The tumor suppressor PTEN regulates cell migration, growth, and survival by removing the 3′-phosphate of phosphoinositides. Exposure of purified PTEN or of cells to H2O2 resulted in inactivation of PTEN in a time- and H2O2concentration-dependent manner. Analysis of various cysteine mutants, including mass spectrometry of tryptic peptides, indicated that the essential Cys124 residue in the active site of PTEN specifically forms a disulfide with Cys71during oxidation by H2O2. The reduction of H2O2-oxidized PTEN in cells appears to be mediated predominantly by thioredoxin. Thus, thioredoxin was more efficient than glutaredoxin, glutathione, or a 14-kDa thioredoxin-like protein with regard to the reduction of oxidized PTEN in vitro. Thioredoxin co-immunoprecipitated with PTEN from cell lysates; and incubation of cells with 2,4-dinitro-1-chlorobenzene (an inhibitor of thioredoxin reductase) delayed the reduction of oxidized PTEN, whereas incubation with buthioninesulfoximine (an inhibitor of glutathione biosynthesis) did not. These results suggest that the reversible inactivation of PTEN by H2O2might be important for the accumulation of 3′-phosphorylated phosphoinositides and that the uncontrolled generation of H2O2 associated with certain pathological conditions might contribute to cell proliferation by inhibiting PTEN function. The tumor suppressor PTEN regulates cell migration, growth, and survival by removing the 3′-phosphate of phosphoinositides. Exposure of purified PTEN or of cells to H2O2 resulted in inactivation of PTEN in a time- and H2O2concentration-dependent manner. Analysis of various cysteine mutants, including mass spectrometry of tryptic peptides, indicated that the essential Cys124 residue in the active site of PTEN specifically forms a disulfide with Cys71during oxidation by H2O2. The reduction of H2O2-oxidized PTEN in cells appears to be mediated predominantly by thioredoxin. Thus, thioredoxin was more efficient than glutaredoxin, glutathione, or a 14-kDa thioredoxin-like protein with regard to the reduction of oxidized PTEN in vitro. Thioredoxin co-immunoprecipitated with PTEN from cell lysates; and incubation of cells with 2,4-dinitro-1-chlorobenzene (an inhibitor of thioredoxin reductase) delayed the reduction of oxidized PTEN, whereas incubation with buthioninesulfoximine (an inhibitor of glutathione biosynthesis) did not. These results suggest that the reversible inactivation of PTEN by H2O2might be important for the accumulation of 3′-phosphorylated phosphoinositides and that the uncontrolled generation of H2O2 associated with certain pathological conditions might contribute to cell proliferation by inhibiting PTEN function. Hydrogen peroxide (H2O2) is produced by all mammalian cells as a by-product of normal metabolism, including the oxidative phosphorylation of ADP and the conversion of arachidonic acid to leukotrienes, as well as by phagocytic cells in the host defense response to noxious stimuli. In addition, ultraviolet and γ irradiation of cells results in the production of H2O2. A substantial increase in the intracellular concentration of H2O2 is generally associated with deleterious effects, including cell death by apoptosis or necrosis, in pathophysiological conditions such as inflammation and ischemia-reperfusion. The generation of H2O2 also appears to be required, however, for many normal cellular functions, including propagation of receptor signaling (1Rhee, S. G., Bae, Y. S., Lee, S.-R., and Kwon, J. (2000)Science stke(www.stke.org/cgi/contentfull/OC_sigtrans;2000/53/pe1)Google Scholar, 2Finkel T. Curr. Opin. Cell Biol. 1998; 10: 248-253Crossref PubMed Scopus (1019) Google Scholar). Ligands that induce an increase in the intracellular concentration of H2O2 include peptide growth factors such as platelet-derived growth factor (PDGF) 1The abbreviations used are: PDGFplatelet-derived growth factorDTTdithiothreitolTrxthioredoxinTrxRthioredoxin reductaseDMEMDulbecco's modified Eagle's mediumHAhemagglutininGrxglutaredoxinNEMN-ethylmaleimidePI(34,5)P3, phosphatidylinositol 3,4,5-trisphosphateHPLChigh-performance liquid chromatographyDNCB2,4-dinitro-1-chlorobenzeneBSObuthioninesulfoximineGSHreduced glutathioneMALDI-TOFmatrix-assisted laser desorption ionization time-of-flightPI(45)P2, phosphatidylinositol 4,5-biphosphatePI(34)P2, phosphatidylinositol 3,4-biphosphate and epidermal growth factor, cytokines such as transforming growth factor β1 and tumor necrosis factor α, and agonists of heterotrimeric GTP-binding protein (G protein)-coupled receptors such as N-formyl-methionyl-leucyl-phenylalanine (fMLP) and angiotensin II (1Rhee, S. G., Bae, Y. S., Lee, S.-R., and Kwon, J. (2000)Science stke(www.stke.org/cgi/contentfull/OC_sigtrans;2000/53/pe1)Google Scholar, 2Finkel T. Curr. Opin. Cell Biol. 1998; 10: 248-253Crossref PubMed Scopus (1019) Google Scholar). The essential role of H2O2 production in intracellular signaling triggered by PDGF (3Sundaresan M. Yu Z.X. Ferrans V.J. Irani K. Finkel T. Science. 