Combination therapy with protein kinase inhibitor H89 and Tetrandrine elicits enhanced synergistic antitumor efficacy

The human breast cancer cell lines (MDA-MB-231, MDA-MB-468, and MCF-7) were purchased from ATCC (Manassas, VA, USA). The human hepatoma cell lines (Hep3B and Huh7) and the normal cell lines (L02, HBL-100, and HEK293T) were purchased from CCTCC (Wuhan, China). The cell line HCCLM9 was purchased from the Liver Cancer Institute (Fudan University, China). These various cell lines were cultured in DMEM. The human renal carcinoma cell lines (769-P, ACHN, and 786-O) and the human lung cancer cell line (A549) were purchased from ATCC and maintained in 1640 RPMI. The human colon cancer cell lines (LOVO and HCT116) were purchased from ATCC and cultured in McCoy’s 5A. All cells were cultured in media supplemented with 10% fetal bovine serum (FBS, HyClone), penicillin 100 U·mL^− 1 and 100 μg·ml^− 1 streptomycin and were incubated at 37 °C in a humidified atmosphere with 5% CO₂.

The reagents used in this study are listed as follows: H89 2HCl (10 mM, dissolved in DMSO, S1582, Selleck, Houston, TX), forskolin (FSK) (10 mM, dissolved in DMSO, S2449, Selleck, Houston, TX), PKI (14–22) amide myristoylated (PKI) (0.5 mg, dissolved in DMSO, Sc471154, Santa Cruz, CA), Tetrandrine (10 mM, dissolved in DMSO, CAS:518–34-3, Shanghai Ronghe Medical, Shanghai, China), Caspase inhibitor z-VAD-fmk (10 mM, dissolved in DMSO, S7023, Selleck, Houston, TX), Bafilomycin A1 (dissolved in DMSO, S1413, Selleck, Houston, TX), Chloroquine (CQ) (C6628, Sigma-Aldrich, USA), PD98059 (Beverly, MA, USA), N-acetyl-L-cysteine (NAC) (dissolved in ddH₂O, A0150000, Sigma-Aldrich, USA), 3-Methyladenine (3-MA) (dissolved in ddH₂O, M9281-100MG, Sigma-Aldrich, USA), 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA) (Invitrogen Carlsbad, CA), and an FITC Annexin V Apoptosis Detection Kit (556,547, BD Pharmingen). All reagents were formulated as recommended by their suppliers.

To measure viability following H89 and/or tetrandrine treatment, cells were seeded on 96-well plates at a density of 4 × 10³ cells per well. Cells were allowed to attach overnight in complete media and were subsequently treated with the indicated concentrations of tetrandrine and/or H89 for an additional 72 h. Control cells received DMSO (< 0.1%) that contained medium. The cell viability was determined using the trypan blue dye exclusion assay according to established protocols.

A drug combination analysis was performed using the method described by Chou and Talalay [31]. Briefly, cells were plated in 96-well plates and treated with 1–5 μM tetrandrine and 1–10 μM H89 alone or in combination for 72 h; the cell viability was subsequently assessed. Multiple drug dose-effect calculations and the combination index plots were generated using Calcusyn 2.1 software (Biosoft, Cambridge, UK). CI values < 1, = 1 and > 1 indicate synergism, additive and antagonism between two drugs, respectively.

Apoptosis was determined using an Annexin V-FITC/PI apoptosis detection kit (BD Biosciences, San Jose, CA, USA) according to the manufacturer’s instructions. Briefly, the untreated and treated cells were washed with PBS and gently suspended in Annexin V binding buffer, followed by incubation with Annexin V-FITC and PI at room temperature for 15 min. Finally, the fluorescent intensities were determined by a flow cytometer (Beckman Coulter, Indianapolis, CA, USA), and the data were analyzed using FlowJo software (Tree Star Inc., San Carlos, CA, USA).

Cell were seeded at 2500 cells per well in 6-well plates and treated with H89, tetrandrine or the combination. Cells were washed with fresh medium after 24 h, allowed to grow for 8 days under drug-free conditions, and stained with crystal violet (Sigma-Aldrich, USA). Colonies with more than 50 cells were counted.

