Genetic variation in the genome-wide predicted estrogen response element-related sequences is associated with breast cancer development.

1.3 MB·English

by  Yu, Jyh-Cherng

Yuetal.BreastCancerResearch2011,13:R13
http://breast-cancer-research.com/content/13/1/R13
RESEARCH ARTICLE Open Access
Genetic variation in the genome-wide predicted
estrogen response element-related sequences is
associated with breast cancer development
Jyh-Cherng Yu1, Chia-Ni Hsiung2, Huan-Ming Hsu1, Bo-Ying Bao3, Shou-Tung Chen4, Giu-Cheng Hsu5,
Wen-Cheng Chou2, Ling-Yueh Hu2, Shian-Ling Ding6, Chun-Wen Cheng7, Pei-Ei Wu2, Chen-Yang Shen2,8*
Abstract
Introduction: Estrogen forms a complex with the estrogen receptor (ER) that binds to estrogen response elements
(EREs) in the promoter region of estrogen-responsive genes, regulates their transcription, and consequently
mediates physiological or tumorigenic effects. Thus, sequence variants in EREs have the potential to affect the
estrogen-ER-ERE interaction. In this study, we examined the hypothesis that genetic variations of EREs are
associated with breast cancer development.
Methods: This case-control study involved 815 patients of Asian descent with incident breast cancer and 821
healthy female controls. A total of 13,737 ERE sites in the whole genome predicted by a genome-wide
computational algorithm were blasted with single-nucleotide polymorphism (SNP) sequences. Twenty-one SNPs
located within 2,000 bp upstream or within introns 1 and 2 of putative genes and with a minor allele frequency
greater than 5% were identified and genotyped. Frequencies of SNPs were compared between cases and controls
to identify SNPs associated with cancer susceptibility.
Results: A significant combined effect of rs12539530, an ERE SNP in intron 2 of NRCAM which codes for a cell
adhesion molecule, and SNPs of ESR1, the gene coding for ER, on breast cancer risk was found. Interestingly, this
combined effect was more significant in women who had experienced a longer period of lifetime estrogen
exposure, supporting a hormonal etiology of this SNP in breast tumorigenesis.
Conclusions: Our findings provide support for a role of genetic variation in ERE-ESR1 in determining susceptibility
of breast cancer development.
Introduction repressing their transcription and consequently mediat-
The roles of estrogen receptor a (ERa) in initiating ing physiological or tumorigenic effects. Given that
tumordevelopmentinbreastcancer,regulatingprogres- sequencevariants,suchassingle-nucleotidepolymorph-
sion and determining therapeutic protocols and efficacy isms (SNPs), located inthe promoters ofgenes have the
are well documented [1-3]. Although ERa can be acti- potentialtoaffecttheprotein(transcriptionfactor)-DNA
vated in an estrogen-independent manner, the classical (promoter)interaction, resultinginalteredexpressionof
activation mechanism involves ERa binding to estrogen target genes [6,7], we decided that it was meaningful to
and other coactivator proteins to form the estrogen- examine the hypothesis that genetic variations of EREs
bound ER complex, which functionsasa transcriptional mightbeassociatedwithbreastcancerdevelopment.
regulator [4,5]. The DNA-binding domainof ERa binds Early work on the Xenopus vitellogenin gene identified
to estrogen response elements (EREs) in the promoter a minimal ERE core sequence of 5’-GGTCANNNT-
region of estrogen-responsive genes, activating or GACC-3’ [8]. Since then, several computational
approaches have been used to map EREs on a genome-
*Correspondence:[email protected] wide level on the basis of the presence of EREs within
8GraduateInstituteofEnvironmentalSciences,ChinaMedicalUniversity,
promoter proximal regions [9,10]. By specifically
Hsueh-ShihRoad,Taichung,404,Taiwan
Fulllistofauthorinformationisavailableattheendofthearticle
©2011Yuetal.;licenseeBioMedCentralLtdThisisanopenaccessarticledistributedunderthetermsoftheCreativeCommons
AttributionLicensehttp://creativecommons.org/licenses/by/2.0,whichpermitsunrestricteduse,distribution,andreproductioninany
medium,providedtheoriginalworkisproperlycited
Yuetal.BreastCancerResearch2011,13:R13 Page2of8
http://breast-cancer-research.com/content/13/1/R13
focusing on promoter regions, 12,515 EREs have been Specimen collection and SNP selection and genotyping
identified in the human genome [10-12]. To distinguish At the end of the interview, blood was taken for DNA
between real binding sites and noise, several attempts isolation and genotyping. All samples were examined by
have been made to improve the specificity of prediction. laboratory personnel who were blinded to the case/con-
For instance, by eliminating EREs that are not conserved trol status of specimens. DNA was extracted from the
between the human and mouse genomes, the number of peripheral blood samples of patients and controls using
gene proximal EREs has been reduced to 660 [10,11]. In DNA purification kits (Promega, Madison, WI, USA).
