Abstract
Background/Aim: Most glioblastoma multiforme (GBM) patients experience tumor recurrence due to their resistance to therapy, especially temozolomide (TMZ). Our previous studies showed the potential of honokiol, a small biphenolic molecule, for treatment of GBM. Recently, the roles of androgen receptor (AR) in GBM malignance were further identified. This study was aimed to further evaluate the effects of enzalutamide, an inhibitor of AR, on honokiol-induced killing to drug-resistant glioblastoma cells and the possible mechanisms.
Materials and Methods: TMZ-resistant human and mouse glioblastoma cells, prepared from respective drug-sensitive cells, were pretreated with enzalutamide and then exposed to honokiol. Death events and apoptotic mechanisms were subsequently determined.
Results: AR was immunodetected in TMZ-sensitive and -resistant glioblastoma cells. Pretreatment of human TMZ-resistant glioblastoma cells with enzalutamide enhanced honokiol-induced DNA fragmentation and cell apoptosis. As to the mechanisms, combined treatment of enzalutamide and honokiol selectively stimulated release of cytochrome c from mitochondria and succeeding activation of initiator caspase-9. Furthermore, the honokiol-induced augmentation of proapoptotic Bak production, disruption of the mitochondrial membrane potential, and activation of executioner caspase-3 were significantly enlarged by enzalutamide. Interestingly, suppressing caspase-9 activity concurrently attenuated the enhanced effects of enzalutamide on honokiol-triggered apoptotic insults. Moreover, the synergistic effects of enzalutamide and honokiol were further confirmed in mouse drug-resistant glioblastoma cells.
Conclusion: Suppressing AR activity by enzalutamide improved honokiol-induced apoptosis of TMZ-resistant glioblastoma cells via an intrinsic Bak-mitochondrion-cytochrome c-dependent caspase cascade pathway. Combined treatment of enzalutamide and honokiol could represent a promising adjuvant strategy for treatment of drug-resistant glioblastomas.
Introduction
Glioblastoma multiforme (GBM) is the most common and aggressive brain tumor in adults. The global annual incidence of GBM is about 3.2 cases per 100,000 population (1). In Taiwan, the incidence of GBM is around 3.3 per 100,000 per year with an average annual 0.1% change of the standardized incidence (2, 3). Traditionally, GBM patients receive surgical tumor removal and followed by concurrent chemoradiotherapy (CCRT) (4). Unfortunately, the median overall survival of GBM patients who have been treated with standard therapy are only 12-15 months, and their 5-year survival rate is even less than 10%. Most (50-90%) of GBM patients experience tumor recurrence due to their resistance to therapy, especially drug resistance (5). Temozolomide (TMZ), an alkylating reagent, is the first-line chemotherapeutic drug for treatment of GBM (6). TMZ can trigger massive methylation of chromosomal DNA, eventually resulting in DNA fragmentation and cell apoptosis (7). However, molecular mechanisms of drug resistance in GBM are multifarious and known little till now. Consequently, how to overcome drug resistance is an urgent and challenging issue in GBM therapy.
Honokiol (3′,5-di-(2-propenyl)-1,1′-biphenyl-2,4′-diol, C18H18O2), a small biphenolic molecule derived from the bark of Magnolia officinalis (Houpo). Honokiol and magnolol are isomer and also identified as bioactive ingredients of Houpo (8). A series of studies done in our laboratory have shown the potential of honokiol to be clinically applied for treatment of GBM. At first, we demonstrated that honokiol can traverse across the blood-brain barrier (BBB) in vitro and in vivo (9). Subsequently, the preclinical effects of honokiol on suppression of GBM growth by triggering autophagic and apoptotic death of glioblastoma cells were confirmed (10, 11). In drug-resistant glioblastoma cells, honokiol selectively stimulated activation of caspase-9 and cell apoptosis (12). Recently, we showed the synergistic effects of honokiol with camptothecin on killing human glioblastoma cells (13, 14). Lin et al. used in vitro and animal models to prove that hydroxyapatite-honokiol particles can work as a drug delivery system for treatment of GBM (15). More fascinatingly, a case report showed an efficient and safe response of liposomal honokiol to a GBM patient with recurrent glioblastomas (16). All of these studies exhibit the potence of honokiol for treatment of drug-sensitive and -resistant GBM patients.