After various treatments, both floating and adherent cells were harvested and subsequently washed with cold PBS. The cells were then lysed with 1% SDS on ice. The cell lysates were subsequently heated at 95 °C for 20 min and centrifuged at 12,000×g for 10 min, and the supernatant was collected. For the tumor tissue, the samples were homogenized and sonicated in RIPA buffer (Beyotime, Nantong, China) in the presence of protease inhibitor cocktail on ice. The tissue lysates were subsequently centrifuged at 12,000×g for 15 min at 4 °C, and the supernatant was collected for Western blotting analysis. Protein was quantified using a BCA Protein Assay Kit (Thermo Scientific, MA, USA). The proteins were separated by 8–12% SDS-PAGE and transferred to PVDF membranes (Millipore, Billerica, MA, USA). The blots were blocked for 2 h at room temperature with freshly prepared 5% nonfat milk (Bio-Rad, USA) in TBST and were subsequently incubated with specific primary antibodies overnight at 4 °C. The membranes were then washed with TBST and incubated with HRP conjugated secondary antibodies for 1 h at room temperature. After washing with TBST, the immunoblots were visualized by Immobilon™ Western HRP Substrate peroxide (Millipore, Billerica, MA, USA). The following antibodies were employed: anti-PARP (# 9542), anti-Caspase-3 (# 9662), anti-Caspase-9 (# 9502), anti-Bcl-2 (# 2872), anti-Bcl-xL (# 2762), anti-Mcl-1 (# 4572), anti-Bim (# 2819), anti-Bid (# 2002), anti-cytochrome c (# 4272), anti-p-ERK1/2 (Thr202/Tyr204, # 9101), anti-T-ERK (# 9102), p-MEK (# 9121S), anti-p-CREB1-s133 (# 9198), and anti-CREB1 (# 9197), which were obtained from Cell Signaling Technology. Antibodies for LC3B (# L7543) and anti-β-actin (# A2228) were obtained from Sigma-Aldrich. Antibody for c-Myc (9E10) (sc-40) was obtained from Santa Cruz Biotechnology.

The intracellular ROS levels were detected by a flow cytometer utilizing Dichloro-dihydro-fluorescein diacetate (DCFH-DA). Briefly, 7 × 10⁴ cells were plated on 12-well plates, allowed to attach overnight and then treated with the indicated concentrations of tetrandrine and/or H89 for the indicated times. NAC pretreatment, where indicated, was conducted for 1 h. Cells were stained with DCFH-DA (1 μM) in serum-free media at 37 °C for 30 min in the dark. DCF fluorescence (produced in the presence of ROS) was analyzed using flow cytometry. For the detection of the mitochondrial membrane potential, the cell pretreatments were performed as described for the ROS detection protocol. A 5 μL solution of 0.1 mg·mL^− 1 Rho123 (Sigma-Aldrich, USA) was added, and the cells were incubated for 30 min at 37 °C. The fluorescence was measured via flow cytometry.

Cells were transfected with the mRFP-GFP-LC3 plasmid (Addgene, # 21074) and were subsequently treated with DMSO or H89/tetrandrine. Autophagy flux was assessed by counting cells that were mRFP⁺GFP⁺LC3 (yellow puncta), which represents autophagosomes, and mRFP⁺GFP⁻LC3 (red puncta), which represents autolysosomes.

pLKO.1 plasmids that contained shRNA sequences targeting c-Myc (shc-Myc#1: target sequence: CCTGAGACAGATCAGCAACAA; shc-Myc#2: target sequence CAGTTGAAACACAAACTTGAA) were established. The empty vector was used as a negative control. PHAGE-puro-c-Myc and the control plasmids were kindly provided by Dr. Youjun Li (Wuhan University). A PHAGE-puro-Mcl-1 plasmid was constructed. To generate virus, 293 T cells were seeded in each 10 cm dish. After 24 h, the control and shRNA constructs that targeted c-Myc (12 μg each), packaging plasmid (psPAX2; 6 μg) and envelope plasmid (pMD2.G; 6 μg) were diluted with 0.5 mL Opti-MEM medium (Gibco, Invitrogen, Carlsbad, CA, USA) and mixed with transfection reagent FuGENE™ HD (Roche, USA). Twelve hours after transfection, the culture medium was changed to fresh culture medium; after 48 h, the virus-containing supernatants were collected and used to infect cells in the presence of 8 μg·mL^− 1 polybrene (Sigma-Aldrich, USA). The cells were subsequently selected in the presence of puromycin (5 μg·mL^− 1; Sigma-Aldrich, USA) to establish stable clones.

