Cellular Signalling 26 (2014) 1532–1538
Contents lists available at ScienceDirect
Cellular Signalling journal homepage: www.elsevier.com/locate/cellsig
Rlim, an E3 ubiquitin ligase, influences the stability of Stathmin protein in human osteosarcoma cells☆ Xi Chen a,1, Jianjun Shen a,1, Xingyu Li b, Xi Wang a, Min Long a, Fang Lin a, Junxia Wei a, Longfei Yang a, Chinglai Yang c, Ke Dong a,⁎, Huizhong Zhang a,⁎ a b c
Department of Medical Laboratory and Research Center, Tangdu Hospital, Fourth Military Medical University, Xi'an, China Department of Ophthalmology, Xi'an No. 4 Hospital, Xi'an, China Department of Microbiology & Immunology, Emory University School of Medicine, Atlanta, USA
a r t i c l e
i n f o
Article history: Received 19 February 2014 Received in revised form 14 March 2014 Accepted 14 March 2014 Available online 29 March 2014 Keywords: Stathmin Rlim E3 ubiquitin-protein ligase Degradation
a b s t r a c t Stathmin is an oncoprotein and is expressed at high levels in a wide variety of human malignancies, which plays important roles in maintenance of malignant phenotypes. The regulation of Stathmin gene overexpression has been wildly explored, but the exact mechanism still needs to be elucidated. It is believed that regulation of an oncogene protein abundance through post-translational modifications is essential for maintenance of malignant phenotypes. Here we identified the Rlim, a Ring H2 zinc finger protein with intrinsic ubiquitin ligase activity, as a Stathmin-interacting protein that could increase Stathmin turnover through binding with this targeted protein and then induce its degradation by proteasome in a ubiquitin-dependent manner. Inhibition of endogenous Rlim expression by siRNA could increase the level of Stathmin protein, which further led to cell proliferation and cell cycle changes in human osteosarcoma cell lines. On the other hand, forced overexpression of Rlim could decrease the level of Stathmin protein. These results demonstrate that Rlim is involved in the negative regulation of Stathmin protein level through physical interaction and ubiquitin-mediated proteolysis. Hence, Rlim is a novel regulator of Stathmin protein in a ubiquitin-dependent manner, and represents a new pathway for malignant phenotype turnover by modulating the level of Stathmin protein in human osteosarcomas. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Stathmin (STMN1), a 19 kDa cytosolic protein and also known as Op18, is a major phosphoprotein that plays crucial roles in the control of cell division and proliferation by regulating the dynamics of the microtubules [1–3]. Stathmin is also an oncoprotein and is expressed at high levels in a wide variety of human malignancies including osteosarcoma, lung cancer, uterine cervix cancer, bladder cancer, and breast cancer, which demonstrates that the overexpression of Stathmin may play an important role in maintenance of malignant phenotypes in malignant tumors [4–8]. Our previous studies on Stathmin have shown that Stathmin overexpression is associated with cellular proliferation, and inhibition of Stathmin expression can abolish the transformed
☆ Grant support: This study was supported by a grant from the National Natural Science Foundation of China (No. 81001195). ⁎ Corresponding authors at: Department of Medical Laboratory and Research Center, Tangdu Hospital, Fourth Military Medical University, Xinsi Road 710038, Xi'an, Shaanxi Province, China. Tel.: +86 29 84777470; fax: +86 29 84777654. E-mail addresses: [email protected] (K. Dong), [email protected] (H. Zhang). 1 These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.cellsig.2014.03.018 0898-6568/© 2014 Elsevier Inc. All rights reserved.
phenotypes in malignant cells [9–11]. In addition, inhibition of Stathmin expression in malignant tumor cells interferes with their proliferation through the cell cycle arrest and abrogates their transformed phenotypes [12–14]. Thus, Stathmin provides an attractive molecular target for the treatment of malignant tumors. In a search for proteins capable of interacting with Stathmin, we identified Rlim as a candidate of such proteins. Rlim (RING finger LIM domain-binding protein), encoded by the Rnf12 gene and located around 500 kb telomeric to the Xist gene on the X chromosome [15, 16], is originally identified as an E3 ubiquitin ligase. Rlim acts as a negative coregulator for LIM homeodomain transcription factors by mediating the ubiquitination and subsequent degradation of LIM cofactors LDB1 and LDB2 [17–19]. It has been reported that Rlim also plays a role in telomere length-mediated growth suppression by mediating the ubiquitination and degradation of TRF1, and acts as an activator of random inactivation of X chromosome in the embryo by targeting ZFP42 for degradation [20,21]. Furthermore, study has shown that Rlim also enhances transcriptional activation of endogenous estrogen receptor alpha (ERalpha) target genes [22,23]. Since signaling pathways controlled by ERalpha are profoundly implicated in mammary oncogenesis, Rlim may play an important role during the development of human cancers.
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
However, whether and how Rlim affects Stathmin protein levels and further influences the Stathmin functional roles during carcinogenesis have not been determined so far. In this study, we found that Rlim increased Stathmin turnover by binding it for degradation through the proteasome in a ubiquitin-dependent manner. Depletion of endogenous Rlim expression by siRNA stabilized Stathmin protein, whereas overexpression of Rlim promoted degradation of Stathmin and further impaired cell growth. In sum, these results demonstrated that Rlim was a critical negative regulator of Stathmin protein abundance and represented a new target for cancer therapy. 2. Materials and methods 2.1. Chemicals and antibodies MG132 (CAS Number: 133407-82-6) and cycloheximide (CAS Number: 66-81-9) were purchased from Sigma (MO, USA). MG132 was dissolved in DMSO and the dosage of MG132 treatment was 10 μM for 4 h unless otherwise mentioned. Cycloheximide was dissolved in water and the concentration for cell treatment was 25 μM at indicated time points. Antibodies against Stathmin (ab52630), Rlim (ab22813), Ubiquitin (ab7780) and Flag tags (ab122902) were purchased from Abcam (UK). The antibodies against His (sc-804), GST (sc-459) as well as all secondary antibodies were purchased from Santa Cruz Biotechnology (CA, USA).
1533
inserted sequences were verified by DNA sequencing. The siRNA designed towards a non-specific (NS) sequence was also used as a negative control. BLAST search against EST libraries was performed to confirm that no other human gene was targeted.
2.4. Cell culture, transfections and cell lysis Human osteosarcoma cell lines MG-63 and Saos-2, and human embryonic kidney cell line HEK-293 were purchased from ATCC (American Type Culture Collection) and cultured in complete Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum, 2 mM L-glutamine in 5% CO2 incubator at 37 °C. MG-63, Saos-2 and HEK 293 cells were seeded in 6-well plates at 2.5 × 10 4 cells per well, respectively and grown overnight to 60% confluence prior to transfection. The plasmids were transfected into above cells with Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. To select neomycin-resistant cells, 600 μg/ml G418 medium (Invitrogen, Carlsbad, CA, USA) was applied. The cells were harvested by trypsin digestion, then lysed using lysis buffer (20 mM Tris (pH 7.5), 150 mM NaCl, 1% Triton X-100) with proteinase inhibitors, as described previously [24]. Protein quantitation was done by bicinchoninic acid assay (Bio-Rad, Hercules, CA, USA).
2.2. Plasmid construction
2.5. Co-immunoprecipitation
The pcDNA3.1-Rlim-Flag and pcDNA3.1–Stathmin–Flag expression vectors were constructed by inserting the full-length Rlim cDNA and Stathmin cDNA (generated by RT-PCR with the appropriate synthetic primers) into pcDNA3.1-Flag vectors (Invitrogen, Carlsbad, CA, USA) [16], respectively. The expression vectors for pGSTag-Rlim and pGSTag-Stathmin were constructed by cloning the full-length Rlim cDNA and Stathmin cDNA into pGSTag vectors (Amersham Biosciences), respectively. The primer sequences used in this part of experiments are shown in Table 1.
Co-immunoprecipitation assay was performed as previously described [25]. In brief, cells were washed with PBS and lysed in modified RIPA buffer (20 mM Tris (pH 7.5); 150 mM NaCl, 0.5% Triton X-100, and protease inhibitor PMSF). The lysates were incubated with antibodies (3–5 μg/sample) overnight at 4 °C with gentle agitation. Protein A/G Plus Agarose (Santa Cruz, CA, USA) (20 μl/tube) was aliquoted to microcentrifuge tubes, followed by the addition of antibody conjugated lysates. Beads and lysates were incubated at 4 °C for 1 h with constant agitation. Then samples were washed and boiled for 5 min in 60 μl 2× SDS loading buffer. Then the beads were collected by centrifugation and SDS-PAGE was performed with the supernatant fraction.
