- Email: [email protected]
Differential regulation of insulin receptor substrate-1 degradation during mannitol and okadaic acid induced apoptosis in human neuroblastoma cells
Cellular Signalling 17 (2005) 769 – 775 www.elsevier.com/locate/cellsig
Differential regulation of insulin receptor substrate-1 degradation during mannitol and okadaic acid induced apoptosis in human neuroblastoma cells Bhumsoo Kim, SangSu Oh, Cynthia M. van Golen, Eva L. Feldman* Department of Neurology, University of Michigan, 4414 Kresge III, 200 Zina Pitcher Place, Ann Arbor, MI 48109-0588, United States Received 1 September 2004; received in revised form 29 October 2004; accepted 1 November 2004 Available online 23 November 2004
Abstract Insulin receptor substrate (IRS) proteins are major docking molecules for the type I insulin like growth factor (IGF) receptor (IGF-IR) and mediate their effects on downstream signaling molecules. In this report, we investigated IRS-1 regulation during apoptosis in human neuroblastoma SH-EP cells. Treatment of SH-EP cells with mannitol or okadaic acid (OA) induces apoptosis with the typical characteristics of anoikis. Mannitol treatment results in IRS-1 degradation with concomitant appearance of smaller fragments, likely representing caspase cleavage products. In contrast OA-induced IRS-1 degradation is accompanied by a mobility shift in IRS-1, suggesting IRS-1 serine/threonine phosphorylation. Mannitol-induced, but not OA-induced, degradation is blocked by IGF-I. Pretreatment of the cells with caspase or proteasome inhibitors also partially blocks mannitol-induced IRS-1 degradation. These results suggest two independent pathways are involved in IRS-1 degradation; one pathway is dependent on caspase activation and is blocked by IGF-I, while a second pathway is caspaseindependent and IGF-I-insensitive. D 2004 Elsevier Inc. All rights reserved. Keywords: Insulin receptor substrate; Apoptosis; Caspase; Proteasome; Neuroblastoma
1. Introduction Insulin receptor substrate (IRS) proteins are key signaling molecules that mediate the physiological actions of the insulin receptor and the type I insulin like growth factor (IGF) receptor (IGF-IR) [1]. Four members of this family have been identified (IRS-1 through -4), each containing characteristic N-terminal pleckstrin homology and phosphotyrosine binding domains. IRS proteins lack intrinsic kinase activity; however, subsequent to binding by activated receptors, IRS proteins are tyrosine phosphorylated and act as docking proteins for downstream signaling molecules [2]. IRS-1 is the first cloned and most studied member of the IRS family. IRS-1 plays critical roles in cellular * Corresponding author. Tel.: +1 734 763 7274; fax: +1 734 763 7275. E-mail address: [email protected] (E.L. Feldman). 0898-6568/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.cellsig.2004.11.001
differentiation and proliferation. Overexpression of IRS-1 results in neoplastic transformation of NIH 3T3 cells [3,4] and expression of antisense IRS-1 RNA inhibits cell growth [5]. IRS-1 is constitutively activated in various tumors including breast cancers, sarcomas, Wilm’s tumors, and adrenal cortical carcinomas; in contrast, transfection of dominant-negative IRS-1 greatly reduces tumor growth [6]. RS-1 has 18 tyrosine residues that serve as docking sites for downstream signaling molecules such as phosphatidylinositol 3-kinase (PI 3-K), phosphotyrosine phosphatase SHP2, and the adapter proteins Grb2, Nck, Crk and Shb [2]. IRS-1 also has more than 30 serine/threonine phosphorylation sites. IRS-1 serine/threonine phosphorylation prevents its tyrosine phosphorylation and inhibits insulin receptor or IGF-IR signaling. For example, phosphorylation of Ser307 on mouse IRS-1 (the equivalent of Ser 312 on human IRS-1) prevents the interaction
770
B. Kim et al. / Cellular Signalling 17 (2005) 769–775
of IRS-1 with the insulin receptor and induces insulin resistance [7,8]. IRS-1 is phosphorylated at serine/threonine residues by multiple kinases including c-Jun Nterminal kinase [9], inhibitor nB kinase complex [10], Rho kinase [11] and PKC ~ [12]. PI3-K/Akt/mTOR pathway [13,14] and the suppressor of cytokine signaling (SOCS) family of proteins [15] also play pivotal roles in serine/threonine phosphorylation. Chronic stimulation by insulin/IGF-I or tumor necrosis factor (TNF)-a results in increased phosphorylation at inhibitory serine/threonine residues [8]. After serine/threonine phosphorylation, IRS-1 is degraded by a ubiquitin/proteasome pathway [16,17], although some reports also show calpain-dependent degradation [18,19]. Our laboratory studies apoptosis in primary and transformed neuronal cells. We have shown that high concentrations of glucose or mannitol induce an apoptotic cell death in dorsal root ganglia and glial cells, as well as in SH-SY5Y and SH-EP human neuroblastoma cells [20– 23]. Mannitol-induced apoptosis in SH-EP cells is accompanied by dephosphorylation and degradation of focal adhesion kinase (FAK), disruption of actin stress fibers, loss of focal adhesion sites, and cell detachment, all of which are prevented by IGF-I treatment, overexpression of IGF-IR or caspase inhibition [24]. SH-EP cells also undergo apoptosis induced by treatment with okadaic acid (OA), an inhibitor of the serine/threonine protein phosphatases 1 and 2A. OA-induced apoptosis also involves FAK dephosphorylation and degradation, morphological changes, cell detachment and caspase activation [25]. However, unlike mannitol-induced apoptosis, there is a long lag between FAK dephosphorylation/ cell detachment and FAK degradation/apoptosis. OAinduced apoptosis is not prevented by IGF-I treatment or caspase-inhibition. Several studies reported that IRS proteins play important roles during apoptosis [26,27]. In this study, we investigated the degradation of IRS-1 during mannitol- and OA-induced apoptosis. Mannitol treatment results in time- and concentration-dependent degradation of IRS-1 without any apparent molecular weight shift. The disappearance of IRS-1 is accompanied by the appearance of smaller degradation fragments, suggesting IRS-1 is cleaved by caspases. In contrast, OA-induced degradation of IRS-1 results in an upward mobility shift of IRS-1 indicating IRS-1 serine/threonine phosphorylation. IGF-I treatment, caspase inhibition, and proteasome inhibition all block mannitol-induced IRS-1 degradation. Consistent with our previous study of apoptosis [25], OA-induced degradation of IRS-1 is not prevented by IGF-I or inhibitors of caspase or proteasome. These results show that IRS-1 can be degraded by at least two independent pathways depending on the apoptotic stimuli; one pathway is caspase dependent and can be prevented by IGF-I and the other pathway is caspase-independent and IGF-I-insensitive.
2. Materials and methods 2.1. Materials Anti-IRS-1 and anti-actin antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies against Shc and integrin h-1 were purchased from Transduction Laboratory (Lexington, KY) and antibodies against Akt and mitogen-activated protein kinase (MAPK) were from Cell Signaling Technology (Beverly, MA). Anti-caspase-3 antibody was from BD Pharmingen (San Diego, CA). OA and proteasome inhibitors were purchased from Calbiochem (La Jolla, CA). Caspase inhibitors were from Enzyme System Products (Livermore, CA). Recombinant human IGF-I was kindly provided by Cephalon (Westchester, PA). 2.2. Cell culture SH-EP human neuroblastoma cells were maintained in DMEM containing 10% calf serum as previously described [28]. The cells were serum starved 4–6 h before treatment with mannitol or OA with or without IGF-I. The inhibitors were added 30 min before mannitol or OA treatment. 2.3. Immunoblotting and immunoprecipitation Immunoblotting was performed as described previously [28]. All experiments were repeated at least three times except Fig. 4B which was repeated twice. Typical representative results are presented in the figures.
