Vitamin D Receptor Gene BsmI and FokI Polymorphisms in Relation to Ovarian Cancer Risk in the Polish Population

Abstract

Background: The role of vitamin D receptor (VDR) single-nucleotide polymorphisms (SNPs) in ovarian cancer has been studied in various populations; however, these results are discordant between different ethnicities. Method: Using the polymerase chain reaction–restriction fragment length polymorphism method, we studied the prevalence of the VDR FokI (rs2228570) and BsmI (rs1544410) SNPs in women with ovarian cancer (n=168) and controls (n=182) in a Polish population. Results: We found a significant contribution of the BsmI SNP Bb+BB-versus-bb dominant inheritance model to ovarian cancer development (p=0.0221, p_corr=0.0442, odds ratio [OR]=1.648 [95% confidence intervals, CI=1.073–2.532]). However, we did not observe an association of the BsmI SNP BB versus Bb+bb recessive inheritance model in patients (p=0.8059, OR=1.093 [95% CI=0.538–2.218]). Moreover, there was no association of FokI SNPs either in Ff+ff versus FF dominant or ff versus Ff+FF recessive inheritance models with ovarian cancer development (p=0.9924, OR=1.002 [95% CI=0.628–1.599] and p=0.1123, OR=1.542 [95% CI=0.901–2.638], respectively). The p-values of the trend test observed for the VDR BsmI and FokI SNPs in patients with ovarian cancer were p_trend=0.0613 and p_trend=0.3655, respectively. Conclusion: Our study indicates that the VDR B gene variant might be a moderate risk factor of ovarian cancer development in the Polish population.

Introduction

Ovarian cancer is the one of the most lethal gynecological malignancies in developed countries, with 225,500 new cases and 140,200 estimated deaths annually worldwide (Jemal et al., 2011). Ovarian cancer may develop in different parts of the ovary; however, ∼90% of malignant ovarian tumors arise from ovarian epithelium (Romero and Bast, 2012). Risk factors for ovarian cancer include advancing age, infertility, inflammation, environmental factors, positive family history of ovarian, uterine, breast, or colon tumors associated with mutations of BRCA1 or BRCA2, mismatch repair genes, or TP53 (Sueblinvong and Carney, 2009). Risk is also related to the number of ovulatory cycles and is halved in women using oral contraceptives, those with greater parity, or those who breast-fed long-term (Brekelmans, 2003; Sueblinvong and Carney, 2009; Romero and Bast, 2012).

Approximately 85% of ovarian cancer cases are sporadic, and 15% are familial, suggesting a significant role in the interaction between genetic factors and environmental exposure (Romero and Bast, 2012). The environmental factors may include diet, lifestyle, and exposure to chemicals or other toxins (Brekelmans, 2003). Recently, some studies have suggested the possible role of vitamin D in the development of cancers, including ovarian carcinogenesis (Grant, 2012).

The active form of vitamin D, 1,25-dihydroxyvitamin D3 [1,25(OH)₂D₃) transduces signals to target cells exploiting the vitamin D receptor (VDR) (Miyamoto et al., 1997).VDR forms heterodimers with the related retinoid X receptors and initiates the transcription of various genes (Zhang et al., 2011). It has been demonstrated that1,25(OH)₂D₃ is able to regulate the expression of tumor-related genes, mediating inhibition of growth of ovarian cancer cells (Zhang et al., 2006). This may suggest that genetic variants of the VDR gene modulating the action of 1,25(OH)₂D₃ may play a significant role in ovarian tumorigenesis. The VDR gene is situated on chromosome 12q, and some of the variants of this gene may affect the function of 1,25(OH)₂D₃ in target cells (Uitterlinden et al., 2004). The most frequently studied VDR single-nucleotide polymorphisms (SNPs) include rs10735810/rs2228570 (FokI) and three other SNPs, namely rs1544410 (BsmI), rs731236 (TaqI), and rs7975232 (ApaI), situated in the same linkage disequilibrium (LD) block (Uitterlinden et al., 2004). In recent years, the relevance of VDR polymorphisms as risk factors for the development of various types of cancer, including ovarian cancer, has been evaluated in numerous population studies (Lurie et al., 2007, 2011; Clendenen et al., 2008; Köstner et al., 2009; Tamez et al., 2009; Tworoger et al., 2009). However, these studies have demonstrated variable and inconsistent results (Lurie et al., 2007, 2011; Clendenen et al., 2008; Tamez et al., 2009; Tworoger et al., 2009). Therefore, we aimed to study whether the FokI and BsmI SNPs situated in the VDR gene can be a genetic risk factor of ovarian cancer in the Polish population.

