Abstract
Background/Aim: The over-expression of enhancer of zeste homolog 2 (EZH2) protein is found in oral cancer tissues. However, the genetic role of the enhancer of EZH2 in the etiology of oral cancer is unknown. The aim of this study was to evaluate the association of EZH2 genotypes with oral cancer risk among Taiwanese. Materials and Methods: Three polymorphic variants of EZH2, rs887569 (C to T), rs41277434 (A to C), and rs3757441 (T to C), were analyzed regarding their association with oral cancer risk among 958 oral cancer patients and the same number of healthy controls by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). In addition, the interaction of EZH2 rs887569, rs41277434, and rs3757441 genotypes with personal behaviors such as smoking, alcohol drinking, and betel quid chewing were also examined. Results: The EZH2 genotypes rs887569, rs41277434, and rs3757441, were not significantly associated with oral cancer risk (p for trend=0.1735, 0.5658, and 0.4606, respectively). The analysis of allelic frequency distribution also supported the findings that the variant alleles at EZH2 rs887569, rs41277434, and rs3757441 may not serve as determinants of oral cancer risk (all p>0.05). There was no interaction between EZH2 rs887569, rs41277434, or rs3757441 genotypes with personal smoking, alcohol drinking or betel quid chewing behaviors. Conclusion: EZH2 genotypes cannot predict oral cancer risk in Taiwan.
Worldwide, oral cancer is ranked sixth among common cancers and has the incidence in Taiwan (1, 2). It has been reported that cancer of the oral cavity and oropharynx represented more than 475,000 newly diagnosed cases all of the world in 2020 (3, 4). In addition to environmental factors, such as tobacco smoking, alcohol drinking, betel quid chewing, and virus infection (5), genetic factors have been suggested to affect an individual’s risk for oral cancer (6-13). Although the development of modern facilities is rapid and health care of cancer has been greatly improved in recent years, more than 450,000 patients worldwide are diagnosed with oral cancer annually, and oral cancer maintains a five-year survival rate of under 50% (14, 15). The fact that the newly diagnosed cases of oral cancer have been increasing for at least three decades has urged genomic scientists to find more practical and feasible markers for early detection of oral cancer to help decrease its incidence (16).
Human enhancer of zeste homolog 2 (EZH2) gene is located on the long arm of chromosome 7 at 7q35, contains 20 exons, and encoded a 746-amino acid protein (17). EZH2 protein is an important epigenetic chromatin modifier that regulates target genes involved in cell proliferation, differentiation, and tumorigenesis (18). In the early twentieth century, EZH2 over-expression began to be reported in hematological malignancies (17, 19). After that, up-regulation and/or mutation of EZH2 gene were observed in a panel of human cancers, such as breast, lung, liver, colorectal, prostate, bladder, and endometrial cancers, as well as melanoma and lymphoma (20-27). Over-expression of EZH2 can lead to disruptions in cell proliferation, apoptosis, migration, and invasion, and abnormally highly expressed EZH2 has been reported to closely correlate with poor prognosis in several types of cancer, including breast cancer (20, 22, 28), esophageal cancer (29), gastric cancer (30-34), colorectal cancer (35, 36), hepatocellular carcinoma (37), pancreatic cancer (38), endometrial cancer (22), prostate cancer (21, 22), and particularly oral cancer (39).
In 2011, it was reported that up to 80% of oral leukoplakia cases, which showed strong expression of EZH2 in immunohistochemistry, developed oral squamous cell carcinoma within 5 years after their first diagnosis, whereas only 24% of the cases with moderate or weak to almost no expression of EZH2 (40). Thus, EZH2 may serve as one of the best predictors of oral cancer risk for patients with oral leukoplakia (41, 42), and may also become a potential target in anticancer drug development (40, 43). However, EZH2 protein expression varies dramatically, and the contribution of EZH2 genotypes to susceptibility for oral carcinogenesis has never been studied. The current work focused on revealing the association of genotypes of EZH2 at rs887569 (C to T), rs41277434 (A to C) and rs3757441 (T to C) among 958 oral cancer patients and 958 healthy controls in Taiwan, a highly genetically-conserved and highly oral cancer prevalent population. The genomic positions of these EZH2 polymorphic sites are shown in Figure 1.
