Cancer Genetics and Cytogenetics 190 (2009) 71e74

A functional promoter polymorphism in the TERT gene does not affectinherited susceptibility to breast cancer

Verena Varadia,*, Annika Brendlea, Ewa Grzybowskab, Robert Johanssonc, Kerstin Enquistd,Dorota Butkiewicze, Jolanta Pamula-Pilatb, Wioletta Pekalab, Kari Hemminkia,f, Per Lennerc,

Asta Forstia,f

aDivision of Molecular Genetic Epidemiology, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 580, D-69120 Heidelberg, GermanybDepartment of Molecular Biology, Center of Oncology, Maria Sklodowska-Curie Institute, Gliwice, Poland

cDepartment of Oncology, Norrlands University Hospital, Umea, SwedendDepartment of Public Health and Clinical Medicine, Nutritional Research, Umea University, Umea, Sweden

eDepartment of Tumor Biology, Center of Oncology, Maria Sklodowska-Curie Institute, Gliwice, PolandfCenter for Family and Community Medicine, Karolinska Institute, Huddinge, Sweden

Received 2 September 2008; accepted 14 November 2008

Abstract Telomere dysfunction is a key mechanism in c

* Corresponding a

E-mail address: v

0165-4608/09/$ e see

doi:10.1016/j.cancerg

ancer development. The human telomerase reversetranscriptase (TERT) is the rate-limiting catalytic subunit of the telomerase enzyme, which is neces-sary for the maintenance of telomere DNA length, chromosomal stability, and cellular immortality.In our attempt to identify functional polymorphisms in the TERT gene and their effect on breastcancer risk, we sequenced the promoter of the gene and identified three single nucleotide polymor-phisms (SNPs) with a frequency of at least 10%. One of these SNPs, rs2853669 (d244 T O C), hasbeen shown to affect telomerase activity and telomere length. Recently, this SNP has been sug-gested to affect familial breast cancer risk. In our caseecontrol study using two large breast cancersample series, including one with 841 cases with inherited susceptibility to breast cancer, we did notfind any association with familial or sporadic breast cancer risk. This well-powered study excludesan effect of the functional d244 T O C SNP and two other correlated SNPs on breast cancerrisk. � 2009 Elsevier Inc. All rights reserved.

1. Introduction

Recently, Savage et al. [1] reported an association betweenthree correlated single nucleotide polymorphisms (SNPs) inthe human telomerase reverse transcriptase (TERT ) geneand familial breast cancer risk as part of a large population-based caseecontrol study. One of the SNPs, rs2853669(d244 T O C), is located at a binding site for ETS2, whichhas been suggested to be a positive regulator of the TERT geneand to be required for telomerase activation during tumorprogression and cellular immortalization [2,3]. A decreasedtelomerase activity and an increased tumor-to-normal tissueterminal restriction fragment ratio have been related to theC allele in a recent study on non-small cell lung cancer [3].

TERT is the rate-limiting catalytic subunit of the telome-rase enzyme, which is necessary for the maintenance oftelomere DNA length, chromosomal stability, and cellular

uthor: Tel.: þ49 6221 42 1811; fax: þ49 6221 1810.

.varadi@dkfz-heidelberg.de (V. Varadi).

front matter � 2009 Elsevier Inc. All rights reserved.

encyto.2008.12.006

immortality [4e6]. It is normally not expressed in mosthuman somatic cells; however, in approximately 85e90%of human cancers telomerase is reactivated, most likelyby transcriptional regulation [4e6]. Thus, polymorphismslocated within transcription factor binding sites in thepromoter are good candidates for low-penetrance cancersusceptibility alleles. Additionally, nonsynonymous SNPs(nsSNPs) can affect the function of the coded protein dueto the change of an amino acid. However, genetic variationin the TERT gene is very limited [7] and only two nsSNPswith allele frequencies !2.5% have been reported in theEuropean population in the NCBI SNP database (http://www.ncbi.nlm.nih.gov/SNP/).

In our investigation of the effect of common, potentiallyfunctional polymorphisms in the TERT gene on breastcancer risk, we focused on the promoter of the gene. Weidentified three relatively highly correlated SNPs with anallele frequency of >10%. Because the SNP rs2853669(d244 T O C) was the most promising candidate fora functional SNP, we investigated whether it affects the risk

72 V. Varadi et al. / Cancer Genetics and Cytogenetics 190 (2009) 71e74

of breast cancer and especially whether it affects inheritedbreast cancer susceptibility.