Materials and Methods

Patients and controls

The patient group is composed of 168 women with histologically recognized ovarian carcinoma according to the International Federation of Gynecology and Obstetrics (FIGO). Histopathological classification, including the stage, grade, and tumor type, was performed by an experienced pathologist (Table 1). The controls encompassed 182 unrelated healthy female volunteers who were matched by age to the cancer patients (Table 1). Written informed consent was obtained from all participating individuals. The procedures of the study were approved by the Local Ethics Committee of the Poznan University of Medical Sciences. Patients and controls were Caucasian from the Wielkopolska area of Poland.

Table 1.

Clinical Characteristics of Ovarian Cancer Patients and Healthy Controls

Characteristic	Patients (n=168)	Controls (n=182)
Mean age±SD	55.6±9.6	54.7±8.7
Histological grade
G1	31 (18.4%)
G2	48 (28.6%)
G3	46 (27.4%)
Gx	43 (25.6%)
Clinical stage
I	46 (27.4%)
II	37 (22.0%)
III	58 (34.5%)
IV	27 (16.1%)
Histological type
Serous	46 (27.4%)
Mucinous	29 (17.3%)
Endometrioid	50 (29.8%)
Clear cell	19 (11.3%)
Brenne	3 (1.8%)
Mixed	11 (6.5%)
Untyped carcinoma	10 (5.9%)

Genotyping

Genomic DNA was isolated from peripheral blood leukocytes by salt extraction. DNA samples were genotyped for the BsmI (rs1544410) and FokI (rs2228570) VDR SNPs (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/gtmb). Genotyping of the BsmI and FokI VDR variants was conducted by polymerase chain reaction (PCR), followed by the appropriate restriction enzyme digestion (PCR–restriction fragment length polymorphism [RFLP]) according to the manufacturer's instructions (Fermentas, Vilnius, Lithuania). DNA fragments were separated in 2% agarose gels and visualized by ethidium bromide staining. Primer sequences and conditions for PCR-RFLP analyses are presented in Supplementary Table S2.

Statistical analysis

For each SNP, the Hardy–Weinberg equilibrium (HWE) was assessed by Pearson's goodness-of-fit chi-square (χ²) statistic. The differences in the allele and genotype frequencies between cases and controls were determined using χ² tests. SNPs were tested for association with ovarian cancer using the Cochran-Armitage trend test. The odds ratio (OR) and associated 95% confidence intervals (95% CI) were also calculated. The data were analyzed under recessive and dominant inheritance models. To adjust for the multiple testing, we employed a Bonferroni correction. Pairwise LD between selected SNPs was computed as both D′ and r² values using HaploView 4.0 software (Broad Institute, Cambridge, MA). HaploView 4.0. software was also used for haplotype analysis. Only haplotypes with a frequency above 0.01 were analyzed. Statistical significance was assessed using the 10,000-fold permutation test.

Results

Association of the VDR BsmI and FokI SNPs with ovarian cancer development

The frequency of all studied genotypes did not exhibit deviation from the HWE between all investigated groups (p>0.05). The number of genotypes, OR, and 95% CI evaluation for the VDR BsmI and FokI polymorphisms are presented in Table 2. The p-values of the trend test observed for the VDR BsmI and FokI SNPs were p_trend=0.0613 and p_trend=0.3655, respectively (Table 2). We found a significant contribution of BsmI SNP in a dominant inheritance model to ovarian cancer development (p=0.0221, p_corr=0.0442, OR=1.648 [95% CI=1.073–2.532]) (Table 2). However, we did not observe an association of BsmI SNP in a recessive inheritance model (p=0.8059, OR=1.093 [95% CI=0.538–2.218]). Moreover, there was no association of the FokI SNP, either in dominant or recessive inheritance models, with ovarian cancer development (p=0.9924, OR=1.002 [95% CI=0.628–1.599] and p=0.1123, OR=1.542 [95% CI=0.901–2.638], respectively) (Table 2).

Table 2.

