- Email: [email protected]
Resveratrol induces cellular senescence with attenuated mono-ubiquitination of histone H2B in glioma cells
Biochemical and Biophysical Research Communications 407 (2011) 271–276
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Resveratrol induces cellular senescence with attenuated mono-ubiquitination of histone H2B in glioma cells q Zhen Gao, Michael S. Xu, Tamara L. Barnett, C. Wilson Xu ⇑ Nevada Cancer Institute, Las Vegas, NV 89135, USA
a r t i c l e
i n f o
Article history: Received 1 February 2011
Keywords: Resveratrol Glioma Cellular senescence Mono-ubiquitinated histone H2B RNF20 Histone modifications
a b s t r a c t Resveratrol (3,40 ,5-trihydroxy-trans-stilbene), a polyphenol naturally occurring in grapes and other plants, has cancer chemo-preventive effects and therapeutic potential. Although resveratrol modulates multiple pathways in tumor cells, how resveratrol or its affected pathways converge on chromatin to mediate its effects is not known. Using glioma cells as a model, we showed here that resveratrol inhibited cell proliferation and induced cellular hypertrophy by transforming spindle-shaped cells to enlarged, irregular and flatten-shaped ones. We further showed that resveratrol-induced hypertrophic cells expressed senescence-associated-b-galactosidase, suggesting that resveratrol-induced cellular senescence in glioma cells. Consistent with these observations, we demonstrated that resveratrol inhibited clonogenic efficiencies in vitro and tumor growth in a xenograft model. Furthermore, we found that acute treatment of resveratrol inhibited mono-ubiquitination of histone H2B at K120 (uH2B) in breast, prostate, pancreatic, lung, brain tumor cells as well as primary human cells. Chronic treatment with low doses of resveratrol also inhibited uH2B in the resveratrol-induced senescent glioma cells. Moreover, we showed that depletion of RNF20, a ubiquitin ligase of histone H2B, inhibited uH2B and induced cellular senescence in glioma cells in vitro, thereby recapitulated the effects of resveratrol. Taken together, our results suggest that uH2B is a novel direct or indirect chromatin target of resveratrol and RNF20 plays an important role in inhibiting cellular senescence programs that are intact in glioma cells. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction Glioblastoma multiforme is the most malignant of the primary brain tumors with no effective treatment currently available [1] (http://www.cbtrus.org). Resveratrol, which is synthesized de novo in food and medicinal plants presumably in response to stress or infection, has been shown to inhibit proliferation in a variety of tumor cells such as skin, breast, prostate, and lung through a multiple pathways [2–4]. However, the effects of resveratrol have not been adequately evaluated in human glioma cells. It is also not clear whether resveratrol-modulated pathways converge on chromatin to mediate its inhibitory effects. Histone modifications are intimately involved in transcriptional regulation. Among histone acetylation, methylation, phosphoryla-
Abbreviations: RV, resveratrol; uH2B, mono-ubiquitinated histone H2B; GBM, glioblastoma multiforme; SA-b-gal, senescence-associated-b-galactosidase. q We thank Giuseppe Pizzorno, Hui Zhang, Hong Sun for providing tumor cell lines, Fei Lu for graphic analysis and members of the Xu Lab for discussion. The work is supported in part by the Stephen & Mary Birch Foundation. ⇑ Corresponding author. Address: Nevada Cancer Institute, One Breakthrough Way, Las Vegas, NV 89135, USA. E-mail address: [email protected] (C.W. Xu).
0006-291X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2011.02.008
tion and mono-ubiquitination, the role of mono-ubiquitination of histones is least understood, despite the fact that histone H2A is the first protein shown to be covalently conjugated with ubiquitin [5]. In budding yeast, H2B is the only histone that is mono-ubiquitinated at K123 (uH2B) [5], which constitutes approximately 15% of the total H2B during exponential phase, but is not detectable in stationary phase [6]. Bre1, a Rad6-associated RING finger protein, is the E3 ubiquitin ligase of H2B in yeast [7–9]. RNF20, a human homolog of yeast Bre1, has been identified in humans [7,10], suggesting that mono-ubiquitination of histone H2B is conserved from yeast to humans. Using mono-ubiquitination of histone H2B at K123 (uH2B) in yeast as a model, we have previously discovered that glucose induces uH2B through glycolysis, revealing a novel paradigm of nutritional regulation of histone modifications [6]. We have further demonstrated that glycolysis is also required for mono-ubiquitination of histone H2B at K120, the orthologous site of K123 of yeast histone H2B, in both human primary and tumor cells [11]. In this report, we found that resveratrol inhibited proliferation of human glioma cells by inducing cellular senescence. We further found that mono-ubiquitination of histone H2B at K120 (uH2B) was inhibited in resveratrol-induced senescent glioma cells. Moreover, we showed that like resveratrol treatment, shRNA depletion
272
Z. Gao et al. / Biochemical and Biophysical Research Communications 407 (2011) 271–276
of RNF20 inhibited uH2B and induced cellular senescence in glioma cells, demonstrating that loss of H2B ubiquitin ligase recapitulated the inhibitory effects of resveratrol. Taken together, our results suggest that resveratrol induces cellular senescence by targeting mono-ubiquitination of histone H2B, and RNF20 may plays an important role in cell proliferation by suppressing cellular senescence pathways in glioma cells.
2.4. Senescence associated-b-galactosidase (SA-b-gal) staining Senescence-b-galactosidase staining was done with a kit (Cat. # 9860, Cell signaling, Danvers, MA) according to manufacturer’s protocol. Briefly, cells in 35 mm well were washed with 2 mL PBS, fixed for 10–15 min in 2 mL fixative solution, washed twice with 2 mL PBS, and then incubated in SA-b-gal staining solution overnight at 37 °C.
2. Materials and methods 2.1. Cells and cell culture
2.5. Clonogenic assays
U87 MG and U118 MG (hereafter abbreviated as U87, U118, respectively) human glioblastoma (glioma) cell lines were kindly provided by Dr. Hui Zhang and Dr. Hong Sun. MCF7 (human breast cancer), H1299 (non-small cell lung carcinoma), and PC3 (human prostate cancer) cell lines were obtained from Dr. Giuseppe Pizzorno. Normal human dermal fibroblasts, HDF and normal human mammary epithelial cells, HMEpC, were purchased from Cell Applications (San Diego, CA). U118, U87, and MCF7 cultures were grown in high-glucose DMEM supplemented with 10% fetal bovine serum (FBS, HyClone, Waltham, MA) and 1% penicillin/streptomycin. H1299 was grown in RPMI with 10% FBS and 1% penicillin/ streptomycin. PC3 cells were in F12K Medium with 10% FBS and 1% antibiotics. HDF was grown in Human Fibroblast Growth Medium (Cell Applications, San Diego, CA) and HMEpC in Mammary Epithelial Growth Medium (Cell Applications). All cells were grown at 37 °C in a humidified atmosphere of 95% air and 5% CO2. Cell number was measured by a hemacytometer (Hausser Scientific, Horsham, PA). Cell population doubling time (Td) was calculated from the exponential portion of the growth curve using the equation Td = 0.693t/ln(Nt/N0), where t is time, and Nt and N0 represent cell numbers at time t and initial time, respectively [12].
Monolayer cultures of glioma cells were trypsinized and appropriate cell dilutions were re-plated. After 14 days, cultured cells were stained with 0.4% Coomassie blue G250 in 50% methanol and 5% acetic acid. Colonies of 50 cells or more were counted. Plating efficiencies (PE) were calculated using the formula PE = (number of colonies counted/number of cells seeded) 100%.