We conducted a caseecontrol study using two differentsample sets, the first consisting of 841 breast cancersamples from Polish women with familial or early onsetof breast cancer together with 460 matched controls andthe second consisting of 815 Swedish incident breast cancercases together with 1559 controls.

2. Materials and methods

2.1. Study population

The analyses were performed on genomic DNA fromtwo different study populations. The first population included841 Polish breast cancer cases which were selected based onthe criteria used in Poland for BRCA1/2 mutation screening.The inclusion criteria were (i) at least two first-degree rela-tives with breast and/or ovarian cancer regardless of age,(ii) breast cancer diagnosed below the age of 35 withoutfamily history, (iii) bilateral breast cancer regardless of thefamily history, (iv) breast and ovarian cancer diagnosed inone patient regardless of the family history and (v) breastcancer diagnosed below 50 years of age regardless of familyhistory. The cases were collected through the ChemotherapyClinics, the Genetic Counseling Service and the SurgeryClinics (Gliwice, Poland) between the years 1997e2006.The mean age of the cases was 46 years (range, 17e81 years),and that of the 460 regionally and ethnically matchedcontrols was 40 years (range, 16e76 years). Further informa-tion can be found in Jin et al. [8].

The second population comprised 815 Swedish breastcancer cases and 1,559 controls. Of these, 782 cases withage- and sex-matched controls were drawn from the popula-tion-based Vasterbotten intervention project and themammary screening project, which contain blood samplescollected between January 1990 and January 2001 from anethnically homogenous population living in a geographicallydefined region in the north of Sweden [9]. Prospective caseswere identified from the cohorts by record linkage to theregional cancer registry. The controls were selected fromthe same cohort as the corresponding case. They werematched with the case by age at baseline (66 months) andthe time of sampling (62 months). The controls had to be aliveat the time of diagnosis of the corresponding case and withoutany previous cancer diagnosis, except carcinoma in situ ofcervix uteri. For this prospective group, 33 samples werecollected consecutively during the same time period fromuntreated patients referred to the Department of Oncologyfor newly diagnosed breast cancer. Their controls, alsomatched for age and sex, were selected from the Vasterbottenintervention cohort. The mean age of the cases was 58.0 years(range, 30.6e86.2 years). All participants gave informedconsent to the use of their samples for research purpose.The study was approved by the ethical committee of

Karolinska Institute, Syd and Umed University. Further infor-mation can be found in Brendle et al. [10].

2.2. SNP screening by sequencing

We screened the promoter (up to d1800 bp relative to tran-scription start site) of the TERT gene for SNPs in a set of 32randomly chosen breast cancer samples by sequencing. Poly-merase chain reaction (PCR) was performed with 5 ng DNA ina 10-mL reaction volume as described earlier by Vaclaviceket al. [11]. Primer sequences and annealing temperatures areavailable on request. The PCR product was purified with Exo-SapIT (USB Amersham, Uppsala, Sweden) at 37�C for 40minutes, followed by 85�C for 15 minutes. The BigDye Termi-nator cycle sequencing ready reaction kit (Applied Biosys-tems, Foster City, CA) was used for the sequencing reaction.The ABI PRISM 3100 genetic analyzer (Applied Biosystems)was used, and the obtained sequences were aligned withDNAStar SeqMan II software (DNASTAR, Madison, WI).The linkage disequilibrium (LD) between the observed SNPswas calculated with Haploview software [12]

2.3. Genotyping

Genotyping was performed using an allele-specificPCR-based KASPar SNP genotyping system (KBioscien-ces, Hoddesdon, UK). Thermocycling was performed ac-cording to the KBiosciences PCR conditions with 5%dimethyl sulfoxide and 1.8 mmol/L Mg2þ. Detection wasperformed using an ABI PRISM 7900 HT sequence detec-tion system with SDS 2.2 software (Applied Biosystems,Weiterstadt, Germany).

2.4. Statistical analysis

The genotype distribution was tested for HardyeWeinbergequilibrium using a c2 test. Odds ratios (OR) with 95% confi-dence intervals (CI) were calculated for association betweenthe genotypes and breast cancer. The calculations were per-formed using the online HardyeWeinberg test tool offeredby the Institute for Human Genetics, Technical Univer-sity Munich (http://ihg2.helmholtz-muenchen.de/cgi-bin/hw/hwa1.pl). Power calculation was performed using PS softwareversion 2.1.31 for power and sample size calculations (http://biostat.mc.vanderbilt.edu/twiki/bin/view/Main/PowerSampleSize).

