Integration Analysis of Epigenetic-related m ⁶A-SNPs Associated with Atrial Fibrillation

Abstract

Objective: Numerous single nucleotide polymorphisms (SNPs) have been identified as genetic contributors to atrial fibrillation (AF). The aim of this study was to investigate the effects of genome-wide N⁶-methyladenosine (m⁶A)-SNPs on AF.

Method: m⁶A-SNPs were identified by analysis of raw data from published AF GWAS datasets and the list of m⁶A-SNPs from the m⁶AVar database. Expression quantitative trait loci (eQTL) analysis was conducted to evaluate the effects of m⁶A-SNPs on gene expression. The expression of linked genes was validated in three independent AF-associated gene expression datasets (GSE14975, GSE108660 and GSE2240).

Results: A total of 1429 (6.2%) unique m⁶A-SNPs that were significantly associated with AF were identified. Seventeen m⁶A-SNPs in 14 genes reached genome-wide significance. Eight m⁶A-SNPs demonstrated eQTL signals. Four m⁶A-SNPs (rs383692, rs3211105, rs1061259 and rs1152582) exhibited strong cis-eQTL signals associated with the gene expression levels of SMIM8, JMJD1C and SYNE2. SYNE2 and SMIM8 had differential gene expression levels between the AF and sinus rhythm groups. In addition, SYNE2 expression was uniformly downregulated in AF samples compared with normal control samples in the three datasets.

Conclusions: Our results provide the first demonstration that m⁶A-SNPs are strongly associated with AF, and extend understanding of m⁶A modification as a potential biological pathway underlying AF.

Main article text

Significance Statement

The role of N6-methyladenosine (m⁶A)-associated SNPs in AF remains unknown. Through integrative analysis of a GWAS dataset and m⁶A-SNPs, we successfully identified 1429 unique m⁶A-SNPs significantly associated with AF, 17 of which reached genome-wide significance. Gene expression analysis indicated that genes affected by these m⁶A-SNPs were differentially expressed between AF samples and normal control samples. The results suggest that m⁶A-SNPs may play an important role in the genetics of AF.

Introduction

Atrial fibrillation (AF) is a prevalent type of cardiac arrhythmia with unclear molecular genetics [1, 2]. Linkage analysis in patients with familial or early-onset AF has identified rare pathogenic variants in ion channel genes such as KCNQ1 and SCN5A [2]. Furthermore, population-based genome-wide association studies (GWAS) have identified numerous rare or common single-nucleotide polymorphisms (SNPs) in more than 100 loci. These findings suggest that, in addition to known common factors, genetic components are involved in the development of AF, and the large amount of available genetic data can accelerate functional research on AF pathogenesis. The identification of genetic variations associated with AF susceptibility can provide a better understanding of underlying pathophysiological mechanisms, aid in the development of more robust risk prediction models and improve therapies. Although most AF-associated SNPs are found in noncoding regions, some of these SNPs have been demonstrated to increase AF risk by regulating the expression levels of effector genes [3–5]. However, most SNPs still require functional characterization.

N⁶-methyladenosine (m⁶A) RNA methylation, a prevalent modification in various RNA types, plays essential roles in multiple biological processes, such as RNA-nuclear export, splicing, stability, translation efficiency, RNA-protein interaction and microRNA biogenesis [6]. Recently, research has demonstrated that m⁶A modification of mRNA is crucial for maintaining cardiac function, and its aberrant regulation is associated with cardiac remodeling, ventricular arrhythmia, ischemia-reperfusion injury and heart failure [7–9]. In addition, a recent study has conducted a systematic evaluation of the RNA modification patterns mediated by differential m⁶A regulators in AF versus non-AF samples, and six key differential m⁶A regulators have been identified. These findings have suggested that m⁶A RNA methylation has a diagnostic value as high as 82.5% in distinguishing patients with AF from healthy participants [10]. Similarly, another study has reported that m⁶A RNA modification-associated key regulatory genes, including METTL3, RBM15B and METTL14, show significantly higher expression in the left atrial tissues of patients with AF than patients with sinus rhythm, on the basis of data mining analyses [11]. These results suggest that m⁶A RNA methylation might play a role in the pathogenesis of AF and might be a potential therapeutic target for AF. Further investigation of m⁶A RNA methylation modification in AF is necessary to uncover its specific roles and mechanisms in the pathogenesis of AF.

