ClpXP controls the expression of LEE genes in enterohaemorrhagic Escherichia coli

Abstract

Enterohaemorrhagic Escherichia coli (EHEC) contains a 36-kb pathogenicity island termed the locus of enterocyte effacement (LEE), which encodes a type III secretion system (TTSS) and virulence proteins. In this paper, we show that the O157:H7 Sakai clpPX mutant strongly impaired the secretion of virulence proteins by TTSS and repressed transcription from all the LEE promoters. The rpoS mutation in O157:H7 Sakai enhanced the transcription from all the LEE promoters and the secretion of virulence proteins, and it could partially suppress the defects of the clpPX mutation. These data indicate that the O157:H7 Sakai ClpXP protease is a positive regulator for LEE expression and that this regulation occurs by two pathways: the σ^S-dependent and -independent pathways.

EHEC, LEE, clpXP, TTSS, rpoS

1 Introduction

Enterohaemorrhagic Escherichia coli (EHEC) is a human pathogen that enters the intestinal tract and causes several outbreaks of severe bloody diarrhea (hemorrhagic colitis) that can subsequently lead to hemolytic uraemic syndrome (HUS) [1,2]. The hallmark of EHEC infection is the ability of the bacteria to attach intimately to the host cell membrane, destroy the microvilli, and induce the formation of pedestal structures, which are composed of cytoskeletal proteins, upon which the bacteria sit. This histopathological lesion is termed as an attaching/effacing (A/E) lesion. Factors responsible for the A/E lesion formation are encoded by the locus of enterocyte effacement (LEE). This locus encodes a type III secretion system (TTSS), a regulator (Ler), an adhesin (intimin) and its receptor (Tir) that is responsible for intimate attachment, several secreted proteins, and their chaperones [3–6]. The proteins secreted by the TTSS consist of effectors (Tir, EspG, EspF, Map, and EspH), which are involved in modulating the host cytoskeleton, as well as translocators (EspA, EspD, and EspB) required for translocating effectors into host cells [7,8]. The genes within the LEE pathogenicity island are arranged in five large polycistronic operons that are designated as LEE1 through LEE5 [9].

ATP-dependent proteases (ClpXP, Lon, ClpAP and HslVU) carry out more than 90% of all cellular proteolysis, and they modulate the cellular amounts of several important transcriptional regulators in E. coli [10]. All ATP-dependent proteases that have been identified so far belong to the AAA+ (ATPases associated with diverse cellular activity plus) superfamily [11]. We have recently reported that ClpXP and Lon control the expression of two TTSSs in Salmonella enterica serovar Typhimurium one of which is encoded by the flagellar regulon and the other by Salmonella Pathogenicity Island 1 [12,13]. In order to investigate the possible role of the ClpXP protease in the production of LEE-encoded proteins, we constructed a ΔclpPX::Cm mutant from EHEC strain O157:H7 Sakai and examined the transcription of the LEE genes and the secretion of effector and translocator proteins by the TTSS.

2 Materials and methods

2.1 Bacterial strains and culture conditions

The bacterial strains used in this study are listed in Table 1. Bacteria were grown in Luria–Bertani (LB) medium (1% tryptone, 0.5% yeast extract, and 1% NaCl) or DMEM without phenol red (Sigma, D5921) containing 0.45% (w/v) glucose and 2 mM l-glutamine. Antibiotics were added at the following concentrations: ampicillin (Ap, 50 μg/ml), chloramphenicol (Cm, 20 μg/ml), kanamycin (Km, 20 μg/ml), and nalidixic acid (NA, 25 μg/ml).

