Skip to main content
Download PDF

Epigenome-wide association study identifies novel genes associated with ischemic stroke

Clinical Epigenetics volume 15, Article number: 106 (2023) Cite this article

Abstract

Background

DNA methylation has previously been associated with ischemic stroke, but the specific genes and their functional roles in ischemic stroke remain to be determined. Here we aimed to identify differentially methylated genes that play a functional role in ischemic stroke in a Chinese population.

Results

Genome-wide DNA methylation assessed with the Illumina Methylation EPIC Array in a discovery sample including 80 Chinese adults (40 cases vs. 40 controls) found that patients with ischemic stroke were characterized by increased DNA methylation at six CpG loci (individually located at TRIM6, FLRT2, SOX1, SOX17, AGBL4, and FAM84A, respectively) and decreased DNA methylation at one additional locus (located at TLN2). Targeted bisulfite sequencing confirmed six of these differentially methylated probes in an independent Chinese population (853 cases vs. 918 controls), and one probe (located at TRIM6) was further verified in an external European cohort (207 cases vs. 83 controls). Experimental manipulation of DNA methylation in engineered human umbilical vein endothelial cells indicated that the identified differentially methylated probes located at TRIM6, TLN2, and FLRT2 genes may play a role in endothelial cell adhesion and atherosclerosis.

Conclusions

Altered DNA methylation of the TRIM6, TLN2, and FLRT2 genes may play a functional role in ischemic stroke in Chinese populations.

Background

Stroke is one of the leading causes of mortality and morbidity worldwide, second only to ischemic heart disease (IHD) [1]. In China, stroke mortality surpasses that of IHD, accounting for around one-third of all stroke cases worldwide [2]. Ischemic stroke (IS) constitutes more than 80% of all stroke cases [3]. The etiology of IS is multifactorial, involving both genetic and environmental factors [4, 5]. Despite the known contribution of genetic factors to the etiology of stroke, genetic variants identified so far explain less than 10% of the disease variability, suggesting that other factors may also play an important role in IS. Additionally, identified genetic risk variants for stroke confer race-specific effects [6], revealing the existence of underlying biological mechanisms that might explain the variability of stroke prevalence across different ancestries. The epigenome responds to a wide range of environmental stressors, providing a mechanism through which environmental factors affect the susceptibility to human complex diseases, including IS. Several epigenome-wide association studies (EWAS) have reported associations of altered DNA methylation with IS [7–[12]. However, prior studies were mostly conducted in European populations and results were largely inconsistent. Moreover, few studies have attempted to functionally validate the identified epigenetic markers in cellular models [13]. Here, we used a multi-stage design to identify DNA methylation alterations associated with IS in Chinese participants: (1) EWAS in a discovery Chinese cohort (discovery stage, n = 80); (2) confirmation and fine mapping of candidate regions by targeted methylation sequencing in an independent Chinese population (internal validation, n = 1,771); (3) replication in a European population (external replication, n = 390); and (4) functional validation of top hits in an endothelial cell model via the CRISPR/dCas9-Dnmt3a system.

Results

The study design is illustrated in Fig. 1. Genome-wide DNA methylation was assessed with the Illumina Methylation EPIC Array in a discovery sample including 80 Chinese adults (40 cases vs. 40 controls). Differentially methylated probes (DMPs) identified were further confirmed by targeted bisulfite sequencing (TBS) in an independent sample (n = 1,771), followed by external replication in a European cohort (n = 390). The potential functional role of DMPs was verified in a CRISPR/dCas9-Dnmt3a model in human umbilical vein endothelial cells (HUVECs). Clinical characteristics of participants in each stage are shown in Table 1. Compared with healthy controls, IS cases had more risk factors such as hypertension, diabetes, and dyslipidemia in participants at all stages.

Fig. 1
figure 1

A flowchart illustrating the study design and selection of participants. In the discovery stage, 40 CATIS IS cases and 40 PMMS controls were randomly selected to identify potential epigenetic markers of IS. From the remaining CATIS and PMMS participants, 1000 cases and 1000 age- and sex-matched controls were randomly selected as an internal replication sample to further confirm epigenetic markers of IS; after QC, 853 cases and 918 controls remained for downstream analysis. Additionally, an external cohort of 390 Spanish participants was used for further validation of the DMPs identified. To test the causal effect on gene expression of DNA methylation at loci identified in the aforementioned stages, human umbilical vein endothelial cell (HUVEC) models were epigenetically modified via CRISPR/dCas9-Dnmt3a. DMG, differentially methylated gene; DMP, differentially methylated probe; DMR, differentially methylated region; HUVEC, human umbilical vein endothelial cell; IS, ischemic stroke; TBS, targeted bisulfite sequencing

Full size image
Table 1 Clinical characteristics of study participants in the discovery and replication samples at enrollment
Full size table

DMPs identified in association with IS

Of the 815,017 CpG sites tested, single-probe analysis identified 46 differentially methylated positions (DMPs) associated with IS at P < 1.0 × 10–5 (Table 2), after adjusting for potential confounding factors in the discovery sample. Seven (out of 46) DMPs had q < 0.05 (Fig. 2) after further correction for multiple testing via FDR. Of these 7 DMPs, 6 DMPs (cg15137954 in TRIM6, cg11199713 in SOX1, cg07913294 in AGBL4, cg16800165 in FLRT2, cg11533098 in FAM84A, and cg15949239 in TLN2) were successfully confirmed (q ≤ 0.05) by targeted bisulfite sequencing (TBS) in an independent Chinese population comprising 853 cases and 918 controls (Table 3). In addition, altered DNA methylation at neighboring CpGs showed significant associations with IS in the same directions (region-based P < 5 × 10–7). Among these 6 DMPs validated in the Chinese cohorts, one probe (cg15137954) was also significantly associated with IS (same direction) in the European cohort (multiple testing adjusted P = 0.01). Complete statistics for the external validation are included in Table 3.

