Propolis was collected from a hive of Apis mellifera from Chiangmai province, Thailand, in August 2006. A voucher specimen (SWU 0212) was deposited at the Faculty of Pharmacy, Srinakharinwirot University, Nakhon Nayok province, Thailand. Thai propolis (1 kg) was extracted by sonication with methanol (2 liters, 90 min × 3) at room temperature. Removal of methanol yielded the methanol extract (517 g). Part of the methanol extract (140 g) was subjected to silica gel column chromatography using dichloromethane, with increasing concentration of methanol, give 16 fractions. Chrysin (732 mg) and tectochrysin (314 mg) were isolated by crystallization from fractions. Chrysin and tectochrysin were identified by comparison of physical and spectroscopic data with authentic standard and literature values (17,18).

Recombinant human TRAIL and recombinant human tumor necrosis factor (TNF)-α were purchased from Peprotech (London, UK). Curcumin was obtained from Fluka Chemika (Steinheim, Switzerland). N-acetylcysteine (NAC) and 3-(4,5-dimethythiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), and chrysin were purchased from Sigma-Aldrich (St. Louis, MO). STAT3 specific inhibitor (cucurbitacin-I) and IKK-2 inhibitor IV were obtained from Calbiochem (San Diego, CA). Lipofectamine 2000 reagent and hygromycin B were purchased from Invitrogen (Carlsbad, CA). Test compounds were dissolved in dimethyl sulfoxide (DMSO) and kept as a stock solution at −20°C. The final concentration of DMSO was kept below 0.2% throughout the study. Primary antibodies specific to caspase-3, poly (ADP-ribose) polymerase (PARP), Mcl-1, Bcl-XL, XIAP, survivin, FLIP, STAT3 and phosphorylated form of STAT3 (Tyr⁷⁰⁵), Akt (Ser⁴⁷³), ERK1/2 (Thr²⁰²/Tyr²⁰⁴), p38 (Thr¹⁸⁰/Tyr¹⁸²), JNK (Thr¹⁸³/Tyr¹⁸⁵), p65 NF-κB (Ser⁵³⁶) were obtained from Cell Signaling Technology (Beverly, MA). Primary antibodies specific to actin, pan 14-3-3, Akt, ERK1/2, p38, JNK and p65 NF-κB were obtained from Santa Cruz Biotechnology (Santa Cruz, CA).

A549 human lung adenocarcinoma and HeLa human cervical carcinoma are TRAIL resistant cell lines (19). Cell lines were obtained from the American Type Culture Collection (ATCC, Rockville, MD). A549 was cultured in RPMI-1640 supplemented with 10% fetal bovine serum (FBS), 100 U/ml of penicillin and 100 μg/ml of streptomycin. HeLa was cultured in DMEM supplemented with 2 mM L-glutamine, 10% FBS and the antibiotics. The cells were maintained at 37°C in a humidified atmosphere of 5% CO₂.

Cell viability after treatment was determined by MTT method (20). Briefly, cells in 96-well plates (100 μl/well) were pretreated by adding medium (100 μl) containing test compound to the wells, and incubated for 30 min, then aliquots of TRAIL in medium (5 μl) were added to each well (final concentration of 100 ng/ml for A549 cells, or 200 ng/ml for HeLa cells), and further incubated for 24 h. Then, the wells were replaced and incubated with fresh culture media containing MTT (0.5 mg/ml) for 2 h at 37°C. Finally, the media were removed and DMSO (100 μl) was added to the wells, and absorbance was measured at 550 nm in a microplate reader, subtracted with absorbance at 650 nm. The number of viable cells was determined from the absorbance. Assays were performed in triplicate wells. Data were expressed as percent viability compared with control.

A549 cells were transfected with pGL4.32[luc2P/NF-κB-RE/Hygro] vector (Promega, Madison, WI) using Lipofectamine 2000 reagent according to manufacturer’s protocol. After transfection for 24 h, the transfected cells were selected by sub-culturing into new flasks containing RPMI-1640 supplemented with 10% FBS and 800 μg/ml hygromycin B and maintained in this media over 2 weeks to obtain the stable transfected cells (A549/NF-κB).

NF-κB transcriptional activity was determined by luciferase reporter gene assay. A549/NF-κB cells were seeded into a 96-well plate (100 μl/well), and left overnight. Cells were pretreated with test compound (100 μl) for 30 min, followed by adding 5 μl of medium containing TNFα or TRAIL, and further incubated for 3 h. Cells were then lysed with 25 μl of passive lysis buffer (Promega), and 20 μl of cell lysate was mixed with 50 μl of luciferase assay reagent (Promega). Luminescence was determined using a luminometer (Atto, Tokyo, Japan).

Immunoblot analysis was performed as previously described (21). Briefly, whole cell lysates were subjected to electrophoresis in 7.5 or 10% SDS-PAGE, and electrophoretically transferred to Immobilon-P nylon membrane (Millipore, Bedford, MA). The membranes were blocked with BlockAce (Dainippon Pharmaceutical, Co. Ltd., Osaka, Japan) for at least 2 h, and probed with the indicated primary antibodies overnight, followed by horseradish peroxidase-conjugated secondary antibodies (Dako, Glostrup, Denmark). Bands were visualized using ECL reagents (Amersham Bioscience, Piscataway, NJ).

