Vitamin D receptor polymorphisms and related biochemical parameters in various cancer species

Introduction

Vitamin D is a steroid hormone that plays a central role in regulating calcium, phosphate and bone metabolism. 25-Hydroxyvitamin D is a preferred biomarker circulating form of vitamin D from diet and sun exposure that undergoes hydroxylation in the liver. It has antitumor effects on cell differentiation induction, apoptosis and inhibits cell proliferation, angiogenesis and metastasis [1]. Up-regulation of adherence and signaling between epithelial cells, contact inhibition of cell cycle stabilization, differentiation, proliferation, promotion of apoptosis, anti-neoangiogenesis are cellular mechanisms for cancer formation that have been determined in vitro conditions. Mechanisms such as down-regulation of glycogen synthase kinase 3 reduce the proliferation of pancreatic, prostate and colorectal cancers. The Wnt/β-catenin signaling pathway plays crucial roles in various biological processes, including cell proliferation and differentiation, regulation of cell cycle, cell-cell and cell-matrix interactions, angiogenesis, apoptosis and adipogenesis. 1,25-dihydroxyvitamin D₃ [1,25(OH)₂D₃] exerts a complex set of regulatory actions leading to the inhibition of the Wnt/β-catenin pathway in line with its protective effect against several types of cancer [2]. Vitamin D deficiency is associated with a number of severe diseases, such as cancer and autoimmune disorders. This association has led the researchers to focus on the vitamin D receptor, its ligand 1,25(OH)₂D₃, and genetic differences and transactions between these molecules [3]. Studies of vitamin D levels and incidences of various cancer species such as breast, pancreatic, ovarian, colon, renal, prostate, bladder, esophageal, skin, hepatocellular cancer have suggested an inverse relationship between them [1]. Also this observation has directed researchers to investigate the biological effects of vitamin D on cancer prevention.

Vitamin D receptor (VDR) is a hormone receptor for 1,25(OH)₂D₃ which acts as a nuclear transcription factor and regulates the biological activities such as cell differentiation and growth by affecting related nucleotide sequences. The VDR gene is known to contain a number of polymorphisms located on chromosome 12q12–q14. The alleles of Fok1 (T/C rs2228570) are located in exon 2 at the 5′-end of the VDR gene and those of Bsm1 (G/A rs1544410) are located in intron 8 at the 3′-end of the VDR gene. The Fok1 restriction fragment length polymorphism generates three amino acids longer VDR protein than Bsm1. Although no significant differences in ligand affinity, DNA binding or transactivation activity have been found between the two allelic forms of Fok1, the shorter restriction fragment length polymorphism variant displays higher receptor activity than the longer one [4].

VDR has a substantial effect for cellular signal mechanisms of vitamin D in the cancer cell development. Since its discovery in 1969, researchers have reported that VDR is related to insulin-like growth factor signaling, inflammation and estrogen, calcium and phosporus metabolism because of its presence in more than 30 tissues or organs in the human body and genetic relationship with the immune system [5].

Various genes and polymorphisms contributing to vitamin D activity have been observed to be associated with 1,25(OH)₂D₃ concentrations. Genetic variants of vitamin D and relation with the risk of various cancers such as pancreas, prostate, breast and colon cancer have been investigated in earlier studies but, incoherent results were reported [6]. However, the long-term serum levels and genetic variants of vitamin D-related genes were related with the risk of several cancer species [7].

In this study, we have hypothesized that lung, colon, breast and pancreatic cancers may be associated with two frequently analyzed VDR gene polymorphisms (Fok1 and Bsm1). We also measured related biochemical parameters such as calcium and phosphorus levels in these patients.

