Honokiol was purchased from the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China) with >99% purity. Honokiol power was dissolved in dimethyl sulfoxide (DMSO) at 20 mg/ml solution and stored at −20°C. RPMI-1640, fetal bovine serum (FBS), and 0.25% trypsin were purchased from Gibco-BRL (Invitrogen Co., Grand Island, NY, USA). Cyclosporin A (CsA), etoposide (VP-16) and MTT power were purchased from Sigma-Aldrich (St. Louis, MO, USA). Doxorubicin (ADM) and daunorubicin (DNR) were purchased from Pfizer (Actavis Italy S.p.A., Nerviano, MI, Italy). Paclitaxel (taxol) was purchased from Bristol-Myers Squibb (Princeton, NJ, USA). Pan-caspase inhibitor z-VAD-fmk was from Beyotime Institute of Biotechnology (Shanghai, China). Annexin V-FITC/PI kit was purchased from Sigma-Aldrich. Apotosis DNA Ladder Detection Kit was from KeyGen Biotech (Nanjing, China).

Human breast carcinoma cell line MCF-7 and acute promyelocytic leukemia cell line HL-60 were maintained in RPMI-1640 supplemented with 10% FBS. Parallel multidrug resistant cell lines MCF-7/ADR and HL-60/ADR were developed as reported previously (2) and maintained in RPMI-1640 with 1 μg/ml and 100 ng/ml ADM, respectively. Various cells were grown in a humidified CO₂ incubator at 37°C and maintained in exponential growth.

Cells were plated in 96-well micro-titer plates of 4,000–8,000 cells/well. After incubation overnight, cells were incubated with drug-free medium, honokiol or chemotherapeutic agent alone or combination of honokiol and chemotherapeutic agent with different rational concentrations. DMSO according to the highest drug concentration served as control. Quadruple wells were prepared for each drug concentration. After 48 h, MTT assays were performed as previously described (2). Viabilities were normalized to untreated samples. Combination index determinations were performed basing on the method of Chou and Talalay (16). Various cytotoxicity levels of the component drugs and combinations were calculated according to equation 1. Combination index values were calculated using equation 2 at each level of cytotoxicity. (1) D 1 − FA = D x   [ FA / ( 1 − FA ) ] 1 / m (2) Combination   index   ( CI ) = ( D ) 1 / ( D 1 − FA ) 1 + ( D ) 2 / ( D 1 − FA ) 2 + α ( D ) 1 ( D ) 2 / ( D 1 − FA ) 1 ( D 1 − FA ) 2

In the above equations, ‘m’ (stood for sigmoidity) and D_x (is the IC₅₀ for 50% growth inhibition) were calculated by transforming concentration-effect data to a logarithmic scale. FA stood for the growth-inhibitory effect of each drug. (D)₁ and (D)₂ represented the doses of the combination required to produce survival (1-FA), while (D_1-FA)₁ and (D_1-FA)₂ stood for the concentrations of the component drugs required to produce survival (1-FA). α=1 for mutually nonexclusive drugs (17). In analysis, degree of combination effect was determined from the combination index (CI) as follows: CI>1, antagonism; CI=1, additivity; CI<1, synergism.

Two concentrations of honokiol were chosen for each cell line in the evaluation of DRI, which would produce 90–95% (HNK¹) and 70–80% (HNK²) cell survival, respectively. The exact doses of HNK¹ and HNK² for each cell line are shown in Table III. After incubation overnight, cells plated in 96-well plates were treated with different chemotherapeutic agents in concentration-gradient combined with either blank control, HNK¹ or HNK² . After 48 h, MTT assays were performed. IC₅₀ was calculated by transforming the concentration-effect data to a logarithmic scale. DRI values were determined using equation 3. (IC₅₀)₁ and (IC₅₀)₂ stood for IC₅₀ of a chemotherapeutic agent without or with honokiol, respectively. (3) Dose   Reduction   Index   ( DRI ) = ( IC 50 ) 1 / ( IC 50 ) 2

MCF-7/ADR cells were treated with 40 μg/ml honokiol, 100 μg/ml VP-16, or combination, respectively. After 6-h incubation, MTT assay was performed to determine cell viability.

MCF-7/ADR cells were cultured in 6-well culture plates at 8×10⁵/well in 2 ml media for 24 h and treated with 0.1% DMSO (vehicle control), 40 μg/ml honokiol, 100 μg/ml VP-16, or combination for 6 h. Apoptotic rates were analyzed by flow cytometry using Annexin V-FITC/PI kit according to the manufacturer’s instructions. The differentiation among early apoptotic cells, necrotic cells and viable cells were detected as follows: Annexin V positive and PI negative; Annexin V positive and PI positive; and Annexin V negative and PI negative, according to a previous report (9). The experiments were done in triplicate and similar results were obtained in three different experimental setups.

