R-loops’ m6A modification and its roles in cancers

Abstract

R-loops are three-stranded nucleic acid structures composed of an RNA–DNA hybrid and a displaced DNA strand. They are widespread and play crucial roles in regulating gene expression, DNA replication, and DNA and histone modifications. However, their regulatory mechanisms remain unclear. As R-loop detection technology advances, changes in R-loop levels have been observed in cancer models, often associated with transcription-replication conflicts and genomic instability. N6-methyladenosine (m6A) is an RNA epigenetic modification that regulates gene expression by affecting RNA localization, splicing, translation, and degradation. Upon reviewing the literature, we found that R-loops with m6A modifications are implicated in tumor development and progression. This article summarizes the molecular mechanisms and detection methods of R-loops and m6A modifications in gene regulation, and reviews recent research on m6A-modified R-loops in oncology. Our goal is to provide new insights into the origins of genomic instability in cancer and potential strategies for targeted therapy.

Introduction

According to the latest data released by the International Agency for Research on Cancer (IARC), a division of the World Health Organization (WHO), the global number of new cancer cases is staggering at 20 million. It is predicted that this number will surge by 77%, highlighting the increasing burden of cancer worldwide. This alarming trend demands immediate attention on a global scale.

In recent years, the advancement of various biotechnologies has provided us with a deeper understanding of the molecular mechanisms underlying the development of different malignant tumors. This knowledge is of significant importance in the diagnosis and treatment of cancer and brings hope for improving the poor prognosis associated with these tumors [1].

R-loops are a type of common three-stranded nucleic acid structure found in cells. They consist of an RNA strand that pairs with one of the template strands of double-stranded DNA, creating an RNA–DNA hybrid. This process leaves the other unpaired DNA strand free [2]. R-loops are widely present in the genome and often occur at specific sites in the DNA known as promoters and transcriptional termination sites. They account for approximately 5% of the mammalian genome [3,4,5].

For a long time, R-loops were considered harmful by-products of transcription, interfering with the normal processes of gene expression and leading to genomic instability [6]. Currently, R-loops are broadly categorized as either physiological or pathological. Physiological R-loops are the result of programmed processes involving specific genetic elements, whereas pathological R-loops arise erroneously and unexpectedly [6].

Physiological R-loops play crucial roles in various biological processes, including DNA repair, cell proliferation, cell differentiation, RNA transcription and processing, regulation of gene expression, and DNA methylation [7, 8]. When the homeostasis of R-loops is disrupted and the regulatory processes are compromised, pathological R-loops can accumulate, exerting replication pressure and causing genomic instability. These pathological R-loops may play a role in cancer susceptibility and development [9,10,11].

Based on their functions, R-loops can also be categorized as regulatory or unscheduled. Regulatory R-loops are involved in gene regulation and genome stability. They influence gene activity by modulating transcription, replication, recombination, centromere function, and DNA editing [12]. Additionally, regulatory R-loops contribute to genome stability by facilitating the repair of DNA double-strand breaks (DSB) and maintaining the integrity of short telomeres [12]. If the accumulation of unwanted R-loops is not properly managed, unscheduled R-loops can form. These unscheduled R-loops have been associated with aberrant gene regulation, causing genomic instability and potentially leading to neurological diseases such as amyotrophic lateral sclerosis (ALS4), Prader-Willi syndrome (PWS), Angelman syndrome (AS), as well as various other diseases including cancer.

Recently, there has been a discovery of RNA modifications, such as N6-methyladenosine (m6A), which play a role in maintaining the stability of R-loops [13]. m6A refers to the methylation of adenine's sixth nitrogen atom in an RNA molecule [14]. Among the various methods of RNA modification that have been identified, m6A modification is the most prevalent and abundant epigenetic modification [14]. Changes in the level of m6A modification in eukaryotic cells have a regulatory effect on RNA metabolism, including translation, splicing, export, decay, and degradation [15]. With the advancement of detection methods and high-throughput sequencing technology, there has been further understanding and exploration of the m6A modification of RNA. Many studies have shown that m6RNA modification is associated with different types of cancer (such as lung, gastric, hepatic, breast, acute myeloid leukemia, and some gynecologic tumors) and resistance to antineoplastic therapies [16,17,18,19,20,21,22].