2.6. Growth of human tumor xenograft U87 or U118 Cells (2,000,000 in 0.1 ml PBS) were injected subcutaneously in the right flank of the female nude (nu/nu) mice (4– 6 weeks old; Charles River Laboratories, Wilmington, MA). Five nude mice were used for each group. When tumors were palpable, they were measured by a Vernier caliper in indicated days after injection. Tumor volume (TV) was calculated as follows: TV (mm3) = L W2/2, where L is the longest dimension of the tumor (in mm), and W is the shortest dimension of the tumor.
2.7. Lentiviral shRNA knockdown experiments 2.2. Resveratrol treatment For short-term resveratrol treatment, cells in 60 mm dishes were allowed to grow to 40–50% confluency. Their original media were changed to 5 ml DMEM media containing glucose (4.5 g/L) as the only carbon source (without both L-glutamine and sodium pyruvate). After 1 h of media change, cells were collected as nontreated or treated with 5 lL acetone or resveratrol in 5 lL acetone vehicle for 4 h and then collected for Western analysis. For longterm resveratrol treatment, 3 105 of U87 or U118 glioma cells were plated in 100 mm dishes. One day later, the culture media were replaced with fresh medium (10 mL/dish). After one more day, the cells were mock-treated, or 10 lL vehicle acetone, or treated with 6, 10, and 20 lM resveratrol (final concentration in medium) for U118 cells, or 6, 20, and 60 lM resveratrol (final concentration in medium) for U87 cells. The media (with or without resveratrol) were changed every 3 days until collection. 2.3. Western analysis Cultured cells were harvested on ice (3–5 min) by scraping in the medium, centrifuged with about 300 relative centrifugal forces (RCF) for 2 min. The cell pellets were then re-suspended in 2 SDS–PAGE sample buffer containing 2-mercaptoethanol and heated at 100 °C for 5 min, and stored at 80 °C until use. Total protein was separated by 12% SDS polyacrylamide gel electrophoresis and analyzed with antibodies to H2B (Millipore, Billerica, MA), mono-ubiquitinated H2B (uH2B, MediMabs, Montreal, Quebec, Canada), RNF20 (Abcam, Cambridge, MA), p53 (Santa Cruz Biotechnology, Santa Cruz, CA). Immunoreactive bands were detected using SuperSignal kit (Thermo Scientific, Waltham, MA).
Lentiviral particles for knockdown of RNF20 (TRCN 0000033877, hereafter is abbreviated as shRNF20) were purchased from Sigma–Aldrich (Mission collection, St. Louis, MO) and contain the inserted sequence CCGGGCCAATGAAATCAAGTCTAAACTCGAGTTTAGACTTGATTTCATTGGCTTTTTG. Control lentiviral particles are Luciferase (SHC007, hereafter abbreviated as shLuc) were from Sigma–Aldrich containing sequence: CCGGCGCTGAGTACTTCGAAATGTCTCGAGGACATTT CGAAGTACTCAGCGTTTTT. Lentiviral transduction was according to the manufacturer’s instruction. Briefly, U87 was seeded at 1.5 105 cells per 100 mm dishes while U118 was seeded 3.0 105 cells per 100 mm dishes. Two days later, hexadimethrine bromide (Sigma– Aldrich) was added to medium (final concentration 8 lg/mL) to enhance transduction efficiency. Then shRNF20 or shLuc {multiplicity of infection (MOI) = 10} was added and the plates were swirled gently for mixing. The viral particle-containing medium was replaced with fresh medium next day. After one more day, the media was replaced with fresh medium containing 2 lg/mL puromycin (Sigma–Aldrich) for selecting transfected cells. The puromycin containing media were then changed every 3 days.
2.8. Statistical analysis Tumor xenograft was performed with five mice in each group; Western experiments were repeated at least twice or more; all other experiments were performed at least three times. Data were analyzed by R Integrated Statistical Software (version 2.6.0, Vienna, Austria). Student t test was used to evaluate the significance of difference. p < 0.05 was accepted as significant.
Z. Gao et al. / Biochemical and Biophysical Research Communications 407 (2011) 271–276
3. Results 3.1. Resveratrol inhibited cell proliferation and induced cellular senescence in cultured glioma cells To examine the effects of resveratrol on glioma cells, we plated 5 104 U87 or U118 glioblastoma (Grade IV glioma cell) lines in 60 mm dishes. We treated these cells with mock, acetone, or different doses of resveratrol that were dissolved in acetone. We then trypsinized and counted cell numbers at 3, 6, and 9 days post treatment. As shown in Fig. 1A, resveratrol treatment significantly inhibited proliferation of these two human glioma cell lines on a dose- and time-dependent manner. Closer examination of glioma cells indicated that resveratrol treatment inhibited cell proliferation by inducing cellular hypertrophy in glioma cells. As shown in Fig. S1A and B, resveratrol induced dramatic changes in cell volume and cell morphology in both U87 and U118 cells. Specifically, resveratrol transformed spindle-shaped glioma cells to enlarged and irregular flattenshaped ones. The mean diameter of the resveratrol-treated U87 is 5-fold longer than the diameter parallel to the spindle axis of the mock-or vehicle-treated glioma cells (Fig. S1A and B). In contrast, mock and vehicle (acetone) treatment had no effect on inducing cellular hypertrophy (Fig. S1A and B).
273
To determine whether cellular hypertrophy induced by resveratrol was associated with cellular senescence of glioma cells, we tested expression of senescence-associated-b-galactosidase (SA-bgal) marker [13,14]. As shown Fig. 1B and C, SA-b-gal was detected in these hypertrophic cells. In contrast, mock- or vehicle-treated cells showed minimal expression of SA-b-gal even at a higher density. These results demonstrated that resveratrol inhibited proliferation of the glioma cells by inducing cellular senescence. 3.2. Resveratrol treatment reduced tumorigenicity of glioma cells To determine whether resveratrol-induced senescent cells had an impact on clonogenicity, we treated U87 and U118 cells with indicated concentrations of resveratrol for up to 9 days. We trypsinized monolayer cultures of glioma cells and seeded 1000, 2000, and 4000 cells separately in 60 mm culture dishes for incubation without further resveratrol treatment. After 2 weeks, we stained these cell cultures with 0.4% Coomassie blue G250 in 50% methanol and 5% acetic acid. We counted colonies that contained more than 50 cells. We calculated cell clonogenicity with plating efficiencies (PE) using the formula PE = (number of colonies counted/number of cells seeded) 100%. As shown in Fig. 2A, resveratrol-treated cells showed marked decrease in their propensity for forming colonies in a dose-dependent manner.
Fig. 1. Resveratrol inhibited cell proliferation and induced cellular senescence of glioma cells. (A) U87 or U118 cells (5 104) were seeded in 60 mm dishes and were treated with resveratrol (RV), acetone, or mock. Viable (Trypan blue excluded) cells were counted on D3, D6, and D9. The means and standard deviations were derived from cell count on three dishes. (B) Expression of senescence-associated-b-galactosidase (SA-b-gal) in resveratrol-treated U87 glioma cells. A resveratrol-induced senescent cell is highlighted. (C) Expression of senescence-associated-b-galactosidase (SA-b-gal) in U118 glioma cells. A resveratrol-induced senescent cell is highlighted. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
274
Z. Gao et al. / Biochemical and Biophysical Research Communications 407 (2011) 271–276
Fig. 2. Resveratrol inhibited clonogenic efficiency of human glioma cells in vitro and in nude mice. (A) Resveratrol-treated U87 and U118 (9-day treatment, 1000–4000 cells) were re-plated. After 14 days, the cultured cells were stained with 0.4% Coomassie blue G250 in 50% methanol and 5% acetic acid. The upper panel shows the stained dishes plated with 1000 cells. The lower panel shows the cell plating efficiencies (PE), which were calculated using the formula: PE = (number of colonies counted/number of cells seeded) 100%. (B) Tumor growth volume from subcutaneously-injected U87 and U118. U87 tumor volume was measured at 27 days post injection while U118 tumor volume at 125 days post injection. Five mice were used for each treatment.
