- Email: [email protected]
ERK signaling regulates autophagy and apoptosis through the dual phosphorylation of glycogen synthase kinase 3β
Biochemical and Biophysical Research Communications 418 (2012) 759–764
Contents lists available at SciVerse ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Proteasome inhibition-induced p38 MAPK/ERK signaling regulates autophagy and apoptosis through the dual phosphorylation of glycogen synthase kinase 3b Cheol-Hee Choi a,b, Byung-Hoon Lee c, Sang-Gun Ahn d, Seon-Hee Oh a,⇑ a
Research Center for Resistant Cells, Chosun University, Seosuk-dong, Dong-gu, Gwangju 501-759, Republic of Korea Department of Pharmacology, College of Medicine, Chosun University, Seosuk-dong, Dong-gu, Gwangju 501-759, Republic of Korea c College of Pharmacy and Multiscreening Center for Drug Development, Seoul National University, Seoul 151-742, Republic of Korea d Department of Pathology, College of Dentistry, Chosun University, Gwangju 501-759, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 20 January 2012 Available online 28 January 2012 Keywords: GSK3 p70S6K MAPK MG132 Autophagy
a b s t r a c t Proteasome inhibition is a promising approach for cancer treatment; however, the underlying mechanisms involved have not been fully elucidated. Here, we show that proteasome inhibition-induced p38 mitogen-activated protein kinase regulates autophagy and apoptosis by modulating the phosphorylation status of glycogen synthase kinase 3b (GSK3b) and 70 kDa ribosomal S6 kinase (p70S6K). The treatment of MDA-MB-231 cells with MG132 induced endoplasmic reticulum stress through the induction of ATF6a, PERK phosphorylation, and CHOP, and apoptosis through the cleavage of Bax and procaspase-3. MG132 caused the phosphorylation of GSK3b at Ser9 and, to a lesser extent, Thr390, the dephosphorylation of p70S6K at Thr389, and the phosphorylation of p70S6K at Thr421 and Ser424. The specific p38 inhibitor SB203080 reduced the p-GSK3bSer9 and autophagy through the phosphorylation of p70S6KThr389; however, it augmented the levels of p-ERK, p-GSK3bThr390, and p-70S6KThr421/Ser424 induced by MG132, and increased apoptotic cell death. The GSK inhibitor SB216763, but not lithium, inhibited the MG132induced phosphorylation of p38, and the downstream signaling pathway was consistent with that in SB203580-treated cells. Taken together, our data show that proteasome inhibition regulates p38/ GSKSer9/p70S6KThr380 and ERK/GSK3bThr390/p70S6KThr421/Ser424 kinase signaling, which is involved in cell survival and cell death. Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction Protein homeostasis is critical to cellular processes involved in the maintenance of cell survival. The ubiquitin–proteasome system is the principal mechanism for the degradation of proteins involved in various cellular processes, including cell cycle progression, stress responses, and apoptosis [1]. The proteasome is also a part of the endoplasmic reticulum (ER)-associated protein degradation (ERAD) system, which removes abnormal proteins from the ER and mitigates ER stress [2]. Proteasome inhibition thus increases the levels of ubiquitin-protein conjugates and abnormal proteins and further impairs ER function, thereby increasing cytotoxicity [3–5]. Additionally, ER stress can activate other protein degradation systems, such as autophagy, to remove bulky abnormal proteins, indicating that ER stress-induced autophagy counterbalances ER stress and is involved in the protection of cells [6]. Dysregulated growth in cancer cells causes the accumulation of abnormal proteins, whose clearance requires a higher level of proteasome function than that in normal cells [7]. The molecular ⇑ Corresponding author. Fax: +82 62 233 6052. E-mail addresses: [email protected], [email protected] (S.-H. Oh). 0006-291X/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2012.01.095
mechanisms of proteasome inhibition have been thus explored as a novel therapeutic strategy for cancer treatment. Proteasome inhibition-induced cytotoxicity is caused by the production of reactive oxygen species [8], the activation of p38 [9], the downregulation of ERK and Akt/mTOR signaling [10], the activation of NF-jB [1], and the inhibition of GSK3b [11]. Proteasome inhibitors may prolong cell survival by conferring resistance to these effects. Bortezomib-induced heat shock protein 27 is associated with bortezomib resistance, which is reversed by p38 inhibition [12,13]. These data show that several signaling pathways are involved in proteasome inhibition-induced cytotoxicity, and they indicate an opposing effect of proteasome inhibition on cytotoxicity. However, the molecular mechanisms remain unclear. GSK3 is a ubiquitously expressed protein serine/threonine kinase involved in diverse cellular processes, including glycogen metabolism, Wnt signaling, the cell cycle, and apoptosis [14]. The critical role of GSK3b in ER stress-mediated apoptosis has been described in various cell lines. The ER stress inducers 6-hydroxydopamine, brefeldin A, thapsigargin, and tunicamycin were shown to induce caspase-dependent or CHOP-mediated apoptosis, which is regulated by GSK3 [15–19]. Although GSK3 is involved in diverse proapoptotic events, it also has anti-apoptotic effects that contribute to tumor
760
C.-H. Choi et al. / Biochemical and Biophysical Research Communications 418 (2012) 759–764
survival. It has been reported that inactivation of GSK3b, which may promote tumor cell survival, is responsible for a worse clinical outcome in breast cancer [20]. However, the molecular mechanism underlying the involvement of GSK3 in cancer treatment has not been defined. In this study, we investigated the involvement of GSK3 in MG132-induced proteasome inhibition in breast cancer cells, and we found that proteasome inhibition-induced cell survival and cell death are regulated by the phosphorylation status of different GSK3b residues through the compensatory activation of p38 and ERK.
A
MG132
B
0 0.25 0.5 1 2 (µM)
0
1
3
6
12 24 (h)
P-PERK ATF6a CHOP -actin
C
DMSO
12 h
24h
Sub-G1: 8.31 G0/G1: 59.11 S: 9.05 G2/M: 17.43
Sub-G1: 26.23 G0/G1: 36.50 S: 11.54 G2/M: 17.39
Sub-G1: 1.39 G0/G1: 51.81 S: 15.29 G2/M: 16.48
2. Materials and methods 2.1. Cell culture and chemicals MDA-MB-231 human breast epithelial cells were maintained at 37 °C in a 5% CO2/95% air humidified incubator in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum, 50 lg/ml penicillin, and 50 lg/ml streptomycin. 3-MA was obtained from Sigma (St. Louis, MO, USA). Rapamycin and SB203589 were obtained from Cell Signaling (Danvers, MA, USA). SB216763 was purchased Santa Cruz Biotechnology (Santa Cruz, CA, USA). Other chemicals used were of the highest grade available from Sigma.
D MG132
2.3. Immunoblot analysis Immunoblotting was performed as described previously [21]. Antibodies specific for Atg8/LC3, phospho (p)-ERK, p38, p-p38, mTOR, p-mTOR, GSK3a/b, p-GSK3Thr390, p-GSK3a/bSer9/Ser21, p-GSK3a/bTyr279/Tyr216, p70S6K, and p-p70S6KThr389 were purchased from Cell Signaling. Antibodies specific for p-p70S6KThr421/Ser424 were obtained from Epitomics (Burlingame, CA, USA). Antibodies against ERK2, CHOP, ATF6, p-PERK, p21, Bax, procaspase-3, and b-actin were purchased from Santa Cruz Biotechnology. Immobilized proteins were incubated with goat anti-mouse IgG and goat anti-rabbit IgG (Santa Cruz Biotechnology), and the signals were detected using a chemiluminescence kit (Amersham Biosciences, Amersham, UK). 2.4. Statistical analysis All experiments were repeated at least three times. The significance of the differences between the treatments and respective controls was analyzed using Student’s t-test. All values are expressed as the mean ± SD. 3. Results 3.1. MG132 induces apoptosis and cell cycle arrest in MDA-MB-231 cells To define the ER stress response in MDA-MB-231 cells, cells were treated with various concentrations of MG132 for 18 h or with 1 lM MG132 for different periods of time. Western blot analysis showed that the expression of ER stress sensor proteins, including ATF6a, p-PERK, and CHOP, was induced in a dose- and time-dependent manner (Fig. 1A and B). Flow cytometric analysis showed an increase in the percentage of cells in G0/G1 phase at
E
2 (µM)
0
1
3
6
12
24 (h)
p21 Bcl2 Bax Caspase-3 Parp-1
2.2. Cytotoxicity assay and flow cytometric analysis MTT assay for cell viability measurement, and the percentage of apoptotic cells and the percentage of cells in each phase of the cell cycle were determined as described previously [21].
0 0.25 0.5 1
29 kDa
29 kDa -actin
Fig. 1. Effect of MG132 on ER stress and apoptosis in MDA-MB-231cells. Cells were treated with increasing concentrations of MG132 for 18 h (A and D) or with 1 lM MG132 for different periods of time (B and E) and were then harvested, lysed, and analyzed by Western blotting for the indicated proteins. b-actin was used as a loading control. (C) Cells were treated with 1 lM MG132 for 12 or 24 h, harvested, stained with propidium iodide (5 lg/ml), and analyzed by flow cytometry. Three separate experiments showed similar results.