Substantial progress in the field of bioinformatics, and the increasing availability of public databases and powerful algorithms, has enabled identification of genomic m⁶A-related single nucleotide polymorphisms (m⁶A-SNPs) and their association with various human diseases, such as cancer, cardiovascular diseases, kidney diseases, neurological diseases and immune-associated diseases [12]. Nonetheless, the relationship between m⁶A-SNPs and AF remains poorly understood. To address this knowledge gap, we conducted an integrative analysis of a published GWAS dataset and a comprehensive list of m⁶A-SNPs. Our study was aimed at exploring the association between m⁶A-SNPs and AF, to advance understanding of the molecular basis of AF and identify novel therapeutic targets for this disease.

Methods

Identification of m⁶A-SNPs in Patients with AF

To identify potential AF-associated m⁶A SNPs, we acquired two datasets. The first (AF2018, http://csg.sph.umich.edu/willer/public/afib2018/) was a published GWAS dataset composed of 60,620 AF cases and 970,216 controls, which included association P-values between 34,740,186 genetic variants with a minor allele frequency>2.5×10⁻⁵ and AF [13]. The second was a list of m⁶A-SNPs from the m⁶AVar database (http://m6avar.renlab.org/) [14].

Subsequently, we selected AF associated m⁶A-SNPs by combining the GWAS summary dataset and the list of m⁶A-SNPs (Figure 1). Only SNPs with P-values<0.05 were considered in the following analyses. To determine whether the number of m⁶A-SNPs associated with AF was significantly enriched, we randomly selected non-m⁶A-SNPs from the AF GWAS dataset 1000 times, then compared the number of non-m⁶A-SNPs with a GWAS P-value<0.05 in the 1000 randomizations with the number of m⁶A-SNPs with a GWAS P-value<0.05.

Figure 1

Workflow of the Study Design and the Main Results.

An in silico replication study was conducted with data from a multi-ethnic GWAS on AF that was published in 2018 [15] and is available at http://www.kp4cd.org/datasets/mi. m⁶A-SNPs were replicated if they were associated with AF with P<0.05 in both GWAS datasets used in this work.

Expression Quantitative Trait Locus Analysis of AF-associated m⁶A-SNPs

The main function of m⁶A has been proposed to be the post-transcriptional fine-tuning of gene expression [16]; therefore, m⁶A may be involved in the progression of AF by regulating gene expression. To determine which m⁶A-SNPs might result in gene expression changes, we performed expression quantitative trait locus (eQTL) analysis in different cells and tissues by using HaploReg [17], a web-based variant annotation tool available at https://pubs.broadinstitute.org/mammals/haploreg/haploreg.php. eQTLs are empirically divided into two classes: cis and trans. A cis-eQTL coincides with the location of the underlying gene, whereas the observed location of a trans-eQTL does not coincide with the location of the gene.

Differential Expression Analysis

For m⁶A-SNPs showing eQTL signals, we further investigated whether the expression levels of the genes on which these m⁶A-SNPs showed eQTL effects were associated with AF. To this end, we first downloaded three normalized expression profile datasets (GSE14975, GSE108660 and GSE2240) from the GEO database (https://www.ncbi.nlm.nih.gov/geo/) [18]. GSE14975 included five atrial tissue samples from patients with AF and five samples from normal controls [19]. GSE108660 detected the expression profiles of mRNAs in five patients with AF and five patients without AF. The GSE2240 dataset contained ten samples obtained from patients with permanent AF and 20 samples from patients in sinus rhythm [20, 21]. Subsequently, for each gene m⁶A-SNP showed eQTL effects, we performed differential expression analyses between AF and non-AF samples using t tests in R language (Version 3.6.1). Genes with P-values<0.05 were identified as differentially expressed.

