Protein-tyrosine Phosphatase α, RPTPα, Is aHelicobacter pyloriVacA Receptor

Helicobacter pylori vacuolating cytotoxin, VacA, induces vacuolation, mitochondrial damage, cytochrome c release, and apoptosis of gastric epithelial cells. To detect gastric proteins that serve as VacA receptors, we used VacA co-immunoprecipitation techniques following biotinylation of the cell surface and identified p250, a receptor-like protein-tyrosine phosphatase β (RPTPβ) as a VacA-binding protein (Yahiro, K., Niidome, T., Kimura, M., Hatakeyama, T., Aoyagi, H., Kurazono, H., Imagawa, K., Wada, A., Moss, J., and Hirayama, T. (1999) J. Biol. Chem. 274, 36693–36699). VacA causes vacuolation of G401 cells, a human kidney tumor cell line, although they do not express RPTPβ. By co-immunoprecipitation with VacA, we identified p140 as a potential receptor in those cells. p140 purified by chromatography on a peanut agglutinin affinity matrix contained internal amino acid sequences of RGEENTDYVNASFIDGYRQK and AEGILDVFQTVK, which are identical to those in RPTPα. The peptide mass fingerprinting of p140 by time of flight-MS analysis also supported this identification. Treatment of G401 cells with RPTPα-morpholino antisense oligonucleotide before exposure to toxin inhibited vacuolation. These data suggest that RPTPα acts as a receptor for VacA in G401 cells. Thus, two receptor tyrosine phosphatases, RPTPα and RPTPβ, serve as VacA receptors.

Persistent infection with Helicobacter pylori causes chronic active gastritis, which predisposes the mucosa to peptic ulceration, and is believed to participate in the pathogenesis of gastric carcinoma and primary gastric lymphoma (i.e. mucosaassociated lymphoid tissue, or MALT type) (1,2). Although H. pylori is a noninvasive bacterium that survives in the stomach mucosa, pathogenic strains of H. pylori produce and secrete a potent vacuolating cytotoxin, VacA. Both epidemiological study (3) and animal experiments (4 -7) have demonstrated that VacA is a major virulence factor and is involved in the pathogenesis of inflammation in H. pylori-induced gastritis and ul-ceration. Oral administration of VacA to mice caused acute inflammation of the gastric mucosa with accumulation of mast cells, the activation of which by VacA resulted in production of proinflammatory cytokines (8). Intense infiltration of granulocytes and lymphocytes was observed following mast cell activation, consistent with the hypothesis that mast cells contribute to the gastric inflammation in H. pylori-infected peptic gastritis and ulceration (9).
Purified VacA has a molecular mass of 87-95 kDa under denaturating conditions, whereas the native toxin is an oligomeric complex of about 1000 kDa (10). VacA proved to be capable of directly inducing progressive vacuolation (11), mitochondrial damage (12), cytochrome c release (13), and apoptosis of epithelial cells (14). Although the detailed mechanism of its toxic activity is still unknown, bafilomycin A1, a specific inhibitor of the vacuolar ATPase proton pump, can inhibit VacAinduced vacuolation. It did not, however, block mitochondrial damage by VacA (12). In addition, a mutant VacA lacking vacuolating activity induced an apoptosis of AGS cells (14). These results led to the current model in which the mitochondrial damage occurs via a pathway different from vacuolation induced by VacA, possibly through the formation of pores or selective anion channels in endosomal membranes (15,16) and alteration of endolysosomal function (17,18).
VacA binding to specific high-affinity cell-surface receptors was shown by using indirect immunofluorescence and flow cytometry; high-affinity toxin binding was necessary for cell intoxication (19,20). VacA can interact with target cells by binding to receptor-like protein-tyrosine phosphatase ␤ (RPTP␤) 1 (21,22). In addition to RPTP␤, other proteins of lesser abundance on the surface of AZ-521 cells were also immunoprecipitated with anti-VacA antibody (23), consistent with the presence of alternative receptors. G401 cells, although responsive to VacA, do not express RPTP␤. Here we report purification of another VacA receptor, p140, and its identification as the receptor-like tyrosine phosphatase ␣, RPTP␣, a ubiquitously expressed protein (24). Thus, RPTP␣ and RPTP␤, two receptor-like protein-tyrosine phosphatases that differ widely in structure, can both serve as VacA receptors.
