The Association between TNF-alpha Gene Polymorphisms and Endometriosis in An Iranian Population

Endometriosis is developed as a result of endometrial tissue exposing outside the uterine cavity. Studies have reported the pelvic and the peritoneum as the most common sites of replacement (1, 2). This highly prevalent disease can be really enervating (about 30% in infertile and 10% in fertile women) (3). Approximately, 25-50% of infertile women develop endometriosis while 30-50% of women with endometriosis are infertile (4).

This polygenic disease with its complex genetic background (5, 6) occurs as a result of interactions between genetically determined factors and environment. The genetic component of endometriosis has been shown through studying the kinship of patients (7, 8). To date, the most common method for investigating genetic factors underlying complex diseases is the hypothesis-based candidate gene studies (8). One of the most important factors in endometriosis is mutations in cytokine genes. The tumor necrosis factor-alpha (TNF-α) is an important cytokine in acute inflammatory response to infective factors and is genetically variable. Based on multiple studies, TNF-α is thought to be a molecular indicator for gynecological-related diseases. It is suggested that the inflammatory response in endometriosis increases because of cytokines such as TNF-α (8, 9).

Studies on patients diagnosed with endometriosis have highlighted that TNF-α is a likely factor in developing endometriosis as suggested by elevated levels of TNF-α in peritoneal fluid and the up-regulation of TNF-α in peritoneal macrophages and peripheral blood monocytes (10, 11). However, the exact role which TNF-α plays in endometrial tissue is ambiguous (12). Thus far, some polymorphisms in the promoter region of the TNF-α gene have been examined in patients with endometriosis (12-20). Therefore, for the first time, we aimed to examine the relationship of TNF-α -238G/A, -308G/A, -857C/T and -863C/A polymorphisms with risk of developing endometriosis in Iranian women.

According to the analysis of descriptive variables, the age range was from 19 to 50 years (mean=31, SD=6.1) in the patients, and from 19 to 44 years (mean=29.2, SD=5.2) in the control group. The mean BMI (Kg/m²) in the case and control groups were 25.2 (SD=3.7) and 26.2 (SD=4) respectively. Genotypes of the TNF-α -238G/A, -308G/A, -857C/T and -863C/A polymorphisms were obtained in 150, 150, 148, 150 patients and 149, 150, 143, 150 control samples respectively. Genotype frequencies of the TNF-α -238G/A, -308G/A, -857C/T and -863C/A polymorphisms in the case and control groups were in Hardy-Weinberg equilibrium. Genotype and allele frequencies for the TNF-α -238G/A, -308G/A, -857C/T and -863C/A are shown in Table 2.

The TNF-α -863C/A allele A frequency between case and control groups represented a significant difference (P=0.047) but the result is not significant when adjusting for multiple testing (P=0.188). However, no significant difference was observed in the allele frequencies of the -238G/A (P=0.862), -308G/A (P=0.438) and -857C/T (P=0.878) polymorphisms in TNF-α between the case and control groups. We adjusted all four polymorphism genotypes by age and BMI but according to the results, no significant difference was discovered between the groups. But there was an association between the case and control and BMI when adjusting by age (OR=1.082, 95% CI=1.009-1.162, P=0.028).

TNF-α interacts with 10 other proteins according to STRING (Fig .2). In specific, they are i. TGF-beta activated kinase 1/MAP3K7 binding protein 2 (TAB2), ii. Nuclear factor of kappa light polypeptide gene enhancer in B-cells 1 (NFKB1), iii. TNF receptor-associated factor 2 (TRAF2), iv. TNFRSF1A-associated via death domain (TRADD), v. Inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase beta (IKBKB), vi. Receptor interacting serine-threonine kinase 1 (RIPK1), vii. Inhibitor of kappa light polypeptide gene enhancer in B-cells, kinase gamma (IKBKG), viii. Baculoviral IAP repeat containing 2 (BIRC2), ix. Tumor necrosis factor receptor superfamily, member 1A (TNFRSF1A), and x. Tumor necrosis factor receptor superfamily, member 1B; (TNFRSF1B).

