Bitter Melon Extract Impairs Prostate Cancer Cell-Cycle Progression and Delays Prostatic Intraepithelial Neoplasia in TRAMP Model

Prostate cancer is the most commonly diagnosed cancer in men and one of the leading causes of cancer death in the United States (1). Although prostate cancer is frequently curable in its early stage by surgical or radiation ablation, many patients present locally advanced or metastatic disease for which there are currently no curative treatment options (2, 3). Therefore, more effective therapies that can cure localized tumors and prevent their metastasis are urgently needed. Cancer cells acquire resistance to apoptosis by overexpression of antiapoptotic proteins and/or by the downregulation or mutation of proapoptotic proteins. The intervention of multistage cancer by modulating intracellular signaling pathways may provide a molecular basis for chemoprevention with a wide variety of dietary phytochemicals (4). Therefore, an ideal approach to inhibit the progression of cancer is to induce cell-cycle arrest or apoptosis using dietary chemopreventive compounds.

Momordica charantia, also known as bitter melon, balsam pear, or karela, is widely cultivated in Asia, Africa, and South America and extensively used in folk medicines as a remedy for diabetes, specifically in India, China, and Central America (5). Animal studies have employed either fresh bitter melon extract (BME) or crude organic fractions to evaluate its beneficial effects on glucose metabolism and on plasma and hepatic lipids (6–11).

The antitumor activity of crude BME in murine lymphoma was reported earlier (12), although the mechanisms are poorly understood. Momordica protein of 30 kDa (MAP 30), isolated from bitter melon seeds, displayed antitumor activity in breast cancer model (13). The chemical modification of BME RIP significantly reduced its in vivo immunogenicity but retained its antiproliferative activity, as measured by DNA fragmentation and caspase-3 activation (14). The bitter melon seed extracts displayed anticancer activity in a rat colonic aberrant crypt foci model and a mouse mammary tumor model (15, 16). However, the acetone extract of bitter melon seed, which is probably rich in α-eleostearic acid, did not suppress the colon cancer growth in xenograft model (17). Therefore, more work will be necessary to understand the in vivo activity of bitter melon.

To our knowledge, this is the first study to show that BME prevents prostate cancer progression in TRAMP (transgenic adenocarcinoma of the mouse prostate) mice model. Mechanistic studies further indicate that treatment of prostate cancer cells with BME accumulates cells at the S phase of the cell cycle and induces cell-cycle arrest and apoptosis.

Prostate cancer cells, PC3 and LNCaP (obtained from American Type Culture Collection), were maintained in Dulbecco's modified Eagle's medium and RPMI with 10% fetal calf serum, respectively. Primary human prostate epithelial cells were obtained from Lonza and maintained for short time in specified media. Cell lines have not been independently authenticated. BME was prepared from the Chinese variety of young bitter melons (raw and green) as discussed previously (10). Briefly, BME was extracted using a household juicer and centrifuged at 560 × g at 4°C for 30 minutes, freeze dried at −45°C for 72 hours, and stored at −80°C until used for feeding studies. We prepared a stock of 0.1 g/mL in water, aliquoted, used 2% (vol/vol) for in vitro cell culture work, and 100 μL/mouse for oral gavage.

PC3 and LNCaP cells were treated with BME for 24 hours. Cells were trypsinized and fixed in ice-cold 70% ethanol overnight at 4°C. Cells were washed, stained with propidium iodide for overnight, and subjected to fluorescence-activated cell sorting (FACS) analysis on a FACScan flow cytometer (BD PharMingen) as described previously (18, 19). Data were analyzed using the CellQuest and ModFit software.

Prostate cancer cells were untreated or treated with BME at different time points, and cell lysates were prepared in 2× SDS sample buffer. Cell lysates were analyzed for Western blot analysis using PARP, caspase-3, caspase -7, cyclin D1, cyclin E, p21, phospho-MEK1/2, total MEK1/2, phospho-p38, total p38, phospho-ERK1/2, total ERK1/2, proliferating cell nuclear antigen (PCNA), and Bax antibodies (Santa Cruz Biotechnology or Cell Signaling), followed by enhanced chemiluminescence (Amersham Biosciences). Blots were reprobed with actin to compare protein load in each lane.

