- Email: [email protected]
E-cadherin gene polymorphisms and haplotype associated with the occurrence of epithelial ovarian cancer in Chinese
Available online at www.sciencedirect.com
Gynecologic Oncology 108 (2008) 409 – 414 www.elsevier.com/locate/ygyno
E-cadherin gene polymorphisms and haplotype associated with the occurrence of epithelial ovarian cancer in Chinese Yan Li a,⁎, Jun Liang b , Shan Kang b , Zhiming Dong a , Na Wang a , Huimin Xing b , Rongmiao Zhou a , Xiulan Li b , Xiwa Zhao b a
Department of Molecular Biology, Hebei Cancer Institute, Hebei Medical University, Fourth Hospital, Jiankanglu 12, Shijiazhuang 050011, China b Department of Obstetrics and Gynaecology, Hebei Medical University, Fourth Hospital, Shijiazhuang, China Received 2 August 2007 Available online 26 November 2007
Abstract Backgrounds and aims. E-cadherin plays an important role in the origin of epithelial ovarian cancer. However, the exact molecular mechanism by which this occurs is unknown. The polymorphisms located at the E-cadherin may contribute to an increased risk for certain cancers. In this paper, we studied the association between polymorphisms of E-cadherin and the risk of epithelial ovarian cancer. Methods. We assessed the −160C/A, −347G/GA polymorphism within the promoter region and 3′-UTR + 54C/T polymorphism of E-cadherin in epithelial ovarian cancer and control women. We also tested the expression of E-cadherin protein in ovarian cancer tissue among three genotype (3′UTR + 54C/T polymorphism) carriers. Results. There was no significant difference in genotype distribution of the −160C/A and − 347G/GA SNPs in the E-cadherin gene promoter region between ovarian cancer patients and controls, but haplotype − 160A/− 347GA relative to haplotype −160C/−347G was 48.6 (95% CI = 2.9–806.2) for epithelial ovarian cancer risk. The C/C genotype of the 3′-UTR + 54C/T polymorphism relative to the C/T + T/T genotype was 1.85 (95% CI = 1.27–2.69) for epithelial ovarian cancer risk. E-cadherin protein expression in was lower in C/C genotype carriers than T allele carriers in ovarian cancer tissue (P = 0.02). Conclusions. The C/C genotype of 3′-UTR C/T SNP and −160C/−374GA haplotype in E-cadherin gene may be a potential susceptibility factor for risk of epithelial ovarian cancer in Chinese, which indicated that the lower expression of E-cadherin might play an important role in the pathogenesis of epithelial ovarian cancer. © 2007 Elsevier Inc. All rights reserved. Keywords: E-cadherin; Single nucleotide polymorphism; Expression; Epithelium ovarian cancer; Susceptibility
Introduction Ovarian cancer is one of the most prevalent causes of malignant tumors in the female reproductive system. Its incidence rate is third, surpassed only by cervical and uterine cancer. Nevertheless, it is the leading cause of death in all types of gynecological malignancy cancer in women. Epithelial ovarian cancer originates from ovarian surface epithelium and occupies 90% of all ovarian tumors in women. So far, the molecular mechanisms underlying this process are not well
⁎ Corresponding author. Fax: +86 311 86077634. E-mail address: [email protected] (Y. Li). 0090-8258/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2007.10.024
characterized. Recent studies showed that E-cadherin plays an important role in the origin of epithelial ovarian cancer [1]. E-cadherin is a member of the cadherin family of Ca2+dependent cell–cell adhesion molecules and maps to chromosome 16q22.1. Gene products are transmembrane glycoproteins. E-cadherin, which is expressed in almost all epithelial cells [2], mediates homotypic cell-to-cell adhesions and maintains epithelial cell polarity and integrity. E-cadherin is important for regulating epithelial development, maintaining organization and cell integrity. Loss of E-cadherin expression is associated with invasion and dedifferentiation of carcinoma cells [3]. As a metastatic suppressor molecule in carcinomas, the decrease of E-cadherin expression is not only associated with tumour progression and metastases formation in a series of
410
Y. Li et al. / Gynecologic Oncology 108 (2008) 409–414
different cancers [4–7] but also results in the development of cancer by activating the WNT/β-catenin signal transduction pathway and specifically promoting over-expression of oncogenes such as c-myc, cyclinD1 and so on [8]. Aberrant expression of E-cadherin may be due to loss of heterozygosity, gene mutation and methylation of the Ecadherin promoter region [9–11]. Moreover, polymorphisms in the E-cadherin gene might influence transcriptional activity and therefore be associated with risk for a series of different cancers [12–16]. The relationship between E-cadherin gene polymorphisms and risk of ovarian cancer has not been reported. In this paper, we studied the genotype and allele frequencies of three polymorphisms (− 160C/A, − 347G/GA and 3′-UTR +54C/T) within the E-cadherin gene in ovarian cancer patients and healthy controls. To this date, because the function of 3′-UTR + 54C/T polymorphism is unclear, we tested whether the 3′-UTR single nucleotide polymorphism (SNP) genotype modified E-cadherin expression in ovarian cancer tissue of patients. Materials and methods Study participants Two hundred seven patients with ovarian cancer admitted for tumor resection in the Fourth Affiliated Hospital, Hebei Medical University from January 2001 to December 2006 were selected for the study. Epithelial ovarian cancer was identified through histopathologic examination by the Department of Pathology of the same hospital. The demographic and pathologic features of patients were summarized in Table 1. The control group consisted of 256 women without any malignant disease as confirmed by surgical exploration during voluntary abortion, cesarean section, uterine prolapse or pathologically confirmed after hysterectomy for uterine bleeding. All women in the control group had no previous oophorectomy and no history of cancer or genetic disease. The mean age of patients was 53 years (range 20–78 year) and of controls was 50 years (range 21–73 year). There was no statistically significant difference in age distribution between the two groups (P N 0.05). All cancer patients and control subjects were unrelated and of Han nationality from Shijiazhuang, Hebei and the surrounding regions. The Ethics Committee of the Hebei Cancer Institute approved the study and informed consent was obtained from all recruited subjects.
Table 1 Demographic and pathologic features of patients with ovarian cancer Epithelial ovarian cancer Age (years) ≤ 40 41–50 51–60 N60 Tumor histology Serous Mucinous Endometrioid Undifferentiated Tumor stage I/II III/IV
Patients (n = 207) n
%
23 64 76 44
11.1 30.9 36.7 21.3
79 23 77 28
38.2 11.1 37.2 13.5
67 140
32.4 67.6
Fig. 1. E-cadherin − 160 bp C/A genotyping by PCR–RFLP analysis followed by separation on 3% agarose gel as described in text. Lane M = 100 bp ladder; lanes 2, 3, 6 = C/C; lanes 1, 5 = C/A; lanes 4, 7 = A/A.
DNA extraction Venous blood (5 ml) was collected from each subject into Vacutainer tubes containing EDTA and stored at 4 °C. After sampling, genomic DNA was extracted within 1 week by proteinase K (Merck, Darmstadt, Germany) digestion followed by a salting out procedure according to the method published by Miller et al. [17].
