Variation in the vitamin D receptor gene is not associated with risk of colorectal cancer in the Czech Republic

ORIGINAL RESEARCH

Variation in the Vitamin D Receptor Gene is not Associated

with Risk of Colorectal Cancer in the Czech Republic

David J. Hughes&Ivona Hlavatá&Pavel Soucek&

Barbara Pardini&Alessio Naccarati&

Ludmila Vodickova&Mazda Jenab&Pavel Vodicka

# Springer Science+Business Media, LLC 2010

Abstract

Purpose Increased levels of vitamin D may protect against colorectal cancer (CRC) development and recurrence. Accumulating epidemiologic evidence suggests these effects may be partly mediated by genetic variants of the vitamin D receptor (VDR) proposed to be associated with altered risk of CRC. We wished to determine if common

VDR polymorphisms affected CRC risk in the Czech Republic, a homogenous European population with a high CRC incidence rate.

Methods Frequencies of the common VDR gene poly-morphisms rs2238136, rs1544410 (BsmI), rs7975232 (ApaI), and rs731236 (TaqI) were determined using allele-specific PCR in a case control analysis of a series of 754 CRC patients and 627 patients without malignant disease recruited from centers throughout the Czech Republic. Unconditional logistic regression was used to calculate odds ratios and 95% confidence intervals for the association between these variants and risk of CRC.

Results None of the four polymorphisms tested had any significant effect on CRC risk. No significant differences were observed in susceptibility when the population was stratified by anatomical sub-site, sex, BMI, smoking, alcohol, or presence of polyps.

Conclusions We conclude that common variation in the VDR gene had little effect on its own on predisposition to sporadic CRC in the Czech population.

Keywords colorectal cancer . cancer genetics . association study . vitamin D receptor polymorphisms . cancer prevention

Introduction

Vitamin D is an important factor for calcium homeostasis and modulation of cell cycle kinetics and is thought to protect against colorectal cancer (CRC) in sufficient concentrations (reviewed in [1–4]). The vitamin D receptor (VDR) (OMIM 601769), a member of the steroid nuclear receptor superfamily, is believed to mediate several vitamin D actions by regulation of genes active in intestinal calcium

Electronic supplementary material The online version of this article (doi:10.1007/s12029-010-9168-6) contains supplementary material, which is available to authorized users.

D. J. Hughes

Department of Clinical Medicine, Trinity College Centre for Health Sciences, Adelaide and Meath Hospital,

Dublin 24, Ireland

I. Hlavatá

:

:

Toxicogenomics Unit, National Institute of Public Health, Prague, Czech Republic

I. Hlavatá

:

Third Medical Faculty, Charles University, Prague, Czech Republic

B. Pardini

:

:

:

Academy of Sciences of the Czech Republic, Prague, Czech Republic

M. Jenab

Section of Nutrition and Metabolism, International Agency for Research on Cancer, Lyon, France

D. J. Hughes (*)

Department of Clinical Medicine, Trinity Centre for Health Sciences Building, AMNCH Hospital,

Tallaght,

Dublin 24, Ireland e-mail: hughesd4@tcd.ie DOI 10.1007/s12029-010-9168-6

absorption, inflammation, immune function, insulin-like growth factor-I signaling, and estrogen metabolism [5, 6]. The importance of the VDR in relation to cancer risk has been experimentally demonstrated in vitro and in vivo [5]. The anti-carcinogenic effects of vitamin D could possibly be modified due to inherited variation in the VDR and metabolism of its 1-alpha,25-dihydroxycholecalciferol. Though we lack detailed knowledge of how variation in VDR affects carcinogenesis, several common VDR poly-morphisms have been identified, some of which are thought to be functionally important [7,8].

