A study of the role of GATA2 gene polymorphism in coronary artery disease risk traits

Gene 544 (2014) 152–158

Contents lists available at ScienceDirect

Gene journal homepage: www.elsevier.com/locate/gene

A study of the role of GATA2 gene polymorphism in coronary artery disease risk traits Nzioka P. Muiya a, Salma Wakil a, Mohammed Al-Najai a, Asma I. Tahir a, Batoul Baz a, Editha Andres a, Olyan Al-Boudari a, Nada Al-Tassan a, Maie Al-Shahid b, Brian F. Meyer a, Nduna Dzimiri a,⁎ a b

Genetics Department, King Faisal Specialist Hospital and Research Centre, P. O. Box 3354, Riyadh 11211, Saudi Arabia King Faisal Heart Institute, King Faisal Specialist Hospital and Research Centre, P. O. Box 3354, Riyadh 11211, Saudi Arabia

a r t i c l e

i n f o

Article history: Received 16 October 2013 Received in revised form 17 March 2014 Accepted 27 April 2014 Available online 28 April 2014 Keywords: GATA2 polymorphism Coronary artery disease Obesity Hyperlipidemia Haplotypes

a b s t r a c t The GATA2 is a multi-catalytic transcription factor believed to play an important role in regulating inflammatory processes, largely contributory to cardiovascular-related events. However, its role in coronary artery disease (CAD) risk traits remains poorly understood. In a preliminary study using Affymetrix 250K, we established a link on chromosome (chr) 3, which harbors the GATA2 gene, to early onset of CAD in two families with heterozygous familial hyperlipidemia (HFH), suggesting a role for the gene in metabolic-related CAD in the general population. We then sequenced the gene in the families and an additional 200 individuals in the general population, followed by an association study for 8 SNPs on CAD metabolic risk traits in a total of 4557 individuals (2386 CAD cases versus 2171 angiographed controls) by the Applied Biosystems real-time PCR system. The rs1573949_C [1.15(1.00–1.32); p = 0.049] was associated with MI, rs7431368_AA [5.2(1.05–26.60); p = 0.43] conferred risk for harboring low high density lipoprotein, and obesity was linked to rs10934857_AA [5.69(1.04–30.98); p = 0.045] following Bonferroni corrections and multivariate adjustments for confounders. Furthermore, a haplotype CCCGGGTC (χ2 = 4.23; p = 0.04) constructed from the eight studied SNPs and its 6-mer derivative CGGGTC (χ2 = 5.05; p = 0.025) were associated with CAD. Obesity was associated with the 6-mer CATAAA (χ2 = 3.66; p = 0.049), and hypercholesterolemia was linked to the 8-mer CCTGGACC (χ2 = 6.02; p = 0.014), but most significantly so with its 5-mer derivative, CTGGA (χ2 = 6.75; p = 0.009). On the other hand, high low density lipoprotein was linked to TGG (χ2 = 4.48; p = 0.034). Our study points to an association of GATA2 at both SNP and haplotype levels with important metabolic risk traits for atherosclerosis. © 2014 Elsevier B.V. All rights reserved.

1. Introduction The formation and development of blood cells are controlled largely by cell-restricted transcription factors (TFs), such as the GATA binding protein family (Fujikura et al., 2002; Morrisey et al., 1997a, 1997b; Narita et al., 1997), which cooperate with more widely expressed factors to direct lineage-specific gene expression (Charron and Nemer, 1999; Molkentin et al., 1997; Pikkarainen et al., 2004; Watt et al.,

Abbreviations: ANOVA, analysis of variance; CAD, coronary artery disease; Chol, cholesterol; hChol, hypercholesterolemia; C.I., confidence interval; S.E., standard error; CON, control; FH, familial hyperlipidemia; hTG, hypertriglyceridemia; HDL-C, high density lipoprotein-cholesterol; HFH, heterozygous familial hypercholesterolemia; HTN, hypertension; LD, linkage disequilibrium; LDL-C, low density lipoprotein-cholesterol; lHDLC, low high density lipoprotein-cholesterol; MI, myocardial infarction; O.R., odds ratio; PCR, polymerase chain reaction; SNP, single nucleotide polymorphism; T2DM, type 2 diabetes mellitus; TF, transcription factor; UTR, untranslated region; VD, vessel disease. ⁎ Corresponding author at: MBC-03-05, Genetics Department, King Faisal Specialist Hospital and Research Centre, P. O. Box 3354, Riyadh 11211, Saudi Arabia. E-mail address: [email protected] (N. Dzimiri).

http://dx.doi.org/10.1016/j.gene.2014.04.064 0378-1119/© 2014 Elsevier B.V. All rights reserved.

2004; Zeisberg et al., 2005). This family of TFs comprises six typically developmental/cell type specific factors, GATA1–6 (Charron and Nemer, 1999; Molkentin, 2000; Orkin, 1998), of which GATA1–3 are important regulators of hematopoietic stem cells and their derivatives, while GATA4–6 are expressed in various mesoderm and endoderm-derived tissues such as the heart, liver, lung, gonad and gut (Charron and Nemer, 1999; Molkentin, 2000). Specifically, the human endothelial transcription factor, GATA2, plays a vital role in controlling growth factor responsiveness, proliferative capacity of early hematopoietic cells as well as establishment and maintenance of adult hematopoiesis in hematopoietic stem cells, aortic endothelial cells and smooth muscle cells (Tsai et al., 1994). The involvement of these TFs in cell formation renders mutations in family members, such as GATA2, natural candidates for cardiac malformations and smooth muscle dysfunction. Indeed, a number of studies have documented associations of defects in genes coding for some of these TFs with congenital and various other cardiac malformation (Brewer and Pizzey, 2006; Nemer, 2008; Tsai et al., 1994). Since GATA2 is expressed in vascular tissue, its local function may also be