Table 2 Significant DMPs associated with ischemic stroke in the discovery sample (P < 1 × 10–5)
Full size table
Fig. 2
figure 2

EWAS in the discovery stage. Manhattan (a) and volcano (b) plots displaying the associations of DNA methylation at 815,017 CpGs with IS in a Chinese sample (40 cases and 40 matched controls). In the Manhattan plot, X-axis indicates the chromosome position; Y-axis indicates the –log10 P-values. The red line denotes the significance threshold at q-value < 0.05. In the volcano plot, blue and red dots highlight significantly hypomethylated and hypermethylated CpG sites, respectively, in cases with respect to controls. c Box plots showing the differences in mean DNA methylation level between cases and controls at the 7 CpG sites significantly associated with IS in the discovery stage

Full size image
Table 3 DMPs significantly associated with ischemic stroke in the discovery (q < 0.05) and replication samples
Full size table

Functional validation in cellular models

To test the potential impact of DNA methylation at the CpG sites identified in a relevant tissue for stroke pathogenesis, the CRISPR/dCas9-Dnmt3a system was used to experimentally induce increased DNA methylation levels in the gene regions of interest in human umbilical endothelial cells (HUVECs). After generating the epigenetically modified cells, pyrosequencing was used to check the validity of the model, i.e., to ascertain increased DNA methylation levels in the CRISPR-modified cells. Of the 7 DMPs identified in the discovery stage, DNA methylation levels of 3 CpG sites (cg15137954, cg16800165, and cg15949239) in the engineered HUVECs were two times higher when compared to the negative control cells, denoting successful construction of the cellular models by the CRISPR/dCas9-Dnmt3a system (Fig. 3). The other 4 CpG sites (cg24891539, cg11199713, cg07913294, and cg11533098) exhibited similar methylation levels when comparing the two groups (Fig. 3). Gene expression analysis by qRT-PCR revealed that, compared to the negative control cells, the dCas9-Dnmt3a engineered HUVECs with an increased methylation level at cg15137954 (located in the TRIM6 gene) resulted in decreased expression of the TRIM6, VCAM1, and SELE genes (Fig. 4a). In contrast, enhanced methylation at cg16800165 (located in FLRT2) led to reduced expression of the FLRT2 gene and increased expression of the endothelial cell adhesion marker genes ICAM1, VCAM1, and SELE (Fig. 4b). In addition, increased DNA methylation at cg15949239 (located at TLN2) was associated with a heightened expression of the TLN2 gene together with decreased expression of the cell adhesion marker genes ICAM1, VCAM1, and SELE (Fig. 4c). Together, these findings suggest that DNA methylation levels at the identified CpG sites regulate gene expression of their corresponding genes; additionally, it points to a functional role of the identified DMPs in endothelial cell activation and vascular inflammation.

Fig. 3
figure 3

Validation of the CRISPR/dCas9-Dnmt3a model construction in human umbilical vein endothelial cells (HUVEC). Pyrosequencing was used to measure DNA methylation levels in the engineered and control cells. Successful model construction was considered when DNA methylation level of the engineered cells was higher than that of the negative control (i.e., TRIM6, FLRT2, and TLN2 genes). No significant differences in DNA methylation levels were observed for SOX1, FAM84A, SOX17, and AGBL4 genes

Full size image
Fig. 4
figure 4

The expression of differentially methylated genes and genes encoding endothelial cell adhesion markers (i.e., ICAM1, VCAM1, and SELE) in site-specific methylation model HUVECs (dCase9-Dnmt3a) and negative control HUVECs. (a) TRIM6 model: (left) methylation level at cg15137954 is higher in dCas9-Dnmt3a engineered cells than in negative control cells; (center) higher methylation is associated with decreased TRIM6 expression; (right) decreased expression of VCAM1 and SELE can be observed in dCas9-Dnmt3a engineered cells. (b) FLRT2 model: (left) methylation level at cg16800165 is higher in dCas9-Dnmt3a engineered cells than in negative control cells; (center) FLRT2 expression is significantly decreased in dCas9-Dnmt3a engineered cells; (right) expression of genes encoding endothelial cell adhesion markers is significantly increased in dCas9-Dnmt3a cells. (c) TLN2 model: (left) methylation at cg15949239 is higher in dCas9-Dnmt3a cells than in the negative control cells; (center) TLN2 expression is significantly increased in dCas9-Dnmt3a cells; (right) expression of genes encoding endothelial cell adhesion markers is significantly decreased in dCas9-Dnmt3a engineered cells when compared to negative control cells. Relative mRNA expression of dCase9-Dnmt3a to negative control was calculated using the 2(-Delta Delta C(T)) method and normalized with the reference gene Actin (3 repeated measurements per group). Data are mean ± SEM (standard error of mean). A Student t test was used to compare the group difference in gene expression levels. *P < 0.05; **P < 0.01; ***P < 0.001

Full size image

DMRs identified in association with IS in the discovery sample

Region-based analysis identified 149 DMRs (annotated to 168 unique genes) significantly associated with IS at q < 0.05 (Fig. 5a, Additional file 1: Table S1). Of these, 129 DMRs were hypomethylated and 20 DMRs were hypermethylated in relation to IS (Fig. 5b). The identified DMRs are mainly located in the first exon and promoter regions. With respect to CpG context, the identified DMRs are largely located within CpG Islands (Fig. 5c).

Fig. 5
figure 5

Manhattan (a) and volcano (b) plots displaying the differentially methylated regions (DMRs) associated with ischemic stroke in the discovery sample of 80 Chinese adults (40 cases vs. 40 matched controls). In the Manhattan plot, X-axis indicates the chromosome position; Y-axis indicates the –log10 P-values. The red line denotes the significance threshold at q-value < 0.05. In the volcano plot, blue and red dots highlight significantly hypomethylated and hypermethylated regions, respectively, in cases with respect to controls. Genomic distribution and CpG context (c) of the identified DMRs indicated that these DMRs are mainly located in the first exon, promoter, and CpG Islands

Full size image

Functional annotation of the differentially methylated genes associated with IS

The 7 DMPs and 149 DMRs identified in the discovery sample were annotated to 170 unique genes. These genes were significantly enriched in 351 GO biological processes including neuron differentiation, neurogenesis, and cell–cell signaling (Additional file 1: Table S2), 37 GO cellular components including neuron and cell junction (Additional file 1: Table S3), and 12 GO molecular functions including DNA binding transcription factor activity (Additional file 1: Table S4). The same gene set was also enriched in several GTEx tissues, including brain, thyroid, nerves, and blood vessels (Additional file 1: Table S5). Finally, the identified differentially methylated genes were significantly enriched in GWAS risk genes associated with stroke-related phenotypes such as chronotype, obesity-related traits, coronary artery calcification, and blood pressure (Additional file 1: Table S6). These findings provide further support for the potential functional role of the identified genes in IS.

