BCL2-BH4 antagonist BDA-366 suppresses human myeloma growth

Multiple myeloma (MM) is a heterogeneous plasma cell malignancy and remains incurable. B-cell lymphoma-2 (BCL2) protein correlates with the survival and the drug resistance of myeloma cells. BH3 mimetics have been developed to disrupt the binding between BCL2 and its pro-apoptotic BCL2 family partners for the treatment of MM, but with limited therapeutic efficacy. We recently identified a small molecule BDA-366 as a BCL2 BH4 domain antagonist, converting it from an anti-apoptotic into a pro-apoptotic molecule. In this study, we demonstrated that BDA-366 induces robust apoptosis in MM cell lines and primary MM cells by inducing BCL2 conformational change. Delivery of BDA-366 substantially suppressed the growth of human MM xenografts in NOD-scid/IL2Rγnull mice, without significant cytotoxic effects on normal hematopoietic cells or body weight. Thus, BDA-366 functions as a novel BH4-based BCL2 inhibitor and offers an entirely new tool for MM therapy.


INTRODUCTION
Multiple myeloma (MM) is a common malignancy characterized by excessive proliferation of abnormal clonal plasma cells in the bone marrow (BM). In recent years, substantial progress has been made in the development of effective therapeutic approaches to treat MM, including autologous and allogeneic stem cell transplants, monoclonal antibodies, proteasome inhibitors, immunomodulatory drugs and recent chimeric antigen receptor-expressing T cells [1][2][3][4][5][6][7]. These approaches have significantly improved clinical outcomes and prolonged patient survival. However, due to the high phenotypic heterogeneity and plasticity of MM cells [8], none of these agents is curative, and the disease eventually relapses and becomes refractory to treatment [9]. Thus, the development of innovative therapeutics that can advance the field toward the cure of MM is very attractive.
The N-terminal amphipathic helix BH4 domain (aa6-31) of BCL2 is known as a survival domain and is essential for the anti-apoptotic function of BCL2 protein [25,26]. Removal of the BH4 domain by caspaseactivated cleavage or genetic mutation converts BCL2 from an anti-apoptotic molecule into a death effector [27]. The BCL2 BH4 domain has been shown to directly bind and inhibit inositol 1,4, 5-trisphosphate receptors (IP3R) [28], which serve as the main intracellular Ca2 + -release channel of the ER [29], a decoy peptide binding the BH4 domain by blocking the interaction between BCL2 and IP3R induces Ca 2+ -mediated apoptosis in MM cells [30]. We recently identified a small molecule, BDA-366, which is a member of the anthraquinone (1,4-diamino-9,10anthraquinone) class, as a selective BCL2 BH4 antagonist. BDA-366 induces a conformational change in the BCL2 molecule that converts it to a death protein, and inhibits lung cancer growth in vitro and in vivo [31,32]. In this study, we report that BDA-366 has potent anti-myeloma activity, inducing robust apoptotic death of MM cells in vitro and suppressing the growth of human MM cells in vivo.

BDA-366 suppresses human myeloma cell growth in vitro
BDA-366 is a small molecule antagonist that specifically binds the BCL2 BH4 domain, and has the ability to convert anti-apoptotic BCL2 into a pro-apoptotic death molecule [31]. To test the effect of the BH4 antagonist on myeloma cell growth, we treated human MM RPMI8226 and U266 cell lines with increasing concentrations of BDA-366 ( Figure 1A) for 48 hours. FACS analyses of the treated MM cells showed that BDA-366 induced robust apoptosis in a dose-dependent manner in both RPMI8226 ( Figure 1B and 1C) and U266 cells ( Figure 1B and 1D). The percentages of apoptotic cells in RPMI8226 and U266 cell lines treated with 0.5μM of BDA-366 were 84.2% and 60.6% on average, respectively ( Figure 1B), which were significantly higher than in cells treated with lower concentrations (i.e. 0.1μM or 0.25μM), or with DMSO control ( Figure 1B). DMSO treatment had little apoptotic effect on the two cell lines ( Figure 1B). Consequently, BDA-366 treatment markedly reduced the absolute live cell number in both RPMI8226 ( Figure  1E) and U266 ( Figure 1F) cell lines after 48-hour culture in comparison with DMSO control, with only 0.1 fold (RPMI8226, Figure 1E) and 0.5 fold (U266, Figure 1F) of the initial live cell numbers remaining in the groups treated with 0.5μM BDA-366. In contrast, DMSO treated groups showed an increase in absolute live cell number of more than 4-fold (RPMI8226, Figure 1E) and 3-fold (U266, Figure 1F) after 48-hour culture.