3. Results 3.1. IRS-1 is degraded during mannitol- and OA-induced apoptosis We have recently shown that treatment of SH-EP human neuroblastoma cells with mannitol or OA results in apoptosis with the characteristics of anoikis, including actin stress fiber disruption and cell detachment [24,25]. Apoptosis was accompanied by the degradation of several signaling molecules including FAK and Akt, which serve important roles in IGF-I signaling. Since several reports suggested the involvement of IRS proteins during apoptosis [19,26,27], we examined changes in IRS-1 during mannitoland OA-induced apoptosis. When serum-starved SH-EP cells are treated with 300 mM mannitol there is a timedependent disappearance of IRS-1 at 160 kDa (Fig. 1A, arrow). The degradation is noticeable from 1.5 h after mannitol treatment, the time when most of the cells are rounding up and starting to show membrane blebbing [24]. IRS-1 degradation is almost complete after 3 h when the majority of cells are detached. The degradation of the primary IRS-1 band is accompanied by the appearance of
B. Kim et al. / Cellular Signalling 17 (2005) 769–775
771
mannitol-induced degradation, OA-induced IRS-1 degradation is accompanied by an upward mobility shift, suggesting serine/threonine phosphorylation (Fig. 2A). We could not detect any smaller fragments during degradation as in mannitol treatment, even with analysis with higher percent PAGE (Fig. 3). The effect of OA is also time- and concentration-dependent with almost complete disappearance of the IRS-1 band at 4 h and 250 nM OA (Fig. 2B and C). The effect is specific to IRS-1 since there are essentially no changes in levels of Akt, MAPK, actin, or integrin h1 (Figs. 2A and 3). We compared IRS-1 band densities over time in treated SH-EP cells with untreated cells using the Scion Image Program (Frederick, Maryland). The half life of IRS-1 degradation is 1.5 h and 4 h for mannitol and OA treatment, respectively. 3.2. IGF-I prevents mannitol-induced, but not OA-induced IRS-1 degradation
Fig. 1. Mannitol treatment induces the degradation of IRS-1. SH-EP neuroblastoma cells are serum starved for 5 h and then treated with 300 mM mannitol for the indicated times (A) or for 4 h with increasing concentrations of mannitol (B). Equal amounts of protein from whole cell lysates are separated by SDS-PAGE, transferred to nitrocellulose and analyzed by immunoblotting with the anti-IRS-1 antibody. Arrows indicate intact IRS-1 and arrowheads indicate the fragments.
distinct smaller fragments around 150, 120, 95, 65 and 35 kDa (arrowheads). The degradation of IRS-1 is also mannitol concentration-dependent, with maximum degradation at 500 mM (Fig. 1B). However, mannitol has no effect on actin or integrin h1 levels (Figs. 3 and 4A). OA treatment also induces IRS-1 degradation; however, in contrast to
We have shown that IGF-I prevents mannitol-induced, but not OA-induced, apoptosis in SH-EP cells [24,25]. Therefore, we next examined the effect of IGF-I on IRS-1 degradation. In agreement with our previous findings [29], when SH-EP cells are treated with 10 nM IGF-I for 4 h there are no changes in IRS-1 levels (Fig. 3). Mannitol-induced degradation of IRS-1 is almost completely blocked by simultaneous treatment of 10 nM IGF-I (Fig. 3). In contrast, OA-induced IRS-1 degradation cannot be prevented by IGF-I treatment. Treatment of mannitol, OA, and IGF-I has no effect on the expression of h1 integrin or actin (Fig. 3, bottom panels). 3.3. Effect of inhibition of caspase or proteasome activity on IRS-1 degradation Caspases are important mediators of apoptosis [30]. Caspase-3 is synthesized as an inactive precursor that is proteolytically cleaved to generate an active enzyme [31]. We have shown that caspases are involved in neuronal apoptosis [23,32,33] and FAK is degraded in a caspasedependent manner during mannitol-induced apoptosis [24].
Fig. 2. OA-induced degradation of IRS-1 is accompanied with mobility shift. (A) Serum-starved SH-EP cells are treated with 250 nM OA for 4 h. Whole cell lysates were analyzed by immunoblotting with the indicated antibodies. The effect of the time course (B) and concentration (C) of OA treatment are analyzed in serum-starved SH-EP cells using anti-IRS-1 antibody.
772
B. Kim et al. / Cellular Signalling 17 (2005) 769–775
caspase type-specific inhibitors are only partially effective in preventing mannitol-induced IRS-1 degradation (Fig. 4B). Among proteasome inhibitors, MG132 is most effective followed by epoxomycine and PS1; lactacystine is not effective at all (Fig. 4C). In contrast, ZVAD and MG132 cannot prevent OA-induced IRS-1 degradation (Fig. 4D). The effects of proteasome inhibitors are closely related to the cleavage of caspase-3. In agreement with our previous results [24], mannitol treatment induces caspase-3 cleavage, which is blocked by IGF-I addition (Fig. 5A). Like the effect on IRS-1 degradation, MG132 and epoxomycine partially prevent caspase-3 cleavage; however, lactacystine is ineffective. As expected, ZVAD blocks mannitol-induced caspase-3 cleavage (Fig. 5B).
4. Discussion In this study, we report the differential degradation of IRS-1 during mannitol- and OA-induced apoptosis. Our previous work shows that treatment of SH-EP neuro-
Fig. 3. IGF-I prevents mannitol-induced, but not OA-induced, IRS-1 degradation. Serum-starved SH-EP cells are treated with 300 mM mannitol or 250 nM OA for 4 h in the absence or presence of 10 nM IGF-I. Equal amounts of whole cell lysates were separated by SDS-PAGE, transferred to nitrocellulose, and analyzed by immunoblotting with the anti-IRS-1 antibody. The arrowhead indicates the fragments appearing during mannitol-induced degradation. The blot is stripped and reprobed with anti-integrin h1 and anti-acting antibodies (lower panels).
IRS-1 undergoes ubiquitin/proteasome-dependent degradation during insulin, IGF-I or TNF-a treatment [8,16,17]. Our previous report showed that caspase and ubiquitin/ proteasome pathways are activated within 4 h after mannitol, but not OA, treatment [24,25]. Depending on the cell type or stimuli, proteasome inhibitors can act as either pro-apoptotic or anti-apoptotic agents by activating or inhibiting caspase activation [34]. When our cells are incubated with ZVAD, a pan-caspase inhibitor, mannitolinduced IRS-1 degradation is greatly reduced (Fig. 4A). The proteasome inhibitor MG132 also blocks mannitol-induced degradation of IRS-1; however, the effect is not as complete as IGF-I or ZVAD treatment, since we can still observe the existence of smaller IRS-1 fragments (arrowhead). Other
Fig. 4. Inhibitors of caspase and proteasome blocks mannitol-induced IRS-1 degradation. Serum-starved SH-EP cells are treated with 300 mM mannitol with or without 10 nM IGF-I, 20 AM MG132, or 20 AM ZVAD for 4 h (A). The specificity of caspase inhibitors (B) and proteasome inhibitors (C) are also tested. ZVAD and MG132 cannot block OA-induced IRS-1 degradation (D). Equal amounts of protein from whole cell lysates are analyzed by immunoblotting with the indicated antibodies. Caspase and proteasome inhibitors are used at 20 AM. Lact, lactacystine; Epox, epoxomycine; MG, MG132.