Cells were seeded and treated as described above for flow cytometric analysis for apoptosis. The exact procedure was processed according to previously reported methods (9).

Cells were seeded and treated as indicated above. After treatment, cells were collected and washed twice with chilled PBS. Afterwards, DNA gel electrophoresis was carried out after DNA extraction according to the manufacturer’s instructions of DNA Ladder Detection kit. Experimental conditions were tested in triplicate and similar results were obtained at least three times.

For detection of nuclear factor-κB (NF-κB) proteins, cells were treated with 0.1% DMSO (vehicle control), 40 μg/ml honokiol, 100 μg/ml VP-16, or combination, respectively, for 2 h. Cells were harvested for isolation of nuclear and cytoplasm extracts using Nuclear Extract Kit (Activ Motif, Carlsbad, CA, USA). The concentration of each protein was then determined by the standard BCA protein assay. About 40 μg of each sample was loaded in gel for immunoblot assay according to standard protocol. After cell lysate preparation, proteins were applied to a 10% SDS-PAGE and electro-transferred to a nitrocellulose membrane. Blots were probed by the proper primary antibodies (mouse monoclonal anti-human NF-κB and rabbit polyclonal anti-human lamin-B and α-tubulin (Cell Signaling Technology Inc., Danvers, MA, USA) overnight at 4°C and then exposed to secondary antibodies (HRP-conjugated anti-rabbit IgG (Santa Cruz Biotechnology, Santa Cruz, CA, USA) for 2 h. Visualization was finally detected by chemiluminescence using ECL Plus Western Blotting Substrate (Thermo Fisher Scientific, Rockford, IL, USA). Lamin B and α-tubulin was used to ensure equal loading of protein on the gel. Densitometry was performed to quantify the relative expression levels of NF-κB protein.

The data represent mean ± SD from three independent experiments. Student’s t-test was used for comparison between in vitro responses to different drugs. All analyses were two-tailed.

We used multiple drug effect analysis method of Chou and Talalay (16) to determine the nature of interaction between honokiol and various chemotherapeutic agents. Four different anticancer drugs (etoposide, doxorubicine, taxol and daunorubicin), commonly used in the treatment of breast cancer and leukemia were selected in the present study. Human breast carcinoma cell line MCF-7, acute promyelocytic leukemia cell line HL-60 and parallel multidrug resistant cell lines MCF-7/ADR, HL-60/ADR were treated with serial dilutions of honokiol and various chemotherapeutic agents at fixed ratios spanning the IC₅₀ of each drug. Combination index values at different dose-effect levels were calculated on the basis of parameters derived from effect plots of honokiol alone, the chemotherapeutic agent alone and the combination of the two. CI<1, CI=1 and CI>1 indicate synergism, additivity and antagonism effect, respectively. Honokiol synergistically enhanced cytotoxicity of ADM and VP-16 in MCF-7/ADR as the CI values were <1.0 at all dose-effect levels (50, 75 and 85% inhibition rate) (Table I, Fig. 1). CI values for MCF-7/ADR at IC₅₀ and IC₇₅ were <1.0 for taxol indicating a synergistic interaction, but CI₈₅ indicated antagonism (Table I, Fig. 1). The interaction for MCF-7 was synergism between honokiol and ADM at all dose-effect levels, whereas antagonism between honokiol and VP-16. It was synergism between honokiol and taxol for MCF-7 at IC₅₀ and IC₇₅ but antagonism at IC₈₅ (Table I, Fig. 1). Honokiol and daunorubicin (DNR) had synergistic interactions against HL-60 and HL-60/ADR (Table II, Fig. 1).

To compare combination effects among different groups, we calculated the mean CI values [mean CI = (CI₅₀ + CI₇₅ + CI₈₅)/3] (Fig. 1B). Honokiol synergized most chemotherapeutic agents both in P-gp (MCF-7/ADR) and MRP1 (HL-60/ADR) mediated MDR cancer cells. Surprisingly, the most marked mean CI values existed in honokiol plus VP-16 combination in MCF-7/ADR and parallel cell line MCF-7, which indicated conflicting effects.