Map of EZH2 rs3757441, rs41277434, and rs887569 polymorphic sites.
Materials and Methods
Oral cancer case and control recruiting methodology. All the oral cancer cases were recruited in our hospital in central Taiwan. Patients with a history of any appearance of malignancy, metastasized tumor, genetic inherited diseases, or cancer-like diseases, such as pterygium and endometriosis, were excluded from the study. All patients signed inform consent form prior inclusion in the study. All participants were Taiwanese. At the same time, the same number of healthy volunteers, matched for age and sex, from the Health Examination Cohort of China Medical University Hospital were selected as controls. They also signed informed consent forms. The study was approved by IRB of China Medical University Hospital (DMR101-IRB1-306). All the protocols have been carefully checked and conducted following the principles of the Helsinki Declaration. More details on the sampling of the cases and controls are available in our previous studies (12, 44). The demographic characteristics of all participants are listed in Table I.
Characteristics of 958 oral cancer patients and 958 healthy subjects.
EZH2 genotyping settings. Each participant provided 3-5 ml venous blood for genotyping. Genomic DNA was extracted from their peripheral blood leucocytes following the protocol guidance of QIAamp Blood Mini Kit (Blossom, Taipei, Taiwan, ROC) and further processed as our routine processes (45-47). Concisely, the polymerase chain reaction (PCR) conditions for EZH2 genotyping were set as: 94°C for 5 min; 35 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s, and a final extension at 72°C for 10 min using BIO-RAD Mycycler PCR machine (BIO-RAD, Hercules, CA, USA). Amplified and digested DNA adducts for EZH2 rs887569 and rs41277434 were monitored with 3% agarose gel electrophoresis, and EZH2 rs3757441 was sequenced as we have previously published (48).
Statistical analysis. Demographic characteristics were compared by Student’s t-test (specifically for age) and Chi-square test (other indexes). Hardy Weinberg Equilibrium (HWE) test was based on Chi-square test of goodness of fit between observed and expected EZH2 genotypes. Pearson’s Chi-square test was used for evaluating the distribution of the EZH2 genotypic and allelic frequencies. The odds ratio (OR) together with 95% confidence interval (CI) were calculated for assessing the relative risk conferred by a particular EZH2 allele and genotype. The results were assumed as significant whenever p-value was less than 0.05.
Results
Description and comparison of demographic indexes between oral cancer cases and controls. There is no difference between the oral cancer case and control groups regarding age and sex, since they were matched during recruitment (p=0.3755 and 1.0000, respectively) (Table I). There are more smokers (p=0.0107), alcohol drinkers (p=0.0377), and betel quid chewers (p=0.0001) in the oral cancer compared to the control group, consistent with the fact that cigarrete smoking, alcohol drinking, and betel quid chewing are the major risk factors for oral cancer in Taiwan (Table I). Among the oral cancer cases, the most prevalent primary tumor sites were the tongue (397 cases, 41.4%) and buccal mucosa (356 cases, 37.2%) (Table I). As for oral cancer subtypes, 93.1% (892 out of 958) of the oral cancer cases are of the squamous cell carcinoma subtype (Table I).
Comparison of EZH2 rs887569 genotype distribution among oral cancer patients and healthy controls. First, the genotypic frequency distribution of EZH2 rs887569 satisfied the HWE (p for HWE=0.1767, Table II). Second, there was no difference between oral cancer case and control groups as for the distribution of EZH2 rs887569 genotypic frequency (p for trend=0.1735, Table II). Third, the ORs for people with CT and TT genotypes were non-significant, 1.05 (95%CI=0.87-1.27, p=0.6588) and 0.76 (95%CI=0.54-1.06, p=0.1236) compared with those with the wild-type CC genotype (Table II). Fourth, in recessive (CC+CT versus TT) and dominant (CC versus CT+TT) comparisons, the ORs were 0.74 (95%CI=0.54-1.03, p=0.0846) and 0.99 (95%CI=0.83-1.19, p=0.9636), respectively (Table II).
