Functional analysis of EspB from enterohaemorrhagic Escherichia coli

Hao-Jie Chiu and Wan-Jr Syu

Institute of Microbiology and Immunology, National Yang-Ming University, Beitou 112, Taipei, Taiwan

Correspondence
Wan-Jr Syu
wjsyu{at}ym.edu.tw


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In enterohaemorrhagic Escherichia coli (EHEC), the type III secretion protein EspB is translocated into the host cells and plays an important role in adherence, pore formation and effector translocation during infection. The secretion domain of EspB has been mapped previously. To define the other functional determinants of EspB, several plasmids encoding different fragments of EspB were created and analysed to see which of them lost the functions of the full-length molecule. One finding was that residues 118–190 of EspB were required for both efficient translocation of EspB and interaction of EspB with EspA. Additionally, the segment consisting of residues 217–312 was necessary for bacterial adherence. Furthermore, a predicted transmembrane domain (residues 99–118) was found to be critical for EHEC to cause red blood cell haemolysis, presumably by forming pores in the cell membrane. The same segment was also important for actin accumulation induced beneath the bacterial-attachment site. Taken together, these data indicate that the EspB protein (312 residues in total) has functions associated with its different regions. These regions may interact with each other or with other components of the type III system to orchestrate the intricate actions of EHEC during infection.


Abbreviations: A/E lesion, attaching and effacement lesion; EHEC, enterohaemorrhagic E. coli; EPEC, enteropathogenic E. coli; LEE, locus of enterocyte effacement; RBC, red blood cell; TTSS, type III secretion system


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Escherichia coli O157 : H7 is the major pathogen that causes a serious illness known as enterohaemorrhagic diarrhoea and haemolytic uraemic syndrome (Nataro & Kaper, 1998). Several virulence factors have been reported to contribute to the pathogenesis of this microbe. One of the factors is a type III secretion system (TTSS) that induces the attaching and effacement lesion (A/E lesion) of enterocytes. The TTSS is encoded by the locus of enterocyte effacement (LEE) island, which contains a total of 41 genes. Homologous LEE islands have been found in Gram-negative pathogens, such as enteropathogenic E. coli (EPEC) and Citrobacter rodentium (McDaniel & Kaper, 1997; Schauer & Falkow, 1993), but not reported to exist in other, non-pathogenic E. coli.

Most of the LEE genes are organized into five major operons: LEE1, LEE2, LEE3, LEE4 and LEE5. EspA (Kenny et al., 1996), EspB (Donnenberg et al., 1993) and EspD (Donnenberg et al., 1993; Lai et al., 1997) are encoded by LEE4 and secreted by the TTSS; deletion of these three genes abolished the translocation of effector proteins, such as Tir (Kenny et al., 1997), EspF (Crane et al., 2001) and MAP (Kenny & Jepson, 2000). EspA is believed to form a hollow, filamentous structure on the bacterial surface to deliver the effector proteins (Knutton et al., 1998) and it may also play a role in adherence of enterohaemorrhagic E. coli (EHEC) to the host cell (Cleary et al., 2004). EspD has been found to target the host-cell membrane (Wachter et al., 1999), whereas EspB is translocated to both membrane and cytoplasm (Wolff et al., 1998). Both EspB and EspD are believed to be involved in pore formation on the membranes of the infected cells (Ide et al., 2001) and have been classified as translocators (Roe et al., 2003). EspB has also been reported to interact with EspA (Hartland et al., 2000) and EspD (Ide et al., 2001), and complexes formed by these three proteins may participate in the initial step of bacterial adherence (Nougayrède et al., 2003). On the host membrane, the N-terminal region of EspB has been found to bind directly to the intracellular {alpha}-catenin and to result in the recruitment of {alpha}-catenin underneath the bacterial-adherence site (Kodama et al., 2002). Recently, it has also been demonstrated that host-cell {alpha}1-antitrypsin interacts with EspB and EspD, and this interaction has been implicated to block the formation of the EspB/EspD translocation pore (Knappstein et al., 2004).

In our previous study on the functional domain of EHEC EspB, one essential region (residues 1–118) and two auxiliary segments (residues 118–190 and 191–282) were defined for EspB secretion (Chiu et al., 2003). When appropriate conditions are supplied, such as the provision of M9 medium for cell cultivation, EspB and other TTSS-secreted proteins can be found in abundance in the medium. In this study, we further dissected the EspB molecule and determined the regions required for its own translocation and its interaction with EspA. As bacteria lacking espB lose adherence ability (Donnenberg et al., 1993), we also examined whether there is a relationship between the EspB structure and bacterial adherence. We report here the results from a series of deletion analyses.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and cell culture.
EHEC strain 43888, which does not produce either Shiga-like toxin I or II and lacks stx genes, was used in this study; this strain and its mutant with espB deleted (E. coli 43888 {Delta}B) have been described previously (Chiu et al., 2003). The tir deletion mutant (E. coli 43888 {Delta}T) was constructed similarly. In brief, bacterial chromosomal DNA was used as the template for PCR amplification. Primers Lee-22239 (5'-CAGAAAGTTGATGAGCAGTGCGGC-3') and NTir-171R (5'-TATCAATCAGCCATAGAATTCC-3') were used to amplify a fragment covering a 5' flanking region and a small 5' sequence of tir. The obtained PCR fragment was digested with BamHI, which made a single cut near the 5' end of the amplified sequence. A second PCR using primers NTir-F (5'-TTATGCCTATTGGTAACCTTGG-3') and NTir-R(PstI) (5'-CAGCTGCAGACGAAACGATGGGATCCCGG-3') was performed to generate a fragment covering a 3' segment of tir; the nucleotide sequences recognized by restriction enzymes are underlined. This PCR product was digested with BamHI and EcoRV. Then, these two PCR products were three-way-ligated into BamHI-restricted PinPoint Xa-3 (Promega) to result in PinPoint Xa-3-{Delta}Tir. To construct pKO3-{Delta}Tir, PinPoint Xa-3-{Delta}Tir was digested with NotI/NruI and the fragment containing the tir-flanking regions was ligated into NotI/SmaI-digested pKO3 (kindly provided by G. M. Church, Harvard University, MA, USA). The resulting plasmid was electroporated into E. coli 43888. Bacteria with homologous recombination were then selected and confirmed with an expected deletion at tir as described previously (Chiu et al., 2003).