Discussion

In this large-scale EWAS, we employed a multi-stage approach to identify peripheral blood DNA methylation patterns associated with IS in Chinese population. Specifically, we identified DNA methylation at 7 CpGs (annotated to 7 unique genes) significantly associated with IS. Of these, 6 CpGs (cg15137954 in TRIM6, cg07913294 in AGBL4, cg11533098 in FAM84A, cg24891539 in SOX17, cg16800165 in FLRT2, and cg15949239 in TLN2) were confirmed in an independent Chinese population, and one CpG (cg15137954, located in the TRIM6 gene) was further verified in an independent cohort of European ancestry. Functional analysis in cellular models via the CRISPR/dCas9-Dnmt3a system revealed that DNA methylation at three genes (TRIM6, FLRT2, and TLN2) harboring the aberrantly methylated probes could regulate the expression of three endothelial cell adhesion markers (i.e., ICAM1, VCAM1, and SELE) and thus be functionally implicated in endothelial cell activation and leukocyte transendothelial migration [28], one of the mechanisms underlying atherosclerosis [29, 30] and stroke [31].

Of the differentially methylated genes identified, the observed association of DNA hypomethylation of the tripartite motif containing 6 (TRIM6) gene with IS is especially interesting. This gene belongs to the TRIM gene family, which has been associated with innate immunity, a mechanism known to be implicated in IS and atherosclerosis [32]. In this study, we found that experimentally induced increased DNA methylation at the TRIM6 locus resulted in decreased expression of the TRIM6 gene as well as reduced expression of genes encoding two endothelial cell adhesion markers (VCAM1 and SELE). These results point to an increased expression of TRIM6, VCAM1, and SELE in IS cases. The TRIM6 gene was confirmed in both internal and external replications and further verified in the cellular models. Previous research in human cells has shown that several TRIM family genes (e.g., TRIM5, TRIM34) were upregulated in response to interferon gamma (IFN-γ) and lipopolysaccharide (LPS), while TRIM5 protein was involved in the containment of viral spread in rhesus monkeys [33, 34]. Of note, acute viral infections have been suggested as potential triggers for IS due to destabilization and rupture of atherosclerotic plaques in humans [35, 36]. In addition, genetic polymorphisms in the TRIM5/TRIM6 locus have previously been associated with coronary artery disease [37]. Moreover, TRIM6 has been reported to aggravate myocardial ischemia/reperfusion injury in mice [38]. A genetic polymorphism at another TRIM family member (TRIM36) has recently been associated with cardioembolic stroke, a subtype of ischemic stroke, in a genome-wide association analysis [39]. Together, these findings support a causal role of altered TRIM6 gene methylation in the pathogenesis of IS.

The Talin 2 (TLN2) gene encodes a protein related to talin 1, a cytoskeletal protein that is involved in the assembly of actin filaments and the spreading and migration of various cell types, and may play an important role in cell adhesion [40]. Differential expression of the TLN2 gene was previously reported to modulate carotid intima media thickness in European individuals [41]. Moreover, downregulation of the TLN2 gene has been associated with atherosclerosis in human arterial plaque tissue samples [42]. In the present study, we found that stroke patients were characterized by TLN2 hypermethylation, while in vitro modulation of DNA methylation at the TLN2 locus via CRISPR/dCas9-Dnmt3a system revealed a positive association between TLN2 methylation and TLN2 expression. Moreover, TLN2 methylation was negatively correlated with the expression of genes encoding endothelial cell adhesion markers. The FLRT2 (fibronectin leucine-rich transmembrane protein 2) gene is expressed in pancreas, skeletal muscle, brain, and heart [43]. This gene was previously found to be involved in the regulation of early embryonic vascular and neural development [43], and a genetic polymorphism of the FLRT2 gene has been associated with IS in African-Americans [44]. In this study, we found that stroke patients exhibited lower methylation level of the FLRT2 gene compared to controls. Additionally, increased FLRT2 methylation was associated with decreased FLRT2 expression and increased expression of genes encoding endothelial cell adhesion markers. Together, these findings suggest that the observed changes in DNA methylation in TLN2 and FLRT2 genes in IS patients are associated with decreased gene expression levels of cell adhesion markers. These observations support the role of DNA methylation in responding to ischemic injury and highlight its potential role in stroke prognosis rather than its pathogenesis.

With regard to DMR findings, out of 149 DMRs identified, 2 DMRs were located in genes previously associated with stroke in a multi-ancestry GWAS: FBN2 and PDE3A genes [39]. Interestingly, the SNP rs55670004 previously associated with IS has been identified as a methylation quantitative trait loci (mQTL) for CpG sites cg02271895 and cg17564775, both of which are included in the FBN2 DMR (chr5:127,872,329–127,875,163) associated with IS in the current study. No mQTLs were found for PDE3A gene. These results highlight that at least some of the identified DMRs are driven by the underlying genetic background of assessed participants suggesting that DNA methylation at these regions might play a causal role in stroke etiology. Additionally, 17 out of 168 unique genes located in close proximity to identified DMRs were already reported by a previous EWAS of IS in a Chinese population [11]. Of note, one of these 17 overlapping genes was FBN2, further highlighting its role in IS etiology.

