Transcriptome-wide identification of N6-methyladenosine modifications for aortic dissection

doi:10.21203/rs.3.rs-3972169/v1

Background: N6-methyladenosine (m⁶A) plays important roles in many biological processes such as gene expression control and may have functional roles in aortic dissection (AD). The aim of this study was to identify N6-methyladenosine (m⁶A) modification and the expressions of the m⁶A regulatory genes related to AD.

Methods: Aortic tissue samples were obtained from AD and controls and MeRIP-seq and RNA-seq experiments were performed to detect m⁶A methylation and mRNA expression profiles, respectively. The differentially RNA methylation peaks were validated by MeRIP-PCR in AD cases and controls.

Results: Compared with the control samples, 3,318 up methylated and 1,573 down methylated coding genes in AD were detected. These genes were mainly enriched in focal adhesion, ECM-receptor interaction and regulating the transcription such as splicing. Significant differentially methylated m⁶A sites in some well-known susceptibility genes for AD were identified, including FBN1, TGFB1, TGFBR1/2, LOXL3, COL3A1, SMAD3, VEGFA and MAPK1/3. A total of 651 differentially expressed genes, including 594 protein-coding genes (96 upregulated and 498 downregulated), and 57 lncRNAs (20 upregulated and37 downregulated) were identified. Integrated analysis of the data from MeRIP-seq and RNA-Seq identified 74 genes that changed significantly in both m⁶A level and mRNA abundance in AD cases compared with the controls. We observed the same m⁶A-level changes in 14 out of the 16 selected m⁶A methylated transcripts in the independent sample.

Conclusions: This study identified m⁶A changes in critical AD susceptibility genes. The identified m⁶A modification may play a role in critical AD-related pathways, thereby regulating the pathogenesis of AD.

Health sciences/Cardiology/Cardiovascular biology/Cardiovascular diseases/Vascular diseases/Aortic diseases

Health sciences/Biomarkers/Predictive markers

Aortic dissection

N6-methyladenosine modification

Focal adhesion

Extracellular matrix

Aortic dissections (ADs) are extremely dangerous cardiovascular diseases that affecting the aorta. AD is a life-threatening disease begins with a tear in the aortic intima or bleeding within the aortic wall, which resulting a false lumen. AD is associated with a high mortality rate, with about 2.78 per 100 000 people each year^{1, 2}. The disease is most commonly seen in people aged 65–75 years old, and is more common in men than women².

AD can be inherited and genetic factors play roles in its development. Family studies have demonstrated the familial aggregation of AD, and genetic association studies have uncovered many susceptibility genes^3–6. The pathogenic variants were found in FBN1, PHACTR1, EDN1⁷, COL3A1⁸, SMAD3^{8, 9}, TGFBR1/2, TGFB2/3¹⁰, TSR1^{11, 12}, TLN1¹³ and others^{5, 13, 14}. Recent studies have demonstrated that epigenetic factors (e.g., DNA methylation, histone modification and non-coding RNA) were key regulators in AD^{15, 16}. Although many susceptibility genes for AD have been identified and the role of epigenetic regulation in AD has been studied, it is still not comprehensively and deeply analyzed. Therefore, how epigenetic factors regulated these genes are in AD has not been known.

RNA modifications are modifiable chemical modifications that are involved in diverse biological regulations¹⁷. In recent years, various types (more than 170) of chemical modifications present in RNA molecules have been identified. N6-methyladenosine (m⁶A) is the most prevalent modification in eukaryotic RNAs. This type of modification can occur within almost all types of RNA molecules such as mRNA, miRNA, circRNA and lncRNA and has been confirmed to be very important in the post-transcriptional regulation of gene expression, RNA stability and homeostasis^18–21. It also has been showed to be critical in the etiology of various diseases²². Recently, m⁶A has been shown to regulate cardiac gene expression and cellular growth and therefore was associated with cardiac function^23–25. It has also been shown to be associated with blood pressure²⁶, the most important risk factor of AD, and coronary artery disease^27–29. This evidence indicated an etiological role of m⁶A in AD. Although some differential m⁶A methylation peaks have been found in AD³⁰, the roles have not been studied still in a deepgoing way by now.

To comprehensively and deeply investigate the transcriptome-wide m⁶A methylome of AD, this study performed the m⁶A specific RNA immunoprecipitation sequencing (MeRIP-seq) and RNA-seq in 9 AD and 9 normal aorta samples. The differential methylation peaks were validated in 20 independent samples by MeRIP-qPCR.

Transcriptome-wide m⁶A methylation

The mean total MeRIP-seq reads for the 9 AD cases and 9 controls was 12,819,119,767 bp. The cut-adapt software remove about 8% of the reads. The mean mapped rate was 80%. The analysis demonstrated that there were 33,792 nonoverlapping m⁶A peaks within 11,817 mRNA transcripts in controls and 36,517 nonoverlapping m6A peaks within 11,458 mRNA transcripts in AD cases. There were 1,259 nonoverlapping m⁶A peaks within 763 lncRNA transcripts in AD group, 1195 nonoverlapping m6A peaks within 758 lncRNA transcripts in control group.

Nearly 20% of the m6A-methylated mRNAs in the two groups contained one or two m⁶A peaks, about 18% and 12% contained three or four m⁶A peaks, respectively, and about 20% contained ≥ 5 m⁶A peaks (Figure 1A). For lncRNAs, more than 60% of the m⁶A-methylated lncRNAs had only one m⁶A modification peak and about 18% contained two m⁶A peaks (Figure 1B). All m⁶A peaks were classified based on their locations in mRNA transcripts: the 5'-UTR, start codon, stop codon, coding DNA sequence (CDS) and 3'-UTR. As the pie chart displayed, m⁶A peaks were most often harbored in the CDS and UTR region in both case and control groups (Figure 1C and 1D, respectively). As shown in Metagene plot, the identified m⁶A peaks were mainly enriched in the CDS, especially in the immediate vicinity of the stop codon (Figure 1E). The m⁶A peaks in lncRNAs were located in the exons (Figure 1F).