To further determine if resveratrol treatment reduced in vivo tumorigenicity of U87 and U118 glioma cells, we treated them with indicated concentrations of resveratrol for 9 days. We then trypsinized glioma cells, collected 2 106 viable cells (by Trypan blue exclusion assay) in 0.1 ml PBS. We subsequently injected them subcutaneously in the right flank of nude mice [female (nu/ nu), 4–6 weeks old, five mice per group]. When the tumor was palpable, we measured the size with a Vernier caliper. We calculated tumor volume (TV) with TV (mm3) = L W2/2, Where L was the longest dimension of the tumor (in mm), and W was the shortest dimension of the tumor. As shown in Fig. 2B, resveratrol inhibited the tumorigenicity of both U87 and U118 in the xenograft model.
3.3. Resveratrol inhibited mono-ubiquitination of histone H2B (uH2B) at K120 in glioma cells We have previously shown that glucose through glycolysis induces mono-ubiquitination of histone H2B in yeast [6] and mammalian cells [11]. Resveratrol has been shown to inhibit glycolysis in human ovarian cancer cells [15] and down-regulates transcripts of glycolytic genes in the livers of the mice on a high calorie diet [16]. We therefore sought to examine whether resveratrol treatment had any effect on uH2B at K120 in human cells, which include glioma cells (U87, U118), breast cancer (MCF7), non-small lung carcinoma cells (H1299), human prostate cells
Z. Gao et al. / Biochemical and Biophysical Research Communications 407 (2011) 271–276
275
(PC3), non-malignant normal human dermal fibroblast and human mammary epithelial cells (HMEpC). We grew these cells to around 50% confluency in regular DMEM media (2–3 days) and replaced media with DMEM media (with glucose 4.5 g/L, but without both glutamine and sodium pyruvate) for 1 h. We then added resveratrol (10, 100, and 500 mM) for 4 h and subsequently collected cells for Western analysis of uH2B. As shown in Fig. S1C, resveratrol inhibited uH2B within 4 h in a dose-dependent manner. These results were consistent with earlier findings that resveratrol inhibits glycolysis [15,16] and glycolysis is required for mono-ubiquitination of histone H2B both in yeast [6] and glioma cells in vitro [11]. Since acute incubation with the high dose of resveratrol (500 mM) was effective in inhibiting uH2B, we then sought to examine whether chronic treatment with lower doses of resveratrol had any effect on uH2B. To test this possibility, we focused only on glioma cells (U87 and U118). As shown in Fig. 3A and B, resveratrol inhibited uH2B of U87 and U118 in a dose-dependent manner. Resveratrol also inhibited uH2B of U87 and U118 in a dose-dependent manner (data not shown). These data suggest that resveratrol could inhibit mono-ubiquitination of histone H2B at K120 in glioma cells. 3.4. Depletion of RNF20, an ubiquitin ligase of histone H2B, phenocopied the effects of resveratrol by inhibiting uH2B and inducing cellular senescence in human glioma cells Since resveratrol inhibited uH2B and induced cellular senescence, we reasoned whether depletion of RNF20, a ubiquitin ligase of histone H2B, could mimick the effects of resveratrol. To test this possibility, we used lentiviral particles (Sigma, St. Louis, MO) expressing short hairpin RNA (shRNA) of RNF20 (referred as shRNF20 thereafter) to knockdown RNF20 in U87 and U118 glioma cells. We used lentiviral particles encoding shRNA for luciferase (referred as shLuc thereafter) as a control. As shown in Fig. 4A, treatment with shRNF20 significantly reduced expression of RNF20 and almost completely abolished uH2B in a time-dependent manner. In contrast, shLuc treatment had no effect on either RNF20 or uH2B levels. These data demonstrate that RNF20 was the major ubiquitin ligase for histone H2B. We then sought to examine whether depletion of RNF20 could induce cellular senescence in glioma cells. As shown in Fig. S1D and E, depletion of RNF20 induced similar cellular hypertrophy as observed with resveratrol treatment. Specifically, depletion of RNF20 transformed spindle-shaped glioma cells into extremely enlarged and irregular flatten-shaped ones (Fig. S1D and E). As shown in Fig. 4B, the hypertrophic cells expressed a high level of SA-b-gal.
M oc k Ac et on e RV 6 µM RV 20 µM RV 60 µM
A uH2B
U87 (D9) H2B
M oc k Ac et on e RV 6 µM RV 10 µM RV 20 µM
B uH2B
U118 (D9) H2B Fig. 3. Resveratrol inhibited mono-ubiquitination of histone H2B at K120 (uH2B) in glioma cells. (A) Chronic treatment (up to 9-day) with lower doses of resveratrol inhibited uH2B of U87. (B) Chronic treatment (up to 9-day) with lower doses of resveratrol inhibited uH2B of U118.
Fig. 4. Depletion of RNF20 induced cellular senescence of human glioma cells. (A) Infection of lentivirus shRNF20 inhibited expression of RNF20 and uH2B. Total cellular proteins were prepared from U87 and U118 infected with either shLuc (MOI = 10) or shRNF20 (MOI = 10). Equal amounts of proteins were separated by SDS–PAGE and immunoblotted for RNF20, uH2B, and H2B. (B) Resveratrol-induced cellular senescence in U87 and U118. U87 and U118 cells were infected with either shLuc (MOI = 10) or shRNF20 (MOI = 10). On Day 6 post-infection, the cells were analyzed for expression of SA-b-gal. Two senescent cells are highlighted.
Taken together, these data suggest that depletion of RNF20 recapitulated the inhibitory effects of resveratrol treatment of glioma cells.
