Combination treatment with doxorubicin and gamitrinib synergistically augments anticancer activity through enhanced activation of Bim

Gamitrinib conjugated with triphenylphosphonium was prepared as described previously [10]. MitoTracker, JC-1, and tetramethylrhodamine methyl ester (TMRM) were purchased from Molecular Probes. All other chemicals were purchased from Sigma.

Human cancer cell types that originated from ovary (SK-OV3), prostate (22Rv1 and PC3), cervix (HeLa), breast (MDA-MB-231), liver (SK-HEP-1), brain (A172), kidney (ACHN), and lung (NCI-H460) were purchased from the Korean Cell Line Bank. Cells were cultured in DMEM or RPMI (GIBCO) medium containing 10% FBS (GIBCO) and 1% penicillin/streptomycin (GIBCO) at 37°C in a 10% CO₂ humidified atmosphere.

Small interfering RNAs (siRNAs) against JNK and CHOP were synthesized by Genolution Inc (Korea). siRNA sequences used in this study were as follows:

JNK1-#1, 5′-AAAGAATGTCCTACCTTCTCT-3′; JNK1-#2, 5′-AAGCCCAGTAATATAGTAGTA-3′; CHOP-#1, 5′-AGAACCAGCAGAGGTCACAA-3′; CHOP-#2, 5′-AAGAGAATGAACGGCTCAAGC-3′; Bim-#1, 5′-GCAACCTTCTGATGTAAGT-3′; Bim-#2, 5′-GACCGAGAAGGTAGACAATT-3′ and control, 5′-ACUCUAUCUGCACGCUGAC-3′. Cells were cultured on 6-well plates to 50–75% confluence, transfected with 40 nM siRNA mixed with G-Fectin (Genolution) for 48 hours, and then analyzed or treated with drugs.

Mitochondrial fractionation from cultured cells was performed with a Mitochondrial Isolation kit (Thermo Scientific) as described in the manufacturer’s instructions. For western blot analysis, proteins were separated on 8-12% SDS-polyacrylamide gels and transferred to polyvinyl difluoride membranes (Millipore). Primary antibodies were diluted 100–5,000-fold, and horseradish peroxidase-conjugated mouse or rabbit secondary antibodies (KLP Inc.) were diluted 5,000-fold. The ECL reagent (GE Healthcare) was used for chemiluminescence detection with a LAS 4000 imager (GE Healthcare).

All experiments involving animals were approved by the Ulsan National Institute of Science and Technology Animal Care and Use Committee (approval number: UNISTIACUC-12-003-A). Cancer cells (7 × 10⁶ 22Rv1 or 1 × 10⁷ MDA-MB-231) were suspended in sterile 200 μl PBS. 22Rv1 cells were injected subcutaneously into both flanks of 8-week-old BALB/c nu/nu male mice (Japan SLC Inc.). MDA-MB-231 cells were orthotopically injected into the mammary fat pad of 8-week-old BALB/c nu/nu female mice. Gamitrinib or vehicle (DMSO) dissolved in 20% Cremophor EL (Sigma) in PBS was injected intraperitoneally (i.p.), and DOX diluted in PBS was injected intravenously (i.v.). The mice were treated with 10 mg/kg gamitrinib and/or 3 mg/kg DOX twice a week according to the group. Tumors were measured daily with a caliper and tumor volume was calculated using the formula V = 1/2 × (width)² × length. At the end of the experiment the animals were euthanized, and organs including brain, heart, kidney, liver, lung, spleen, stomach, intestine, and testis, and tumors were collected for histologic or western blot analyses. Blood was also collected for measurement of serum creatine phosphokinase activity using the Indiko and Konelab System CK (Thermo Scientific) according to the manufacturer’s instructions.

