Abstract
The head and neck region is one of the most important locations predisposed for tobacco-associated cancer. Chemoprevention might offer a chance to decrease the risk for this type of disease. Materials and Methods: Mini-organ cultures (MOC) of macroscopically-healthy pharyngeal tissues from 20 patients with oropharyngeal squamous cell carcinoma (SCC) and from 20 controls were employed in the study. MOC were firstly incubated with Celecoxib, and DNA damage was induced by incubation with Benz[a]pyren-7,8-diol-9,10-epoxid (BPDE), a major representative of tobacco-associated carcinogens. DNA damage was evaluated with the alkaline single-cell microgel electrophoresis (Comet assay). Furthermore, fragmentation of the cyclin D1 gene, a gene of special importance in head and neck carcinogenesis was examined by the Comet-FISH assay. Finally, the chemoprotective potential of Celecoxib was analyzed after incubation with MOC. Results: As expected, BPDE caused significant DNA fragmentation in tumor compared to negative control tissues. No enhanced damage was observed in the cyclin D1 gene. DNA fragmentation was significantly reduced when MOC were incubated with Celecoxib in the tumor group. Surprisingly, these effects were also observed in the group without cancer of the oropharynx, although COX-2 is not expressed in macroscopically-healthy mucosa. Conclusion: Celecoxib showed considerable chemoprotective effeciency against BPDE in both groups and this effect seems to be independent of COX-2 expression. No evidence for higher mutagen sensitivity in the Cyclin D1 gene was observed.
DNA mutations are of fundamental aetiology when considering possible explanations for carcinogenesis. Oxidative DNA damage or the formation of DNA adducts by metabolically-activated carcinogens are common events (1). Tobacco smoke, the major risk factor for development of squamous cell cancer of the head and neck (HNSCC), comprises of various xenobiotics. Up to 1015 free radicals and reactive species are enclosed per puff, demonstrating the high mutagenic potential of cigarette smoke (2). Among those compounds, at least 50 carcinogenic agents have been detected in tobacco smoke so far, polycyclic aromatic hydrocarbons being the most frequent ones (3). In the pharyngeal mucosa, benz[a]pyren-7,8-diol-9,10-epoxide (BPDE), the active and carcinogenic form of polycyclic aromatic carbohydrates, arises from the transformation of benz[a]pyrene by cytochrome P4502E (4). When BPDE reacts with the DNA, it frequently builds adducts, predominantly BPDE-deoxyguanosine adducts (5).
“Chemoprevention” was introduced by Sporn et al. in 1976 (6). It describes the use of specific natural, biological or synthetic agents to reverse, suppress or prevent cancer development (7). In 1986 Hong et al. showed, that treatment of oral leukoplakia with 13-cis-retinoic acid, leads to a decrease in size and a histological downgrading (8). Since then large clinical intervention studies testing for different agents, mostly vitamins, have been undertaken, partly with promising results (9). As a consequence, various synthetic or natural dietary compounds have been identified as effective agents in the prevention of carcinogenesis (7, 10).
In the present study, we evaluated the chemoprotective potential of Celecoxib after incubation with BPDE. Mini-organ cultures (MOC) of macroscopically-healthy pharyngeal tissue of 20 patients with, and 20 patients without oropharyngeal carcinoma were used. DNA damage was quantified using the Comet assay. Furthermore, we evaluated the susceptibility of the Cyclin D1 gene to BPDE-induced DNA damage as a risk factor for DNA mutation in the specific region. Cell-cycle activation is coordinated by cyclin D1, which is rate-limiting and essential for the G1 phase progression of the cell cycle. Mutations in this gene might lead to loss of function and uncontrolled cell proliferation (11). The Comet-FISH assay was applied to quantify for DNA gene damage.
Materials and Methods
Celecoxib. Celecoxib is not available as a pure active agent, and only Celebrex© is distributed by the producer. Besides, Celecoxib consists of the following ingredients: lactosemonohydrate, natriumdodecylsulfate, povidon K 30, croscarmellose sodium, magnesium, gelatine, titandioxide and indigotindisulfonate. It seems highly unlikely for the observed effects not to be caused by any of these ingredients but Celecoxib itself.
Biopsies. Tissue cultures from fresh-biopsied mucosa samples from tumor and control patients were prepared. Control patients suffered from chronic tonsillitis or obstructive sleep apnea syndrome. Samples were harvested during resection of oropharyngeal squamous cell carcinomas and tonsillectomy, respectively. Biopsies of tumor patients were obtained from macroscopically-normal mucosa close to the tumor-free margins. The study was approved by the Ethical Commission of the Medical Department, Ludwig-Maximilians-University (No. 349 05).