Differentially methylated m⁶A sites

In total, 4,988 differentially methylated m⁶A sites (DMMSs) within 3,768 mRNA transcripts were identified (FC > 2, FDR < 0.05) (Figure 2A, Online Table 1). Compared with control group, 3,318 upmethylated m⁶A sites within 2,642 mRNA transcripts and 1,573 downmethylated m⁶A sites within 1,355 mRNA transcripts were found in AD group (Table 2). We also identified 97 DMMSs within 79 lncRNA transcripts between the two groups, among which there were 63 upmethylated and 34 downmethylated m⁶A sites (Online Figure 1, Online Table 2). Tables 3 showed the up (5) and down (11) methylated m⁶A sites within the mRNAs that have >100-fold change between AD cases and controls. The DMMSs were mapped to chromosomes, and chromosome 1 (517), 2 (349) and 3 (308) harbored most DMMSs in mRNA (Online Figure 2A), and chromosome 15 (10), 6 (8) and 19 (8) harbored most DMMSs in lncRNA (Online Figure 2B). We further analyzed the m⁶A distribution patterns of mRNAs and found that most of the DMMSs were located in the CDS and UTR region.

Significant DMMSs in some well-known susceptibility genes for AD were identified (Table 4), including FBN1, TGFB1, TGFBR1/2, LOXL3, COL3A1, SMAD3, VEGFA and MAPK1/3. Three significant DMMSs, including chr15:48474328-48483880, chr15:48499022-48505066 and chr15:48537655-48537806 in FBN1 were found. These three sites were all upmethylated (FC = 2.81, 5.13 and 2.51, respectively). Two significant upmethylated DMMSs in COL3A1, two significant downmethylated DMMSs in TGFBR1 and one significant DMMS in other genes were found.

Differentially methylated mRNAs are involved in specific biological functions and pathways

To uncover the functions of m⁶A in AD, the protein-coding genes containing DMMSs were selected for GO enrichment analysis and KEGG pathway analysis. The analysis showed that genes with up or down methylated m⁶A sites significant (FDR < 0.05) enriched in 1,434 BP (Online Table 3), 292 CC (Online Table 4) and 201 MF (Online Table 5) GO terms. For the CC and MF categories, both up or down methylation of m⁶A sites showed notable enrichment in the focal adhesion. For the BP category, genes with up methylation of m⁶A sites were enriched in splicing (Online Figure 3A), while genes with down methylation of m⁶A sites showed high enrichment in extracellular matrix organization (Online Figure 3B). These results suggested that m⁶A might play an important role in the occurrence and development of AD.

KEGG pathway analysis of DMMS-contained mRNA associated genes was performed and 47 pathways were identified (Online Table 6). The top pathway for both genes with up or down methylation of m⁶A sites was focal adhesion. In addition, genes with up methylation of m⁶A sites were enriched in regulating the transcription (Figure 2B), while genes with down methylation of m⁶A sites were mainly enriched in ECM-receptor interaction, TGF-beta signaling pathway and cardiomyopathy (Figure 2C). These pathways highlighted some genes related to AD, including MAP4K4, DSP, GADD45B, PYGL, COL1A1, VEGFA, TNNC1, STMN1, MYBPC3, TPM1, ACTC1, TNXB, F2R and RYR2.

The expression of m⁶A methylated genes

RNA-seq dataset was used to discover the differentially expressed lncRNAs/mRNAs between the two groups. A total of 651 DEGs were identified (FC > 2, FDR < 0.05), including 594 protein-coding genes (96 upregulated and 498 downregulated) (Figure 3A, Online Table 7) and 57 lncRNAs (20 upregulated and 37 downregulated) (Figure 3B, Online Table 8). Some of the known AD susceptibility genes such as VEGFA (FC = 8.44, FDR = 2.42×10^-6) and IL6 (FC = 15.22, FDR = 1.26×10^-6) were differentially expressed between Ads and controls (Figure 3C). The lncRNAs H19 (FC = 31.20, FDR = 3.57×10^-5) and NEAT1 (FC = 2.15, FDR = 0.04) were differentially expressed between cases and controls (Figure 3C). DEGs such as COL1A1, VEGFA, TNNC1, SLC2A1, COX4I1, MYBPC3, TPM1, CREB3L1, ACTC1, TNXB and RYR2 were enriched in specific pathways such as focal adhesion, ECM-receptor interaction, insulin secretion, cardiac muscle contraction, dilated cardiomyopathy, oxidative phosphorylation, adrenergic signaling in cardiomyocytes and calcium signaling pathway.

Integration of DMMSs and DEGs

Integrated analysis of the data from MeRIP-seq and RNA-Seq identified 74 genes that changed significantly (FC > 2, FDR < 0.05) in both m⁶A level and mRNA abundance in AD cases compared with the controls (Online Table 9). These genes were classified into four groups, including 10 hypermethylated and upregulated genes (hyper-up), 21 hypermethylated but downregulated genes (hyper-down), 22 hypomethylated but upregulated genes (hypo-up) and 21 hypomethylated and downregulated genes (hypo-down) (Figure 4A). Based on the above data, we deduced that m⁶A modification tended to have a positive correlation with mRNA expression in AD. DSP (DMMSs FC = 25.81, FDR = 1.45×10^-6; DEG FC = 91.92, FDR = 2.29×10^-7), TNNC1 (DMMSs FC = 113.77, FDR = 1.00×10^-7; DEG FC = 9.83, FDR = 2.64×10^-3) and PCDH1 (DMMSs FC = 448.82, FDR = 2.95×10^-7; DEG FC = 4.68, FDR = 1.30×10^-2) were representative hypo-down genes in AD group. MAP4K4 (DMMSs FC = 28.44, FDR = 7.94×10^-15; DEG FC = 2.11, FDR = 2.13×10^-2) and VEGFA (DMMSs FC = 7.89, FDR = 1.45×10^-5; DEG FC = 8.44, FDR = 5.80×10^-4) and lncRNA NEAT1 (DMMSs FC = 2.77, FDR = 4.57×10^-10; DEG FC = 2.15, FDR = 4.12×10^-2) were one of hyper-up genes in AD group.

