The ARG Tyrosine Kinase Interacts with Siva-1 in the Apoptotic Response to Oxidative Stress
2001; Elsevier BV; Volume: 276; Issue: 15 Linguagem: Inglês
10.1074/jbc.c100050200
ISSN1083-351X
AutoresCheng Cao, Xinping Ren, Surender Kharbanda, Anthony J. Koleske, Kartik Prasad, Donald Küfe,
Tópico(s)Kruppel-like factors research
ResumoThe Abl family of mammalian nonreceptor tyrosine kinases consists of c-Abl and ARG (Abl-related gene). Certain insights are available regarding the involvement c-Abl in the response of cells to stress. ARG, however, has no known function in cell signaling. The present studies demonstrate that ARG associates with the proapoptotic Siva-1 protein. The functional significance of the ARG-Siva-1 interaction is supported by the finding that ARG is activated by oxidative stress and that this response involves ARG-mediated phosphorylation of Siva-1 on Tyr48. The proapoptotic effects of Siva-1 are accentuated in cells stably expressing ARG and are inhibited in ARG-deficient cells. Moreover, the proapoptotic effects of Siva-1 are abrogated by mutation of the Tyr48site. We also show that the apoptotic response to oxidative stress is attenuated in ARG-deficient cells and that this defect is corrected by reconstituting ARG expression. These findings support a model in which the activation of ARG by oxidative stress induces apoptosis by a Siva-1-dependent mechanism. The Abl family of mammalian nonreceptor tyrosine kinases consists of c-Abl and ARG (Abl-related gene). Certain insights are available regarding the involvement c-Abl in the response of cells to stress. ARG, however, has no known function in cell signaling. The present studies demonstrate that ARG associates with the proapoptotic Siva-1 protein. The functional significance of the ARG-Siva-1 interaction is supported by the finding that ARG is activated by oxidative stress and that this response involves ARG-mediated phosphorylation of Siva-1 on Tyr48. The proapoptotic effects of Siva-1 are accentuated in cells stably expressing ARG and are inhibited in ARG-deficient cells. Moreover, the proapoptotic effects of Siva-1 are abrogated by mutation of the Tyr48site. We also show that the apoptotic response to oxidative stress is attenuated in ARG-deficient cells and that this defect is corrected by reconstituting ARG expression. These findings support a model in which the activation of ARG by oxidative stress induces apoptosis by a Siva-1-dependent mechanism. The mammalian c-Abl and ARG 1The abbreviations used are:ARGAbl-related geneDNA-PKDNA-dependent protein kinaseROSreactive oxygen speciesPKCprotein kinase CGSTglutathioneS-transferaseGFPgreen fluorescence proteinPAGEpolyacrylamide gel electrophoresisMEFmouse embryo fibroblastSH2/SH3Src homology 2 and 3 (domains) nonreceptor tyrosine kinases are ubiquitously expressed in adult tissues (1Goff S.P. Gilboa E. Witte O.N. Baltimore D. Cell. 1980; 22: 777-785Abstract Full Text PDF PubMed Scopus (279) Google Scholar, 2Kruh G.D. Perego R. Miki T. Aaronson S.A. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 5802-5806Crossref PubMed Scopus (135) Google Scholar). These proteins contain N-terminal SH3, SH2, and kinase domains that share ∼90% identity. The C-terminal regions of c-Abl and ARG share 29% identity and are distinguished from other nonreceptor tyrosine kinases by the presence of globular (G) and filamentous (F) actin-binding domains (3Van Etten R.A. Jackson P.K. Baltimore D. Sanders M.C. Matsuddaira P.T. Janmey P.A. J. Cell Biol. 1994; 124: 325-340Crossref PubMed Scopus (237) Google Scholar). The c-Abl protein is expressed in the nucleus and cytoplasm, whereas ARG has been detected predominately in the cytoplasm (4Wang B. Kruh G.D. Oncogene. 1996; 13: 193-197PubMed Google Scholar). In addition, the C-terminal region of c-Abl differs from ARG by the presence of a nuclear localization signal (5Van Etten R.A. Jackson P. Baltimore D. Cell. 1989; 58: 669-678Abstract Full Text PDF PubMed Scopus (335) Google Scholar), sites for phosphorylation by the Cdc2 kinase (6Kipreos E.T. Wang J.Y. Science. 1990; 248: 217-220Crossref PubMed Scopus (118) Google Scholar), and DNA binding sequences (7Kipreos E.T. Wang J.Y. Science. 1992; 256: 382-385Crossref PubMed Scopus (177) Google Scholar). The structural differences of the C-terminal regions have suggested that c-Abl and ARG may share only certain cellular functions.Mice with targeted disruption of the c-abl gene are born runted with head and eye abnormalities and succumb as neonates to defective lymphopoiesis (8Tybulewicz V.L.J. Crawford C.E. Jackson P.K. Bronson R.T. Mulligan R.C. Cell. 1991; 65: 1153-1163Abstract Full Text PDF PubMed Scopus (1158) Google Scholar, 9Schwartzberg P.L. Stall A.M. Hardin J.D. Bowdish K.S. Humaran T. Boast S. Harbison M.L. Robertson E.J. Goff S.P. Cell. 1991; 65: 1165-1175Abstract Full Text PDF PubMed Scopus (299) Google Scholar). Mice deficient in ARG develop normally but exhibit behavioral abnormalities (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar). Embryos deficient in both c-Abl and ARG exhibit defects in neurolation and die before 11 days postcoitum (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar). These findings and the observation thatabl−/−,arg−/− neuroepithelial cells exhibit an altered actin cytoskeleton have supported roles for c-Abl and ARG in the regulation of actin microfilaments (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar).Other studies have demonstrated that c-Abl is involved in the cellular response to stress (11Kharbanda S. Ren R. Pandey P. Shafman T.D. Feller S.M. Weichselbaum R.R. Kufe D.W. Nature. 1995; 376: 785-788Crossref PubMed Scopus (458) Google Scholar). Nuclear c-Abl associates with the DNA-dependent protein kinase (DNA-PK) complex (12Kharbanda S. Pandey P. Jin S. Inoue S. Bharti A. Yuan Z.-M. Weichselbaum R. Weaver D. Kufe D. Nature. 1997; 386: 732-735Crossref PubMed Scopus (237) Google Scholar, 13Jin S. Kharbanda S. Mayer B. Kufe D. Weaver D.T. J. Biol. Chem. 1997; 272: 24763-24766Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar) and with the product of the gene mutated in ataxia telangiectasia (14Shafman T. Khanna K.K. Kedar P. Spring K. Kozlov S. Yen T. Hobson K. Gatei M. Zhang N. Watters D. Egerton M. Shiloh Y. Kharbanda S. Kufe D. Lavin M.F. Nature. 