1995; 270: 296-299Crossref PubMed Scopus (2342) Google Scholar, 4Brar S.S. Kennedy T.P. Whorton A.R. Murphy T.M. Chitano P. Hoidal J.R. J. Biol. Chem. 1999; 274: 20017-20026Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar), epidermal growth factor (5Bae Y.S. Kang S.W. Seo M.S. Baines I.C. Tekle E. Chock P.B. Rhee S.G. J. Biol. Chem. 1997; 272: 217-221Abstract Full Text Full Text PDF PubMed Scopus (1109) Google Scholar), angiotensin II (6Ushio-Fukai M. Alexander R.W. Akers M. Yin Q. Fujio Y. Walsh K. Griendling K.K. J. Biol. Chem. 1999; 274: 22699-22704Abstract Full Text Full Text PDF PubMed Scopus (501) Google Scholar), and cell-cell contact (7Pani G. Colavitti R. Bedogni B. Anzevino R. Borrello S. Galeotti T. J. Biol. Chem. 2000; 275: 38891-38899Abstract Full Text Full Text PDF PubMed Scopus (131) Google Scholar) has been demonstrated by the observation that corresponding receptor-mediated events are abrogated by blocking the accumulation of H2O2 with enzymes such as catalase or small molecules such as N-acetylcysteine. platelet-derived growth factor dithiothreitol thioredoxin thioredoxin reductase Dulbecco's modified Eagle's medium hemagglutinin glutaredoxin N-ethylmaleimide 4,5)P3, phosphatidylinositol 3,4,5-trisphosphate high-performance liquid chromatography 2,4-dinitro-1-chlorobenzene buthioninesulfoximine reduced glutathione matrix-assisted laser desorption ionization time-of-flight 5)P2, phosphatidylinositol 4,5-biphosphate 4)P2, phosphatidylinositol 3,4-biphosphate The addition of exogenous H2O2 or the intracellular production of this metabolite in response to receptor stimulation affects the function of a variety of proteins including transcription factors, protein kinases and phosphatases, phospholipases, ion channels, and G proteins (1Rhee, S. G., Bae, Y. S., Lee, S.-R., and Kwon, J. (2000)Science stke(www.stke.org/cgi/contentfull/OC_sigtrans;2000/53/pe1)Google Scholar, 2Finkel T. Curr. Opin. Cell Biol. 1998; 10: 248-253Crossref PubMed Scopus (1019) Google Scholar). However, the mechanisms by which H2O2 achieves these effects remain unknown. It is unlikely that H2O2specifically binds proteins and thereby affects their functions. On the other hand, H2O2 is a mild oxidant that is able to oxidize cysteine residues in proteins to cysteine sulfenic acid or to disulfide, both of which are readily reduced back to cysteine by various cellular reductants. Because the p Ka (where Ka is the acid constant) of the sulfhydryl group (Cys–SH) of most cysteine residues is ∼8.5 (8Besse D. Siedler F. Diercks T. Kessler H. Moroder L. Angew. Chem. Int. Ed. Engl. 1997; 36: 883-885Crossref Scopus (118) Google Scholar) and because this group is less readily oxidized by H2O2 than the cysteine thiolate anion (Cys–S–), few proteins might be expected to possess a Cys–SH that is vulnerable to oxidation by H2O2 in cells. However, certain protein cysteine residues do exist as thiolate anions at neutral pH as a result of the lowering of their p Ka values by charge interactions between the negatively charged thiolate and nearby positively charged amino acid residues (9Kim J.R. Yoon H.W. Kwon K.S. Lee S.R. Rhee S.G. Anal. Biochem. 2000; 283: 214-221Crossref PubMed Scopus (249) Google Scholar). Proteins with low p Ka cysteine residues include protein tyrosine phosphatases (10Jia Z. Barford D. Flint A.J. Tonks N.K. Science. 1995; 268: 1754-1758Crossref PubMed Scopus (566) Google Scholar, 11Denu J.M. Dixon J.E. Curr. Opin. Chem. Biol. 1998; 2: 633-641Crossref PubMed Scopus (340) Google Scholar, 12Lee S.R. Kwon K.S. Kim S.R. Rhee S.G. J. Biol. Chem. 1998; 273: 15366-15372Abstract Full Text Full Text PDF PubMed Scopus (848) Google Scholar). All protein tyrosine phosphatases contain an essential cysteine residue (p Ka, 4.7–5.4) in the signature active site motif, HC XXG XXRS/T (where X is any amino acid residue), that exists as a thiolate anion at neutral pH (11Denu J.M. Dixon J.E. Curr. Opin. Chem. Biol. 1998; 2: 633-641Crossref PubMed Scopus (340) Google Scholar). This thiolate anion contributes to formation of a thiol-phosphate intermediate in the catalytic mechanism of protein tyrosine phosphatases. The active site cysteine is the target of specific oxidation by various oxidants, including H2O2, and this modification is reversed by incubation with thiol compounds such as dithiothreitol (DTT) and reduced glutathione (GSH). The ability of intracellularly produced H2O2 to inhibit protein tyrosine phosphatase activity was demonstrated by the observation that stimulation of A431 cells with epidermal growth factor resulted in a selective reduction in the extent of subsequent labeling of the active site cysteine residue of protein tyrosine phosphatase 1B by [3H]iodoacetic acid in cell lysates (12Lee S.R. Kwon K.S. Kim S.R. Rhee S.G. J. Biol. Chem. 1998; 273: 15366-15372Abstract Full Text Full Text PDF PubMed Scopus (848) Google Scholar). PTEN is a member of the protein tyrosine phosphatase family and reverses the action of phosphoinositide 3-kinase by catalyzing the removal of the phosphate attached to the 3′-hydroxyl group of the phosphoinositide inositol ring (13Maehama T. Dixon J.E. J. Biol. Chem. 1998; 273: 13375-13378Abstract Full Text Full Text PDF PubMed Scopus (2646) Google Scholar, 14Maehama T. Dixon J.E. Trends Cell Biol. 1999; 9: 125-128Abstract Full Text Full Text PDF PubMed Scopus (513) Google Scholar, 15Maehama T. Taylor G.E. Dixon J.E. Annu. Rev. Biochem. 