In line with previous research [12], we found that the majority of the genes identified in our study, both in the DMP and DMR approaches, are hypomethylated with respect to stroke status (i.e., DNA methylation was lower in cases than in controls). The identified genes are significantly enriched in biological pathways involved in neurogenesis, neuron differentiation, and neuronal development, as well as nervous cellular components, such as neurons and synapses. These findings suggest that peripheral blood DNA methylation may serve as a proxy for DNA methylation in the brain, the target organ of stroke. Further studies in postmortem brain tissue will reveal whether the observed DNA methylation changes in blood of IS survivors are also present in brain tissue of deceased participants that underwent IS. Other pathways highlighted included cell adhesions, cell junctions, and extracellular matrix, all of which are known to be involved in leukocyte transendothelial migration and atherogenesis, a major contributor to IS [45]. Several pathways identified in the present study were also reported in a previous EWAS of IS in a Chinese population, including synapse assembly, synapse organization, and homophilic cell adhesion [11]. GWAS trait enrichment further revealed that genes associated with chronotype and sleep disturbances might be differentially methylated in IS participants. Chronotype, also known as diurnal preference, refers to activity-rest preference in a given 24-h period [46]. Of note, several sleep disturbances have previously been identified as a risk factor for stroke onset [47], while sleep duration has been linked with subclinical atherosclerosis [48]. Disruption of circadian rhythms in shift workers and in experimental models in mice and sheep increased reperfusion injury after myocardial infarction providing experimental evidence of the causal link between poor sleep patterns and worse prognosis after ischemic events [49]. Finally, the observation that several risk factors for stroke (i.e., obesity, body mass index, blood pressure, and smoking status, as summarized in Additional file 1: Table S6) were enriched in gene pathway analysis suggests that DNA methylation might mediate the deleterious effects of such risk factors, thus increasing IS risk.

Several limitations of our study merit discussion. First, DNA methylation was measured after the onset of stroke. Due to the dynamic nature of epigenetic mechanisms, identified differentially methylated genes could be involved in either stroke pathogenesis, ischemic injury, or neuroprotection following IS. Further prospective longitudinal studies are warranted to test causality of observed DNA methylation changes. However, findings from our CRISPR/dCas9-Dnmt3a cell model suggest that the observed DNA methylation changes have antagonistic effects suggesting that we identified a mixture of stroke inducing (e.g., TRIM6 gene) as well as neuroprotective (e.g., TLN2 and FLRT2 genes) epigenetic biomarkers. Second, given the tissue and cell type specificity of DNA methylation, blood DNA methylation may not reflect the changes in human brain. Nonetheless, previous research has shown that blood DNA methylation may partially reflect changes in the targeted tissue and thus can serve as proxy non-invasive markers [50]. Third, participants were not stratified by stroke etiology (i.e., large artery, small vessel, cardioembolic strokes). Fourth, it is well established that nutrition could not only affect DNA methylation but also contribute to the risk of stroke. Unfortunately, data on diet were not available in the cohorts involved in our study. We cannot prevent the confounding effects caused by nutrition, as well as other unmeasured confounding, on the association between DNA methylation and stroke in an observational study. Fifth, the sample size of the discovery cohort is relatively smaller than prior EWAS studies. However, we conducted an independent replication study with a larger sample size to exclude the probable false-positive results. All DMPs identified in the discovery cohort were successfully replicated, indicating that the small sample size of the discovery cohort may not affect our findings a lot. Finally, although we included an external validation, only one DMP was nominally associated with IS (same direction) in the European population. The lack of replication could be attributed to differences in genetic makeup, clinical diagnosis, environmental exposures, and recruitment procedures between the study populations. For example, blood collection following stroke was performed within 12 h in the European cohort and within 48 h in the Chinese cohort. Given the known epidemiological differences in stroke incidence among different countries, with stroke being the leading cause of mortality in China, it is possible that some of the reported genes herein are specific to the Chinese population. The robustness of our multi-stage study design, including initial discovery, technical verification by TBS, internal and external validation in human populations, as well as functional validation in vascular endothelial cell models supports the strength of the reported findings.

In summary, we identified novel genes associated with IS in Chinese population. These findings enhance the understanding of DNA methylation involvement in both stroke pathogenesis and response to ischemic injury and highlight the importance of functional validation in epigenome-wide association studies in future research.

Methods

Participants

The study design and selection of participants are illustrated in Fig. 1. In the discovery stage, 40 IS cases and 40 controls were randomly selected from the China Antihypertensive Trial in Acute Ischemic Stroke (CATIS, 2009–2016) and the Prevention of Metabolic syndrome and Multi-metabolic disorders Study (PMMS, 2000–2004), to identify potential epigenetic markers of IS. From the remaining CATIS and PMMS participants, 1000 cases and 1000 age- and sex-matched controls were randomly selected as an internal replication sample to further confirm epigenetic markers of IS; after QC, 853 cases and 918 controls remained for downstream analysis. Additionally, an external cohort of 390 Spanish participants was used for further validation of the DMPs identified. To test the causal effect on gene expression of DNA methylation at loci identified in the aforementioned stages, human umbilical vein endothelial cell (HUVEC) models were epigenetically modified via CRISPR/dCas9-Dnmt3a.

Chinese cohorts (discovery and internal validation) included adults from the China Antihypertensive Trial in Acute Ischemic Stroke (CATIS, 2009–2016) and the Prevention of Metabolic syndrome and Multi-metabolic disorders Study (PMMS, 2000–2004). Detailed study design and methods for both cohorts have been described elsewhere [14, 15]. Briefly, CATIS is a multicenter randomized clinical trial (NCT01840072) designed to test whether moderate lowering of blood pressure in the acute phase after the onset of IS can reduce mortality and major disability at 14 days or hospital discharge. A total of 4,071 patients who had first-ever IS within 48 h of symptom onset and had an elevated systolic blood pressure between 140 mmHg and less than 220 mmHg were recruited to the CATIS. PMMS is a community-based cohort study including 5,888 Chinese adults free of overt cardiovascular disease (CVD) at the time of enrollment. All participants in the CATIS and PMMS reside in the Changshu city, Jiangsu Province, China.

The European cohort (external replication) included Spanish individuals of European ancestry from the BasicMar Register and the Girona Heart Registry study (REGICOR). Detailed information for these cohorts has been described elsewhere [16, 17]. Briefly, BasicMar is a hospital-based registry of stroke patients admitted to the Hospital del Mar in Barcelona, Spain [16]. REGICOR is a population-based registry recruiting participants residing in Girona, Spain [17]. REGICOR participants with no previous history of IS and no previous history of acute myocardial infarction were included as controls in the current analysis.

Study approval

The study protocols of CATIS were approved by the Institutional Review Boards at Soochow University in China and Tulane University in the USA. The study protocols of PMMS were approved by the Soochow University Ethics Committee. Written informed consent was obtained from all study participants. The study protocols of BasicMar and REGICOR were approved by the ethics committees of all participating hospitals, and all participants provided written informed consent to participate.

