Cloning and Characterization of Secretory Tyrosine Phosphatases of Mycobacterium tuberculosis

Abstract

Two genes with sequence homology to those encoding protein tyrosine phosphatases were cloned from genomic DNA of Mycobacterium tuberculosis H₃₇Rv. The calculated molecular masses of these two putative tyrosine phosphatases, designated MPtpA and MPtpB, were 17.5 and 30 kDa, respectively. MPtpA and MPtpB were expressed as glutathione S-transferase fusion proteins in Escherichia coli. The affinity-purified proteins dephosphorylated the phosphotyrosine residue of myelin basic protein (MBP), but they failed to dephosphorylate serine/threonine residues of MBP. The activity of these phosphatases was inhibited by sodium orthovanadate, a specific inhibitor of tyrosine phosphatases, but not by okadaic acid, an inhibitor of serine/threonine phosphatases. Mutations at the catalytic site motif, cysteine 11 of MPtpA and cysteine 160 of MPtpB, abolished enzyme activity. Southern blot analysis revealed that, while mptpA is present in slow-growing mycobacterial species as well as fast-growing saprophytes, mptpB was restricted to members of the M. tuberculosis complex. These phosphatases were present in both whole-cell lysates and culture filtrates of M. tuberculosis, suggesting that these proteins are secreted into the extracellular medium. Since tyrosine phosphatases are essential for the virulence of several pathogenic bacteria, the restricted distribution of mptpB makes it a good candidate for a virulence gene of M. tuberculosis.

With one-third of the world population infected with tubercle bacilli that cause 3 million deaths every year, tuberculosis continues to be the most important source of deaths from infectious diseases (30). The problem is exacerbated by the spread of AIDS and the development of resistance against most of the antibiotics used in the treatment of tuberculosis. The need to focus on global tuberculosis control through basic and applied research in its diagnosis, treatment, and prevention cannot be overemphasized. An important prerequisite for rapid development in these areas is understanding the host-pathogen interaction and its contribution to the development of disease. Currently, our knowledge of how Mycobacterium tuberculosis enters the host cell, circumvents host defenses, and spreads to neighboring cells is inadequate.

Pathogenicity of a microorganism normally depends on the ability of the organism to survive and replicate in the host. M. tuberculosis has evolved mechanisms to circumvent the hostile environment of the macrophage (29). These mechanisms include inhibition of phagosome-lysosome fusion (1), inhibition of acidification of phagosomes (32), and recruitment and retention of a tryptophan aspartate-containing host protein to phagosomes that prevents their delivery to lysosomes (10). These activities, which allow mycobacteria to escape the bactericidal action of macrophages, require live bacteria (10). Therefore, bacteria may be able to trigger specific signals within the host cell that interfere with its normal functioning.

The importance of tyrosine phosphorylation in eukaryotic cells is well established. For example, reversible phosphorylation of tyrosine residues has been shown to represent a key mechanism for the transduction of signals that regulate cell growth, differentiation, mobility, metabolism, and survival (36). The level of phosphorylation on tyrosine residues required for normal cell function is maintained by the opposing actions of tyrosine kinases and phosphatases (31). In some bacteria, protein phosphorylation plays an important role in sensing extracellular signals and coordinating intracellular events (20). Thus, it is not surprising that in pathogenic bacteria, such as Yersinia pseudotuberculosis (13, 15), Salmonella enterica serovar Typhimurium (19), and enteropathogenic Escherichia coli (27), tyrosine kinases and phosphatases act as major virulence determinants. In Yersinia, for example, expression of a tyrosine phosphatase disrupts the host signal transduction processes involved in bacterial killing (4). For mycobacteria, the significance of protein phosphorylation for intracellular survival, propagation, and pathogenicity is not understood.

In the present study, we report the cloning and characterization of two tyrosine phosphatases from M. tuberculosis. The proteins were expressed in E. coli as glutathione S-transferase (GST) fusion proteins, purified, and characterized with respect to catalytic activity. In addition, we show that these phosphatases are secreted into the culture medium. Based on the knowledge of phosphatase function in other pathogens, we suggest that these enzymes may play an important role in the pathogenicity of mycobacteria by interfering with phosphotyrosine-mediated signals in macrophages.

MATERIALS AND METHODS

Bacterial strains, plasmids, and antibodies.

