2.1 Cells and reagents
EMT-6 cells were a gift from the Department of Molecular Biology, College of Basic Medicine, Jilin University. ICR female mice were purchased from the Liaoning Changsheng Biotechnology, Co. Limited, China. GBE was produced by Chi Sheng Chemical Corporation, Taiwan.
2.2 EMT-6 Cell culture
EMT-6 cells were incubated in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS, Hyclone USA), penicillin 100 IU / mL, streptomycin 100 ug / mL at 37℃ with 5% CO2.
2.3 Animal experiment
To investigate whether the application of GBE at an early stage of tumor development could help CTX inhibit the growth of EMT-6 cells, we administered GBE to the mice 24 hours after EMT-6 cells were injected into the mice. All procedures used in handling animals were approved by the Institutional Review Board of the School of Public Health, Jilin University. ICR mice were housed in the animal center for ten days to adapt to the environment. After adaptation, the mice were injected with EMT-6 cells at the concentration of 1X106 cells/mouse in the right thigh followed by random distribution of these mice into 4 treatment groups of 12 mice per group. The groups included PBS, GBE, CTX, and GBE+CTX. After 24 h, the mice in the PBS group were injected with PBS intraperitoneally at a volume of 0.2 ml; the mice in the GBE group were injected with GBE intraperitoneally at the concentration of 3mg/kg body weight; the CTX mice were injected intraperitoneally with CTX at the dose of 2mg/kg body weight; the mice in the GBE+CTX group were injected intraperitoneally with both CTX at the concentration of 2mg/kg body weight and GBE at the concentration of 3mg/kg body weight. The injections were conducted consecutively for 10 days when all the mice in the PBS group developed a palpated tumor, except for the CTX group which was injected every other day. A fifth group of 12 mice (negative controls) were not injected with EMT-6 cells but were included to compare body weight gain and serum malondialdehyde (MDA) and superoxide dismutase (SOD) levels in cancer-free mice. Euthanasia: All mice were anesthetized with 10% chloral hydrate before the mice were sacrificed. The changes in body weight, the latency of tumor onset, the growth of a tumor, tumor weight, serum MDA levels and SOD activities, and the key elements in the NF-κB signaling pathway, PI3K/AKT pathway and apoptosis were measured and compared among different groups.
2.4 Body weight measurement
Body weight is a basic parameter of body status. It is a marker of the effect of the tumors on the growth of the mice. The body weight of the mice was measured every day until the mice were sacrificed. The growth trends of body weight were compared among treatment groups and the control group.
2.5 Tumor onset and growth
To explore whether GBE and CTX would prolong the onset of tumor and inhibit the growth of EMT-6 cells, we determined the tumor onset and tumor growth. Tumor onset was determined by palpation. The site of tumor injection was observed and palpated each day. The day of tumor onset was recorded for each and every mouse to compare the tumor latency among the different groups. The length and width of the tumor was measured. The tumor volume was calculated based on the formula V=(a*b2)/2, where a is the length of the tumor and b is the width of the tumor. When each and every mouse in the PBS group developed a tumor, all the mice from different groups were sacrificed and the tumor from each mouse was removed and weighed. Tumor weight was compared among the four groups.
2.6 Determination of malondialdehyde (MDA) and superoxide dismutase (SOD) in serum
Tumor growth is closely associated with ROS production, which is also an underlying mechanism of tumor growth. We measured the level of MDA and activity of SOD in serum. Serum MDA and SOD as described in detail in Ren et al [13], was measured in each mouse in the four treatment groups and in the untreated control mice.
2.7 Western blot
The NF-κB signaling pathway, PI3K/AKT pathway and apoptosis are implicated in the multiple aspects of tumor development. To reveal the mechanisms underlying GBE and CTX inhibiting the growth of EMT-6 cells, we determined the expression of key elements in the NF-κB signaling pathway, PI3K/AKT pathway and apoptosis. Western blot analysis was conducted as in the protocols described in Ren et al [13]. The antibody information was NF-κB (1:1000, CST USA), p-NF-κB (1:1000, CST USA), PI3K (1:1000, CST USA), Akt (1:1000, CST USA), IκB (1:1000, CST USA), caspase 8 (1:1000, CST USA), CytC (1:1000, Santa Crue Biotechnology, USA), FADD (1:1000, Santa Cruz Biotechnology, USA), p53 (1:1000, Santa Cruz Biotechnology, USA), Bax (1: 500, Bioss China), Bcl-2 (1:500, Bioss China), actin (1:1000, CST, USA). Detection was done by measuring the chemiluminescence of ECL reagent (PIERCE, USA). The photographs generated were quantitatively analyzed with a Quantity One image densitometer. Protein levels were standardized by comparison with anti-β-actin antibody.
2.8 Statistical analysis
2.8.1 ROS biochemical analyses
Results are presented as mean±SD. SPSS 21.0 software was used to do the statistical analysis by ANOVA and Duncan multiple comparisons were carried out for comparison between groups. A p <0.05 was considered significant.
2.8.2 Differences in median tumor weights and latency times between treatment groups
Normality was tested using quantile plots. We used a Kruskal-Wallis statistic to generate a 100,000 bootstrapped samples from which to estimate the p-value for testing for overall differences among groups. Pairwise comparisons of medians were conducted with an exact Wilcoxon statistic and a Dunn’s test was used to adjust for multiple comparisons.
2.8.3 Survival analysis
We tested whether the data met the proportional hazards assumption and used the Cox proportional
hazards model to assess differences in time-to-event analysis with uncensored data because all outcomes
were observed at 10 days. The four treatments groups were coded as 1=PBS, 2=CTX, 3=GBE and
4=GBE+CTX. Exponentiated hazard ratios (HR) with 95% confidence intervals (CI) were calculated. A
HR less than 1.0 indicates that the lower coded value demonstrated a shorter latency period. We used the
method of Peto and Peto (1972), a modified log rank test, to generate the p-value because it works well
in small samples and is better at detecting early differences in survival. We plotted hazard curves for the
treatment groups [14]. Due to a small sample size and truncated survival period, we compared the effect sizes for the hazard ratios between the GBE+CTX vs PBS groups to the CTX vs PBS groups.