Androgen receptor (AR), a member of the steroid hormone nuclear receptor family, is activated following binding with androgens. Following initiation, AR can trigger consecutive intracellular signals for development and maintenance of reproductive, cardiovascular, hemopoietic, musculoskeletal, immune, and neural systems (17). Moreover, the AR-mediating signals are involved in tumorigenesis of prostate, esophagus, ovary, bladder, lung, liver, head and neck, and kidney (18). In GBM, our previous study showed the contribution of the testosterone-AR-PARD3B signaling axis to tumorigenesis and malignance of GBM (19). Remarkably, targeting AR signaling has been potentially applied to treat multiple types of cancers, especially prostate cancer (20, 21). Enzalutamide, a nonsteroidal antiandrogen medication, is a specific inhibitor of AR that is traditionally used for therapy of prostate cancer (22). Our previous study demonstrated the effects of enzalutamide to induce apoptosis of human TMZ-sensitive and -resistant glioblastoma cells via an intrinsic apoptosis pathway (23). There are two distinguished intrinsic mitochondrion- and extrinsic Fas ligand/death receptor-dependent pathways involved in regulation of cell apoptosis (24). Specifically, upstream initiator caspase-8 and -9 play key roles in response to intracellular and extracellular apoptotic signals, respectively (12, 24). In an intracellular apoptosis pathway, augmentation and translocation of proapoptotic Bak, one of Bcl-2 family proteins, from the cytoplasm to the outer membrane of mitochondria induces continuous programmed cell death events, including disruption of the mitochondria membrane potential (MMP), cytochrome c release, caspase-9-initiated activation of downstream executioner caspases-3 and-6, DNA fragmentation, and cell collapse (25). Targeting these intrinsic and extrinsic apoptosis signals have been widely investigated for cancer therapy (26). Based on our previous findings, this study was aimed to further evaluate the effects of AR suppression by enzalutamide on honokiol-induced insults to drug-resistant glioblastoma cells and the possible mechanisms.
Materials and Methods
Selection and preparation of TMZ-resistant human and mouse glioblastoma cells. Human U87MG and mouse GL261 glioblastoma cells, purchased from American Type Culture Collection (Manassas, VA, USA), were seeded in Dulbecco’s modified Eagle medium (DMEM; Gibco-BRL Life Technologies, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS), L-glutamine (2 mM), penicillin (100 IU/ml), streptomycin (100 mg/ml), sodium pyruvate (1 mM), and nonessential amino acids (1 mM) at 37°C in a humidified atmosphere of 5% CO2. TMZ-resistant human U87MG-R and mouse GL261-R cells were selected and prepared from drug-sensitive U87MG and GL261 cells as described previously (28). Briefly, U87MG and GL261 glioblastoma cells (105 cells) were treated with 50 μM TMZ with a purity >98% obtained from Sigma-Aldrich (St. Louis, MO, USA) for 48 h. Cells were harvested, diluted, and seeded in rich DMEM containing 100 μM TMZ. The colonies of TMZ-resistant U87MZ-R and GL261-R glioblastoma cells were further sub-cultured. To maintain drug resistance, U87MG-R and GL261-R glioblastoma cells were cultured in rich DMEM containing 100 μM TMZ. After exposure to 100 μM TMZ for 72h, viabilities of U87MG and GL261 cells were significantly decreased by 54% and 55%, respectively. In comparison, treatment of U87MG-R and GL261-R cells to 100 μM TMZ for 72 h did not affect cell viability. Thus, the resistance was validated by evaluating the effects of TMZ on viabilities of U87MG-R and GL261-R cells.
Drug treatment. Enzalutamide and honokiol with purities >98%, purchased from Sigma-Aldrich, were freshly dissolved in dimethyl sulfoxide (DMS). TMZ-resistant human U87MG-R and mouse GL261-R glioblastoma cells were pretreated with 12.5, 25, or 50 μM enzalutamide for 1 h and then exposed to 25 μM honokiol for 72 h. The death events were successively evaluated.