Isolation of total cellular RNA from cells was performed using RNeasy mini kit (Qiagen, Valencia, CA), and 2 μg of the total cellular RNA was used to synthesize first-strand cDNA by using SuperScript III Reverse Transcriptase Kit (Invitrogen). Quantitative determination of gene expression was performed by real-time PCR on a LightCycler^® 2.0 Instrument (Roche). PCR primer sequences were obtained from following references; human Mcl-1 (22); human β-actin (23). Mcl-1 primer (fwd) 5’-CGG TAA TCG GAC TCA ACC TC-3’; and Mcl-1 primer (rev) 5’-CCT CCT TCT CCG TAG CCA A-3’. β-actin primer (fwd) 5’-GAC CTG ACT GAC TAC CTC ATG A-3’; and β-actin primer (rev) 5’-AGC ATT TGC GGT GGA CGA TGG AG-3’. The PCR reaction mixture (20 μl) was composed of 1 μl of 5-fold diluted cDNA solution, 10 μl QuantiTect™ SYBR-Green PCR Master mix (Qiagen), PCR primers and sterile distilled water. An initial activation step at 95°C for 15 min was followed by 40 cycles comprising denaturation at 94°C/15 sec, annealing at 55°C/30 sec, and extension at 72°C/30 sec. Expression level of Mcl-1 gene was normalized to β-actin gene, and calculated from crossing point (C_p) value of the sample, relative gene expression level = 2^{(C_pactin− C_ptarget)}. The relative change in gene expression compared with untreated conditions was expressed as fold change, calculated by 2^−ΔΔCp method (24).

Silencing of Mcl-1 expression in A549 cells was performed as previously described (19). Human Mcl-1 Stealth RNAi™ siRNA (HSS181042) was purchased from Invitrogen. Firefly luciferase (GL2) siRNA was obtained from Hokkaido System Science Co. Ltd. (Sapporo, Japan), and used as control siRNA (19). A549 cells were transfected with siRNAs at a final concentration of 20 nM, using Lipofectamine 2000 reagent. Cells were treated with TRAIL at 24 h after transfection.

Data are expressed as mean + SD, and analyzed by Student’s two-tailed t-test to determine the significance of differences between groups. A p-value of lower than 0.05 was considered to be significant.

Initially, TRAIL sensitization effect of Thai propolis extract was examined in a TRAIL resistant cell line, A549. Pretreatment with propolis extract (25–100 μg/ml) for 30 min, followed by TRAIL stimulation for 24 h, enhanced TRAIL-induced cell death. At the highest propolis concentration test (100 μg/ml), cell viability of the combination treatment was 38%, compared with 63 and 98% viability with propolis alone and TRAIL alone, respectively (Fig. 1A). Chrysin and tectochrysin were found to be two major constituents of the Thai propolis extract. Chrysin enhanced TRAIL-induced cell death, with cell viability being 47% in the combination treatment, compared to 86 and 92% viability with chrysin alone and TRAIL alone, respectively, while tectochryin did not show significant effect (Fig. 1B). TRAIL sensitization effect of chrysin was observed in a dose-dependent manner over the tested ranges (20–60 μM) in both A549 and HeLa cell lines, which originated from different organs (Fig. 1C).

Hallmarks of cells undergoing apoptosis are cleavage of procaspase-3 and poly(ADP-ribose) polymerase (PARP). Massive cell death of A549 and HeLa cells could be observed as early as 3 h after treatment with chrysin/TRAIL combination (Fig. 1D), concordant with enhancement of apoptosis as clearly shown by increased cleavage of pro-caspase-3 and PARP (Fig. 1E). Our results revealed that apoptosis-inducing effect of TRAIL was enhanced by chrysin.

In addition to inducing apoptosis signaling, TRAIL and TNFα receptors also activate survival signaling such NF-κB which regulates expression of several anti-apoptotic proteins to counteract the apoptosis signaling (1). Since chrysin was shown to sensitize TNFα-induced apoptosis in human cancer cells by inhibiting NF-κB activation (15), thus, we examined effect of chrysin on TNFα- or TRAIL-induced NF-κB activation in A549/NF-κB stable cell line. Stimulation of TNFα for 3 h induced activation and increased transcriptional activity of NF-κB by 20.8-fold compared to control (Fig. 2A). TNFα-induced NF-κB activation was diminished by chrysin pretreatment (60 μM), with 79% inhibition observed, while IKK-2 inhibitor pretreatment (10 μM) led complete inhibition (Fig. 2A). In contrast, TRAIL stimulation slightly induced NF-κB activation by 1.1-fold, while NF-κB activation was more increased by chrysin alone or chrysin/TRAIL combination treatment, with 1.8- and 1.9-fold induction respectively (Fig. 2A). These results indicated that chrysin did not inhibit TRAIL-induced NF-κB activation in A549 cells. Furthermore, inhibition of TRAIL-induced NF-κB activation by IKK-2 inhibitor did not significantly enhance cell death in A549 cells (Fig. 2B), indicating that inhibition of NF-κB activation is unlikely to be the mechanism of chrysin for TRAIL sensitization.