E-cadherin − 160C/A, −347G/GA, 3′-UTRC/T genotyping E-cadherin − 160C/A, −347G/GA and +54C/T genotypes were determined by the polymerase chain reaction–restriction fragment length polymorphism (PCR–RFLP) assay. The primers used for amplifying the − 160C/A, − 347G/GA in the E-cadherin promoter fragment were 5′-CGCCCCGACTTGTCTCTCTAC-3′ (forward) and 5′-GGCCA CAGCCAATCAGCA-3′ (reverse) and for amplifying the +54C/T SNP in the E-cadherin 3′untranslated region were 5′-CAGACAAAGACCAGGACTAT-3′ (forward) and 5′-AAGGGAGCTGAAAAACCACCAGC-3′ (reverse). PCR was performed in a 25-μl volume containing 100 ng of DNA template, 2.5 μl of 10× PCR buffer, 1.5 μl of 25 mmol/L MgCl2, 2.5 U of Taq-DNA-polymerase (BioDev-Tech, Beijing, China), 0.5 μl of 10 mmol/L dNTPs and 200 nM of each primer. PCR cycling conditions were 5 min at 94 °C followed by 35 cycles of 30 s at 94 °C, 30 s at 66 °C for C-160A, G-347GA, 30 s at 56 °C for +54C/T and 60 s at 72 °C, with a final step at 72 °C for 10 min to allow for complete extension of all PCR fragments. The PCR product for the E-cadherin promoter region SNP was 448 bp. Two 8-μl products were subjected to digestion at 37 °C overnight in a 10-μl reaction containing 10 U of HincII (SBS Genetech Co., Ltd., Beijing, China) or 10 U of BanII (TakaRa Biotechnology Co., Ltd., Dalian, China), respectively. After digestion, products were separated on a 3% agarose gel stained with ethidium bromide. For the − 160C/A polymorphism, A alleles were represented by DNA bands of 368 bp and 80 bp and C alleles were represented by a DNA band of 448 bp, whereas heterozygotes displayed a combination of both alleles (448 bp, 368 bp and 80 bp) (Fig. 1). For the −347G/GA polymorphism, GA alleles were represented by DNA bands with sizes of 332 bp and 116 bp, and G alleles were represented by a DNA band with sizes of 263 bp, 116 bp and 68 bp, whereas heterozygotes displayed a combination of both alleles (332 bp, 263 bp, 116 bp and 68 bp) (Fig. 2). The PCR product of for the 3′-UTR C/T was 172 bp. After digestion with PmacI (TakaRa Biotechnology Co., Ltd., Dalian, China), the product was separated on a 3% agarose gel stained with ethidium bromide. As a result, a DNA band of 172 bp represented the 3′-UTR T allele and the 3′-UTR C allele was represented by DNA bands of
Fig. 2. E-cadherin − 347 bp G/GA genotyping by PCR–RFLP analysis followed by separation on 3% agarose gel as described in text. Lane M = 100 bp ladder; lanes 3, 5 = G/GA; lanes 2, 6 = C/A; lane 4 = PCR product.
Y. Li et al. / Gynecologic Oncology 108 (2008) 409–414
411
146 bp and 26 bp, whereas heterozygotes displayed a combination of both alleles (172 bp, 146 bp and 26 bp) (Fig. 3). As a negative control, distilled water was used instead of DNA template in the reaction system for each panel of PCR. PCR reactions of the 10% samples were run in duplicate for quality control, with a reproducibility of 100%.
Immunohistochemistry section preparation and pre-treatment Immunohistochemistry was carried out using the avidin–biotin peroxidase complex method. Tissues were sectioned at 4 μm and mounted on APES-treated glass slides (Sigma, St. Louis, MO, USA), deparaffinized and rehydrated using xylene and a series of graded ethanols (100%, 90%, 80% and 70%). Endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide for 20 min at room temperature. For antigen unmasking, sections were immersed in 0.01 M sodium citrate buffer solution (pH 6.0) and heated in an 800-W microwave on full power for four periods of 4 min, pausing to ensure that there was no fluid lost due to evaporation. The slides were then left to cool for 30 min and rinsed in phosphate-buffered saline solution (PBS, pH 6.0).
Antibodies and immunohistochemical staining Non-specific binding was reduced by incubation of the tissue section in normal rabbit serum diluted 1:5 in 12.5% bovine serum albumin (BSA) (Sigma) for 30 min. The slides were then incubated at 4 °C overnight with 80 μl of the monoclonal antibody anti-E-cadherin (5 Ha) 1:100 (Abcam, UK). After washing in PBS for 10 min, the slides were incubated in biotinylated rabbit anti-mouse IgG (Dako, Denmark), diluted 1:200 for 30 min, washed in PBS and then incubated for another 30 min with avidin–biotin-complexed labeling antibody (ABC Complex kit, Dako, Denmark) at room temperature. After washing in PBS, chromogen visualization was achieved by the addition of peroxidase substrate solution [0.05% DAB, 3,3-diaminobenzidine tetrahydrochloride (Sigma) and 0.03% H2O2 in PBS, pH 6.0]. After 7 min, the DAB reaction was stopped under running tap water. Sections were counterstained with modified Mayer's hematoxylin for 30 s and mounted for microscopic examination. To ensure accurate and reproducible staining, normal gastric glandular epithelium was used as a positive control. Strong and homogenous expression of E-cadherin was observed at the cell membrane throughout the epithelium. Normal gastric glandular epithelium without primary antibody was used as a negative control.
Fig. 4. Immunoreactivity of E-cadherin observed in primary ovarian carcinomas. (A) Negative expression of E-cadherin; (B) positive expression of E-cadherin; magnification ×100.
Evaluation and quantification of immunostaining The immunoexpression of tumours was scored semiquantitatively according to the staining pattern (membranous staining) on a three-point scale of − to +++ (−: complete absence of expression; +: ≤10%; ++: N10% and ≤ 50%; +++: N50%). The immunoexpression of E-cadherin in epithelial ovarian cancer is shown in Fig. 4 (negative: A; positive: B).
Statistical analysis Statistical analysis was performed using the SPSS11.5 software package (SPSS Company, Chicago, Illinois, USA). Hardy–Weinberg analysis was
performed to compare the observed and expected genotype frequencies using the chi-square test. Comparison of the E-cadherin −160C/A, −347G/GA and 3′-UTR C/T genotype distribution in the study groups was performed by means of two-sided contingency tables using the chi-square test. Comparison of the protein expression in the E-cadherin 3′-UTR C/T different genotypes was performed by means of two-sided contingency tables using the chi-square test. The E-cadherin − 160C/A and − 347G/GA haplotype frequencies and linkage disequilibrium coefficient were estimated using the EH linkage software (version 1.2, Rockefeller University, New York) and 2LD program. The odds ratio (OR) and 95% confidence interval (CI) were calculated using an unconditional logistic regression model. A probability level of 5% was considered significant.
Results Association of E-cadherin − 160C/A, − 347G/GA, 3′-UTR C/T SNP with susceptibility to epithelial ovarian cancer
Fig. 3. E-cadherin 3′-UTR C/T genotyping by PCR–RFLP analysis followed by separation on 3% agarose gel as described in text. Lane M = 100 bp ladder; lane 2 = T/T; lanes 3, 4 = C/T; lane 1 = PCR product.
The distribution of the E-cadherin − 160C/A, − 347G/GA, 3′-UTR C/T genotypes in the control group did not significantly deviate from that expected for Hardy–Weinberg equilibrium (χ2 = 0.46, 0.13 and 1.52, respectively; all P value N 0.05). There was no significant difference in genotype and allelotype distribution of the E-cadherin − 160C/A between epithelial
412
Y. Li et al. / Gynecologic Oncology 108 (2008) 409–414
Table 2 Distribution of genotypes and alleles of three single nucleotide polymorphisms in E-cadherin
Table 4 Haplotypes of − 160C/A and − 347G/GA SNP in E-cadherin with the risk of ovarian cancer
SNP
Haplotypes
Controls, n (%)
Cases, n (%)
OR (95% CI)
−160C/− 347G −160A/− 347G −160C/− 347GA −160A/− 347GA
278 (54.3) 140 (27.3) 94 (18.4) 0 (0)
246 (59.4) 92 (22.2) 55 (13.3) 21 (5.1)
0.74 (0.54–1.02) 0.66 (0.45–0.96) 48.6 (2.9–806.2)
− 160C/A Genotype
Allelotype − 347G/GA Genotype
Allelotype 3′-UTRC/T Genotype
Allelotype
Genotype/allele
Controls, n (%)
Cases, n (%)
C/C C/A A/A C A
169 (66.0) 80 (31.3) 7 (2.7) 418 (81.6) 94 (18.4)
140 (67.6) 63 (30.4) 4 (2.0) 343 (82.9) 71 (17.1)
G/G G/GA GA/GA G GA
134 (52.4) 104 (40.6) 18 (7.0) 372 (72.7) 140 (27.3)
118 (57.0) 79 (38.2) 10 (4.8) 315 (76.1) 99 (23.9)
C/C C/T T/T C T
129 (50.4) 111 (43.4) 16 (6.2) 369 (72.1) 143 (27.9)
135 (65.2) 60 (29.0) 12 (5.8) 330 (79.7) 84 (20.3)
P value
0.83 0.63
0.46 0.24
0.004 0.007
ovarian cancer patients and controls (Table 2, all P value N 0.05). Compared to the C/C genotype, the ‘A’ allele (C/A + A/ A genotype) were not significantly modified the risk of developing epithelial ovarian cancer; odds ratios were 1.08 (95% CI = 0.73–1.59) (Table 3). The frequencies of the three genotypes and two allelotype of the E-cadherin − 347G/GA in epithelial ovarian cancer patients were not different from controls (Table 2, all P value N0.05). Compared with the G/G genotype, the G/GA + GA/GA genotype did not significantly influence the risk of developing epithelial ovarian cancer. Odds ratios were 1.21 (95% CI = 0.84–1.75) (Table 3). There are significant differences in the genotype and allelotype distribution of the E-cadherin 3′-UTR C/T between epithelial ovarian cancer patients and controls (Table 2, all P value b 0.05). Distribution of the C/C genotype in epithelial ovarian cancer patients (65.2%) was higher than in controls (50.4%). Compared with the T/T + C/T genotype, the C/C genotype significantly increased the risk for development of