Bacteria were regularly grown in LB medium (Difco). To activate the type III secretion, bacteria were transferred into M9 medium. Media were supplemented with ampicillin (100 µg ml–1) or chloramphenicol (34 µg ml–1) when necessary. HeLa cells were maintained in Dubelcco's modified Eagle medium (DMEM) supplemented with 10 % (v/v) fetal calf serum and cultured at 37 °C in the presence of 5 % CO2.

General recombinant DNA techniques.
Unless otherwise stated, restriction endonucleases, DNA-modifying enzymes and polymerases were purchased from New England BioLabs. DNA-manipulation procedures were followed either as described by Sambrook & Russell (2001) or as recommended by the manufacturers. DNA was purified from a mini-column (Qiagen) and sequenced automatically by a contract service (Mission Biotech).

Plasmids.
To create plasmid pBP312D, pB312D (Chiu et al., 2003) was amplified with primers PQE60-2R (5'-TGACTGCAGGGTTAATTTCTCCTC-3') and PEspB-9 (5'-TAACTGCAGATGAATACTATTGATAAT-3'). After amplification, the inverted PCR product was digested with PstI and self-ligated by using T4 DNA ligase. To create plasmids pB1–312, pB1–282, pB1–250, pB1–220 and pB1–190, pBP312D was used as the template and PCR-amplified with primer pairs PB939RS (5'-CGCGGATCCTTACCCAGCTAAGCGACCCG-3')/PEspB-9, PB846RS (5'-CGCGGATCCTTAATCATCCTGCGTTCTGCG-3')/PEspB-9, PB750RS (5'-GCGGGATCCTTAAGTCGATTTGACGGACTC-3')/PEspB-9, PB660RS (5'-GCGGGATCCTTACAGTTTATCTACGGAATTCAA-3')/PEspB-9 and PB570RS (5'-CGCGGATCCTTAAACATCATCTGCAACGCC-3')/PEspB-9, respectively. After amplification, the PCR products were digested with PstI/BamHI and ligated with the vector derived from pBP312D digested with the same enzymes. Plasmids pB{Delta}191–253, pB{Delta}CC, pB{Delta}118–190 and pB{Delta}TM were created by an inverted PCR method (Vandeyar et al., 1988; Weiner et al., 1993) using pB312 as template. The primer pairs used were PB570R (5'-TTAAACATCATCTGCAACGCC-3')/PB760 (5'-AATGAACAACGTGCGAAG-3'), PB558R (5'-AACGCCAGATGCACGGCT)/PB649 (5'-GTAGATAAACTGACCAATACC-3'), PEspB-14R (5'-TGCTGCAAAAGAACCTAA-3')/PEspB-19 (5'-GCGAAAGCCACTGAC-3') and PB294R (5'-AGCGGTTGCCGCGGCTTT-3')/PB355 (5'-AACAACGCGGCTAAAGGG-3'), respectively. The PCR products were phosphorylated by T4 DNA kinase and then self-ligated to produce the designated constructs. After so doing, these plasmid-encoded EspB constructs had authentic EspB N termini and carried no tag at the C termini of the molecules.

To create pEspA–His, the entire espA sequence was PCR-amplified from the chromosomal DNA of E. coli 43888; the primers used were PEspA-1 and PEspA-3R. These had the nucleotide sequences 5'-CTAACCATGGATACATCAAATGCA-3' and 5'-CGCAGATCTTTTACCAAGGGATATTGC-3', respectively. After PCR amplification, the obtained DNA fragment was digested with NcoI/BamHI and cloned into NcoI/BamHI-restricted pQE60 (Qiagen); the resulting pEspA–His expressed a recombinant EspA tagged with a hexahistidine epitope at the C terminus (i.e. EspA–Hisx6).

Fractionation of cell components after bacterial infection.
A previously described method (Gauthier et al., 2000) was followed with a slight modification. In brief, HeLa cells in 100 mm plates cultured to 80 % confluence were washed with PBS. Overnight-grown bacteria were diluted 1 : 100 (v/v) in DMEM and added to the HeLa cells for 6 h. After infection, the cells were washed twice with PBS and scraped off the plates into a cold imidazole buffer (3 mM, pH 7·4) containing 250 mM sucrose, 0·5 mM EDTA, 1 mM sodium orthovanadate, 1 mM sodium fluoride and 1 µM pepstatin. The mixture was passed through a 22-gauge needle eight times and the unbroken cells with the bacteria were pelleted down by centrifugation twice at 3500 g for 15 min. To collect the disrupted membrane fraction of the cells, the recovered supernatant was further centrifuged at 20 000 g for 45 min. Cytosolic proteins in the supernatant were precipitated by adding 1/9 vol. trichloroacetic acid and further incubated at 4 °C for 1 h. Proteins in the precipitates were then collected by centrifugation. Proteins from each sample were dissolved in SDS sample buffer and separated by SDS-PAGE (12 % gel), followed by analysis using Western blotting.