1997; 387: 520-523Crossref PubMed Scopus (419) Google Scholar, 15Baskaran R. Wood L.D. Whitaker L.L. Xu Y. Barlow C. Canman C.E. Morgan S.E. Baltimore D. Wynshaw-Boris A. Kastan M.B. Wang J.Y.J. Nature. 1997; 387: 516-519Crossref PubMed Scopus (485) Google Scholar). Activation of c-Abl by DNA-PK and ataxia telangiectasia mutated gene product in cells exposed to genotoxic agents contributes to DNA damage-induced apoptosis by mechanisms in part dependent on p53 and its homolog p73 (11Kharbanda S. Ren R. Pandey P. Shafman T.D. Feller S.M. Weichselbaum R.R. Kufe D.W. Nature. 1995; 376: 785-788Crossref PubMed Scopus (458) Google Scholar, 16Yuan Z.M. Huang Y. Whang Y. Sawyers C. Weichselbaum R. Kharbanda S. Kufe D. Nature. 1996; 382: 272-274Crossref PubMed Scopus (209) Google Scholar, 17Yuan Z.M. Shioya H. Ishiko T. Sun X. Huang Y. Lu H. Kharbanda S. Weichselbaum R. Kufe D. Nature. 1999; 399: 814-817Crossref PubMed Scopus (539) Google Scholar, 18Gong J. Costanzo A. Yang H. Melino G. Kaelin JR W. Levrero M. Wang J.Y.J. Nature. 1999; 399: 806-809Crossref PubMed Scopus (830) Google Scholar, 19Agami R. Blandino G. Oren M. Shaul Y. Nature. 1999; 399: 809-813Crossref PubMed Scopus (504) Google Scholar). In the cellular response to reactive oxygen species (ROS), the cytoplasmic form of c-Abl is activated by protein kinase C δ (PKCδ) (20Sun X. Wu F. Datta R. Kharbanda S. Kufe D. J. Biol. Chem. 2000; 275: 7470-7473Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar). Activation of cytoplasmic c-Abl by ROS transduces signals that induce release of mitochondrial cytochromec and thereby apoptosis (21Sun X. Majumder P. Shioya H. Wu F. Kumar S. Weichselbaum R. Kharbanda S. Kufe D. J. Biol. Chem. 2000; 275: 17237-17240Abstract Full Text Full Text PDF PubMed Scopus (143) Google Scholar).No functional role has been ascribed to ARG as a cell signaling molecule. The present studies demonstrate that ARG interacts with the proapoptotic Siva-1 protein (22Prasad K.V. Ao Z. Yoon Y. Wu M.X. Rizk M. Jacquot S. Schlossman S.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6346-6351Crossref PubMed Scopus (250) Google Scholar, 23Yoon Y. Ao Z. Cheng Y. Schlossman S.F. Prasad K.V. Oncogene. 1999; 18: 7174-7179Crossref PubMed Scopus (44) Google Scholar). ARG phosphorylates Siva-1 in the cellular response to oxidative stress and induces apoptosis by a Siva-1-dependent mechanism.RESULTS AND DISCUSSIONTo extend the findings in the yeast two-hybrid system that ARG associates with the pro-apoptotic Siva-1 protein (data not shown), lysates from MCF-7 cells expressing Flag-tagged ARG and GFP-tagged human Siva-1 were subjected to immunoprecipitation with anti-Flag. Analysis of the precipitates by immunoblotting with anti-GFP demonstrated the presence of ARG-Siva-1 complexes (Fig.1A, left). Analysis of anti-Flag immunoprecipitates from cells expressing Flag-Siva-1 by immunoblotting with anti-ARG provided further support for binding of ARG and Siva-1 (Fig. 1A, right). To extend these findings, lysates from cells expressing Flag-ARG were incubated with a GST-Siva-1 fusion protein. Analysis of the adsorbates with anti-Flag confirmed the binding of ARG and Siva-1 (Fig. 1B). Other studies with GST fusion proteins prepared from the ARG SH2 and ARG SH3 domains demonstrated that both confer binding to Siva-1 (Fig. 1C). By contrast, binding of GST-ARG SH2, but not GST-ARG SH3, was detectable to the shorter, nonapoptotic Siva-2 protein (Fig.1D). Human Siva-1, but not Siva-2, contains a proline-rich PESP sequence (amino acids 82–85) for potential binding to ARG SH3. Notably, however, the PESP site is not conserved in mouse Siva-1 (22Prasad K.V. Ao Z. Yoon Y. Wu M.X. Rizk M. Jacquot S. Schlossman S.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6346-6351Crossref PubMed Scopus (250) Google Scholar,23Yoon Y. Ao Z. Cheng Y. Schlossman S.F. Prasad K.V. Oncogene. 1999; 18: 7174-7179Crossref PubMed Scopus (44) Google Scholar). In this regard, GST-ARG SH2, and not GST-ARG SH3, binds to mouse Siva-1 (data not shown). These findings demonstrate that the ARG SH2 domain interacts with Siva-1 and Siva-2 of both human and mouse origin and that binding of the ARG SH3 domain is also detectable with human Siva-1.To determine whether Siva-1 is a substrate for ARG, GST-Siva-1 was incubated with kinase-active ARG (Fig.2A, left) in the presence of [γ-32P]ATP. Analysis of the reaction products by SDS-PAGE and autoradiography demonstrated phosphorylation of Siva-1 (Fig. 2A, right). As a control, there was no detectable phosphorylation when Siva-1 was incubated with kinase-inactive ARG(K-R) in which Lys337 in the ATP binding site was mutated to Arg (Fig. 2A). There are two potential tyrosine phosphorylation sites in Siva-1 that are located at Tyr48 and Tyr67. Mutation of these sites to Phe and then incubation of the mutant proteins with ARG demonstrated abrogation of phosphorylation with GST-Siva-1(Y48F), but not with GST-Siva-1(Y67F) (Fig. 2A, right). To assess whether ARG phosphorylates Siva-1 in vivo, Flag-Siva-1 was coexpressed with ARG and lysates were analyzed by immunoblotting with anti-P-Tyr. The results show that Siva-1 is phosphorylated by ARG in cells (Fig.2B). By contrast, there was no detectable tyrosine phosphorylation of Flag-Siva-1 when this vector was coexpressed with ARG(K-R) (data not shown). Although Flag-Siva-1(Y67F) was also phosphorylated by ARG, there was no detectable tyrosine phosphorylation of Flag-Siva-1(Y48F) (Fig. 2B). In concert with these findings and the presence of Tyr48 in Siva-2, coexpression of Siva-2 with ARG, but not ARG(K-R), also resulted in detectable tyrosine phosphorylation (Fig. 2B). These findings thus provided support for ARG-mediated phosphorylation of Siva-1 and Siva-2 on Tyr48in vitro and in vivo. To determine whether ARG, like c-Abl (20Sun X. Wu F. Datta R. Kharbanda S. Kufe D. J. Biol. Chem. 2000; 275: 