Berberine Exerts Its Anti-tumor Effects By Targeting Cancer Cells As Well As Interfering With The Tumor Microenvironment In Prostate Cancer

Prostate cancer, one of the most frequently diagnosed cancers of men, leads to poor quality of life. Almost three-quarters of cases registered occur in developed countries; however, the mortality rates worldwide are less varied than that observed for incidence. While prostatectomy and androgen-deprivation are the choices of treatment for patients with early-stage prostate cancer, chemotherapy is often resorted to for patients with advanced prostate cancer metastatic to bone. Because conventional cancer chemotherapeutic agents are usually of high cost and have high toxic side effects, it is desirable to develop alternative chemotherapeutics that have relatively low toxicity and are less costly, such as those natural compounds in herbs.Berberine, a major bioactive compound in Huanglian, is one of the most commonly used herbal medicines. It is widely present in medicinal plants from Berberidaceae, Ranunculaceae, Papaveraceae and Rutaceae, and can be obtained easily. Traditionally used for treatment of gastroenteric discomfort, berberine has also been sought for the therapy of diabetes and cardiovascular diseases.There are a growing number of reports documenting the anti-tumor activity of berberine. Those studies invariably showed that berberine inhibits tumor cell growth either by inducing cell cycle arrest and/or apoptosis. However, the relative contribution of G1arrest, G2/M arrest or apoptosis to the inhibition of cancer cell proliferation may vary depending on the cells tested, dose and duration of treatment. In2009, Liu et al reported that berberine can induce G1arrest in a p53-p21dependent pathway that is activated by DNA damage, and at high dose berberine can induce G2/M arrest, which is not dependent on p53-p21cascade. Some studies showed that the levels of proteins that are associated with G2/M progression, such as CDK1, Cyclin B1, CDC25c,14-3-3σ and Weel, were altered in cancer cells treated with berberine. However, the signaling pathway that lies upstream of these factors remains to be identified.Epidemiological studies have shown that chronic inflammation can lead to various types of cancer. Cancer, on the other hand, can also generate an inflammatory microenvironment that promotes cancer progression. Tumor microenvironment consists of fibroblasts, endothelial cells, mesenchymal stem cells, lymphocytes and myeloid cells, as well as many types of soluble factors. It is reported that myeloid cells play a vital role in promoting tumor progression as important inflammatory mediators. Recent studies indicate that myeloid-derived suppressor cells (MDSCs) are required for the generation and maintenance of cancer-related inflammatory microenvironment. It is reported that MDSCs are elevated in peripheral blood and tumor tissues of cancer patients and also in bone marrow, spleen, blood and tumor tissues of tumor-bearing mice. MDSCs can promote tumor progression not only by suppressing host immune response but also by promoting tumor angiogenesis through regulating the bioavailability of VEGF via the production of MMP9.Considering the importance of cancer-induced host inflammatory microenvironment in tumor progression, it is logical to explore strategies that target the formation and maintenance of tumor microenvironment. Because berberine possesses anti-inflammatory properties, has low toxicity and is inexpensive, we set out to test whether berberine can contribute to tumor suppression by inhibiting cancer-related inflammatory response.In this study, we evaluated the anti-tumor effects of berberine on RM-1prostate cancer cells and RM-1transplanted tumor and characterized the underlying mechanisms.PART ONEIn order to investigate the effect of berberine on proliferation of RM-1cells in vitro and characterize the underlying mechanism, we employed MTT assay, flow cytometry, western blotting, Alexa Fluor(?)488annexin V and Propidium iodide (PI) assay and immunofluorescence staining method. We also employed specific inhibitors and small interfering RNA to test the functions of the factors involved.1. MTT results showed that berberine inhibited proliferation of murine RM-1prostate cancer cells in a time-and dose-dependent manner.2. Cell cycle analysis showed that treatment of RM-1cells with berberine (5,10,20μM) for24h or48h resulted in a significantly higher percentage of cells in the G1phase, with a concomitant reduction in the number of cells in the S phase. Berberine at50μM induced a significant accumulation of cells in the G2/M phase. These data suggested that the proliferation-inhibiting effect of berberine on RM-1cells was mediated by the induction of cell cycle arrest, a G1arrest at lower concentrations, and a G2/M arrest at a high dose.3. Western blotting analysis showed an activation of p53-p21cascade especially at lower berberine doses (5,10,20μM),which followed a time-and dose-dependent manner, as was previously reported for osteosarcoma cells.4. Apoptosis of RM-1cells was assessed by employing the Alexa Fluor(?)488annexin V and Propidium iodide (PI). The results indicated that berberine could induce