記住我
Research ArticleDermatologyOncology
Open Access | 
10.1172/JCI171063
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Wu, M. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Hanly, A. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Gibson, F. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Fisher, R. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Rogers, S. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Park, K. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Zuger, A. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Kuang, K. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by
Kalin, J.
in:
JCI
|
PubMed
|
Google Scholar
|
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Nocco, S. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Cole, M. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Xiao, A. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Agus, F. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by
Labadorf, A.
in:
JCI
|
PubMed
|
Google Scholar
|
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by
Beck, S.
in:
JCI
|
PubMed
|
Google Scholar
|
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Collard, M. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Cole, P. in: JCI | PubMed | Google Scholar
1Department of Dermatology, Boston University Chobanian and Avedisian School of Medicine, Boston, Massachusetts, USA.
2Division of Genetics, Departments of Medicine and Biological Chemistry and Molecular Pharmacology, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA.
3Bioinformatics Program, Boston University, Boston, Massachusetts, USA.
4Department of Neurology, Boston University School of Medicine, Boston, Massachusetts, USA.
Address correspondence to: Rhoda M. Alani, 609 Albany Street, J-507, Boston, Massachusetts 02190, USA. Phone: 617.358.9770; Email: alani@bu.edu. Or to: Philip A. Cole, New Research Building, 77 Avenue Louis Pasteur, Room 168C, Boston, Massachusetts 02115, USA. Phone: 617.525.5208; Email: pacole@bwh.harvard.edu.
Find articles by Alani, R. in: JCI | PubMed | Google Scholar
Published February 1, 2024 - More info
Published in Volume 134, Issue 6 on March 15, 2024
AbstractVirtually all patients with BRAF-mutant melanoma develop resistance to MAPK inhibitors largely through nonmutational events. Although the epigenetic landscape is shown to be altered in therapy-resistant melanomas and other cancers, a specific targetable epigenetic mechanism has not been validated. Here, we evaluated the corepressor for element 1–silencing transcription factor (CoREST) epigenetic repressor complex and the recently developed bivalent inhibitor corin within the context of melanoma phenotype plasticity and therapeutic resistance. We found that CoREST was a critical mediator of the major distinct melanoma phenotypes and that corin treatment of melanoma cells led to phenotype reprogramming. Global assessment of transcript and chromatin changes conferred by corin revealed specific effects on histone marks connected to epithelial-mesenchymal transition–associated (EMT-associated) transcription factors and the dual-specificity phosphatases (DUSPs). Remarkably, treatment of BRAF inhibitor–resistant (BRAFi-R) melanomas with corin promoted resensitization to BRAFi therapy. DUSP1 was consistently downregulated in BRAFi-R melanomas, which was reversed by corin treatment and associated with inhibition of p38 MAPK activity and resensitization to BRAFi therapies. Moreover, this activity was recapitulated by the p38 MAPK inhibitor BIRB 796. These findings identify the CoREST repressor complex as a central mediator of melanoma phenotype plasticity and resistance to targeted therapy and suggest that CoREST inhibitors may prove beneficial for patients with BRAFi-resistant melanoma.
Graphical Abstract
Introduction
Melanomas exhibit tremendous intratumoral heterogeneity and phenotype plasticity, which allows them to switch between distinctive transcriptional programs in response to external stressors, including targeted therapies (1). These transcriptional phenotypes, which are mediated largely through epigenetic mechanisms (1, 2), are characterized by altered differentiation and metabolic states including a proliferative/differentiated/MITFhi/AXLlo phenotype and an undifferentiated/invasive/MITFlo/AXLhi phenotype (3, 4) with associated changes in the epigenetic landscape (5, 6). Moreover, cellular plasticity is a driver of resistance to targeted therapies in melanoma and other cancers, with dynamic transitions between distinctive molecular phenotypes promoting MAPK inhibitor (MAPKi) bypass mechanisms (7). As reversible transcriptional reprogramming dictates the plasticity of molecular phenotypes, research efforts have focused on the role of epigenetic regulation in this process (8–10).