Furthermore, gene set enrichment analysis was performed to deduce the functional pathway for genes identified in the integrative analysis. These 74 genes were mainly enriched in cardiac muscle contraction, adrenergic signaling in cardiomyocytes, hypertrophic cardiomyopathy and dilated cardiomyopathy KEGG pathways (Figure 4B). The GO enrichment analysis of the 74 genes showed the most enriched terms of biological processes were mainly included in the cardiac muscle tissue development and morphogenesis (Figure 4C). MYBPC3, RYR2, ACTC1, CREB3L1, TNNC1, COX4I1, TPM1, UQCRC1, DSP, FHOD3 and VEGFA were representative genes involved in these pathways.

MeRIP-qPCR of differentially methylated transcripts

To further confirm the results of MeRIP-seq, we conducted MeRIP-qPCR assays for differentially methylated transcripts of representative DMGs, including 15 protein-coding genes (COL1A1, VEGFA, F2R, PKD1, PKD2, SLC2A1, RYR2, ACTC1, CREB3L1, TNNC1, COX4I1, TNXB, CP, PCOLCE2 and TM4SF1) and one lncRNA NEAT1 (Online Table 10). These 16 genes were identified by the above integration and enrichment analyses. We observed the same m⁶A-level changes in 13 out of the 16 m⁶A methylated transcripts (Table 5), including 13 (COL1A1, VEGFA, F2R, SLC2A1, RYR2, ACTC1, CREB3L1, TNNC1, COX4I1, PKD2, CP, PCOLCE2 and TM4SF1) protein-coding genes and the lncRNA NEAT1. The changes of six of these genes were nominally significant, including COL1A1, VEGFA, F2R, ACTC1, TNNC1 and PKD2 (Figure 5). After further immunohistochemical validation of paraffin sections of aortic tissue, the results showed a similar protein expression trend. ACTC1 and TNNC1 were found to be downregulated in AD patients, while COL1A1, VEGFA, F2R and PKD2 were found to be upregulated (Figure 6).

In this study, we describe the m⁶A RNA methylation landscape and reveal the potential functions of this methylation in the regulation of RNA metabolism in AD. We identified that abnormal methylation in FBN1, TGFB1, TGFBR1/2, LOXL3, COL3A1, SMAD3, VEGFA and MAPK1/3 might be involved in the pathophysiology of AD. Conjoint analysis of m⁶A-RIP-seq and RNA-seq data resulted in the identification of differentially expressed hyper-methylated or hypo-methylated mRNA m⁶A peaks. Differential m⁶A methylation in COL1A1, VEGFA, F2R, ACTC1, TNNC1 and PKD2 was validated in independent sample.

AD is currently a serious public health problem worldwide and patients with AD are associated with extremely high mortality rates. Convincing evidence has suggested that VSMCs play a key role in the development and progression of AD. As shown in previous study, mRNA and protein expression level of METTL14 was up-regulated, while the mRNA level of FTO was down-regulated in AD tissue samples compared with normal samples³⁰. The differences among the expressions of METTL14 and FTO were found in the other types of inflammatory cells that exist in the AD tissues³² and m⁶A mRNA modification can regulate T cell homeostasis³³. Our study identified many genes that were abnormally methylated in AD aortic tissues, including many well-known AD susceptibility genes such as FBN1, TGFB1, VEGFA, LOXL3, COL3A1 and SMAD3. Abnormal methylation TGFB2, TGFBR1, TGFBR1, TGFBR2, TGFB1I1, VEGFB, VEGFC and MAPK1/3 was also found. Therefore, m⁶A modification may have an impaction on the expression of genes in aortic tissues, which was consistent with previous findings.

It has been shown that m⁶A plays a role in the extracellular matrix^{30, 34}. The results of GO and KEGG analyses in previous study indicated that AD-related methylated genes are mainly enriched in the processes and pathways related to cellular matrix, such as the extracellular matrix organization, and ECM-receptor interaction signaling pathways. Our findings also supported the importance of m⁶A in the ECM-receptor interaction signaling pathway. But the outstanding pathway identified in this study was focal adhesion. In addition, differentially expressed hyper-methylated or hypo-methylated genes identified in the present study are showed enrichment in cardiomyopathy and cardiac muscle tissue development and morphogenesis. The canonical TGF-β signaling pathway promotes aortic development and maintains aortic wall homeostasis³⁵. Although TGF-β signaling pathway was not significant in functional enrichment analysis, TGFB1, TGFB2, TGFBR1, TGFBR1, TGFBR2 and TGFB1I1 were abnormally methylated in aorta of AD patients. Genes in these pathways are critical molecules that contribute to the development and progression of AD. The findings provides further evidence that AD is a heterogeneous and multifactorial disease and that a variety of pathogenic mechanisms are implicated in its occurrence and development and m⁶A modification might be involved in.

It is known that m⁶A methylation plays a critical role in the regulation of coordinate transcriptional and post-transcriptional gene expression, including mRNA splicing, export, localisation and stability. In this study, many abnormal m⁶A methylated genes were also found to be enriched in biological processes of RNA splicing, RNA processing and RNA localization, which uncovers emerging links between m⁶A mRNA methylation and genome transcription. Thus, RNA modification by m⁶A promotes the interaction of the methylated genes with nuclear transcription factors. The distortion of RNA splicing and localization induced by abnormal m⁶A modification in these gene may be a cause of AD.