N.P. Muiya et al. / Gene 544 (2014) 152–158

influenced by alterations in its gene sequence. However, little is known about the role of GATA2 in coronary artery disease (CAD). From the few studies available to date, only a couple have postulated its association with both early-onset of CAD (Connelly et al., 2006; Hauser et al., 2004; Shah et al., 2006) and sporadic disease (Alberti and Zimmet, 1998), while some others could not establish any relationship at all with the disease (Dandona et al., 2009). In a recent linkage study in two families harboring heterozygous familial hypercholesterolemia (HFH), we found a linkage for both HFH and early onset CAD with a locus on chr 3. This locus has also been linked to CAD in both a genome-wide association study (Erdmann et al., 2009) and some population-based studies (Jiang et al., 2011), but the genes possibly involved in CAD and its risk traits in this genomic region have not been fully deciphered yet. Given this lack of information on the subject, we sought to extensively investigate the potential relevance of GATA2 as a predisposing factor for CAD and its important risk traits, such as type 2 diabetes mellitus (T2DM), hypertension (HTN) and dyslipidemia, using a homogeneous Saudi population as a study model. To achieve this, we first identified the sequence variations in both coding and non-coding regions of the gene and then determined their relationships in a larger cohort of angiographically established CAD patients compared with CAD-free individuals. 2. Materials and methods The study was performed in three stages in which three groups of Saudi individuals were recruited. The first group was recruited for the linkage study. This group comprised two families of 11 individuals each in which HFH was rampant (Supplementary data). In the first family, the index case (S3), the third of seven sons and two daughters born to unrelated parents underwent triple coronary artery bypass surgery at the age of 15 years (Supplementary data). On admission, this propositus had reported with a chest wound and xanthomas, clinical features of bilateral carotid artery disease and a very severe form of familial hyperlipidemia, harboring cholesterol (Chol) level of 10.1 mmol/l (desirable: b5.2 mmol/l) and LDL-C level of 7.9 mmol/l (optimal: b2.59 mmol/l), in addition to low HDL-C (0.51 mmol/l; normal: 1.04– 1.55 mmol/l). The father had the features of borderline HFH (Chol, 6.1 mmol/l; LDL 4.0 mmol/l), while the mother, two other sons (S4 and S6) and a daughter (D2) presented with clinical phenotypes of the disease [combination of high Chol (≥6.2 mmol/l) and high LDL-C (≥ 4.12 mmol/l) (Table 1)]. In addition to the four affected siblings and the mother, two other sons having otherwise normal Chol and

153

LDL-C levels also exhibited low HDL-C levels (b1.04 mmol/l), while none of the family members had isolated elevated LDL-C levels. The second family also consisted of 11 members in which HFH was also distinctly evident. These individuals shared some common patterns of chromosomal linkage to the disease with the first. Of these, the locus embracing chr 3, also isolated at least two of the affected siblings in each family therefore strongly pointing to the region as harboring several hotspots for the observed manifestations. The criteria for the diagnosis of familial hyperlipidemia (FH) were adapted from our Institutional Guidelines, employing the reference values approved by the USA National Cholesterol Education Program (NCEP). Following the identification of potentially informative SNPs by sequencing, we embarked on a case–control study in a total of 2557 individuals consisting of 2386 CAD individuals (1825 males and 529 females, mean age 60.8 ± 0.4 yr) with angiographically determined narrowing of the coronary vessels by at least 50% versus 2171 angiographed controls (1159 males and 919 females, mean age 50.0 ± 0.5 yr). Exclusion criteria for the disease cases were major cardiac rhythm disturbances, incapacitating or life-threatening illness, major psychiatric illness or substance abuse, history of cerebral vascular disease, neurological disorder, and administration of psychotropic medication. The controls were a group of individuals undergoing surgery for heart valvular diseases, and those who may have reported with chest pain, but were established to have no significant coronary stenosis by angiography (CON). Exclusion criteria for this group were among others diseases such as cancer, autoimmune disease, or any other disorders likely to interfere with variables under investigation. This study was performed in accordance with the regulations laid down by the Hospital Ethics Committee and all participants signed an informed consent. Within the study population, cardiovascular risk traits for atherosclerosis, particularly hypertension (HTN), type 2 diabetes mellitus (T2DM) and dyslipidemia, were rampant enough to warrant independent association analyses. We therefore elected to study the possible role of GATA2 as an independent risk factor for these disease traits. Individual candidates were characterized as hypertensive if their systole/ diastole blood pressure was ≥140/90 mm Hg (Bertoia et al., 2007). Obesity was defined as having body mass index of ≥30.0 kg/m2, while type 2 diabetes mellitus (T2DM) was characterized by combinations of decreased insulin secretion and sensitivity (also defined as insulin resistance) (Kuzuya et al., 2002). The study candidates for T2DM fulfilled the World Health Organization criteria (Mellitus WECoD, 1980) and the American Association for Diabetes Guidelines (Alberti and Zimmet, 1998; Kuzuya et al., 2002; Mellitus ECotDaCoD, 2003;