Association of Polymorphisms of the Vitamin D Receptor Gene with Ovarian Cancer

			Genotypec
rs no.	Allelesa	MAFb	Cases	Controls	ORdominant (95% CI); p-valued	ORrecessive (95% CI); p-valued	ptrendvaluee
rs1544410	A/G (B/b)	0.31	60/91/17	87/78/17	AG+AA vs. GG (Bb+BB vs. bb)	AA vs. AG+GG (BB vs. Bb+bb)
					1.648 (1.073–2.532); 0.0221	1.093 (0.538–2.218); p=0.8059	0.0613
rs2228570	C/T (F/f)	0.44	47/83/38	51/102/29	CT+TT vs. CC (Ff+ff vs. FF)	TT vs. CT+CC (ff vs. Ff+FF)
					1.002 (0.628–1.599); 0.9924	1.542 (0.901–2.638); 0.1123	0.3555

Bold values are significant p<0.05.

^aUnderline denotes the minor allele in the control samples.

^bMAF, minor allele frequency calculated from the control samples.

^cThe order of genotypes: DD/Dd/dd (d is the minor allele).

^dChi-square analysis.

^eCochran-Armitage trend test.

CI, confidence interval; OR, odds ratio.

Association of the VDR BsmI and FokI haplotypes with ovarian cancer development

Haplotype analysis of VDR BsmI and FokI polymorphisms did not reveal statistically significant differences in haplotype frequencies between cases and controls. The lowest p-value (p=0.0910, p_corr=0.1736) was observed for haplotypes composed of the Bf alleles (Table 3). However, these results were not statistically significant when permutations were used to generate empiric p-values. The VDR BsmI and FokI SNPs were in weak pairwise LD. This was calculated from the control samples, and had D′ ranges of 0.004–0.106 for the VDR BsmI and FokI SNPs (Supplementary Table S3).

Table 3.

Analysis of Haplotypes in the VDR Gene and the Risk of Ovarian Cancer

Polymorphisms	Haplotypes	Frequency	Case, control ratios	χ2	p-Value	pcorrvaluea
rs1544410_ rs2228570	GC (bF)	0.353	0.327, 0.378	1.995	0.1578	0.3149
	GT (bf)	0.307	0.301, 0.313	0.108	0.7422	0.9756
	AC (BF)	0.191	0.200, 0.183	0.331	0.5650	0.8896
	AT (Bf)	0.148	0.172, 0.126	2.857	0.0910	0.1736

^ap-Value calculated using permutation test and a total of 10,000 permutations.

VDR, Vitamin D receptor.

Discussion

The basic role of 1,25(OH)₂D₃ is the homeostasis of calcium and maintenance of proper bone mineral densities (Uitterlinden et al., 2004; Maruotti and Cantatore, 2010). The VDR has been found in various tissues, including cardiac myocytes, β-cells in the pancreas, immune cells, and other peripheral tissues that indicate the broad role of 1,25(OH)₂D₃ in human physiology, cardioprotection, immune system regulation, and cancer prevention (Uitterlinden et al., 2004; Maruotti and Cantatore, 2010). Moreover, the VDR has been also identified in different malignancies, which may draw attention to the role of vitamin D in growth control of various cancer cells (Wolden-Kirk et al., 2012). Recently, it has been demonstrated that 1,25(OH)₂D₃ suppresses motility, invasion, and metastasis of squamous cell carcinoma in the murine model (Ma et al., 2012). In addition to these findings, Swami et al. (2012) observed that dietary vitamin D₃ and 1,25(OH)₂D₃ displayed equivalent anticancer activity in murine xenograft models of breast and prostate cancer. Preclinical studies conducted by Zhang et al. (2006) demonstrated that the synthetic vitamin D analog EB1089 increased the apoptotic rate and decreased cell proliferation in human ovarian tumor xenografts in mice.

Vitamin D deficiency has been observed in patients with various cancer types and has been correlated with advanced stage of disease (Churilla et al., 2012). Reduced vitamin D status has been associated with either poor prognosis or development of lung, thyroid, breast, gastric, colon, and head and neck cancers (Cheng and Neuhouser, 2012; Imtiaz et al., 2012; Orell-Kotikangas et al., 2012; Pereira et al., 2012; Roskies et al., 2012; Ren et al., 2012). Decreased levels of vitamin D have also been found in patients with ovarian cancer as compared to the general population (Bakhru et al., 2010). Moreover, intake of total vitamin D was inversely associated with the risk of developing serous borderline and mucinous histological subtypes of ovarian tumors (Merritt et al., 2012). This may indicate that interaction between vitamin D levels and genetic variants of VDR may play a pivotal role in ovarian carcinogenesis.