Assays of cell viability and survival. Toxicities of TMZ, enzalutamide, and honokiol to drug-resistant human and mouse glioblastoma cells were assayed as described previously (28). After drug treatment, cell viability and cell survival were examined using a colorimetric method and a trypan blue exclusion technique, respectively.
Quantification of DNA fragmentation. Damage of chromosomal DNA was analyzed using an enzyme-linked immunosorbent assay (ELISA) kit (Boehringer Mannheim, Indianapolis, IN, USA) (29). Briefly, after drug treatment, the BrdU-labeled DNA in TMZ-resistant glioblastoma cells was measured using a microplate photometer (Anthos Labtec Instruments, Wals/Salzburg, Austria) at a wavelength of 450 nm.
Quantification of apoptotic cells. Apoptotic cells were measured with flow cytometry as described previously (30). Briefly, following drug treatment, TMZ-resistant human and mouse glioblastoma ells were harvested and fixed in cold 80% ethanol. Fixed cells were then stained with propidium iodide (PI) and analyzed using a flow cytometer (FACScan, Beckman Coulter, Fullerton, CA, USA).
Assays of caspases-3, -8, and -9. Activities of caspases-3, -8, and -9 were assayed using fluorogenic methods as described previously (12). After drug treatment, cell lysates were prepared and incubated with specific fluorogenic peptide substrates, Asp-Glu-Val-Asp (DEVD), Ile-Glu-Thr-Asp (IETD), and Leu-Glu-His-Asp (LEHD) of caspases-3, -8, and -9, respectively, and the fluorescent intensities were analyzed with a fluorescent spectrophotometer. To suppress caspase-9 activity, cells were pretreated with Z-LEHD-FMK, an inhibitor of caspase-9, for 1 h and then exposed to enzalutamide and honokiol.
Analysis of MMP. The potential of mitochondrial membrane in human and mouse TMZ-resistant glioblastoma cells was examined using flow cytometry (16). After drug treatment, drug-resistant U87MG-R and GL261-R glioblastoma cells were harvested and incubated with 3,30-dihexyloxacarbocyanine (DiOC6), a positively charged dye, at 37 °C for 30 min. The cell pellets were prepared for analysis of flow cytometry (Beckman Coulter).
Immunoblotting analyses. Levels of AR, cytochrome c, Fas ligand, Bak, and β-actin proteins in human TMZ-resistant glioblastoma cells were immunodetected (13). After drug treatment, cell lysates of human TMZ-resistant glioblastoma cells were prepared. Protein concentrations were measured with a bicinchonic acid protein assay kit (Thermo Fisher Scientific, San Jose, CA, USA). Cell lysates (100 μg/ml) were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis. After being transferred to nitrocellulose membranes, levels of AR, cytochrome c, Fas ligan, and Bak were immunodetected using rabbit polyclonal antibodies (Cell Signaling Technology, Danvers, MA, USA). β-Actin was analyzed with a polyclonal antibody (Cell Signaling) as an internal loading control. Immunoreactive protein signals were detected using an enhanced chemiluminescence reagent (PerkinElmer, Waltham, MA, USA) and then imaged using a digital analyzer (Syngene, Cambridge, UK).
Statistical analysis. The statistical significance of differences between control and honokiol- or enzalutamide-treated groups was evaluated using Student′s t-test, and differences were considered statistically significant at p<0.05. Differences between drug-treated groups were considered significant when the p-value of Duncan′s multiple-range test was <0.05. Statistical analyses between groups over time were carried out by a two-way analysis of variance (ANOVA). Values are expressed as the mean±standard derivation (SD).