7470-7473Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar, 21Sun X. Majumder P. Shioya H. Wu F. Kumar S. Weichselbaum R. Kharbanda S. Kufe D. J. Biol. Chem. 2000; 275: 17237-17240Abstract Full Text Full Text PDF PubMed Scopus (143) Google Scholar), is activated by ROS, cells expressing Flag-ARG were treated with H2O2. Analysis of anti-Flag immunoprecipitates for phosphorylation of GST-Siva-1 demonstrated H2O2concentration-dependent induction of ARG activity (Fig.2C). As a control, the same anti-Flag immunoprecipitates failed to phosphorylate GST-Siva-1(Y48F) (Fig. 2C). In studies of wild-type and arg−/−MEFs, expression of Flag-Siva-1 resulted in little if any detectable phosphorylation (Fig. 2D). Treatment of the wild-type cells with H2O2, however, was associated with tyrosine phosphorylation of Flag-Siva-1 (Fig. 2D). By contrast, there was no detectable phosphorylation of Flag-Siva-1 in H2O2-treatedarg−/− MEFs (Fig. 2D). These findings demonstrate that activation of ARG by H2O2 is associated with phosphorylation of Siva-1 on Tyr48.Figure 2ARGphosphorylation of Siva-1 in response to oxidative stress.A, Flag-ARG and Flag-ARG(K-R) (left panel) were incubated with GST-Siva-1, GST-Siva-1(Y67F), or GST-Siva-1(Y48F). GST-Crk-(120–225) was used as a positive control. Reaction products were analyzed by SDS-PAGE and autoradiography (right panel). B, 293 cells were cotransfected with ARG and Flag-Siva-1, Flag-Siva-1(Y48F), Flag-Siva-1(Y67F), or Flag-Siva-2. Lysates were subjected to immunoblotting (IB) with anti-P-Tyr and anti-Flag.C, 293 cells were transfected to express Flag-ARG. At 36 h after transfection, the cells were treated with the indicated concentrations of H2O2 for 2 h. Anti-Flag immunoprecipitates (IP) were subjected to ARG kinase assays using Siva-1 or Siva-1(Y48F) as substrate. D, wild-type (arg+/+) andarg−/− MEFs expressing Flag-Siva-1 were treated with 1 mm H2O2 for 2 h. Anti-Flag immunoprecipitates were analyzed by immunoblotting with anti-P-Tyr and anti-Flag.View Large Image Figure ViewerDownload Hi-res image Download (PPT)To assess the functional significance of the ARG-Siva-1 interaction, MCF-7 cells were prepared that stably express ARG or the ARG(K-R) mutant. There was no detectable effect of ARG or ARG(K-R) expression on MCF-7 cell growth (data not shown) or cell cycle distribution (Fig.3A). Expression of Siva-1 in the wild-type MCF-7 cells was associated with the appearance of apoptotic cells containing sub-G1 DNA (Fig. 3A). Whereas Siva-1-induced apoptosis was more pronounced in MCF-7/ARG cells, expression of Siva-1 had little if any effect on MCF-7/ARG(K-R) cells (Fig. 3A). As a control, expression of Siva-2 resulted in substantially less apoptosis in MCF-7/ARG cells as compared with that obtained with Siva-1, and had no effect on the wild-type MCF-7 and MCF-7/ARG(K-R) cells (Fig. 3A). To extend these findings, Siva-1 was expressed in the wild-type andarg−/− MEFs. The results demonstrate that, although Siva-1 induces apoptosis in wild-type cells, there was little effect of Siva-1 in the absence of ARG expression (Fig. 3B). In concert with the finding that ARG phosphorylates Siva-1 on Tyr48, the expression of Siva-1(Y48F) had little effect on the induction of apoptosis, whereas the proapoptotic effects of Siva-1(Y67F) were similar to those obtained with wild-type Siva-1 (Fig. 3C). These findings demonstrate that Siva-1-induced apoptosis is dependent on the ARG kinase function and that ARG-mediated phosphorylation of Siva-1 on Tyr48 is a proapoptotic signal.Figure 3Interaction ofARGand Siva-1 in the apoptotic response to oxidative stress.A, MCF-7, MCF-7/ARG, and MCF-7/ARG(K-R) cells were transfected to express GFP-Siva-1 or GFP-Siva-2. At 36 h after transfection, GFP-positive cells were analyzed for sub-G1 DNA (left panels). Cells with sub-G1 DNA are depicted in theshaded profiles. GO/G1 and G2/M cells are shown in the dark profiles and S phase cells in the hatched profiles. Cells were subjected to immunoblot analysis (IB) with the indicated antibodies (right panels). B, wild-type (arg+/+) andarg−/− MEFs were infected with empty or Siva-1 retroviral vectors for 24 h. The cells were analyzed by flow cytometry (left panels) and immunoblotting (right panels). C, MCF-7 (open bars), MCF-7/ARG (hatched bars), and MCF-7/ARG(K-R) (solid bars) were transfected with GFP-Siva-1, GFP-Siva-2, GFP-Siva-1(Y48F), or GFP-Siva-1(Y67F). GFP-positive cells were analyzed for DNA content. The results are expressed as the percentage (mean ± S.E. of two independent experiments each performed in duplicate) of GFP-positive cells with sub-G1 DNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT)As the findings demonstrate that ARG is activated by ROS, H2O2-induced apoptosis was assessed in the MCF-7, MCF-7/ARG, and MCF-7/ARG(K-R) cells. The results demonstrate that the apoptotic effects of H2O2 are attenuated in MCF-7/ARG(K-R) as compared with wild-type MCF-7 cells (Fig. 4A). Notably, however, there was a marked increase in H2O2-induced apoptosis in MCF-7/ARG cells (Fig. 4A). The apoptotic effects of H2O2 were attenuated inarg−/− (two separate embryos) as compared with wild-type MEFs (Fig. 4B). Moreover, expression of ARG in arg−/− cells corrected the defect in H2O2-induced apoptosis (Fig.4C).Figure 4Apoptotic response to oxidative stress isARGkinase-dependent.A–C, the indicated cells were left untreated (open bars) or treated with 1 mm H2O2 for 2 h (hatched bars), washed, and then cultured for 18 h. In C,arg+/− cells were prepared by transduction ofarg−/− MEFs with a retroviral vector expressing ARG (transduction efficiency, >95% as determined by GFP expression). Cells were analyzed for DNA content. The results are expressed as the percentage (mean ± S.E. of two independent experiments performed in duplicate) of cells with sub-G1DNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT)The ARG tyrosine kinase has, like c-Abl, been associated with the development of leukemia (25Cazzaniga G. Tosi S. Aloisi A. Giudici G. Daniotti M. Pioltelli P. Kearney L. Biondi A. Blood. 