Results

Identification of 17 m⁶A-SNPs Associated with AF with Genome-wide Significance

By mapping 13,703 high, 54,621 medium and 247,269 low confidence level m⁶A-SNPs in human samples from the m⁶AVar database to 34,740,186 genetic variants associated with AF in the AF GWAS dataset, we obtained a total of 22,982 unique m⁶A-SNPs associated with 24,581 m⁶A sites for AF. For the 22,982 unique m⁶A-SNPs identified in the GWAS data, 774, 3883 and 18,325 m⁶A-SNPs were annotated with high, medium and high confidence levels, respectively. Among these m⁶A-SNPs, 1429 (6.2%) unique SNPs were nominally (P-value<0.05) associated with AF. Seventeen m⁶A-SNPs had a genome-wide significant (P-value<5×10⁻⁸) association with AF (Figure 2). Detailed information for the 17 m⁶A-SNPs is provided in Table 1. The locations of exonic m⁶A-SNPs in the target transcripts are shown in Figure 3. For the 17 genome-wide significant AF-associated m⁶A-SNPs, rs1061259 (P-value=5.48×10⁻⁹) and rs1881193 (P-value=2.34×10⁻⁸) were mapped with a medium confidence level, and the remaining 15 were mapped with a low confidence level.

Figure 2

Manhattan Plot of Genome-wide m⁶A-SNPs Identified in AF Samples (P-value<5×10⁻⁸).

The Manhattan plot presents -log₁₀ (P-value) for each of the 22,982 m⁶A-SNPs associated with AF. SNPs with associated P-values less than 5.0×10⁻⁸ are labeled with identifiers.

Table 1

m⁶A-SNPs with Genome-wide Significant Associations with AF.

SNP	Chromosome	Position	Mutation type	P-value	Gene(s)	m6A ID	m6A position	m6A function	Cis-eQTL	Trans-eQTL
rs1061259	chr10	65225244	Exon	5.48E-09	JMJD1C-AS1	m6A_ID_25666	65225245	Loss	1	10
rs1152582	chr14	64692630	3′-UTR	1.75E-23	SYNE2	m6A_ID_276113	64692631	Loss	1	9
rs13220764	chr6	18366699	Exon	2.40E-12	piR-50760, piR-35183, piR-50759	m6A_ID_189225	18366697	Gain	–	–
rs133885	chr22	26159289	Nonsynonymous	2.22E-09	MYO18B	m6A_ID_347214	26159291	Loss	–	–
rs13390491	chr2	179582327	Nonsynonymous	5.00E-18	TTN	m6A_ID_148344	179582345	Loss	0	2
rs1881193	chr17	44248769	Synonymous	2.34E-08	KANSL1	m6A_ID_46559	44248752	Loss	1	3
rs2163009	chr2	179455207	Synonymous	7.74E-15	TTN	m6A_ID_147926	179455191	Loss	0	4
rs2846675	chr11	128787036	3′-UTR	4.38E-09	KCNJ5	m6A_ID_255272	128787035	Loss	–	–
rs3211105	chr10	64945364	Synonymous	1.10E-08	JMJD1C	m6A_ID_235300	64945365	Loss	1	6
rs3812132	chr6	87970301	Synonymous	1.08E-09	ZNF292	m6A_ID_195511	87970328	Loss	2	6
rs3812132	chr6	87970301	Synonymous	1.08E-09	ZNF292	m6A_ID_195504	87970328	Loss	2	6
rs3813250	chr2	179395958	Synonymous	2.53E-15	TTN	m6A_ID_147095	179395966	Loss	0	4
rs383692	chr6	88051444	3′-UTR	1.86E-08	SMIM8	m6A_ID_195543	88051446	Gain	7	18
rs4707358	chr6	87994572	Nonsynonymous	6.11E-09	GJB7	m6A_ID_195538	87994573	Loss	0	9
rs4894029	chr2	179447848	Synonymous	8.35E-15	TTN	m6A_ID_147800	179447846	Loss	0	4
rs4963770	chr12	24720617	Exon	3.08E-10	LINC00477	m6A_ID_258424	24720616	Loss	–	–
rs7729926	chr5	137514641	Intron	5.88E-16	KIF20A	m6A_ID_183887	137514641	Loss	0	5
rs774854	chr3	111688578	Synonymous	9.80E-11	PHLDB2	m6A_ID_161546	111688577	Loss	0	1

Figure 3

Locations of Exonic m⁶A-SNPs in Target Transcripts.

rs774854, rs3813250, rs4707358, rs133885 and rs3211105 are located in coding regions. rs383692, rs2846675 and rs1152582 are located in 3′-UTRs. The pink arrow in each transcript indicates the coding region.