Materials and methods

Results

Mean ages and levels of biochemical parameters for patient and healthy control groups and comparison of parameters were shown in Table 1 with statistical significance values (p-values). Serum calcium levels (8.95±1.41 mg/dL) of lung cancer patients were not significantly different than control group levels (9.27±0.53 mg/dL) (p=0.10). However, serum phosphorus levels of lung cancer patients (3.25±0.70 mg/dL) were found to be significantly lower than those of the control group (3.53±0.57 mg/dL) (p=0.02) and vitamin D levels of the lung cancer patients (10.97±5.60 ng/mL) were significantly decreased when compared with controls (20.94±5.28 ng/mL) (p<0.01). Serum calcium (9.38±0.97 mg/dL) and phosphorus (3.56±0.94 mg/dL) levels of colon cancer patients were not significantly different when compared with control group (9.27±0.53 mg/dL and 3.53±0.57 mg/dL), (p=0.44 and p=0.82), respectively. But vitamin D (10.04±5.04 ng/mL) levels of colon cancer group were significantly lower than controls (20.94±5.28 ng/mL), (p<0.01). Serum calcium (9.21±0.80 mg/dL) and phosphorus (3.44±0.66 mg/dL) levels of the breast cancer group were not significantly different when compared with controls (9.27±0.53 mg/dL and 3.53±0.57 mg/dL) (p=0.64 and p=0.45), respectively. However, vitamin D levels were (9.85±2.71 ng/mL) found to be lower than controls (20.94±5.28 ng/mL) (p<0.01). Serum calcium levels (8.52±0.92 mg/dL) were significantly lower in pancreas cancer patients, than in controls (p=0.01). But phosphorus levels (3.34±0.75 mg/dL) were not significantly different when compared with control group (3.53±0.57 mg/dL), (p=0.14). Serum vitamin D levels were significantly decreased in pancreas cancer group (8.04±6.00 ng/mL) when compared with controls (20.94±5.28 ng/mL) (p<0.01).

Genotype and allele frequencies of the VDR Fok1 gene SNPs (CC, CT or TT) for patients and control subjects are given in Table 2. The CC allele was observed in 30 subjects in the control group (54.5%) and 31 patients in the lung cancer group (52.5%). The CT allele was detected in 23 subjects in the control group (41.8%) and 20 patients in the lung cancer group (33.9%). The TT allele was only present in two subjects in the control group (3.6%) and eight patients in the lung cancer group (13.6%). This study showed that the VDR Fok1 allele frequencies had a significant difference between lung cancer and control group subjects according to the logistic regression model (p=0.042). TT allele frequencies of the VDR Fok1 gene polymorphism were detected to be higher in lung cancer patients compared with CC allele frequencies; results were significantly different at the 95% confidence interval (p=0.03). The CC allele was detected in 30 subjects in the control group (54.5%) and 29 patients in colon cancer group (55.8%). CT allele was observed in 23 subjects in the control group (41.8%) and 19 patients in the colon cancer group (36.5%). The TT allele was observed in two subjects in the control group (3.6%) and four patients in the colon cancer group (7.7%). The CC allele was observed in 30 subjects in the control group (54.5%) and 33 patients in the breast cancer group (64.7%). The CT allele was observed in 23 subjects in the control group (41.8%) and 16 patients in the breast cancer group (31.4%). The TT allele was observed in two subjects in the control group (3.6%) and two patients in the breast cancer group (3.9%). The CC allele was observed in 30 subjects in the control group (54.5%) and 23 patients in the pancreas cancer group in (46.0%). The CT allele was observed in 23 subjects in the control group (41.8%) and 21 patients in the pancreas cancer group (42.0%). The TT allele was observed in two subjects in the control group (3.6%) and six patients in the pancreatic cancer group (12.0%).