Total RNA was prepared from cells suspended in cold PBS using the RNeasy mini kit (QIAGEN), and cDNA was synthesized using the ProtoScript® First Strand cDNA Synthesis Kit (New England Biolabs) using an oligo(dT) primer. The PCR reaction was performed in a Mastercycler PCR machine (Eppendorf) with the following sets of oligonucleotide primers: NOXA, 5′-GTGCCCTTGGAAACGGAAGA-3′ and 5′-CCAGCCGCCCAGTCTAATCA-3′; PUMA, 5′-CAGACTGTGAATCCTGTGCT-3′ and 5′-ACAGTATCTTACAGGCTGGG-3′; DR5, 5′-TGCAGCCGTAGTCTTGATTG-3′ and 5′-GAGTCAAAGGGCACCAAGTC-3′; Bcl-2, 5′-TTTTAGGAGACCGAAGTCCG-3′ and 5′-AGCCAACGTGCCATGTGCTA-3′; Bim, 5′- ATGGCAAAGCAACCTTCTGA-3′ and 5′-GGAAGCCATTGCACTGAGA-3′; CHOP, 5′-CTTTCTCCTTCGGGACACTG-3′ and 5′-AGCCGTTCATTCTCTTCAGC-3′ GAPDH, 5′-GGGAAGCTTGTCATCAATG-3′ and 5′-GCAGTGATGGCATGGACT-3′.

Data from MTT assay (triplicate experiments independently repeated at least two times) were averaged and statistically analyzed by unpaired t-test using Prism 5.0 (GraphPad). A p-value less than 0.05 was considered significant. To investigate the synergistic efficacy of the drug combination, the combination index (CI) was determined according to the Chou-Talalay method using CalcuSyn software version 2.1 (Biosoft) [23].

To investigate the effect of combination treatment with DOX and gamitrinib, cancer cell lines originating from cervix (HeLa), ovary (SK-OV3), and prostate (22Rv1) were treated with the drugs as single agents or in combination. Gamitrinib sensitized HeLa, SK-OV3, and 22Rv1 cells to a wide range of DOX concentrations (Figure 1A) and the drug combination resulted in cancer cell death at suboptimal concentrations (Additional file 1: Figure S1A). In contrast to the effect on cancer cells, gamitrinib did not sensitize cardiomyocytes to DOX treatment (Figure 1B). Next, we calculated the combination index (CI) using the Chou-Talalay method [23]. The DOX and gamitrinib combination showed synergistic anticancer activity (CI < 0.9) in all cancer cell types tested at a 50% effective dose: high synergism (CI < 0.7) for HeLa (Figure 1C), A172 (glioblastoma), ACHN (renal cell carcinoma), SK-HEP-1 (hepatocellular carcinoma), NCI-H460 (lung carcinoma), and SK-OV-3; and moderate synergism (0.7 < CI < 0.9) for 22Rv1 and MDA-MB-231 cells (Table 1).

Single drug treatment at a suboptimal concentration did not increase caspase activity significantly compared with the DMSO-treated control, whereas the drug combination dramatically increased caspase activity and concomitant cell death in 22Rv1 and MDA-MB-231 cells (Figure 2A). Consistently, cell extracts from 22Rv1 and SK-OV3 showed caspase activation for the combined drug treatment but not for single agent treatment (Figures 2B; Additional file 1: Figure S1B). Activation of caspase by the drug combination was completely inhibited by the pan-caspase inhibitor zVAD-fmk in 22Rv1 (Figure 2B). In addition, the enhanced cytotoxic activity of the drug combination was significantly abrogated by zVAD-fmk (Figures 2C; Additional file 1: Figure S1C). These data suggest that the drug combination activates caspases and triggers the apoptotic cell death program.