12 h (accompanied by a slight increase in the apoptotic sub-G1 population), as well as a decrease in the G0/G1 population and a marked increase in the sub-G1 population at 24 h (Fig. 1C), which is reflected by our results for p21 expression (Fig. 1E). MG132 induced apoptosis by reducing the level of Bcl2, full-length Bax, and procaspase-3, leading to PARP-1 cleavage (Fig. 1D and E). Time-course experiments showed a marked apoptotic sign at 12 h post-MG132 treatment, as evidenced by PARP-1 cleavage. These results indicate that MG132 cytotoxicity is dependent on ER-stress mediated apoptosis and cell cycle arrest. 3.2. Effects of MG132 on the phosphorylation of GSK3 and p70S6K, and LC3II levels MG132 activated p38 MAPK in a dose-dependent manner, while ERK expression remained at baseline levels, but slightly increased at 0.25 lM-treated cells (Fig. 2A). In a time-course experiment (Fig. 2B), p-p38 increased in a time-dependent manner. The level of p-ERK showed a biphasic pattern, gradually decreasing before beginning to recover after 6 h of treatment. GSK3a/b is activated as a result of phosphorylation at Tyr276/Tyr216 and inhibited by phosphorylation at Ser21/Ser9 [22]. To examine the effect of MG132 on GSK3 inhibition, GSK3 phosphorylation was assessed using antibodies against p-GSK3a/bser21/Ser9. MG132 strongly increased p-GSK3bSer9 levels, but had no obvious effect on the p-GSK3aSer21. Consistent with this, the levels of p-GSK3bTyr216/Tyr279 (Try279 showed a high basal level) were comparable in MG132treated and untreated cells. Interestingly, phosphorylation of GSK3b
761
C.-H. Choi et al. / Biochemical and Biophysical Research Communications 418 (2012) 759–764
A
0 0.25 0.5 1
B
2 (µM)
0
1
3
6
12 24 (h)
p-p38 T-p38 p-ERK ERK2 p-Ser21-GSK3 p-Ser9-GSK3 p-Try279-GSK3 p-Try216-GSK3 p-Thr390-GSK3 GSK3 -actin
C
0 0.25 0.5 1
D
2 (µM)
0
p-mTOR mTOR
1
3
6 12 24 (h)
E MG132 mTOR
p-Thr421/Ser424 -70S6K
p- Thr389 -70S6K p- Thr421/Ser424 -70S6K p70S6K
LC3I/II -actin
- +
p-mTOR
p-Thr389-70S6K
p70S6K
Rapamycin
- +
-actin
Fig. 2. Effects of MG132 on the phosphorylation of MAPK, GSK3, and autophagy induction. Cells were treated with increasing concentrations of MG132 for 18 h (A and C) or with 1 lM MG132 for different periods of time (B and D). (E) Cells treated with 100 nM rapamycin for 2 h were further treated with 1 lM MG132 for 18 h. After treatment, cells were harvested, lysed, and then analyzed by Western blotting with the indicated antibodies. b-actin was used as a loading control.
at Thr390 was also detected. The phosphorylation of GSK3b in MG132-treated cells was further confirmed by the results of a time-course experiment. The level of p-GSK3bSer9 increased in a time-dependent manner, but not p-GSK3aSer21. The level of GSK3a/bTyr279/Tyr216 was unaffected by MG132 treatment; however, p-GSK3bThr390 was detected from 6 h post-treatment. These data indicate that MG132-induced MAPK activation is correlated with the dual phosphorylation of GSK3b at Ser9 and Thr390. p70S6K regulates protein synthesis by phosphorylating ribosomal protein S6, which lies downstream of mTOR [23]. MG132 induced the dephosphorylation of p70S6KThr389 and mTOR, and the generation of LC3II in a dose- and time-dependent manner (Fig. 2C and D), indicating that MG132 induces autophagy via mTOR/p70S6KThr389 signaling. However, the level of p-p70S6KThr421/Ser424 was increased in a dose- and time-dependent manner. Thus, to determine which phosphorylation sites in p70S6K are regulated by mTOR, cells were treated with MG132 in the presence or absence of the mTOR inhibitor rapamycin (Fig. 2E). Rapamycin further increased the MG132-induced dephosphorylation of p70S6KThr389 and mTOR. However, p-p70S6KThr421/Ser424 levels were not altered in the presence of rapamycin, indicating that the phosphorylation of 70S6K at Thr389, but not Thr421/Ser424, is dependent on mTOR. These data indicate that MG132 induces mTOR-dependent and -independent p70S6K phosphorylation. 3.3. Role of MAPK in the MG132-induced phosphorylation of GSK3b and p70S6k Pharmacological inhibitors of p38 and ERK were used to evaluate the effects of MAPK on GSK3 and p70S6K. Cells were treated with the specific p38 inhibitor SB203580 and the specific MAPK kinase (MEK) inhibitor PD98059 for 2 h and then treated with MG132 for 18 h (Fig. 3A). SB203580 almost completely blocked
the MG132-induced p-GSK3bSer9, but did not affect p-GSK3a/ bTyr216/Tyr217 levels, indicating that the serine phosphorylation of GSK3b is regulated by p38. PD98059 did not affect the MG132-induced p-GSK3abSer9/Ser21. SB203580 further augmented the levels of p-GSK3bThr390 and p-p70S6KThr421/Ser424 in cells treated with MG132, and reversed the MG132-induced dephosphorylation of p-p70S6KThr389, indicating that a signaling pathway involving GSK3bThr390 and p70S6K Thr421/Ser424 is induced by the inactivation of p38. These data led us to examine the activation of ERK when p38 was inactivated. As expected, treatment with SB203580 increased the activation of ERK compared to MG132-treated cells, indicating that the function and activation of ERK are inhibited by p38. Consistent with this, PD98059 markedly inhibited the MG132-induced phosphorylation of GSK3b at Thr390 and p70S6K at Thr389. Next, the effect of MAPK on MG132-induced cytotoxicity was assessed by MTT assay (Fig. 3B). MG132-induced cytotoxicity was significantly increased in the presence of SB203580, and reduced in the presence of PD98059, but not significant (P < 0.423). Western blot analysis showed that the cleavage of procaspase-3 and Bax, and the level of CHOP enhanced in the presence of SB203580. PD98059 blocked the cleavage of procaspase-3 and Bax, but not CHOP. Next, we examined whether MG132-induced autophagy is regulated by MAPK. Consistent with the phosphorylation of p70S6K at Thr389 (Fig. 3A), LC3 conversion was partially reduced by SB203580 (Fig. 3C). The above data suggest that autophagy affects cell survival by activating p38/p-GSKSer9. To confirm this hypothesis, we treated cells with the autophagy inhibitor 3-methyladenine (3-MA) and MG132, and then assessed the degree of cytotoxicity by MTT assay (Fig. 3D). As expected, 3-MA enhanced MG132-induced cytotoxicity, and increased the MG132-induced cleavage of procaspase-3 and Bax (Fig. 3E). These data indicate that the activation of p38 seems to regulate the phosphorylation status of GSK3b and
762
C.-H. Choi et al. / Biochemical and Biophysical Research Communications 418 (2012) 759–764
A
B
C
D
E
Fig. 3. Modulation of autophagy and apoptosis by MAPK-mediated GSK3. Cells pretreated with SB205309 (10 lM) or PD98059 (10 lM) for 2 h were further treated with 1 lM MG132 for 18 h for Western blot analysis (A and C), and treated for 24 h for MTT assay to measure cell viability (B). The data are presented as the mean ± SD of the percent increase relative to untreated control cells in three independent experiments. ⁄⁄P < 0.0002, ⁄P < 0.005. (D) Cells pretreated with 10 mM 3-MA for 2 h and treated with 1 lM MG132 for a further 24 h for MTT assay (D), and treated with 1 lM MG132 for 18 h and then analyzed by Western blotting for the indicated proteins (E). The data are presented as the mean ± SD of the percent increase relative to untreated control cells in three independent experiments. b-actin was used as a loading control. ⁄⁄P < 0.0002, ⁄ P < 0.02.
p70S6K through ERK downregulation and cell survival during proteasome inhibition. 3.4. The effect of GSK-3 inhibitors on MG132-mediated proteasome inhibition The above data suggest that cellular toxicity caused by proteasome inhibition is regulated by p38 and ERK. Molecular crosstalk between p38 and ERK was further confirmed using the GSK inhibitors SB216763 and lithium. The effects of GSK3 inhibition on cytotoxicity was quantified by MTT assay. SB216763 treatment further enhanced MG132-induced cell death, while lithium treatment tended to inhibit it (P < 0.169) (Fig. 4A). Western blot analysis showed that 10 lM SB216763 almost completely blocked the MG132-induced p-GSK3a/bTyr216/Tyr217 (Fig. 4B). SB216763 treatment abrogated the MG132-induced p-GSK3bSer9 and induced the p-GSK3bThr390. Furthermore, SB216763 treatment attenuated p-p38 levels while inducing strong phosphorylation of ERK and enhanced the p-p70S6KThr421/Ser424, and it subsequently inhibited MG132-induced LC3 conversion, and augmented the cleavage of procaspase-3 and CHOP expression. Ten mM lithium treatment did not affect the phosphorylation of GSK3a/b at serine, tyrosine, or threonine residues, and it had no effect on the phosphorylation of MAPK or p70S6K. These results indicate that the compensatory activation of p38 and ERK regulates proteasome inhibition-induced autophagy and apoptosis through phosphorylation status of GSK3b and p70S6K, as well as show the possibility for interaction between p38 and p-GSK3bSer9.