DRI parameters represent the order of magnitude of dose reduction which is determined from the relative proportions between the dose of combination and each drug alone at 50% inhibitory effect. Cell viabilities of each cancer cell line after treatments of the two given doses of honokiol (HNK¹ and HNK², exact doses are shown in Table III) alone were 90–95 and 70–80%, respectively. On the other hand, HNK¹ and HNK² do no harm to human normal cells, such as normal PBMC (18). Dose-response curves demonstrated dose-dependent growth-inhibition by a single chemotherapeutic agent or combined with a given dose of Honokiol (HNK¹ or HNK²) in all cell lines (Fig. 2). IC₅₀ was calculated by transforming the concentration-effect data to a logarithmic scale. Based on the IC₅₀s, DRI for each chemotherapeutic agent was determined using equation 3.

Conspicuous DRI values from 1.85 to 17.32 were observed in the combination setting compared with chemotherapeutic agents alone except that of MCF-7 treated with VP-16 (Table IV). A dose of 12.5 μg/ml honokiol had no synergistic effect in combination with VP-16 in MCF-7 cell line as the DRI value was −1.20 (Table IVD). The given doses of honokiol had greater dose-reducing effects in settings of VP-16 treatment in MCF-7/ADR than in MCF-7, and of DNR in HL-60/ADR than in HL-60 (Table IVB and D).

Above results indicated that in most combination groups calculated DRI values were >1, whose trendency coincided well with CI values. Conspicuously, the most prominent DRI (17.32) was measured in honokiol plus VP-16 combination in MCF-7/ADR cells, which matched exactly to the lowest CI arisen group (as shown in the section: Combination effects of honokiol with different chemotherapeutic agents at rational combination doses in different cell lines). Accordingly, honokiol plus VP-16 combination treatment in MCF-7/ADR was chosen as representative in assessing synergetic mechanism.

Based on the above results, we further tested the potential synergistic effect of honokiol on VP-16 induced MCF-7/ADR cell death at highest rational doses. Concentration of honokiol used in the following experiment was 40 μg/ml, and corresponding rational concentration of VP-16 was 100 μg/ml. After 6-h incubation, cell viabilities after 100 μg/ml VP-16, 40 μg/ml honokiol, or combination treatment were 82.33±3.73, 56.21±2.97 and 48.18±1.38%, respectively. VP-16 alone only slightly affected cell viability of MCF-7/ADR. Honokiol plus VP-16 had dramatically enhanced cell toxicity compared to that of VP-16 or honokiol alone (P<0.005) (Fig. 3).

Honokiol (40 μg/ml) induces programmed cell death with necrotic characteristics in MCF-7 cells as we reported previously (9). We hypothesized honokiol at the same concentration could primarily induce programmed necrotic cell death in the parallel drug resistant cell line MCF-7/ADR. This hypothesis was supported by results of DNA ladder detection as only clumping degradation of nuclear DNA was detected after 6-h incubation with 40 μg/ml honokiol (Fig. 4). Compared with normal cells (Fig. 5A) and 100 μg/ml VP-16-treated cells (Fig. 5B), electron microscopy morphological manifestations revealed majority of cells underwent necrotic cell death characterized manifestations of increased cell volume due to swelling of cytoplasmic organelles, dilatation of mitochondria and endoplasmic reticulum, vacuoles and nuclear modifications (Fig. 5C). Not surprisingly, flow cytometry demonstrated that honokiol alone induced a dominant death characterized by loss of plasma membrane integrity (PI-positive), 39.07% of cells were PI-positive (upper quadrants) and 6.16% were Annexin V positive and PI negative (lower right quadrant) (Fig. 6A).

We previously demonstrated that honokiol-induced necrotic cell death in MCF-7 cells was modulated by cyclophilin D (a major component of the mitochondrial permeability transition pore, CypD), which can be prevented by cyclosporin A (CsA, inhibitor of CypD) (9). Accordingly, we used Annexin V-FITC/PI assay to determine the effect of CsA on honokiol-induced necrotic cell death in MCF-7/ADR cells. Honokiol-induced necrotic cell death was reversed by 14.97% with pretreatment of 25 μM CsA for 2 h (Fig. 6A). These results indicated honokiol-induced necrotic cell death in MCF-7/ADR was partly associated with CypD.

The above results provided substantial evidence for the original assumptions that honokiol at a higher dose (40 μg/ml) induced dominant programmed necrotic cell death in MCF-7/ADR cells and could be prevented by CsA, which accords with that in the parallel cell line MCF-7 (9).

Honokiol (40 μg/ml) induced cell death with necrotic characteristics in MCF-7/ADR cells as demonstrated above. Consequently, we hypothesized the combination of honokiol would enhance the cytotoxic effect of VP-16 through additional necrotic cell death. To authenticate this hypothesis, several markers of cell death modality were analyzed, including DNA ladder, morphological manifestations and Annexin V/PI staining.