Whole-cell lysates and culture filtrate proteins of M. tuberculosis (H₃₇Rv and H₃₇Ra) were provided by John T. Belisle (Fort Collins, Colo.) under the Tuberculosis Research Material and Vaccine Testing Program of the National Institute of Allergy and Infectious Diseases, National Institutes of Health (contract no. AI-75320). Genomic DNA of M. tuberculosis H₃₇Rv and H₃₇Ra, Mycobacterium bovis BCG, and Mycobacterium smegmatis was prepared as described previously (8). The expression plasmid (pGEX-5X-3) was purchased from Pharmacia (Uppsala, Sweden). Rabbit polyclonal antisera against ERK2 and anti-Src antibodies (mouse monoclonal antibodies) were purchased from Santa Cruz Biotechnology (Santa Cruz, Calif.) and Upstate Biotechnology Inc. (Lake Placid, N.Y.), respectively.

Plasmid construction and mutagenesis.

M. tuberculosis H₃₇Rv genomic DNA was used as a template for amplification of two putative tyrosine phosphatase genes by PCR. The two genes were designated mptpA (492 bp) and mptpB (831 bp). The sequences of the two PCR primers for cloning mptpA were 5′ GGAATTCCATGTCTGATCCGCTGCACGTCACATTC-3′ for the 5′ end (carrying an EcoRI site) and 5′ CCGCTCGAGTCAACTCGGTCCGTTCCGCGCGAGAC-3′ for the 3′ end of the gene (carrying an XhoI site). To clone mptpB, the sequences of the two primers were 5′-CGGGATCCCGATGGCTGTCCGTGAACTGCCGGG-3′ for the 5′ end of the gene (containing a BamHI site) and 5′-CGAATTCTCATCCGAGCAGCACCCCGCGCATCCG-3′ for the 3′ end of the gene (containing an EcoRI site). The amplified product of mptpA was digested with EcoRI and XhoI, and the resulting fragment was inserted into the pGEX-5X-3 plasmid, which was previously digested with the same restriction enzymes. The resulting plasmid was designated pGEX-mptpA. Similarly, the PCR-amplified product of mptpB was digested with BamHI and EcoRI and inserted by ligation into the pGEX-5X-3 plasmid, also digested with BamHI and EcoRI. The resulting plasmid was designated pGEX-mptpB.

Site-directed mutagenesis of cysteine 11 of MPtpA and cysteine 160 of MPtpB to serine was carried out as described previously (21). The oligonucleotide for mutating cysteine 11 to serine in MPtpA was 5′-GTCACATTCGTTAGTACGGGCAACATC-3′, and the oligonucleotide for mutating cysteine 160 to serine in MPtpB was 5′-CCGGTGCTCACCCACAGCTTCGCGGGTAAGGATC-3′ (the underlined bases indicate the alteration to encode serine rather than cysteine). The plasmids with the mutant genes were designated pGEX-mptpA-C11S and pGEX-mptpB-C160S. The nucleotide sequence of each gene was confirmed by sequencing using the dideoxynucleotide method (28).

Expression and purification of MPtpA and MPtpB.

E. coli strain BL21 was separately transformed with pGEX-mptpA, pGEX-mptpB, pGEX-mptpA-C11S, and pGEX-mptpB-C160S plasmids. Transformants were grown in 2YT medium containing 100 μg of ampicillin per ml at 37°C until the A₆₀₀ reached 0.5. Isopropyl-1-thio-β-d-galactopyranoside (IPTG) was then added to a final concentration of 0.5 mM, and cultures were further grown for 5 h at 37°C with shaking. Cells were harvested by centrifugation at 5,000 × g for 15 min and suspended in 20 ml of sonication buffer (50 mM Tris-Cl [pH 7.4], containing 150 mM NaCl, 1 mM EDTA, 10% glycerol, 1 mM phenylmethylsulfonyl fluoride, and 10 μg of aprotinin per ml). The cells were then sonicated on ice for 2 min, and the sonicate was supplemented with Triton X-100 to a final concentration of 1% before centrifugation at 30,000 × g for 30 min at 4°C. The supernatant was incubated overnight at 4°C with glutathione-Sepharose 4B matrix (Pharmacia Biotech). The resin bound to protein was packed into a column and washed with 5 bed volumes of phosphate-buffered saline. Protein was eluted with 50 mM Tris-Cl, pH 8.0, containing 1 mM dithiothreitol, 5 mM MgCl₂, and 15 mM glutathione. Fractions were analyzed by sodium dodecyl sulfate (SDS)–12.5% polyacrylamide gel electrophoresis (PAGE) (22). Fractions containing purified fusion proteins were pooled and dialyzed against phosphate-buffered saline containing 20% glycerol and stored at −20°C.