VacA Preparation-The toxin-producing H. pylori strain ATCC49503 was used as the source of VacA for purification according to a modification of our published procedure (23). In brief, after growth of H. pylori in Brucella broth containing 0.1% ␤-cyclodextran at 37°C for 3-4 days with vigorous shaking in a controlled microaerophilic atmosphere of 10% O 2 and 10% CO 2 , VacA was precipitated from culture supernatant with 50% saturated ammonium sulfate and purified by affinity column chromatography. Ammonium sulfate precipitates were dialyzed against RX buffer (10 mM KCl, 0.3 mM NaCl, 0.35 mM MgCl 2 , and 0.125 mM EGTA in 1 mM HEPES, pH 7.3) and applied to an anti-VacA-specific IgG antibody column (10 mg of IgG/2 mg of dried resin) equilibrated with RX buffer. After washing the column with RX buffer, VacA was eluted with 50 mM glycine-HCl buffer (pH 1.0), which was subsequently neutralized with 1 M Tris-HCl (pH 10). After gel filtration on Superose 6HR 10/30 equilibrated with TBS buffer (60 mM Tris-HCl buffer, pH 7.7, containing 0.1 M NaCl), purified VacA was concentrated and stored (200 g/ml).
RNase A Protection Assay-RNA quantification was performed using the Ribonuclease Protection Assay kit (RPAIII; Ambion) as previously described (22). Briefly, 20 g of total RNA from G401 cells, positive control cells such as AZ-521 cells, and HL60 cells treated with 20 nM phorbol 12-myristate 13-acetate for 48 h were used in the RNase A protection assay (22). Total RNA was hybridized overnight with RPTP␤ and glyceraldehyde-3-phosphate dehydrogenase probes in hybridization buffer at 42°C, followed by treatment with RNase, which was then inactivated. Protected fragments were separated in 5% sequencing gels, and quantified by Fuji film autoradiography. Sizes of protected fragments were 230 nucleotides for RPTP␤ and 184 nucleotides for glyceraldehyde-3-phosphate dehydrogenase.
Assay for Vacuolating Activity-Vacuolating activity was assessed using AZ-521 or G401 cells as previously described (21). Cells (1 ϫ 10 5 cells/well, 250 l) were grown in 24-well culture plates as monolayers for 24 h in a 5% CO 2 atmosphere at 37°C. VacA was added and cells were incubated for an additional 8 h at 37°C. To quantify the vacuolating activity, the uptake of neutral red into the vacuoles was determined as described previously (21).
Glycosidase Treatment-After biotinylation of proteins on the surface, G401 cells were lysed in 1 ml of Sol buffer for 15 min on ice. After centrifugation (20 min, 15,000 ϫ g), 10 l of each supernatant (1 g/ml) was added to a mixture of 34.5 l of 20 mM sodium phosphate buffer (pH 7.2), 5 l of 10% Nonidet P-40, and 0.5 l of N-glycosidase F (0.1 unit), neuraminidase (0.1 unit), heparitinase (0.5 unit), or chondroitinase ABC (0.5 unit), followed by incubation for 3 h, and immunoprecipitation with anti-VacA after addition of native VacA or a heat-inactivated VacA. Immunoprecipitated proteins were separated by SDS-PAGE in 5% gels and transferred to PVDF membranes.