RNA was isolated from cultured cells using an OMEGA-RNA Miniprep kit, and RNA was reverse-transcribed into cDNA molecules using a cDNA synthesis kit (Roche Applied Science, USA). The numbers of Mcl-1 and β-actin molecules were monitored in real time on a 7500 Fast Real-Time PCR System (Applied Biosystems) by measuring the fluorescence increases of SYBR Green (Roche Applied Science, USA). The primer sequences for PCR were as follows: Mcl-1 forward 5’-CCAAGAAAGCTGCATCGAACCAT-3′ and Mcl-1 reverse 5’-CAGCACATTCCTGATGCCACCT-3′; β-actin forward 5’-GGCATGGGTCAGAAGGATT-3′ and β-actin reverse 5’-AGGATGCCTCTCTTGCTCTG-3′. To determine the relative abundance of Mcl-1 in relation to β-actin, the Δ-ΔCT (cycle threshold) method was utilized.

Animals were handled according to the Guidelines of the China Animal Welfare Legislation, as provided by the Committee on Ethics in the Care and Use of Laboratory Animals of Wuhan University. The experimental protocols were approved by the Experimental Animal Centre of Wuhan University. Female nude nu/nu BALB/c mice (14–16 g; 4–5 weeks of age) were purchased from Hunan SJA Laboratory Animal Co., Ltd. (Changsha, China). Animals were housed at a constant room temperature with a 12/12 h light/dark cycle and were fed a standard rodent diet.

MDA-MB-231 cells (5 × 10⁶), MDA-MB-231 control or Mcl-1 cells (5 × 10⁶), or MDA-MB-231 shCtrl or shc-Myc cells (5 × 10⁶) were subcutaneously implanted into the right flank of each mouse in 0.2 mL PBS. Once the tumor volume of MDA-MB-231 cells reached 50–100 mm³, the tumor-bearing mice were randomly separated into four groups (n = 6) and treated via gavage of 25 mg·kg^− 1tetrandrine with 0.5% sodium carboxyl methylcellulose, i.p. injection of 10 mg·kg^− 1 H89 (in a solution of PBS with 1% DMSO+ 30% polyethylene glycol+ 1%Tween 80) or a combination of H89 and tetrandrine (10 mg·kg^− 1 H89 and 25 mg·kg^− 1 tetrandrine) every other day for 28 days. The control group received the same vehicle. For the MDA-MB-231 control and Mcl-1 cells, tumor-bearing mice were randomized into two test groups (n = 6) and were administered vehicle (0.5% carboxymethylcellulose sodium) or H89 (10 mg·kg^− 1) and tetrandrine (25 mg·kg^− 1) every other day for 36 days. For the MDA-MB-231 shCtrl and shc-Myc cells, tumor-bearing mice were randomized into two test groups (n = 6) and were administered vehicle (0.5% carboxymethylcellulose sodium) or H89 (10 mg·kg^− 1) and tetrandrine (25 mg·kg^− 1) every other day for 30 days. The tumor volumes were determined by measuring the length (l) and width (w) and calculating the volume (V = 0.5 × l × w²) every other day. At the end of the experiments, the mice were euthanized after being anesthetized with an i.p. injection with pentobarbital (50 mg·kg^− 1); their tumors were isolated by dissection, weighed and used for in vitro experiments. Samples were prepared for histology and protein assays. All studies involving animals are reported in accordance with the ARRIVE guidelines for reporting experiments involving animals [32].

Tumor samples from mice were homogenized and sonicated. Tissue lysates were subsequently centrifuged at 12000×g for 10 min at 4 °C to collect the supernatant. The total protein content was determined using the Bradford assay. The MDA levels were measured by the Lipid Peroxidation MDA assay kit (Beyotime Institute of Biotechnology).

