SW982 cells were obtained from the American Type Culture Collection (Manassas, VA, USA) cultured in L-15 medium (Gibco Life Technologies, Grand Island, N Y, USA) containing 10% FBS (Gibco Life Technologies), 100 U/ml penicillin and 100 U/ml streptomycin (Shandong Lukang Record Pharmaceutical Co., Ltd., Jining, China), and were incubated at 37°C with 5% CO₂. L-gossypol was provided by Professor Tang Hui, University of Shihezi (Shihezi, China). Doxorubicin (Bio Basic, Inc., Amherst, N Y, USA).

The experimental groups were divided as follows: i) Control group, cells cultured in the medium described; ii) L-gossypol group, 2.5 μmol/l L-gossypol; iii) doxorubicin group, 0.2 μmol/l doxorubicin; iv) combination group, 2.5 μmol/l L-gossypol and 0.2 μmol/l doxorubicin. The culture medium was replenished every 3–4 days and the cells were trypsinized (Gibco Life Technologies) and replated at 85% confluence.

MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; MP Biomedicals, LLC, Santa Ana, CA, USA] assays were performed in accordance with the manufacturer's instructions. The effect of five different concentrations of doxorubicin (0.01, 0.02, 0.05, 0.1 and 0.2 μmol/l) and L-gossypol (0.0, 0.5, 1.0, 1.5, 2.0 and 2.5 μmol/l) were investigated. The drug combinations were designed in accordance with the factorial design and are listed in Table I.

The cells were seeded during the logarithmic growth phase onto 96-well plates at a density of 5×10⁴ cells/ml (100 μl/well) with five parallel wells. The experimental drugs were added, resulting in a final volume of 200 μl/well. Subsequently, 20 μl MTT/well was incubated for 4 h at 37°C, then the supernatant was removed, 150 μl dimethyl sulfoxide was added, and the plates were incubated for a further 10 min. The absorbance was then measured using an xMark Microplate Absorbance Spectrophotometer (Bio-Rad Laboratories, Inc., Hercules, CA, USA) at a wavelength of 490 nm. The absorbance values at 24, 48, 72 and 96 h were measured in order to calculate the IC₅₀ (the concentration resulting in 50% inhibition) and inhibitory rates using the following formula: Inhibitory rate = (1 - absorbance of the experimental group/absorbance of the control group) × 100%.

Wright-Giemsa staining (Nanjing Jiangcheng Bioengineering Institute, Nanjing, China) was conducted according to the manufacturer's instructions on 1×10⁵ cells/ml seeded in 6-well plates, cultured for 24 h with the experimental drug. Cellular morphology was observed under an IX71 inverted optical microscope (Olympus Corporation, Tokyo, Japan). Hoechst 33258 (Sigma-Aldrich, St. Louis, MO, USA) staining was conducted on 1×10⁶ cells per experimental group under the microscope. Ultrathin sections (70–90 nm) were cut from samples of the control and 24 h treatment groups using a Reichert Ultracut-E ultramicrotome (Reichert-Jung, Vienna, Austria). Cells from the prepared sections were visualized using transmission electron microscopy (TEM) on a JEOL 1230 Transmission Electron Microscope (JEOL, Ltd., Tokyo, Japan).

A total of 1×10⁶ cells from each of the experimental and drug-treated (24 and 48 h) groups were collected and resuspended in 1 ml 10X annexin V binding buffer from a FITC Annexin V Apoptosis Detection Kit I (BD Biosciences, San Jose, CA, USA). A total of 100 l f of the solution was transferred into a cell culture tube; 5 μl fluorescein isothiocyanate (FITC)-annexin V and 5 μl propidium iodide (PI) from the kit were added. Simultaneously, FITC and PI were added to the blank well. Subsequent to mixing, the solution was cultured at room temperature in the dark for 15 min, 400 μl combined buffer was added into the cell culture tube, and flow cytometry was conducted using an Epics Altra Flow Cytometer (Beckman Coulter, Inc., Miami, FL, USA).