There has been much published debate examining if variation in the VDR gene modifies cancer risk in various tissues, including CRC (reviewed extensively recently by

[9, 10]). The VDR BsmI (rs1544410), PolyA, and Fok1

(rs10735810) rare allele genotypes were reported to be associated with weak protective effects against CRC [11,

12] in a large North American CRC cohort (N= 2,306

cases and 2,749 controls). In a nested case (N=1,248)–control (N=1,248) study within the European Prospective Investi-gation of Cancer and Nutrition (EPIC) cohort, the VDR BsmI polymorphism but not Fok1 was shown to be associated with a reduced risk of colon cancer [13]. Conversely, a Scottish case (N=2070)–control (N=2793) study saw no association with any of the four VDR variants they examined (BsmI, Cdx2 rs11568820, ApaI rs7975232, Fok1) and CRC risk [14]. The Cdx2 polymorphism was also found to have no effect on cancer risk in two multi-ethnic studies of colon (N=1,574 cases and 1,970 controls) and rectal (N=791 cases and 999 controls) cancer [15].

A recent meta-analysis by Raimondi and colleagues (2009) of 67 reports (14 on CRC) including the two most studied VDR polymorphisms (FokI and BsmI) concluded that they may modulate the risk of cancer of the breast, skin, and prostate and possibly affect cancer risk at other sites, including CRC, in Caucasians [16]. In a review published in 2009 by McCullough and colleagues, they examined associations between poly-morphisms in VDR (Fok1, BsmI, TaqI, ApaI, and Cdx2, poly (A)) and selected genes in the vitamin D pathway in relation to colorectal, breast, and prostate cancer risk. They concluded that there is no strong, consistent epidemiologic evidence for substantial influences of single variants in vitamin D pathway genes on risk for

these cancers [9]. Köstner and colleagues (2009)

reviewed the literature to assess the relevance of these same VDR polymorphisms (plus Bgl1) for cancer risk, concluding that conflicting data have been reported for most malignancies and that there were no clear cut associations with CRC [10].

Therefore, the relation of genetic variation in the VDR gene to CRC risk remains controversial and there is very little data on Eastern European populations. In order to

address this, we genotyped four common VDR polymor-phisms rs2238136; BsmI: rs1544410; ApaI: rs7975232; TaqI: rs731236) in a large CRC case (N=754)–control (N= 627) series from the Czech Republic, a country that currently has the highest CRC mortality rate in the developed world [17].

Materials and Methods Study Population

The studied cohort for genetic analysis comprised 754 cases with CRC and 627 controls with no evidence of colorectal malignancy, all of Czech origin, and aged >29 years. The study cases with histologically confirmed CRC diagnosis were recruited (between September 2004 and February 2006) from patients attending nine oncol-ogy departments throughout the Czech Republic (two in Prague, the others in the towns of Benesov, Brno, Liberec, Ples, Pribram, Usti nad Labem, and Zlin) [18,

19]. During the study period, a total of 968 cases with CRC provided blood samples and 16 individuals were excluded because they met the Amsterdam criteria I and II for hereditary CRC [20]. Controls were selected over the same time period from individuals admitted for colono-scopy in five gastroenterological departments (Prague, Brno, Jihlava, Liberec, and Pribram) due to macroscopic bleeding, positive fecal occult blood test, or abdominal pain of unknown origin and showed negative results for malignancy or idiopathic bowel diseases. Controls had no diagnosis of chronic disease necessitating repeated admit-tance to hospital [18,19]. Among 739 recruited controls, 627 controls were included in the study. In total, 214 cases and 112 controls were excluded because there was incomplete lifestyle and potential risk factor information or biological material was lacking. The participating subjects were properly informed and signed a written consent and approval form for genetic analysis in accord with the Helsinki declaration. The design of the study was approved by the Ethical Committee of the Institute of Experimental Medicine, Prague, Czech Republic. All subjects were interviewed using a structured questionnaire to determine demographic characteristics and potential risk factors for CRC. Study subjects provided information on their living area, education, lifestyle habits (smoking, drinking, diet, etc.), body mass index (BMI), diabetes, family/personal history of cancer, and long-term (at least six consecutive months) drug use. The median age of the cases was 61 (range, 27–85) and of the controls 53 (range, 29–91). The characteristics of the study population have been reported previously [18,19].

