Genetic polymorphisms in RNA binding proteins contribute to breast cancer survival
-
Upload
rohit-upadhyay -
Category
Documents
-
view
213 -
download
0
Transcript of Genetic polymorphisms in RNA binding proteins contribute to breast cancer survival
Genetic polymorphisms in RNA binding proteins contribute tobreast cancer survival
Rohit Upadhyay, Sandhya Sanduja, Vimala Kaza, and Dan A. Dixon
Department of Biological Sciences and Cancer Research Center, University of South Carolina, Columbia, SC
The RNA-binding proteins TTP and HuR control expression of numerous genes associated with breast cancer pathogenesis by
regulating mRNA stability. However, the role of genetic variation in TTP (ZFP36) and HuR (ELAVL1) genes is unknown in breast
cancer prognosis. A total of 251 breast cancer patients (170 Caucasians and 81 African–Americans) were enrolled and
followed up from 2001 to 2011 (or until death). Genotyping was performed for 10 SNPs in ZFP36 and 7 in ELAVL1 genes. On
comparing both races with one another, significant differences were found for clinical and genetic variables. The influence of
genetic polymorphisms on survival was analyzed by using Cox-regression, Kaplan-Meier analysis and the log-rank test.
Univariate (Kaplan-Meier/Cox-regression) and multivariate (Cox-regression) analysis showed that the TTP gene polymorphism
ZFP36*2 A > G was significantly associated with poor prognosis of Caucasian patients (HR 5 2.03; 95% CI 5 1.09–3.76;
p 5 0.025; log-rank p 5 0.022). None of the haplotypes, but presence of more than six risk genotypes in Caucasian patients,
was significantly associated with poor prognosis (HR52.42; 95% CI 5 1.17–4.99; p 5 0.017; log-rank p 5 0.007). The effect
of ZFP36*2 A > G on gene expression was evaluated from patients’ tissue samples. Both TTP mRNA and protein expression
was significantly decreased in ZFP36*2 G allele carriers compared to A allele homozygotes. Conversely, upregulation of the
TTP-target gene COX-2 was observed ZFP36*2 G allele carriers. Through its ability to attenuate TTP gene expression, the
ZFP36*2 A > G gene polymorphism has appeared as a novel prognostic breast cancer marker in Caucasian patients.
Breast cancer alone is expected to account for 30% of all newcancer cases among women and 39,520 deaths in 2011 in theUnited States.1 Breast cancer is a multi-factorial disease withseveral environmental and genetic factors contributing to itsoccurrence and progression,2 with �28% of familial casesattributed to mutations in breast cancer susceptibility loci.3
In addition, several reports show that the breast cancer inci-dence, progression and mortality vary between Caucasiansand African–Americans, with heightened incidence and lower
mortality observed in Caucasian than African–Americanspatients.1,4–6 These observations indicate the need to identifynovel genetic factors that can contribute to the occurrenceand progression as well as race-specific differential prognosisof breast cancer.
A critical point in the regulation of many pro-inflamma-tory cytokines, growth factors and proto-oncogenes occursthrough post-transcriptional mechanisms that regulatemRNA degradation.7 A prominent cis-acting RNA elementpresent in a majority of these cancer-associated transcripts isthe adenylate- and uridylate (AU)-rich element (ARE) con-tained within the mRNA 30-untranslated region (30UTR).8
The importance of this particular RNA element is evident,since estimates ranging from 8 to 16% of all human protein-coding genes contain a 30UTR ARE sequence.9,10 AREs medi-ate their regulatory function through their association withRNA-binding proteins that display high affinity for AREs.The best studied ARE-binding proteins can promote rapidmRNA decay, mRNA stabilization, or translational silencing.7
Through these mechanisms, ARE-binding proteins exhibitwide-ranging effects on gene expression, since a single ARE-binding protein can target multiple distinct transcripts.
The ARE-binding proteins TTP (Tristetraprolin; ZFP36)and HuR (Hu antigen R; ELAVL1) regulate gene expressionthrough opposing post-transcriptional activities. TTP proteinis a member of a small family of tandem Cys3His zinc fingerproteins and promotes rapid decay of ARE-containingmRNAs.11,12 In contrast, HuR protein is a ubiquitouslyexpressed member of the ELAV-like family of RNA-binding
Key words: TTP, HuR, polymorphism, breast cancer, RNA-binding
protein
Abbreviations: ARE: adenylate- and uridylate-rich element; HR:
hazard ratio; htSNP: haplotype-tag SNP; HuR: Hu antigen R; LD:
linkage disequilibrium; OS: overall survival; qPCR: real-time PCR;
RFLP: restriction fragment length polymorphism; SNP: single
nucleotide polymorphism; TTP: tristetraprolin; 30UTR: 30-
untranslated region
Additional Supporting Information may be found in the online
version of this article.
Grant sponsor: NIH; Grant number: R01CA134609; Grant
sponsor: American Cancer Society Research Scholar grant; Grant
number: RSG-06-122-01-CNE
DOI: 10.1002/ijc.27789
History: Received 28 Jan 2012; Accepted 7 Aug 2012; Online 21
Aug 2012
Correspondence to: Dan A. Dixon, Department of Cancer Biology,
University of Kansas Medical Center, Kansas City, KS 66160, USA,
Tel.: 913-945-8120, Fax: þ913-588-4701, E-mail: [email protected]
Can
cerGenetics
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC
International Journal of Cancer
IJC
proteins and can function to stabilize ARE-containingmRNAs when overexpressed in cells.13–15 In the context ofcancer, HuR has been demonstrated to promote expressionof many breast cancer-related genes including COX-2, VEGF,HIF1a, TSP1, ERa, IL-8, Cyclin D1, Cyclin E1, MMP-9 andBRCA-1.16–22 Whereas, TTP has been shown to promotedownregulation of various cytokines (e.g., TNFa, IL-3, IL-6,IL-8, IL-10, IL-12, IL-23 and GM-CSF) and oncogenes andgrowth factors (e.g., COX-2, VEGF, Cyclin D1, uPA, uPAR,MMP-1 and c-Myc).23–34
A critical feature impacting the ability of TTP and HuR tofunction properly occurs through alterations in their respec-tive expression. In various cancer types, including breast can-cer, TTP expression is commonly lost and HuR levels arecommonly upregulated in tumor tissue.12,13 Through thesecombined defects, aberrant mRNA stabilization of ARE-con-taining mRNAs can occur in breast cancer cells leading tooverexpression of growth-promoting genes. However, thegenetic factors contributing to the loss of TTP and HuRoverexpression in breast cancer are not understood. In thisstudy, we have examined novel genetic polymorphisms inZFP36 and ELAVL1 genes and determined their possibleassociations with breast cancer prognosis in two native popu-lations of United States. The genetic variant ZFP36*2 A > Gwas identified as a marker of poorer overall survival in Cau-casian breast cancer patients, with the G allele attenuatingTTP gene expression. These findings indicate a causal role ofthis SNP in prognosis of breast cancer through the suppres-sion of TTP expression and thus allowing for pathogenicgene overexpression during tumor development.
Material and MethodsPatients
This study consisted of 251 histologically confirmed primarybreast cancer patients (170 Caucasian and 81 African–American). Patient recruitment and demographic/clinicaldata retrieval was accomplished with University of SouthCarolina Cancer Research Center Biorepository in collabora-tion with Palmetto Health Tissue Bank, Columbia, SC.Patients who had received their treatment in Palmetto HealthCenter during the period from 2001 to 2005 and followed upuntil March, 2011 were included. Patients were followed at 6-month intervals through Palmetto Health Tumor Bank regis-try from the time of enrollment until the end of study or thepatients’ final outcomes (death). All-cause deaths were con-sidered as events for survival analysis. Study approval was
obtained from the Institutional Review Board of Universityof South Carolina.
