From August, 2010 to December, 2012, 66 glioma tissue samples were obtained from patients who underwent resection of cerebral gliomas. Normal brain tissue samples were obtained from 5 patients with brain trauma. This study was approved by the Institutional Research Board of Harbin Medical University and a signed consent form was obtained from each patient prior to obtaining the samples. Histopathological examination revealed that of the 66 patients, 15 had pilocytic astrocytoma, 8 had oligodendroglioma, 13 had anaplastic oligoastrocytoma and 30 had glioblastoma. According to the WHO classification of tumors of the central nervous system (1), our cases were graded as follows: 23 cases belonged to the grade I–II low-level group and 43 cases belonged to the grade III–IV high-level group. Of the 66 cases, 35 were males and 31 were females. The age of the patients ranged from 5 to 65 years (mean age, 36.7±15.2 years). None of the cancer patients had received radiotherapy or chemotherapy prior to surgery (Table I). Furthermore, another set of 10 human glioblastoma (WHO grade IV) patient tissue samples were collected at the same hospital and were randomly selected for further analysis. The same patient selection and diagnosis criteria, as well as the informed consent form, were applied to this set of specimens. All specimens confirmed by neuropathological diagnosis were stored at −80°C after freezing in liquid nitrogen.

In the present study, we used various human cancer cell lines (originally obtained from ATCC and maintained in our laboratory) as follows: the U87MG, RG, T98, LN229, LN18 and U343 glioma cell lines; the HT29, LoVo, HCT116, SW620 and SW1116 colorectal cancer cell lines; the DU145 and PC3 prostate cancer cell lines; the HeLa cervical caner cells; the A431 human epidermoid carcinoma cells; and the K562 leukemia cells. The cells were grown in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS). All cell lines were cultured in an atmosphere of 5% CO₂ at 37°C.

TRIzol reagent (Invitrogen, Carlsbad, USA) was used to extract total RNA from the tissues and cell lines. SuperScript II transcription enzyme, oligonucleotide(dT), and random primers (high capactiy cDNA reverse transcription kit; Applied Biosystems) were incubated with total RNA for reverse transcription. The entire process was carried out in accordance with the manufacturer's instructions. The synthesized cDNA was stored at −70°C. The primers used to amplify the caspase-8 gene and the internal reference genes, glyceraldehyde-phosphate dehydrogenase (GAPDH) in tissues and β-actin in cell lines, are listed in Table II. The size of the amplified fragments for caspase-8, GAPDH and β-actin were 427, 240 and 336 bp, respectively. The 50 µl PCR reaction included 1X buffer, 1.5 mM MgCl₂, 10 pmol bidirectional primers, 0.2 mM deoxynucleoside triphosphates (dNTPs), 1 unit TaqDNA polymerase, cDNA and double-distilled water. A PCR reaction without a template was used as the negative control. The PCR reaction conditions were as follows: denaturation at 94°C for 5 min; denaturation at 94°C for 30 sec, annealing at 60°C for 45 sec, extension at 72°C for 45 sec, 28 cycles; extension at 72°C for 10 min. PCR products were separated on a 1.2% agarose gel (with ethidium bromide staining). The resultant gel image was analyzed using the AlphaImager gel analysis system. Each analysis was repeated 3 times, and the mean was obtained in order to reduce error. The level of caspase-8 mRNA expression in the glioma and normal brain tissues was determined by semi-quantitative RT-PCR. The semi-quantitative value was expressed as an integrated optical density ratio (irOD), where irOD = (average caspase-8 electrophoresis optical density × area)/(average GAPDH electrophoresis optical density × area). A ratio of >0.5 was considered as positive expression, and a ratio of ≤0.5 was considered as negative expression.

The Qiagen DNeasy kit (Qiagen Co. Ltd., Shanghai, China) was used to extract DNA from the tissues and cell lines. Extracted DNA was first subjected to agarose gel electrophoresis to detect whether there was any degradation. Subsequently, the quality and quantity of the extracted DNA was determined using an ultraviolet (UV) spectrophotometer (BioPhotometer plus; Eppendorf Co., Ltd., Hamburg, Germany). An OD value of 260/280 nm between 1.8 and 2.0 was used as the quality standard.