All experiments were randomized and repeated at least three times. Data analysis was performed using Microsoft Excel and GraphPad Prism Software version 5.0 (GraphPad Software, La Jolla, CA, USA). All data are expressed as the mean ± SD. Student’s two-tailed t-tests were performed to calculate P values unless otherwise specified. P < 0.05 was considered statistically significant.

To examine whether a cooperative effect exists between H89 and tetrandrine in tumor chemotherapy, we treated cancer cells with a series of concentrations of H89 and tetrandrine alone or in combination. Dose-response studies indicated that 0–5 μM tetrandrine or 0–10 μM H89 were only minimally toxic by themselves (Fig. 1, a and b). However, 4 μM tetrandrine substantially increased the cell death of H89 at 6–10 μM (Fig. 1c). Analogously, the lethality of the marginally toxic tetrandrine concentrations (3–5 μM) was significantly increased by co-exposure to 6 μM H89 (Fig. 1d). In addition, we determined that 4 μM tetrandrine plus 6 μM H89 had a clear effect on most human cancer cells (Fig. 1e). In contrast, normal cells, such as HBL-100, L02 and HEK-293 T, were substantially less sensitive to H89/tetrandrine, which suggests that the H89/tetrandrine combined treatment is relatively selective towards cancer cells (Fig. 1f). The use of the Chou-Talalay method to calculate the H89/tetrandrine interaction showed combination indexes < 1, which indicated their synergistic effects under a number of the treatment conditions (Fig. 1g). Furthermore, the analysis of long-term cell survival via colony formation assay indicated that the combination of H89 and tetrandrine suppressed colony formation significantly more than tetrandrine or H89 alone in cancer cells, whereas the effect on normal cells was not significant (Fig. 1h). These data suggest that a combination treatment of H89 and tetrandrine has a significant synergistic anti-tumor effect on cancerous cells.

To determine whether the reduction in cell survival induced by H89/tetrandrine was associated with cell apoptosis, cancer cells were treated with H89 and tetrandrine alone or in combination for 48 h and were subsequently analyzed via flow cytometry with Annexin V/PI. As shown in Fig. 2a and Additional file 1: Figure S1A, H89/tetrandrine combination treatment increased apoptotic cell death compared with each agent alone. Consistently, apoptosis was also evident from Western blots, which indicated that the cleavages of PARP, caspase-3, caspase-9 and cytochrome c were substantially increased in the combination treatment (Fig. 2b). To ascertain whether H89/tetrandrine induced cell apoptosis was dependent on caspase activation, we pre-treated cells with z-VAD-fmk for 1 h. The results showed that z-VAD-fmk partially rescued H89/tetrandrine induced cell death (Fig. 2c). Our previous studies have shown that tetrandrine is a potent autophagy agonist. We subsequently examined whether the H89/tetrandrine combination synergistically induced autophagy. Western blot and GFP-LC3 fluorescence assays showed that the H89/tetrandrine combination treatment synergistically increased autophagy (Fig. 2d and e, and Additional file 1: Figure S1B). In addition, we used an mRFP-EGFP-LC3 tandem-tagged fluorescent protein (ptf-LC3) to determine the dynamic process of autophagic flux [33, 34]. The results suggested that the combination of H89/tetrandrine promotes functional autophagy in cancer cells (Fig. 2f). Next, to investigate the role of autophagy in cell death, we pre-treated cells with 3-MA or bafilomycin A1 for 1 h. As shown in Fig. 2g, the autophagy inhibitor also partially inhibited H89/tetrandrine induced cell death. Collectively, these data indicate that apoptosis and autophagy act simultaneously in H89/tetrandrine induced cancer cell death.