Available clinicopathological data include: (i) patient-relatedcharacteristics (e.g., age, race, family history, tobacco/alcoholintake habit), (ii) clinical follow-up (e.g., treatment regimen,overall survival) and (iii) tumor-based properties (e.g., side ofpaired organ, pathology, grade, stage, size, tumor marker sta-tus), along with tumor marker status including estrogen/pro-gesterone receptor (ER/PR) and human epidermal growth fac-tor receptor (HER)-2/neu. All patients enrolled underwentsurgical treatment with or without systemic treatments(including chemo/radio/hormone therapies); delays werereported during diagnosis to first surgical/systemic treatments.The impact of these delays upon patient survival was calcu-lated by adding ‘‘delay in surgical treatment’’ and ‘‘delay insystemic treatment’’ to yield ‘‘total delay.’’
DNA extraction and genotyping
Human breast tumors and histologically normal tissue wereobtained from surgical remnants through the University ofSouth Carolina Cancer Research Center Biorepository. Tissuewas snap-frozen in liquid nitrogen and kept at �80�C untilprocessed. Blood was collected from 25 patients (17 Cauca-sian and 8 African–American), for those the tissue was notavailable. All patients were informed and had provided writ-ten consent. Genomic DNA was extracted from 50 mg of tis-sue samples (178 histologically normal and 48 tumor sam-ples) and blood samples using Qiagen DNA mini kitaccording to the vendor’s protocol (Qiagen, Valencia, CA)and quantitated using a NanoDrop analyzer (Thermo Scien-tific, Wilmington, DE).
Genotyping of ZFP36 and ELAVL1 SNPs involved PCRamplification followed by restriction fragment length poly-morphism (PCR-RFLP) and/or DNA sequencing. Details forgenotyping primers and restriction enzymes used are given inSupporting Information Table 1. As a quality control mea-sure, 5% of cases from each genotype that were assayed byPCR-RFLP were randomly selected for sequencing and theresults were in 100% of concordance.
RNA extraction and qPCR
Total RNA was isolated from 50 mg of histologically normalbreast tissue samples using Trizol reagent (Invitrogen,Carlsbad, CA). Complementary DNA (cDNA) synthesis wasperformed using 1 lg of total RNA in combination with oli-go(dT) and Improm-II reverse transcriptase (Promega, Madi-son, WI). Real-time PCR (qPCR) analysis was performed as
What’s new?
RNA-binding proteins control important genes associated with breast cancer pathogenesis through post-transcriptional
regulation. The authors identify a new single nucleotide polymorphism in the gene encoding for the mRNA decay factor TTP
(ZFP36*2 A>G) that is significantly associated with poor prognosis in patients with breast cancer. This SNP functions to
attenuate expression of TTP, allowing for increased expression of pro-inflammatory factors. These results point to ZFP36*2 as
a genetic marker that may help identify breast cancer patients at high risk for poor disease outcome.
Can
cerGenetics
2 TTP and HuR gene polymorphisms and breast cancer
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC
described25 using Taqman probes for TTP (ZFP36), COX-2(PTGS2) and GAPDH purchased from Applied Biosystems(Foster City, CA) using the 7300 PCR Assay System (AppliedBiosystems); GAPDH was used as control for normalization.Mean of fold changes for all the genotypes were calculatedand compared by independent two-sample t-test.
Protein analysis
Western blots were performed as described25 using a polyclo-nal anti-TTP antibody (ab36558; Abcam, Cambridge, MA).Blots were stripped and then probed with b-actin antibody(Clone C4; MP Biomedicals, Aurora, OH). Detection andquantitation of blots were carried out as previouslydescribed.25 Cell lysates (50 lg/sample) were obtained fromCaucasian normal breast tissue samples (ZFP36*2 AA andGG genotypes). Tissue was homogenized in M-PER mamma-lian protein extraction reagent (Thermo Scientific, Wilming-ton, DE) supplemented with protease inhibitors (Sigma, 50Xprotease inhibitor cocktail). Samples were dounced in micro-centrifuge tubes, kept on ice for 30 min and centrifuged at13,000 rpm for 30 min at 4�C.
Statistical analysis
A comparison between two populations for different variables(clinical, genetic and exposure with environmental risk fac-tors) were performed by cross tabulation and chi-square test.Mean age of onset and total delay in treatment were com-pared through independent two-sample t-test. Demographicand clinical characteristics of patients were stated as percen-tages or summary measures. The primary outcome for thisstudy was overall survival (OS) which was estimated usingthe Kaplan-Meier method. A log-rank test was used to assessthe association between the factors and OS. Univariate Cox’s-regression analysis was used to assess the association betweeneach potential prognostic factor and OS. Factors found to berelatively significant (p < 0.1) in the univariate analysis wereincluded in the multivariate Cox’s proportional hazardsregression model to evaluate the effect of different variableson OS with adjustments for age and known prognostic fac-tors of tumor. The relative risk (hazard ratio [HR]) and 95%CI were calculated from the Cox model for all significantpredictors from cancer diagnosis to the end point of study(event). Analyses were also conducted after stratifying thedata by cancer prognostic factors to examine the potentialinteractive effects. A two-tailed p-value of < 0.05 was consid-ered significant. Due to the exploratory nature of this study,no attempt was made to correct for multiplicity of analysesand nominal p values were reported.
Statistical tests for survival analyses were performed usingSPSS software version 15.0 (SPSS, Chicago, IL). Haplotypeswere constructed, linkage disequilibria were measured and D0
values were calculated to measure indices of linkage disequili-brium (LD) using SNPAnalyzer Version 1.0 (ISTECH). Hap-lotypes were compared between dichotomized patients (withan OS time of 5 or less years versus patients with an OS timeof more than 5 years).
To maintain quality control, Levene’s test for equality ofvariance was performed before the comparison of means toassess the assumption of the equality of variances in differentsamples. In addition, the ‘‘proportional hazard model’’assumption by ‘‘log-minus-log’’ survival plot for Cox-regres-sion was evaluated and found that survival lines do not inter-sect indicating that the ‘‘proportional hazard assumption’’ wassatisfied and therefore this study was not subjected to time de-pendent correlation for Cox regression to analyze the data.
ResultsSurvival analysis and comparison of clinical characteristics
between Caucasian and African–American breast cancer
patients
The distribution of demographic and clinical characteristicsin breast cancer patients are summarized in Table 1. Therewere 170 Caucasian patients and 81 African–Americanpatients and the mean age of onset was significantly higherin Caucasian than African–American breast cancer patients(60.42 years vs. 52.54 years, respectively; p ¼ 1.80 � 10–6).However, we found a significant higher mean of total delayin treatment in African–American than Caucasian breast can-cer patients (p ¼ 0.009). Even though the status of survival(live vs. dead) and 5-year survival (�5 year survival vs. >5year survival) was similar between both of the populations,the median survival was poorer in African–American thanCaucasian breast cancer patients (116 months vs. 124months). In addition, various other factors such as higher tu-mor grade, ER/PR negativity, eligibility for chemotherapy,non-eligibility for hormone therapy, less frequency of familialcancer, nondrinking habit of alcohol and higher Elston histo-logical score were significantly more prevalent in African–American than Caucasian breast cancer patients (Table 1).
Patients’ survival was taken as continuous variable in theanalysis with clinical characteristics (log-rank test); however,for survival analysis of haplotypes dichotomous survival datawas used. ‘‘Higher extent of disease at diagnosis,’’ ‘‘higher tu-mor grade,’’ and ‘‘higher Elston’s score’’ were found toimpart significant negative effect on survival of breast cancerpatients in both Caucasian and African–American popula-tions. Whereas, ‘‘higher AJCC staging,’’ ‘‘ER negativity,’’ ‘‘nohormone therapy’’ and ‘‘no radio therapy’’ variables were sig-nificantly associated with poor survival in Caucasian breastcancer patients (Table 1).