4. Discussion This is the first report demonstrating that resveratrol modulates histone modifications. Resveratrol has a variety of activities including anti-cancer, anti-inflammatory, anti-angiogenic, and chemosensitizing properties in vitro and in vivo models by targeting multiple cellular targets [2,3,17,18]. However, it is not known whether resveratrol-mediated signals ultimately converge on chromatin to confer the effects of resveratrol. Our previous studies show that glucose induces mono-ubiquitination of histone H2B (uH2B) both in yeast and mammalian cells through glycolysis [6,11]. Since resveratrol can inhibit glycolysis [15], it is not unanticipated that both acute and chronic treatment with resveratrol inhibited uH2B in glioma cells. uH2B occurs in a rapidly-dividing (exponential-phase) yeast but is not detectable in stationary phase (G0) yeast [6]. Consistent with this, global uH2B levels in proliferating tumor cells were much higher than those in resveratrol-induced senescent tumor cells (Fig. 3). Furthermore, it has been shown that deletion of Bre1 gene increases cell size in yeast [7], which resembles that of stationaryphase yeast under carbon starvation (unpublished). Remarkably, depletion of RNF20 induced flat and enlarged senescent glioma cells, suggesting an evolutionary conservation of the cell size con-
276
Z. Gao et al. / Biochemical and Biophysical Research Communications 407 (2011) 271–276
trol from yeast to tumor cells. Moreover, our results also implicate a proto-oncogenic activity of RNF20, which is consistent with the role of RNF20 in antagonizing Ebp1 function in breast tumor cells [19]. Therefore, the role of RNF20 in glioma and breast cancer cells are in contrast with the proposed role of RNF20 as a tumor suppressor in HeLa cells [20]. Compelling evidence suggests that therapy-induced cellular senescence correlates with better clinical outcome than therapyinduced apoptosis [21]. Cellular senescence was first noted in the 1960s as a permanent arrest of cell cycle of non-malignant cells, which undergo limited number of cell divisions before entering cellular senescence, a permanent and irreversible proliferation arrest [22]. Bypassing cellular senescence is thought to be an important step in tumorigenesis [23]. In this regard, resveratrol-induced cellular senescence in human glioma cells could represent a potential novel attractive therapy for glioma, which currently lacks an effective treatment. It has been reported that resveratrol induces cellular senescence in colon carcinoma cells, osteosarcoma cells and C6 cells [24–26]. However, molecular mechanisms underlying resveratrol-induced cellular senescence in those cells were not clearly defined. We show here that depletion of RNF20 inhibited uH2B and induced cellular senescence of malignant human glioma cells, and thereby recapitulated the effects of resveratrol treatment. Our study therefore suggests that uH2B is a novel direct or indirect chromatin target of resveratrol. Moreover, our study demonstrates the critical role of RNF20 in inhibiting cellular senescence programs that are intact in glioma cells. Therefore, RNF20 may serve as a potential therapeutic target. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc.2011.02.008. References [1] E.C. Holland, Gliomagenesis: genetic alterations and mouse models, Nat. Rev. Genet. 2 (2001) 120–129. [2] M. Jang, L. Cai, G.O. Udeani, K.V. Slowing, C.F. Thomas, C.W. Beecher, H.H. Fong, N.R. Farnsworth, A.D. Kinghorn, R.G. Mehta, R.C. Moon, J.M. Pezzuto, Cancer chemopreventive activity of resveratrol, a natural product derived from grapes, Science 275 (1997) 218–220. [3] S. Pervaiz, A.L. Holme, Resveratrol: its biological targets and functional activity, Antioxid. Redox Signal. 11 (2009) 2851–2894. [4] A. Bishayee, Cancer prevention, treatment with resveratrol: from rodent studies to clinical trials, Cancer Prev. Res. (Phila Pa) 2 (2009) 409–418. [5] M.A. Osley, Regulation of histone H2A and H2B ubiquitylation, Brief. Funct. Genomic Proteomic 5 (2006) 179–189. [6] L. Dong, C.W. Xu, Carbohydrates induce mono-ubiquitination of H2B in yeast, J. Biol. Chem. 279 (2004) 1577–1580. [7] W.W. Hwang, S. Venkatasubrahmanyam, A.G. Ianculescu, A. Tong, C. Boone, H.D. Madhani, A conserved RING finger protein required for histone H2B monoubiquitination and cell size control, Mol. Cell 11 (2003) 261–266.
[8] H. Huang, A. Kahana, D.E. Gottschling, L. Prakash, S.W. Liebman, The ubiquitinconjugating enzyme Rad6 (Ubc2) is required for silencing in Saccharomyces cerevisiae, Mol. Cell. Biol. 17 (1997) 6693–6699. [9] A. Wood, N.J. Krogan, J. Dover, J. Schneider, J. Heidt, M.A. Boateng, K. Dean, A. Golshani, Y. Zhang, J.F. Greenblatt, M. Johnston, A. Shilatifard, Bre1, an E3 ubiquitin ligase required for recruitment and substrate selection of Rad6 at a promoter, Mol. Cell 11 (2003) 267–274. [10] J. Kim, M. Guermah, R.K. McGinty, J.S. Lee, Z. Tang, T.A. Milne, A. Shilatifard, T.W. Muir, R.G. Roeder, RAD6-mediated transcription-coupled H2B ubiquitylation directly stimulates H3K4 methylation in human cells, Cell 137 (2009) 459–471. [11] Z. Gao, C.W. Xu, Glucose metabolism induces mono-ubiquitination of histone H2B in mammalian cells, Biochem. Biophys. Res. Commun. 404 (2011) 428– 433. [12] Y. Zhang, W. Zhao, H.J. Zhang, F.E. Domann, L.W. Oberley, Overexpression of copper zinc superoxide dismutase suppresses human glioma cell growth, Cancer Res. 62 (2002) 1205–1212. [13] G.P. Dimri, X. Lee, G. Basile, M. Acosta, G. Scott, C. Roskelley, E.E. Medrano, M. Linskens, I. Rubelj, O. Pereira-Smith, et al., A biomarker that identifies senescent human cells in culture and in aging skin in vivo, Proc. Natl. Acad. Sci. USA 92 (1995) 9363–9367. [14] G.P. Dimri, A. Testori, M. Acosta, J. Campisi, Replicative senescence, aging and growth-regulatory transcription factors, Biol. Signals 5 (1996) 154–162. [15] A. Kueck, A.W. Opipari Jr., K.A. Griffith, L. Tan, M. Choi, J. Huang, H. Wahl, J.R. Liu, Resveratrol inhibits glucose metabolism in human ovarian cancer cells, Gynecol. Oncol. 107 (2007) 450–457. [16] J.A. Baur, K.J. Pearson, N.L. Price, H.A. Jamieson, C. Lerin, A. Kalra, V.V. Prabhu, J.S. Allard, G. Lopez-Lluch, K. Lewis, P.J. Pistell, S. Poosala, K.G. Becker, O. Boss, D. Gwinn, M. Wang, S. Ramaswamy, K.W. Fishbein, R.G. Spencer, E.G. Lakatta, D. Le Couteur, R.J. Shaw, P. Navas, P. Puigserver, D.K. Ingram, R. De Cabo, D.A. Sinclair, Resveratrol improves health and survival of mice on a high-calorie diet, Nature 444 (2006) 337–342. [17] K.B. Harikumar, B.B. Aggarwal, Resveratrol: a multitargeted agent for ageassociated chronic diseases, Cell Cycle 7 (2008) 1020–1035. [18] S.K. Goswami, D.K. Das, Resveratrol and chemoprevention, Cancer Lett. 284 (2009) 1–6. [19] Z. Liu, S.M. Oh, M. Okada, X. Liu, D. Cheng, J. Peng, D.J. Brat, S.Y. Sun, W. Zhou, W. Gu, K. Ye, Human BRE1 is an E3 ubiquitin ligase for Ebp1 tumor suppressor, Mol. Biol. Cell 20 (2009) 757–768. [20] E. Shema, I. Tirosh, Y. Aylon, J. Huang, C. Ye, N. Moskovits, N. Raver-Shapira, N. Minsky, J. Pirngruber, G. Tarcic, P. Hublarova, L. Moyal, M. Gana-Weisz, Y. Shiloh, Y. Yarden, S.A. Johnsen, B. Vojtesek, S.L. Berger, M. Oren, The histone H2B-specific ubiquitin ligase RNF20/hBRE1 acts as a putative tumor suppressor through selective regulation of gene expression, Genes Dev. 22 (2008) 2664–2676. [21] I.B. Roninson, Tumor cell senescence in cancer treatment, Cancer Res. 63 (2003) 2705–2715. [22] L. Hayflick, The limited in vitro lifetime of human diploid cell strains, Exp. Cell Res. 37 (1965) 614–636. [23] G.P. Dimri, What has senescence got to do with cancer?, Cancer Cell 7 (2005) 505–512 [24] E.H. Heiss, Y.D. Schilder, V.M. Dirsch, Chronic treatment with resveratrol induces redox stress- and ataxia telangiectasia-mutated (ATM)-dependent senescence in p53-positive cancer cells, J. Biol. Chem. 282 (2007) 26759– 26766. [25] M. Rusin, A. Zajkowicz, D. Butkiewicz, Resveratrol induces senescence-like growth inhibition of U-2 OS cells associated with the instability of telomeric DNA and upregulation of BRCA1, Mech. Ageing Dev. 130 (2009) 528–537. [26] L.L. Zamin, E.C. Filippi-Chiela, P. Dillenburg-Pilla, F. Horn, C. Salbego, G. Lenz, Resveratrol and quercetin cooperate to induce senescence-like growth arrest in C6 rat glioma cells, Cancer Sci. 100 (2009) 1655–1662.