1

Bacterial strains used in this study

Strain	Relevant characteristics	Reference or source
DH5α	F−endA1 hsdR17supE44 thi-1 recA1 gyrA(Nalr) relA1Δ(lacZYA-argF) U169deoR(Φ80d lacΔ(lacZ)M15)	Lab. collection
SM10 λpir	thi, thr, leu, tonA, lacY, supE, recA::RP4-2 Tc::Mu(Kmr)λpir	Lab. collection
O157:H7 Sakai	Escherichia coli O157:H7 derived from Sakai outbreak (RIMD 0509952)	[36]
CS5295	O157:H7 Sakai NAr	Present study
CS5297	CS5295 ΔclpPX::Cm	Present study
CS5524	CS5295 ΔrpoS::Km	Present study
CS5525	CS5297 ΔrpoS::Km	Present study

2.2 Construction of ΔclpPX and ΔrpoS mutant

In order to inactivate clpP and clpX genes, we used Gateway™ technology (Invitrogen™). The attB-flanked PCR products of the clpPX operon was amplified by PCR from the O157:H7 Sakai genomic DNA template using attB sequence-attached primers (Table 2). The BP reaction used for cloning PCR products into donor vector (pODNR201) to construct entry clones was essentially carried out according to the manufacturer's instructions (Invitrogen™). The resulting plasmid, pTKY655 (clpPX), was cleaved with Bst XI and Bss HII. This restriction digestion removed a central fragment of the clpPX operon. Subsequently, the cleaved ends were filled by T4 DNA polymerase and ligated to a filled fragment that encoded the Cm-resistance gene that had been generated from Bam HI-digested pTKY702. The resulting entry clones, pTKY663 (ΔclpPX::Cm) of the target DNA were cloned by the LR reaction into destination vector, pABB-CR2, which is a transferable suicide vector [14]. The resulting destination clones pTKY667 (ΔclpPX::Cm) was used to construct O157:H7 Sakai ΔclpPX mutant. ΔclpPX::Cm mutant was selected based on the resistance to Cm and NA. A double crossover event that resulted in ΔclpPX::Cm mutants was assessed by its sensitivity to Ap and its resistance to sucrose. PCR and immunoblotting analysis confirmed replacement of the wild-type gene by ΔclpPX::Cm.

2

PCR primers used in this study

Gateway cloning
ClpPX	F:	GGGGACAAGTTTGTACAAAAAAGCAGGCTGAACTGATGGACAACATGCGCAATG
ClpPX	R:	GGGGACCACTTTGTACAAGAAAGCTGGGTGTACGTCAGTATATGGGGATGTTTC
Lambda red system
RpoS	F:	ATGAGTCAGAATACGCTGAAAGTTCATGATTTAAATGAAGGTGTAGGCTGGAGCTGCTTC
RpoS	R:	TTACTCGCGGAACAGCGCTTCGATATTCAGCCCCTGCGTTCATATGAATATCCTCCTTAG
Northern blotting
LEE1 (ler)	F:	AATTCACATACAACAAGTCC
LEE1 (ler)	R:	CTAGCCATTCATCATGAACC
LEE2 (sepZ)	F:	TCCTTCTGGTGCAGTAATGC
LEE2 (sepZ)	R:	TGGATAGCAACAGTGCTTGC
LEE3 (orf12)	F:	GTTGTTTGTTAACCAATATCG
LEE3 (orf12)	R:	GCCACATACCAATGAGCACG
LEE4 (espB)	F:	TATTTGTAAGCTGATGCTGG
LEE4 (espB)	R:	GGCTGCTTTCGTTGTTGTGG
LEE5 (tir)	F:	GTGAATAATTCAATTCCTCC
LEE5 (tir)	R:	CTCTAACAGCTCCAGTATCC
gadE	F:	CCTGTCAGAGATCAAAAAAGTAGGC
gadE	R:	CTAAAAATAAGATGTGATACCCAGG
gadX	F:	GAATTGTCTAATTGCGTACG
gadX	R:	GTCATCCCATAATAATTTCG
16S rRNA	F:	GGCCTAACACATGCAAGTCGAAC
16S rRNA	R:	ACAACACCTTCCTCCCCGCT
clpPX operon
clpPX	F:	GCATAACGCGCTGGATCCGCACAAAGGCCC
clpPX	R:	TCAGATGACACGACTAAGCTTCACGCCATC

