Purpose: The ubiquitin proteasome pathway is a validated therapeutic target in multiple myeloma. Deubiquitylating enzyme USP1 participates in DNA damage response and cellular differentiation pathways. To date, the role of USP1 in multiple myeloma biology is not defined. In the present study, we investigated the functional significance of USP1 in multiple myeloma using genetic and biochemical approaches.
Experimental Design: To investigate the role of USP1 in myeloma, we utilized USP1 inhibitor SJB3-019A (SJB) for studies in myeloma cell lines and patient multiple myeloma cells.
Results: USP1-siRNA knockdown decreases multiple myeloma cell viability. USP1 inhibitor SJB selectively blocks USP1 enzymatic activity without blocking other DUBs. SJB also decreases the viability of multiple myeloma cell lines and patient tumor cells, inhibits bone marrow plasmacytoid dendritic cell–induced multiple myeloma cell growth, and overcomes bortezomib resistance. SJB triggers apoptosis in multiple myeloma cells via activation of caspase-3, caspase-8, and caspase-9. Moreover, SJB degrades USP1 and downstream inhibitor of DNA-binding proteins as well as inhibits DNA repair via blockade of Fanconi anemia pathway and homologous recombination. SJB also downregulates multiple myeloma stem cell renewal/survival-associated proteins Notch-1, Notch-2, SOX-4, and SOX-2. Moreover, SJB induced generation of more mature and differentiated plasma cells. Combination of SJB and HDACi ACY-1215, bortezomib, lenalidomide, or pomalidomide triggers synergistic cytotoxicity.
Conclusions: Our preclinical studies provide the framework for clinical evaluation of USP1 inhibitors, alone or in combination, as a potential novel multiple myeloma therapy. Clin Cancer Res; 23(15); 4280–9. ©2017 AACR.
This article is featured in Highlights of This Issue, p. 3975
Proteasome inhibitors are effective therapy for multiple myeloma; however, the development of resistance and relapse of disease are common. Recent research discovered novel drugs that modulate protein ubiquitin-conjugating/deconjugating enzymes rather than the proteasome itself, thus leading to less toxic side effects, and can overcome proteasome inhibitor resistance in multiple myeloma. In this study, we show that USP1 inhibition by SJB3-019A decreases cell viability in multiple myeloma cells. The modulation of ID proteins and DNA repair proteins by USP1 inhibitor SJB3-019A shown here has further biologic implications and clinical applications. SJB3-019A–mediated inhibition of USP1/ID/Notch/Sox2 pathway triggers immature plasma cells to differentiate, mature, and undergo apoptosis. Isobologram analysis demonstrated that the combination of SJB3-019A with bortezomib, HDAC6i ACY1215, lenalidomide, or pomalidomide triggers synergistic anti–multiple myeloma activity. Our preclinical data support clinical investigation of USP1 inhibitors alone or in combination with other agents in multiple myeloma.
Proteasome inhibitors are effective therapy for relapsed/refractory, relapsed, and newly diagnosed multiple myeloma; however, the development of resistance and relapse of disease is common (1–3). Recent research discovered novel drugs that modulate protein ubiquitin-conjugating/deconjugating enzymes rather than the proteasome itself. We recently showed that targeting deubiquitylating enzymes (DUBs) USP14, UCHL5, and USP7 can overcome proteasome inhibitor resistance in multiple myeloma (4–8).
DUBs deconjugate ubiquitin from targeted proteins and facilitate regeneration of free ubiquitin pools (4). In addition, DUBs maintain cellular protein homeostasis by modulating protein activation, turnover rate, recycling, and localization (4). Alteration in DUBs activity has been linked with several diseases, including cancer (5, 6). The human genome encodes about 100 DUBs, which are classified into 5 families: the USP (the ubiquitin-specific processing protease), UCH (the ubiquitin C-terminal hydrolase), OTU (the ovarian tumor), MJD (the Josephin domain), and JAMM (the Jab1/Mov34 metalloenzyme). The first 4 families are cysteine proteases, whereas the fifth family is metalloproteases, and to date, USP and UCH are the best characterized families (4).