Table 3 Association of three single nucleotide polymorphisms in E-cadherin with the risk of ovarian cancer SNP
Genotype
Controls (n)
Cases (n)
− 160C/A
C/C C/A + A/A G/G G/GA + GA/GA T/T + C/T C/C
169 87 134 122 127 129
140 67 118 89 72 135
− 347G/GA 3′-UTR C/T
epithelial ovarian cancer; odds ratios were 1.85 (95% CI = 1.27– 2.69) (Table 3). When stratified for histology types and clinical stage epithelial ovarian cancer, frequencies of the − 160C/A, − 347G/GA and 3′-UTR C/T SNP genotypes among patients were also not significantly different from controls. Haplotype association of E-cadherin − 160C/A and − 347G/GA with susceptibility to epithelial ovarian cancer Results of the 2LD program analysis showed that E-cadherin − 160C/A and − 347G/GA polymorphism were link disequilibrium (D' = 0.999). The − 160C/− 347G is the most common haplotype in control women (54.3%), followed by the − 160A/ − 347G (27.3%) and the −160C/− 347GA (18.4%) haplotype. The − 160A/− 347GA haplotype is only detected in epithelial ovarian cancer patients (5.1%). Compared with the most common haplotype − 160C/− 347G, haplotype − 160A/ − 347GA significantly modified the risk of epithelial ovarian cancer. Odds ratios were 48.6 (95% CI = 2.9–806.2) (Table 4). 3′-UTR C/T genotype-dependent expression of E-cadherin protein in epithelial ovarian cancer The 54 subjects for immunohistochemical staining, 17 carried the C/C genotype, 19 carried the C/T genotype and 18 carried the T/T genotype, were randomly selected from 207 epithelial ovarian cancer patients. Table 5 shows E-cadherin protein expression in ovarian cancer tissue of different genotype carriers. Negative E-cadherin expression (− expression plus + expression) was observed in fifteen (88.2%) cases and positive (++ expression plus +++ expression) was observed in two (11.8%) cases in C/C genotype carriers. Negative expression was observed in twenty-one (56.8%) cases and positive was observed in sixteen (43.2%) case in C/T or T/T genotype carriers. E-cadherin protein expression was significantly lower
OR (95% CI) 1.08 (0.73–1.59) a 1.21 (0.84–1.75) b
Table 5 A 3′-UTR C/T genotype-dependent expression of E-cadherin protein in epithelial ovarian cancer Genotype
n
E-cadherin expression −
+
++
+++
T/T C/T C/C
18 19 17
3 4 7
7 7 8
7 6 2
1 2 0
1.85 (1.27–2.69) c
a
The odds ratio of the C/A + A/A genotype against the C/C genotype for − 160C/A polymorphism. b The odds ratio of the G/GA + GA/GA genotype against the C/C genotype for − 347G/GA polymorphism. c The odds ratio of the C/C genotype against the T/T + C/T genotype for 3′-UTR C/T polymorphism.
P value
0.02 a
a E-cadherin expression of C/C genotype carriers vs. C/T + T/T genotype carriers.
Y. Li et al. / Gynecologic Oncology 108 (2008) 409–414
in ovarian cancer patients with C/C genotype compared to those with C/T or T/T genotypes (χ2 = 5.19, P = 0.02). Discussion In this paper, the study results showed that the E-cadherin 3′-UTR C/T single nucleotide polymorphism was associated with risk of epithelial ovarian cancer. The C/C homozygosity may significantly increase the risk for developing ovarian cancer, and the protein expression of E-cadherin in C/C genotype carriers was lower than that of T allele (C/T or T/T genotype) carriers in ovarian cancer. Although there was no significant difference in genotype distribution of the − 160C/A and of the − 347G/GA SNP in the E-cadherin gene promoter region between ovarian cancer patients and controls, the − 160A/− 347GA haplotype of the E-cadherin gene may enhance the risk of epithelial ovarian cancer. The C/T polymorphism was at nucleotide 2797 of the E-cadherin cDNA, located 54 nucleotides downstream of the TAG stop codon [18]. Although the function of the 3′-UTR + 54C/T polymorphism is unclear, studies have shown that this polymorphism is associated with cancer and other disease. Wu et al. [14] found that the ‘C/C’ homozygote was at a relatively higher risk for developing prostate cancer than other genotypes. Our study also showed that the frequency of the ‘C/C’ homozygote in epithelial ovarian cancer patients (65.2%) was higher than in controls (50.4%), the ‘C/C’ homozygote carriers have a higher incidence of epithelial ovarian cancer (1.85-fold) than individuals with the T allele. As we know, the 3′-UTR is not translated in the protein, but it may affect the relative mRNA stability of gene. Through a study in vitro, Keirsebilck et al. [19] revealed that downregulation of E-cadherin protein expression was caused by mRNA instability triggered by particular 3′-UTR sequences. In order to evaluate the association between genotypes and the expression of E-cadherin, we examined the expression of E-cadherin in ovarian cancer tissue of three genotype carriers by immunohistochemical methods. Results showed that E-cadherin expression in C/C homozygote carriers in ovarian cancer patients was lower than in the other genotypes (C/T or T/T) (P = 0.02), suggesting that the C/C genotype may be associated with downregulation of E-cadherin expression. In fact, there are many examples demonstrating that the 3′-UTR may change the expression of a gene. For example, a 6-bp deletion/insertion polymorphism in the 3′-untranslated region of the TYMS gene affected TYMS mRNA stability and translation [20]. The 3′-UTR polymorphism in the CYP2A6 gene played an important role in CYP2A6 mRNA stabilization and enzyme expression [21]. The − 160C/A and − 347G/GA are two common SNPs upstream of the transcriptional start site of the E-cadherin gene promoter that have a significant effect on transcriptional activity in transient transfection studies [12,13]. Several major cisacting elements have been identified within a short section of the proximal promoter. Among these are two E boxes, a CAAT box and an SP1 binding site [22]. The E-cadherin gene promoter thus exhibits a modular structure, suggesting that the strict control of epithelium-specific E-cadherin expression
413
might result from interactions among various regulatory elements [23]. Therefore, the molecular mechanism of different transcriptional activity may well be explained as the difference in affinity of the DNA-binding protein(s) to the two allelic forms of the E-cadherin promoters. The ‘A’ allele of the − 160C/A polymorphism decreased transcriptional efficiency by 68% compared with the C allele [12]. The ‘GA’ allele of the − 347G/GA polymorphism decreased transcriptional efficiency 10-fold compared with the ‘G’ allele [16]. The − 160C/A and − 347G/GA allelic variation may be a potential genetic marker that can help identify those individuals at higher risk for invasive/metastatic diseases. Recently, studies have investigated the association between the E-cadherin −160C/A polymorphism and the risk of several types of cancer, although results were inconsistent. Verhage et al. [24] found that ‘A’ allele carriers had a higher risk of prostate cancer compared with ‘C’ allele carriers. Jonsson et al. [25], who studied Swedish patients with hereditary prostate cancer, also found a higher risk among ‘A’ allele carriers. However, there is no association between the E-cadherin −160C/A polymorphism and the occurrence or progression of prostate cancer in Japanese populations [26]. Park et al. suggest that the − 160C/A polymorphism of the E-cadherin has no direct effect on the risk of gastric cancer and on its histological classification. However, Wu et al. [27] suggested that individuals with the −160A/A genotype of E-cadherin have a decreased risk of gastric carcinoma. The E-cadherin − 347G/GA polymorphism has been reported to be associated with risk of developing familial gastric cancer and sporadic colorectal cancers (CRC) [28,29]. The GA allele may be a risk factor for certain types of cancer. Our studies showed that the genotype frequencies distribution of − 160C/A and − 347G/GA in the E-cadherin promoter region were not significantly different in ovarian cancer patients and controls. However, a significant linkage disequilibrium between the − 160C/A and − 347G/GA polymorphism, i.e. the − 160C allele tended to be linked to the − 374G allele. Analysis of haplotypes showed that distribution of the four haplotypes was significantly different between epithelial ovarian cancer patients and controls (P b 0.01). The − 160C/− 347G was the most common haplotype in North Chinese women (54.3%); the − 160A/− 347GA haplotype was only detected in ovarian cancer patients (5.1%). Compared with the − 160C/− 347G haplotype, the − 160A/− 347GA haplotype significantly modified the risk of ovarian cancer; the odds ratios were 48.6 (95% CI = 2.9– 806.2). The − 160C/− 374GA haplotype reduced the risk of ovarian cancer; the odds ratios were 0.66 (95% CI = 0.45–0.96). This result suggested that although the − 160A and − 374GA alleles reduced E-cadherin gene transcriptional activity, the allele alone could not increase the risk of ovarian cancer by reducing E-cadherin expression. The − 160A combined with − 374GA might significantly reduce E-cadherin mRNA transcriptional activity and might result in increased risk of ovarian cancer. The low frequency of the − 160C/− 374GA haplotype in ovarian cancer patients might be due to a higher frequency of − 160A/− 374GA in cases. So we thought that the OR result which − 160C/− 374GA haplotype significantly reduced risk of ovarian cancer may not significant.
