Association of rs17042171 with chromosome 4q25 with atrial fibrillation in Chinese Han populations.

OBJECTIVE: A recent genome-wide association study (GWAS) identified a susceptibility single nucleotide polymorphism (SNP), rs17042171 on 4q25 for atrial fibrillation (AF). The aim of the present study was to investigate whether this association between rs17042171 and AF also exists in Chinese Han populations.
METHODS: It was a case-control study. We enrolled a total of 1,593 Chinese Han origin individuals in the study, including 597 AF patients and 996 AF-free controls. Genotyping was performed using the TaqMan allelic discrimination Assay. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated in logistic regression models.
RESULTS: There was strongly significant difference between AF patients and control subjects regarding rs17042171 assumption of additive model (OR=2.20, 95% CI: 1.88-2.57, p=2.00 × 10(-22)), dominant model (OR=2.99; 95% CI: 2.19-4.09; p=6.47 × 10(-12)) and a recessive (OR=2.75; 95% CI: 2.21-3.43; p=1.30 × 10(-19)). In the stratification analysis based on age, gender, hypertension, diabetes and coronary artery disease, there was no significant difference of the associations for rs17042171 among the subgroups.
CONCLUSION: Our results indicated that rs17042171 confers an increased risk of AF in Chinese Han Populations and expanded the association to non-European ancestry populations for the first time.

Introduction

Atrial fibrillation (AF) is the most common sustained arrhythmia in clinical practice. Independent of preexisting diseases, AF can increase both cardiovascular mortality and morbidity (1-3). Arrhythmia has become an important public health problem because of its association with significantly increased risks of stroke, heart failure, and death (4). Recently, radiofrequency catheter ablation has been developed as an aggressive management of AF. Moreover, morphological and electrophysiological alterations that develop and maintain AF have been studied extensively (5). However, the limited effectiveness of the management and the high recurrence rate of catheter ablation indicate that the mechanism of AF is still not fully understood. There is a growing body of studies demonstrating that genetic factors play an important role in the pathogenesis of AF (6), especially in patients with lone AF (7). Scientific studies conducted in the last decade described an increased risk in offspring of patients with AF (8, 9). Moreover, the first genome-wide association study (GWAS) of AF identified some noncoding single-nucleotide polymorphisms (SNPs) located on chromosome 4q25 that are associated with an increased AF risk (10). Till date, several follow-up studies have replicated these GWAS results in populations with different racial backgrounds (11-14). However, it is worth noting that there were some drastic differences in individuals of different populations. The frequency of the risk allele T of SNP rs2200733 in the European population deviates greatly from that in the Chinese population and SNP rs2200733 is a more common genetic risk factor in the Chinese population (12-15). These results suggest that the common AF is a complex disease resulting from the interaction among multiple genes, environmental factors, and gene-environment interactions (16, 17).

Then, Benjamin et al. (18) performed an independent GWAS and revealed variants in the ZFHX3 gene on chromosome 4q25 that are associated with AF in various individuals of European ancestry and identified a new locus for AF on chromosome 4q25 (rs17042171, OR=2.46, p=6.9×10-51), which was approximately 150 kb telomeric to the transcription factor gene PITX2. Accordingly, we conducted a large-scale case-control association study in 597 AF patients and 996 non-AF controls in Chinese Han populations to investigate or replicate whether the rs17042171 confers a risk of AF in a non-European ancestry cohort.

Methods

Study subjects

For this study, cases were selected from patients enrolled in our cardiovascular ward or had visited our cardiovascular clinic who had self-reported ethnic Han origin. The study recruited 597 consecutive patients with AF formed the case group. The diagnostic criterion of AF was made by expert cardiologists and according to ACC/AHA/ESC 2006 guidelines for the management of patients with AF (19). Physical examinations were performed by either a 12-lead electrocardiogram (ECG) or by Holter ECG recordings. On the ECG, AF is characterized by the replacement of consistent P waves by rapid oscillations or fibrillatory waves that vary in amplitude, shape, and timing, associated with an irregular, frequently rapid ventricular response. According to clinical characteristics, AF can be classified into paroxysmal AF (episodes that generally last 7 days or less), persistent AF (episodes that sustain beyond 7 days), and permanent AF (ongoing long-term episodes, in which cardioversion has failed or has not been attempted). AF patients who are under 60 years of age without clinical or echocardiographic evidence of cardiopulmonary disease, including hypertension, were considered as lone AF. Patients with other types of cardiac arrhythmias, cardiomyopathies, and valvular disease were excluded from this study. We acquired relevant information such as age; gender; and history of hypertension, diabetes, coronary artery disease (CAD), and hyperthyroidism from the medical records. The remaining 996 patients were assigned to control group on the basis of the ECG or no history of AF. Similar to the cases, the controls were selected without consideration of the presence or absence of hypertension, diabetes, coronary artery disease, and hyperthyroidism. The study protocol was approved by local Ethics Committees and all participants signed consent forms.