1999; 94: 4370-4373Crossref PubMed Google Scholar, 26Iijima Y. Ito T. Oikawa T. Eguchi M. Eguchi-Ishimae M. Kamada N. Kishi K. Asano S. Sakaki Y. Sato Y. Blood. 2000; 95: 2126-2131PubMed Google Scholar). Despite the relatedness to c-Abl and a role in neurolation as defined inarg−/− mice (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar), a function for ARG in cell signaling has remained obscure. Genetic studies inDrosophila have indicated that Abl family kinases regulate cellular morphology through interactions with the cytoskeleton (27Gertler F.B. Bennett R.L. Clark M.J. Hoffmann F.M. Cell. 1989; 58: 103-113Abstract Full Text PDF PubMed Scopus (199) Google Scholar,28Gertler F.B. Hill K.K. Clark M.J. Hoffman F.M. Genes Dev. 1993; 7: 441-453Crossref PubMed Scopus (113) Google Scholar). The identification of actin-binding domains in the C-terminal region of ARG (3Van Etten R.A. Jackson P.K. Baltimore D. Sanders M.C. Matsuddaira P.T. Janmey P.A. J. Cell Biol. 1994; 124: 325-340Crossref PubMed Scopus (237) Google Scholar) and localization of ARG with actin microfilaments (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar) have supported a role in regulation of the actin cytoskeletion. The present results provide evidence for involvement of ARG in the cellular response to oxidative stress. Moreover, the induction of apoptosis by oxidative stress is attenuated in ARG-deficient cells. These findings indicate that, in addition to regulating the actin cytoskeleton, ARG functions in ROS-mediated signals that induce an apoptotic response.Siva-1 interacts with members of the tumor necrosis factor receptor family and induces apoptosis in diverse cells (22Prasad K.V. Ao Z. Yoon Y. Wu M.X. Rizk M. Jacquot S. Schlossman S.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6346-6351Crossref PubMed Scopus (250) Google Scholar). Siva-1 has also been implicated in the induction of Coxsackievirus-induced apoptosis (29Henke A. Launhardt H. Klement K. Stelzner A. Zell R. Munder T. J. Virol. 2000; 74: 4284-4290Crossref PubMed Scopus (115) Google Scholar). Full-length Siva-1 but not Siva-2, which lacks sequences encoded by exon 2, induces the apoptotic response (23Yoon Y. Ao Z. Cheng Y. Schlossman S.F. Prasad K.V. Oncogene. 1999; 18: 7174-7179Crossref PubMed Scopus (44) Google Scholar). The present studies demonstrate that ARG phosphorylates both Siva-1 and Siva-2. By contrast, the interaction between ARG and Siva-1, but not Siva-2, is associated with the induction of apoptosis. Our results also demonstrate that mutation of the Siva-1 Tyr48 site abrogates the apoptotic function of Siva-1 and that apoptosis induced by Siva-1 is dependent on expression of kinase-active ARG. These findings thus define ARG as an upstream effector of Siva-1 in the apoptotic response to oxidative stress. The mammalian c-Abl and ARG 1The abbreviations used are:ARGAbl-related geneDNA-PKDNA-dependent protein kinaseROSreactive oxygen speciesPKCprotein kinase CGSTglutathioneS-transferaseGFPgreen fluorescence proteinPAGEpolyacrylamide gel electrophoresisMEFmouse embryo fibroblastSH2/SH3Src homology 2 and 3 (domains) nonreceptor tyrosine kinases are ubiquitously expressed in adult tissues (1Goff S.P. Gilboa E. Witte O.N. Baltimore D. Cell. 1980; 22: 777-785Abstract Full Text PDF PubMed Scopus (279) Google Scholar, 2Kruh G.D. Perego R. Miki T. Aaronson S.A. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 5802-5806Crossref PubMed Scopus (135) Google Scholar). These proteins contain N-terminal SH3, SH2, and kinase domains that share ∼90% identity. The C-terminal regions of c-Abl and ARG share 29% identity and are distinguished from other nonreceptor tyrosine kinases by the presence of globular (G) and filamentous (F) actin-binding domains (3Van Etten R.A. Jackson P.K. Baltimore D. Sanders M.C. Matsuddaira P.T. Janmey P.A. J. Cell Biol. 1994; 124: 325-340Crossref PubMed Scopus (237) Google Scholar). The c-Abl protein is expressed in the nucleus and cytoplasm, whereas ARG has been detected predominately in the cytoplasm (4Wang B. Kruh G.D. Oncogene. 1996; 13: 193-197PubMed Google Scholar). In addition, the C-terminal region of c-Abl differs from ARG by the presence of a nuclear localization signal (5Van Etten R.A. Jackson P. Baltimore D. Cell. 1989; 58: 669-678Abstract Full Text PDF PubMed Scopus (335) Google Scholar), sites for phosphorylation by the Cdc2 kinase (6Kipreos E.T. Wang J.Y. Science. 1990; 248: 217-220Crossref PubMed Scopus (118) Google Scholar), and DNA binding sequences (7Kipreos E.T. Wang J.Y. Science. 1992; 256: 382-385Crossref PubMed Scopus (177) Google Scholar). The structural differences of the C-terminal regions have suggested that c-Abl and ARG may share only certain cellular functions. Abl-related gene DNA-dependent protein kinase reactive oxygen species protein kinase C glutathioneS-transferase green fluorescence protein polyacrylamide gel electrophoresis mouse embryo fibroblast Src homology 2 and 3 (domains) Mice with targeted disruption of the c-abl gene are born runted with head and eye abnormalities and succumb as neonates to defective lymphopoiesis (8Tybulewicz V.L.J. Crawford C.E. Jackson P.K. Bronson R.T. Mulligan R.C. Cell. 1991; 65: 1153-1163Abstract Full Text PDF PubMed Scopus (1158) Google Scholar, 9Schwartzberg P.L. Stall A.M. Hardin J.D. Bowdish K.S. Humaran T. Boast S. Harbison M.L. Robertson E.J. Goff S.P. Cell. 1991; 65: 1165-1175Abstract Full Text PDF PubMed Scopus (299) Google Scholar). Mice deficient in ARG develop normally but exhibit behavioral abnormalities (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar). Embryos deficient in both c-Abl and ARG exhibit defects in neurolation and die before 11 days postcoitum (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar). These findings and the observation thatabl−/−,arg−/− neuroepithelial cells exhibit an altered actin cytoskeleton have supported roles for c-Abl and ARG in the regulation of actin microfilaments (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar). Other studies have demonstrated that c-Abl is involved in the cellular response to stress (11Kharbanda S. Ren R. Pandey P. Shafman T.D. Feller S.M. Weichselbaum R.R. Kufe D.W. Nature. 