In silico Replication of the m⁶A-SNPs

We randomly selected 22,982 non-m⁶A-SNPs from the AF GWAS dataset and calculated the number of SNPs with GWAS P-values<0.05. This randomization process was repeated 1000 times. On the basis of the 1000 randomizations, the proportion of AF associated m⁶A-SNPs with a GWAS P-value<0.05 was significantly higher than that of non-m⁶A-SNPs (95% CI: [5.76%, 5.78%], P-value<2.2×10⁻¹⁶). For the 1429 m⁶A-SNPs nominally associated with AF, we also performed an in silico replication by using a multi-ethnic GWAS dataset for AF in 2018, which contained fewer SNPs but more AF samples than the AF2018 dataset [15]. A total of 623 unique SNPs were nominally (P-value<0.05) associated with AF, and 86 m⁶A-SNPs overlapped with those from the AF2018 dataset.

A SNP is defined as an m⁶A-SNP if it has the potential to alter the DRACH motif [22] (where D=A, G or U; R=A or G; H=A, C or U) or other sequence features essential for m⁶A modification [14]. When examining the sequences around m⁶A sites for the 17 m⁶A-SNPs with genome-wide significant associations with AF, we observed that ten SNPs were located in the DRACH motif (Figure 4A). Notably, five SNPs (rs2846675, rs3211105, rs4707358, rs7729926 and rs774854) altered the core AC motif (Figure 4B).

Figure 4

The Signature of Alterations of the m⁶A Motif by the SNPs.(A) Nucleotide sequence composition around m⁶A residues for the 11 m⁶A-SNPs. The nucleotide where the SNP is located is shown in blue. (B) Frequency of mutations located in the DRACH motif (D=A, G or U; R=A or G; H=A, C or U).

Identification of 13 m⁶A-SNPs Showing eQTL Signals

To investigate the possible functional mechanisms underlying the 17 identified m⁶A-SNPs with genome-wide significant associations with AF, we tested their association with gene expression levels by using HaploReg. In total, 13 of the 17 AF-associated m⁶A-SNPs showed eQTL signals in various tissues or cells (Table 1; details in Table S1). Six m⁶A-SNPs displayed strong cis-eQTL signals with the gene being located, whereas the remaining seven m⁶A-SNPs showed only trans-eQTL signals. In particular, rs383692 (SMIM8) had strong cis-eQTL signals with SMIM8 in seven different tissues (Table S1). rs3211105 and rs1061259 in JMJD1C were found to be associated with JMJD1C expression in the testis, with P-values of 3.63×10⁻⁶ and 1.51×10⁻⁵, respectively. rs1152582 (SYNE2) was associated with SYNE2 expression in the left ventricle, with a P-value of 3.37×10⁻⁸.

Differential Expression Analysis

For the 13 m⁶A-SNPs displaying eQTL signals, we assessed the expression levels of their associated genes in three expression profile datasets (GSE14975, GSE108660 and GSE2240). After removing genes that were not detected on the microarray, we finally performed differential expression analysis for 19 associated genes (Table S2). Four genes, SYNE2, SMIM8, ZNF292 and JMJD1C-AS1, were differentially expressed in at least one of the three datasets. The gene SYNE2 was significantly down-regulated in AF samples compared with sinus rhythm samples in all three datasets (Figure 5A). Furthermore, the gene SMIM8 was significantly downregulated in two of the three datasets (Figure 5B). According to the eQTL results from HaploReg, the expression levels of SYNE2 were associated with m⁶A-SNP rs1152582 (cis-eQTL, P-value=3.37×10⁻⁸). The expression level of SMIM8 was associated with rs383692 (cis-eQTL), rs4707358 (trans-eQTL) and rs3812132 (trans-eQTL).