Clinical data and blood sample collection

For participants in the Chinese cohorts, demographic and lifestyle information and disease history were collected using standard questionnaires, as previously described [18]. For CATIS participants, blood samples were collected within 48 h of stroke onset and then stored at -80 °C for biochemical and DNA methylation analysis. Fasting plasma glucose and blood lipids including total cholesterol, triglycerides, high-density lipoprotein cholesterol (HDL-C), and low-density lipoprotein cholesterol (LDL-C) were measured by standard laboratory methods [19]. IS was ascertained by computed tomography (CT) or magnetic resonance imaging (MRI) of the brain.

For participants in the BasicMar and REGICOR cohorts, information for disease history including vascular risk factors was obtained from personal interviews to the patients, relatives, or caregivers, or abstracted from medical records. IS diagnosis was confirmed by CT or MRI. Blood samples were collected at hospital arrival, in the acute phase of the stroke (within 12 h of symptoms onset).

Epigenome-wide association study and confirmation by targeted bisulfite sequencing

Discovery stage

Forty cases and 40 age- and sex-matched controls were randomly selected from the CATIS and PMMS, respectively, for epigenome-wide association analysis. Briefly, genomic DNA was extracted from whole peripheral blood using the EZgene™ Blood DNA Kit (Biomega, Inc.), followed by bisulfite treatment using the EZ DNA Methylation-Gold Kit (Zymo Research, Orange, CA, USA). DNA methylation was quantified by the Infinium HumanMethylationEPIC BeadChip array (Illumina, Inc., San Diego, CA). Raw DNA methylation data were processed using the R Bioconductor package “minfi” (v1.30.0) [20]. After quality control and exclusion of probes on the sex chromosomes, a total of 815,017 probes were used in the downstream analysis. Blood cell proportions (CD8+ T cells, CD4+ T cells, natural killer cells, B cells, monocytes, and granulocytes) were estimated using the Houseman’s estimation method [21].

Fine mapping and internal replication

In an independent sample comprising 1000 cases from the CATIS and 1000 age- and sex-matched controls from the PMMS, targeted bisulfite sequencing (TBS) [22] was performed to confirm and discover more differentially methylated probes (DMPs) associated with IS. Briefly, primers were designed to detect the maximum number of additional CpG loci in the vicinity of identified DMPs. TBS target gene sequences and primers can be found in Additional file 1: Table S7. Genomic DNA was bisulfite-treated using the EZ DNA Methylation-Gold Kit (Zymo Research, Inc., CA, USA) according to the manufacturer’s protocol. Samples were then amplified, barcoded, and sequenced by Illumina Hiseq 2000 (Illumina, Inc., CA, USA) using the paired-end sequencing protocol according to manufacturer’s guidelines. Methylation level at each CpG locus was calculated as the percentage of the methylated alleles over the sum of methylated and unmethylated alleles. For quality control, the samples with bisulfite conversion rate < 98% and the cytosine sites with average coverage less than 20 × were filtered out. After quality control, 853 cases and 918 controls remained for downstream analysis; all CpG sites passed quality control. Demographic and clinical characteristics of included (n = 1771) vs. excluded participants (n = 229) can be found in Additional file 1: Table S8.

External replication

DNA methylation in the BasicMar and REGICOR was assessed by Infinium MethylationEPIC Beadchip arrays (Illumina Netherlands, Eindhoven, Netherlands), as previously reported [9]. A total of 390 individuals, including 207 cases from BasicMar and 183 controls from REGICOR, were included in the external replication analysis.

Functional validation of DMPs in cellular models

To examine the potential functional role of the identified CpGs in vascular endothelial cell activation and inflammation, we employed the CRISPR/dCas9-Dnmt3a system to develop site-specific methylation models in a culture of human umbilical vein endothelial cells (HUVEC).

Construction of CRISPR/dCas9-Dnmt3a plasmids

The dCas9-5xGCN4 fusion protein expression vector (ZP509, pEFS-dCas9-5xGCN4-T2A-EGFP) was generated by fusing cDNA encoding the catalytically inactive nuclease codon-optimized S. pyogenes Cas9 (dCas9) to quintuple tandem short peptide GCN4 in an expression vector using the EFS promoter. The 5xGCN4 fragment was synthesized by Genewiz and cloned into pEFS-dCas9-EGFP (Pregen plasmid: ZP508) with BamHI sites by Gibson Assembly to package lentiviruses (Pregen plasmid: ZP509). The aa sequence of GCN4 used was EELLSKNYHLENEVARLKK. The sequences of the linker between each GCN4 peptide unit were GSGSGGSGSGGSGSGSGGSGSGGSGSG. All the fusion proteins were expressed under the control of the EFS promoter. The DNMT3L CD-DNMT3ACD fragment was amplified from Addgene plasmid: 112,210. The scFv GCN4 fragment was amplified from Addgene plasmid 60,904. The two PCR fragments were cloned into pEFS-mCherry (Pregen plasmid: ZP510) with AgeI and RsrII sites by Gibson Assembly to package lentiviruses (Pregen plasmid: ZP512). Once expressed, the scFv GCN4 minimal antibody fused to DNMT3A binds to the GCN4 tandem repeats fused to dCas9, thus allowing colocalization of the inactive nuclease dCas9 with the DNA methyltransferase in transfected cells.

Construction of gRNAs

The single-guide RNA (sgRNA) expression vectors for TRIM6, SOX17, TLN2, SOX1, AGBL4, FLRT2, FAM84A, and negative control (NC) were cloned by inserting the target sequences into ZP512 plasmid. Cloning was performed by linearization of an PacI site and Gibson Assembly-mediated incorporation of the gRNA insert fragment. All constructs were sequenced before transfection. The sequences for primers and sgRNA used for CRISPR/dCas9-Dnmt3a can be found in Additional file 1: Table S9.