1995; 376: 785-788Crossref PubMed Scopus (458) Google Scholar). Nuclear c-Abl associates with the DNA-dependent protein kinase (DNA-PK) complex (12Kharbanda S. Pandey P. Jin S. Inoue S. Bharti A. Yuan Z.-M. Weichselbaum R. Weaver D. Kufe D. Nature. 1997; 386: 732-735Crossref PubMed Scopus (237) Google Scholar, 13Jin S. Kharbanda S. Mayer B. Kufe D. Weaver D.T. J. Biol. Chem. 1997; 272: 24763-24766Abstract Full Text Full Text PDF PubMed Scopus (91) Google Scholar) and with the product of the gene mutated in ataxia telangiectasia (14Shafman T. Khanna K.K. Kedar P. Spring K. Kozlov S. Yen T. Hobson K. Gatei M. Zhang N. Watters D. Egerton M. Shiloh Y. Kharbanda S. Kufe D. Lavin M.F. Nature. 1997; 387: 520-523Crossref PubMed Scopus (419) Google Scholar, 15Baskaran R. Wood L.D. Whitaker L.L. Xu Y. Barlow C. Canman C.E. Morgan S.E. Baltimore D. Wynshaw-Boris A. Kastan M.B. Wang J.Y.J. Nature. 1997; 387: 516-519Crossref PubMed Scopus (485) Google Scholar). Activation of c-Abl by DNA-PK and ataxia telangiectasia mutated gene product in cells exposed to genotoxic agents contributes to DNA damage-induced apoptosis by mechanisms in part dependent on p53 and its homolog p73 (11Kharbanda S. Ren R. Pandey P. Shafman T.D. Feller S.M. Weichselbaum R.R. Kufe D.W. Nature. 1995; 376: 785-788Crossref PubMed Scopus (458) Google Scholar, 16Yuan Z.M. Huang Y. Whang Y. Sawyers C. Weichselbaum R. Kharbanda S. Kufe D. Nature. 1996; 382: 272-274Crossref PubMed Scopus (209) Google Scholar, 17Yuan Z.M. Shioya H. Ishiko T. Sun X. Huang Y. Lu H. Kharbanda S. Weichselbaum R. Kufe D. Nature. 1999; 399: 814-817Crossref PubMed Scopus (539) Google Scholar, 18Gong J. Costanzo A. Yang H. Melino G. Kaelin JR W. Levrero M. Wang J.Y.J. Nature. 1999; 399: 806-809Crossref PubMed Scopus (830) Google Scholar, 19Agami R. Blandino G. Oren M. Shaul Y. Nature. 1999; 399: 809-813Crossref PubMed Scopus (504) Google Scholar). In the cellular response to reactive oxygen species (ROS), the cytoplasmic form of c-Abl is activated by protein kinase C δ (PKCδ) (20Sun X. Wu F. Datta R. Kharbanda S. Kufe D. J. Biol. Chem. 2000; 275: 7470-7473Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar). Activation of cytoplasmic c-Abl by ROS transduces signals that induce release of mitochondrial cytochromec and thereby apoptosis (21Sun X. Majumder P. Shioya H. Wu F. Kumar S. Weichselbaum R. Kharbanda S. Kufe D. J. Biol. Chem. 2000; 275: 17237-17240Abstract Full Text Full Text PDF PubMed Scopus (143) Google Scholar). No functional role has been ascribed to ARG as a cell signaling molecule. The present studies demonstrate that ARG interacts with the proapoptotic Siva-1 protein (22Prasad K.V. Ao Z. Yoon Y. Wu M.X. Rizk M. Jacquot S. Schlossman S.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6346-6351Crossref PubMed Scopus (250) Google Scholar, 23Yoon Y. Ao Z. Cheng Y. Schlossman S.F. Prasad K.V. Oncogene. 1999; 18: 7174-7179Crossref PubMed Scopus (44) Google Scholar). ARG phosphorylates Siva-1 in the cellular response to oxidative stress and induces apoptosis by a Siva-1-dependent mechanism. RESULTS AND DISCUSSIONTo extend the findings in the yeast two-hybrid system that ARG associates with the pro-apoptotic Siva-1 protein (data not shown), lysates from MCF-7 cells expressing Flag-tagged ARG and GFP-tagged human Siva-1 were subjected to immunoprecipitation with anti-Flag. Analysis of the precipitates by immunoblotting with anti-GFP demonstrated the presence of ARG-Siva-1 complexes (Fig.1A, left). Analysis of anti-Flag immunoprecipitates from cells expressing Flag-Siva-1 by immunoblotting with anti-ARG provided further support for binding of ARG and Siva-1 (Fig. 1A, right). To extend these findings, lysates from cells expressing Flag-ARG were incubated with a GST-Siva-1 fusion protein. Analysis of the adsorbates with anti-Flag confirmed the binding of ARG and Siva-1 (Fig. 1B). Other studies with GST fusion proteins prepared from the ARG SH2 and ARG SH3 domains demonstrated that both confer binding to Siva-1 (Fig. 1C). By contrast, binding of GST-ARG SH2, but not GST-ARG SH3, was detectable to the shorter, nonapoptotic Siva-2 protein (Fig.1D). Human Siva-1, but not Siva-2, contains a proline-rich PESP sequence (amino acids 82–85) for potential binding to ARG SH3. Notably, however, the PESP site is not conserved in mouse Siva-1 (22Prasad K.V. Ao Z. Yoon Y. Wu M.X. Rizk M. Jacquot S. Schlossman S.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6346-6351Crossref PubMed Scopus (250) Google Scholar,23Yoon Y. Ao Z. Cheng Y. Schlossman S.F. Prasad K.V. Oncogene. 1999; 18: 7174-7179Crossref PubMed Scopus (44) Google Scholar). In this regard, GST-ARG SH2, and not GST-ARG SH3, binds to mouse Siva-1 (data not shown). These findings demonstrate that the ARG SH2 domain interacts with Siva-1 and Siva-2 of both human and mouse origin and that binding of the ARG SH3 domain is also detectable with human Siva-1.To determine whether Siva-1 is a substrate for ARG, GST-Siva-1 was incubated with kinase-active ARG (Fig.2A, left) in the presence of [γ-32P]ATP. Analysis of the reaction products by SDS-PAGE and autoradiography demonstrated phosphorylation of Siva-1 (Fig. 2A, right). As a control, there was no detectable phosphorylation when Siva-1 was incubated with kinase-inactive ARG(K-R) in which Lys337 in the ATP binding site was mutated to Arg (Fig. 2A). There are two potential tyrosine phosphorylation sites in Siva-1 that are located at Tyr48 and Tyr67. Mutation of these sites to Phe and then incubation of the mutant proteins with ARG demonstrated abrogation of phosphorylation with GST-Siva-1(Y48F), but not with GST-Siva-1(Y67F) (Fig. 2A, right). To assess whether ARG phosphorylates Siva-1 in vivo, Flag-Siva-1 was coexpressed with ARG and lysates were analyzed by immunoblotting with anti-P-Tyr. The results show that Siva-1 is phosphorylated by ARG in cells (Fig.2B). By contrast, there was no detectable tyrosine phosphorylation of Flag-Siva-1 when this vector was coexpressed with ARG(K-R) (data not shown). Although Flag-Siva-1(Y67F) was also phosphorylated