In addition to the genes that were previously associated with AD, we also identified a number of novel genes. Some of the most essential and crucial pathological changes in the AD process include focal adhesion, ECM-receptor interaction, actin cytoskeleton, TGF-beta signaling pathway and Hypertrophic cardiomyopathy. These pathways can regulate several biological processes such as proliferation, migration, invasion, and apoptosis. Conjoint analysis of m⁶A-RIP-seq and RNA-seq data identified m6A-modified mRNA transcripts that were hyper-methylated or hypo-methylated and significantly differentially expressed. This may shed light on a new layer of gene regulation at the RNA level, ultimately giving rise to the field of m⁶A epitranscriptomics. Studies of the human aorta not only enrich our understanding of the mechanisms underlying AD but also provide a theoretical basis for the prevention and treatment. To provide further insight, it will be necessary, in a future study, to elucidate the biological function of the identified genes in the context of AD and verify if m⁶A methylation in these genes may provide a prospective therapeutic target for the prevention and treatment of AD.

In summary, we reveal a region-specific m⁶A methylation map and identified numerous m⁶A changes in critical AD susceptibility genes (e.g., FBN1, TGFB1, VEGFA, LOXL3, COL3A1 and SMAD3) in AD patients compared to healthy controls. The identified m⁶A modification might regulate critical AD-related pathways such as focal adhesion, ECM-receptor interaction signaling pathway, cardiomyopathy and cardiac muscle tissue development and morphogenesis. The findings provided new insights into the pathogenesis of AD by m⁶A methylation. Further research on m⁶A target genes in AD might contribute to the clinical application of AD molecular targeted therapy.

Study participants

All aorta samples used in this study were collected from the department of Cardiac Surgery, Fujian Medical University Union Hospital. A total of 142 participants have been enrolled in this study. Diseased aortic tissue samples were obtained from AD patients, and control aortic tissue samples were obtained from recipients of heart transplants or lung donors. The aorta samples were obtained safely during surgery. Tissues were immediately snap-frozen and kept in liquid nitrogen. The cause of the donor’s death was cerebrovascular or traffic accident. All donors had no history of cardiovascular diseases or active infection at the time of transplantation. Participants with hereditary aortic diseases, connective tissue disorders, cancers, drug history, infections, or any other immune-related diseases were excluded. 9 AD cases and nine controls for MeRIP-seq and RNA-seq (The clinical characteristics of the two groups were shown in Table 1), and 10 AD cases and 10 controls for MeRIP-qPCR validation were selected from the 142 participants. This study has been approved by the Ethics Committee of Fujian Medical University Union Hospital. Written informed consent was collected from the participant or their families, conforming to the Declaration of Helsinki.

RNA-seq

Total RNA from the aorta was extracted using TRIzol reagent (Invitrogen Corporation, CA, USA) following the manufacturer’s instructions. After removing the cytoplasmic ribosomal RNA and mitochondrial rRNA from the total RNA by Ribo-Zero rRNA Removal Kit (MRZG12324; Illumina, CA, USA), mRNAs and lncRNAs were fragmented to 100-200 nt fragments using fragmentation buffer (Illumina). For RNA-seq, 10 ng RNA fragment was used to constructed strand-specific RNA library in accordance with a dUTP method instead of dTTP to label the second-strand cDNA during synthesis. The quality of the library was done using a Bioptic Qsep100 analyzer. RNA-seq was performed on Illumina HiSeq 4000 sequencer (Illumina). The cut-adapt software (v1.9.3) was used to remove the 3’ adaptor trimming and low-quality reads, and then the clean high-quality trimmed reads were aligned to the reference genome (hg38) with HISAT2 software (v2.0.4)³¹. The expression profiles were obtained with the cuffdiff software (v2.2.1) and differentially expressed genes (DEGs) were assessed. The DEGs were determined with the threshold of |log2FC| > 1 and FDR < 0.05. The genomic assembly version used in this study was GRCh38.

MeRIP-seq

RNA fragments were incubated with anti-m6A polyclonal antibody (202003; Synaptic Systems, Goettingen, Germany) in immunoprecipitation (IP) buffer for 2 h at 4◦C, and subsequently, the mixture was immunoprecipitated by incubation with protein-A beads (Thermo Fisher Scientific, MA, USA) at 4◦C for an additional 2 h. The purified immunoprecipitated RNA and input RNA were collected for the generation of an RNA-seq library with NEBNext R UltraTM RNA Library Prep Kit (New England Biolabs, MA, USA). The quality of the library was done using a Bioptic Qsep100 analyzer. Then, paired-end library sequencing was performed on Illumina HiSeq 4000 sequencer (Illumina). The quality of the paired-end reads was analyzed using fastQC (v0.11.8).

After the 3’ adaptor trimming and low-quality reads removing by cut-adapt software (v1.9.3), the clean high-quality trimmed reads were aligned to the reference genome (hg38) with HISAT2 software (v2.0.4)³¹. Methylated sites on RNAs (peaks) were detected by MACS software and diffReps software was used to detect differentially methylated lncRNAs/mRNAs. Overlapped exons of mRNA and lncRNA were identified and chosen by home-made scripts. For differentially methylated m6A peaks, HOMER was used for de novo motif prediction. The input RNA-seq and MeRIP-seq data were analyzed jointly to screen genes with significant differences in both RNA m6A methylation and abundance between the two groups (|log2FC| > 1, FDR < 0.05). Differential methylation sites were identified using the exomePeak R package (v2.13.2) with default parameters. The DEGs were assessed using the DESeq2 R package (v1.10.1) between the AD group and the normal individuals. m6A peak visualization of representative genes was performed by IGV software.

Gene set enrichment analysis

Next, molecular pathways of differentially methylated genes (DMGs) with differential expression were determined by gene set enrichment analysis. Gene symbol and log2(FC) were sorted in descending order to create the ranking list. Then, the ranking list was used to calculate normalized enrichment scores and p-value. The C2 gene set ‘c2.cp.v7.1.symbols.gmt’ from the Molecular Signature Database version 7.0 was selected as the reference gene set. The number of permutations was set to 1000, and a gene set with nominal p-value < 0.05 was treated as significantly enriched. The data supporting the findings of the present study are available from corresponding author upon reasonable request.