USP1 regulates DNA repair through Fanconi anemia (FA) pathway by deubiquitylating DNA repair proteins, FANCD2-Ub and PCNA-Ub (9). For example, USP1 deubiquitylates monoubiquitylated PCNA, which inhibits recruitment of DNA polymerases in the absence of DNA damage, and thereby leads to regulated DNA repair. USP1 also regulates DNA break repair via homologous recombination (HR) pathway (10). Conversely, inhibition of USP1 sensitizes cancer cells to chemotherapy and radiation (9). As USP1 participates in DNA damage response pathways, USP1-knockout mice are genetically unstable and highly sensitive to DNA damage (11). Finally, USP1 inhibits cell differentiation by stabilizing tumor-promoting inhibitor of DNA-binding (ID) proteins (12, 13). To date, the role of USP1 in multiple myeloma biology is undefined. In the present study, we investigated the functional significance of USP1 in multiple myeloma using genetic and biochemical approaches.
Materials and Methods
Cell culture and reagents
Multiple myeloma cell lines and normal donor–derived peripheral blood mononuclear cells (PBMC) were cultured in complete medium containing 10% FBS and antibiotics. All cell lines were tested for mycoplasma contamination using MycoAlertTM mycoplasma detection kit (Lonza). Plasmacytoid dendritic cells (pDC), bone marrow stromal cells (BMSC), or tumor cells from patients with multiple myeloma were purified and cultured as described previously (14). All patient samples were obtained with prior informed consent in accordance Helsinki protocol. Bone marrow mononuclear cells (MNC) were purchased from Allcells (USA). SJB3-019A (SJB), bortezomib, lenalidomide, pomalidomide, and ACY-1215 were obtained from Selleck chemicals (USA).
Cell cycle, cell viability, and apoptosis assays
Western blotting assays
Immunoblotting was performed as previously described (15) using antibodies against USP1, USP2, USP5, USP7, USP14, caspase-3/8/9, p21, FANCD2, FANCI, PCNA, Rad51, GAPDH (Cell Signaling); ID1, ID2, ID3, ID4, Notch-1, Notch-2, Sox-4, and Sox-2 (Bethyl Laboratories).
MM.1S cells were transiently transfected with control scr-siRNA or USP1-siRNA using the Cell Line Nucleofector Kit V (Amaxa Biosystems). Cells were harvested 24 hours posttransfection, followed by analysis using both immunoblotting and cell viability assays.
Ubiquitin vinyl sulfone labeling, Ub-AMC, and tetraubiquitin chain cleavage assays
Cells were treated with or without SJB for 3 hours and were harvested and lysed. Total protein (25 μg) was labeled with hemaglutinnin (HA)-linked Ub-VS probe (1 μmol/L) for 30 minutes at 37°C and analyzed with immunoblotting.
recombinant DUBs (rDUB; USP1/UAF1, USP2, USP5, USP7, or UCH37) were incubated with SJB for 30 minutes at 37°C, and then UB-AMC was added for another 30 minutes, followed by measurement of fluorescence intensity.
Ubiquitin-linked K48 chain cleavage assay
Purified rDUBs were incubated with SJB for 30 minutes, followed by the addition of K48-linked tetraubiquitin chains. The reaction was terminated after 30 minutes by addition of reducing buffer, and samples were analyzed by Western blotting (17).
Multiple myeloma cells were stained with Rad51 Ab and Giemsa stain as described previously, and sections were then imaged by microscopy (18).
USP1 gene expression analysis
The exon-1.0 ST array data for 170 newly diagnosed patients with multiple myeloma were quality-controlled and normalized with aroma Affymetrix package. Gene expression was estimated with a PLM model. The survival analysis was carried out using the R package.
The raw data for expression profiling and the CEL files can be found at the website Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/) under accession numbers: GSE5900 and GSE2658. Survival data can be accessed at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE39754.
The Student t test was utilized to derive statistical significance. Synergistic cytotoxic activity of combination regimes was assessed with isobologram analysis and CalcuSyn software program (19).