We chose TRAMP mice for our studies because TRAMP males develop spontaneous multistage prostate carcinogenesis that exhibits both histologic and molecular features recapitulating many salient aspects of human prostate cancer (20, 21). Furthermore, TRAMP model has been used successfully to test chemopreventive efficacy of several natural agents (22–25). Tg(TRAMP)8247 Ng mice) male mice (∼5 weeks old) were purchased from the Jackson Laboratory and divided into 2 groups (10 mice in each group) randomly to examine the effects of BME treatment on prostatic intraepithelial neoplasia (PIN) in TRAMP mice. Mice at 6 weeks received 0.1 mL water by oral gavage (control group) and 0.1 mL BME by oral gavage (experimental group), 5 d/wk for a period of 15 weeks. The mice were sacrificed 24 hours after the last administration of the water or BME by CO₂ inhalation. A portion of the prostate/tumor tissue was placed in 10% neutral buffered formalin and paraffin embedded. The other portion of the dorsolateral prostate was snap-frozen and stored at −80°C. During the study, animals were permitted free access to food and drinking water, and the animals were monitored daily for their general health. All animal treatments were according to the protocol approved by the Institutional Animal Care and Use Committee, and NIH guidelines.

Randomly selected fields (10–20) of hematoxylin and eosin (H&E)–stained sections (5-μm thick) of the dorsolateral prostate (DLP) from individual mice of both control and BME-fed groups were independently scored (blinded) by 2 researchers for incidence of prostate cancer. The percentage of area corresponding to each pathologic stage was also counted. Pathologic grading was done based on Shappell and colleagues (26).

The paraffin-embedded sections (5-μm thick) were deparaffinized and stained using antibody against PCNA or SV40 T-antigen, followed by 3,3′-diaminobenzidine staining. Ten fields of DLP (each animal) from the control and BME-fed groups were scored by 2 researchers, and the relative percentage of PCNA expression cells was calculated.

All statistical analyses were carried out with Statistical Analysis System software (GraphPad Prism 4.03), and P values less than 0.05 were considered significant.

We had initially treated prostate cancer cells with BME at different time points, and cell viability was determined by trypan blue exclusion. Prostate cancer cells died (>90%) within 96 hours of BME treatment, whereas primary prostate epithelial cells exhibited very little effects. We selected 2 prostate cancer cell lines, PC3 (androgen receptor negative and mutant p53) and LNCaP (androgen receptor positive and wild-type p53) for subsequent studies. To understand the mechanism of BME-mediated proliferation inhibition, we examined the effect of BME on the cell cycle by FACS as described previously (18, 19). Cell populations in the G₀–G₁ and S phases were 51% and 35% in control PC3 cells. However, after 24 hours incubation with BME, the S population was noticeably enhanced to 45%, whereas the G₀–G₁ population was decreased to 35% (Fig. 1A). We did not observe a significant change in the proportions of cells in G₂–M phase. Similarly, BME treatment in LNCaP cells also displayed enhanced S phase, 24% versus 12% in control untreated cells (Fig. 1B). This result suggests that BME treatment on prostate cancer cells induces S-phase arrest.

To identify cell-cycle regulatory molecules, we examined the expression of cyclins and cyclin-dependent kinase (CDK) inhibitors. Cells were treated with BME for 48 and 72 hours, and cell lysates were prepared for Western blot analysis. Our results suggested that the treatment of BME led to a significant downregulation of cyclin D1 and cyclin E in prostate cancer cell lines (Fig. 2A). We have also observed a significant increase of p21 protein in both the prostate cancer cell lines, suggesting an S-phase arrest (Fig. 2B). Interestingly, the expression levels of p27 protein remained similar in BME-treated PC3 and LNCaP cells as compared with untreated cells (Fig. 2B). Together, these results suggested that BME treatment in prostate cancer cells strongly modulates cyclins and inhibits their normal regulation of cell-cycle progression.

To further determine molecular pathways, we examined the activation of p38MAPK (mitogen-activated protein kinase) and ERK1/2, which are important upstream regulators of p21 (27, 28). Activation of p38MAPK and ERK1/2 are also correlated with the promotion of cell-cycle arrest in prostate cancer (29, 30). Western blot analysis revealed that phospho-p38, the activated form of p38MAPK, was significantly induced following BME treatment in both PC3 and LNCaP cell lines (Fig. 3A). Next, activation of ERK1/2 was examined. Prostate cancer cells treated with BME displayed upregulation of phospho-ERK1/2 expression (Fig. 3B). We further examined the status of MEK1/2, an upstream regulator of ERK1/2. Activation of MEK1/2 was observed in BME-treated cells as compared with control (Fig. 3C). Together, these results suggested that BME treatment in prostate cancer cells activates MEK–ERK or p38MAPK signaling pathway.