Immunoblotting.
Western blotting analysis of antigens was performed as described previously (Hsu et al., 2000). The anti-EspB antibody has been described previously (Chuang et al., 2001) and anti-EspA was prepared by immunizing rabbits with gel-purified EspA from the secreted proteins of EHEC. Anti-OmpA was a mouse mAb prepared previously (Yu et al., 2000). Mouse mAbs against syndecan (Santa Cruz) and actin (Chemicon) were obtained commercially. To detect the primary antibodies bound to the antigens on the nitrocellulose membrane, horseradish peroxidase-conjugated goat secondary antibodies (Sigma) were used. The membranes were finally developed by a chemiluminescence reagent (Hsu et al., 2000).

Affinity purification with an Ni2+ column.
Ni2+–NTA agarose beads (Qiagen) were packed into columns and charged with Ni2+ by passing through five bed volumes of 100 mM NiSO4, and finally equilibrated in Tris-buffered saline (TBS). E. coli lysates derived from bacteria harbouring pEspA–His, pQE60, pB1–312, pB1–282, pB1–190, pB{Delta}118–190 or pB{Delta}CC were prepared in TBS and clarified by centrifugation. The lysate containing EspB or its truncated form was mixed with an equal volume of that containing EspA–Hisx6. The mixture was inverted gently at 4 °C for 1 h and then passed through the TBS-equilibrated Ni2+ column. After washing with TBS containing 150 mM imidazole, the proteins retained in the columns were eluted with 500 mM imidazole in TBS and analysed by Western blotting.

Adherence assay.
Analysis of the adherence of EHEC to cells was modified from the method described by Gansheroff et al. (1999). In brief, 2x105 HeLa cells were plated in 12-well plates. After incubation overnight, the cells were washed with PBS and maintained in DMEM without any additives. Overnight-grown bacteria were diluted 1 : 100 (v/v) and added to the cells, and infection was maintained for 6 h. Thereafter, the cells were harvested and lysed in PBS containing 10 % (w/v) saponin solution (5 mM Tris/HCl, pH 7·4, 0·4 mM NaVO4, 0·1 mM PMSF) at 4 °C for 10 min. The lysates were then diluted serially in LB medium and plated on LB agar plates supplemented with ampicillin (100 µg ml–1). After 16 h incubation at 37 °C, c.f.u. were scored and the relative adherence efficiencies were calculated.

Red blood cell (RBC) haemolysis assay.
A previously described method (Warawa et al., 1999) was followed. In brief, human RBCs, type B, were washed with PBS three times and suspended in PBS to a final concentration of 3 % (v/v). The RBCs were then plated on poly-lysine-coated 12-well plates (700 µl per well) for 20 min at 37 °C. After two washes with PBS, the cells were kept in DMEM without phenol red. Overnight-grown bacteria were diluted 1 : 100 (v/v) to the RBC culture. After 6 h incubation at 37 °C in the presence of 5 % CO2, the culture medium was collected and centrifuged. The released haemoglobin was evaluated by measuring OD543 of the supernatant with an ELISA reader (TECAN RainBow). Experiments were carried out in triplicate and the degrees of RBC haemolysis were calculated accordingly (Warawa et al., 1999).

Immunofluorescence staining.
HeLa cells were infected with bacteria as described above except for cultivation on glass coverslips in six-well plates. After infection, the cells were washed gently with cold PBS twice and fixed with 4 % (v/v) paraformaldehyde for 20 min. The cells were then permeated with 0·5 % Triton X-100 in PBS and blocked with 3 % BSA in PBS (Lin et al., 1999). To stain the bacteria, rabbit anti-O157 antibody (Difco) was used and the bound primary antibodies were in turn detected by FITC-labelled anti-rabbit immunoglobulin (Jackson Laboratory). To stain the actin filaments, cells were treated with tetramethylrhodamine isothiocyanate (TRITC)-labelled phalloidin at 5 µg ml–1 (Jackson Laboratory) in PBS containing 1 % (w/v) BSA. The stained cells on coverslips were examined by using a fluorescence microscope (Olympus BX51) and pictures were taken with a CCD camera (Photometrics CoolSNAP).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The segment of EspB critical for translocation
EspB is a type III-secreted protein that translocates into the membrane and cytoplasm of bacteria-infected cells (Wolff et al., 1998). Our previous study demonstrated that the C-terminal region (residues 283–312) of EspB is completely dispensable for its secretion (Chiu et al., 2003). To study whether the translocation capacities of various constructs of EspB are in parallel with that of secretion, we examined the membrane and cytosolic fractions of the infected host cells for the presence of the molecules. As the N-terminal 118 aa segment of EspB is essential for secretion (Chiu et al., 2003), it was preserved in almost all of the constructs (Fig. 1a) except for pB{Delta}TM, in which residues 99–118, predicted to form a transmembrane domain (http://www.ch.embnet.org/software/TMPRED_form.html), were deleted. Previously, a coiled-coil domain of EspD was found to be functionally important for EPEC infectivity (Daniell et al., 2001). Thus, a construct named pB{Delta}CC was included in the tests and, within this construct, residues 187–216 of EspB, predicted to form a coiled-coil structure, were deleted. All of the constructs summarized in Fig. 1(a) carry no tag at either end of the molecules. They have been examined for proper expression in bacterial lysates (Fig. 1b) and demonstrated to retain secretion capacity (Fig. 1c).



View larger version (45K):
[in this window]
[in a new window]
 
Fig. 1. Secretion competence of EspB constructs. (a) Schematic illustration of the EspB truncation constructs. In EspB, a hydrophobic region predicted to form a transmembrane structure is hatched, whereas a region predicted to form a coiled-coil structure is labelled by a filled box. (b) EspB constructs present in bacterial lysates. Expression plasmids, based on pQE60, were transformed separately into E. coli 43888 {Delta}B, and the transformants were cultured in M9 medium, harvested and lysed directly in SDS sample buffer. EspB in the bacterial lysates was detected with Western blotting using anti-EspB antibody. (c) EspB secreted in the medium. Proteins in the medium spent for sample preparations in (b) were concentrated by trichloroacetic acid (Chiu et al., 2003) and analysed with Western blotting as in (b). Open circles indicate EspB molecules with expected sizes from the expression plasmid.