A total of 1×10⁶ cells from each of the experimental and drug-treated (24 and 48 h) groups were collected for flow cytometry. The cells were centrifuged at 300 × g for 5 min, washed twice with phosphate-buffered saline (PBS), fixed with 1 ml ice cold 70% ethanol and mixed, then maintained at 4°C for 24 h. Prior to testing, the mixture was centrifuged at 1,500 rpm for 5 min and washed twice with PBS prior to the addition of 400 μl RNA enzyme (30 μg/ml; Sigma-Aldrich). The cells were then incubated at 37°C for 30 min, mixed with 100 μl PI (200 μg/ml) and incubated in the dark at 4°C for 30 min prior to flow cytometry. The cell distribution percentages for the G₁, S and G₂ phases were then calculated using EXPO 32 MultiComp software, version 1.2 (Beckman Coulter, Inc.).

A total of 100 μg protein was collected from the experimental and 24 h drug-treated groups. Protein samples underwent polyacrylamide gel electrophoresis with 8% gel at 80 V (Wuhan Boster Biological Technology, Ltd., Wuhan, China) and were subsequently transferred onto polyvinylidene difluoride membranes. The samples were then blocked with nonfat milk and incubated overnight with polyclonal rabbit anti-human Bcl-2 (sc-782) or Bax (sc-493) primary antibodies (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) overnight at 4°C. Blots were incubated with the horseradish peroxidase-conjugated secondary antibody (1:5,000; anti-IgG; Santa Cruz Biotechnology, Inc.). The secondary antibody was added and incubated at 37°C for 2 h and visualized using staining with 3,3′-diaminobenzidine (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China). Quantity One 1-D analysis software, version 4.6.2 (Bio-Rad Laboratories, Inc.) was used for analysis of the grayscale values. The relative expression level of the target protein was calculated as follows: Target protein band gray value/internal reference (β-actin) gray value.

A total of 1×10⁶ cells were collected from each of the experimental and the drug-treated groups at 0, 6, 12, 18 and 24 h. These were analyzed using the Caspase-3 and Caspase-9 Colorimetric Assay kits (R&D Systems, Inc., Minneapolis, MN, USA), at a wavelength of 405 nm. The activity levels of caspase-3 and -9 in the control group were set as 1, while the optical density (OD) value/the control group OD was used to represent the activity of caspase-3 and -9 in the other groups. The formula used was as follows: Caspase-9 and -3 activity = (OD⁴⁰⁵ of sample - OD⁴⁰⁵ of blank)/(OD⁴⁰⁵ of control - OD⁴⁰⁵ of blank).

The current study used factorial design and the experiment was repeated three times with the results presented as the mean ± standard deviation. All data obtained were analyzed using SPSS software, version 18.0 (SPSS, Inc., Chicago, IL, USA). The data met the criteria for independence, normality and homogeneity of variance, therefore a t-test was selected for analysis. In addition, the χ² test was used when appropriate, with the test level set as α=0.05. P<0.05 was considered to indicate a statistically significant difference.

L-gossypol and LCD exhibited different degrees of time-dependent inhibition on the proliferation of SW982 cells (Fig. 1). The IC₅₀s of L-gossypol were 16.2, 4.7, 3.9 and 2.5 μmol/l at 24, 48, 72 and 96 h, respectively, whereas those of LCD were 5.12, 1.06, 0.32 and 0.20 μmol/l at 24, 48, 72 and 96 h, respectively. In addition, the combined inhibitory effect of L-gossypol and LCD exhibited a clear increase in a concentration- and time-dependent manner (P<0.001), indicating a strong synergy between the two drugs (Fig. 1, Table II).

Wright-Giemsa staining was conducted on SW982 HSSCs cocultured in 2.5 μmol/l L-gossypol and/or 0.2 μmol/l doxorubicin for 48 h, which revealed distinct morphological differences between the groups (Fig. 2). Control cells were tightly packed, large and predominantly appeared as clostridial forms with few cells exhibiting the polygonal form. The nuclei were large and centrally located, and a small number of cells deviated from the standard form. The majority of cells in the control group were observed to have two nuclei and the cytoplasm was abundant. However, in the experimental groups, the number and size of the cells was significantly reduced. In addition, a larger number of cells appeared to be irregular in shape and a small number were circular. The nuclei were observed to shrink in certain cells, while in others the nuclei fragmented, the fragments of which were observed moving towards the outer part of cells, forming film-coated apoptotic bodies (Fig. 2).