by ARG, there was no detectable tyrosine phosphorylation of Flag-Siva-1(Y48F) (Fig. 2B). In concert with these findings and the presence of Tyr48 in Siva-2, coexpression of Siva-2 with ARG, but not ARG(K-R), also resulted in detectable tyrosine phosphorylation (Fig. 2B). These findings thus provided support for ARG-mediated phosphorylation of Siva-1 and Siva-2 on Tyr48in vitro and in vivo. To determine whether ARG, like c-Abl (20Sun X. Wu F. Datta R. Kharbanda S. Kufe D. J. Biol. Chem. 2000; 275: 7470-7473Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar, 21Sun X. Majumder P. Shioya H. Wu F. Kumar S. Weichselbaum R. Kharbanda S. Kufe D. J. Biol. Chem. 2000; 275: 17237-17240Abstract Full Text Full Text PDF PubMed Scopus (143) Google Scholar), is activated by ROS, cells expressing Flag-ARG were treated with H2O2. Analysis of anti-Flag immunoprecipitates for phosphorylation of GST-Siva-1 demonstrated H2O2concentration-dependent induction of ARG activity (Fig.2C). As a control, the same anti-Flag immunoprecipitates failed to phosphorylate GST-Siva-1(Y48F) (Fig. 2C). In studies of wild-type and arg−/−MEFs, expression of Flag-Siva-1 resulted in little if any detectable phosphorylation (Fig. 2D). Treatment of the wild-type cells with H2O2, however, was associated with tyrosine phosphorylation of Flag-Siva-1 (Fig. 2D). By contrast, there was no detectable phosphorylation of Flag-Siva-1 in H2O2-treatedarg−/− MEFs (Fig. 2D). These findings demonstrate that activation of ARG by H2O2 is associated with phosphorylation of Siva-1 on Tyr48.To assess the functional significance of the ARG-Siva-1 interaction, MCF-7 cells were prepared that stably express ARG or the ARG(K-R) mutant. There was no detectable effect of ARG or ARG(K-R) expression on MCF-7 cell growth (data not shown) or cell cycle distribution (Fig.3A). Expression of Siva-1 in the wild-type MCF-7 cells was associated with the appearance of apoptotic cells containing sub-G1 DNA (Fig. 3A). Whereas Siva-1-induced apoptosis was more pronounced in MCF-7/ARG cells, expression of Siva-1 had little if any effect on MCF-7/ARG(K-R) cells (Fig. 3A). As a control, expression of Siva-2 resulted in substantially less apoptosis in MCF-7/ARG cells as compared with that obtained with Siva-1, and had no effect on the wild-type MCF-7 and MCF-7/ARG(K-R) cells (Fig. 3A). To extend these findings, Siva-1 was expressed in the wild-type andarg−/− MEFs. The results demonstrate that, although Siva-1 induces apoptosis in wild-type cells, there was little effect of Siva-1 in the absence of ARG expression (Fig. 3B). In concert with the finding that ARG phosphorylates Siva-1 on Tyr48, the expression of Siva-1(Y48F) had little effect on the induction of apoptosis, whereas the proapoptotic effects of Siva-1(Y67F) were similar to those obtained with wild-type Siva-1 (Fig. 3C). These findings demonstrate that Siva-1-induced apoptosis is dependent on the ARG kinase function and that ARG-mediated phosphorylation of Siva-1 on Tyr48 is a proapoptotic signal.Figure 3Interaction ofARGand Siva-1 in the apoptotic response to oxidative stress.A, MCF-7, MCF-7/ARG, and MCF-7/ARG(K-R) cells were transfected to express GFP-Siva-1 or GFP-Siva-2. At 36 h after transfection, GFP-positive cells were analyzed for sub-G1 DNA (left panels). Cells with sub-G1 DNA are depicted in theshaded profiles. GO/G1 and G2/M cells are shown in the dark profiles and S phase cells in the hatched profiles. Cells were subjected to immunoblot analysis (IB) with the indicated antibodies (right panels). B, wild-type (arg+/+) andarg−/− MEFs were infected with empty or Siva-1 retroviral vectors for 24 h. The cells were analyzed by flow cytometry (left panels) and immunoblotting (right panels). C, MCF-7 (open bars), MCF-7/ARG (hatched bars), and MCF-7/ARG(K-R) (solid bars) were transfected with GFP-Siva-1, GFP-Siva-2, GFP-Siva-1(Y48F), or GFP-Siva-1(Y67F). GFP-positive cells were analyzed for DNA content. The results are expressed as the percentage (mean ± S.E. of two independent experiments each performed in duplicate) of GFP-positive cells with sub-G1 DNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT)As the findings demonstrate that ARG is activated by ROS, H2O2-induced apoptosis was assessed in the MCF-7, MCF-7/ARG, and MCF-7/ARG(K-R) cells. The results demonstrate that the apoptotic effects of H2O2 are attenuated in MCF-7/ARG(K-R) as compared with wild-type MCF-7 cells (Fig. 4A). Notably, however, there was a marked increase in H2O2-induced apoptosis in MCF-7/ARG cells (Fig. 4A). The apoptotic effects of H2O2 were attenuated inarg−/− (two separate embryos) as compared with wild-type MEFs (Fig. 4B). Moreover, expression of ARG in arg−/− cells corrected the defect in H2O2-induced apoptosis (Fig.4C).Figure 4Apoptotic response to oxidative stress isARGkinase-dependent.A–C, the indicated cells were left untreated (open bars) or treated with 1 mm H2O2 for 2 h (hatched bars), washed, and then cultured for 18 h. In C,arg+/− cells were prepared by transduction ofarg−/− MEFs with a retroviral vector expressing ARG (transduction efficiency, >95% as determined by GFP expression). Cells were analyzed for DNA content. The results are expressed as the percentage (mean ± S.E. of two independent experiments performed in duplicate) of cells with sub-G1DNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT)The ARG tyrosine kinase has, like c-Abl, been associated with the development of leukemia (25Cazzaniga G. Tosi S. Aloisi A. Giudici G. Daniotti M. Pioltelli P. Kearney L. Biondi A. Blood. 1999; 94: 4370-4373Crossref PubMed Google Scholar, 26Iijima Y. Ito T. Oikawa T. Eguchi M. Eguchi-Ishimae M. Kamada N. Kishi K. Asano S. Sakaki Y. Sato Y. Blood. 