Flow Cytometric Analysis-AZ-521 or G401 cells (1 ϫ 10 7 ) were harvested in TNE buffer, washed twice in PBS, and suspended in 1 ml of PBS containing 2% BSA. Samples (90 l) were treated with neuraminidase (0.1 unit) for the indicated times. After two washes with PBS containing 2% BSA, cells (1 ϫ 10 5 /90 l) were incubated at 4°C for 30 min with 10 l (1 g) of VacA or PBS as a control. Cells were washed twice with 500 l of PBS containing 2% BSA, followed by addition of anti-VacA antibody and incubation at 4°C for 30 min. After two washes with PBS containing 2% BSA, cells were suspended in 400 l of the same solution containing Fluoro Link TM Cy TM 2-labeled goat anti-rabbit IgG (1:400) (Amersham Biosciences) and incubated at 4°C for 30 min. After cells were washed twice with the same PBS buffer, samples of 10,000 cells were analyzed by flow cytometry (BD Pharmingen immunocytometry system) with excitation at 488 nm and emission at 530 nm.
Neuraminidase Treatment-AZ-521 or G401 cells (5 ϫ 10 5 ) were harvested in TNE buffer, washed twice in PBS, suspended in 90 l of PBS containing 2% BSA, followed by addition of 10 l (0.1 unit) of neuraminidase, and incubation for 1 h at 37°C. Samples were then washed twice with PBS containing 2% BSA. Cells were suspended in 300 l of PBS containing 2% BSA and incubated with 120 nM VacA at 4°C for 30 min, before washing twice with 500 l of PBS containing 2% BSA and addition of 300 l of DMEM containing 10% FCS. Samples (100 l) were incubated in 96-well plates for 3 h at 37°C and vacuolating activity was quantified by NRU assay.
Characterization of Carbohydrate Moieties of p140 -Specific binding of lectins was used to identify carbohydrate moieties of p140 from G401 cells, according to the manufacturer's specifications (Roche Diagnostics; DIG glycan differentiation kit) as previously described (21). The lectins selectively recognize terminal sugars; MAA and PNA recognize sialic acid terminally linked ␣(2-3) to galactose and galactose-␤-(1-3)-Nacetylgalactosamine, respectively. PVDF membranes with proteins that had been separated by SDS-PAGE after immunoprecipitation were incubated at 25°C for 1 h with lectin conjugated to the steroid hapten digoxigenin. After washing the membrane, glycoproteins were incubated with anti-digoxigenin Fab fragments conjugated with alkaline phosphatase, followed by detection with 4-nitro blue tetrazolium chloride/5-bromo-4-chloro-3-indolyl phosphate.
Purification and Identification of p140 -To examine binding of p140 to PNA-agarose (5 ml bed volume), biotinylated G401 cells (1 ϫ 10 8 , from ten 10-cm dishes) were washed twice with PBS, and suspended in 10 ml of Sol buffer without Triton X-100. After sonification, cells were centrifuged at 4°C for 15 min at 15,000 ϫ g, and the precipitated fractions were lysed in 10 ml of Sol buffer for 15 min on ice. After centrifugation (20 min, 15,000 ϫ g), the supernatant was filtered and the filtrate (9.5 ml) was applied to a PNA-agarose column (Seikagaku) (5 ml of bed volume). The column was washed with 25 ml of Sol buffer, and then Sol buffer containing 0.3 M D-galactose was used to elute the carbohydrate-containing proteins in 1-ml fractions. After SDS-PAGE in 5% gels, proteins in effluents were transferred to PVDF membranes, followed by incubation with avidin-HRP. Biotinylated p140 was detected using enhanced ECL.
