Synergistic Induction of Apoptosis in Multiple Myeloma Cells by Bortezomib and Hypoxia-Activated Prodrug TH-302, In Vivo and In Vitro

Multiple myeloma is a malignant plasma-cell disorder that accounts for 1% of all cancers and 10% to 15% of all hematologic malignancies (1). In the past decade, there have been major advances in the treatment of multiple myeloma. The introduction of novel agents such as bortezomib (proteasome inhibitor) and thalidomide (immunomodulatory drug) has dramatically improved the outcome of patients with multiple myeloma (2–4). Moreover, significant advances in both basic and translational research have enhanced understanding of disease pathogenesis and guided the development of new and more effective therapies (5). It has been shown previously by us and other groups that hypoxia is a critical microenvironmental factor in multiple myeloma (6–11). Our data further support that targeting the hypoxic niche with hypoxia-activated prodrug (HAP) TH-302 (Fig. 1A) is a potential new treatment option for multiple myeloma (6). TH-302 as a 2-nitroimidazole prodrug of the cytotoxic bromo-isophosphoramide mustard (Br-IPM), exhibits hypoxia-selective cytotoxicity against a broad spectrum of human cancer cell lines in vitro, and shows efficacy across a large panel of human tumor xenografts in vivo. TH-302 is being evaluated in clinical trials for the treatment of solid tumors as a monotherapy and in combination with chemotherapeutic agents (12).

Bortezomib (PS-341), the first therapeutic proteasome inhibitor, was approved for refractory, relapsed, and newly diagnosed multiple myeloma (13). Treatment of multiple myeloma with bortezomib has dramatically improved survival for patients with multiple myeloma; however, some patients do not benefit from this treatment, and those responding ultimately acquire resistance (14). On the other hand, severe treatment-related toxicities such as peripheral neuropathy have been observed in conjunction with the use of bortezomib (15). Combination therapy using lower doses of bortezomib is an approach to increase efficacy and reduce side effects, thereby enhancing the likelihood of survival for patients with multiple myeloma (16).

In the current study, we investigated the anti–multiple myeloma effects of TH-302 combined with bortezomib in hypoxic conditions. Our results show that the combination of TH-302 and bortezomib synergistically induced apoptosis in multiple myeloma cells in vitro, associated with the activation of caspases, the decrease of antiapoptotic Bcl-2 family members, and importantly, the increase of proapoptotic BH3-only proteins PUMA (p53-upregulated modulator of apoptosis) and NOXA. Further experiments in the 5T33MMvv mouse model in vivo showed that the combination of TH-302 and bortezomib can improve multiple disease parameters and result in a significant prolonged survival. Taken together, these findings provide preclinical rationale for further evaluation of the combination of TH-302 and bortezomib in clinical trials.

TH-302 was provided by Threshold Pharmaceuticals. TH-302 was dissolved in sterile PBS at 10 mmol/L for in vitro studies and at 5 mg/mL for in vivo studies. Bortezomib was from LC Laboratories. For in vitro studies, bortezomib was reconstituted in dimethyl sulfoxide (DMSO) at a stock concentration of 10 mmol/L and stored at −20°C until use, and this stock was diluted in medium just before use so that the concentration of DMSO never exceeded 0.1%. For in vivo studies, clinical-grade vials of bortezomib were reconstituted using a sterile NaCl 0.9% solution. Solutions were sterilized by filtration through a 0.22-μm syringe filter.

The human LP1, OPM2 (both from American Type Culture Collection), MMS1 (a kind gift from Y. Okuno, Kyoto University, Kyoto, Japan) and murine myeloma cell line 5T33MMvt (a kind gift of J. Radl, TNO, Leiden, the Netherlands) were used in this study. A stock was prepared of all authenticated cell lines and experiments were carried out within 2 months after thawing vials. The murine cell line was authenticated by its expression of a specific idiotype. The multiple myeloma cells were cultured in normoxic (20% O₂) and hypoxic conditions (1% O₂) as described previously (6).