Regulatory effect of R-loop on gene expression regulation

R-loops are nucleic acid structures consisting of three strands, including an RNA–DNA hybrid and a displaced DNA strand [23]. In 1976, Thomas et al. observed the R-loop structure for the first time using electron microscopy [24]. R-loops are non-B-type DNA structures that are formed during transcription when nascent RNA anneals with template DNA strands, thereby displacing non-template DNA strands [25]. R-loops can be formed in cis and trans. Cis R-loops form when RNA polymerase transcribes and nascent RNA anneals to the DNA template strand behind RNA Pol II. Trans R-loops form through the hybridization of RNA and distant DNA strands, such as those based on non-coding RNAs and guide RNAs [26]. Initially, R-loops were thought to be rare by-products of transcription, but they have now been found in various genomes, including bacteria, yeast, higher plants, and human cells [27,28,29,30]. R-loops play a significant role in the regulation of gene expression.

R-loop mediates transcriptional regulation

R-loop exerts a "promoter-like" function to promote gene transcription

A single-stranded DNA component in the R-loop structure has the potential to directly promote the transcription of Pol II antisense RNA, without the need for conventional transcription factors to activate the double-stranded DNA. When the nascent mRNA transcript invades the DNA duplex to form the R-loop structure, the elongated RNA Pol II can act as a promoter element of the antisense lncRNA, and the single-stranded DNA in the R-loop structure can serve as a transcription template, thereby promoting antisense transcription in the mammalian genome [31].

R-loop regulates transcription initiation by facilitating/hindering transcription factors

R-loops play a pivotal role in regulating transcription initiation by influencing the binding of transcription factors. This regulatory effect primarily occurs in the promoter region, where R-loops can hinder the binding of transcription factors, thereby inhibiting transcription initiation. On the other hand, R-loops can also facilitate transcription factor binding or block the binding sites of transcriptional repressors, leading to enhanced transcription. An illustrative example of this regulatory mechanism is observed in the VIM gene region, where the antisense lncRNA VIM-AS1 forms an R-loop in the promoter region of the protein-encoding gene VIM. This R-loop recruits the nuclear factor-κB (NF-κB) and enhances the binding of the transcriptional activator NF-κB, thereby promoting the expression of the protein-encoding gene VIM [32].

Resident R-loop inhibits transcriptional elongation

R-loops also have an impact on transcriptional elongation by acting as residents in various organisms. For instance, the formation of an R-loop structure in the non-coding RNA Snord116 gene region inhibits the extension of transcription at the non-coding RNA Ube3a-ATS gene, leading to the repression of Ube3a-ATS expression. The presence of GC-biased repeat units in the Snord116 gene predisposes this region to R-loop formation. Under normal physiological conditions, R-loop formation promotes the expression of the transcriptional complex in neurons through Ube3a-ATS, which subsequently silences the protein-encoding gene Ube3a. However, treatment with the topoisomerase inhibitor topotecan can result in excessive R-loop formation, causing the excess transcriptional complexes to arrest and subsequently leading to the inhibition of Ube3a-ATS expression. Consequently, this inhibition promotes the expression of the protein-coding gene Ube3a [33]. Variants in Snord116 or Ube3a have been associated with neurodevelopmental genetic disorders such as PWS and AS [33].

The formation of the R-loop causes pausing transcription termination

Many studies have shown that the R-loop can cause transcription termination through various mechanisms, such as promoting the suspension of Pol II at the 3' end [34]. In addition, the R-loop formed by the G-rich region after Pol II can recruit the human helicase Senataxin to relieve the R-loop, which in turn allows the XRN2 protein to degrade the nascent RNA strand using its 5'-3' exonuclease activity, and removes Pol II out of the DNA strand to terminate transcription [35]. The R-loop at gene termini may also recruit specific termination proteins to achieve transcriptional termination. Recent findings indicate that RNA m6A modification may play a role in enhancing the formation of the R-loop, thereby aiding in the process of transcription termination [36]. For example, researchers investigating breast cancer have discovered that circSMARCA5 binds to its maternal locus to generate an R-loop, resulting in the cessation of transcription in exon 15 of SMARCA5. This disturbance leads to the production of defective transcripts, translating into truncated non-functional proteins that are subsequently degraded, ultimately disrupting the DNA damage repair process [37]. However, the regulatory mechanism of transcriptional pause caused by the R-loop is still unclear and needs to be further explored and verified.