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Resveratrol induces cellular senescence with attenuated mono-ubiquitination of histone H2B in glioma cells q Zhen Gao, Michael S. Xu, Tamara L. Barnett, C. Wilson Xu ⇑ Nevada Cancer Institute, Las Vegas, NV 89135, USA
a r t i c l e
i n f o
Article history: Received 1 February 2011
Keywords: Resveratrol Glioma Cellular senescence Mono-ubiquitinated histone H2B RNF20 Histone modifications
a b s t r a c t Resveratrol (3,40 ,5-trihydroxy-trans-stilbene), a polyphenol naturally occurring in grapes and other plants, has cancer chemo-preventive effects and therapeutic potential. Although resveratrol modulates multiple pathways in tumor cells, how resveratrol or its affected pathways converge on chromatin to mediate its effects is not known. Using glioma cells as a model, we showed here that resveratrol inhibited cell proliferation and induced cellular hypertrophy by transforming spindle-shaped cells to enlarged, irregular and flatten-shaped ones. We further showed that resveratrol-induced hypertrophic cells expressed senescence-associated-b-galactosidase, suggesting that resveratrol-induced cellular senescence in glioma cells. Consistent with these observations, we demonstrated that resveratrol inhibited clonogenic efficiencies in vitro and tumor growth in a xenograft model. Furthermore, we found that acute treatment of resveratrol inhibited mono-ubiquitination of histone H2B at K120 (uH2B) in breast, prostate, pancreatic, lung, brain tumor cells as well as primary human cells. Chronic treatment with low doses of resveratrol also inhibited uH2B in the resveratrol-induced senescent glioma cells. Moreover, we showed that depletion of RNF20, a ubiquitin ligase of histone H2B, inhibited uH2B and induced cellular senescence in glioma cells in vitro, thereby recapitulated the effects of resveratrol. Taken together, our results suggest that uH2B is a novel direct or indirect chromatin target of resveratrol and RNF20 plays an important role in inhibiting cellular senescence programs that are intact in glioma cells. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction Glioblastoma multiforme is the most malignant of the primary brain tumors with no effective treatment currently available [1] (http://www.cbtrus.org). Resveratrol, which is synthesized de novo in food and medicinal plants presumably in response to stress or infection, has been shown to inhibit proliferation in a variety of tumor cells such as skin, breast, prostate, and lung through a multiple pathways [2–4]. However, the effects of resveratrol have not been adequately evaluated in human glioma cells. It is also not clear whether resveratrol-modulated pathways converge on chromatin to mediate its inhibitory effects. Histone modifications are intimately involved in transcriptional regulation. Among histone acetylation, methylation, phosphoryla-
Abbreviations: RV, resveratrol; uH2B, mono-ubiquitinated histone H2B; GBM, glioblastoma multiforme; SA-b-gal, senescence-associated-b-galactosidase. q We thank Giuseppe Pizzorno, Hui Zhang, Hong Sun for providing tumor cell lines, Fei Lu for graphic analysis and members of the Xu Lab for discussion. The work is supported in part by the Stephen & Mary Birch Foundation. ⇑ Corresponding author. Address: Nevada Cancer Institute, One Breakthrough Way, Las Vegas, NV 89135, USA. E-mail address: [email protected] (C.W. Xu).
0006-291X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2011.02.008
tion and mono-ubiquitination, the role of mono-ubiquitination of histones is least understood, despite the fact that histone H2A is the first protein shown to be covalently conjugated with ubiquitin [5]. In budding yeast, H2B is the only histone that is mono-ubiquitinated at K123 (uH2B) [5], which constitutes approximately 15% of the total H2B during exponential phase, but is not detectable in stationary phase [6]. Bre1, a Rad6-associated RING finger protein, is the E3 ubiquitin ligase of H2B in yeast [7–9]. RNF20, a human homolog of yeast Bre1, has been identified in humans [7,10], suggesting that mono-ubiquitination of histone H2B is conserved from yeast to humans. Using mono-ubiquitination of histone H2B at K123 (uH2B) in yeast as a model, we have previously discovered that glucose induces uH2B through glycolysis, revealing a novel paradigm of nutritional regulation of histone modifications [6]. We have further demonstrated that glycolysis is also required for mono-ubiquitination of histone H2B at K120, the orthologous site of K123 of yeast histone H2B, in both human primary and tumor cells [11]. In this report, we found that resveratrol inhibited proliferation of human glioma cells by inducing cellular senescence. We further found that mono-ubiquitination of histone H2B at K120 (uH2B) was inhibited in resveratrol-induced senescent glioma cells. Moreover, we showed that like resveratrol treatment, shRNA depletion
272
Z. Gao et al. / Biochemical and Biophysical Research Communications 407 (2011) 271–276
of RNF20 inhibited uH2B and induced cellular senescence in glioma cells, demonstrating that loss of H2B ubiquitin ligase recapitulated the inhibitory effects of resveratrol. Taken together, our results suggest that resveratrol induces cellular senescence by targeting mono-ubiquitination of histone H2B, and RNF20 may plays an important role in cell proliferation by suppressing cellular senescence pathways in glioma cells.
2.4. Senescence associated-b-galactosidase (SA-b-gal) staining Senescence-b-galactosidase staining was done with a kit (Cat. # 9860, Cell signaling, Danvers, MA) according to manufacturer’s protocol. Briefly, cells in 35 mm well were washed with 2 mL PBS, fixed for 10–15 min in 2 mL fixative solution, washed twice with 2 mL PBS, and then incubated in SA-b-gal staining solution overnight at 37 °C.
2. Materials and methods 2.1. Cells and cell culture
2.5. Clonogenic assays
U87 MG and U118 MG (hereafter abbreviated as U87, U118, respectively) human glioblastoma (glioma) cell lines were kindly provided by Dr. Hui Zhang and Dr. Hong Sun. MCF7 (human breast cancer), H1299 (non-small cell lung carcinoma), and PC3 (human prostate cancer) cell lines were obtained from Dr. Giuseppe Pizzorno. Normal human dermal fibroblasts, HDF and normal human mammary epithelial cells, HMEpC, were purchased from Cell Applications (San Diego, CA). U118, U87, and MCF7 cultures were grown in high-glucose DMEM supplemented with 10% fetal bovine serum (FBS, HyClone, Waltham, MA) and 1% penicillin/streptomycin. H1299 was grown in RPMI with 10% FBS and 1% penicillin/ streptomycin. PC3 cells were in F12K Medium with 10% FBS and 1% antibiotics. HDF was grown in Human Fibroblast Growth Medium (Cell Applications, San Diego, CA) and HMEpC in Mammary Epithelial Growth Medium (Cell Applications). All cells were grown at 37 °C in a humidified atmosphere of 95% air and 5% CO2. Cell number was measured by a hemacytometer (Hausser Scientific, Horsham, PA). Cell population doubling time (Td) was calculated from the exponential portion of the growth curve using the equation Td = 0.693t/ln(Nt/N0), where t is time, and Nt and N0 represent cell numbers at time t and initial time, respectively [12].
Monolayer cultures of glioma cells were trypsinized and appropriate cell dilutions were re-plated. After 14 days, cultured cells were stained with 0.4% Coomassie blue G250 in 50% methanol and 5% acetic acid. Colonies of 50 cells or more were counted. Plating efficiencies (PE) were calculated using the formula PE = (number of colonies counted/number of cells seeded) 100%.