this study, we used PReMod [13], a new database of The PReMod database is a genome-wide/transcription
genome-wide cis-regulatory modules, to predict all pos- factor-wide collection of more than 100,000 computa-
sible EREs in the genome. The prediction algorithm of tional predicted transcriptional regulatory modules
PReMod takes into account the fact that, in higher within the human genome [13,14]. These modules are
eukaryotes, cis-regulatory regions often contain several specific sequences potentially regulated by 229 tran-
phylogenetically conserved binding sites for different scription factor families, and the PReMod algorithm pre-
transcription factors [13,14], and thus it has proven to dicts that a total of 13,737 sites within the human
be more reliable than other methods. Using the SNPs of genome are bound and/or regulated by the ER [13,14].
the human genome available in databases, we searched We compared these sites with conventional SNP data-
for SNPs within these genome-wide predicted EREs and bases [22-24] and identified ER-binding sites potentially
explored their association with breast cancer. harboring SNPs. The following three criteria were then
used to determine the SNPs to be genotyped: (1)
Materials and methods because of statistical considerations (that is, considera-
Study participants tion of study power), the minor allele frequency of the
This case-control study is part of an ongoing coopera- selected SNPs had to be higher than 0.05; (2) because of
tive study aimed at understanding the causes of breast biological considerations, the selected SNPs had to be
cancer in Taiwan. Breast cancer in Taiwan is charac- located within 2,000 bp upstream or located within
terized by low incidence, early tumor onset, hormone introns 1 and 2 of putative genes; and (3) because of
dependency and novel genomic alterations [15-17]. We technical considerations, SNPs having the potential to
studied 815 female breast cancer patients with patholo- yield a false signal using the iPLEX high-throughput
gically confirmed incident primary breast cancer seen genotyping platform were excluded. As a result, a total
at the Tri-Service General Hospital or the Changhua of 21 SNPs were chosen for genotyping.
Christian Hospital between March 2002 and August SNPs were genotyped in all samples using Sequenom
2007. The 821 healthy female controls were selected iPLEX technology (Sequenom, Hamburg, Germany).
from among women attending the health examination Positive, negative and duplicate controls were included
clinics of the same hospitals during the same period. on all plates, with genotypes being autocalled by specia-
The characteristics of these study participants have lized software (MassARRAY Typer version 3.4; Seque-
already been described in detail [18-21], and some par- nom) and subsequently confirmed by visual assessment
ticipants (551 patients and 727 controls) have recently of the data. All assays were performed by individuals
been genotyped for polymorphism of ESR1 [21], the blinded to the case-control status of the samples. As a
gene coding for the ER. This study was approved by quality control, we repeated the genotyping on 10% of
the Ethics Committee of the Institutional Review the samples, and all genotype scoring was performed
Board of the Academia Sinica, Taiwan, and informed and checked separately by one reviewer who was una-
consent was obtained from all study participants before ware of the case-control status. The concordance rate
the collection of epidemiologic data by personal inter- for replicate samples was 100%.
view. Considerations regarding methodological issues
in the present study (such as study design, sampling Statistical analysis
scheme, and potential bias) have been described in Univariate and multivariate analyses were used to deter-
detail previously [18-21]. mine the risk factors for breast cancer in this series of
study participants, and the odds ratio (OR) and corre-
Questionnaire sponding 95% confidence intervals (95% CIs) were esti-
Experienced research nurses were assigned to administer mated. For individual ERE SNPs, genotype frequencies
a structured questionnaire to both patients and controls. were assessed for departure from the Hardy-Weinberg
The information collected has been described, and the equilibrium using either a c2 goodness-of-fit test or a
validity of the questionnaire has been addressed and Fisher’s exact test. The c2 test for a 2 × 3 contingency
confirmed, in our previous studies [18-21]. table was used to compare genotype frequency between
Yuetal.BreastCancerResearch2011,13:R13 Page3of8
http://breast-cancer-research.com/content/13/1/R13
patients and controls. To take account of multiple com- Results
parisons, these associations were also assessed using the The risk profile of breast cancer in our study partici-
permutation test provided in the Haploview software pants was similar to that found in our previous studies
tool (Broad Institute of the Massachusetts Institute of [18-21] and in other breast cancer studies. The develop-
Technology and Harvard University, Cambridge, MA, ment of breast cancer was found to be highly associated
USA), run using 10,000 permutations. The association with reproductive risk factors, including early menarche,
of susceptibility genotypes and breast cancer risk was nulliparity, lower number of full-term pregnancies
further evaluated with simultaneous consideration of (FTPs) and older age at FFTP (Table 1). Compared to
established risk factors for breast cancer or other signifi- controls, patients were younger at menarche (≤ 14 years
cant risk factors in multivariate logistic regression mod- vs. >14 years, aOR, 1.52; 95% CI, 1.18 to 1.96) and older
els. Biologic plausibility was the most important at FFTP (>23 years vs. ≤ 23 years, aOR, 1.30; 95% CI,
criterion for inclusion of variables in the model. There- 1.00 to 1.69). Significant protection was conferred by a
fore, we included all established risk factors for breast history of FTP (parous women vs. nulliparous women,
cancer in the statistical models: age, family history of aOR, 0.67; 95% CI, 0.46 to 0.97) and a greater number
breast cancer, age at menarche, parity and age at first of FTPs (history of more than two FTPs vs. history of
full-term pregnancy (FFTP). Adjusted ORs (aORs) and two or fewer FTPs, aOR, 0.46, 95% CI, 0.34 to 0.61). No
95% CIs for genotypes were then estimated. association was found between cancer risk and a history
We made use of the information on the ESR1 poly- of oral contraceptive use or between cancer risk and
morphism in our study participants that we published body mass index. Significant risk factors were included
recently [21] andexplored the effectof a possibleESR1- in the multivariate logistic regression models when we
ERE interaction or estrogen-ESR1-ERE interaction in examined the association between SNPs and cancer risk.
