We thank Professor James Will for assistance with the English-language presentation of the article.
It is well known that 1α,25-Dihydroxyvitamin D3(1,25[OH]2D3) restrains cell proliferation and induces differentiation and apoptosis in normal and tumor cells. The authors of this report recently demonstrated that 1,25(OH)2D3 effectively inhibits the proliferation of cholangiocarcinoma (CCA) cell lines. The antitumor activity and the underlying mechanism of 22-oxa-D3, an analog of vitamin D, in mice and in tissue cultures from patients with CCA were further explored in the current study.
Cell growth and cell cycle distribution were examined in CCA cells by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay and flow cytometry. Mice were injected subcutaneously with 4 × 106 CCA cells at both flank sides and intraperitoneal injections with phosphate-buffered saline or 22-oxa-D3(15 μg/kg/day) for 17 days thereafter. Tumors were removed the next day. The expression levels of cyclin D1 and the cyclin-dependent kinase inhibitor p21 were determined by Western blot analysis and immunohistochemistry. Growth inhibition of 22-oxa-D3 in fresh tissue samples from patients with CCA was analyzed by using a histodrug response assay.
22-Oxa-D3 effectively suppressed the growth of CCA cell lines in a time-dependent and dose-dependent manner. 22-Oxa-D3 arrested CCA cells at G1 phase to S phase by the suppression of cyclin D1 expression and the up-regulation of p21. Supplementation of 22-oxa-D3 to CCA-inoculated mice significantly inhibited tumor growth without hypercalcemia or serious side effects. The treatment also induced cellular apoptosis in tissue samples from patients with CCA.
Cholangiocarcinoma (CCA), an aggressive and lethal cancer, arises from biliary epithelium within either the intrahepatic or the extrahepatic biliary tract. The global incidence of CCA is low compared with that of hepatoma, although the rates of mortality and incidence of CCA have been noted increasingly worldwide.1 Because CCA is difficult to diagnose at an early stage, almost all patients with CCA present with advanced, incurable disease. Even in patients who have undergone complete surgical resection, the recurrence rate remains quite high; consequently, the low 5-year survival rate ranges from 0% to 40%.2, 3 Therefore, novel treatment strategies directed against this malignancy constitute an urgent need.
Apart from the regulation of calcium homeostasis, it has been demonstrated that 1α,25-dihydroxyvitamin D3(1,25[OH]2D3), the active form of vitamin D, controls the differentiation and proliferation of many human cancer cells, including cancers of the breast,4 prostate,5 liver,6 and colon.7 Several clinical trials using 1,25(OH)2D3 either as a single agent or in combination with chemotherapeutic drugs have indicated the effectiveness of vitamin D in controlling tumor growth.8, 9 The clinical application of this compound is limited, however, because of its hypercalcemia-inducing activity.10 For this reason, various synthetic vitamin D3 compounds with reduced calcemic activity have been developed. A vitamin D analog, 22-oxa-1α,25-dihydroxyvitamin D3 (22-oxa-D3), has reduced calcemic activity but strong action for controlling cell proliferation and differentiation.11, 12
In a previous report on up-regulation of the vitamin D receptor (VDR) in tumor bile duct epithelium from patients with CCA, we reported that supplementation with 1,25(OH)2D3 significantly reduced the proliferation of CCA cells in a dose-dependent manner.13 Moreover, the action of 1,25(OH)2D3 depended on the basal level of cellular VDR. In the current study, we demonstrated further that 22-oxa-D3 can significantly suppress the growth of CCA cells in inoculated mice without a hypercalcemic effect. The effectiveness of 22-oxa-D3 in controlling the growth of primary tissue cultures from patients with CCA also was demonstrated, and we explored the molecular mechanism by which 22-oxa-D3 controlled the growth of CCA cells.
MATERIALS AND METHODS
Two CCA cell lines, KKU-M213 and KKU-M214, were established as described previously.14 Cells were cultured in Dulbecco modified Eagle medium supplemented with 10% heat inactivated fetal calf serum (FCS), 1% L-glutamine, and 1% penicillin-streptomycin at 37°C in a 5% CO2 atmosphere.
