Induction of zinc finger protein RNF6 auto-ubiquitination for the treatment of myeloma and chronic myeloid leukemia
2022; Elsevier BV; Volume: 298; Issue: 9 Linguagem: Inglês
10.1016/j.jbc.2022.102314
ISSN1083-351X
AutoresHaixia Zhuang, Ying Ren, Chenyu Mao, Yueya Zhong, Zubin Zhang, Biyin Cao, Yuming Zhang, Jinqi Huang, Guoqiang Xu, Zhenqian Huang, Yujia Xu, Xinliang Mao,
Tópico(s)Chronic Lymphocytic Leukemia Research
ResumoThe zinc finger ubiquitin ligase RNF6 has been proposed as a potential therapeutic target in several cancers, but understanding its molecular mechanism of degradation has been elusive. In the present study, we find that RNF6 is degraded via auto-ubiquitination in a manner dependent on its Really Interesting New Gene (RING) domain. We determine that when the RING domain is deleted (ΔRING) or the core cysteine residues in the zinc finger are mutated (C632S/C635S), the WT protein, but not the ΔRING or mutant RNF6 protein, undergoes polyubiquitination. We also identify USP7 as a deubiquitinase of RNF6 by tandem mass spectrometry. We show that USP7 interacts with RNF6 and abolishes its K48-linked polyubiquitination, thereby preventing its degradation. In contrast, we found a USP7-specific inhibitor promotes RNF6 polyubiquitination, degradation, and cell death. Furthermore, we demonstrate the anti-leukemic drug Nilotinib and anti-myeloma drug Panobinostat (LBH589) induce RNF6 K48-linked polyubiquitination and degradation in both multiple myeloma (MM) and leukemia cells. In agreement with our hypothesis on the mode of RNF6 degradation, we show these drugs promote RNF6 auto-ubiquitination in an in vitro ubiquitination system without other E3 ligases. Consistently, reexpression of RNF6 ablates drug-induced MM and leukemia cell apoptosis. Therefore, our results reveal that RNF6 is a RING E3 ligase that undergoes auto-ubiquitination, which could be abolished by USP7 and induced by anti-cancer drugs. We propose that chemical induction of RNF6 auto-ubiquitination and degradation could be a novel strategy for the treatment of hematological malignancies including MM and leukemia. The zinc finger ubiquitin ligase RNF6 has been proposed as a potential therapeutic target in several cancers, but understanding its molecular mechanism of degradation has been elusive. In the present study, we find that RNF6 is degraded via auto-ubiquitination in a manner dependent on its Really Interesting New Gene (RING) domain. We determine that when the RING domain is deleted (ΔRING) or the core cysteine residues in the zinc finger are mutated (C632S/C635S), the WT protein, but not the ΔRING or mutant RNF6 protein, undergoes polyubiquitination. We also identify USP7 as a deubiquitinase of RNF6 by tandem mass spectrometry. We show that USP7 interacts with RNF6 and abolishes its K48-linked polyubiquitination, thereby preventing its degradation. In contrast, we found a USP7-specific inhibitor promotes RNF6 polyubiquitination, degradation, and cell death. Furthermore, we demonstrate the anti-leukemic drug Nilotinib and anti-myeloma drug Panobinostat (LBH589) induce RNF6 K48-linked polyubiquitination and degradation in both multiple myeloma (MM) and leukemia cells. In agreement with our hypothesis on the mode of RNF6 degradation, we show these drugs promote RNF6 auto-ubiquitination in an in vitro ubiquitination system without other E3 ligases. Consistently, reexpression of RNF6 ablates drug-induced MM and leukemia cell apoptosis. Therefore, our results reveal that RNF6 is a RING E3 ligase that undergoes auto-ubiquitination, which could be abolished by USP7 and induced by anti-cancer drugs. We propose that chemical induction of RNF6 auto-ubiquitination and degradation could be a novel strategy for the treatment of hematological malignancies including