The one-step method of chromosomal gene inactivation that utilizes the Lambda Red recombinase in E. coli K-12 [15] was used to construct O157:H7 Sakai ΔrpoS::Km and ΔrpoS::Km ΔclpPX::Cm double mutants. O157:H7 Sakai wild-type and ΔclpPX::Cm mutant cells were electroporated with plasmid pKD46 containing the Lambda Red recombinase. These cells were subsequently electroporated with a linear DNA fragment containing ΔrpoS::Km that had been amplified with the primers rpoS F and rpoS R (Table 2) using pKD4 [15] as the template. Km-resistant clones were selected and streaked at 37 °C, and Km-resistant clones that had lost pKD46 were isolated. PCR and immunoblotting analysis using anti-σ^S antiserum confirmed replacement of the wild-type gene by ΔrpoS::Km and the formation of O157:H7 Sakai ΔrpoS::Km and ΔclpPX::Cm ΔrpoS::Km double mutants.

2.3 Construction of pTKY672

Plasmid pTKY672, in which clpPX operon expression is under the control of its own promoter, was constructed. This was carried out by amplifying the clpPX operon, using the PCR primers clpPX F and clpPX R (Table 2) and subsequently cloning the fragment into Bam H1–Hin dIII-digested pBR322.

2.4 Isolation of secreted proteins and gel electrophoresis

Bacterial strains were grown overnight as standing cultures in LB medium at 30 °C. The preculture was diluted 1:100 in DMEM containing 0.45% (w/v) glucose and 2 mM l-glutamine. Cells were further cultured with shaking for 5 h at 37 °C and harvested by centrifugation. The supernatant was centrifuged again to remove any remaining cells and the proteins in the supernatant were precipitated by addition of ice-cold trichloroacetic acid (TCA) to a final concentration of 10% and incubation on ice for 30 min. After centrifugal separation, the pellets were washed twice with ice-cold acetone and resuspended in Laemmli sample buffer. Equal amounts of total cell lysates (20 μg) and the concentrated samples of secreted proteins from cells corresponding to 400 μg total protein were analyzed by SDS–PAGE and immunoblotting. Quantification of the amounts of total and secreted proteins was performed using the Bradford assay reagent (Bio-Rad) with BSA as standard. Gel electrophoresis was carried out according to the Laemmli method using SDS–polyacrylamide gels and staining with Coomassie brilliant blue.

2.5 Isolation of membrane fraction

Bacterial cells were grown at 37 °C for 4 h in 25 ml of DMEM containing 0.45% (w/v) glucose and 2 mM l-glutamine; subsequently, they were harvested by centrifugal separation. Cells were resuspended in 50 mM Tris buffer (pH 8.0) containing 2 mM EDTA, 1 mM PMSF, 5 mg/ml lysozyme, and 20% sucrose, and were then incubated for 30 min on ice. Cell lysis was performed by the addition of a 10-fold volume of 10 mM Tris buffer (pH 8.0) containing 2 mM EDTA, followed by mixing and then sonication. Intact cells were removed by centrifugation at 10,000g for 15 min at 4 °C. The membrane fraction was isolated by subsequent centrifugation at 100,000g for 1 h at 4 °C. An equal amount of membrane proteins (20 μg) was analyzed by SDS–PAGE.

2.6 Immunoblotting and quantifications

Gel-resolved proteins were transferred to a PVDF membrane (Millipore). Immunoblotting analysis was performed by using a guinea pig anti-EscC [14], a rabbit anti-σ^S [16], -EspA, -EspB, -intimin, -Tir [17] or -Ler [18] antibodies coupled with a goat anti-guinea pig or a goat rabbit IgG conjugated to alkaline phosphatase secondary antibody. Detection was carried out by using BCIP/NBT (Wako). The immunoblots were scanned and quantified using MacBAS software (Fuji Film). Standard deviation (SD) was calculated using three different data sets.