2000; 95: 2126-2131PubMed Google Scholar). Despite the relatedness to c-Abl and a role in neurolation as defined inarg−/− mice (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar), a function for ARG in cell signaling has remained obscure. Genetic studies inDrosophila have indicated that Abl family kinases regulate cellular morphology through interactions with the cytoskeleton (27Gertler F.B. Bennett R.L. Clark M.J. Hoffmann F.M. Cell. 1989; 58: 103-113Abstract Full Text PDF PubMed Scopus (199) Google Scholar,28Gertler F.B. Hill K.K. Clark M.J. Hoffman F.M. Genes Dev. 1993; 7: 441-453Crossref PubMed Scopus (113) Google Scholar). The identification of actin-binding domains in the C-terminal region of ARG (3Van Etten R.A. Jackson P.K. Baltimore D. Sanders M.C. Matsuddaira P.T. Janmey P.A. J. Cell Biol. 1994; 124: 325-340Crossref PubMed Scopus (237) Google Scholar) and localization of ARG with actin microfilaments (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar) have supported a role in regulation of the actin cytoskeletion. The present results provide evidence for involvement of ARG in the cellular response to oxidative stress. Moreover, the induction of apoptosis by oxidative stress is attenuated in ARG-deficient cells. These findings indicate that, in addition to regulating the actin cytoskeleton, ARG functions in ROS-mediated signals that induce an apoptotic response.Siva-1 interacts with members of the tumor necrosis factor receptor family and induces apoptosis in diverse cells (22Prasad K.V. Ao Z. Yoon Y. Wu M.X. Rizk M. Jacquot S. Schlossman S.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6346-6351Crossref PubMed Scopus (250) Google Scholar). Siva-1 has also been implicated in the induction of Coxsackievirus-induced apoptosis (29Henke A. Launhardt H. Klement K. Stelzner A. Zell R. Munder T. J. Virol. 2000; 74: 4284-4290Crossref PubMed Scopus (115) Google Scholar). Full-length Siva-1 but not Siva-2, which lacks sequences encoded by exon 2, induces the apoptotic response (23Yoon Y. Ao Z. Cheng Y. Schlossman S.F. Prasad K.V. Oncogene. 1999; 18: 7174-7179Crossref PubMed Scopus (44) Google Scholar). The present studies demonstrate that ARG phosphorylates both Siva-1 and Siva-2. By contrast, the interaction between ARG and Siva-1, but not Siva-2, is associated with the induction of apoptosis. Our results also demonstrate that mutation of the Siva-1 Tyr48 site abrogates the apoptotic function of Siva-1 and that apoptosis induced by Siva-1 is dependent on expression of kinase-active ARG. These findings thus define ARG as an upstream effector of Siva-1 in the apoptotic response to oxidative stress. To extend the findings in the yeast two-hybrid system that ARG associates with the pro-apoptotic Siva-1 protein (data not shown), lysates from MCF-7 cells expressing Flag-tagged ARG and GFP-tagged human Siva-1 were subjected to immunoprecipitation with anti-Flag. Analysis of the precipitates by immunoblotting with anti-GFP demonstrated the presence of ARG-Siva-1 complexes (Fig.1A, left). Analysis of anti-Flag immunoprecipitates from cells expressing Flag-Siva-1 by immunoblotting with anti-ARG provided further support for binding of ARG and Siva-1 (Fig. 1A, right). To extend these findings, lysates from cells expressing Flag-ARG were incubated with a GST-Siva-1 fusion protein. Analysis of the adsorbates with anti-Flag confirmed the binding of ARG and Siva-1 (Fig. 1B). Other studies with GST fusion proteins prepared from the ARG SH2 and ARG SH3 domains demonstrated that both confer binding to Siva-1 (Fig. 1C). By contrast, binding of GST-ARG SH2, but not GST-ARG SH3, was detectable to the shorter, nonapoptotic Siva-2 protein (Fig.1D). Human Siva-1, but not Siva-2, contains a proline-rich PESP sequence (amino acids 82–85) for potential binding to ARG SH3. Notably, however, the PESP site is not conserved in mouse Siva-1 (22Prasad K.V. Ao Z. Yoon Y. Wu M.X. Rizk M. Jacquot S. Schlossman S.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6346-6351Crossref PubMed Scopus (250) Google Scholar,23Yoon Y. Ao Z. Cheng Y. Schlossman S.F. Prasad K.V. Oncogene. 1999; 18: 7174-7179Crossref PubMed Scopus (44) Google Scholar). In this regard, GST-ARG SH2, and not GST-ARG SH3, binds to mouse Siva-1 (data not shown). These findings demonstrate that the ARG SH2 domain interacts with Siva-1 and Siva-2 of both human and mouse origin and that binding of the ARG SH3 domain is also detectable with human Siva-1. To determine whether Siva-1 is a substrate for ARG, GST-Siva-1 was incubated with kinase-active ARG (Fig.2A, left) in the presence of [γ-32P]ATP. Analysis of the reaction products by SDS-PAGE and autoradiography demonstrated phosphorylation of Siva-1 (Fig. 2A, right). As a control, there was no detectable phosphorylation when Siva-1 was incubated with kinase-inactive ARG(K-R) in which Lys337 in the ATP binding site was mutated to Arg (Fig. 2A). There are two potential tyrosine phosphorylation sites in Siva-1 that are located at Tyr48 and Tyr67. Mutation of these sites to Phe and then incubation of the mutant proteins with ARG demonstrated abrogation of phosphorylation with GST-Siva-1(Y48F), but not with GST-Siva-1(Y67F) (Fig. 2A, right). To assess whether ARG phosphorylates Siva-1 in vivo, Flag-Siva-1 was coexpressed with ARG and lysates were analyzed by immunoblotting with anti-P-Tyr. The results show that Siva-1 is phosphorylated by ARG in cells (Fig.2B). By contrast, there was no detectable tyrosine phosphorylation of Flag-Siva-1 when this vector was coexpressed with ARG(K-R) (data not shown). Although Flag-Siva-1(Y67F) was also phosphorylated by ARG, there was no detectable tyrosine phosphorylation of Flag-Siva-1(Y48F) (Fig. 2B). In concert with these findings and the presence of Tyr48 in Siva-2, coexpression of Siva-2 with ARG, but not ARG(K-R), also resulted in detectable tyrosine phosphorylation (Fig. 2B). These findings thus provided support for ARG-mediated phosphorylation of Siva-1 and Siva-2 on Tyr48in vitro and in vivo. To determine whether ARG, like c-Abl (20Sun X. Wu F. Datta R. Kharbanda S. Kufe D. J. Biol. Chem. 2000; 275: 7470-7473Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar, 21Sun X. Majumder P. Shioya H. Wu F. Kumar S. Weichselbaum R. Kharbanda S. Kufe D. J. Biol. Chem. 2000; 275: 17237-17240Abstract Full Text Full Text PDF PubMed Scopus (143) Google Scholar), is activated by ROS, cells expressing Flag-ARG were treated with H2O2. Analysis of anti-Flag immunoprecipitates