Growth inhibition of head and neck squamous cell carcinoma cells in vitro with 7β-hydroxycholesterol (7β-HC). Cell-Counting Kit 8 (Dojindo Molecular Technologies, Gaithersburg, Maryland) assays were performed on SCC9 (A), SCC25 (B), CAL27 (C), and FaDu (D) cells after 72 hours of treatment with increasing doses of 7β-HC dissolved in dimethyl sulfoxide (DMSO). Cells treated with DMSO served as control. Treatment with 7β-HC resulted in significantly increased rates of cell death compared with control (P < .001). There was near-complete cell death with 25-μmol/L 7β-HC, and 40-μmol/L 7β-HC was required to achieve complete cell death in FaDu cells. Error bars indicate standard errors of the mean.
Head and neck squamous cell carcinoma cell lines SCC9, SCC25, CAL27, and FaDu were treated with 10-, 15-, and 20-μmol/L 7β-hydroxycholesterol (7β-HC) for 24, 48, and 72 hours. The cells were labeled with probidium iodide and Anti-Annexin V antibody (BenderMedSystems, Vienna, Austria) and then analyzed with flow cytometry. Graphs show an increase in apoptotic and necrotic cell death in a dose- and time-dependent manner. Error bars indicate standard errors of the mean.
Cultured head and neck squamous cell carcinoma (HNSCC) cells. Cultured HNSCC cells (SCC9, A and E; SCC25, B and F; CAL27, C and G; and FaDu, D and H) exposed to the M30 monoclonal antibody after treatment with 15-μmol/L 7β-hydroxycholesterol (E-H). Cells treated with dimethyl sulfoxide served as control (A-D). Cells undergoing apoptosis show cytoplasmatic staining for M30. Clearly, an increase in apoptosis is visible after treatment (original magnification, ×200).
Cyclooxygenase (COX)-1 (left panel) and COX-2 (right panel) expression before and after treatment with 7β-hydroxycholesterol (7β-HC) in SCC9, SCC25, CAL27, and FaDu cell lines. COX expression before treatment is visible in the leftmost lanes (0 hours).Treatment with 7β-HC results in a time-dependent down-regulation of COX-2 in all cell lines (48 hours). A brief up-regulation of COX-2 was detected in SCC25 after 12 hours; COX-1 expression was not influenced by 7β-HC or by tubulin, which served as control.
Heiduschka G, Erovic BM, Vormittag L, Skoda C, Martinek H, Brunner M, Ehrenberger K, Thurnher D. 7β-Hydroxycholesterol Induces Apoptosis and Regulates Cyclooxygenase 2 in Head and Neck Squamous Cell Carcinoma. Arch Otolaryngol Head Neck Surg. 2009;135(3):261-267. doi:10.1001/archoto.2008.558
To determine whether treatment with 7β-hydroxycholesterol (7β-HC) would trigger cell death in head and neck squamous cell carcinoma (HNSCC) cell lines in a dose-dependent fashion.
In vitro study.
The study included HNSCC cell lines SCC9, SCC25, CAL27, and FaDu.
We treated HNSCC cell lines with increasing doses of 7β-HC. Proliferation assays were performed to assess cell viability after treatment. Western blots were carried out to evaluate cyclooxygenase (COX)-1 and -2 expression levels.
Using proliferation assays and immunocytochemical analysis, we detected significant growth inhibition via apoptosis in 4 different HNSCC cell lines after treatment with 7β-HC (P < .001). The 50% inhibitory concentration levels were between 13.19 and 20.79 μmol/L after 72 hours. Western analysis indicated that COX-2, but not COX-1, levels were suppressed after treatment.
Treatment with 7β-HC resulted in suppression of HNSCC growth in vitro. Our data warrant further investigations for the potential use of 7β-HC as a cytotoxic agent in head and neck cancer.
Each year, more than 600 000 new cases of head and neck squamous cell carcinoma (HNSCC) are reported worldwide.1 Current treatment protocols for advanced head and neck cancer involve surgical, chemotherapeutic, and/or radiotherapeutic procedures that ultimately result in high morbidity rates.2 Indeed, the 5-year-survival rate of patients with advanced head and neck cancer is less than 50%, a number that has improved only marginally over the past 3 decades,3 and it is apparent that a different approach to treatment is needed. Therefore, identification of novel antineoplastic agents for head and neck tumors that boost the ability of tumor cells to undergo apoptosis is continuing.