Increasing evidence suggests that chemotherapeutic agents activate intracellular reactive oxygen species (ROS) accumulation and subsequently induce cell death [35, 36]. We assessed the intracellular ROS levels via flow cytometry with DCFH-DA, which specifically detects peroxides. As shown in Fig. 3a, the H89/tetrandrine combination treatment substantially increased intracellular ROS compared with either agent alone in cancer cells. Pretreatment of cells with the ROS scavenger NAC not only strikingly abrogated H89/tetrandrine-induced ROS production (Fig. 3b) but also significantly rescued them from cell death induced by combination treatment (Fig. 3c). Moreover, NAC rescued cells from apoptosis (Fig. 3d) and blocked both PARP cleavage and caspase 3/caspase 9 activation in AGS and MDA-MB-231 cells (Fig. 3e). In addition, pretreatment with NAC also significantly decreased the intracellular LC3-II expression and GFP-LC3 fluorescent punctate dots induced by H89/tetrandrine (Fig. 3f and g). Therefore, these results clearly demonstrated that the H89/tetrandrine combination treatment-induced apoptosis and autophagy were initiated by ROS production.

H89 is a strong inhibitor of cyclic AMP-dependent protein kinase A (PKA) [37]. To determine whether PKA activity is related to synergistic antitumor activity, we initially detected the levels of phosphorylated cAMP-dependent response element (CREB ser133). As shown in Fig. 4a, H89 alone or combined with tetrandrine significantly decreased the phosphorylation of the transcription factor CREB in cancer cells. In contrast, forskolin (FSK), an established activator of adenylyl cyclase, significantly reduced the proportion of cells undergoing apoptosis and autophagy (Fig. 4b and c, and Additional file 1: Figure S2, A and B). Our previous studies have shown that tetrandrine activates intracellular ERK [38]. We subsequently examined the role of MEK/ERK signaling in H89/tetrandrine combination treatment. As shown in Fig. 4d, H89/tetrandrine transiently activated MEK/ERK1/2 signaling, and these effects mainly originated from tetrandrine (Additional file 1: Figure S2C). Inhibition of ERK1/2 with PD98059 could partially rescue H89/tetrandrine induced cell apoptosis (Fig. 4e and f), whereas it had no effect on autophagy (Fig. 4g). However, concomitant pretreatment of cells with FSK and PD98059 almost completely restored cell viability, which implies that PKA and ERK signaling are simultaneously involved in H89/tetrandrine induced cell death (Fig. 4h). Moreover, NAC could mitigate H89/tetrandrine regulated PKA and ERK activities, whereas FSK and PD98059 had no effect on H89/tetrandrine increased ROS (Fig. 4i and j), which indicates that PKA and ERK signaling are involved in H89/tetrandrine induced cell death and are initiated by ROS. Therefore, we demonstrated that the inhibition of PKA and activation of ERK activity play a role in H89/tetrandrine induced cell death.

Bcl-2-family proteins play a role in chemotherapy induced cell death, including apoptosis and autophagy [39]. We first detected the Bcl-2, Bcl-xl and Mcl-1 expression levels by Western blotting after AGS and MDA-MB-231 cells underwent treatment with H89 and tetrandrine alone or in combination for 48 h. As shown in Fig. 5a, the expression of Mcl-1 was significantly decreased after H89/tetrandrine treatment in both AGS and MDA-MB-231 cells. Consistent with its protein level, the Mcl-1 mRNA expression was downregulated by H89/tetrandrine (Fig. 5b). We subsequently assessed the role of Mcl-1 in H89/tetrandrine mediated cell death. AGS and MDA-MB-231 were established stably expressing a control or Mcl-1 and were subsequently treated with H89/tetrandrine. In the presence of H89/tetrandrine, overexpression of Mcl-1 diminished cell sensitivity to the combined treatment (Fig. 5c) and attenuated cell apoptosis (Fig. 5d and e); however, it had no effect on autophagy (Fig. 5g). As an anti-apoptotic Bcl-2 family member, Mcl-1 appeared to inhibit apoptosis by preventing mitochondrial dysfunction [40]. We determined that overexpression of Mcl-1 significantly rescued the H89/tetrandrine-induced mitochondrial transmembrane potential drop (Fig. 5f). In addition, NAC could partially impede the H89/tetrandrine triggered Mcl-1 decrease (Fig. 5h); however, Mcl-1 overexpression had no significant effect on the ROS increases (Fig. 5i), which demonstrates that the induction of intracellular ROS occurs upstream of the inhibition of Mcl-1 activity. Therefore, we concluded that the activity of the anti-apoptotic protein Mcl-1 considerably contributes to H89/tetrandrine induced cell death.

