Ursolic Acid Inhibits Proliferation and Induces Apoptosis of Cancer Cells In Vitro and In Vivo

The aims of the study are to explore the effect of ursolic acid (UA) on the growth of gastric cancer cell line BGC-803 and hepatocellular cancer cell H22 xenograft and to understand the mechanism. UA inhibits growth of BGC-803 cells in vitro in dose-dependent and time-dependent manner. Treated with UA in vivo, tumor cells can be arrested to G0/G1 stage. The apoptotic rate was significantly increased in tumor cells treated with UA both in vitro and in vivo. DNA fragmentation was found in BGC-803 cells exposed to UA. UA activated caspase-3, -8, and -9 and down regulated expression of Bcl-2 in BGC-803 cells. The expression of caspase-3 and -8 was elevated in tumor cells from xenograft treated with UA. 18F-FLT PET-CT imaging confirmed tumor model and UA effectiveness. Our results indicated that UA inhibits growth of tumor cells both in vitro and in vivo by decreasing proliferation of cells and inducing apoptosis.


Introduction
Gastric cancer is the second cause of cancer-related death worldwide, and it has now become the first cancer-related death in China. Hepatocellular carcinoma is a primary ma-lignancy of the liver and is the third leading cause of cancer-related death worldwide. There are about 110,000 people died of gastric cancer and hepatocellular carcinoma in China every year, which accounts for 45% (for both cancer combined) mortality worldwide. In addition to surgery, radiotherapy, and chemotherapy, it is essential to find a more effective way to treat gastric cancer and hepatocellular carcinoma.
More and more clinical practice shows that the Chinese medicinal herbs have antitumor activity, which sheds a light on new therapeutic strategy for cancer treatment [1][2][3]. Ursolic acid (UA) is a pentacyclic triterpene compound and exists in medicinal herbs such as Oldenlandia diffusa and Radix actinidiae. UA has been shown to have the effects of anti-inflammatory, antioxidant, and antitumor [4][5][6]. Studies have found that UA can inhibit the activities of DNA polymerase and DNA topoisomerase and decrease the rate of cell proliferation [7]. Moreover, UA can induce apoptosis of tumor cell by increasing the level of intracellular calcium ion [8], suppressing the expression of FoxM1 [9], and upregulating of death receptors [10]. In this study, we evaluated the effect of UA treatment on the gastric cancer cell and the hepatocellular carcinoma xenograft. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and DMSO were purchased from SIGMA Company. Annexin-V-FITC apoptosis kit was purchased from BD Biosciences. RPMI 1640 and calf serum was purchased from GIBCO, Invitorgen. All the antibodies used in western blot (rabbit antihuman β-actin, mouse antihuman procaspase-3, mouse antihuman procaspase-8, mouse antihuman procaspase-9, mouse antihuman Bcl-2, and mouse antihuman Bax antibodies) were purchased from SANTA CRUZ Biotechnology. Far infrared-marked goat antirabbit or mouse secondary antibodies were purchased from ROCK-LAND Immunochemicals. Odyssey far infrared scanner was purchased from LI-COR Biosciences (USA).

Cell Lines and Culture
Conditions. BGC-803 cell and hepatocellular carcinoma cell line H22 (ATCC, USA) were generous gifts from Dr. Yang Ke at the Cancer Research Center, Peking University. Gastric cancer cell line BGC-803 was maintained in RPMI-1640 with 10% calf serum. Cells were grown at 37 • C in a CO 2 incubator. The cells in the exponential growth phase were collected for cell proliferation, cell cycle, and apoptosis assay.

Mouse
Model. Ten male and ten female kunming mice were bought from The Research Center for Laboratory Animal Science in Beijing, and H22 cells of exponential growth phase were injected subcutaneously into right front axilla of kuming mice. The mice were divided into two groups with equal numbers of male and female mice. One group was given 0.2 mL/mouse physiological saline by oral administration as negative control; another group was given 2.53 mg/mouse UA everyday for 10 days. Mice were sacrificed at the end of the treatment. Tumors tissues were either fixed in 75 ethanol filtered after treated by 500 mesh to get the single cell suspension for flow cytometry analysis or fixed in 4% paraformaldehyde for pathological analysis. Samples prepared from tumor tissues were also flash frozen in liquid nitrogen for Western blot analysis. All animal experiments were approved by Animal Care and Use Committee, Inner Mongolia Medical College. Tumor cells obtained from xenograft were also fixed in 75% ethanol after filtered by 500 mesh. The single-cell suspension obtained above was used for apoptosis analysis using flow cytometry and for cell cycle analysis using ModFit analysis. DNA was extracted from BGC-803 cells according to protocol provided by manufacturer and then isolated in the 1.5% agarose gel by electrophoresis (65 V, 1.75 h). Photographs were taken using a gel-imaging system to detect DNA Ladder.