Cell Cycle Analysis
CCA cells (50,000 cells per well) were seeded in a 6-well plate and treated with either 1 μM 22-oxa-D3 or vehicle (0.05% ethanol) as a control for 3 days. Cells were scraped into culture medium, washed once with phosphate-buffered saline (PBS), and fixed with ice-cold 70% ethanol overnight at 4°C. After centrifuging for 5 minutes at 1500 revolutions per minute, the cells were resuspended in 5 mL fluorescence-activated cell sorting washing medium (3% FCS/PBS) and centrifuged as described above. Cell pellets were resuspended in 1 mL of 1 μg/mL propidium iodine and incubated at room temperature for 1 hour. DNA content in each cell was analyzed on an LSR II flow cytometer (BD Bioscience, San Jose, Calif). Data were analyzed using FlowJo software (Tree Star, San Carlos, Calif).
3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays were applied to test cell viability. Briefly, cells were seeded at a density of 3 × 103 cells per well in a 96-well dish. The cells were incubated either with 0.05% ethanol or with different concentrations of 22-oxa-D3(0.25 μM, 0.5 μM, 1 μM, and 2 μM) at 37°C in a 5% CO2 atmosphere for 1 day, 2 days, and 3 days. Subsequently, 10 μL MTT (0.5 mg/mL final concentration) were added to each well. After 4 hours of additional incubation, 100 μL of 0.01 N HCl in isopropanol were added to dissolve the crystals. The absorption values at 570 nm were determined with an automatic enzyme-linked immunosorbent assay plate reader (Tecan Austria GmbH, Salzburg, Austria).
In Vivo Assay
NOD/Scid/Jak3 deficient (NOJ) mice were established by backcrossing Janus kinase 3 (Jak3)-deficient mice with the severe combined immunodeficiency (SCID) NOD.Cg.-Prkdcscid strain for 10 generations.15 NOJ male mice (ages 8-10 weeks) were housed and monitored in the animal research facility according to institutional guidelines. All experimental procedures and protocols were approved by the Institutional Animal Care and Use Committee at Kumamoto University. Mice were injected subcutaneously with 4 × 106 KKU-M213 cells or KKU-M214 cells in the flank on both sides. Then, the mice (n = 5 mice per group) received intraperitoneal injections of either PBS or 22-oxa-D3(15 μg/kg per day) for 17 days thereafter. Tumors were removed 18 days after inoculation.
Serum Calcium Determination
To determine the effect of active vitamin D compounds on serum calcium, blood samples from CCA-inoculated mice that were treated with either placebo or 22-oxa-D3 were collected at the end of treatment. Serum calcium levels were measured using a colorimetric method (Aqua-Auto Kainos Ca; Kainos Laboratories, Tokyo, Japan)16 according to the manufacture's instruction.
Western Blot Analysis
CCA cells were lysed with NP-40 lysis buffer (50 mM Tris, pH 7.4; 150 mM NaCl; 1% NP-40), and the whole cell lysate was separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis.17 The electrophoresed proteins were blotted onto polyvinylidine fluoride membranes (GE Healthcare, Tokyo, Japan) and were detected with a specific primary antibody using the enhanced chemiluminescence (ECL) Western Blotting Detection System (GE Healthcare Bio-Science, Buckinghamshire, United Kingdom). The primary antibodies used were as follows: anti-VDR (1:500 dilution; C-20), anticyclin D1 (1:250 dilution; H-295; both from Santa Cruz Biotechnology, Santa Cruz, Calif), p21 (1:250 dilution; EA10; Zymed Laboratories, South San Francisco, Calif), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (1:5000 dilution; MoAb374; Chemicon, Temecula, Calif). Quantification of the Western blots was performed using GelPro 32 (Media Cybernetics, Bethesda, Md). Relative density was evaluated and normalized with GAPDH.
Expression of cell cycle-related genes was detected on formalin-fixed, paraffin-embedded sections from CCA-inoculated mouse tissues according to the standard immunohistochemistry technique. Briefly, deparaffinized tissue sections were blocked with 3% H2O2 and nonimmune horse serum for 30 minutes and then heated in 10 mM citrate buffer, pH 6.0, in a boiling pressure chamber for 4 minutes. The slides were reacted at room temperature overnight with either 1:200 anti-VDR (C-20), 1:50 anticyclin D1 (H-295), 1:100 anti-p21 (EA10), or 1:200 antiproliferating nuclear antigen (anti-PCNA) (PC-10; Chemicon) and 1:2000 peroxidase at room temperature for 1 hour. The peroxidase activity was observed using diaminobenzidine tetrahydroxychloride solution (DAB; Dako, Glostrup, Denmark) as the substrate. The sections were counterstained with hematoxylin.