ZFP36 (TTP) and ELAVL1 (HuR) gene polymorphisms in
Caucasian and African–American breast cancer patients
and their association with survival outcome
In this study, ZFP36 and ELAVL1 SNPs were selected basedon their minor allele frequencies (MAFs) in normal Cauca-sian and African–American populations according to dbSNP,SNP’s for which stratified data was unavailable, global MAFswere used. Based on this, three ZFP36 gene polymorphismsand seven ELAVL1 gene polymorphisms that had >5%
Can
cerGenetics
Upadhyay et al. 3
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC
Table
1.
Un
iva
ria
tesu
rviv
al
an
aly
sis
of
clin
ica
lch
ara
cte
rist
ics
inC
au
casi
an
(CA
)a
nd
Afr
ica
n–
Am
eri
can
(AA
)b
rea
stca
nce
rp
ati
en
tsa
nd
com
pa
risi
on
be
twe
en
bo
thra
ces
Characteristics
Caucasians(N
¼170)
African–Americans(N
¼81)
Comparative
pvalue(CAvs.AA)
N(%
)Logrank
pvalue
N(%
)Logrank
pvalue
Me
an
ag
eo
fo
nse
t(y
ea
rs6
SD
)6
0.4
26
14
.52
–5
2.5
46
10
.39
–1.80�
10�6(t-test)
Me
an
of
tota
ld
ela
yin
tre
atm
en
t1(d
ays
6S
D)
66
.786
47
.33
–8
8.1
66
68
.40
–0.009(t-test)
Sta
tus
of
surv
iva
l(l
ive
:d
ea
d)
11
8(6
9.4
%):
52
(30
.6%
)–
56
(69
.1%
):2
5(3
0.9
%)
–0
.96
43
Me
dia
nsu
rviv
al
(mo
nth
s)1
24
–1
16
––
5ye
ars
surv
iva
l(�
5:>
5)
62
(36
.5%
):1
08
(63
.5%
)–
30
(37
%):
51
(63
%)
–0
.92
87
Re
curr
en
ceo
fd
ise
ase
28
(16
.5%
)–
12
(14
.8%
)–
0.7
37
9
Ext
en
to
fd
ise
ase
at
dia
gn
osi
s2(l
oca
lize
d:
reg
ion
al
toly
mp
hn
od
es:
dis
tan
tm
eta
sta
sis)
10
1(6
2.0
%):
51
(31
.3%
):1
1(6
.7%
)8.5�10�10
39
(50
.6%
):2
8(3
6.4
%):
10
(13
.0%
)0.002
0.1
43
5
Tum
or
gra
de
(we
ll:
mo
d:
po
or:
un
de
term
ine
d)
34
(20
.0%
):6
6(3
8.8
%):
67
(39
.4%
):3
(1.8
%)
0.015
5(6
.2%
):2
3(2
8.4
%):
52
(64
.2%
):1
(1.2
%)
0.022
0.0013
AJC
Cst
ag
ing
2(S
tag
eI:
II:I
II)
57
(35
.8%
):7
0(4
4.0
%):
32
(20
.1%
)4.46�
10�5
22
(27
.5%
):3
6(4
5.0
%):
22
(27
.5%
)0
.10
90
.29
89
Tum
or
size
(dia
me
ter)
inm
m(m
ea
n6
SD
)2
6.7
26
24
.47
7–
29
.596
30
.87
4–
0.4
29
ER
sta
tus
(po
siti
ve:
ne
ga
tive
:m
issi
ng
)1
21
(71
.2%
):4
5(2
6.5
%):
4(2
.4%
)0.010
38
(46
.9%
):4
1(5
0.6
%):
2(2
.5%
)0
.13
90.0007
PR
sta
tus
(po
siti
ve:
ne
ga
tive
:m
issi
ng
)9
7(5
7.1
%):
67
(39
.4%
):6
(3.5
%)
0.2
21
33
(40
.7%
):4
6(5
6.8
%):
2(2
.5%
)0
.43
50.0352
ER
/PR
sta
tus2
ERþ
/PRþ
95
/16
4(5
7.9
2%
)0
.06
22
9/7
9(3
6.7
%)
0.4
14
0.0017
ERþ
/PR�
24
/16
4(1
4.6
3%
)9
/79
(11
.4%
)
ER�
/PRþ
2/1
64
(1.2
1%
)4
/79
(5.1
%)
ER�
/PR�
43
/16
4(2
6.2
2%
)3
7/7
9(4
6.8
%)
HE
R2
/ne
ust
atu
s2(n
eg
ati
ve/b
ord
erl
ine
/p
osi
tive
/un
kn
ow
n)
11
0(6
4.7
%):
9(5
.3%
):2
0(1
1.8
%):
31
(18
.2%
)0
.16
15
7(7
0.4
%):
5(6
.2%
):7
(8.6
%):
12
(14
.8%
)0
.66
00
.74
99
Ch
em
oth
era
py
(giv
en
:n
ot
giv
en
)9
5(5
5.9
%):
75
(44
.1%
)0
.20
76
3(7
7.8
%):
18
(22
.2%
)0
.64
90.0008
Ho
rmo
ne
the
rap
yg
ive
no
rn
ot
10
5(6
1.8
%):
65
(38
.2%
)5.46�
10�7
37
(45
.7%
):4
4(5
4.3
%)
0.2
09
0.0162
Ra
dio
the
rap
yg
ive
no
rn
ot
84
(49
.4%
):8
6(5
0.6
%)
0.015
43
(53
.1%
):3
8(4
6.9
%)
0.6
51
0.5
86
4
Kn
ow
nfa
mil
yh
isto
ryo
fca
nce
r1
34
(78
.8%
)0
.48
95
4(6
6.7
%)
0.0
86
0.0378
Fam
ily
his
tory
of
bre
ast
can
cer
88
(51
.8%
)0
.95
02
7(3
3.3
%)
0.7
75
0.0061
Tob
acc
osm
ok
ing
(no
nsm
ok
er:
smo
ke
r:u
nk
no
wn
)8
3(4
8.8
%):
77
(45
.3%
):1
0(5
.9%
)0
.64
64
7(5
8.0
%):
30
(37
.0%
):4
(4.9
%)
0.6
03
0.3
94
2
Alc
oh
ol
inta
ke
(dri
nk
er:
no
nd
rin
ke
r:u
nk
no
wn
)7
1(4
1.8
%):
40
(23
.5%
):5
9(3
4.7
%)
0.6
76
37
(45
.7%
):3
3(4
0.7
%):
11
(13
.6%
)0
.81
20.00069
Els
ton
gra
de
2(G
rad
eI:
Gra
de
II:
Gra
de
III)
28
(21
.2%
):5
4(4
0.9
%):
50
(37
.9%
)0.020
4(5
.8%
):2
0(2
9.0
%):
45
(65
.2%
)0.034
0.00039
Late
rali
tyo
fp
air
ed
org
an
(rig
ht:
left
:b
oth
)8
0(4
7.1
%):
88
(51
.8%
):2
(1.2
%)
0.1
49
40
(49
.4%
):3
9(4
8.1
%):
2(2
.5%
)0
.26
00
.67
74
1To
tal
de
lay
intr
ea
tme
nt¼
tim
efr
om
dia
gn
osi
sto
surg
ery
þti
me
fro
md
iag
no
sis
tofi
rst
syst
em
ictr
ea
tme
nt.
2S
om
ed
ata
mis
sin
g;
sig
nifi
can
tva
lue
ssh
ow
nin
bo
ld.
Can
cerGenetics
4 TTP and HuR gene polymorphisms and breast cancer
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC
variant allelic frequency in corresponding control populationwere chosen (Supporting Information Table 1) and examinedin genomic DNA extracted from breast tumor and histologi-cally normal tissue obtained from surgical remnants; genomicDNA extracted from blood samples was used in 25 caseswhere tissue was not available. To our knowledge, there areno reports to indicate these SNPs result as a tumor-associatedsomatic mutation and DNA sequencing of PCR productsderived from tumor tissue found no tumor-associated so-matic mutations to occur in the ZFP36*2 polymorphic site orthe surrounding sequence constituting the restriction site(data not shown).
