Articles were extracted from the Web of Science Core Collection (WoSCC) and were downloaded on November 06, 2022. The search terms were as follows: TS = (RNA methylation) AND (“circulation” OR “heart” OR “cardiovascular”), and the publication date: “1990-01-01” to “2022-11-06”, resulting in 889 records. We excluded invalid documents, including editorial material (13 articles), book chapters (12 articles), proceeding paper (nine articles), meeting abstract (seven articles), and letters (one article). Finally, 847 documents were included, and all records were exported and saved in plain text format. Figure 1 displays the flow chart.

All valid documents retrieved from the WoSCC are exported into.txt files for visual analysis. We input the information into Microsoft Excel 2016 (Microsoft, United States) for the number of publications, the number of literature citations, and the analysis of research fields. CiteSpace 6.1.R4 advanced (CiteSpace for short, Drexel University, United States) and VOSviewer 1.6.18 (VOSviewer for short, the Centre for Science and Technology Studies, Holland) were used for literature clustering and visualization analysis. CiteSpace is a bibliometric analysis software developed by Chaomei Chen (20). It is a visual analytic tool for visualizing landmarks, critical paths, and emerging trends in a field of research based on relevant scholarly publications (21). We used CiteSpace to visualize the author and institutional collaboration, co-cited references and citation burst, keyword co-occurrence, and cluster and burst. The file was named starting with “download_” and imported into CiteSpace to remove duplicate files. There were no duplicate files. We imported the deduplicated files to CiteSpace for visual analysis. For institutional analysis, journal analysis, literature analysis, and author analysis, the parameter was set as “time slice 1 year” and “N = 50”; that is, the top 50 entries in each year were included in the analysis. The connection cable-cutting modes were pathfinder & pruning networks. To display image information more clearly, the time slice was set as two years in the keyword timeline view analysis, while other parameters remained unchanged. VOSviewer is an application for visual literature analysis, capable of coupling analysis of countries, journals, authors, citations, and more (22). We used VOSviwer to establish the item density of the countries. In the country analysis, the minimum number of articles per country was set at 1. In addition, the 2021 journal impact factors (IF) and JCR were obtained from the Web of Science.

The number of articles published in a field over the years can display the trend and popularity of research. Before 2010, the number of publications and citations related to RNA methylation in CVD was very small, with no papers in 2004 and only five in 2009. A turning point occurred in 2011, when 30 research papers were published, indicating that the field had captured the attention of researchers. Since then, the number of papers published has risen steadily. Since 2012, the number of citations of RNA methylation literature has increased rapidly, with a growth spurt from 2019 to 2021. Combined with the number of published articles and references cited, we infer that RNA methylation in CVD is rising rather than peaking. We expect that more articles will emerge in the next few years.

In national and institutional studies, the United States and China have an absolute advantage in the number of published articles; however, the institutions that publish articles are more scattered, and the number of published articles by a single research institution is not high. In contrast, though less than the United States in terms of the total number of articles published, China has the most among the top 20 institutions. This finding suggests that Chinese institutions such as Nanjing Medical University and Fudan University have a better research foundation and may make future breakthroughs in CVD RNA methylation.

In the journal analysis, the top ten journals were all in the CRJ Q1/Q2 area, suggesting that the outstanding journals have a high degree of recognition for the research results of RNA methylation. These results suggest that the study of RNA methylation in CVD has high credibility.

The co-cited literature with high citations represents the research basis of a discipline. We interpreted the top five co-cited references.

The first co-cited paper was the experimental article published in Circulation by Mathiyalagan et al. (26). The functional roles of m⁶A and FTO in the heart and cardiomyocytes were investigated using clinical samples, pig and mouse models, and primary cardiomyocyte cultures. The authors found that the expression of the demethylation enzyme FTO was decreased in the HF model, and the increase of FTO expression in HF mice reduced the decline of systolic function induced by m⁶A. These findings suggest the functional importance of the FTO-dependent m⁶A methyl group in cardiac contraction in HF.

Second place went to Dorn et al. (27), published in Circulation. The authors showed that METTL3-mediated methylation of m⁶A mRNA is a dynamic modification. Enhanced m⁶A leads to compensatory myocardial hypertrophy, while inhibition of m⁶A leads to muscle cell remodeling and dysfunction, highlighting the importance of METTL3-mediated methylation of m⁶A mRNA for maintaining normal cardiac function.