Genotype and allele frequencies of VDR Bsm1 gene SNPs (AA, AG or GG) for patients and control subjects are given in Table 2. The AA allele was observed in three subjects in the control group (5.5%) and three patients in the lung cancer group (5.1%). The AG allele was observed in 23 subjects in the control group (41.8%) and 19 patients in the lung cancer group (32.2%). The GG allele was observed in 29 subjects in the control group (52.7%) and 37 patients in the lung cancer group (62.7%). This study showed that the VDR Bsm1 gene polymorphism has a significant difference in the GG/AA allele ratio between lung cancer and control group subjects (p=0.04). The AA allele was observed in three subjects in the control group (5.5%) and two patients in the colon cancer group (3.8%). The AG allele was observed in 23 subjects in the control group (41.8%) and 11 patients in the colon cancer group (21.2%). GG allele was observed in 29 subjects in the control group (52.7%) and 39 patients in the colon cancer group (75.0%). The VDR Bsm1 gene polymorphism allele frequencies were not significantly different between the colon cancer and control group subjects (p>0.05). The AA allele was observed in three subjects in the control group (5.5%) and four patients in the breast cancer group (7.8%). The AG allele was detected in 23 subjects in the control group (23.5%) and 12 patients in the breast cancer group (23.5%). The GG allele was observed in 29 subjects in the control group (52.7%) and 35 patients in the breast cancer group (68.6%). The VDR Bsm1 gene polymorphism had a significant difference in allele frequency between the breast cancer and control group subjects according to the logistic regression model (p=0.038). The VDR Bsm1 gene polymorphism AG/GG allelle ratios were found to be significantly different (p=0.012) and GG allele frequency was higher than AG allelle (p=0.012) in 95% confidence interval in the breast cancer group. The AA allele was observed in three subjects in the control group (5.5%) and three patients in pancreatic cancer group (6.0%). The AG allele was observed in 23 subjects in the control group (41.8%) and 17 patients in the pancreatic cancer group (34.0%). GG allele was observed in 29 subjects in the control group (52.7%) and 30 patients in the pancreatic cancer group (60.0%). No significant difference was observed in the VDR Bsm1 gene polymorphism allele frequencies between pancreatic cancer and control group subjects (p>0.05).

Discussion

Epidemiological studies have shown that there is a relationship between vitamin D status and the risk of various malignencies [10]. It is also associated with autoimmune diseases, inflammatory bowel diseases, rheumatoid arthritis, multiple sclerosis, diabetes, heart and brain diseases, obesity and respiratory diseases. The presence of vitamin D receptors in most tissues and organs such as the ovary, skin, stomach, lung, pancreas, pituitary gland, breast, kidney, thymus, leukocytes and parathyroid gland indicates that vitamin D has crucial roles in the body [3].

According to the International Agency for Research on Cancer, by the year 2025, there will be an estimated 19.3 million new cancer cases and 11.4 million cancer deaths. Lung and prostate cancers are the most common cancers in men, followed by colorectal, stomach, and liver cancers. In terms of mortality in men, lung cancer has the highest rates followed by liver and stomach cancers. Breast cancer is by far the most frequently diagnosed and has the highest mortality in females, followed by colorectal, cervix, and lung cancer [11].

Nakagawa et al. [12] carried out studies on rats and reported that calcitriol is thought to inhibit the metastasis of lung cancer due to antiproliferative effect of vitamin D. People living in northern latitudes have higher mortality from lung cancer probably due to the lower levels of circulating 1,25(OH)₂D₃ resulting from inadequate sun light exposure [13].

In our study, calcium levels were similar for lung cancer patients and controls in the reference range. The data showed that vitamin D and serum phosphorus levels in lung cancer patients were significantly lower (p<0.01 and p=0.02, respectively), than healthy individuals, these results impelled us to consider their possible roles in pathogenesis.

Although there have been many studies on the effect of VDR polymorphisms for risk of developing cancer, there is very little information about prognostic value. The VDR G-T-C (Cdx-2-Fok1-Bsm1) haplotype, which is thought to represent lower receptor activity, is related to worse prognosis in lung cancer [14]. A study in Tunisia showed that, VDR Fok1 and Apa1 polymorphisms might be risk factors for lung cancer development but Bsm1 polymorphism was not a risk factor [15]. According to our findings, lung cancer risk was influenced by VDR Fok1 and Bsm1 polymorphisms. There were associations between Fok1 polymorphisms (p=0.042), VDR Fok1 TT/CC allele frequencies (p=0.03) and VDR Bsm1 GG/AA allele frequencies (p=0.04) when compared with control subjects.