Results
Establishment of drug-resistant human and mouse glioblastoma cell models. TMZ-resistant human U87MG-R and mouse GL261-R glioblastoma cell models were selected and prepared from drug-sensitive U87MG and GL261 cells, respectively (Figure 1). After exposure to 25, 50, 75, and 100 μM TMZ for 72 h, viabilities of human U87MG glioblastoma cells were significantly decreased by 14%, 21%, 37%, and 54%, respectively (Figure 1A). In parallel, TMZ at 25, 50, 75, and 100 μM TMZ triggered 10%, 21%, 38%, and 55% cell apoptosis in human U87MG cells, respectively (Figure 1B). Furthermore, exposure of mouse TMZ-sensitive GL261 glioblastoma cells to 25, 50, 75, and 100 μM for 72 h resulted in apoptotic death in a concentration-dependent manner (Figure 1C, D). In comparison, treatment of drug-resistant human U87MG-R and mouse GL261-R glioblastoma cells to 25, 50, 75, and 100 μM TMZ for 72 h did not affect cell viability and apoptotic insults (Figure 1A-D).
Selection and preparation of TMZ-resistant human and mouse glioblastoma cells. TMZ-resistant human U87MG-R and mouse GL261-R glioblastoma cells were selected from drug-sensitive U87MG and GL261 cells, respectively. These drug-sensitive and -resistant glioblastoma cells were treated with 25, 50, 75, and 100 μM TMZ for 72 h. Cell viability was assayed using a colorimetric method (A, C). Apoptotic cells were analyzed using flow cytometry (B, D). Each value represents the mean±SD for n=6. *p<0.05, indicates that the value significantly differed from the respective control group. ENZ: Enzalutamide; TMZ: temozolomide; HNK: honokiol.
Enzalutamide enhanced honokiol-induced killing to human TMZ-resistant glioblastoma cells via an apoptotic mechanism. AR could be detected in drug-sensitive and -resistant human and mouse glioblastoma cells (Figure 2A). There was not significant difference between TMZ-sensitive U87MG and GL261 glioblastoma cells and TMZ-resistant U87MG-R and GL261-R cells (Figure 2B). Exposure of human U87MG-R glioblastoma cells to 25 μM honokiol caused a 13% decrease in cell viability (Figure 2C). Pretreatment with 12.5, 25, and 50 μM enzalutamide enlarged honokiol-triggered death of human TMZ-resistant U87MG-R glioblastoma cells by 62%, 269%, and 569%, respectively. Moreover, pretreatment of human U87MG-R glioblastoma cells with 12.5, 25, and 50 μM enzalutamide led to 53%, 194%, and 341% augmentations in honokiol-induced reduction in cell survival (Figure 2D). As to the mechanisms, treatment with 25 μM honokiol and 25 μM enzalutamide induced DNA fragmentation in human U87MZ-R cells by 68% and 48% (Figure 2E). Interestingly, combined treatment of enzalutamide and honokiol led to a 146% enhancement in DNA fragmentation compared to the honokiol-treated group. Exposure of human U87MG-R glioblastoma cells to honokiol and enzalutamide induced 21% and 15% apoptotic cells (Figure 2F). Pretreatment with enzalutamide amplified honokiol-induced apoptotic insults to human drug-resistant glioblastoma cells by 156%.
ENZ effectively enhanced HNK-induced apoptotic insults to human drug-resistant glioblastoma cells. Levels of AR in human U87MG and U87MG-R and mouse GL261 and GL261-R glioblastoma cells were immunodetected (A, top panel). β-Actin was analyzed as an internal loading control (bottom panel). These protein bands were quantified and statistically analyzed (B). Human temozolomide-resistant U87MG-R glioblastoma cells were pretreated with 12.5, 25, and 50 μM ENZ for 1 h and then exposed to 25 μM HNK for 72 h. Cell viability (C) and cell survival (D) were analyzed using a colorimetric method and a trypan blue exclusion method, respectively. Human U87MG-R cells were pretreated with 25 μM ENZ for 1 h and then exposed to 25 μM HNK for 72 h. DNA damage was assayed with a cellular DNA fragmentation ELISA kit (E). Apoptotic cells were examined using flow cytometry (F). Each value represents the mean±SD for n=6. *p<0.05 and #p<0.05, indicate that the values significantly differed from control and HNK-treated groups, respectively. AR: Androgen receptor; ENZ: enzalutamide; HNK: honokiol.