The distributions of ZFP36 and ELAVL1 genotypes of allselected SNPs were consistent with Hardy-Weinbergequilibrium in both Caucasian and African–American breastcancer patients, except for ELAVL1 rs35986520 genotypes inCaucasian breast cancer patients. Genotypic frequencies of allselected SNPs are shown in Table 2. We compared genotypicfrequencies between the two populations studied to see theethnic variability for selected SNPs and found that distribu-tion of genotypes for most of the SNPs (rs251864,rs17879933, rs12983784, rs14394, rs12985234 and rs2042920)was significantly different in both populations (Table 2). Oneof the SNPs in ELAVL1 (rs74369359) was detected to bemonomorphic in all Caucasian and African–American breastcancer patients. Genotypic frequencies of one ZFP36*8(rs3746083) and two ELAVL1 gene polymorphisms(rs35986520 and rs10402477) were not significantly differentin Caucasian or African–American breast cancer patients.
To check the independent effect of each SNP on survivalof breast cancer patients, we performed log-rank test andestimated hazard for death using univariate Cox regressionanalysis. Since the frequencies of homozygous variants werelow, especially with final outcome of disease, minor allele-containing genotypes were grouped according to the domi-nant model and the data was analyzed according to both log-additive as well as dominant models (Table 2). Out of all 10SNPs evaluated in ZFP36 and ELAVL1 genes, only oneZFP36 gene polymorphism ZFP36*2 (rs251864) was found tobe significantly associated with poor survival of breast cancerpatients in Caucasian population but not with African–Amer-ican population. As shown in Figure 1a and Table 2, Cauca-sian breast cancer patients who are carriers of ZFP36*2 G al-leles (AG þ GG) were found to be at a two-fold morehazard for death than those carrying AA genotypes (HR ¼2.03; 95% CI ¼ 1.09–3.76; p ¼ 0.025; log-rank p ¼ 0.022).This in contrast to African–American breast cancer patientswhere this effect of the presence of ZFP36*2 G allele was notobserved to impact overall survival (HR ¼ 1.20; 95% CI ¼0.45–3.23; p ¼ 0.711; log-rank p ¼ 0.710; Figure 1b).
Multivariate analysis for survival outcome
Based on the univariate survival analysis indicating the presenceof multiple factors which could affect patient survival, multivari-ate Cox regression analysis was employed to identify important
factors associated with overall survival in Caucasian patients(Table 3). Significant modulators of survival arrived throughmultivariate analysis were ZFP36*2 A > G gene polymorphism,age of disease diagnosis, tumor grade and AJCC staging andhormone therapy received. Furthermore, interactions betweenall significant variables were performed but none were detectedas significant, indicating that these are independent prognosticfactors for breast cancer in Caucasian patients (Table 3). Whenperformed in African–American breast cancer patients, multi-variate Cox regression did not identify any factors to be signifi-cantly associated with patient survival (data not shown).
ZFP36*2 A > G gene polymorphism and its effect on gene
expression
ZFP36*2 polymorphism exists within the promoter region ofTTP and the presence of the minor G allele can inhibit pro-moter activity,35 suggesting that ZFP36*2 genetic variation couldbe a factor contributing to the loss of TTP expression. To assessthis, normal tissue of Caucasian breast cancer patients weregenotyped for ZFP36*2, and TTP protein and mRNA levelswere assayed. As shown in Figure 2a, TTP protein was detectedin three per four samples with the ZFP36*2 AA genotype,whereas limited expression was observed in tissues bearing theZFP36*2 GG genotype. In agreement, there was a significantdecrease in TTP mRNA in tissue that correlated with the pres-ence of the G allele. Heterozygote carriers for the ZFP36*2 AGgenotype and homozygotes for the ZFP36*2 GG genotype werefound to express 0.41- and 0.31-fold less TTP mRNA comparedwith the ZFP36*2 AA genotypes, respectively (Fig. 2b).
Previous work has demonstrated overexpression of theprostaglandin synthase COX-2 to be a factor in breast cancerpathogenesis.37,38 The COX-2 mRNA contains an AREwithin its 30UTR39 and based on our previous findings dem-onstrating the ability of TTP to target COX-2 mRNA forrapid degradation,25 we hypothesized that COX-2 expressionlevels would be inversely correlated with ZFP36*2 genotype.Shown in Figure 2b, COX-2 mRNA levels were increased intissue samples from ZFP36*2 G allele carriers, with ZFP36*2AG heterozygotes and GG homozygotes showing a 2.5-foldand 5.2-fold increase in COX-2 expression compared withthe ZFP36*2 AA genotypes, respectively. These findings indi-cate that the presence of ZFP36*2 G allele attenuates TTPexpression in breast tissue allowing for enhanced expressionof the TTP target gene COX-2.
ZFP36 and ELAVL1 haplotypes and their impact on survival
Since the ZFP36 and ELAVL1 genes are located on the samechromosome, we tested LD for all possible pairs of loci (Sup-porting Information Table 2). Both populations had variableLD scores and significance. Intragenic loci displayed a highdegree of LD with greater significance, whereas only oneintergenic locus showed significant LD. Haplotype frequen-cies were estimated and compared between patients with an
Can
cerGenetics
Upadhyay et al. 5
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC
Table
2.
Dis
trib
uti
on
of
ZFP
36
(TTP
)a
nd
ELA
VL1
(Hu
R)
ge
ne
po
lym
orp
his
ms
inC
au
casi
an
(CA
)a
nd
Afr
ica
n–
Am
eri
can
(AA
)b
rea
stca
nce
rp
ati
en
tsa
nd
un
iva
ria
tesu
rviv
al
an
aly
sis
Genotypes/alleles
Caucasians(N
¼170)
African–Americans(N
¼81)
Comparative
pvalue
(CAvs.AA)
N(%
)live:dead
HR(95%CI)pvalue
Logrank
pvalue
N(%
)Live:
dead
HR(95%CI)Pvalue
Logrank
pvalue
ZFP36
*2A>
G(r
s25
18
64
)A
A7
0(4
1.1
8)
55
:15
Re
f.1
0.0477
19
(23
.46
)1
4:
5R
ef.
10
.81
90.0006
AG
80
(47
.06
)4
9:3
12.19(1.15–4.14)0.016
38
(46
.91
)2
5:
13
1.3
2(0
.47
–3
.72
)0
.60
3
GG
20
(11
.76
)1
4:6
1.5
0(0
.58
–3
.92
)0
.40
42
4(2
9.6
3)
17
:7
1.0
4(0
.33
–3
.29
)0
.94
4
AGþ
GG
10
0(5
8.8
2)
63
:37
2.03(1.09–3.76)0.025
0.022
62
(76
.54
)4
2:
20
1.2
0(0
.45
–3
.23
)0
.71
10
.71
0
ZFP36
*8C>
T(r
s37
46
08
3)
CC
15
5(9
1.1
8)
11
1:4
4R
ef.
10
.43
47
4(9
1.3
6)
52
:2
2R
ef.
10
.38
70
.78
35
CT
14
(8.2
3)
7:7
1.6
8(0
.75
–3
.74
)0
.20
47
(8.6
4)
4:
31
.69
(0.5
0–
5.7
2)
0.3
94
TT1
(0.5
9)
0:1
1.2
0(0
.13
9–
10
.33
5)
0.8
68
0–
NC
CTþ
TT1
5(8
.82
)7
:81
.61
(0.7
5–
3.4
6)
0.2
21
0.2
16
7(8
.64
)4
:3
1.6
9(0
.50
–5
.72
)0
.39
40
.38
7
ZFP36
*10
2b
pd
ele
tio
n(r
s17
87
99
33
)II
12
4(7
2.9
4)
88
:36
Re
f.1
0.2
37
77
(95
.06
)5
3:
24
Re
f.1
0.9
07
0.0002
ID4
3(2
5.2
9)
27
:16
1.5
1(0
.83
–2
.74
)0
.17
54
(4.9
4)
3:
10
.89
(0.1
2–
6.5
8)
0.9
07
DD
3(1
.77
)3
:0N
C0
NC
IDþ
DD
46
(27
.06
)3
0:1
60
.72
(0.3
9–
1.2
9)
0.2
68
4(4
.94
))3
:1
0.8
9(0
.12
–6
.58
)0
.90
70
.90
7
ELA
VL1
rs1
29
83
78
4T>
CTT
93
(54
.71
)6
2:3
1R
ef.
