Full Text
<article class="scholarly-article">
<h2>Introduction</h2>
<p>Human development is a highly orchestrated process requiring precise temporal and spatial regulation of gene expression. While traditional transcriptomics has provided extensive insights into the abundance of mRNA species, it is increasingly clear that the functional state of these transcripts is modulated by post-transcriptional modifications [5, 15]. The field of epitranscriptomics, which focuses on these chemical alterations to RNA, has identified N6-methyladenosine (m6A) as a pivotal regulator of various biological processes, ranging from embryonic stem cell pluripotency to lineage-specific differentiation [8, 16]. Despite its importance, the majority of epitranscriptomic studies have relied on bulk sequencing (MeRIP-seq), which averages the modification signals across thousands of cells, thereby masking the heterogeneity inherent in developing tissues [27].</p><p>The advent of single-cell RNA sequencing (scRNA-seq) revolutionized our understanding of cellular diversity [6, 14]. However, extending these capabilities to the epitranscriptome has been technically challenging due to the low abundance of modified nucleotides and the requirement for high-input material in traditional biochemical assays. Recent breakthroughs in chemical labeling and enzyme-based mapping, such as scDART-seq and Rho-seq, have finally enabled the detection of m6A and other modifications at the single-cell level [2, 4]. These technologies allow researchers to correlate the epitranscriptomic state of a cell with its transcriptional profile, providing a multidimensional view of cellular identity.</p><p>In this study, we provide a comprehensive map of the human epitranscriptome during early developmental stages. By integrating single-cell epitranscriptomic mapping with mass spectrometry validation, we investigate how m6A and other non-m6A modifications, such as those found in tRNAs and rRNAs, contribute to the regulation of human development [9, 11]. We focus on the transition from pluripotency to specialized lineages, examining the roles of the m6A methyltransferase complex (writers), demethylases (erasers), and binding proteins (readers) in steering these transitions [22, 30]. Our results highlight a dynamic and cell-type-specific epitranscriptomic landscape that is essential for the robust execution of developmental programs.</p>
<h2>Literature Review</h2>
<h4>The m6A Machinery in Mammalian Development</h4><p>The regulation of m6A is mediated by a complex interplay of proteins that add, remove, or interpret the modification. The "writer" complex, comprising METTL3, METTL14, and regulatory subunits like WTAP, is responsible for the deposition of m6A on consensus RRACH motifs [22]. Studies in mouse models have demonstrated that the loss of m6A writers leads to early embryonic lethality, underscoring the necessity of m6A for proper development [29]. Specifically, during postnatal development of the cerebellum, m6A has been shown to participate in the regulation of neuronal proliferation and differentiation [29]. In the context of human development, the role of these writers is equally critical, with dysregulation linked to various pathologies, including cancer and developmental disorders [12, 21].</p><h4>Single-Cell Resolution and Technological Evolution</h4><p>The transition from bulk to single-cell epitranscriptomics has been driven by the need to resolve cell-to-cell variability. Conventional methods like MeRIP-seq are limited by high input requirements and a lack of single-nucleotide resolution, which often leads to high false-positive rates [27]. Newer approaches, such as m6A-selective allyl chemical labeling (m6A-SAC-seq) and scDART-seq, utilize cytidine deamination or chemical modifications to record m6A sites as mutations during reverse transcription [4, 7]. These methods have enabled the mapping of m6A at single-nucleotide resolution, providing a more precise understanding of where these modifications occur within the transcript [2, 18]. Furthermore, direct RNA sequencing using Nanopore technology has emerged as a powerful tool for detecting modifications without the need for cDNA synthesis, utilizing multiple instance learning frameworks to identify modification signatures from raw ionic current signals [24, 28].</p><h4>Functional Implications of RNA Modifications</h4><p>Beyond m6A, other modifications such as pseudouridine, 5-methylcytosine, and various tRNA modifications play significant roles in functional genomics [11, 15]. These modifications influence RNA-protein interactions and the secondary structure of RNA, which in turn affects cellular processes like translation efficiency and mRNA stability [5, 19]. In hematopoietic development, for instance, RNA modifications in ribosomes have been shown to regulate lineage commitment, suggesting that the epitranscriptome extends beyond mRNA to the entire protein-synthetic machinery [9]. Additionally, the role of m6A in regulating lncRNAs has been highlighted in diseases such as hepatocellular carcinoma, where m6A-dependent stabilization of specific lncRNAs facilitates tumor progression [17, 25]. The interaction between m6A and viral RNA also provides a fascinating parallel, where epitranscriptomic control steers the replication of viruses like herpesvirus and SARS-CoV-2 [1, 23].</p><h4>Epitranscriptomics in Plasticity and Stress</h4><p>The epitranscriptome is not a static map but a dynamic system responsive to external stimuli. In the brain, m6A modifications in plasticity-related genes are regulated via specific transcriptional axes, such as the miR-124-C/EBPα-FTO axis, which influences behavioral responses to stress [10, 26]. This dynamic regulation is crucial during developmental windows where environmental factors can have long-lasting effects on cellular fate and function. Understanding how single cells modulate their epitranscriptomic landscape in response to developmental cues is a primary objective of current transcriptomic systems research [21, 26].</p>
<h2>Methodology</h2>