Table 1 Clinical and demographic characteristics of the studied groups. Controls

CAD Age BMI MI T2DM HTN lHDLC hLDL hTG hChol FH OBS Smokers VD One Two NTwo

Cases

All

Male

Female

All

Male

Female

2171 50.0 ± 0.5 29.0 ± 0.2 1533 2051 1076 2365 3624 3108 2680 3646 2531 2741

1159 (52.9) 50.6 ± 0.5 27.97 ± 0.2 754 (49.2) 1309 (63.8) 714 (66.4) 1350 (57.1) 2398 (66.3) 1986 (63.9) 1753 (65.4) 2390 (65.6) 1842 (72.8) 1289 (47.0)

919 (47.1) 49.5 ± 0.5 30.3 ± 0.3 779 (50.8) 742 (36.2) 362 (33.6) 1015 (42.9) 1226 (33.8) 1122 (36.1) 927 (34.6) 1256 (34.4) 689 (27.2) 1452 (53.0)

2386 60.8 ± 0.4 28.9 ± 0.1 3024 2506 3481 1883 619 1140 1663 911 1727 1728

1825 (76.7) 59.7 ± 0.3 28.3 ± 0.1 2230 (73.7) 1675 (66.8) 2270 (65.2) 1447 (76.8) 397 (64.1) 810 (71.1) 1103 (66.3) 594 (65.2) 947 (54.8) 1654 (95.7)

529 (23.3) 61.9 ± 0.5 31.0 ± 0.3 794 (26.3) 831 (33.2) 1211 (34.8) 436 (23.2) 222 (35.9) 330 (28.9) 560 (33.7) 317 (34.8) 780 (45.2) 74 (4.3)

0 0 0

0 0 0

0 0 0

898 476 1012

650 (72.4) 373 (78.4) 802 (79.2)

248 (27.6) 103 (21.6) 210 (20.8)

The numbers in brackets give the percentages of the total (all) values of the group. BMI, body mass index; FH, family history of CAD; MI, myocardial infarction; lHDLC, low high density lipoprotein-cholesterol level; hTG, hypertriglyceridemia; hChol, hypercholesterolemia; HTN, hypertension; T2DM, type 2 diabetes mellitus; VD, number of diseased vessels. Gender, age, and vessel disease are relative to the coronary artery disease group.

154

N.P. Muiya et al. / Gene 544 (2014) 152–158

Teuscher and Jarrett, 1984) for the disease. This study was performed in accordance with the regulations laid down by the Institutional Ethics Committee, and in accordance with the principles of the Declaration of Helsinki as well as Title 45, Part 46 of the U.S. Code of Federal Regulation on Protection of Human Subjects. All participants signed an informed consent.

first cycle occurs at 50 °C for 2 min, and 95 °C for 10 min, followed by 40 cycles of 94 °C for 15 s, and 60 °C for 30 s. The plates were then scanned for FRET signal using the 7900HT sequence detection system and data analyzed using SDS 2.0 software (Applied Biosystems, Foster City, CA, USA). 2.3. Statistical analysis

2.1. Linkage analysis for familial dyslipidemia and early onset of coronary artery disease Five milliliters of peripheral blood were sampled in EDTA tubes from each of the study individuals after obtaining their written consent. Genomic DNA was extracted from leukocytes by the standard salt methods using PUREGENE DNA isolation kit (Gentra system, Minneapolis, MN, USA), and the genome-wide scanning performed using the Affymetrix Gene Chip 250 sty1 mapping array (Affymetrix, Inc., Santa Clara, CA, USA). Briefly, 250 ng of genomic DNA was digested with the restriction endonuclease StyI, mixed with Sty1 adaptors and ligated with T4 DNA ligase. The mixture was then added to four separate PCR reactions, amplified, pooled and purified to remove the unincorporated dNTPs. The PCR product was then fragmented, biotinylated, hybridized to the 250 sty1 array for 18 h, washed, stained and scanned as recommended by the manufacturer. SNP genotypes, linear chromosomal locations and marker ordering were performed with the Affymetrix GeneChip® Genotyping Analysis Software (GTYPE) Version 4.0. Multipoint parametric linkage analysis was performed using the GeneHunter Easy Linkage analysis software 4.0 (Affymetrix, Inc., Santa Clara, CA, USA) for estimating the LOD scores. The disease was assumed to be an autosomal dominant trait with 100% penetrance. Copy Number Analyzer for GeneChip® Ver. 3.0 (CNAG) (Affymetrix, Inc., Santa Clara, CA, USA) was employed to check the shared chromosomal regions for homozygosity. All exons and intron–exon junctions of the GATA2 gene, including the untranslated regions, were genotyped by sequencing using the MegaBACE DNA analysis system (Amersham Biosciences, Sunnyvale, CA, USA). Briefly, the DNA was subjected to PCR by standard methods described elsewhere. Five microliters of PCR product were treated with 2 μl of ExoSAP-IT (USB Corporation, Cleveland, Ohio, USA) at 37 °C for 30 min to allow the hydrolytic removal of excess primers and dNTPs by Exonuclease 1 and Shrimp Alkaline phosphatase. The enzymes were inactivated at 80 °C for 15 min, and the sequencing reaction was initiated by mixing 2 μl DNA, 1 μl of 5 μmol primer, 8 μl of DYEnmic ET Dye Terminator (Amersham Biosciences, Buckinghamshire, UK) and 9 μl of distilled water. The mixture was thermally cycled 40× at 95 °C for 20 s, 50 °C for 15 s, and 60 °C for 1 min. Unincorporated dye-labeled terminators were removed by gel-filtration through the DyeEx 96 plate (Qiagen, GmbH, Hilden, Germany). The eluent was vacuum-dried and dissolved in 10 μl of loading solution (GE Healthcare UK Ltd, Little Chalfont, Buckinghamshire, UK) for sequencing. Data were analyzed for SNPs by Lasergene software (DNASTAR, Inc. Madison, WI, USA). 2.2. Genotyping by real-time polymerase chain reaction The selected SNPs were genotyped by TaqMan chemistry using the Applied Biosystems (ABI) Real-time PCR system (ABI Inc. CA, USA). Genotyping was achieved by TaqMan chemistry using the ABI Prism 7900HT Sequence Detection System. Primers and the TaqMan fluorogenic probes bearing a suitable reporter dye on the 5′-end and a quencher dye on the 3′-end were designed using the primer Express software V2.0 (ABI Inc., Foster City, CA, USA) and procured from Applied Biosystems (ABI, Warrington, UK). Serial dilutions were run to determine the optimal working concentration. For each reaction, a 25 μl reaction was prepared by mixing 5 μl containing 50 ng DNA, 12.5 μl of 2 × Universal mix (Eurogentec, Liege Science Park, Seraing, Belgium), 1.25 μl of 20× probe Assay mix and 6.25 μl DNase-free distilled water. Three no-template controls were included in each plate for normalization of emission signal. The thermal profile for amplification for the