B. Kim et al. / Cellular Signalling 17 (2005) 769–775
Fig. 5. Proteasome inhibitors prevent caspase-3 activation. Serum-starved SH-EP cells are treated with 300 mM mannitol with or without 10 nM IGF-I, or indicated inhibitors for 4 h. The whole cell lysates are analyzed for caspase-3 cleavage by anti-caspase-3 antibody.
blastoma cells with either mannitol or OA induces anoikislike apoptosis [24,25]. Disruption of actin filaments, dephosphorylation and degradation of FAK, loss of focal contacts, and simultaneous activation of caspase and ubiquitin/proteasome pathways all accompany mannitolinduced apoptosis. These changes are prevented by IGF-I [24]. OA treatment also induces similar morphological changes and cell detachment; however, apoptosis with the activation of the caspase and ubiquitin/proteasome pathways is evident only at much later time points and is not protected by IGF-I treatment [25]. In this study, we extend this work, finding that IRS-1 degradation follows a similar pattern; IGF-I treatment and inhibition of caspase and proteasome activation prevent mannitol-induced degradation, whereas OA-induced degradation is not reversed by these treatments. These results illustrate that two distinct pathways are engaged in IRS-1 degradation, which in turn suggests that different strategies must be considered for targeting IRS-1 for therapeutic applications. IRS proteins play critical roles in insulin or IGF-I signaling [2,35]. Manipulation of IRS-1 or -2 levels results in a diabetic phenotype in vivo and in vitro. IRS proteins also regulate cellular differentiation and proliferation. IRS-1 is constitutively activated in many tumors [6] and overexpression of IRS-1 results in neoplastic transformation [3,4]. Recently, involvement of IRS proteins in apoptosis has been reported. IRS-1 levels are decreased in a calpaindependent manner during thapsigargin-induced apoptosis in CPTX human prostate epithelial cell lines [19]. The level of IRS-2 is directly proportional to the degree of apoptosis in
773
pancreatic h cells [26] and IRS-1 knock-out cells are more sensitive to serum-starvation induced apoptosis [27]. Moreover, IRS-1 and -2 bind to the anti-apoptotic protein Bcl-2 and up-regulate cell survival [36]. In our current report, we show the degradation of IRS-1 during mannitol- and OAinduced apoptosis. Even though the pathways involved in the degradation are different, these results suggest the possible involvement of IRS-1 in apoptosis in neuroblastoma cells. IRS-1 may in part regulate mannitol- and OA-induced apoptosis through its interaction with integrins. Both mannitol and OA treatment result in morphological and biochemical changes typical to anoikis [24,25]. Anoikis is an apoptotic process which occurs in cells denied cellmatrix or cell–cell attachment [37]. IRS-1 associates with integrins and regulates cell adhesion [38–40]. In this report, we show that the apparent changes in IRS-1 expression levels after mannitol treatment start to occur after 60 min. OA treatment initiates IRS-1 degradation between 2 and 4 h. Interestingly, these are the time points at which we observe a majority of cells rounding up and starting to detach [24,25]. Even though IRS-1 degradation may not be the sole instigator of apoptosis, the possible dissociation of integrin-IRS-1 after degradation may enhance the apoptotic process by facilitating cell detachment. This cell detachment would then initiate additional anoikis-related apoptotic pathways, thereby ensuring cell death. IRS-1 degradation is caspase-dependent during mannitol-induced apoptosis. During the preparation of this manuscript, Green et al. [41] published a paper supporting our results of caspase-mediated cleavage of IRS. They showed that caspase-10, but none of the other caspases, is responsible for the degradation of IRS-1 and -2 during the apoptosis of mammary epithelial cells. Our results show that of the four caspase-specific inhibitors we tested LEHD (caspase-9 specific inhibitor) is most effective other than general caspase inhibitor ZVAD. Therefore, it is possible that different caspases are involved in IRS degradation depending on cell type or apoptotic stimulus. Even though we show that proteasome inhibitors block IRS-1 degradation, the proteasome is likely not directly involved in IRS-1 degradation. Proteasome-mediated degradation requires prior ubiquitination of the target proteins [42]. IRS-1 is degraded by a ubiquitination/ proteasome-dependent pathway during insulin, IGF-I or TNF-a treatment, which is preceded by serine/threonine phosphorylation [8,16,17,43]. In our study, we do not detect a mobility shift in IRS-1 by mannitol treatment, suggesting a lack of serine/threonine phosphorylation. We also show that proteasome inhibitors only partially prevent IRS-1 degradation, which is well correlated with the level of cleaved (i.e. active) caspase-3. Proteasome inhibitors either enhance or inhibit apoptosis depending on the cell type and stimuli [34]. This specificity depends on whether proteasome inhibitors activate or inhibit caspase cleavage, which may explain our discrepancy with a previous study
774
B. Kim et al. / Cellular Signalling 17 (2005) 769–775
where sorbitol-induced IRS-1 degradation is not prevented by lactacystine in 3T3-L1 adipocytes [44]. Therefore, we suggest that proteasome inhibitors regulate caspase activation, which cleaves IRS-1. Indeed we found that there are several potential caspase cleavage sites in IRS-1. We are currently investigating whether the mutation of these sites can block mannitol-induced IRS-1 degradation. In contrast to mannitol, OA-induced IRS-1 degradation is not prevented by the inhibition of caspases or the proteasome, suggesting the involvement of a completely different pathway. Recently, Zhang et al. [19] have shown that IRS-1 is degraded by calcium-activated protease calpain, and this process cannot be prevented by caspase or proteasome inhibitors. Chi et al. [45] have shown that overexpression of calpastatin, an endogenous calpain inhibitor, prevents OA-induced apoptosis in human fibroblasts. However, as with proteasome inhibition, blocking calpain is either pro-apoptotic or anti-apoptotic depending on cell type and stimuli [46]. Another possible pathway responsible for OA-induced IRS-1 degradation involves lysosomal proteases. Recent studies show that lysosomal proteases actively participate in apoptosis as well as necrosis [47]. When mouse hepatocytes are treated with TNF-a, stimulation of the apoptotic pathway requires the release of cathepsin B from lysosomes [48]. This multistep process results in a long lag between TNF-a stimulation and the execution of apoptosis. Interestingly, we have also observed that there is a temporal discrepancy between OA treatment, morphological changes, and the start of apoptosis [25]. In summary, we have shown in this report that IRS-1 is degraded by at least two completely different pathways during mannitol- and OA-induced apoptosis. Mannitolinduced IRS-1 degradation is dependent on caspase activation and prevented by IGF-I treatment. IRS-1 degradation by OA treatment is independent of caspase or proteasome pathways and cannot be rescued by IGF-I treatment.
