Rafoxanide, an organohalogen drug, triggers apoptosis and cell cycle arrest in multiple myeloma by enhancing DNA damage responses and suppressing the p38 MAPK pathway

A B S T R A C T
Rafoxanide is used in veterinary medicine for the treatment of fascioliasis. We previously repositioned the drug as the inhibitor of B-Raf V600E, but its anti-tumor effect in human cancer has never been reported. In this study, we investigated the effects of rafoxanide in multiple myeloma (MM) in vitro and in vivo. We found that rafoxanide inhibited cell proliferation and overcame the protective effect of the bone marrow (BM) microenvironment on MM cells. Rafoxanide induced cell apoptosis by reducing mitochondrial membrane potential (MMP) and reg- ulating the caspase pathway, while having no apparent toxic effect on normal cells. Rafoxanide also inhibited DNA synthesis and caused cell cycle arrest by regulating the cdc25A-degradation pathway. In addition, rafox- anide enhanced the DNA damage response by up-regulating the expression of γ-H2AX, and suppressed activation of the p38 MAPK pathway by down-regulating p38 MAPK phosphorylation and Stat1 phosphorylation. Rafoxanide treatment inhibited tumor growth, with no significant side effects, in an MM mouse xenograft model. Combination of rafoxanide with bortezomib or lenalidomide significantly induced synergistic cytotoxicity in MM cells. Finally, rafoxanide had anti-proliferation effect on both wild type and B-Raf V600E mutated MM cells. And the weaker anti-MM activity of rafoxanide than vemurafenib may indicate other potential mechanisms besides targeting B-Raf V600E mutation. Collectively, our results provide a rationale for use of this drug in MM treatment.

1.Introduction
Multiple myeloma (MM), a clonal plasma cell malignancy that ac- counts for approximately 10% of all hematological cancers, is asso- ciated with characteristic clinical complications, including skeletal de- struction, renal impairment, anemia, and hypercalcemia [1,2].Approximately 120,000 new cases of myeloma are diagnosed annually worldwide, an incidence rate that is expected to rise along with the aging world population [3]. Over the last few decades, proteasome inhibitors (bortezomib) and immunomodulatory drugs (thalidomide and lenalidomide) have greatly prolonged the survival of MM patients [4]. However, even though the prognosis of MM patients hassignificantly improved after the introduction of novel treatment agents, it remains incurable, with a high rate of relapse and refractory disease [5]. Thus, novel treatment agents with different mechanisms need to be identified.Rafoxanide, N-[3-chloro-4-(4-chloro-phenoxy)phenyl]-2-hydroxy- 3,5-diiodobenzamide, is an anthelmintic used in veterinary medicine for the treatment of fascioliasis in cattle and sheep [6]. Moreover, ra- foxanide is also active against gastrointestinal nematodes and against nasal bot fly [7]. Matsubara et al. used an improved thyroid hormone reporter assay and found that rafoxanide has thyroid hormone-like activity [8]. Alamri et al. have shown that rafoxanide inhibits SPAK (STE20/SPS1-related proline/alanine-rich kinase) and OSR1 (oxidative- stress-responsive kinase 1) kinases [9]. Although rafoxanide has been used extensively in veterinary medicine, there is little information available concerning its biological effects on humans.

However, a pre- vious study reported the therapeutic use of rafoxanide in a child with fascioliasis, which is the first report in the literature concerning the usage of rafoxanide in humans [10]. Meanwhile, a recent study into the effects of rafoxanide on two human cell lines indicated that these ve- terinary anthelmintics have potential for the treatment of human dis- ease [11]. With molecular docking and bioassay, we demonstrated that rafoxanide, an organohalogen drug, is a potent B-Raf V600E inhibitor [12].Raf, a serine/threonine kinase, is a component of the mitogen-ac-tivated protein kinase (MAPK)/extracellular signal-regulated kinase (ERK) pathway, and it has attracted attention given the success of targeted therapy in many human cancer [13]. B-Raf has been found to be mutated in approximately 4% of MM cases, with the mutation of B- Raf V600E being the most common [14]. Although the B-Raf mutation is rare in MM, a case report showed that combination treatment with the small-molecular B-Raf V600E inhibitor vemurafenib and cobime- tinib achieved a rapid and sustained response in a young patient with highly resistant and rapidly progressing MM haboring the B-raf V600E mutation [15]. The B-Raf V600E mutation causes constitutive activa- tion of the Ras-Raf-MEK-ERK (RAS) and MAPK signaling pathway, further stimulating cell proliferation, differentiation and survival [16]. Therefore, in this study, we investigated the antitumor activities of rafoxanide in MM cell lines both in vitro and in vivo. Interestingly, we found that rafoxanide could induce significant cytotoxicity in MM cell lines in vitro and inhibit tumor growth in an MM mouse xenograft model in vivo. Furthermore, we examined the molecular mechanism of anti- MM activities induced by rafoxanide. In addition, we constructed B-Raf V600E overexpressed MM cells by lentivirus transduction, and B-Raf V600E knocked down MM cells by siRNA transduction to further ex- plore the effect of rafoxanide.