SNP Genotyping

Germline DNA was isolated from peripheral leucocytes within 4 weeks after the blood sample was collected using the standard proteinase K digestion, phenol/chloroform extraction, and ethanol precipitation and stored at −80°C. Four SNPs were chosen within the VDR gene based on these criteria: (a) published evidence in previous genetic epidemiology studies, (b) minor allele frequencies (at least 30%) sufficient to provide adequate study power in our sample size to see modest genetic effects, (c) functional relevance, importance, and gene location. The four poly-morphisms we examined included rs2238136 (−4817G→A) which lies in the promoter region upstream of exon 1 in LD block C distinct to the widely examined Fok1 allele located in a 1.3 kb LD breaking spot separating blocks B and C [21]. Thus, rs2238136 is a common independent marker to test for association with the 5′ end of the VDR gene. The

three other variants, rs1544410 (BsmI, G→A), rs7975232

(ApaI, C→A) both in intron 8, and rs731236 in exon 9 (TaqI, T→C, I352I), are located at the 3′ flanking end of the VDR gene in LD block B, in weak LD with rs2238126. They have often been studied previously, and are referred to as ApaI, BsmI, and TaqI, respectively, throughout this report. Genotyping was performed simultaneously for cases and controls by K Biosciences (Hoddesdon, Hertfordshire, EN11 OEX, UK) using a competitive allele-specific PCR system (KASPar). Details of allele probe sequences are available on request. The genotyping assay was validated using a random 10% of samples as duplicate quality controls with complete concordance. The genotyping success rate was 96.5% and samples with unclear or failed genotype calls were excluded from the analysis.

Statistical Analysis

Hardy–Weinberg equilibrium (HWE) was tested for the

VDR polymorphisms in the control subjects using a χ2

test. Age- and sex-adjusted unconditional logistic regression was used to estimate the cancer risk association and 95% confidence intervals (95% CI) for the VDR SNPs (SAS statistical software, version 9, SAS Institute, Cary, NC). For all analyses, the reference genotype was defined as the homozygous (wild type) allele. All main effects analysis models were run for colon and rectum combined (i.e., CRC) as well as separately. Tests for linear trend were performed using a score variable with values from 1 to 3 consistent with the genotype groupings. Data were also analyzed with and without adjustment for age, gender, BMI, diabetes, smoking, and alcohol consumption.

A study power calculation, using the Quanto software

[22], indicated that our study had 80% power (alpha =

0.05) to detect odds ratio (OR) risk reductions of 0.7 and 0.6 in the dominant and recessive models, respectively, for the VDR SNPs with minor allele frequencies of at least 30%.

Results

All of the four tested VDR variants were in HWE for both cases and controls. The VDR rare allele frequencies (rs2238136 =32%, BsmI = 38%, ApaI = 49%, TaqI = 37%) are comparable to other Caucasian populations [23–27].

Association of individual SNPs with CRC risk was assessed by logistic regression. We found no association for any of the four VDR variants rs2238136, BsmI, ApaI, and TaqI with risk of CRC in the 754 cases and 627 controls we examined from the Czech Republic (results

summarized in Table 1). Furthermore, no significant

associations were found when we looked at colon or rectal cancers separately (Electronic supplementary material, Table 1), or sub-group analyses stratified by sex, or other variables such as BMI, smoking, and alcohol (data not shown).

Discussion

Variations in the activity of the VDR could conceivably alter individual responses to Vitamin D and the VDR gene represents a strong candidate gene for CRC susceptibility. However, studies examining the associations between specific VDR polymorphisms and CRC risk have shown conflicting results (reviewed by [9,10,16]).

The aim of the present study was to determine if common VDR polymorphisms increased CRC risks in a homogenous Caucasian European population with a high CRC incidence rate [17]. We tested the association of 4 VDR variants (rs2238136, BsmI, ApaI, and TaqI) with the risk of CRC using a hospital-based case–control design with 754 cases and 627 controls from the Czech Republic. We found no significant association between any of the polymorphisms and risk of CRC for either heterozygotes or rare allele homozygotes (see Table1). We also found no significant differences in our results when we examined cancer site (colon or rectum), sex, or for other major factors in our dataset such as BMI, smoking, and alcohol. We found no deviation from HWE for rs2238136 as previously reported by Xiong et al. (2005) in a study reporting association of VDR genotypes with human height in a familial study of Caucasian Americans [23]. The deviation found by Xiong and colleagues was probably a result of population stratification in the families sampled.