4. Discussion The present study demonstrated that proteasome inhibition induces the phosphorylation of GSK3b at Ser9 and Thr390 through the compensatory activation of p38 MAPK and ERK. p38/p-GSK3bSer9 and ERK/p-GSK3bThr390 signaling causes the dephosphorylation of
B
A * 120 100 80 60 40 20 0
**
MG132
-
SB216763 Lithium + - + - +
p-p38 p38 p-ERK ERK2 p-Ser21-GSK3 p-Ser9-GSK3 p-Try279-GSK3 p-Try216-GSK3 p-Thr390-GSK3 GSK3 p- Thr421/Ser424 -70S6K 70S6K LC3I/II Procaspase-3 CHOP -actin
Fig. 4. Effects of GSK3 inhibitors on the phosphorylation of MAPK and p70S6K and MG132-induced cytotoxicity. Cells pretreated with SB216763 (10 lM) or lithium (10 mM) for 2 h were treated with 1 lM MG132 for a further 24 h for MTT assay (A), or treated for 18 h (B). The data are presented as the mean ± SD of the percent increase relative to untreated control cells in three independent experiments. ⁄ P < 0.002, ⁄⁄P < 0.0002. b-actin was used as a loading control.
p70S6KThr389 and phosphorylation of p70S6K at Thr421/Ser424, and it appears to regulate autophagy and apoptosis. MG132 causes apoptosis through diverse signaling pathways, including mitochondrion-dependent caspase activation and ER stress [1]. In the present study, MDA-MB-231 cells treated with MG132 showed Bax and caspase-3cleavage, and subsequent PARP-1 cleavage, as well as ER stress-mediated expression of the proapoptotic protein CHOP. Our results together with those of
C.-H. Choi et al. / Biochemical and Biophysical Research Communications 418 (2012) 759–764
previous reports indicate that ER stress-mediated, mitochondriondependent apoptosis is involved in MG132-induced cytotoxicity. Autophagy, a bulk protein degradation system, can effectively remove accumulated proteins and it may play a critical role in cell protection [6]. MDA-MB-231 cells treated with MG132 showed dose- and time-dependent LC3II induction, which may contribute to cell protection. The inhibition of LC3II conversion by treatment with 3-MA, and the blockade of p38 using SB203580 and SB216763, increased procaspase-3 cleavage and CHOP levels, and subsequently increased cell death compared to MG132-treated cells. Together with previous reports, the present data link proteasome inhibition to apoptosis and autophagy. To better understand proteasome inhibition-induced cytotoxicity, it is necessary to define the downstream effects of proteasome inhibition. The involvement of MAPK proteins, including ERK, p38, and Jun N-terminal kinases (JNK), in proteasome inhibition has been reported. In Na2 neural cells, ERK and JNK differentially regulated the toxic effects of proteasome inhibition: ERK activation ameliorated neurotoxicity, while JNK promoted cell death [24]. In MG132treated human pulmonary fibroblast cells, JNK is involved in MG132-induced growth inhibition and death, while p38 is involved in cell death [9]. Furthermore, MG132-induced apoptosis is dependent on the activation of p38 and JNK in renal cell carcinoma [25] and on ERK in HT22 mouse hippocampal cells [26]. In contrast, bortezomib-induced p38 activation is involved in cell survival/resistance [13,27]. This discrepancy suggests that the roles of MAPK are cell-type specific and depend on the experimental system used. In the present study, MG132 induced sustained activation of p38, while the level of phosphorylated ERK returned to basal levels at later time points. Furthermore, the inhibition of p38 by SB203580 or SB216763 induced the phosphorylation of ERK, indicating that full activation of p38 requires ERK inactivation or downregulation. Thus, identification of the downstream targets of p38 will be critical for defining the different roles of MG132-induced MAPK. GSK3 can be regulated by different signaling pathways, including those involving PKA, PKB, p90RSK, and p70S6K [14]. Although the best-characterized mechanism of GSK3b inactivation is its phosphorylation at Ser9 by Akt [22], the treatment of MDA-MB-321 cells with MG132 did not induce Akt phosphorylation (data not shown). Thornton [28] described the p38-mediated phosphorylation of GSK3b at Thr390 as an alternative pathway of GSK3b inactivation, which is involved in cell survival. Hepatitis B virus-X protein-mediated ERK activation regulates the phosphorylation of GSK3b at Ser9 [29]. In the present study, MG132 treatment led to the phosphorylation of GSK3b at Ser9 and, to a lesser extent, at Thr390. The inactivation of p38 by SB203580 or SB216763 was correlated with ERK phosphorylation, and strongly inhibited the phosphorylation of GSK3b at Ser9, but markedly enhanced its phosphorylation at Thr390. Furthermore, treatment with the MEK inhibitor PD98059 did not affect the MG132-induced phosphorylation of GSK3b at Ser9, but markedly inhibited its phosphorylation at Thr390. Thus, our data indicate that MAPK regulates the dual phosphorylation of GSK3b; p38 targets p-GSK3bSer9 and ERK likely targets p-GSK3bThr390. The mitogen-activated serine/threonine kinase p70S6K plays a critical role in cell growth and survival [30], and its activation is tightly regulated by phosphorylation at multiple serine and threonine residues [31]. The present study demonstrated for the first time that proteasome inhibition induces the phosphorylation of p70S6K at Thr421/Ser424 and dephosphorylation of p70S6KThr389. p70S6KThr389 has been shown to be regulated by mTOR [32], and, consistent with this, the extent of MG132-mediated dephosphorylation of p70S6KThr389 was increased in the presence of rapamycin. MG132 increased the phosphorylation of p70S6K at Thr421/Ser424 in a dose- and time-dependent manner. This response was not
763
inhibited by rapamycin, indicating that the phosphorylation of p70S6K at Thr421/Ser424 is independent of rapamycin and highlighting the dual role of p70S6K in proteasome inhibition. In fact, it has been reported that the phosphorylation of p70S6K at different residues (Thr389 and Thr421/Ser424) is dependent on developmental stage [33] and growth status [34]. Thus, we attempted to determine the molecular mechanisms of phosphorylation by the kinase. MG132-induced LC3II generation occurred through the dephosphorylation of mTOR and p70S6KThr389 signaling, which was reversed by SB203580 and which increased cell death, indicating that p70S6KThr389 modulates MG132-induced autophagy downstream of p38. However, the MG132-induced phosphorylation of p70S6K at Thr421/Ser424 was further augmented when p38 was inactivated, a response accompanied by the phosphorylation of p-ERK, but was completely inhibited by PD98059, suggesting that the phosphorylation of p70S6K at Thr421/Ser424 is regulated by ERK and is rapamycin-independent. Consistent with this, Lehman [35] reported that rapamycin-resistant p70S6KThr421/Ser424 phosphorylation was dependent on ERK. Furthermore, the GSK3 inhibitor SB216763 augmented the MG132-induced phosphorylation of p70S6K at Thr421/Ser424 corresponding with p-ERK induction. However, in the present study, we did not identify the downstream target of ERK/p70S6KThr421/Ser424. MG132 causes a cell cycle arrest through the induction of p21 [36]. The blockade of p70S6K by neutralizing antibodies or rapamycin prevents cell cycle progression through G1 [37]. In the present study, MG132 induced p21 accumulation in a dose- and time-dependent manner. However, it remains to be confirmed whether proteasome inhibitionmediated p21 induction is correlated with the phosphorylation of p70S6K at Thr421/Ser424. The results of the present study show that the dual phosphorylation of GSK3 is regulated by the compensatory activation of p38 and ERK, which appears to play a significant role in MG132-induced autophagy and cell death. Proteasome inhibition promotes p38 signaling, the activation of which appears to have different effects on proteasome inhibition-induced toxicity as a result of the phosphorylation of GSK3b and p70S6K at different residues. p38 MAPK regulates the phosphorylation of GSK3b at Ser9 and the dephosphorylation of GSK3bThr389, and it subsequently induces the production of LC3II. ERK activation was augmented in cells treated with the p38 inhibitor SB203580, which also increased the phosphorylation of GSK at Thr390 and p70S6K at Thr421/Ser421, and cell death. Acknowledgments This work was supported by National Research Foundation of Korea (NRF) Grant funded by the Ministry of Education, Science and Technology (MEST) through the Research Center for Resistant Cells (R13-2003-009). References [1] L.J. Crawford, B. Walker, A.E. Irvine, Proteasome inhibitors in cancer therapy, J. Cell Commun. Signal. 5 (2011) 101–110. [2] B. Meusser, C. Hirsch, E. Jarosch, et al., ERAD: the long road to destruction, Nat. Cell Biol. 7 (2005) 766–772. [3] A. Szokalska, M. Makowski, D. Nowis, Proteasome inhibition potentiates antitumor effects of photodynamic therapy in mice through induction of endoplasmic reticulum stress and unfolded protein response, Cancer Res. 69 (2009) 4235–4243. [4] D.H. Hyun, M. Lee, B. Halliwell, et al., Proteasomal inhibition causes the formation of protein aggregates containing a wide range of proteins, including nitrated proteins, J. Neurochem. 86 (2003) 363–373. [5] D. Chen, Q.P. Dou, The ubiquitin-proteasome system as a prospective molecular target for cancer treatment and prevention, Curr. Protein Pept. Sci. 11 (2010) 459–470. [6] Y. Cheng, J.M. Yang, Survival and death of endoplasmic-reticulum-stressed cells: role of autophagy, World J. Biol. Chem. 2 (2011) 226–231. [7] J. Adams, The development of proteasome inhibitors as anticancer drugs, Cancer Cell 5 (2004) 417–421.