Acknowledgements The authors thank Ms. Judy Boldt for expert secretarial assistance. This work was supported by grants from the National Institutes of Health (NS38849 and NS36778), the Juvenile Diabetes Research Foundation Center for the Study of Complications in Diabetes, and the Program for Understanding Neurological Diseases (PFUND).
References [1] M.F. White, Am. J. Physiol.: Endocrinol. Metab. 283 (2002) E413. [2] M.G. Myers Jr., M.F. White, Annu. Rev. Pharmacol. Toxicol. 36 (1996) 615. [3] S. Tanaka, T. Ito, J.R. Wands, J. Biol. Chem. 271 (1996) 14610.
[4] T. Ito, Y. Sasaki, J.R. Wands, Mol. Cell. Biol. 16 (1996) 943. [5] M. Taouis, J. Dupont, A. Gillet, M. Derouet, J. Simon, Mol. Cell. Endocrinol. 137 (1998) 177. [6] Q. Chang, Y. Li, M.F. White, J.A. Fletcher, S. Xiao, Cancer Res. 62 (2002) 6035. [7] V. Aguirre, E.D. Werner, J. Giraud, Y.H. Lee, S.E. Shoelson, M.F. White, J. Biol. Chem. 277 (2002) 1531. [8] L. Rui, V. Aguirre, J.K. Kim, G.I. Shulman, A. Lee, A. Corbould, A. Dunaif, M.F. White, J. Clin. Invest. 107 (2001) 181. [9] Y.H. Lee, J. Giraud, R.J. Davis, M.F. White, J. Biol. Chem. 278 (2003) 2896. [10] Z. Gao, D. Hwang, F. Bataille, M. Lefevre, D. York, M.J. Quon, J. Ye, J. Biol. Chem. 277 (2002) 48115. [11] N. Begum, O.A. Sandu, M. Ito, S.M. Lohmann, A. Smolenski, J. Biol. Chem. 277 (2002) 6214. [12] Y.F. Liu, K. Paz, A. Herschkovitz, A. Alt, T. Tennenbaum, S.R. Sampson, M. Ohba, T. Kuroki, D. LeRoith, Y. Zick, J. Biol. Chem. 276 (2001) 14459. [13] O.N. Ozes, H. Akca, L.D. Mayo, J.A. Gustin, T. Maehama, J.E. Dixon, D.B. Donner, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 4640. [14] J. Li, K. DeFea, R.A. Roth, J. Biol. Chem. 274 (1999) 9351. [15] L. Rui, M. Yuan, D. Frantz, S. Shoelson, M.F. White, J. Biol. Chem. 277 (2002) 42394. [16] R. Zhande, J.J. Mitchell, J. Wu, X.J. Sun, Mol. Cell. Biol. 22 (2002) 1016. [17] A.V. Lee, J.L. Gooch, S. Oesterreich, R.L. Guler, D. Yee, Mol. Cell. Biol. 20 (2000) 1489. [18] L.K. Smith, K.M. Rice, C.W. Garner, Mol. Cell. Endocrinol. 122 (1996) 81. [19] H. Zhang, H. Hoff, C. Sell, Exp. Cell Res. 289 (2003) 352. [20] J.W. Russell, E.L. Feldman, Horm. Metab. Res. 31 (1999) 90. [21] C.C. Matthews, H. Odeh, E.L. Feldman, Neuroscience 79 (1997) 525. [22] C.M. van Golen, E.L. Feldman, J. Cell. Physiol. 182 (2000) 24. [23] C.L. Delaney, J.W. Russell, H.-L. Cheng, E.L. Feldman, J. Neuropathol. Exp. Neurol. 60 (2001) 147. [24] B. Kim, E.L. Feldman, J. Biol. Chem. 277 (2002) 27393. [25] B. Kim, C.M. van Golen, E.L. Feldman, Neoplasia 5 (2003) 405. [26] M.K. Lingohr, L.M. Dickson, C.E. Wrede, I. Briaud, J.F. McCuaig, M.G. Myers Jr., C.J. Rhodes, Mol. Cell. Endocrinol. 209 (2003) 17. [27] Y.H. Tseng, K. Ueki, K.M. Kriauciunas, C.R. Kahn, J. Biol. Chem. 277 (2002) 31601. [28] P.S. Leventhal, E.A. Shelden, B. Kim, E.L. Feldman, J. Biol. Chem. 272 (1997) 5214. [29] B. Kim, C. van Golen, EL. Feldman, Oncogene 23 (2004) 130. [30] H.Y. Chang, X. Yang, Microbiol. Mol. Biol. Rev. 64 (2000) 821. [31] D.W. Nicholson, A. Ali, N.A. Thornberry, J.P. Vaillancourt, C.K. Ding, M. Gallant, Y. Gareau, P.R. Griffin, M. Labelle, Y.A. Lazebnik, N.A. Munday, S.M. Raju, M.E. Smulson, T.-T. Yamin, V.L. Yu, D.K. Miller, Nature 376 (1995) 37. [32] J.W. Russell, A.J. Windebank, A. Schenone, E.L. Feldman, J. Neurobiol. 36 (1998) 455. [33] C.M. van Golen, V.P. Castle, E.L. Feldman, Cell Death Differ. 7 (2000) 654. [34] C. Wojcik, Cell. Mol. Life Sci. 56 (1999) 908. [35] D. LeRoith, C.T. Roberts Jr., Cancer Lett. 195 (2003) 127. [36] H. Ueno, E. Kondo, R. Yamamoto-Honda, K. Tobe, T. Nakamoto, K. Sasaki, K. Mitani, A. Furusaka, T. Tanaka, Y. Tsujimoto, T. Kadowaki, H. Hirai, Mol. Biol. Cell 11 (2000) 735. [37] S.M. Frisch, R.A. Screaton, Curr. Opin. Cell Biol. 13 (2001) 555. [38] K. Vuori, E. Ruoslahti, Science 266 (1994) 1576. [39] A. Guilherme, K. Torres, M.P. Czech, J. Biol. Chem. 273 (1998) 22899. [40] K. Reiss, J.Y. Wang, G. Romano, X. Tu, F. Peruzzi, R. Baserga, Oncogene 20 (2001) 490. [41] K.A. Green, M.J. Naylor, E.T. Lowe, P. Wang, E. Marshman, C.H. Streuli, J. Biol. Chem. 279 (2004) 25149.
B. Kim et al. / Cellular Signalling 17 (2005) 769–775 [42] V. Jesenberger, S. Jentsch, Nat. Rev., Mol. Cell Biol. 3 (2002) 112. [43] T.M. Pederson, D.L. Kramer, C.M. Rondinone, Diabetes 50 (2001) 24. [44] P. Gual, T. Gonzalez, T. Gremeaux, R. Barres, Y. Marchand-Brustel, J.F Tanti, J. Biol. Chem. 278 (2003) 26550. [45] X.J. Chi, T. Hiwasa, M. Maki, S. Sugaya, J. Nomura, K. Kita, N. Suzuki, FEBS Lett. 459 (1999) 391.
775
[46] R.W. Neumar, Y.A. Xu, H. Gada, R.P. Guttmann, R. Siman, J. Biol. Chem. 278 (2003) 14162. [47] G.S. Salvesen, J. Clin. Invest. 107 (2001) 21. [48] M.E. Guicciardi, J. Deussing, H. Miyoshi, S.F. Bronk, P.A. Svingen, C. Peters, S.H. Kaufmann, G.J. Gores, J. Clin. Invest. 106 (2000) 1127.