The corepressor for element 1–silencing transcription factor (CoREST) epigenetic repressor complex is a member of the class I histone deacetylase family of repressor complexes that was originally identified as a cofactor for REST repression (11) and regulation of neuron-specific gene silencing during development (12), but more recently has been connected with the snail family of transcription factors (13). RE1-silencing transcription factor (REST) corepressor 1 (RCOR1) functions as a scaffold for the CoREST repressor complex promoting crosstalk between histone deacetylases 1 and 2 (HDAC1/2) and lysine demethylase 1A (LSD1), an H3K4 demethylase (14), and has been shown to regulate Treg function and antitumor immunity (15). We have recently described corin, a potent and specific dual-warhead inhibitor of the CoREST complex targeting HDAC1/2 and LSD1, that demonstrates growth inhibition in melanoma (16), cutaneous squamous cell carcinoma (16), breast cancers (17), and diffuse intrinsic pontine glioma (18). We therefore hypothesized that CoREST inhibition may elicit synergistic growth inhibition with BRAF inhibitor (BRAFi) therapies through epigenetic reprogramming of BRAF-mutant melanoma.
Here, we show that CoREST inhibition in human melanoma cell lines reversed the 2 major melanoma cell phenotypes, those characterized as either MITFhi/AXLlo or MITFlo/AXLhi, and resensitized BRAFi-resistant (BRAFi-R) melanoma cells to BRAFi therapy. Additionally, we explored the transcriptomic and epigenomic landscapes regulated by CoREST inhibition and found specific alteration of epithelial-mesenchymal transition–associated (EMT-associated) transcription factors following corin treatment of BRAFi-R melanomas in addition to upregulation of the dual-specificity phosphatases and downstream inhibition of p38 MAPK activity. We further noted specific reactivation of BRAFi sensitivity in BRAFi-R melanoma cells following treatment with the p38 inhibitor BIRB 796, suggesting a specific mechanism of action for corin-associated resensitization of BRAFi-R melanomas to BRAFi therapies in this setting. In vivo studies demonstrate enhanced inhibition of BRAFi-R melanoma growth following treatment with the combination of corin plus BRAFi versus corin alone, further supporting a role for CoREST inhibition in resensitizing BRAFi-R melanomas to BRAF therapies and suggesting a potential role for CoREST inhibition as a therapeutic modality to enhance MAPK-targeted therapies in patients with advanced melanoma.
ResultsThe CoREST repressor complex mediates phenotype switching in melanoma. In order to further explore the role of the CoREST repressor complex in human melanoma development, we treated a panel of phenotypically distinct melanoma cell lines with the CoREST inhibitor corin for 24 hours (Figure 1). Remarkably, all tumor cell lines demonstrating an MITFhi/AXLlo melanoma phenotype showed decreased expression of MITF and increased expression of histone H3K4me2, a common readout of LSD1 inhibition (19, 20), and H3K9ac/K27ac marks following corin treatment, without significant effects on the MEK/ERK pathway (Figure 1A). In addition, corin treatment of MITFlo/AXLhi melanoma cells led to decreased AXL expression and increased histone acetylation and methylation marks in all cell lines evaluated, without affecting the MEK/ERK pathway (Figure 1B). These data suggest a specific reversion of melanoma differentiation phenotypes to intermediate, non-MITFhi, and non-AXLhi cellular programs following corin treatment. Moreover, treatment of melanoma cells for 24 hours with corin inhibited tumor cell growth in the majority of MITFhi/AXLlo melanoma cells evaluated; however, minimal growth inhibition and even enhanced cellular proliferation was seen in MITFlo/AXLhi cells following 24 hours of treatment with corin (Figure 1C), consistent with conversion of both distinctive melanoma cell line phenotypes to intermediate proliferation states following early corin treatment. Corin treatment also led to increased cellular invasion and expression of focal adhesions in MITFhi/AXLlo cells, while decreasing invasion and expression of focal adhesions in MITFlo/AXLhi cells (Figure 1, D and E). Notably, extended corin treatment of all melanoma cell lines (72 hours) led to substantial) growth inhibition with IC50 values consistently in the submicromolar range (Figure 1F, Table 1, Supplemental Figure 1, A–C, and Supplemental Table 1; supplemental material available online with this article; https://doi.org/10.1172/JCI171063DS1). We also observed morphological changes following corin treatment, with both 1205Lu and 451Lu cells displaying an elongated morphology (Figure 2A). Notably, corin-treated MITFlo/AXLhi (1205Lu) cells demonstrated a more melanocytic phenotype compared with the DMSO control–treated cells, whereas MITFhi/AXLlo (451Lu) cells had a more senescent cellular phenotype, which was confirmed by β-galactosidase staining (Figure 2B).