determining breast cancer development. A combination More importantly, these significant associations between
of the joint method and stratified analysis [18-21] was reproductive risk factors and breast cancer reveal the
applied to determine whether this interaction between importance of the estrogen-related etiology of breast
ESR1 and ERE was associated with breastcancer forma- cancer in our participants, providing the opportunity to
tion.AjointeffectofESR1-EREonincreasedbreastcan- examine the contribution of EREs during breast
cer risk was explored using conventional logistic tumorigenesis.
regression,atestevaluatingwhetherastatisticallysignifi- A total of 13,737 sites (274,740 bp of DNA) in the
cantincreaseinriskwasobservedwithspecificcombina- whole genome were predicted to be ERE-related
tions of putative high-risk genotypes in these SNPs sequences using the PReMod algorithm. After blasting
(measured by the b estimates from this regression these data with online information available from SNP
model). In addition, we stratified our study participants data sets (University of California, Santa Cruz, National
on the basis of ESR1 genotype and examined whether Center for Biotechnology Information and HapMap),
breastcancerriskassociatedwithERESNPswasparticu- 322 ER-binding sites were identified as potentially har-
larly significant in specific ESR1 genotype subsets of boring SNPs. One hundred ten of these 322 SNPs were
women.Becausewewereespeciallyinterestedintherela- found to contain no variant allele in the Chinese popu-
tionshipbetweenthejointeffectofESR1-EREandbreast lation, resulting in 212 SNPs, 21 of which met our cri-
cancerriskwithin categoriesofriskfactorsrepresenting teria and were genotyped in patients and controls
differentlevelsofestrogenexposure,weperformedstrati- (Figure 1).
fiedanalysistotestthishypothesis.Therefore,iftheiden- To determine the breast tumorigenic contribution of
tified joint effect of ESR1-ERE SNPs initiated breast ERE SNPs, we examined whether the genotypic distribu-
cancer by the formation of the estrogen-ER complex, tion of individual SNPs differed between the cases and
thentherelationshipbetweenbreastcancerriskand the controls (Table S1 in Additional file 1). The frequencies
joint effect would not be the same in women who had of all SNPs in the controls agreed with those expected
experienceddifferentlengthsofestrogenexposure.This on the basis of the Hardy-Weinberg equilibrium, sug-
wasevaluatedbycalculatingtherisk(aOR)ofbreastcan- gesting that genotyping error was relatively unlikely.
cerassociated withthejointeffectof ESR1-ERESNPsin The result for the genotypic analysis of two SNPs
women witha longer or shorterperiod oftotal estrogen (rs12539530 and rs9527676) was important, as it showed
exposure. For menopausal women, total estrogen expo- that women carrying the homozygous variant genotype
surewascalculatedusingtheformula(ageatmenopause had an significantly increased OR (P < 0.05) compared
-ageatmenarche-yearsoffull-termpregnancy)and,for with women carrying the homozygous wild-type geno-
premenopausalwomen,ageatmenopauseinthisformula type, as well as that the carrying of one additional risk
wasreplacedbyageatrecruitmentintothisstudy. allele was associated with a significant increase in risk
Yuetal.BreastCancerResearch2011,13:R13 Page4of8
http://breast-cancer-research.com/content/13/1/R13
Table 1Frequency ofriskfactorsin breast cancer patients andcontrols andthe adjusted oddsratios in relation to
breast cancerriska
Riskfactor Patients,%(n=814) Controls,%(n=821) aOR(95%CI)b
Ageatmenarche
>14yr 34.1 37.6 1.00(reference)
≤14yr 65.9 62.4 1.52(1.18to1.96)
Menopause
No 49.6 59.5 1.00(reference)
Yes 50.4 40.5 0.60(0.42to0.84)
Familyhistoryofbreast
cancerinfirst-degree
relatives
No 88.7 90.7 1.00(reference)
Yes 11.3 9.3 1.20(0.83to1.73)
Nulliparity
No 11.6 10.2 1.00(reference)
Yes 88.4 89.8 0.67(0.46to0.97)
No.offull-term
pregnancies
≤2 51.0 45.6 1.00(reference)
>2 49.0 54.4 0.46(0.34to0.61)
Ageatfirstfull-term
pregnancy
≤23years 62.6 64.0 1.00(reference)
>23years 37.4 36.0 1.30(1.00to1.69)
Bodymassindex
≤23 56.9 59.1 1.00(reference)
>23 43.1 40.9 0.90(0.71to1.15)
Useoforalcontraceptives
No 81.4 82.1 1.00(reference)
Ever 18.6 17.9 1.05(0.78to1.40)
aaOR,adjustedoddsratio;95%CI,95%confidenceinterval;baORand95%CIwereestimatedinalogisticregressionmodelinwhichtheageofstudy
participantswasincluded.