Comparison of genotypes and alleles between different groups for continuous dependent variables was achieved by Analysis of Variance (ANOVA) or Student's test as appropriate. Categorical variables were analyzed by Chi-Square test, and logistic regression analysis was used to compute odds ratios and their 95% confidence intervals. The haplo.stats package (http://mayoresearch.mayo.edu/mayo/research/ schaid_lab/software.cfm) in the R Statistical Computing software (http://www.r-project.org/) was used to perform haplotype-based association analysis. Odd ratios for haplotypes were calculated using as reference the baseline haplotype AGGAGAGA, and the Haplotype Score statistic for the association of a haplotype with the binary trait was calculated as in Schaid et al. (2002) and Lake et al. (2003). Significance of association was determined between haplotypes and the case–control status — a binomial trait denoting whether or not a patient had CAD. All other statistical analyses were performed using the SPSS software version 14 (SPSS Inc., Chicago, USA), and data are expressed as mean ± SEM. Associations with a two-tailed p value b 0.05 were considered statistically significant. 3. Results 3.1. Genotyping and coronary heart disease The first step of this three-stage study was to identify chromosomal loci linked to HFH and early onset of CAD by the whole-genome scanning approach using the Affymetrix Gene Chip 250 sty1 mapping array. The linkage study led to the identification of different chromosomal regions that were linked with early onset of CAD as well as HFH. These included loci on chr 3, isolating at least three of the affected siblings with the autosomal recessive trait and one affected individual in family two. We then elected to study further the GATA2 gene, residing on chr 3, for its role in CAD and dyslipidemia risk traits in the general population. The demographic and clinical data of the study population are given in Table 1. The extent of coronary artery stenosis is depicted as single, double or multiple diseased vessels. As shown in the table, our population displayed a significantly high prevalence of MI, T2DM, dyslipidemia and hypertension. Hence we were interested to test the association of the GATA2 gene polymorphisms with the individual disease traits as the underlying causes of atherosclerosis in individuals harboring these disorders. We first sequenced the whole gene in the two families and other 200 individuals from the general population, in order to identify SNPs of potential interest within the GATA2 gene. Our data was then compared with the HapMap (http://hapmap.org) and Perlegen (http://genome. perlegen.com) databases to facilitate the SNP selection for the study. From the several identified novel findings, a great number resided in the 3′UTR of the gene, raising our curiosity as to the possible relevance of this region with respect to disease pathways leading to atherosclerosis. In all, a total of 8 SNP, rs2953128_TNC also denoted as 1, rs7431368_ANC (2), rs1573949_TNC (3), rs2713579_ANC (4), rs3803_GNA (5), rs10934857_GNA (6), rs2713604_CNT (7) and rs2335052_CNT (8) with frequency (MAF) of N5% in the general population were then selected for the study. Apart from the rs2335052_CNT found in exon 3 and rs2713604_CNT in intron 5, the rest of the selected SNPs reside in the 3′-UTR, which was the primary focus of the study (Fig. 1). The inclusion of rs2335052_CNT and rs2713604_CNT was based on their prevalence in our general population and designed to facilitate a comparison with available data in the literature.

N.P. Muiya et al. / Gene 544 (2014) 152–158

155

ATG 5’

1

2

3

4

5’ UTR

5

6

3’

3’ UTR

Fig. 1. The figure represents a schematic diagram of the GATA2 gene (not to scale) showing the eight SNPs selected for the study.

Multiple regression analysis revealed significant association of the rs1573949_C allele [1.15(1.00–1.32); p = 0.049] with MI, following Bonferroni correction and adjustment for confounding effects of coexisting risk traits (Table 2; Supplementary data 1). Other SNPs, including the rs3803 which initially exhibited a strong association with disease (p = 0.007) and four others, rs37431368, rs2713604, rs2953128 and rs2713579A displaying protective properties in the univariate analysis, failed the multivariate association test. Interestingly, the recessive mode of inheritance for rs7431368_CNA (rs7431368_AA) [5.29(1.05– 26.60); p = 0.043] conferred risk for harboring low HDL-C, but was protective against obesity [0.10(0.02–0.48); p = 0.004] and hypertriglyceridemia hTG [0.23(0.06–0.83); p = 0.026]. On the other obesity was linked to the recessive rs10934857_AA [5.69(1.04–30.98); p = 0.045]. Another variant rs2713579_AA displayed protective properties [0.15(0.03–0.78); p = 0.24] towards hypertension and another rs2953128_T allele [0.09(0.02–0.48); p = 0.005] against type 2 diabetes mellitus, while the two others rs2713579_ANG and rs2713604 showed Table 2 Association of GATA2 polymorphism with cardiovascular disease traits. Beta

S.E.