Mini-organ cultures (MOC). Specimens were dissected into 1-mm3 cubes excluding deeper layers and washed three times in bronchial epithelial cell basal medium (BEGM; Promocell, Heidelberg, Germany). Cubes were placed in 24-well plates, coated with 0.75% Agar Noble (Difco, Detroit, MI, USA), and dissolved in Dulbecco's Modified Eagle's Medium (Gibco; Eggenstein, Germany), 10% fetal calf serum (Gibco), non-essential amino acids (Gibco) and amphotericine B (Gibco). After about 20 days in 250 μl BEGM, each well at 37°C, 5% CO2 and 100% relative humidity, MOC were completely coated with epithelium. BEGM was replaced every second day during cultivation. Multi-well plates were changed every week. After complete epithelialization, most cells were of epithelial origin, and there was just a limited number of stromal cells, vessels and glands.
Incubation and cell separation. MOC were incubated with 25 μl of Celebrex© (0,1 μmol/ml; Pfizer, NY, USA) for 30 min at 37°C and washed twice afterwards. Dimethylsulfoxide (166 mM) (DMSO; Merck; Darmstadt, Germany) served as the negative control. Part of the MOC were incubated with 25 μl BPDE (9 μM) (Midwest Research, Kansas, USA) for 60 min. Again, DMSO served as a negative control. MOC were washed twice with BEGM before further treatment. After microscopic preparation all biopsies underwent enzymatic digestion with a solution containing 10 mg hyaluronidase (Boehringer; Mannheim, Germany), 10 mg collagenase (Roche; Mannheim, Germany), 50 mg proteases (Sigma; Steinheim, Germany) for 45 min at 37°C. To preserve the physiological character of the samples, no metabolic activation was implemented before the incubation period. Viability was tested with trypan blue staining (12).
Comet assay. The alkaline version of the Comet assay was carried out according to a standard protocol (13). Lysis of cellular and nuclear membranes was performed under alkaline conditions in a solution containing 2.5 M NaCl (Sigma), 10 mM Trizma-base (Merck), 100 mM Na2EDTA (Serva, Heidelberg, Germany) and 1% N-laurosylsarcosin sodium salt (Sigma) at pH 10, with 1% Triton X-100 (Sigma) and 10 % DMSO (Merck) added before use. Subsequently, slides were positioned in a horizontal electrophoresis chamber (Renner, Darmstadt, Germany), close to the anode. Slides were covered with a 4°C-cold alkaline buffer solution (pH 13.2) consisting of 300 mM naOH (Merck) and 1 mM Na2EDTA (Serva). The DNA unwinded for 20 min just before migration within an electric field (25 V, 1.0 V/cm, 300 mA, 20 min). The alkaline buffer was then neutralized with Trizma base solution (Merck; 400 mM, pH 7.5). Fluorescent DNA staining was performed with 75 μl ethidiumbromide (Sigma). After staining, slides were analyzed with a DMLB microscope (Leica, Bensheim, Germany). 80 cell nuclei per slide were selected randomly and digitized with the attached monochrome CCD camera (Cohu Inc., San Diego, CA, USA).
Comet-FISH. For hybridization, the protocol of McKelvey-Martin et al. (14) was used under minor modifications. After neutralization and treatment with SSC-buffer (saline sodium citrate buffer) (0.3 M NaCl, 30 mM sodium citrate), the slides were dehydrated with alcohol (70, 85 and 100%) and dried at 37°C. A hybridization mixture was added, containing (all quantities are listed per slide) hybridization-buffer (formamide with dextran sulfate, 14 μl), DNA-probes (2 μl, LSI EGFR Dual Color Probe-Hyb Set) (Abbott; Abbott Park, IL, USA)) and Aqua bidest (4 μl). The DNA probes hybridized to the centromere of chromosome 7 and the EGFR gene on the same chromosome simultaneously. The centromere served as a reference gene, due to its close location on the same chromosome as the EGFR gene.
After coverage and sealing of the prepared slides and incubation at 74°C for 5 min on a precision hot plate, the slides were placed into a wet-chamber for 12-16 h at 37°C. Before detection of probes, the slides were washed three times each in 50% formamide in 2xSSC (Abbott) and incubated for 10 min in 2xSSC and 0.1% detergent tergitol NP-40 in 2xSSC for 5 min.
Staining and analysis. 10 μl DAPI (42 ng/ml) with Antifade (Abbott) were applied after air-drying the slides, followed by −20°C storage protected from light. The DNA fragmentation was visualized using a fluorescence microscope and digital analysis (Comet++, Kinetic Imaging™; Liverpool, UK). 40 cells per slide and 2 slides per patient were analyzed. To describe the degree of chromosomal damage in the cell, the Munich chromosomal tail moment (MCTM) was used. The MCTM is calculated by the median chromosomal migration distance and the chromosomal fluorescence in the tail of the comet devided by the entire chromosomal fluorescence measured in a cell.
Statistical analysis. Statistical analysis was performed using SPSS 21.0™. DNA damage of all patients was compared using the Wilcoxon's test. The level of significance was set at p≤0.05.
Results
The average age of patients in the group without HNSCC was 46.4 years (range: 29-62: 9 males, 6 females). In the group with HNSCC the average age was 55.6 years (ranging: 46-72 years: 11 males, 4 females).