Figure 1The CoREST repressor complex mediates phenotype switching in melanoma cell lines. (A and B) Western blot analysis of the MITFhi/AXLlo melanoma cell lines 451Lu, SKMel28, WM35, and WM983B (A) and the MITFlo/AXLhi melanoma cell lines Sbcl2, WM1552C, 1205Lu, and A375 (B) following 24 hours of treatment with DMSO or 2.5 μM corin. 1205Lu lysates were used as a positive control for AXL in A. Western blots were run contemporaneously. (C) Cellular proliferation of MITFhi/AXLlo (upper panel) and MITFlo/AXLhi (lower panel) melanoma cell lines following 24 hours of treatment with DMSO or 2.5 μM corin (n = 3). (D) Invasion assay and quantification of MITFhi/AXLlo (WM983B, upper panel) and MITFlo/AXLhi (1205Lu, lower panel) melanoma cells following 24 hours treatment with DMSO or 2.5 μM corin (n = 4–8). Representative images are shown. Scale bar: 100 μm. (E) Vinculin staining of focal adhesions in MITFhi/AXLlo melanoma cells (WM983B, upper panel) and MITFlo/AXLhi melanoma cells (1205Lu, lower panel) following a 24-hour treatment with DMSO or 2.5 μM corin (n = 5–6), with quantification of focal adhesions (puncta/cell) on the right (n = 5–6). Representative images are shown. Scale bar: 20 μm. (F) Dose-response proliferation assays of MITFhi/AXLlo (WM35, WM983B, 451Lu, SkMel28) and MITFlo/AXLhi (Sbcl2, WM1552C, 1205Lu, A375) melanoma cell lines treated with increasing doses of corin for 72 hours. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001, by 2-tailed, unpaired t test compared with DMSO controls.
Figure 2CoREST inhibition induces morphological changes in melanoma cells. (A) Average cell length for MITFhi/AXLlo 451Lu and MITFlo/AXLhi 1205Lu melanoma cells treated with 1 μM corin for 0, 24, and 72 hours (n = 20). Representative images shown. Scale bars: 100 μm. (B) Senescence-associated β-galactosidase staining of 451Lu melanoma cells treated with DMSO or 1 μM corin for 72 hours and quantification. Scale bar: 50 μm. **P < 0.01, ***P < 0.001, and ****P < 0.0001, by 2-tailed, unpaired t test compared with DMSO controls.
Table 1IC50 values of melanoma cells treated with corin for 72 hours
CoREST inhibition resensitizes BRAFi-R melanoma cells to BRAFi therapy. As tumor cell plasticity and phenotype switching is associated with resistance to targeted therapies in melanoma and other cancers, and intermediate phenotypes are generally considered to be treatment sensitive (21, 22), we next investigated the effect of CoREST inhibition combined with the BRAFi PLX4032 (vemurafenib, referred to herein as PLX4032) on BRAFi-R melanoma cell proliferation. Notably, we found that corin markedly increased the antiproliferative effects of PLX4032 in all BRAFi-R melanoma cells evaluated (Figure 3A). In addition, silencing of the CoREST scaffolding protein RCOR1 also led to resensitization of BRAFi-R MITFhi/AXLlo and MITFlo/AXLhi melanoma cells to PLX4032 (Figure 3, B and C). Interestingly, we also found that PLX4032 enhanced the antiproliferative effects of corin treatment alone in both BRAFi-R MITFhi/AXLlo and MITFlo/AXLhi melanoma cells (Figure 3D), with synergy identified between PLX4032 and corin in MITFhi/AXLlo and MITFlo/AXLhi melanoma cells (combination index [CI] <1.0 [ref. 23]) (Figure 3E, Table 2, Supplemental Figure 1D, and Supplemental Table 2). Of note, growth inhibition in BRAFi-R melanoma cells treated with hi-dose or low-dose PLX4032 was increased following corin treatment versus treatment with the LSD1 inhibitor Cpd7 or the HDAC1 inhibitor MS275 alone, suggesting specific synergies with PXL4032 activity relevant to targeting of the CoREST complex (Figure 3F). In addition, corin treatment significantly reduced colony formation and increased apoptosis in BRAFi-R melanoma cells (Figure 4, A and B, Supplemental Figure 1, E and F), which was further enhanced in combination with PLX4032 (Figure 4, A and B). While the synergy between corin and PLX was not as evident in the colony formation assays compared with the proliferation assays, this may reflect differences between corin’s efficacy in bulk cell proliferation compared with single-cell clonogenic potential. Corin treatment also induced cellular senescence in BRAFi-R MITFhi/AXLlo melanoma cells (Figure 4C) but not in MITFlo/AXLhi melanoma cells, without obvious effects on autophagy (Supplemental Figure 1, G–I).