Distribution of enhancer of zeste homolog 2 (EZH2) rs887569 genotypes among oral cancer patients and healthy subjects.
Comparison of EZH2 rs41277434 genotype distribution among oral cancer patients and healthy controls. First, the genotypic frequency distribution of EZH2 rs41277434 satisfied the HWE (p for HWE=0.1474, Table III). Second, there was no difference between oral cancer case and control groups as for the distribution of EZH2 rs41277434 genotypic frequency (p for trend=0.5658, Table III). Third, the ORs for people with variant AC and CC genotypes were non-significant, 0.90 (95%CI=0.73-1.10, p=0.3135) and 0.94 (95%CI=0.58-1.53, p=0.9041), compared with those with the wild-type AA genotype (Table III). Fourth, in recessive (AA+AC versus CC) and dominant (AA versus AC+CC) comparisons, the ORs were 0.97 (95%CI=0.60-1.57, p=0.9024) and 0.90 (95%CI=0.74-1.10, p=0.3178), respectively (Table III).
Distribution of enhancer of zeste homolog 2 (EZH2) rs41277434 genotypes among oral cancer patients and healthy subjects.
Comparison of EZH2 rs3757441 genotype distribution among oral cancer patients and healthy controls. First, the genotypic frequency distribution of EZH2 rs3757441 satisfied the HWE (p for HWE=0.9108, Table IV). Second, there was no difference between oral cancer case and control groups as for the distribution of EZH2 rs3757441 genotypic frequency (p for trend=0.4606, Table IV). Third, the ORs for people with CT and CC genotypes were non-significant, 0.92 (95%CI=076-1.11, p=0.4170) and 0.84 (95%CI=0.62-1.14, p=0.2997), compared with those with the wild-type TT genotype (Table IV). Fourth, in recessive (TT+CT versus CC) and dominant (TT versus CT+CC) comparisons, the ORs were 0.87 (95%CI=0.65-1.17, p=0.4091) and 0.90 (95%CI=0.76-1.08, p=0.2919), respectively (Table IV).
Distribution of enhancer of zeste homolog 2 (EZH2) rs3757441 genotypes among oral cancer patients and healthy subjects.
Comparison of EZH2 allelic frequency distribution among oral cancer patients and healthy controls. Last, the results of the allelic frequency analysis showed that there was no altered OR for neither people carrying the variant T allele at EZH2 rs887569 (OR=0.94, 95%CI=0.82-1.08, p=0.4122), variant C allele at EZH2 rs41277434 (OR=0.92, 95%CI=0.78-1.09, p=0.3631), nor variant C allele at rs3757441 (OR=0.92, 95%CI=0.80-1.05, p=0.2395, Table V). The tentative conclusion deduced from Table II, Table III, Table IV and Table V is that among the representative population containing 958 cases and 958 controls, neither EZH2 rs887569, rs41277434, nor rs3757441 may serve as a good biomarker for early oral cancer detection.
Distribution of enhancer of zeste homolog 2 (EZH2) allelic frequencies among oral cancer patients and healthy subjects.
Discussion
EZH2 plays a role in cell-cycle regulation, proliferation, tissue maturation, and tumorigenesis, and accumulated studies have shown that over-expression of EZH2 is closely related to the invasive capacity and poor prognosis of many types of human cancers (49, 50). To the best of our knowledge, in the current study, we have chosen three polymorphic sites of EZH2, rs887569, rs41277434 and rs3757441, and clarified their associations with oral cancer risk in Taiwan for the first time. EZH2 rs887569 is located in the 19th intron of the EZH2 gene, which has been reported to bind with the PPAR-α/RXR-α complex. The PPAR-α/RXR-α complex can down-regulate the expression level of EZH2, trigger the cell to undergo apoptosis, and inhibit cell proliferation (51). Our results showed that variant CT and TT genotypes of EZH2 rs887569 were not associated with risk for oral cancer among the investigated Taiwanese population, compared to those with CC wild-type genotypes (Table II). This is different from previous findings reporting that the T allele of EZH2 rs887569 is associated with decreased susceptibility for colorectal cancer in a Chinese population (36) and bladder cancer in a Taiwanese population (48). Furthermore, the genotypes of EZH2 rs3757441, located in the 6th intron of the EZH2 gene, were also associated with colorectal cancer risk (36), while negative association was found in the study on bladder cancer (48). In our results, the genotypes of EZH2 rs3757441 were not associated with oral cancer risk (Table IV). Aside from the different populations investigated, possible sampling bias and technical errors that may have happened, it is more reasonable that EZH2 SNPs regulate different signaling networks in different types of cancer cells. Further investigations are needed to clarify this.