Deregulated and enhanced expression of c-Myc contributes to the genesis of a substantial fraction of human tumors [41]. Paradoxically, high expression levels of c-Myc not only promote transformation but also sensitize cells against a broad range of pro-apoptotic stimuli, such as growth factor deprivation, hypoxia, ionizing radiation, or exposure to chemotherapy [42‐44]. Consistently, in this study, we determined that H89/tetrandrine sensitive cells showed higher levels of c-Myc expression than resistant cells (Fig. 6a, Fig. 1e, and Additional file 1: Figure S3A). To further verify this observation, we ectopically expressed c-Myc in HCT116 and A549 cells, which express low levels of c-Myc and are resistant to H89/tetrandrine treatment. Interestingly, the overexpression of c-Myc enhanced the effects of combination treatment induced cell death and apoptosis (Fig. 6b and c). In contrast, the knockdown of c-Myc in AGS and MDA-MB-231, which express high levels of c-Myc, decreased the sensitivity of cells to H89/tetrandrine (Fig. 6d-f). Importantly, we determined that c-Myc significantly regulated Mcl-1 expression (Fig. 6g). Furthermore, we determined that knockdown of c-Myc in AGS and MDA-MB-231 cells, cell autophagy, and intracellular ROS generation triggered by H89/tetrandrine combination treatment were suppressed (Fig. 6h and i). In contrast, ectopically expressed c-Myc in HCT116 and A549 cells significantly elevated intracellular ROS (Additional file 1: Figure S3B). Together, these data indicate that c-Myc amplification downregulates Mcl-1 expression and increases intracellular ROS, which contributes to H89/tetrandrine sensitivity.

To assess the therapeutic efficacy of H89 and tetrandrine combination therapy in vivo, we established MDA-MB-231 subcutaneous tumor xenograft models with female athymic nude mice. After six days, tumor-bearing mice were randomly assigned to four groups and were administered vehicle, H89 (10 mg·kg^− 1), tetrandrine (25 mg·kg^− 1) or the combination treatment every other day for 28 days. At 18 days, treatment with H89 or tetrandrine inhibited the growth of tumor xenografts; however, there was an enhanced effect in the combination group (Fig. 7a). Consistent with the tumor volumes, the mean tumor weights substantially decreased in the H89/tetrandrine group compared with the other groups (Fig. 7b). Notably, the body weight measurement indicated that the dose of H89/tetrandrine combination was tolerable to the animals (Additional file 1: Figure S4A). Furthermore, TUNEL assays showed significant apoptosis in the tumor tissue from the animals treated with H89/tetrandrine (Fig. 7c). The level of lipid peroxidation product MDA, which indicates the level of oxidative stress in tissues, was increased in the H89/tetrandrine group (Fig. 7d). Consistent with the observations in vitro, H89/tetrandrine increased the levels of cleaved PARP and LC3-II and decreased the levels of Mcl-1 in vivo (Fig. 7e). Next, to examine the role of Mcl-1 in H89/tetrandrine combination treatment in vivo, we performed xenograft assays in MDA-MB-231 ectopically expressing Mcl-1. The tumor regression in response to H89/tetrandrine was statistically significant in the control tumors, but not in the Mcl-1 overexpressing tumors (Fig. 7f and g). Tumor lysates showed the Mcl-1 overexpression group exhibited decreased cleavage of PARP in response to H89/tetrandrine treatment (Fig. 7h). Again, the combination therapy was well tolerated by evidence of weight sustainability (Additional file 1: Figure S4B). To be consistent with the previous experiment, we continued to experiment with MDA-MB-231 and established xenograft models using c-Myc knockdown MDA-MB-231 cells. The in vivo results showed that c-Myc depletion exhibited resistance to H89/tetrandrine treatment (Fig. 7i and j). Consistent with the in vitro findings, c-Myc knockdown increased the Mcl-1 expression in vivo (Fig. 7k). The combination-treated mice did not exhibit a reduction in weight gain during the treatment period (Additional file 1: Figure S4C).