Wald

p-Value

O.R.

95% C.I. (lower–upper)

Myocardial infarction rs1573949_C 0.14

0.07

3.868

0.049⁎

1.150

1.00–1.32

Hypertension rs2713579_AA rs10934857_AA rs10934857_G

−1.92 −1.47 −1.59

0.85 0.79 0.78

5.08 3.50 4.13

0.024⁎ 0.061 0.042⁎

0.15 0.25 0.20

0.03–0.78 0.05–1.07 0.04–0.94

Type 2 diabetes mellitus rs2953128_T −2.45 rs10934857_AA 1.21 rs2335052_C 1.11

0.88 0.71 0.64

7.81 2.92 2.97

0.005⁎⁎ 0.087 0.085

0.09 3.35 3.02

0.02–0.48 0.84–13.38 0.86–10.62

Obesity rs7431368_AA rs2713579_G rs10934857_AA rs2713604_C

−2.30 1.25 1.74 0.15

0.80 0.67 0.87 0.08

8.20 3.50 4.04 3.44

0.004⁎⁎ 0.061 0.045⁎ 0.064

0.10 3.49 5.69 1.16

0.02–0.48 0.94–12.93 1.04–31.08 0.99–1.36

Hypertriglyceridemia rs7431368_AA −1.49

0.67

4.99

0.026⁎

0.23

0.06–0.83

Low high density lipoprotein rs7431368_AA 1.67 0.82

4.09

0.043⁎

5.29

1.05–26.60

The table displays the association of different variants with disease traits that remain significant after Bonferroni test and multivariate analysis to exclude the confounding effects of age, sex and other classical risk factors for the respective analysis. Details of the test are given in the Supplementary data. O.R., odds ratio; S.E., standard error; C.I., confidence interval. ⁎ p b 0.05. ⁎⁎ p b 0.005.

weak association with disease after adjustment for the confounding effects of coexisting disease traits (Table 2). 3.2. Haplotype analysis Since several of the studied variants showed an association with the important risk traits in a causative fashion, we were interested in establishing further the potential impact of haplotypes created from these SNPs as predisposing factors to the disease traits. We performed staged haplotyping involving various sets of SNPs, resulting in multiple haplotypes that displayed strong association with the different disorders. Fig. 2 represents the linkage disequilibrium for the 8 studied SNPs, indicating that the rs7431368 displayed the weakest link with all other variants. Tables 3 and 4 list the most significant haplotypes for CAD/MI and the important metabolic risk traits. The most frequent haplotype created from all 8 SNPs was CCCCGGCC with a frequency of 0.167. Employing this sequence as the baseline for comparing the relationship for created haplotypes, an 8-mer haplotype CCCGGGTC (χ2 = 4.23; p = 0.04) constructed from the eight studied variants and its 6-mer derivative CGGG TC (blocks 3–8; χ2 = 5.05; p = 0.025) conferred risk for CAD. In

Fig. 2. Linkage disequilibrium (LD) structure of the eight studied SNPs. D′ = coefficient of linkage disequilibrium, r = regression coefficient.

156

N.P. Muiya et al. / Gene 544 (2014) 152–158

Table 3 GATA2 haplotyping versus cardiovascular disease traits. Block

Haplotype

Pooled

Controls

Table 4 GATA2 haplotyping versus dyslipidemia disease traits. Cases

χ2

p-Value

Coronary artery disease 1–8 CCCGGGTC 2–8 ATAGGTC 3–8 CGGGTC TAGGTC

0.051 0.017 0.184 0.019

0.047 0.020 0.176 0.022

0.056 0.014 0.194 0.015

4.23 4.50 5.05 5.51

0.040 0.034 0.025 0.019

Myocardial infarction 2–8 ATAGGTC 4–8 GGGTC AGGTC 2–7 ATAGGT 5–8 GATC 3–7 TAGGT 4–7 AGGT 6–8 ATC ACT

0.017 0.185 0.02 0.026 0.013 0.028 0.029 0.021 0.017

0.019 0.179 0.023 0.028 0.015 0.032 0.033 0.024 0.019

0.013 0.196 0.016 0.020 0.010 0.021 0.023 0.016 0.013

4.23 3.57 5.10 5.47 4.13 7.89 6.55 6.93 4.01

0.040 0.059 0.024 0.019 0.042 0.005 0.011 0.009 0.045

Hypertension 6–8 ATC 3–6 TAGA

0.021 0.033

0.023 0.035

0.015 0.027

5.10 3.45

0.024 0.063

Type 2 diabetes mellitus 1–8 TATAGACC 1–7 TATAGAC 2–8 ATAGACC 1–6 TATAGA 3–8 TAGACC 4–8 AGACC 2–6 ATAGA 4–7 AGAC 3–6 TAGA