DNA damage caused by DMSO, Celecoxib and BPDE in tumor vs. non-tumor-group (n=30).
In the analysis of all 30 patients, benzo[a]pyren-diolepoxid (BPDE), as expected caused significant DNA damage compared to controls incubated with DMSO and Celecoxib. BPDE caused a median OTM of 8.9 in non-tumor, and of 8.6 in tumor samples. Only OTM values >2 were considered to represent significant DNA damage (15). DMSO and Celecoxib showed values >1 for both groups (Figure 1). The p-value for genotoxicity after incubation with BPDE was p≤0.001 for patients without and p≤0.001 for patients with carcinoma of the oropharynx. No significant differences between the patient groups could be found comparing OTM values after incubation with BPDE (p=0.970), respectively.
When MOC were treated with Celecoxib before DNA damage was caused, a significant damage reduction could be observed in both groups. Damage was reduced to 4.6 (non-tumor) vs. 4.4. (tumor) OTM. No significant difference was found between the two groups (Figure 2).
Significant damage could also be detected in the cyclin D1 gene for both groups. MCTMs after DNA fragmentation were 8.8 (non-tumor group) vs. 10.2 (tumor group) and were comparable for the entire DNA. Although values were higher in the tumor group, there was no significant difference (p=0.095).
Incubation with Celecoxib reduced DNA fragmentation in cyclin D1 in both groups. While in the non-tumor group an MCTM of 4.2 was detected, the tumor-group showed MCTM of 4.7 (Figure 3).
Discussion
Approximately two-thirds of head and neck squamous cell carcinomas can be attributed to heavy tobacco and alcohol consumption (16). Tobacco carcinogens, like BPDE, exert their harmful effects following decades of smoking. Throughout this period genetic alterations are accumulated in the mucosa finally resulting in malignant transformation of the cells (17). Cancer chemoprevention, the use of natural or synthetic compounds to prevent, arrest, or reverse the process of carcinogenesis, aims to reverse pre-malignancies and prevent second primary tumors (18). To be useful in humans, a compound must have an acceptable safety profile in addition to being effective at a dose low enough not to cause significant toxicity (7). The term “chemoprevention” was introduced by Sporn et al. in 1976 and since then a variety of compounds have been tested for their chemopreventive potential (6). COX-2 inhibitors have been evaluated on this respect before. As COX-2 influences cell proliferation and neo-angiogenesis, inhibition of COX-2 was found to reduce these effects (19). On the other hand direct effects of Celecoxib, independent of COX-2 have also been observed. Following Celecoxib treatment, certain tumor cell lines were arrested in G1 phase and reduced expression of cyclin D1 was detected (20). Narayan et al. found inhibitory effects of Celecoxib on DNA synthesis (21), while other groups found anti-proliferative effects in vitro and in vivo (22). In our study Celecoxib showed chemoprotective effects in patients with cancer of the oropharynx. In these patients, expression of COX-2 in adjacent tissue is described in up to 95% of the cases (24, 25). On the otherhand, expression of COX-2 in patients without malignancy of the head and neck is a rare event in oropharyngeal mucosa (23). Thus, chemopreventive effects detected in the tumor-free control group, must be independent from COX-2 expression as well. These findings correlate with previous findings of other groups where chemopreventive potential of Celecoxib against oxidative DNA damage was detected (24).
Chemoprevention by Celecoxib in tumor and tumor-group (n=30).
DNA damage and chemoprevention by Celecoxib in the cyclin D1 gene in tumor and tumor-group (n=30).
Combining the Comet assay and fluorescence in situ hybridization, human upper-aerodigestive tract cells can be used to identify DNA damage and specific alterations by hybridizing genetic sites of interest. In our study we evaluated DNA damage in the cyclin D1 gene located in 11q13, a gene of special interest in the head and neck area. Its protein is of significant importance in the regulation of the cell cycle in the G1 phase. Thus, abnormal changes in cyclin D1 can result in uncontrolled cell growth, finally ending up in malignant transformation (25). Mutations in this gene are a frequently event in HNSCC. Allelic gain, as well as allelic loss are reported in this area and gene mutations occur more frequent than in other DNA areas (26, 27). As Celecoxib interacts with cyclin D1 (28) and down-regulates the protein expression rate, this gene was of special interest in our study. Nevertheless, no enhanced mutagen sensitivity was revealed for the gene, neither in the tumor- nor in the non-tumor group. Incubation with Celecoxib led to a significant reduction of DNA fragmentation in both groups, but the effect was comparable to the effects in the entire DNA. As a consequence, enhanced mutagenic sensitivity in cyclin D1 does not seem to play a significant role in the accumulation of DNA mutations in the gene.
- Received August 29, 2013.
- Revision received November 27, 2013.
- Accepted November 28, 2013.
- Copyright © 2014 The Author(s). Published by the International Institute of Anticancer Research.