A total of 1 × 10⁶ cells were washed twice with PBS and stained with 2 μL 7-amino-actinomycin D (7-AAD; BD Biosciences) and 5 μL Annexin V–fluorescein isothiocyanate (FITC; BD Biosciences) in 100 μL of binding buffer, and incubated at 4°C for 15 minutes. Then, cells were resuspended in 400 μL of binding buffer and immediately analyzed using a FACSCanto flow cytometer (Becton Dickinson).

Multiple myeloma cells were treated simultaneously with various concentrations of TH-302 and bortezomib for 16 hours in hypoxic conditions. Apoptosis induced by each agent alone and by the combination treatment was monitored by flow-cytometric analysis of multiple myeloma cells stained with Annexin V–FITC and 7-AAD as described earlier. The degree of synergism between the two compounds was determined using the CompuSyn Software (ComboSyn Inc.) developed by Chou (17). The combination index (CI) for each combination was calculated at a nonconstant ratio.

Western blot analysis was conducted as previously described (6). The primary antibodies against Bcl-2, Bcl-xL, Mcl-1, BAD, BID, BAX, caspase-3, caspase-8, caspase-9, PARP, phospho-p53 (Ser15), and β-actin were from Cell Signaling Technology and diluted at 1:1,000. The primary antibodies against PUMA, NOXA, p53, p21^(WAF1/CIP1), ATF4, GRP-78, and CHOP (C/EBP homologous protein; also known as DNA damage–inducible gene 153) were from Santa Cruz Biotechnology and diluted at 1:500.

RNA extraction was conducted using the RNeasy Kit (Qiagen). Total RNA was reverse transcribed using the Verso cDNA Synthesis Kit (Thermo Scientific) according to the manufacturer's instructions. Real-time PCR was conducted with Maxima SYBR Green/ROX qPCR Master Mix (Fermentas) on an ABI Prism 7900 Fast instrument using gene-specific primers. Primer sequences used for amplifications are as follows: β-actin, forward: 5′-ATCGTGCGTGACATTAAGGAGAAG-3′; reverse: 5′-AGGAAGGAAGGCTGGAAGAGTG-3′. BRCA1, forward: 5′-ACCGTTGCTACCGAGTGTCTG-3′; reverse: 5′-GTGATGTTCCTGAGATGCCTTTGC-3′. BRCA2, forward: 5′-AACCGTGTGGAAGTTGCGTATTG-3′; reverse: 5′-GGCTCCCGTGGCTGGTAAATC-3′. FANCD2, forward: 5′-CAGGAGAGCACAGCAGATGAGAG-3′; reverse: 5′-AGGCAGGAGAATCGCTTGAACC-3′. RAD51, forward: 5′-TCAAGTGGATGGAGCAGCGATG-3′; reverse: 5′-GGCAGTCACAACAGGAAGAGG-3′. p53, forward: 5′-TTGCGTGTGGAGTATTTGGATGAC-3′; reverse: 5′-AGTGTGATGATGGTGAGGATGGG-3′. p21^(WAF1/CIP1), forward: 5′-CCAGCGACCTTCCTCATCCAC-3′; reverse: 5′-CCATAGCCTCTACTGCCACCATC-3′. PUMA, forward: 5′-CTGCCGCCCACCACCATC-3′; reverse: 5′-TGAAGGAGCACCGAGAGGAGAG-3′. NOXA, forward: 5′-ACCGTGTGTAGTTGGCATCTCC-3′; reverse: 5′-AGGTTCCTGAGCAGAAGAGTTTGG-3′. GRP-78, forward: 5′-AGGAGGAGGACAAGAAGGAGGAC-3′; reverse: 5′-CAGGAGTGAAGGCGACATAGGAC-3′. CHOP, forward: 5′-TGCTTCTCTGGCTTGGCTGAC-3′; reverse: 5′-CCGTTTCCTGGTTCTCCCTTGG-3′. The thermal cycling conditions included 2 minutes at 50°C and 10 minutes at 95°C followed by 40 cycles of 95°C for 0.15 minutes and 60°C for 1 minute. C_t values were collected for β-actin and the genes of interest during the log phase of the cycle. Quantification of given genes expressed as mRNA level was normalized to β-actin RNA using the ΔΔC_t method.