Production of lentiviral particles

HEK293T immortalized cells were incubated in DMEM with 10% FBS and 1% P/S. All cells were grown at 5% CO2 and 37 °C incubation. The HEK293T cells were transfected with either the pEFS-dCas9-5xGCN4-EGFP plasmid or the pU6-sgRNA-EFS-scFvGCN4-DNMT3L-DNMT3a-mCherry plasmid with different sgRNAs sites using polyethyIenimine (PEI polysciences #23966). Lentiviral particles were collected at 48 h and 72 h after transfection. Viral particles were then concentrated by centrifugation at 25,000 g for 2 h at 4 °C and resuspended in PBS medium.

Lentivirus transfection and flow sorting

HUVEC cells (LONZA #CC-2519) were grown in EBM™ Basal medium (LONZA, #CC-3121) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (P/S). HUVEC cells were co-infected with various lentiviral mixtures, transfected with both plasmids with different gRNAs sites for 72 h. EGFP and mCherry double-positive cells were obtained by flow sorting and were cultured and expanded for 5 generations. After expansion, the HUVEC cells were used in downstream functional analysis.

Pyrosequencing

We then verified the presence of site-specific methylation at our region of interest by pyrosequencing to check the validity of the cellular model. Briefly, genomic DNA of the isolated cells was extracted using a DNA Mini Kit (QIAGEN 51304) and then bisulfite converted using an EpiTect Bisulfite Kit (Qiagen 59,104) according to the manufacturer’s instructions. Bisulfite PCR primers and their target gene sequences are listed in Additional file 1: Table S10. Twenty-five microliter 2xTaq DNA polymerase premix, 1 μl for each forward and reverse primer, 21 μl double-distilled water (ddH2O), and 2 μl bisulfite-treated DNA were mixed and run according to the following PCR program. First, samples were heat activated at 95 °C for 3 min; for the second step, they were kept at 94 °C for 30 s, then at 56 °C for 30 s, and then at 72 °C for 1 min. Then the process was repeated from the second step for 40 cycles and finally extended at 72 °C for 7 min. Bisulfite PCR products were immobilized on streptavidin beads, and single-stranded DNA was prepared, sequenced, and finally analyzed on the PyroMark Q96 ID system (Qiagen). For quality control, each experiment included non-CpG cytosines as an internal control to verify completion of sodium bisulfite DNA conversion.

Quantitative real-time PCR (qRT-PCR)

Finally, we quantified the expression levels of candidate genes harboring the putative DMPs by quantitative real-time polymerase chain reaction (qRT-PCR). Additionally, the expression levels of genes encoding endothelial cell adhesion markers, including ICAM1 (intercellular adhesion molecule 1), VCAM1 (vascular cell adhesion molecule 1), and SELE (selectin E), were also measured by qRT-PCR in the human umbilical vein endothelial cells (HUVECs). The Actin gene of human species was used as a control. Specific primers were designed based on cDNA sequences. Primer sequences for qRT-PCR experiments can be found in Additional file 1: Table S11. The qRT-PCR was performed on a Bio-Rad S1000 with Bestar SYBR Green RT-PCR Master Mix (TOYOBO). PCR conditions consisted of denaturing at 95 °C for 1 min and 40 cycles of denaturing at 95 °C for 15 s followed by annealing and extension at 60 °C for 30 s. Relative gene expression was calculated using the Livak and Schmittgen 2 − ΔΔCt method [23], normalized with the reference gene Actin. PCR amplifications were performed in triplicate for each sample.

Statistical analysis

Single-probe analysis

To identify DMPs associated with IS, we constructed multivariate linear regression models via the R package limma [24]. In the model, DNA methylation level at each CpG was the dependent variable and status of IS was the independent variable. The model adjusted for age, sex, BMI, smoking, alcohol consumption, hypertension, LDL-C, HDL-C, fasting plasma glucose, estimated cell counts, and surrogate variables derived from surrogate variable analysis (SVA) [25]. Multiple testing was controlled by false discovery rate (FDR), and FDR-adjusted P-value (i.e., q-value) < 0.05 was considered statistically significant.

Region-based analysis

Differentially methylated regions (DMRs) were identified using the program DMRcate [26]. The program identifies genomic regions harboring putative disease-associated CpGs and takes into consideration the correlations between adjacent probes. Here we defined a DMR as a region containing at least 2 correlated probes (adjacent peak probes [P < 0.05] spaced less than 1,000 bp). A DMR with P-value < 5.0 × 10–7 was considered statistically significant.

Enrichment and pathway analysis

To functionally annotate the identified differentially methylated genes, we conducted functional enrichment analysis using the computational program FUMA [27]. We defined differentially methylated genes as unique genes containing either DMPs or DMRs identified above. P-value < 0.05 was used to determine statistical significance for the annotated biological pathways.

Availability of data and materials

The datasets used during the current study are available from the corresponding author on reasonable request.

References

  1. GBD 2016 Stroke Collaborators. Global, regional, and national burden of stroke, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol. 2019;18(5):439–458.

  2. Li Z, Jiang Y, Li H, et al. China’s response to the rising stroke burden. BMJ. 2019;364: l879.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Tsao CW, Aday AW, Almarzooq ZI, et al. Heart disease and stroke statistics-2022 update: a report from the American Heart Association. Circulation. 2022;145(8):e153–639.

    Article  PubMed  Google Scholar 

  4. O’Donnell MJ, Chin SL, Rangarajan S, et al. Global and regional effects of potentially modifiable risk factors associated with acute stroke in 32 countries (INTERSTROKE): a case-control study. Lancet. 2016;388(10046):761–75.

    Article  PubMed  Google Scholar 

  5. Bevan S, Traylor M, Adib-Samii P, et al. Genetic heritability of ischemic stroke and the contribution of previously reported candidate gene and genomewide associations. Stroke. 2012;43(12):3161–7.

    Article  CAS  PubMed  Google Scholar 

  6. Malik R, Chauhan G, Traylor M, et al. Multiancestry genome-wide association study of 520,000 subjects identifies 32 loci associated with stroke and stroke subtypes. Nat Genet. 2018;50(4):524–37.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Deng G-X, Xu N, Huang Q, et al. Association between promoter DNA methylation and gene expression in the pathogenesis of ischemic stroke. Aging (Albany, NY). 2019;11(18):7663–77.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Shen Y, Peng C, Bai Q, et al. Epigenome-wide association study indicates hypomethylation of MTRNR2L8 in large-artery atherosclerosis stroke. Stroke. 2019;50(6):1330–8.