R-loop mediates epigenomics regulation

R-loop regulates DNA methylation

Genome-wide assays have revealed that R-loops are more abundant at sites with reduced DNA methylation and increased DNase hypersensitivity [38]. R-loops can impede DNA methylation in the promoter region by hindering the binding of DNA methyltransferases (DNMTs) to DNA [39, 40]. Researchers have discovered that nascent RNAs can form R-loops in the gene promoter region, which safeguard the gene promoter of BAMBI against DNA methylation alterations and facilitate further transcription by preventing DNA from binding to DNMTs [39]. In patients with ALS4, senataxin mutations can deplete R-loops, leading to transcriptional disruption and aberrant TGFβ signaling, which may contribute to motor neuron dysfunction and demise [39].

R-loop regulates chromatin conformational changes

R-loops can impact chromatin conformational changes by directly recruiting chromatin modification complexes. The long non-coding RNA (lncRNA) HOTTIP is frequently upregulated in acute myeloid leukemia (AML) and plays a role in the formation of HOXA topology-associated domains (TADs), which are high-order structures formed by the spatial folding of chromatin DNA. These structures can influence the function of promoters and enhancers, leading to enhanced gene expression within the TADs and promoting tumorigenesis and progression [41]. Mechanistic studies have demonstrated that HOTTIP can form an R-loop at two different CTCF protein binding sites on either side of the β-catenin gene through trans action at the TAD boundary. This R-loop strengthens the CTCF chromatin boundary and facilitates the formation of TADs containing the β-catenin gene by recruiting CTCF and other proteins involved in TAD formation to the relevant locations. Consequently, this drives oncogene transcription and leukemia development [41]. Moreover, R-loops can act as anchors for certain single-stranded lncRNAs to recruit chromatin remodeling complexes. The antisense lncRNA GTAT3-AS1, for instance, is capable of forming an R-loop in the co-promoter region of the GATA3 gene and GATA3-AS1 [42]. Simultaneously, some single-stranded GATA3-AS1 molecules can recruit MLL proteins to the promoter region of the GATA3 gene and enhance GATA3 gene transcription by trimethylating lysine 4 of histone H3 (H3K4me3) [42].

Trans-inducible R-loop is involved in the regulation of gene expression

Since cis-inducible R-loops can only form in regions where RNA transcription occurs, trans-inducible R-loops can form in multiple locations throughout the genome, leading to the emergence of multiple hotspots of genomic instability. As a result, trans-induced R-loops pose a greater threat to genome integrity compared to cis-formed R-loops. In Arabidopsis, lncRNA APOLO (auxin-regulated promoter loop) forms R-loops through trans interactions, which is part of a broad range of regulatory mechanisms for auxin response genes. Normally, the target gene of APOLO is silenced by Polycomb factor like heterochromatin protein 1 (LHP1). However, upon activation of the auxin-responsive APOLO, it recognizes a specific motif in the promoter region of its target gene, binds to that region, and forms an R-loop. This R-loop is anchored to the promoter region of the target gene, with the single-stranded APOLO RNA acting as a decoy for LHP1. Consequently, this promotes the expression of the target gene [43].

Standard technology for detecting R-loops

Currently, the localization of R-loops in the genome is primarily detected using genome unit points or whole genome sequencing. These methods enable the detection of specific information such as the locations and sequences of R-loops in the genome at different levels. However, the single-point R-loop detection method has a low throughput and cannot meet the research demands for identifying genome-wide R-loops comprehensively. To address this limitation, scientists have developed high-throughput detection technologies for genome-wide R-loop analysis using second-generation high-throughput sequencing. These methods mainly exploit proteins [40, 44,45,46,47] or antibodies [38, 40, 48,49,50,51] that recognize RNA:DNA hybrid chains. By enriching the RNA:DNA hybrid strand, and subsequently performing library preparation and sequencing analysis, precise mapping of R-loops and semi-quantitative information can be obtained. Nonetheless, accurately dissociating RNA:DNA hybrids from the genome and enhancing the resolution of R-loop high-throughput sequencing remain significant challenges for researchers. Additionally, the progress made in single-cell sequencing technology in recent years offers an alternative avenue for the development of R-loop detection methods. This advancement will facilitate a more accurate understanding of the biological functions of R-loops in relation to temporal and spatial dimensions [52,53,54]. As sequencing technology and bioinformatics continue to advance, numerous studies have demonstrated a direct association between R-loops and various diseases, including neurological disorders, cancer (Table 1), and other diseases. These findings may provide new targets and strategies for the treatment of related diseases.