Preparation of ³²P-labeled phosphoprotein substrate.

Human 293 embryonic kidney cells were obtained from the American Type Culture Collection and grown in Dulbecco's modified Eagle's medium supplemented with 2 mM glutamine and 10% fetal calf serum. The cells were then transfected separately with plasmid p60^c-Src, carrying Src kinase (a tyrosine kinase), or with a plasmid carrying the ERK2 kinase gene (a serine/threonine kinase gene) as described previously (5). Cells overexpressing the desired proteins were lysed, and Src kinase and ERK2 kinase were immunoprecipitated from the cell lysates using anti-Src or anti-ERK2 antibodies as described previously (39). The immunoprecipitate containing each protein was washed three times with 0.5 ml of washing buffer (20 mM HEPES [pH 7.5], 150 mM NaCl, 1% Triton X-100, 10% glycerol, 10 mM NaF, and 1 mM sodium orthovanadate) and then once with kinase buffer (20 mM HEPES [pH 7.5], 10 mM MgCl₂, 1 mM dithiothreitol, and 200 μM sodium orthovanadate).

The substrate, myelin basic protein (MBP), was phosphorylated either at tyrosine residues by immunoprecipitated Src kinase or at serine/threonine residues by immunoprecipitated ERK2 kinase in separate reactions. In brief, MBP (10 μg) was incubated at 30°C for 30 min with kinase in the kinase buffer (20 μl) containing 20 μCi of [γ-³²P]ATP. The reaction was stopped, and unincorporated ATP was removed by adding ice-cold trichloroacetic acid (25% final concentration). The precipitate was washed twice with 10% trichloroacetic acid and once with acetone. The phosphorylated substrates were dissolved in 25 mM imidazole, pH 7.4, and used for dephosphorylation assays. The phosphorylated substrates were analyzed for phosphorylated amino acids as described previously (33).

Phosphatase assay.

The phosphatase assay measured release of ³²P_i from ³²P-labeled substrates. The activities of purified MPtpA and its mutant derivative were assayed by incubating phosphorylated MBP (0.5 μg) for 120 min at 37°C in an imidazole buffer (25 mM, pH 7.0) containing 0.05% β-mercaptoethanol and 0.1 mg of bovine serum albumin per ml. Similarly, the activities of MPtpB and its mutant protein were determined by using sodium acetate buffer (50 mM, pH 5.6). The reactions were terminated by the addition of SDS sample buffer and analyzed by SDS–12.5% PAGE. The gel was electroblotted to a nitrocellulose membrane and autoradiographed to determine dephosphorylation.

Purification of rabbit antibodies against MPtpA and MPtpB.

Purified GST-MPtpA fusion protein (500 μg) and GST-MPtpB fusion protein (200 μg) were separately solubilized in 1 ml of Freund's complete adjuvant and injected into rabbits. Subsequently, three injections of 250 μg each in 1 ml of Freund's incomplete adjuvant were given after an interval of 15 days. Ten days after the final injection, animals were bled and titers of anti-GST-MPtpA and anti-GST-MPtpB were determined by enzyme-linked immunosorbent assay as described previously (16). The antibodies specific to MPtpA and MPtpB were isolated by passaging the immunized rabbit serum on Sepharose resin coupled to MPtpA and MPtpB. The coupling of Sepharose to phosphatases and purification of antibodies were performed as described previously (16).

Southern blot analysis.

Genomic DNAs (7 μg each) from M. tuberculosis H₃₇Rv and H₃₇Ra, M. bovis BCG, and M. smegmatis were digested with restriction enzymes (HincII and XmnI for mptpA and HincII and XmaI for mptpB). Digested products were separated by electrophoresis in a 1% agarose gel at 25 to 30 V for 16 h and transferred to nitrocellulose membranes. Hybridization was performed at 66°C using 6× SSC (1× SSC is 150 mM sodium chloride and 15 mM sodium citrate, pH 7.2) and ³²P-labeled mptpA and mptpB probes as described previously (26), and hybrids were subjected to autoradiography.