To purify p140, G401 cells (1.6 ϫ 10 9 ) were harvested from 160 10-cm dishes with TNE buffer, washed twice with PBS, and suspended in 80 ml of Sol buffer without Triton X-100. After sonification, cell lysate was applied to a PNA-agarose column (5 ml of bed volume) using the same method described for purification of biotinylated p140. The column was washed with 25 ml of Sol buffer, followed by elution of carbohydratecontaining proteins with 0.3 M D-galactose in Sol buffer. Fractions (1 ml) in which p140 were detected by immunoprecipitation using VacA and anti-VacA antibody were pooled. Proteins were precipitated with 10% trichloroacetic acid, washed five times with ice-cold acetone, and dried. Samples (5 g) were solubilized by heating in 50 l of SDS-PAGE sample buffer and separated in 5% gels, which were stained with Coomassie Brilliant Blue. The stained p140 band was cut from the gels, digested with Achromobacter protease I (Wako Pure Chemical Institutes, Japan), and subjected to peptide sequence analysis (21) and time of flight mass spectrophotometric analyses (25).
Expression and Identification of RPTP␣-cDNA for human RPTP␣ in a pcDNA3.1 vector (purchased from Invitrogen) has a V5 epitope at the C terminus. COS-7 cells were plated 24 h before transfection using GenePorter transfection reagent (Gene Therapy System) following the manufacturer's protocol. After transfection, cells were incubated at 37°C for 48 h, harvested in TNE buffer, washed with ice-cold PBS buffer, lysed in Sol buffer, and centrifuged at 15,000 ϫ g for 20 min. The supernatant was incubated at 4°C for 1 h with 0.5 g of native or heat-inactivated VacA. The mixtures were then incubated at 4°C overnight with 1 l of an antibody raised against purified VacA. After addition of 25 l of protein A-Sepharose CL-4B (Amersham Bio-sciences), 0.1 g/ml in Sol buffer, and incubation at 4°C for 1 h, beads were washed three times with Sol buffer. Bound proteins were solubilized by heating in 5 l of SDS-PAGE sample buffer, separated by SDS-PAGE in 5% gels, and transferred to PVDF membranes (Millipore; Immobilon-P membrane). Blots were incubated with HRP-conjugated anti-V5 monoclonal antibody (Invitrogen), which recognizes the 14amino acid (GKPIPNPLLGLDST) sequence of V5 (26), and proteins were detected using the ECL system (Amersham Biosciences).
Transfection of G401 Cells with RPTP␣-morpholino Antisense Oligonucleotide-RPTP␣-morpholino antisense oligonucleotide (5Ј-GAA-CAAGAATGAACCAGGAATCCAT-3Ј) was designed and purchased from Gene Tools, LLC (Philomath, Oregon). G401 cells were seeded in 24-well culture plates (5 ϫ 10 4 cells in 1 ml of DMEM per well) and incubated overnight at 37°C. The gene delivery agent EPEI and 1.4 M RPTP␣-morpholino antisense oligonucleotide were incubated in 500 l of FCS-free DMEM for 20 min at room temperature before addition to G401 cells in 24-well plates, followed by incubation for 3 h at 37°C. After the culture medium was replaced with 500 l of DMEM containing 10% FCS, cells were incubated for 16 h. Vacuolation of cells caused by incubation with 120 nM VacA for 0, 2, and 4 h was quantified by NRU.
Internalization Assay-Internalization of VacA into the cells was quantified using VacA labeled with sulfo-NHS-SS-biotin (Pierce) according to the instructions provided by the manufacturer. Briefly, 25 l of 1 M sodium bicarbonate buffer (pH 8.5) was added to 500 l (100 g in PBS) of VacA followed by addition of 10 l (10 g) of sulfo-NHS-SSbiotin and incubated on ice for 2 h. After addition of 465 l of 1 M Tris-HCl (pH 7.5) and dialysis against PBS, VacA-SS-biotin was stored (100 g/ml). After G401 cells (

Identification of p140 on G401 Cells-The
Wilms' tumor cell line G401 was susceptible to VacA intoxication (Fig. 1). In Western blot analysis using anti-RPTP␤ antibody, RPTP␤ was, however, not detected in G401 cell lysates; the lack of RPTP␤ expression in G401 cells was confirmed by RNase protection assays ( Fig. 2A). Thus, G401 cells did not express RPTP␤ protein or mRNA as suggested by Qi et al. cell lysate with VacA, but not with inactivated VacA (Fig. 2B). These results were consistent with the hypothesis that not only RPTP␤, but also p140, can be a functional VacA receptor and led us to use G401 cells for purification of p140.