USP1 expression analysis in multiple myeloma cells
Examination of gene expression datasets showed a higher USP1 in clonal plasma cells from patients with monoclonal gammopathy of undetermined significance (MGUS), smoldering multiple myeloma (SMM), and active multiple myeloma versus normal plasma cells (Fig. 1A). Immunoblot analysis showed elevated USP1 levels in a panel of multiple myeloma cell lines versus normal healthy donor–derived bone marrow MNCs or PBMCs (Fig. 1B). The prognostic relevance of USP1 was assessed by correlating baseline USP1 expression in bone marrow biopsy samples with overall and event-free survival of 170 patients with multiple myeloma. All patients analyzed in this study were newly diagnosed, and no therapy was administered at the time of expression profiling (20). We found a statistically significant inverse correlation between USP1 levels and both overall and event-free survival (Fig. 1C). These findings indicate a role of USP1 in multiple myeloma pathogenesis.
To determine the functional significance of USP1 in multiple myeloma, we performed loss-of-function studies with USP1-siRNA. Transfection of USP1-siRNA, but not scr-siRNA, reduces multiple myeloma cell viability (Fig. 1D, bar graph; P < 0.05). These data show a role for USP1 in survival of multiple myeloma cells.
Biochemical characterization of USP1 inhibitor SJB
SJB, 2-(pyridin-3-yl)naphtho[2,3-d]oxazole-4,9-dione), targets USP1 in an irreversible manner (Fig. 2A). We utilized Ub-AMC (ubiquitin 7-amino-4-methycoumar) assays (Fig. 2B) to assess the effect of SJB on USP1 or other DUBs using both rDUB proteins and cellular DUBs. In vitro assays using purified rDUBs showed that SJB inhibits USP1 (>90% inhibition at 1 μmol/L; Fig. 2C). Importantly, SJB did not significantly affect the activity of other rDUBs (USP2/USP2/USP5/USP7/USP14 or UCH37; Fig. 2C and D). Moreover, a USP7 inhibitor P5091 did not inhibit USP1 DUB activity (Fig. 2C). ML323 was used as a positive control for USP1 inhibition assays (Supplementary Fig. S1).
We next examined the effect of SJB against cellular DUB enzymatic activity. SJB blocked cellular USP1 activity in a concentration-dependent manner, without markedly affecting activity of other DUBs (USP2, USP7, USP14; Fig. 2E, bar graph). A higher concentration of SJB (1 μmol/L) modestly inhibited USP5 activity. Immunoblotting shows that equivalent USP1 immunoprecipitates were analyzed for DUB activity assay (Fig. 2E). Together, these data suggest that SJB is a specific inhibitor of USP1.
To ascertain the effect of SJB on DUB activity, we performed competitive labeling between SJB and ubiquitin-active site probe HA–Ub-VS. Untreated cell lysates incubated with Ub-VS probe showed Ub-VS–USP conjugate formation, represented by an increase in mass of USPs of about 10 kDa (Fig. 3A). Importantly, this conjugate formation was inhibited in the presence of SJB for USP1 enzyme, but not other DUBs. Immunoblotting with anti-HA antibody showed equal protein loading of control and treated protein lysates (Supplementary Fig. S2). These data suggest that SJB binds to cellular USP1 and blocks its deubiquitylating activity. Furthermore, analysis using K48-linked ubiquitin tetramer chain cleavage assays showed that SJB inhibits USP1-mediated ubiquitin chain disassembly in a concentration-dependent manner (Fig. 3B). Proteasome inhibitor marizomib served as negative control in these assays. No significant inhibitory effect of SJB was noted in USP7- or USP2-mediated ubiquitin chain disassembly (Supplementary Fig. S3). Taken together, these findings provide evidence for the specificity of SJB against USP1.