RESULTS

Expression and purification of MPtpA and MPtpB.

The complete sequence of the M. tuberculosis genome has revealed two DNA sequences that have homology to those of protein tyrosine phosphatases (PTPs) (7). They were expected to encode translation products of 17.5 kDa (MPtpA) and 30 kDa (MPtpB). Both of these genes were amplified by PCR using oligonucleotide primers deduced from the genomic sequence of M. tuberculosis (7). The amplified DNA products of mptpA and mptpB were cloned into pGEX-5X-3. The resulting plasmids (pGEX-mptpA and pGEX-mptpB) were used to transform E. coli, and the transformants expressed MPtpA and MPtpB fused with GST (29 kDa) at its NH₂ terminus. An in vitro transcription and translation assay was carried out in order to confirm that pGEX-mptpA and pGEX-mptpB encoded translation products of 46.5 kDa (GST-MPtpA) and 59 kDa (GST-MPtpB), respectively (data not shown).

The expressed GST-fusion proteins (GST-MPtpA and GST-MPtpB) were purified using a glutathione-Sepharose 4B matrix and analyzed by SDS-PAGE (Fig. 1). The size of the fusion proteins was found to be consistent with the calculated molecular mass of these proteins. The typical yield of purified proteins was about 2 mg from 1 liter of bacterial culture.

FIG. 1.

Electrophoretic analysis of recombinant tyrosine phosphatases. Affinity-purified tyrosine phosphatases were separated by SDS–12.5% PAGE and stained with Coomassie blue. Lane 1, GST protein; lane 2, GST-MPtpA fusion protein; lane 3, GST-MPtpB fusion protein.

Phosphotyrosine activity of MPtpA and MPtpB.

The tyrosine phosphatase activity of the purified proteins was determined by their ability to dephosphorylate tyrosine-phosphorylated MBP. Analysis of phosphorylated amino acids of MBP was performed to identify specific phosphorylated residues. Labeled MBP was acid hydrolyzed and analyzed by two-dimensional thin-layer chromatography. Incubation of MBP with immunoprecipitated Src kinase led to the phosphorylation of tyrosine residues (Fig. 2a), whereas MBP incubated with immunoprecipitated ERK2 phosphorylated serine/threonine residues (Fig. 2b). Incubation of purified MPtpA with tyrosine-phosphorylated MBP led to efficient dephosphorylation of tyrosine residues at pH 7.0 (Fig. 3a). Similarly, MPtpB dephosphorylated tyrosine residues of phosphorylated MBP (Fig. 3b). The optimum dephosphorylation of tyrosine residues of MBP by MPtpB was observed at pH 5.5 to 5.8 (data not shown).

FIG. 2.

Analysis of phosphorylated residues of MBP. MBP was phosphorylated by either Src kinase (a) or ERK2 kinase (b) using [γ-³²P]ATP. Phosphorylated MBP was run on an SDS–15% polyacrylamide gel and electroblotted on a polyvinylidene fluoride membrane, and bands containing proteins were excised and acid hydrolyzed in 5.7 M HCl for 90 min at 110°C. The acid-stable phosphoamino acids liberated by hydrolysis were separated by two-dimensional electrophoresis and autoradiographed. ³²P_i was produced by partial acid hydrolysis of labeled amino acids. Samples of nonradioactive phosphotyrosine, phosphoserine, and phosphothreonine were run in parallel and visualized by ninhydrin staining.

FIG. 3.

Protein dephosphorylation assays. The substrate, MBP, was phosphorylated at either tyrosine or serine/threonine residues by immunoprecipitated Src kinase or ERK2 kinase as described in Materials and Methods. Equal amounts of ³²P-Tyr-labeled MBP (0.5 μg, 650 cpm/μg of MBP) and ³²P-Ser/Thr-labeled MBP (0.5 μg, 800 cpm/μg of MBP) were incubated with purified native and mutant tyrosine phosphatases (0.3 μg) for 120 min at 37°C. The samples were loaded on SDS–15% polyacrylamide gels, electroblotted, and autoradiographed to determine dephosphorylation. (a) Activity of MPtpA (lane 1, MBP alone; lanes 2 and 3, MBP incubated with native and mutant MPtpA, respectively). (b) Activity of MPtpB (lane 1, MBP alone; lanes 2 and 3, MBP incubated with mutant and native MPtpB, respectively). (c) Activity of MPtpA and MPtpB with ³²P-Ser/Thr-labeled MBP (lane 1, MBP alone; lanes 2 to 5, MBP incubated with native MPtpA, mutant MPtpA, native MPtpB, and mutant MPtpB, respectively).