Histoculture drug response assay
Liver tissues from patients with intrahepatic CCA (n = 10) were obtained after surgery. The research protocols were approved by the Human Research Ethics Committee, Khon Kaen University (HE471214), and informed consent was obtained from each patient before surgery.
The histoculture drug response assay (HDRA) was used as an in vitro drug-sensitivity test. Collagen sponge gels manufactured from pig skin were purchased from Sumitomo Pharma (Osaka, Japan). Cancerous portions of specimens were minced into pieces, and approximately 10 mg were placed on the prepared collagen sponge gels in a 24-well microplate with 1 μM or 2 μM 22-oxa-D3 dissolved in RPMI-1640 medium containing 20% FCS. Tissues on sponge gels that were incubated in culture medium alone served as controls. Plates were incubated at 37°C in a 5% CO2 atmosphere for 4 days. Thereafter, tumor tissues were fixed in buffered formalin and processed into paraffin-embedded tissues according to standard protocol.
Histologic analysis of DNA fragmentation was used to identify apoptotic cells in the paraffin sections of the HDRA. In situ terminal deoxynucleotide transferase-mediated 2′-deoxyuridine, 5′-triphosphate nick-end labeling (TUNEL) was done using the DeadEnd Colorimetric TUNEL System (Promega, Madison, Wis). TUNEL-positive cells were quantified in at least 4 high-power fields (at ×40 magnification) from each randomly selected tissue section. The protocol was approved by the Human Research Ethics Committee, Khon Kaen University (HE471214), and informed consent was obtained from each participant.
The results of multiple observations are presented as the means ± standard deviation of at least 3 separate experiments. Statistical significance was determined with the Student t test. P values < .05 were considered significant.
22-Oxa-D3 Suppressed the Growth of CCA Cells by Induced Cell Cycle Arrest at G1 to S Phase
The antitumor activity of 22-oxa-D3 in 2 CCA cell lines (KKU-M213 and KKU-M214) was investigated. CCA cells were treated with various concentrations of 22-oxa-D3(0.25 μM, 0.5 μM, 1 μM, and 2 μM) or with 0.05% ethanol a as control. The percentage of viable cells, as determined by the MTT assay, indicated that supplementation with 22-oxa-D3 significantly decreased the growth of both CCA cell lines in a dose-dependent and time-dependent manner (P < .001) (Fig. 1).
22-Oxa-D3 Arrested Cells at G1 Phase
Next, we investigated the effect of 22-oxa-D3 on cell cycle distribution by flow cytometry with propidium iodine staining. KKU-M213 and KKU-M214 cells that were treated with 1 μM 22-oxa-D3 caused an accumulation of cells in G1 phase compared with controls that were treated with vehicle alone (Fig. 2A). Approximately 50% to 56% of control cells were accumulated at G1 phase, whereas accumulation increased to 67% to 71% after the cells were incubated with 22-oxa-D3 for 72 hours (Fig. 2B).
Further examination of the expression levels of candidate genes that mediate the transition from G1 phase to S phase of the cell cycle was undertaken. Expression levels of the cell cycle-regulated genes cyclin D1 and cyclin-dependent kinase inhibitor p21 as well as VDR were analyzed by Western blot analysis. The addition of 1 μM 22-oxa-D3 significantly suppressed cyclin D1 expression, whereas treatment induced the up-regulation of VDR and p21, the cyclin-dependent kinase that inhibits progression of the cell cycle from G1 phase to S phase (Fig. 2C).
22-Oxa-D3 Inhibits the Growth of CCA-Inoculated Mice Without Hypercalcemia
To investigate whether the growth-suppressing activity of 22-oxa-D3 could be demonstrated in vivo, the CCA cell lines KKU-M213 and KKU-M214 were implanted subcutaneously into NOJ knockout mice, and 22-oxa-D3(15 μg/kg/day) was injected intraperitoneally into a group of mice every day for 17 days. For comparison, CCA-inoculated mice supplemented with PBS were used as controls. The efficient intervention of growth inhibition by 22-oxa-D3 on primary tumors was obviously in the mice. Figure 3A,B demonstrates clearly that the tumor sizes and tumor weights from 22-oxa-D3–treated mice were reduced significantly compared with those in the placebo-treated mice (P < .01). Mice that received 22-oxa-D3 did not have signs of morbidity during treatment. Their body weight (Fig. 3C) and serum calcium levels (Fig. 3D) were not affected by the treatment with 22-oxa-D3.