2.7 Northern blot analysis

Bacterial cells were grown at 37 °C in 25 ml of DMEM containing 0.45% (w/v) glucose and 2 mM l-glutamine. Total RNAs were prepared by the hot-phenol extraction method [19]. The Probes to detect the transcript from LEE1 through LEE5, gadE, gadX, and rRNA promoters were amplified by PCR, and the primers used are listed in Table 2. The same amount of RNAs were resolved by 1.0% agarose gel electrophoresis in the presence of formaldehyde [19] and blotted onto a Immobilon-NY+ membrane (Millipore). The membrane was hybridized to a DNA probe labeled with alkaline phosphatase by using a Gene Images AlkPhos Direct Labelling and Detection system (Amersham) and washed. The signals were visualized with a CDP-Star™ Detection Reagent (Amersham) in accordance with the manufacturer's protocol. Developed X-ray films were scanned and quantified using the MacBAS software (Fuji Film). Values were normalized to the levels of 16S RNA in each sample. SD was calculated using three different data sets.

3 Results

3.1 ClpXP is a positive regulator for the expression of LEE genes

In order to investigate the possible role of ClpXP in the production of LEE-encoded proteins, we constructed a ΔclpPX::Cm mutant from O157:H7 Sakai. We examined the secretion of LEE-encoded translocators (EspA, B) and an effector protein (Tir) from the clpPX mutant in the culture medium by immunoblotting analysis. The secretion of Tir, EspA, and EspB was strongly inhibited by the clpPX mutation (less than 5% when compared with that in the wild-type cells) (Figs. 1 and 2). We also investigated the cellular amount of LEE translocators, effector proteins, and a main component of the TTSS (EscC) in the clpPX mutant (Figs. 1 and 2) [20]. Our data show that Intimin, Tir, EspA, and EscC proteins were markedly reduced to less than 10% when compared with that in the wild-type cells. However, a significant amount of EspB was observed in the clpPX mutant cells (21.7 ± 6.0% when compared with that in the wild-type cells) (Fig. 1). A plasmid encoding clpPX could complement defects in Intimin, Tir, EspA, and EspB synthesis and secretion caused by the clpPX mutation in O157:H7 Sakai (Fig. 1). It has been shown that Ler is encoded by the LEE1 operon that can activate the expression of LEE2 through LEE5 promoters [21–23]. Therefore, we analyzed the cellular amounts of Ler by immunoblotting analysis. A significant reduction in Ler was observed in the clpPX mutant cells (31.4 ± 5.4% when compared with that in the wild-type cells) (Fig. 2).

1

Inactivation of ClpXP impaired the synthesis and secretion of LEE-encoded effectors. The cells were grown in DMEM for 5 h at 37 °C. Equal amounts of total cell lysates (20 μg) (A) and concentrated samples of secreted proteins from cells corresponding to 400 μg of total proteins (B) were analyzed by SDS–PAGE and immunoblotting with anti-σ^S, -intimin, -Tir, -EspA, and -EspB antibodies. WT, wild type; ΔPX, ΔclpPX mutant; ΔPX+ pPX, ΔclpPX mutant complemented by a plasmid coding clpPX; Total, total cell lysate; Secreted, secreted proteins; M, molecular weight marker.

2

σ^S negatively regulates the synthesis and secretion of LEE-encoded proteins. The cells were grown in DMEM for 5 h at 37 °C. Equal amounts of total cell lysates (20 μg) (A), membrane (20 μg) (B), and concentrated samples of secreted proteins from cells corresponding to 400 μg of total proteins (C) were analyzed by SDS–PAGE and immunoblotting with anti-Ler, -EscC, -Tir, and -EspB antibodies. Asterisk indicates non-specific background bands. WT, wild type; ΔPX, ΔclpPX mutant; ΔS, ΔrpoS mutant; ΔPX+ΔS, ΔclpPXΔrpoS double mutant; Total, total cell lysate; Secreted, secreted proteins; M, molecular weight marker; % of WT ± SD; The relative quantity of the proteins to the wild-type cells is shown along with the SD.

To assess whether clpPX plays a positive role in the expression, we examined the transcription from the LEE promoters using Northern blotting. Our results show that the levels of transcript from LEE2 through LEE5 promoters were strongly repressed (less than 5% when compared with that in the wild-type cells) by the clpPX mutation (Fig. 3). However, a low level of transcript from the LEE1 (26 ± 4.4%) promoter was observed in the clpPX mutant (Fig. 3).