 
To address the translocation property of mutated EspB, bacteria must interact well with host cells so that a failure of translocation could not simply arise from a lack of bacterium–host contact. Therefore, the expression plasmids were transformed into wild-type E. coli 43888. By doing so, detection of the authentic EspB in the host-cell lysates affirms a proper bacterium–cell interaction. Experimentally, the infected cells with the attached bacteria were prepared as a total lysate (Fig. 2a) in SDS gel-sample buffer. In a parallel investigation, bacteria were removed together with unbroken cells by low-speed centrifugation after cells were scraped off. The remains were then fractionated into cell-membrane (Fig. 2b) and cytosolic (Fig. 2c) fractions. These preparations were analysed by Western blotting using anti-EspB antibodies. The authentic EspB encoded by the chromosomal espB was detected in all samples (Fig. 2a) and the truncated EspB molecules, as seen with sizes smaller than that of the authentic molecule, were also detected.



View larger version (66K):
[in this window]
[in a new window]
 
Fig. 2. Translocation capacity of EspB constructs. Expression plasmids shown in Fig. 1(a) were transformed into E. coli 43888, and the corresponding bacteria were then used to infect HeLa cells. The infected HeLa cells were prepared as (a) a total lysate (TL) derived from both bacteria and HeLa cells, or fractionated into (b) host-cell membrane (CM) fraction and (c) cytosolic (CS) fraction. EspB molecules present in different samples were detected by Western blotting using anti-EspB antibody. Open circles indicate the plasmid-expressed EspB, whilst arrows label the authentic EspB encoded by the bacterial chromosome; open boxes highlight the expected migration positions of products if translocated. Note: the EspB construct derived from pB{Delta}TM was barely separable from the authentic EspB in lane 9. (d) Control Western blotting analysis of the cell content fractionated above. Protein components marked by antibodies are: OmpA, a bacterial outer-membrane protein; syndecan, a host-cell membrane protein; actin, a host-cell cytosolic protein.

 
The infected cells were fractionated into a membrane fraction and a cytosolic part after washing off the bacteria attached. When the membrane fraction of the infected cells was examined, all but pB1–190 and pB{Delta}118–190 gave products that could be detected by anti-EspB (Fig. 2b, lanes 2–9). As neither bacterial outer-membrane protein A (OmpA) nor the cytosolic molecule actin was detected in the sampled cellular-membrane fraction (Fig. 2d), the possibility of bacterial or cytosolic contamination in this fraction was excluded. Thus, a positive detection of EspB on the membrane fractions represented a proper translocation of EspB. The same set of EspB constructs seen in the membrane fraction was also detected in the cytosolic fraction (Fig. 2c), a result consistent with the translocation results observed with the membrane fraction.

The products derived from pB1–190 and pB{Delta}118–190 were not found in the membrane fractions of the infected cells; nor were they seen in the cytosol (lanes 5 and 8, Fig. 2b, c). The failure to detect these proteins could not be attributed to a low level of expression, as pB1–250, expressed equally well as pB1–190 and poorer than pB{Delta}118–190, was translocated successfully. These facts indicate that the loss of translocation ability with these constructs was because regions critical for translocation had been deleted or altered. Intriguingly, these two constructs are secreted and have been demonstrated repeatedly in the spent medium (Chiu et al., 2003; Fig. 1c). Thus, at least for EspB, secretion does not warrant successful translocation and, taken together with the observation that the construct encoded by pB1–220 was translocated (lane 4, Fig. 2b and c), the EspB region spanning residues 118–220 must be critical for EspB translocation. Additional products with sizes smaller than expected (Fig. 2a) were seen with some of the total lysates. They were present in a relatively small amount and may represent minor degradation products.

Interaction of EspB with EspA
EspB has been reported to interact with EspA when secreted into the medium (Hartland et al., 2000) and deletion of espA in EPEC has been found to abolish the translocation of EspB (Knutton et al., 1998). For these reasons, we investigated whether EspB defective in translocation also fails to interact with EspA. Proteins from pB1–190 and pB{Delta}118–190 were tested, along with three translocatable controls, in an affinity column-retention assay (Fig. 3). In the assay, EspA tagged with a C-terminal Hisx6 epitope was retained in a nickel-ion column and tested for the co-eluted EspB molecules. The results (Fig. 3b) indicated that the full-length EspB, encoded by pB1–312, and those derived from pB1–282 and pB{Delta}CC were co-eluted from the column with EspA (Fig. 3c). In contrast, the products resulting from pB1–190 and pB{Delta}118–190 were not eluted, a fact suggesting that constructs that do not translocate have also lost the ability to interact with EspA.



View larger version (69K):
[in this window]
[in a new window]
 
Fig. 3. Interaction of truncated EspB with EspA. (a) Analysis of the EspB constructs contained in the bacterial lysates. (b) Analysis of EspB that interacts with EspA–Hisx6 bound on an Ni2+–NTA column. After washing the column with TBS containing 150 mM imidazole, the proteins retained were eluted with the same buffer containing 500 mM imidazole, and EspB molecules in the eluants were detected by Western blotting with anti-EspB antibodies. (c) The EspA–Hisx6 in the same eluants as in (b) was analysed with anti-EspA antibodies. Arrows indicate the specific products detected.

 
Region of EspB important for adherence of EHEC to host cells
In EPEC, deletion of espB affects the adherence of bacteria to the host cell (Donnenberg et al., 1993), but not the length of EspA filaments or the translocation of EspD (Delahay et al., 1999; Hartland et al., 2000; Knutton et al., 1998; Wachter et al., 1999). Therefore, we used E. coli 43888 {Delta}B to investigate which EspB constructs could restore adherence of bacteria to the host cells.