We investigated the effect of combined DOX and gamitrinib on stress signaling and found that c-Jun N-terminal kinase (JNK) activation (phosphorylation) and C/EBP homologous protein (CHOP) induction were increased more by the drug combination than by single agent treatment in 22Rv1 cells (Figure 3A). Treatment with CHOP-specific siRNA reduced the enhanced cytotoxicity after combined drug treatment but did not affect cytotoxicity of single agent treatment (Figure 3B), whereas JNK-specific siRNA did not affect the cytotoxic activity of the drugs singly or in combination (Additional file 1: Figure S2A). These data suggest that, in 22Rv1 cells, induction of CHOP is required for the combination effect of the drugs. As a downstream effector, the expression of Bim was enhanced by the drug combination in 22Rv1 cells and knockdown of CHOP compromised the up-regulation of Bim (Figure 3C). Another prostate cancer cell PC3 similarly showed synergistic induction of cell death and enhanced expression of CHOP and Bim by the drug combination (Additional file 1: Figure S2B). DOX alone slightly increased expression of Bim (Figure 3C; Additional file 1: S2B), which can occur through a CHOP-independent mechanism as previously reported [24, 25]. Inactivation of Bim by siRNA treatment compromised the enhanced cytotoxicity of the drug combination (Figure 3D). Bim transcription was elevated, but expression of other Bcl-2 family proteins, such as Bcl-2, Puma, and Noxa, was not affected by the drug combination (Figure 3E). Expression of death receptor 5 (DR5) was elevated by the drug combination in a CHOP-dependent manner (Figure 3E) as described previously [26], while procaspase-8, recruited to the death inducing signaling complex (DISC) after DR5 activation [27], was not cleaved at all (Figure 3A). These data further suggest the crucial role of Bim expression in the drug combination.

In contrast to 22Rv1 cells, the drug combination did not induce CHOP expression in HeLa and MDA-MB-231 cells (Figure 4A). In HeLa cells, knockdown of JNK compromised the enhanced cytotoxicity of the drug combination (Figure 4B) whereas knockdown of CHOP did not affect the drug synergism (Additional file 1: Figure S2C). These data suggest that there are context-dependent disparate stress responses to the drug combination, and the JNK pathway, but not CHOP, can be critically involved in the synergistic combination effect in certain cancer cell types. Previous studies have shown that JNK can activate Bim through phosphorylation to trigger Bax-dependent mitochondrial apoptosis [28, 29]. Phosphorylated Bim was detected in HeLa and MDA-MB-231 cells, but not in 22Rv1 (Figures 3C and 4A). Bim protein in HeLa cells was reduced by the drug combination, but the proteasomal inhibitor MG132 largely elevated the Bim protein level (Additional file 1: Figure S3A). Similarly, CHOP depletion in the drug combination was also recovered by MG132 treatment (Additional file 1: Figure S3A). The data suggest that turnover rates of CHOP and phosphorylated Bim can be higher in certain cancer cell types as a result of proteasomal degradation [30, 31]. Treatment with the drug combination caused increased accumulation of Bim and Bax in the mitochondria, which was significantly reduced by the JNK inhibitor SP600125 (Figure 4C; Additional file 1: Figure S3B) [32]. The Bim that accumulated in mitochondria was a slow-migrating phospho-form of the protein (Figure 4C, right panel). The combination effect was compromised by treatment with Bim siRNA (Figure 4D), further supporting the important role of the proapoptotic Bcl-2 protein in the drug-induced stress response [33].

We examined the anticancer activity of gamitrinib in the presence of DOX using a prostate cancer xenograft model with the hormone-independent relapsed human prostate cancer cell line 22Rv1 [34]. Single treatment with gamitrinib or DOX resulted in a slight reduction in tumor volume, whereas combination treatment dramatically suppressed tumor growth (Figure 5A). Because DOX is frequently used to treat early and metastatic breast cancers in the clinic [35], we tested the effect of the drug combination on an orthotopic xenograft model with the triple negative (lack of ER, PR, and HER2 expression) metastatic breast cancer cell line MDA-MB-231 [36]. The tumor growth was strongly inhibited by the drug combination but not by single agent treatment (Figure 5B). Histologic analysis did not show any prominent differences among the groups (Additional file 1: Figure S4) except for the heart: a cardiotoxic phenotype of cytoplasmic vacuolization for DOX alone and in combination treatment (Figure 5C). The serum creatine phosphokinase (CPK) level was measured as an index of cardiotoxicity [37] at the end of the experiment (Figure 5D). DOX treatment alone increased the level of CPK, but there was no significant further increase in CPK levels by the drug combination (Figure 5D). These data suggest that the DOX-induced cardiotoxicity is not aggravated by the drug combination in vivo. The mechanism underlying the activity of the drug combination in vivo was consistent with the in vitro data: the drug combination synergistically increased the phosphorylation of JNK in whole tumor tissue extracts (Figure 5E) and the accumulation of Bim and Bax in mitochondrial fractions (Figure 5F).