Cytotoxic activity of SJB against multiple myeloma
SJB reduces the viability of all multiple myeloma cell lines in a concentration-dependent manner (IC50 ranges from 100 to 500 nmol/L; Fig. 4A). Treatment of purified multiple myeloma cells from patients showed a concentration-dependent decrease in the viability of all patient multiple myeloma cells (Fig. 4B). In contrast, no significant effect of SJB was noted against normal PBMCs (Fig. 4C). However, the highest concentration of SJB (1 μmol/L) decreases viability of PBMCs by 10% to 20%. These data suggest a favorable therapeutic index for SJB in multiple myeloma. A comparative analysis of SJB with other reported USP1 inhibitors (ML323, SJB2-043, pimozide, GW7647) showed a more potent and selective anti–multiple myeloma activity of SJB (Supplementary Fig. S4).
SJB activity in the multiple myeloma bone marrow microenvironment
Previous studies have shown that BMSCs mediate multiple myeloma cells growth and block drug-induced cytotoxicity (21–24). Treatment of patient BMSCs for 24 hours with SJB does not decrease their viability. Importantly, SJB significantly inhibits BMSC-induced MM.1S cell growth (Fig. 4D). Our earlier reports highlighted the growth-promoting role of pDCs in the multiple myeloma bone marrow microenvironment (14, 25, 26). Using pDC–multiple myeloma co-culture model, we found that SJB inhibits pDC-induced MM.1S cell growth (Fig. 4E). These data demonstrate that SJB targets multiple myeloma cells even in the cytoprotective multiple myeloma–host bone marrow microenvironment.
Mechanistic insight into the SJB activity
Cell-cycle analysis showed that SJB induces G1 phase growth arrest associated with decrease in G2–M and S-phases (Fig. 5A). In concert with these findings, SJB upregulated G1-associated cell-cycle protein p21 (Fig. 5A). Moreover, treatment of MM.1S and Dox-40 cells with SJB induced significant apoptosis associated with a marked increase in PARP cleavage, caspase-3, caspase-8, and caspase-9 (Fig. 5B and Supplementary Fig. S5A).
SJB disrupts HR in multiple myeloma cells
USP1 regulates DNA repair and FA pathways through its association with binding partner UAF1 (27). In particular, USP1 regulates DNA damage response by deubiquitylation of DNA repair proteins, Ub-FANCD2 and Ub-PCNA (9). We therefore examined whether SJB-mediated inhibition of USP1 affects FA signaling pathway in multiple myeloma cells. MM.1S cells were treated with SJB for 12 hours, and cellular levels of Ub-FANCD2, Ub-FANCI, and Ub-PCNA were then analyzed by Western blotting. As shown in Fig. 5C, USP1 inhibition by SJB led to an increase in Ub-FANCD2, Ub-FANCI, and Ub-PCNA levels (Fig. 5C).
Previous studies showed that USP1 inhibition affects HR (10), and we next examined the effects of SJB on HR in multiple myeloma cells. For these studies, we assessed whether SJB alters expression of HR-associated RAD51 protein (28). RAD51 accumulates at discrete foci on chromosomal DNA during meiotic prophase; and importantly, DNA damage triggers RAD51 foci formation (28). Our study showed high levels of RAD51 protein in multiple myeloma cells, indicating ongoing DNA damage in these cells (29). Treatment of multiple myeloma cells with SJB decreased RAD51 foci formation, assessed by immunofluorescent studies (Fig. 5D). Quantification of immunofluorescent data showed decreased RAD51 foci/nuclei in SJB-treated MM.1S cells (Fig. 5D). Together, these findings suggest that SJB abrogates HR mechanisms in multiple myeloma cells.
We next determined whether the cytotoxic effects induced by SJB are irreversible in drug-washout experiments. MM.1S cells were treated with SJB for a short interval (4 hours); cells were then washed to remove SJB and cultured in complete medium for another 24 hours, followed by analysis of cell viability. Results showed that short-term (4 hours) treatment with SJB triggers significant cytotoxicity in MM.1S cells (Fig. 5E). Although the extent of MM.1S cell death triggered by short-term (4 hours) exposure to SJB is less than long-term continuous treatment, it is sufficient to block DNA repair- and HR-related signaling pathways as well as induce cytotoxicity.