Characteristics of Sugar Moiety of p140 -PNA and MAA bound p140 (Fig. 3) through interaction with terminal sialic acid linked to ␣(2-3)galactose and galactose-␤-(1-3)-N-acetylgalactosamine, respectively. After incubation with glycosidases, biotinylated p140 on G401 cells was subjected to immunoprecipitation with anti-VacA antibodies after incubation with native VacA or heat-inactivated VacA (Fig. 4A). The samples immunoprecipitated via their association with native VacA before (Fig. 4a, lane 2) and after (Fig. 4a, lane 4) Nglycosidase F treatment contained 140-and 135-kDa proteins, respectively. No significant change in molecular size of p140 was observed after treatment with chondroitinase ABC, or heparitinase. As shown in lane 8 (Fig. 4b), neuraminidase treatment inhibited p140 binding to VacA, an effect confirmed by flow cytometric analysis of VacA binding to G401 cells treated with neuraminidase (Fig. 5). Treatment of G401 cells with neuraminidase before incubation with VacA decreased fluorescence intensity to control levels. In addition, two kinds of sialyllactosamine with terminal sialic acids, 3Ј-sialyllactosamine and 6Ј-sialyllactosamine, did not interfere with VacA binding to p140 when the biotinylated p140 on G401 cells was immunoprecipitated using native VacA and anti-VacA antibody (data not shown). These results indicate that not only the  A and B), detected by reaction with lectin conjugated to digoxigenin followed by incubation with antidigoxigenin Fab fragments conjugated to alkaline phosphatase, were visualized after reaction with 4-nitro blue tetrazolium chloride/5-bromo-4-chloro-3-indolyl phosphate. Panel C shows p140 immunoprecipitated with anti-VacA antibody, separated by SDS-PAGE in 5% gels, transferred to PVDF membrane, and incubated with HRP-conjugated streptavidin followed by ECL detection. Data are representative of three separate experiments. terminal sialic acid but also specific sugar structure and p140 sequence may be required for VacA binding.
Inhibition of VacA-induced Vacuolation in AZ-521 and G401 Cells by Neuraminidase Treatment-As treatment of AZ-521 and G401 cells with neuraminidase diminished the binding of VacA, we examined whether neuraminidase treatment also diminished VacA-induced vacuolation (Fig. 6). VacA-induced vacuolation of both AZ-521 and G401 cells treated with neuraminidase was found to be significantly less than that of untreated cells.
Purification and Sequence Analysis of p140 -By using PNA lectin affinity chromatography with solubilized G401 cells, we obtained partially purified p140 from fractions (E1-3) eluted with Sol buffer containing 0.3 M D-galactose (Fig. 7). Proteins in fractions, in which p140 was detected by immunoprecipitation (Fig. 7a), were precipitated with 10% trichloroacetic acid, solubilized with SDS-PAGE sample buffer, and separated in 5% gels, which were stained with Coomassie Brilliant Blue (Fig.   7b). Internal sequences obtained from two peptides separated by SDS-PAGE after Achromobacter protease I hydrolysis of p140 and excised from the gel were RGEENTDYVNASFIDGY-RQK and AEGILDVFQTVK, which are identical to positions 581-600 and 754 -765 in the RPTP␣ sequence reported by Kaplan et al. (29). In agreement with these data, peptide mass fingerprinting of p140 by time of flight-MS analysis after proteolysis (25) showed that all fragments were identical to those predicted for RPTP␣ (data not shown). Thus, these analytical data lead to the conclusion that p140 is RPTP␣.