Differential regulation of insulin receptor substrate-1 degradation during mannitol and okadaic acid induced apoptosis in human neuroblastoma cells Bhumsoo Kim, SangSu Oh, Cynthia M. van Golen, Eva L. Feldman* Department of Neurology, University of Michigan, 4414 Kresge III, 200 Zina Pitcher Place, Ann Arbor, MI 48109-0588, United States Received 1 September 2004; received in revised form 29 October 2004; accepted 1 November 2004 Available online 23 November 2004
Abstract Insulin receptor substrate (IRS) proteins are major docking molecules for the type I insulin like growth factor (IGF) receptor (IGF-IR) and mediate their effects on downstream signaling molecules. In this report, we investigated IRS-1 regulation during apoptosis in human neuroblastoma SH-EP cells. Treatment of SH-EP cells with mannitol or okadaic acid (OA) induces apoptosis with the typical characteristics of anoikis. Mannitol treatment results in IRS-1 degradation with concomitant appearance of smaller fragments, likely representing caspase cleavage products. In contrast OA-induced IRS-1 degradation is accompanied by a mobility shift in IRS-1, suggesting IRS-1 serine/threonine phosphorylation. Mannitol-induced, but not OA-induced, degradation is blocked by IGF-I. Pretreatment of the cells with caspase or proteasome inhibitors also partially blocks mannitol-induced IRS-1 degradation. These results suggest two independent pathways are involved in IRS-1 degradation; one pathway is dependent on caspase activation and is blocked by IGF-I, while a second pathway is caspaseindependent and IGF-I-insensitive. D 2004 Elsevier Inc. All rights reserved. Keywords: Insulin receptor substrate; Apoptosis; Caspase; Proteasome; Neuroblastoma
1. Introduction Insulin receptor substrate (IRS) proteins are key signaling molecules that mediate the physiological actions of the insulin receptor and the type I insulin like growth factor (IGF) receptor (IGF-IR) [1]. Four members of this family have been identified (IRS-1 through -4), each containing characteristic N-terminal pleckstrin homology and phosphotyrosine binding domains. IRS proteins lack intrinsic kinase activity; however, subsequent to binding by activated receptors, IRS proteins are tyrosine phosphorylated and act as docking proteins for downstream signaling molecules [2]. IRS-1 is the first cloned and most studied member of the IRS family. IRS-1 plays critical roles in cellular * Corresponding author. Tel.: +1 734 763 7274; fax: +1 734 763 7275. E-mail address: [email protected] (E.L. Feldman). 0898-6568/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.cellsig.2004.11.001
differentiation and proliferation. Overexpression of IRS-1 results in neoplastic transformation of NIH 3T3 cells [3,4] and expression of antisense IRS-1 RNA inhibits cell growth [5]. IRS-1 is constitutively activated in various tumors including breast cancers, sarcomas, Wilm’s tumors, and adrenal cortical carcinomas; in contrast, transfection of dominant-negative IRS-1 greatly reduces tumor growth [6]. RS-1 has 18 tyrosine residues that serve as docking sites for downstream signaling molecules such as phosphatidylinositol 3-kinase (PI 3-K), phosphotyrosine phosphatase SHP2, and the adapter proteins Grb2, Nck, Crk and Shb [2]. IRS-1 also has more than 30 serine/threonine phosphorylation sites. IRS-1 serine/threonine phosphorylation prevents its tyrosine phosphorylation and inhibits insulin receptor or IGF-IR signaling. For example, phosphorylation of Ser307 on mouse IRS-1 (the equivalent of Ser 312 on human IRS-1) prevents the interaction
770
B. Kim et al. / Cellular Signalling 17 (2005) 769–775
of IRS-1 with the insulin receptor and induces insulin resistance [7,8]. IRS-1 is phosphorylated at serine/threonine residues by multiple kinases including c-Jun Nterminal kinase [9], inhibitor nB kinase complex [10], Rho kinase [11] and PKC ~ [12]. PI3-K/Akt/mTOR pathway [13,14] and the suppressor of cytokine signaling (SOCS) family of proteins [15] also play pivotal roles in serine/threonine phosphorylation. Chronic stimulation by insulin/IGF-I or tumor necrosis factor (TNF)-a results in increased phosphorylation at inhibitory serine/threonine residues [8]. After serine/threonine phosphorylation, IRS-1 is degraded by a ubiquitin/proteasome pathway [16,17], although some reports also show calpain-dependent degradation [18,19]. Our laboratory studies apoptosis in primary and transformed neuronal cells. We have shown that high concentrations of glucose or mannitol induce an apoptotic cell death in dorsal root ganglia and glial cells, as well as in SH-SY5Y and SH-EP human neuroblastoma cells [20– 23]. Mannitol-induced apoptosis in SH-EP cells is accompanied by dephosphorylation and degradation of focal adhesion kinase (FAK), disruption of actin stress fibers, loss of focal adhesion sites, and cell detachment, all of which are prevented by IGF-I treatment, overexpression of IGF-IR or caspase inhibition [24]. SH-EP cells also undergo apoptosis induced by treatment with okadaic acid (OA), an inhibitor of the serine/threonine protein phosphatases 1 and 2A. OA-induced apoptosis also involves FAK dephosphorylation and degradation, morphological changes, cell detachment and caspase activation [25]. However, unlike mannitol-induced apoptosis, there is a long lag between FAK dephosphorylation/ cell detachment and FAK degradation/apoptosis. OAinduced apoptosis is not prevented by IGF-I treatment or caspase-inhibition. Several studies reported that IRS proteins play important roles during apoptosis [26,27]. In this study, we investigated the degradation of IRS-1 during mannitol- and OA-induced apoptosis. Mannitol treatment results in time- and concentration-dependent degradation of IRS-1 without any apparent molecular weight shift. The disappearance of IRS-1 is accompanied by the appearance of smaller degradation fragments, suggesting IRS-1 is cleaved by caspases. In contrast, OA-induced degradation of IRS-1 results in an upward mobility shift of IRS-1 indicating IRS-1 serine/threonine phosphorylation. IGF-I treatment, caspase inhibition, and proteasome inhibition all block mannitol-induced IRS-1 degradation. Consistent with our previous study of apoptosis [25], OA-induced degradation of IRS-1 is not prevented by IGF-I or inhibitors of caspase or proteasome. These results show that IRS-1 can be degraded by at least two independent pathways depending on the apoptotic stimuli; one pathway is caspase dependent and can be prevented by IGF-I and the other pathway is caspase-independent and IGF-I-insensitive.
2. Materials and methods 2.1. Materials Anti-IRS-1 and anti-actin antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Antibodies against Shc and integrin h-1 were purchased from Transduction Laboratory (Lexington, KY) and antibodies against Akt and mitogen-activated protein kinase (MAPK) were from Cell Signaling Technology (Beverly, MA). Anti-caspase-3 antibody was from BD Pharmingen (San Diego, CA). OA and proteasome inhibitors were purchased from Calbiochem (La Jolla, CA). Caspase inhibitors were from Enzyme System Products (Livermore, CA). Recombinant human IGF-I was kindly provided by Cephalon (Westchester, PA). 2.2. Cell culture SH-EP human neuroblastoma cells were maintained in DMEM containing 10% calf serum as previously described [28]. The cells were serum starved 4–6 h before treatment with mannitol or OA with or without IGF-I. The inhibitors were added 30 min before mannitol or OA treatment. 2.3. Immunoblotting and immunoprecipitation Immunoblotting was performed as described previously [28]. All experiments were repeated at least three times except Fig. 4B which was repeated twice. Typical representative results are presented in the figures.