Figure 5

Expression Patterns of the Genes SYNE2 and SMIM8 in the AF and Sinus Rhythm Samples From the Datasets GSE14975, GSE108660 and GSE2240.

(A) The expression of SYNE2 was significantly downregulated in the AF group compared with the sinus rhythm group in all three datasets. (B) The expression of SMIM8 was significantly downregulated in the AF group compared with the sinus rhythm group according to the GSE108660 and GSE2240 datasets.

Discussion

Despite recent advances in GWAS data that have identified many AF-associated loci, methods to explore their functional implications have been less explored. To date, no genetic or experimental evidence has demonstrated an association between m⁶A RNA methylation and AF. The detection of m⁶A-SNPs enables the potential function of a subset of SNPs to be interpreted from the perspective of RNA methylation. In this study, we report the first exploration of the relationship between m⁶A-SNPs and AF through mining data from a large-scale GWAS. We identified 1429 m⁶A-SNPs associated with AF, among which 17 unique m⁶A-SNPs reached genome-wide significance. Among these AF-associated SNPs, five m⁶A-SNPs showed strong cis-eQTL signals with the genes in which they were located. Differential expression analysis indicated that SYNE2 was significantly down-regulated in AF. We hypothesized that m⁶A methylation modifications might participate in AF development by regulating SYNE2 expression. Our study provides population genetic evidence suggesting that m⁶A modifications are likely to be involved in the development of AF and to be associated with AF susceptibility.

Appropriate m⁶A RNA modification relies on the precise recognition and binding of RNA to methylation enzymes. SNPs influence RNA modification by affecting RNA secondary structures, RNA-protein interactions and exon splicing [23–25]. In particular, m⁶A is affected by SNPs occurring in the specific motif of the target RNA or the key flanking nucleotides [26]. Here, we identified 17 AF-associated m⁶A-SNPs in 14 genes (JMJD1C, JMJD1C-AS1, KANSL1, Pir-50760, SYNE2, MYO18B, TTN, KCNJ5, ZNF292, SMIM8, GJB7, LINC00477, KIF20A and PHLDB2). Among these genes, seven (SYNE2 [13, 15], MYO18B [13, 15], TTN [27], KCNJ5 [13, 15, 28], ZNF292 [15], LINC00477 [15] and PHLDB2 [15]) have been reported to be associated with AF in previous studies, whereas JMJD1C, JMJD1C-AS1, Pir-50760, KANSL1, SMIM8, GJB7 and KIF20A have not yet been reported. According to the Bgee database (https://bgee.org) [29], these genes are expressed in heart tissue. However, the functions of these genes in the heart, particularly in AF, remain to be further explored. The results of our study suggest novel potential links between each of the seven genes (JMJD1C, JMJD1C-AS1, Pir-50760, KANSL1, SMIM8, GJB7 and KIF20A) and AF development. Additionally, m⁶A RNA methylation may be the underlying mechanism.

Recent transcriptome-wide m⁶A maps suggest that m⁶A modification is restricted predominantly to adenosines in the DRACH motif [30, 31]. We compared the ten m⁶A-SNPs containing motifs to human assembly GRCh37 reference sequences and found that “AC” nucleotides were relatively evolutionarily conserved. This finding indicated that “AC” nucleotides in the m⁶A motif play a crucial role in m⁶A modification, in agreement with findings from previous studies [9, 22]. In addition, five of ten m⁶A-SNPs occurred in “AC” nucleotides, thereby suggesting that variants in “AC” in the m⁶A motif may substantially affect the physiological functions of m⁶A modification.