DNA isolation and genotyping

Blood samples were drawn from study participants and genomic DNA was extracted from EDTA-preserved whole blood, using the standard phenol-chloroform method (20). The SNP was genotyped using the TaqMan allelic discrimination Assay (Applied Biosystems, Inc., USA). Genotyping was performed in 25 µL of standard PCR volume containing 1 µL of LC Green dye, 5 pmoL of each primer, 25 ng of genomic DNA, 2.5 µL of 10× PCR buffer with 1.5 mmoL/L MgCl2, 5 mmoL deoxynucleotide triphosphates, and 1 U of Taq polymerase.

Statistical analysis

SNP rs17042171 genotypes were tested for Hardy-Weinberg equilibrium among controls using PLINK v1.05. Continuous variables were presented as mean and standard deviation (SD); normality tests (Kolmogorov-Smirnov) were used. Categorical variables compared using the χ2 test. The χ2 test was used to determine deviations of the genotype distribution between two groups. The Hardy-Weinberg equilibrium of the genotype distribution of polymorphisms in the case and control groups was determined using the χ2 test. Odds ratios (ORs) and 95% confidence intervals (CIs) were estimated when we used logistic regression to adjust for covariant factors, including age; gender; and presence or absence of hypertension, diabetes, coronary artery disease, or hyperthyroidism, to assess the strength of the relationship of the genotype distribution of SNPs between the AF and control groups. The results have statistical significance if p<0.05 or the range of 95% CI did not include unity. All statistical analyses were performed using SPSS 20.0 software (SPSS Inc., Chicago, IL, USA).

Results

Characteristics of study patients

We performed genetic analysis in 597 patients with AF vs. 996 control subjects to detect the association between SNP rs17042171 and the risk of AF in the Chinese Han population. The general and clinical characteristics of the AF patients and controls are summarized in Table 1. The average age for AF cases and controls was 58.37±11.46 and 58.99±10.21 years, respectively. There were no significant differences between AF patients and control subjects regarding age and gender (p=0.428 and p=0.630, respectively). Among the AF group, 64.2% patients were paroxysmal AF, 32.8% were persistent, and permanent AF accounted for 11.9%. The hypertensive patients in the AF and control groups amounted to 260 and 267, respectively, and 48 had CAD. In the AF group, 53 patients had diabetes. Common cardiovascular risk factors for AF, including hypertension, coronary artery disease, and diabetes, were more prevalent in the AF group than in the control group (p=6.18×10-12, p=0.019, and p=9.55×10-8, respectively).

Table 1

Clinical characteristics of the study population

	AF (n=597)	Control (n=996)	P
Male gender, n (%)	397 (66.5)	674 (67.7)	0.630
Ager* (Mean±SD)	5a8.37±11.46	58.9±10.21	0.428
Paroxysmal AF, n (%)	383 (64.2)	NA	-
Persistent AF, n (%)	196 (32.8)	NA	-
Permanent AF, n (%)	18 (3.0)	NA	-
Lone AF, n (%)	71 (11.9)	NA	-
Hypertension, n (%)	260 (43.6)	267 (26.8)	6.18×10-12
CAD, n (%)	53 (8.0)	28 (5.1)	0.019
Diabetes, n (%)	48 (8.9)	51 (2.8)	9.55×10-8

Age was defined as the age at the sample collection. Data are presented as mean ± standard deviation or number (percentage); AF - atrial fibrillation; CAD - coronary artery disease; NA - not available; SD - standard deviation

Hardy-Weinberg equilibrium of the genotype distribution of polymorphisms

The distribution of genotypes for SNP rs17042171 was not significantly deviated from Hardy-Weinberg equilibrium in the control group (p=0.823).