A previous study in Japan reported that higher 1,25(OH)₂D₃ levels were associated with a better survival status in operated colorectal cancer patients, but they reported that 97% of patients had insufficient vitamin D concentrations [16]. On the other hand a meta-analysis has suggested that 1,25(OH)₂D₃ levels were inversely proportional with colorectal adenoma risk [17]. As reported by Rasool et al. [18] the Bsm1 SNPs may lead to 2.7-fold raising of colorectal cancer risk in individuals found homozygous for the presence of the ‘b’ allele, in comparison to subjects homozygous for the ‘B’ allele (Bsm1).

A recent review reported that increasing 1,25(OH)₂D₃ levels to the normal range, by supplementation with an 1000 IU daily dose of vitamin D has provided reduced colorectal cancer risk [19]. In another study on 1248 colorectal cancer patients and 1248 controls the Bsm1 (rs1544410) and the Fok1 (rs2228570) polymorphisms have been investigated. The results of that study showed an inverse correlation between colorectal cancer risk and the BB genotype of the Bsm1 (rs1544410) polymorphism; but, no colorectal cancer risk associations were observed for the Fok1 (rs2228570) [20]. According to our research there was no association between calcium and phosphorus levels or VDR Fok1 and Bsm1 polymorphisms in colon cancer patients, but vitamin D concentrations were significantly lower.

Evidence is accumulating proofs that vitamin D is protective against breast cancer through effects carried out via the VDR. A study from the Roswell Park Cancer Institute examined 579 women with breast cancer and 574 healthy women between 2003 and 2008 and detected a correlation between higher 1,25(OH)₂D₃ serum levels and decreased breast cancer risk [21].

In many meta-analyses a significant association has been declared between VDR Fok1 polymorphisms and most cancer species across multiple ethnic studies. Although the published findings are conflicting, it is usually reported that breast cancer is influenced by some VDR gene polymorphisms [22]. A study has found no significant correlation between the Taq1, Bsm1 and Apa1 variants and the risk of breast cancer. Also, a huge meta-analysis including 23.020 subjects found no association between the Bsm1 polymorphism and breast cancer [23]. Another meta-analysis by Tang et al. suggested that Fok1 could be a sensitive biomarker for breast cancer in the European population. However, dietary factors may alter the effects of VDR genotype for women under low risk of breast cancer who consume higher calcium [24]. On the other hand, according to research by Bretherton-Watt et al. [25] among VDR gene polymorphisms the poly (A) variant was related but, Fok1 was not related with breast cancer risk in the United Kingdom Caucasian population. A previous meta-analysis, after evaluation of 19 SNPs of the VDR, concluded that several polymorphisms may affect breast cancer risk. A poor association was found between breast cancer risk and Fok1, Cdx2, Apa1 and Poly-A, but strong association with Taq1 and Bsm1 polymorphisms [26]. The Bsm1 polymorphism was also related with breast cancer risk in the current study (p=0.038). Another study also supports an increased risk of breast cancer associated with the VDR Bsm1 polymorphism [27].