2.Materials and methods
H929, MM1S, U266 and BMSC line HS-5 were purchased from the American Type Culture Collection (ATCC) (Manassas, VA, USA). ARP1, OCI-MY5, OPM2 and RPMI8226 were kindly provided by Fenghuang Zhan (Department of Internal Medicine, University of Iowa, Iowa City, IA, USA). The bortezomib-resistant H929R was gained by our culture. HS-5 cells were stably transduced with GFP expressing lentivirus, which were kindly provided by professor Ping Wang (Tongji University School of Medicine, Shanghai, China). Primary cells were isolated from MM patient BM samples and peripheral blood of healthy donors using lymphoprep (Stemcell Technologies, Vancouver, BC, Canada) by Ficoll- Hypaque density gradient centrifugation. Then the bone marrow mononuclear cells (BMMCs) and peripheal blood mononuclear cells (PBMCs) were obtained. CD138+ MM cells in BMMCs were then dis- tinguished using the APC conjugated anti-CD138 (BioLegend, San Diego, CA, USA) according to the manufacturer’s instructions. All pri- mary samples were obtained from MM patients and healthy donorsafter informed consent was obtained in accordance with the Declaration of Helsinki protocol and approved by the institutional review board of The Tenth People’s Hospital of Shanghai, Tongji University.All MM cell lines, primary CD138+ MM cells and PBMCs were cultured in RPMI-1640 medium (Gibco, Carlsbad, CA, USA) containing 10% fetal bovine serum (FBS; Gibco, BRL, USA) and 1% penicillin- streptomycin (PS; Gibco, Carlsbad, CA, USA) at 37 °C, 5% carbon-di- oxide. Human BMSC line HS-5 was cultured in DMEM/HIGH GLUCOSE medium (Gibco, Carlsbad, CA, USA) containing 10% FBS and 1% PS at 37 °C, 5% carbon-dioxide.

Culture medium was changed every other day.Rafoxanide was purchased from J&K Scientific Ltd (Shanghai, China). Recombination humaninterleukin (IL)-6 and recombination human insulin-like growth factor (IGF)-1 were purchased from R&D system (Minneapolis, MN, USA). IL-6 (25 ng/mL) and IGF-1 (25 ng/mL) were used to induce MM cell growth and proliferation. Pan-caspase inhibitor, Z-VAD-FMK, and vemurafenib were purchased from Selleck Chemicals (Houston, TX, USA). Bortezomib and lenalidomide were purchased from Sigma (Sigma-Aldrich, St, Louis, MO, USA).MM cell lines were seeded into 96-well plates in 100 μL complate media at a density of 2 × 105 cells/mL and treated with rafoxanide (0, 5, 10, 20, 40, or 80 μM) or vemurafenib (0, 5, 10, 20, 40, or 80 nM) for indicated time. For detecting the effect of this drug on MM cells at the presence of BM microenvironment, H929 and ARP1 cells were cultured with rafoxanide (0, 10, or 20 μM) alone or in the presence of BMSC (HS- 5), or cell cytokines (IL-6 and IGF-1) for 48 h. H929 and ARP1 cells were cultured with rafoxanide in combination with bortezomib or le- nalidomide for 48 h. Cell viability was evaluated by Cell Counting Kit-8 (CCK8, Dojindo, Kumamoto, Japan) according to the manufacturer’s instructions. Half maximal inhibitory concentration (IC50) values and combination index (CI) were evaluated by using CalcuSyn software, Version 2.0 (CI < 1 indicates synergistic effect).

MM cell lines and PBMCs were cultured in 24-well plates at a density of 2 × 105 cells/mL and treated with rafoxanide (0, 10, or 20 μM) alone or in the presence of BMSC (GFP-HS-5), or Z-VAD-FMK (50 μM) for indicated time. Then cells were collected and washed in PBS, and incubated at room temperature for 15 min with an Annexin-V/ propidium iodide (PI) (BD Pharmingen, Franklin Lakes, NJ, USA) double staining. BMMCs were stained with APC conjugated anti-CD138 (BioLegend, San Diego, CA, USA), Annexin-V, and 7-aminoactino- maycin D (7-AAD, KeyGen Biotech, Nanjing, China). Then the staining was performed according to the manufacturer's protocol. Finally, cell apoptosis were detected by a BD FASCCanto II flow cytometer (BD BioScience, San Jose, CA, USA).H929 and ARP1 cells were seeded in 24-well plates at a density of 2× 105 cells/mL and treated with rafoxanide (0, 10, or 20 μM) for 24 h. Then the 5-ethynyl-2′-deoxyuridine (EdU) incorporation analysis wasperformed according to the manufacturer's instruction by using EDU kit(RiboBio, Guangzhou, China). MM cells were exposed to 50 μM EDU for 2 h at 37 °C, collected and washed with PBS and fixed with 4% paraf- ormaldehyde, followed by permeabilization with 0.5% Triton-100X (Sigma-Aldrich, St. Louis, MO, USA).