3

Analysis of mRNA generated in the LEE operons by Northern blotting. The cells were grown in DMEM for 5 h at 37 °C, and the total RNA was extracted. LEE1 (ler), LEE2 (sepZ), LEE3 (orf12), LEE4 (espB), LEE5 (tir), and 16S rRNA probes were used to detect each mRNA. The arrow indicates the approximate size of the transcript. WT, wild type; ΔPX, ΔclpPX mutant; ΔS, ΔrpoS mutant; ΔPX+ΔS, ΔclpPXΔrpoS double mutant. % of WT ± SD. The relative quantity of the mRNAs to the wild-type cells is shown along with the SD.

3.2 ClpXP regulates LEE expression in a σ^S-dependent and -independent manner

One of the substrates for ClpXP is a stationary phase-specific transcription factor (σ^S) [24]. σ^S is present at very low levels during exponential cell growth and its stability increases 10-fold following transition to the stationary phase or when subjected to other stress treatments in wild-type cells, [25]. Since the rapid degradation of σ^S is largely due to ClpXP, the clpPX mutation results in the accumulation of large amounts of σ^S in the exponential phase cells. The immunoblotting data show that exponential phase cells of the O157:H7 Sakai clpPX mutant used in this study also accumulated large amounts of σ^S (Fig. 1A). σ^S promotes expression of more than 50 genes involved in responses to many stresses, including starvation, acid shock, and oxidative damage, as well as the transition to stationary phase [26]. It has been reported that the expression of LEE in EHEC:H7 EDL933 is activated at the mid-exponential phase and repressed at the stationary phase [27]. Therefore, we hypothesized that σ^S might control the expression of the negative regulator(s) that is required for the repression of LEE-encoded genes. In order to investigate this possibility, we disrupted the rpoS gene in the wild-type and clpPX mutant cells and analyzed the secretion of LEE effector proteins and transcription from the LEE promoters. Interestingly, the amount of Ler and the level of all LEEs mRNA in rpoS single mutant cells increased when compared with those in the wild-type cells (Figs. 2 and 3). These results clearly indicated that ClpXP controls the expression of the unidentified negative regulator(s) of LEE1 expression by regulating the cellular amount of σ^S.

We also examined whether the rpoS mutation could completely suppress the defects of the clpPX mutation. Our results showed that the rpoS mutation partially suppressed the defect of clpPX mutation. The secretion of Tir and EspB recovered to almost the wild-type level (Fig. 2). The levels of LEE1 mRNA in rpoS single mutant and clpPX rpoS double mutant cells were approximately the same (Fig. 3). The amounts of Ler in both rpoS single mutant and clpPX rpoS double mutant cells were also approximately the same and increased to more than 3-fold when compared to that in the wild-type cells (Fig. 2). However, the levels of LEE2 through LEE5 mRNA in clpPX rpoS double mutant cells (Fig. 3) and the secretion of Tir and EspB from these cells (Fig. 2) were still lower (less than 50%) than those in the rpoS single cells, although the cellular amount of Ler was similar in both single and double mutant cells (Fig. 2). These data indicate that ClpXP regulates the expression of LEE2 through LEE5 promoters through an unidentified negative regulator(s) that is not controlled by σ^S.

It has been reported that σ^S controls the expression of several stress-resistant genes, including the glutamate-dependent acid resistance (GDAR) system [28]. Recently, GadE and GadX that belong to the GDAR system have been found to be the negative regulators of LEE expression [29,30]. It is suggested that the accumulation of σ^S due to the clpPX mutation might result in the overproduction of GadE and GadX. Indeed, the level of transcript from the gadE and gadX genes increased due to the clpPX mutation and decreased due to the rpoS mutation (Fig. 4).

4

Analysis of mRNA generated in the gadE and gadX genes by Northern blotting. The cells were grown in DMEM for 4 h at 37 °C, and the total RNA was extracted. The gadE and gadX gene specific probes were used to detect each mRNA. The arrow indicates the approximate size of the transcript. WT, wild type; ΔPX, ΔclpPX mutant; ΔS, ΔrpoS mutant; ΔPX+ΔS, ΔclpPXΔrpoS double mutant. % of WT ± SD. The relative quantity of the mRNAs to the wild-type cells is shown along with the standard deviation.