BDA-366 induces conformational change and reduces the phosphorylation of BCL-2
To examine whether BDA-366 induces BCL2 conformational change by exposure of the BCL2 BH3 domain in MM cells, we treated human RPMI8226 and U266 cells with BDA-366 at concentrations of 0.25 and 0.5μM, and stained the cells with anti-BCL2 BH3 domain specific antibody [31]. Flow cytometry assay demonstrated that BDA-366 treatment indeed induced the conformational change in BCL2 molecule that promotes the exposure of its BCL2 BH3 domain with significantly higher anti-BH3 fluorescence intensity in the treated RPMI8226 (Figure 2A-2B) and U266 ( Figure 2C-2D) cells. It was previously demonstrated that phosphorylation of BCL2 at Ser70 is required to stabilize the anti-apoptotic activity of BCL2 and its function as a pro-survival molecule [33][34][35]. To examine whether BDA-366-induced BCL2 conformational change reduces BCL2 phosphorylation, we performed a Western blot on BDA-366-treated MM cells with an anti-pBCL2 (Ser70) specific antibody [31]. Western blotting showed that both RPMI8226 and U266 cells have a high baseline level of BCL2 phosphorylation (

BDA-366 induces apoptosis in human primary myeloma cells
To test the effect of the small molecule BDA-366 as a BCL2 BH4 antagonist on the apoptosis of primary myeloma cells, we treated the whole bone marrow (BM) cells harvested from myeloma patients with BDA-366. The patients were either untreated or had relapsed/ refractory disease, with varying genetic backgrounds (Table 1). Primary myeloma cells were gated on the CD45 -CD38 + CD138 + population. FACS analyses of the treated BM cells demonstrated that BDA-366 treatment at concentrations of both 0.25 and 0.5 μM robustly induced the apoptosis of primary myeloma cells within a 24-hour culture period ( Figure 4A), with 32.4% and 63.4% apoptotic cells on average respectively ( Figure  4B). In control primary myeloma cells without BDA-366 treatment, there was a baseline level of 4.7% apoptotic cells. In non-MM cells, BDA-366 treatment did not cause significant apoptotic death in comparison with control treatment with DMSO ( Figure 4C-4D). To examine www.impactjournals.com/oncotarget

Administration of BDA-366 inhibited human myeloma cell growth in vivo
To test whether the efficacy of BDA-366 in the suppression of human MM cell growth in vivo reflects that seen in vitro, we implanted RPMI8226 or U266 myeloma cells into NSG mice, and treated the mice with BDA-366 at a dose of 10mg per kg mouse weight by intraperitoneal injection every 2 days from day 4-12 ( Figure 6A). Measurement of myeloma tumors showed that delivery of BDA-366 into the mice significantly inhibited both RPMI8226 ( Figure 6B) and U266 ( Figure 6C) myeloma cell growth in vivo. Control DMSO treatment had no effect on the tumor growth ( Figure 6B-6C). In parallel, we also assessed the cytotoxicity of BDA-366 in naive NSG mice following 5 treatment doses. Monitoring mouse body weight showed that BDA-366 treatment had no significant toxic side effects in the mice; the three groups of mice treated with the antagonist, DMSO, or no treatment had similar weights on day 23 ( Figure 6D). Peripheral blood cell counts demonstrated that there were similar levels of white blood cells, red blood cells and platelets in the blood collected from the three groups of mice ( Figure 6E), confirming that BDA-366 treatment had no significant toxic side effects on the murine hematopoietic system at a therapeutic dose.