Then cells were incubated with azide-conjugated Alexa Fluor 567 dye and Hoechst 33342 for 30 min respectively and visualized under a confocal laser scanning microscope(LSM710; Zeiss, Germany) and Zen2011 software (Carl-Zeiss, Jena, Germany).MM cells were cultured in 24-well plates at a density of 2× 105 cells/mL and treated with rafoxanide (0 or 20 μM) for indicated time. Then cells were collected and washed in cold PBS, and fixed in 70% ethanol at −20 °C overnight. Next day, the fixed-cells were wa- shed in PBS and incubated with 300 μL PI/RNase staining buffer (BD Pharmingen, Franklin Lakes, NJ, USA) at room temperature for 15 min and analyzed by flow cytometry.The loss of mitochondrial membrane potential (MMP) occurring in apoptosis was detected by flow cytometry using JC-1 kit (Beyotime Insititute of Biotechnology, Haimen, China) according to the manu- facturer's instructions. Briefly, MM cells were cultured in 24-well plates and treated with rafoxanide (0, 10, or 20 μM) for 48 h. Then cells were collected and washed in PBS, and incubated with JC-1 working solution at 37 °C for 20 min. Finally, cells were washed in JC-1 staining buffer and analyzed by flow cytometry.MM cells were cultured in 24-well plates at a density of 2× 105 cells/mL and treated with rafoxanide (0, 10, or 20 μM) for 24 h.

Then cells were collected, washed in PBS, and fixed in 4% paraf- ormaldehyde for 20 min. After washing in PBS, cells were blocked for 30 min with 5% BSA and ruptured with 0.2% Trinton-100X, and in- cubated γ-H2AX (1:200 dilution; Abcam, Cambridge, MA, USA) at 37 °C for 1 h in the incubator. Then cells were washed in PBS and followed by immunofluorescence detection using a donkey anti-rabbit antibody conjugated with fluorochrome Alexa FluorH 488 (1:400 dilution; Jackson ImmunoResearch Laboratory, USA) for 1 h in the incubator at 37 °C. After washing in PBS, cells were incubated with 49, 6-diamidino- 2-pheny-lindole (DAPI) (Sigma-Aldrich, St. Louis, MO, USA) for 10 min and fluorescence analysis was performed using a confocal laser scan- ning microscope and Zen2011 software.MM cells were collected, washed with cold PBS, and lysed with lysis buffer (100 mM Tris-HCl, pH 6.8, 4% SDS, 20% glycerol) on ice for 30 min. Protein concentrations were detected using the BCA method (Beyotime Insititute of Biotechnology, Haimen, China). Proteins were electrophoresed using sodium dodecyl sulfate/polyacrylamide gel electrophoresis (SDS-PAGE Bio-Rad, CA, USA) and transferred electro- phoretically to membranes. The membranes blocked with 5% non-fat milk at room temperature for 1 h and were incubated with primary antibodies overnight at 4 °C. The next day, membranes were washed and incubated with the appropriate Fluorescence-conjugated secondary antibodies at room temperature for 1 h.

Finally, membranes were wa- shed and developed using the Odyssey two-color infrared laser imaging system (LICOR, Lincoln, NE, USA). Primary antibodies were as follow: anti-cleaved-caspase 3, anti-cleaved-caspase 8, anti-caspase 9, anti-Bcl- 2, anti-Bcl-xl, anti-Bax, and anti-B-Raf were purchased from Cell Signaling Technology (CST, Beverly, USA); anti-CDK4, anti-CDK6, anti- cyclinD1, anti-cdc25A, anti-phospho-CHK2 (Thr68), anti-phospho- H2AX (γ-H2AX), p38 MAPK, anti-phospho-p38 MAPK, anti-phospho- Stat1, anti-phospho-ERK1/2, and anti-phospho-JNK were purchased from Abcam (Cambridge, MA, USA); anti-β-actin was purchased from Sigma (Sigma-Aldrich, St. Louis, MO, USA).Male BALB/C nude mice (5 weeks old) were purchased from Shanghai Laboratory Animal Center (Shanghai, China). Mice were maintained under a 12 h light–dark cycle at 22 °C, provided water ad libitum, fed standard laboratory chow, and allowed to acclimatize for a minimum of one week. 2.5 × 106 H929 cells in 100 μL serum-free cul-ture medium were inoculated subcutaneously into the right flank of each nude mice. When the tumors were measurable, mice were ran- domly divided into control and treatment groups: the vehicle-treated group (DMSO and saline) and 15 mg/kg rafoxanide-treated group (dissolved in DMSO and saline solution). Mice were administrated of DMSO, saline, and with or without rafoxanide by intraperitoneal in- jection every two days for a total of 14 days.