0.016 0.016 0.021 0.019 0.027 0.026 0.027 0.028 0.033

0.019 0.019 0.026 0.023 0.033 0.032 0.032 0.034 0.040

0.012 0.012 0.016 0.015 0.020 0.020 0.020 0.021 0.025

6.49 6.72 11.51 7.88 13.66 13.20 13.45 13.51 14.63

0.011 0.010 0.0007 0.005 0.0002 0.0003 0.0002 0.0002 0.0001

Obesity 1–8 1–7 1–6

0.077 0.083 0.087

0.070 0.076 0.080

0.081 0.088 0.092

3.66 3.82 3.88

0.056 0.051 0.049

CATAAACC CATAAAC CATAAA

The table shows selected haplotypes associated with disease. The most frequent 8-mer haplotype (0.16) was employed as the baseline to determine the relative effects of the other haplotypes. The studied SNPs are rs2953128TNC also denoted as 1, rs7431368ANC (2), rs1573949TNC (3), rs2713579ANC (4), rs3803GNA (5), rs10934857GNA (6), rs2713604CNT (7) and rs2335052CNT (8) arranged sequentially by their chromosomal positions, whereby blocks represent the range of variants constituting the respective haplotypes. *p b 0.01; **p b 0.005 by χ2 test.

contrast the 6-mer haplotype ATAGGTC (blocks 2–8) was protective against both CAD (χ2 = 4.50; p = 0.034) and MI (χ2 = 4.23; p = 0.040). Thereby, the shorter derivative TAGGTC (blocks 3–8) showed slightly stronger protective effects against CAD (χ2 = 5.51; p = 0.019) and ATAGGT (blocks 2–7) against MI (χ2 = 5.47; p = 0.019), respectively. Interestingly, these properties augmented as the stretches of the constituent nucleotides became shorter, culminating in the 5′mer TAGGT (χ2 = 7.89; p = 0.005) displaying the strongest properties. Put together, these observations indicate that the core of these interactions is embedded in this haplotype. Another 4-mer GATC (χ2 = 4.14; p = 0.042) and its 3-mer derivative ATC (χ2 = 6.93; p = 0.009) also displayed significant protective properties against MI (Table 3). Obesity was only weakly linked to the 8-mer haplotype CATAAACC, but significantly associated with its 6-mer CATAAA (χ2 = 3.66; p = 0.049). Furthermore, hypercholesterolemia (hChol) was linked to a completely different set of haplotypes derived from the 8-mer CCTG GACC (χ2 = 6.02; p = 0.014), which by itself was associated with the disease. Among these, the most significantly associated sequence with the disease was the 5-mer, CTGGA (blocks 2–6; χ2 = 6.75; p = 0.009) (Table 4). On the other hand, high low density lipoprotein was linked to TGG (χ2 = 4.48; p = 0.034), while hTG was only weakly linked to the 5-mer CGGGC. Interestingly, a set of haplotypes derived from the 8-mer TATAGACC (χ2 = 6.49; p = 0.011) was protective against acquiring T2DM, with the 4-mer TAGA (blocks 3–6) (χ2 = 14.63; p = 0.0001) exhibiting by far the most significant association (Table 3). Altogether, the various risk traits were linked to different haplotypes.

Pooled

Controls

Cases

χ2

p-Value

0.034 0.034 0.025 0.036 0.041 0.037 0.035 0.037 0.037 0.037 0.037 0.021 0.037

0.028 0.028 0.021 0.029 0.036 0.030 0.029 0.030 0.030 0.030 0.031 0.027 0.031

0.038 0.038 0.027 0.040 0.045 0.041 0.039 0.041 0.041 0.041 0.041 0.017 0.041

6.02 6.21 3.72 6.38 4.83 6.51 6.62 6.13 6.75 6.58 5.82 9.91 6.22

0.014 0.013 0.054 0.012 0.028 0.011 0.010⁎

High low density lipoprotein-cholesterol 3–5 TGG 0.049

0.037

0.051

4.48

0.034

Hypertriglyceridemia 3–7 CGGGC 2–5 CTAG

0.009 0.087

0.014 0.074

3.74 4.26

0.053 0.039

Block

Haplotype

Hypercholesterolemia 1–8 CCTGGACC 2–8 CTGGACC CTAAACC 1–7 CCTGGAC 3–8 TGGACC 1–6 CCTGGA 2–7 CTGGAC 1–5 CCTGG 2–6 CTGGA 2–5 CTGG 1–4 CCTG 6–8 ATC 2–4 CTG

0.013 0.077

0.013 0.009⁎ 0.010 0.016 0.002⁎⁎ 0.013

The table shows selected haplotypes associated with disease. The most frequent 8-mer haplotype (0.18) was employed as the baseline to determine the relative effects of the other haplotypes. The studied SNPs are rs2953128TNC also denoted as 1, rs7431368ANC (2), rs1573949TNC (3), rs2713579ANC (4), rs3803GNA (5), rs10934857GNA (6), rs2713604CNT (7) and rs2335052CNT (8) arranged sequentially by their chromosomal positions, whereby blocks represent the range of variants constituting the respective haplotypes. ⁎ p b 0.01 by χ2 test. ⁎⁎ p b 0.005 by χ2 test.