SJB induces the morphologic differentiation and maturation of multiple myeloma cells by blocking USP1-associated downstream signaling via ID proteins
Inhibitor of ID proteins belongs to a class of helix-loop-helix (HLH) family of transcriptional regulatory proteins, which consists of 4 members (ID1–ID4; ref. 30). ID proteins inhibit bHLH protein–mediated transactivation of genes involved in cellular differentiation pathways (30). Importantly, ID1 is downstream target of USP1 (31). Treatment of MM.1S cells with SJB decreased USP1 and ID1 protein levels (Fig. 6A). Besides ID1, SJB treatment also decreased other ID protein family members ID2, ID3, and ID4 (Fig. 6A). Treatment of resistant cell line Dox-40 slightly reduced USP1 levels and did not affect ID protein levels (Supplementary Fig. S5B). Importantly, GEP analysis of ID1 and ID2 proteins showed an inverse correlation with survival in patients with multiple myeloma (Supplementary Fig. S6).
Earlier studies showed that ID proteins are highly expressed in various cancer types including multiple myeloma and promote stem cell survival (30, 32). As SJB downregulates ID proteins, we examined whether SJB affects differentiation of multiple myeloma cells using Giemsa staining assays. A significantly decreased nuclear–cytoplasmic ratio was observed in SJB- versus DMSO/control-treated MM.1S cells (Fig. 6B). SJB-treated cells showed a more mature and differentiated state versus untreated cells: immature plasmablasts with large nuclei in untreated cells (Fig. 6B, left) versus condensed nuclei with rough chromatin pattern and a more prominent cytoplasm in SJB-treated cells (Fig. 6B, right). These data indicate that SJB induces plasma cell differentiation with prominent cytoplasm and a reduced nuclear–cytoplasmic ratio.
Previous studies showed that ID proteins contribute to chemoresistance and cancer stemness (32). For example, ID4 regulates the expression of stemness-associated Sox2 and Notch signaling in glioma cells (33, 34). We found that SJB-mediated inhibition of USP1/ID signaling led to downregulation of stemness-associated Sox-4/Sox-2/Notch-1/Notch-2 signaling proteins in multiple myeloma cells (Fig. 6C). In additionally, Notch signaling promotes the maintenance and proliferation of hematopoietic stem cells (35). Previous studies have reported presence of multiple myeloma clonogenic side population (MM-SP) with characteristic stem-like features (36). We therefore here evaluated the effect of SJB on MM-SP using flow cytometric analysis of Hoechst 33342–stained cells. SJB significantly (38% decrease) lowered the fraction of MM-SP (Fig. 6D). Verapamil and reserpine served as positive controls. Importantly, SJB exhibited cytotoxic activity against MM-SP.
Combining SJB with bortezomib, lenalidomide, pomalidomide, or the HDAC6 inhibitor ACY-1215 induces synergistic cytotoxicity
Preclinical studies laid the framework for clinical trials of bortezomib in combination with other anti–multiple myeloma agents (37). Isobologram analysis for synergistic anti–multiple myeloma activity showed that the combination of SJB and bortezomib triggers synergistic anti–multiple myeloma activity (Supplementary Fig. S7A). The mechanism of synergy between SJB and bortezomib may include enhanced accumulation of ubiquitylated proteins, which are not degraded and triggering of distinct cell death signaling pathways.
Protein degradation also utilizes HDAC-dependent aggressome autophagic cell death signaling pathway. Interestingly, the combination of SJB and HDAC6 inhibitor ACY-1215 triggers synergistic anti–multiple myeloma cytotoxicity (Supplementary Fig. S7B). Similarly, we found that combination of SJB and lenalidomide (Supplementary Fig. S7C) or pomalidomide (Supplementary Fig. S7D) triggered synergistic cytotoxicity against multiple myeloma cells. These combination studies were also performed using primary multiple myeloma tumor cells (Supplementary Fig. S8). These data provide the framework for scientifically informed clinical trials combining strategies to inhibit USP1 with standard-of-care anti–multiple myeloma agents.