The 5T33MM model originated spontaneously in aging C57BL/KaLwRij mice and has since been propagated in vivo by intravenous transfer of the diseased marrow in young syngeneic mice (18). C57BL/KaLwRijHsd mice were purchased from Harlan CPB. Male mice were 6 to 10 weeks old when used, and housed and treated following the conditions approved by the Ethical Committee for Animal Experiments, VUB (license no. LA1230281).

To study the effects of TH-302 alone, bortezomib alone, and the combination on myeloma progression, four groups of C57BL/KaLwRij mice (n = 10) were injected intravenously with 0.5 × 10⁶ 5T33MMvv cells; one group of 10 naïve mice was included as negative control. One week after tumor cell inoculation, mice were treated with either TH-302 alone [100 mg/kg, twice weekly, intraperitoneally (i.p.) injection], or bortezomib alone (0.8 mg/kg, twice weekly, s.c. injection), or bortezomib in combination with TH-302 (twice weekly, TH-302 was injected 4 hours after bortezomib), or vehicle (0.9% NaCl) until the first mouse showed signs of morbidity. Serum paraprotein concentration was assessed using standard electrophoretic techniques. Bone marrow tumor burden was assessed by determining plasmacytosis on cytosmears. Bone marrow angiogenesis was assessed by determining microvessel density (MVD) as previously described (19).

Five days after tumor cell inoculation, four groups (n = 10) of mice injected with 5T33MMvv cells were treated with either TH-302 alone (100 mg/kg, twice weekly, i.p. injection), or bortezomib alone (0.8 mg/kg, twice weekly, s.c. injection), or bortezomib in combination with TH-302 (twice weekly, TH-302 was injected 4 hours after bortezomib), or vehicle (0.9% NaCl). Kaplan–Meier analysis was used for analyzing overall survival.

For statistical analysis of the in vivo data, differences between two groups were determined by Mann–Whitney test. Kaplan–Meier analysis was used to determine the effects on the survival time of the tumor-bearing mice, the differences between groups were tested for statistical significance using the two-tailed log-rank test. For qRT-PCR data analysis, a two-tailed paired t test was conducted for the analysis of differences in cells with or without treatment with a software package GraphPad Prism (GraphPad software). P < 0.05 was considered statistically significant.

In our previous study, we showed that the TH-302 exhibits hypoxia-selective cytotoxicity against multiple myeloma cells in vitro and shows attractive efficacy in multiple myeloma in vivo (6). In the present study, we investigated whether bortezomib can potentiate TH-302–induced apoptosis of multiple myeloma cells in hypoxic conditions. To examine the effects of the combination, three multiple myeloma cell lines were exposed to increasing concentrations of the two agents. Cell apoptosis was measured after 16 hours by flow cytometry. As shown in Fig. 1, bortezomib in combination with TH-302 increased the apoptosis levels in 5T33MMvt, OPM2, and MMS1 cells. The interaction between bortezomib and TH-302 was further analyzed using the Chou–Talalay method (17), to determine whether this combination exhibited additive or synergistic cytotoxicity. Using CompuSyn software, we calculated the CIs of varying concentrations of TH-302 with bortezomib. In 5T33MMvt cells, at doses ranging from 2.5 to 10 μmol/L TH-302 combined with 5 to 20 nmol/L bortezomib, CI ranged from 0.308 to 0.891, suggesting that this combination was synergistic (Table 1). Similar results were also observed in OPM2 and MMS1 cells (Supplementary Tables S1 and S2).