(Table S1 in Additional file 1). The possibility of false- with previous findings using microarray technology
positives due to multiple testing is less likely, because [25,26]. In addition, cell-cell adhesion, in which NRCAM
the permutation test showed on the basis of 10,000 ran- is known to play a part, has been well documented as
dom permutations that these two associations were bor- being involved in cancer formation [27,28]. All of these
derline significant or significant (P = 0.08 andP = 0.03, lines of evidence support the biological plausibility of
respectively) (Table S1 in Additional file 1). Both SNPs our findings, suggesting that rs12539530, found in an
are located in regulatory regions of genes coding for cell ERE-related sequence and possibly regulating NRCAM
adhesion molecules, as rs12539530 is located in intron 2 expression, is associated with breast cancer
of NRCAM, a gene coding for a neuron-related cell susceptibility.
adhesion molecule, and rs9527676 is located in intron 1 The well-known mechanism in which ER binds to EREs
of PCDH17, a gene coding for protocadherin-17. To to mediate the expression of ER-regulated genes [2-4]
gain initial clues for further analysis, we examined the prompted us to speculate whether the SNPs of EREs
expression of these two genes in breast cancer cell lines. and ESR1 might be jointly associated with breast cancer.
To this end, we checked the expression of NRCAM and We made use of the information on the ESR1 genotype
PCDH17 in breast cancer cell lines expressing ESR1 of our study participants that we published recently [21]
(MCF-7 cells) or not expressing ESR1 (MDA-MB-231 and examined this possibility using both stratified analy-
cells) and examined whether the expression of these sis and the joint method. Three SNPs (rs3778609,
putative ER-regulated genes is ER-dependent. The rs12665044, and rs827421) located in one cluster in
results for NRCAM were more promising, as they intron 1 within the sequence coding for the activation
showed that this gene could be expressed in an ER-posi- function (AF)-1 domain of the ER and one SNP
tive breast cancer cell (Figure 2), which is consistent (rs7739506) located in intron 4 within the sequence
Yuetal.BreastCancerResearch2011,13:R13 Page5of8
http://breast-cancer-research.com/content/13/1/R13
Figure2Transcription(mRNA)ofNRCAMdetectedbyreverse
transcriptase-polymerasechainreaction(RT-PCR)assayisseen
Figure 1 Flow diagram showing selection from the genome-
intheestrogenreceptor(ER)-positivebreastcancercellline
wide-predictedestrogenresponseelement-relatedsequences
MCF-7,butnotintheER-negativebreastcancercelllineMDA-
ofsingle-nucleotidepolymorphisms(SNPs)forgenotyping.ER,
MB-231.GAPDHisthepositive(mRNA)controlofRT-PCR.ForRT-
estrogenreceptor;MULDI-TOF,matrix-assistedlaserdesorption
PCR,totalRNAwasextractedfromcelllinesusingtheArcturus
inoization-timeofflightmassspectrometry.
PicoPurRNAIsolationKit(AppliedBiosystems,ForsterCity,CA,USA),
andtoconvertRNAintocDNA,reversetranscriptionwasperformed
for70minutesat42°Cinareactionvolumeof20μlcontaining1
coding for the AF-2/ligand-binding domain have pre- μgofRNA,10mMrandomoligo(dT)primer(Promega),and5Uof
viously been found to show significant or borderline sig- SuperScriptIIreversetranscriptase(Gibco-BRL,Gaithersburg,MD,
USA).ThepresenceofacDNAbandoftheappropriatemolecular
nificant associations with breast cancer susceptibility in
weightwasdeterminedon1%agarosegelafterelectrophoresis.
our population [21]. If this ERE SNP were linked to
breast cancer susceptibility via the suspected ER-related
mechanism, the association between rs12539530 and ESR1 SNPs in all women combined (Figure 3). More
breast cancer would differ between women harboring interestingly, estrogen promotes breast tumorigenesis by
different ESR1 genotypes. Our findings are consistent forming a complex with the ER which then binds to
with this speculation, and the association between high- EREs[1,2],andthustherelationshipbetweenbreastcan-
risk genotypes of rs12539530 and an increased breast cerriskandthejointeffectofrs12539530andESR1poly-
cancer risk was significant in only one subset of women morphism might be modified by estrogen exposure,
carrying specific genotypes of ESR1, but not in the other which is consistent with what we observed (Figure 3).