10
.89
15
8(7
1.6
0)
39
:1
9R
ef.
10
.52
20.0130
CT
63
(37
.06
)4
6:1
70
.86
(0.4
7–
1.5
7)
0.6
32
22
(27
.16
)1
6:
60
.73
(0.2
9–
1.8
3)
0.4
98
CC
14
(8.2
3)
10
:40
.95
(0.3
3–
2.6
9)
0.9
17
1(1
.24
)1
:0
NC
CTþ
TT7
7(4
5.2
9)
56
:21
0.8
8(0
.50
–1
.54
)0
.65
10
.65
02
3(2
8.4
0)
17
:6
0.6
7(0
.26
–1
.69
)0
.39
50
.39
1
ELA
VL1
rs1
43
94
T>
CTT
86
(50
.59
)6
3:2
3R
ef.
10
.70
65
5(6
7.9
0)
40
:1
5R
ef.
10
.46
90.0150
TC7
2(4
2.3
5)
47
:25
1.2
6(0
.71
–2
.26
)0
.42
52
5(3
0.8
6)
15
:1
01
.54
(0.6
9–
3.4
3)
0.2
92
CC
12
(7.0
6)
8:4
1.2
7(0
.44
–3
.67
)0
.66
11
(1.2
4)
1:0
NC
TCþ
CC
84
(49
.41
)5
5:2
91
.27
(0.7
3–
2.2
1)
0.4
07
0.4
04
26
(32
.10
)1
6:
10
1.4
7(0
.66
–3
.27
)0
.34
70
.34
4
ELA
VL1
rs7
43
69
35
9G>
CG
G1
70
(10
0)
11
8:5
2R
ef.
1N
C8
1(1
00
)5
6:
25
Re
f.1
NC
NC
GCþ
CC
0–
NC
0N
C
ELA
VL1
rs3
59
86
52
01
G>
AG
G1
44
(85
.71
)9
9:4
5R
ef.
10
.72
57
4(9
2.5
)5
2:
22
Re
f.1
0.2
21
0.2
02
7
GA
20
(11
.91
)1
5:5
0.6
9(0
.27
–1
.77
)0
.44
66
(7.5
)3
:3
2.0
9(0
.62
–7
.02
)0
.23
2
AA
4(2
.38
)3
:10
.76
(0.1
1–
5.5
6)
0.7
92
0–
NC
GAþ
AA
24
(14
.29
)1
8:6
0.7
1(0
.29
–1
.67
)0
.42
70
.42
46
(7.5
)3
:3
2.0
9(0
.62
–7
.02
)0
.23
20
.22
1
ELA
VL1
rs1
04
02
47
71
C>
TC
C1
62
(96
.43
)1
12
:50
Re
f.1
0.5
51
73
(91
.25
)5
1:
22
Re
f.1
NC
0.1
36
6
CT
6(3
.57
)5
:10
.55
(0.0
7–
4.0
1)
0.5
58
6(7
.5)
4:
21
.0(0
.23
–4
.37
)1
.00
TT0
–N
C1
(1.2
5)
0:1
NC
CTþ
TT6
(3.5
7)
5:1
0.5
5(0
.07
–4
.01
)0
.55
80
.55
17
(8.7
5)
4:
31
.63
(0.4
8–
5.4
8)
0.4
21
ELA
VL1
rs1
29
85
23
41
A>
GA
A8
8(5
2.3
8)
64
:24
Re
f.1
0.8
68
55
(68
.75
)4
0:
15
Re
f.1
0.3
50
0.0183
Can
cerGenetics
6 TTP and HuR gene polymorphisms and breast cancer
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC
OS time of 5 or less years versus patients with a greaterthan 5 year OS. Five haplotypes were constructed forZFP36 gene, but none were found to be associated with OS(global p ¼ 0.1089 and 0.1337 for Caucasian and African–American breast cancer patients, respectively). Similarly, forthe 17 haplotypes constructed with ELAVL1, none of themwas associated with OS (global p ¼ 0.8978 and 0.4923 forCaucasian and African–American breast cancer patients,respectively; Supporting Information Table 3). Ta
ble
2.
Dis
trib
uti
on
of
ZFP
36
(TTP
)a
nd
ELA
VL1
(Hu
R)
ge
ne
po
lym
orp
his
ms
inC
au
casi
an
(CA
)a
nd
Afr
ica
n–
Am
eri
can
(AA
)b
rea
stca
nce
rp
ati
en
tsa
nd
un
iva
ria
tesu
rviv
al
an
aly
sis
(Co
nti
nu
ed
)
Genotypes/alleles
Caucasians(N
¼170)
African–Americans(N
¼81)
Comparative
pvalue
(CAvs.AA)
N(%
)live:dead
HR(95%CI)pvalue
Logrank
pvalue
N(%
)Live:
dead
HR(95%CI)Pvalue
Logrank
pvalue
AG
67
(39
.88
)4
4:2
31
.17
(0.6
5–
2.1
0)
0.6
00
24
(30
.00
)1
4:
10
1.6
9(0
.76
–3
.77
)0
.19
7
GG
13
(7.7
4)
9:4
1.1
2(0
.39
–3
.22
)0
.84
01
(1.2
5)
1:0
NC
AGþ
GG
80
(47
.62
)5
3:2
71
.16
(0.6
6–
2.0
4)
0.6
02
0.6
01
25
(31
.25
)1
5:
10
1.6
1(0
.72
–3
.58
)0
.24
40
.23
9
ELA
VL1
rs2
04
29
20
T>
GTT
11
5(6
7.6
5)
84
:31
Re
f.1
0.4
48
73
(90
.12
)5
0:
23
Re
f.1
0.7
40
TG4
9(2
8.8
2)
30
:19
1.4
5(0
.81
–2
.58
)0
.21
08
(9.8
8)
6:
20
.78
(0.1
8–
3.3
3)
0.7
41
GG
6(3
.53
)4
:21
.24
(0.2
9–
5.1
9)
0.7
72
0–
NC
0.0005
TGþ
GG
55
(32
.35
)3
4:2
11
.42
(0.8
1–
2.4
9)
0.2
16
0.2
13
8(9
.88
)6
:2
0.7
8(0
.18
–3
.33
)0
.74
10
.74
0
NC
,N
ot
calc
ula
ted
;si
gn
ifica
nt
valu
es
sho
wn
inb
old
.1G
en
oty
pin
gco
uld
no
tb
ed
on
ein
two
Ca
uca
sia
na
nd
on
eA
fric
an
–A
me
rica
nb
rea
stca
nce
rp
ati
en
ts.
Figure 1. Kaplan-Meier survival curves in Caucasian and African–
American breast cancer patients according to ZFP36*2 A > G
genotypes: AA versus AG þ GG genotype. Vertical ticks show
censored cases and each step down represents an event (death).
(a) Survival curves for Caucasian breast cancer patients. (b)
Survival curves for African–American breast cancer patients.
Median survivals could not be calculated for African–American
breast cancer patients so mean survivals are indicated.