MPtpB showed 26.8% sequence homology to tyrosine/serine phosphatase (IphP) of Nostoc commune (25). The N. commune phosphatase has been shown to display phosphatase activity toward both tyrosine and serine residues. Thus, the substrate specificity of purified MPtpA and MPtpB was determined using the MBP substrate phosphorylated at serine/threonine residues. Both MPtpA and MPtpB failed to dephosphorylate serine/threonine residues of MBP, unlike IphP (Fig. 3c). These results suggest that mycobacterial phosphatases are specific for tyrosine residues.

Role of catalytic cysteines of MPtpA and MPtpB.

MPtpA is a low-molecular-weight (LMW) phosphatase having a striking similarity to other LMW phosphatases with respect to sequence homology in the catalytic domain (Fig. 4a). The conserved catalytic site cysteine of LMW phosphatases has been shown previously to be essential for their activity (14). In order to determine the role of cysteine 11 present in the catalytic domain of MPtpA, the residue was changed to serine. The mutant protein (GST-MPtpA-C11S) was expressed, purified, and assayed for activity. The mutant protein had no enzymatic activity, suggesting that cysteine 11 is crucial for the enzymatic activity (Fig. 3a).

FIG. 4.

Comparison of MPtpA and MPtpB with other known tyrosine phosphatases. (a) Alignment of MPtpA with LMW phosphatases from Streptomyces coelicolor (PTPA) (23), S. pombe (PPAL) (24), and bovine heart (PPAC) (35). (b) Alignment of MPtpB with N. commune tyrosine phosphatase (IphP) (25). Identities between catalytic site residues of MPtpA and MPtpB and those of other tyrosine phosphatases are shown by boxes. The catalytic site domain of MPtpA is located a few amino acids upstream from the N terminus. Identical amino acids are indicated by asterisks, and high similarity is indicated by double dots. Hyphens indicate gaps introduced to optimize alignment.

Comparison of the catalytic site of MPtpB with those of other bacterial and eukaryotic tyrosine phosphatases also revealed a similarity in amino acid sequence (Fig. 4b). The catalytic site cysteine (Cys-160) of MPtpB was replaced with serine by site-directed mutagenesis, and the mutant protein (GST-MPtpB-C160S) was expressed and purified. The mutant protein failed to dephosphorylate tyrosine-phosphorylated MBP, indicating loss of enzymatic activity (Fig. 3b).

Inhibition of enzymatic activities of MPtpA and MPtpB.

The activities of MPtpA and MPtpB toward tyrosine-phosphorylated MBP were blocked by 1 mM sodium orthovanadate, an inhibitor of PTPs. However, okadaic acid, a potent inhibitor of protein serine/threonine phosphatases; tetramisole, an inhibitor of alkaline phosphatase; tartrate, an acid phosphatase inhibitor; and sodium fluoride, a nonspecific inhibitor of serine/threonine phosphatases, had no significant effect on the activity of MPtpA or MPtpB (Fig. 5).

FIG. 5.

Effect of various inhibitors on the activity of MPtpA and MPtpB. ³²P-Tyr-labeled MBP (0.5 μg) was incubated with MPtpA (0.2 μg) in imidazole buffer (pH 7.0) or MPtpB (0.2 μg) in sodium acetate buffer (pH 5.6) in the presence of sodium orthovanadate (1 mM), okadaic acid (100 nM), sodium tartrate (5 mM), sodium fluoride (1 mM), and tetramisole (1 mM) for 30 min at 30°C. Samples were electrophoresed on an SDS–15% polyacrylamide gel and electroblotted, and dephosphorylation was quantitated using a PhosphorImager. Activity is reported as the percentage of phosphorylated MBP remaining after incubation with enzyme in the presence of indicated inhibitors. Each value is the average of two individual reactions.

Western blot analysis of mycobacterial tyrosine phosphatases.