MeRIP-qPCR validation

Transcripts revealed by the integration analysis of DEGs and DMGs were tested by reverse transcription-qPCR in 10 AD cases and 10 controls. The RNA sample was fragmented (300 nt) after incubation with fragmentation buffer at 94◦C for 4 min. A total of 5% of the fragmented RNA was saved as the input control. The procedure of m6A-IP sample preparation was the same as that of MeRIP-seq. Both input control and m6A-IP samples were subjected to reverse transcription-qPCR using Universal SYBR qPCR Master Mix (Q711-02/03, Vazyme Bio).

Immunohistochemistry Staining validation

The obtained AD and control aortic tissues were paraffin-embedded and sectioned into 3µm slices. After baking the slices for 2 hours (60 °C), they were then subjected to gradient immersion in xylene and ethanol. The slices were processed using an Immunohistochemistry Kit (UItraSensitive SP mouse/rabbit; MXB) and incubated overnight with corresponding primary antibodies (4°C). The next day, DAB staining (DAB-0031; MXB) was performed, and the results were observed under the Nikon fluorescence microscope (Nikon ECLIPSE Ni). Image-J (Version.1.53a) was used for result analysis.

Statistical analysis

The comparison between the two groups was performed using the GraphPad software with an unpaired Student’s t-test. ∗P < 0.05 was considered statistically significant.

ADs

Aortic dissections

m6A

N6-methyladenosine

MeRIP-seq

m⁶A specific RNA immunoprecipitation sequencing

DEGs

differentially expressed genes

DMGs

differentially methylated genes

DMMSs

differentially methylated m⁶A sites

Ethical approval: The studies involving human participants were reviewed and approved by the Ethics Committee of the Fujian Medical University Union Hospital (IACUC FJMU 2023-0053). This study complied with The Code of Ethics of the World Medical Association (Declaration of Helsinki). Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin. This study did not involve any animal experiments.

Consent for publication: Not applicable.

Data availability statement: The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

Competing Interests Statement: The authors declare that they have no competing interests.

Funding：This work was supported by the National Natural Science Foundation of China [U2005202], [82370470] and [82241209], and the Fujian Provincial Special Reserve Talents Fund [2021-25].

Author contributions: TC.Chai and LK.Ma contributed to the conceptualization. TC.Chai and LK.Ma contributed to data curation and writing of the original draft. LK.Ma and JK.Li contributed to the formal analysis. RM.Xie and J.He contributed to the investigation. TC.Chai, LK.Ma, LL.Tang contributed to the methodology and the visualization. ZH.Qiu and LW.Chen contributed to the writing, review, and editing. All authors contributed to the article and approved the submitted version.

Acknowledgments: Not applicable.