As per the reported literature (22), bisulfite modification was performed as follows. First, 2 µg of DNA were diluted with 25 ml distilled water and heated for 5 min in a water bath maintained at 95°C. The denaturation DNA was then mixed with 2% low-melting-point agarose gel. Second, agarose beads were prepared using paraffin sealing film. The beads were placed into microtubes, and 1 ml of fresh sulfite solution (3.0 M sodium bisulfite, 1 mM hydroquinone, pH 5.0) was added to each microtube. The samples were kept at 55°C in the dark in a water bath for 16 h. The samples were then washed twice for 15 min each using 1 ml wash buffer [10 mM Tris-HCl, pH 8.0, 10 mM ethylenediaminetetraacetate (EDTA)] and then washed 3 times for 15 min each using 1 ml 10 N NaOH solution to remove acid. The samples were then neutralized with 200 µl of 1 N HCl 3 times, for 15 min each time. Finally, the samples were washed twice with the wash buffer. Agarose beads were cut into 4–6 equal-sized sections to be used as the methylation-specific PCR (MSP) template.

The primers used to detect the methylation status were synthesized according to the procedures outlined in the study by Hopkins-Donaldson et al (11). The primer sequences for caspase-8 methylation and demethylation are listed in Table II. The sizes of the amplified products were 320 and 322 bp. PCR reaction conditions were in accordance with the methods described in the study by Hopkins-Donaldson et al (11). Known methylated DNA was used as the positive control; the negative control was the PCR reaction without a template. PCR products were separated on a 1.5% agarose gel (with ethidium bromide staining) and then photographed and analyzed under UV light. In addition, purified PCR products were ligated to the plasmid pGEMT at 4°C at an appropriate molecular ratio. Plasmids with high purity and quality were selected and delivered to a gene sequencing company (BGI Life Technologies Co., Ltd., Beijing, China).

Caspase-8 activity in the HT29 and SW620 cells was assayed after the start of treatment with 5-Aza-2-deoxycytidine (5adc, 0.5 µM) for 24/48 h and in LN18 cells for 18/24/48 h. The cells were collected and analyzed using the Caspase-8 Colorimetric assay kit (R&D Systems, Minneapolis, MN, USA). The cells were lysed to collect their intracellular contents. The cell lysate was tested for protease activity by the addition of a caspase-specific peptide that was conjugated to the color reporter molecule p-nitroanaline (pNA). The cleavage of the peptide by the caspase releases the chromophore pNA, which was quantified spectrophotometrically at a wavelength of 405 nm using a UV spectrophotometer (BioPhotometer plus; Eppendorf Co., Ltd.). The level of caspase enzymatic activity in the cell lysate was directly proportional to the color reaction. 5-Aza-2-deoxycytidine was purchased from Sigma-Aldrich (Shangai, China).

The apoptotic cells were assayed after the use of camptothecin (CPT; 50 µM), or the combination of CPT (50 µM) and 5adc (0.5 µM). These agents were respectively used with the HT29, LN18 and SW620 cells for 24 h. Apoptotic cells were assessed by Annexin V staining using the Annexin V-FITC apoptosis detection kit (BD Bioscience, Beijing, China) according to the manufacturer's instructions and analyzed by flow cytometry using CellQuest (FACSAria I; BD Biosciences, Franklin Lakes, NJ, USA). In the Annexin V assay, the cells with staining only for Annexin V-FITC (early apoptosis) were located in the lower right quadrant, whereas the cells with staining for both Annexin V and PI (late apoptosis) were located in the upper right quadrant. The percentage of positive cells for PI staining only in the top left quadrant represented the necrotic population. CPT was purchased from Sigma-Aldrich.

The statistical package for social sciences (SPSS) 19.0 statistical analysis software was used. The χ² test was performed on the count data, and analysis of variance (ANOVA) and the t-test were performed on the measured data. A value of P<0.05 was considered to indicate a statistically significant difference.

We used RT-PCR to analyze the expression of caspase-8 at the mRNA level in tissue of human gliomas and normal brain specimens, and semi-quantified the expression levels using the housekeeping gene, GAPDH, as the internal reference control. The results revealed that 80.0% of the normal brain tissue samples (4/5) had an intermediate to a high level of caspase-8 expression (positive) (Fig. 1A), whereas only 18.2% of the human glioma tissue samples (12/66) had a high level of caspase-8 expression (positive) (Fig. 1A). The remaining 54 glioma tissue samples [81.8% (54/66)] had very low to null levels of caspase-8 expression (negative) (Table I). Therefore, caspase-8 expression at the mRNA level in the human glioma tissues was significantly lower than that in the normal brain tissues (P<0.01; Fig. 1B).