apoptosis in a time-and dose-dependent manner.5. Berberine treatment led to an increased level of y-H2AX in a dose-dependent manner as determined by immunofluorescence staining or flow cytometry method. These results suggested that RM-1cells treated with berberine were indeed experiencing considerable genotoxic stress.6. Flow cytometry analysis indicated that G2/M arrest, not G1arrest, caused by berberine could be abrogated when RM-1cells were pretreated with2mM caffeine (an inhibitor of ATM/ATR) for1h, suggesting that G2/M arrest induced by berberine might be dependent on ATM/ATR pathway.7. Western blotting results suggested that berberine treatment resulted in an increase in the phosphorylation of Chkl (Ser345) in RM-1cells, which lies downstream of ATM/ATR and plays an essential role in the activation of G2/M checkpoint.We pretreated RM-1cells with UCN-01(a Chkl inhibitor) prior to berberine treatment to test whether it would have any effect on the berberine-induced G2arrest. The results indicated that the G2/M arrest induced by berberine (50μM) treatment for24h was indeed abrogated. As expected, berberine-induced G2arrest was significantly attenuated in Chkl siRNA-treated RM-1cells. These results indicated that berberine-induced G2/M arrest was Chk1dependent.8. We employed KU55933, a specific ATM inhibitor, to test whether ATM lies upstream of Chkl in establishing G2checkpoint in berberine-treated cells. The result showed that the G2/M arrest induced by berberine (50μM) treatment for24h was abrogated after pretreatment with10μM KU55933for2h. Together, these results above indicated that berberine-induced G2/M arrest was ATM/Chkl-dependent.9. The Alexa Fluor(?)488annexin V and Propidium iodide (PI) were employed to assess berberine-induced apoptosis when the G2/M checkpoint was abolished. The results showed that when compared with group treated with berberine alone for24h, apoptosis of RM-1cells was increased significantly when cells were pretreated with2mM caffeine for1h or10μM KU55933for2h before berberine treatment. Inhibition of Chk1, by applying UCN-01at300nM, had little effect on the apoptosis induced by berberine, although it efficiently abrogated the berberine-induced G2/M checkpoint. However, berberine-induced apoptosis was greatly enhanced by UCN-01when p53was inhibited, suggesting that abrogation of G2/M arrest by Chkl inhibitor sensitized the cells to berberine only when p53function was compromised.In conclusion, berberine inhibited proliferation of the murine RM-1prostate cancer cells in vitro, by inducing cell cycle arrest, including G1arrest and G2/M arrest, and apoptosis. G1arrest was mediated by the p53-p21cascade, and G2/M arrest, induced by higher dose of berberine (50μM), was dependent on ATM/Chkl pathway. Combined administration of berberine and caffeine, or ATM inhibitor, enhanced the berberine-induced apoptosis and accelerate the killing of RM-1cancer cells. Furthermore, UCN-01could sensitize the RM-1cells to berberine when p53function was compromised.PART TWOWe next investigated the effect of berberine, with anti-inflammatory properties, on tumor growth in vivo. We first established a prostate cancer-induced tumor microenvironment model by subcutaneously injecting murine RM-1prostate cancer cells into the right hand side rear flank of C57BL/6male mice. Berberine was administered by oral gavage. We employed anti-CD31immunohistochemistry and TUNEL assay to assess the tumor vascularity and apoptosis. In addition, the underlying mechanism of effect of berberine on RM-1tumor was studied by flow cytometry, real-time PCR and RNA interference.10. Mice,6～8week old, were inoculated subcutaneously with6×104RM-1prostate cancer cells, and then randomly divided into three groups. From the second day on, all three groups were treated by oral gavage with vehicle (sodium carboxymethycellulose, SCMC),100or200mg berberine/kg body weight daily, respectively. The results showed that200mg berberine/kg body weight significantly inhibited RM-1tumor growth after berberine treatment for17consecutive days.11. Berberine at200mg/kg body weight was used for further experiments. Vehicle and berberine (200mg/kg body weight) were administered daily for20 consecutive days by oral gavage. The volumes of RM-1tumors were measured with caliper every2days from the eleventh day on and the growth curves of RM-1transplanted tumors were obtained. The results showed that RM-1transplanted tumor growth was significantly inhibited with administration of berberine at a dose of200mg/kg body weight, with a50%decrease in tumor volume compared with vehicle-treated mice.12. We next investigated whether the decrease in RM-1tumor growth was associated with a decrease in tumor vascularity. RM-1tumors were harvested16days after tumor inoculation and frozen tumor sections were incubated with an anti-CD31antibody. A significantly lower blood vessel density was observed in tumors derived from mice treated with berberine than those derived from vehicle-treated mice, with27.21±1.7and36.47±2.5(P<0.01) blood vessels per field respectively (P<0.01). This decrease was accompanied by a significant increase in apoptosis in tumors derived from berberine-treated mice compared with those derived