The causes of oral cancer are complex and largely unrevealed. Almost 90% of oral cancers are linked to cigarrete smoking, alcoholism, and poor diet (52, 53). In Taiwan, betel quid chewing has been reported to be the most dangerous contributor to oral cancer many times (11-13, 54-57). As a meta-analysis has suggested that smokers are five times more likely to develop oral cancer than non-smokers (58), our team has further found that smoking habits can synergistically enhance the influence of specific risk genotypes (9, 11-13, 56, 58-62). In this study, we have also accessed the interaction of smoking behavior with EZH2 rs887569, rs41277434, and rs3757441 genotypes on oral cancer risk, but no interaction was found (data not shown).
Alcohol drinking has been reported to increase the risk of oral cancer by 1.56-fold, and higher rates of oral cancer are associated with combined consumption of tobacco and alcohol (63, 64). In this study, we also found that alcoholism seems to be one of the risk factors for oral cancer in the investigated population (Table I). However, in our previous findings, a less positive combined influence of alcohol in association with specific genotypes was found (11, 12). More frequently, no combined effect was found (6-8, 10, 65-69). Therefore, alcoholism may play a minor role in enhancing the contribution of these genotypes to oral cancer, or it may not be so powerful as cigarrete smoking and/or betel quid chewing.
Previous studies have shown that betel quid chewing behaviors are also one of oral cancer contributors (63, 70). This is much more serious in Taiwan (5, 71, 72). Although the genotypes of EZH2 rs887569, rs41277434, and rs3757441 were not directly associated with oral cancer risk, we are still interested in the interaction between the genotypes of EZH2 and betel quid chewing habits. The results showed no interaction between EZH2 rs887569, rs41277434, and rs3757441 and betel quid chewing behavior (data not shown). As for prognosis indexes such as metastasis status and 5-year survival, no interaction with EZH2 rs887569, rs41277434, and rs3757441 was found (data not shown).
In summary, to the best of our knowledge, this is the first study to focus on EZH2 genotype and its interaction with behavioral status, including smoking and alcohol drinking habit on oral cancer risk in Taiwan. In this representative population, the polymorphic genotypes of EZH2 rs887569, rs41277434, and rs3757441 seemed not to be a good marker for early oral cancer detection. The contribution of EZH2 to oral cancer etiology at the post-translational level are worth further investigation.
Acknowledgements
This study was supported mainly by China Medical University, Taichung, Taiwan and Asia University, Taichung, Taiwan (CMU111-ASIA-01) and partially by research grant from Taichung Veterans General Hospital, Taichung, Taiwan (TCVGH-1117201B). The Authors thank their colleagues in the Tissue bank of China Medical University Hospital for their technical assistance.
Footnotes
Authors’ Contributions
Research design was performed by LCS, CWT, and TCL. Patient and questionnaire collections were conducted by LCS and CLH. Experimental work was performed by YCW, JLH, JSY, and YCW. Statistical analysis was conducted and checked by TCH, FJT, and YMH. DTB and WSC wrote the manuscript, whereas DTB, LCS, CWS, and WSC reviewed it and were responsible for the revision.
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
- Received September 12, 2022.
- Revision received September 22, 2022.
- Accepted September 23, 2022.
- Copyright © 2022 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