DISCUSSION
In this study, we demonstrate that the non-peptidic small molecule BDA-366 functions as a novel BCL2 inhibitor and effectively induces robust apoptotic death of human MM cells. To our knowledge, this is the first report that BH4 antagonist-induced BCL2 conformational change leads to the suppression of MM growth.
The mechanistic action of current BH3 mimetics for cancer therapy is designed to block the binding between BCL2 and BIM, BAX and BAK. Utilizing BDA-366 as a BH4-based BCL2 inhibitor to treat MM is distinct from this BH3 mimetics strategy, since BDA-366 not only blocks the activities of the BH4 domain, but also changes the conformation of BCL2 in the hydrophobic groove. During BCL2 conformational change, the hydrophobic pocket undergoes a large-scale realignment, causing the exposure of the BH3 domain [36], which can be detected by a BCL2 BH3-specific antibody [31]. Indeed, using anti-BCL2 BH3-specific antibodies, we were able to White blood cells (WBC), red blood cells (RBC) and platelets (PLT) in peripheral blood were collected from the NSG mice on day 23 before sacrifice, and profiled on an automatic blood cell counter. Data in (E) were from five mice per group, and presented as mean± SEM. www.impactjournals.com/oncotarget show this conformational change in BCL2 in MM cells treated with BDA-366, confirming the induction of BCL2 conformational change by the BH4 antagonist.
BCL2 phosphorylation at Ser70 promotes tumor cell survival [33,35]. In this study, we found that BDA-366 treatment significantly reduced BCL2 phosphorylation at Ser70, indicating an additional mechanistic effect of BDA-366 on BCL2. BDA-366 was previously shown to disrupt the noncanonical BCL2-BH4-BAX interaction, inhibit the binding of BCL2 with IP3R and enhance Ca 2+ release and apoptosis in tumor cells [31]. BCL2 BH4 domain also binds to calcineurin, CED-4, HIF-1α, c-Myc, paxillin, Raf-1, Ras and VDAC, promoting tumorigenesis or tumor survival and contributing to both anti-apoptosis and pro-tumorgenesis activities of BCL2 [37]. BDA-366induced suppression of MM growth may also result from the inhibition of those functions of the BH4 domain.
MM is considered a heterogeneous disease in terms of malignant plasma cell subclones and its sensitivity to inhibition by anti-apoptotic BCL2 family proteins [38]. Using a BH3-profiling method [39], RPMI8226 cells were shown to be sensitive to the BCL2 inhibitor ABT-737 (a BCL2 BH3 mimetic), but insensitive to ABT-199, while U266 cells were resistant to the BH3 mimetics, since ABT-737 treatment had no significant suppressive effect on U266 cells [38]. In this study, we observed that RPMI8226 cells were very sensitive to BDA-366 treatment in cell culture. However, we also found that BDA-366 treatment potently induced apoptosis in U266 cells, suggesting that the BH4 antagonist employs distinct actions in malignant plasma cells compared to the BH3-based mimetics such as ABT-737 and ABT-199. BDA-366-induced robust apoptosis in primary myeloma cells from untreated patients or patients with relapsed/refractory myeloma with varying genetic background further indicates that BDA-366 as a novel BCL2 BH4 antagonist possesses potent therapeutic effect for relapsed/refractory myeloma. We also evaluated the anti-MM effect of BDA-366 in NSG immune deficient mice xenografted with RPMI8226 or U266 cells. Consistent with our in vitro results, BDA-366 treatment for 8 days efficiently suppressed human MM growth in the mice. Analyses of cytotoxicity of BDA-366 revealed that the BH4 antagonist had minimal side effects in terms of both body weight and the hematopoietic cellular system in NSG mice. Collectively, our data demonstrate that BDA-366, as a novel BH4-based BCL2 inhibitor and an inducer of BCL2 conformational change could offer an entirely new tool for MM therapy.