2001; 70: 247-279Crossref PubMed Scopus (411) Google Scholar). By negatively modulating the phosphoinositide 3-kinase-Akt signaling pathway, PTEN functions as an important tumor suppressor (15Maehama T. Taylor G.E. Dixon J.E. Annu. Rev. Biochem. 2001; 70: 247-279Crossref PubMed Scopus (411) Google Scholar, 16Furnari F.B. Lin H. Huang H.S. Cavenee W.K. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 12479-12484Crossref PubMed Scopus (386) Google Scholar, 17Stambolic V. Suzuki A. de la Pompa J.L. Brothers G.M. Mirtsos C. Sasaki T. Ruland J. Penninger J.M. Siderovski D.P. Mak T.W. Cell. 1998; 95: 29-39Abstract Full Text Full Text PDF PubMed Scopus (2146) Google Scholar, 18Myers M.P. Pass I. Batty I.H. Van der Kaay J. Stolarov J.P. Hemmings B.A. Wigler M.H. Downes C.P. Tonks N.K. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 13513-13518Crossref PubMed Scopus (1023) Google Scholar, 19Li D.M. Sun H. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 15406-15411Crossref PubMed Scopus (441) Google Scholar, 20Cantley L.C. Neel B.G. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 4240-4245Crossref PubMed Scopus (1768) Google Scholar). We now show that H2O2 induces reversible inactivation of PTEN through oxidation of the essential Cys124 and the formation by this residue of a disulfide with Cys71 and that the PTEN disulfide is reduced by thioredoxin (Trx). Given that stimulation of various receptors induces H2O2 production, we propose that the receptor-mediated activation of phosphoinositide 3-kinase may not be sufficient for the accumulation of 3′-phosphorylated phosphoinositides; the concomitant inactivation of PTEN by H2O2produced in response to receptor stimulation might also be necessary for this effect. Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum, penicillin, and streptomycin were from Invitrogen. Both PDGF-BB and rabbit polyclonal antibodies to PTEN were obtained from Upstate Biotechnology, and monoclonal antibodies to PTEN and to the hemagglutinin (HA) epitope tag were from Santa Cruz Biotechnology. Horseradish peroxidase-conjugated antibodies to mouse or rabbit immunoglobulin G were from Amersham Biosciences. Recombinant Trx and glutaredoxin (Grx) were prepared as described previously (12Lee S.R. Kwon K.S. Kim S.R. Rhee S.G. J. Biol. Chem. 1998; 273: 15366-15372Abstract Full Text Full Text PDF PubMed Scopus (848) Google Scholar). Thioredoxin reductase (TrxR) was prepared as described previously (21Lee S.R. Bar-Noy S. Kwon J. Levine R.L. Stadtman T.C. Rhee S.G. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 2521-2526Crossref PubMed Scopus (228) Google Scholar). Glucose oxidase, GSH, DTT, and N-ethylmaleimide (NEM) were from Sigma. Bovine xanthine oxidase and catalase were from Roche Molecular Biochemicals, and xanthine was from Calbiochem. A 30% solution of H2O2 was from Fisher. Human PTEN cDNA cloned in the pQE30 vector (Qiagen) for expression of the protein with a histidine tag at the NH2 terminus was kindly provided by K. M. Yamada (National Institutes of Health). The cDNAs encoding the PTEN mutants C71S, C83S, C105S, C124S, and C136S were generated by polymerase chain reaction-mediated site-directed mutagenesis and verified by DNA sequencing. The histidine-tagged wild-type and mutant proteins were expressed in Escherichia coli according to standard procedures and purified with the use of an immobilized nickel resin (Qiagen). The purified proteins were dialyzed against 50 mm Tris-HCl buffer (pH 7.5) containing 1 mmEDTA, 10% glycerol, and 10 mm 2-mercaptoethanol and then stored at –80 °C. The phosphatase activity of PTEN was assayed with32P-labeled phosphatidylinositol 3,4,5-trisphosphate (PI(3,4,5)P3) as described previously (13Maehama T. Dixon J.E. J. Biol. Chem. 1998; 273: 13375-13378Abstract Full Text Full Text PDF PubMed Scopus (2646) Google Scholar). HeLa and NIH 3T3 cells were grown at 37 °C under an atmosphere of 5% CO2 in DMEM supplemented with 10% fetal bovine serum. The cDNAs encoding wild-type PTEN or the C124S mutant tagged at their NH2 termini with HA were subcloned into the pCGN plasmid (22Tanaka M. Herr W. Cell. 1990; 60: 375-386Abstract Full Text PDF PubMed Scopus (599) Google Scholar), and the resulting vectors were introduced into NIH 3T3 cells by transfection with the use of Effectene (Qiagen). After stimulation, cells (1 × 106 cells in 1 ml of DMEM) were scraped into 0.2 ml of ice-cold 50% trichloroacetic acid and transferred to microfuge tubes. The cell suspensions were sonicated briefly and then centrifuged at 2000 × g for 5 min. The supernatants were removed, and the pellets were washed with acetone and then solubilized in 0.2 ml of 100 mm Tris-HCl (pH 6.8) buffer containing 2% SDS and 40 mm NEM. Portions (5 μl) of the solubilized pellets were subjected to SDS-PAGE under nonreducing conditions, and the separated proteins were transferred electrophoretically to a nitrocellulose membrane. The membrane was then subjected to immunoblot analysis with either rabbit antibodies to PTEN or a monoclonal antibody to HA. Immune complexes were detected with horseradish peroxidase-conjugated secondary antibodies and enhanced chemiluminescence reagents (AmershamBiosciences or Pierce). The intensity of PTEN bands was quantitated with an Image Station 440 instrument (Kodak Digital Science). Oxidized PTEN (100 μg) was obtained by incubating the purified protein for 5 min at room temperature in a final volume of 20 μl containing 50 mm Hepes-NaOH (pH 7.0), 150 mm NaCl, and 1 mm H2O2; the reaction was stopped by the addition of 1 μg of catalase. The protein was then denatured with 6 m guanidine-HCl in 100 mm Tris-HCl (pH 9.0), and the free cysteine residues were alkylated by incubation for 2 h with 10 mmiodoacetamide in an anaerobic chamber. Alkylated PTEN was digested overnight at 37 °C with Lys-C (0.5 μg/ml) in 100 mmTris-HCl (pH 8.5) containing 10% acetonitrile. The digestion products were then fractionated by HPLC on a C18 column with a linear gradient (0–60%) of acetonitrile in 0.1% trifluoroacetic acid at a flow rate of 1 ml/min. A peak that eluted at 30.5 min was exposed or not exposed to 10 mm DTT for 10 min and then analyzed by MALDI-TOF mass spectrometry. Purified human PTEN was inactivated by H2O2 in a concentration-dependent manner (Fig. 1A). PTEN was also inactivated on incubation with xanthine oxidase and xanthine, the combination of which results in the generation of superoxide anions that are subsequently converted to H2O2 (Fig. 1B). The inactivation of PTEN by the xanthine oxidase-xanthine system was inhibited by up to ∼80% by catalase (Fig. 1B), suggesting that PTEN oxidation was mediated predominantly by H2O2 rather than by superoxide anions. PTEN that had been treated with H2O2 migrated faster on SDS-PAGE under nonreducing conditions than did the untreated protein (Fig. 2A). This observation suggested that H2O2 induces the oxidation of sulfhydryl groups of PTEN to a disulfide, a reaction that generally results in a more compact protein structure. The PTEN protein comprises 403 amino acids, with the catalytic domain located in the NH2-terminal region and a C2 domain present in the COOH-terminal region (23Lee J.O. Yang H. Georgescu M.M., Di Cristofano A. Maehama T. Shi Y. Dixon J.E. Pandolfi P. Pavletich N.P. Cell. 1999; 99: 323-334Abstract Full Text Full Text PDF PubMed Scopus (894) Google Scholar). The catalytic domain of human PTEN contains cysteines at positions 71, 83, 105, 124, and 136 (15Maehama T. Taylor G.E. Dixon J.E. Annu. Rev. Biochem. 2001; 70: 247-279Crossref PubMed Scopus (411) Google Scholar). To identify the cysteine residues that form a disulfide, we changed each of these five cysteines in the catalytic domain individually to serine, thereby generating the mutants C71S, C83S, C105S, C124S, and C136S. The mutant proteins were incubated with H2O2 and then analyzed by gel electrophoresis. Like the wild-type protein, the C83S, C105S, and C136S mutants migrated faster after exposure to H2O2, whereas the mobility of the C71S and C124S mutants was not affected by H2O2 (Fig. 2A). These results thus suggested that, on oxidation with H2O2, the Cys124 residue in the active site of PTEN forms a disulfide bond with Cys71. To demonstrate directly the formation of a disulfide bond between Cys71 and Cys124 of PTEN, we subjected the H2O2-treated protein to alkylation under denaturing conditions with iodoacetamide (to block free cysteines) and then to digestion with the endoproteinase Lys-C. The resulting peptides were fractionated by reversed-phase HPLC. A peak that eluted at 30.5 min (data not shown) revealed several components on analysis by MALDI-TOF mass spectrometry, including one of 4432.86 mass units (Fig. 2B); this size is highly similar to the value of 4435.88 mass units calculated for a 37-residue fragment of PTEN in which a Cys71-containing peptide and a Cys124-containing peptide are bridged by a disulfide bond. When the HPLC peak fraction that eluted at 30.5 min was treated with DTT before analysis by mass spectrometry, no mass peak corresponding to the 37-residue fragment was detected; instead, fragments of 1696.37 and 2741.02 mass units, similar to the values of 1695.90 and 2741.98 mass units calculated for the 14-residue Cys71-containing peptide