for phosphorylation of GST-Siva-1 demonstrated H2O2concentration-dependent induction of ARG activity (Fig.2C). As a control, the same anti-Flag immunoprecipitates failed to phosphorylate GST-Siva-1(Y48F) (Fig. 2C). In studies of wild-type and arg−/−MEFs, expression of Flag-Siva-1 resulted in little if any detectable phosphorylation (Fig. 2D). Treatment of the wild-type cells with H2O2, however, was associated with tyrosine phosphorylation of Flag-Siva-1 (Fig. 2D). By contrast, there was no detectable phosphorylation of Flag-Siva-1 in H2O2-treatedarg−/− MEFs (Fig. 2D). These findings demonstrate that activation of ARG by H2O2 is associated with phosphorylation of Siva-1 on Tyr48. To assess the functional significance of the ARG-Siva-1 interaction, MCF-7 cells were prepared that stably express ARG or the ARG(K-R) mutant. There was no detectable effect of ARG or ARG(K-R) expression on MCF-7 cell growth (data not shown) or cell cycle distribution (Fig.3A). Expression of Siva-1 in the wild-type MCF-7 cells was associated with the appearance of apoptotic cells containing sub-G1 DNA (Fig. 3A). Whereas Siva-1-induced apoptosis was more pronounced in MCF-7/ARG cells, expression of Siva-1 had little if any effect on MCF-7/ARG(K-R) cells (Fig. 3A). As a control, expression of Siva-2 resulted in substantially less apoptosis in MCF-7/ARG cells as compared with that obtained with Siva-1, and had no effect on the wild-type MCF-7 and MCF-7/ARG(K-R) cells (Fig. 3A). To extend these findings, Siva-1 was expressed in the wild-type andarg−/− MEFs. The results demonstrate that, although Siva-1 induces apoptosis in wild-type cells, there was little effect of Siva-1 in the absence of ARG expression (Fig. 3B). In concert with the finding that ARG phosphorylates Siva-1 on Tyr48, the expression of Siva-1(Y48F) had little effect on the induction of apoptosis, whereas the proapoptotic effects of Siva-1(Y67F) were similar to those obtained with wild-type Siva-1 (Fig. 3C). These findings demonstrate that Siva-1-induced apoptosis is dependent on the ARG kinase function and that ARG-mediated phosphorylation of Siva-1 on Tyr48 is a proapoptotic signal. As the findings demonstrate that ARG is activated by ROS, H2O2-induced apoptosis was assessed in the MCF-7, MCF-7/ARG, and MCF-7/ARG(K-R) cells. The results demonstrate that the apoptotic effects of H2O2 are attenuated in MCF-7/ARG(K-R) as compared with wild-type MCF-7 cells (Fig. 4A). Notably, however, there was a marked increase in H2O2-induced apoptosis in MCF-7/ARG cells (Fig. 4A). The apoptotic effects of H2O2 were attenuated inarg−/− (two separate embryos) as compared with wild-type MEFs (Fig. 4B). Moreover, expression of ARG in arg−/− cells corrected the defect in H2O2-induced apoptosis (Fig.4C). The ARG tyrosine kinase has, like c-Abl, been associated with the development of leukemia (25Cazzaniga G. Tosi S. Aloisi A. Giudici G. Daniotti M. Pioltelli P. Kearney L. Biondi A. Blood. 1999; 94: 4370-4373Crossref PubMed Google Scholar, 26Iijima Y. Ito T. Oikawa T. Eguchi M. Eguchi-Ishimae M. Kamada N. Kishi K. Asano S. Sakaki Y. Sato Y. Blood. 2000; 95: 2126-2131PubMed Google Scholar). Despite the relatedness to c-Abl and a role in neurolation as defined inarg−/− mice (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar), a function for ARG in cell signaling has remained obscure. Genetic studies inDrosophila have indicated that Abl family kinases regulate cellular morphology through interactions with the cytoskeleton (27Gertler F.B. Bennett R.L. Clark M.J. Hoffmann F.M. Cell. 1989; 58: 103-113Abstract Full Text PDF PubMed Scopus (199) Google Scholar,28Gertler F.B. Hill K.K. Clark M.J. Hoffman F.M. Genes Dev. 1993; 7: 441-453Crossref PubMed Scopus (113) Google Scholar). The identification of actin-binding domains in the C-terminal region of ARG (3Van Etten R.A. Jackson P.K. Baltimore D. Sanders M.C. Matsuddaira P.T. Janmey P.A. J. Cell Biol. 1994; 124: 325-340Crossref PubMed Scopus (237) Google Scholar) and localization of ARG with actin microfilaments (10Koleske A.J. Gifford A.M. Scott M.L. Nee M. Bronson R.T. Miczek K.A. Baltimore D. Neuron. 1998; 21: 1259-1272Abstract Full Text Full Text PDF PubMed Scopus (333) Google Scholar) have supported a role in regulation of the actin cytoskeletion. The present results provide evidence for involvement of ARG in the cellular response to oxidative stress. Moreover, the induction of apoptosis by oxidative stress is attenuated in ARG-deficient cells. These findings indicate that, in addition to regulating the actin cytoskeleton, ARG functions in ROS-mediated signals that induce an apoptotic response. Siva-1 interacts with members of the tumor necrosis factor receptor family and induces apoptosis in diverse cells (22Prasad K.V. Ao Z. Yoon Y. Wu M.X. Rizk M. Jacquot S. Schlossman S.F. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6346-6351Crossref PubMed Scopus (250) Google Scholar). Siva-1 has also been implicated in the induction of Coxsackievirus-induced apoptosis (29Henke A. Launhardt H. Klement K. Stelzner A. Zell R. Munder T. J. Virol. 2000; 74: 4284-4290Crossref PubMed Scopus (115) Google Scholar). Full-length Siva-1 but not Siva-2, which lacks sequences encoded by exon 2, induces the apoptotic response (23Yoon Y. Ao Z. Cheng Y. Schlossman S.F. Prasad K.V. Oncogene. 1999; 18: 7174-7179Crossref PubMed Scopus (44) Google Scholar). The present studies demonstrate that ARG phosphorylates both Siva-1 and Siva-2. By contrast, the interaction between ARG and Siva-1, but not Siva-2, is associated with the induction of apoptosis. Our results also demonstrate that mutation of the Siva-1 Tyr48 site abrogates the apoptotic function of Siva-1 and that apoptosis induced by Siva-1 is dependent on expression of kinase-active ARG. These findings thus define ARG as an upstream effector of Siva-1 in the apoptotic response to oxidative stress. The ARG tyrosine kinase interacts with Siva-1 in the apoptotic response to oxidative stress.Journal of Biological ChemistryVol. 287Issue 43PreviewVOLUME 276 (2001) PAGES 11465–11468 Full-Text PDF Open Access
Referência(s)