Figure 3CoREST inhibition resensitizes BRAFi-R melanoma cells to BRAFi therapy. (A) Proliferation assays of 451Lu-R and SkMel28-R (MITFhi/AXLlo) and 1205Lu-R and A375-R (MITFlo/AXLhi) BRAFi-R melanoma cell lines treated with increasing doses of PLX4032 with or without 1 μM corin for 72 hours (n = 3). (B) RCOR1 knockdown by shRNA in 1205Lu-R melanoma cells, confirmed by RT-qPCR (data are representative of 2 independent experiments). (C) PLX4032 (PLX) dose-response curves (72-hour treatment) in 451Lu-R and 1205Lu-R melanoma cells following knockdown of RCOR1 (n = 3). (D) Proliferation assays of 451Lu-R and 1205Lu-R BRAFi-R melanoma cell lines treated with increasing doses of corin with or without 1 μM or 5 μM PLX4032 for 72 hours (n = 3). (E) Drug synergy graphs for corin and PLX4032 in 451Lu-R (left) and 1205Lu-R (right) BRAFi-R melanoma lines. (F) Proliferation assays of A375-R melanoma cells treated with 0.1 μM (top panel) or 1 μM (bottom panel) PLX4032 with or without corin, MS275, or compound 7 (Cpd7) at 0.1 μM and 1 μM for 72 hours (n = 1).
Figure 4CoREST inhibition reduces colony formation and induces apoptosis in BRAFi-R melanoma cells with or without BRAFi. (A) Colony formation assay quantification in 451Lu-R (left) and 1205Lu-R (right) melanoma cells treated with DMSO, 5 μM PLX4032 alone, 2.5 μM corin alone, or 2.5 μM corin plus 5 μM PLX4032 for 10 days (n = 3). (B) Quantification of TUNEL+ cells in 451Lu-R and 1205Lu-R melanoma cells following 72 hours of treatment with DMSO, 5 μM PLX4032 alone, 1 μM corin alone, or 1 μM corin plus 5 μM PLX4032 (n = 3). (C) Senescence-associated β-galactosidase staining of 451Lu-R melanoma cells treated with DMSO, 5 μM PLX4032 alone, 1 μM corin alone, or 1 μM corin plus 5 μM PLX4032 for 72 hours and quantification (n = 2–3). Scale bar: 50 μm. *P < 0.05, **P < 0.01, and ****P < 0.0001, by 1-way ANOVA with Tukey’s test.