and the 23-residue Cys124-containing peptide, respectively, were apparent (Fig. 2B). These results thus demonstrate that Cys71 and Cys124 of PTEN form a disulfide on exposure to H2O2. This conclusion is consistent with the crystal structure of PTEN, which reveals that, among the four other cysteine residues in the catalytic domain, only Cys71 is sufficiently close (a distance of 5.9 Å) to Cys124 to form a disulfide with this residue (23Lee J.O. Yang H. Georgescu M.M., Di Cristofano A. Maehama T. Shi Y. Dixon J.E. Pandolfi P. Pavletich N.P. Cell. 1999; 99: 323-334Abstract Full Text Full Text PDF PubMed Scopus (894) Google Scholar). We examined whether PTEN is oxidized in cells exposed to H2O2. NIH 3T3 cells were incubated for 5 min with various concentrations of H2O2, after which cell extracts were exposed to NEM to block free sulfhydryls and then subjected to immunoblot analysis with antibodies to PTEN. Exposure of cells to H2O2 at concentrations as low as 50 μm resulted in the appearance of the higher mobility (oxidized) form of PTEN, and the intensity of this band increased as the concentration of H2O2 increased (Fig. 3A). The amount of the oxidized form of PTEN also increased with time of incubation of cells with 0.5 mm H2O2, reaching a maximum at 10–40 min and decreasing gradually thereafter (Fig. 3B). The amount of the reduced form of the protein decreased as the amount of the oxidized form increased and vice versa, suggesting that PTEN was inactivated by H2O2 and then reactivated by cellular reductants as the concentration of H2O2 declined. When extracts derived from NIH 3T3 cells treated with H2O2 were incubated with DTT before gel electrophoresis, only the lower mobility PTEN band was detected (Fig. 3, A and B), consistent with the notion that the higher mobility form of the protein contains a disulfide. To demonstrate that Cys124 contributes to the disulfide responsible for the increase in PTEN mobility, we transfected NIH 3T3 cells separately with vectors encoding either the C124S mutant of human PTEN tagged with HA or the HA-tagged wild-type protein. Exposure of the transfected cells to H2O2revealed that the HA-tagged wild-type protein, but not the HA-tagged C124S mutant, underwent reversible oxidation (Fig. 3C), confirming that Cys124 forms the disulfide bond responsible for the mobility shift of PTEN. The reversible inactivation of PTEN was also apparent in HeLa cells exposed to 1.5 mm H2O2 (Fig. 4A). When HeLa cells were incubated in a medium in which H2O2 was produced continuously as a result of the presence of glucose oxidase (which converts molecular oxygen to H2O2 with electrons derived from the glucose also present in the medium), the oxidized form of PTEN persisted (Fig. 4B). Two prominent physiological electron donors for protein reduction are GSH and Trx (24Holmgren A. Antioxid. Redox Signal. 2000; 2: 811-820Crossref PubMed Scopus (434) Google Scholar). GSH reduces proteins either directly or indirectly through the regeneration of Grx after its reaction with oxidized proteins. Oxidized Trx is reduced by TrxR with the use of electrons supplied by NADPH. Purified PTEN that had been inactivated by exposure to glucose and glucose oxidase was incubated with electron donors, and the extent of its reduction was monitored on the basis of the mobility shift in nonreducing gels (Fig. 5A). The most rapid reduction was achieved with the nonphysiological electron donor DTT (5 mm). The Trx system, comprising 5 μm Trx and saturating concentrations of TrxR and NADPH, was almost as efficient as DTT; the Trx system devoid of Trx was completely inactive (data not shown). GSH (5 mm) was markedly less effective than the Trx system, and the presence of 5 μm Grx in addition to 5 mm GSH (the Grx system) did not substantially increase the efficiency of the latter (Fig. 5A). Similar results were obtained when PTEN reduction was monitored on the basis of enzyme activity (data not shown). We also measured the rate of PTEN reactivation in the presence of various concentrations of Trx (Fig. 5B). Half-maximal reactivation was apparent at 1 μm Trx. Furthermore, immunoblot analysis revealed that Trx was co-immunoprecipitated with PTEN from NIH 3T3 cell extracts (Fig. 6). Estimation of the relative amounts of PTEN and Trx in the immunoprecipitates by comparison of the band intensities with those of protein standards yielded a molar ratio of PTEN to Trx of 1:0.07. Mammalian cells contain a 14-kDa protein, which is 20% identical to Trx in amino acid sequence. This Trx-related protein, named TRP14, contains a WC XXC motif, which is conserved among the members of the thiol-disulfide oxidoreductase superfamily that includes Trx, Grx, and protein disulfide isomerases. The reduction potentials of TRP14 (−0.257 V) and Trx (−0.274 V) are similar, and oxidized TRP14 and Trx were reduced with equal efficiency by TrxR. 2W. Jeong and S. G. Rhee, unpublished data. As such, we