Oxysterols form a large family of oxygenated derivatives of sterols. Studies have focused on their biologic properties as these compounds are found in the human diet.4,5 Recent studies showed that some of the oxysterols lead to oxygenation of low-density lipoprotein and thus contribute to the development of atherosclerosis6,7 via apoptotic pathways. Certain oxysterols have been shown to induce apoptosis via generation of oxidative stress8- 10; loss of transmembrane potential11,12; release of cytochrome c13; activation of caspase-9, leading to subsequent activation of caspase-36,14; polyadenosine diphosphate–ribose polymerase degradation15; and DNA fragmentation.16 Interestingly, this class of compounds is also able to induce apoptosis in malignant cell lines. In vitro treatment of several human cancer cell lines has led to growth inhibition and induction of apoptosis.17- 19
The compound used in this study, 7β-hydroxycholesterol (7β-HC), which can be isolated from the Chinese traditional anticancer drug Bombyx cum Botryte,20 is one of the main substances in the group of oxysterols. The objective of this study was to determine whether 7β-HC could qualify as a possible anticancer drug in head and neck cancer. We found that 7β-HC induces apoptosis in HNSCC cell lines.
Cyclooxygenase (COX) is a key regulatory enzyme in prostaglandin biosynthesis. So far, 2 human isoforms with COX activity have been identified. While COX-1 is apparently expressed constitutively in all healthy tissues, COX-2 expression is almost undetectable under normal conditions in most tissues. However, COX-2 can strongly be induced by proinflammatory agents, oncogenes, and carcinogens. Overexpression of COX-2 is described in cancers of the colon,21,22 pancreas,23 and head and neck.24,25 Cyclooxygenase-2 is related to carcinogenesis via angiogenesis, increasing invasiveness, and inhibition of apoptosis.26 To our knowledge, an influence of 7β-HC on COX expression in HNSCC cell lines or any other cancer cell line has not yet been described.
Cell lines SCC9, SCC25, CAL27, and FaDu were obtained from the American Type Culture Collection (ATCC, Manassas, Virginia) or from the German Collection of Microorganisms and Cell Cultures (DSMZ, Braunschweig, Germany); RPMI medium was obtained from Cambrex Corp (Walkersville, Maryland); fetal bovine serum, from PAA Laboratories (Linz, Austria); penicillin-streptomycin, from Gibco BRL (Gaithersburg, Maryland); Cell-Counting Kit 8, a cell-counting system, from Dojindo Molecular Technologies (Gaithersburg); and 7β-HC, from Sigma-Aldrich Corp (St Louis, Missouri) at a purity of more than 95%.
The cancer cell lines SCC9, SCC25, CAL27, and FaDu were grown as monolayers and maintained in RPMI medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin. The cells were maintained in a humidified atmosphere enriched with 5% carbon dioxide at 37°C; 3×103 cells per well were seeded into 96-well plates and allowed to grow for 24 hours. The initial stock solution was 100-μmol/L 7β-HC dissolved in dimethyl sulfoxide (DMSO) and diluted in RPMI to final concentrations ranging from 5 to 40 μmol/L. The final DMSO concentrations ranged from 0.04% for 10-, 20-, and 40-μmol/L 7β-HC to 0.06% for 15- and 30-μmol/L 7β-HC. The 7β-HC-doses were chosen because treatment of 5 μmol/L for 72 hours resulted in minimal effect on the cells, whereas treatment of 40 μmol/L for 72 hours resulted in almost total cell death. Control cells were incubated to corresponding DMSO doses. The cells were incubated in a drug-containing medium for 24, 48, or 72 hours at 37°C, and cell viability was measured using the Cell-Counting Kit 8 cell proliferation assay system according to the manufacturer's protocol.
For visualization of apoptosis, we used a mouse monoclonal antibody raised against M30,27 and 106 cells were grown on glass slides and treated with 15-μmol/L 7β-HC for 72 hours. This concentration was used because it caused approximately 50% cell death in our cell lines. The DMSO-treated cells served as control. The slides were fixed by ice-cold methanol (−20°C) for 30 minutes, washed twice in phosphate-buffered saline, and blocked with 1% bovine serum albumin in Tris-buffered saline for 30 minutes. They were then incubated overnight with M30 CytoDeath antibody (1:200, Roche Applied Science, Vienna, Austria). As a control, slides were exposed to IgG1 antibody (Ancell, Bayport, Minnesota). After being washed with Tris-buffered saline 3 times, they were incubated with a multilink antibody (Dako Corp, Glostrup, Denmark) for 1 hour at room temperature, washed, and again exposed to alkaline phosphatase–conjugated streptavidin with 10% human serum (Dako Corp) for 1 hour at room temperature. Visualization was performed with 4-chloro-2-methylbenzenediazonium salt (Fast Red TR; Sigma-Aldrich Corp), counterstained with Hemalaun (Sigma-Aldrich Corp), dehydrated, and mounted.