2.3. SiRNA expression vector construction DNA template encoding Rlim shRNA was designed and synthesized (Genepharma Inc, Suzhou, China) as follows: the 21-nt target sequence (NM_016120, 1299–1319 bp) was chosen as a sense strand followed by a 9-nt spacer and complementary antisense strand sequence as shown in Table 1 [20]. The underlined sequences are BamHI and HindIII enzyme sites. The shRNAs were annealed and subcloned into pSilencer4.1-CMVneo (Ambion, Austin, TX, USA) BamHI and HindIII enzyme sites downstream of CMV promoter, and recombinant vector was named as pSilencer4.1-Rlim-siRlim. The recombinant vector was confirmed by the digestion analysis of restriction endonuclease and all Table 1 The primer sequences used in the construction of recombinant vectors. Vectors
Primers
pcDNA3.1-Rlim-Flag
Forward: 5′-CGGGATCCATGGAAAACTCAGATTCC-3′ Reverse: 5′-CGGCTCGAGTTACACAACACTTTCTCTGC-3′ Forward: 5′-CGGGATCCATGGCTTCTTCTGATATCCAGG-3′ Reverse: 5′-CGGCTCGAGTTAGTCAGCTTCAGTCTCG-3′ Forward: 5′-CGCCATGGGTGAAAACTCAGATTCC-3′ Reverse: 5′-CGCTC GAG TTACACAACACTTTCTCTGC-3′ Forward: 5′-CGCCATGGGT GCTTCTTCTGATATCCAGG-3′ Reverse: 5′-CGCTCGAG TTAGTCAGCTTCAGTCTCG-3′ Sense: 5′-GATCCGGTCTCAGACACCAAACAAC ATTCAAGAGATGTTGTTTGGTGTCTGAGACCTAA-3′ Antisense: 5′-AGCTTATGGTCTCAGACACCAAACAACATC TCTTGAATGTTGTTTG GTGTCTGAGACCG-3′
pcDNA3.1–Stathmin–Flag pGSTag–Rlim pGSTag–Stathmin pSilencer4.1-Rlim-siRlim
2.6. Protein purification and GST-pulldown The prokaryotic expression vectors of GST-Stathmin, GST-Rlim and GST were expressed in BL21 (DE3) pLysS Escherichia coli, respectively. The cell lysates were prepared following the manufacturer's protocol (GE Healthcare Bio-Sciences AB, Sweden) and were allowed to bind to Glutathione Sepharose 4B for 1 h at room temperature, washed thrice with Binding Buffer TN1 (140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.3), and then eluted with 600 μl of elution buffer TN2 (50 mM Tris–HCl, 10 mM reduced glutathione, pH 8.0). The eluted proteins GST-Stathmin, GST-Rlim and GST were analyzed by SDS-PAGE and stained with Coomassie Blue to check for the protein purity [26]. Then, 50 μg of GST or GST-Stathmin was incubated with 50 μl (50% slurry) of Glutathione HiCap Matrix (Qiagen) in Binding Buffer TN3 (50 mM Tris–HCl, 200 mM NaCl, 1 mM EDTA, 1% NP40, 1 mM DTT, 10 mM MgCl2, pH 8.0 and protease inhibitor PMSF) at room temperature with rocking for preclearing. After 2 h, the mixtures were collected in separate tubes. 500 μg of pcDNA3.1-Rlim-Flag transfected MG-63 cell lysate was added to each tube containing Glutathione HiCap Matrix treated GST or GST-Stathmin, and kept overnight under similar conditions. The beads were washed 3 times with buffer TN1 and eluted with 50 μl of elution buffer TN4 (50 mM Tris–HCl pH 8.0, 400 mM NaCl, 50 mM reduced glutathione, 1 mM EDTA, 1 mM DTT) and then subjected to Western blot analysis with anti-Rlim antibody [27].
1534
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
2.7. Protein identification by mass spectrometry The analysis of proteins by liquid chromatography combined with tandem mass spectrometry (LC/MS/MS) was conducted as previously reported [28], and it was performed by Beijing Proteome Research Center, China. 2.8. Ubiquitin ligase activity assay
for 24 h in 96-well plate and then transfected with pcDNA3.1 vector or pcDNA3.1-Rlim-Flag vector. At different time points (after 12, 24, 48, 72 h of transfection), the medium was replaced with 100 μl of 5% MTT (5 mg/ml), and the plate was incubated for another 4 h at 37 °C. After incubation, the culture medium was removed gently, and 100 μl of DMSO was added. Finally, the absorbance was determined on a microreader (Bio-Rad) at 570 nm. The cell proliferation curve was plotted using the absorbance at each time point and each sample was done in triplicate. Approximately 1.0 × 106 cells were cultured for 24 h in 6-well plates and transfected with pcDNA3.1 vector or pcDNA3.1-Rlim-Flag vector. After 48 h of transfection, the cells were collected, fixed with 70% ethanol and stained with propidiumiodide (PI), and then cell cycle distribution was analyzed on a FACSCalibur system (BD Bioscience, Bedford, MA, USA) by ModFIT software (Verity Software House, Topsham, ME, USA) [31].
For in vivo ubiquitination assay, the transfected MG-63 cells were treated with or without proteasome inhibitor of MG132 at a concentration of 10 μm for 4 h, and then lysed with modified RIPA buffer (20 mM Tris (pH 7.5); 150 mM NaCl, 0.5% Triton X-100, and protease inhibitor PMSF). Identical aliquots of cell lysates (2 mg) were used in immunoprecipitation assays with anti-Stathmin monoclonal antibody. Immunocomplexes were separated on 12% gradient gels and immunoblotting was performed using anti-Ub antibody. An in vitro ubiquitination assay was performed using the Ubiquitinylation kit (Enzo life sciences) following the manufacturer's protocol [29]. In brief, 2 μg of purified GST-Stathmin or GST was used as the substrate, a ubiquitination cocktail was prepared in the presence or absence of either GST-Rlim (100 nM) or ‘F1 and F2’ (containing ubiquitin activating enzyme E1 (100 nM) and ubiquitin carrier protein E2 (2.5 μM) alone or in combination), then incubated at 37 °C for 4 h along with ATP (5 mM) and Ub (2.5 μM), and the total reaction volume was 50 μl. The incubated samples were then checked by anti-Stathmin antibody.
3. Results
2.9. Cell proliferation assay and flow cytometry analysis of cell cycle
3.1. Identification of Rlim as a partner of Stathmin in osteosarcoma cells
Cell proliferation was determined by MTT (Sigma, St. Louis, MO, USA) assay as described elsewhere [30]. 6 × 103 cells were cultured
To identify the natural protein partners of Stathmin in osteosarcoma cells, we expressed a Flag–Stathmin in the human osteosarcoma cell
2.10. Statistical analysis Each experiment was repeated 3 times and all data were indicated as the mean ± standard deviation (±s). Statistical analyses were performed using SPSS version 17.0 (IBM Company, Chicago, IL, USA). Student's t-test and one-way analysis of variance (ANOVA) followed by Dunnett's multiple comparison tests were adopted. Values of P b 0.05 were considered as significant and indicated by asterisks in the figures.
Fig. 1. GST pull down, co-immunoprecipitation assay and endogenous Stathmin level detection results. (A) Identification of Rlim in Stathmin complexes from human osteosarcoma cell line. Silver staining of affinity purified Flag–Stathmin complexes from MG-63 cells. Specific Stathmin-interacting proteins were identified by mass spectrometry and are indicated. (B) GST and GST-Stathmin were used for GST pull down assay in 500 μg of pcDNA3.1-Rlim-Flag transfected MG-63 cell lysates and Western blotting results showed positive bands in all three different primary antibodies of anti-Stathmin, anti-Rlim, and anti-GST. (C) Proteins were immunoprecipitated from MG-63 whole cell lysates with antibodies against Stathmin and checked for Rlim. The Western blotting results showed positive bands in both anti-Stathmin and anti-Rlim antibodies panels. (D) Proteins were immunoprecipitated from MG-63 whole cell lysates with antibodies against Rlim and checked for Stathmin. The Western blotting results showed positive bands in both anti-Stathmin and anti-Rlim antibodies panels. (E) Left figure, MG-63 cells were transfected with plasmids expressing the full-length Rlim or the empty vectors. The level of endogenous Stathmin was examined by Western blotting and the results showed that with the increase of Rlim protein expression, the Stathmin protein level decreased compared to control group. Right figure, densitometry of the immunoblots shown in D (*P b 0.05, **P b 0.01 vs control, n = 3).
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
1535
Fig. 2. Results of co-immunoprecipitation and ubiquitin ligase activity assay in vivo and vitro. (A) MG-63 cells were transfected with Stathmin alone or in combination in the presence or absence of Rlim as shown in the figure, then immunoprecipitation was performed with anti-Stathmin antibody, and antibodies against Ub, Stathmin were used in Western blotting, respectively. The results showed that in the presence of Rlim and Stathmin, the Stathmin ubiquitination level was enhanced compared to the control groups (the third lane from left). (B) MG-63 cells were transfected with Stathmin alone or in combination in the presence or absence of Rlim as shown in the figure, followed by a treatment with MG132, then immunoprecipitation was performed with anti-Stathmin antibody, and antibodies against Ub, Stathmin were used in Western blotting, respectively. The results showed that, in the presence of Rlim and Flag–Stathmin, the Stathmin ubiquitination level was increased compared with control groups (the third lane from left), and the level of poly-ubiquitination of Stathmin was enhanced upon treatment with MG132 (top panel). (C) Left figure, HEK293 cells were transfected with either pSilencer4.1-Rlim-siR or pSilencer4.1-CMVneo (negative control) vector, respectively. After proteasome inhibitor MG132 treatment, the whole cell lysates were detected by Western blotting with anti-Stathmin antibody and the results showed that with the decrease of Rlim protein expression, the Stathmin ubiquitination level decreased accordingly compared to control group. Right figure, densitometry of the immunoblots shown in C (*P b 0.05, **P b 0.01 vs control, n = 3). (D) GST-Rlim, GST-Stathmin and GST were purified using GST column. The purified proteins were then checked by SDS-PAGE and stained with Coomassie on the left figure. GST or GST-Stathmin was incubated for 4 h at 37 °C with or without GST-Rlim, F1 and F2 in the presence of Ub and ATP. The incubated mixtures were then checked by Western blotting with anti-Stathmin antibody and the results showed that in the presence of Rlim, Ub, F1, F2, ATP and GST-Stathmin, poly-ubiquitinated Stathmin appears as a ladder with higher molecular weight forms on the right figure (fourth lane from left).