Genotypic association of rs1704217 with AF

The allele and genotype associations in patients with AF and control subjects are shown in Table 2. The distributions of the SNP rs17042171 genotype was significantly different between the AF group and the controls (p=7.57×10-24). The SNP rs17042171 AA genotype was found was more in the AF group than in the control group (9.7% vs. 24.6%, p=7.57×10-24). We found a significant allelic association between SNP rs1704217 and AF. The frequency of the allele A of SNP rs17042171 was 0.70 compared with 0.30 in the unaffected controls (OR=2.25, 95% CI: 1.93-2.62, p=5.15×10-26).

Table 2

Genotype distribution and allelic analysis of the association of rs17042171 with AF

Phenotype	CC, n (%)	AC, n (%)	AA, n (%)	P	Frequency (A/C)	P	OR (95% CI)
AF	58 (9.7)	242 (40.6)	296 (49.7)	7.57×10-24	0.70/0.30	5.15×10-26	2.25 (1.93–2.62)
Control	245 (24.6)	488 (49.0)	263 (26.4)		0.51/0.49

A - adenine; AF - atrial fibrillation; C - cytosine; CI - confidence interval; OR - odds ratio. Genotyping calling rate: 99.9%

Adjusting for gender, age, hypertension, diabetes, coronary artery disease, and hyperthyroidism, logistic regression analysis of SNP revealed that patients with the AA genotype were at a higher risk (2. 8-fold) of developing AF than those with the CC genotype (OR=2.18, 95% CI: 1.84-2.58, p=2.72×10-19). Patients with the heterozygous genotype (AC) were 2.07 times more likely to develop AF (OR=2.07, 95% CI: 1.49-2.88; p=2.76×10-12). To test the association between rs17042171 and AF, genotypic frequencies under three genetic models (dominant, recessive, and additive) were also applied. Indeed, rs17042171 was strongly associated with the AF assumption of additive model (OR=2.20, 95% CI: 1.88-2.57, p=2.00×10-22), dominant model (OR=2.99; 95% CI: 2.19-4.09; p=6.47×10-12), and recessive model (OR=2.75; 95% CI: 2.21-3.43; p=1.30×10-19) (Table 3).

Table 3

Genotypic analysis of SNP rs17042171 with AF under three genetic models

SNP	Genotype	Adjusted OR* (95% CI)	P*	P **
rs17042171	AA	2.18 (1.84-2.58)	2.72×10-19	0.823
	AC	2.07 (1.49-2.88)	2.76×10-12
	CC	1.00
	Dominant model	2.99 (2.19-4.09)	6.47×10-12
	Recessive model	2.75 (2.21-3.43)	1.30×10-19
	Additive model	2.20 (1.88-2.57)	2.00×10-22

Obtained in logistic regression models with adjustment for gender, age, hypertension, diabetes, coronary artery disease, and hyperthyroidism;

P for Hardy–Weinberg equilibrium test. A - adenine; AF - atrial fibrillation; C - cytosine; CI - confidence interval; OR - odds ratio

To further validate the associations of the significance, stratified analysis was also performed for rs17042171 on the basis of age and gender and hypertension, diabetes, or CAD. As shown in Table 4, risk effects were also observed between subgroups stratified by age, gender, hypertension, and CAD. When the cohort was divided into different age groups, the OR in AF cases diagnosed over 59 years of age was slightly higher than that in AF cases under 59 years of age (2.52 vs. 1.93), but the difference was not significant (p for heterogeneity was 0.104). The association between rs17042171and AF was significant in the AF group without diabetes but not in the AF group with diabetes (Table 4).