MM and leukemia. The zinc finger protein RNF6 is an oncogenic ubiquitin ligase that has been overexpressed in various cancers including multiple myeloma (MM) cells (1Ren Y. Xu X. Mao C.Y. Han K.K. Xu Y.J. Cao B.Y. et al.Rnf6 promotes myeloma cell proliferation and survival by inducing glucocorticoid receptor polyubiquitination.Acta Pharmacol. Sin. 2020; 41: 394-403Crossref PubMed Scopus (21) Google Scholar) and leukemia cells (2Xu X. Han K. Tang X. Zeng Y. Lin X. Zhao Y. et al.The ring finger protein rnf6 induces leukemia cell proliferation as a direct target of pre-B-cell leukemia homeobox 1.J. Biol. Chem. 2016; 291: 9617-9628Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar). In MM cells, RNF6 as a ubiquitin ligase binds to and mediates K63-linked polyubiquitination toward the glucocorticoid receptor and promotes its oncogenic transcriptional activity, therefore contributing to MM cell survival and drug resistance (1Ren Y. Xu X. Mao C.Y. Han K.K. Xu Y.J. Cao B.Y. et al.Rnf6 promotes myeloma cell proliferation and survival by inducing glucocorticoid receptor polyubiquitination.Acta Pharmacol. Sin. 2020; 41: 394-403Crossref PubMed Scopus (21) Google Scholar). In leukemia cells, RNF6 is upregulated by the Pre-B-Cell Leukemia Transcription Factor 1 (PBX1) and promotes leukemia progression (2Xu X. Han K. Tang X. Zeng Y. Lin X. Zhao Y. et al.The ring finger protein rnf6 induces leukemia cell proliferation as a direct target of pre-B-cell leukemia homeobox 1.J. Biol. Chem. 2016; 291: 9617-9628Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar). Knockdown of RNF6 leads to shrinkage of human leukemia xenografts in mice. Moreover, downregulation of RNF6 by the natural product saponins from Paris forrestii induces leukemia cell apoptosis (3Lu Q. He Y. Wang Y. Gao L. Zheng Y. Zhang Z. et al.Saponins from Paris forrestii (Takht.) H. Li display potent activity against acute myeloid leukemia by suppressing the RNF6/AKT/mTOR signaling pathway.Front. Pharmacol. 2018; 9: 673Crossref PubMed Scopus (15) Google Scholar). All these studies suggest RNF6 could be a potential therapeutic target for both MM and leukemia. RNF6 belongs to the C2H2-type zinc finger ubiquitin ligases containing a Really Interesting New Gene (RING) motif and this class of ubiquitin ligases can mediate themselves ubiquitination or auto-ubiquitination (4Canning M. Boutell C. Parkinson J. Everett R.D. A RING finger ubiquitin ligase is protected from autocatalyzed ubiquitination and degradation by binding to ubiquitin-specific protease USP7.J. Biol. Chem. 2004; 279: 38160-38168Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar). A number of such ubiquitin ligases have been well elucidated including XIAP (5Yang Y. Fang S. Jensen J.P. Weissman A.M. Ashwell J.D. Ubiquitin protein ligase activity of IAPs and their degradation in proteasomes in response to apoptotic stimuli.Science. 2000; 288: 874-877Crossref PubMed Scopus (875) Google Scholar), TRAF6 (6Park Y. Pang K. Park J. Hong E. Lee J. Ooshima A. et al.Destablilization of TRAF6 by DRAK1 suppresses tumor growth and metastasis in cervical cancer cells.Cancer Res. 2020; 80: 2537-2549Crossref PubMed Scopus (14) Google Scholar), TRIM26 (7Ran Y. Zhang J. Liu L.L. Pan Z.Y. Nie Y. Zhang H.Y. et al.Autoubiquitination of TRIM26 links TBK1 to NEMO in RLR-mediated innate antiviral immune response.J. Mol. Cell Biol. 2016; 8: 31-43Crossref PubMed Scopus (58) Google Scholar), and RNF115 (7Ran Y. Zhang J. Liu L.L. Pan Z.Y. Nie Y. Zhang H.Y. et al.Autoubiquitination of TRIM26 links TBK1 to NEMO in RLR-mediated innate antiviral immune response.J. Mol. Cell Biol. 2016; 8: 31-43Crossref PubMed Scopus (58) Google Scholar). Similar to general ubiquitination, auto-ubiquitination can be classified as mono-ubiquitination or