Tumor size and body weight were measured every other day. Tumor volume = 0.5 (a × b2) where a is the long diameter and b is the short diameter. At the end of the treatment, mice were sacrificed. Tumors were fixed with 4% par- aformaldehyde for 24 h. Then hematoxylin and eosin (HE), Ki67, cleaved-caspase 3, TUNEL, γ-H2AX, and p-p38 MAPK staining were performed. All animal-related procedures in vivo were approved by the Animal Care and Use Committee of The Tenth People's Hospital of Shanghai, Tongji University. This study was also approved by the Science and Technology Commission of Shanghai Municipality (ID:SYXK 2007–0006) under the permit number 2011-0111.The tumors were fixed in 4% paraformaldehyde for 24 h, dehy- drated via a graduated ethanol series, and embedded in paraffin blocks. All sections (5 μm) were dewaxed in xylene, hydrated through an up- graded ethanol series, and stained with HE. Morphological changes were examined under a light microscope at a magnification of × 200 by three pathologists who were unaware of the original specimens (CTR 6000; Leica, Wetzlar, Germany).Tumor sections (5 μm) were dewaxed in xylene, hydrated through an upgraded ethanol series. For antigen retrieval, slides were boiled in EDTA (1 mM, pH 8.0) for 15 min in a microwave oven. Endogenous peroxidase activity was quenched with 0.3% hydrogen peroxide solu- tion for 10 min at room temperature. After rinsing with PBS, slides were blocked with 5% BSA for 30 min.

Slides were subsequently incubated with a polyclonal antibody against Ki-67 (1:200 dilution), cleaved- caspase 3 (1:200 dilution), γ-H2AX (1:200 dilution), and p-p38 MAPK (1:200 dilution) overnight at 4 °C respectively. Antibody binding was detected with an Envision Detection Kit, Peroxidase/DAB, Rabbit/ Mouse (Gene Tech, Shanghai, China). Sections were counterstained with hematoxylin. Positive areas stained with Ki-67, and cleaved-cas- pase 3 were observed in all specimens under a microscope at a mag- nification of × 400 by three pathologists who were unaware of spe- cimen origins.Tumor sections (5 μm) were dewaxed in xylene, hydrated through an upgraded ethanol series, and detected the cell apoptosis via using TUNEL kit (Roche, Basel, Switzerland) according to the manufacturer's protocols. Cell apoptosis was evaluated by use a light microscope at a magnification of × 400 by three pathologists who were unaware of the original specimens.MM cells (2 × 105 cells) were prepared and infected at amultiplicity of infection (MOI) of 50 with control, or B-Raf V600E overexpression lentiviruses (GeneChem, Shanghai, China) according to the manufacturer's instructions, and western blot was used to validate the efficiency of B-Raf V600E overexpression.Three small interfering RNAs (siRNAs) targeting human B-Raf were designed and constructed by RiboBio (Guangzhou, China). The siRNA sequences (5′–3′) were as follows: GGAGCATAATCCACCATCA, GGAGAATGTTCCACTTACA, and CAAGCTAGATGCACTCCAA. The negativecontrol siRNA and B-Raf siRNAs were transfected into B-Raf V600E-OE MM cells using Lipofectamine 3000 transfection reagent (Invitrogen, Carlsbad, CA, USA). The first siRNA sequence was used to construct the knockdown of B-Raf V600E, and western blot was used to validate the efficiency of B-Raf V600E knockdown.The data were expressed as mean ± standard deviation (SD). Student's t-test was performed as appropriate using SPSS v20.0 software (IBM, Armonk, NY, USA). Significance was established at a p value of less than 0.05.