subset (Table S2 in Additional file 1). Furthermore, on The riskofbreast cancer associatedwith the joint effect
the basis of a very stringent multiplicative model, the ofpolymorphismsofrs12539530andESR1wasevaluated
borderline significance P values for the interaction in different groups of women stratified by years of total
between rs12539530 and rs827421 (P = 0.07) and estrogen exposure and a significant and increased joint
between rs12539530 and rs7739506 (P = 0.09) are in effect was seen only in the subgroup of women with
line with the suggestion that an interaction between more than 30 years of estrogen exposure (Figure 3). In
rs12539530 and ESR1 polymorphism is linked to breast contrast, in the subsetof women with less than 30 years
cancer risk. To confirm this interaction, we used the ofestrogenexposure,thesamejointeffectwasassociated
joint method and calculated the risk of breast cancer withanonsignificantaOR(Figure3).
associated with both rs12539530 and ESR1 SNPs using a
set of dummy variables representing different combina- Discussion
tions of genotypes of ESR1 and rs12539530. In contrast Of the predicted ERE-related sequences found through-
to the increased, but not significant, risk associated with out the whole genome, the present study has identified
either a high-risk genotype of ESR1 or rs12539530 a genetic variation of rs12539530, a SNP located in the
alone, the greatest risk was found in those patients har- putative ERE site in intron 2 of NRCAM, as an impor-
boring high-risk genotypes of both rs12539530 and tant factor in determining susceptibility to breast cancer
Yuetal.BreastCancerResearch2011,13:R13 Page6of8
http://breast-cancer-research.com/content/13/1/R13
Figure3Associationofthejointeffectofgenotypesofrs12539530,asingle-nucleotidepolymorphism(SNP)associatedwithestrogen
responseelements(EREs),andofsingle-nucleotidepolymorphismsofESR1,theestrogenreceptorgene,withbreastcancerriskinall
womencombinedandinwomenstratifiedbytotalyearsofestrogenexposure(indicatedas“Estrogen”).aOR,adjustedoddsratio;95%
CI,95%confidenceinterval.
development. We attempted to address the possibility of we checked the LD block in which rs12539530 is
false-positives and the effects of multiple testing by located in the Chinese population in HapMap [24],
demonstrating a borderline significant P value in the which spans 22 kb in chromosome region 7q31, and
permutation test. Even so, given the lack of strongly sig- found that the whole block is within intron 2 of
nificant results, the power of the present study should NRCAM. In other words, no well-defined genes are in
be an issue of particular concern, and our suggestion the same haplotype block as this SNP. To address the
that rs12539530 is involved in determining breast cancer second possibility, we examined the sequence of this
susceptibility certainly needs to be confirmed in other LD block and found that it contains more than 200
studies with larger sample sizes. predicted transcription factor binding sites, including
In considering whether our findings represent a true those for specificity protein 1 (Sp1) and activator pro-
association between this SNP in ERE and breast can- tein 1 (AP-1). After blasting these binding sites with
cer, the most important issue is the interpretation of the sequences of 30 SNPs known to reside in this
the identified association between SNPs and the trait. block as reported by HapMap, some SNPs were found
Because the SNP identified is in an intron, it does not to be located within these binding sites, so the second
affect amino acid coding and therefore probably does possibility cannot be totally excluded. However, we are
not directly affect protein function. In addition, the prompted to suggest the breast tumorigenic contribu-
observed association between breast cancer and this tion of rs12539530 on the basis of the findings that (1)
SNP could be due to the presence of linkage disequili- rs12539530 and ESR1 SNPs jointly increased breast
brium (LD) between this SNP and other SNPs in cancer susceptibility and (2) this joint effect was more
exons (resulting in functional change) or in regulatory significant in women with a longer period of estrogen
regions (affecting the expression of these genes). How- exposure. In addition, this suggestion is mechanistically
ever, the first of these two possibilities is less likely, as plausible and is similar to the finding that rs10736303,
Yuetal.BreastCancerResearch2011,13:R13 Page7of8
http://breast-cancer-research.com/content/13/1/R13
the SNP generating a putative ERE in intron 2 of suggests that this is a promising approach to identifying
FGFR2, has been identified as the most significant thebreasttumorigeniccontributionofEREsonagenome-
gene determining breast cancer susceptibility in recent widescaleonthebasisofinformationgeneratedbyrecent
genome-wide association studies (GWASs) [29-31]. In technological advances. For example, in our ongoing
addition, expression of NRCAM, the gene putatively study, we are exploring the breast tumorigenic role of
regulated by rs12539530, has been suggested to be putativeEREsidentifiedusingchromatinimmunoprecipi-
upregulated in ER-positive, but not in ER-negative, tation-basedmethods[25,32,33]. The simultaneous con-
breast cancer cell lines [25,26], and this hypothesis was siderationoftheseER-associatedEREsandER-regulating
confirmed by the present study (Figure 2). genesthat drivebreasttumorigenesismightbecriticalin
Recent GWASs based on technological advances in thedevelopmentofnewanticancer drugtargetsandnew
high-throughput genotyping have provided information therapeuticanddiagnosticapproaches.