line MG-63. We analyzed proteins that co-purified with Flag–Stathmin by LC/MS/MS after sequential Flag immunoprecipitation and peptide elution [28] and discovered that Stathmin complexes contain Rlim (Fig. 1A). Rlim (Ring finger LIM domain-binding protein, also known as RNF12), is originally identified as an E3 ubiquitin ligase and capable of targeting cofactors of LIM-HD proteins (CLIM) for proteasomedependent degradation, thereby inhibiting developmental LIM homeodomain activity [17,18]. In our studies, we firstly discovered that Rlim could also interact with Stathmin in osteosarcoma cells. By using purified GST-Stathmin in a pull-down assay, we demonstrated that Flag–Rlim interacted specifically with Stathmin in vitro (Fig. 1B), which confirmed the partner relationship between Rlim and Stathmin. To further reveal if there was an endogenous binding property between Rlim and Stathmin, we performed co-immunoprecipitation assays in MG-63 cells and found that there was an integration band of Rlim and Stathmin proteins (Fig. 1C, D), which demonstrated that Rlim could bind to the Stathmin protein in vivo and suggested that Rlim might play important role in Stathmin post-translational modifications. Based on the above findings, we overexpressed Rlim in MG-63 cells to determine whether Rlim could affect the steady-state levels of Stathmin, and found that overexpression of full-length Rlim could lead to a marked reduction in the steady-state levels of endogenous Stathmin protein (Fig. 1E). Altogether, it could be concluded that Rlim
might be a potential E3 ligase for Stathmin and played important roles in Stathmin post-translational regulation. 3.2. Rlim induces ubiquitination of Stathmin Since Rlim interacted with Stathmin, it was a matter of interest to check whether Rlim played a ubiquitin ligase role for Stathmin protein degradation. To confirm this possibility, we performed ubiquitination assay with Rlim as an E3 ubiquitin ligase. Under in-vivo conditions, we overexpressed Stathmin together with or without Rlim in MG-63 cells. Whole cell lysates prepared from 48 h post-transfection cells were immunoprecipitated using anti-Stathmin antibody. Western blotting results showed that the presence of poly-ubiquitinated Stathmin adducts and the level of poly-ubiquitination of Stathmin was markedly increased in the presence of Rlim compared with control groups, while this phenomenon was enhanced upon treatment with MG132, a proteasome inhibitor, which suggested that Stathmin could be polyubiquitinated in part by Rlim (E3 ligase) and degraded by the Ubproteasome system. (Fig. 2A, B). To confirm the role of Rlim in ubiquitinating Stathmin, we firstly tried to find out whether this function of Rlim was existent under endogenous condition. Accordingly, Rlim was knocked down in HEK293 cells by siRNA and the lysates, with the proteasome inhibitor MG132, were subjected to
1536
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
Fig. 3. Stathmin protein level and half-life detection results. (A) Left figure, HEK293 cells were transfected with pSilencer4.1-Rlim-siR or pSilencer4.1-CMVneo (negative control) vectors, respectively. Western blotting was done using anti-Stathmin antibody and the results showed that with the decrease of Rlim protein expression, the Stathmin protein level increased compared to control groups. Right figure, densitometry of the immunoblots shown in A (*P b 0.05, **P b 0.01 vs control, n = 3). (B) MG-63 cells were transfected with plasmids expressing the full-length Rlim or the empty vectors, respectively followed by a treatment with 50 μM cycloheximide (CHX) for a time period as indicated in the figure; the lysates were then checked for Stathmin level to determine the protein half-life. The half life of Stathmin protein was about 10 h in natural condition, but it was reduced to ~6 h in the condition of Rlim overexpression. (C) HEK293 cells were transfected with pSilencer4.1-Rlim-siR or pSilencer4.1-CMVneo vectors, respectively, followed by a treatment with 50 μM cycloheximide (CHX) for a time period as indicated in the figure, the lysates were then checked for Stathmin level. The results showed that the Stathmin protein level did not change obviously even in the condition of decreased Rlim protein expression compared to control group. (D) MG-63 cells were transfected with plasmids expressing the full-length Rlim or the empty vectors, and then treated with or without MG132. The cell lysates were then checked for Stathmin protein level. The results showed that, with the MG132 treatment, the reduction of Stathmin protein level was ablated even in the condition of increased Rlim expression compared to MG132 untreated cells (top panel).
immunoprecipitation with anti-Stathmin antibody. Western blotting analysis revealed that depletion of endogenous Rlim significantly reduced the Stathmin ubiquitination level (Fig. 2C). Then, we performed in vitro ubiquitination assay to rule out the involvement of any other E3 ligase(s) in the ubiquitination of Stathmin as well as to establish Rlim as a novel E3 ligase for Stathmin. GST-Stathmin, GST-Rlim and GST were purified using Glutathione Sepharose 4B. With either GST-Stathmin or GST as the substrate, a ubiquitination cocktail was prepared in the presence or absence of either GST-Rlim or ‘F1 and F2’ (containing ubiquitin activating enzyme E1 and ubiquitin carrier protein E2) alone or in combination, incubated at 37 °C for 4 h along with ATP and Ub. The incubated samples were then checked by anti-Stathmin antibody. The results quite clearly highlighted the capability of Rlim to independently ubiquitinate Stathmin (Fig. 2D). Collectively, these results indicated that Rlim could ubiquitinate Stathmin under both in vivo and in vitro conditions.
3.3. Rlim promotes proteasome mediated degradation of Stathmin With the findings that Stathmin was ubiquitinated by Rlim, we examined the Stathmin protein level under the influence of Rlim. Our results demonstrated that forced overexpression of Rlim in MG-63 cells decreased protein level of Stathmin (Fig. 1D), conversely, knock down of Rlim by siRNA in HEK293 cells produced the anticipated antagonistic effect (Fig. 3A). It was established that Rlim negatively regulated the steady state level of Stathmin protein. To further investigate this function, we observed endogenous level of Stathmin protein under the condition of ectopic overexpression of Rlim in MG-63 cells or knocked down Rlim by siRNA in HEK293 cells, followed by treatment with cycloheximide (protein synthesis inhibitor). The results showed that Stathmin protein half life was either decreased in the condition of overexpressed Rlim or increased in the condition of knock down of Rlim, respectively (Fig. 3B,
Fig. 4. The results of cell proliferation analysis. Human osteosarcoma cells (MG-63 and Saos-2) were transfected with plasmids expressing the full-length Rlim or the empty vectors (4 μg). Cell viability was evaluated by MTT assay for a time period as indicated in the figure. Compared with the control cells, the highest inhibitory rate of cell proliferation was 27.28 ± 3.21% and 32.72 ± 3.63%, respectively in MG-63 cells and Saos-2 cells on 72 h (P b 0.05).
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
1537
C). The half life of Stathmin protein was about 10 h in natural condition, but it was reduced to ~6 h in the condition of overexpression of Rlim. To verify the involvement of the proteasome system in Rlim mediated degradation of Stathmin protein, we treated MG-63 cells with MG132, which stabilized Stathmin and ablated the reduction of Stathmin protein level even in the presence of Rlim (Fig. 3D). All these results confirmed the involvement of the proteasome system in Stathmin degradation by Rlim.
percentage of G2/M phase increased by 11.01 ± 1.7% and 11.55 ± 1.9% respectively in Rlim-overexpression MG-63 and Saos-2 cells (P b 0.05) (Fig. 5). These results demonstrated that Rlim–Stathmin ubiquitination pathway was effective in reversing the malignant phenotype of osteosarcoma cells caused by overexpression of Stathmin.
3.4. The Rlim–Stathmin ubiquitination pathway controls cell proliferation and cell cycle
Stathmin, a kind of small molecular weight cytosolic phosphoprotein, regulates microtubule dynamics and influences cell cycle through integrating and transducing various transduction signals within cells. Expression studies have shown that Stathmin is aberrantly overexpressed in the majority of human malignancies, and plays an important role in maintaining malignant phenotype of tumors [33]. Stathmin expression and its function are controlled by multiple steps and different regulating levels from transcription, phosphorylation and ubiquitination. The regulation of Stathmin expression and function is very complicated, and the mechanisms for its overexpression in the malignant tumors have not been fully determined [34–37]. Historically, the focus has been on the transcriptional regulation of Stathmin overexpression, while in this study, we conducted researches on the posttranslational modifications of Stathmin. We, for the first time, verified that Rlim was a novel Stathmin binding protein, and further found that Rlim regulated Stathmin degradation in a ubiquitin-dependent manner, which might explain, in part, the reason of high level Stathmin protein in most of the malignant tumors cells. We firstly examined the biological role of Rlim, a ring-domain E3 ubiquitin ligase, and demonstrated the existence of a Rlim–Stathmin
To analyze phenotypic changes caused by Rlim–Stathmin ubiquitination pathway, we investigated the effects of ectopic overexpression of Rlim on cell proliferation in osteosarcoma cells. Cell proliferation was evaluated by MTT assay and the results showed a significant inhibition of the proliferation both in MG-63 cells and Saos-2 cells, and the highest inhibitory rate was 27.28 ± 3.21% and 32.72 ± 3.63% compared with control vector transfected cells, respectively on 72 h after transfection (P b 0.05) (Fig. 4). It has been reported that the proliferation inhibition of tumor cells by knockdown of Stathmin expression is caused mainly by cell cycle arrest and microtubule assembly in other types of mammalian cells [9,32]. To reveal the mechanisms underlying ubiquitination-mediated proliferation inhibition, we used flow cytometric analysis to detect cell cycle changes of MG-63 and Saos-2 cells that were transient transfected with 4 μg plasmids expressing the fulllength Rlim, respectively. The population of G2/M phase was significantly increased in both MG-63 and Saos-2 cells 72 h later after plasmid transfection compared with control cells. As shown in Fig. 5, the
4. Discussion
Fig. 5. The results of cell cycle analysis. Cells were harvested 72 h after transfection of plasmids expressing the full-length Rlim or the empty vectors (4 μg) and subsequently analyzed for their DNA content by flow cytometry. Compared with control cells, the percentage of G2/M phase increased by 11.01 ± 1.7% and 11.55 ± 1.9%, respectively in MG-63 and Saos-2 cells in the condition of Rlim-overexpression (P b 0.05).