Cell-cycle arrest was correlated with apoptosis (31). Because we observed cell-cycle arrest and ERK/p38 activation, we next examined whether BME treatment in prostate cancer cells induces apoptosis. Cleavage of the DNA repair enzyme PARP from a 116-kDa protein to a signature peptide of 86-kDa fragment is associated with a variety of apoptotic responses. PARP is a nuclear protein and a downstream substrate of activated caspase-3/7. Prostate cancer cells treated with BME displayed a cleaved 86-kDa signature peptide (Fig. 4A). We observed an induction of cleaved caspase-3/7 and 9 in prostate cancer cells treated with BME (Figs. 4B and C). Mitochondria play a pivotal role in cell homeostasis, and the mitochondria pathway is tightly regulated by the Bcl-2 family proteins. To better understand the anticancer effect of BME treatment including growth inhibitory and proapoptosis effects, we next tested whether BME treatment induces the expression of proapoptosis markers in prostate cancer cells. Western blot analysis indicated that Bax was induced by BME treatment in both PC3 and LNCaP cell lines (Fig. 4D). These results showed that BME treatment in prostate cancer cells induces apoptosis.

The TRAMP mouse model is one of the popular mouse models for prostate cancer chemoprevention studies (22–25). These male mice develop spontaneous multistage prostate carcinogenesis that exhibits both histologic and molecular features recapitulating many salient aspects of human prostate cancer. The transgene construct in these mice is a PB-Tag gene consisting of the minimal −426/+28 bp regulatory sequence of the rat probasin promoter directing prostate-specific epithelial expression of the SV40 T-antigen gene introduced into the C57BL/6 mouse (32). TRAMP males characteristically express the PB-Tag transgene by 8 weeks of age and display PIN by 10 to 12 weeks of age (20). TRAMP [Tg(TRAMP)8247 Ng mice] male mice (5 weeks old) were purchased from the Jackson Laboratory and divided into 2 groups (10 mice in each group) randomly to examine the effects of BME treatment on prostate cancer development in TRAMP mice. The experimental group was gavaged BME (0.1 g/mL) 5 days a week, and control group were similarly fed with water. After 15 weeks of treatment with BME or water, TRAMP mice were sacrificed, prostates glands were removed, and formalin-fixed for immunohistochemistry analysis. The body weights of the control and BME-fed mice did not differ significantly throughout the experimental protocol. The mice were monitored everyday for signs of stress throughout the experimental period. At the time of sacrifice, all animals were examined for gross pathology, and there was no evidence of global edema, abnormal organ size, or appearance in nontarget organs. A significant weight difference of the prostate (70%, P < 0.013) was observed between control and experimental groups.

Histopathologic analysis of the prostate displayed less aggressiveness in the BME-fed group (Fig. 5). H&E-stained sections of prostate glands were microscopically examined and classified based on Shappell and colleagues (26): (a) low-grade PIN having foci with 2 or more layers of atypical cells with elongated hyperchromatic nuclei and intact gland profiles, and (b) high-grade PIN having increased epithelial stratification, foci of atypical cells fill, or almost fill the lumen of the ducts, enlarged diameter of glands, distorted duct profiles, increase in nuclear pleomorphism, hyperchromatic nuclei, and cribriform structures. We randomly selected 10 areas of each TRAMP mouse prostate section and scored the incidence of normal, low-grade PIN and high-grade PIN of prostate glands. The incidence of normal gland on the prostate section from BME-fed group was 64% significantly higher than observed in control group (23%, P < 0.0002). The incidence of low-grade PIN was similar from both groups. But the incidence of high-grade PIN from BME-fed group was 47% significantly lower than 78% from control group (Fig. 5A). We have observed that approximately 21% area of prostate gland was histologically normal in BME-fed group, as compared with approximately 12% normal glands in the control group (Fig. 5B). The area covered by low-grade PIN lesions in BME-fed group was 41% more than that in the positive control group (29%), and high-grade PIN in BME-fed group was 38% less than that in the control group (59%). We did not observe invasive cancer, and very rarely, the well-differentiated stages were observed. A representative H&E staining in sections of DLP from control- and BME-fed mice are shown in Fig. 5C. We did not observe a difference in SV40 T-antigen expression in control and BME-fed mice prostate tissues (Fig. 5D). Our results showed that control group displayed more high-grade PIN, whereas BME-fed group displayed low-grade PIN, suggesting that BME treatment significantly delayed the progression of PIN in TRAMP mice.

To assess the in vivo effect of BME feeding on the proliferation index in the dorsolateral prostate, tissue samples from TRAMP mice were analyzed by PCNA immunostaining. A representative immunostaining slides displayed significant decrease in PCNA-positive cells in BME-fed group as compared with the control group (Fig. 6A). Quantitative microscopic examination of PCNA-stained sections showed that PCNA-positive cells in BME-fed group were 49% as compared with control group (Fig. 6B). Western blot analysis of PCNA was further confirmed that PCNA expression was indeed inhibited in BME-fed mice as compared with control mice (Fig. 6C). We have also observed PARP cleavage (as observed decreased expression of full-length PARP) in BME-fed mice as compared with control TRAMP mice prostate tissues (Fig. 6D). Together, these results suggested that BME treatment in TRAMP mice strongly reduces PCNA expression that may prevent prostate tumor progression in TRAMP mice.