various polyubiquitination types, therefore leading to self-degradation or functional modification. It is reported that mono-ubiquitination of TRIM26 by itself occurs upon viral infection, thus leading to the activation of TBK1 innate antiviral immune response (7Ran Y. Zhang J. Liu L.L. Pan Z.Y. Nie Y. Zhang H.Y. et al.Autoubiquitination of TRIM26 links TBK1 to NEMO in RLR-mediated innate antiviral immune response.J. Mol. Cell Biol. 2016; 8: 31-43Crossref PubMed Scopus (58) Google Scholar). In contrast, XIAP undergoes degradation upon self-ubiquitination induced by its antagonist, therefore leading to the activation of Caspase-3 and cancer cell apoptosis (5Yang Y. Fang S. Jensen J.P. Weissman A.M. Ashwell J.D. Ubiquitin protein ligase activity of IAPs and their degradation in proteasomes in response to apoptotic stimuli.Science. 2000; 288: 874-877Crossref PubMed Scopus (875) Google Scholar). Given that auto-ubiquitination is a characteristic of RING E3 ligases, whether RNF6 could undergo auto-ubiquitination and how this auto-ubiquitination is modulated remains elusive. It will be of interest to know whether induction of RNF6 auto-ubiquitination benefits the treatment of hematological malignancies. In the present study, we found that RNF6 is degraded via the proteasomes upon self-directed polyubiquitination. Moreover, we identified that the ubiquitin-specific protease USP7 stabilizes RNF6 by preventing its auto-ubiquitination. Inhibiting USP7 leads to RNF6 degradation and cell apoptosis. Furthermore, we found that anti-cancer drugs can trigger RNF6 with K48-linked auto-ubiquitination and proteasomal degradation. This study provides an in-depth understanding on the modulation of RNF6 auto-ubiquitination and initiated the concept that chemical induction of auto-ubiquitination represents a novel strategy for the treatment of leukemia and myeloma. Given that RNF6 is a member of the RING family E3 ligases featured with auto-ubiquitination and degradation (4Canning M. Boutell C. Parkinson J. Everett R.D. A RING finger ubiquitin ligase is protected from autocatalyzed ubiquitination and degradation by binding to ubiquitin-specific protease USP7.J. Biol. Chem. 2004; 279: 38160-38168Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar), we first examined whether RNF6 could be degraded via the proteasomes. To this end, we first evaluated RNF6 stability in HEK293T cells that were treated with proteasomal inhibitors or lysosomal inhibitors for 12 h, and the subsequent assays demonstrated that RNF6 protein was markedly increased by proteasome inhibitors (MG132 and bortezomib) but not by a typical lysosomal inhibitor chloroquine (Fig. 1A). Moreover, the RNF6 protein stability was increased by proteasomal inhibition in a concentration- and time-dependent manner (Fig. 1, B and C). Furthermore, MG132 also strikingly accumulated RNF6 polyubiquitination in a concentration-dependent manner (Fig. 1D), suggesting RNF6 was modified by polyubiquitination and it was processed in proteasomes. Given that RNF6 is a RING domain E3 ligase, we wondered whether RNF6 polyubiquitination is self-directed. To this end, we generated a ΔRNF6 construct lacking the RING domain that determines E3 ligase activity and evaluated its ubiquitination level. We found that in the presence of intact RNF6, ΔRNF6 could be polyubiquitinated (Fig. 1E), but no polyubiquitination was observed on ΔRNF6 in the absence of intact RNF6, suggesting RNF6 might direct the polyubiquitination of ΔRNF6; in other words, RNF6 might undergo auto-ubiquitination. To confirm this hypothesis, we further constructed a RNF6 double mutant with C632S/C635S (RNF6C/S) to disrupt the zinc finger scaffold, thereby inactivating RNF6 E3 ligase activity. We purified wtRNF6, ΔRNF6, or RNF6C/S and further measured