1538
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
pathway, in which Rlim operated on a linear pathway epistatic to Stathmin. Rlim interacted with Stathmin and promoted its ubiquitination in vitro and in vivo, thus acting as a negative regulator of Stathmin. Whereas, overexpression of Rlim reduced the half-life of endogenous Stathmin protein from ~ 10 h to ~ 6 h, and depletion of Rlim resulted in an increase of endogenous Stathmin protein level. These results supported the hypothesis that Rlim played a critical role in regulating the level of Stathmin protein within the cells. It was reported that Stathmin played important roles in maintenance of malignant phenotypes in human malignancies and the inhibition of Stathmin gene expression could lead to proliferation and cell-cycle arrest in tumor cells [9,11,37]. Our results showed that the interaction of Rlim with Stathmin resulted in converting of tumor cell malignant phenotypes as this interaction could down regulate Stathmin protein level, and subsequently lead to inhibition of cell proliferation and accumulation of G2/M phase of osteosarcoma MG-63 and Saos-2 cells (Figs. 4 and 5). These findings suggested that Rlim might be a potential therapeutic target for the treatment of osteosarcomas. We have demonstrated that Stathmin is subject to regulation via the ubiquitin proteasome through interaction with the E3-ubiquitin ligase Rlim. The importance of this type of interaction between regulatory proteins and E3-ligases has recently been emphasized in a number of reports [38,39]. For example, E3-ligase Huwe1 interacts with N-Myc and mediates its degradation, this interaction is deemed essential for it restrains proliferation and enables neuronal differentiation [39]. Furthermore, the ubiquitin-mediated degradation of PARP-1 plays an important role in cell cycle regulation and cancer therapy [38]. As such, we propose that Rlim may represent a new pathway for the regulation of cell cycle and cell proliferation in osteosarcoma cells by acting as a negative regulator of Stathmin protein level. In conclusion, our experiment results demonstrated that Rlim could increase Stathmin turnover by targeting it for degradation through the proteasome in a ubiquitin-dependent manner, which led to proliferation and cell-cycle arrest and malignant phenotype changes in osteosarcoma cells. We believe that the interaction of Rlim and Stathmin will be a promising strategy in targeted tumor gene therapy and may have potential for clinical use in tumor treatment. Acknowledgment Thanks to everyone from the Department of Clinical Laboratory for their sincere help and excellent technical assistance. This study was supported by a grant from the National Natural Science Foundation of China (No. 81001195). References [1] P. Niethammer, P. Bastiaens, E. Karsenti, Science 303 (2004) 1862–1866. [2] C.I. Rubin, G.F. Atweh, J. Cell. Biochem. 93 (2004) 242–250. [3] L. Cassimeris, Cell Biol. 14 (2002) 18–24.
[4] G. Chen, H. Wang, T.G. Gharib, C.C. Huang, D.G. Thomas, K.A. Shedden, R. Kuick, J.M. Taylor, S.L. Kardia, D.E. Misek, T.J. Giordano, M.D. Iannettoni, M.B. Orringer, S.M. Hanash, D.G. Beer, Mol. Cell. Proteomics 2 (2003) 107–116. [5] X. Wang, J.H. Ren, F. Lin, J.X. Wei, M. Long, L. Yan, H.Z. Zhang, Cancer Biol. Ther. 10 (2010) 632–643. [6] S. Rana, P.B. Maples, N. Senzer, J. Nemunaitis, Expert. Rev. Anticancer. Ther. 8 (2008) 1461–1470. [7] S.J. Mistry, G.F. Atweh, Mt Sinai J. Med. 69 (2002) 299–304. [8] H.Z. Zhang, P. Gao, L. Yan, F. Lin, Ai Zheng 23 (2004) 493–496. [9] H.Z. Zhang, Y. Wang, P. Gao, F. Lin, L. Liu, B. Yu, J.H. Ren, H. Zhao, R. Wang, Cancer Biol. Ther. 5 (2006) 1457–1461. [10] F. Lin, P. Gao, L. Liu, R. Wang, H.Z. Zhang, J. Mod. Oncol. 15 (2007) 485–487. [11] R. Wang, K. Dong, F. Lin, X. Wang, P. Gao, S.H. Wei, S.Y. Cheng, H.Z. Zhang, Mol. Med. 13 (2007) 567–575. [12] B. Belletti, G. Baldassarre, Expert Opin. Ther. Targets 15 (2011) 1249–1266. [13] M.O. Steinmetz, J. Struct. Biol. 158 (2007) 137–147. [14] L. Lee, J.S. Tirnauer, J. Li, S.C. Schuyler, J.Y. Liu, D. Pellman, Science 287 (2000) 2260–2262. [15] I. Jonkers, T.S. Barakat, E.M. Achame, K. Monkhorst, A. Kenter, E. Rentmeester, F. Grosveld, J.A. Grootegoed, J. Gribnau, Cell 139 (2009) 999–1011. [16] H.P. Ostendorff, M. Bossenz, A. Mincheva, N.G. Copeland, D.J. Gilbert, N.A. Jenkins, P. Lichter, I. Bach, Genomics 69 (2000) 120–130. [17] I. Bach, E.C. Rodriguez, C. Carrière, A. Bhushan, A. Krones, D.W. Rose, C.K. Glass, B. Andersen, J.C. Izpisúa Belmonte, M.G. Rosenfeld, Nat. Genet. 22 (1999) 394–399. [18] H.P. Ostendorff, R.I. Peirano, M.A. Peters, A. Schlüter, M. Bossenz, M. Scheffner, I. Bach, Nature 416 (2002) 99–103. [19] C. Güngör, N.I. Taniguchi, H. Ma, A. Drung, B. Tursun, H.P. Ostendorff, M. Bossenz, C. G. Becker, T. Becke, I. Bach, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 15000–15005. [20] Y.R. Her, I.K. Chung, J. Biol. Chem. 284 (2009) 8557–8566. [21] I. Bach, Cell Res. 22 (2012) 1524–1526. [22] R.R. Singh, C.J. Barnes, A.H. Talukder, S.A. Fuqua, R. Kumar, Cancer Res. 65 (2005) 10594–10601. [23] S.A. Johnsen, C. Güngör, T. Prenzel, S. Riethdorf, L. Riethdorf, N.I. Taniguchi, T. Rau, B. Tursun, J.D. Furlow, G. Sauter, M. Scheffne, K. Pantel, F. Gannon, I. Bach, Cancer Res. 69 (2009) 128–136. [24] R. Nikolay, T. Wiederkehr, W. Rist, G. Kramer, M.P. Mayer, B. Bukau, J. Biol. Chem. 279 (2004) 2673–2678. [25] L. Kashima, M. Toyota, H. Mita, H. Suzuki, M. Idogawa, K. Ogi, Y. Sasaki, T. Tokino, Oncogene 28 (2009) 2643–2653. [26] X. Chen, K. Dong, M. Long, F. Lin, X. Wan, J.X. Wei, J.H. Ren, H.Z. Zhang, Oncol. Lett. 2 (2012) 319–323. [27] S.F. Ahmed, S. Deb, I. Paul, A. Chatterjee, T. Mandal, U. Chatterjee, M.K. Ghosh, J. Biol. Chem. 287 (2012) 15996–16006. [28] M. Idogawa, M. Masutani, M. Shitashige, K. Honda, T. Tokino, Y. Shinomura, K. Imai, S. Hirohashi, T. Yamada, Cancer Res. 67 (2007) 911–918. [29] X. Li, D.F. Lu, F.J. He, H.B. Zhou, Q. Liu, Y. Wang, C.S. Shao, Y.Q. Gong, J. Biol. Chem. 286 (2011) 32344–32354. [30] Ch.D. Fan, J. Zhao, B.X. Zhao, Sh.l. Zhang, J.y. Miao, Chem. Res. Toxicol. 22 (2009) 1517–1525. [31] K. Ogi, M. Toyota, H. Mita, A. Satoh, L. Kashima, Y. Sasaki, H. Suzuki, K. Akino, N. Nishikawa, M. Noguchi, Y. Shinomura, K. Imai, H. Hiratsuka, T. Tokino, Cancer Biol. Ther. 4 (2005) 773–780. [32] P. Gao, F. Lin, J. Modern Oncol. 2 (2006) 210–212. [33] S. Rana, P.B. Maples, N. Senzer, J. Nemunaitis, Expert. Rev. Anticancer 8 (2008) 1461–1470. [34] R.G. Polzin, H. Benlhabib, J. Trepel, J.E. Herrera, Gene 341 (2004) 209–218. [35] S. Polager, D. Ginsberg, J. Biol. Chem. 278 (2003) 14431449. [36] H. Benlhabib, J.E. Herrera, Gene 377 (2006) 177–185. [37] F. Lin, M. Long, L. Yan, P. Gao, R. Wang, Y. Zhang, X. Wang, T. He, L. Liu, K. Dong, J.H. Reng, H.Zh. Zhang, Cell. Signal. 22 (2010) 166–173. [38] L. Kashima, M. Idogawa, H. Mita, M. Shitashige, T. Yamada, K. Ogi, H. Suzuki, M. Toyota, H. Ariga, Y. Sasaki, T. Tokino, J. Biol. Chem. 287 (2012) 12975–12984. [39] X. Zhao, J.I. Heng, D. Guardavaccaro, R. Jiang, M. Pagano, F. Guillemot, A. Iavarone, A. Lasorella, Nat. Cell Biol. 10 (2008) 643–653.