764
C.-H. Choi et al. / Biochemical and Biophysical Research Communications 418 (2012) 759–764
[8] P. Perez-Galan, G. Roué, N. Villamor, et al., The proteasome inhibitor bortezomib induces apoptosis in mantle-cell lymphoma through generation of ROS and Noxa activation independent of p53 status, Blood 107 (2006) 257– 264. [9] W.H. Park, Mitogen-activated protein kinase inhibitors differently affect the growth inhibition and death of a proteasome inhibitor, carbobenzoxy-Leu-Leuleucinal-treated human pulmonary fibroblast cells, Hum. Exp. Toxicol. 30 (2011) 1945–1954. [10] J.K. Ko, C.H. Choi, Y.K. Kim, et al., The proteasome inhibitor MG-132 induces AIF nuclear translocation through down-regulation of ERK and Akt/mTOR pathway, Neurochem. Res. 36 (2011) 722–731. [11] S.J. Stoll, S.C. Pitt, H. Chen, Follicular thyroid cancer cell growth inhibition by proteosome inhibitor MG132, J. Surg. Res. 156 (2009) 39–44. [12] P.G. Richardson, B. Barlogie, J. Berenson, et al., A phase 2 study of bortezomib in relapsed, refractory myeloma, N. Engl. J. Med. 348 (2003) 2609–2617. [13] T. Hideshima, K. Podar, D. Chauhan, et al., p38 MAPK inhibition enhances PS341 (bortezomib)-induced cytotoxicity against multiple myeloma cells, Oncogene 23 (2004) 8766–8776. [14] C.A. Grimes, R.S. Jope, The multifaceted roles of glycogen synthase kinase 3b in cellular, Signaling Prog. Neurobiol. 65 (2001) 391–426. [15] G. Chen, K.A. Bower, C. Ma, et al., Glycogen synthase kinase 3beta (GSK3beta) mediates 6-hydroxydopamine-induced neuronal death, FASEB J. 18 (2004) 1162–1164. [16] J.L. Brewster, D.A. Linseman, R.J. Bouchard, et al., Endoplasmic reticulum stress and trophic factor withdrawal activate distinct signaling cascades that induce glycogen synthase kinase-3 and a caspase-9-dependent apoptosis in cerebellar granule neurons, Mol. Cell Neurosci. 32 (2006) 242–253. [17] W.C. Huang, Y.S. Lin, C.L. Chen, et al., Glycogen Synthase Kinase-3b mediates endoplasmic reticulum stress-induced lysosomal apoptosis in leukemia, J. Pharmacol. Exp. Therapeutics 329 (2009) 524–531. [18] T. Takadera, M. Fujibayashi, H. Kaniyu, et al., Caspase-dependent apoptosis induced by thapsigargin was prevented by glycogen synthase kinase-3 inhibitors in cultured rat cortical neurons, Neurochem. Res. 32 (2007) 1336– 1342. [19] G.P. Meares, Glycogen synthase kinase-3 regulates endoplasmic reticulum (ER) stress-induced CHOP expression in neuronal cells, Exp. Cell Res. 317 (2011) 1621–1628. [20] H. Armanious, J. Deschenes, P. Gelebart, et al., Clinical and biological significance of GSK-3b inactivation in breast cancer-an immunohistochemical study, Hum. Pathol. 41 (2010) 1657–1663. [21] C.H. Choi, Y.K. Jung, S.H. Oh, Selective induction of catalase-mediated autophagy by dihydrocapsaicin in lung cell lines, Free Radical Bio. Med. 49 (2010) 245–257. [22] P. Cohen, S. Frame, The renaissance of GSK3, Nat. Rev. Mol. Cell Biol. 2 (2001) 769–776.
[23] H. Kawasome, P. Papst, S. Webb, et al., Targeted disruption of p70(s6k) defines its role in protein synthesis and rapamycin sensitivity, Proc. Natl. Acad. Sci. USA 95 (1998) 5033–5038. [24] L. Zhang, P.J. Ebenezer, K. Dasuri, et al., Ebenezer Proteasome inhibition modulates kinase activation in neural cells: relevance to ubiquitination, ribosomes, and survival, J. Neurosci. Res. 87 (2009) 3231–3238. [25] J. Ishizawa, S. Yoshida, M. Oya, et al., Inhibition of the ubiquitin-proteasome pathway activates stress kinases and induces apoptosis in renal cancer cells, Int. J. Oncol. 25 (2004) 697–702. [26] M. Stanciu, D.B. DeFranco, Prolonged nuclear retention of activated extracellular signal-regulated protein kinase promotes cell death generated by oxidative toxicity or proteasome inhibition in a neuronal cell line, J. Biol. Chem. 277 (2002) 4010–4017. [27] Y.Y. Shi, G.W. Small, R.Z. Orlowski, Proteasome inhibitors induce a p38 mitogen-activated protein kinase (MAPK)-dependent anti-apoptotic program involving MAPK phosphatase-1 and Akt in models of breast cancer, Breast Cancer Res. Treat. 100 (2006) 33–47. [28] T.M. Thornton, G. Pedraza-Alva, B. Deng, et al., Phosphorylation by p38 MAPK as an alternative pathway for GSK3b inactivation, Science 320 (2008) 667–670. [29] Q. Ding, W. Xia, J.C. Liu, et al., Erk associates with and rimes GSK-3b for its inactivation resulting in upregulation of b-catenin, Mol. Cell 19 (2005) 159– 170. [30] T.R. Fenton, I.T. Gout, Functions and regulation of the 70kDa ribosomal S6 kinases, Int. J. Biochem. Cell Biol. 43 (2011) 47–59. [31] T.J. Ragan, D.B. Ross, M.M. Keshwani, et al., Expression, purification, and characterization of a structurally disordered and functional C-terminal autoinhibitory domain (AID) of the 70 kDa 40S ribosomal protein S6 kinase1 (S6K1), Protein Expres. Purif. 57 (2008) 271–279. [32] B.A. Moser, P.B. Dennis, N. Pullen, et al., Dual requirement for a newly identified phosphorylation site in p70s6k, Mol. Cell Biol. 17 (1997) 5648–5655. [33] A. Musnier, D. Heitzler, T. Boulo, et al., Developmental regulation of p70 S6 kinase by a G protein-coupled receptor dynamically modelized in primary cells, Cell. Mol. Life Sci. 66 (2009) 3487–3503. [34] S. Duchêne, E. Audouin, C. Berri, et al., Tissue-specific regulation of S6K1 by insulin in chickens divergently selected for growth, Gen. Comp. Endocrinol. 156 (2008) 190–198. [35] J.H. Lehman, V. Calvo, J. Gomez-Cambronero, Mechanism of ribosomal p70S6 kinase activation by granulocyte macrophage colony-stimulating factor in neutrophils: cooperation of a MEK-related, THR421/SERr424 kinase and a rapamycin-sensitive, mTOR-related THR389 kinase, J. Biol. Chem. 278 (2003) 28130–28138. [36] M. Li, X. Dong, Y. Liu, et al., Inhibition of ubiquitin proteasome function suppresses proliferation of pulmonary artery smooth muscle cells, NaunynSchmiedeberg’s Arch. Pharmacol. 384 (2011) 517–523. [37] H.A. Lane, A. Fernandez, N.J. Lamb, et al., P70s6k function is essential for G1 progression, Nature 363 (1993) 170–172.