Can
cerGenetics
Upadhyay et al. 7
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC
Combined effect of variant genotypes on patient survival
To detect the combined effect of multiple risk genotypes ofZFP36 and ELAVL1 gene polymorphisms, we categorizedpatients according to number of risk genotypes present ineach patient. Genotypes showing HR > 1 in dominant modelwere considered as risk genotypes (Table 2). For ZFP36 genepolymorphisms, patients were categorized into ‘‘0–1’’ vs. ‘‘2–3’’ risk genotype carriers, however for ELAVL1 gene poly-morphisms, patients were categorized into ‘‘0–3’’ vs. ‘‘4–7’’risk genotype carriers. Furthermore, patients were also com-pared as ‘‘0–2’’ vs. ‘‘3–5’’ or ‘‘6–8’’ total risk genotype carriers(Table 4). We found that, in Caucasian breast cancer patients‘‘2–3’’ ZFP36 risk genotype carriers show poorer survival(HR ¼ 1.79; 95%CI ¼ 1.04–3.11; p ¼ 0.037; log-rank p value¼ 0.034) than ‘‘0–1’’ ZFP36 risk genotype carriers and thisrisk increases in patients who carry total number of ‘‘6–8’’risk genotypes of ZFP36 and ELAVL1 gene polymorphisms(HR ¼ 2.42; 95% CI ¼ 1.17–4.99; p ¼ 0.017; log-rank p ¼0.007). The P-trend analysis also showed that with increasednumber of risk genotypes, a respective increase in Caucasianbreast cancer patient HR was observed (ptrend ¼ 0.0109). TheAfrican–American breast cancer patients did not show simi-lar trend of poor prognosis with increased number of totalrisk genotypes, however, when we compared ‘‘0–3’’ versus‘‘4–7’’ ELAVL1 risk genotype carriers, synergic effect ofELAVL1 gene polymorphisms (more than three risk geno-types) appeared to have borderline modest effect in prognosisof African–American breast cancer patients (HR ¼ 2.18; 95%CI ¼ 0.94–5.08; p ¼ 0.070; log-rank p ¼ 0.063).
DiscussionBreast cancer is the most common type of malignant canceramong women, with various factors contributing to its highmortality rate.1 While current efforts utilizing gene expres-sion profiling (e.g., Oncotype DX and MammaPrint) are
gaining acceptance as clinical predictors,40 further insightinto the genetic causes underlying pathogenic gene expressionin breast cancer is needed. Various gene products associatedwith promoting the various facets of tumorigenesis are fre-quently overexpressed in cancer cells. A consistent featurepresent within these gene transcripts is the ARE sequence.However, the ability of the ARE to target these mRNAs forpost-transcriptional regulation is defective in tumor cells,allowing for aberrant gene overexpression and the acquisitionof neoplastic traits during breast cancer development.8
Through their ability to bind ARE sequences, the RNA-binding proteins TTP and HuR are pleiotropic regulators ofseveral genes associated with breast cancer.17–20,34 Whilechanges in the expression pattern of HuR and TTP are com-monly observed during tumorigenesis resulting in enhancedmRNA stabilization,12,13 the underlying causes promotingthese changes are not well understood. Here, we examinedwhether 10 common genetic polymorphisms present in HuRand TTP genes (ELAVL1 and ZFP36, respectively) play a piv-otal role influencing corresponding gene expression andmore significantly, impact disease outcomes using a cohort of251 breast cancer patients of Caucasian and African–Ameri-can origins. To our knowledge, this is the first study explor-ing the association of genetic variations in ELAVL1 andZFP36 genes with cancer patient outcomes.
Several studies have shown that African–Americans havepoor prognosis in comparison with Caucasian breast cancerpatients. Various reasons underlying this difference can beattributed to social, environmental and genetic factors in Afri-can–American breast cancer patients.4,5,41–44 In this study, wealso found poor survival in African–Americans than Caucasianbreast cancer patients (116 months vs. 124 months) and com-parison between clinical characteristics of patients suggestedthat treatment delay, higher grade and extent of tumor andER/PR negativity were primary clinical factors contributing topoor African–American patient survival outcomes. Our
Table 3. Multivariate survival analysis for Caucasian breast cancer patients
Variables HR1 95% CI p value
Age at diagnosis 1.028 1.003–1.054 0.030
Total delay in treatment 0.997 0.988–1.005 0.415
AJCC (2 vs. 1) 1.680 0.681–4.148 0.260
AJCC staging (3 vs. 1) 5.696 2.319–13.989 0.001
Tumor grade (2 þ 3 vs. 1) 5.585 1.061–29.385 0.042
ER status (negative vs. positive) 1.059 0.334–3.363 0.922
Hormone therapy (without vs. with) 3.840 1.517–9.722 0.005
Radiotherapy (without vs. with) 1.602 0.696–3.686 0.268
ZFP36*2 (AG þ GG vs. AA) 3.230 1.510–6.909 0.030
‘‘ZFP36*A > G’’ interaction ‘‘tumor grade 2 þ 3’’ 3.647 0.418–31.882 0.242
‘‘ZFP36*A > G’’ interaction ‘‘AJCC stage III’’ 0.998 0.976–1.021 0.874
‘‘ZFP36*2 A > G’’ interaction ‘‘hormone therapy not given’’ 1.085 0.220–5.345 0.920
1HR is the Hazard ratio; significant values shown in bold.
Can
cerGenetics
8 TTP and HuR gene polymorphisms and breast cancer
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC
findings are consistent with previous reports4–6 except thatthere was a significant difference between mean age of diseaseonset in patients of both races (60.42 years in Caucasians vs.52.54 years in African–Americans; p ¼ 1.80 � 10–6).
Together with the comparison of breast cancer patients’prognosis between two races, this study examined associations
between genetic variations in ELAVL1 and ZFP36 genes andsurvival outcomes. The ELAVL1 gene has more than 400SNPs, whereas ZFP36 has 49 SNPs according to dbSNP(http://www.ncbi.nlm.nih.gov/projects/SNP/). With the major-ity of ELAVL1 SNPs having less than 5% MAF, 7 ELAVL1SNPs were selected that had a MAF >5%. An establishedfunctional role for these selected SNPs has yet to be deter-mined, although their location within the ELAVL1 mRNA sug-gests a possible role in mRNA stability, microRNA recognitionand splicing regulation. Previously, Carrick et al. had exploredgenetic polymorphisms and haplotypes of ZFP36 gene andidentified four specific haplotype-tag SNPs (htSNPs) for Cau-casians and five for African–Americans that were predicted todistinguish 95% or more of the haplotypes.45 However, out ofthese htSNPs only three polymorphisms (ZFP36*2, ZFP36*8and ZFP36*10) have >5% MAF and therefore chosen in thisstudy. ZFP36*2 is a promoter region polymorphism previouslyshown to impact TTP promoter activity using a luciferasereporter assay, with the presence of the G allele inhibitingTTP promoter activity two-fold.35 In order to predict theimpact of ZFP36*2 SNP on TTP gene expression, the ZFP36promoter sequence surrounding this SNP was identified tobind several putative transcription factors whose binding couldbe negatively impacted due to the ZFP36*2 G allele (Fig. 2c).For instance, liver X receptor (LXR), a member of the nuclearreceptor family of transcription factors that plays a role inlipid metabolism, has a putative binding site in the wild-typesequence that could be disrupted when the SNP is present.This is interesting since LXR agonists have been shown to in-hibit expression of inflammatory mediators in cultured macro-phages and be used to limit inflammation.46 In contrast, theZFP36*8 polymorphism in protein coding domain is not pre-dicted to alter the amino acid sequence of TTP, and its func-tional consequence impacts protein translation presumablythrough a rare codon phenomenon.45,47
The genotypic frequencies were consistent with Hardy-Weinberg equilibrium but differ significantly between bothraces for 60% SNPs selected indicating strong ethnic variabil-ity. On analyzing the independent effect of each SNP onpatient survival, ZFP36*2 G allele carriers were found to havesignificantly lower median survival (101 months vs. 132months) and higher risk for death (HR ¼ 2.03; 95% CI ¼1.09–3.76; p ¼ 0.025; log-rank p ¼ 0.022) in comparisonwith ZFP36*2 AA genotype carriers in Caucasian race. Theseresults were still valid for multivariate analysis, however thispolymorphism did not show any significant interaction withother factors, indicating ZFP36*2 A > G gene polymorphismas independent prognostic factor for Caucasian breast cancerpatients. By contrast, this polymorphism was not found to beassociated with survival outcomes of African–Americanpatients.