EspB-expressing plasmids were transformed into E. coli 43888 {Delta}B and recovery of the lost bacterial adherence was examined. Fig. 4 shows that E. coli 43888 {Delta}B, when complemented with the full-length EspB expressed from pB1–312, had restored bacterial adherence to a level similar to that of the wild-type strain transformed with a vector control (i.e. pQE60). Constructs with the C-terminal region of EspB deleted, such as those from pB1–282, pB1–250, pB1–220 and pB1–190, did not recover bacterial adherence. Furthermore, the construct from pB{Delta}191–253, with a deletion of residues 191–253 of EspB but retaining the C-terminal 59 aa, did not restore the adherence of E. coli 43888 {Delta}B (Fig. 4). In contrast, truncation at residues 118–190 (in pB{Delta}118–190), which abolished both translocation and interaction with EspA, apparently did not damage EspB's bacterial-adherence property. Deletion of the hypothetical coiled-coil domain (residues 187–216) of EspB or the possible transmembrane domain (residues 99–118) alone also did not affect bacterial adherence (Fig. 4, pB{Delta}CC and pB{Delta}TM). Therefore, the region of EspB in residues 216–312 that locates C-terminally to the hypothetical coiled-coil domain of EspB (see summary in Fig. 7) must contribute to bacterial adherence, and this region of EspB obviously differs from that required for the functions of translocation and EspA interaction seen above.



View larger version (29K):
[in this window]
[in a new window]
 
Fig. 4. Ability of various EspB constructs to restore the adherence activity lost in E. coli 43888 {Delta}B, an espB-deleted strain of EHEC. HeLa cells were infected with plasmid-transformed bacteria for 6 h. Then, the cells were washed, harvested in PBS and lysed with PBS containing saponin. The cell lysates were diluted serially and plated on LB plates supplemented with ampicillin (100 µg ml–1). After overnight incubation, c.f.u. were scored and presented as a percentage of those observed with the wild-type EHEC (Wt). Note: E. coli 43888 {Delta}T ({Delta}Tir) was included as a control, showing that this adherence assay was dependent on EspB rather than on Tir. Experiments were performed in triplicate. Significant differences compared with the wild-type strain 43888 after using Student's t-test (P<0·01) are marked with an asterisk.

 


View larger version (32K):
[in this window]
[in a new window]
 
Fig. 7. Summary of the structure–function relationship of EspB. (a) Summarized results observed from Figs 1–6. Abbreviations used for the phenotypes/properties described in the text are: Sec, secretion (Chiu et al., 2003); Translo., translocation of EspB; EspA, interaction of EspB with EspA; Adhe., adherence of bacteria to host cells; Haemo., RBC haemolysis; Actin, focused actin accumulation. Dotted lines are used simply for easy association of the relative relationship. NT, Not tested. (b) Schematic illustration of the functional-analysis results in EspB. TM and CC label the predicted regions forming transmembrane and coiled-coil structures, respectively. It should be noted that secretion is the principal requirement for the other functions to be achieved.

 
Tir is a type III effector protein, translocated into the host cells to function as a receptor for the bacterial membrane protein intimin. Tir–intimin interaction has been reported to be important for the intimate attachment of bacteria to the host cells (Kenny et al., 1997). However, deletion of espB also blocks the translocation of Tir (Kenny & Finlay, 1995) and, thus, it is important to clarify the role that Tir plays in this experiment. For this purpose, an isogenic strain with Tir deleted was included in the assays. Results in Fig. 4 showed that E. coli 43888 {Delta}T did not lose this adherence activity and, apparently, EspB was the major determinant monitored in this system. Therefore, this assay (Gansheroff et al., 1999) may represent a measurement of bacterium–cell adherence before intimin binds to the translocated Tir.

Regions of EspB critical for causing RBC haemolysis
EspB, together with EspD, is thought to form pores at the membranes of the infected cells (Ide et al., 2001; Kenny et al., 1997; Warawa et al., 1999). In EPEC, this activity has been assayed by using haemolysis of human RBCs, and deletion of either espA, espD or espB attenuates the haemolytic activity of EPEC (Ide et al., 2001; Shaw et al., 2001; Warawa et al., 1999). To examine the domain of EHEC EspB possibly involved in pore formation, RBCs were incubated with the bacteria used in Fig. 4, and haemoglobin release was measured after 6 h incubation. Fig. 5 shows that E. coli 43888 {Delta}B transformed with pQE60 lost the haemolytic activity found with the wild-type EHEC. The constructs of EspB derived from pB1–282, pB1–250, pB1–220, pB1–190 and pB{Delta}191–253 failed to show the complementation activity seen with the full-length EspB. Therefore, truncated EspB that failed in the adherence assay also failed to show haemolysis in RBCs (compare Figs 4 and 5). Thus, the results with the adherence and haemolysis assays seemed to be in parallel, except for the results with pB{Delta}TM, in which deletion of residues 99–118 apparently did not affect adherence, but did abolish RBC-haemolysis activity. This observation suggests that residues 99–118 of EspB may be involved in pore formation, and this notion is consistent with a transmembrane domain predicted for these residues. Thus, it is proposed that EspB may mediate adherence through a structure that involves the C-terminal region and, after adherence, the transmembrane domain of EspB may further participate in pore formation. In a control with E. coli 43888 {Delta}T, transformation with pQE60 decreased the haemolysis only to a small degree (Fig. 5), a result similar to that reported previously (Warawa et al., 1999).



View larger version (25K):
[in this window]
[in a new window]
 
Fig. 5. Haemolytic activity of E. coli 43888 {Delta}B after complementation with various EspB constructs. Human RBCs in 12-well plates were incubated in DMEM without phenol red. Overnight-grown bacteria were added to the RBC culture in triplicate and incubated for 6 h. The whole culture was harvested and centrifuged, and OD543 of the supernatant was measured with an ELISA reader. The results are shown as a percentage of the readings obtained with the wild-type 43888 strain; asterisks indicate significant differences at P<0·01 compared with the wild-type strain 43888.