Sampson UK, Norman PE, Fowkes FG, Aboyans V, Yanna S, Harrell FE, Jr., Forouzanfar MH, Naghavi M, Denenberg JO, McDermott MM, Criqui MH, Mensah GA, Ezzati M, Murray C. Global and regional burden of aortic dissection and aneurysms: Mortality trends in 21 world regions, 1990 to 2010. Glob Heart. 2014;9:171-180 e110
Sampson UK, Norman PE, Fowkes FG, Aboyans V, Song Y, Harrell FE, Jr., Forouzanfar MH, Naghavi M, Denenberg JO, McDermott MM, Criqui MH, Mensah GA, Ezzati M, Murray C. Estimation of global and regional incidence and prevalence of abdominal aortic aneurysms 1990 to 2010. Glob Heart. 2014;9:159-170
Guo DC, Grove ML, Prakash SK, Eriksson P, Hostetler EM, LeMaire SA, Body SC, Shalhub S, Estrera AL, Safi HJ, Regalado ES, Zhou W, Mathis MR, Eagle KA, Yang B, Willer CJ, Boerwinkle E, Milewicz DM. Genetic variants in lrp1 and ulk4 are associated with acute aortic dissections. Am J Hum Genet. 2016;99:762-769
LeMaire SA, McDonald ML, Guo DC, Russell L, Miller CC, 3rd, Johnson RJ, Bekheirnia MR, Franco LM, Nguyen M, Pyeritz RE, Bavaria JE, Devereux R, Maslen C, Holmes KW, Eagle K, Body SC, Seidman C, Seidman JG, Isselbacher EM, Bray M, Coselli JS, Estrera AL, Safi HJ, Belmont JW, Leal SM, Milewicz DM. Genome-wide association study identifies a susceptibility locus for thoracic aortic aneurysms and aortic dissections spanning fbn1 at 15q21.1. Nat Genet. 2011;43:996-1000
Turley TN, O'Byrne MM, Kosel ML, de Andrade M, Gulati R, Hayes SN, Tweet MS, Olson TM. Identification of susceptibility loci for spontaneous coronary artery dissection. JAMA Cardiol. 2020;5:1-10
van 't Hof FN, Ruigrok YM, Lee CH, Ripke S, Anderson G, de Andrade M, Baas AF, Blankensteijn JD, Böttinger EP, Bown MJ, Broderick J, Bijlenga P, Carrell DS, Crawford DC, Crosslin DR, Ebeling C, Eriksson JG, Fornage M, Foroud T, von Und Zu Fraunberg M, Friedrich CM, Gaál EI, Gottesman O, Guo DC, Harrison SC, Hernesniemi J, Hofman A, Inoue I, Jääskeläinen JE, Jones GT, Kiemeney LA, Kivisaari R, Ko N, Koskinen S, Kubo M, Kullo IJ, Kuivaniemi H, Kurki MI, Laakso A, Lai D, Leal SM, Lehto H, LeMaire SA, Low SK, Malinowski J, McCarty CA, Milewicz DM, Mosley TH, Nakamura Y, Nakaoka H, Niemelä M, Pacheco J, Peissig PL, Pera J, Rasmussen-Torvik L, Ritchie MD, Rivadeneira F, van Rij AM, Santos-Cortez RL, Saratzis A, Slowik A, Takahashi A, Tromp G, Uitterlinden AG, Verma SS, Vermeulen SH, Wang GT, Han B, Rinkel GJ, de Bakker PI. Shared genetic risk factors of intracranial, abdominal, and thoracic aneurysms. J Am Heart Assoc. 2016;5
Adlam D, Olson TM, Combaret N, Kovacic JC, Iismaa SE, Al-Hussaini A, O'Byrne MM, Bouajila S, Georges A, Mishra K, Braund PS, d'Escamard V, Huang S, Margaritis M, Nelson CP, de Andrade M, Kadian-Dodov D, Welch CA, Mazurkiewicz S, Jeunemaitre X, Wong CMY, Giannoulatou E, Sweeting M, Muller D, Wood A, McGrath-Cadell L, Fatkin D, Dunwoodie SL, Harvey R, Holloway C, Empana JP, Jouven X, Olin JW, Gulati R, Tweet MS, Hayes SN, Samani NJ, Graham RM, Motreff P, Bouatia-Naji N. Association of the phactr1/edn1 genetic locus with spontaneous coronary artery dissection. J Am Coll Cardiol. 2019;73:58-66
Kaadan MI, MacDonald C, Ponzini F, Duran J, Newell K, Pitler L, Lin A, Weinberg I, Wood MJ, Lindsay ME. Prospective cardiovascular genetics evaluation in spontaneous coronary artery dissection. Circ Genom Precis Med. 2018;11:e001933
Solomonica A, Bagur R, Choudhury T, Lavi S. Familial spontaneous coronary artery dissection and the smad-3 mutation. Am J Cardiol. 2019;124:313-315
Verstraeten A, Perik M, Baranowska AA, Meester JAN, Van Den Heuvel L, Bastianen J, Kempers M, Krapels IPC, Maas A, Rideout A, Vandersteen A, Sobey G, Johnson D, Fransen E, Ghali N, Webb T, Al-Hussaini A, de Leeuw P, Delmotte P, Lopez-Sublet M, Pappaccogli M, Sprynger M, Toubiana L, Van Laer L, Van Dijk FS, Vikkula M, Samani NJ, Persu A, Adlam D, Loeys B. Enrichment of rare variants in loeys-dietz syndrome genes in spontaneous coronary artery dissection but not in severe fibromuscular dysplasia. Circulation. 2020;142:1021-1024
Sun Y, Chen Y, Li Y, Li Z, Li C, Yu T, Xiao L, Yu B, Zhao H, Tao M, Jiang J, Yan J, Wang Y, Zeng H, Shen X, Zhou Y, Jin L, Song W, Dou K, Wang DW. Association of tsr1 variants and spontaneous coronary artery dissection. J Am Coll Cardiol. 2019;74:167-176
Grond-Ginsbach C, Böckler D, Newton-Cheh C. Pathogenic tsr1 gene variants in patients with spontaneous coronary artery dissection. J Am Coll Cardiol. 2019;74:177-178
Turley TN, Theis JL, Sundsbak RS, Evans JM, O'Byrne MM, Gulati R, Tweet MS, Hayes SN, Olson TM. Rare missense variants in tln1 are associated with familial and sporadic spontaneous coronary artery dissection. Circ Genom Precis Med. 2019;12:e002437
Saw J, Yang ML, Trinder M, Tcheandjieu C, Xu C, Starovoytov A, Birt I, Mathis MR, Hunker KL, Schmidt EM, Jackson L, Fendrikova-Mahlay N, Zawistowski M, Brummett CM, Zoellner S, Katz A, Coleman DM, Swan K, O'Donnell CJ, Zhou X, Li JZ, Gornik HL, Assimes TL, Stanley JC, Brunham LR, Ganesh SK. Chromosome 1q21.2 and additional loci influence risk of spontaneous coronary artery dissection and myocardial infarction. Nat Commun. 2020;11:4432
Zhuang J, Luan P, Li H, Wang K, Zhang P, Xu Y, Peng W. The yin-yang dynamics of DNA methylation is the key regulator for smooth muscle cell phenotype switch and vascular remodeling. Arterioscler Thromb Vasc Biol. 2017;37:84-97
Ryer EJ, Ronning KE, Erdman R, Schworer CM, Elmore JR, Peeler TC, Nevius CD, Lillvis JH, Garvin RP, Franklin DP, Kuivaniemi H, Tromp G. The potential role of DNA methylation in abdominal aortic aneurysms. Int J Mol Sci. 2015;16:11259-11275
Malbec L, Zhang T, Chen YS, Zhang Y, Sun BF, Shi BY, Zhao YL, Yang Y, Yang YG. Dynamic methylome of internal mrna n(7)-methylguanosine and its regulatory role in translation. Cell Res. 2019;29:927-941
Meyer KD, Jaffrey SR. The dynamic epitranscriptome: N6-methyladenosine and gene expression control. Nat Rev Mol Cell Biol. 2014;15:313-326
Wang X, Lu Z, Gomez A, Hon GC, Yue Y, Han D, Fu Y, Parisien M, Dai Q, Jia G, Ren B, Pan T, He C. N6-methyladenosine-dependent regulation of messenger rna stability. Nature. 2014;505:117-120
Edupuganti RR, Geiger S, Lindeboom RGH, Shi H, Hsu PJ, Lu Z, Wang SY, Baltissen MPA, Jansen P, Rossa M, Muller M, Stunnenberg HG, He C, Carell T, Vermeulen M. N(6)-methyladenosine (m(6)a) recruits and repels proteins to regulate mrna homeostasis. Nat Struct Mol Biol. 2017;24:870-878
Fu Y, Dominissini D, Rechavi G, He C. Gene expression regulation mediated through reversible m⁶a rna methylation. Nat Rev Genet. 2014;15:293-306
Visvanathan A, Somasundaram K. Mrna traffic control reviewed: N6-methyladenosine (m(6) a) takes the driver's seat. Bioessays. 2017
Kmietczyk V, Riechert E, Kalinski L, Boileau E, Malovrh E, Malone B, Gorska A, Hofmann C, Varma E, Jürgensen L, Kamuf-Schenk V, Altmüller J, Tappu R, Busch M, Most P, Katus HA, Dieterich C, Völkers M. M(6)a-mrna methylation regulates cardiac gene expression and cellular growth. Life Sci Alliance. 2019;2
Mathiyalagan P, Adamiak M, Mayourian J, Sassi Y, Liang Y, Agarwal N, Jha D, Zhang S, Kohlbrenner E, Chepurko E, Chen J, Trivieri MG, Singh R, Bouchareb R, Fish K, Ishikawa K, Lebeche D, Hajjar RJ, Sahoo S. Fto-dependent n(6)-methyladenosine regulates cardiac function during remodeling and repair. Circulation. 2019;139:518-532
Dorn LE, Lasman L, Chen J, Xu X, Hund TJ, Medvedovic M, Hanna JH, van Berlo JH, Accornero F. The n(6)-methyladenosine mrna methylase mettl3 controls cardiac homeostasis and hypertrophy. Circulation. 2019;139:533-545
Mo XB, Lei SF, Zhang YH, Zhang H. Examination of the associations between m(6)a-associated single-nucleotide polymorphisms and blood pressure. Hypertens Res. 2019;42:1582-1589
Deng K, Ning X, Ren X, Yang B, Li J, Cao J, Chen J, Lu X, Chen S, Wang L. Transcriptome-wide n6-methyladenosine methylation landscape of coronary artery disease. Epigenomics. 2021;13:793-808
Mo XB, Lei SF, Zhang YH, Zhang H. Detection of m(6)a-associated snps as potential functional variants for coronary artery disease. Epigenomics. 2018;10:1279-1287
Chai T, Tian M, Yang X, Qiu Z, Lin X, Chen L. Genome-wide identification of rna modifications for spontaneous coronary aortic dissection. Front Genet. 2021;12:696562
Zhou X, Chen Z, Zhou J, Liu Y, Fan R, Sun T. Transcriptome and n6-methyladenosine rna methylome analyses in aortic dissection and normal human aorta. Front Cardiovasc Med. 2021;8:627380
Kim D, Langmead B, Salzberg SL. Hisat: A fast spliced aligner with low memory requirements. Nat Methods. 2015;12:357-360
He Y, Xing J, Wang S, Xin S, Han Y, Zhang J. Increased m6a methylation level is associated with the progression of human abdominal aortic aneurysm. Ann Transl Med. 2019;7:797
Li HB, Tong J, Zhu S, Batista PJ, Duffy EE, Zhao J, Bailis W, Cao G, Kroehling L, Chen Y, Wang G, Broughton JP, Chen YG, Kluger Y, Simon MD, Chang HY, Yin Z, Flavell RA. M(6)a mrna methylation controls t cell homeostasis by targeting the il-7/stat5/socs pathways. Nature. 2017;548:338-342
Liu Q, Li M, Jiang L, Jiang R, Fu B. Mettl3 promotes experimental osteoarthritis development by regulating inflammatory response and apoptosis in chondrocyte. Biochem Biophys Res Commun. 2019;516:22-27
Shen YH, LeMaire SA, Webb NR, Cassis LA, Daugherty A, Lu HS. Aortic aneurysms and dissections series: Part ii: Dynamic signaling responses in aortic aneurysms and dissections. Arterioscler Thromb Vasc Biol. 2020;40:e78-e86