Vitamin D deficiency is commonly detected in pancreatic adenocarcinoma patients and vitamin D levels may be a prognostic factor for patients with advanced pancreatic adenocarcinoma [28]. Although numerous studies of vitamin D status showed that patients taking vitamin D supplements were protected against the risk of pancreatic cancer with increased vitamin D levels, conversely another case-control study showed a significant three-fold increased risk for pancreatic cancer patients with higher vitamin D levels [29]. VDR gene expression may be a potential prognostic factor for individuals with pancreatic adenocarcinoma and this receptor has been studied widely on pancreatic cancer patients [30]. When the cell lines obtained from these patients were treated with vitamin D analogs at high concentrations, they responded with a reduction in the cell number [31]. Our results suggested that serum calcium and vitamin D levels of pancreatic cancer patients were significantly lower then controls (p=0.01 and p<0.01, respectively). Also, no association has been found between VDR Fok1 and Bsm1 polymorphisms and pancreatic cancer.

It has been reported that VDR polymorphisms were significantly associated with most cancer species such as breast [Fok1, Bsm1, Taq1, Apa1, poly (A)], skin (Fok1, Bsm1, A-1210), colorectal (Fok1, Bsm1), prostate [Fok1, Bsm1, Taq1, poly (A)], bladder (Fok1), ovary (Fok1, Apa1), malignant melanoma (Fok1, Bsm1) and in renal cell carcinoma (Taq1, Apa1). However, the indicated associations between VDR polymorphisms and cancer prognosis were strongest for prostate cancer (Fok1), breast cancer (Bsm1, Taq1), malignant melanoma (Bsm1) and renal cell carcinoma (Taq1) [32]. Thus, a study has suggested that these polymorphisms may alter the polyadenylation of the VDR mRNA transcript, and thereby affects mRNA stability [33]. Certain VDR gene polymorphisms alter the translational process, which may cause vitamin D resistance. Isojima et al. reported that the Fok1 polymorphism (F/f) affects both VDR translation and VDR activation [34].

Serum levels of Vitamin D under 12 ng/mL were defined as hypovitaminosis D, 12–20 ng/mL as deficiency, 20–30 ng/mL as insufficiency and concentrations above 30 ng/mL as sufficent for adults [28]. In our study groups, cancer patients had lower vitamin D levels that are probably associated with a higher incidence of cancer, while the healthy group were in the range of insufficiency, but not hypovitaminosis. We collected samples from indoor patients who were taking prolonged bedrest and not frequently exposed to the sun, in winter months. So, the healthy subjects were affected by the lack of sunlight and had lower vitamin D concentrations than might be expected in summer months. However, different types of dressing may also affect vitamin D levels in individuals.

Limitations of the current study include a lower number of patients, possible different forms of cancer in each organ, the need for a separate female control group for breast cancer, and possible differences in sun exposure and life-style between patients and healthy individuals. Therefore, further studies should be carried out to assess the diagnostic and prognostic value of vitamin D levels and VDR expression.

Conclusion

Although a lot of studies about the protection mechanisms of vitamin D have not clearly demonstrated that it has a role in improvement of malignencies, evidence suggest that vitamin D contributes to cell growth regulation, apoptosis and differentiation. Also, vitamin D has been shown to be related to invasion, angiogenesis and metastasis of cancer cells.

Vitamin D has anti-cancer effects through promotion of differentiation and inhibition of proliferation, inflammation, angiogenesis, etc. Vitamin D expresses its biological activities via vitamin D receptors that are found in many healthy human tissues and also cancer tissues derived from various locations. The VDR regulates both transcriptional responses and post transcriptional mechanisms. The role of vitamin D in calcium and bone metabolism has been proven and its impact on carcinogenesis in many cellular levels such as apoptosis, angiogenesis, proliferation and metastasis has also been demonstrated at the genomic level. We investigated the VDR gene polymorphism Fok1 and Bsm1 and levels of vitamin D and related biochemical parameters such as calcium and phosphorus in lung, colon, breast and pancreatic cancer. In the present study, there were significant valuable results that may contribute to literature and shed light on ongoing researchs.

To conclude, our findings suggest that VDR polymorphisms may modulate the risk of cancer in some tissues and in future studies VDR genetic variations should be integrated with vitamin D status as a prediagnostic indicator. All cancer patients have had lower vitamin D levels than the healthy group.