 
Regions of EspB critical for actin-accumulation activity
Deletion of espB may abolish the membrane-pore formation and could theoretically affect translocation of the type III effectors (Foubister et al., 1994), including Tir, which is necessary for actin accumulation underneath the bacterial-attachment site (Kenny et al., 1997). To test this possibility, cells infected with the plasmid-transformed E. coli 43888 {Delta}B were stained for condensed actin by using TRITC-labelled phalloidin (Fig. 6). Strain 43888 {Delta}B transformed with a vector control (i.e. pQE60) showed no condensed actin accumulation beneath the bacteria, whereas the similarly treated wild-type bacteria clearly showed this phenotype (Fig. 6a and b). As expected, the actin-accumulation property was restored when E. coli 43888 {Delta}B was transformed with pB1–312 (Fig. 6c). In contrast, when the same bacteria were transformed separately with pB1–282, pB1–250, pB1–220 and pB{Delta}191–253, which failed to make strains adhere to the host cells, it is clear from Fig. 6(d–g) that there is no actin condensation beneath the sites where bacteria can be seen. Furthermore, the transformant derived from pB{Delta}TM, which adhered well to the cells but failed to produce lysis of RBCs, also yielded a negative result for actin condensation (Fig. 6j). On the other hand, strain 43888 {Delta}B transformed with pB{Delta}CC or pB{Delta}118–190, which showed haemolytic activity against RBCs, retained actin-accumulation activity (Fig. 6h and i). Therefore, to assist in the translocation of Tir, as reflected by actin accumulation, the EspB constructs of bacteria must be able to adhere to the host membrane and to form the pores.



View larger version (109K):
[in this window]
[in a new window]
 
Fig. 6. Accumulation of actin beneath E. coli 43888 {Delta}B after complementation with EspB constructs. HeLa cells were plated on glass coverslips and cultured in DMEM containing 10 % (v/v) fetal bovine serum. After infection for 6 h, the cells were washed and fixed with 4 % (w/v) paraformaldehyde. The fixed cells were then permeated with 0·5 % (v/v) Triton X-100. Bacteria adhering to the cells were detected with rabbit anti-O157 antibodies followed by staining with FITC-conjugated goat anti-rabbit antibodies. To stain the condensed actin, the cells were treated with TRITC-labelled phalloidin. Samples were observed by using a fluorescence microscope and the images were taken with a CCD camera; the plasmid carried by infected bacteria in each panel is shown in parentheses.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The LEE comprises 41 ORFs that are involved in the A/E lesion. Systematic deletion of individual genes has recently been performed with C. rodentium (Deng et al., 2004) and the results indicated that bacterial virulence was reduced to a certain degree when any of these genes was deleted. The implication, then, is the existence of a gene-regulation network and the presence of numerous interplays between the gene products. Deletion of EspB completely abolishes bacterial adhesion, actin rearrangement, translocation of Tir and, thus, bacterial virulence. Therefore, EspB may play multiple roles during the process of host–bacterium interaction. To unveil the functional complexity built into the structure of the EspB molecule, we have dissected the molecule and generated different constructs of EspB. By complementation of an EspB-deleted mutant, the general structure–function relationship of EspB was established; the results are summarized in Fig. 7.

As shown previously, the N-terminal 118 aa region of EspB is critical, but not sufficient, for EspB secretion (Chiu et al., 2003). To serve as a translocator, EspB must retain secretion competence. Therefore, any construct that lacks the N-terminal 118 aa is not secreted at all and could not possibly function for other purposes.

Translocation of EspB into the host cells apparently requires the same region of EspB that is also critical to interact with EspA (Fig. 7a). Thus, these two functions are currently assigned to the same domain. Interaction of EspA and EspB has been reported (Hartland et al., 2000) and here we have further demonstrated that residues 118–190 in the middle region of EspB are needed for this interaction. Less clear is the function of EspB residues 187–216, which are predicted to form a coiled-coil structure. In comparison of pB1–190 with pB1–220 (Fig. 7a), residues 191–216 seem to be crucial for EspB transloction. However, in pB{Delta}191–253, the deletion includes residues 191–216 of EspB, but does not abolish EspB transloction activity. As many factors, such as conformational changes, could account for this inconsistency, we currently prefer a conservative way and assign no function for this region (Fig. 7b).

Adherence of bacteria to HeLa cells apparently involves EspB, particularly the C-terminal region, and we have narrowed this down to the region spanning from residue 216 to the C terminus (Fig. 7a). This region is also needed for pore formation on the membrane, as measured by the RBC-haemolysis experiment. The latter activity of causing RBC lysis appeared to be associated with an extra domain in residues 99–118 that has a high degree of hydrophobicity, as evidenced by the data from pB{Delta}TM. Intriguingly, both the C-terminal region and the hydrophobic residues important for RBC haemolysis matched with the regions critical for the formation of the condensed actin structure underneath the sites where bacteria are attached (Fig. 7a). Actin accumulation has been correlated with successful translocation of Tir, a function that results from the correct formation of membrane pores. Therefore, residues 99–118 and 216–312 in EspB must play an essential role to function effectively as a translocator.

The functional domains of EspB described above seem to be discernible, and they are grouped accordingly. It is worth noting that, without the EspA-interacting region, the truncated EspB molecule (from pB{Delta}118–190) remains active in helping bacterial adherence (Fig. 7a). Similarly, EspB losing the pore-forming activity (from pB{Delta}TM) did not lose its bacterial-adherence activity. On the other hand, EspB mutants that lost the activity of assisting bacterial adherence all gave a negative result in the pore-formation assay, an observation suggesting that the pore-formation activity of EspB relies on its active function of adherence. It is then proposed that EspB's assistance in bacterial adherence is an action prior to the formation of membrane pores. On the other hand, bacterial adherence and EspB translocation/EspB–EspA interaction seem to be independent events, as no clear association could be made (Fig. 7a).