Monospecific polyclonal antibodies raised against MPtpA and MPtpB were used to analyze the expression of tyrosine phosphatases in growing mycobacterial cultures. Equal amounts of mycobacterial whole-cell lysates and culture filtrate proteins from M. tuberculosis H₃₇Rv and H₃₇Ra strains were separated by SDS–15% PAGE and electroblotted on nitrocellulose membranes. The membranes were incubated with monospecific antibodies and visualized using an enhanced chemiluminescence kit (Dupont, NEN Research Products, Boston, Mass.). The affinity-purified antibodies were specific for MPtpA and MPtpB as seen by immunoblot analysis (Fig. 6). Both MPtpA and MPtpB were present in whole-cell lysates of M. tuberculosis H₃₇Rv and H₃₇Ra. The culture filtrate, which was prepared from M. tuberculosis grown to mid-log phase, also showed the presence of MPtpA and MPtpB proteins (Fig. 6). Thus, both phosphatases are secreted into the culture medium by growing mycobacterial cells. The secreted MPtpA in M. tuberculosis H₃₇Rv and H₃₇Ra was slightly smaller than the cytosolic MPtpA.

FIG. 6.

Expression of tyrosine phosphatases in M. tuberculosis. Equal amounts of whole-cell lysates (40 μg) and culture filtrate proteins (40 μg) from M. tuberculosis strains H₃₇Rv and H₃₇Ra were loaded on an SDS–15% polyacrylamide gel and electroblotted. Blots were probed with anti-MPtpA (a) or anti-MPtpB (b) antibodies and developed using enhanced chemiluminescence reagents (NEN).

Analysis of prevalence of tyrosine phosphatases in other species of mycobacteria.

The PCR products of mptpA (492 bp) and mptpB (831 bp) were used in Southern hybridization experiments to determine the prevalence of mptpA and mptpB homologs in various species of mycobacteria. Hybridization results revealed that an mptpA-homologous gene was present in all the members of the M. tuberculosis complex analyzed in this study as well as in M. smegmatis, a saprophytic organism. However, sequences homologous to mptpB were found to be present exclusively among the members of the M. tuberculosis complex (Fig. 7).

FIG. 7.

Presence of tyrosine phosphatase genes in other mycobacteria. Genomic DNAs (7 μg each) from various strains of M. tuberculosis (H₃₇Rv and H₃₇Ra), M. bovis BCG, and M. smegmatis were digested with restriction enzymes, resolved on a 1% agarose gel at 25 to 30 V for 16 h, and transferred to a nitrocellulose membrane. The hybridization was performed using ³²P-labeled mptpA (a) and mptpB (b) probes, and the hybrids were autoradiographed. Lanes 1, M. smegmatis; lanes 2, M. bovis BCG; lanes 3, H₃₇Ra; lanes 4, H₃₇Rv. Numbers at left are molecular sizes in kilobase pairs.

DISCUSSION

PTPs have long been considered to be confined to eukaryotes. Only recently have genes encoding PTPs been found in bacteria (20, 23). For example, Y. pseudotuberculosis secretes a PTP (YopH) which is essential for survival in the host cells (15). YopH is secreted into the extracellular medium by the bacterium and is targeted to the inner surface of macrophages, where it dephosphorylates host proteins that are implicated in bactericidal action (3, 4). With S. enterica serovar Typhimurium, a secreted tyrosine phosphatase (SptP) is required for full virulence in a murine model (11, 19). The entry and survival of intracellular pathogens in host cells require a complex dialogue of signaling events between the host cells and the pathogenic bacteria (12). Therefore, understanding the mechanisms involved in the signal cross talk between bacterial pathogens and their host cells may help us in the development of effective therapeutic targets against these diseases.

The entry of M. tuberculosis into macrophages and subsequent events appear to involve specific signals between the host cell and the bacterium, suggesting that molecules such as tyrosine phosphatases and kinases may be necessary for the reprogramming of the host signaling network that helps the bacterium in its propagation. We characterized the PTPs from mycobacteria as part of an effort to understand the pathogenesis of M. tuberculosis. Two genes with sequence homology to genes encoding PTPs were cloned from the genomic DNA of M. tuberculosis (7). The putative PTPs were expressed in E. coli, and after affinity purification, the proteins were characterized. It has been observed previously that several PTPs like IphP in N. commune and Stp1 in Schizosaccharomyces pombe can dephosphorylate both tyrosine and serine/threonine residues of substrates (25, 38). In order to determine the substrate specificity of the PTPs from M. tuberculosis, MBP phosphorylated at either tyrosine or serine/threonine residues was used as a substrate in a dephosphorylation reaction with purified MPtpA or MPtpB protein. Both MPtpA and MPtpB were specific for phosphotyrosine residues and showed no activity for phosphoserine or phosphothreonine.