3. Results

We screened the promoter of the TERT gene for polymor-phisms in a set of 32 Swedish breast cancer samples andfound three SNPs with a minor allele frequency above10%, rs2853669 (d244 T O C), rs2735940 (d1381C O T)and rs2736109 (d1654 G O A). Two of the SNPs,rs2853669 (d244 T O C) and rs2735940 (d1381C O T),showed a relatively high LD with a D’ value of 1.0 andan r2 value of 0.63. The LD between the rs2736109

73V. Varadi et al. / Cancer Genetics and Cytogenetics 190 (2009) 71e74

(d1654 G O A) SNP with the two other SNPs was lower (D0

5 0.79, r2 5 0.63 and D05 1, r2 5 0.44, respectively). This isin agreement with previously published data [7]. Based onthese data and published data about the functionality of thers2853669 (d244 T O C) SNP [1e3], we selected thed244 T O C SNP for further analyses in two large caseecontrol sample sets.

For the Polish sample set, the distribution of the geno-types for the SNP rs2853669 (d244 T O C) was in agree-ment with HardyeWeinberg equilibrium. The caseecontrolanalysis showed no statistically significant associationbetween the SNP and the risk of breast cancer (Table 1).In the Swedish sample set, women homozygous for theminor allele C seem to have a lower risk of breast cancerthan women homozygous for the T allele (Table 1).However, the genotype distribution of the controls deviatedfrom HardyeWeinberg equilibrium (P < 0.001). Becausethe genotypes of the 104 duplicate samples included inthe sample set agreed with each other, genotyping errorscould be excluded. If the genotype distribution in controlshad been according to HardyeWeinberg equilibrium, noeffect on breast cancer risk would have been observed,and the findings for association would have been (TC) vs(TT) OR 5 0.99, 95% CI 0.82e1.18 and (CC) vs. (TT)OR 5 0.78, 95% CI 5 0.54e1.11.

4. Discussion

Our finding of no effect of the TERT SNP rs2853669(d244 T O C) on breast cancer in the Swedish populationwith incident cases of breast cancer agrees with the data ofSavage et al. [1] on a large population-based study amongPolish women. However, we could not confirm thedecreased risk observed by Savage et al. [1] for the carriersof the minor allele in a small subgroup of 204 women witha family history of breast cancer (per allele, OR 5 0.66,

Table 1

Association of the TERT promoter SNP rs2853669 (d244C O T) with

breast cancer

Genotype

Cases Controls

P-valueN % N % OR 95% CI

Polish samples

TT 411 54 244 56 1

TC 299 39 154 35 1.15 0.90e1.48 0.27

CC 58 7 38 9 0.91 0.58e1.41 0.66

Totala 768 100 460 100

Swedish samples

TT 409 53 818 54 1

TC 310 41 558 37 1.1 0.92e1.32 0.31

CC 47 6 143 9 0.65 0.46e0.92 0.02

Totala 766 100 1519 100

Abbreviations: CI, confidence interval; N, number of individuals with

genotype data; OR, odds ratio.a The total number of cases and controls is smaller than the number

enrolled in the study because of missing genotype information for 122 cases

(73/841 Polish cases, 49/815 Swedish) and 64 controls (24/460 Polish

control subjects, 40/1,559 Swedish).

95% CI 5 0.46e0.95). In both the present study withfamilial and early-onset breast cancer cases and the studyby Savage et al. [1], the breast cancer cases and healthycontrols were Polish women living in upper Silesia and inWarsaw or qodz, respectively. The genotype distributionwas similar in both studies, but Savage et al. [1] reportedthe C allele to be the major allele and the CC genotypethe most common genotype. In the present study, for bothpopulations, the T allele was the major one and the TTgenotype the most common genotype (Table 1). The NCBISNP database (http://www.ncbi.nlm.nih.gov/SNP/) alsoreports the T allele to be the ancestral and the major alleleamong people of European ancestry.

The power of our study to detect the odds ratio of 0.66reported by Savage et al. [1] for women with a family historyof breast cancer was O95%, considering that the Polish caseshad an inherited susceptibility to breast cancer [13]. With thispower, we can also exclude an effect of the SNP rs2735940(d1381C O T) on breast cancer risk because of its relativelyhigh LD with the studied SNP. Savage et al. [1] also reporteda protective effect for a synonymous SNP rs2736098(Ala305Ala), most likely because of its LD with the SNPrs2853669 (d244 T O C). Thus, our finding of no effectof the d244 T O C SNP on breast cancer risk also suggestsno effect for the rs2736098 (Ala305Ala). The much lowerpower (~50%) of the Savage et al. [1] study may in partexplain the different outcome of the two studies.