In our experimental results shown in Fig. 4, we measured the bacteria attached to the cells after 6 h incubation and calculated the relative activity after comparison with the parental strain. To ensure that the measured values did not result from the rate differences of bacterial growth, parallel experiments were performed with individual bacteria, to follow their growth curves. The fact that no apparent differences were found thus excludes this possibility. However, adherence activity normally monitors the ratio of bacteria attached to the number of bacteria inoculated. Our values of the relative adhering activity of different strains should then reflect more than a simple association at the early stage. Away from a direct input of EspB, an indirect effect of EspB on other adherence-assisting molecules, such as EspA (Cleary et al., 2004) and EspD, may have been measured collectively.


   ACKNOWLEDGEMENTS
 
We thank Ralph Kirby for critical reading of this manuscript. This research was supported in parts by grant 89-B-FA22-2-4 (Program for Promoting Academic Excellence of Universities) from the Department of Education and NSC 92-2320-B-010-048 from the National Science Council, Taiwan, R.O.C.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Chiu, H.-J., Lin, W.-S. & Syu, W.-J. (2003). Type III secretion of EspB in enterohemorrhagic Escherichia coli O157 : H7. Arch Microbiol 180, 218–226.[CrossRef][Medline]

Chuang, C.-H., Hsu, S.-C., Hsu, C.-L., Hsu, T.-C. & Syu, W.-J. (2001). Construction of a tagging system for subcellular localization of proteins encoded by open reading frames. J Biomed Sci 8, 170–175.[CrossRef][Medline]

Cleary, J., Lai, L.-C., Shaw, R. K., Straatman-Iwanowska, A., Donnenberg, M. S., Frankel, G. & Knutton, S. (2004). Enteropathogenic Escherichia coli (EPEC) adhesion to intestinal epithelial cells: role of bundle-forming pili (BFP), EspA filaments and intimin. Microbiology 150, 527–538.[CrossRef][Medline]

Crane, J. K., McNamara, B. P. & Donnenberg, M. S. (2001). Role of EspF in host cell death induced by enteropathogenic Escherichia coli. Cell Microbiol 3, 197–211.[CrossRef][Medline]

Daniell, S. J., Delahay, R. M., Shaw, R. K., Hartland, E. L., Pallen, M. J., Booy, F., Ebel, F., Knutton, S. & Frankel, G. (2001). Coiled-coil domain of enteropathogenic Escherichia coli type III secreted protein EspD is involved in EspA filament-mediated cell attachment and hemolysis. Infect Immun 69, 4055–4064.[Abstract/Free Full Text]

Delahay, R. M., Knutton, S., Shaw, R. K., Hartland, E. L., Pallen, M. J. & Frankel, G. (1999). The coiled-coil domain of EspA is essential for the assembly of the type III secretion translocon on the surface of enteropathogenic Escherichia coli. J Biol Chem 274, 35969–35974.[Abstract/Free Full Text]

Deng, W., Puente, J. L., Gruenheid, S. & 12 other authors (2004). Dissecting virulence: systematic and functional analyses of a pathogenicity island. Proc Natl Acad Sci U S A 101, 3597–3602.[Abstract/Free Full Text]

Donnenberg, M. S., Yu, J. & Kaper, J. B. (1993). A second chromosomal gene necessary for intimate attachment of enteropathogenic Escherichia coli to epithelial cells. J Bacteriol 175, 4670–4680.[Abstract]

Foubister, V., Rosenshine, I., Donnenberg, M. S. & Finlay, B. B. (1994). The eaeB gene of enteropathogenic Escherichia coli is necessary for signal transduction in epithelial cells. Infect Immun 62, 3038–3040.[Abstract]

Gansheroff, L. J., Wachtel, M. R. & O'Brien, A. D. (1999). Decreased adherence of enterohemorrhagic Escherichia coli to HEp-2 cells in the presence of antibodies that recognize the C-terminal region of intimin. Infect Immun 67, 6409–6417.[Abstract/Free Full Text]

Gauthier, A., de Grado, M. & Finlay, B. B. (2000). Mechanical fractionation reveals structural requirements for enteropathogenic Escherichia coli Tir insertion into host membranes. Infect Immun 68, 4344–4348.[Abstract/Free Full Text]

Hartland, E. L., Daniell, S. J., Delahay, R. M., Neves, B. C., Wallis, T., Shaw, R. K., Hale, C., Knutton, S. & Frankel, G. (2000). The type III protein translocation system of enteropathogenic Escherichia coli involves EspA–EspB protein interactions. Mol Microbiol 35, 1483–1492.[CrossRef][Medline]

Hsu, S.-C., Lin, H.-P., Wu, J.-C., Ko, K.-L., Sheen, I.-J., Yan, B.-S., Chou, C.-K. & Syu, W.-J. (2000). Characterization of a strain-specific monoclonal antibody to hepatitis delta virus antigen. J Virol Methods 87, 53–62.[CrossRef][Medline]

Ide, T., Laarmann, S., Greune, L., Schillers, H., Oberleithner, H. & Schmidt, M. A. (2001). Characterization of translocation pores inserted into plasma membranes by type III-secreted Esp proteins of enteropathogenic Escherichia coli. Cell Microbiol 3, 669–679.[CrossRef][Medline]

Kenny, B. & Finlay, B. B. (1995). Protein secretion by enteropathogenic Escherichia coli is essential for transducing signals to epithelial cells. Proc Natl Acad Sci U S A 92, 7991–7995.[Abstract/Free Full Text]