BDA-366
Small molecule BDA-366 (NSC639366) [31] was obtained from the Drug Synthesis and Chemistry Branch, Developmental Therapeutic Program, Division of Cancer Treatment and Diagnosis, National Cancer Institute (Bethesda, MD), dissolved in DMSO as stock and diluted with PBS for in vitro and in vivo experimental use.

Cell culture
Human myeloma cell lines RPMI8226 and U266 (obtained from the American Type Culture Collection) were cultured and maintained in RPMI-1640 medium (Hyclone, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS) (Gibco, USA) and penicillin / streptomycin antibiotics at 37 o C in a 5% CO 2 incubator. For cell apoptosis assay, U266 or 8226 cells (10,000 cells/well) were cultured in 96-well plates with 200ul/ well of RPMI-1640 medium in the presence of different concentrations of BDA-366 (0, 0.1, 0.25, 0.5μM) for 48 hours.

Patient sample processing
BM aspirates from patients with MM (n=5) ( Table 1) were diluted to 20ml with 1× PBS, and processed as previously described [38]. White blood cells in the marrow were isolated with lymphocyte separation medium (Mediatech Inc., Manassas, VA), and cultured (10 6 cells/ml) in complete RPMI medium supplemented with BDA-366 (0, 0.25, 0.5μM) for 24 hours. Informed consent was obtained from all human subjects. The use of human samples was approved by the Institutional Review Board of Emory University.

Western blot
RPMI8226 or U266 cells (10 6 /ml) were harvested 8 hours after BDA-366 treatment (0, 0.1, 0.25, 0.5μM), and washed with cold PBS. Total proteins were extracted from the cells lysed with protein lysate buffer supplemented with protease and phosphatase inhibitors as previously described [40]. The proteins were separated on 10% SDS polyacrylamide gels and electrophoretically transferred to polyvinylidene difluoride membrane (Millipore, USA). The membranes were treated with rabbit anti-pBCL2 (Ser70), BCL2 (1:100), and anti-β-actin antibodies (1:1000 dilution) (Cell Signaling Technology, USA); then incubated with HRP-conjugated secondary antibodies before being subjected to enhanced chemiluminescent (ECL) detection on an ECL machine (Pierce, USA). The blots were scanned and the band density was measured using the Quantity One imaging software.

Blood counting
Peripheral blood was collected from immune deficient NOD-scid IL2Rgamma null (NSG) mice directly to Microvette tubes containing EDTA-tripotassium salt (Sarstedt AG & Co, Nümbrecht, Germany), and subjected to blood cell counting on a Vet-ABC Animal Blood Counter (Scil, Gurnee, IL, USA) using mouse-specific software provided by the company.

Myeloma xenografts in NSG mice
Human RPMI8226 or U266 myeloma cells (5×10 6 cells/mouse) in 100μl of MatriGel solution (StemCell Technologies, Inc, Bedford, MA) were subcutaneously implanted in the right rear flanks of NSG immune deficient mice [42]. The mice were then treated with BDA-366 (10mg/Kg/day) on days 4, 6, 8, 10 and 12 after implantation. On day 15, myeloma tumors in the three groups of mice were measured with a digital caliper, and tumor size was calculated based on the length and width (micrometer) of the tumor: volume=length×(width) 2 /2. The mice used were all female (6-8 weeks old) purchased from Jackson Laboratory (Bar Harbor, ME), and maintained in compliance with an IACUC protocol approved by Emory University (Atlanta, USA).

Statistical analysis
Data are shown as mean ± SEM. P values were calculated using the one-way analysis of variance test. P value of less than 0.05 was considered significant (*: P<0.05; **: P<0.01; *** P<0.001).