Enzalutamide selectively stimulated caspase-9 activation and cytochrome c release from mitochondria in human drug-resistant glioblastoma cells. Treatment of human TMZ-resistant U87MG-R glioblastoma cells with enzalutamide and honokiol did not change caspase-8 activity (Figure 3A). Combined treatment of enzalutamide and honokiol caused a slight 39% increase in activity of caspase-8. Exposure to enzalutamide and honokiol amplified caspase-9 activities by 46% and 54%, respectively (Figure 3B). However, pretreatment of human U87MG-R glioblastoma cells with enzalutamide caused a 101% enhancement in honokiol-triggered caspase-9 activation. Furthermore, combined treatment of enzalutamide and honokiol did not change levels of Fas ligand in human U87MG-R glioblastoma cells (Figure 3C, top panel). In contrast, levels of cytochrome c in human U87MG-R glioblastoma cells were obviously elevated after exposure to a combination of enzalutamide and honokiol. These protein bands were quantified and statistically analyzed using β-actin as an internal loading control (Figure 3D). A combination of enzalutamide and honokiol caused a 2.3-fold elevation in levels of cytochrome c but did not affect Fas ligan (Figure 3D).
Combined treatment of ENZ and HNK selectively stimulated caspase-9 activation and cytochrome c release from mitochondria in human drug-resistant glioblastoma cells. Human temozolomide-resistant U87MG-R glioblastoma cells were pretreated with 25 μM ENZ for 1 h and then exposed to 25 μM HNK for 72 h. Activities of caspase-8 and caspase-9 were assayed using fluorogenic methods (A, B). Levels of Fas and cytochrome c were immunodetected (C, top panels). β-Actin was analyzed as an internal loading control (bottom panel). These protein bands were quantified and statistically analyzed (D). Each value represents the mean±SD for n=6. *p<0.05 and #p<0.05, indicate that the values significantly differed from control and HNK-treated groups, respectively. ENZ: Enzalutamide; HNK: honokiol.
The Bak-involved mitochondrial dysfunction contributed to the enhanced effects of enzalutamide on honokiol-induced apoptosis of human TMZ-resistant glioblastoma cells. Treatment of human U87MG-R glioblastoma cells with honokiol and enzalutamide led to slight 18% and 31% amplifications in levels of proapoptotic Bak protein (Figure 4A, B). However, combined treatment of enzalutamide and honokiol raised levels of Bak by 124% compared to the honokiol-treated group. Consecutively, exposure to honokiol and enzalutamide decreased the MMP of human U87MG-R cells by 11% and 10%, respectively (Figure 4C). Pretreatment with enzalutamide caused a 114% enlargement in honokiol-induced reduction of the MMP. Moreover, administration of human TMZ-resistant U87MG-R glioblastoma cells with honkiol and enzalutamide caused 33% and 40% increases in activities of caspase-3, respectively (Figure 4D). In parallel, combined exposure to enzalutamide and honokiol improved honokiol-induced activation of caspase-3 by 203% (Figure 4D).
A combination of ENZ and HNK triggered successive alterations in proapoptotic Bak levels, the MMP, and caspase-3 activation in human drug-resistant glioblastoma cells. Human temozolomide-resistant U87MG-R glioblastoma cells were pretreated with 25 μM ENZ for 1 h and then exposed to 25 μM HNK for 72 h. Levels of proapoptotic Bak protein were immunodetected (A). These protein bands were quantified and statistically analyzed using β-actin as an internal control (B). The MMP was measured using flow cytometry (C). Activity of caspase-3 was assayed using a fluorogenic method (D). Each value represents the mean±SD for n=6. *p<0.05 and #p<0.05, indicate that the values significantly differed from control and HNK-treated groups, respectively. ENZ: Enzalutamide; HNK: honokiol; MMP: mitochondrial membrane potential.
Suppressing caspase-9 activity concurrently attenuated the enhanced effects of enzalutamide on honokiol-induced apoptotic insults to human drug-resistant glioblastoma cells. Pretreatment of human TMZ-resistant U87MG-R glioblastoma cells with Z-LEHD-FMK, an inhibitor of caspase-9, successively resulted in 76%, 83%, 77%, 70%, and 70% alleviations in the enlarged actions of enzalutamide on honokiol-triggered caspasee-9 activation, caspase-3 stimulation, DNA fragmentation, apoptotic insults, and cell death, respectively (Figure 5A-E). In contrast, treatment of human U87MG-R glioblastoma cell with Z-LEHD-FMK, enzalutamide, honokiol, and a combination of enzalutamide and honokiol did not initiate cell necrosis (Figure 5F).