Third place went to Song et al. (28) for an article published in Autophagy. This study found that METTL3 silencing enhanced autophagy and inhibited hypoxia/reoxygenation-induced apoptosis of cardiomyocytes. However, overexpression of METTL3 had the opposite effect, suggesting that METTL3 inhibits autophagy. The transcription factor EB activates its transcription by binding to the ALKBH5 promoter, while the inhibition of METTL3 downregulates the stability of mRNA and does not involve transcription inhibition. The essential function of RNA methylation in the bidirectional regulation of ischemic heart disease was revealed.

In fourth place was a study by Berulava et al. (29) published in the European Journal of Heart Failure. Changes in m⁶A RNA methylation outweighed changes in gene expression in mammalian models of HF. Moreover, changes in m⁶A methylation of mRNA affected the polymeric binding of corresponding transcripts in the heart, affecting protein balance in a transcription-independent manner.

Roundtree et al. (30) described RNA methylation with different modifications, including adenosine methylation and cytosine modifications. Adenosine methylation in studies includes m⁶A and m¹A, while cytosine modifications include m⁵C and 5-hydroxymethylcytosine. This paper points out the technical difficulties of RNA methylation. Taking m⁶A methylation as an example, although m⁶A accounts for 0.2%–0.6% of all adenosine in mammalian mRNAs, only semi-quantitative detection can be performed within the transcriptome.

These studies explored the clinical, animal, and cellular levels. At present, there have been many studies on the methodology of RNA methylation. For example, XNA probe and CRISPR/Cas12a-powered flexible fluorescent and electrochemical dual-mode biosensor (38), mG-quant-seq (39), direct RNA sequencing (40), and other techniques provide technical support for the study of RNA methylation in CVD. We should also recognize the shortcomings of the research. The most prominent problems are the deviation of detection results due to low RNA methyl content (41, 42) and the lack of international standards for methylation studies. Although RNA methylation includes various types, current studies focus on the effects of m⁶A methylation on HF (43), ischemic cardiomyopathy (44, 45), and myocardial hypertrophy (46). Other diseases in CVD require further exploration.

Ten co-cited papers in the author's analysis were published by Francesco Paneni, assistant professor of cardiovascular epigenetics at the Schlieren Campus of the University of Zurich and team leader. A primary goal of Paneni's lab is to characterize the epigenetic mechanisms of the myocardium and vascular damage in the setting of cardiometabolic disturbances (47, 48).

Paneni's research team believes that epigenetic biomarkers should be regarded as the target of molecular intervention to treat diseases rather than blindly clinging to complex biological models of diseases (49). The aim is to apply epigenetic research to clinical applications rather than focusing on microscopic gene changes. One of these studies applied the central regulator SIRT1 to obese patients of different ages and found that SIRT1 affects the complex epigenetic control of p66sch or arginase II. NO availability and mitochondrial respiratory chain protein expression were affected to treat obesity and age-related early microvascular endothelial dysfunction (50). The team also reviewed drugs with known epigenetic effects and potential use in HF (51), including resveratrol (52), curcumin (53), and danshen (54).

The runner-up author, Claudio Napoli, published nine articles. Napoli's research on epigenetics is related to histones, microRNA, and DNA methylation, dedicated to the precise treatment of CVD and its high-risk pathogenic factors from the level of gene methylation (55–57). The Napoli group performed reduced-representation bisulfite sequencing and found that aortic endothelial cells exposed to high glucose had higher methylation differences of vascular endothelial growth factor (3.6 times), nitric oxide synthase 3 (1.6 times), and gene expression than normal aortic endothelial cells (58).

Third, Sarah Costantino suggested that studying cell-specific epigenetic information and developing chromatin modification drugs could provide a personalized risk assessment and treatment for patients with cardiometabolic disorders (51). The team examined the effects of hypoxia and hypertrophy stimulation on cardiomyocyte function and methylation levels and found that both resulted in epigenetic-driven upregulation of tenascin C and subsequent damage to cardiomyocyte energy metabolism, revealing the role of RNA methylation in the pathogenesis of CVD (59).

Gene-targeted therapy (60), precision medicine (61), and personalized therapy (62) were frequently used in co-cited literature. It is not difficult to see that the current research on RNA methylation in CVD is no longer limited to the simple mechanism study of disease but is committed to the clinical application of research results.