In this study, we also investigated the mechanism by which 22-oxa-D3 exerts antitumor activity. The expression of cell cycle-related genes, p21, cyclin D1, PCNA and VDR in tumor tissues from CCA-inoculated mice was determined by immunohistochemistry (Fig. 4A) and Western blot analysis (Fig. 4B). Figure 4 reveals that comparing the results from immunohistochemistry and Western blot analyses of tumor tissues from the treated mice with the results from placebo mice, supplementation with 22-oxa-D3, significantly reduced the expression of cyclin D1 and up-regulated the expression of p21. Consequently, the expression of PCNA, which is used as a proliferative index, was reduced in tumors from the treated mice. Increased expression of VDR in the treated group also was noted.
22-Oxa-D3 Stimulated Apoptosis and Necrosis in CCA Patient Tissues
The HDRA is representative of an in vitro drug-response assay for anticancer agents. Several clinical studies have revealed that inhibition rates obtained from HDRA can predict the clinical responses to chemotherapy of corresponding patients. In the current study, HDRA was performed on 3 tissue samples from patients with CCA. Figure 5A reveals that culturing tumor tissues in the presence of 1 μM or 2 μM 22-oxa-D3 for 4 days significantly induced cell death, as determined by TUNEL assay. The percentages of apoptotic cells observed in control tissues were 21.76% ± 16% compared with 34.72% ± 21% and 40.27% ± 19%, respectively, in tissues that were incubated with 1 μM and 2 μM 22-oxa-D3. Increased apoptosis of CCA tissues that were treated with 22-oxa-D3 was significant (P < .05) (Fig. 5B). The results indicate that 22-oxa-D3 has an antitumor effect in human CCA tissues.
In the past 2 decades, an increasing body of evidence has demonstrated 1,25(OH)2D3 effectively inhibits the proliferation of a variety of malignant cells.18 In our previous report, we observed that the overexpression of VDR in CCA tissues was a good prognostic indication.13 The addition of 1,25(OH)2D3 to CCA cell cultures effectively inhibited the growth of CCA cell lines in a dose-dependent manner. In the current study, we demonstrated further that supplementation with 22-oxa-D3, an analog of 1,25(OH)2D3, suppressed the growth of CCA cell lines in culture, in CCA-inoculated mice, and in histocultures of patient tissues. This preclinical study indicates the potential of using 22-oxa-D3 in a clinical trial of patients with CCA.
A pitfall of using 1,25(OH)2D3 is its hypercalcemia-inducing activity, which limits its application in clinical practice. To provide alternatives, various synthetic vitamin D3 compounds with reduced calcemic activity have been developed. An analog of vitamin D3, 22-oxa-D3, has been synthesized to reduce hypercalcemia-inducing activity and to enhance differentiation-inducing and antiproliferative effects. In the current study, we used 22-oxa-D3 to clarify the growth-inhibitory effects of vitamin D on CCA, and 22-oxa-D3 significantly suppressed the growth of 2 CCA cell lines in a dose-dependent and time-dependent manner. We demonstrated that the mechanism by which 22-oxa-D3 inhibited the growth of CCA cell lines was the induction of cell cycle arrest at G1 to S phase through up-regulation of p21 and suppression of cyclin D1. This observation is in accordance with previous reports in cancers of the breast,11, 19, 20 pancreas,12, 21 and thyroid.22
In our current study, the suppression of tumor growth by 22-oxa-D3 was demonstrated clearly in CCA-inoculated mice by decreased tumor weights and sizes without creating significant hypercalcemic effects or other serious side effects. The results from our immunohistochemical study and Western blot analyses of the tumor tissues obtained from these mice agreed well with the in vitro results in CCA cell lines, indicating that 22-oxa-D3 enhanced the expression of the cyclin-dependent kinase inhibitor p21WAF and suppressed the expression of cyclin D1. These modifications may reduce the proliferation rate of tumor cells, as indicated by the reduction of PCNA-positive cells in tumor tissues from 22-oxa-D3-treated mice. These results are supported by several studies indicating that 1,25(OH)2D3 and its analogs can inhibit tumor growth. The antitumor activity of 22-oxa-D3 was more obvious in our mouse model than in cell cultures. The enhancement of 22-oxa-D3 activity and its role in suppressing tumor growth in vivo can be explained by the finding that, apart from inducing cell cycle arrest as a result of enhanced expression of the cyclin-dependent kinase inhibitors p21WAF1 and p27Kip1,23, 24, a variety of mechanisms, including the regulation of angiogenesis,25 increased apoptosis,26 and reduced tumor invasiveness,25, 27 reportedly responsible for the actions of 1,25(OH)2D3. These latter 3 mechanisms could not be demonstrated in our CCA cell culture studies. The multiple actions of 22-oxa-D3 in vivo, however, could be demonstrated in the mouse model and may be responsible for the impressive suppression of tumor growth that we demonstrated in our mice. Further studies may reveal more significant applications for synthetic 22-oxa-D3 compounds in the clinical management of patients with CCA.
