DNA and RNA methylation directly linked for the first time
DNA and RNA methylation, initially thought to operate as distinct pathways, have been shown to interact closely, shifting our understanding of gene regulation.
A team of researchers from the Laboratory of Cancer Epigenetics at the Université libre de Bruxelles (Belgium), led by François Fuks, has discovered a paradigm-shifting link between DNA and RNA methylation. Their exposure of the connection between epigenetics and epitranscriptomics challenges our current understanding of these systems, revealing how they work closely to control fundamental pathways like embryo formation.
Epigenetics, the mechanisms that control gene expression via modifications made to DNA, operate through two key approaches in eukaryotes. The first, histone modification, involves the modification of histones through processes including methylation, influencing their conformation and, therefore, the accessibility of specific genes for transcription. The second, DNA methylation, primarily involves the direct methylation of cytosine bases (5mC) in DNA by enzymes known as DNA methyltransferases, with methylation of promoter regions diminishing gene expression and methylation in the gene body promoting it. Epitranscriptomics modifications, meanwhile, are primarily conducted by the methylation of adenosine by the methyltransferase complex METTL3-METTL14, forming N6-methyladenosine (m6A), which destabilizes RNA transcripts, enabling them to be broken down more readily, ultimately reducing gene expression.
The connection between histone modifications and DNA methylation has been well established, with the histone modification H3K36me3 leading to the enrichment of gene body methylation. The link between RNA methylation and histone methylation has also been observed, as METTL3-METTL14 has been shown to promote demethylation of H3K9me2 and that conversely, METTL3-METTL14 can be recruited to chromatin by histone modifications. However, a direct link between RNA and DNA methylation has yet to be shown and discussion around the topic is vanishingly small.
AI uncovers cancer-specific DNA methylation patterns
An AI model has been trained to recognize DNA methylation signatures of 13 different types of cancer.
To investigate this link, the team first conducted a NanoBRET (bioluminescence resonance energy transfer) binding assay with NanoLuc-tagged METTL3 or METTL14 and a HaloTag-tagged DNA methyltransferase DNMT1, which established a close association between the two molecules. Through further tests the team was able to reveal that chromatin-bound METTL3-METTL14 recruits DNMT1 to chromatin for gene-body methylation and identified a suite of genes that are regulated by both gene-body 5mC and post-transcriptional m6A. Using immunostaining, the team revealed co-localization of the DNMT1 and METTL3 or METTL14 in the nucleus. This suggests a regulatory mechanism that both promotes gene transcription and destabilizes the transcripts, reducing translation.
Exploring these identified genes further in wild-type and with knock-out Mettl3 or knock-out Mettl14 embryonic stem cells, the team was able to show that this regulatory mechanism plays a key role in the differentiation of embryonic stem cells to form the embryoid body. During differentiation, the team observed that an increase in 5mC and a decrease in m6A occurred in differentiation-associated genes, such as EOMES, NOTCH2 and SMAD3.
Next, the team intends to explore the potential of these mechanisms in drug design, looking to create ‘epigenetic drugs’ that simultaneously target DNA and RNA to restore regulation in disrupted pathways associated with disease.