Table 1. Clinical Characteristics of Matched Patients With Aortic Dissection and Controls.

Characteristics	AD (n = 9, % / mean±SD)	Con (n = 9, % / mean±SD)	X² / t	P value
Age, y	53.78±15.13	53.78±15.13	<0.001	1
Male	6 (66.7)	6 (66.7)	<0.001	1
BMI, kg/m²	25.6±1.75	23.05±2.37	2.598	0.019
≥25	7 (77.8)	1 (11.1)	8.1	0.015
Standford A	8 (88.9)	--	--	--
Diameter of ascending aorta, cm (tracheal carina level)	4.57±0.75	3.28±0.32	4.73	<0.001
Diameter of descending aorta , cm (tracheal carina level)	3.93±1.02	2.58±0.19	3.9	0.001
EF, %	68.2±6.43	71.9±1.98	-1.66	0.13
Risk factors
Hypertension	8 (88.9)	1 (11.1)	10.889	0.003
Diabetes	1 (11.1)	1 (11.1)	<0.001	1
Smoking	4 (44.4)	3 (33.3)	0.234	1
Drinking	5 (55.6)	3 (33.3)	0.9	0.637
Blood lipids
TG, mmol/L	1.41±0.73	1.14±0.47	0.942	0.36
TC, mmol/L	3.97±0.93	4.85±0.67	-2.29	0.036
HDL—C, mmol/L	1.07±0.25	1.51±0.3	-3.34	0.004
LDL—C, mmol/L	2.52±0.89	3.09±0.63	-1.56	0.139
APOA, g/L	1.18±0.22	1.52±0.19	-3.53	0.003
APOB, g/L	0.86±0.19	0.89±0.17	-0.413	0.685
Dyslipidemia	7 (77.8)	4 (44.4)	2.104	0.335
Patient history
Marfan syndrome	1 (11.1)	0 (0)	1.059	1
CAD	1 (11.1)	1 (11.1)	<0.001	1
HbA1c	5.97±0.89	4.84±0.79	2.825	0.012
HbA1c disorder	5 (55.6)	1 (11.1)	4	0.131

Table 2 General numbers of differentially methylated m⁶A peaks and associated genes

Item	Upmethylated		Downmethylated		Total
Item	Peak	Gene	Peak	Gene	Peak	Gene*
mRNA	3,318	2,642	1,573	1,355	4,891	3,689
lncRNA	63	51	34	32	97	79
Total	3,381	2,693	1,607	1,387	4,988	3,768

*Some genes contain both upmethylated and downmethylated peaks, therefore the total number of genes do not equals to the number of upmethylated genes plus the number of downmethylated genes.