their ubiquitination levels in a cell-free ubiquitination system in the absence of any other E3 ligases. The result showed that wtRNF6 but not ΔRNF6 or RNF6C/S was polyubiquitinated (Fig. 1F), further confirming the conclusion that RNF6 undergoes auto-ubiquitination. Subsequently, we measured the protein stability of RNF6 and its mutants in the presence of MG132 for 24 h and found that MG132 stabilized wtRNF6 but not its RING-deleting counterparts (Fig. 1, G and H). Moreover, when cycloheximide (CHX), an inhibitor of protein synthesis de novo, was added, wtRNF6 was almost completely degraded within 6 h, but ΔRNF6 remained unchanged within 24 h (Fig. 1, I and J), further suggesting RNF6 stability relied on its RING domain–related E3 ligase activity. However, the C/S mutant was found to have a slight increase by MG132 in 24 h (Fig. 1, G and H), which was confirmed in the CHX chase assay (Fig. 1, I and J). To be noted, C/S mutant was not markedly altered in 12 h (Fig. 1, I and J), suggesting that the C/S mutant might still undergo degradation in proteasomes with a yet-to-know mechanism. All the above results therefore collectively demonstrated that RNF6 undergoes self-directed polyubiquitination and degradation in proteasomes. The above study clearly demonstrated that RNF6 undergoes auto-ubiquitination. Given that protein ubiquitination is a dynamic and reversible process, we wondered whether there is a deubiquitinase that modulates RNF6 auto-ubiquitination. To this end, we performed a HPLC-coupled tandem mass spectrometry assay against ΔRNF6-interactomes (given ΔRNF6 is more stable), from which two deubiquitinases USP7 and USP9x were identified with more than two unique peptides in the RNF6-interacting proteome from the ΔRNG but not in the empty-vector–transfected cells (Fig. 2A). To confirm the interaction between RNF6 and the Dubs, we overexpressed Flag-RNF6 or Flag-USP7 in HEK293T cells, respectively, followed by immunoprecipitation (IP) with a specific antibody against Flag, the subsequent immunoblot (IB) showed that both USP7 and USP9x proteins were found in the RNF6 immunoprecipitates (Fig. 2B) and both RNF6 and USP9x were present in the USP7 immunoprecipitates (Fig. 2C). Moreover, this interaction was recapitulated in multiple MM cell lines (Fig. 2, D and E). Lastly, we examined the detailed interaction between USP7 and RNF6 and found that the N-terminal TRAF or the C-terminal UBL domains were critical for USP7 to interact with RNF6 (Fig. 2F). We also found that RNF6 interacted with USP7 via its undefined domain (aa. 87–482) (Fig. 2G). Given that both USP7 and USP9x interact with RNF6, we wondered whether these two Dubs deubiquitinated RNF6. To this end, RNF6 was cotransfected with USP7 or USP9x into HEK293T cells, followed by IP/IB assays. The results revealed that USP7 but not USP9x markedly reduced the polyubiquitination level of RNF6 (Fig. 3A). In contrast, USP9x increased the polyubiquitination levels of RNF6 (Fig. 3A). This finding suggested that USP7 but not USP9x was a potential Dub of RNF6. We thus subsequently evaluated the effects of USP7 on RNF6 ubiquitination. As shown in Figure 3B, USP7 downregulated the ubiquitination level of RNF6 in a concentration-dependent manner. A previous study has shown that a specific domain in USP7 might be sufficient to stabilize its substrate protein (8Ma J. Martin J.D. Xue Y. Lor L.A. Kennedy-Wilson K.M. Sinnamon R.H. et al.C-terminal region of USP7/HAUSP is critical for deubiquitination activity and contains a second mdm2/p53 binding site.Arch. Biochem. Biophys. 