from vehicle-treated mice (P<0.01). These results suggested that berberine might inhibit tumor progression by modifying tumor-induced microenvironment in addition to its direct growth-inhibiting effect on RM-1cells.13. Because the expansion of CD11b+Gr1+myeloid-derived suppressor cells (MDSCs) is critical for tumor angiogenesis, we further determined the percentage of MDSCs in BM, spleen, blood and tumor tissues of tumor-bearing mice treated with vehicle as well as those treated with berberine. FACS data showed that the percentages of CD11b+Gr1+myeloid-derived suppressor cells in spleen and blood, but not in tumor tissues (about4%), were significantly decreased in tumor-bearing mice administered with berberine compared with those in vehicle-treated mice. MDSCs infiltrated into tumor tissues probably have transdifferentiated into other types of cells such as tumor-associated macrophages (TAMs), which are also known to promote tumor growth through enhancing tumor angiogenesis.14. It has been reported that the expression of stem cell factor (SCF) is correlated with the expansion of MDSCs and tumor angiogenesis. SCF, also known as c-kit ligand, has also been reported to be upregulated by inflammatory conditions in vitro, such as IL-1(3, and is highly expressed in many types of cancer cells. We found that SCF expression was significantly downregulated in tumor tissues from tumor-bearing mice administered with berberine compared with those from vehicle-treated mice (p<0.01), suggesting that the reduced production of MDSCs in berberine-treated mice might be caused by the downregulation of SCF expression. To test whether the suppression of host inflammatory response by berberine is mediated by the downregulation of SCF expression, we performed the following experiments.(1) Adherent RM-1cells derived from tumor tissue were cultured in vitro and were assayed for SCF expression. SCF expression was found to be decreasing with culture time from day1to day3, but to remain stable after day3, suggesting that tumor-induced inflammatory conditions might be required to maintain a high level of SCF expression in tumor cells.(2) The LPS-treated RAW264.7-conditioned medium was collected and used for culturing tumor tissue-derived adherent RM-1cells. In corresponding to the upregulation of inflammatory factors in LPS-treated RAW264.7cells, SCF expression in adherent RM-1cells treated with RAW264.7-conditioned medium for2.5hours or20hours was upregulated indeed (p<0.01). When LPS-treated RAW264.7cells were pretreated with berberine, the conditioned medium no longer had the ability to upregulate SCF expression in adherent RM-1cells.(3) In order to exclude the possibility that berberine directly suppresses SCF expression in RM-1tumor cells, we treated adherent RM-1cells derived from tumor tissues with different doses of berberine in vitro and found that berberine could not suppress SCF expression. Therefore, the downregulation of SCF expression in RM-1tumor tissues from berberine-treated mice is mediated by the suppression of RM-1 tumor-induced inflammatory response.15. SCF siRNA treatment was employed to investigate whether the reduced accumulation of MDSCs in tumor-bearing mice treated with berberine was dependent on the downregulation of SCF expression. RM-1cells were inoculated at24h after SCF siRNA treatment and MDSCs were measured on day6after tumor inoculation. The results showed that SCF interference reduced the accumulation of MDSCs in peripheral blood, indicating that SCF is required for the expansion of MDSCs.In conclusion, berberine treatment reduced SCF expression in RM-1tumor tissues. Downregulation of SCF expression resulted in the decrease of accumulation of MDSCs in peripheral blood and spleen, which might contribute to the reduced angiogenesis and increased apoptosis in RM-1tumor tissues, thus inhibiting the growth of RM-1tumors. Berberine might down-regulate SCF expression in RM-1cells through suppressing RM-1tumor-induced inflammatory response. These results indicated that suppression of cancer-related inflammatory microenvironment by drugs with anti-inflammation properties may contribute to the cancer therapy.In summary, our study demonstrated that berberine can target both tumor cells and tumor microenvironment in exerting its anti-tumor effect. We first reported that berberine-induced G2/M arrest was dependent on the ATM-Chkl pathway. Berberine-induced apoptosis can be enhanced by pharmacological inhibition of ATM. Chk1inhibition rendered cancer cells more sensitive to apoptosis when p53function was compromised. The usage of inhibitors targeting DNA repair response may improve the therapeutic effect of drugs, which can induce DNA damage. Due to the tumor-associated inflammatory microenvironment, SCF was upregulated in tumor tissues, which promoted the expansion of MDSCs that are conducive to tumor progression. By inhibiting the inflammatory response, berberine was able to downregulate the expression of SCF, thus attenuating the expansion of MDSCs. This finding illustrates the importance of targeting the tumor microenvironment in cancer therapy. Together, our results provided several insights into the potential of berberine as a cancer therapeutic drug.