tested whether TRP14 can reactivate oxidized PTEN in the presence of TrxR and NADPH. TRP14 was nearly as ineffective as GSH, further suggesting the specificity of electron donor molecules (Fig. 7). To evaluate the relative importance of GSH and Trx with regard to reactivation of PTEN in cells, we incubated HeLa cells with either 2,4-dinitro-1-chlorobenzene (DNCB), which prevents recycling of Trx by inhibiting TrxR activity (25Nordberg J. Zhong L. Holmgren A. Arner E.S. J. Biol. Chem. 1998; 273: 10835-10842Abstract Full Text Full Text PDF PubMed Scopus (176) Google Scholar), or buthioninesulfoximine (BSO), which inhibits GSH biosynthesis (26Griffith O.W. Meister A. J. Biol. Chem. 1979; 254: 7558-7560Abstract Full Text PDF PubMed Google Scholar), before exposure to H2O2. Treatment of cells with 100 μm DNCB for 15 min resulted in the accumulation of >80% of Trx in its oxidized form, and treatment with 300 μmBSO for 24 h resulted in the depletion of GSH by >90% (data not shown). Whereas BSO did not substantially affect the reduction of H2O2-oxidized PTEN, DNCB induced a marked delay in PTEN reduction (Fig. 8), suggesting that the reduction of PTEN in cells is mediated predominantly by Trx. DNCB is known to be toxic and induce cell death upon incubation for several hours (27Ishikawa A. Kubota Y. Murayama T. Nomura Y. Neurosci. Lett. 1999; 277: 99-102Crossref PubMed Scopus (24) Google Scholar). Incubation of HeLa cells with 100 μmDNCB for 15 min, however, did not cause any detectable cell death as measured by the trypan blue exclusion assay, and detectable levels of cell death were apparent after incubation for 90 min (data not shown). A variety of cellular stimuli induce the activation of phosphoinositide 3-kinase, resulting in the conversion of PI(4,5)P2 to PI(3,4,5)P3 (28Fruman D.A. Meyers R.E. Cantley L.C. Annu. Rev. Biochem. 1998; 67: 481-507Crossref PubMed Scopus (1328) Google Scholar, 29Vanhaesebroeck B. Leevers S.J. Ahmadi K. Timms J. Katso R. Driscoll P.C. Woscholski R. Parker P.J. Waterfield M.D. Annu. Rev. Biochem. 2001; 70: 535-602Crossref PubMed Scopus (1383) Google Scholar). The latter lipid is dephosphorylated back to PI(4,5)P2 by PTEN (13Maehama T. Dixon J.E. J. Biol. Chem. 1998; 273: 13375-13378Abstract Full Text Full Text PDF PubMed Scopus (2646) Google Scholar) or to PI(3,4)P2 by 5-phosphatases such as SHIP (30Majerus P.W. Kisseleva M.V. Norris F.A. J. Biol. Chem. 1999; 274: 10669-10672Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar). There are at least nine distinct mammalian 5-phosphatases (31Kisseleva M.V. Wilson M.P. Majerus P.W. J. Biol. Chem. 2000; 275: 20110-20116Abstract Full Text Full Text PDF PubMed Scopus (118) Google Scholar), and the catalytic cores of all of them contain two signature active site motifs, (F/I)W XGD XN(F/Y)R and (R/N)XP(S/A)(W/Y)(C/T)DR; these motifs, unlike the active site motif of the 3-phosphatase PTEN, do not possess a conserved cysteine residue, however. The 3′-phosphorylated phosphoinositides bind and modulate numerous proteins that regulate a variety of cellular functions (15Maehama T. Taylor G.E. Dixon J.E. Annu. Rev. Biochem. 2001; 70: 247-279Crossref PubMed Scopus (411) Google Scholar, 29Vanhaesebroeck B. Leevers S.J. Ahmadi K. Timms J. Katso R. Driscoll P.C. Woscholski R. Parker P.J. Waterfield M.D. Annu. Rev. Biochem. 2001; 70: 535-602Crossref PubMed Scopus (1383) Google Scholar). One of the best-characterized targets of PI(3,4,5)P3 and PI(3,4)P2 is the protein kinase Akt. Both PI(3,4,5)P3 and PI(3,4)P2 activate Akt (20Cantley L.C. Neel B.G. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 4240-4245Crossref PubMed Scopus (1768) Google Scholar), although their relative contribution to this process in vivo remains unclear (32Kisseleva M.V. Cao L. Majerus P.W. J. Biol. Chem. 2002; 277: 6266-6272Abstract Full Text Full Text PDF PubMed Scopus (85) Google Scholar). Activated Akt promotes cell survival by phosphorylating the proapoptotic factors BAD and caspase-9 and thereby inhibiting their functions (33Datta S.R. Dudek H. Tao X. Masters S., Fu, H. Gotoh Y. Greenberg M.E. Cell. 1997; 91: 231-241Abstract Full Text Full Text PDF PubMed Scopus (4999) Google Scholar, 34Cardone M.H. Roy N. Stennicke H.R. Salvesen G.S. Franke T.F. Stanbridge E. Frisch S. Reed J.C. Science. 1998; 282: 1318-1321Crossref PubMed Scopus (2748) Google Scholar). Unchecked activation of Akt may thus lead to tumor formation (17Stambolic V. Suzuki A. de la Pompa J.L. Brothers G.M. Mirtsos C. Sasaki T. Ruland J. Penninger J.M. Siderovski D.P. Mak T.W. Cell. 1998; 95: 29-39Abstract Full Text Full Text PDF PubMed Scopus (2146) Google Scholar, 35Wu X. Senechal K. Neshat M.S. Whang Y.E. Sawyers C.L. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 15587-15591Crossref PubMed Scopus (598) Google Scholar). PTEN opposes this proliferation signaling pathway by reducing the cellular abundance of PI(3,4,5)P3 and PI(3,4)P2 and blocking Akt activation (17Stambolic V. Suzuki A. de la Pompa J.L. Brothers G.M. Mirtsos C. Sasaki T. Ruland J. Penninger J.M. Siderovski D.P. Mak T.W. Cell. 1998; 95: 29-39Abstract Full Text Full Text PDF PubMed Scopus (2146) Google Scholar, 35Wu X. Senechal K. Neshat M.S. Whang Y.E. Sawyers C.L. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 15587-15591Crossref PubMed Scopus (598) Google Scholar). Exogenous H2O2 induces the activation of Akt (6Ushio-Fukai M. Alexander R.W. Akers M. Yin Q. Fujio Y. Walsh K. Griendling K.K. J. Biol. Chem. 1999; 274: 22699-22704Abstract Full Text Full Text PDF PubMed Scopus (501) Google Scholar, 36Konishi H. Matsuzaki H. Tanaka M. Takemura Y. Kuroda S. Ono Y. Kikkawa U. FEBS Lett. 1997; 410: 493-498Crossref PubMed Scopus (240) Google Scholar, 37Wang X. McCullough K.D. Franke T.F. Holbrook N.J. J. Biol. Chem. 2000; 275: 14624-14631Abstract Full Text Full Text PDF PubMed Scopus (408) Google Scholar). However, the 3′-phosphorylated phosphoinositide that accumulates in response to H2O2 is almost exclusively PI(3,4)P2 rather than PI(3,4,5)P3(24Holmgren A. Antioxid. Redox Signal. 2000; 2: 811-820Crossref PubMed Scopus (434) Google Scholar, 38Van der Kaay J. Beck M. Gray A. Downes C.P. J. Biol. Chem. 1999; 274: 35963-35968Abstract Full Text Full Text PDF PubMed Scopus (107) Google Scholar). Given that H2O2 activates receptor-type protein tyrosine kinases such as the PDGF receptor and the epidermal growth factor receptor (37Wang X. McCullough K.D. Franke T.F. Holbrook N.J. J. Biol. Chem. 2000; 275: 14624-14631Abstract Full Text Full Text PDF PubMed Scopus (408) Google Scholar, 39Knebel A. Rahmsdorf H.J. Ullrich A. Herrlich P. EMBO J. 1996; 15: 5314-5325Crossref PubMed Scopus (470) Google Scholar, 40Goldkorn T. Balaban N. Matsukuma K. Chea V. Gould R. Last J. Chan C. Chavez C. Am. J. Respir. Cell Mol. Biol. 1998; 19: 786-798Crossref PubMed Scopus (176) Google Scholar), it also likely activates phosphoinositide 3-kinase, resulting in the production of PI(3,4,5)P3. In support of this notion, H2O2-dependent activation of Akt is blocked by the phosphoinositide 3-kinase inhibitor wortmannin (6Ushio-Fukai M. Alexander R.W. Akers M. Yin Q. Fujio Y. Walsh K. Griendling K.K. J. Biol. Chem. 1999; 274: 22699-22704Abstract Full Text Full Text PDF PubMed Scopus (501) Google Scholar, 37Wang X. McCullough K.D. Franke T.F. Holbrook N.J. J. Biol. Chem. 2000; 275: 14624-14631Abstract Full Text Full Text PDF PubMed Scopus (408) Google Scholar,41Salsman S. Felts N. Pye Q.N. Floyd R.A. Hensley K. Arch. Biochem. Biophys. 2001; 386: 275-280Crossref PubMed Scopus (22) Google Scholar). We have now shown that PTEN is inactivated by H2O2. This observation, together with the fact that PTEN, but not 5-phosphatases, contains an oxidizable cysteine residue at its active site, suggests that in cells exposed to H2O2, PI(3,4,5)P3 synthesized by phosphoinositide 3-kinase activity is rapidly dephosphorylated by 5-phosphatases that are insensitive to H2O2, whereas dephosphorylation at the 3′ position is markedly inhibited as a result of inactivation of PTEN by H2O2. The H2O2-dependent inhibition of the lipid phosphatase activity of the tumor suppressor PTEN suggests that H2O2 produced under pathological conditions, such as during chronic inflammation, might contribute to inhibition of apoptosis, to hyperplasia, and to tumor formation. Indeed, many normal cells exposed to H2O2exhibit increased proliferation and express growth-related genes (42Burdon R.H. Free Radic. Biol. Med. 1995; 18: 775-794Crossref PubMed Scopus (1075) Google Scholar, 43Nose K. Shibanuma M. Kikuchi K. Kageyama H. Sakiyama S. Kuroki T. Eur. J. Biochem. 1991; 201: 99-106Crossref PubMed Scopus (275) Google Scholar, 44Datta R. Hallahan D.E. Kharbanda S.M. Rubin E. Sherman M.L. Huberman E. Weichselbaum R.R. Kufe D.W. Biochemistry. 1992; 31: 8300-8306Crossref PubMed Scopus (157) Google Scholar). Furthermore, given that tumor cells produce large amounts of H2O2 (45Szatrowski T.P. Nathan C.F. Cancer Res. 1991; 51: 794-798PubMed Google Scholar, 46Irani K. Xia Y. Zweier J.L. Sollott S.J. Der C.J. Fearon E.R. Sundaresan M. Finkel T. Goldschmidt-Clermont P.J. Science. 1997; 275: 1649-1652Crossref PubMed Scopus (1449) Google Scholar), this metabolite might also promote tumor cell proliferation. Previously, the tumor-promoting activity of reactive oxygen species has been explained mainly in terms of the mutagenic activity of the hydroxyl radicals derived from them. A variety of cell surface receptor ligands induce the production of H2O2, which is necessary for receptor-mediated cellular events such as chemotaxis, DNA synthesis, and cell cycle progress (3Sundaresan M. Yu Z.X. Ferrans V.J. Irani K. Finkel T. Science. 