414
Y. Li et al. / Gynecologic Oncology 108 (2008) 409–414
In conclusion, our studies showed that polymorphism and haplotype of E-cadherin might play an important role in the occurrence of ovarian cancer by regulation of protein expression. The C/C genotype of 3′-UTR C/T SNP and − 160C/ − 374GA haplotype in E-cadherin gene may be a potential susceptibility factor for risk of epithelial ovarian cancer, at least in North Chinese women. To the best of our knowledge, this is the first study to look for an association between polymorphisms in the E-cadherin gene and the risk of developing epithelial ovarian cancer. Acknowledgment This work is supported by funds for the potentially distinguished scientific project construction in program in Hebei Universities. References [1] Sundfeldt K, Piontkewitz Y, Ivarsson K, Nilsson O, Hellberg P, Brannstrom M, et al. E-cadherin expression in human epithelial ovarian cancer and normal ovary. Int J Cancer 1997;74:275–80. [2] Shimoyama Y, Hirohashi S, Hirano S, Noguchi M, Shimosato Y, Takeichi M, et al. Cadherin cell-adhesion molecules in human epithelial tissues and carcinomas. Cancer Res 1989;49:2128–33. [3] Frixen UH, Behrens J, Sachs M, Eberle G, Voss B, Warda A, et al. E-cadherin-mediated cell–cell adhesion prevents invasiveness of human carcinoma cells. J Cell Biol 1991;113:173–85. [4] Frixen UH, Behrens J, Sachs M, Eberle G, Voss B, Warda A, et al. E-cadherin-mediated cell–cell adhesion prevents invasiveness of human carcinoma cells. J Cell Biol 1991;113:173–85. [5] Birchmeier W, Behrens J. Cadherin expression in carcinomas: role in the formation of cell junctions and the prevention of invasiveness. Biochim Biophys Acta 1994;27:11–26. [6] Yoshida R, Kimura N, Harada Y, Ohuchi N. The loss of E-cadherin, alphaand beta-catenin expression is associated with metastasis and poor prognosis in invasive breast cancer. Int J Oncol 2001;18:513–20. [7] Huiping C, Kristjansdottir S, Jonasson JG, Magnusson J, Egilsson V, Ingvarsson S. Alterations of E-cadherin and beta-catenin in gastric cancer. BMC Cancer 2001;1:16. [8] Chan AO. E-cadherin in gastric cancer. World J Gastroenterol 2006;12: 199–203. [9] Park WS, Cho YG, Park JK, Kim CJ, Kim HS, Lee JW, et al. A single nucleotide polymorphism in the E-cadherin gene promoter-160 is not associated with risk of Korean gastric cancer. J Korean Med Sci 2003;18: 501–4. [10] Liu YC, Shen CY, Wu HS. Mechanisms inactivating the gene for E-cadherin in sporadic gastric carcinomas. World J Gastroenterol 2006;12: 2168–73. [11] Caldeira JR, Prando EC, Quevedo FC, Neto FA, Rainho CA, Rogatto SR, et al. CDH1 promoter hypermethylation and E-cadherin protein expression in infiltrating breast cancer. BMC Cancer 2006;2:48.
[12] Li LC, Chui RM, Sasaki M, Nakajima K, Perinchery G, Au HC, et al. A single nucleotide polymorphism in the E-cadherin gene promoter alters transcriptional activities. Cancer Res 2002;60:873–6. [13] Nakamura A, Shimazaki T, Kaneko K, Shibata M, Matsumura T, Nagai M, et al. Characterization of DNA polymorphisms in the E-cadherin gene (CDH1) promoter region. Mutat Res 2002;502:19–24. [14] Wu HC, Lai MT, Wu CI, Chen HY, Wan L, Tsai FJ, et al. Chen WC, E-cadherin gene 3′-UTR C/T polymorphism is associated with prostate cancer. Urol Int 2005;75:350–335. [15] Verhage BA, Van Houwelingen K, Ruijter TE, Kiemeney LA, Schalken JA. Single-nucleotide polymorphism in the E-cadherin gene promoter modifies the risk of prostate cancer. Int J Cancer 2002;100:683–5. [16] Shin Y, Kim LJ, Kang HC, Park JH, Park HR, Park HW, et al. The E-cadherin − 347G → GA promoter polymorphism and its effect on transcriptional regulation. Carcinogenesis 2004;25:895–9. [17] Miller SA, Dybes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acid Res 1988;16:1215. [18] Becker HF, Reich U, Schott C, Hofler H. Single superoxide nucleotide polymorphisms in the human E-cadherin gene. Hum Genet 1995;96:739–40. [19] Keirsebilck A, Van Hoorde L, Gao Y, De Bruyne G, Bruyneel E, Vermassen P, et al. Mechanisms of downregulation of transfected E-cadherin cDNA during formation of invasive tumors in syngeneic mice. Invasion Metastasis 1998;18:44–56. [20] Ulrich CM, Bigler J, Velicer CM, Greene EA, Farin FM, Potter JD, et al. Searching expressed sequence tag databases: discovery and confirmation of a common polymorphism in the thymidylate synthase gene. Cancer Epidemiol Biomarkers Prev 2000;9:1381–5. [21] Wang J, Pitarque M, Ingelman-Sundberg M. 3′-UTR polymorphism in the human CYP2A6 gene affects mRNA stability and enzyme expression. Biochem Biophys Res Commun 2006;10:491–7. [22] Giroldi LA, Bringuier PP, De Weijert M, Jansen C, Van Bokhoven A, Schalken JA, et al. Role of E boxes in the repression of E-cadherin expression. Biochem Biophys Res Commun 1997;241:453–8. [23] Behrens J, Lowrick O, Klein-hitpass L, Birchmeier W. The E-cadherin promoter: functional analysis of a G.C-rich region and an epithelial cell-specific palindromic regulatory element. Proc Natl Acad Sci 1991;88:11495–9. [24] Verhage BA, Van Houwelingen K, Ruijter TE, Kiemeney LA, Schalken JA. Single-nucleotide polymorphism in the E-cadherin gene promoter modifies the risk of prostate cancer. Int J Cancer 2002;100:683–5. [25] Jonsson BA, Adami HO, Hagglund M, Bergh A, Goransson I, Stattin P. −160C/A polymorphism in the E-cadherin gene promoter and risk of hereditary, familial and sporadic prostate cancer. Int J Cancer 2004;109:348–52. [26] Tsukino H, Kuroda Y, Imai H, Nakao H, Qiu D, Komiya Y, et al. Lack of evidence for the association of E-cadherin gene polymorphism with increased risk or progression of prostate cancer. Urol Int 2004;72:203–7. [27] Wu MS, Huang SP, Chang YT, Lin MT, Shun CT, Chang MC, et al. Association of the −160 C → A promoter polymorphism of E-cadherin gene with gastric carcinoma risk. Cancer 2002;94:1443–8. [28] Shin Y, Kim LJ, Kang HC, Park JH, Park HR, Park HW, et al. The E-cadherin − 347G → GA promoter polymorphism and its effect on transcriptional regulation. Carcinogenesis 2004;25:895–9. [29] Shin Y, Kim IJ, Kang HC, Park JH, Park HW, Jang SG, et al. A functional polymorphism (− 347 G → GA) in the E-cadherin gene is associated with colorectal cancer. Carcinogenesis 2004;25:2173–6.