regarding the LD of neighboring polymorphisms. These
GWASs have made possible the use of a few hundred Conclusions
thousand SNPs as tags for all other variants, as well as Though the promise of personalizedmedicine, in which
thebiobankingoftensofthousandsofspecimens,allow- the risk and the course of diseases and the efficacy of
ing their immediate use in whole genome research. treatmentprotocolswouldbepredictedonthebasisofa
These studies have ledtothe mappingof novelsuscept- person’s genotype, must been tempered with caution,
ibilitylociformanycommontraits,includingbreastcan- validatedmoleculartestsassessingthepatient’sgermline
cer[30,31]. Interestingly, none of theselociidentifiedas DNA already drive therapeutic decision-making.On the
significant in determining breast cancer risk in GWASs basisofthewell-documentedroleofERinbreastcancer
havebeenidentifiedassignificantinpreviousassociation development and progression, this study explored
studiesbasedonacandidategeneapproach,andveryfew whethergeneticvariationsinEREs,thesequencesbound
of the genes involved in the most plausible mechanisms by ER to activate the transcriptional regulationof target
ofbreasttumorigenesis,includingthoseinvolvedinDNA genes,areassociatedwithsusceptibilityforbreastcancer.
repairandsexhormone synthesisandmetabolismpath- Notably,theEREsitesgenotypedwerebasedongenome-
ways, have been reported as important in GWASs. wideprediction,providingauniqueopportunitytocom-
Region 7q31, in which NRCAM is located, has never prehensivelyexamineputativeEREsiteswithoutdepend-
beenreportedtobeimportantwithregardtobreastcan- ing on a prior hypothesis. A significant combined effect
cerriskdeterminationinGWASs.Thismightbepartially of rs12539530, an ERE SNP in intron 2 of NRCAM,
explainedbytherelativelylow-penetranceeffectofSNPs which codes for a cell adhesion molecule, and SNPs of
of estrogen-regulating genes. Distinct from those alleles ESR1, the gene coding for ER, on breast cancer risk was
identifiedbycurrentGWAS,thesepolymorphicallelesof found. Our findings provide support for a role of ESR1-
estrogen-regulating genes would predispose carriers to ERE polymorphism in determining susceptibility of
only a moderately increased risk of developing cancer. breastcancerdevelopment.Thisknowledgewillbehelp-
Thus,thesignificanceofsuchSNPsdependsnotonlyon fulfordirectingthefocusoffutureexperimentalstudies.
their own effect but also on the interaction between the
target genes and other functionally related genes (for Additional material
example, ESR1) or the promoting effect of reproductive
riskfactorsreflectingestrogenexposure.Ourfindingsare Additionalfile1:TablesS1andS2.TableS1presentsgenotype
inlinewiththissuggestion. frequenciesofsequencevariantsofestrogenresponseelement(ERE)-
relatedsequencesinbreastcancerpatientsandcontrolsandthe
This study is a hybrid of candidate gene and genome-
adjustedoddsratio(aOR)inrelationtobreastcancerrisk.TableS2
wide approaches in which we took advantage of both presentsbreastcancerriskassociatedwithgenotypicpolymorphismof
designs. The well-defined roles of the ER during breast rs12539530,asingle-nucleotidepolymorphism(SNP)intheestrogen
responseelement(ERE)-relatedsequence,stratifiedbygenotypesofESR1,
tumorigenesis makes it reasonable to assume that poly-
theestrogenreceptorgene.
morphicgeneticvariantsofEREs,centralnodesintheER
pathway,mightunderliethevariationsseenbetweenindi-
vidualsintheirsusceptibilitytobreastcancer.Thiscandi-
Abbreviations
date mechanism lends critical support to the biological
CI:confidenceinterval;ER:estrogenreceptor;ERα:estrogenreceptorα;ERE:
plausibilityandtumorigenic relevanceofour findings. In estrogenresponseelement;FFTP:firstfull-termpregnancy;FTP:full-term
addition,ourgenotypingofSNPsonthebasisofgenome- pregnancy;GWAS:genome-wideassociationstudies;LD:linkage
disequilibrium;OR:oddsratio;SNP:singlenucleotidepolymorphism.
wide predicted EREs provided a unique opportunity to
comprehensively examine putative ERE sites without
Acknowledgements
dependingonapriorhypothesis.Thesuccessfulidentifica-
FinancialsupportwasprovidedbygrantNSC98-2628-B-016-001-MY3(toJCY)
tion of rs12539530 by using these combined methods fromtheNationalScienceCouncil,TaiwanandbygrantsTSGH-C97-7-S03,
Yuetal.BreastCancerResearch2011,13:R13 Page8of8
http://breast-cancer-research.com/content/13/1/R13
DOD-98-28-03andDOD-98-36fromtheTri-ServiceGeneralHospital(toJCY 14. BlanchetteM,BatailleAR,ChenX,PoitrasC,LaganièreJ,LefèbvreC,
andHMH). DebloisG,GiguèreV,FerrettiV,BergeronD,CoulombeB,RobertF:
Genome-widecomputationalpredictionoftranscriptionalregulatory
Authordetails modulesrevealsnewinsightsintohumangeneexpression.GenomeRes
1DepartmentofSurgery,Tri-ServiceGeneralHospital,Ming-ChengEastRoad, 2006,16:656-668.