VacA Binding to RPTP␣ in RPTP␣ Gene-transfected COS-7 Cells-Using COS-7 cells, which we transfected with V5-tagged RPTP␣ cDNA with relatively high efficiency, we examined the relationship of RPTP␣ expression to VacA binding; of note, COS-7 cells are sensitive to VacA and exhibit baseline VacA binding (23). COS-7 cells transfected with V5-tagged RPTP␣ cDNA exhibited greater VacA binding than did untransfected cells or cells transfected with vector only. Overexpression of V5-tagged RPTP␣ in the cell lysate or cell membrane was confirmed by visualization using Western blotting with HRPconjugated anti-V5 antibody and its binding activity was assayed by immunoprecipitation using VacA and anti-VacA antibody, followed by visualization of V5-tagged RPTP␣ in immunoprecipitates (Fig. 8). Anti-V5 antibody recognized two expressed V5-tagged RPTP␣ products of ϳ140 and 100 kDa in lysates of transfected cells (Fig. 8b). Using immunoprecipitation analysis, however, only the 140-kDa form could apparently bind VacA (Fig. 8a). To characterize the effects of VacA in COS-7 cells expressing RPTP␣, we assayed VacA-induced vacuolation in cells transfected with vector or RPTP␣ gene after incubation with 120 nM VacA for 8 h (data not shown). Both untransfected and transfected cells with the same magnitude or the vacuole formation in transfected cells were slightly increased by treatment with VacA, indicating that additional binding of VacA did not result in a remarkable increase in vacuoles in RPTP␣-expressing cells.
RPTP␣ Antisense Oligonucleotide-To examine the relationship between VacA sensitivity and RPTP␣ expression in G401 cells, cells were treated with 1.4 M RPTP␣-morpholino antisense oligonucleotide followed by incubation with 120 nM VacA  (15, 000 ϫ g, 15 min), the sedimented material was dissolved in Sol buffer on ice. After centrifugation and filtration, the filtrate was applied to a PNA-agarose column (bed volume 5 ml) that was washed with 25 ml of Sol buffer to yield 25 1-ml fractions (P1-P25). Bound proteins were eluted with three 1-ml volumes of Sol buffer containing 0.3 M D-galactose, collected as fractions E1, E2, and E3. VacA binding activity by proteins in fractions P1, E1, E2, and E3 was tested using VacA and immunoprecipitation with anti-VacA antibodies. Data are representative of three separate experiments (a). Using the same protocol, 3 ml of effluents were prepared using PNA column chromatography, for separation of non-biotinylated proteins from G401 cells. Fractions were pooled and proteins precipitated with 10% trichloroacetic acid were washed five times with ice-cold acetone, dried, and separated by SDS-PAGE in 5% gels, which were stained with Coomassie Brilliant Blue (b). Positions of molecular mass standards (kDa) and p140 are indicated on the right. for 0, 2, or 4 h. VacA-induced vacuolation was decreased in cells treated with antisense oligonucleotide, suggesting that it was mediated by RPTP␣ (Fig. 9). In additional control experiments, RPTP␤ antisense oligonucleotide (22), which was used instead of RPTP␣ antisense oligonucleotide, had no effect (data not shown).
Inhibition of VacA Internalization into G401 Cells by Treatment with RPTP␣ Antisense Oligonucleotide-To assess the effect of treatment with RPTP␣ antisense oligonucleotide on VacA internalization, we quantified VacA internalization into G401 using VacA-SS-biotin. As shown in Fig. 10, in cells transfected with RPTP␣ antisense oligonucleotide there was a marked decrease in VacA internalization. DISCUSSION Identification and functional analysis of the receptors for bacterial toxins is important not only to understand their roles in cell intoxication but also to develop methods to neutralize toxicity following bacterial infection. Our previous studies showed that RPTP␤ serves as a receptor for VacA on AZ-521 cells, a human gastric cancer cell line (23,24); a second protein, p140, was also commonly detected in AZ-521 and AGS, including non-gastric VacA-sensitive cells such as the monkey kidney line COS-7 (19). Here we report that internal amino acid sequence analysis and time of flight-mass analysis of p140, which was isolated from G401 cells, identified their receptor as RPTP␣. In agreement with this finding, the binding of V5tagged RPTP␣ to VacA was demonstrated by immunoprecipitation using VacA and anti-VacA antibody as shown in Fig. 8. In addition, treatment of G401 cells with RPTP␣-morpholino antisense oligonucleotide before exposure to VacA inhibited its induction of vacuolation (Fig. 9). These results support the involvement of RPTP␣ in VacA-induced vacuolation in G401 cells.