Gynecologic Oncology 108 (2008) 409 – 414 www.elsevier.com/locate/ygyno
E-cadherin gene polymorphisms and haplotype associated with the occurrence of epithelial ovarian cancer in Chinese Yan Li a,⁎, Jun Liang b , Shan Kang b , Zhiming Dong a , Na Wang a , Huimin Xing b , Rongmiao Zhou a , Xiulan Li b , Xiwa Zhao b a
Department of Molecular Biology, Hebei Cancer Institute, Hebei Medical University, Fourth Hospital, Jiankanglu 12, Shijiazhuang 050011, China b Department of Obstetrics and Gynaecology, Hebei Medical University, Fourth Hospital, Shijiazhuang, China Received 2 August 2007 Available online 26 November 2007
Abstract Backgrounds and aims. E-cadherin plays an important role in the origin of epithelial ovarian cancer. However, the exact molecular mechanism by which this occurs is unknown. The polymorphisms located at the E-cadherin may contribute to an increased risk for certain cancers. In this paper, we studied the association between polymorphisms of E-cadherin and the risk of epithelial ovarian cancer. Methods. We assessed the −160C/A, −347G/GA polymorphism within the promoter region and 3′-UTR + 54C/T polymorphism of E-cadherin in epithelial ovarian cancer and control women. We also tested the expression of E-cadherin protein in ovarian cancer tissue among three genotype (3′UTR + 54C/T polymorphism) carriers. Results. There was no significant difference in genotype distribution of the −160C/A and − 347G/GA SNPs in the E-cadherin gene promoter region between ovarian cancer patients and controls, but haplotype − 160A/− 347GA relative to haplotype −160C/−347G was 48.6 (95% CI = 2.9–806.2) for epithelial ovarian cancer risk. The C/C genotype of the 3′-UTR + 54C/T polymorphism relative to the C/T + T/T genotype was 1.85 (95% CI = 1.27–2.69) for epithelial ovarian cancer risk. E-cadherin protein expression in was lower in C/C genotype carriers than T allele carriers in ovarian cancer tissue (P = 0.02). Conclusions. The C/C genotype of 3′-UTR C/T SNP and −160C/−374GA haplotype in E-cadherin gene may be a potential susceptibility factor for risk of epithelial ovarian cancer in Chinese, which indicated that the lower expression of E-cadherin might play an important role in the pathogenesis of epithelial ovarian cancer. © 2007 Elsevier Inc. All rights reserved. Keywords: E-cadherin; Single nucleotide polymorphism; Expression; Epithelium ovarian cancer; Susceptibility
Introduction Ovarian cancer is one of the most prevalent causes of malignant tumors in the female reproductive system. Its incidence rate is third, surpassed only by cervical and uterine cancer. Nevertheless, it is the leading cause of death in all types of gynecological malignancy cancer in women. Epithelial ovarian cancer originates from ovarian surface epithelium and occupies 90% of all ovarian tumors in women. So far, the molecular mechanisms underlying this process are not well
⁎ Corresponding author. Fax: +86 311 86077634. E-mail address: [email protected] (Y. Li). 0090-8258/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2007.10.024
characterized. Recent studies showed that E-cadherin plays an important role in the origin of epithelial ovarian cancer [1]. E-cadherin is a member of the cadherin family of Ca2+dependent cell–cell adhesion molecules and maps to chromosome 16q22.1. Gene products are transmembrane glycoproteins. E-cadherin, which is expressed in almost all epithelial cells [2], mediates homotypic cell-to-cell adhesions and maintains epithelial cell polarity and integrity. E-cadherin is important for regulating epithelial development, maintaining organization and cell integrity. Loss of E-cadherin expression is associated with invasion and dedifferentiation of carcinoma cells [3]. As a metastatic suppressor molecule in carcinomas, the decrease of E-cadherin expression is not only associated with tumour progression and metastases formation in a series of
410
Y. Li et al. / Gynecologic Oncology 108 (2008) 409–414
different cancers [4–7] but also results in the development of cancer by activating the WNT/β-catenin signal transduction pathway and specifically promoting over-expression of oncogenes such as c-myc, cyclinD1 and so on [8]. Aberrant expression of E-cadherin may be due to loss of heterozygosity, gene mutation and methylation of the Ecadherin promoter region [9–11]. Moreover, polymorphisms in the E-cadherin gene might influence transcriptional activity and therefore be associated with risk for a series of different cancers [12–16]. The relationship between E-cadherin gene polymorphisms and risk of ovarian cancer has not been reported. In this paper, we studied the genotype and allele frequencies of three polymorphisms (− 160C/A, − 347G/GA and 3′-UTR +54C/T) within the E-cadherin gene in ovarian cancer patients and healthy controls. To this date, because the function of 3′-UTR + 54C/T polymorphism is unclear, we tested whether the 3′-UTR single nucleotide polymorphism (SNP) genotype modified E-cadherin expression in ovarian cancer tissue of patients. Materials and methods Study participants Two hundred seven patients with ovarian cancer admitted for tumor resection in the Fourth Affiliated Hospital, Hebei Medical University from January 2001 to December 2006 were selected for the study. Epithelial ovarian cancer was identified through histopathologic examination by the Department of Pathology of the same hospital. The demographic and pathologic features of patients were summarized in Table 1. The control group consisted of 256 women without any malignant disease as confirmed by surgical exploration during voluntary abortion, cesarean section, uterine prolapse or pathologically confirmed after hysterectomy for uterine bleeding. All women in the control group had no previous oophorectomy and no history of cancer or genetic disease. The mean age of patients was 53 years (range 20–78 year) and of controls was 50 years (range 21–73 year). There was no statistically significant difference in age distribution between the two groups (P N 0.05). All cancer patients and control subjects were unrelated and of Han nationality from Shijiazhuang, Hebei and the surrounding regions. The Ethics Committee of the Hebei Cancer Institute approved the study and informed consent was obtained from all recruited subjects.
Table 1 Demographic and pathologic features of patients with ovarian cancer Epithelial ovarian cancer Age (years) ≤ 40 41–50 51–60 N60 Tumor histology Serous Mucinous Endometrioid Undifferentiated Tumor stage I/II III/IV
Patients (n = 207) n
%
23 64 76 44
11.1 30.9 36.7 21.3
79 23 77 28
38.2 11.1 37.2 13.5
67 140
32.4 67.6
Fig. 1. E-cadherin − 160 bp C/A genotyping by PCR–RFLP analysis followed by separation on 3% agarose gel as described in text. Lane M = 100 bp ladder; lanes 2, 3, 6 = C/C; lanes 1, 5 = C/A; lanes 4, 7 = A/A.
DNA extraction Venous blood (5 ml) was collected from each subject into Vacutainer tubes containing EDTA and stored at 4 °C. After sampling, genomic DNA was extracted within 1 week by proteinase K (Merck, Darmstadt, Germany) digestion followed by a salting out procedure according to the method published by Miller et al. [17].