The novel m⁶A motifs derived by SNPs have a higher frequency than matched control alleles or neutral sites, thus suggesting that some of these SNPs are subject to positive selection [32]. We observed no significant difference in allele frequency between the loss-of-function m⁶A-SNPs and reference alleles, thus suggesting that AF-associated m⁶A-SNPs did not have a selection trend in the population. This result might have been be partly due to the limited number of variants that we identified. Our eQTL analysis revealed that the AF-associated m⁶A-SNP rs1152582 was highly correlated with the expression level of SYNE2 (cis-eQTL). The SNP rs1152582 is located in the 3’-UTR of the SYNE2 gene, in agreement with previous observations that most m⁶A site enrichment occurs in 3´-terminal exons and the 3’-UTR [22]. m⁶A modification regulates mRNA stability and target gene expression. We further used three independent GEO datasets to confirm that SYNE2 expression was significantly diminished in the AF group, thus suggesting that differences in SYNE2 expression may contribute to AF development and that m⁶A methylation modifications may be involved in the regulation of SYNE2 expression.

The SYNE2 gene is ubiquitously expressed and consists of 116 exons that encode the Nesprin-2 protein. Alternative splicing of gene produces multiple isoform transcripts with tissue-specific expression patterns [33]. Notably, Nesprin-2α1, a small isoform encoded by exons 106–116, is highly expressed in cardiac and skeletal muscle cells [33]. Nesprin-2, located mainly in the outer nuclear membrane of the nuclear envelope, contains two major domains: the N-terminal actin-binding domain and the C-terminal Klarsicht/Anc-1/Syne homology (KASH) domain [34]. Nesprin-2 functions as a scaffold linking the nucleoskeleton and cytoskeleton, and forms part of the LINC complex [35]. Interactions with various partners, such as F-actin, SUN2, Lamin A/C and Emerin, regulate Nesprin-2’s functions [33, 36, 37]. In cardiomyocytes, Nesprin-2 is crucial for maintaining nuclear morphology, positioning, mechanotransduction and nuclear trafficking [38–41]. Knocking out Nesprin-1/2 has been shown to cause early-onset cardiomyopathy and cardiac dysfunction [41]. These results suggest that SYNE2 plays a critical role in cardiac physiology. However, further investigations are necessary to determine the additional functions of this gene.

Conclusion

In conclusion, our study identified several m⁶A-SNPs that may contribute to the genetic risk of AF, thus highlighting the potential role of m⁶A modification in AF pathogenesis. Our findings also emphasize the crucial role of SYNE2 in AF. However, the lack of in vivo experimental verification of the effects of genetic variations on m⁶A modification in candidate transcripts and their subsequent functions in the pathogenesis of AF is a substantial limitation of this study. Future studies should focus on functional experimental verification to investigate the specific mechanisms involved in these genetic contributions. Comprehensive elucidation of the relationship between m⁶A modification and AF would deepen understanding of AF pathogenesis and potentially lead to the development of targeted therapies for this disease.

Conflict of interests

The authors declare no conflicts of interest.

Citation Information

Download Citation.

References

1. W Wang, Y Tian, W Wang, H Yin, D Yin, Y Tian. Role of epicardial adipose tissue in triggering and maintaining atrial fibrillation. Cardiovasc Innov Appl 2022. https://doi.org/10.15212/CVIA.2022.0012 .

2. JA Kim, MG Chelu, N Li. Genetics of atrial fibrillation. Curr Opin Cardiol 2021;36:281–7.

3. S Mahida, PT Ellinor. New advances in the genetic basis of atrial fibrillation. J Cardiovasc Electrophysiol 2012;23:1400–6.

4. P Kirchhof, PC Kahr, S Kaese, I Piccini, I Vokshi, H-H Scheld, et al. PITX2c is expressed in the adult left atrium, and reducing Pitx2c expression promotes atrial fibrillation inducibility and complex changes in gene expression. Circ Cardiovasc Genet 2011;4:123–33.

5. F Syeda, P Kirchhof, L Fabritz. PITX2-dependent gene regulation in atrial fibrillation and rhythm control. J Physiol (Lond) 2017;595:4019–26.

6. K Boulias, EL Greer. Biological roles of adenine methylation in RNA. Nat Rev Genet 2022. https://doi.org/10.1038/s41576-022-00534-0 .

7. LE Dorn, L Lasman, J Chen, X Xu, TJ Hund, M Medvedovic, et al. The N6-Methyladenosine mRNA methylase METTL3 controls cardiac homeostasis and hypertrophy. Circulation 2019;139:533–45.