Taipei,114,Taiwan.2InstituteofBiomedicalSciences,AcademiaSinica, 15. LoYL,YuJC,HuangCS,TsengSL,ChangTM,ChangKJ,WuCW,ShenCY:
AcademyStreet,Taipei,115,Taiwan.3DepartmentofPharmacy,China AlleliclossoftheBRCA1andBRCA2genesandotherregionson17q
MedicalUniversity,Hsueh-ShihRoad,Taichung,404,Taiwan.4Departmentof and13qinbreastcanceramongwomenfromTaiwan(areaoflow
Surgery,ChanghuaChristianHospital,NanxiaoStreet,Changhua,500, incidencebutearlyonset).IntJCancer1998,79:580-587.
Taiwan.5DepartmentofRadiology,Tri-ServiceGeneralHospital,Ming-Cheng 16. ShenCY,YuJC,LoYL,KuoCH,YueCT,JouYS,HuangCS,LungJC,
EastRoad,Taipei,114,Taiwan.6DepartmentofNursing,Kang-NingJunior WuCW:Genome-widesearchforlossofheterozygosityusinglaser
CollegeofMedicalCareandManagement,KangningRoad,Taipei,114, capturemicrodissectedtissueofbreastcarcinoma:animplicationfor
Taiwan.7InstituteofBiochemistryandBiotechnology,ChungShanMedical mutatorphenotypeandbreastcancerpathogenesis.CancerRes2000,
University,JianguoNorthRoad,Taichung,402,Taiwan.8GraduateInstituteof 60:3884-3892.
EnvironmentalSciences,ChinaMedicalUniversity,Hsueh-ShihRoad, 17. ChengTC,ChenST,HuangCS,FuYP,YuJC,ChengCW,WuPE,ShenCY:
Taichung,404,Taiwan. Breastcancerriskassociatedwithgenotypepolymorphismofthe
catecholestrogen-metabolizinggenes:amultigenicstudyoncancer
Authors’contributions susceptibility.IntJCancer2005,113:345-353.
JCYandCYSparticipatedinthestudydesignandcoordination,carriedout 18. DingSL,YuJC,ChenST,HsuGC,KuoSJ,LinYH,WuPE,ShenCY:Genetic
theparticipantrecruitmentanddraftedthemanuscript.CNHandPEW variantsofBLMinteractwithRAD51toincreasebreastcancer
carriedoutthegenotypinganddataanalysis.BYBcontributedtosearching susceptibility.Carcinogenesis2009,30:43-49.
EREwebsitesusingPReMod.HMH,STCandGCHcarriedouttheparticipant 19. DingSL,YuJC,ChenST,HsuGC,ShenCY:Geneticvariationinthe
recruitment.WCCandLYHcontributedtothevalidationofNRCAM prematureaginggeneWRN:acase-controlstudyonbreastcancer
expressioninbreastcancercelllines.SDandCWCperformedESR1 susceptibility.CancerEpidemiolBiomarkersPrev2007,16:263-269.
genotyping. 20. YuJC,DingSL,ChangCH,KuoSH,ChenST,HsuGC,HsuHM,HouMF,
JungLY,ChengCW,WuPE,ShenCY:Geneticsusceptibilitytothe
Competinginterests developmentandprogressionofbreastcancerassociatedwith
Theauthorsdeclarethattheyhavenocompetinginterests. polymorphismofcell-cycleandubiquitinligasegenes.Carcinogenesis
2009,30:1562-1570.
Received:27October2010 Revised:24December2010 21. DingSL,YuJC,ChenST,HsuHM,HoJY,LinYH,ChangCC,FannCS,
Accepted:31January2011 Published:31January2011 ChengCW,WuPE,ShenCY:DiverseassociationsbetweenESR1
polymorphismandbreastcancerdevelopmentandprogression.Clin
References CancerRes2010,16:3473-3484.
1. AliS,CoombesRC:Endocrine-responsivebreastcancerandstrategiesfor 22. UCSCGenomeBioinformatics.[http://genome.ucsc.edu/].
combatingresistance.NatRevCancer2002,2:101-112. 23. NCBIPubMed.[http://www.ncbi.nlm.nih.gov/sites/entrez].
2. YagerJD,DavidsonNE:Estrogencarcinogenesisinbreastcancer.NEnglJ 24. HapMapProject.[http://hapmap.ncbi.nlm.nih.gov/index.html.en/].
Med2006,354:270-282. 25. LinCY,VegaVB,ThomsenJS,ZhangT,KongSL,XieM,ChiuKP,LipovichL,
3. OsborneCK,SchiffR:Estrogen-receptorbiology:continuingprogressand BarnettDH,StossiF,YeoA,GeorgeJ,KuznetsovVA,LeeYK,CharnTH,
therapeuticimplications.JClinOncol2005,23:1616-1622. PalanisamyN,MillerLD,CheungE,KatzenellenbogenBS,RuanY,
4. GreenKA,CarrollJS:Oestrogen-receptor-mediatedtranscriptionandthe BourqueG,WeiCL,LiuET:Whole-genomecartographyofestrogen
influenceofco-factorsandchromatinstate.NatRevCancer2007, receptorαbindingsites.PLoSGenet2007,3:e87.