3. Results 3.1. IRS-1 is degraded during mannitol- and OA-induced apoptosis We have recently shown that treatment of SH-EP human neuroblastoma cells with mannitol or OA results in apoptosis with the characteristics of anoikis, including actin stress fiber disruption and cell detachment [24,25]. Apoptosis was accompanied by the degradation of several signaling molecules including FAK and Akt, which serve important roles in IGF-I signaling. Since several reports suggested the involvement of IRS proteins during apoptosis [19,26,27], we examined changes in IRS-1 during mannitoland OA-induced apoptosis. When serum-starved SH-EP cells are treated with 300 mM mannitol there is a timedependent disappearance of IRS-1 at 160 kDa (Fig. 1A, arrow). The degradation is noticeable from 1.5 h after mannitol treatment, the time when most of the cells are rounding up and starting to show membrane blebbing [24]. IRS-1 degradation is almost complete after 3 h when the majority of cells are detached. The degradation of the primary IRS-1 band is accompanied by the appearance of
B. Kim et al. / Cellular Signalling 17 (2005) 769–775
771
mannitol-induced degradation, OA-induced IRS-1 degradation is accompanied by an upward mobility shift, suggesting serine/threonine phosphorylation (Fig. 2A). We could not detect any smaller fragments during degradation as in mannitol treatment, even with analysis with higher percent PAGE (Fig. 3). The effect of OA is also time- and concentration-dependent with almost complete disappearance of the IRS-1 band at 4 h and 250 nM OA (Fig. 2B and C). The effect is specific to IRS-1 since there are essentially no changes in levels of Akt, MAPK, actin, or integrin h1 (Figs. 2A and 3). We compared IRS-1 band densities over time in treated SH-EP cells with untreated cells using the Scion Image Program (Frederick, Maryland). The half life of IRS-1 degradation is 1.5 h and 4 h for mannitol and OA treatment, respectively. 3.2. IGF-I prevents mannitol-induced, but not OA-induced IRS-1 degradation
Fig. 1. Mannitol treatment induces the degradation of IRS-1. SH-EP neuroblastoma cells are serum starved for 5 h and then treated with 300 mM mannitol for the indicated times (A) or for 4 h with increasing concentrations of mannitol (B). Equal amounts of protein from whole cell lysates are separated by SDS-PAGE, transferred to nitrocellulose and analyzed by immunoblotting with the anti-IRS-1 antibody. Arrows indicate intact IRS-1 and arrowheads indicate the fragments.
distinct smaller fragments around 150, 120, 95, 65 and 35 kDa (arrowheads). The degradation of IRS-1 is also mannitol concentration-dependent, with maximum degradation at 500 mM (Fig. 1B). However, mannitol has no effect on actin or integrin h1 levels (Figs. 3 and 4A). OA treatment also induces IRS-1 degradation; however, in contrast to
We have shown that IGF-I prevents mannitol-induced, but not OA-induced, apoptosis in SH-EP cells [24,25]. Therefore, we next examined the effect of IGF-I on IRS-1 degradation. In agreement with our previous findings [29], when SH-EP cells are treated with 10 nM IGF-I for 4 h there are no changes in IRS-1 levels (Fig. 3). Mannitol-induced degradation of IRS-1 is almost completely blocked by simultaneous treatment of 10 nM IGF-I (Fig. 3). In contrast, OA-induced IRS-1 degradation cannot be prevented by IGF-I treatment. Treatment of mannitol, OA, and IGF-I has no effect on the expression of h1 integrin or actin (Fig. 3, bottom panels). 3.3. Effect of inhibition of caspase or proteasome activity on IRS-1 degradation Caspases are important mediators of apoptosis [30]. Caspase-3 is synthesized as an inactive precursor that is proteolytically cleaved to generate an active enzyme [31]. We have shown that caspases are involved in neuronal apoptosis [23,32,33] and FAK is degraded in a caspasedependent manner during mannitol-induced apoptosis [24].
Fig. 2. OA-induced degradation of IRS-1 is accompanied with mobility shift. (A) Serum-starved SH-EP cells are treated with 250 nM OA for 4 h. Whole cell lysates were analyzed by immunoblotting with the indicated antibodies. The effect of the time course (B) and concentration (C) of OA treatment are analyzed in serum-starved SH-EP cells using anti-IRS-1 antibody.
772
B. Kim et al. / Cellular Signalling 17 (2005) 769–775
caspase type-specific inhibitors are only partially effective in preventing mannitol-induced IRS-1 degradation (Fig. 4B). Among proteasome inhibitors, MG132 is most effective followed by epoxomycine and PS1; lactacystine is not effective at all (Fig. 4C). In contrast, ZVAD and MG132 cannot prevent OA-induced IRS-1 degradation (Fig. 4D). The effects of proteasome inhibitors are closely related to the cleavage of caspase-3. In agreement with our previous results [24], mannitol treatment induces caspase-3 cleavage, which is blocked by IGF-I addition (Fig. 5A). Like the effect on IRS-1 degradation, MG132 and epoxomycine partially prevent caspase-3 cleavage; however, lactacystine is ineffective. As expected, ZVAD blocks mannitol-induced caspase-3 cleavage (Fig. 5B).
4. Discussion In this study, we report the differential degradation of IRS-1 during mannitol- and OA-induced apoptosis. Our previous work shows that treatment of SH-EP neuro-
Fig. 3. IGF-I prevents mannitol-induced, but not OA-induced, IRS-1 degradation. Serum-starved SH-EP cells are treated with 300 mM mannitol or 250 nM OA for 4 h in the absence or presence of 10 nM IGF-I. Equal amounts of whole cell lysates were separated by SDS-PAGE, transferred to nitrocellulose, and analyzed by immunoblotting with the anti-IRS-1 antibody. The arrowhead indicates the fragments appearing during mannitol-induced degradation. The blot is stripped and reprobed with anti-integrin h1 and anti-acting antibodies (lower panels).
IRS-1 undergoes ubiquitin/proteasome-dependent degradation during insulin, IGF-I or TNF-a treatment [8,16,17]. Our previous report showed that caspase and ubiquitin/ proteasome pathways are activated within 4 h after mannitol, but not OA, treatment [24,25]. Depending on the cell type or stimuli, proteasome inhibitors can act as either pro-apoptotic or anti-apoptotic agents by activating or inhibiting caspase activation [34]. When our cells are incubated with ZVAD, a pan-caspase inhibitor, mannitolinduced IRS-1 degradation is greatly reduced (Fig. 4A). The proteasome inhibitor MG132 also blocks mannitol-induced degradation of IRS-1; however, the effect is not as complete as IGF-I or ZVAD treatment, since we can still observe the existence of smaller IRS-1 fragments (arrowhead). Other
Fig. 4. Inhibitors of caspase and proteasome blocks mannitol-induced IRS-1 degradation. Serum-starved SH-EP cells are treated with 300 mM mannitol with or without 10 nM IGF-I, 20 AM MG132, or 20 AM ZVAD for 4 h (A). The specificity of caspase inhibitors (B) and proteasome inhibitors (C) are also tested. ZVAD and MG132 cannot block OA-induced IRS-1 degradation (D). Equal amounts of protein from whole cell lysates are analyzed by immunoblotting with the indicated antibodies. Caspase and proteasome inhibitors are used at 20 AM. Lact, lactacystine; Epox, epoxomycine; MG, MG132.