4. Discussion The present study investigated the potential impact of the GATA2 polymorphism on CAD and its metabolic risk traits, based on a preliminary analysis that linked chr 3 with the two diseases in two families with HFH and clinical evidence of early onset of CAD. As mentioned above, this genomic locus has also been recently linked to CAD (Erdmann et al., 2009), but there is still a lack of information as to the identity of the genes involved. Based on available knowledge on the cardiovascular function of GATA2, and the discrepancies in the literature on its role in cardiovascular disease (Connelly et al., 2006; Dandona et al., 2009; Hauser et al., 2004; Shah et al., 2006), this gene stood out as the most attractive candidate at this chromosomal locus to evaluate for a potential role in atherosclerosis and its metabolic risk traits in our population. Subsequent sequencing of the gene led to the striking identification of several potentially informative variants, largely in the 3′UTR, which became the primary focus of the present study. Initially, the univariate analysis linked several of the studied SNPs with CAD/MI. However, only the rs1573949 passed the multivariate analysis and Bonferroni adjustment tests, independently implicating this SNP in the disease. Thus, despite the fact that associations of majority of these SNPs become weaker in the presence of confounding factors, it can nonetheless be concluded that changes at the GATA2 locus play an important role in processes leading to atherosclerosis. Our study therefore furnishes further support for the notion that this genomic locus is associated with CAD onset as suggested by some previous studies (Connelly et al., 2006; Hauser et al., 2004; Shah et al., 2006). It should be mentioned however, that some other studies were similarly not able to replicate such relationships for the rs2713604 and the rs1573949 with sporadic CAD (Dandona et al., 2009). This heterogeneity in the patterns for the GATA2 impact on CAD seems to point to other factors as the ultimate underlying contributory entities to the causative relationship of the gene with the disease in the different populations. Nonetheless, the fact that this locus was linked to early CAD onset in HFH may also point to a direct link for the latter, rather than the CAD outcome per se, to these genetic changes.

N.P. Muiya et al. / Gene 544 (2014) 152–158

Hence, our focus turned towards investigating the likelihood of the GATA2 gene interacting with metabolic disorders, such as dyslipidemia, obesity, T2DM and HTN as independent end-points as well as potential underlying triggers for CAD onset under these disease conditions. Interestingly, the study implicated one variant, rs7431368_AA in harboring of low HDL-C, which was protective against hTG and obesity. Hence, our results seem to confirm the notion of GATA2 as a risk gene for predisposing individuals to some underlying metabolic risk for sporadic onset of CAD. Although there is dearth of data in the literature linking this gene to dyslipidemia or obesity, the finding of protective effects on obesity is not surprising, since GATA2 function is also linked to fat cell formation and the gene is thought to be involved in the regulation of adipocyte differentiation through some molecular control of preadipocyte–adipocyte transition (Tong et al., 2000, 2003). Besides, another variant, rs10934857, was also associated with obesity. Put together therefore, our findings point to an association of the GATA2 polymorphisms with metabolic disease traits for atherosclerosis. To our knowledge, no other evidence exists yet in the literature directly implicating the GATA2 gene polymorphism in these disorders. Most importantly, haplotypes constructed from the suspected alterations led to even more fascinating results. To begin with, it was interesting to note that a haplotype constructed from the eight studied variants (CCCGGGTC) and its shorter derivatives conferred risk for CAD, despite that none of the variants was independently implicated in the disease. This suggests that CAD is affected by changes related to the function of this genomic region, rather than individual loci per se. Furthermore, obesity was significantly associated with a 6-mer CATA AA, and hChol linked to a completely different set of haplotypes derived from the 8-mer CCTGGACC. Notably, the most significant associations were observed with shorter nucleotide stretches within the 3′UTR. Furthermore, these haplotypes also exhibited similarities with the trends shown by the individual SNPs, in that the different traits were associated with different haplotypes, confirming that the GATA2 locus constitutes a risk for different metabolic risk traits for atherosclerosis. Without offering unwarranted speculation, it can be suggested that the various ways in which the gene variants are linked to the different traits relate to the modes by which these traits interact with each other leading to atherosclerosis. The mechanisms underlying the influence of GATA2 gene on CAD remain to be elucidated. To date, the only cardiac-related events associated with this gene are malformation and early onset CAD, but no speculations have been offered yet on possible mechanisms. While such mechanisms might explain some involvement of the transcriptional pathways in the disease process of the risk traits, they appear to be separable from those leading to CAD. Based on our results, such a mechanism may not necessarily involve functional changes in the GATA2 protein product per se. In this regard, the most interesting finding of the study was the observation of a link for these traits with variants residing in the 3′UTR, bearing important implications for the involvement of this genic region in metabolic disease pathways. Furthermore, the various modes by which these traits share causative entities point to yet undiscovered mechanisms inherent to the function of this genomic locus in disease. Therefore, the novel finding implicating the GATA2 3′ UTR in these traits points to the likelihood of this genic region harboring important players for yet undefined pathways involved in events leading to atherosclerosis through mechanisms not directly ascribable to GATA2 functional changes, but rather to some gene regulatory or mRNA maturation processes. Further studies are therefore necessary to explain such common or interdependent pathways for these processes. In conclusion, our study suggests an association of GATA2 as a risk for atherosclerosis resulting from its impact on its cardiovascular risk traits. The observation of an involvement of the 3′ UTR of the gene strongly points to the relevance of this region in disease pathways. Potential pathways involved in the atherosclerotic process are likely to engage entities residing in this genomic region regulating individual signaling systems affecting the disease risk traits.