3.Results
We investigated the effect of rafoxanide in MM cell lines (H929, H929R, ARP1, OCI-MY5, MM1S, RPMI8226, OPM2, and U266) usingthe CCK8 assay. Our data showed that the viability of MM cells was significantly inhibited after treatment with rafoxanide for 48 h (Fig. 1A). The half-maximal inhibitory concentration (IC50) of rafox- anide at doses of 5, 10, 20, 40, and 80 μM was determined in each of these MM cell lines using CalcuSyn software, Version 2.0, and the re- sults were 19.2 μM (H929), 40.1 μM (H929R), 19.3 μM (ARP1), 27.8 μM (OCI-MY5), 47.2 μM (MM1S), 27.3 μM (RPMI8226), 22.4 μM (OPM2),and 21.6 μM (U266). Interestingly, we found that rafoxanide induces a dose- and time-dependent cytotoxicity in H929, ARP1, and OCI-MY5 cells (Fig. 1B–D). Additionally, rafoxanide could induce significant cy- totoxicity in lymphoma cells and other human cancer cells via detecting the cell viability using the CCK8 assay (Supplementary Fig. S1). Thesefindings suggested that rafoxanide has potent anticancer activity.MM is a BM microenvironment-dependent malignancy, and BMSC, IL-6 and IGF-1 promote MM growth, migration, and survival [17,18]. We investigated whether rafoxanide could protect against the effect of the BM microenvironment on MM cells. H929 and ARP1 cells were cultured with rafoxanide (0, 10, or 20 μM) alone or in the presence of BMSC, IL-6 and IGF-1 for 48 h.

Our results showed that BMSC and cy- tokines (IL-6 and IGF-1) promoted the growth of MM cells, while ra- foxanide treatment could induce cell cytotoxicity in the presence of BMSC (Fig. 1E), as well as IL-6 and IGF-1 (Fig. 1F and G). As shown in Fig. 1H and I, rafoxanide treatment also could induce MM cells apop- tosis even in the presence of BMSC. Moreover, rafoxanide treatment does not affect the viability of BMSC and has no apparent apoptosis in BMSC (Fig. 1E, H and I). These data suggested that rafoxanide not only directly targets MM cells, but also overcomes the protective effect of the MM-host BM microenvironment.We next evaluated the effect of rafoxanide on apoptosis in MM cells via Annexin-V/PI double staining using flow cytometry. After treatmentwith rafoxanide (0, 10, or 20 μM) for 24, 48, or 72 h, analysis of apoptosis showed that rafoxanide induces marked apoptosis in H929, ARP1, and OCI-MY5 cells in a dose-dependent manner (Fig. 2A–D). Western blot analysis further confirmed that rafoxanide induced apoptosis by increasing the expression of cleaved-caspase 3, cleaved-caspase 8, and cleaved-caspase 9, as well as reducing Bcl-2 and Bcl-xl and up-regulating Bax (Fig. 2E). Meanwhile, we analyzed the MMP, an indicator of cell apoptosis, in MM cells by flow cytometry using a JC-1 MMP kit. As shown in Fig. 3A and B, MMP was significantly reduced in H929, ARP1, and OCI-MY5 cells after treatment with rafoxanide for 48 h.

Moreover, to further determine the dependence of rafoxanide- induced apoptosis on the caspase pathway, we next analyzed the effect of a pan-caspase inhibitor named Z-VAD-FMK. Importantly, using flow cytometry with Annexin-V/PI double staining, our results indicated that Z-VAD-FMK significantly blocked rafoxanide-induced apoptosis of H929, ARP1 and OCI-MY5 cells (Fig. 3C and D). These findings sug- gested that rafoxanide triggers both extrinsic and intrinsic apoptotic pathways, and that rafoxanide-induced apoptosis in MM is dependent on the caspase pathway.To further confirm the anti-MM activity of rafoxanide, we evaluatedthe effect of rafoxanide on CD138+ MM cells isolated from the BM of MM patients (Table 1) by flow cytometric analysis. As shown in Fig. 3E and F, rafoxanide induced significant apoptosis in CD138+ MM cells, while no apparent apoptosis was observed in normal PBMCs (Fig. 3G and H), indicating that rafoxanide has no toxicity in normal PBMCs and is a favorable drug for the treatment of MM.As rafoxanide was able to inhibit MM cell viability, we next eval- uated whether this drug could affect DNA synthesis in MM cells using EdU incorporation assay. Our results showed that rafoxanide greatly reduced the level of EDU in H929 and ARP1 cells (Fig. 4A and B), suggesting that the DNA synthesis of MM cells was markedly inhibited after treatment with rafoxanide for 24 h.