2.6. Growth of human tumor xenograft U87 or U118 Cells (2,000,000 in 0.1 ml PBS) were injected subcutaneously in the right flank of the female nude (nu/nu) mice (4– 6 weeks old; Charles River Laboratories, Wilmington, MA). Five nude mice were used for each group. When tumors were palpable, they were measured by a Vernier caliper in indicated days after injection. Tumor volume (TV) was calculated as follows: TV (mm3) = L W2/2, where L is the longest dimension of the tumor (in mm), and W is the shortest dimension of the tumor.
2.7. Lentiviral shRNA knockdown experiments 2.2. Resveratrol treatment For short-term resveratrol treatment, cells in 60 mm dishes were allowed to grow to 40–50% confluency. Their original media were changed to 5 ml DMEM media containing glucose (4.5 g/L) as the only carbon source (without both L-glutamine and sodium pyruvate). After 1 h of media change, cells were collected as nontreated or treated with 5 lL acetone or resveratrol in 5 lL acetone vehicle for 4 h and then collected for Western analysis. For longterm resveratrol treatment, 3 105 of U87 or U118 glioma cells were plated in 100 mm dishes. One day later, the culture media were replaced with fresh medium (10 mL/dish). After one more day, the cells were mock-treated, or 10 lL vehicle acetone, or treated with 6, 10, and 20 lM resveratrol (final concentration in medium) for U118 cells, or 6, 20, and 60 lM resveratrol (final concentration in medium) for U87 cells. The media (with or without resveratrol) were changed every 3 days until collection. 2.3. Western analysis Cultured cells were harvested on ice (3–5 min) by scraping in the medium, centrifuged with about 300 relative centrifugal forces (RCF) for 2 min. The cell pellets were then re-suspended in 2 SDS–PAGE sample buffer containing 2-mercaptoethanol and heated at 100 °C for 5 min, and stored at 80 °C until use. Total protein was separated by 12% SDS polyacrylamide gel electrophoresis and analyzed with antibodies to H2B (Millipore, Billerica, MA), mono-ubiquitinated H2B (uH2B, MediMabs, Montreal, Quebec, Canada), RNF20 (Abcam, Cambridge, MA), p53 (Santa Cruz Biotechnology, Santa Cruz, CA). Immunoreactive bands were detected using SuperSignal kit (Thermo Scientific, Waltham, MA).
Lentiviral particles for knockdown of RNF20 (TRCN 0000033877, hereafter is abbreviated as shRNF20) were purchased from Sigma–Aldrich (Mission collection, St. Louis, MO) and contain the inserted sequence CCGGGCCAATGAAATCAAGTCTAAACTCGAGTTTAGACTTGATTTCATTGGCTTTTTG. Control lentiviral particles are Luciferase (SHC007, hereafter abbreviated as shLuc) were from Sigma–Aldrich containing sequence: CCGGCGCTGAGTACTTCGAAATGTCTCGAGGACATTT CGAAGTACTCAGCGTTTTT. Lentiviral transduction was according to the manufacturer’s instruction. Briefly, U87 was seeded at 1.5 105 cells per 100 mm dishes while U118 was seeded 3.0 105 cells per 100 mm dishes. Two days later, hexadimethrine bromide (Sigma– Aldrich) was added to medium (final concentration 8 lg/mL) to enhance transduction efficiency. Then shRNF20 or shLuc {multiplicity of infection (MOI) = 10} was added and the plates were swirled gently for mixing. The viral particle-containing medium was replaced with fresh medium next day. After one more day, the media was replaced with fresh medium containing 2 lg/mL puromycin (Sigma–Aldrich) for selecting transfected cells. The puromycin containing media were then changed every 3 days.
2.8. Statistical analysis Tumor xenograft was performed with five mice in each group; Western experiments were repeated at least twice or more; all other experiments were performed at least three times. Data were analyzed by R Integrated Statistical Software (version 2.6.0, Vienna, Austria). Student t test was used to evaluate the significance of difference. p < 0.05 was accepted as significant.
Z. Gao et al. / Biochemical and Biophysical Research Communications 407 (2011) 271–276
3. Results 3.1. Resveratrol inhibited cell proliferation and induced cellular senescence in cultured glioma cells To examine the effects of resveratrol on glioma cells, we plated 5 104 U87 or U118 glioblastoma (Grade IV glioma cell) lines in 60 mm dishes. We treated these cells with mock, acetone, or different doses of resveratrol that were dissolved in acetone. We then trypsinized and counted cell numbers at 3, 6, and 9 days post treatment. As shown in Fig. 1A, resveratrol treatment significantly inhibited proliferation of these two human glioma cell lines on a dose- and time-dependent manner. Closer examination of glioma cells indicated that resveratrol treatment inhibited cell proliferation by inducing cellular hypertrophy in glioma cells. As shown in Fig. S1A and B, resveratrol induced dramatic changes in cell volume and cell morphology in both U87 and U118 cells. Specifically, resveratrol transformed spindle-shaped glioma cells to enlarged and irregular flattenshaped ones. The mean diameter of the resveratrol-treated U87 is 5-fold longer than the diameter parallel to the spindle axis of the mock-or vehicle-treated glioma cells (Fig. S1A and B). In contrast, mock and vehicle (acetone) treatment had no effect on inducing cellular hypertrophy (Fig. S1A and B).
273
To determine whether cellular hypertrophy induced by resveratrol was associated with cellular senescence of glioma cells, we tested expression of senescence-associated-b-galactosidase (SA-bgal) marker [13,14]. As shown Fig. 1B and C, SA-b-gal was detected in these hypertrophic cells. In contrast, mock- or vehicle-treated cells showed minimal expression of SA-b-gal even at a higher density. These results demonstrated that resveratrol inhibited proliferation of the glioma cells by inducing cellular senescence. 3.2. Resveratrol treatment reduced tumorigenicity of glioma cells To determine whether resveratrol-induced senescent cells had an impact on clonogenicity, we treated U87 and U118 cells with indicated concentrations of resveratrol for up to 9 days. We trypsinized monolayer cultures of glioma cells and seeded 1000, 2000, and 4000 cells separately in 60 mm culture dishes for incubation without further resveratrol treatment. After 2 weeks, we stained these cell cultures with 0.4% Coomassie blue G250 in 50% methanol and 5% acetic acid. We counted colonies that contained more than 50 cells. We calculated cell clonogenicity with plating efficiencies (PE) using the formula PE = (number of colonies counted/number of cells seeded) 100%. As shown in Fig. 2A, resveratrol-treated cells showed marked decrease in their propensity for forming colonies in a dose-dependent manner.
Fig. 1. Resveratrol inhibited cell proliferation and induced cellular senescence of glioma cells. (A) U87 or U118 cells (5 104) were seeded in 60 mm dishes and were treated with resveratrol (RV), acetone, or mock. Viable (Trypan blue excluded) cells were counted on D3, D6, and D9. The means and standard deviations were derived from cell count on three dishes. (B) Expression of senescence-associated-b-galactosidase (SA-b-gal) in resveratrol-treated U87 glioma cells. A resveratrol-induced senescent cell is highlighted. (C) Expression of senescence-associated-b-galactosidase (SA-b-gal) in U118 glioma cells. A resveratrol-induced senescent cell is highlighted. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
274
Z. Gao et al. / Biochemical and Biophysical Research Communications 407 (2011) 271–276
Fig. 2. Resveratrol inhibited clonogenic efficiency of human glioma cells in vitro and in nude mice. (A) Resveratrol-treated U87 and U118 (9-day treatment, 1000–4000 cells) were re-plated. After 14 days, the cultured cells were stained with 0.4% Coomassie blue G250 in 50% methanol and 5% acetic acid. The upper panel shows the stained dishes plated with 1000 cells. The lower panel shows the cell plating efficiencies (PE), which were calculated using the formula: PE = (number of colonies counted/number of cells seeded) 100%. (B) Tumor growth volume from subcutaneously-injected U87 and U118. U87 tumor volume was measured at 27 days post injection while U118 tumor volume at 125 days post injection. Five mice were used for each treatment.