Suppressing caspase-9 activation concurrently attenuated ENZ-induced enlargements in HNK-triggered apoptotic insults to human drug-resistant glioblastoma cells. Human temozolomide-resistant U87MG-R glioblastoma cells were pretreated with caspase-9 inhibitor LEHD for 1 h and then exposed to a combination of ENZ and HNK for 72 h. Activities of caspase-9 and caspase-3 were measured using fluorogenic methods (A, B). DNA injury was evaluated with a cellular DNA fragmentation kit (C). Apoptotic (D) and necrotic cells (F) were examined using flow cytometer and a photometric immunoassay, respectively. Cell viability was assayed using a colorimetric method (E). Each value represents the mean±SD for n=6. T* and #p<0.05, indicate that the values significantly differed from control and ENZ+HNK-treated groups, respectively. ENZ: Enzalutamide; HNK: honokiol; LEHD: Z-LEHD-FMK.
The enhanced effects of enzalutamide on honokiol-induced apoptotic insults were further confirmed in mouse drug-resistant glioblastoma cells. Pretreatment of mouse TMZ-resistant GL261-R glioblastoma cells with enzalutamide for 1 h resulted in a 66% amplification in honokiol-induced reduction of the MMP (Table I). In addition, combined treatment with enzalutamide and honokiol slightly increased caspase-8 activity by 35%. In contrast, pretreatment of mouse GL261-R glioblastoma cells with enzalutamide caused significant 180% and 140% augmentations in honokiol-triggered activation of caspases-9 and -3, respectively (Table I). Consequently, the honokiol-induced DNA fragmentation and cell apoptosis in mouse GL261-R glioblastoma cells were enlarged by 171% and 205% following enzalutamide pretreatment. The honokiol-induced killing to mouse GL261-R cells was meaningfully enhanced by 161% by enzalutamide (Table I). Exposure of mouse TMZ-resistant GL261-R glioblastoma cells to honokiol, enzalutamide, and their combinations did not influence cell necrosis (Table I).
Enhanced effects of ENZ on HNK-induced insults to mouse temozolomide (TMZ)-resistant glioblastoma cells.
Discussion
The present study has shown the enhanced effects of AR suppression by enzalutamide on honokiol-induced insults to TMZ-resistant human and mouse glioblastoma cells. In adult, GBM is the most common and aggressive brain tumor. Even though GBM patients have received standard therapy of surgical tumor removal and followed CCRT, most of them experience tumor recurrence due to therapy resistance (4). Till now, TMZ is the first-line chemotherapeutic drug for treatment of GBM. Unfortunately, more than 50% of GBM patients are ultimately TMZ-resistant (6). Thus, overcoming drug resistance is an urgent and challenging issue in GBM therapy. This study showed that pretreatment with enzalutamide inhibited AR activity in TMZ-resistant glioblastoma cells and concurrently enhanced honokiol-induced killing to tumor cells. Enzalutamide can block AR signaling by competing the binding of androgens to this hormone receptor (22). In a mouse model of castration-resistant prostate cancer, interference with the AR signaling resulted in cell apoptosis and tumor suppression (31). Our previous study proved the roles of the testosterone-AR-PARD3B signaling axis in tumorigenesis and malignance of GBM (19). Moreover, suppressing AR signals by enzalutamide consequently led to death of human and mouse glioblastoma cells (23). A series of studies done in our laboratory showed the safety of honokiol and its capacity to suppress growth of glioblastomas in vitro and in vivo (10-14). In this study, we further demonstrated the improved effects of AR blocking by enzalutamide on honokiol-induced insults to TMZ-resistant glioblastoma cells. As a result, combined treatment of enzalutamide and honokiol may represent a de novo strategy for treatment of drug-resistant GBM.
