/function analysis of PilX reveals how minor pilins can modulate the virulence properties of type IV pili. Proc Natl Acad Sci U S A. 2007;104:15888–15893. doi: 10.1073/pnas.0707581104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Humphries RM, Donnenberg MS, Strecker J, Kitova E, Klassen JS, Cui L, Griener TP, Mulvey GL, Armstrong GD. From alpha to beta: identification of amino acids required for the N-acetyllactosamine-specific lectin-like activity of bundlin. Mol Microbiol. 2009a;72:859–868. doi: 10.1111/j.1365-2958.2009.06679.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Humphries RM, Strecker JG, Kitov PI, Cui L, Paszkiewicz E, Mulvey GL, Armstrong GDA. AICCS. Banff; Alberta, Canada: 2009b. Development of a tailored, polymeric LacNAc inhibitor of EPEC adherence. [Google Scholar]
  18. Humphries RM, Waterhouse CCM, Mulvey GL, Beck P, Armstrong GD. Interactions of enteropathogenic Escherichia coli with pediatric and adult intestinal biopsy specimens during early adherence. Infect Immun. 2009c;77:4463–4468. doi: 10.1128/IAI.00686-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hyland RM, Beck P, Mulvey GL, Kitov PI, Armstrong GD. N-acetyllactosamine conjugated to gold nanoparticles inhibits enteropathogenic Escherichia coli colonization of the epithelium in human intestinal biopsy specimens. Infect Immun. 2006a;74:5419–5421. doi: 10.1128/IAI.00739-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hyland RM, Griener TP, Mulvey GL, Kitov PI, Srivastava OP, Marcato P, Armstrong GD. Basis for N-acetyllactosamine-mediated inhibition of enteropathogenic Escherichia coli localized adherence. J Med Microbiol. 2006b;55:669–675. doi: 10.1099/jmm.0.46344-0. [DOI] [PubMed] [Google Scholar]
  21. Hyland RM, Sun J, Griener TP, Mulvey GL, Klassen JS, Donnenberg MS, Armstrong GD. The bundlin pilin protein of enteropathogenic Escherichia coli is an N-acetyllactosamine-specific lectin. Cell Microbiol. 2008;10:177–187. doi: 10.1111/j.1462-5822.2007.01028.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Iwanczyk J, Damjanovic D, Kooistra J, Leong V, Jomaa A, Ghirlando R, Ortega J. Role of the PDZ domains in Escherichia coli DegP protein. J Bacteriol. 2007;189:3176–3186. doi: 10.1128/JB.01788-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kitagawa M, Ara T, Arifuzzaman M, Ioka-Nakamichi T, Inamoto E, Toyonaga H, Mori H. Complete set of ORF clones of Escherichia coli ASKA library (a complete set of E. coli K-12 ORF archive): unique resources for biological research. DNA Res. 2005;12:291–299. doi: 10.1093/dnares/dsi012. [DOI] [PubMed] [Google Scholar]
  24. Knutton S, Shaw RK, Anantha RP, Donnenberg MS, Zorgani AA. The type IV bundle-forming pilus of enteropathogenic Escherichia coli undergoes dramatic alterations in structure associated with bacterial adherence, aggregation and dispersal. Mol Microbiol. 1999;33:499–509. doi: 10.1046/j.1365-2958.1999.01495.x. [DOI] [PubMed] [Google Scholar]
  25. Krojer T, Pangerl K, Kurt J, Sawa J, Stingl C, Mechtler K, Huber R, Ehrmann M, Clausen T. Interplay of PDZ and protease domain of DegP ensures efficient elimination of misfolded proteins. Proc Natl Acad Sci U S A. 2008;105:7702–7707. doi: 10.1073/pnas.0803392105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lemieux RU, Baker A, Bundle DR. A methodology for the production of carbohydrate-specific antibody. Can J Biochem. 1977;55:507–512. doi: 10.1139/o77-072. [DOI] [PubMed] [Google Scholar]