Contents lists available at SciVerse ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Proteasome inhibition-induced p38 MAPK/ERK signaling regulates autophagy and apoptosis through the dual phosphorylation of glycogen synthase kinase 3b Cheol-Hee Choi a,b, Byung-Hoon Lee c, Sang-Gun Ahn d, Seon-Hee Oh a,⇑ a
Research Center for Resistant Cells, Chosun University, Seosuk-dong, Dong-gu, Gwangju 501-759, Republic of Korea Department of Pharmacology, College of Medicine, Chosun University, Seosuk-dong, Dong-gu, Gwangju 501-759, Republic of Korea c College of Pharmacy and Multiscreening Center for Drug Development, Seoul National University, Seoul 151-742, Republic of Korea d Department of Pathology, College of Dentistry, Chosun University, Gwangju 501-759, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 20 January 2012 Available online 28 January 2012 Keywords: GSK3 p70S6K MAPK MG132 Autophagy
a b s t r a c t Proteasome inhibition is a promising approach for cancer treatment; however, the underlying mechanisms involved have not been fully elucidated. Here, we show that proteasome inhibition-induced p38 mitogen-activated protein kinase regulates autophagy and apoptosis by modulating the phosphorylation status of glycogen synthase kinase 3b (GSK3b) and 70 kDa ribosomal S6 kinase (p70S6K). The treatment of MDA-MB-231 cells with MG132 induced endoplasmic reticulum stress through the induction of ATF6a, PERK phosphorylation, and CHOP, and apoptosis through the cleavage of Bax and procaspase-3. MG132 caused the phosphorylation of GSK3b at Ser9 and, to a lesser extent, Thr390, the dephosphorylation of p70S6K at Thr389, and the phosphorylation of p70S6K at Thr421 and Ser424. The specific p38 inhibitor SB203080 reduced the p-GSK3bSer9 and autophagy through the phosphorylation of p70S6KThr389; however, it augmented the levels of p-ERK, p-GSK3bThr390, and p-70S6KThr421/Ser424 induced by MG132, and increased apoptotic cell death. The GSK inhibitor SB216763, but not lithium, inhibited the MG132induced phosphorylation of p38, and the downstream signaling pathway was consistent with that in SB203580-treated cells. Taken together, our data show that proteasome inhibition regulates p38/ GSKSer9/p70S6KThr380 and ERK/GSK3bThr390/p70S6KThr421/Ser424 kinase signaling, which is involved in cell survival and cell death. Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction Protein homeostasis is critical to cellular processes involved in the maintenance of cell survival. The ubiquitin–proteasome system is the principal mechanism for the degradation of proteins involved in various cellular processes, including cell cycle progression, stress responses, and apoptosis [1]. The proteasome is also a part of the endoplasmic reticulum (ER)-associated protein degradation (ERAD) system, which removes abnormal proteins from the ER and mitigates ER stress [2]. Proteasome inhibition thus increases the levels of ubiquitin-protein conjugates and abnormal proteins and further impairs ER function, thereby increasing cytotoxicity [3–5]. Additionally, ER stress can activate other protein degradation systems, such as autophagy, to remove bulky abnormal proteins, indicating that ER stress-induced autophagy counterbalances ER stress and is involved in the protection of cells [6]. Dysregulated growth in cancer cells causes the accumulation of abnormal proteins, whose clearance requires a higher level of proteasome function than that in normal cells [7]. The molecular ⇑ Corresponding author. Fax: +82 62 233 6052. E-mail addresses: [email protected], [email protected] (S.-H. Oh). 0006-291X/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2012.01.095
mechanisms of proteasome inhibition have been thus explored as a novel therapeutic strategy for cancer treatment. Proteasome inhibition-induced cytotoxicity is caused by the production of reactive oxygen species [8], the activation of p38 [9], the downregulation of ERK and Akt/mTOR signaling [10], the activation of NF-jB [1], and the inhibition of GSK3b [11]. Proteasome inhibitors may prolong cell survival by conferring resistance to these effects. Bortezomib-induced heat shock protein 27 is associated with bortezomib resistance, which is reversed by p38 inhibition [12,13]. These data show that several signaling pathways are involved in proteasome inhibition-induced cytotoxicity, and they indicate an opposing effect of proteasome inhibition on cytotoxicity. However, the molecular mechanisms remain unclear. GSK3 is a ubiquitously expressed protein serine/threonine kinase involved in diverse cellular processes, including glycogen metabolism, Wnt signaling, the cell cycle, and apoptosis [14]. The critical role of GSK3b in ER stress-mediated apoptosis has been described in various cell lines. The ER stress inducers 6-hydroxydopamine, brefeldin A, thapsigargin, and tunicamycin were shown to induce caspase-dependent or CHOP-mediated apoptosis, which is regulated by GSK3 [15–19]. Although GSK3 is involved in diverse proapoptotic events, it also has anti-apoptotic effects that contribute to tumor
760
C.-H. Choi et al. / Biochemical and Biophysical Research Communications 418 (2012) 759–764
survival. It has been reported that inactivation of GSK3b, which may promote tumor cell survival, is responsible for a worse clinical outcome in breast cancer [20]. However, the molecular mechanism underlying the involvement of GSK3 in cancer treatment has not been defined. In this study, we investigated the involvement of GSK3 in MG132-induced proteasome inhibition in breast cancer cells, and we found that proteasome inhibition-induced cell survival and cell death are regulated by the phosphorylation status of different GSK3b residues through the compensatory activation of p38 and ERK.
A
MG132
B
0 0.25 0.5 1 2 (µM)
0
1
3
6
12 24 (h)
P-PERK ATF6a CHOP -actin
C
DMSO
12 h
24h
Sub-G1: 8.31 G0/G1: 59.11 S: 9.05 G2/M: 17.43
Sub-G1: 26.23 G0/G1: 36.50 S: 11.54 G2/M: 17.39
Sub-G1: 1.39 G0/G1: 51.81 S: 15.29 G2/M: 16.48
2. Materials and methods 2.1. Cell culture and chemicals MDA-MB-231 human breast epithelial cells were maintained at 37 °C in a 5% CO2/95% air humidified incubator in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum, 50 lg/ml penicillin, and 50 lg/ml streptomycin. 3-MA was obtained from Sigma (St. Louis, MO, USA). Rapamycin and SB203589 were obtained from Cell Signaling (Danvers, MA, USA). SB216763 was purchased Santa Cruz Biotechnology (Santa Cruz, CA, USA). Other chemicals used were of the highest grade available from Sigma.
D MG132
2.3. Immunoblot analysis Immunoblotting was performed as described previously [21]. Antibodies specific for Atg8/LC3, phospho (p)-ERK, p38, p-p38, mTOR, p-mTOR, GSK3a/b, p-GSK3Thr390, p-GSK3a/bSer9/Ser21, p-GSK3a/bTyr279/Tyr216, p70S6K, and p-p70S6KThr389 were purchased from Cell Signaling. Antibodies specific for p-p70S6KThr421/Ser424 were obtained from Epitomics (Burlingame, CA, USA). Antibodies against ERK2, CHOP, ATF6, p-PERK, p21, Bax, procaspase-3, and b-actin were purchased from Santa Cruz Biotechnology. Immobilized proteins were incubated with goat anti-mouse IgG and goat anti-rabbit IgG (Santa Cruz Biotechnology), and the signals were detected using a chemiluminescence kit (Amersham Biosciences, Amersham, UK). 2.4. Statistical analysis All experiments were repeated at least three times. The significance of the differences between the treatments and respective controls was analyzed using Student’s t-test. All values are expressed as the mean ± SD. 3. Results 3.1. MG132 induces apoptosis and cell cycle arrest in MDA-MB-231 cells To define the ER stress response in MDA-MB-231 cells, cells were treated with various concentrations of MG132 for 18 h or with 1 lM MG132 for different periods of time. Western blot analysis showed that the expression of ER stress sensor proteins, including ATF6a, p-PERK, and CHOP, was induced in a dose- and time-dependent manner (Fig. 1A and B). Flow cytometric analysis showed an increase in the percentage of cells in G0/G1 phase at
E
2 (µM)
0
1
3
6
12
24 (h)
p21 Bcl2 Bax Caspase-3 Parp-1
2.2. Cytotoxicity assay and flow cytometric analysis MTT assay for cell viability measurement, and the percentage of apoptotic cells and the percentage of cells in each phase of the cell cycle were determined as described previously [21].
0 0.25 0.5 1
29 kDa
29 kDa -actin
Fig. 1. Effect of MG132 on ER stress and apoptosis in MDA-MB-231cells. Cells were treated with increasing concentrations of MG132 for 18 h (A and D) or with 1 lM MG132 for different periods of time (B and E) and were then harvested, lysed, and analyzed by Western blotting for the indicated proteins. b-actin was used as a loading control. (C) Cells were treated with 1 lM MG132 for 12 or 24 h, harvested, stained with propidium iodide (5 lg/ml), and analyzed by flow cytometry. Three separate experiments showed similar results.