Western Blot Analysis.
Expression of procaspase-3, -8, and -9, Bcl-2, and Bax in BGC-803 treated with UA for 36 h at IC 50 and expression of caspase-3 and -8 in tumor cells obtained from xenograft was determined by western blot. Cells were treated with RIPA buffer (1% deoxycholate, 0.1% SDS). Concentration of protein was determined by BCA assay. Proteins were transferred to PVDF membrane after separation by SDS-PAGE and incubated with proper primary and secondary antibodies. The signal was detected by far infrared scanner. Result was analyzed by Quantity One. The relative amount of protein expressed was determined by using β-actin as an internal control.

Statistical
Analysis. Statistical analysis was performed using the program SPSS13.0. In each experimental group, one-factor analysis of variance and MANOVA of repeated measuring were used. The t-test was used for comparison between groups. A value of P < .05 was considered to be statistically significant. IC 50 values at different time points were calculated by Probit Analysis.

Induction of Apoptosis in BGC-803 Cells.
Morphological changes of BGC-803 cells after UA treatment. Cells not treated with UA showed single-layer adherence growth with rich cytoplasm, round or oval nuclei. Cells exposed to UA rounded up, lacked cell-cell contact with low density of cell (Figure 1(b)). Furthermore, H&E staining revealed that cells treated with UA showed typical morphology for apoptosis: nuclear fragmentation, chromatin condensation, and loss of membrane asymmetry (Figure 1(c)). In addition, fragments of degraded DNA were observed in agarose gel electrophoresis for BGC-803 cells treated with 43.78 and 35.94 μM UA. This was an indication that apoptosis occurred in the cells. For the control cells without treatment, no fragment of DNA was observed (Figure 1(d)). We further evaluated the apoptotic rate by Annexin-V staining. BGC-803 cells treated with UA had more apoptotic cells than cells treated with DMSO or without treatment (F = 72.579, P < .01, Figures 1(e) and 1(f)), while there was no significant difference in apoptosis between cells treated with DMSO or without treatment indicating that apoptosis was specifically induced by UA treatment. We investigated the mechanism how UA induced apoptosis. Expression of Pro caspase-3, -8, and -9, Bcl-2, Bax was detected by Western blot (Figure 1(g)). Relative amount of protein was calculated (Figure 1(h)). Our data showed that expression of pro caspase-3, -8, and -9 protein declined after UA treatment (P < .05), indicating that activated caspase-3, -8, -9 proteins were induced by UA treatment. Expression of Bcl-2 protein was downregulated by UA treatment (P < .05), while Bax protein expression did not show significant change with and without UA treatment. Our result suggested that UA may activate caspase cascade and downregulate Bcl-2 protein to induce apoptosis in gastric cancer cell line BGC-803.

Inhibition of Tumor Growth in Mouse Xenograft Model.
Compared to negative control, xenografts which were treated with UA were robust and had more weight (Table 2). There was 52.8% inhibition rate of tumor growth in UA treatment xenografts. Tumor tissue was examined under optical microscope after H&E staining. Tumor cells in negative control resembled nests or cords and showed vigorous mitosis. Nuclei of tumor cells were round or oval; while there was more necrosis and apoptosis in UA treatment group, density of tumor cells decreased. We examined the proliferation of tumor cells in xenografts by PET-CT. Xenografts was injected with 18 F-FLT, and the whole mouse was imaged by PET-CT. Focal 18 F-FLT uptake could be observed in tumor and abdominal cavity in xenografts of PS control (Figure 2 Percentange of 38.71±3.27 of the tumor cells treated with UA were in G0/G1 stage (P < .05, Figure 2(b)), but 48.97±3.96% and 23.53 ± 5.97% cells were in S or G2/M stage for the negative control, suggesting that UA treatment decreased proliferation of tumor cells in mouse xenograft model.