Contents lists available at ScienceDirect
Cellular Signalling journal homepage: www.elsevier.com/locate/cellsig
Rlim, an E3 ubiquitin ligase, influences the stability of Stathmin protein in human osteosarcoma cells☆ Xi Chen a,1, Jianjun Shen a,1, Xingyu Li b, Xi Wang a, Min Long a, Fang Lin a, Junxia Wei a, Longfei Yang a, Chinglai Yang c, Ke Dong a,⁎, Huizhong Zhang a,⁎ a b c
Department of Medical Laboratory and Research Center, Tangdu Hospital, Fourth Military Medical University, Xi'an, China Department of Ophthalmology, Xi'an No. 4 Hospital, Xi'an, China Department of Microbiology & Immunology, Emory University School of Medicine, Atlanta, USA
a r t i c l e
i n f o
Article history: Received 19 February 2014 Received in revised form 14 March 2014 Accepted 14 March 2014 Available online 29 March 2014 Keywords: Stathmin Rlim E3 ubiquitin-protein ligase Degradation
a b s t r a c t Stathmin is an oncoprotein and is expressed at high levels in a wide variety of human malignancies, which plays important roles in maintenance of malignant phenotypes. The regulation of Stathmin gene overexpression has been wildly explored, but the exact mechanism still needs to be elucidated. It is believed that regulation of an oncogene protein abundance through post-translational modifications is essential for maintenance of malignant phenotypes. Here we identified the Rlim, a Ring H2 zinc finger protein with intrinsic ubiquitin ligase activity, as a Stathmin-interacting protein that could increase Stathmin turnover through binding with this targeted protein and then induce its degradation by proteasome in a ubiquitin-dependent manner. Inhibition of endogenous Rlim expression by siRNA could increase the level of Stathmin protein, which further led to cell proliferation and cell cycle changes in human osteosarcoma cell lines. On the other hand, forced overexpression of Rlim could decrease the level of Stathmin protein. These results demonstrate that Rlim is involved in the negative regulation of Stathmin protein level through physical interaction and ubiquitin-mediated proteolysis. Hence, Rlim is a novel regulator of Stathmin protein in a ubiquitin-dependent manner, and represents a new pathway for malignant phenotype turnover by modulating the level of Stathmin protein in human osteosarcomas. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Stathmin (STMN1), a 19 kDa cytosolic protein and also known as Op18, is a major phosphoprotein that plays crucial roles in the control of cell division and proliferation by regulating the dynamics of the microtubules [1–3]. Stathmin is also an oncoprotein and is expressed at high levels in a wide variety of human malignancies including osteosarcoma, lung cancer, uterine cervix cancer, bladder cancer, and breast cancer, which demonstrates that the overexpression of Stathmin may play an important role in maintenance of malignant phenotypes in malignant tumors [4–8]. Our previous studies on Stathmin have shown that Stathmin overexpression is associated with cellular proliferation, and inhibition of Stathmin expression can abolish the transformed
☆ Grant support: This study was supported by a grant from the National Natural Science Foundation of China (No. 81001195). ⁎ Corresponding authors at: Department of Medical Laboratory and Research Center, Tangdu Hospital, Fourth Military Medical University, Xinsi Road 710038, Xi'an, Shaanxi Province, China. Tel.: +86 29 84777470; fax: +86 29 84777654. E-mail addresses: [email protected] (K. Dong), [email protected] (H. Zhang). 1 These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.cellsig.2014.03.018 0898-6568/© 2014 Elsevier Inc. All rights reserved.
phenotypes in malignant cells [9–11]. In addition, inhibition of Stathmin expression in malignant tumor cells interferes with their proliferation through the cell cycle arrest and abrogates their transformed phenotypes [12–14]. Thus, Stathmin provides an attractive molecular target for the treatment of malignant tumors. In a search for proteins capable of interacting with Stathmin, we identified Rlim as a candidate of such proteins. Rlim (RING finger LIM domain-binding protein), encoded by the Rnf12 gene and located around 500 kb telomeric to the Xist gene on the X chromosome [15, 16], is originally identified as an E3 ubiquitin ligase. Rlim acts as a negative coregulator for LIM homeodomain transcription factors by mediating the ubiquitination and subsequent degradation of LIM cofactors LDB1 and LDB2 [17–19]. It has been reported that Rlim also plays a role in telomere length-mediated growth suppression by mediating the ubiquitination and degradation of TRF1, and acts as an activator of random inactivation of X chromosome in the embryo by targeting ZFP42 for degradation [20,21]. Furthermore, study has shown that Rlim also enhances transcriptional activation of endogenous estrogen receptor alpha (ERalpha) target genes [22,23]. Since signaling pathways controlled by ERalpha are profoundly implicated in mammary oncogenesis, Rlim may play an important role during the development of human cancers.
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
However, whether and how Rlim affects Stathmin protein levels and further influences the Stathmin functional roles during carcinogenesis have not been determined so far. In this study, we found that Rlim increased Stathmin turnover by binding it for degradation through the proteasome in a ubiquitin-dependent manner. Depletion of endogenous Rlim expression by siRNA stabilized Stathmin protein, whereas overexpression of Rlim promoted degradation of Stathmin and further impaired cell growth. In sum, these results demonstrated that Rlim was a critical negative regulator of Stathmin protein abundance and represented a new target for cancer therapy. 2. Materials and methods 2.1. Chemicals and antibodies MG132 (CAS Number: 133407-82-6) and cycloheximide (CAS Number: 66-81-9) were purchased from Sigma (MO, USA). MG132 was dissolved in DMSO and the dosage of MG132 treatment was 10 μM for 4 h unless otherwise mentioned. Cycloheximide was dissolved in water and the concentration for cell treatment was 25 μM at indicated time points. Antibodies against Stathmin (ab52630), Rlim (ab22813), Ubiquitin (ab7780) and Flag tags (ab122902) were purchased from Abcam (UK). The antibodies against His (sc-804), GST (sc-459) as well as all secondary antibodies were purchased from Santa Cruz Biotechnology (CA, USA).
1533
inserted sequences were verified by DNA sequencing. The siRNA designed towards a non-specific (NS) sequence was also used as a negative control. BLAST search against EST libraries was performed to confirm that no other human gene was targeted.
2.4. Cell culture, transfections and cell lysis Human osteosarcoma cell lines MG-63 and Saos-2, and human embryonic kidney cell line HEK-293 were purchased from ATCC (American Type Culture Collection) and cultured in complete Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum, 2 mM L-glutamine in 5% CO2 incubator at 37 °C. MG-63, Saos-2 and HEK 293 cells were seeded in 6-well plates at 2.5 × 10 4 cells per well, respectively and grown overnight to 60% confluence prior to transfection. The plasmids were transfected into above cells with Lipofectamine 2000 transfection reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. To select neomycin-resistant cells, 600 μg/ml G418 medium (Invitrogen, Carlsbad, CA, USA) was applied. The cells were harvested by trypsin digestion, then lysed using lysis buffer (20 mM Tris (pH 7.5), 150 mM NaCl, 1% Triton X-100) with proteinase inhibitors, as described previously [24]. Protein quantitation was done by bicinchoninic acid assay (Bio-Rad, Hercules, CA, USA).
2.2. Plasmid construction
2.5. Co-immunoprecipitation
The pcDNA3.1-Rlim-Flag and pcDNA3.1–Stathmin–Flag expression vectors were constructed by inserting the full-length Rlim cDNA and Stathmin cDNA (generated by RT-PCR with the appropriate synthetic primers) into pcDNA3.1-Flag vectors (Invitrogen, Carlsbad, CA, USA) [16], respectively. The expression vectors for pGSTag-Rlim and pGSTag-Stathmin were constructed by cloning the full-length Rlim cDNA and Stathmin cDNA into pGSTag vectors (Amersham Biosciences), respectively. The primer sequences used in this part of experiments are shown in Table 1.
Co-immunoprecipitation assay was performed as previously described [25]. In brief, cells were washed with PBS and lysed in modified RIPA buffer (20 mM Tris (pH 7.5); 150 mM NaCl, 0.5% Triton X-100, and protease inhibitor PMSF). The lysates were incubated with antibodies (3–5 μg/sample) overnight at 4 °C with gentle agitation. Protein A/G Plus Agarose (Santa Cruz, CA, USA) (20 μl/tube) was aliquoted to microcentrifuge tubes, followed by the addition of antibody conjugated lysates. Beads and lysates were incubated at 4 °C for 1 h with constant agitation. Then samples were washed and boiled for 5 min in 60 μl 2× SDS loading buffer. Then the beads were collected by centrifugation and SDS-PAGE was performed with the supernatant fraction.