Additionally, we analyzed the effect of rafoxanide on the MM cell cycle using flow cytometric analysis. As shown in Fig. 4C and D, rafoxanide treatment caused H929, ARP1, and OCI-MY5 cells to accumulate in G0G1 phase. Meanwhile, western blot analysis further confirmed rafoxanide-induced MM cell cycle arrest in G0G1 phase by reduction of the protein expression of cyclinD1, CDK4, CDK6, decreased cdc25A, and increased phosphorylation of CHK2, further suggesting that rafoxanide-induced cell cycle arrest is mediated by the cdc25A-degradation pathway (Fig. 4E).To determine whether rafoxanide affects DNA damage, we next analyzed the expression of γ-H2AX (a marker of DNA damage) after treatment of MM cells with rafoxanide for 24 h. The results of im- munofluorescence analysis showed that the presence of γ-H2AX nuclear foci was markedly increased in H929 and ARP1 cells after administra- tion of rafoxanide (Fig. 5A). Meanwhile, we evaluated the protein level of γ-H2AX in H929 and ARP1 cells. Importantly, consistent with the results of immunofluorescence analysis, the protein expression of γ- H2AX in H929 and ARP1 cells was increased after rafoxanide treatment (Fig. 5B), further indicating that rafoxanide can enhance the DNA da- mage response in MM cells. Besides DNA damage, the MAPK pathway also plays an important role in MM. Thus, we analyzed changes in the MAPK pathway after treatment with rafoxanide. Interestingly, we found that rafoxanide was able to significantly suppress the p38 MAPK pathway by reducing the phosphorylation of p38 MAPK, whereas the expression of phosphorylated ERK1/2 and phosphorylated JNK re- mained unchanged. The down-regulation of Stat1 phosphorylation, a downstream protein of p38 MAPK, further confirmed the inhibition ofthe p38 MAPK pathway after rafoxanide treatment (Fig. 5C).

We established a human myeloma xenograft model to further in- vestigate the anti-MM activity of rafoxanide in vivo. Five-week-old nude mice were injected subcutaneously with 2.5 × 106 H929 cells. Mice were treated with or without rafoxanide (15 mg/kg body weight every two days) by intraperitoneal injection for a total of 14 days. Tumorgrowth and tumor weight in rafoxanide-treated groups was inhibited compared with the vehicle-treated group (Fig. 6A–C). Meanwhile, we evaluated the body weight of vehicle-treated and rafoxanide-treated mice. Results showed no significant body weight changes in these two groups, indicating that rafoxanide was well tolerated (Fig. 6D). In ad-dition, HE staining showed an obvious increase of cell shrinkage and fragmentation in harvested tumors from the rafoxanide-treated group compared with the vehicle-treated group (Fig. 6E). Meanwhile, we found no obvious histological changes in the heart, liver, lung, or kidney in any of the mice (Supplementary Fig. S2), further suggesting that the side-effects of rafoxanide were minimal. In addition, rafoxanide treatment markedly inhibited tumor proliferation, indicated by reduced Ki-67 staining, and induced tumor apoptosis, indicated by increased numbers of cleaved-caspase 3- and TUNEL-positive cells (Fig. 6F).

Meanwhile, we examined the expression of γ-H2AX and p-p38 MAPK in harvest tumors, results showed that rafoxanide enhances DNA damage response by increasing the expression of γ-H2AX and inhibits the acti- vation of p38 MAPK pathway via suppressing the expression of p-p38 MAPK (Fig. 6F), consistent with our in vitro results. These findings suggested that rafoxanide has potent anti-MM activity in vivo.To examine whether rafoxanide could be used in combination therapy, we detected MM cells viability of rafoxanide in combination with bortezomib or lenalidomide. As shown in Fig. 7A–D, combination of rafoxanide and bortezomib or lenalidomide significantly induced synergistic cytotoxicity in H929 and ARP1 cells, with the combinationindex (CI) < 1.Our previous study has indicated rafoxanide is a potent B-Raf V600Einhibitor with IC50 value of 0.07 μM, which is appreciable compared to the positive control vemurafenib (a marketed drug targeting B-Raf V600E; IC50: 0.17 μM). Moreover, vemurafenib inhibited mutated and wild type B-Raf with the same IC50 level, while rafoxanide showed high potency against the wild type B-Raf and B-Raf V600E [12], suggesting that rafoxanide could inhibit both wide type B-Raf and B-Raf V600E.

To determine whether rafoxanide is a B-Raf V600E inhibitor, we firstly examined the expression of B-Raf in MM cell lines by using western blot analysis (Supplementary Fig. S3), then constructed B-Raf V600E over- expressed H929 and ARP1 cells by lentivirus transduction (Fig. 8A), and B-Raf V600E knocked down H929 and ARP1 cells by siRNA transduc- tion (Fig. 8B). Our CCK8 and flow cytometry analysis showed that ra- foxanide suppresses the growth of H929 and ARP1 cells while over- expression of B-Raf V600E could rescue cells from rafoxanide-induced proliferative inhibition and apoptosis. Additionally, knockdown of B- Raf V600E could induce cytotoxicity and apoptosis towards rafoxanide compared with B-Raf V600E-OE H929 and ARP1 cells, while no sig-nificant difference compared with the wild type group (Fig. 8C–F).However, CCK8 analysis showed the IC50 of rafoxanide in these MM cells is 20–40 μM, while 10–20 nM for vemurafenib, and vemurafenib has the similar IC50 level in these cells (Fig. 8C and D). These data suggested that rafoxanide has anti-proliferation effect on both wild typeand B-Raf V600E mutated MM cells. And the weaker anti-MM activity of rafoxanide than vemurafenib in vitro may indicate other potential mechanisms besides targeting B-Raf V600E mutation.

