Berberine suppresses tumorigenicity and growth of nasopharyngeal carcinoma cells by inhibiting STAT3 activation induced by tumor associated fibroblasts

Berberine Chloride (C₂₀H₁₈ClNO₄) was purchased from Sigma Chemicals (St. Louis, MO, USA). It was dissolved in sterile milli-Q water at a 80°C water bath for 10 mins to a stock concentration of 5 mM and stored at -70˚C before use. The primary antibodies used to detect p-STAT3 (Tyr 705), STAT3, cleaved-PARP-1, and β-actin were purchased from Cell Signaling Technology (Beverly, MA, USA). Antibody for detecting cytokeratin (clones AE1/AE3) was purchased from Dako (Carpinteria, CA, USA). The antibodies to detect Mcl-1, IL-6 (clone H-183) and the horseradish peroxidase (HPR)-linked secondary antibodies goat anti-mouse and goat anti-rabbit IgG were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). IL-6 and anti-IL-6-receptor antibody were purchased from R&D system (Minneapolis, MN, USA).

C666-1 is a subclone of its parental cell line, C666, derived from an undifferentiated NPC xenograft of southern Chinese origin [25]. HONE-1 is derived from a poorly differentiated NPC from Chinese patient [26]. HK1 was established from a recurrent well-differentiated NPC of a Chinese patient after radiation therapy [27]. C666-1, HONE1 and HK1 cells were cultured in RPMI-1640 medium (Sigma) supplemented with 10% fetal bovine serum (FBS) (Sigma), 100 μg/ml penicillin/streptomycin (Invitrogen, Carlsbad, CA, USA) and maintained at 37˚C in a humidified atmosphere of 5% CO₂. NP460 is an immortalized nasopharyngeal epithelial cell line established in our laboratory [28]. It is a non-tumorigenic cell line derived from normal nasopharyngeal tissue. It was cultured in a 1:1 mixture of Defined Keratinocyte-SFM (Invitrogen) and EpiLife™ medium with full supplements (Invitrogen). Tumour-associated fibroblasts were derived from primary cultures of NPC biopsies. Prior patient consents were obtained for the use of biopsied tissues for research investigation. The collection and use of the specimen have been approved by the Human Research Ethic Committee of the University of Hong Kong. The sample was collected in Queen Mary Hospital in Hong Kong. The NPC tissue was cut into small pieces (1 mm³) and left to grow in RPMI-1640 medium (Sigma) supplemented with 10% FBS (Sigma). After one week, the fibroblasts grew out from the NPC biopsies will be frozen down in liquid nitrogen for future use.

The in vivo nude mouse tumorigenicity assay was performed by injecting a total of 1 X 10⁶ C666-1 cells subcutaneously into the flank of 6–8 week old male nude mice. At the same day of injection of tumor cells, the drug was then administrated into the mice intraperitoneally (i.p.). The mice were either injected with saline (control group) or with berberine concentrations of 5 mg/kg body weight (low dose group) and 10 mg/kg body weight (high dose group) every other two days. Once tumor growth was established in the control mice (i.e. 14 days post-injection of tumor cells), tumor sizes were measured every other day. Tumor volume (mean ± SD) was calculated as length x width²/2 [29].

Cell lysates were collected by scraping the cells with RIPA lysis buffer from the culture dishes, and the protein concentrations were determined using the DC Protein Assay Kit (Bio-Rad, Hercules, CA, USA) according to the manufacturer’s protocol. Equal amount of protein lysate (20 μg) per sample was resolved on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a polyvinylidene fluoride (PVDF) membrane (Amersham, Piscataway, NJ, USA). Membranes were probed with desired primary antibodies followed by detection of chemiluminescent signals of the peroxidase-conjugated secondary antibody using ECL Plus Western blotting detection system (Amersham, Buckinghamshire, UK). β-actin was used as an internal control to verify basal expression levels and equal protein loading. The ratio of the specific proteins to β-actin was calculated.

Subcutaneous tumors developed in nude mice were excised on 43 days post-injection of C666-1 cells. Sections for immunohistochemistry were dewaxed in xylene and rehydrated in graded alcohol. Endogenous peroxidase activity was blocked by incubating slides with 3% hydrogenous peroxide for 10 min. For antigen retrieval, all slides were incubated with 10 mmol/L citrate buffer (pH 6.0) for 93°C for 10 min, and then cooled down to room temperature. After that, the sections were rinsed with PBS and treated with normal blocking serum (Vector Laboratories, Inc., Burlingame, CA, USA) for 30 min. Anti-p-STAT3 (1:100, Cell signaling) and anti-cytokeratin antibodies (1:200, Dako) diluted in PBS were applied to the sections and incubated at 4°C overnight. After rinsing, all sections were further incubated for 1 hr with biotin-conjugated secondary antibody and horseradish peroxidase-conjugated streptavidin followed by using diaminobenzidine (Dako) as a chromogen. Counterstaining was performed by hematoxylin before dehydration and mounting.