2010; 503: 207-212Crossref PubMed Scopus (56) Google Scholar). To find out which domain was critical for USP7 to deubiquitinate RNF6, we structured a series of USP7 truncates (Fig. 2F) and these truncates were subjected to co-transfection into HEK293T cells with RNF6. The resultant IP/IB assay demonstrated that the catalytic domain was not sufficient to deubiquitinate RNF6 (Fig. 3C); it must act together with either the TRAF or the UBL domain (Fig. 3C). Interestingly, the truncate with the TRAF domain remained equal deubiquitinase activity comparable to the WT form (Fig. 3C). It is known that there are several types of ubiquitination upon the lysine residue in ubiquitin; to find out which type of ubiquitination could be prevented by USP7, we cotransfected USP7, RNF6, and individual Ub plasmids with single lysine residues into HEK293T cells, and the subsequent IP/IB assay revealed that RNF6 was processed with high polyubiquitination in the presence of Ub. Notably, USP7 abolished RNF6 from K48- but not other Ub-linked polyubiquitination (Fig. 3D), indicating that RNF6 might mainly undergo K48-linked auto-ubiquitination and USP7 deubiquitinates RNF6 for its K48-linked ubiquitination. To confirm this hypothesis, we next examined the endogenous K48-linked polyubiquitination of RNF6 in both the MM cell line RPMI-8226 and chronic myelogenous leukemia CML cell line K562. The specific IP/IB assays indicated that USP7 downregulated RNF6 for its K48-linked polyubiquitination in both cell lines in a concentration-dependent manner (Fig. 3E). In accordance with this finding, when USP7 was knocked out by its specific sgRNA, RNF6 was found to be modified with increased K48-linked polyubiquitination (Fig. 3F). Therefore, these results collectively demonstrated that RNF6 might undergo K48-linked auto-ubiquitination that could be abolished by the Dub USP7. The above studies have demonstrated that USP7 but not USP9x decreases RNF6 for its K48-linked polyubiquitination that is necessary for protein degradation in proteasomes; therefore, we wondered whether USP7 prevents RNF6 from degradation via the ubiquitin-proteasomal pathway. To this end, Flag-USP7 was transfected into cells followed by evaluating RNF6 at the protein and mRNA levels. The results showed that USP7 failed to alter RNF6 mRNA but upregulated its protein in a concentration- and time-dependent manner (Fig. 4, A and B). These findings were consistent with the previously mentioned study that USP7 prevented RNF6 from K48-linked polyubiquitination in association with protein stability (Fig. 3). The effects of USP7 on RNF6 protein were further recapitulated in MM and leukemia cell lines (Fig. 4C). When USP7 was introduced into RPMI-8226, LP1, and K562 cells, RNF6 protein was increased (Fig. 4C); in contrast, when USP7 was knocked out by its specific sgRNA, RNF6 protein was markedly reduced (Fig. 4D). To further find out whether USP7 stabilizes RNF6, we examined the effect of USP7 on the half-life of RNF6 in the presence of CHX. As shown in Figure 4E, RNF6 was almost completely degraded within 6 h, in a manner as shown in Figure 1I; however, the introduction of USP7 markedly extended its degradation time. Given that RNF6 undergoes auto-ubiquitination, we next examined the effects of USP7 on the metabolism of wtRNF6 as well as its ΔRING and C/S mutants. The results showed that USP7 stabilized the wtRNF6 protein but showed no activity to increase the protein stability of the ΔRING and C/S mutants (Fig. 4F), further suggesting that USP7 antagonizes RNF6 auto-ubiquitination and subsequent degradation. To further characterize the correlation between USP7 and RNF6 protein levels in MM and leukemia cells, a panel of MM and leukemia cell lines as well as primary MM bone marrow cells were subjected to IB to measure USP7 and RNF6 proteins. The results showed that the RNF6 protein level was highly correlated to USP7 levels in both types of cell lines (Fig. 4G) and primary cells (Fig. 4H). Therefore, there results