E-cadherin − 160C/A, −347G/GA, 3′-UTRC/T genotyping E-cadherin − 160C/A, −347G/GA and +54C/T genotypes were determined by the polymerase chain reaction–restriction fragment length polymorphism (PCR–RFLP) assay. The primers used for amplifying the − 160C/A, − 347G/GA in the E-cadherin promoter fragment were 5′-CGCCCCGACTTGTCTCTCTAC-3′ (forward) and 5′-GGCCA CAGCCAATCAGCA-3′ (reverse) and for amplifying the +54C/T SNP in the E-cadherin 3′untranslated region were 5′-CAGACAAAGACCAGGACTAT-3′ (forward) and 5′-AAGGGAGCTGAAAAACCACCAGC-3′ (reverse). PCR was performed in a 25-μl volume containing 100 ng of DNA template, 2.5 μl of 10× PCR buffer, 1.5 μl of 25 mmol/L MgCl2, 2.5 U of Taq-DNA-polymerase (BioDev-Tech, Beijing, China), 0.5 μl of 10 mmol/L dNTPs and 200 nM of each primer. PCR cycling conditions were 5 min at 94 °C followed by 35 cycles of 30 s at 94 °C, 30 s at 66 °C for C-160A, G-347GA, 30 s at 56 °C for +54C/T and 60 s at 72 °C, with a final step at 72 °C for 10 min to allow for complete extension of all PCR fragments. The PCR product for the E-cadherin promoter region SNP was 448 bp. Two 8-μl products were subjected to digestion at 37 °C overnight in a 10-μl reaction containing 10 U of HincII (SBS Genetech Co., Ltd., Beijing, China) or 10 U of BanII (TakaRa Biotechnology Co., Ltd., Dalian, China), respectively. After digestion, products were separated on a 3% agarose gel stained with ethidium bromide. For the − 160C/A polymorphism, A alleles were represented by DNA bands of 368 bp and 80 bp and C alleles were represented by a DNA band of 448 bp, whereas heterozygotes displayed a combination of both alleles (448 bp, 368 bp and 80 bp) (Fig. 1). For the −347G/GA polymorphism, GA alleles were represented by DNA bands with sizes of 332 bp and 116 bp, and G alleles were represented by a DNA band with sizes of 263 bp, 116 bp and 68 bp, whereas heterozygotes displayed a combination of both alleles (332 bp, 263 bp, 116 bp and 68 bp) (Fig. 2). The PCR product of for the 3′-UTR C/T was 172 bp. After digestion with PmacI (TakaRa Biotechnology Co., Ltd., Dalian, China), the product was separated on a 3% agarose gel stained with ethidium bromide. As a result, a DNA band of 172 bp represented the 3′-UTR T allele and the 3′-UTR C allele was represented by DNA bands of
Fig. 2. E-cadherin − 347 bp G/GA genotyping by PCR–RFLP analysis followed by separation on 3% agarose gel as described in text. Lane M = 100 bp ladder; lanes 3, 5 = G/GA; lanes 2, 6 = C/A; lane 4 = PCR product.
Y. Li et al. / Gynecologic Oncology 108 (2008) 409–414
411
146 bp and 26 bp, whereas heterozygotes displayed a combination of both alleles (172 bp, 146 bp and 26 bp) (Fig. 3). As a negative control, distilled water was used instead of DNA template in the reaction system for each panel of PCR. PCR reactions of the 10% samples were run in duplicate for quality control, with a reproducibility of 100%.
Immunohistochemistry section preparation and pre-treatment Immunohistochemistry was carried out using the avidin–biotin peroxidase complex method. Tissues were sectioned at 4 μm and mounted on APES-treated glass slides (Sigma, St. Louis, MO, USA), deparaffinized and rehydrated using xylene and a series of graded ethanols (100%, 90%, 80% and 70%). Endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide for 20 min at room temperature. For antigen unmasking, sections were immersed in 0.01 M sodium citrate buffer solution (pH 6.0) and heated in an 800-W microwave on full power for four periods of 4 min, pausing to ensure that there was no fluid lost due to evaporation. The slides were then left to cool for 30 min and rinsed in phosphate-buffered saline solution (PBS, pH 6.0).
Antibodies and immunohistochemical staining Non-specific binding was reduced by incubation of the tissue section in normal rabbit serum diluted 1:5 in 12.5% bovine serum albumin (BSA) (Sigma) for 30 min. The slides were then incubated at 4 °C overnight with 80 μl of the monoclonal antibody anti-E-cadherin (5 Ha) 1:100 (Abcam, UK). After washing in PBS for 10 min, the slides were incubated in biotinylated rabbit anti-mouse IgG (Dako, Denmark), diluted 1:200 for 30 min, washed in PBS and then incubated for another 30 min with avidin–biotin-complexed labeling antibody (ABC Complex kit, Dako, Denmark) at room temperature. After washing in PBS, chromogen visualization was achieved by the addition of peroxidase substrate solution [0.05% DAB, 3,3-diaminobenzidine tetrahydrochloride (Sigma) and 0.03% H2O2 in PBS, pH 6.0]. After 7 min, the DAB reaction was stopped under running tap water. Sections were counterstained with modified Mayer's hematoxylin for 30 s and mounted for microscopic examination. To ensure accurate and reproducible staining, normal gastric glandular epithelium was used as a positive control. Strong and homogenous expression of E-cadherin was observed at the cell membrane throughout the epithelium. Normal gastric glandular epithelium without primary antibody was used as a negative control.
Fig. 4. Immunoreactivity of E-cadherin observed in primary ovarian carcinomas. (A) Negative expression of E-cadherin; (B) positive expression of E-cadherin; magnification ×100.
Evaluation and quantification of immunostaining The immunoexpression of tumours was scored semiquantitatively according to the staining pattern (membranous staining) on a three-point scale of − to +++ (−: complete absence of expression; +: ≤10%; ++: N10% and ≤ 50%; +++: N50%). The immunoexpression of E-cadherin in epithelial ovarian cancer is shown in Fig. 4 (negative: A; positive: B).
Statistical analysis Statistical analysis was performed using the SPSS11.5 software package (SPSS Company, Chicago, Illinois, USA). Hardy–Weinberg analysis was
performed to compare the observed and expected genotype frequencies using the chi-square test. Comparison of the E-cadherin −160C/A, −347G/GA and 3′-UTR C/T genotype distribution in the study groups was performed by means of two-sided contingency tables using the chi-square test. Comparison of the protein expression in the E-cadherin 3′-UTR C/T different genotypes was performed by means of two-sided contingency tables using the chi-square test. The E-cadherin − 160C/A and − 347G/GA haplotype frequencies and linkage disequilibrium coefficient were estimated using the EH linkage software (version 1.2, Rockefeller University, New York) and 2LD program. The odds ratio (OR) and 95% confidence interval (CI) were calculated using an unconditional logistic regression model. A probability level of 5% was considered significant.
Results Association of E-cadherin − 160C/A, − 347G/GA, 3′-UTR C/T SNP with susceptibility to epithelial ovarian cancer
Fig. 3. E-cadherin 3′-UTR C/T genotyping by PCR–RFLP analysis followed by separation on 3% agarose gel as described in text. Lane M = 100 bp ladder; lane 2 = T/T; lanes 3, 4 = C/T; lane 1 = PCR product.
The distribution of the E-cadherin − 160C/A, − 347G/GA, 3′-UTR C/T genotypes in the control group did not significantly deviate from that expected for Hardy–Weinberg equilibrium (χ2 = 0.46, 0.13 and 1.52, respectively; all P value N 0.05). There was no significant difference in genotype and allelotype distribution of the E-cadherin − 160C/A between epithelial
412
Y. Li et al. / Gynecologic Oncology 108 (2008) 409–414
Table 2 Distribution of genotypes and alleles of three single nucleotide polymorphisms in E-cadherin
Table 4 Haplotypes of − 160C/A and − 347G/GA SNP in E-cadherin with the risk of ovarian cancer
SNP
Haplotypes
Controls, n (%)
Cases, n (%)
OR (95% CI)
−160C/− 347G −160A/− 347G −160C/− 347GA −160A/− 347GA
278 (54.3) 140 (27.3) 94 (18.4) 0 (0)
246 (59.4) 92 (22.2) 55 (13.3) 21 (5.1)
0.74 (0.54–1.02) 0.66 (0.45–0.96) 48.6 (2.9–806.2)
− 160C/A Genotype
Allelotype − 347G/GA Genotype
Allelotype 3′-UTRC/T Genotype
Allelotype
Genotype/allele
Controls, n (%)
Cases, n (%)
C/C C/A A/A C A
169 (66.0) 80 (31.3) 7 (2.7) 418 (81.6) 94 (18.4)
140 (67.6) 63 (30.4) 4 (2.0) 343 (82.9) 71 (17.1)
G/G G/GA GA/GA G GA
134 (52.4) 104 (40.6) 18 (7.0) 372 (72.7) 140 (27.3)
118 (57.0) 79 (38.2) 10 (4.8) 315 (76.1) 99 (23.9)
C/C C/T T/T C T
129 (50.4) 111 (43.4) 16 (6.2) 369 (72.1) 143 (27.9)
135 (65.2) 60 (29.0) 12 (5.8) 330 (79.7) 84 (20.3)
P value
0.83 0.63
0.46 0.24
0.004 0.007
ovarian cancer patients and controls (Table 2, all P value N 0.05). Compared to the C/C genotype, the ‘A’ allele (C/A + A/ A genotype) were not significantly modified the risk of developing epithelial ovarian cancer; odds ratios were 1.08 (95% CI = 0.73–1.59) (Table 3). The frequencies of the three genotypes and two allelotype of the E-cadherin − 347G/GA in epithelial ovarian cancer patients were not different from controls (Table 2, all P value N0.05). Compared with the G/G genotype, the G/GA + GA/GA genotype did not significantly influence the risk of developing epithelial ovarian cancer. Odds ratios were 1.21 (95% CI = 0.84–1.75) (Table 3). There are significant differences in the genotype and allelotype distribution of the E-cadherin 3′-UTR C/T between epithelial ovarian cancer patients and controls (Table 2, all P value b 0.05). Distribution of the C/C genotype in epithelial ovarian cancer patients (65.2%) was higher than in controls (50.4%). Compared with the T/T + C/T genotype, the C/C genotype significantly increased the risk for development of