Table 1 R-loops in cancer

Mechanism of m6A modification in regulating gene expression

m6A, a methylated form of RNA, was initially discovered in mRNA during the 1970s and has since been found to be widely present in various organisms. The formation of m6A involves the transfer of methyl groups to the N6 position of RNA adenine glycotide using S-adenosylmethionine (SAM) as a methyl donor. This process is catalyzed by a methyltransferase complex composed of METTL3, METTL14, Wilms tumor-associated protein, RBM15/15B, Virma, and ZC3H13 [15, 59, 60]. On the other hand, m6A demethylation is achieved by the action of the demethylases human AlkB homolog 5 (ALKBH5) and fat mass and obesity-associated protein (FTO) [61, 62].

RNA methylation is a dynamic and reversible process regulated by various methyltransferases and demethyltransferases. Moreover, the biological functions of m6A modification require specific "reader" proteins. Class I readers, such as YTHDC1/2 and YTHDF1/2/3, contain YTH domains that enable them to bind to transcripts containing m6A [63]. Class II readers, including hnRNPC and hnRNPA2B1, are heterogeneous nuclear ribonucleoproteins that regulate transcript alternative splicing or processing [64]. Class III readers, like insulin-like growth factor 2 mRNA-binding proteins (IGF2BPs), contain RNA recognition motifs (RRMs) and KH domains that recognize m6A modifications [65]. Other readers, such as eukaryotic initiation factor 3I (EIF3I) and human antigen R (HuR), also play roles in translation and transcript stability, respectively [66, 67].

The majority of m6A modifications occur in the shared motif "RRm6ACH" (R = G or A; H = A, C, or U). This modification pathway is crucial for regulating gene expression and affects multiple stages of RNA function, including localization, splicing, translation, and degradation [15, 68]. The effects of m6A modification on gene expression are achieved through various mechanisms.

m6A modification affects RNA processing and export

m6A can promote alternative splicing of precursor mRNA by recruiting RNA-binding proteins (RBPs) that regulate splicing or by directly affecting RBP-target RNA interactions [69]. Furthermore, m6A modification plays a role in selective polyadenylation (APA) within the nucleus [70,71,72,73] and miRNA biogenesis [18, 74, 75]. Additionally, m6A modification may impact the export of mRNA from the nucleus to the cytoplasm by facilitating the formation of nuclear export complexes through specific binding proteins [76,77,78,79,80]. For instance, the m6A reader protein YTHDC1 forms an RNA–protein complex with the downstream adaptor protein SRSF3, which selectively binds to NXF1 and promotes the export of m6A-modified mRNA to the cytosol [78].

m6A modification regulates RNA translation

m6A modification has been found to play a role in mRNA translation. YTHDF1 can recruit eukaryotic translation initiation factor 3 (eIF3) to bind m6A-tagged mRNA and facilitate cap-dependent translation [81]. Moreover, eIF3 can directly bind to m6A in the 5'-UTR of transcripts, enabling cap-independent translation [82]. YTHDC2 serves as a bridge between mRNAs containing m6A and ribosomes, enhancing the efficient translation of target mRNAs [83]. YTHDF3 binds to the transcript and transports it to YTHDF1, facilitating mRNA translation [84]. Conversely, m6A modification can also inhibit translation. FMR1 can interact with mRNAs containing m6A, thereby preventing ribosomal translocation and inhibiting translation [85]. Additionally, m6A modification can influence the translation of circRNA [86].

m6A modification regulates RNA stability and decay

m6A modification also regulates RNA stability and decay. Modified mRNAs are more likely to be recognized by specific RNA-binding proteins, such as the YTHDF family, resulting in mRNA degradation or stabilization [81]. RBPs also contribute to the decay of mRNAs with m6A modifications. IGF2BP, for example, stabilizes specific mRNAs in a m6A-dependent manner by adding mRNA stabilizers like HuR and MATR3 [65].