4 Discussion

TTSSs have been identified in several divergent mammalian and plant pathogens [31,32]. They are functionally conserved and act as “molecular syringes” by injecting effector proteins into eukaryotic target cells in order to modulate various cellular functions [33]. We have reported that the AAA⁺ proteases (ClpXP and Lon) in Salmonella enterica serovar Typhimurium control the expression of the TTSSs, which are responsible for flagella formation and translocation of SPI1 effector proteins [12,13]. In order to elucidate whether ClpXP protease regulates TTSSs in O157:H7 Sakai, we inactivated this protease and investigated the secreted proteins.

We observed significant cellular amounts of EspB (Fig. 1) in the clpPX mutant. It is known that a major portion of the TTSS apparatus is encoded on the LEE1, LEE2, and LEE3 operons [21]. We observed that cellular amounts of a TTSS component (EscC) were greatly reduced by the clpPX mutation (Fig. 2). Therefore, the clpPX mutant appears to lack the TTSS apparatus in the cells. We were able to detect a 1.2-kDa transcript from the LEE4 operon, although only a small amount is present (Fig. 3). The clpPX mutant may possibly produce a lower amount of EspB in the cells without secretion. This might be the reason for the significant accumulation of EspB in the clpPX mutant cells.

How does the ClpXP protease regulate LEE expression? It was previously reported that the transcription of the LEE4 promoter was activated by σ^S, and the 10 sequences of this promoter in O157:H7 (EDL933) exhibit a high degree of homology to those of the osmE promoter, which is σ^S dependent [34]. We compared the promoter sequences of O157:H7 (EDL933) and our strain O157:H7 Sakai (RIMD 0509952); these were found to be completely identical. However, northern blotting data obtained by us suggest that the contribution of σ^S accumulated as a result of the clpPX mutation to the transcriptional regulation of the LEE4 promoter is very small. The clpPX mutation might sequester the positive effect of σ^S on the LEE4 promoter by an unknown mechanism.

We showed that the levels of mRNA from the gadE and gadX genes were increased by the clpPX mutation and decreased by the rpoS mutation. (Fig. 4) Shin et al. [29] showed that GadX could repress the transcription of LEE genes through Per in enteropathogenic E. coli (EPEC). The PerC protein of EPEC, encoded by the pEAF plasmid, is an activator of the LEE pathogenicity island via the LEE1 promoter. PerC1 (also termed PchABC) – the PerC-like protein family identified from EHEC genome analyses – can also activate both promoters in a manner similar to that of EPEC PerC [35]. Therefore, GadX in O157:H7 Sakai might also regulate the expression of the LEE genes via PerC1. This might be a reason for the σ^S-dependent LEE repression by the clpPX mutation.

We showed that despite the high amount of Ler in the clpPX rpoS double mutant cells, the level of LEE2 to LEE5 mRNA was lower in these cells than in the rpoS single mutant cells (Figs. 2 and 3). These data indicate that ClpXP might control the amount or activity of the repressor protein(s) of the LEE2 to LEE5 promoters, with the exception of those of the LEE1 promoter. Alternatively, it might control the negative regulator(s) of Ler activity that is not controlled by σ^S. This might explain the strong repression of the LEE2 to LEE5 promoters, which is more than that of the LEE1 promoter, in the clpPX mutant cells.

In summary, our results indicate that ClpXP is an important regulator for the modulation of LEE expression. This regulation occurs by two pathways: the σ^S-dependent and -independent pathways. Future studies will define the control mechanisms for the LEE expression by ClpXP.

Acknowledgments

We thank Ms. Y. Akema for her helpful assistance. This work was supported by Grants-in-Aid for scientific research (13470058 to T.Y. and 17590387 to T.T.) from the Ministry of Education, Culture, Sports, Science and Technology of the Japanese Government.

Author notes

Section on DNA Replication, Repair & Mutagenesis, NICHD, NIH, Building 6, Room 1A11, 9000 Rockville Pike, Bethesda, MD 20892-2725, USA.