The concentrations of IL-6 secreted from the NPC fibroblasts were quantitated using ELISA assay for IL-6 (R&D Systems) according to manufacturer instructions. Triplicated samples were estimated and the averages were used in the analysis.

The effect of berberine on cell viability/ proliferation was determined using the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide) assay. MTT, a tetrazolium salt, is reduced to a purple blue formazan product by dehydrogenases in the mitochondria of living cells. The cell viability can thus be determined by quantifying the purple blue formazan product based on UV absorbance. Briefly, 5000 cells/well were plated in 96-well plates and incubated for 24 h. The cells were then treated with berberine at indicated concentrations for 24, 48 and 72 hr. The cells were then treated with 10 μl of 5 mg/ml MTT (Sigma) and incubated for 4 hr at 37˚C. The medium was then discarded, and 200 μl of dimethyl sulfoxide (DMSO) (Sigma-Aldrich, MO, USA) was added to dissolve the resulting formazan crystals. The absorbance was measured at 570 nm by the Multiskan MS microplate reader (Labsystems, Finland) with a blank reference at 650 nm.

The data from each experiment were expressed as mean ± standard deviation (SD). Student’s t test was used to assess the differences between experimental groups. A p value < 0.05 was considered as statistically significant throughout this study.

The use of animals in this study was approved by the Committee on the use of live animals in teaching and research, the University of Hong Kong, Hong Kong.

Corptis rhizoma is commonly used in several treatment remedies for cancer and berberine is the major active ingredient extracted from it [6, 7]. In vitro inhibition of growth of human cancer cells by berberine has been reported by multiple studies [7]. It is crucial to examine any anti-tumor effects of natural product using in vivo tumor models. In this study, we have first assessed whether berberine could suppress the tumorigenic growth of NPC cells in immune deficient mice. The C666-1 cell line used for in vivo investigation harbours EBV infection which is representative of NPC from southern Chinese. To examine for inhibition of tumorigenicity and growth in vivo, 1 X 10⁶ C666-1 cells were injected subcutaneously into the flank of nude mice. On the same day of injection, the mice were either injected intraperitoneally (i.p.) with saline or low or high doses of berberine (5 mg/ and 10 mg/kg respectively) every two days. In the control mice (saline-treated), tumors started to appear after 14 days post-injection, and substantial growth of the tumor masses could be observed (Figure 1a). However, no measurable tumor masses could be detected in the berberine-treated mice until 30 days after injection of tumor cells in the 5 mg/kg berberine treatment group and 43 days in the 10 mg/kg berberine treatment group respectively (Figure 1a). By examination of the growth curves of individual tumor developed in control and berberine-treated groups, we could clearly observe a steady increase of tumor size in 4 out of 4 mice in control group after 14 days post injection (Figure 1b). Significant inhibition of in vivo tumorigenic growth of C666-1 was observed in both berberine treatment groups (5 mg/kg and 10 mg/kg). In treatment group injected with berberine (5 mg/kg), tumor growth was detected at 31 ± 5 days post-injection. In treatment group injected with higher dose (10 mg/kg), tumor growth was detected at 37 ± 5 days post infection (Figure 1b). Furthermore, tumorigenic growth of C666-1 was observed in only 2 out of 5 mice injected with 10 mg/kg (Figure 1b). These data suggest that berberine effectively suppresses the establishment and subsequent tumorigenic growth of NPC cells in vivo. Berberine is a natural compound found in traditional Chinese medicine and generally regarded as having low toxicity. Nonetheless, we have examined if there are any toxicity induced in berberine-treated mice in terms of mortality of the animals and alteration in body weight. The body weight and well-being of the control and berberine-treated mice were monitored periodically starting from day 1 of berberine administration. We did not observe significant alteration of body weight, morbidity and mortality after berberine treatment (Figure 1c). These results confirmed the findings from other reports stating that berberine is low in toxicity and does not induce adverse or detrimental effects on the general healthiness of mice.