8. P Mathiyalagan, M Adamiak, J Mayourian, Y Sassi, Y Liang, N Agarwal, et al. FTO-dependent N6-methyladenosine regulates cardiac function during remodeling and repair. Circulation 2019;139:518–32.

9. T Berulava, E Buchholz, V Elerdashvili, T Pena, MR Islam, D Lbik, et al. Changes in m6A RNA methylation contribute to heart failure progression by modulating translation. Eur J Heart Fail 2020;22:54–66.

10. P-F Zheng, S-Y Zhou, C-Q Zhong, Z-F Zheng, Z-Y Liu, H-W Pan, et al. Identification of m6A regulator-mediated RNA methylation modification patterns and key immune-related genes involved in atrial fibrillation. Aging (Albany NY) 2023;15:1371–93.

11. Q Wang, S Zhang, X Xu, J Liu, P Tan, C Wang, et al. The nine RNA methylation regulatory gene signature is associated with the pathogenesis of atrial fibrillation by modulating the immune microenvironment in the atrial tissues. Dis Markers 2023;2023:e7277369.

12. A Cayir. RNA modifications as emerging therapeutic targets. Wiley Interdiscip Rev RNA 2022;13:e1702.

13. N Jb, T Rb, F Lg, Z W, S Mw, G Se, et al. Biobank-driven genomic discovery yields new insight into atrial fibrillation biology. Nat Genet 2018;50:1234–9.

14. S Jiang, Y Xie, Z He, Y Zhang, Y Zhao, L Chen, et al. m6ASNP: a tool for annotating genetic variants by m6A function. Gigascience 2018;7:giy035.

15. C Roselli, MD Chaffin, L-C Weng, S Aeschbacher, G Ahlberg, CM Albert, et al. Multi-ethnic genome-wide association study for atrial fibrillation. Nat Genet 2018;50:1225–33.

16. JY Roignant, M Soller. m⁶A in mRNA: an ancient mechanism for fine-tuning gene expression. Trends Genet 2017;33:380–90.

17. LD Ward, M Kellis. HaploReg: a resource for exploring chromatin states, conservation, and regulatory motif alterations within sets of genetically linked variants. Nucleic Acids Res 2012;40 Database issue:D930–4.

18. E Clough, T Barrett. The gene expression omnibus database. Methods Mol Biol 2016;1418:93–110.

19. O Adam, D Lavall, K Theobald, M Hohl, M Grube, S Ameling, et al. Rac1-induced connective tissue growth factor regulates connexin 43 and N-cadherin expression in atrial fibrillation. J Am Coll Cardiol 2010;55:469–80.

20. AS Barth, S Merk, E Arnoldi, L Zwermann, P Kloos, M Gebauer, et al. Functional profiling of human atrial and ventricular gene expression. Pflugers Arch 2005;450:201–8.

21. AS Barth, S Merk, E Arnoldi, L Zwermann, P Kloos, M Gebauer, et al. Reprogramming of the human atrial transcriptome in permanent atrial fibrillation: expression of a ventricular-like genomic signature. Circ Res 2005;96:1022–9.

22. S Ke, EA Alemu, C Mertens, EC Gantman, JJ Fak, A Mele, et al. A majority of m6A residues are in the last exons, allowing the potential for 3’ UTR regulation. Genes Dev 2015;29:2037–53.

23. X Wu, LD Hurst. Determinants of the usage of splice-associated cis-motifs predict the distribution of human pathogenic SNPs. Mol Biol Evol 2016;33:518–29.

24. F Mao, L Xiao, X Li, J Liang, H Teng, W Cai, et al. RBP-Var: a database of functional variants involved in regulation mediated by RNA-binding proteins. Nucleic Acids Res 2016;44:D154–63.

25. G Ramaswami, P Deng, R Zhang, M Anna Carbone, TFC Mackay, J Billy Li. Genetic mapping uncovers cis-regulatory landscape of RNA editing. Nat Commun 2015;6:8194.