Endometriosis is a multifactorial disease with both genetic and environmental components (8). Studies have scanned the genome and specific candidate genes to determine the genetic aetiology of this disease (22), reporting on endometriosis and related genes involved in "detoxification, galactose metabolism, steroid hormone production and inflammation" (15, 21, 22). Studies have also shown that any change in function and number of immune cells as well as high levels of inflammatory cytokines may lead to endometriosis (23). The present study aimed at examining the association of four TNF-α polymorphisms, -238G/A, -308G/A, -857C/T and -863C/A with endometriosis in an Iranian sample.

The effects of polymorphisms on cytokines genes have been examined by many studies (22) as well as the possible association of TNF-α polymorphisms with augmented endometriosis risk (12, 13, 15, 17-20). So far, many polymorphic variants have been examined in different populations, leading to various results (24). A number of polymorphisms have been associated with the disease in many populations, however, in some studies no association with the disease was observed. The reason for these observed differences may include different diagnostic criteria in selection of patients and controls, distinct living setting, number of samples, and varying populations and ethnicities (24, 25).

All in all, endometriosis has been reported to be linked with a number of polymorphisms of TNF-α. The study by Asghar et al. (12) and two other investigations by Lee et al. (17) concluded that -238G/A, -308G/A, -857C/T and -863C/A polymorphisms in the promoter region of TNF-α had no impact on women developing endometriosis. In contrast, these studies and that by Saliminejad et al. (15) showed that the frequency of the -1031C allele in the TNF-α gene in patients suffering severe endometriosis was significantly lower than that of their control group. Additional genetic studies on promoter polymorphisms in TNF-α by Wieser et al. (18) (238 G/A and-308G/A), and Hsieh et al. (13) and Lu et al. (14) (308G/A) found no association with endometriosis in the Asian population. Another study by Zhao et al. (19) in Australia reported that the -863C/A polymorphisms in TNF-α had no effect on patients with endometriosis. Moreover, Abutorabi et al. (16) found a positive association between the -1031 T/C polymorphism with endometriosis. However, no significant association was observed between the -238 G/A and -308 G/A polymorphisms with the disease. A similar study by Teramoto et al. (20) discovered an over- representation between the TNF-U01 haplotype (1031T-863C and, 857C) and endometriosis in Japanese women.

According to the studies mentioned, TNF-α -238G>A has been inspected in three studies, -308G>A in five articles, -857C>T in three articles and -863C>A in four articles, all of which showed compatible findings where no significant association was reported between these four polymorphisms and endometriosis in any of the models. Our study reported an association between the -863C/A polymorphism in the promoter region of TNF-α and endometriosis. We also observed a direct relationship between the case and control and BMI when adjusting by age (Table 3).

It has been shown that the presence of polymorphism in promoter regions can affect gene expression (26). STRING showed that TNF-α interacts with 10 other molecules. These interactions are likely to be functionally important, especially those with proteins involved in cell survival and apoptosis such as TRAF2, which regulates activation of NF-Kappa-B and JNK and has a central role in the regulation of cell survival and apoptosis (27, 28). Also, the interaction with TNFRSF1B (receptor with high dependency for TNFSF2/TNF-α) is essential for mediating most of the metabolic efficacy of TNF-α (27, 29).

This study has some limitations in spite of its strengths. The limitations of our study on endometriosis were not only the difficulty in choosing the controls, but also in recruiting patients. This is because laparoscopy should be undertaken to confirm the disease and its steady state that was a matter of time to collect samples.

We investigated the association of four polymorphisms in the promoter region of TNF-α in Iranian women with endometriosis (stages I-IV of disease). TNF-α -863 A allele was significantly lower in women with endometriosis than controls, suggesting that the -863 A allele may play a role in incidence of endometriosis among Iranian women. Development of endometriosis among those people with -863 A allele seems low although it should be noted that the calculations show is not significant when adjusting for multiple testing. According to the results, the current study indicates that there might be a direct relationship between BMI and progression of endometriosis.