Suppressed expression of TTP is associated with poorprognosis of breast cancer,29,48 indicating that the poor prog-nosis in ZFP36*2 ‘‘AG þ GG’’ genotype carriers may be dueto lower TTP expression. Our results showed a significant
Figure 2. Relative TTP and COX-2 expression among different
genotypes of ZFP36*2 A > G polymorphism. (a) Protein lysates
isolated from normal Caucasian breast tissue genotyped for the
ZFP36*2 A > G polymorphism were assayed for TTP expression by
western blot. Actin was used as a loading control. (b) Total RNA
was isolated from normal Caucasian breast tissue genotyped for
the ZFP36*2 A > G polymorphism (n ¼ 5 samples of each
genotype) and assayed for TTP and COX-2 mRNA expression by
qPCR. Relative mRNA levels were normalized to GAPDH internal
control. (a) *p ¼ 0.0419 and 0.0206 for AA vs. AG and AA vs. GG
genotypes, respectively. (b) *p ¼ 0.016 for AA vs. GG genotypes.
(c) Schematic representation of the ZFP36 promoter containing the
ZFP36*2 A > G SNP. Transcription factor binding sites containing
the ZFP36*2 A allele (shown in bold) were identified36 and
indicated with the consensus binding motif shown in uppercase.
Can
cerGenetics
Upadhyay et al. 9
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC
difference between mRNA expression between ZFP36*2 AAvs. ZFP36*2 AG and GG genotypes. Furthermore, ZFP36*2genotype-dependent loss of TTP expression was reflected inenhanced expression of the TTP-target mRNA COX-2 andmay additionally may allow for upregulation other TTP-tar-get genes encoding pro-inflammatory cytokines, oncogenesand growth factors. These findings are in agreement with astudy examining rheumatoid arthritis (RA) where a trendwas observed with the ZFP36*2 GG genotype to have anearly age of disease onset compared to the AA/AG genotypes,however no significant differences were observed in ZFP36*2allele frequencies between healthy individuals and RApatients.35 Taken together, these findings indicate the abilitythis SNP to modulate disease activity by negatively impactingexpression of TTP on a transcriptional level and advocateZFP36*2 A > G polymorphism as a new prognostic markerfor breast cancer patients.
This was the first study exploring the role of commongenetic variations in ELAVL1 and ZFP36 genes in prognosisof breast cancer patients. While the findings are comprehen-
sive and make a causal link between pathogenic gene expres-sion and a regulatory SNP in the mRNA decay factor TTP,some limitations should be noted. The main limitation of ourstudy was low sample size especially in African–Americanbreast cancer patients. Along with this limitation was missingdata for some variables such as menopausal status, progres-sion free survival and specific therapeutic details. Nonethe-less, the novel findings presented here provide the basis forfuture similar and replicative studies in larger cohorts. Inconclusion, ZFP36*2 A > G gene polymorphism has emergedas novel prognostic marker for Caucasian breast cancerpatients and ELAVL1 gene polymorphisms may have somecontributing role in determining survival outcome of breastcancer patients.
AcknowledgementsWe thank Dr. Kristin Wallace and Dr. Edsel Pena for critical review of thismanuscript and helpful comments and the University of South CarolinaCancer Research Center Biorepository for providing study samples and clin-ical details.
References
1. Siegel R, Ward E, Brawley O, et al. Cancerstatistics, 2011: the impact of eliminatingsocioeconomic and racial disparities onpremature cancer deaths. CA Cancer J Clin 2011;61:212–36.
2. Lichtenstein P, Holm NV, Verkasalo PK, et al.Environmental and heritable factors in thecausation of cancer—analyses of cohorts of twinsfrom Sweden, Denmark, and Finland. N Engl JMed 2000;343:78–85.
3. Turnbull C, Ahmed S, Morrison J,et al. Genome-wide association study
identifies five new breast cancersusceptibility loci. Nat Genet 2010;42:504–7.
4. Adams SA, Smith ER, Hardin J, et al. Racialdifferences in follow-up of abnormalmammography findings among economicallydisadvantaged women. Cancer 2009;115:5788–97.
5. Smith ER, Adams SA, Das IP, et al. Breast cancersurvival among economically disadvantagedwomen: the influences of delayed diagnosis andtreatment on mortality. Cancer EpidemiolBiomarkers Prev 2008;17:2882–90.
6. Cunningham JE, Montero AJ, Garrett-Mayer E, etal. Racial differences in the incidence of breastcancer subtypes defined by combined histologicgrade and hormone receptor status. CancerCauses Control 2010;21:399–409.
7. Garneau NL, Wilusz J, Wilusz CJ. The highwaysand byways of mRNA decay. Nat Rev Mol CellBiol 2007;8:113–26.
8. Lopez de Silanes I, Quesada MP, Esteller M.Aberrant regulation of messenger RNA 30-untranslated region in human cancer. Cell Oncol2007;29:1–17.
Table 4. Number of risk genotypes and survival of breast cancer patients
Number ofrisk genotypes
Caucasians African–Americans
N (live: dead)HR (95%CI)p value
Log rankp value N (live: dead)
HR (95%CI)p value
Log rankp value
ZFP36
0–1 112 (84:28) Ref. 1 0.034 70 (49:21) Ref. 1 0.531
2–3 58 (34:24) 1.79 (1.04–3.11) 0.037 11 (7:4) 1.41 (0.48–4.12) 0.534
ELAVL1
0–3 112 (82:30) Ref. 1 0.245 64 (47:17) Ref. 1 0.063
4–7 56 (35:21) 1.39 (0.79–2.44) 0.248 16 (8:8) 2.18 (0.94–5.08) 0.070
Total risk genotypes
0–2 54 (41:13) Ref. 1 0.007Ptrend ¼ 0.0109
23 (17:6) Ref. 1 0.256Ptrend ¼ 0.1837
3–5 82 (61:21) 0.95 (0.47–1.92) 0.883 40 (29:11) 1.19 (0.44–3.27) 0.733
6–8 32 (15:17) 2.42 (1.17–4.99) 0.017 17 (9:8) 2.21 (0.77–6.36) 0.143
Significant values shown in bold; underlined values show a borderline significance; risk genotypes for Caucasians: ZFP36*2 AG þ GG, ZFP36*8 CTþ TT, ZFP36*10II, ELAVL1 rs12983784 TT, ELAVL1 rs14394 TC þ CC, ELAVL1 rs35986520 GG, ELAVL1 rs10402477 CC, ELAVL1 rs12985234 AG þGG, ELAVL1 rs2042920 TG þ GG; risk genotype for African–Americans: ZFP36*2 AG þ GG, ZFP36*8 CT þ TT, ZFP36*10 II, ELAVL1 rs12983784 TT,ELAVL1 rs14394 TC þ CC, ELAVL1 rs35986520 GA þ AA, ELAVL1 rs10402477 CT þ TT, ELAVL1 rs12985234 AG þ GG, ELAVL1 rs2042920 TT (riskgenotypes were derived from Table 2).
Can
cerGenetics
10 TTP and HuR gene polymorphisms and breast cancer
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC
9. Gruber AR, Fallmann J, Kratochvill F, et al.AREsite: a database for the comprehensiveinvestigation of AU-rich elements. Nucleic AcidsRes 2010;39:D66–9.
10. Bakheet T, Williams BR, Khabar KS. ARED 3.0:the large and diverse AU-rich transcriptome.Nucleic Acids Res 2006;34:D111–14.
11. Lai WS, Carballo E, Thorn JM, et al. Interactionsof CCCH zinc finger proteins with mRNA.Binding of tristetraprolin-related zinc fingerproteins to AU-rich elements and destabilizationof mRNA. J Biol Chem 2000;275:17827–37.