7:713-722. 26. RaeJM,JohnsonMD,ScheysJO,CorderoKE,LariosJM,LippmanME:
5. ShangY:MolecularmechanismsofoestrogenandSERMsinendometrial GREB1isacriticalregulatorofhormonedependentbreastcancer
carcinogenesis.NatRevCancer2006,6:360-368. growth.BreastCancerResTreat2005,92:141-149.
6. BondGL,HuW,BondEE,RobinsH,LutzkerSG,ArvaNC,BargonettiJ, 27. MicalizziDS,FarabaughSM,FordHL:Epithelial-mesenchymaltransitionin
BartelF,TaubertH,WuerlP,OnelK,YipL,HwangSJ,StrongLC,LozanoG, cancer:parallelsbetweennormaldevelopmentandtumorprogression.
LevineAJ:AsinglenucleotidepolymorphismintheMDM2promoter JMammaryGlandBiolNeoplasia2010,15:117-134.
attenuatesthep53tumorsuppressorpathwayandacceleratestumor 28. ThieryJP,AcloqueH,HuangRY,NietoMA:Epithelial-mesenchymal
formationinhumans.Cell2004,119:591-602. transitionsindevelopmentanddisease.Cell2009,139:871-890.
7. LiLC,ChuiRM,SasakiM,NakajimaK,PerincheryG,AuHC,NojimaD, 29. EastonDF,PooleyKA,DunningAM,PharoahPD,ThompsonD,
CarrollP,DahiyaR:AsinglenucleotidepolymorphismintheE-cadherin BallingerDG,StruewingJP,MorrisonJ,FieldH,LubenR,WarehamN,
genepromoteralterstranscriptionalactivities.CancerRes2000, AhmedS,HealeyCS,BowmanR,MeyerKB,HaimanCA,KolonelLK,
60:873-876. HendersonBE,LeMarchandL,BrennanP,SangrajrangS,GaborieauV,
8. Klein-HitpassL,SchorppM,WagnerU,RyffelGU:Anestrogen-responsive OdefreyF,ShenCY,WuPE,WangHC,EcclesD,EvansDG,PetoJ,
elementderivedfromthe5’flankingregionoftheXenopusvitellogenin FletcherO,etal:Genome-wideassociationstudyidentifiesnovelbreast
A2genefunctionsintransfectedhumancells.Cell1986,46:1053-1061. cancersusceptibilityloci.Nature2007,447:1087-1093.
9. BajicVB,TanSL,ChongA,TangS,StrömA,GustafssonJA,LinCY,LiuET: 30. VargheseSJ,EastonDF:Genome-wideassociationstudiesincommon
DragonEREFinderversion2:atoolforaccuratedetectionandanalysis cancers:whathavewelearnt?GenetDevelop2010,20:201-209.
ofestrogenresponseelementsinvertebrategenomes.NucleicAcidsRes 31. MavaddatN,AntoniouAC,EastonDF,Garcia-ClosasM:Genetic
2003,31:3605-3607. susceptibilitytobreastcancer.MolOncol2010,4:174-191.
10. BourdeauV,DeschenesJ,MetivierR,NagaiY,NguyenD,BretschneiderN, 32. WelborenWJ,vanDrielMA,Janssen-MegensEM,vanHeeringenSJ,SweepFC,
GannonF,WhiteJH,MaderS:Genome-wideidentificationofhigh-affinity SpanPN,StunnenbergHG:ChIP-SeqofERαandRNApolymeraseIIdefines
estrogenresponseelementsinhumanandmouse.MolEndocrinol2004, genesdifferentiallyrespondingtoligands.EMBOJ2009,28:1418-1428.
18:1411-1427. 33. WelborenWJ,SweepFC,SpanPN,StunnenbergHG:Genomicactionsof
11. WelborenWJ,StunnenbergHG,SweepFC,SpanPN:Identifyingestrogen estrogenreceptorα:whatarethetargetsandhowaretheyregulated?
receptortargetgenes.MolOncol2007,1:138-143. EndocrRelatCancer2009,16:1073-1089.
12. CarrollJS,BrownM:Estrogenreceptortargetgene:anevolvingconcept.
doi:10.1186/bcr2821
MolEndocrinol2006,20:1707-1714.
Citethisarticleas:Yuetal.:Geneticvariationinthegenome-wide
13. FerrettiV,PoitrasC,BergeronD,CoulombeB,RobertF,BlanchetteM:
predictedestrogenresponseelement-relatedsequencesisassociated
PReMod:adatabaseofgenome-widemammaliancis-regulatorymodule
withbreastcancerdevelopment.BreastCancerResearch201113:R13.
predictions.NucleicAcidsRes2007,35:D122-D126.