12 h (accompanied by a slight increase in the apoptotic sub-G1 population), as well as a decrease in the G0/G1 population and a marked increase in the sub-G1 population at 24 h (Fig. 1C), which is reflected by our results for p21 expression (Fig. 1E). MG132 induced apoptosis by reducing the level of Bcl2, full-length Bax, and procaspase-3, leading to PARP-1 cleavage (Fig. 1D and E). Time-course experiments showed a marked apoptotic sign at 12 h post-MG132 treatment, as evidenced by PARP-1 cleavage. These results indicate that MG132 cytotoxicity is dependent on ER-stress mediated apoptosis and cell cycle arrest. 3.2. Effects of MG132 on the phosphorylation of GSK3 and p70S6K, and LC3II levels MG132 activated p38 MAPK in a dose-dependent manner, while ERK expression remained at baseline levels, but slightly increased at 0.25 lM-treated cells (Fig. 2A). In a time-course experiment (Fig. 2B), p-p38 increased in a time-dependent manner. The level of p-ERK showed a biphasic pattern, gradually decreasing before beginning to recover after 6 h of treatment. GSK3a/b is activated as a result of phosphorylation at Tyr276/Tyr216 and inhibited by phosphorylation at Ser21/Ser9 [22]. To examine the effect of MG132 on GSK3 inhibition, GSK3 phosphorylation was assessed using antibodies against p-GSK3a/bser21/Ser9. MG132 strongly increased p-GSK3bSer9 levels, but had no obvious effect on the p-GSK3aSer21. Consistent with this, the levels of p-GSK3bTyr216/Tyr279 (Try279 showed a high basal level) were comparable in MG132treated and untreated cells. Interestingly, phosphorylation of GSK3b
761
C.-H. Choi et al. / Biochemical and Biophysical Research Communications 418 (2012) 759–764
A
0 0.25 0.5 1
B
2 (µM)
0
1
3
6
12 24 (h)
p-p38 T-p38 p-ERK ERK2 p-Ser21-GSK3 p-Ser9-GSK3 p-Try279-GSK3 p-Try216-GSK3 p-Thr390-GSK3 GSK3 -actin
C
0 0.25 0.5 1
D
2 (µM)
0
p-mTOR mTOR
1
3
6 12 24 (h)
E MG132 mTOR
p-Thr421/Ser424 -70S6K
p- Thr389 -70S6K p- Thr421/Ser424 -70S6K p70S6K
LC3I/II -actin
- +
p-mTOR
p-Thr389-70S6K
p70S6K
Rapamycin
- +
-actin
Fig. 2. Effects of MG132 on the phosphorylation of MAPK, GSK3, and autophagy induction. Cells were treated with increasing concentrations of MG132 for 18 h (A and C) or with 1 lM MG132 for different periods of time (B and D). (E) Cells treated with 100 nM rapamycin for 2 h were further treated with 1 lM MG132 for 18 h. After treatment, cells were harvested, lysed, and then analyzed by Western blotting with the indicated antibodies. b-actin was used as a loading control.
at Thr390 was also detected. The phosphorylation of GSK3b in MG132-treated cells was further confirmed by the results of a time-course experiment. The level of p-GSK3bSer9 increased in a time-dependent manner, but not p-GSK3aSer21. The level of GSK3a/bTyr279/Tyr216 was unaffected by MG132 treatment; however, p-GSK3bThr390 was detected from 6 h post-treatment. These data indicate that MG132-induced MAPK activation is correlated with the dual phosphorylation of GSK3b at Ser9 and Thr390. p70S6K regulates protein synthesis by phosphorylating ribosomal protein S6, which lies downstream of mTOR [23]. MG132 induced the dephosphorylation of p70S6KThr389 and mTOR, and the generation of LC3II in a dose- and time-dependent manner (Fig. 2C and D), indicating that MG132 induces autophagy via mTOR/p70S6KThr389 signaling. However, the level of p-p70S6KThr421/Ser424 was increased in a dose- and time-dependent manner. Thus, to determine which phosphorylation sites in p70S6K are regulated by mTOR, cells were treated with MG132 in the presence or absence of the mTOR inhibitor rapamycin (Fig. 2E). Rapamycin further increased the MG132-induced dephosphorylation of p70S6KThr389 and mTOR. However, p-p70S6KThr421/Ser424 levels were not altered in the presence of rapamycin, indicating that the phosphorylation of 70S6K at Thr389, but not Thr421/Ser424, is dependent on mTOR. These data indicate that MG132 induces mTOR-dependent and -independent p70S6K phosphorylation. 3.3. Role of MAPK in the MG132-induced phosphorylation of GSK3b and p70S6k Pharmacological inhibitors of p38 and ERK were used to evaluate the effects of MAPK on GSK3 and p70S6K. Cells were treated with the specific p38 inhibitor SB203580 and the specific MAPK kinase (MEK) inhibitor PD98059 for 2 h and then treated with MG132 for 18 h (Fig. 3A). SB203580 almost completely blocked
the MG132-induced p-GSK3bSer9, but did not affect p-GSK3a/ bTyr216/Tyr217 levels, indicating that the serine phosphorylation of GSK3b is regulated by p38. PD98059 did not affect the MG132-induced p-GSK3abSer9/Ser21. SB203580 further augmented the levels of p-GSK3bThr390 and p-p70S6KThr421/Ser424 in cells treated with MG132, and reversed the MG132-induced dephosphorylation of p-p70S6KThr389, indicating that a signaling pathway involving GSK3bThr390 and p70S6K Thr421/Ser424 is induced by the inactivation of p38. These data led us to examine the activation of ERK when p38 was inactivated. As expected, treatment with SB203580 increased the activation of ERK compared to MG132-treated cells, indicating that the function and activation of ERK are inhibited by p38. Consistent with this, PD98059 markedly inhibited the MG132-induced phosphorylation of GSK3b at Thr390 and p70S6K at Thr389. Next, the effect of MAPK on MG132-induced cytotoxicity was assessed by MTT assay (Fig. 3B). MG132-induced cytotoxicity was significantly increased in the presence of SB203580, and reduced in the presence of PD98059, but not significant (P < 0.423). Western blot analysis showed that the cleavage of procaspase-3 and Bax, and the level of CHOP enhanced in the presence of SB203580. PD98059 blocked the cleavage of procaspase-3 and Bax, but not CHOP. Next, we examined whether MG132-induced autophagy is regulated by MAPK. Consistent with the phosphorylation of p70S6K at Thr389 (Fig. 3A), LC3 conversion was partially reduced by SB203580 (Fig. 3C). The above data suggest that autophagy affects cell survival by activating p38/p-GSKSer9. To confirm this hypothesis, we treated cells with the autophagy inhibitor 3-methyladenine (3-MA) and MG132, and then assessed the degree of cytotoxicity by MTT assay (Fig. 3D). As expected, 3-MA enhanced MG132-induced cytotoxicity, and increased the MG132-induced cleavage of procaspase-3 and Bax (Fig. 3E). These data indicate that the activation of p38 seems to regulate the phosphorylation status of GSK3b and
762
C.-H. Choi et al. / Biochemical and Biophysical Research Communications 418 (2012) 759–764
A
B
C
D
E
Fig. 3. Modulation of autophagy and apoptosis by MAPK-mediated GSK3. Cells pretreated with SB205309 (10 lM) or PD98059 (10 lM) for 2 h were further treated with 1 lM MG132 for 18 h for Western blot analysis (A and C), and treated for 24 h for MTT assay to measure cell viability (B). The data are presented as the mean ± SD of the percent increase relative to untreated control cells in three independent experiments. ⁄⁄P < 0.0002, ⁄P < 0.005. (D) Cells pretreated with 10 mM 3-MA for 2 h and treated with 1 lM MG132 for a further 24 h for MTT assay (D), and treated with 1 lM MG132 for 18 h and then analyzed by Western blotting for the indicated proteins (E). The data are presented as the mean ± SD of the percent increase relative to untreated control cells in three independent experiments. b-actin was used as a loading control. ⁄⁄P < 0.0002, ⁄ P < 0.02.
p70S6K through ERK downregulation and cell survival during proteasome inhibition. 3.4. The effect of GSK-3 inhibitors on MG132-mediated proteasome inhibition The above data suggest that cellular toxicity caused by proteasome inhibition is regulated by p38 and ERK. Molecular crosstalk between p38 and ERK was further confirmed using the GSK inhibitors SB216763 and lithium. The effects of GSK3 inhibition on cytotoxicity was quantified by MTT assay. SB216763 treatment further enhanced MG132-induced cell death, while lithium treatment tended to inhibit it (P < 0.169) (Fig. 4A). Western blot analysis showed that 10 lM SB216763 almost completely blocked the MG132-induced p-GSK3a/bTyr216/Tyr217 (Fig. 4B). SB216763 treatment abrogated the MG132-induced p-GSK3bSer9 and induced the p-GSK3bThr390. Furthermore, SB216763 treatment attenuated p-p38 levels while inducing strong phosphorylation of ERK and enhanced the p-p70S6KThr421/Ser424, and it subsequently inhibited MG132-induced LC3 conversion, and augmented the cleavage of procaspase-3 and CHOP expression. Ten mM lithium treatment did not affect the phosphorylation of GSK3a/b at serine, tyrosine, or threonine residues, and it had no effect on the phosphorylation of MAPK or p70S6K. These results indicate that the compensatory activation of p38 and ERK regulates proteasome inhibition-induced autophagy and apoptosis through phosphorylation status of GSK3b and p70S6K, as well as show the possibility for interaction between p38 and p-GSK3bSer9.