Characteristic DNA fragmentation detected by DNA ladder is a direct indication of the existence of apoptosis. We used DNA ladder to determine if apoptosis was the major cell death pathway for MCF-7/ADR cells treated with honokiol plus VP-16. As shown in Fig. 4, no DNA ladder but clumping degradation of nuclear DNA was detected in cells treated with honokiol plus VP-16, which indicated apoptosis was not the major cell death pathway (Fig. 4).

Further electron microscopy morphological examinations demonstrated cells treated with VP-16 plus honokiol displayed complex features composed of necrotic cell death (severe vacuolation and dilatation of mitochondria) (Fig. 5D and E) and apoptosis (chromatin condensation and margination) (Fig. 5F). Parts of cells treated with VP-16 exhibited early staged apoptotic manifestations, while most cells had normal morphological manifestations (Fig. 5B).

The effects of VP-16 combined with honokiol on induction of cell death were then examined with Annexin V-FITC/PI assay. Inhibition of cell viability was more efficient after incubation with honokiol plus VP-16 (40.96±2.18%) than honokiol (51.90±3.46%) (P<0.05) (Fig. 6B). Honokiol alone induced MCF-7/ADR cell death with obvious characteristics of necrotic cell death as shown in Honokiol-induced programmed necrotic cell death in MCF-7/ADR cells, which is reversed by cyclosporin A. While treatment with combination of honokiol and VP-16 increased the ratio of cell death, including 37.48% of cells PI positive and 19.97% Annexin V-positive and PI-negative (Fig. 6A), which increased necrosis by 27.60% and apoptosis by 16.80% from basal level. Taken together, these results indicated that the combination treatment of honokiol plus VP-16 induced complex cell death including apoptosis and programmed necrotic cell death in MCF-7/ADR cells.

Above data indicated: i) the combination treatment of VP-16 and honokiol induced programmed cell death containing apoptosis and necrotic cell death in MCF-7/ADR cells; ii) CsA reversed honokiol-induced programmed necrotic death in MCF-7/ADR cells. CsA and z-VAD-fmk were used separately or together to confirm whether caspase-dependent apoptosis and necrotic cell death constituted the complex cell death induced by honokiol plus VP-16. After CsA pretreatment, morphological examinations demonstrated that more cells exhibited apoptotic manifestations (Fig. 5G) than that shown in treatment of VP-16 plus honokiol (Fig. 5D–F), whereas still some cells exhibited necrotic manifestations. Moreover, Fig. 6A revealed honokiol plus VP-16-induced necrotic cell death was reversed by 19.97% with CsA, while apoptosis was reversed by z-VAD-fmk 13.40%, but when pretreated with both CsA and z-VAD-fmk, honokiol plus VP-16-induced necrotic cell death and apoptosis were increasingly reversed by 26.85 and 15.67%. Viable cells were increased after pretreated with CsA, z-VAD-fmk, or both (79.02±8.12, 75.79±6.41 and 84.47±4.31%, respectively) (Fig. 6B). Which indicates that honokiol plus VP-16 induced complex cell death was almost completely shut down by pretreatment with both CsA and z-VAD-fmk. These results indicated combination treatment of VP-16 and honokiol induced programmed cell death contained caspase-dependent apoptosis and programmed necrotic cell death, which were prevented by CsA and z-VAD-fmk.

Activated NF-κB binds to specific DNA sequences in target genes associated with anti-apoptotic effects or expression of multiple drug resistant proteins, such as P-gp, Bcl-2, Bcl-xl and xIAP. Constitutively activated NF-κB is associated with cell resistance to various chemotherapeutic agents. Nuclear translocation of NF-κB is momentous in NF-κB signal transduction pathway. Blocking nuclear translocation of NF-κB is a potential target in the reversal of chemoresistance. We used western blot analysis to explore the differences of NF-κB expression in nuclear and cytoplasm after VP-16 incubation with or without honokiol. Although no changes were detected in cytoplasmic NF-κB, expressions of nuclear NF-κB were different between different treatments (Fig. 7A). Expression of nuclear NF-κB was slightly increased after incubated with VP-16 for 2 h (1.16±0.02-fold compared with control), but decreased after incubated with honokiol alone for 2 h (0.55±0.01-fold). And the expression of nuclear NF-κB after co-incubated with VP-16 and honokiol (0.95±0.03-fold) was decreased compared to that of VP-16 treatment alone (P<0.001) (Fig. 7B). These results revealed that honokiol not only decreased the primarily enhanced nuclear expression of NF-κB in multi-drug resistant cell line MCF-7/ADR, but also reversed VP-16-induced nuclear translocation and activation of NF-κB.