The BsmI, ApaI, and TaqI variants have been frequently examined for associations with cancer risk (reviewed in [9,

10,16]). Two other studies on CRC risk with the rs2238136 variant reported a similar lack of association, a German cohort (256 cases, 256 controls; Nb. in this report

rs2238136 was mislabeled as Cdx2/rs11568820) [24] and

recently a report from the colon cancer family registry [28]. Restricting consideration of the published reports to the better designed studies with an appreciable sample size (>500 cases) strongly supports our conclusion of no marked association for any of these variants and CRC risk [9,10,

16, 28]. The best evidence for a likely association in Western European populations for common variants in VDR and decreased risk of CRC derives from the well designed EPIC cohort study with a large sample size of 1,248 matched case–control sets [13]. This study suggested that the presence of the rare allele of the BsmI polymorphism decreased the risk of colon cancer (RR=0.69, 95% CI= 0.45–0.95; p=0.05) but not rectal cancer (RR=0.97, 95% CI=0.62–1.49). Our odds ratio estimations for the Czechs agree with the absence of an association for rectal cancer

(OR= 1.11, 95% CI = 0.67–1.82), but we also find no

significant risk decrease for colon cancer (OR=0.87, 95% CI=0.59–1.26), although our CIs overlap. However the association for colon cancers reported by EPIC for the rare BsmI genotypes was of borderline significance (A/A cases N=109, p=0.05) and neither of our studies had high power to see a significant effect on colon cancer risk for the rare BsmI genotype. A recent report from the Colon Cancer

Family Registry of a population-based case—unaffected sibling control design of 1,750 sibships found no associa-tion with any of the 43 tagSNPs tested in the VDR gene

(including BsmI) and CRC risk [28]. The summary

estimates from this study are in close agreement to what we report here, and their results were also not significantly affected by tumor location.

Our study limitations include a low statistical power for detecting small effects on CRC risk of the tested VDR variants, and the absence of serum measures of vitamin D (25OHD) for analysis of gene–nutrient inter-actions. However, the EPIC analysis found no significant effects on CRC risk for interactions of VDR BsmI or Fok1

genotypes with serum 25OHD concentration [13]. It is

possible to look at haplotype combinations of the VDR variants to further explore putative CRC risk associations. However, we had little power to adequately examine this, compounded with the high LD with the analyzed SNPs. Indeed, Slattery and colleagues with 2,716 cases and 2,130 controls reported that haplotype variables (for BsmI, poly A, and Fok1) had only slightly better ability to explain case–control differences than genotype variables [25], and other results to date from various populations have been inconsistent [9,10,15,24].

Further studies are needed to increase understanding of VDR-mediated vitamin D function in cancer and to examine the possibility that vitamin D may also have some

important roles independent of the VDR [29] and that

alternative receptors for vitamin D metabolites may exist

VDR Genotype Casesa Controls OR (95% CI)b

Intron1 (−4817A/G) (rs2238136) 732 618 GG 330 289 1.00 GA 328 268 1.00 (0.79–1.26) AA 74 61 1.12 (0.76–1.64) P for trend 0.836 BsmI (rs1544410) 725 614 GG 296 246 1.00 GA 328 275 1.02 (0.80–1.30) AA 101 93 0.93 (0.66–1.30) P for trend 0.851 ApaI (rs7975232) 699 611 CC 201 153 1.00 CA 333 309 0.84 (0.64–1.10) AA 165 149 0.89 (0.65–1.21) P for trend 0.460 TaqI (rs731236) 717 615 TT 298 249 1.00 TC 321 277 1.07 (0.76–1.50) CC 98 89 1.07 (0.76–1.51) P for trend 0.917