Table 2CI for 451Lu-R and 1205Lu-R cells treated with PLX4032 and corin
Inhibition of the CoREST complex in BRAFi-R melanoma promotes transcriptional changes associated with the phenotype switch and increased expression of DUSP family MAPK inhibitors. In order to investigate mechanisms of corin resensitization of BRAFi-R melanoma cells to PLX4032, RNA-Seq was performed on 451Lu-R (MITFhi/AXLlo) and 1205Lu-R (MITFlo/AXLhi) cells treated with PLX4032 alone, corin alone, or corin in combination with PLX4032 for 24 hours. Differentially regulated genes were identified by comparison with PLX4032-treated controls (Figure 5A). Not surprisingly, we found that corin treatment led to substantially greater numbers of upregulated genes than downregulated genes, consistent with the repressive functions of the CoREST complex. K-means clustering of differentially regulated genes revealed common corin-affected gene clusters as well as cell line–specific, corin-affected gene clusters (Figure 5B), which were characterized ontologically (Supplemental Figure 2, A and B). Although ontological analysis paired with gene set enrichment analysis (GSEA) using the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway gene sets revealed common corin-regulated gene sets (Figure 5C), we also identified distinct pathways specific to each cell line (Figure 5D). Cell-cycle and DNA repair–related gene sets were commonly downregulated in both cell lines with corin treatment (Figure 5, C and E, and Supplemental Figure 2A), confirmed by real-time quantitative PCR (RT-qPCR) (Supplemental Figure 2, C and D), while MAPK signaling, cell adhesion molecules, axon guidance, hedgehog signaling, and neuronal pathways were commonly upregulated with corin treatment (Figure 5, C and E and Supplemental Figure 2A). Differential expression of key genes involved in focal adhesion, axon guidance, and EMT were confirmed by RT-qPCR (Supplemental Figure 2, E and F and Supplemental Figure 3, A–C), suggesting a cellular phenotype switch similar to what was observed in the BRAFi-sensitive (BRAFi-S) melanoma cells treated with corin.
Figure 5CoREST complex inhibition in BRAFi-R melanoma cells induces global transcriptional upregulation and causes changes in cell-cycle, MAPK, and axon guidance signaling pathways. (A) RNA-Seq profiling of 451Lu-R and 1205Lu-R melanoma cells following 24 hours of treatment with 5 μM PLX4032 versus 2.5 μM corin plus 5 μM PLX4032. Venn diagram illustrates corin plus PLX4032–induced upregulated and downregulated expression of cell line–specific and common genes compared with PLX4032 treatment alone (fold change [FC] ≥4, FDR <0.001). (B) K-means clustered (K = 6) heatmap of all genes showing significant expression changes (Padj < 0.01, |log2 FC| >2) upon corin treatment in at least 1 condition/cell line (i.e., with or without [w/o] PLX4032; 451Lu-R or 1205Lu-R). In each cell line, expression in the absence of PLX4032 and corin was set as reference (i.e., 0), and the relative expression changes were summarized as SD (σ). Numbers in the brackets indicate the gene counts in clusters. (C) GSEA of corin-induced common enriched pathways in 451Lu-R and 1205Lu-R melanoma cells. All gene sets displayed are significantly (P < 0.05) enriched with corin plus PLX4032 versus PLX4032 treatment alone. (D) GSEA of corin-induced distinct enriched pathways in 451Lu-R (top) and 1205Lu-R (bottom) melanoma cells. All gene sets displayed are significantly (P < 0.05) enriched with corin plus PLX4032 versus PLX4032 treatment alone. (E) GSEA in samples treated with corin plus PLX4032 versus those treated with PLX4032 alone illustrating representative common enriched gene sets in 451Lu-R (top) and 1205Lu-R (bottom) melanoma cells. NES, normalized enrichment score. n = 2 for all panels in this figure.
Comparison of corin-associated RNA-Seq data with publicly available data sets (3, 24, 25) supported a phenotype switch signature following corin treatment in both 451Lu-R and 1205Lu-R cells; genes associated with the corin-induced phenotype switch included AXL, MITF, SOX10, WNT5A, PAX3, ZEB1, ZEB2, PGC1a, DUSP1, and DUSP5 (Figure 6A) and were significantly associated with the intermediate melanoma phenotype signature recently reported by Wouters et al. (26) (Figure 6B and Supplemental Figure 3D). Consistent with RNA-Seq results, Western blot analysis of BRAFi-R cell lines treated with corin showed reduced MITF protein expression in MITFhi/AXLlo cells and reduced AXL protein expression in MITFlo/AXLhi cells (Supplemental Figure 4, A and B). Of note, the dual-specificity phosphatases DUSP1 and DUSP5 were among the genes whose expression was most highly upregulated in the MAPK signaling GSEA and significantly (adjusted P [Padj] < 1 × 10–31 ) upregulated following corin treatment in both BRAFi-R cell lines (Figure 6C) as well as additional melanoma cell lines (
留言 (0)