    Article  CAS  PubMed  Google Scholar 

  9. Soriano-Tárraga C, Lazcano U, Giralt-Steinhauer E, et al. Identification of 20 novel loci associated with ischaemic stroke. Epigenome-wide association study. Epigenetics. 2020;15(9):988–97.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Davis Armstrong NM, Chen W-M, Hsu F-C, et al. DNA methylation analyses identify an intronic ZDHHC6 locus associated with time to recurrent stroke in the Vitamin Intervention for Stroke Prevention (VISP) clinical trial. PLoS ONE. 2021;16(7): e0254562.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Sun H, Xu J, Hu B, et al. Association of DNA methylation patterns in 7 novel genes with ischemic stroke in the Northern Chinese Population. Front Genet. 2022;13

  12. Cullell N, Soriano-Tárraga C, Gallego-Fábrega C, et al. DNA methylation and ischemic stroke risk: an epigenome-wide association study. Thromb Haemost 2022;(EFirst)

  13. Zeng M, Zhen J, Zheng X, et al. The role of DNA methylation in ischemic stroke: a systematic review. Front Neurol. 2020;11: 566124.

    Article  PubMed  PubMed Central  Google Scholar 

  14. He J, Zhang Y, Xu T, et al. Effects of immediate blood pressure reduction on death and major disability in patients with acute ischemic stroke: the CATIS randomized clinical trial. JAMA. 2014;311(5):479–89.

    Article  CAS  PubMed  Google Scholar 

  15. Zhang L, Guo Z, Wu M, et al. Interaction of smoking and metabolic syndrome on cardiovascular risk in a Chinese cohort. Int J Cardiol. 2013;167(1):250–3.

    Article  PubMed  Google Scholar 

  16. Roquer J, Rodríguez-Campello A, Jiménez-Conde J, et al. Sex-related differences in primary intracerebral hemorrhage. Neurology. 2016;87(3):257LP – 262.

    Article  Google Scholar 

  17. Grau M, Subirana I, Elosua R, et al. Trends in cardiovascular risk factor prevalence (1995–2000–2005) in northeastern Spain. Eur. J. Cardiovasc. Prev. Rehabil. Off. J. Eur. Soc. Cardiol. Work. Groups Epidemiol. Prev. Card. Rehabil. Exerc. Physiol. 2007;14(5):653–659.

  18. Peng H, Fan Y, Li J, et al. DNA methylation of the natriuretic peptide system genes and ischemic stroke. Neurol Genet. 2022;8(3): e679.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Peng H, Zhang Q, Cai X, et al. Association between high serum soluble corin and hypertension: a cross-sectional study in a general population of China. Am J Hypertens. 2015;28(9):1141–9.

    Article  CAS  PubMed  Google Scholar 

  20. Aryee MJ, Jaffe AE, Corrada-Bravo H, et al. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics. 2014;30(10):1363–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Houseman E, Accomando WP, Koestler DC, et al. DNA methylation arrays as surrogate measures of cell mixture distribution. BMC Bioinform. 2012;13(1):86.

    Article  Google Scholar 

  22. Pu W, Wang C, Chen S, et al. Targeted bisulfite sequencing identified a panel of DNA methylation-based biomarkers for esophageal squamous cell carcinoma (ESCC). Clin Epigenetics. 2017;9:129.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-delta delta C(T)) method. Methods. 2001;25(4):402–8.

    Article  CAS  PubMed  Google Scholar 

  24. Ritchie ME, Phipson B, Wu D, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7): e47.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Leek JT, Johnson WE, Parker HS, et al. The sva package for removing batch effects and other unwanted variation in high-throughput experiments. Bioinformatics. 2012;28(6):882–3.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Peters TJ, Buckley MJ, Statham AL, et al. De novo identification of differentially methylated regions in the human genome. Epigenetics Chromatin. 2015;8:6.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Watanabe K, Taskesen E, van Bochoven A, Posthuma D. Functional mapping and annotation of genetic associations with FUMA. Nat Commun. 2017;8(1):1826.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Yang L, Froio RM, Sciuto TE, et al. ICAM-1 regulates neutrophil adhesion and transcellular migration of TNF-alpha-activated vascular endothelium under flow. Blood. 2005;106(2):584–92.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Galkina E, Ley K. Vascular adhesion molecules in atherosclerosis. Arterioscler Thromb Vasc Biol. 2007;27(11):2292–301.

    Article  CAS  PubMed  Google Scholar 

  30. Gross MD, Bielinski SJ, Suarez-Lopez JR, et al. Circulating soluble intercellular adhesion molecule 1 and subclinical atherosclerosis: the Coronary Artery Risk Development in Young Adults Study. Clin Chem. 2012;58(2):411–20.

    Article  CAS  PubMed  Google Scholar 

  31. Banerjee C, Chimowitz MI. Stroke caused by atherosclerosis of the major intracranial arteries. Circ Res. 2017;120(3):502–13.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  32. Courties G, Moskowitz MA, Nahrendorf M. The innate immune system after ischemic injury: lessons to be learned from the heart and brain. JAMA Neurol. 2014;71(2):233–6.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Ozato K, Shin D-M, Chang T-H, Morse HC 3rd. TRIM family proteins and their emerging roles in innate immunity. Nat Rev Immunol. 2008;8(11):849–60.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Stremlau M, Owens CM, Perron MJ, et al. The cytoplasmic body component TRIM5α restricts HIV-1 infection in Old World monkeys. Nature. 2004;427(6977):848–53.

    Article  CAS  PubMed  Google Scholar 

  35. Bahouth MN, Venkatesan A. Acute viral illnesses and ischemic stroke: pathophysiological considerations in the era of the COVID-19 pandemic. Stroke. 2021;52(5):1885–94.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Smeeth L, Thomas SL, Hall AJ, et al. Risk of myocardial infarction and stroke after acute infection or vaccination. N Engl J Med. 2004;351(25):2611–8.

    Article  CAS  PubMed  Google Scholar 

  37. van der Harst P, Verweij N. Identification of 64 novel genetic loci provides an expanded view on the genetic architecture of coronary artery disease. Circ Res. 2018;122(3):433–43.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Zeng G, Lian C, Yang P, et al. E3-ubiquitin ligase TRIM6 aggravates myocardial ischemia/reperfusion injury via promoting STAT1-dependent cardiomyocyte apoptosis. Aging (Albany, NY). 2019;11(11):3536–50.