Table 1 VDR genotype and risk of colorectal cancer

a_{All case and control entries refer}

to the number of successfully genotyped subjects

b_{All listed odds ratios (OR) refer}

to the adjusted age and sex esti-mations

[30]. New studies are also warranted to estimate interactions of VDR variants with other factors such as calcium and vitamin D intake, 25OHD plasma levels and UV radiation exposure. In addition to comprehensive investigation of polymorphisms in VDR there is a need to examine other vitamin D pathway genes, such as the vitamin D binding protein gene (Gc) and CYP27B1 and CYP24A1, which code for enzymes that metabolize 1-alpha,25-dihydroxycholecalciferol [9]. Recently, Dong and colleagues addressed this lack of knowledge by examining variation in the CYP24A1 and CYP27B1 in a US population-based case–control study of 1,600 cases

and 1,949 controls [31]. The CYP24A1 polymorphism

IVS4-66T>G was found to be associated with risk of CRC, particularly for proximal colon cancer and three CYP24A1 variants were found to modify risk of distal colon cancer. These, along with the spectrum of VDR variation and haplotypes, are strong candidate association studies to be addressed in a much larger sample size, taking into account population heterogeneity and environ-mental exposures where possible.

In summary, we found no association of four common VDR gene variants and risk of CRC, or for colon or rectal cancer separately, in 754 cases and 627 controls from throughout the Czech Republic. Established risk factors postulated to modify cancer risk depending on the VDR genotype such as smoking, alcohol, and obesity did not change the null results. Thus, there seems to be no major effect for common variation in the VDR gene alone on CRC risk in this Czech population.

Acknowledgements This study was supported by the following grants: Czech Science Foundation, no.: GACR 310/07/1430 and GACR 305/09/P194, and Grant Agency of Charles University in Prague, no.: GAUK 15109/2009.

References

1. Giovannucci E. Epidemiological evidence for vitamin D and colorectal cancer. J Bone Miner Res. 2007;22 Suppl 2:V81–5. 2. Garland CF, Gorham ED, Mohr SB, Garland FC. Vitamin D for

cancer prevention: global perspective. Ann Epidemiol. 2009;19:468– 83.

3. Pilz S, Tomaschitz A, Obermayer-Pietsch B, Dobnig H, Pieber TR. Epidemiology of vitamin D insufficiency and cancer mortality. Anticancer Res. 2009;29:3699–704.

4. Rhee HV, Coebergh JW, Vries ED. Sunlight, vitamin D and the prevention of cancer: a systematic review of epidemiological studies. Eur J Cancer Prev. 2009;18(6):458–75.

5. Thorne J, Campbell MJ. The vitamin D receptor in cancer. Proc Nutr Soc. 2008;67:115–27.

6. Uitterlinden AG, Fang Y, van Meurs JB, van Leeuwen H, Pols HA. Vitamin D receptor gene polymorphisms in relation to Vitamin D related disease states. J Steroid Biochem Mol Biol. 2004;89–90:187–93.

7. Uitterlinden AG, Fang Y, Van Meurs JB, Pols HA, Van Leeuwen JP. Genetics and biology of vitamin D receptor polymorphisms. Gene. 2004;338:143–56.

8. Dusso AS, Brown AJ, Slatopolsky E. Vitamin D. Am J Physiol Renal Physiol. 2005;289:F8–28.

9. McCullough ML, Bostick RM, Mayo TL. Vitamin D gene pathway polymorphisms and risk of colorectal, breast, and prostate cancer. Annu Rev Nutr. 2009;29:111–32.

10. Kostner K, Denzer N, Muller CS, Klein R, Tilgen W, Reichrath J. The relevance of vitamin D receptor (VDR) gene polymorphisms for cancer: a review of the literature. Anticancer Res. 2009;29:3511–36.

11. Slattery ML, Neuhausen SL, Hoffman M, Caan B, Curtin K, Ma KN, et al. Dietary calcium, vitamin D, VDR genotypes and colorectal cancer. Int J Cancer. 2004;111:750–6.

12. Slattery ML, Murtaugh M, Caan B, Ma KN, Wolff R, Samowitz W. Associations between BMI, energy intake, energy expenditure, VDR genotype and colon and rectal cancers (United States). Cancer Causes Control. 2004;15:863–72.