To determine whether the expression of caspase-8 is involved in the development of human glioma, we examined the association between the caspase-8 expression level and clinicopathological indexes using the χ² test. Statistical analysis revealed that the expression level of caspase-8 was closely related to the tumor grade (P<0.05; Table I); however, no significant correlation was observed with gender, age or tumor size (P>0.05).

In order to investigate the mechanisms underlying the downregulation of the caspase-8 gene in human glioma tissue, we used MSP to detect the methylation status of the caspase-8 gene at the CpG island. The results indicated that, among the 66 samples, 52 samples exhibited caspase-8 gene methylation at the CpG islands, as confirmed by repeated experiments. Among the 52 samples, 9 of them expressed medium to high levels of caspase-8 at the mRNA level, whereas 43 samples expressed very low to null levels of caspase-8; yet, 25 of the 43 samples exhibited a status of high gene methylation at the CpG islands (Fig. 2A and B). Furthermore, statistical analysis revealed that caspase-8 methylation was closely related to its expression at the mRNA level in human glioma tissue (P<0.01; Fig. 2B).

Additionally, PCR product sequencing with the MSP assay confirmed that the CpG islands of the caspase-8 gene in the methylation-positive samples were indeed methylated (Fig. 2C). Moreover, among a separate set of WHO grade IV GBM patient samples that were further examined for DNA methylation (Fig. 2D) and the expression status of caspase-8 at the mRNA level (Fig. 2E), 6 out of a total of 10 GBM samples had gene methylation at the CpG islands (Fig. 2D). In addition, the mRNA expression levels of caspase-8 significantly correlated with the gene methylation status (Fig. 2D and E).

To verify the caspase-8 gene methylation status in human cancer cells, we first detected the status in 6 glioma cell lines and found that only the LN18 cells had relatively low caspase-8 gene methylation (Fig. 3A). We then examined 6 cell lines of other cancer types, including HT29 colorectal cancer cells, HeLa cervical cancer cells, K562 myelogenous leukemia cells, A431 epidermoid carcinoma cells, and DU145 and PC3 prostate cancer cell lines. The results revealed that, among these 6 cancer cell lines, only the HT29 colorectal cancer cells exhibited high caspase-8 gene methylation at the CpG islands, whereas the PC3 prostate cancer cells appeared to have a relatively low methylation status. The remaining cell lines were not found to have a clear methylation (Fig. 3B). To further verify the caspase-8 gene methylation status, we examined 4 other colorectal cancer cell lines, including LoVo, HCT116, SW620 and SW1116 cells. The data indicated that these colorectal cancer cells hardly exhibited caspase-8 gene methylation at the CpG islands (Fig. 3C).

We then used RT-PCR to analyze the expression of caspase-8 at the mRNA level in 5 colorectal cancer cell lines and to further semi-quantify the gene expression levels using the housekeeping gene, β-actin, as the internal reference control. The results revealed that the HT29 cells, in which the caspase-8 gene was highly methylated, exhibited a lower level of caspase-8 expression (Fig. 4A); the other colorectal cancer cell lines exhibited a higher level of caspase-8 expression, but relatively low gene methylation, as compared to the HT29 cells (Figs. 3B and C and 4A).

We further verified whether caspase-8 protein kinase activity was highly relevant to its gene methylation status by treating the tumor cells with 5adc, a demethylation reagent. The data indicated that treatment with 5adc restored caspase-8 protein kinase activity in the HT29 and LN18 cells, but not in the SW620 cells. In the HT29 cells, the expression of caspase-8 was upregulated at both 24 and 48 h (P<0.05; Fig. 4B, panel a); in the LN18 cell, the upregulation occurred only at 24 h (P<0.05; Fig. 4B, panel b); however, in the SW620 cells, caspase-8 activity was not restored (Fig. 4B, panel c) following treatment with 5adc.

To confirm the function of the upregulated caspase-8 protein activity by treatment with a demethylation agent, we assayed the apoptotic cells in the HT29, LN18 and SW620 cell lines following treatment with CPT (50 µM) alone, or in combination with 5adc (0.5 µM) for 24 h. The apoptosis assay carried out by flow cytometry revealed that the HT29 cell apoptotic rate in the combination group was 79.3±8.07%, which was significantly higher than that in the group treated with CPT alone (18.67±3.52%; P<0.05; Fig. 4C, panel a). These values (79.3±8.07% and 18.67±3.52%) are presented as mean values of at least 3 experiments. In Fig. 4C, the values are presented as the percentage of one experiment. However, no difference was observed as regards the apoptotic rate of the LN18 and SW620 cells between CPT treatment alone and the combination therapy (P>0.05; Fig. 4C, panels b and c).