Table 4

Stratification analysis for the association between rs17042171 and AF risk

Variables	Case, n	Control, n	Adjusted OR* (95% CI)	P*	P **
SNP	rs17042171 (AA/AC/CC)
Age
≤59	33/117/157	111/221/146	1.93 (1.55-2.40)	3.74×10-9	0.104
>59	25/125/139	134/267/117	2.52 (1.99-3.19)	1.91×10-14
Gender
Male	43/154/199	167/329/178	2.14 (1.77-2.59)	6.24×10-15	0.653
Female	15/88/97	78/159/85	2.32 (1.72-3.11)	2.48×10-8
Hypertension
Yes	21/105/134	66/139/62	2.68 (2.03-3.53)	2.64×10-12	0.075
No	37/137/162	179/349/201	1.97 (1.62-2.40)	9.81×10-12
CAD
Yes	3/21/23	11/30/10	3.15 (1.42-6.99)	0.005	0.358
No	55/221/273	234/458/253	2.15 (1.82-2.53)	1.65×10-9
Diabetes
Yes	4/24/25	7/14/7	2.32 (0.96-5.60)	0.062	0.900
No	54/218/271	238/474/256	2.19 (1.86-2.58)	5.47×10-21

Associations were evaluated using logistic regression with adjustment for gender, age, hypertension, diabetes mellitus, coronary artery disease, and hyperthyroidism;

P for heterogeneity test using the chi-square-based Q test. A - adenine; AF - atrial fibrillation; C - cytosine; CAD - coronary artery disease; CI - confidence interval; n - number of genotypes. OR - odds ratio; SNP - single nucleotide polymorphism

Discussion

We conducted a case-control association study of 597 subjects with AF vs. 996 controls for SNP rs1704217 on chromosome 4 locus. Consistent with previous studies (OR=2.46, p=6.9×10-51) (18), our results indicated that rs17042171 confers an increased risk of AF in Chinese Han populations (OR=2.20, p=2.00×10-22).

In the present control subjects, the frequency of the A allele of SNP rs17042171 was approximately 51%, which is comparable to that reported in the NCBI dbSNP database, ranging from 42% to 54% in other Asian ethnicities, but much more than that in European-descent populations (the allelic frequency is 0.115). In the study on European ancestry AF subjects by Benjamin et al. (18), the allele frequency of the allele A of SNP rs17042171 was approximately 16%, much lower than that of 70% in the present AF subjects of Chinese Han populations. Although the cause and significance of the apparent higher frequency of the allele C of SNP rs1704217 are uncertain, it may account for the different estimated risks in the two populations.

The limited success of various therapies for AF and differences between the individual treatment effects suggested that the pathogenesis of AF is multifactorial (21). Subsequently, data that have emerged from independent studies strongly support the notion that genetic variants are associated with AF risk, including ion channel genes (6, 11). A genome-wide association study has been instrumental in the identification of common variants on 4q25, 16q22, and 1q21 for the nonfamilial AF (22) and a strong association between variants on chromosome 4q25 and AF was identified in three populations of European descent and a Chinese population from Hong Kong (10). The most popular and widely accepted hypothesis suggests the close proximity of these variants on 4q25 to the PITX2 (paired-like homeodomain transcription factor 2), known to be critical in the embryonic development of the cardiac conduction system and pulmonary venous myocardium, which is a major source of ectopic activity related to the initiation and maintenance of AF (23). Previous studies revealed that reducing PITX2c expression can promote atrial fibrillation inducibility and the mutant PITX2c is associated with significantly reduced transcriptional activity (24-26). By examining GWAS data for AF, Benjamin et al. (18) replicated the previously reported association with chromosome 4q25 variants and identified a new locus rs17042171 on chromosome 4 locus, which was approximately 150 kb telomeric to PITX2 (p=6.9×10-51). Rs17042171 may be expected to exert influence on the transcriptional activity of PITX2c. Further studies are warranted to evaluate the precise mechanism by which rs1704217 regulates AF risk.

Study limitations

This study has several limitations. One limitation is that this is the first time that SNP rs17042171 was found to be associated with AF in the Chinese Han population; therefore, further investigation should be conducted in other independent Chinese Han populations. Second, the mean age of the AF patients was 58.37±11.46 years, which was lesser than that of typical AF in the community. Moreover, our findings suggested rs17042171 is an AF-susceptibility SNP; however, its role in AF was rarely examined.

Conclusion

In summary, to the best of our knowledge, this is the first study to investigate whether rs17042171 confers a highly significant risk of AF in the Chinese Han population. The results expand the association of SNPrs17042171 with AF previously identified in a cohort of European descent to a non-European ancestry population.