Constitutive activation of STAT3 was commonly observed in NPC tissues and was known to promote growth, survival and progression of NPC [19, 22, 23]. In this study, we examined if berberine may suppress STAT3 activation in NPC grown as xenografted tumor in nude mice. The C666-1 NPC cells grown as subcutaneous tumor harvested from mice in control and treatment groups were embedded in paraffin, sectioned, and examined by immunohistochemical staining. Activated STAT3 will undergo phosphorylation and translocate to cell nucleus. Using antibody specific for phosphorylated STAT3 (p-STAT3), activated STAT3 could be demonstrated as brown nuclei stain in the tumor cells (Figure 2). We have also performed staining using antibody against cytokeratin to identify the C666-1 cells (epithelial in origin) from the stromal cells. For each section of tumors harvested from control and treatment groups, 5 random microscopic fields which contain over 95% of C666-1 cells were chosen for examination. The percentages of STAT3-activated C666-1 cells were counted. The tumor sections in the control group revealed 75% ± 12.6 of positive staining of p-STAT3 while only around 12.4 ± 8.6 and 5.4 ± 4.5% of cells were stained positive for p-STAT3 in the treatment groups injected with 5 mg/kg and 10 mg/kg of berberine respectively (p <0.05). Representative images showing the effective downregulation of p-STAT3 in the xenografted tumors treated with 5 mg/kg and 10 mg/kg of berberine are shown (Figure 2).

We then examined if berberine could suppress STAT3 in NPC cell grown in vitro. We first examined STAT3 activation in C666-1. Interestingly, STAT3 activation was not prominent in C666-1 grown in culture despite of its strong activation when grown in vivo. We postulated that STAT3 activation in C666-1 grown as xenograft may be induced by inflammatory signals from the host stroma. We have also investigated the effects of berberine on HONE1 which is an NPC cell line with high level of constitutive activation of STAT3 (Figure 3). High expression of p-STAT3 was detected in HONE1 cells. We observed that berberine could effectively suppress the level of p-STAT3 in HONE1 cells (Figure 3). Furthermore, the downregulation of STAT3 activation was associated with the suppressed expression of Mcl-1 (a downstream survival protein of STAT3) and an increased level of cleaved PARP-1 (an apoptotic marker) (Figure 3). This indicates that berberine could suppress constitutive activation of STAT3 in NPC cells and inhibited their survival ability by activating apoptosis.

Next, we examined if activation of STAT3 by extracellular stimulus, such as IL-6, could also be suppressed by berberine. IL-6 is one of the major inflammatory cytokine present in NPC tissues. It has been reported that there is a positive feed-back loop of IL-6/STAT3 signaling in NPC to potentiate STAT3 signaling which enhances malignant properties of NPC [30]. In this study, IL-6 (20 ng/ml) was added to two NPC cell lines (C666-1 and HK1), which have low basal level of STAT3 activation, for 0, 4 and 6 hr in the presence or absence of 50 μM of berberine. Figure 4 shows that expression of p-STAT3 in both cell lines could be induced by IL-6 in a time-dependent manner. Berberine could effectively suppress the IL-6-induced p-STAT3 in both NPC cell lines (Figure 4).

We have previously demonstrated that berberine could potently suppress the IL-6-induced or constitutive activated STAT3 in NPC cells. We speculated that berberine may be able to target cancer cells which are dependent on STAT3 activation for growth and survival. We performed MTT assays to assess the cytotoxicity of berberine on three NPC cell lines (HK1, C666-1 and HONE1) and also one normal immortalized nasopharyngeal epithelial cell line (NP460) . Interestingly, berberine had high toxicity towards the HONE1 cell line, which has constitutively activated STAT3 (Figure 5). The IC50 after 24 hr berberine treatment for HONE1 was around 100 μM, while the IC50 after similar treatment for HK1, C666-1 and NP460 was around 400 μM (Figure 5).

In vivo, fibroblasts in tumor tissue have been postulated to serve as a source of pro-oncogenic cytokines, including IL-6 [31]. Here, we used fibroblasts isolated from primary human NPC tumor biopsy to develop an in vitro system to mimic the in vivo secretion of IL-6 by stromal cells and the subsequent activation of STAT3 in cancer cells. The primary fibroblasts derived from NPC tissue were cultured in serum free medium for 4 days and culture supernatants from day 1 to day 4 were collected for investigation. The concentrations of IL-6 in the culture supernatants were measured by ELISA, and was found to progressively increase to a concentration of around 1,500 pg/ml at day 4 (Figure 6a). This concentration of IL-6 was sufficient to stimulate STAT3 activation in C666-1 cells. STAT3 activation as indicated by the phosphorylation of STAT3 was induced at a time-dependent manner in C666-1 cells after incubation with fibroblast-conditioned medium (Figure 6b). The STAT3 activation in the NPC cells by fibroblast-conditioned medium could be suppressed by treating the cells with 50 μM berberine (Figure 6c). To further confirm that IL-6 in the fibroblast-conditioned medium was involved in the activation of STAT3 in C666-1 cells, anti-IL-6 and anti-IL-6-receptor antibodies were added to the fibroblast-conditioned medium. Both antibodies could effectively inhibit the activation of STAT3 (Figure 6d). It could be deduced that IL-6 secreted from fibroblasts play a key role in the activation of STAT3 in C666-1 cells and activation of IL-6-receptor on NPC cells was involved. Furthermore, IL-6 treatment also increased the growth rate of C666-1 cells (Figure 6e).