2.3. SiRNA expression vector construction DNA template encoding Rlim shRNA was designed and synthesized (Genepharma Inc, Suzhou, China) as follows: the 21-nt target sequence (NM_016120, 1299–1319 bp) was chosen as a sense strand followed by a 9-nt spacer and complementary antisense strand sequence as shown in Table 1 [20]. The underlined sequences are BamHI and HindIII enzyme sites. The shRNAs were annealed and subcloned into pSilencer4.1-CMVneo (Ambion, Austin, TX, USA) BamHI and HindIII enzyme sites downstream of CMV promoter, and recombinant vector was named as pSilencer4.1-Rlim-siRlim. The recombinant vector was confirmed by the digestion analysis of restriction endonuclease and all Table 1 The primer sequences used in the construction of recombinant vectors. Vectors
Primers
pcDNA3.1-Rlim-Flag
Forward: 5′-CGGGATCCATGGAAAACTCAGATTCC-3′ Reverse: 5′-CGGCTCGAGTTACACAACACTTTCTCTGC-3′ Forward: 5′-CGGGATCCATGGCTTCTTCTGATATCCAGG-3′ Reverse: 5′-CGGCTCGAGTTAGTCAGCTTCAGTCTCG-3′ Forward: 5′-CGCCATGGGTGAAAACTCAGATTCC-3′ Reverse: 5′-CGCTC GAG TTACACAACACTTTCTCTGC-3′ Forward: 5′-CGCCATGGGT GCTTCTTCTGATATCCAGG-3′ Reverse: 5′-CGCTCGAG TTAGTCAGCTTCAGTCTCG-3′ Sense: 5′-GATCCGGTCTCAGACACCAAACAAC ATTCAAGAGATGTTGTTTGGTGTCTGAGACCTAA-3′ Antisense: 5′-AGCTTATGGTCTCAGACACCAAACAACATC TCTTGAATGTTGTTTG GTGTCTGAGACCG-3′
pcDNA3.1–Stathmin–Flag pGSTag–Rlim pGSTag–Stathmin pSilencer4.1-Rlim-siRlim
2.6. Protein purification and GST-pulldown The prokaryotic expression vectors of GST-Stathmin, GST-Rlim and GST were expressed in BL21 (DE3) pLysS Escherichia coli, respectively. The cell lysates were prepared following the manufacturer's protocol (GE Healthcare Bio-Sciences AB, Sweden) and were allowed to bind to Glutathione Sepharose 4B for 1 h at room temperature, washed thrice with Binding Buffer TN1 (140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, pH 7.3), and then eluted with 600 μl of elution buffer TN2 (50 mM Tris–HCl, 10 mM reduced glutathione, pH 8.0). The eluted proteins GST-Stathmin, GST-Rlim and GST were analyzed by SDS-PAGE and stained with Coomassie Blue to check for the protein purity [26]. Then, 50 μg of GST or GST-Stathmin was incubated with 50 μl (50% slurry) of Glutathione HiCap Matrix (Qiagen) in Binding Buffer TN3 (50 mM Tris–HCl, 200 mM NaCl, 1 mM EDTA, 1% NP40, 1 mM DTT, 10 mM MgCl2, pH 8.0 and protease inhibitor PMSF) at room temperature with rocking for preclearing. After 2 h, the mixtures were collected in separate tubes. 500 μg of pcDNA3.1-Rlim-Flag transfected MG-63 cell lysate was added to each tube containing Glutathione HiCap Matrix treated GST or GST-Stathmin, and kept overnight under similar conditions. The beads were washed 3 times with buffer TN1 and eluted with 50 μl of elution buffer TN4 (50 mM Tris–HCl pH 8.0, 400 mM NaCl, 50 mM reduced glutathione, 1 mM EDTA, 1 mM DTT) and then subjected to Western blot analysis with anti-Rlim antibody [27].
1534
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
2.7. Protein identification by mass spectrometry The analysis of proteins by liquid chromatography combined with tandem mass spectrometry (LC/MS/MS) was conducted as previously reported [28], and it was performed by Beijing Proteome Research Center, China. 2.8. Ubiquitin ligase activity assay
for 24 h in 96-well plate and then transfected with pcDNA3.1 vector or pcDNA3.1-Rlim-Flag vector. At different time points (after 12, 24, 48, 72 h of transfection), the medium was replaced with 100 μl of 5% MTT (5 mg/ml), and the plate was incubated for another 4 h at 37 °C. After incubation, the culture medium was removed gently, and 100 μl of DMSO was added. Finally, the absorbance was determined on a microreader (Bio-Rad) at 570 nm. The cell proliferation curve was plotted using the absorbance at each time point and each sample was done in triplicate. Approximately 1.0 × 106 cells were cultured for 24 h in 6-well plates and transfected with pcDNA3.1 vector or pcDNA3.1-Rlim-Flag vector. After 48 h of transfection, the cells were collected, fixed with 70% ethanol and stained with propidiumiodide (PI), and then cell cycle distribution was analyzed on a FACSCalibur system (BD Bioscience, Bedford, MA, USA) by ModFIT software (Verity Software House, Topsham, ME, USA) [31].
For in vivo ubiquitination assay, the transfected MG-63 cells were treated with or without proteasome inhibitor of MG132 at a concentration of 10 μm for 4 h, and then lysed with modified RIPA buffer (20 mM Tris (pH 7.5); 150 mM NaCl, 0.5% Triton X-100, and protease inhibitor PMSF). Identical aliquots of cell lysates (2 mg) were used in immunoprecipitation assays with anti-Stathmin monoclonal antibody. Immunocomplexes were separated on 12% gradient gels and immunoblotting was performed using anti-Ub antibody. An in vitro ubiquitination assay was performed using the Ubiquitinylation kit (Enzo life sciences) following the manufacturer's protocol [29]. In brief, 2 μg of purified GST-Stathmin or GST was used as the substrate, a ubiquitination cocktail was prepared in the presence or absence of either GST-Rlim (100 nM) or ‘F1 and F2’ (containing ubiquitin activating enzyme E1 (100 nM) and ubiquitin carrier protein E2 (2.5 μM) alone or in combination), then incubated at 37 °C for 4 h along with ATP (5 mM) and Ub (2.5 μM), and the total reaction volume was 50 μl. The incubated samples were then checked by anti-Stathmin antibody.
3. Results
2.9. Cell proliferation assay and flow cytometry analysis of cell cycle
3.1. Identification of Rlim as a partner of Stathmin in osteosarcoma cells
Cell proliferation was determined by MTT (Sigma, St. Louis, MO, USA) assay as described elsewhere [30]. 6 × 103 cells were cultured
To identify the natural protein partners of Stathmin in osteosarcoma cells, we expressed a Flag–Stathmin in the human osteosarcoma cell
2.10. Statistical analysis Each experiment was repeated 3 times and all data were indicated as the mean ± standard deviation (±s). Statistical analyses were performed using SPSS version 17.0 (IBM Company, Chicago, IL, USA). Student's t-test and one-way analysis of variance (ANOVA) followed by Dunnett's multiple comparison tests were adopted. Values of P b 0.05 were considered as significant and indicated by asterisks in the figures.
Fig. 1. GST pull down, co-immunoprecipitation assay and endogenous Stathmin level detection results. (A) Identification of Rlim in Stathmin complexes from human osteosarcoma cell line. Silver staining of affinity purified Flag–Stathmin complexes from MG-63 cells. Specific Stathmin-interacting proteins were identified by mass spectrometry and are indicated. (B) GST and GST-Stathmin were used for GST pull down assay in 500 μg of pcDNA3.1-Rlim-Flag transfected MG-63 cell lysates and Western blotting results showed positive bands in all three different primary antibodies of anti-Stathmin, anti-Rlim, and anti-GST. (C) Proteins were immunoprecipitated from MG-63 whole cell lysates with antibodies against Stathmin and checked for Rlim. The Western blotting results showed positive bands in both anti-Stathmin and anti-Rlim antibodies panels. (D) Proteins were immunoprecipitated from MG-63 whole cell lysates with antibodies against Rlim and checked for Stathmin. The Western blotting results showed positive bands in both anti-Stathmin and anti-Rlim antibodies panels. (E) Left figure, MG-63 cells were transfected with plasmids expressing the full-length Rlim or the empty vectors. The level of endogenous Stathmin was examined by Western blotting and the results showed that with the increase of Rlim protein expression, the Stathmin protein level decreased compared to control group. Right figure, densitometry of the immunoblots shown in D (*P b 0.05, **P b 0.01 vs control, n = 3).
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
1535
Fig. 2. Results of co-immunoprecipitation and ubiquitin ligase activity assay in vivo and vitro. (A) MG-63 cells were transfected with Stathmin alone or in combination in the presence or absence of Rlim as shown in the figure, then immunoprecipitation was performed with anti-Stathmin antibody, and antibodies against Ub, Stathmin were used in Western blotting, respectively. The results showed that in the presence of Rlim and Stathmin, the Stathmin ubiquitination level was enhanced compared to the control groups (the third lane from left). (B) MG-63 cells were transfected with Stathmin alone or in combination in the presence or absence of Rlim as shown in the figure, followed by a treatment with MG132, then immunoprecipitation was performed with anti-Stathmin antibody, and antibodies against Ub, Stathmin were used in Western blotting, respectively. The results showed that, in the presence of Rlim and Flag–Stathmin, the Stathmin ubiquitination level was increased compared with control groups (the third lane from left), and the level of poly-ubiquitination of Stathmin was enhanced upon treatment with MG132 (top panel). (C) Left figure, HEK293 cells were transfected with either pSilencer4.1-Rlim-siR or pSilencer4.1-CMVneo (negative control) vector, respectively. After proteasome inhibitor MG132 treatment, the whole cell lysates were detected by Western blotting with anti-Stathmin antibody and the results showed that with the decrease of Rlim protein expression, the Stathmin ubiquitination level decreased accordingly compared to control group. Right figure, densitometry of the immunoblots shown in C (*P b 0.05, **P b 0.01 vs control, n = 3). (D) GST-Rlim, GST-Stathmin and GST were purified using GST column. The purified proteins were then checked by SDS-PAGE and stained with Coomassie on the left figure. GST or GST-Stathmin was incubated for 4 h at 37 °C with or without GST-Rlim, F1 and F2 in the presence of Ub and ATP. The incubated mixtures were then checked by Western blotting with anti-Stathmin antibody and the results showed that in the presence of Rlim, Ub, F1, F2, ATP and GST-Stathmin, poly-ubiquitinated Stathmin appears as a ladder with higher molecular weight forms on the right figure (fourth lane from left).