Chromatin regulation and transcriptional regulation of m6A modifications

m6A modification is involved in chromatin and transcriptional regulation. It can impact various histone modifications, including H3K27me3, H3K27ac, H3K9me2, and H3K4me3, which significantly influence gene expression [87, 88]. m6A modification also plays a role in the formation and maintenance of heterochromatin, affecting the organization of RNA/protein complexes that regulate gene expression and genome integrity [89, 90]. It has been found that m6A modification regulates RNA Pol II pausing and influences the transcription of nascent RNA, thereby precisely controlling gene expression [91]. Additionally, m6A modification is implicated in the transcriptional regulation of lncRNAs. m6A-tagged lncRNA x inactivity-specific transcripts (XISTs), for instance, recruit the YTHDC1 protein to promote XIST-mediated gene inhibition [92].

m6A detection methods

The first comprehensive map of m6A-modified RNA was published in 2012. Researchers utilized m6A antibodies to enrich RNA fragments containing m6A and generated profiles with a resolution of 100–200 nt [93, 94]. Currently, the predominant sequencing methods for m6A still rely on antibody-based RNA methylation sequencing, such as Methylated RNA Immunoprecipitation (MeRIP). However, these methods have notable limitations, including low resolution, lack of quantification, limited ability to compare m6A differences under different conditions, requirement for a large amount of RNA for library construction, and inapplicability to single-cell level research. Over the years, several m6A sequencing methods have been developed to address these limitations partially, including m6A-REF-seq/MAZTER-seq, DART-seq, and m6A-SEAL [95]. In 2022, Hu et al. proposed a novel method called m6A-SAC-Seq. This method enables direct labeling of m6A, covers almost all m6A classical motifs, and allows single-base resolution quantitative analysis of captured m6A sites [96]. The m6A-SAC-Seq technique is independent of antibodies and enables analysis of trace RNA samples, tracking dynamic changes in m6A distribution and content at the single-base resolution level [96]. This technology has broad applicability to various biological contexts and holds great potential for basic biological research and clinical applications.

With the continuous application and improvement of m6A-seq technology, significant advancements have been made in our understanding of m6A modification at the whole transcriptome level. It has been successfully applied to various functional studies, revealing the critical role of m6A modification in regulating RNA splicing, translation, and stability. Additionally, it reveals evolutionary conservation, tissue specificity, response to stress, and other distinctive features of m6A modification.

m6A-modified R-loops in tumors

The relationship between m6A modification and R-loop has become a research hotspot in recent years. Intracellularly, m6A modification is a common RNA modification that plays an important role in the regulation of gene expression by influencing RNA stability, transport, translation, and degradation. R-loop, on the other hand, is a structure formed by the interaction of RNA and DNA strands, usually consisting of one RNA strand bonded to one strand of DNA, while the other strand remains single-stranded. The relationship between m6A modification and R-loop is mainly reflected in the following aspects: m6A modification may affect the formation of R-loop during RNA transcription. Studies have shown that m6A modification can change the secondary structure of RNA, making it easier for RNA to interact with DNA during transcription, thereby promoting the generation of the R-loop [36]. The m6A modification also promotes the co-transcriptional R-loop to inhibit the readthrough activity of pol II, thereby promoting transcriptional termination [36]. Nevertheless, it has been suggested that m6A modification can promote the removal of the R-loop and prevent R-loop-mediated genotoxic stress [13]. The m6A modification also regulates the stability of the R-loop and plays a role in maintaining genomic stability [13]. The interaction between m6A modification and R-loop may play an important role in DNA repair and gene expression response under cellular stress or injury conditions. In some cases, R-loop may promote DNA repair, while m6A modification can modulate this process [97]. Overall, the interaction between m6A modification and R-loop provides a new perspective for understanding RNA biology and gene expression regulation, and future studies may reveal more about the mechanism of these interactions and their potential role in all kinds of diseases.