MPtpA, a 17.5-kDa protein, displayed sequence homology with LMW tyrosine phosphatases isolated from bovine heart and the yeast S. pombe (24, 35, 37). LMW phosphatases (previously called acid phosphatases) lack regulatory domains, unlike other tyrosine phosphatases that contain both catalytic and regulatory domains (9). The catalytic domain of LMW phosphatases is located within a few amino acids of the N terminus of the protein. Site-directed mutagenesis of cysteine 11 to serine in MPtpA completely abolished enzymatic activity, suggesting that cysteine 11 is the conserved catalytic site residue present adjacent to the N terminus of the protein. MPtpB, a 30-kDa protein, exhibited sequence homology with the PTP (IphP) of N. commune, whose catalytic site, unlike that in LMW phosphatases, is located near the C-terminal portion of the protein (25). Sequence homology between MPtpB and IphP suggests an evolutionary connection between mycobacterial and other prokaryotic tyrosine phosphatases. The catalytic cysteine is highly conserved in all PTPs, and it is required for the formation of covalent phosphoenzyme intermediates (6). When cysteine 11 of MPtpA and cysteine 160 of MPtpB were changed to serine, both mutant proteins failed to dephosphorylate tyrosine-phosphorylated MBP. Thus, cysteine 11 of MPtpA and cysteine 160 of MPtpB are required for enzyme activity consistent with MPtpA and MPtpB having the same catalytic mechanism as that of other PTPs. Moreover, MPtpA and MPtpB enzymatic activities were unaffected by okadaic acid, an inhibitor of protein serine/threonine phosphatases 1 and 2A. Both MPtpA and MPtpB were also insensitive to tetramisole or tartrate, indicating that their enzymatic activities did not arise from contaminating E. coli alkaline or acid phosphatases. These results indicate that mycobacterial tyrosine phosphatases are specific for phosphotyrosine residues.

M. tuberculosis is known to secrete a large number of proteins into the extracellular medium. These secreted proteins play an important role in the interaction of mycobacteria with the host cell (18), and they are thought to be prime candidates for the development of subunit vaccines and new antimycobacterial drugs (2). Both tyrosine phosphatases, MPtpA and MPtpB, were secreted into the culture medium, as revealed by Western blot analysis. The secreted MPtpA was slightly smaller than the cytosolic MPtpA. The reason for the difference is not clear, but it is possible that MPtpA may be processed before secretion. In most cases, secreted proteins have an N-terminal sequence encoding a signal peptide responsible for the transport of the proteins to the outside of the cell. The export of these proteins occurs after cleavage of the signal peptides by a specific peptidase. In the case of MPtpA, however, the catalytic domain is located only a few amino acids from the N terminus, suggesting that this phosphatase lacks a secretory signal peptide. Comparison of known signal sequences of mycobacterial proteins and conserved signal sequences of tyrosine phosphatases with sequences of both MPtpA and MPtpB using the BLAST search program provided no evidence for a signal peptide. Nevertheless, both tyrosine phosphatases were secreted in the extracellular medium by mycobacterial cells growing in mid-log phase. The mechanism employed by mycobacteria in exporting these proteins is presently unclear. Other mycobacterial proteins such as glutamine synthetase and superoxide dismutase are also secreted into the extracellular medium in the absence of signal peptides at the N terminus (17). Chaperone proteins may be involved in the secretion of mycobacterial tyrosine phosphatases, as has been found previously with Y. pseudotuberculosis for export of a PTP (YopH) into the host macrophages (34).

The gene coding for MPtpA was present in the members of the M. tuberculosis complex analyzed in this study as well as in M. smegmatis. However, the gene coding for MPtpB was restricted to members of the M. tuberculosis complex, suggesting that it may have a role in processes specific to the members of the M. tuberculosis complex.

In summary, the present study showed that mycobacteria express two active tyrosine phosphatases that are secreted into the culture medium. We are now examining the possibility that these phosphatases are translocated into host macrophages where they modify host phosphorylation patterns and thereby interfere with the host cell signal transduction pathways essential for the survival and pathogenicity of M. tuberculosis.