The enlarged effects of enzalutamide on honokiol-triggered death occurred via an apoptotic mechanism. Exposure of TMZ-resistant human and mouse glioblastoma cells led to cell shrinkage. In addition, activities of executioner caspase-3 were significantly augmented following treatment with a combination of enzalutamide and honokiol. Caspase-3 is a master regulator of cell apoptosis by contributing to formation of apoptotic bodies (32). Our present study also showed induction of DNA fragmentation in human TMZ-resistant glioblastoma cells following exposure to enzalutamide and honokiol. Wolf et al. reported that caspase-3 can initiate caspase-activated DNase (CAD), an endonuclease, to degrade chromosomal DNA by proteolytically inactivating DNA fragmentation factor-45/inhibitor of CAD (33). Hence, the enhanced effects of enzalutamide on honokiol-induced DNA fragmentation are due to activation of caspase-3. Cleavage of chromosomal DNA into oligonucleosomal size fragments is a biochemical hallmark of cell apoptosis (34). Accordingly, pretreatment of human TMZ-resistant glioblastoma cells with enzalutamide amplified honokiol-induced apoptotic insults. In addition, the enhanced effects of enzalutamide on honokiol-induced caspase-3 activation, DNA fragmentation, and cell apoptosis were further confirmed in mouse drug-resistant glioblastoma cells. Thus, enzalutamide could improve honokiol-triggered killing to TMZ-resistant glioblastoma cells occurs via an apoptotic mechanism. Resistance to death is a typical feature of cancer/tumor cells so targeting apoptosis has been widely applied for cancer therapy (35). Fascinatingly, exposure to a combination of enzalutamide and honokiol did not trigger necrotic insults to TMZ-resistant human and mouse glioblastoma cells. As a result, targeting apoptosis by combined treatment of enzalutamide and honokiol can be an effective strategy for treatment of drug-resistant GBM cells.
Combined treatment of enzalutamide and honokiol selectively activated an intracellular cytochrome c/caspase-9 mechanism rather than an extracellular Fas ligand/caspase-8 pathway. Compared to caspase-8, exposure of human TMZ-resistant glioblastoma cells to enzalutamide and honokiol stimulated much more activation of caspase-9. Caspase-9, an upstream initiator caspase, is activated by a complex of cytosolic cytochrome c and apoptotic protease-activating factor-1 (Apaf-1) (32). In this study, levels of cytochrome c were augmented in human TMZ-resistant glioblastoma cells following exposure to enzalutamide and honokiol. Thus, the enhanced effect of enzalutamide on honokiol-induced activation of caspase-9 is because of release of cytochrome c from mitochondria to the cytoplasm. In contrast, caspase-8 is activated by extracellular Fas ligand/death receptor signaling (36, 37). Combined treatment of enzalutamide and honokiol did not affect levels of Fas ligand and caspase-8 activity in human TMZ-resistant glioblastoma cells. Being an upstream initiator caspase, caspase-9 can be targeted for treatment of apoptosis-related diseases, especially cancer (38). Therefore, combined treatment of enzalutamide and honokiol can be clinically applied for treatment of drug-resistant GBM by specifically triggering caspase-9-dependent apoptosis of glioblastoma cells.
The proapoptotic Bak protein-involved disruption of the MMP contributed to the enhanced effects of enzalutamide on honokiol-induced apoptosis of drug-resistant glioblastoma cells. Levels of Bak protein in human drug-resistant glioblastoma cells were obviously elevated following exposure to a combination of enzalutamide and honokiol. Bak, a proapoptotic protein belonging to the Bcl-2 family, is activated and translocated to the outer membrane of mitochondria to induce permeabilization after oligomerization (25). In this study, we report a reduction in the potential of the mitochondrial membrane after administration of enzalutamide and honokiol. The augmented Bak protein can translocate to the outer membrane of mitochondria from the cytoplasm to produce oligomerized pores, leading to reduction of the MMP in human drug-resistant glioblastoma cells. Sequentially, cytochrome c is released from these mitochondrial pores to the cytoplasm (39). In this study, we accordingly detected obvious elevations in levels of cytochrome c and activation of caspase-9 in human TMZ-resistant glioblastoma cells following administration of enzalutamide and honokiol. Pretreatment with enzalutamide could amplify honokiol-induced activation of caspase-3 in human TMZ-resistant glioblastoma cells. After being proteolytically cleaved by upstream caspase-9, activated caspase-3 functionally stimulates cascade activation of downstream caspase-6 (32). These executioner caspase-3 and caspase-6 can degrade critical nuclear structural proteins, like lamin A/C, and cytoskeletal components such as PARD3, finally leading to cellular collapse (40). Therefore, the proapoptotic Bak protein-mediated disruption of the MMP could consecutively initiate cytochrome c release, cascade activations of caspases-9 and -3, DNA fragmentation, and cell apoptosis in human TMZ-resistant glioblastoma cells.