Aberrant accumulation of R-loops derived from various RNAs has been observed in cancer cells and may contribute to cancer progression by altering gene expression and exacerbating genomic instability [23]. m6A modification also plays a significant regulatory role in the occurrence and progression of tumors. It has been shown that m6A modification, facilitated by methyltransferase-like 3 (METTL3), mediates the formation and buildup of R-loops, thereby impacting transcriptional termination and maintaining genome stability [23]. Kang et al. discovered that the tension-responsive enhancer-binding protein (TonEBP) recruits METTL3 into the R-loop using the Rel homology domain (RHD) to facilitate m6A RNA methylation [98]. Depletion of TonEBP or METTL3 under conditions of UV or CPT exposure resulted in increased R-loop formation and decreased cell survival [98]. Furthermore, Hao et al. identified that DDX21 is co-recruited to chromatin by recognizing the R-loop, interacting with METTL3. During transcription, when the nascent transcript hybridizes with the DNA template strand, the co-transcriptional R-loop facilitates transcriptional termination and promotes genomic stability [99]. Any interruptions in these steps, such as DDX21, METTL3, or a loss of their enzymatic activity, can lead to a termination defect that induces DNA damage [99]. Considering the close association between m6A modification and R-loop, as both participate in gene expression regulation, tumorigenesis, and development, it is essential to investigate the mechanism of m6A-modified R-loop action in tumors. Such investigations may provide valuable insights into the origins of genomic instability in cancer cells and contribute to targeted therapy approaches. However, there is a scarcity of existing studies on this topic. Thus, our aim is to shed light on the significant role of m6A-modified R-loop in the development of tumors by focusing on the pathogenesis of prostate cancer, myeloproliferative syndrome, neuroblastoma, and AT-rich interaction domain 1A (ARID1A) abnormal cancer.

Prostate cancer

Prostate cancer (PCa) is the second most prevalent cancer and the fifth leading cause of cancer-related deaths in men [100]. In recent years, significant advancements have been made in comprehending the genomic landscape and biological functions associated with it. Various targeted drugs have been developed for PCa treatment, leading to improved prognosis [101]. Therefore, it is crucial to further investigate the molecular mechanism underlying PCa. Ying et al. revealed that co-transcriptional R-loop mediated epigenetics can regulate the growth retardation observed in advanced PCa [102]. They demonstrated that IGF2BPs recognize R-loops in an m6A-dependent manner, where IGF2BPs preferentially bind to m6A-modified R-loops [102]. Additionally, the KH domain of IGF2BPs is essential for recognizing m6A-containing R-loops and exerting tumor suppressor functions [102]. The potential mechanisms are illustrated in Fig. 1. Firstly, METTL3 and RBM15 act as key regulators of m6A modification, together inducing m6A methylation of R-loops in PCa. Subsequently, IGF2BPs selectively bind to m6A-modified R-loops, preventing the binding of YTHDF2 to these R-loops and resulting in increased R-loop levels in PCa [102]. Simultaneously, IGF2BPs also hinder the binding of DNA methyltransferase 1 (DNMT1) to the semaphorin 3F (SEMA3F) promoter, thereby safeguarding the SEMA3F gene from interference and promoting the production of SEMA3 protein [102]. Recent research has demonstrated that SEMA3 possesses numerous unique abilities such as activating the Hippo pathway, inhibiting tumorigenesis, angiogenesis, and tissue growth suppression [103]. Hence, it can be concluded that IGF2BPs, functioning as epigenetic R-loop readers, regulate R-loop metabolism in an m6A-dependent manner, leading to inhibition of cell migration and growth retardation in PCa [102].

Fig. 1

IGF2BPs regulate R-loop metabolism in an m6A-dependent manner, leading to inhibition of cell migration and growth retardation in PCa. A Without IGF2BPs, YTHDF2 preferentially binds to R-loops containing m6A, leading to the elimination of R-loops. DNMT1 binds directly to the promoter of SEMA3F, forming CpG islands and inhibiting SEMA3F transcription. B IGF2BPs selectively bind to m6A-modified R-loops and cause R‑loop accumulation. IGF2BPs also upregulate SEMA3F expression via repelling DNMT1 and YTHDF2. SEMA3s, the expression products of SEMA3F, activate the Hippo pathway and inhibit tumorigenesis, angiogenesis, and tissue growth suppression