    Article  CAS  PubMed  Google Scholar 

  39. Mishra A, Malik R, Hachiya T, et al. Stroke genetics informs drug discovery and risk prediction across ancestries. Nature. 2022;611(7934):115–23.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Rangarajan ES, Primi MC, Colgan LA, et al. A distinct talin2 structure directs isoform specificity in cell adhesion. J Biol Chem. 2020;295(37):12885–99.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Castaneda AB, Petty LE, Scholz M, et al. Associations of carotid intima media thickness with gene expression in whole blood and genetically predicted gene expression across 48 tissues. Hum Mol Genet. 2022;31(7):1171–82.

    Article  CAS  PubMed  Google Scholar 

  42. von Essen M, Rahikainen R, Oksala N, et al. Talin and vinculin are downregulated in atherosclerotic plaque; Tampere Vascular Study. Atherosclerosis. 2016;255:43–53.

    Article  Google Scholar 

  43. Lacy SE, Bönnemann CG, Buzney EA, Kunkel LM. Identification of FLRT1, FLRT2, and FLRT3: a novel family of transmembrane leucine-rich repeat proteins. Genomics. 1999;62(3):417–26.

    Article  CAS  PubMed  Google Scholar 

  44. Carty CL, Keene KL, Cheng Y-C, et al. Meta-analysis of genome-wide association studies identifies genetic risk factors for stroke in African Americans. Stroke. 2015;46(8):2063–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Libby P, Buring JE, Badimon L, et al. Atherosclerosis. Nat Rev Dis Prim. 2019;5(1):56.

    Article  PubMed  Google Scholar 

  46. Adan A, Archer SN, Hidalgo MP, et al. Circadian typology: a comprehensive review. Chronobiol Int. 2012;29(9):1153–75.

    Article  PubMed  Google Scholar 

  47. Fan M, Sun D, Zhou T, et al. Sleep patterns, genetic susceptibility, and incident cardiovascular disease: a prospective study of 385 292 UK biobank participants. Eur Heart J. 2020;41(11):1182–9.

    Article  PubMed  Google Scholar 

  48. Domínguez F, Fuster V, Fernández-Alvira JM, et al. Association of sleep duration and quality with subclinical atherosclerosis. J Am Coll Cardiol. 2019;73(2):134–44.

    Article  PubMed  Google Scholar 

  49. Zhao Y, Lu X, Wan F, et al. Disruption of circadian rhythms by shift work exacerbates reperfusion injury in myocardial infarction. J Am Coll Cardiol. 2022;79(21):2097–115.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Braun PR, Han S, Hing B, et al. Genome-wide DNA methylation comparison between live human brain and peripheral tissues within individuals. Transl Psychiatry. 2019;9(1):47.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We gratefully acknowledge the cooperation and participation of the members of the current study. We especially thank the clinical staff at all participating hospitals for their support and contribution to this project. Without their contribution, this research would not have been possible. Targeted bisulfite sequencing was performed by Genesky Biotechnologies.

Funding

This study was supported by the National Natural Science Foundation of China (Nos. 82173596, 81903384, and 81872690) and a Project of the Priority Academic Program Development of Jiangsu Higher Education Institutions. Drs. Zhao and Palma-Gudiel were supported by the National Institute of Health Grants R01DK107532, RF1AG052476, and R01AG064786.

Author information

Author notes

  1. Hao Peng and Helena Palma-Gudiel should be noted as co-first authors.

Authors and Affiliations

  1. Department of Epidemiology, School of Public Health and Jiangsu Key Laboratory of Preventive and Translational Medicine for Geriatric Diseases, Medical College of Soochow University, 199 Renai Road, Suzhou, 215123, China

    Hao Peng, Mingzhi Zhang & Yonghong Zhang

  2. Department of Epidemiology, College of Public Health and Health Professions and College of Medicine, University of Florida, 2004 Mowry Road, Gainesville, FL, 32610, USA

    Helena Palma-Gudiel & Jinying Zhao

  3. Neurovascular Research Group, Department of Neurology of Hospital del Mar-IMIM (Institut Hospital del Mar d’Investigacions Mèdiques), Universitat Autònoma de Barcelona/DCEXS, Universitat Pompeu Fabra, Barcelona, Spain

    Carolina Soriano-Tarraga & Jordi Jimenez-Conde

  4. Department of Psychiatry, Washington University School of Medicine, St. Louis, MO, USA

    Carolina Soriano-Tarraga

  5. Hope Center for Neurological Disorders, Washington University School of Medicine, St. Louis, MO, USA

    Carolina Soriano-Tarraga

  6. NeuroGenomics and Informatics, Department of Psychiatry, Washington University in St. Louis, St. Louis, USA

    Carolina Soriano-Tarraga

Authors
  1. Hao Peng
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Helena Palma-Gudiel
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Carolina Soriano-Tarraga
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Jordi Jimenez-Conde
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Mingzhi Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Yonghong Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Jinying Zhao
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

HP, YZ, and JZ conceptualized and designed the study. HP, HPG, and CST conducted the analyses. HP, MZ, CST, and JJC acquired the data. JZ, HP, and HPG drafted the manuscript. All authors critically reviewed the article and contributed to data interpretation. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Yonghong Zhang or Jinying Zhao.

Ethics declarations

Ethics approval and consent to participate

The study protocols of CATIS were approved by the Institutional Review Boards at Soochow University in China and Tulane University in the USA. The study protocols of PMMS were approved by the Soochow University Ethics Committee. Written informed consent was obtained from all study participants. The study protocols of BasicMar and REGICOR were approved by the ethics committees of all participating hospitals, and all participants provided written informed consent to participate.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1

. The supplemental tables showing the additional results.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Peng, H., Palma-Gudiel, H., Soriano-Tarraga, C. et al. Epigenome-wide association study identifies novel genes associated with ischemic stroke. Clin Epigenet 15, 106 (2023). https://doi.org/10.1186/s13148-023-01520-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s13148-023-01520-x

Keywords

Clinical Epigenetics

ISSN: 1868-7083

Contact us