4.Discussion
Rafoxanide, an anthelmintic drug of the salicylanilide class, is used for the treatment of both mature and immature stages of liver fluke in cattle and sheep [19]. An important finding indicated that rafoxanide exhibits desirable properties for repositioning as a human agent [11]. We previously demonstrated that rafoxanide is a potent B-Raf V600E inhibitor via molecular docking with halogen bonding scoring function and bioassay [12]. The B-Raf mutation causes constitutive activation of the MAPK pathway and both of B-Raf and MAPK play significant roles in MM cell proliferation and growth [13]. Despite advances in treat- ment options over recent decades, MM is still an incurable hematolo- gical malignancy with accumulation of plasma cells in the BM [20]. Therefore, we investigated the effects of rafoxanide on MM cells. Our present study is the first to demonstrate that this anthelmintic drug has antitumor activity and exerts effective cytotoxicity on MM cells both in vitro and in vivo (Fig. 7E). In our in vitro study, we found that rafoxanide markedly reduces the viability of a series of MM cell lines and has a dose- and time-dependent effect in H929, ARP1 and OCI-MY5 cells, further suggesting that ra- foxanide has anti-proliferative activity in MM cells. There is a growing body of evidence indicating that MM is highly dependent on the BM microenvironment [21,22], so there is an urgent need to develop novel anti-MM drugs to overcome the protective effect of the BM micro- environment. Our present data show that rafoxanide significantly in- hibits the growth of MM cells even in the presence of BMSC and cyto- kines IL-6 and IGF-1. Importantly, rafoxanide treatment also induces the apoptosis of MM cells in the presence of BMSC, while has no sig- nificant effect in BMSC. These findings suggest that rafoxanide not only directly targets MM cells, but also confers the ability to overcome the protective effect of the BM microenvironment on MM cells.

Several studies have indicated that anti-proliferative activity of drugs is associated with apoptosis and cell cycle arrest [23,24]. Apop- tosis, a main mechanism in cell death, could be triggered by intrinsically or extrinsically via death signal pathways, including cas- pases, inhibitors of apoptosis proteins, B cell lymphoma (Bcl)-2 family, tumor necrosis factor (TNF) receptor gene family, or p53 gene [25]. Therefore, we first evaluated apoptosis of MM cells after treatment with rafoxanide. Consistent with the results from CCK8 assay, rafoxanide induced cell apoptosis in both MM cell lines and primary CD138+ MM cells, whereas no apparent apoptosis was observed in healthy PBMCs. Moreover, we further confirmed rafoxanide-induced cell apoptosis by activation of caspases, including up-regulation of the expression of cleaved-caspase 3, cleaved-caspase 8, and cleaved-caspase 9, as well as down-regulation of the protein levels of Bcl-2 and Bcl-xl along with increasing levels of Bax in MM cells. MMP, an indicator of mitochon- drial membrane permeability, is decreased during early apoptosis [26]. Interestingly, we showed that rafoxanide could regulate the loss of MMP. Moreover, we also found that a pan-caspase inhibitor (Z-VAD- FMK) impaired the effect of rafoxanide in inducing MM apoptosis.

These data suggest that rafoxanide induces MM apoptosis via both ex- trinsic and intrinsic apoptotic pathways, and that rafoxanide-induced cell apoptosis is regulated via MMP levels and the caspase pathway. Cell cycle arrest induced by anti-tumor inhibitor could result in cell death via apoptosis in MM [27]. We found that rafoxanide induced inhibition of DNA synthesis, as well as arrest of MM cells in G0G1 phase. CyclinD1/CDK4/CDK6 complex, an important complex in the G0G1 phase, plays a vital role in regulating the progression from G0G1 phase to the G2M phase [28]. Thus, we detected the expression of this com- plex in MM cells after rafoxanide treatment. Importantly, our western blot results indicated that rafoxanide-induced G0G1 phase arrest is as- sociated with reduced protein expression of cyclinD1, CDK4, and CDK6, further suggesting that rafoxanide could induce G0G1 phase arrest of MM cells. Meanwhile, the results showed that rafoxanide decreases the expression of cdc25A, while increases p-CHK2. It is known that CHK2 plays a significant role in the DNA damage response pathway, which can be activated in response to DNA damage and directly regulate cdc25A that associated with cell cycle control [29]. These findings show that rafoxanide-induced cell cycle arrest is mediated by the cdc25A-degradation pathway, and this arrest could further trigger DNA damage response. γ-H2AX also has an important role in the DNA da- mage response, and the presence of γ-H2AX nuclear foci is the hallmark of DNA double-strand breaks [30]. Thus, we analyzed the expression of γ-H2AX in MM cells after treatment with rafoxanide. The data showed that rafoxanide increases the presence of γ-H2AX nuclear foci in MM cells and activates the phosphorylation of H2AX, suggesting that ra- foxanide enhances the DNA damage response. Several series of studies have reported that the induction of cell apoptosis is associated with DNA damage response pathway or MAPK pathway [31,32].