Myelodysplastic syndrome

Myelodysplastic syndrome (MDS) is a form of myeloid neoplastic disease characterized by the gradual expansion of malignant hematopoietic clones, resulting in the failure of normal hematopoiesis. Pathological hematopoiesis of blood cells is also a characteristic feature of MDS, and about 30% of patients with MDS will progress to AML, earning MDS the nickname "pre-leukemia" [104, 105]. In recent years, genetic screening has led to increased recognition of hereditary MDS, with the DDX41 mutation being the most common susceptibility gene for this form of the disease [106, 107]. A recent study has revealed that CD34 + cells from MDS patients with DDX41 mutations show increased levels of m6A-modified R-loops and DNA damage [108]. In the R-loop structure, m6A modification is catalyzed by the METTL3 and METTL14 complexes, whereas the YTHDC1 protein is responsible for recognizing m6A and recruiting DNA repair proteins [109, 110]. In normal cells, DDX41 promotes the recruitment of YTHDC1 to the R-loops by facilitating the binding of the m6A complex to YTHDC1, thus participating in the DNA damage response (DDR) induced by R-loops (Fig. 2) [108]. However, this binding is not regulated in DDX41-deficient cells, leading to impaired binding affinity between YTHDC1 and METTL3/METTL14, which results in the accumulation of R-loops in DDX41-mutated MDS and increased DNA damage. This ultimately leads to genome instability and the onset of MDS (Fig. 2) [108]. These findings offer valuable insights into the molecular mechanisms underlying the development of DDX41-mutated MDS. Additionally, targeting m6A regulators may hold therapeutic potential for the treatment of DDX41-mutated MDS. Despite these findings, the exact mechanisms underlying m6A methylation in MDS development remain unclear, and further research on m6A-related molecules and their potential association with R-loops in MDS patients is still necessary.

Fig. 2

m6A-modified R-loops and DNA damage are elevated in DDX41-mutated CD34 + cells of MDS patients. A In normal cells, DDX41 promotes the recruitment of YTHDC1 to the R-loops by facilitating the binding of the m6A complex to YTHDC1, thus participating in the DDR induced by R-loops and maintaining genome stability. B In DDX41-deficient cells, the binding affinity between YTHDC1 and METTL3/METTL14 is impaired, which results in the accumulation of R-loops and increased DNA damage, ultimately leading to genome instability and the onset of MDS

Neuroblastoma

Neuroblastoma (NB) is the most prevalent solid tumor found outside the brain in children and exhibits a significant degree of clinical and biological diversity [111]. A crucial factor contributing to the heterogeneity of NB tumors and their disease results is believed to be the mechanism by which telomeres are maintained [112]. In approximately one-third of high-risk NB cases, telomeres are primarily sustained through the alternative lengthening of telomeres (ALT) pathway [112, 113]. It has been reported that only 10 to 15 percent of cancer cells extend telomeres via the ALT pathway, and these ALT-positive cancer cells exhibit substantial intrinsic DNA damage. They heavily rely on frequent homologous recombination (HR) which results in sister chromatin exchange at telomeres, thereby ensuring telomere length maintenance [114,115,116,117,118]. Telomeric repeat-containing RNA (TERRA) is a long non-coding RNA (lncRNA) derived from telomere ends. It localizes to telomeres in a manner dependent on a structure called the R-loop and plays a critical role in telomere maintenance. Vaid R et al. discovered that in ALT + NB, there is an abundance of m6A modifications in TERRA that is rich in R-loops. The m6A-mediated recruitment of hnRNPA2B1 to TERRA is essential for the formation of R-loops. The underlying mechanism is as follows: METTL3 mediates m6A modification of TERRA repeats and the m6A reader hnRNPA2B1 recognizes m6A. Subsequently, the m6A-modified TERRA invades telomeric DNA, leading to the formation of R-loops. This, in turn, promotes telomere maintenance in ALT + cells through HR [57] (Fig. 3). The researchers hypothesized that m6A modification of telomeric repeats containing UUAGGG may facilitate local unfolding of TERRA during transcription. This, in turn, enhances the recruitment of hnRNPA2B1 and other repair proteins such as RAD51, ultimately promoting R-loop formation [57]. Chen et al. further reported that METTL3 catalyzes the m6A modification of TERRA and that the m6A reader YTHDC1 recognizes and stabilizes TERRA. The m6A-modified TERRA then forms an R-loop, promoting HR, which contributes to the substitution and elongation of the ALT pathway in ALT + cancer cells [119]. Furthermore, in vitro experiments conducted by Vaid R et al. showed that treatment of ALT + NB cells with METTL3 inhibitors led to impaired telomere targeting and increased telomeric DNA damage [57]. Thus, targeting METTL3 to reduce R-loop formation and induce telomere shortening and instability could potentially be a novel approach for treating ALT-positive cancers. However, the existence of a similar mechanism of action in other ALT-positive cancers and the sensitivity of ALT-positive cancer cells to METTL3 inhibition still requires further exploration and verification.