4. Discussion The present study demonstrated that proteasome inhibition induces the phosphorylation of GSK3b at Ser9 and Thr390 through the compensatory activation of p38 MAPK and ERK. p38/p-GSK3bSer9 and ERK/p-GSK3bThr390 signaling causes the dephosphorylation of
B
A * 120 100 80 60 40 20 0
**
MG132
-
SB216763 Lithium + - + - +
p-p38 p38 p-ERK ERK2 p-Ser21-GSK3 p-Ser9-GSK3 p-Try279-GSK3 p-Try216-GSK3 p-Thr390-GSK3 GSK3 p- Thr421/Ser424 -70S6K 70S6K LC3I/II Procaspase-3 CHOP -actin
Fig. 4. Effects of GSK3 inhibitors on the phosphorylation of MAPK and p70S6K and MG132-induced cytotoxicity. Cells pretreated with SB216763 (10 lM) or lithium (10 mM) for 2 h were treated with 1 lM MG132 for a further 24 h for MTT assay (A), or treated for 18 h (B). The data are presented as the mean ± SD of the percent increase relative to untreated control cells in three independent experiments. ⁄ P < 0.002, ⁄⁄P < 0.0002. b-actin was used as a loading control.
p70S6KThr389 and phosphorylation of p70S6K at Thr421/Ser424, and it appears to regulate autophagy and apoptosis. MG132 causes apoptosis through diverse signaling pathways, including mitochondrion-dependent caspase activation and ER stress [1]. In the present study, MDA-MB-231 cells treated with MG132 showed Bax and caspase-3cleavage, and subsequent PARP-1 cleavage, as well as ER stress-mediated expression of the proapoptotic protein CHOP. Our results together with those of
C.-H. Choi et al. / Biochemical and Biophysical Research Communications 418 (2012) 759–764
previous reports indicate that ER stress-mediated, mitochondriondependent apoptosis is involved in MG132-induced cytotoxicity. Autophagy, a bulk protein degradation system, can effectively remove accumulated proteins and it may play a critical role in cell protection [6]. MDA-MB-231 cells treated with MG132 showed dose- and time-dependent LC3II induction, which may contribute to cell protection. The inhibition of LC3II conversion by treatment with 3-MA, and the blockade of p38 using SB203580 and SB216763, increased procaspase-3 cleavage and CHOP levels, and subsequently increased cell death compared to MG132-treated cells. Together with previous reports, the present data link proteasome inhibition to apoptosis and autophagy. To better understand proteasome inhibition-induced cytotoxicity, it is necessary to define the downstream effects of proteasome inhibition. The involvement of MAPK proteins, including ERK, p38, and Jun N-terminal kinases (JNK), in proteasome inhibition has been reported. In Na2 neural cells, ERK and JNK differentially regulated the toxic effects of proteasome inhibition: ERK activation ameliorated neurotoxicity, while JNK promoted cell death [24]. In MG132treated human pulmonary fibroblast cells, JNK is involved in MG132-induced growth inhibition and death, while p38 is involved in cell death [9]. Furthermore, MG132-induced apoptosis is dependent on the activation of p38 and JNK in renal cell carcinoma [25] and on ERK in HT22 mouse hippocampal cells [26]. In contrast, bortezomib-induced p38 activation is involved in cell survival/resistance [13,27]. This discrepancy suggests that the roles of MAPK are cell-type specific and depend on the experimental system used. In the present study, MG132 induced sustained activation of p38, while the level of phosphorylated ERK returned to basal levels at later time points. Furthermore, the inhibition of p38 by SB203580 or SB216763 induced the phosphorylation of ERK, indicating that full activation of p38 requires ERK inactivation or downregulation. Thus, identification of the downstream targets of p38 will be critical for defining the different roles of MG132-induced MAPK. GSK3 can be regulated by different signaling pathways, including those involving PKA, PKB, p90RSK, and p70S6K [14]. Although the best-characterized mechanism of GSK3b inactivation is its phosphorylation at Ser9 by Akt [22], the treatment of MDA-MB-321 cells with MG132 did not induce Akt phosphorylation (data not shown). Thornton [28] described the p38-mediated phosphorylation of GSK3b at Thr390 as an alternative pathway of GSK3b inactivation, which is involved in cell survival. Hepatitis B virus-X protein-mediated ERK activation regulates the phosphorylation of GSK3b at Ser9 [29]. In the present study, MG132 treatment led to the phosphorylation of GSK3b at Ser9 and, to a lesser extent, at Thr390. The inactivation of p38 by SB203580 or SB216763 was correlated with ERK phosphorylation, and strongly inhibited the phosphorylation of GSK3b at Ser9, but markedly enhanced its phosphorylation at Thr390. Furthermore, treatment with the MEK inhibitor PD98059 did not affect the MG132-induced phosphorylation of GSK3b at Ser9, but markedly inhibited its phosphorylation at Thr390. Thus, our data indicate that MAPK regulates the dual phosphorylation of GSK3b; p38 targets p-GSK3bSer9 and ERK likely targets p-GSK3bThr390. The mitogen-activated serine/threonine kinase p70S6K plays a critical role in cell growth and survival [30], and its activation is tightly regulated by phosphorylation at multiple serine and threonine residues [31]. The present study demonstrated for the first time that proteasome inhibition induces the phosphorylation of p70S6K at Thr421/Ser424 and dephosphorylation of p70S6KThr389. p70S6KThr389 has been shown to be regulated by mTOR [32], and, consistent with this, the extent of MG132-mediated dephosphorylation of p70S6KThr389 was increased in the presence of rapamycin. MG132 increased the phosphorylation of p70S6K at Thr421/Ser424 in a dose- and time-dependent manner. This response was not
763
inhibited by rapamycin, indicating that the phosphorylation of p70S6K at Thr421/Ser424 is independent of rapamycin and highlighting the dual role of p70S6K in proteasome inhibition. In fact, it has been reported that the phosphorylation of p70S6K at different residues (Thr389 and Thr421/Ser424) is dependent on developmental stage [33] and growth status [34]. Thus, we attempted to determine the molecular mechanisms of phosphorylation by the kinase. MG132-induced LC3II generation occurred through the dephosphorylation of mTOR and p70S6KThr389 signaling, which was reversed by SB203580 and which increased cell death, indicating that p70S6KThr389 modulates MG132-induced autophagy downstream of p38. However, the MG132-induced phosphorylation of p70S6K at Thr421/Ser424 was further augmented when p38 was inactivated, a response accompanied by the phosphorylation of p-ERK, but was completely inhibited by PD98059, suggesting that the phosphorylation of p70S6K at Thr421/Ser424 is regulated by ERK and is rapamycin-independent. Consistent with this, Lehman [35] reported that rapamycin-resistant p70S6KThr421/Ser424 phosphorylation was dependent on ERK. Furthermore, the GSK3 inhibitor SB216763 augmented the MG132-induced phosphorylation of p70S6K at Thr421/Ser424 corresponding with p-ERK induction. However, in the present study, we did not identify the downstream target of ERK/p70S6KThr421/Ser424. MG132 causes a cell cycle arrest through the induction of p21 [36]. The blockade of p70S6K by neutralizing antibodies or rapamycin prevents cell cycle progression through G1 [37]. In the present study, MG132 induced p21 accumulation in a dose- and time-dependent manner. However, it remains to be confirmed whether proteasome inhibitionmediated p21 induction is correlated with the phosphorylation of p70S6K at Thr421/Ser424. The results of the present study show that the dual phosphorylation of GSK3 is regulated by the compensatory activation of p38 and ERK, which appears to play a significant role in MG132-induced autophagy and cell death. Proteasome inhibition promotes p38 signaling, the activation of which appears to have different effects on proteasome inhibition-induced toxicity as a result of the phosphorylation of GSK3b and p70S6K at different residues. p38 MAPK regulates the phosphorylation of GSK3b at Ser9 and the dephosphorylation of GSK3bThr389, and it subsequently induces the production of LC3II. ERK activation was augmented in cells treated with the p38 inhibitor SB203580, which also increased the phosphorylation of GSK at Thr390 and p70S6K at Thr421/Ser421, and cell death. Acknowledgments This work was supported by National Research Foundation of Korea (NRF) Grant funded by the Ministry of Education, Science and Technology (MEST) through the Research Center for Resistant Cells (R13-2003-009). References [1] L.J. Crawford, B. Walker, A.E. Irvine, Proteasome inhibitors in cancer therapy, J. Cell Commun. Signal. 5 (2011) 101–110. [2] B. Meusser, C. Hirsch, E. Jarosch, et al., ERAD: the long road to destruction, Nat. Cell Biol. 7 (2005) 766–772. [3] A. Szokalska, M. Makowski, D. Nowis, Proteasome inhibition potentiates antitumor effects of photodynamic therapy in mice through induction of endoplasmic reticulum stress and unfolded protein response, Cancer Res. 69 (2009) 4235–4243. [4] D.H. Hyun, M. Lee, B. Halliwell, et al., Proteasomal inhibition causes the formation of protein aggregates containing a wide range of proteins, including nitrated proteins, J. Neurochem. 86 (2003) 363–373. [5] D. Chen, Q.P. Dou, The ubiquitin-proteasome system as a prospective molecular target for cancer treatment and prevention, Curr. Protein Pept. Sci. 11 (2010) 459–470. [6] Y. Cheng, J.M. Yang, Survival and death of endoplasmic-reticulum-stressed cells: role of autophagy, World J. Biol. Chem. 2 (2011) 226–231. [7] J. Adams, The development of proteasome inhibitors as anticancer drugs, Cancer Cell 5 (2004) 417–421.