Suppression of caspase-9 activation concurrently attenuated the enhanced effects of enzalutamide on honokiol-induced apoptosis of drug-resistant glioblastoma cells. Z-LEHD-FMK, a chemical peptide, is usually used to specifically inhibit the proteolytic activity of caspase-9 (41). In the present study, pretreatment of human TMZ-resistant glioblastoma cells with Z-LEHD-FMK suppressed activity of caspase-9 could attenuate the improved actions of enzalutamide on honokiol-triggered cascade activation of caspases-9 and -3. Subsequently, downregulation of caspase-9 activation by Z-LEHD-FMK simultaneously lowered DNA fragmentation and cell apoptosis in human drug-resistant glioblastoma cells treated with enzalutamide and honokiol. This study showed the critical roles of caspase-9 in the synergistic effects of enzalutamide and honokiol on induction of apoptotic insults to human TMZ-resistant glioblastoma cells. Our previous study reported the major role of caspase-9 in honokiol-induced apoptosis of drug-resistant glioblastoma cells (24). A systems medicine investigation indicated that the caspase-9-initiated activation of executioner caspases can be a predictor of progression-free survival in GBM patients (42). Caspase-9 is required for intracellular apoptosis signaling. By targeting caspase-9 apoptosis pathway, combined treatment of enzalutamide and honokiol can effectively kill drug-resistant GBM cells.
Conclusion
The present study successfully created TMZ-resistant human U87MG-R and mouse GL261-R glioblastoma cells from their respective drug-sensitive cells as our experimental models. Pretreatment with enzalutamide enhanced honokiol-induced killing to human TMZ-resistant glioblastoma cells via an apoptotic mechanism without affecting cell necrosis. As to the mechanisms, suppressing AR activity by enzalutamide selectively triggered an intracellular cytochrome c-dependent activation of initiator caspase-9. Sequentially, a combination of enzalutamide and honokiol successively induced the proapoptotic Bak protein-mediated mitochondrial dysfunction, cytochrome c release, and cascade action of executioner caspase-3. In contrast, suppressing caspase-9 by its specific peptide inhibitors concurrently lessened the enhanced effects of enzalutamide on honokiol-induced apoptotic insults to human TMZ-resistant glioblastoma cells. The improved actions of enzalutamide on honokiol-induced apoptotic events were further confirmed in TMZ-resistant glioblastoma cells. Therefore, a combination of enzalutamide and honokiol represents a de novo strategy for treatment of drug-resistant glioblastomas.
Footnotes
Authors’ Contributions
Conceptualization, C.Y.C., J.T.C., and R.M.C.; methodology, C.Y.C., J.T.C., and E.C.; software, S.H.L. and E.C.; validation, S.H.L. and R.M.C.; formal analysis, C.Y.C. and J.T.C.; investigation, C.Y.C. and J.T.C.; resources, S.H.L. and R.M.C.; data curation, E.C.; writing-original draft preparation, C.Y.C. and J.T.C.; writing-review and editing, R.M.C.; visualization, S.H.L. and R.M.C.; supervision, R.M.C.; project administration, J.T.C.; funding acquisition, R.M.C. All Authors have read and agreed to the published version of the manuscript.
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
Funding
This research was funded by Yuan’s General Hospital (109YGH-TMU-03), Wan Fang Hospital (115-wf-swf-05), and the Ministry of Science and Technology (MOST 114-2314-B-038-087-MY3), Taipei, Taiwan.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received February 21, 2026.
- Revision received April 7, 2026.
- Accepted April 9, 2026.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.