Induction of Apoptosis in Mouse Xenograft
Model. In addition to cell cycle, we also examined the apoptosis of tumor cells in xenografts. Tumor cells in xenografts treated with UA had more apoptosis as shown in Figure 2(b) (sub-G1). The apoptotic tumor cells were quantified by flow cytometry after cells were incubated with FITC conjugated Annexin-V. Comparison of apoptotic rate of the PS control to the UA treatment group revealed statistically significant different (12.91 ± 1.43 for PS control versus 37.24 ± 3.85 for UA treatment, P < .05, Figure 2(c)), indicating that UA treatment could induce apoptosis of tumor cells in mouse xenograft model. Furthermore, we examined the expression of Caspase-3, -8 in these tumor cells. The result of western blot showed that the protein level of Caspase-3, -8 was elevated after treatment with UA ( Figure 2(d)). Our data suggested that UA may induce apoptosis through upregulation of Caspase-3 and 8.

Discussion
UA is one anticancer active compound in Chinese anticancer herbal medicines. Previous studies showed that UA could inhibit growth of colon cancer cells, endometrial cancer cells, and melanoma cells [11][12][13]. It has been reported that UA inhibited growth of tumor cells through multiple functions, such as cytotoxicity [14,15], induction of apoptosis [9,10,16,17], and prevention of angiogenesis [18,19].
Our results showed that UA decreased proliferation of tumor cells both in vitro and in vivo. UA inhibited proliferation of BGC-803 cells in dose-and time-dependent manners. Inhibition rate of tumor growth is 52.8% in H22 xenografts treated with UA. Cell cycle of tumor cells from xenagrafts treated with UA was arrested in G0/G1 stage. Our data are consistent with the results of lung cancer cell A549 treated with UA [20], Hsu showed that UA up-regulated P21 expression through a P53-dependent manner and decreased expression of cyclins and their activating partner cdks. UA may block cell cycle progression in G0/G1 stage in gastric cancer cell and tumor cells from xenografts by activating P53/P21 pathway concomitant with inactivating cyclins/cdks. Mammalian cell apoptosis is initiated either by mitochondria-mediated or Fas receptor coupled extrinsic signals mediated pathways. Many apoptotic-related genes involved in these processes, including proapoptotic genes (Bax, Bid, and Bak), antiapoptotic genes (Bcl-Xl and Bcl-2), and cysteine proteases called caspases (caspase-3, -8, and -9). Bcl-2 family plays an important role in preventing apoptosis. The ratio of Bax to Bcl-2 can be used to assess apoptosis. The higher this ratio is, the more possibility apoptosis would happen, and vice versa. We found that the expression of Bcl-2 protein decreased, but the level of Bax protein did   [20]. Expression of Bax protein significantly increased in human breast cancer cell line MCF-7 [21]. The results above are different from ours, suggesting that UA may function differently in different type of cancer cells to induce apoptosis. Caspase-3 is a key protease in mammalian cell apoptosis. We found that caspase-3 was activated in BGC-803 cells and H22 cell xenograft treated with UA. It is reported that cytochrome C was released and caspase-3 was activated in BGC-823 cells with UA treatment [22]. Our data showed that caspase-8 and -9 were all activated in BGC-803 cells, indicating that UA induced apoptosis through two classical apoptotic pathways. Studies have demonstrated that caspase-3, -8, and -9 are necessary for UA inducing tumor cell apoptosis [23]. Our findings of decreased level of Bcl-2 with no change of Bax and activated caspase-8 and -9 in cell line BGC-803 enrich the understanding of mechanisms how UA induces apoptosis in gastric cancer cell line BGC-803. Furthermore, we found expression of caspase-8 was elevated in tumor cells obtained from H22 xenograft, indicating that UA also can induce apoptosis in vivo.

Conclusions
In summary, we found that UA inhibited growth of gastric cancer cell line BGC-803 in dose-dependent and timedependent manner in vitro. UA treatment arrested tumor cells from xenograft to G0/G1 stage in vivo. The apoptotic rate was significantly increased in tumor cells treated with UA both in vitro and in vivo (P < .05). DNA fragmentation was found in BGC-803 cells exposed to UA. UA activated caspase-3, -8, -9 and downregulated expression of Bcl-2 in BGC-803 cells, but had no effect on expression of Bax. Expression of caspase-3 and -8 were elevated in tumor cells from xenograft treated with UA. UA inhibited growth of tumor cells in vitro and vivo by decreasing proliferation and inducing apoptosis. Understanding the mechanism of UA will help us fully understand how Chinese medical herbs work.