We found a moderate association of the BsmI VDR B gene variant with ovarian cancer in the Polish population. There was no BsmI VDR polymorphism contribution to ovarian cancer development in the study conducted by Clendenen et al. (2008) in a Caucasian population (Clendenen et al., 2008). Moreover, the BsmI VDR SNP was not associated with ovarian cancer development in cohorts from Massachusetts and New Hampshire in the United States (Tworoger et al., 2009).

However, we did not observe significant differences in the distribution of the VDR FokI polymorphism between patients with ovarian cancer and controls. Our findings are consistent with the study results of Clendenen et al. (2008), who did not find that the FokI SNP is a risk of ovarian cancer in a Caucasian population (Clendenen et al., 2008). However, in some studies, the FokI polymorphism has been recognized as a risk factor of ovarian cancer (Lurie et al., 2007, 2011; Tworoger et al., 2009). Tworoger et al. (2009) observed that the TT (ff) genotype of the FokI SNP was significantly associated with ovarian cancer risk in a population studied in Massachusetts and New Hampshire in the United States (Tworoger et al., 2009). Moreover, Caucasian women, but not Japanese women, having the heterozygous FokI T (f) allele were also at increased risk for ovarian carcinoma compared with the homozygous CC (FF) carriers (Lurie et al., 2007). A recently conducted pooled analysis of five studies, including 1820 white non-Hispanic cases and 3479 controls, confirmed that the FokI T (f) allele is associated with ovarian cancer (Lurie et al., 2011). In addition, Tamez et al. (2009) demonstrated that, in a Japanese population, the FokI CC (FF) genotype was associated with better survival compared to the CT (Ff) or TT (ff) genotypes (Tamez et al., 2009).

The differences on the effect of the BsmI and FokI VDR SNPs on the increased risk of ovarian cancer development in various studies may be due to exposure of the analyzed groups to disparate environmental factors, group size, and genetic heterogeneity. Moreover, these differences might also result from confounding factors that are taken into consideration in these studies and include food habits, age, education, menopausal status, parity, body–mass index, use of oral contraceptives or hormone replacement therapy, tubal ligation, and others (Lurie et al., 2007, 2011; Tworoger et al., 2009).

The BsmI or FokI SNPs in VDR have been associated with breast, bladder, renal, carcinoma, and cutaneous malignant melanoma and nonmelanoma skin cancer (Mittal et al., 2007; Gandini et al., 2009; Tang et al., 2009; Arjumand et al., 2012). These polymorphisms were also recognized as risk factors of hepatocellular, head and neck, thyroid, prostate, and colorectal cancers (Liu et al., 2005; Mishra et al., 2005; Bai et al., 2009, 2012; Penna-Martinez et al., 2009; Falleti et al., 2010).

The effects of BsmI and FokI SNPs on the VDR protein function have already been studied (Arai et al., 1997; Luo et al., 2012; Monticielo et al., 2012). The FokI SNP situated in exon 2 results in the formation of a second methionine start site that leads to the production of a shorter protein receptor (Arai et al., 1997). This VDR isoform displayed higher transcriptional activity than the longer-type receptor (Arai et al., 1997). The role of the FokI SNP has recently been reinforced by Monticielo et al. (2012). They found that vitamin D concentration was remarkably elevated in bearers of the TT (ff) genotype versus individuals having the CC (FF) genotype (Monticielo et al., 2012). The BsmI SNP may be associated with the different length polyadenylate sequence within the 3′-untranslated region of the VDR gene (Uitterlinden et al., 2004). Recently, studies conducted by Luo et al. (2012) have indicated that the level of VDR mRNA was remarkably reduced in patients with the VDR BsmI A (B) allele versus individuals bearing the GG (bb) genotype (Luo et al., 2012).

Our study demonstrates that the BsmI, but not the FokI, VDR SNP is a risk factor of ovarian cancer in the Polish population. However, this evaluation was conducted in a small size group, and this study should be replicated in other independent cohorts.

Acknowledgments

Supported by grant no 502-01-01124182-07474, Poznań University of Medical Sciences. The technical assistance of Ms. Sylwia Matuszewska and Joanna Jurkowlaniec-Salazar.

Author Disclosure Statement

No competing financial interests exist.