In transfected COS-7 cells, two forms of overexpressed V5tagged RPTP␣ have higher molecular masses (140 and 110 kDa) than the mass calculated (90,559) from RPTP␣ cDNA (29), suggesting that they have undergone glycosylation. Inter-estingly, V5-tagged RPTP␣ of 140 kDa, but not the 110-kDa form, bound VacA, suggesting that the sugar modification of RPTP␣ is important for binding to VacA. In agreement with this suggestion, neuraminidase treatment diminished the binding to VacA and VacA-induced vacuolation as shown in Figs. 5 and 6.
For many bacterial toxins, interaction with specific cell-surface receptors results in toxin internalization. VacA is known to be internalized by eukaryotic cells (19,31), and several lines of evidence indicate that VacA-induced cell vacuolation results from VacA activity at an intracellular site (17,18,32). Fig. 10 suggests that VacA interaction with RPTP␣ is required for toxin internalization by G401 cells. FIG. 8. Immunoprecipitation of V5-tagged RPTP␣ expressed in COS-7 cells using heat-inactivated or native VacA. After transfection of COS-7 cells with vector or V5-tagged RPTP␣, proteins in cell lysates were separated by SDS-PAGE in 5% gels, transferred to PVDF membranes, and detected using HRP-conjugated anti-V5 monoclonal antibodies followed by ECL detection (b). To assess VacA binding, lysates were incubated with heat-inactivated (I) or native VacA (A) at 4°C for 1 h before immunoprecipitation with anti-VacA antibody. Immunoprecipitated proteins were separated by SDS-PAGE in 5% gels and transferred to PVDF membranes. RPTP␣ was identified by its reaction with HRP-conjugated anti-V5 monoclonal antibody (1:2,000) and detected by ECL (a). Positions of molecular mass standards (kDa) and V5-tagged RPTP␣ are indicated. Data are representative of three separate experiments. Most RPTPs such as RPTP␣ and RPTP␤ contain two cytoplasmic PTP domains, a membrane proximal domain (D1) and a membrane distal domain (D2), and, in addition, have a single transmembrane segment and an extracellular domain. Structural variability of the extracellular domains of RPTPs suggests selective ligand interactions. The extracellular domain of RPTP␤ has recognition sites for many ligands including tenascin, pleiotrophin, contactin, and N-CAM (31). As VacA can bind RPTP␣ as well as RPTP␤, we compared the sequences of two extracellular regions of significant similarity, which could serve as VacA-binding domains. The amino acid sequence of RPTP␣ positions 24 -74 is 26.7% identical to that of 661-724 in RPTP␤, suggesting that this region of RPTP␣ may function in VacA binding.
Interestingly, PTPs are a key group of signal transduction enzymes together with protein-tyrosine kinases, and control the levels of cellular protein tyrosine phosphorylation. The capacity of protein-tyrosine phosphatases to dephosphorylate phosphotyrosine residues selectively on their substrates plays a pivotal role in initiating, sustaining, and terminating cellular signaling. Impaired protein-tyrosine phosphatase signal transduction caused by bacterial toxin action may be linked to disease. The combination of structural studies, kinetic analysis of protein-tyrosine phosphatase domains, and studies involving substrates in VacA-treated cells may provide valuable information regarding the mechanism of VacA toxicity through RPTP␣ as well as RPTP␤.