The SCC9, SCC25, CAL27, and FaDu cells were seeded in 6-well plates (100 000 cells per well). After 24 hours, 7β-HC (10, 15, and 20 μmol/L), cisplatin as positive control (10 μmol/L), or medium as negative control was added. Apoptosis was measured after 24, 48, and 72 hours using a commercially available apoptosis detection kit (Anti-Annexin-V antibody; BenderMedSystems, Vienna, Austria).
An SCC9, SCC25, CAL27, and FaDu cell monolayer (105 cells in a 6-cm tissue culture dish) was rapidly rinsed twice with ice-cold phosphate-buffered saline and lysed in 0.5 mL of ice-cold lysis buffer. The lysis buffer contained 1% phenol-containing detergent (NP-40), 0.1% sodium dodecyl sulfate, 150-mmol/L sodium chloride, 50-mmol/L Tris at a pH of 7.4, 10-mmol/L edetic acid, 10-mmol/L p-nitrophenolphosphate, 250-U/L aprotinin, 40-μg/mL leupeptin, 1-mmol/L phenylmethylsulfonyl fluoride, 1-mmol/L sodium orthovanadate, 10-mmol/L sodium fluoride, and 40-mmol/L β-glycerolphosphate. The lysates were centrifuged at 14 000 rpm at 4°C for 20 minutes, and the supernatants were collected. Aliquots of supernatants containing 20 μg of protein were subjected to sodium dodecyl sulfate–polyacrylamide gel in 10% gels and transferred to nitrocellulose membranes. After blocking nonspecific binding by incubation with 5% bovine serum albumin in Tris-buffered saline (Bio-Rad Laboratories, San Diego, California), the membranes were incubated with anti–COX-1 or anti–COX-2 (Santa Cruz Biotechnologies, Santa Cruz, California) as primary antibodies. Bound antigen was visualized with the ECL Western blotting analysis system (Amersham Life Sciences, Buckinghamshire, England). The density of bands was measured by Quantity One Basic Software (Bio-Rad Laboratories).
The data were analyzed by both 2-way analysis of variance and 2-sided t tests (SPSS version 11.5; SSPS Inc, Chicago, Illinois). P values of less than .05 were considered statistically significant.
Four HNSCC cell lines (SCC9, SCC25, CAL27, and FaDu) were tested for growth inhibition with varying concentrations of 7β-HC. Dimethyl sulfoxide as the solvent for the compound was used as control. For each cell line, at least 3 independent experiments were carried out in triplicate using 96-well plates. Treatment of SCC9, SCC25, CAL27, and FaDu cells with 7β-HC resulted in a dose-dependent growth inhibition compared with treatment with the DMSO control.
There was a dose-dependent cell death with a maximal killing at 40-μmol/L 7β-HC in all cell lines tested. After 24, 48, and 72 hours, 40-μmol/L 7β-HC led to cell death ranging from 74.24% to 84.16%, 86.92% to 93.13%, and 87.00% to 97.50%, respectively. Curves after 72 hours of treatment are shown in Figure 1 (data for 24 and 48 hours not shown). The 50% inhibitory concentration (IC50) values ranged from 15.02- to 18.31-μmol/L at 24 hours, from 13.54- to 23.93-μmol/L at 48 hours, and from 13.19- to 20.79-μmol/L at 72 hours (Table). Statistical analysis showed that the correlation between cell death and increasing concentrations of 7β-HC was highly significant (P < .001) in treated cell lines compared with control.
Cell death was analyzed with flow cytometry after incubation of SCC9, SCC25, CAL27, and FaDu with 10-, 15-, and 20-μmol/L 7β-HC for 24, 48, and 72 hours (Figure 2). Experiments were conducted in triplicate. There was a dose- and time-dependent increase in apoptosis as well as necrosis. The highest amount of apoptotic cells was measured in CAL27, which is described by the American Type Culture Collection as a multidrug-resistant (vindesine, cisplatin, and actinomycin) cell line. The lowest amount of apoptotic cells was measured in SCC9. Furthermore, apoptotic cells were visualized using the M30 antibody, which is specific for the caspase-cleaved cytokeratin 18 fragment (Figure 3). Experiments were carried out in all 4 HNSCC cell lines. The DMSO-treated cells were used as control and showed almost no apoptosis (Figure 3 A-D). The 7β-HC–treated cells showed a higher rate of apoptosis (Figure 3 E-H).
Western blot analysis of COX-1 and COX-2 expression was performed in 4 HNSCC cell lines before and after 12, 24, and 48 hours of treatment (Figure 4). The SCC9 cell line showed only weak COX-1 and COX-2 expression, whereas the SCC25, CAL27, and FaDu cell lines strongly expressed both isoenzymes. Over time, COX-2 expression was down-regulated in all cell lines, whereas COX-1 was not affected by treatment.