line MG-63. We analyzed proteins that co-purified with Flag–Stathmin by LC/MS/MS after sequential Flag immunoprecipitation and peptide elution [28] and discovered that Stathmin complexes contain Rlim (Fig. 1A). Rlim (Ring finger LIM domain-binding protein, also known as RNF12), is originally identified as an E3 ubiquitin ligase and capable of targeting cofactors of LIM-HD proteins (CLIM) for proteasomedependent degradation, thereby inhibiting developmental LIM homeodomain activity [17,18]. In our studies, we firstly discovered that Rlim could also interact with Stathmin in osteosarcoma cells. By using purified GST-Stathmin in a pull-down assay, we demonstrated that Flag–Rlim interacted specifically with Stathmin in vitro (Fig. 1B), which confirmed the partner relationship between Rlim and Stathmin. To further reveal if there was an endogenous binding property between Rlim and Stathmin, we performed co-immunoprecipitation assays in MG-63 cells and found that there was an integration band of Rlim and Stathmin proteins (Fig. 1C, D), which demonstrated that Rlim could bind to the Stathmin protein in vivo and suggested that Rlim might play important role in Stathmin post-translational modifications. Based on the above findings, we overexpressed Rlim in MG-63 cells to determine whether Rlim could affect the steady-state levels of Stathmin, and found that overexpression of full-length Rlim could lead to a marked reduction in the steady-state levels of endogenous Stathmin protein (Fig. 1E). Altogether, it could be concluded that Rlim
might be a potential E3 ligase for Stathmin and played important roles in Stathmin post-translational regulation. 3.2. Rlim induces ubiquitination of Stathmin Since Rlim interacted with Stathmin, it was a matter of interest to check whether Rlim played a ubiquitin ligase role for Stathmin protein degradation. To confirm this possibility, we performed ubiquitination assay with Rlim as an E3 ubiquitin ligase. Under in-vivo conditions, we overexpressed Stathmin together with or without Rlim in MG-63 cells. Whole cell lysates prepared from 48 h post-transfection cells were immunoprecipitated using anti-Stathmin antibody. Western blotting results showed that the presence of poly-ubiquitinated Stathmin adducts and the level of poly-ubiquitination of Stathmin was markedly increased in the presence of Rlim compared with control groups, while this phenomenon was enhanced upon treatment with MG132, a proteasome inhibitor, which suggested that Stathmin could be polyubiquitinated in part by Rlim (E3 ligase) and degraded by the Ubproteasome system. (Fig. 2A, B). To confirm the role of Rlim in ubiquitinating Stathmin, we firstly tried to find out whether this function of Rlim was existent under endogenous condition. Accordingly, Rlim was knocked down in HEK293 cells by siRNA and the lysates, with the proteasome inhibitor MG132, were subjected to
1536
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
Fig. 3. Stathmin protein level and half-life detection results. (A) Left figure, HEK293 cells were transfected with pSilencer4.1-Rlim-siR or pSilencer4.1-CMVneo (negative control) vectors, respectively. Western blotting was done using anti-Stathmin antibody and the results showed that with the decrease of Rlim protein expression, the Stathmin protein level increased compared to control groups. Right figure, densitometry of the immunoblots shown in A (*P b 0.05, **P b 0.01 vs control, n = 3). (B) MG-63 cells were transfected with plasmids expressing the full-length Rlim or the empty vectors, respectively followed by a treatment with 50 μM cycloheximide (CHX) for a time period as indicated in the figure; the lysates were then checked for Stathmin level to determine the protein half-life. The half life of Stathmin protein was about 10 h in natural condition, but it was reduced to ~6 h in the condition of Rlim overexpression. (C) HEK293 cells were transfected with pSilencer4.1-Rlim-siR or pSilencer4.1-CMVneo vectors, respectively, followed by a treatment with 50 μM cycloheximide (CHX) for a time period as indicated in the figure, the lysates were then checked for Stathmin level. The results showed that the Stathmin protein level did not change obviously even in the condition of decreased Rlim protein expression compared to control group. (D) MG-63 cells were transfected with plasmids expressing the full-length Rlim or the empty vectors, and then treated with or without MG132. The cell lysates were then checked for Stathmin protein level. The results showed that, with the MG132 treatment, the reduction of Stathmin protein level was ablated even in the condition of increased Rlim expression compared to MG132 untreated cells (top panel).
immunoprecipitation with anti-Stathmin antibody. Western blotting analysis revealed that depletion of endogenous Rlim significantly reduced the Stathmin ubiquitination level (Fig. 2C). Then, we performed in vitro ubiquitination assay to rule out the involvement of any other E3 ligase(s) in the ubiquitination of Stathmin as well as to establish Rlim as a novel E3 ligase for Stathmin. GST-Stathmin, GST-Rlim and GST were purified using Glutathione Sepharose 4B. With either GST-Stathmin or GST as the substrate, a ubiquitination cocktail was prepared in the presence or absence of either GST-Rlim or ‘F1 and F2’ (containing ubiquitin activating enzyme E1 and ubiquitin carrier protein E2) alone or in combination, incubated at 37 °C for 4 h along with ATP and Ub. The incubated samples were then checked by anti-Stathmin antibody. The results quite clearly highlighted the capability of Rlim to independently ubiquitinate Stathmin (Fig. 2D). Collectively, these results indicated that Rlim could ubiquitinate Stathmin under both in vivo and in vitro conditions.
3.3. Rlim promotes proteasome mediated degradation of Stathmin With the findings that Stathmin was ubiquitinated by Rlim, we examined the Stathmin protein level under the influence of Rlim. Our results demonstrated that forced overexpression of Rlim in MG-63 cells decreased protein level of Stathmin (Fig. 1D), conversely, knock down of Rlim by siRNA in HEK293 cells produced the anticipated antagonistic effect (Fig. 3A). It was established that Rlim negatively regulated the steady state level of Stathmin protein. To further investigate this function, we observed endogenous level of Stathmin protein under the condition of ectopic overexpression of Rlim in MG-63 cells or knocked down Rlim by siRNA in HEK293 cells, followed by treatment with cycloheximide (protein synthesis inhibitor). The results showed that Stathmin protein half life was either decreased in the condition of overexpressed Rlim or increased in the condition of knock down of Rlim, respectively (Fig. 3B,
Fig. 4. The results of cell proliferation analysis. Human osteosarcoma cells (MG-63 and Saos-2) were transfected with plasmids expressing the full-length Rlim or the empty vectors (4 μg). Cell viability was evaluated by MTT assay for a time period as indicated in the figure. Compared with the control cells, the highest inhibitory rate of cell proliferation was 27.28 ± 3.21% and 32.72 ± 3.63%, respectively in MG-63 cells and Saos-2 cells on 72 h (P b 0.05).
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
1537
C). The half life of Stathmin protein was about 10 h in natural condition, but it was reduced to ~6 h in the condition of overexpression of Rlim. To verify the involvement of the proteasome system in Rlim mediated degradation of Stathmin protein, we treated MG-63 cells with MG132, which stabilized Stathmin and ablated the reduction of Stathmin protein level even in the presence of Rlim (Fig. 3D). All these results confirmed the involvement of the proteasome system in Stathmin degradation by Rlim.
percentage of G2/M phase increased by 11.01 ± 1.7% and 11.55 ± 1.9% respectively in Rlim-overexpression MG-63 and Saos-2 cells (P b 0.05) (Fig. 5). These results demonstrated that Rlim–Stathmin ubiquitination pathway was effective in reversing the malignant phenotype of osteosarcoma cells caused by overexpression of Stathmin.
3.4. The Rlim–Stathmin ubiquitination pathway controls cell proliferation and cell cycle
Stathmin, a kind of small molecular weight cytosolic phosphoprotein, regulates microtubule dynamics and influences cell cycle through integrating and transducing various transduction signals within cells. Expression studies have shown that Stathmin is aberrantly overexpressed in the majority of human malignancies, and plays an important role in maintaining malignant phenotype of tumors [33]. Stathmin expression and its function are controlled by multiple steps and different regulating levels from transcription, phosphorylation and ubiquitination. The regulation of Stathmin expression and function is very complicated, and the mechanisms for its overexpression in the malignant tumors have not been fully determined [34–37]. Historically, the focus has been on the transcriptional regulation of Stathmin overexpression, while in this study, we conducted researches on the posttranslational modifications of Stathmin. We, for the first time, verified that Rlim was a novel Stathmin binding protein, and further found that Rlim regulated Stathmin degradation in a ubiquitin-dependent manner, which might explain, in part, the reason of high level Stathmin protein in most of the malignant tumors cells. We firstly examined the biological role of Rlim, a ring-domain E3 ubiquitin ligase, and demonstrated the existence of a Rlim–Stathmin
To analyze phenotypic changes caused by Rlim–Stathmin ubiquitination pathway, we investigated the effects of ectopic overexpression of Rlim on cell proliferation in osteosarcoma cells. Cell proliferation was evaluated by MTT assay and the results showed a significant inhibition of the proliferation both in MG-63 cells and Saos-2 cells, and the highest inhibitory rate was 27.28 ± 3.21% and 32.72 ± 3.63% compared with control vector transfected cells, respectively on 72 h after transfection (P b 0.05) (Fig. 4). It has been reported that the proliferation inhibition of tumor cells by knockdown of Stathmin expression is caused mainly by cell cycle arrest and microtubule assembly in other types of mammalian cells [9,32]. To reveal the mechanisms underlying ubiquitination-mediated proliferation inhibition, we used flow cytometric analysis to detect cell cycle changes of MG-63 and Saos-2 cells that were transient transfected with 4 μg plasmids expressing the fulllength Rlim, respectively. The population of G2/M phase was significantly increased in both MG-63 and Saos-2 cells 72 h later after plasmid transfection compared with control cells. As shown in Fig. 5, the
4. Discussion
Fig. 5. The results of cell cycle analysis. Cells were harvested 72 h after transfection of plasmids expressing the full-length Rlim or the empty vectors (4 μg) and subsequently analyzed for their DNA content by flow cytometry. Compared with control cells, the percentage of G2/M phase increased by 11.01 ± 1.7% and 11.55 ± 1.9%, respectively in MG-63 and Saos-2 cells in the condition of Rlim-overexpression (P b 0.05).