Caspase activation plays a central role in the execution of apoptosis. This proteolytic cascade, in which one caspase can activate other caspases, amplifies the apoptotic signaling pathway and thus leads to rapid cell death (20). To determine whether caspase activation is involved in the augmented apoptosis by the combination treatment with bortezomib and TH-302, two cell lines 5T33MMvt and OPM2, exhibiting good and moderate response to the combination treatment, were chosen for further studies. We examined the activation of caspase cascades as defined by their cleavage at specific aspartate residues by Western blot analysis. As seen in Fig. 2, treating the cells with the combination induced a greater processing of caspase-3, -8, and -9 and PARP in both 5T33MMvt and OPM2 cells, coincided with the appearance of bands representing cleavage products of caspase-3, -8, and -9 and PARP.

To characterize the mechanism responsible for bortezomib–TH-302–mediated synergism in multiple myeloma, we next investigated the changes of Bcl-2 family members at the protein level. In Fig. 3A and B, the changes of antiapoptotic protein Bcl-2, Bcl-xL, and Mcl-1 in two multiple myeloma cell lines are shown. TH-302 alone causes a decrease in the expression levels of all three antiapoptotic proteins, whereas bortezomib alone does not affect the expression of Bcl-xL, slightly decreases the expression of Bcl-2, and markedly increases the expression of Mcl-1. When treated with the combination, the expression of Bcl-xL is further decreased as compared with TH-302 alone, whereas Mcl-1 and Bcl-2 are similarly decreased as compared with TH-302 alone.

In addition, the effects of TH-302 and bortezomib on proapoptotic Bcl-2 family proteins were investigated. As shown in Fig. 3C and D, no significant changes were observed in BAX proteins. However, among the proapoptotic BH3-only members, both TH-302 and bortezomib induced the cleavage of BAD and BID, thereby consequently producing the truncated BAD and BID proteins, which are stronger inducer of apoptosis than the full-length proteins (21–23). In contrast to BAD and BID, the expression levels of the proapoptotic BH3-only protein PUMA and NOXA were increased by both TH-302 and bortezomib.

To further address the mechanisms underlying the induction of PUMA and NOXA, we investigated the roles of p53-dependent and -independent pathways in response to TH-302 and bortezomib treatment. When exposed to hypoxic conditions, TH-302 reduction at the nitroimadazole site of the prodrug by intracellular reductases leads to the release of the DNA alkylating agent bromo-isophosphoramidate mustard, which is genotoxic (24). In response to genotoxic stress, the activation of p53 plays a crucial role in governing cell apoptosis. Because PUMA and NOXA are direct targets in p53-mediated apoptosis (25), we then investigated whether p53 is involved in TH-302- and bortezomib-induced apoptosis. As shown in Fig. 4A, we confirmed that p53 was upregulated and activated by TH-302 in OPM2 cells, which express the wild-type p53. In parallel to the activation of p53, as monitored by the phosphorylation on Ser-15, the expression level of the target gene p21^WAF1/CIP1 was also increased. Similar results were also observed in bortezomib-treated multiple myeloma cells in hypoxic condition (Fig. 4C), and the results confirmed the finding that the induction and activation of p53 is one of the important molecular mechanisms mediating antimyeloma activity of bortezomib (26, 27). Moreover, by quantitative measuring of the transcriptional expression levels of the p53-related DNA repair genes such as BRCA1, BRCA2, RAD51, and FANCD2, we found that TH-302 upregulates the transcription of these DNA repair genes, suggesting the activation of DNA repair machinery. However, when treated with bortezomib, the expression levels of DNA repair genes were decreased (Fig. 4B and D). Interestingly, in the p53-mutated multiple myeloma cell line LP1, we still observed the induction of PUMA and NOXA, implicating the role of p53-independent pathways in the regulation of PUMA and NOXA. We then investigated the contribution of endoplasmic reticulum (ER) stress to the induction of PUMA and NOXA. As expected, we observed upregulated expression levels of the ER stress markers GRP-78 and CHOP in response to the treatment by both bortezomib and TH-302. Moreover, the expression of ATF4, one major regulator of ER stress signaling pathway, was also increased (Fig. 4E and G). Furthermore, by using qRT-PCR, we show that the expression of ER stress markers and PUMA as well as NOXA were all upregulated at transcription level (Fig. 4F and H).