The novel findings in this study are that (i) BME treatment in prostate cancer cells induces cell-cycle arrest and eventually cell death, (ii) BME feeding in TRAMP mice inhibits prostate tumor progression without any toxicity, and (iii) the chemopreventive efficacy of BME is accompanied by delays of tumor progression at high-grade PIN, and a decrease in cell proliferation. The potential molecular mechanisms of BME-mediated inhibition of prostate cancer growth are a decrease in expression of critical cell-cycle regulatory proteins cyclin D1 and cyclin E as well as a block in S phase, with a concomitant increase in p21 protein level. This is the first article to our knowledge describing the mechanism of BME-mediated prostate cancer growth inhibition.

Uncontrolled cell growth and resistance to apoptosis are major defects in neoplasia. Development of approaches that induce the apoptotic machinery within cancer cells could be effective against their proliferation and invasive potential (33). A number of agents such as γ-irradiation, immunotherapy, and chemotherapy induce apoptosis in tumor cells as the primary mode of action for most anticancer therapies. Impairment of this pathway is implicated in treatment resistance (34). In fact, many chemopreventive agents of natural origin have shown promising anticancer properties by induction of the apoptotic pathway in transformed or tumor cells (4, 35). In the last few decades, considerable progress has been made in this direction leading to identification of many cancer chemopreventive agents, one of them being bitter melon, which has shown anticancer effects in other cancer cell types (12–17, 19, 36). Over the years, there has been a worldwide interest in BME as a dietary supplement because of its various health benefits, including lowering diabetes and lipidemia (9, 11). In a different study, we have shown that BME is well tolerated and has been termed as relatively safe in acute, subchronic, and chronic doses in animals (9).

The CDK inhibitors (p21, p27, and p57) causes cell-cycle arrest and inhibit the growth of cancer cells (37–40). Our in vitro results indicated that cyclin D1 and cyclin E were downregulated in BME-treated prostate cancer cells, and cyclin A was decreased in BME-treated PC3 cells (data not shown), whereas the CDK inhibitor p21 was upregulated, suggesting that BME inhibited prostate cell growth through the arrest of cell cycle and inhibition of proliferation. Interestingly, BME treatment in prostate cancer cells did not alter survivin expression as observed in breast cancer cells (19). We also observed the activation of p38 and ERK1/2 in prostate cancer cells treated with BME. p38MAPK and ERK1/2 are reported to be the upstream regulator of p21 (37) and involved in p21 dependent cell-cycle arrest (41). In prostate cancer, activation of p38 and ERK1/2 are also reported to promote apoptosis (42, 43). MEK1/2 is the upstream regulator of ERK1/2, which is also involved in the induction of apoptosis in prostate cancer (44). Our result indicated that BME treatment activates MEK–ERK and p38MAPK pathway, thereby inducing cell-cycle arrest and apoptosis.

In humans, prostate cancer progression is a multistage process involving an onset as a small carcinoma of low histologic grade, which progresses slowly toward metastatic lesions of higher grade disease. In TRAMP model, prostate cancer development closely mimics this human prostate cancer progression (20, 32). We have shown that BME-fed mice displayed lower incidence of high-grade as compared with control mice. This suggests that BME feeding for 15 weeks starting from the 6th week of age causes inhibition or delay of the tumor progression at the neoplastic stage. We also found that oral administration of BME causes a statistically significant decrease in cell proliferation in the dorsolateral prostate of TRAMP mice, as evidenced by reduced expression of PCNA and PARP cleavage. Therefore, the antitumor progression effect of BME in TRAMP mice could most likely be mediated, at least in part, via its effect on Cdk–cyclin–Cdk inhibitor axis.

In conclusion, the results of this study indicates, for the first time, that BME modulates cell-cycle regulatory molecules, thereby causing cell-cycle arrest and, eventually, apoptosis in prostate cancer cells. Oral administration of BME significantly inhibits prostate cancer progression in TRAMP mice without any adverse health effects or reducing T-antigen expression. Together, these results suggest that BME modulates several signal transduction pathways, which additively or synergistically restrict prostate cancer growth, and can be utilized as dietary supplement for prevention of prostate cancer.

No potential conflicts of interest were disclosed.

The authors thank Dr. Cheri West, Frank Speck, Frank Strebeck, and Anna Knobeloch for helping BME feeding to TRAMP mice and prostate tissues isolation, and Peter Humphrey for helpful discussion.

This work was supported by research grant R21CA137424 from the NIH.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.