Table 3 Association of three single nucleotide polymorphisms in E-cadherin with the risk of ovarian cancer SNP
Genotype
Controls (n)
Cases (n)
− 160C/A
C/C C/A + A/A G/G G/GA + GA/GA T/T + C/T C/C
169 87 134 122 127 129
140 67 118 89 72 135
− 347G/GA 3′-UTR C/T
epithelial ovarian cancer; odds ratios were 1.85 (95% CI = 1.27– 2.69) (Table 3). When stratified for histology types and clinical stage epithelial ovarian cancer, frequencies of the − 160C/A, − 347G/GA and 3′-UTR C/T SNP genotypes among patients were also not significantly different from controls. Haplotype association of E-cadherin − 160C/A and − 347G/GA with susceptibility to epithelial ovarian cancer Results of the 2LD program analysis showed that E-cadherin − 160C/A and − 347G/GA polymorphism were link disequilibrium (D' = 0.999). The − 160C/− 347G is the most common haplotype in control women (54.3%), followed by the − 160A/ − 347G (27.3%) and the −160C/− 347GA (18.4%) haplotype. The − 160A/− 347GA haplotype is only detected in epithelial ovarian cancer patients (5.1%). Compared with the most common haplotype − 160C/− 347G, haplotype − 160A/ − 347GA significantly modified the risk of epithelial ovarian cancer. Odds ratios were 48.6 (95% CI = 2.9–806.2) (Table 4). 3′-UTR C/T genotype-dependent expression of E-cadherin protein in epithelial ovarian cancer The 54 subjects for immunohistochemical staining, 17 carried the C/C genotype, 19 carried the C/T genotype and 18 carried the T/T genotype, were randomly selected from 207 epithelial ovarian cancer patients. Table 5 shows E-cadherin protein expression in ovarian cancer tissue of different genotype carriers. Negative E-cadherin expression (− expression plus + expression) was observed in fifteen (88.2%) cases and positive (++ expression plus +++ expression) was observed in two (11.8%) cases in C/C genotype carriers. Negative expression was observed in twenty-one (56.8%) cases and positive was observed in sixteen (43.2%) case in C/T or T/T genotype carriers. E-cadherin protein expression was significantly lower
OR (95% CI) 1.08 (0.73–1.59) a 1.21 (0.84–1.75) b
Table 5 A 3′-UTR C/T genotype-dependent expression of E-cadherin protein in epithelial ovarian cancer Genotype
n
E-cadherin expression −
+
++
+++
T/T C/T C/C
18 19 17
3 4 7
7 7 8
7 6 2
1 2 0
1.85 (1.27–2.69) c
a
The odds ratio of the C/A + A/A genotype against the C/C genotype for − 160C/A polymorphism. b The odds ratio of the G/GA + GA/GA genotype against the C/C genotype for − 347G/GA polymorphism. c The odds ratio of the C/C genotype against the T/T + C/T genotype for 3′-UTR C/T polymorphism.
P value
0.02 a
a E-cadherin expression of C/C genotype carriers vs. C/T + T/T genotype carriers.
Y. Li et al. / Gynecologic Oncology 108 (2008) 409–414
in ovarian cancer patients with C/C genotype compared to those with C/T or T/T genotypes (χ2 = 5.19, P = 0.02). Discussion In this paper, the study results showed that the E-cadherin 3′-UTR C/T single nucleotide polymorphism was associated with risk of epithelial ovarian cancer. The C/C homozygosity may significantly increase the risk for developing ovarian cancer, and the protein expression of E-cadherin in C/C genotype carriers was lower than that of T allele (C/T or T/T genotype) carriers in ovarian cancer. Although there was no significant difference in genotype distribution of the − 160C/A and of the − 347G/GA SNP in the E-cadherin gene promoter region between ovarian cancer patients and controls, the − 160A/− 347GA haplotype of the E-cadherin gene may enhance the risk of epithelial ovarian cancer. The C/T polymorphism was at nucleotide 2797 of the E-cadherin cDNA, located 54 nucleotides downstream of the TAG stop codon [18]. Although the function of the 3′-UTR + 54C/T polymorphism is unclear, studies have shown that this polymorphism is associated with cancer and other disease. Wu et al. [14] found that the ‘C/C’ homozygote was at a relatively higher risk for developing prostate cancer than other genotypes. Our study also showed that the frequency of the ‘C/C’ homozygote in epithelial ovarian cancer patients (65.2%) was higher than in controls (50.4%), the ‘C/C’ homozygote carriers have a higher incidence of epithelial ovarian cancer (1.85-fold) than individuals with the T allele. As we know, the 3′-UTR is not translated in the protein, but it may affect the relative mRNA stability of gene. Through a study in vitro, Keirsebilck et al. [19] revealed that downregulation of E-cadherin protein expression was caused by mRNA instability triggered by particular 3′-UTR sequences. In order to evaluate the association between genotypes and the expression of E-cadherin, we examined the expression of E-cadherin in ovarian cancer tissue of three genotype carriers by immunohistochemical methods. Results showed that E-cadherin expression in C/C homozygote carriers in ovarian cancer patients was lower than in the other genotypes (C/T or T/T) (P = 0.02), suggesting that the C/C genotype may be associated with downregulation of E-cadherin expression. In fact, there are many examples demonstrating that the 3′-UTR may change the expression of a gene. For example, a 6-bp deletion/insertion polymorphism in the 3′-untranslated region of the TYMS gene affected TYMS mRNA stability and translation [20]. The 3′-UTR polymorphism in the CYP2A6 gene played an important role in CYP2A6 mRNA stabilization and enzyme expression [21]. The − 160C/A and − 347G/GA are two common SNPs upstream of the transcriptional start site of the E-cadherin gene promoter that have a significant effect on transcriptional activity in transient transfection studies [12,13]. Several major cisacting elements have been identified within a short section of the proximal promoter. Among these are two E boxes, a CAAT box and an SP1 binding site [22]. The E-cadherin gene promoter thus exhibits a modular structure, suggesting that the strict control of epithelium-specific E-cadherin expression
413
might result from interactions among various regulatory elements [23]. Therefore, the molecular mechanism of different transcriptional activity may well be explained as the difference in affinity of the DNA-binding protein(s) to the two allelic forms of the E-cadherin promoters. The ‘A’ allele of the − 160C/A polymorphism decreased transcriptional efficiency by 68% compared with the C allele [12]. The ‘GA’ allele of the − 347G/GA polymorphism decreased transcriptional efficiency 10-fold compared with the ‘G’ allele [16]. The − 160C/A and − 347G/GA allelic variation may be a potential genetic marker that can help identify those individuals at higher risk for invasive/metastatic diseases. Recently, studies have investigated the association between the E-cadherin −160C/A polymorphism and the risk of several types of cancer, although results were inconsistent. Verhage et al. [24] found that ‘A’ allele carriers had a higher risk of prostate cancer compared with ‘C’ allele carriers. Jonsson et al. [25], who studied Swedish patients with hereditary prostate cancer, also found a higher risk among ‘A’ allele carriers. However, there is no association between the E-cadherin −160C/A polymorphism and the occurrence or progression of prostate cancer in Japanese populations [26]. Park et al. suggest that the − 160C/A polymorphism of the E-cadherin has no direct effect on the risk of gastric cancer and on its histological classification. However, Wu et al. [27] suggested that individuals with the −160A/A genotype of E-cadherin have a decreased risk of gastric carcinoma. The E-cadherin − 347G/GA polymorphism has been reported to be associated with risk of developing familial gastric cancer and sporadic colorectal cancers (CRC) [28,29]. The GA allele may be a risk factor for certain types of cancer. Our studies showed that the genotype frequencies distribution of − 160C/A and − 347G/GA in the E-cadherin promoter region were not significantly different in ovarian cancer patients and controls. However, a significant linkage disequilibrium between the − 160C/A and − 347G/GA polymorphism, i.e. the − 160C allele tended to be linked to the − 374G allele. Analysis of haplotypes showed that distribution of the four haplotypes was significantly different between epithelial ovarian cancer patients and controls (P b 0.01). The − 160C/− 347G was the most common haplotype in North Chinese women (54.3%); the − 160A/− 347GA haplotype was only detected in ovarian cancer patients (5.1%). Compared with the − 160C/− 347G haplotype, the − 160A/− 347GA haplotype significantly modified the risk of ovarian cancer; the odds ratios were 48.6 (95% CI = 2.9– 806.2). The − 160C/− 374GA haplotype reduced the risk of ovarian cancer; the odds ratios were 0.66 (95% CI = 0.45–0.96). This result suggested that although the − 160A and − 374GA alleles reduced E-cadherin gene transcriptional activity, the allele alone could not increase the risk of ovarian cancer by reducing E-cadherin expression. The − 160A combined with − 374GA might significantly reduce E-cadherin mRNA transcriptional activity and might result in increased risk of ovarian cancer. The low frequency of the − 160C/− 374GA haplotype in ovarian cancer patients might be due to a higher frequency of − 160A/− 374GA in cases. So we thought that the OR result which − 160C/− 374GA haplotype significantly reduced risk of ovarian cancer may not significant.