B. Kim et al. / Cellular Signalling 17 (2005) 769–775
Fig. 5. Proteasome inhibitors prevent caspase-3 activation. Serum-starved SH-EP cells are treated with 300 mM mannitol with or without 10 nM IGF-I, or indicated inhibitors for 4 h. The whole cell lysates are analyzed for caspase-3 cleavage by anti-caspase-3 antibody.
blastoma cells with either mannitol or OA induces anoikislike apoptosis [24,25]. Disruption of actin filaments, dephosphorylation and degradation of FAK, loss of focal contacts, and simultaneous activation of caspase and ubiquitin/proteasome pathways all accompany mannitolinduced apoptosis. These changes are prevented by IGF-I [24]. OA treatment also induces similar morphological changes and cell detachment; however, apoptosis with the activation of the caspase and ubiquitin/proteasome pathways is evident only at much later time points and is not protected by IGF-I treatment [25]. In this study, we extend this work, finding that IRS-1 degradation follows a similar pattern; IGF-I treatment and inhibition of caspase and proteasome activation prevent mannitol-induced degradation, whereas OA-induced degradation is not reversed by these treatments. These results illustrate that two distinct pathways are engaged in IRS-1 degradation, which in turn suggests that different strategies must be considered for targeting IRS-1 for therapeutic applications. IRS proteins play critical roles in insulin or IGF-I signaling [2,35]. Manipulation of IRS-1 or -2 levels results in a diabetic phenotype in vivo and in vitro. IRS proteins also regulate cellular differentiation and proliferation. IRS-1 is constitutively activated in many tumors [6] and overexpression of IRS-1 results in neoplastic transformation [3,4]. Recently, involvement of IRS proteins in apoptosis has been reported. IRS-1 levels are decreased in a calpaindependent manner during thapsigargin-induced apoptosis in CPTX human prostate epithelial cell lines [19]. The level of IRS-2 is directly proportional to the degree of apoptosis in
773
pancreatic h cells [26] and IRS-1 knock-out cells are more sensitive to serum-starvation induced apoptosis [27]. Moreover, IRS-1 and -2 bind to the anti-apoptotic protein Bcl-2 and up-regulate cell survival [36]. In our current report, we show the degradation of IRS-1 during mannitol- and OAinduced apoptosis. Even though the pathways involved in the degradation are different, these results suggest the possible involvement of IRS-1 in apoptosis in neuroblastoma cells. IRS-1 may in part regulate mannitol- and OA-induced apoptosis through its interaction with integrins. Both mannitol and OA treatment result in morphological and biochemical changes typical to anoikis [24,25]. Anoikis is an apoptotic process which occurs in cells denied cellmatrix or cell–cell attachment [37]. IRS-1 associates with integrins and regulates cell adhesion [38–40]. In this report, we show that the apparent changes in IRS-1 expression levels after mannitol treatment start to occur after 60 min. OA treatment initiates IRS-1 degradation between 2 and 4 h. Interestingly, these are the time points at which we observe a majority of cells rounding up and starting to detach [24,25]. Even though IRS-1 degradation may not be the sole instigator of apoptosis, the possible dissociation of integrin-IRS-1 after degradation may enhance the apoptotic process by facilitating cell detachment. This cell detachment would then initiate additional anoikis-related apoptotic pathways, thereby ensuring cell death. IRS-1 degradation is caspase-dependent during mannitol-induced apoptosis. During the preparation of this manuscript, Green et al. [41] published a paper supporting our results of caspase-mediated cleavage of IRS. They showed that caspase-10, but none of the other caspases, is responsible for the degradation of IRS-1 and -2 during the apoptosis of mammary epithelial cells. Our results show that of the four caspase-specific inhibitors we tested LEHD (caspase-9 specific inhibitor) is most effective other than general caspase inhibitor ZVAD. Therefore, it is possible that different caspases are involved in IRS degradation depending on cell type or apoptotic stimulus. Even though we show that proteasome inhibitors block IRS-1 degradation, the proteasome is likely not directly involved in IRS-1 degradation. Proteasome-mediated degradation requires prior ubiquitination of the target proteins [42]. IRS-1 is degraded by a ubiquitination/ proteasome-dependent pathway during insulin, IGF-I or TNF-a treatment, which is preceded by serine/threonine phosphorylation [8,16,17,43]. In our study, we do not detect a mobility shift in IRS-1 by mannitol treatment, suggesting a lack of serine/threonine phosphorylation. We also show that proteasome inhibitors only partially prevent IRS-1 degradation, which is well correlated with the level of cleaved (i.e. active) caspase-3. Proteasome inhibitors either enhance or inhibit apoptosis depending on the cell type and stimuli [34]. This specificity depends on whether proteasome inhibitors activate or inhibit caspase cleavage, which may explain our discrepancy with a previous study
774
B. Kim et al. / Cellular Signalling 17 (2005) 769–775
where sorbitol-induced IRS-1 degradation is not prevented by lactacystine in 3T3-L1 adipocytes [44]. Therefore, we suggest that proteasome inhibitors regulate caspase activation, which cleaves IRS-1. Indeed we found that there are several potential caspase cleavage sites in IRS-1. We are currently investigating whether the mutation of these sites can block mannitol-induced IRS-1 degradation. In contrast to mannitol, OA-induced IRS-1 degradation is not prevented by the inhibition of caspases or the proteasome, suggesting the involvement of a completely different pathway. Recently, Zhang et al. [19] have shown that IRS-1 is degraded by calcium-activated protease calpain, and this process cannot be prevented by caspase or proteasome inhibitors. Chi et al. [45] have shown that overexpression of calpastatin, an endogenous calpain inhibitor, prevents OA-induced apoptosis in human fibroblasts. However, as with proteasome inhibition, blocking calpain is either pro-apoptotic or anti-apoptotic depending on cell type and stimuli [46]. Another possible pathway responsible for OA-induced IRS-1 degradation involves lysosomal proteases. Recent studies show that lysosomal proteases actively participate in apoptosis as well as necrosis [47]. When mouse hepatocytes are treated with TNF-a, stimulation of the apoptotic pathway requires the release of cathepsin B from lysosomes [48]. This multistep process results in a long lag between TNF-a stimulation and the execution of apoptosis. Interestingly, we have also observed that there is a temporal discrepancy between OA treatment, morphological changes, and the start of apoptosis [25]. In summary, we have shown in this report that IRS-1 is degraded by at least two completely different pathways during mannitol- and OA-induced apoptosis. Mannitolinduced IRS-1 degradation is dependent on caspase activation and prevented by IGF-I treatment. IRS-1 degradation by OA treatment is independent of caspase or proteasome pathways and cannot be rescued by IGF-I treatment.
Acknowledgements The authors thank Ms. Judy Boldt for expert secretarial assistance. This work was supported by grants from the National Institutes of Health (NS38849 and NS36778), the Juvenile Diabetes Research Foundation Center for the Study of Complications in Diabetes, and the Program for Understanding Neurological Diseases (PFUND).