157

Conflict of interest All authors declare that there is no conflict of interest to disclose. Acknowledgment This study was supported by the Royal Cardiovascular Research Grant No 2030020 provided through the King Faisal Specialist Hospital and Research Centre, for which the investigators are deeply indebted. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.gene.2014.04.064. References Alberti, K.G., Zimmet, P.Z., 1998. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabetic Medicine 15 (7), 539–553. Bertoia, M.L., Waring, M.E., Gupta, P.S., Roberts, M.B., Eaton, C.B., 2007. Implications of new hypertension guidelines in the United States. Hypertension 58 (3), 361–366. Brewer, A., Pizzey, J., 2006. GATA factors in vertebrate heart development and disease. Expert Reviews in Molecular Medicine 8 (22), 1–20. Charron, F., Nemer, M., 1999. GATA transcription factors and cardiac development. Seminars in Cell and Developmental Biology 10 (1), 85–91. Connelly, J.J., Wang, T., Cox, J.E., et al., 2006. GATA2 is associated with familial early-onset coronary artery disease. PLoS Genetics 2 (8), e139. Dandona, S., Chen, L., Fan, M., et al., 2009. The transcription factor GATA-2 does not associate with angiographic coronary artery disease in the Ottawa Heart Genomics and Cleveland Clinic GeneBank Studies. Human Genetics 127 (1), 101–105. Erdmann, J., Grosshennig, A., Braund, P.S., et al., 2009. New susceptibility locus for coronary artery disease on chromosome 3q22.3. Nature Genetics 41 (3), 280–282. Fujikura, J., Yamato, E., Yonemura, S., et al., 2002. Differentiation of embryonic stem cells is induced by GATA factors. Genes and Development 16 (7), 784–789. Hauser, E.R., Crossman, D.C., Granger, C.B., et al., 2004. A genomewide scan for early-onset coronary artery disease in 438 families: the GENECARD Study. American Journal of Human Genetics 75 (3), 436–447. Jiang, F., Dong, Y., Wu, C., et al., 2011. Fine mapping of chromosome 3q22.3 identifies two haplotype blocks in ESYT3 associated with coronary artery disease in female Han Chinese. Atherosclerosis 218 (2), 397–403. Kuzuya, T., Nakagawa, S., Satoh, J., et al., 2002. Report of the Committee on the classification and diagnostic criteria of diabetes mellitus. Diabetes Research and Clinical Practice 55 (1), 65–85. Lake, S.L., Lyon, H., Tantisira, K., et al., 2003. Estimation and tests of haplotype–environment interaction when linkage phase is ambiguous. Human Heredity 55 (1), 56–65. Mellitus ECotDaCoD, 2003. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 26 (Suppl. 1), S5–S20. Mellitus WECoD, 1980. WHO Expert Committee on Diabetes Mellitus: second report. World Health Organization Technical Report Series 646, 1–80. Molkentin, J.D., 2000. The zinc finger-containing transcription factors GATA-4, -5, and -6. Ubiquitously expressed regulators of tissue-specific gene expression. Journal of Biological Chemistry 275 (50), 38949–38952. Molkentin, J.D., Lin, Q., Duncan, S.A., Olson, E.N., 1997. Requirement of the transcription factor GATA4 for heart tube formation and ventral morphogenesis. Genes and Development 11 (8), 1061–1072. Morrisey, E.E., Ip, H.S., Tang, Z., Parmacek, M.S., 1997a. GATA-4 activates transcription via two novel domains that are conserved within the GATA-4/5/6 subfamily. Journal of Biological Chemistry 272 (13), 8515–8524. Morrisey, E.E., Ip, H.S., Tang, Z., Lu, M.M., Parmacek, M.S., 1997b. GATA-5: a transcriptional activator expressed in a novel temporally and spatially-restricted pattern during embryonic development. Developmental Biology 183 (1), 21–36. Narita, N., Bielinska, M., Wilson, D.B., 1997. Wild-type endoderm abrogates the ventral developmental defects associated with GATA-4 deficiency in the mouse. Developmental Biology 189 (2), 270–274. Nemer, M., 2008. Genetic insights into normal and abnormal heart development. Cardiovascular Pathology 17 (1), 48–54. Orkin, S.H., 1998. Embryonic stem cells and transgenic mice in the study of hematopoiesis. International Journal of Developmental Biology 42 (7), 927–934. Pikkarainen, S., Tokola, H., Kerkela, R., Ruskoaho, H., 2004. GATA transcription factors in the developing and adult heart. Cardiovascular Research 63 (2), 196–207. Schaid, D.J., Rowland, C.M., Tines, D.E., Jacobson, R.M., Poland, G.A., 2002. Score tests for association between traits and haplotypes when linkage phase is ambiguous. American Journal of Human Genetics 70 (2), 425–434. Shah, S.H., Kraus, W.E., Crossman, D.C., et al., 2006. Serum lipids in the GENECARD study of coronary artery disease identify quantitative trait loci and phenotypic subsets on chromosomes 3q and 5q. Annals of Human Genetics 70 (Pt 6), 738–748. Teuscher, A., Jarrett, R.J., 1984. Diabetes mellitus: diagnostic criteria. Recommendations of the European Diabetes Epidemiology Study Group (EDESG) concerning the Second

158

N.P. Muiya et al. / Gene 544 (2014) 152–158

Report of the WHO Expert Committee on Diabetes Mellitus. Diabetic Medicine 1 (4), 305–307. Tong, Q., Dalgin, G., Xu, H., Ting, C.N., Leiden, J.M., Hotamisligil, G.S., 2000. Function of GATA transcription factors in preadipocyte–adipocyte transition. Science 290 (5489), 134–138. Tong, Q., Tsai, J., Hotamisligil, G.S., 2003. GATA transcription factors and fat cell formation. Drug News & Perspectives 16 (9), 585–588. Tsai, F.Y., Keller, G., Kuo, F.C., et al., 1994. An early haematopoietic defect in mice lacking the transcription factor GATA-2. Nature 371 (6494), 221–226.

Watt, A.J., Battle, M.A., Li, J., Duncan, S.A., 2004. GATA4 is essential for formation of the proepicardium and regulates cardiogenesis. Proceedings of the National Academy of Sciences of the United States of America 101 (34), 12573–12578. Zeisberg, E.M., Ma, Q., Juraszek, A.L., et al., 2005. Morphogenesis of the right ventricle requires myocardial expression of GATA4. Journal of Clinical Investigation 115 (6), 1522–1531.