13. Jenab M, McKay J, Bueno-de-Mesquita HB, et al. Vitamin D receptor and calcium sensing receptor polymorphisms and the risk of colorectal cancer in European populations. Cancer Epidemiol Biomarkers Prev. 2009;18:2485–91.

14. Theodoratou E, Farrington SM, Tenesa A, McNeill G, Cetnarskyj R, Barnetson RA, et al. Modification of the inverse association between dietary vitamin D intake and colorectal cancer risk by a FokI variant supports a chemoprotective action of Vitamin D intake mediated through VDR binding. Int J Cancer. 2008;123:2170–9.

15. Slattery ML, Herrick J, Wolff RK, Caan BJ, Potter JD, Sweeney C. CDX2 VDR polymorphism and colorectal cancer. Cancer Epidemiol Biomarkers Prev. 2007;16:2752–5.

16. Raimondi S, Johansson H, Maisonneuve P, Gandini S. Review and meta-analysis on vitamin D receptor polymorphisms and cancer risk. Carcinogenesis. 2009;30:1170–80.

17. Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin. 2005;55:74–108.

18. Pardini B, Naccarati A, Novotny J, et al. DNA repair genetic polymorphisms and risk of colorectal cancer in the Czech Republic. Mutat Res. 2008;638:146–53.

19. Polakova V, Pardini B, Naccarati A, et al. Genotype and haplotype analysis of cell cycle genes in sporadic colorectal cancer in the Czech Republic. Hum Mutat. 2009;30:661–8.

20. Vasen HF, Watson P, Mecklin JP, Lynch HT. New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative group on HNPCC. Gastroenterology. 1999;116:1453–6.

21. Nejentsev S, Godfrey L, Snook H, et al. Comparative high-resolution analysis of linkage disequilibrium and tag single nucleotide polymorphisms between populations in the vitamin D receptor gene. Hum Mol Genet. 2004;13:1633–9.

22. Gauderman WJ. Sample size requirements for association studies of gene-gene interaction. Am J Epidemiol. 2002;155:478–84. 23. Xiong DH, Xu FH, Liu PY, Shen H, Long JR, Elze L, et al.

Vitamin D receptor gene polymorphisms are linked to and associated with adult height. J Med Genet. 2005;42:228–34. 24. Flugge J, Krusekopf S, Goldammer M, Osswald E, Terhalle W,

Malzahn U, et al. Vitamin D receptor haplotypes protect against development of colorectal cancer. Eur J Clin Pharmacol. 2007;63:997–1005.

25. Sweeney C, Curtin K, Murtaugh MA, Caan BJ, Potter JD, Slattery ML. Haplotype analysis of common vitamin D receptor variants and colon and rectal cancers. Cancer Epidemiol Biomarkers Prev. 2006;15:744–9.

26. Chen L, Davey Smith G, Evans DM, et al. Genetic variants in the vitamin d receptor are associated with advanced prostate cancer at

diagnosis: findings from the prostate testing for cancer and treatment study and a systematic review. Cancer Epidemiol Biomarkers Prev. 2009;18:2874–81.

27. Curran JE, Vaughan T, Lea RA, Weinstein SR, Morrison NA, Griffiths LR. Association of A vitamin D receptor polymorphism with sporadic breast cancer development. Int J Cancer. 1999;83:723–6. 28. Poynter JN, Jacobs ET, Figueiredo JC, et al. Genetic variation

in the vitamin D receptor (VDR) and the vitamin D-binding protein (GC) and risk for colorectal cancer: results from the

Colon Cancer Family Registry. Cancer Epidemiol Biomarkers Prev. 2010;19:525–36.

29. Mehta RG, Mehta RR. Vitamin D and cancer. J Nutr Biochem. 2002;13:252–64.

30. Khanal R, Nemere I. Membrane receptors for vitamin D metabolites. Crit Rev Eukaryot Gene Expr. 2007;17:31–47. 31. Dong LM, Ulrich CM, Hsu L, et al. Vitamin D related genes,

CYP24A1 and CYP27B1, and colon cancer risk. Cancer Epidemiol Biomarkers Prev. 2009;18:2540–8.