1538
X. Chen et al. / Cellular Signalling 26 (2014) 1532–1538
pathway, in which Rlim operated on a linear pathway epistatic to Stathmin. Rlim interacted with Stathmin and promoted its ubiquitination in vitro and in vivo, thus acting as a negative regulator of Stathmin. Whereas, overexpression of Rlim reduced the half-life of endogenous Stathmin protein from ~ 10 h to ~ 6 h, and depletion of Rlim resulted in an increase of endogenous Stathmin protein level. These results supported the hypothesis that Rlim played a critical role in regulating the level of Stathmin protein within the cells. It was reported that Stathmin played important roles in maintenance of malignant phenotypes in human malignancies and the inhibition of Stathmin gene expression could lead to proliferation and cell-cycle arrest in tumor cells [9,11,37]. Our results showed that the interaction of Rlim with Stathmin resulted in converting of tumor cell malignant phenotypes as this interaction could down regulate Stathmin protein level, and subsequently lead to inhibition of cell proliferation and accumulation of G2/M phase of osteosarcoma MG-63 and Saos-2 cells (Figs. 4 and 5). These findings suggested that Rlim might be a potential therapeutic target for the treatment of osteosarcomas. We have demonstrated that Stathmin is subject to regulation via the ubiquitin proteasome through interaction with the E3-ubiquitin ligase Rlim. The importance of this type of interaction between regulatory proteins and E3-ligases has recently been emphasized in a number of reports [38,39]. For example, E3-ligase Huwe1 interacts with N-Myc and mediates its degradation, this interaction is deemed essential for it restrains proliferation and enables neuronal differentiation [39]. Furthermore, the ubiquitin-mediated degradation of PARP-1 plays an important role in cell cycle regulation and cancer therapy [38]. As such, we propose that Rlim may represent a new pathway for the regulation of cell cycle and cell proliferation in osteosarcoma cells by acting as a negative regulator of Stathmin protein level. In conclusion, our experiment results demonstrated that Rlim could increase Stathmin turnover by targeting it for degradation through the proteasome in a ubiquitin-dependent manner, which led to proliferation and cell-cycle arrest and malignant phenotype changes in osteosarcoma cells. We believe that the interaction of Rlim and Stathmin will be a promising strategy in targeted tumor gene therapy and may have potential for clinical use in tumor treatment. Acknowledgment Thanks to everyone from the Department of Clinical Laboratory for their sincere help and excellent technical assistance. This study was supported by a grant from the National Natural Science Foundation of China (No. 81001195). References [1] P. Niethammer, P. Bastiaens, E. Karsenti, Science 303 (2004) 1862–1866. [2] C.I. Rubin, G.F. Atweh, J. Cell. Biochem. 93 (2004) 242–250. [3] L. Cassimeris, Cell Biol. 14 (2002) 18–24.
[4] G. Chen, H. Wang, T.G. Gharib, C.C. Huang, D.G. Thomas, K.A. Shedden, R. Kuick, J.M. Taylor, S.L. Kardia, D.E. Misek, T.J. Giordano, M.D. Iannettoni, M.B. Orringer, S.M. Hanash, D.G. Beer, Mol. Cell. Proteomics 2 (2003) 107–116. [5] X. Wang, J.H. Ren, F. Lin, J.X. Wei, M. Long, L. Yan, H.Z. Zhang, Cancer Biol. Ther. 10 (2010) 632–643. [6] S. Rana, P.B. Maples, N. Senzer, J. Nemunaitis, Expert. Rev. Anticancer. Ther. 8 (2008) 1461–1470. [7] S.J. Mistry, G.F. Atweh, Mt Sinai J. Med. 69 (2002) 299–304. [8] H.Z. Zhang, P. Gao, L. Yan, F. Lin, Ai Zheng 23 (2004) 493–496. [9] H.Z. Zhang, Y. Wang, P. Gao, F. Lin, L. Liu, B. Yu, J.H. Ren, H. Zhao, R. Wang, Cancer Biol. Ther. 5 (2006) 1457–1461. [10] F. Lin, P. Gao, L. Liu, R. Wang, H.Z. Zhang, J. Mod. Oncol. 15 (2007) 485–487. [11] R. Wang, K. Dong, F. Lin, X. Wang, P. Gao, S.H. Wei, S.Y. Cheng, H.Z. Zhang, Mol. Med. 13 (2007) 567–575. [12] B. Belletti, G. Baldassarre, Expert Opin. Ther. Targets 15 (2011) 1249–1266. [13] M.O. Steinmetz, J. Struct. Biol. 158 (2007) 137–147. [14] L. Lee, J.S. Tirnauer, J. Li, S.C. Schuyler, J.Y. Liu, D. Pellman, Science 287 (2000) 2260–2262. [15] I. Jonkers, T.S. Barakat, E.M. Achame, K. Monkhorst, A. Kenter, E. Rentmeester, F. Grosveld, J.A. Grootegoed, J. Gribnau, Cell 139 (2009) 999–1011. [16] H.P. Ostendorff, M. Bossenz, A. Mincheva, N.G. Copeland, D.J. Gilbert, N.A. Jenkins, P. Lichter, I. Bach, Genomics 69 (2000) 120–130. [17] I. Bach, E.C. Rodriguez, C. Carrière, A. Bhushan, A. Krones, D.W. Rose, C.K. Glass, B. Andersen, J.C. Izpisúa Belmonte, M.G. Rosenfeld, Nat. Genet. 22 (1999) 394–399. [18] H.P. Ostendorff, R.I. Peirano, M.A. Peters, A. Schlüter, M. Bossenz, M. Scheffner, I. Bach, Nature 416 (2002) 99–103. [19] C. Güngör, N.I. Taniguchi, H. Ma, A. Drung, B. Tursun, H.P. Ostendorff, M. Bossenz, C. G. Becker, T. Becke, I. Bach, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 15000–15005. [20] Y.R. Her, I.K. Chung, J. Biol. Chem. 284 (2009) 8557–8566. [21] I. Bach, Cell Res. 22 (2012) 1524–1526. [22] R.R. Singh, C.J. Barnes, A.H. Talukder, S.A. Fuqua, R. Kumar, Cancer Res. 65 (2005) 10594–10601. [23] S.A. Johnsen, C. Güngör, T. Prenzel, S. Riethdorf, L. Riethdorf, N.I. Taniguchi, T. Rau, B. Tursun, J.D. Furlow, G. Sauter, M. Scheffne, K. Pantel, F. Gannon, I. Bach, Cancer Res. 69 (2009) 128–136. [24] R. Nikolay, T. Wiederkehr, W. Rist, G. Kramer, M.P. Mayer, B. Bukau, J. Biol. Chem. 279 (2004) 2673–2678. [25] L. Kashima, M. Toyota, H. Mita, H. Suzuki, M. Idogawa, K. Ogi, Y. Sasaki, T. Tokino, Oncogene 28 (2009) 2643–2653. [26] X. Chen, K. Dong, M. Long, F. Lin, X. Wan, J.X. Wei, J.H. Ren, H.Z. Zhang, Oncol. Lett. 2 (2012) 319–323. [27] S.F. Ahmed, S. Deb, I. Paul, A. Chatterjee, T. Mandal, U. Chatterjee, M.K. Ghosh, J. Biol. Chem. 287 (2012) 15996–16006. [28] M. Idogawa, M. Masutani, M. Shitashige, K. Honda, T. Tokino, Y. Shinomura, K. Imai, S. Hirohashi, T. Yamada, Cancer Res. 67 (2007) 911–918. [29] X. Li, D.F. Lu, F.J. He, H.B. Zhou, Q. Liu, Y. Wang, C.S. Shao, Y.Q. Gong, J. Biol. Chem. 286 (2011) 32344–32354. [30] Ch.D. Fan, J. Zhao, B.X. Zhao, Sh.l. Zhang, J.y. Miao, Chem. Res. Toxicol. 22 (2009) 1517–1525. [31] K. Ogi, M. Toyota, H. Mita, A. Satoh, L. Kashima, Y. Sasaki, H. Suzuki, K. Akino, N. Nishikawa, M. Noguchi, Y. Shinomura, K. Imai, H. Hiratsuka, T. Tokino, Cancer Biol. Ther. 4 (2005) 773–780. [32] P. Gao, F. Lin, J. Modern Oncol. 2 (2006) 210–212. [33] S. Rana, P.B. Maples, N. Senzer, J. Nemunaitis, Expert. Rev. Anticancer 8 (2008) 1461–1470. [34] R.G. Polzin, H. Benlhabib, J. Trepel, J.E. Herrera, Gene 341 (2004) 209–218. [35] S. Polager, D. Ginsberg, J. Biol. Chem. 278 (2003) 14431449. [36] H. Benlhabib, J.E. Herrera, Gene 377 (2006) 177–185. [37] F. Lin, M. Long, L. Yan, P. Gao, R. Wang, Y. Zhang, X. Wang, T. He, L. Liu, K. Dong, J.H. Reng, H.Zh. Zhang, Cell. Signal. 22 (2010) 166–173. [38] L. Kashima, M. Idogawa, H. Mita, M. Shitashige, T. Yamada, K. Ogi, H. Suzuki, M. Toyota, H. Ariga, Y. Sasaki, T. Tokino, J. Biol. Chem. 287 (2012) 12975–12984. [39] X. Zhao, J.I. Heng, D. Guardavaccaro, R. Jiang, M. Pagano, F. Guillemot, A. Iavarone, A. Lasorella, Nat. Cell Biol. 10 (2008) 643–653.