The HDRA is a 3-dimensional, native-state histoculture assay that simulates the structure of the tumor in the body. It is believed that the histoculture method has advantages over other methods that use single-cell suspensions and that the high value of HDRA results in its good predictability for clinical responses. Several clinical studies that included cancers of the gastrointestinal tract,28 ovary,29 and breast30 as well as head and neck squamous cell carcinoma31 revealed that inhibition rates obtained from HDRA predicted clinical responses to chemotherapy with excellent results. In the current study, we investigated the effectiveness of 22-oxa-D3 in surgically resected specimens from patients with CCA. The antitumor activity of 22-oxa-D3 in those specimens was confirmed by HDRA. Culturing CCA tissues in the presence of 22-oxa-D3 for 4 days enhanced apoptosis approximately 2-fold compared with that in controls. The effect of 22-oxa-D3 on apoptosis was evident in CCA tissues that were cultured in the presence of 22-oxa-D3 but not in cell cultures. This discrepancy may be explained in part by the finding that 22-oxa-D3 also may affect tumor stromal cells surrounding the tumor tissues. It is well accepted that stromal cells have a supporting role in tumor growth.32
The effect of vitamin D on apoptosis has been reported in many tumor models, including breast cancer,33, 34 colon cancer,35 and prostate cancer.36, 37 Vitamin D and its analogs induce cell apoptosis by regulating the expression of apoptosis protein, Bcl family, p53, and telomerase. Decreased expression of the antiapoptotic protein Bcl-2, increased expression of proapoptotic protein Bax, and release of cytochrome c from the mitochondria may underlie the action of vitamin D.37, 38 A key regulator of apoptotic cell death, p53, also is up-regulated after vitamin D treatment.39, 40 In addition, vitamin D inhibits the expression of telomerase, an enzyme that adds a specific DNA sequence repeat to the 3′ end of DNA strands, resulting in cell death.41, 42
In the current investigation, we noted that supplementation with 22-oxa-D3 also up-regulated the expression of VDR, a receptor that is necessary for the genomic action of vitamin D. This observation is was evident in cell cultures, in the mouse model, and in the HDRA. It also has been demonstrated that the antitumor growth activity of 1,25(OH)2D3 is depends on VDR, and CCA cells with high VDR expression are more sensitive to 1,25(OH)2D3 treatment.13 Therefore, increasing VDR expression as a result of 1,25(OH)2D3 treatment should be beneficial in patients with CCA.
In summary, the current results provide strong encouragement for further investigation of 1,25(OH)2D3 or its analogue, 22-oxa-D3, in the treatment of patients with CCA. The only disadvantage of using 1,25(OH)2D3 in clinical practice is its hypercalcemic effect, which can be minimized by using its analogues, such as 22-oxa-D3. 1,25(OH)2D3 has been used either alone or in combination with anticancer drugs in many trials. A phase 2 clinical trial of 1,25(OH)2D3 in combination with conventional chemotherapeutic drugs has been initiated at the Srinagarindh Hospital, Faculty of Medicine, Khon Kaen University, Thailand.
CONFLICT OF INTEREST DISCLOSURES
Supported by a Research Team Strengthening Grant from the National Genetic Engineering and Biotechnology Center, National Science and Technology Development Agency, Thailand, by the Royal Golden Jubilee-PhD Program for Drs. Seubwai and Wongkham (PHD/0034 of 2549), and was supported in part by Grants-in-Aid for Science Research (Nos. 21107522, 21591209) from the Ministry of Education, Science, Sports, and Culture (MEXT) of Japan.