To further determine if resveratrol treatment reduced in vivo tumorigenicity of U87 and U118 glioma cells, we treated them with indicated concentrations of resveratrol for 9 days. We then trypsinized glioma cells, collected 2 106 viable cells (by Trypan blue exclusion assay) in 0.1 ml PBS. We subsequently injected them subcutaneously in the right flank of nude mice [female (nu/ nu), 4–6 weeks old, five mice per group]. When the tumor was palpable, we measured the size with a Vernier caliper. We calculated tumor volume (TV) with TV (mm3) = L W2/2, Where L was the longest dimension of the tumor (in mm), and W was the shortest dimension of the tumor. As shown in Fig. 2B, resveratrol inhibited the tumorigenicity of both U87 and U118 in the xenograft model.
3.3. Resveratrol inhibited mono-ubiquitination of histone H2B (uH2B) at K120 in glioma cells We have previously shown that glucose through glycolysis induces mono-ubiquitination of histone H2B in yeast [6] and mammalian cells [11]. Resveratrol has been shown to inhibit glycolysis in human ovarian cancer cells [15] and down-regulates transcripts of glycolytic genes in the livers of the mice on a high calorie diet [16]. We therefore sought to examine whether resveratrol treatment had any effect on uH2B at K120 in human cells, which include glioma cells (U87, U118), breast cancer (MCF7), non-small lung carcinoma cells (H1299), human prostate cells
Z. Gao et al. / Biochemical and Biophysical Research Communications 407 (2011) 271–276
275
(PC3), non-malignant normal human dermal fibroblast and human mammary epithelial cells (HMEpC). We grew these cells to around 50% confluency in regular DMEM media (2–3 days) and replaced media with DMEM media (with glucose 4.5 g/L, but without both glutamine and sodium pyruvate) for 1 h. We then added resveratrol (10, 100, and 500 mM) for 4 h and subsequently collected cells for Western analysis of uH2B. As shown in Fig. S1C, resveratrol inhibited uH2B within 4 h in a dose-dependent manner. These results were consistent with earlier findings that resveratrol inhibits glycolysis [15,16] and glycolysis is required for mono-ubiquitination of histone H2B both in yeast [6] and glioma cells in vitro [11]. Since acute incubation with the high dose of resveratrol (500 mM) was effective in inhibiting uH2B, we then sought to examine whether chronic treatment with lower doses of resveratrol had any effect on uH2B. To test this possibility, we focused only on glioma cells (U87 and U118). As shown in Fig. 3A and B, resveratrol inhibited uH2B of U87 and U118 in a dose-dependent manner. Resveratrol also inhibited uH2B of U87 and U118 in a dose-dependent manner (data not shown). These data suggest that resveratrol could inhibit mono-ubiquitination of histone H2B at K120 in glioma cells. 3.4. Depletion of RNF20, an ubiquitin ligase of histone H2B, phenocopied the effects of resveratrol by inhibiting uH2B and inducing cellular senescence in human glioma cells Since resveratrol inhibited uH2B and induced cellular senescence, we reasoned whether depletion of RNF20, a ubiquitin ligase of histone H2B, could mimick the effects of resveratrol. To test this possibility, we used lentiviral particles (Sigma, St. Louis, MO) expressing short hairpin RNA (shRNA) of RNF20 (referred as shRNF20 thereafter) to knockdown RNF20 in U87 and U118 glioma cells. We used lentiviral particles encoding shRNA for luciferase (referred as shLuc thereafter) as a control. As shown in Fig. 4A, treatment with shRNF20 significantly reduced expression of RNF20 and almost completely abolished uH2B in a time-dependent manner. In contrast, shLuc treatment had no effect on either RNF20 or uH2B levels. These data demonstrate that RNF20 was the major ubiquitin ligase for histone H2B. We then sought to examine whether depletion of RNF20 could induce cellular senescence in glioma cells. As shown in Fig. S1D and E, depletion of RNF20 induced similar cellular hypertrophy as observed with resveratrol treatment. Specifically, depletion of RNF20 transformed spindle-shaped glioma cells into extremely enlarged and irregular flatten-shaped ones (Fig. S1D and E). As shown in Fig. 4B, the hypertrophic cells expressed a high level of SA-b-gal.
M oc k Ac et on e RV 6 µM RV 20 µM RV 60 µM
A uH2B
U87 (D9) H2B
M oc k Ac et on e RV 6 µM RV 10 µM RV 20 µM
B uH2B
U118 (D9) H2B Fig. 3. Resveratrol inhibited mono-ubiquitination of histone H2B at K120 (uH2B) in glioma cells. (A) Chronic treatment (up to 9-day) with lower doses of resveratrol inhibited uH2B of U87. (B) Chronic treatment (up to 9-day) with lower doses of resveratrol inhibited uH2B of U118.
Fig. 4. Depletion of RNF20 induced cellular senescence of human glioma cells. (A) Infection of lentivirus shRNF20 inhibited expression of RNF20 and uH2B. Total cellular proteins were prepared from U87 and U118 infected with either shLuc (MOI = 10) or shRNF20 (MOI = 10). Equal amounts of proteins were separated by SDS–PAGE and immunoblotted for RNF20, uH2B, and H2B. (B) Resveratrol-induced cellular senescence in U87 and U118. U87 and U118 cells were infected with either shLuc (MOI = 10) or shRNF20 (MOI = 10). On Day 6 post-infection, the cells were analyzed for expression of SA-b-gal. Two senescent cells are highlighted.
Taken together, these data suggest that depletion of RNF20 recapitulated the inhibitory effects of resveratrol treatment of glioma cells.