  27. Leverton LQ, Kaper JB. Temporal Expression of Enteropathogenic Escherichia coli Virulence Genes in an In Vitro Model of Infection. Infect Immun. 2005;73:1034–1043. doi: 10.1128/IAI.73.2.1034-1043.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Macritchie DM, Ward JD, Nevesinjac AZ, Raivio TL. Activation of the Cpx envelope stress response down-regulates expression of several locus of enterocyte effacement-encoded genes in enteropathogenic Escherichia coli. Infect Immun. 2008;76:1465–1475. doi: 10.1128/IAI.01265-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Merz AJ, So M. Interactions of pathogenic neisseriae with epithelial cell membranes. Annu Rev Cell Dev Biol. 2000;16:423–457. doi: 10.1146/annurev.cellbio.16.1.423. [DOI] [PubMed] [Google Scholar]
  30. Nevesinjac AZ, Raivio TL. The Cpx Envelope Stress Response Affects Expression of the Type IV Bundle-Forming Pili of Enteropathogenic Escherichia coli. J Bacteriol. 2005;187:672–686. doi: 10.1128/JB.187.2.672-686.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Raivio TL. Envelope stress responses and Gram-negative bacterial pathogenesis. Mol Microbiol. 2005;56:1119–1128. doi: 10.1111/j.1365-2958.2005.04625.x. [DOI] [PubMed] [Google Scholar]
  32. Raivio TL, Silhavy TJ. The sigmaE and Cpx regulatory pathways: overlapping but distinct envelope stress responses. Curr Opin Microbiol. 1999;2:159–165. doi: 10.1016/S1369-5274(99)80028-9. [DOI] [PubMed] [Google Scholar]
  33. Raivio TL, Silhavy TJ. Periplasmic stress and ECF sigma factors. Annu Rev Microbiol. 2001;55:591–624. doi: 10.1146/annurev.micro.55.1.591. [DOI] [PubMed] [Google Scholar]
  34. Ramboarina S, Fernandes PJ, Daniell S, Islam S, Simpson P, Frankel G, Booy F, Donnenberg MS, Matthews S. Structure of the Bundle-forming Pilus from Enteropathogenic Escherichia coli. J Biol Chem. 2005;280:40252–40260. doi: 10.1074/jbc.M508099200. [DOI] [PubMed] [Google Scholar]
  35. Saldana Z, Erdem AL, Schuller S, Okeke IN, Lucas M, Sivananthan A, Phillips AD, Kaper JB, Puente JL, Giron JA. The Escherichia coli common pilus and the bundle-forming pilus act in concert during the formation of localized adherence by enteropathogenic E. coli. J Bacteriol. 2009;191:3451–3461. doi: 10.1128/JB.01539-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Spiess C, Beil A, Ehrmann M. A temperature-dependent switch from chaperone to protease in a widely conserved heat shock protein. Cell. 1999;97:339–347. doi: 10.1016/s0092-8674(00)80743-6. [DOI] [PubMed] [Google Scholar]
  37. Tobe T, Sasakawa C. Role of bundle-forming pilus of enteropathogenic Escherichia coli in host cell adherence and in microcolony development. Cell Microbiol. 2001;3:579–585. doi: 10.1046/j.1462-5822.2001.00136.x. [DOI] [PubMed] [Google Scholar]
  38. Vanmaele RP, Finlayson MC, Armstrong GD. Effect of enteropathogenic Escherichia coli on adherent properties of Chinese hamster ovary cells. Infect Immun. 1995;63:191–198. doi: 10.1128/iai.63.1.191-198.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Vanmaele RP, Heerze LD, Armstrong GD. Role of lactosyl glycan sequences in inhibiting enteropathogenic Escherichia coli attachment. Infect Immun. 1999;67:3302–3307. doi: 10.1128/iai.67.7.3302-3307.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Vogt SL, Nevesinjac AZ, Donnenberg M, Armstrong GD, Raivio TL. The Cpx envelope stress response both facilitates and inhibits elaboration of the enteropathogenic Escherichia coli bundle forming pilus. doi: 10.1111/j.1365-2958.2010.07145.x. accompanying article. Submitted. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zhang HZ, Donnenberg MS. DsbA is required for stability of the type IV pilin of enteropathogenic escherichia coli. Mol Microbiol. 1996;21:787–797. doi: 10.1046/j.1365-2958.1996.431403.x. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp Table s1-s2
NIHMS207650-supplement-Supp_Table_s1-s2.pdf (82.6KB, pdf)

RESOURCES