Head and neck cancers include malignant neoplasms of the salivary glands, oral cavity, pharynx, and larynx. Squamous cell carcinoma is the predominant histologic type, representing more than 90% of cases. In the last 3 decades, we did not observe a substantial improvement of disease-specific survival for recurrent and/or metastatic head and neck cancers. Despite novel anticancer strategies, including multiple modalities of treatment such as surgery, radiation therapy, and chemotherapy, the median survival rate for patients with head and neck cancer remains very poor.
In the last few years, however, first breakthroughs were achieved with targeted therapies. Phase 2 studies with tyrosine kinase inhibitors gefitinib28,29 and erlotinib30 showed improvements in patients' quality of life, as well as increased survival rates. In their current study, Bonner et al31 combined radiotherapy with cetuximab, a monoclonal antibody against epidermal growth factor receptor, which resulted in prolonged progression-free survival. However, this outcome could only be demonstrated in cases involving oropharyngeal cancer, a clear tissue-specific effect that shows the limitations of current targeted therapies. Identification of new “anticancer drugs” or small molecules is necessary to overcome the current obstacles of modern therapy for head and neck cancer.
To date, only limited information is available regarding the role of 7β-HC in the biology of solid tumors, and virtually no information is available regarding its role in HNSCC. In this study, we report, for the first time (to our knowledge), an antiproliferative and apoptosis-inducing effect in 4 different HNSCC cell lines. Treatment was found to be very effective, as the IC50 value after 72 hours of exposure was between 13.19 and 20.79 μmol/L depending on the cell line used. Taking into consideration the fact that the IC50 value of the standard first-line chemotherapeutic drug, cisplatin, was found to be between 2 and 5 μmol/L (data not shown) in our 4 HNSCC cell lines, 7β-HC has to be considered a highly potent anticancer agent. As previously described in the literature, 7β-HC induces apoptosis in cancer cells.17,18 We have confirmed this finding by analyzing treated cells with flow cytometry as well as by immunostaining a caspase-cleaved cytokeratin 18 fragment.
Cyclooxygenase-2 has been identified as an enzyme that contributes to carcinogenesis in various ways. Overexpression of COX-2 is commonly found in HNSCC24,25 and various other premalignant and malignant lesions.32,33 Therefore, COX-2 has been identified as a promising pharmacological target in cancer prevention26 and chemotherapy. Nevertheless, a group of COX-2 inhibitors, including celecoxib, rofecoxib, and valdecoxib, is currently under critical investigation. Risk of cardiovascular adverse effects,34- 37 which have appeared after long-term use of rofecoxib and valdecoxib, led to the withdrawal of these drugs from the pharmaceutical market until April 2005.38
So far, the influence of 7β-HC on COX has not been evaluated in cancer cell lines, to our knowledge. We assessed expression of both COX isoenzymes after treatment with 7β-HC and documented a down-regulation of COX-2. The most interesting finding is the substantial inhibition of COX-2 isoenzyme by 7β-HC in all 4 cell lines. Our study shows that 7β-HC could act as a dietary supplement by helping to reduce carcinogenesis via inhibition of COX-2 in HNSCC. Furthermore, 7β-HC had no influence on COX-1 expression in the cell lines tested. These preliminary in vitro results indicate that, in addition to its cytotoxic effects, 7β-HC also exerts anti-inflammatory properties in vivo via down-regulation of COX-2, an enzyme that is strongly expressed in most cases of head and neck cancer and has been linked to carcinogenesis. This dual effect of 7β-HC makes it an interesting candidate for further studies.
In summary, we have identified 7β-HC as a cytotoxic agent that is active against HNSCC cell lines. We found that the use of 7β-HC led to induction of apoptosis in all 4 cell lines tested. The oxysterol 7β-HC seems to be an attractive substance in the treatment of head and neck cancer owing to the combination of its cytotoxic and anti-inflammatory properties. However, further in vitro and in vivo experiments are needed to evaluate a possible combination with current treatment protocols.
Correspondence: Dietmar Thurnher, MD, Department of Otorhinolaryngology–Head and Neck Surgery, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria (firstname.lastname@example.org).
Submitted for Publication: October 2, 2007; final revision received May 5, 2008; accepted May 23, 2008.
Author Contributions: Dr Heiduschka had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Heiduschka, Ehrenberger, and Thurner. Acquisition of data: Heiduschka, Erovic, Vormittag, Skoda, Martinek, and Brunner. Analysis and interpretation of data: Erovic and Thurner. Drafting of the manuscript: Heiduschka. Critical revision of the manuscript for important intellectual content: Erovic, Vormittag, Skoda, Martinek, Brunner, Ehrenberger, and Thurner. Statistical analysis: Brunner and Thurner. Obtained funding: Ehrenberger and Thurner. Administrative, technical, and material support: Erovic, Martinek, Ehrenberger, and Thurner. Study supervision: Erovic and Thurner.
Financial Disclosure: None reported.