Furthermore, these pathways are involved in the cellular responses, such as cell proliferation, differentiation, migration and apoptosis [33,34]. Moreover, as rafoxanide is a potent B-Raf V600E inhibitor [12], the MAPK pathway could be activated via a B-Raf mutation [16], and a recent study reported that PLX8394, a new generation B-Raf in- hibitor, could selectively inhibit B-Raf in colonic adenocarcinoma cells and prevent paradoxical MAPK pathway activation [35]. Therefore, we further investigated whether rafoxanide affects the activation of MAPK pathway. Interestingly, our data showed that rafoxanide could inhibit activation of the p38 MAPK pathway by reducing the phosphorylation of p38 MAPK, as well as down-regulating phosphorylation of Stat1 (a downstream of p38 MAPK). Taken together, our results suggest that enhancement of the DNA damage response and inhibition of the p38 MAPK pathway play important roles in rafoxanide-induced apoptosis in MM cells. To further confirm the anti-MM activity of rafoxanide, we next established an MM xenograft mouse model in vivo. Our data showed that rafoxanide not only markedly inhibits tumor growth, but also has no apparent toxicity in mice. Meanwhile, immunohistochemical staining of harvested tumors confirmed rafoxanide-induced anti-proliferation, pro-apoptotic effects, enhancement of DNA damage response, and suppression of p38 MAPK activation in MM cells. Consistent with our in vitro study, these in vivo results further indicate that rafoxanide has potent anti-MM activity. Additionally, we found that 10–20 μM rafoxanide induced apoptosis in patient CD138+ MM cells but did not affect normal PBMCs. Injection of 15 mg/kg dose of rafoxanide every two days for a total of 14 days retarded tumor growth but did not reduce body weight or cause heart/liver/lung/kidney damage in mice. Therefore, our preliminary results indicate that rafoxanide is a well-
tolerated drug in MM treatment.

Combinational therapy is widely required after primary therapy in MM patients. Therefore, we detected whether rafoxanide could be en- hance the cytotoxicity of anti-MM agents, such as bortezomib and le- nalidomide. Our data showed that combination of rafoxanide with bortezomib or lenalidomide induced synergistic cytotoxicity in MM cells. The synergistic effect suggests that combination of rafoxanide with bortezomib or lenalidomide may be a promising therapeutic strategy in MM patients. Additionally, the in vivo efficacy of combina- tion of rafoxanide and bortezomib and the main mechanism of this synergistic effect is being investigated in our laboratory. To determine whether rafoxanide is a B-Raf V600E inhibitor, we constructed B-Raf V600E mutated MM cells in vitro. Unexpectedly, our data indicated that the anti-proliferation effect of rafoxanide is weaker than vemurafenib in MM cells, and the IC50 level of rafoxanide in vitro is inconsistent with our previous studies. In view of the conflicting results, further studies are necessary to investigate whether these discrepancies are due to specific experiment design, inherent MM cells performance, drug characteristic, or other factors. Nevertheless, our preliminary findings also reveal that rafoxanide has anti-MM effect on both wild type and mutated cells, and provide a theoretical basis for the ther- apeutic dose of rafoxanide in MM patients with or without B-Raf V600E mutation. Certainly, further studies are needed to explore other po- tential mechanisms and the optimal dose for future clinical attempts.

In conclusion, rafoxanide, as an anthelmintic drug, effectively exerts anti-MM activity by enhancing the DNA damage response and sup- pressing the p38 MAPK pathway by inhibiting MM cell proliferation, inducing MM cells apoptosis, reducing MMP, causing accumulation of MM cells in G0G1 phase, retarding tumor growth in vivo, and inducing synergistic cytotoxicity in combination with bortezomib or lenalido- mide. Our study demonstrates that the potential rafoxanide in the treatment of MM. However, the detailed mechanisms of whether its therapeutic effect is related to targeting B-Raf V600E mutation in MM require further investigation.