In conclusion, our well-powered caseecontrol study usingtwo different sample sets, including Polish cases with an in-herited susceptibility to breast cancer, indicates that the func-tional TERT SNP d244 T O C does not affect familial orsporadic breast cancer risk. The findings also suggest thatthe two other correlated SNPs, rs2735940 (d1381C O T)and rs2736098 (Ala305Ala), do not affect breast cancer risk.However, because the telomerase is reactivated in cancerand because its expression has been correlated with poor clin-ical outcome [14], further studies are needed to clarify the roleof TERT polymorphisms in breast cancer prognosis.

Acknowledgments

This study was supported by a grant from the EuropeanUnion (LSHC-CT-2004-503465).

References

[1] Savage SA, Chanock SJ, Lissowska J, Brinton LA, Richesson D,

Peplonska B, Bardin-Mikolajczak A, Zatonski W, Szeszenia-

Dabrowska N, Garcia-Closas M. Genetic variation in five genes impor-

tant in telomere biology and risk for breast cancer. Br J Cancer 2007;97:

832e6.

[2] Meyerson M, Counter CM, Eaton EN, Ellisen LW, Steiner P,

Caddle SD, Ziaugra L, Beijersbergen RL, Davidoff MJ, Liu Q,

Bacchetti S, Haber DA, Weinberg RA. hEST2, the putative human

telomerase catalytic subunit gene, is up-regulated in tumor cells

and during immortalization. Cell 1997;90:785e95.

[3] Hsu C-P, Hsu N-Y, Lee L-W, Ko J-L. Ets2 binding site single nucle-

otide polymorphism at the hTERT gene promoter: effect on

74 V. Varadi et al. / Cancer Genetics and Cytogenetics 190 (2009) 71e74

telomerase expression and telomere length maintenance in non-small

cell lung cancer. Eur J Cancer 2006;42:1466e74.

[4] Cong YS, Wen J, Bacchetti S. The human telomerase catalytic

subunit hTERT: organization of the gene and characterization of

the promoter. Hum Mol Genet 1999;8:137e42.

[5] Cheung AL, Deng W. Telomere dysfunction, genome instability and

cancer. Front Biosci 2008;13:2075e90.

[6] Jefford CE, Irminger-Finger I. Mechanisms of chromosome insta-

bility in cancers. Crit Rev Oncol Hematol 2006;59:1e14.

[7] Savage SA, Stewart BJ, Eckert A, Kiley M, Liao JS, Chanock SJ.

Genetic variation, nucleotide diversity, and linkage disequilibrium

in seven telomere stability genes suggest that these genes may be

under constraint. Hum Mutat 2005;26:343e50.

[8] Jin Q, Hemminki K, Grzybowska E, Klaes R, Soderberg M,

Zientek H, Rogozinska-Szczepka J, Utracka-Hutka B, Pamula J,

Pekala W, Forsti A. Polymorphisms and haplotype structures in

genes for transforming growth factor b1 and its receptors in

familial and unselected breast cancers. Int J Cancer 2004;112:

94e9.

[9] Kaaks R, Lundin E, Rinaldi S, Manjer J, Biessy C, Soderberg S,

Lenner P, Janzon L, Riboli E, Berglund G, Hallmans G. Prospective

study of IGF-I, IGF-binding proteins, and breast cancer risk, in

northern and southern Sweden. Cancer Causes Control 2002;13:

307e16.

[10] Brendle A, Lei H, Brandt A, Johansson R, Enquist K, Henriksson R,

Hemminki K, Lenner P, Forsti A. Polymorphisms in predicted micro-

RNA-binding sites in integrin genes and breast cancer: ITGB4 as

prognostic marker. Carcinogenesis 2008;29:1394e9.

[11] Vaclavicek A, Hemminki K, Bartram CR, Wagner K,

Wappenschmidt B, Meindl A, Schmutzler RK, Klaes R, Untch M,

Burwinkel B, Forsti A. Association of prolactin and its receptor gene

regions with familial breast cancer. J Clin Endocrinol Metab 2006;91:

1513e9.

[12] Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visu-

alization of LD and haplotype maps. Bioinformatics 2005;21:

263e5.

[13] Houlston RS, Peto J. The future of association studies of common

cancers. Hum Genet 2003;112:434e5.

[14] Salhab M, Jiang WG, Newbold RF, Mokbel K. The expression of

gene transcripts of telomere-associated genes in human breast cancer:

correlation with clinico-pathological parameters and clinical

outcome. Breast Cancer Res Treat 2008;109:35e46.