We utilized multiple myeloma cell lines, patient cells, and co-culture models of multiple myeloma cells with BMSCs or pDCs, as well as genetic and biochemical and strategies, to validate DUB enzyme USP1 as a therapeutic target in multiple myeloma. USP1 is highly expressed in multiple myeloma cells compared with normal cells. A similar high USP1 expression was shown in melanoma, sarcoma, cervical, and gastric cancers (9–12). USP1 expression correlates with poor prognosis in patients with multiple myeloma.
RNA interference and biochemical strategies were utilized to determine the functional significance of USP1 in multiple myeloma. USP1-siRNA decreases multiple myeloma cell viability. We utilized a pharmacologic inhibitor of USP1 SJB that targets USP1 (30). The specificity of SJB was confirmed by various experiments: (i) we show that SJB potently and selectively blocks USP1 activity without inhibiting other DUBs (USP2/USP5/USP7/USP14/UCH37); (ii) SJB inhibited binding of USP1 with HA–Ub-VS probe, but it did not affect labeling of other DUBs with probe; and (iii) SJB-inhibited USP1, but not USP2 or USP7, triggered cleavage of ubiquitin tetramer chains.
A recent study demonstrated that USP1 inhibition induces apoptosis in leukemic cells (30). Here, we show that SJB triggers apoptosis in multiple myeloma cells in vitro, without reducing the viability of normal PBMCs. Moreover, SJB triggered more potent anti–multiple myeloma activity versus other available USP1 inhibitors (ML323, SJB2-043, pimozide, GW7647; ref. 31). Of note, we found that SJB retains its activity against multiple myeloma cell lines resistant to conventional and novel therapies. Distinct genetic backgrounds (38) and/or drug resistance characteristics may account for the differences in IC50 of SJB against multiple myeloma cell lines. Differential USP1 expression was observed among various multiple myeloma cell lines; however, we found no direct correlation between USP1 expression and sensitivity to SJB. Other factors, such as expression/function of USP1-binding partners (UAF1) or USP1 downstream signaling proteins, may impact the overall potency of SJB against multiple myeloma cell lines. Our data in multiple myeloma model are consistent with cytotoxic effects of USP1 inhibition observed in leukemic cells (31).
We next assessed whether SJB can overcome bortezomib resistance in multiple myeloma cells. For these studies, we utilized bortezomib-sensitive (ANBL6.WT) and -resistant (ANBL6.BR) multiple myeloma cell lines (39). SJB induced cytotoxicity in ANBL6.BR cells, which confirmed the ability of SJB to overcome bortezomib resistance. In addition, SJB was active against tumor cells from patients with multiple myeloma resistant to novel (bortezomib, lenalidomide), conventional (dexamethasone) agents. The bone marrow microenvironment (BMSCs and pDCs) promotes multiple myeloma cell growth, survival, and drug resistance; importantly, SJB triggers apoptosis in multiple myeloma cells even in the presence of the bone marrow milieu.
USP1–UAF1 regulates DNA damage response signaling pathways (9). We examined the effect of SJB on the 3 essential DNA repair pathways associated with USP1: (i) FA pathway (via FANCD2/FANCI), (ii) translesion synthesis (via PCNA), and (iii) DNA double-strand break repair through HR (via RAD51). USP1–UAF1 deubiquitylates FANCD2 and FANCI proteins; conversely, knockdown of USP1 results in elevated Ub-FANCD2 and Ub-FANCI levels, which in turn disrupts FA repair pathway (31). We found that USP1 inhibition by SJB increased the levels of both Ub-FANCD2 and Ub-FANCI in multiple myeloma cells. These findings suggest that SJB inhibits the FA repair pathway in multiple myeloma cells and that SJB may sensitize multiple myeloma cells to DNA-damaging anti–multiple myeloma therapies. Indeed, a recent study using a non–small cell lung cancer model showed that USP1 inhibitor ML323 sensitizes cisplatin-resistant cells to cisplatin (31, 40). In addition, USP1 plays a role in HR-mediated DNA repair, and USP1 blockade disrupts this pathway. Here, we found that multiple myeloma cells express high levels of HR-associated DNA repair protein RAD51. Treatment of multiple myeloma cells with SJB significantly decreased RAD51 foci formation, suggesting that SJB blocks activation of HR-mediated DNA repair pathways. These findings are consistent with our washout experiments showing that even short treatment duration (4 hours) of multiple myeloma cells triggers significant cytotoxicity (Fig. 5E). Together, the ability of SJB to block multiple DNA repair mechanisms likely contributes to its overall potent anti–multiple myeloma activity.