To test whether the enhanced myeloma cell apoptotic response is observed in vivo, we examined the efficacy of the combination of bortezomib and TH-302 in the 5T33MM mouse model. In the first series of experiments, C57BL/KaLwRijHsd mice inoculated with 5T33MMvv cells were either assigned to receive TH-302, bortezomib, or a combination of both starting at day 7 after inoculation (Fig. 5A). In mice treated with TH-302 alone, bortezomib alone, or the combination, a significant reduction in serum paraprotein concentrations and plasmacytosis in the bone marrow were observed compared with the vehicle-treated mice (Fig. 5B and C). Importantly, the combination-treated group showed more decreases in the serum paraprotein concentrations and plasmacytosis than in the single-drug–treated groups. Immunohistochemical staining for CD31 showed significantly decreased MVD in the treatment groups (Fig. 5D). Similarly, combination treatment led to more significant decrease of MVD than each single-drug group. No significant reduction in body weight was observed in treated mice compared with vehicle-treated controls (data not shown).

To examine whether the combination of TH-302 and bortezomib can prolong the survival of multiple myeloma–inoculated mice, a second series of experiments was carried out. As shown in Fig. 5E, in vivo treatment of 5T33MM mice with a combination of TH-302 and bortezomib resulted in a significantly prolonged survival when compared with the vehicle group (P < 0.0001), the TH-302-alone–treated group (P < 0.0001), and the bortezomib-alone–treated group (P < 0.0001). The median survival time of each group was 25.5 (vehicle), 34 (bortezomib), 41.5 (TH-302), and 53.5 (bortezomib and TH-302) days, respectively. These data clearly show that the combination of TH-302 and bortezomib results in improvements in multiple disease parameters and increased overall survival of the mice.

Despite advances in chemotherapy and stem-cell transplantation, multiple myeloma remains an incurable disease. As shown in our previous report, multiple myeloma cells localize at the hypoxic regions in the bone marrow (6, 10). This finding not only provides us new insights into the multiple myeloma–bone marrow microenvironment, but also provides us a new treatment target taking advantage of hypoxia in multiple myeloma. By using a HAP TH-302, we previously showed that targeting hypoxia is a potential new treatment option for multiple myeloma (6). In the present study, we investigated whether the combination of hypoxia-activated treatment with bortezomib can achieve better therapeutic efficacy, based on a finding that bortezomib has preferential cytotoxicity toward hypoxic solid tumor cells (28). We showed here that the combinatorial use of TH-302 and bortezomib can mediate synergistic antitumor effects in multiple myeloma.

Induction of apoptosis in tumor cells represents a key strategy of cancer chemotherapy. It has been shown that Bcl-2 family members are central regulators of apoptosis (29). All Bcl-2 family members contain at least one of four conserved Bcl-2 homology (24) domains. According to their function and structure, the Bcl-2 family can be divided into two classes: the antiapoptotic and proapoptotic members. The antiapoptotic proteins have all four BH domains and include Bcl-2, Bcl-xL, Mcl-1, Bcl-w, and A1. The proapoptotic Bcl-2 proteins are further functionally divided into two subgroups: the effector molecules such as BAX and BAK, containing BH domains 1–3, and the BH3-only proteins such as BAD, BID, BIK, BIM, BMF, BNIP3, HRK, NOXA, and PUMA, which contain only the BH3 domain (30). In the current study, following the findings that the combination of bortezomib and TH-302 can synergistically induce apoptosis (Fig. 1 and Table 1), we further investigated the changes of some important Bcl-2 family members in multiple myeloma cells. The changes of Bcl-2 family members are shown in Fig. 3 and are summarized in Table 2 based on quantitative analysis.