414
Y. Li et al. / Gynecologic Oncology 108 (2008) 409–414
In conclusion, our studies showed that polymorphism and haplotype of E-cadherin might play an important role in the occurrence of ovarian cancer by regulation of protein expression. The C/C genotype of 3′-UTR C/T SNP and − 160C/ − 374GA haplotype in E-cadherin gene may be a potential susceptibility factor for risk of epithelial ovarian cancer, at least in North Chinese women. To the best of our knowledge, this is the first study to look for an association between polymorphisms in the E-cadherin gene and the risk of developing epithelial ovarian cancer. Acknowledgment This work is supported by funds for the potentially distinguished scientific project construction in program in Hebei Universities. References [1] Sundfeldt K, Piontkewitz Y, Ivarsson K, Nilsson O, Hellberg P, Brannstrom M, et al. E-cadherin expression in human epithelial ovarian cancer and normal ovary. Int J Cancer 1997;74:275–80. [2] Shimoyama Y, Hirohashi S, Hirano S, Noguchi M, Shimosato Y, Takeichi M, et al. Cadherin cell-adhesion molecules in human epithelial tissues and carcinomas. Cancer Res 1989;49:2128–33. [3] Frixen UH, Behrens J, Sachs M, Eberle G, Voss B, Warda A, et al. E-cadherin-mediated cell–cell adhesion prevents invasiveness of human carcinoma cells. J Cell Biol 1991;113:173–85. [4] Frixen UH, Behrens J, Sachs M, Eberle G, Voss B, Warda A, et al. E-cadherin-mediated cell–cell adhesion prevents invasiveness of human carcinoma cells. J Cell Biol 1991;113:173–85. [5] Birchmeier W, Behrens J. Cadherin expression in carcinomas: role in the formation of cell junctions and the prevention of invasiveness. Biochim Biophys Acta 1994;27:11–26. [6] Yoshida R, Kimura N, Harada Y, Ohuchi N. The loss of E-cadherin, alphaand beta-catenin expression is associated with metastasis and poor prognosis in invasive breast cancer. Int J Oncol 2001;18:513–20. [7] Huiping C, Kristjansdottir S, Jonasson JG, Magnusson J, Egilsson V, Ingvarsson S. Alterations of E-cadherin and beta-catenin in gastric cancer. BMC Cancer 2001;1:16. [8] Chan AO. E-cadherin in gastric cancer. World J Gastroenterol 2006;12: 199–203. [9] Park WS, Cho YG, Park JK, Kim CJ, Kim HS, Lee JW, et al. A single nucleotide polymorphism in the E-cadherin gene promoter-160 is not associated with risk of Korean gastric cancer. J Korean Med Sci 2003;18: 501–4. [10] Liu YC, Shen CY, Wu HS. Mechanisms inactivating the gene for E-cadherin in sporadic gastric carcinomas. World J Gastroenterol 2006;12: 2168–73. [11] Caldeira JR, Prando EC, Quevedo FC, Neto FA, Rainho CA, Rogatto SR, et al. CDH1 promoter hypermethylation and E-cadherin protein expression in infiltrating breast cancer. BMC Cancer 2006;2:48.
[12] Li LC, Chui RM, Sasaki M, Nakajima K, Perinchery G, Au HC, et al. A single nucleotide polymorphism in the E-cadherin gene promoter alters transcriptional activities. Cancer Res 2002;60:873–6. [13] Nakamura A, Shimazaki T, Kaneko K, Shibata M, Matsumura T, Nagai M, et al. Characterization of DNA polymorphisms in the E-cadherin gene (CDH1) promoter region. Mutat Res 2002;502:19–24. [14] Wu HC, Lai MT, Wu CI, Chen HY, Wan L, Tsai FJ, et al. Chen WC, E-cadherin gene 3′-UTR C/T polymorphism is associated with prostate cancer. Urol Int 2005;75:350–335. [15] Verhage BA, Van Houwelingen K, Ruijter TE, Kiemeney LA, Schalken JA. Single-nucleotide polymorphism in the E-cadherin gene promoter modifies the risk of prostate cancer. Int J Cancer 2002;100:683–5. [16] Shin Y, Kim LJ, Kang HC, Park JH, Park HR, Park HW, et al. The E-cadherin − 347G → GA promoter polymorphism and its effect on transcriptional regulation. Carcinogenesis 2004;25:895–9. [17] Miller SA, Dybes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acid Res 1988;16:1215. [18] Becker HF, Reich U, Schott C, Hofler H. Single superoxide nucleotide polymorphisms in the human E-cadherin gene. Hum Genet 1995;96:739–40. [19] Keirsebilck A, Van Hoorde L, Gao Y, De Bruyne G, Bruyneel E, Vermassen P, et al. Mechanisms of downregulation of transfected E-cadherin cDNA during formation of invasive tumors in syngeneic mice. Invasion Metastasis 1998;18:44–56. [20] Ulrich CM, Bigler J, Velicer CM, Greene EA, Farin FM, Potter JD, et al. Searching expressed sequence tag databases: discovery and confirmation of a common polymorphism in the thymidylate synthase gene. Cancer Epidemiol Biomarkers Prev 2000;9:1381–5. [21] Wang J, Pitarque M, Ingelman-Sundberg M. 3′-UTR polymorphism in the human CYP2A6 gene affects mRNA stability and enzyme expression. Biochem Biophys Res Commun 2006;10:491–7. [22] Giroldi LA, Bringuier PP, De Weijert M, Jansen C, Van Bokhoven A, Schalken JA, et al. Role of E boxes in the repression of E-cadherin expression. Biochem Biophys Res Commun 1997;241:453–8. [23] Behrens J, Lowrick O, Klein-hitpass L, Birchmeier W. The E-cadherin promoter: functional analysis of a G.C-rich region and an epithelial cell-specific palindromic regulatory element. Proc Natl Acad Sci 1991;88:11495–9. [24] Verhage BA, Van Houwelingen K, Ruijter TE, Kiemeney LA, Schalken JA. Single-nucleotide polymorphism in the E-cadherin gene promoter modifies the risk of prostate cancer. Int J Cancer 2002;100:683–5. [25] Jonsson BA, Adami HO, Hagglund M, Bergh A, Goransson I, Stattin P. −160C/A polymorphism in the E-cadherin gene promoter and risk of hereditary, familial and sporadic prostate cancer. Int J Cancer 2004;109:348–52. [26] Tsukino H, Kuroda Y, Imai H, Nakao H, Qiu D, Komiya Y, et al. Lack of evidence for the association of E-cadherin gene polymorphism with increased risk or progression of prostate cancer. Urol Int 2004;72:203–7. [27] Wu MS, Huang SP, Chang YT, Lin MT, Shun CT, Chang MC, et al. Association of the −160 C → A promoter polymorphism of E-cadherin gene with gastric carcinoma risk. Cancer 2002;94:1443–8. [28] Shin Y, Kim LJ, Kang HC, Park JH, Park HR, Park HW, et al. The E-cadherin − 347G → GA promoter polymorphism and its effect on transcriptional regulation. Carcinogenesis 2004;25:895–9. [29] Shin Y, Kim IJ, Kang HC, Park JH, Park HW, Jang SG, et al. A functional polymorphism (− 347 G → GA) in the E-cadherin gene is associated with colorectal cancer. Carcinogenesis 2004;25:2173–6.