Chrysin has been shown to sensitize A549 cells to chemotherapeutic drugs by depleting intracellular glutathione (13,14). N-acetylcysteine (NAC), a precursor of glutathione synthesis, has been shown to reverse TRAIL sensitization by curcumin (25). We evaluated the involvement of glutathione depletion in TRAIL sensitization effect of chrysin by pretreatment of A549 cells with NAC (10 mM) for 2 h, to increase intracellular glutathione level before exposure to TRAIL combination treatment. If the cytotoxic effect of the combination treatment is mediated by glutathione depletion effect, the NAC pretreatment will abolish the cytotoxic effect. The results showed that TRAIL sensitization effect of chrysin was unaffected by NAC pretreatment, but TRAIL sensitization effect of curcumin was potently reversed by NAC pretreatment (Fig. 3). The results clearly indicated that glutathione depletion does not contribute to the TRAIL sensitization effect of chrysin.

Many naturally occurring polyphenols enhance TRAIL sensitization through downregulation of anti-apoptotic proteins (26). We further examined effect of chrysin on levels of several anti-apoptotic proteins, including Mcl-1, Bcl-XL, XIAP, survivin and 14-3-3, in both A549 and HeLa cells. The results showed that, at 3 h after treatment, chrysin alone (60–100 μM) or in combination with TRAIL selectively decreased Mcl-1 protein level, while the levels of other proteins remained unaffected (Fig. 4A). The dose-dependent decrease in Mcl-1 protein level by chrysin was observed in both A549 and HeLa cell lines (Fig. 4A and B).

Effect of chrysin on another short half-life anti-apoptotic protein, FLIP, was examined compared with Mcl-1. The rapid turnover rate of both proteins in cells is mediated by the proteasome (27). FLIP has 2 isoforms - FLIP_L and FLIP_S, but FLIP_S was not detected in our cell lines. Chrysin treatment (60 μM) for 3 h decreased Mcl-1 level, while FLIP_L levels in both cell lines were not reduced (Fig. 4C).

To explore the cause of Mcl-1 reduction by chrysin, we determined mRNA level of Mcl-1 gene after treatment by using qRT-PCR. Effect of chrysin on Mcl-1 gene expression was clearly demonstrated, chrysin (60 μM) treatment for 3 h reduced Mcl-1 mRNA level by 52% (Fig. 4D). The results suggested that modulation of TRAIL sensitivity of cancer cells by chrysin was occurred through downregulation of Mcl-1 gene expression.

To confirm that reduction of Mcl-1 level was able to sensitize cells to TRAIL-induced apoptosis, siMcl-1 transfection was used to silence Mcl-1 expression in A549 cells, while siLuc was used as control siRNA. At 24 h after transfection, the level of Mcl-1 protein was obviously decreased in siMcl-1 transfected cells compared with siLuc transfected cells (Fig. 5A). After TRAIL stimulation for 3 h, many apoptotic cells were observed in the siMcl-1 transfected cells (Fig. 5B), correlating with cleavage of pro-caspase-3, while the procaspase-3 cleavage was faintly detected in TRAIL-stimulated siLuc transfected cells (Fig. 5A). Taken together, the results confirmed that TRAIL sensitization effect of chrysin was mediated by Mcl-1 downregulation. Thus, we further investigated mechanism of Mcl-1 modulation by chrysin.

Mcl-1 gene expression is regulated by multiple signaling pathways, including STAT3, Akt and p38 pathways (28). We examined the effects of chrysin treatment (60–100 μM) for 3 h, either alone or in combination with TRAIL, on several survival signaling pathways including STAT3, Akt, ERK, p38 and NF-κB. Among the selected pathways, marked inhibition was obtained only in constitutive STAT3 tyrosine phosphorylation (Tyr⁷⁰⁵), in both A549 and HeLa cells (Fig. 6A). Tyr⁷⁰⁵ phosphorylation of STAT3 is necessary for DNA-binding and transcriptional activity of STAT3 (29). Correlation between decreased STAT3 tyrosine phosphorylation and Mcl-1 down-regulation was observed in A549 cells treated with chrysin in a time-dependent manner, while Akt phosphorylation was not affected within the time-course measured (Fig. 6B). The results suggested that STAT3 was the target of chrysin for modulating level of Mcl-1 expression.

To confirm that Mcl-1 expression is regulated by STAT3, A549 cells was treated with a specific inhibitor of STAT3 (cucurbitacin-I) for 3 h, the result showed that the STAT3 inhibitor decreased both STAT3 tyrosine phosphorylation and Mcl-1 level (Fig. 7A). Furthermore, the STAT3 inhibitor enhanced TRAIL-induced cell death (Fig. 7B) in a dose-dependent manner, similar to that observed with chrysin treatment. Collectively, our results indicated that the TRAIL sensitization effect of chrysin was mediated by Mcl-1 down-regulation through inhibition of constitutive STAT3 tyrosine phosphorylation.