1995; 270: 296-299Crossref PubMed Scopus (2342) Google Scholar, 47Page K., Li, J. Wang Y. Kartha S. Pestell R.G. Hershenson M.B. Am. J. Respir. Cell Mol. Biol. 2000; 23: 436-443Crossref PubMed Scopus (60) Google Scholar). Like exogenous H2O2, the receptor-produced H2O2 also likely causes PTEN inactivation and contributes to the accumulation of 3′-phosphorylated phosphoinositides. Thus, the receptor-mediated activation of phosphoinositide 3-kinase may not be sufficient for the accumulation of 3′-phosphorylated phosphoinositides, and the concomitant inactivation of PTEN by receptor-produced H2O2 might also be necessary for this effect. Despite numerous studies demonstrating the importance of PTEN in cell growth, survival, and migration, the mechanisms by which the activity of this protein is regulated have remained unclear (15Maehama T. Taylor G.E. Dixon J.E. Annu. Rev. Biochem. 2001; 70: 247-279Crossref PubMed Scopus (411) Google Scholar). PTEN contains tandem phosphatase and C2 domains, which share an extensive interface (23Lee J.O. Yang H. Georgescu M.M., Di Cristofano A. Maehama T. Shi Y. Dixon J.E. Pandolfi P. Pavletich N.P. Cell. 1999; 99: 323-334Abstract Full Text Full Text PDF PubMed Scopus (894) Google Scholar). The phosphatase domain exhibits substantial structural homology to protein tyrosine phosphatases, but it contains an enlarged active site with several basic residues that are important for the accommodation of the bulky inositol trisphosphate moiety. PTEN also contains a putative PDZ domain-binding motif at its COOH terminus. Although this notion remains controversial (48Georgescu M.M. Kirsch K.H. Akagi T. Shishido T. Hanafusa H. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 10182-10187Crossref PubMed Scopus (281) Google Scholar, 49Tolkacheva T. Chan A.M. Oncogene. 2000; 19: 680-689Crossref PubMed Scopus (63) Google Scholar, 50Leslie N.R. Gray A. Pass I. Orchiston E.A. Downes C.P. Biochem. J. 2000; 346: 827-833Crossref PubMed Scopus (72) Google Scholar), the C2 domain and the PDZ domain-binding motif have been proposed to mediate the translocation of PTEN from the cytosol to sites near to PI(3,4,5)P3 (23Lee J.O. Yang H. Georgescu M.M., Di Cristofano A. Maehama T. Shi Y. Dixon J.E. Pandolfi P. Pavletich N.P. Cell. 1999; 99: 323-334Abstract Full Text Full Text PDF PubMed Scopus (894) Google Scholar, 51Wu X. Hepner K. Castelino-Prabhu S., Do, D. Kaye M.B. Yuan X.J. Wood J. Ross C. Sawyers C.L. Whang Y.E. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 4233-4238Crossref PubMed Scopus (341) Google Scholar, 52Wu Y. Dowbenko D. Spencer S. Laura R. Lee J., Gu, Q. Lasky L.A. J. Biol. Chem. 2000; 275: 21477-21485Abstract Full Text Full Text PDF PubMed Scopus (233) Google Scholar, 53Adey N.B. Huang L. Ormonde P.A. Baumgard M.L. Pero R. Byreddy D.V. Tavtigian S.V. Bartel P.L. Cancer Res. 2000; 60: 35-37PubMed Google Scholar). Phosphorylation of PTEN by casein kinase II has also been shown to increase the susceptibility of the protein to proteasome-mediated degradation (54Torres J. Pulido R. J. Biol. Chem. 2001; 276: 993-998Abstract Full Text Full Text PDF PubMed Scopus (544) Google Scholar). How these modes of regulation might be affected by the activation of cell surface receptors, however, is not known. Although in vitro measurements of its lipid phosphatase activity suggest that PTEN might be constitutively active, no mechanism for its negative regulation has previously been revealed (15Maehama T. Taylor G.E. Dixon J.E. Annu. Rev. Biochem. 2001; 70: 247-279Crossref PubMed Scopus (411) Google Scholar). Constitutive activity of PTEN is likely required to prevent accumulation of PI(3,4,5)P3 and unwanted triggering of proliferation signals in quiescent cells. The oxidative inactivation of PTEN in response to H2O2 generation might therefore be an important determinant of the timing and localization of the production of this important lipid messenger elicited by receptor stimulation. Oxidized PTEN is gradually converted back to the reduced form by intracellular reducing agents, indicating that the reversible oxidation and reduction of the active site cysteine residue is an important mechanism for the regulation of PTEN. The best candidate for the intracellular reducing agent is Trx, given that oxidized PTEN was reduced much more effectively by Trx than by Grx or GSH in vitro. Furthermore, depletion of reduced Trx, but not of GSH, markedly reduced the rate of reactivation of oxidized PTEN in cells. The association of Trx with PTEN revealed by the co-immunoprecipitation of these proteins suggests that the reduction reaction might be governed by a specific interaction between oxidized PTEN and reduced thioredoxin. Such a specific interaction is also supported by the observation that a 14-kDa Trx-related protein (TRP14) was not able to reduce oxidized PTEN despite the fact that it contains a WC XXC motif, which is conserved among members of the thiol-disulfide oxidoreductase superfamily, and that TRP14 and Trx exhibit similar redox potentials and reactivity toward TrxR. We thank K. Yamada for PTEN cDNA cloned in the pQE30 vector.