Fig. 3

METTL3-mediated TERRA m6A modification plays a role in telomere maintenance of ALT + NB. METTL3 catalyzes m6A modification of TERRA repeats containing UUAGGG and the m6A reader hnRNPA2B1 recognizes m6A. Subsequently, the m6A-modified TERRA invades telomeric DNA, leading to the formation of R-loops. This, in turn, promotes telomere maintenance in ALT + cells through HR

ARID1A abnormal cancer

The ARID1A is a member of the SWI/SNF family, which plays a role in altering chromatin structure through helicase and ATPase activity, thereby regulating gene transcription [120]. In addition, ARID1A is involved in genome maintenance by locating damaged sites and clearing nucleosome occupancy quickly [121,122,123,124,125,126]. However, frequently observed mutations in ARID1A contribute to the development of various cancers, including ovarian, gastric, pancreatic, and bile duct cancer [127,128,129]. Zhang et al. confirmed that ARID1A recruits METTL3 and METTL14 to R-loops, catalyzing the m6A modification of RNA on the chromatin of the DNA DSB side in an ATM-dependent manner, and promotes HR repair in response to DNA damage (Fig. 3) [97]. Additionally, RNase H1 binds preferentially to m6A-modified R-loops and assists in their decomposition to maintain genome stability (Fig. 4) [97]. However, ARID1A deficiency results in HR defects, which may lead to tumorigenesis due to disorders in R-loops clearance and chromatin homeostasis disruptions [97]. Therefore, timely and efficient recruitment of RNaseH1 by m6A-modified R-loop RNA ensures genome stability during DNA damage repair, presenting a promising avenue for exploring potential therapeutic strategies for cancers with ARID1A aberration. Nevertheless, the mechanism through which m6A-modified R-loops recognize RNase H1 in a timely and effective manner is not yet clear, warranting further exploration [97]. Additionally, the involvement of this mechanism in the pathogenesis of ARID1A-related cancers, such as ovarian, gastric, and pancreatic cancer, requires verification as there are limited relevant studies available, highlighting the need for further in-depth research.

Fig. 4

The ARID1A-METTL3-m6A axis ensures effective RNase H1-mediated resolution of R-loops and genome stability. A Chromatin-enriched ARID1A recruits METTL3 and METTL14 to R-loops, catalyzing the m6A modification of RNA on the chromatin of the DNA DSB side in an ATM-dependent manner, and promotes HR repair in response to DNA damage. RNase H1 binds preferentially to m6A-modified R-loops and assists in their decomposition to maintain genome stability. B ARID1A deficiency results in HR defects, leading to disorders in R-loops clearance and chromatin homeostasis disruptions

Conclusions and outlook

In recent years, R-loops and m6A modification have gained increasing attention, playing significant roles in cancer and various other diseases, as indicated by advancements in detection technologies. However, the combined study of these two factors and their role in tumors remains unexplored. This article highlights four types of cancer, demonstrating that m6A-modified R-loops can exert either anti-tumor or pro-tumor effects through different molecular mechanisms. These findings signify the importance of m6A modification in R-loops in tumorigenesis and development, showcasing its potential clinical applications. Advancements in theoretical and clinical research will enable researchers to target relevant molecules involved in this mechanism as potential therapeutic targets for developing new drugs and improving the prognosis of malignant tumors. However, current research on this mechanism is still in its early stages, with many unknown aspects requiring further investigation and exploration.