demonstrated that USP7 stabilizes RNF6 protein. The previous studies have shown that RNF6 stability could be modulated by its auto-ubiquitination and USP7 acts as a Dub of RNF6 self-ubiquitination. These findings suggest that inhibition of USP7 might induce RNF6 degradation via the ubiquitin-proteasomal pathway. To validate this hypothesis, MM and leukemia cell lines, including RPMI-8226, LP1, and K562, were treated with P5091, a small chemical molecular inhibitor of USP7 (9Chauhan D. Tian Z. Nicholson B. Kumar K.G. Zhou B. Carrasco R. et al.A small molecule inhibitor of ubiquitin-specific protease-7 induces apoptosis in multiple myeloma cells and overcomes bortezomib resistance.Cancer Cell. 2012; 22: 345-358Abstract Full Text Full Text PDF PubMed Scopus (451) Google Scholar). The subsequent analysis revealed that P5091 downregulated RNF6 at its protein level, along with the cleavage of PARP and Caspase-3, hallmarks of cell apoptosis (Fig. 5A), indicating P5091 induces MM and CML cell apoptosis that might be at least partly contributed by RNF6 degradation. To further confirm this finding, LP1 and K562 were infected with lentiviral RNF6, followed by P5091 treatment and IB assays. The results showed that RNF6 overexpression significantly abolished the action of P5091 in terms of PARP cleavage and Caspase-3 activation in both cell lines (Fig. 5B). Therefore, these results suggested that RNF6 degradation is highly associated with P5091 in MM and CML cell apoptosis. Given that P5091 is an inhibitor of USP7 while USP7 is a Dub of RNF6, we next analyzed the polyubiquitination levels of RNF6 after P5091 treatment by using the IP/IB assay. The results showed that P5091 increased RNF6 polyubiquitination in a concentration-dependent manner (Fig. 5C). Specifically, P5091 markedly increased RNF6 at the K48-linked polyubiquitination form (Fig. 5D), in association with its degradation (Fig. 5A). And this finding was further confirmed by the CHX chase assay in which when protein synthesis de novo was inhibited by CHX, P5091 strikingly increased the turnover rate of RNF6 and the half-life of RNF6 was significantly reduced (Fig. 5E). Notably, P5091 displayed no activity toward the stability of ΔRNF6 or RNF6C/S but the RNF6 (Fig. 5F). Therefore, all these results collectively concluded that USP7 was a Dub of RNF6 auto-ubiquitination and a stabilizer of RNF6 protein. Given that the downregulation of RNF6 protein could trigger MM and leukemia cell apoptosis (1Ren Y. Xu X. Mao C.Y. Han K.K. Xu Y.J. Cao B.Y. et al.Rnf6 promotes myeloma cell proliferation and survival by inducing glucocorticoid receptor polyubiquitination.Acta Pharmacol. Sin. 2020; 41: 394-403Crossref PubMed Scopus (21) Google Scholar, 2Xu X. Han K. Tang X. Zeng Y. Lin X. Zhao Y. et al.The ring finger protein rnf6 induces leukemia cell proliferation as a direct target of pre-B-cell leukemia homeobox 1.J. Biol. Chem. 2016; 291: 9617-9628Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar, 3Lu Q. He Y. Wang Y. Gao L. Zheng Y. Zhang Z. et al.Saponins from Paris forrestii (Takht.) H. Li display potent activity against acute myeloid leukemia by suppressing the RNF6/AKT/mTOR signaling pathway.Front. Pharmacol. 2018; 9: 673Crossref PubMed Scopus (15) Google Scholar) and auto-ubiquitination could lead to RNF6 degradation, we wondered whether induction of RNF6 auto-ubiquitination and degradation could lead to MM and leukemia cell death. To this end, a panel of anti-MM and anti-leukemia drugs including Panobinostat (LBH589) (10Laubach J.P. Tuchman S.A. Rosenblatt J.M. Mitsiades C.S. Colson K. Masone K. et al.Phase 1 open-label study of panobinostat, lenalidomide, bortezomib + dexamethasone in relapsed and relapsed/refractory multiple myeloma.Blood Cancer J. 2021; 11: 20Crossref PubMed Scopus (7) Google Scholar), an FDA-approved anti-MM drug, and Nilotinib, an FDA-approved anti-leukemia drug (11Jain P. Kantarjian H. Alattar M.L. Jabbour E. Sasaki K. Nogueras Gonzalez G. et al.Long-term molecular and cytogenetic response and survival outcomes with imatinib 400 mg, imatinib 800 mg, dasatinib, and nilotinib in patients with chronic-phase chronic myeloid leukaemia: retrospective analysis of patient data from five clinical trials.Lancet Haematol. 