12. Sanduja S, Blanco FF, Dixon DA. The roles ofTTP and BRF proteins in regulated mRNA decay.Wiley Interdiscip Rev RNA 2010;2:42–57.
13. Abdelmohsen K, Gorospe M. Posttranscriptionalregulation of cancer traits by HuR. WileyInterdiscip Rev RNA 2010;1:214–29.
14. Fan XC, Steitz JA. Overexpression of HuR, anuclear-cytoplasmic shuttling protein, increasesthe in vivo stability of ARE-containing mRNAs.EMBO J 1998;17:3448–60.
15. Brennan CM, Steitz JA. HuR and mRNAstability. Cell Mol Life Sci 2001;58:266–77.
16. Gubin MM, Calaluce R, Davis JW, et al.Overexpression of the RNA binding protein HuRimpairs tumor growth in triple negative breastcancer associated with deficient angiogenesis. CellCycle 2010;9:3337–46.
17. Denkert C, Weichert W, Winzer KJ, et al.Expression of the ELAV-like protein HuR isassociated with higher tumor grade and increasedcyclooxygenase-2 expression in human breastcarcinoma. Clin Cancer Res 2004;10:5580–6.
18. Saunus JM, French JD, Edwards SL, et al.Posttranscriptional regulation of the breastcancer susceptibility gene BRCA1 by the RNAbinding protein HuR. Cancer Res 2008;68:9469–78.
19. Pryzbylkowski P, Obajimi O, Keen JC.Trichostatin A and 5 Aza-20 deoxycytidinedecrease estrogen receptor mRNA stability in ERpositive MCF7 cells through modulation of HuR.Breast Cancer Res Treat 2008;111:15–25.
20. Guo X, Hartley RS. HuR Contributes to cyclin E1deregulation in MCF-7 breast cancer cells. CancerRes 2006;66:7948–56.
21. Yuan Z, Sanders AJ, Ye L, et al. Knockdown ofhuman antigen R reduces the growth andinvasion of breast cancer cells in vitro and affectsexpression of cyclin D1 and MMP-9. Oncol Rep2011;26:237–45.
22. Suswam EA, Nabors LB, Huang Y, et al. IL-1betainduces stabilization of IL-8 mRNA in malignantbreast cancer cells via the 30 untranslated region:involvement of divergent RNA-binding factors
HuR, KSRP and TIAR. Int J Cancer 2005;113:911–19.
23. Carballo E, Lai WS, Blackshear PJ. Evidence thattristetraprolin is a physiological regulator ofgranulocyte-macrophage colony-stimulating factormessenger RNA deadenylation and stability.Blood 2000;95:1891–9.
24. Lai WS, Carballo E, Strum JR, et al. Evidencethat tristetraprolin binds to AU-rich elementsand promotes the deadenylation anddestabilization of tumor necrosis factor alphamRNA. Mol Cell Biol 1999;19:4311–23.
25. Young LE, Sanduja S, Bemis-Standoli K, et al.The mRNA binding proteins HuR andtristetraprolin regulate cyclooxygenase 2expression during colon carcinogenesis.Gastroenterology 2009;136:1669–79.
26. Suswam E, Li Y, Zhang X, et al. Tristetraprolindown-regulates interleukin-8 and vascularendothelial growth factor in malignant gliomacells. Cancer Res 2008;68:674–82.
27. Marderosian M, Sharma A, Funk AP, et al.Tristetraprolin regulates Cyclin D1 and c-MycmRNA stability in response to rapamycin in anAkt-dependent manner via p38 MAPK signaling.Oncogene 2006;25:6277–90.
28. Stoecklin G, Gross B, Ming XF, et al. A novelmechanism of tumor suppression by destabilizingAU-rich growth factor mRNA. Oncogene 2003;22:3554–61.
29. Brennan SE, Kuwano Y, Alkharouf N, et al. ThemRNA-destabilizing protein tristetraprolin issuppressed in many cancers, altering tumorigenicphenotypes and patient prognosis. Cancer Res2009;69:5168–76.
30. Patil CS, Liu M, Zhao W, et al. Targeting mRNAstability arrests inflammatory bone loss. Mol Ther2008;16:1657–64.
31. Stoecklin G, Tenenbaum SA, Mayo T, et al.Genome-wide analysis identifies interleukin-10mRNA as target of tristetraprolin. J Biol Chem2008;283:11689–99.
32. Jalonen U, Nieminen R, Vuolteenaho K, et al.Down-regulation of tristetraprolin expressionresults in enhanced IL-12 and MIP-2 productionand reduced MIP-3alpha synthesis in activatedmacrophages. Mediat Inflamm 2006;2006:40691.
33. Qian X, Ning H, Zhang J, et al.Posttranscriptional regulation of IL-23 expressionby IFN-gamma through tristetraprolin. JImmunol 2011;186:6454–64.
34. Al-Souhibani N, Al-Ahmadi W, Hesketh JE, et al.The RNA-binding zinc-finger proteintristetraprolin regulates AU-rich mRNAs involvedin breast cancer-related processes. Oncogene 2010;29:4205–15.
35. Suzuki T, Tsutsumi A, Suzuki H, et al.Tristetraprolin (TTP) gene polymorphisms inpatients with rheumatoid arthritis and healthyindividuals. Mod Rheumatol 2008;18:472–9.
36. Xu Z, Taylor JA. SNPinfo: integrating GWAS andcandidate gene information into functional SNPselection for genetic association studies. NucleicAcids Res 2009;37:W600–5.
37. Howe LR. Inflammation and breast cancer.Cyclooxygenase/prostaglandin signaling andbreast cancer. Breast Cancer Res 2007;9:210–18.
38. Liu CH, Chang SH, Narko K, et al.Overexpression of cyclooxygenase-2 is sufficientto induce tumorigenesis in transgenic mice. J BiolChem 2001;276:18563–9.
39. Dixon DA, Kaplan CD, McIntyre TM,Zimmerman GA, Prescott SM. Post-transcriptional control of cyclooxygenase-2 geneexpression. The role of the 30-untranslated region.J Biol Chem 2000;275:11750–7.
40. Turaga K, Acs G, Laronga C. Gene expressionprofiling in breast cancer. Cancer Control 2010;17:177–82.
41. Shavers VL, Harlan LC, Stevens JL. Racial/ethnicvariation in clinical presentation, treatment, andsurvival among breast cancer patients under age35. Cancer 2003;97:134–47.
42. Ademuyiwa FO, Olopade OI. Racial differencesin genetic factors associated with breast cancer.Cancer Metastasis Rev 2003;22:47–53.
43. Jones BA, Kasl SV, Howe CL, et al. African-American/White differences in breast carcinoma:p53 alterations and other tumor characteristics.Cancer 2004;101:1293–301.
44. Lantz PM, Mujahid M, Schwartz K, et al. Theinfluence of race, ethnicity, and individualsocioeconomic factors on breast cancer stage atdiagnosis. Am J Public Health 2006;96:2173–8.
45. Carrick DM, Chulada P, Donn R, et al. Geneticvariations in ZFP36 and their possiblerelationship to autoimmune diseases. JAutoimmun 2006;26:182–96.
46. Joseph SB, Castrillo A, Laffitte BA, et al.Reciprocal regulation of inflammation and lipidmetabolism by liver X receptors. Nat Med 2003;9:213–19.
47. Griseri P, Bourcier C, Hieblot C, et al. Asynonymous polymorphism of the tristetraprolin(TTP) gene, an AU-rich mRNA-binding protein,affects translation efficiency and response toherceptin treatment in breast cancer patients.Hum Mol Genet 2011;20:4556–68.
48. Gebeshuber CA, Zatloukal K, Martinez J. miR-29a suppresses tristetraprolin, which is aregulator of epithelial polarity and metastasis.EMBO Rep 2009;10:400–5.
Can
cerGenetics
Upadhyay et al. 11
Int. J. Cancer: 000, 000–000 (2012) VC 2012 UICC