Kenny, B. & Jepson, M. (2000). Targeting of an enteropathogenic Escherichia coli (EPEC) effector protein to host mitochondria. Cell Microbiol 2, 579–590.[CrossRef][Medline]

Kenny, B., Lai, L.-C., Finlay, B. B. & Donnenberg, M. S. (1996). EspA, a protein secreted by enteropathogenic Escherichia coli, is required to induce signals in epithelial cells. Mol Microbiol 20, 313–323.[CrossRef][Medline]

Kenny, B., DeVinney, R., Stein, M., Reinscheid, D. J., Frey, E. A. & Finlay, B. B. (1997). Enteropathogenic E. coli (EPEC) transfers its receptor for intimate adherence into mammalian cells. Cell 91, 511–520.[CrossRef][Medline]

Knappstein, S., Ide, T., Schmidt, M. A. & Heusipp, G. (2004). {alpha}1-Antitrypsin binds to and interferes with functionality of EspB from atypical and typical enteropathogenic Escherichia coli strains. Infect Immun 72, 4344–4350.[Abstract/Free Full Text]

Knutton, S., Rosenshine, I., Pallen, M. J., Nisan, I., Neves, B. C., Bain, C., Wolff, C., Dougan, G. & Frankel, G. (1998). A novel EspA-associated surface organelle of enteropathogenic Escherichia coli involved in protein translocation into epithelial cells. EMBO J 17, 2166–2176.[Abstract/Free Full Text]

Kodama, T., Akeda, Y., Kono, G., Takahashi, A., Imura, K., Iida, T. & Honda, T. (2002). The EspB protein of enterohaemorrhagic Escherichia coli interacts directly with {alpha}-catenin. Cell Microbiol 4, 213–222.[CrossRef][Medline]

Lai, L.-C., Wainwright, L. A., Stone, K. D. & Donnenberg, M. S. (1997). A third secreted protein that is encoded by the enteropathogenic Escherichia coli pathogenicity island is required for transduction of signals and for attaching and effacing activities in host cells. Infect Immun 65, 2211–2217.[Abstract]

Lin, H.-P., Hsu, S.-C., Wu, J.-C., Sheen, I.-J., Yan, B.-S. & Syu, W.-J. (1999). Localization of isoprenylated antigen of hepatitis delta virus by anti-farnesyl antibodies. J Gen Virol 80, 91–96.[Abstract]

McDaniel, T. K. & Kaper, J. B. (1997). A cloned pathogenicity island from enteropathogenic Escherichia coli confers the attaching and effacing phenotype on E. coli K-12. Mol Microbiol 23, 399–407.[CrossRef][Medline]

Nataro, J. P. & Kaper, J. B. (1998). Diarrheagenic Escherichia coli. Clin Microbiol Rev 11, 142–201.[Abstract/Free Full Text]

Nougayrède, J.-P., Fernandes, P. J. & Donnenberg, M. S. (2003). Adhesion of enteropathogenic Escherichia coli to host cells. Cell Microbiol 5, 359–372.[CrossRef][Medline]

Roe, A. J., Hoey, D. E. E. & Gally, D. L. (2003). Regulation, secretion and activity of type III-secreted proteins of enterohaemorrhagic Escherichia coli O157. Biochem Soc Trans 31, 98–103.[Medline]

Sambrook, J. & Russell, D. (2001). Molecular Cloning: a Laboratory Manual, 3rd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Schauer, D. B. & Falkow, S. (1993). Attaching and effacing locus of a Citrobacter freundii biotype that causes transmissible murine colonic hyperplasia. Infect Immun 61, 2486–2492.[Abstract]

Shaw, R. K., Daniell, S., Ebel, F., Frankel, G. & Knutton, S. (2001). EspA filament-mediated protein translocation into red blood cells. Cell Microbiol 3, 213–222.[CrossRef][Medline]

Vandeyar, M. A., Weiner, M. P., Hutton, C. J. & Batt, C. A. (1988). A simple and rapid method for the selection of oligodeoxynucleotide-directed mutants. Gene 65, 129–133.[CrossRef][Medline]

Wachter, C., Beinke, C., Mattes, M. & Schmidt, M. A. (1999). Insertion of EspD into epithelial target cell membranes by infecting enteropathogenic Escherichia coli. Mol Microbiol 31, 1695–1707.[CrossRef][Medline]

Warawa, J., Finlay, B. B. & Kenny, B. (1999). Type III secretion-dependent hemolytic activity of enteropathogenic Escherichia coli. Infect Immun 67, 5538–5540.[Abstract/Free Full Text]

Weiner, M. P., Felts, K. A., Simcox, T. G. & Braman, J. C. (1993). A method for the site-directed mono- and multi-mutagenesis of double-stranded DNA. Gene 126, 35–41.[CrossRef][Medline]

Wolff, C., Nisan, I., Hanski, E., Frankel, G. & Rosenshine, I. (1998). Protein translocation into host epithelial cells by infecting enteropathogenic Escherichia coli. Mol Microbiol 28, 143–155.[CrossRef][Medline]

Yu, S.-L., Ko, K.-L., Chen, C.-S., Chang, Y.-C. & Syu, W.-J. (2000). Characterization of the distal tail fiber locus and determination of the receptor for phage AR1, which specifically infects Escherichia coli O157 : H7. J Bacteriol 182, 5962–5968.[Abstract/Free Full Text]

Received 12 April 2005; revised 20 June 2005; accepted 21 July 2005.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Chiu, H.-J.
Articles by Syu, W.-J.
Articles citing this Article
PubMed
PubMed Citation
Articles by Chiu, H.-J.
Articles by Syu, W.-J.
Agricola
Articles by Chiu, H.-J.
Articles by Syu, W.-J.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2005 Society for General Microbiology.