Control HSSCs were observed to be condensed, granular and exhibit clear fluorescence in either the nuclei or the cytoplasm following 48 h of treatment (Fig. 3). In addition, apoptosis was clearly evident. The cells shrank, the chromatin became condensed and marginated and clear fragments were observed inside the cytoplasm. The cell membrane was observed to form bubble-like protrusions and apoptotic bodies were formed.

The results from the TEM demonstrated that the control HSSCs were oval, with a large number of microvilli on the cell surface. The nuclei were clearly visible, the distribution of intranuclear chromatin was uneven and intracytoplasmic mitochondria were common (Fig. 4). Following drug coculture, the cells appeared to morphologically transition to exhibit apoptotic characteristics. These predominantly manifested as disappearance and condensation of the microvilli on the apoptotic cell surface, and the appearance of marginated intranuclear chromatin. Additionally, in the early cellular stages, HSSCs were crescent-shaped, whilst during the late stage the cells swelled and ruptured, the nuclei shrank and fragmented, a large number of vacuoles appeared in the cytoplasm and apoptotic bodies appeared (Fig. 4).

Following coculture for 48 h, the early apoptotic rates of L-gossypol and doxorubicin in the experimental groups were 18.40±2.20 and 23.20±2.45%, respectively, which were significantly higher than the control group (4.30±0.26%; P<0.001) (Table III, Fig. 5). The early apoptotic rate of L-gossypol/doxorubicin combination therapy was (38.70±3.40%). No significant interaction between the two drugs was observed (P=0.623), however, a single-factor analysis of variance revealed that the apoptotic rate in the combination group was significantly increased compared with the other groups (P<0.001).

The proportion of cells in the G₁ phase in the experimental groups increased significantly following 24-h drug treatment (P=0.05 and P<0.001 in the doxorubicin and L-gossypol groups, respectively; Table IV). As presented in Table V, following 48-h coculture, the proportion of L-gossypol group cells in the G₁ phase increased to 88.31±1.26%, compared with 80.95±1.61% in the control group (P=0.048). The numbers of cells in the S phase in the doxorubicin group (15.03±1.29%) and the combination group (26.29±1.16%) were also significantly different compared with the control group (9.48±0.65%; P<0.001) and a significant interaction was observed (P<0.001). These results indicate that L-gossypol predominantly arrests the cells in the G₁ phase, while doxorubicin and the combination therapy predominantly arrest cells in the S phase (Table IV and V, Fig. 6).

Bcl-2 protein levels following L-gossypol and doxorubicin coculture (18.30±1.32% and 19.73±2.01%, respectively) were significantly reduced compared with the control group (32.66±1.58%; P<0.001) (Table VI and Fig. 7). In addition, during combination therapy, the levels of Bcl-2 were further reduced (6.43±0.97%) while the interaction of the two drugs was not significant. Furthermore, L-gossypol and doxorubicin produced significant increases in Bax protein levels (18.08±1.46% and 12.36±1.80%) when compared with the control group (4.36±0.60%; P<0.001). Following combination therapy, Bax levels were further increased (25.31±1.41%), however, no significant interaction between the two drugs was observed (P=0.644).

Significant activation of caspase-9 and -3 in SW982 cells was observed in every treatment group at at least one time point, compared with time point 0 h (P<0.05). Caspase-9 activity reached its maximum value at 6 h subsequent to drug administration. Following 12 h, caspase-9 activity began to decline towards the control level. By contrast, no caspase-3 activity was observed prior to 12 h, and began to reduce at 18 h. The single-factor analysis of variance indicated that the combination group led to a significantly greater increase in caspase activity compared with the L-gossypol group, the doxorubicin group and the control group (P<0.05, Fig. 8). The data also demonstrate that the concentration required to achieve the same inhibitory effect was lower in the combination therapy group (data not shown).