We found that the protein levels of antiapoptotic Bcl-2, Bcl-xl, and Mcl-1 were decreased by TH-302 in hypoxic conditions in all tested multiple myeloma cell lines. Moreover, when treated with bortezomib alone, the expression of Bcl-2 was decreased, the expression of Bcl-xL was not disturbed, and the expression of Mcl-1 was increased. Nevertheless, when the multiple myeloma cells were treated with the combination of drugs, we observed clearly a decrease of all three antiapoptotic Bcl-2 family proteins. Among these three investigated antiapoptotic Bcl-2 family members, the changes of Mcl-1 are noticeable. Mcl-1 is overexpressed in multiple myeloma cells, and plays essential roles for the survival of multiple myeloma cells by its ability to oppose to a wide variety of proapoptotic stimuli (31). Downregulation or reduction of Mcl-1 has been proposed to play very important roles in response to drug-induced apoptotic stimuli (32). The expression of Mcl-1 can be regulated at both transcriptional and posttranslational levels. The accumulation of Mcl-1 upon bortezomib treatment has been shown to induce multiple myeloma cell resistance to bortezomib-induced lethality, and was speculated to be the result of the inhibition of ubiquitation-mediated degradation (33). More recently, we showed that the accumulation of Mcl-1 by bortezomib was transcriptionally upregulated by the selective activation of the ATF4 signaling branch of the unfolded protein response (UPR; ref. 34). Moreover, Mcl-1 was also shown to be a substrate for caspases during induction of apoptosis (35–37). In this regard, the cleavage of Mcl-1 by caspases is related to the decrease of the full length of the Mcl-1 protein. More importantly, the cleaved Mcl-1 shows strong proapoptotic activity by enhancing its turnover and impairing its ability to interact with BH3-only proapoptotic proteins such as NOXA-, PUMA-, and BIM-induced apoptosis (38). In line with the finding of Mcl-1 cleavage, the cleavage of PARP and pro-caspase-3/8/9 shows similar changes when treated with TH-302 and bortezomib either alone or in combination, suggesting a tight correlation between caspase activation and Mcl-1 reduction (Figs. 2 and 3A). Similar to the changes of Mcl-1 antiapoptotic proteins, we also observed that TH-302 and bortezomib can result in the cleavage of several BH3-only proapoptotic proteins such as BAD and BID (Fig. 3B and Table 2). However, the cleavage of BAD and BID does not imply the decrease of proapoptotic function, because previous studies have shown that the cleavage of BAD and BID is a caspase-dependent process, and the cleaved BAD and BID fragments exhibit stronger mitochondrial localization and higher apoptotic capability than intact ones, thus promoting apoptosis (21, 22).

One of the most striking findings of the study was that both TH-302 and bortezomib can induce the expression of proapoptotic BH3-only proteins PUMA and NOXA. The induction of PUMA and NOXA was speculated to be another crucial event mediating the synergistic apoptosis. PUMA and NOXA have been shown to be the transcriptional targets of p53 and essential mediators in p53-induced apoptosis (39–42). In the current study, we found that both TH-302 and bortezomib can activate p53 and its downstream target gene p21^WAF1/CIP1 in multiple myeloma cells expressing wild-type p53 (Fig. 4A and C). However, the mechanisms underlying p53 activation by TH-302 and bortezomib are different. TH-302 is a hypoxia-activated DNA cross-linking agent and under hypoxic condition will lead to genotoxic stress in multiple myeloma cells, inducing p53 signal transduction and apoptosis. In response to the genotoxic stress caused by TH-302, the activation of p53 may play a direct role in triggering the onset of DNA repair processes. Accordingly, we found that the transcription levels of DNA repair genes such as BRCA1, BRCA2, RAD51, and FANCD2 were significantly increased by TH-302 (Fig. 4B). However, in multiple myeloma cells treated with bortezomib, although p53 was activated, the DNA repair machinery induction was delayed by downregulating the transcription of these DNA repair genes (Fig. 4D). Growing evidence has shown that bortezomib is particularly toxic to multiple myeloma cells due to an increased susceptibility of multiple myeloma cells to ER stress-induced apoptosis (43). Treatment of multiple myeloma cells with bortezomib leads to induction of proapoptotic UPR components, including growth arrest and CHOP. A previous study has indicated that the major apoptotic regulator p53 was significantly increased in response to ER stress, and participates in ER stress-induced apoptosis (40, 44). In this regard, our findings are consistent with the previous studies that the antimyeloma activity of bortezomib was partly mediated by the induction and activation of p53, and downregulation of the expression of the DNA repair machinery (26, 27). Very recently, Lin and colleagues further showed that ER stress stimulates p53 expression through NF-κB activation (45). Work by Hideshima and colleagues showed that bortezomib actually activates two upstream NF-κB–activating kinases (RIP2 and IKKβ), promotes downregulation of NF-κB's inhibitor (IκBα), and increases NF-κB DNA binding both in vitro and in vivo (46). Taken together, it suggests that the induction and activation of p53 by bortezomib was mediated by ER stress rather than genotoxic stress.