4. Discussion This is the first report demonstrating that resveratrol modulates histone modifications. Resveratrol has a variety of activities including anti-cancer, anti-inflammatory, anti-angiogenic, and chemosensitizing properties in vitro and in vivo models by targeting multiple cellular targets [2,3,17,18]. However, it is not known whether resveratrol-mediated signals ultimately converge on chromatin to confer the effects of resveratrol. Our previous studies show that glucose induces mono-ubiquitination of histone H2B (uH2B) both in yeast and mammalian cells through glycolysis [6,11]. Since resveratrol can inhibit glycolysis [15], it is not unanticipated that both acute and chronic treatment with resveratrol inhibited uH2B in glioma cells. uH2B occurs in a rapidly-dividing (exponential-phase) yeast but is not detectable in stationary phase (G0) yeast [6]. Consistent with this, global uH2B levels in proliferating tumor cells were much higher than those in resveratrol-induced senescent tumor cells (Fig. 3). Furthermore, it has been shown that deletion of Bre1 gene increases cell size in yeast [7], which resembles that of stationaryphase yeast under carbon starvation (unpublished). Remarkably, depletion of RNF20 induced flat and enlarged senescent glioma cells, suggesting an evolutionary conservation of the cell size con-
276
Z. Gao et al. / Biochemical and Biophysical Research Communications 407 (2011) 271–276
trol from yeast to tumor cells. Moreover, our results also implicate a proto-oncogenic activity of RNF20, which is consistent with the role of RNF20 in antagonizing Ebp1 function in breast tumor cells [19]. Therefore, the role of RNF20 in glioma and breast cancer cells are in contrast with the proposed role of RNF20 as a tumor suppressor in HeLa cells [20]. Compelling evidence suggests that therapy-induced cellular senescence correlates with better clinical outcome than therapyinduced apoptosis [21]. Cellular senescence was first noted in the 1960s as a permanent arrest of cell cycle of non-malignant cells, which undergo limited number of cell divisions before entering cellular senescence, a permanent and irreversible proliferation arrest [22]. Bypassing cellular senescence is thought to be an important step in tumorigenesis [23]. In this regard, resveratrol-induced cellular senescence in human glioma cells could represent a potential novel attractive therapy for glioma, which currently lacks an effective treatment. It has been reported that resveratrol induces cellular senescence in colon carcinoma cells, osteosarcoma cells and C6 cells [24–26]. However, molecular mechanisms underlying resveratrol-induced cellular senescence in those cells were not clearly defined. We show here that depletion of RNF20 inhibited uH2B and induced cellular senescence of malignant human glioma cells, and thereby recapitulated the effects of resveratrol treatment. Our study therefore suggests that uH2B is a novel direct or indirect chromatin target of resveratrol. Moreover, our study demonstrates the critical role of RNF20 in inhibiting cellular senescence programs that are intact in glioma cells. Therefore, RNF20 may serve as a potential therapeutic target. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc.2011.02.008. References [1] E.C. Holland, Gliomagenesis: genetic alterations and mouse models, Nat. Rev. Genet. 2 (2001) 120–129. [2] M. Jang, L. Cai, G.O. Udeani, K.V. Slowing, C.F. Thomas, C.W. Beecher, H.H. Fong, N.R. Farnsworth, A.D. Kinghorn, R.G. Mehta, R.C. Moon, J.M. Pezzuto, Cancer chemopreventive activity of resveratrol, a natural product derived from grapes, Science 275 (1997) 218–220. [3] S. Pervaiz, A.L. Holme, Resveratrol: its biological targets and functional activity, Antioxid. Redox Signal. 11 (2009) 2851–2894. [4] A. Bishayee, Cancer prevention, treatment with resveratrol: from rodent studies to clinical trials, Cancer Prev. Res. (Phila Pa) 2 (2009) 409–418. [5] M.A. Osley, Regulation of histone H2A and H2B ubiquitylation, Brief. Funct. Genomic Proteomic 5 (2006) 179–189. [6] L. Dong, C.W. Xu, Carbohydrates induce mono-ubiquitination of H2B in yeast, J. Biol. Chem. 279 (2004) 1577–1580. [7] W.W. Hwang, S. Venkatasubrahmanyam, A.G. Ianculescu, A. Tong, C. Boone, H.D. Madhani, A conserved RING finger protein required for histone H2B monoubiquitination and cell size control, Mol. Cell 11 (2003) 261–266.
[8] H. Huang, A. Kahana, D.E. Gottschling, L. Prakash, S.W. Liebman, The ubiquitinconjugating enzyme Rad6 (Ubc2) is required for silencing in Saccharomyces cerevisiae, Mol. Cell. Biol. 17 (1997) 6693–6699. [9] A. Wood, N.J. Krogan, J. Dover, J. Schneider, J. Heidt, M.A. Boateng, K. Dean, A. Golshani, Y. Zhang, J.F. Greenblatt, M. Johnston, A. Shilatifard, Bre1, an E3 ubiquitin ligase required for recruitment and substrate selection of Rad6 at a promoter, Mol. Cell 11 (2003) 267–274. [10] J. Kim, M. Guermah, R.K. McGinty, J.S. Lee, Z. Tang, T.A. Milne, A. Shilatifard, T.W. Muir, R.G. Roeder, RAD6-mediated transcription-coupled H2B ubiquitylation directly stimulates H3K4 methylation in human cells, Cell 137 (2009) 459–471. [11] Z. Gao, C.W. Xu, Glucose metabolism induces mono-ubiquitination of histone H2B in mammalian cells, Biochem. Biophys. Res. Commun. 404 (2011) 428– 433. [12] Y. Zhang, W. Zhao, H.J. Zhang, F.E. Domann, L.W. Oberley, Overexpression of copper zinc superoxide dismutase suppresses human glioma cell growth, Cancer Res. 62 (2002) 1205–1212. [13] G.P. Dimri, X. Lee, G. Basile, M. Acosta, G. Scott, C. Roskelley, E.E. Medrano, M. Linskens, I. Rubelj, O. Pereira-Smith, et al., A biomarker that identifies senescent human cells in culture and in aging skin in vivo, Proc. Natl. Acad. Sci. USA 92 (1995) 9363–9367. [14] G.P. Dimri, A. Testori, M. Acosta, J. Campisi, Replicative senescence, aging and growth-regulatory transcription factors, Biol. Signals 5 (1996) 154–162. [15] A. Kueck, A.W. Opipari Jr., K.A. Griffith, L. Tan, M. Choi, J. Huang, H. Wahl, J.R. Liu, Resveratrol inhibits glucose metabolism in human ovarian cancer cells, Gynecol. Oncol. 107 (2007) 450–457. [16] J.A. Baur, K.J. Pearson, N.L. Price, H.A. Jamieson, C. Lerin, A. Kalra, V.V. Prabhu, J.S. Allard, G. Lopez-Lluch, K. Lewis, P.J. Pistell, S. Poosala, K.G. Becker, O. Boss, D. Gwinn, M. Wang, S. Ramaswamy, K.W. Fishbein, R.G. Spencer, E.G. Lakatta, D. Le Couteur, R.J. Shaw, P. Navas, P. Puigserver, D.K. Ingram, R. De Cabo, D.A. Sinclair, Resveratrol improves health and survival of mice on a high-calorie diet, Nature 444 (2006) 337–342. [17] K.B. Harikumar, B.B. Aggarwal, Resveratrol: a multitargeted agent for ageassociated chronic diseases, Cell Cycle 7 (2008) 1020–1035. [18] S.K. Goswami, D.K. Das, Resveratrol and chemoprevention, Cancer Lett. 284 (2009) 1–6. [19] Z. Liu, S.M. Oh, M. Okada, X. Liu, D. Cheng, J. Peng, D.J. Brat, S.Y. Sun, W. Zhou, W. Gu, K. Ye, Human BRE1 is an E3 ubiquitin ligase for Ebp1 tumor suppressor, Mol. Biol. Cell 20 (2009) 757–768. [20] E. Shema, I. Tirosh, Y. Aylon, J. Huang, C. Ye, N. Moskovits, N. Raver-Shapira, N. Minsky, J. Pirngruber, G. Tarcic, P. Hublarova, L. Moyal, M. Gana-Weisz, Y. Shiloh, Y. Yarden, S.A. Johnsen, B. Vojtesek, S.L. Berger, M. Oren, The histone H2B-specific ubiquitin ligase RNF20/hBRE1 acts as a putative tumor suppressor through selective regulation of gene expression, Genes Dev. 22 (2008) 2664–2676. [21] I.B. Roninson, Tumor cell senescence in cancer treatment, Cancer Res. 63 (2003) 2705–2715. [22] L. Hayflick, The limited in vitro lifetime of human diploid cell strains, Exp. Cell Res. 37 (1965) 614–636. [23] G.P. Dimri, What has senescence got to do with cancer?, Cancer Cell 7 (2005) 505–512 [24] E.H. Heiss, Y.D. Schilder, V.M. Dirsch, Chronic treatment with resveratrol induces redox stress- and ataxia telangiectasia-mutated (ATM)-dependent senescence in p53-positive cancer cells, J. Biol. Chem. 282 (2007) 26759– 26766. [25] M. Rusin, A. Zajkowicz, D. Butkiewicz, Resveratrol induces senescence-like growth inhibition of U-2 OS cells associated with the instability of telomeric DNA and upregulation of BRCA1, Mech. Ageing Dev. 130 (2009) 528–537. [26] L.L. Zamin, E.C. Filippi-Chiela, P. Dillenburg-Pilla, F. Horn, C. Salbego, G. Lenz, Resveratrol and quercetin cooperate to induce senescence-like growth arrest in C6 rat glioma cells, Cancer Sci. 100 (2009) 1655–1662.