2015; 2: e118-e128Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar), were applied for the study. As shown in Figure 6A, LBH589 and Nilotinib induced apoptosis of MM cell line LP1 and leukemia cell line K562, respectively, as evidenced by the cleavage of PARP and Caspase-3, hallmarks of cell apoptosis, which was consistent with previous studies (12Maiso P. Carvajal-Vergara X. Ocio E.M. Lopez-Perez R. Mateo G. Gutierrez N. et al.The histone deacetylase inhibitor LBH589 is a potent antimyeloma agent that overcomes drug resistance.Cancer Res. 2006; 66: 5781-5789Crossref PubMed Scopus (225) Google Scholar, 13Sacha T. Saglio G. Nilotinib in the treatment of chronic myeloid leukemia.Future Oncol. 2019; 15: 953-965Crossref PubMed Scopus (34) Google Scholar). To find out whether RNF6 was involved in cell apoptosis induced by these two drugs, LP1 and K562 cells were infected with lentiviral RNF6, followed by drug treatment and IB assays. The results showed that both LBH589 and Nilotinib induced cell apoptosis as evidenced by PARP cleavage and Caspase-3 activation (Fig. 6B). However, when RNF6 was overexpressed, both PARP cleavage and Caspase-3 activation were markedly reduced (Fig. 6B), suggesting RNF6 partly abolished apoptosis induced by these two drugs and that RNF6 was a potential target of the drugs. Next, we wondered whether this kind of degradation of RNF6 was associated with ubiquitination. To find this out, the proteins from drug-treated cells were subjected to IP with an anti-RNF6 antibody followed by ubiquitination evaluation by IB assays with a ubiquitin-specific antibody. It clearly demonstrated that both drugs markedly increased RNF6 polyubiquitination (Fig. 6C). Moreover, we found that both LBH589 and Nilotinib specifically induced the K48- but not the K63-linked form (Fig. 6D). Furthermore, we found that these drugs directly mediated RNF6 polyubiquitination in the in vitro assay in the absence of any extra ubiquitination ligases (Fig. 6E). Therefore, these results together firmly demonstrated that RNF6 auto-ubiquitination could be activated by clinical drugs such as LBH589 and Nilotinib. It also suggests that chemical induction of RNF6 auto-ubiquitination could be a novel therapeutic strategy against hematological malignancies such as myeloma and leukemia. Auto-ubiquitination is a characteristic of the RING E3 ligases (4Canning M. Boutell C. Parkinson J. Everett R.D. A RING finger ubiquitin ligase is protected from autocatalyzed ubiquitination and degradation by binding to ubiquitin-specific protease USP7.J. Biol. Chem. 2004; 279: 38160-38168Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar), and auto-ubiquitination may represent a significant manner by which E3 ligases regulate their own stability within the cell. Induction of auto-ubiquitination and subsequent degradation could be a potential strategy for cancer treatment by targeting a specific oncogenic RING family E3 ligase (14Yang Y.L. Li X.M. The iap family: endogenous caspase inhibitors with multiple biological activities.Cell Res. 2000; 10: 169-177Crossref PubMed Scopus (259) Google Scholar). Therefore, understanding the molecular modulation is critical to develop novel RING E3 ligases as targets for cancer treatment. In the present study, we provided comprehensive evidence that the RING family