Co-cited author analysis can obtain high-influence authors in RNA methylation in CVD. We identified the authors with similar research topics, which authors created liaisons among topics, and which authors have multidisciplinary backgrounds or research fields with interdisciplinary attributes (63, 64). In the analysis of co-cited authors, Zhang, Wang, Mathiyalagan, and Wang were all cited more than 80 times. This finding suggests that their research results substantially influence this field.

Using CiteSpace for burst detection analysis, we discovered the evolutionary trends of research from macro to micro, from single to diversified, reviewed the period when keywords become hot and predicted which keywords will continue the explosive trend in the future (65).

The keyword burst detection analysis showed no keyword emergence before 2015, suggesting that RNA methylation was an emerging field. In 2016–2018, investigators began to study the relationship between RNA methylation and specific types of CVD, including cardiac hypertrophy (66, 67); and coronary artery disease (68, 69), and began to explore mechanisms such as DNA methylation and histone methylation. From 2016 to 2020, RNA methylation-mediated molecular mechanisms, including inflammation and endothelial damage, attracted attention. Endothelial dysfunction and inflammatory infiltration are essential processes that promote the development of atherosclerosis. High blood lipids, smoking, high blood shear stress, and other factors lead to abnormal vascular endothelial function, release adhesion factors and chemokines, such that monocytes in blood migrate to the arterial subendothelial space, differentiate into macrophages to take up lipids, and convert into foam cells, promoting atherosclerosis (70). Studies found that FTO can promote the demethylation of m⁶A by yes-associated protein 1 and reduce apoptosis and inflammation of hypoxia/re-oxygenation-induced cardiomyocytes (71). Lysine methyltransferase Smyd1 reduces IL-6 expression by enhancing methylation of trimethylation of lysine-4 of histone-3 within the IL-6 promoter (72). After 2020, m⁶A RNA methylation, nuclear RNA, inhibition, and other microscopic precise regulation gradually attracted investigator attention (73, 74).

Co-citation burst detection reflects the research nodes in the development of a specific field and is suitable for investigating research hotspots and predicting development trends (75). There were four articles with a burst intensity >20 (Figure 7), all of which were experimental articles published in CRJ Q1 journals. An experimental article published by Mathiyalagan et al. in 2019 (strength = 34.78, Circulation) was also the highest co-citation. From clinical to animal to cell, from in vitro to in vivo research, this paper comprehensively and meticulously demonstrated the role of FTO de-m⁶A methylation. This article represents a turning point in research methods and content (26). The second, an experimental article by Dorn et al. published in 2019 (strength = 29.9, Circulation), inhibited m⁶A RNA methylation in vitro and in vivo, clarifying the importance of RNA methylation for maintaining cardiac homeostasis (27). The authors of the third paper, Song et al. (strength = 22.39, Autophagy), revealed a link between METTL3-ALKBH5 and autophagy, providing a scientific basis for the role of reversible mRNA m⁶A methylation in ischemic heart disease (28). A fourth, published in 2020 by Berulava et al. (strength = 21.37, European Journal of Heart Failure), found that m⁶A methylation is altered in cardiac hypertrophy and HF with corresponding protein abundance but independent of mRNA levels, revealing a novel transcription-independent translation regulatory mechanism (29).

In RNA methylation studies, m⁶A is the most common methylation modification in mRNA, with approximately 25% of mRNA carrying at least one m⁶A site (76), while m⁵C and m¹A are found in tRNA and rRNA at relatively low levels (77, 78). m⁶A methylation is currently the focus of research on RNA methylation in CVD due to the differences in the content of different methylation types and the limited detection techniques.

Combined with citation bursts and keyword bursts, myocardial infarction, myocardial hypertrophy, and HF are current research hot spots. The exploration of methylation regulators related to these diseases and the clinical application of their results, especially the research related to m⁶A methylation, will become the goal of future research. The current RNA methylation detection methods are few and expensive, including liquid chromatography-mass spectrometry (79) and high-throughput sequencing (80). Dot blot represents a cost reduction for RNA methylation detection; however, the procedure is tedious and may result in false positives (81). Currently, the research on RNA methylation in the field of CVD is in its early stages, and optimizing the detection method of methylation to reduce the detection cost is a future direction of efforts.