In multiple myeloma cell lines with p53 mutations, we still observed the induction of PUMA and NOXA by TH-302 and bortezomib, strongly implicating the existence of p53-independent mechanisms. We found that the activation of CHOP by TH-302 and bortezomib through ER stress may be another upstream regulator. As shown in Fig. 4E and G, both TH-302 and bortezomib can trigger ER stress signaling, confirmed by upregulation of ER stress markers GRP-78 and CHOP. As an important ER stress downstream effector, the transcription factor CHOP works as a critical initiator of ER stress-induced apoptosis through upregulating the expression of proapoptotic BH3-only proteins PUMA and NOXA (47, 48). The mechanism how bortezomib induces ER stress has been well shown; however, the reasons why TH-302 can trigger both ER and genotoxic stresses have not been fully clarified. TH-302 consists of two distinct parts: a bis-alkylating effector and an oxygen concentration-activating trigger. The hypoxia-selective release of the Br-IPM may lead to protein alkylation in addition to DNA cross-linking, disturb the homeostasis of the ER, and induce ER stress.

In summary, our findings in the present study show that the combination of TH-302 and bortezomib has synergistic cytotoxicity in multiple myeloma, and the synergy is tightly related to the changes in the balance between proapoptotic and antiapoptotic Bcl-2 proteins favoring induction of apoptosis. Importantly, the experiments carried out in the in vivo murine 5T33MM model showed impressive improvements in multiple disease parameters and significantly prolonged survival after treatment with the combination. Taken together, our data provide a novel insight into the potential application of bortezomib and TH-302 for multiple myeloma, and provide support for further clinical evaluation of the combination of bortezomib and TH-302 for patients with multiple myeloma.

D. Handisides has ownership interest (including patents) in Threshold. No potential conflicts of interest were disclosed by the other authors.

Conception and design: J. Hu, D. Handisides, C.P. Hart, K. Vanderkerken

Development of methodology: J. Hu, E. Van Valckenborgh

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J. Hu, H. De Raeve, S. Xu, K. Vanderkerken

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J. Hu, D. Xu, E. Menu

Writing, review, and/or revision of the manuscript: J. Hu, E. Van Valckenborgh, E. Menu, H. De Raeve, E. De Bryune, B. Van Camp, C.P. Hart, K. Vanderkerken

Study supervision: K. Vanderkerken

The authors thank Prof. Mark De Ridder, Dr. Valeri N. Verovski, and Heng Jiang (Cancer Research Unit of Oncologic Center, Universitair Ziekenhuis Brussel, Brussel, Belgium) for their expertise and equipment used in the hypoxia experiments. The authors also thank A. Willems and C. Seynaeve for expert technical assistance, Prof. F. Gorus (Universitair Ziekenhuis Brussel) for analysis of serum electrophoresis.

This work was supported by Vlaamse Kankerliga (to K.Vanderkerken), FWO-Vlaanderen, Stichting Tegen Kanker (to K. Vanderkerken), Onderzoeksraad VUB (to K. Vanderkerken), and the Fundamental Research Funds for the Central Universities (No. 2012jdhz59; to J. Hu). E. Van Valckenborgh, E. Menu, E. De Bryune are post-doctoral fellows of FWO-Vlaanderen.

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.