764
C.-H. Choi et al. / Biochemical and Biophysical Research Communications 418 (2012) 759–764
[8] P. Perez-Galan, G. Roué, N. Villamor, et al., The proteasome inhibitor bortezomib induces apoptosis in mantle-cell lymphoma through generation of ROS and Noxa activation independent of p53 status, Blood 107 (2006) 257– 264. [9] W.H. Park, Mitogen-activated protein kinase inhibitors differently affect the growth inhibition and death of a proteasome inhibitor, carbobenzoxy-Leu-Leuleucinal-treated human pulmonary fibroblast cells, Hum. Exp. Toxicol. 30 (2011) 1945–1954. [10] J.K. Ko, C.H. Choi, Y.K. Kim, et al., The proteasome inhibitor MG-132 induces AIF nuclear translocation through down-regulation of ERK and Akt/mTOR pathway, Neurochem. Res. 36 (2011) 722–731. [11] S.J. Stoll, S.C. Pitt, H. Chen, Follicular thyroid cancer cell growth inhibition by proteosome inhibitor MG132, J. Surg. Res. 156 (2009) 39–44. [12] P.G. Richardson, B. Barlogie, J. Berenson, et al., A phase 2 study of bortezomib in relapsed, refractory myeloma, N. Engl. J. Med. 348 (2003) 2609–2617. [13] T. Hideshima, K. Podar, D. Chauhan, et al., p38 MAPK inhibition enhances PS341 (bortezomib)-induced cytotoxicity against multiple myeloma cells, Oncogene 23 (2004) 8766–8776. [14] C.A. Grimes, R.S. Jope, The multifaceted roles of glycogen synthase kinase 3b in cellular, Signaling Prog. Neurobiol. 65 (2001) 391–426. [15] G. Chen, K.A. Bower, C. Ma, et al., Glycogen synthase kinase 3beta (GSK3beta) mediates 6-hydroxydopamine-induced neuronal death, FASEB J. 18 (2004) 1162–1164. [16] J.L. Brewster, D.A. Linseman, R.J. Bouchard, et al., Endoplasmic reticulum stress and trophic factor withdrawal activate distinct signaling cascades that induce glycogen synthase kinase-3 and a caspase-9-dependent apoptosis in cerebellar granule neurons, Mol. Cell Neurosci. 32 (2006) 242–253. [17] W.C. Huang, Y.S. Lin, C.L. Chen, et al., Glycogen Synthase Kinase-3b mediates endoplasmic reticulum stress-induced lysosomal apoptosis in leukemia, J. Pharmacol. Exp. Therapeutics 329 (2009) 524–531. [18] T. Takadera, M. Fujibayashi, H. Kaniyu, et al., Caspase-dependent apoptosis induced by thapsigargin was prevented by glycogen synthase kinase-3 inhibitors in cultured rat cortical neurons, Neurochem. Res. 32 (2007) 1336– 1342. [19] G.P. Meares, Glycogen synthase kinase-3 regulates endoplasmic reticulum (ER) stress-induced CHOP expression in neuronal cells, Exp. Cell Res. 317 (2011) 1621–1628. [20] H. Armanious, J. Deschenes, P. Gelebart, et al., Clinical and biological significance of GSK-3b inactivation in breast cancer-an immunohistochemical study, Hum. Pathol. 41 (2010) 1657–1663. [21] C.H. Choi, Y.K. Jung, S.H. Oh, Selective induction of catalase-mediated autophagy by dihydrocapsaicin in lung cell lines, Free Radical Bio. Med. 49 (2010) 245–257. [22] P. Cohen, S. Frame, The renaissance of GSK3, Nat. Rev. Mol. Cell Biol. 2 (2001) 769–776.
[23] H. Kawasome, P. Papst, S. Webb, et al., Targeted disruption of p70(s6k) defines its role in protein synthesis and rapamycin sensitivity, Proc. Natl. Acad. Sci. USA 95 (1998) 5033–5038. [24] L. Zhang, P.J. Ebenezer, K. Dasuri, et al., Ebenezer Proteasome inhibition modulates kinase activation in neural cells: relevance to ubiquitination, ribosomes, and survival, J. Neurosci. Res. 87 (2009) 3231–3238. [25] J. Ishizawa, S. Yoshida, M. Oya, et al., Inhibition of the ubiquitin-proteasome pathway activates stress kinases and induces apoptosis in renal cancer cells, Int. J. Oncol. 25 (2004) 697–702. [26] M. Stanciu, D.B. DeFranco, Prolonged nuclear retention of activated extracellular signal-regulated protein kinase promotes cell death generated by oxidative toxicity or proteasome inhibition in a neuronal cell line, J. Biol. Chem. 277 (2002) 4010–4017. [27] Y.Y. Shi, G.W. Small, R.Z. Orlowski, Proteasome inhibitors induce a p38 mitogen-activated protein kinase (MAPK)-dependent anti-apoptotic program involving MAPK phosphatase-1 and Akt in models of breast cancer, Breast Cancer Res. Treat. 100 (2006) 33–47. [28] T.M. Thornton, G. Pedraza-Alva, B. Deng, et al., Phosphorylation by p38 MAPK as an alternative pathway for GSK3b inactivation, Science 320 (2008) 667–670. [29] Q. Ding, W. Xia, J.C. Liu, et al., Erk associates with and rimes GSK-3b for its inactivation resulting in upregulation of b-catenin, Mol. Cell 19 (2005) 159– 170. [30] T.R. Fenton, I.T. Gout, Functions and regulation of the 70kDa ribosomal S6 kinases, Int. J. Biochem. Cell Biol. 43 (2011) 47–59. [31] T.J. Ragan, D.B. Ross, M.M. Keshwani, et al., Expression, purification, and characterization of a structurally disordered and functional C-terminal autoinhibitory domain (AID) of the 70 kDa 40S ribosomal protein S6 kinase1 (S6K1), Protein Expres. Purif. 57 (2008) 271–279. [32] B.A. Moser, P.B. Dennis, N. Pullen, et al., Dual requirement for a newly identified phosphorylation site in p70s6k, Mol. Cell Biol. 17 (1997) 5648–5655. [33] A. Musnier, D. Heitzler, T. Boulo, et al., Developmental regulation of p70 S6 kinase by a G protein-coupled receptor dynamically modelized in primary cells, Cell. Mol. Life Sci. 66 (2009) 3487–3503. [34] S. Duchêne, E. Audouin, C. Berri, et al., Tissue-specific regulation of S6K1 by insulin in chickens divergently selected for growth, Gen. Comp. Endocrinol. 156 (2008) 190–198. [35] J.H. Lehman, V. Calvo, J. Gomez-Cambronero, Mechanism of ribosomal p70S6 kinase activation by granulocyte macrophage colony-stimulating factor in neutrophils: cooperation of a MEK-related, THR421/SERr424 kinase and a rapamycin-sensitive, mTOR-related THR389 kinase, J. Biol. Chem. 278 (2003) 28130–28138. [36] M. Li, X. Dong, Y. Liu, et al., Inhibition of ubiquitin proteasome function suppresses proliferation of pulmonary artery smooth muscle cells, NaunynSchmiedeberg’s Arch. Pharmacol. 384 (2011) 517–523. [37] H.A. Lane, A. Fernandez, N.J. Lamb, et al., P70s6k function is essential for G1 progression, Nature 363 (1993) 170–172.