Table 3 Top m⁶A peaks associated with AD

Gene name	CHR	Start	End	Position	FC	P value	FDR	Regulation
PCDH1	5	141853769	141853950	3'-UTR	448.82	9.55E-09	2.95E-07	DOWN
CNOT3	19	54154034	54154244	CDS	424.61	1.95E-04	2.40E-03	UP
CHI3L1	1	203178990	203179230	CDS	349.71	1.00E-14	6.31E-13	DOWN
RORC	1	151811347	151811527	CDS	317.37	6.76E-04	7.08E-03	DOWN
WWC2	4	183317005	183317186	3'-UTR	224.41	6.31E-14	3.16E-12	UP
TMEM268	9	114645732	114645883	3'-UTR	221.32	2.00E-04	2.45E-03	UP
LRRC8C	1	89714913	89715094	3'-UTR	186.11	1.58E-28	2.51E-26	DOWN
FZD1	7	91270608	91270788	3'-UTR	144.01	2.45E-07	6.03E-06	UP
PI16	6	36963097	36963397	CDS	143.01	8.51E-08	2.29E-06	DOWN
ARHGAP11A	15	32615678	32616038	5'-UTR	134.36	1.00E-24	1.26E-22	DOWN
KDM2A	11	67256196	67256347	3'-UTR	133.44	7.94E-33	1.58E-30	DOWN
BLNK	10	96223857	96227457	CDS	133.44	3.16E-05	4.90E-04	DOWN
STK32A	5	147387253	147387433	3'-UTR	129.79	2.24E-05	3.63E-04	UP
C10orf128	10	49166612	49166792	3'-UTR	113.77	1.26E-03	1.20E-02	DOWN
TNNC1	3	52452143	52452674	5'-UTR	113.77	2.95E-09	1.00E-07	DOWN
SOCS6	18	70327207	70327388	3'-UTR	110.66	2.24E-10	8.91E-09	DOWN

Table 4 Significant differentially methylated m⁶A peaks in known AD susceptibility genes

Gene name	CHR	Start	End	Position	FC	P value	FDR	Regulation
TGFB2	1	218434278	218442044	3'-UTR	5.13	1.26E-20	1.26E-18	DOWN
LOXL3	2	74532652	74533549	3'-UTR	4.03	1.20E-04	1.62E-03	UP
IL1R1	2	102164924	102165270	CDS	2.03	1.10E-05	1.91E-04	UP
COL3A1	2	188988619	189003441	CDS	2.38	7.94E-93	5.01E-90	UP
COL3A1	2	189004012	189008425	CDS	2.17	2.00E-25	2.51E-23	UP
TGFBR2	3	30671744	30674131	CDS	2.36	2.00E-35	3.98E-33	UP
VEGFA	6	43779990	43780320	CDS	7.89	6.31E-07	1.45E-05	UP
TGFBR1	9	99128892	99132558	CDS	3.10	6.31E-05	8.91E-04	DOWN
TGFBR1	9	99152189	99152578	3'-UTR	2.41	1.00E-12	5.01E-11	DOWN
FBN1	15	48474328	48483880	CDS	2.81	1.45E-10	5.89E-09	UP
FBN1	15	48499022	48505066	CDS	5.13	1.26E-30	2.51E-28	UP
FBN1	15	48537655	48537806	CDS	2.51	2.88E-06	5.62E-05	UP
SMAD3	15	67065992	67078037	CDS	2.45	3.47E-04	3.98E-03	DOWN
MAPK3	16	30118150	30122158	5'-UTR	2.50	5.01E-04	5.37E-03	UP
TGFB1	19	41332258	41342100	CDS	3.53	1.82E-04	2.29E-03	UP
MAPK1	22	21769164	21788787	CDS	3.68	9.33E-08	2.45E-06	UP

Table 5 Validation of differentially methylated transcripts

Gene	Discovery stage		Validation stage
Gene	Log₂(FC)	P value	Log₂(FC)	P value
ACTC1	-2.12	2.19E-04	-3.15	1.87E-02
COL1A1	3.03	1.58E-15	3.90	3.97E-02
COX4I1	2.62	2.95E-05	4.04	7.26E-02
CP	-3.87	2.00E-91	-3.19	5.23E-01
CREB3L1	-2.46	2.19E-04	-3.93	2.18E-01
F2R	2.06	4.17E-03	4.95	2.56E-02
NEAT1	1.47	1.00E-11	2.82	4.90E-02
PCOLCE2	3.38	3.98E-11	5.09	9.03E-02
PKD2	1.24	2.51E-16	3.97	3.48E-02
RYR2	2.90	2.51E-11	4.00	9.52E-01
SLC2A1	-1.89	1.78E-05	-3.71	3.29E-01
TM4SF1	-3.80	1.00E-78	-4.53	6.07E-01
TNNC1	-6.38	2.95E-09	-5.88	4.44E-03
VEGFA	2.98	6.31E-07	2.55	3.94E-02

Fisher's exact test, t-test for continuous variables. Age and gender were used as matching for 1:1 case-control matching study. P values are for comparisons between controls (Con) and patients with aortic dissection（AD). SD, standard deviation; BMI, body mass index; TC, total cholesterol; TG, triglyceride; HDL—C, high density lipoprotein cholesterol; LDL—C, low density lipoprotein cholesterol; APOA, apolipoprotein A; and APOB, apolipoprotein B; EF, ejection fraction; CAD, coronary artery disease; HbA1c, hemoglobin A1c.

No competing interests reported.

Transcriptome-wide identification of N6-methyladenosine modifications for aortic dissection

Status:

Version 1

Abstract

Figures

Introduction

Results

Discussion

Conclusion

Materials and Methods

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1