ubiquitin ligase RNF6 undergoes auto-ubiquitination, and induction of RNF6 auto-ubiquitination could be a promising strategy for the treatment of some hematological malignancies such as MM and leukemia. RNF6 as a ubiquitin ligase has been reported in various cancers that modifies substrate proteins, therefore promoting their oncogenic activity. For example, RNF6 mediates K27-linked polyubiquitination of the androgen receptor (AR), thereby recruiting the chaperone proteins to increase the oncogenic transcriptional activity of AR in prostate cancer (15Xu K. Shimelis H. Linn D.E. Jiang R. Yang X. Sun F. et al.Regulation of androgen receptor transcriptional activity and specificity by RNF6-induced ubiquitination.Cancer Cell. 2009; 15: 270-282Abstract Full Text Full Text PDF PubMed Scopus (180) Google Scholar). In myeloma cells, RNF6 binds to and triggers glucocorticoid receptor for K63-linked polyubiquitination, therefore promoting its transcriptional activity to promote the transcription of prosurvival genes including Bcl2L1 and Mcl-1 (1Ren Y. Xu X. Mao C.Y. Han K.K. Xu Y.J. Cao B.Y. et al.Rnf6 promotes myeloma cell proliferation and survival by inducing glucocorticoid receptor polyubiquitination.Acta Pharmacol. Sin. 2020; 41: 394-403Crossref PubMed Scopus (21) Google Scholar). RNF6 also acts on some tumor suppressors such as SHP-1 (16Liang Q. Ma D. Zhu X. Wang Z. Sun T.T. Shen C. et al.RING-finger protein 6 amplification activates JAK/STAT3 pathway by modifying SHP-1 ubiquitylation and associates with poor outcome in colorectal cancer.Clin. Cancer Res. 2018; 24: 1473-1485Crossref PubMed Scopus (44) Google Scholar) and TLE3 (17Liu L. Zhang Y. Wong C.C. Zhang J. Dong Y. Li X. et al.RNF6 promotes colorectal cancer by activating the Wnt/beta-catenin pathway via ubiquitination of TLE3.Cancer Res. 2018; 78: 1958-1971Crossref PubMed Scopus (60) Google Scholar) by promoting their polyubiquitination and degradation in colorectal cancer. In the present study, we found that RNF6 as a RING family E3 ligase can direct its own polyubiquitination, and this action depends on its RING domain. Both the in vivo and the in vitro ubiquitination assays demonstrate that when the RING domain is deleted, RNF6 loses its auto-ubiquitination ability. It is well known that the cysteine residue cooperates with histidine residues in the RING domain to act as ligands of Zn2+ to stabilize the scaffold of the RING finger E3 ligase (18Cassandri M. Smirnov A. Novelli F. Pitolli C. Agostini M. Malewicz M. et al.Zinc-finger proteins in health and disease.Cell Death Discov. 2017; 317071Crossref PubMed Scopus (404) Google Scholar). We indeed found that the C/S mutant becomes significantly stable; however, we also found that in the extended treatment (24 h), MG132 can increase RNF6C/S stability, which is consistent with the CHX assay, suggesting that, different from the ΔRING mutant, the mutant RNF6C/S probably could be degraded via the proteasomes. But this finding does not occlude the conclusion that loss of either RING or C2H2 structure inactivates RNF6 in its ubiquitin ligase activity because in the absence of any extra E3 ligases in the in vitro ubiquitination assay, only the WT but not the ΔRING or C/S mutant could be polyubiquitinated, suggesting that RNF6 undergoes auto-polyubiquitination and this auto-ubiquitination depends on its RING domain. This finding is consistent with previous reports on protein auto-ubiquitination, such as XIAP (5Yang Y. Fang S. Jensen J.P. Weissman A.M. Ashwell J.D. Ubiquitin protein ligase activity of IAPs and their degradation in proteasomes in response to apoptotic stimuli.Science. 2000; 288: 874-877Crossref PubMed Scopus (875) Google Scholar), TRIM26 (7Ran Y. Zhang J. Liu L.L. Pan Z.Y. Nie Y. Zhang H.Y. et al.Autou
Referência(s)