26. Y Zheng, P Nie, D Peng, Z He, M Liu, Y Xie, et al. m6AVar: a database of functional variants involved in m6A modification. Nucleic Acids Res 2018;46:D139–45.

27. SH Choi, SJ Jurgens, L-C Weng, JP Pirruccello, C Roselli, M Chaffin, et al. Monogenic and polygenic contributions to atrial fibrillation risk: results from a national biobank. Circ Res 2020;126:200–9.

28. J Jabbari, MS Olesen, AG Holst, JB Nielsen, S Haunso, JH Svendsen. Common polymorphisms in KCNJ5 [corrected] are associated with early-onset lone atrial fibrillation in Caucasians. Cardiology 2011;118:116–20.

29. F Bastian, G Parmentier, J Roux, S Moretti, V Laudet, M Robinson-Rechavi. Bgee: integrating and comparing heterogeneous transcriptome data among species. Data Integration in the Life Sciences 2008:124–31.

30. S Schwartz, MR Mumbach, M Jovanovic, T Wang, K Maciag, GG Bushkin, et al. Perturbation of m6A writers reveals two distinct classes of mRNA methylation at internal and 5’ sites. Cell Rep 2014;8:284–96.

31. B Linder, AV Grozhik, AO Olarerin-George, C Meydan, CE Mason, SR Jaffrey. Single-nucleotide-resolution mapping of m6A and m6Am throughout the transcriptome. Nat Methods 2015;12:767–72.

32. H Zhang, X Shi, T Huang, X Zhao, W Chen, N Gu, et al. Dynamic landscape and evolution of m6A methylation in human. Nucleic Acids Res 2020;48:6251–64.

33. Q Zhang, CD Ragnauth, JN Skepper, NF Worth, DT Warren, RG Roberts, et al. Nesprin-2 is a multi-isomeric protein that binds lamin and emerin at the nuclear envelope and forms a subcellular network in skeletal muscle. J Cell Sci 2005;118(Pt 4):673–87.

34. T Libotte, H Zaim, S Abraham, VC Padmakumar, M Schneider, W Lu, et al. Lamin A/C-dependent localization of Nesprin-2, a giant scaffolder at the nuclear envelope. Mol Biol Cell 2005;16:3411–24.

35. M Crisp, Q Liu, K Roux, JB Rattner, C Shanahan, B Burke, et al. Coupling of the nucleus and cytoplasm: role of the LINC complex. J Cell Biol 2006;172:41–53.

36. S Vlcek, R Foisner. Lamins and lamin-associated proteins in aging and disease. Curr Opin Cell Biol 2007;19:298–304.

37. EC Schirmer, R Foisner. Proteins that associate with lamins: many faces, many functions. Exp Cell Res 2007;313:2167–79.

38. MJ Stroud, I Banerjee, J Veevers, J Chen. Linker of nucleoskeleton and cytoskeleton complex proteins in cardiac structure, function, and disease. Circ Res 2014;114:538–48.

39. JA Mellad, DT Warren, CM Shanahan. Nesprins LINC the nucleus and cytoskeleton. Curr Opin Cell Biol 2011;23:47–54.

40. ML Lombardi, J Lammerding. Keeping the LINC: the importance of nucleocytoskeletal coupling in intracellular force transmission and cellular function. Biochem Soc Trans 2011;39:1729–34.

41. I Banerjee, J Zhang, T Moore-Morris, E Pfeiffer, KS Buchholz, A Liu, et al. Targeted ablation of nesprin 1 and nesprin 2 from murine myocardium results in cardiomyopathy, altered nuclear morphology and inhibition of the biomechanical gene response. PLoS Genet 2014;10:e1004114.

Supplementary Material

Supplementary material is available for this paper at the following links: https://cvia-journal.org/wp-content/uploads/2023/04/Table_S1_Detail_information_of_the_14_AF-associated_m6A-SNPs_showed_eQTL_signals_with_local_genes_in_a_variety_of_tissues_or_cells.pdf

https://cvia-journal.org/wp-content/uploads/2023/04/Table_S2_The_expression_pattern_of_genes_displayed_eQTL_signals_with_13_m6A-SNPs_in_three_dataset.pdf