References [1] M.F. White, Am. J. Physiol.: Endocrinol. Metab. 283 (2002) E413. [2] M.G. Myers Jr., M.F. White, Annu. Rev. Pharmacol. Toxicol. 36 (1996) 615. [3] S. Tanaka, T. Ito, J.R. Wands, J. Biol. Chem. 271 (1996) 14610.
[4] T. Ito, Y. Sasaki, J.R. Wands, Mol. Cell. Biol. 16 (1996) 943. [5] M. Taouis, J. Dupont, A. Gillet, M. Derouet, J. Simon, Mol. Cell. Endocrinol. 137 (1998) 177. [6] Q. Chang, Y. Li, M.F. White, J.A. Fletcher, S. Xiao, Cancer Res. 62 (2002) 6035. [7] V. Aguirre, E.D. Werner, J. Giraud, Y.H. Lee, S.E. Shoelson, M.F. White, J. Biol. Chem. 277 (2002) 1531. [8] L. Rui, V. Aguirre, J.K. Kim, G.I. Shulman, A. Lee, A. Corbould, A. Dunaif, M.F. White, J. Clin. Invest. 107 (2001) 181. [9] Y.H. Lee, J. Giraud, R.J. Davis, M.F. White, J. Biol. Chem. 278 (2003) 2896. [10] Z. Gao, D. Hwang, F. Bataille, M. Lefevre, D. York, M.J. Quon, J. Ye, J. Biol. Chem. 277 (2002) 48115. [11] N. Begum, O.A. Sandu, M. Ito, S.M. Lohmann, A. Smolenski, J. Biol. Chem. 277 (2002) 6214. [12] Y.F. Liu, K. Paz, A. Herschkovitz, A. Alt, T. Tennenbaum, S.R. Sampson, M. Ohba, T. Kuroki, D. LeRoith, Y. Zick, J. Biol. Chem. 276 (2001) 14459. [13] O.N. Ozes, H. Akca, L.D. Mayo, J.A. Gustin, T. Maehama, J.E. Dixon, D.B. Donner, Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 4640. [14] J. Li, K. DeFea, R.A. Roth, J. Biol. Chem. 274 (1999) 9351. [15] L. Rui, M. Yuan, D. Frantz, S. Shoelson, M.F. White, J. Biol. Chem. 277 (2002) 42394. [16] R. Zhande, J.J. Mitchell, J. Wu, X.J. Sun, Mol. Cell. Biol. 22 (2002) 1016. [17] A.V. Lee, J.L. Gooch, S. Oesterreich, R.L. Guler, D. Yee, Mol. Cell. Biol. 20 (2000) 1489. [18] L.K. Smith, K.M. Rice, C.W. Garner, Mol. Cell. Endocrinol. 122 (1996) 81. [19] H. Zhang, H. Hoff, C. Sell, Exp. Cell Res. 289 (2003) 352. [20] J.W. Russell, E.L. Feldman, Horm. Metab. Res. 31 (1999) 90. [21] C.C. Matthews, H. Odeh, E.L. Feldman, Neuroscience 79 (1997) 525. [22] C.M. van Golen, E.L. Feldman, J. Cell. Physiol. 182 (2000) 24. [23] C.L. Delaney, J.W. Russell, H.-L. Cheng, E.L. Feldman, J. Neuropathol. Exp. Neurol. 60 (2001) 147. [24] B. Kim, E.L. Feldman, J. Biol. Chem. 277 (2002) 27393. [25] B. Kim, C.M. van Golen, E.L. Feldman, Neoplasia 5 (2003) 405. [26] M.K. Lingohr, L.M. Dickson, C.E. Wrede, I. Briaud, J.F. McCuaig, M.G. Myers Jr., C.J. Rhodes, Mol. Cell. Endocrinol. 209 (2003) 17. [27] Y.H. Tseng, K. Ueki, K.M. Kriauciunas, C.R. Kahn, J. Biol. Chem. 277 (2002) 31601. [28] P.S. Leventhal, E.A. Shelden, B. Kim, E.L. Feldman, J. Biol. Chem. 272 (1997) 5214. [29] B. Kim, C. van Golen, EL. Feldman, Oncogene 23 (2004) 130. [30] H.Y. Chang, X. Yang, Microbiol. Mol. Biol. Rev. 64 (2000) 821. [31] D.W. Nicholson, A. Ali, N.A. Thornberry, J.P. Vaillancourt, C.K. Ding, M. Gallant, Y. Gareau, P.R. Griffin, M. Labelle, Y.A. Lazebnik, N.A. Munday, S.M. Raju, M.E. Smulson, T.-T. Yamin, V.L. Yu, D.K. Miller, Nature 376 (1995) 37. [32] J.W. Russell, A.J. Windebank, A. Schenone, E.L. Feldman, J. Neurobiol. 36 (1998) 455. [33] C.M. van Golen, V.P. Castle, E.L. Feldman, Cell Death Differ. 7 (2000) 654. [34] C. Wojcik, Cell. Mol. Life Sci. 56 (1999) 908. [35] D. LeRoith, C.T. Roberts Jr., Cancer Lett. 195 (2003) 127. [36] H. Ueno, E. Kondo, R. Yamamoto-Honda, K. Tobe, T. Nakamoto, K. Sasaki, K. Mitani, A. Furusaka, T. Tanaka, Y. Tsujimoto, T. Kadowaki, H. Hirai, Mol. Biol. Cell 11 (2000) 735. [37] S.M. Frisch, R.A. Screaton, Curr. Opin. Cell Biol. 13 (2001) 555. [38] K. Vuori, E. Ruoslahti, Science 266 (1994) 1576. [39] A. Guilherme, K. Torres, M.P. Czech, J. Biol. Chem. 273 (1998) 22899. [40] K. Reiss, J.Y. Wang, G. Romano, X. Tu, F. Peruzzi, R. Baserga, Oncogene 20 (2001) 490. [41] K.A. Green, M.J. Naylor, E.T. Lowe, P. Wang, E. Marshman, C.H. Streuli, J. Biol. Chem. 279 (2004) 25149.
B. Kim et al. / Cellular Signalling 17 (2005) 769–775 [42] V. Jesenberger, S. Jentsch, Nat. Rev., Mol. Cell Biol. 3 (2002) 112. [43] T.M. Pederson, D.L. Kramer, C.M. Rondinone, Diabetes 50 (2001) 24. [44] P. Gual, T. Gonzalez, T. Gremeaux, R. Barres, Y. Marchand-Brustel, J.F Tanti, J. Biol. Chem. 278 (2003) 26550. [45] X.J. Chi, T. Hiwasa, M. Maki, S. Sugaya, J. Nomura, K. Kita, N. Suzuki, FEBS Lett. 459 (1999) 391.
775
[46] R.W. Neumar, Y.A. Xu, H. Gada, R.P. Guttmann, R. Siman, J. Biol. Chem. 278 (2003) 14162. [47] G.S. Salvesen, J. Clin. Invest. 107 (2001) 21. [48] M.E. Guicciardi, J. Deussing, H. Miyoshi, S.F. Bronk, P.A. Svingen, C. Peters, S.H. Kaufmann, G.J. Gores, J. Clin. Invest. 106 (2000) 1127.