Besides DNA repair proteins, USP1 also promotes stabilization of ID (ID1–ID4) proteins (31). ID proteins are highly expressed in proliferating cells and promote stem cell–like characteristics in osteosarcoma cells (9, 12). Studies in multiple myeloma have shown that histone methytransferase protein MMSET promotes oncogenic transformation by transcriptional activation of ID1 protein expression (41). In our study, we found that SJB downregulates ID1, ID2, ID3, and ID4 proteins in multiple myeloma cells, suggesting that it may affect the differentiation state of multiple myeloma cells and/or multiple myeloma stem cell–like cell populations. A prior study showed that subpopulations of multiple myeloma with distinct morphology and immunophenotype are consistent with distinct phases of differentiation (42). In addition, the presence of immature multiple myeloma cells portends poor prognosis, and several studies have reported the survival advantage for patients with plasmacytic versus plasmablastic type multiple myeloma (42–44). In our study, we found that SJB induced multiple myeloma cell differentiation and maturation at low concentrations, evidenced by well-developed cytoplasm with reduced nuclear–cytoplasmic ratio. These findings suggest that USP1 inhibition may represent a potential differentiation therapy in multiple myeloma.
The modulation of ID proteins by USP1 inhibitor SJB has further biologic implications and clinical applications. Prior studies have shown that ID proteins function as master regulators of stem cell identity (30). Specifically, loss of ID proteins affects both the self-renewal and the tumor-initiating capacity of cancer stem cells in colorectal cancer and gliomas (33, 45). Conversely, ID4 relieves miR-9–mediated suppression of SOX-2 and enhances stem cell in gliomas (46); and overexpression of ID4 induces activation of Notch signaling and stem cells in gliomas (34). In our study, we showed that SJB inhibits SOX and Notch signaling in multiple myeloma cells. Notch1 and Notch2 are highly expressed in primary multiple myeloma cells and their expression increases with progression from MGUS to multiple myeloma (35). Moreover, overexpression of Notch receptors/ligands is integral to multiple myeloma stem cell self-renewal and proliferation (35). Finally, SOX2 expression is a key feature of clonogenic multiple myeloma cells (47, 48). Importantly, we here found that inhibition of USP1/ID/Notch/Sox2 pathway by SJB decreased the percentage of clonogenic MM-SP. Taken together, our data show that: (i) SJB triggers immature plasma cells to differentiate, mature, and undergo apoptosis and (ii) SJB decreases multiple myeloma stem cell–like clonogenic population (MM-SP).
Finally, we also examined whether SJB enhances the anti–multiple myeloma activity of other agents. Isobologram analysis for synergistic anti–multiple myeloma activity demonstrated that the combination of SJB with bortezomib, HDAC6i ACY1215, lenalidomide, or pomalidomide triggers synergistic anti–multiple myeloma activity. The mechanism of synergy between SJB and other anti–multiple myeloma agents is associated with activation of distinct apoptotic signaling pathways as well as blockade of DNA repair mechanisms. Future studies for SJB will be done using in vivo assays. Our preclinical data support clinical investigation of USP1 inhibitors in multiple myeloma.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Conception and design: D.S. Das, D. Chauhan
Development of methodology: D.S. Das, A. Das
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): D.S. Das, A. Das, A. Ray, Y. Song, N.C. Munshi, K.C. Anderson
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): D.S. Das, A. Das, M.K. Samur, K.C. Anderson
Writing, review, and/or revision of the manuscript: D.S. Das, D. Chauhan, K.C. Anderson
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): D.S. Das, A. Ray, Y. Song, K.C. Anderson
Study supervision: D.S. Das, D. Chauhan
K.C. Anderson is an American Cancer Society Clinical Research Professor.
This study was supported by NIH SPORE grant nos. P50100707, R01CA207237, and RO1 CA050947.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.