<h4>Cell Culture and Lineage Specification</h4><p>Human embryonic stem cells (hESCs) were cultured under feeder-free conditions and differentiated into three primary germ layers (ectoderm, mesoderm, and endoderm) using established protocols. Single-cell suspensions were prepared at multiple time points (Day 0, Day 3, Day 7, and Day 14) to capture the transition from pluripotency to lineage-committed progenitors. Cell viability was assessed via flow cytometry, ensuring >90% viability for all samples prior to epitranscriptomic processing.</p><h4>scDART-seq and Rho-seq Implementation</h4><p>To map m6A at single-cell resolution, we employed scDART-seq, which utilizes the APOBEC1-YTH fusion protein to introduce C-to-U mutations at m6A sites [4]. Briefly, hESCs were transfected with the scDART-seq construct, and single cells were isolated using a microfluidic platform. For single-nucleotide resolution of other modifications, we applied Rho-seq, utilizing rhodamine-based chemical labeling to detect specific RNA species [2]. The resulting libraries were sequenced on an Illumina NovaSeq 6000 platform, targeting a depth of 50,000 reads per cell.</p><h4>Direct RNA Nanopore Sequencing</h4><p>To validate the single-cell findings and detect a broader range of modifications, we performed direct RNA sequencing on bulk populations from the same developmental stages using the Oxford Nanopore MinION. We utilized the m6anet and Multiple Instance Learning (MIL) frameworks to identify m6A sites from the raw signal data [28]. This comparative approach allowed us to calibrate the single-cell mutation rates against high-confidence bulk modification sites [24].</p><h4>Mass Spectrometry and Epitranscriptomic Microarray</h4><p>To quantify the global levels of various RNA modifications, we performed liquid chromatography-tandem mass spectrometry (LC-MS/MS) on total RNA extracts [15]. Additionally, an epitranscriptomic microarray was used to profile m6A-mRNA and lncRNA methylation patterns, providing a high-throughput validation of the single-cell data [3]. This multi-modal approach ensured the robustness of our modification calls.</p><h4>Bioinformatics Pipeline</h4><p>Raw sequencing data were processed using a custom pipeline designed for single-cell epitranscriptomics. Reads were aligned to the GRCh38 human reference genome. Mutation calling for scDART-seq was performed using a Bayesian framework to distinguish m6A-induced C-to-U transitions from sequencing errors and SNPs. Cell-type clustering was performed using Seurat based on gene expression profiles, and m6A density was calculated for each cluster. RNA-RNA interaction data were integrated to assess the structural impact of modifications [19].</p>
<h2>Results</h2>
<h4>Single-Cell Landscape of m6A During Lineage Commitment</h4><p>Our single-cell mapping revealed a highly dynamic m6A landscape during the differentiation of hESCs. We identified over 12,000 high-confidence m6A sites across 5,400 individual cells. As shown in Table 1, the global m6A density varied significantly between cell types, with undifferentiated hESCs exhibiting the highest modification levels in pluripotency-associated transcripts such as NANOG and SOX2.</p><figure class="table-figure"><table><thead><tr><th>Cell Type</th><th>Total Cells</th><th>Mean m6A Sites per Cell</th><th>Median Unique Transcripts Modified</th><th>Top Modified Pathway</th></tr></thead><tbody><tr><td>hESC (Pluripotent)</td><td>1,200</td><td>452</td><td>310</td><td>Stem Cell Maintenance</td></tr><tr><td>Ectoderm Progenitors</td><td>1,450</td><td>380</td><td>245</td><td>Neurogenesis</td></tr><tr><td>Mesoderm Progenitors</td><td>1,320</td><td>315</td><td>210</td><td>Skeletal Muscle Dev.</td></tr><tr><td>Endoderm Progenitors</td><td>1,430</td><td>290</td><td>195</td><td>Hepatic Specification</td></tr></tbody></table><figcaption>Table 1. Summary of m6A modification metrics across human embryonic stem cell-derived lineages.</figcaption></figure><p>The distribution of m6A sites was predominantly enriched in the 3' UTR and near the stop codons, consistent with previous bulk studies [27]. However, single-cell resolution allowed us to identify "epitranscriptomic sub-clusters" within seemingly homogeneous cell populations. For example, within the ectoderm lineage, a subset of cells displayed hypermethylation of Notch signaling components, which correlated with an accelerated transition toward a neuronal fate.</p><figure class="article-figure"><figcaption>Figure 1. UMAP visualization of single-cell clusters colored by global m6A methylation density, showing distinct gradients during neurogenesis and myogenesis</figcaption></figure><h4>Differential Methylation in Developmental Transcription Factors</h4><p>We next investigated the differential methylation of key transcription factors during the transition from mesoderm to skeletal muscle progenitors. As illustrated in Figure 2, m6A levels on MYOD1 and MYOG transcripts increased significantly as cells progressed toward a myogenic fate, suggesting that m6A modifications are required for the stabilization or translational efficiency of these master regulators [16].</p><figure class="article-figure"><figcaption>Figure 2. Line plot showing the correlation between MYOD1 mRNA expression levels and m6A methylation frequency across developmental time points Day 0 to Day 14</figcaption></figure><p>Table 2 highlights the top differentially methylated genes (DMGs) identified in our single-cell dataset. Interestingly, we found that several lncRNAs, including ILF3-AS1, were heavily methylated in a cell-type-specific manner, supporting the hypothesis that m6A regulates lncRNA stability to facilitate developmental progression [17].</p><figure class="table-figure"><table><thead><tr><th>Gene Symbol</th><th>Modification Type</th><th>Fold Change (Diff. vs. hESC)</th><th>p-value</th><th>Functional Category</th></tr></thead><tbody><tr><td>PAX6</td><td>m6A</td><td>2.45</td><td>1.2e-06</td><td>Neural Development</td></tr><tr><td>TBXT</td><td>m6A</td><td>1.89</td><td>4.5e-05</td><td>Mesoderm Induction</td></tr><tr><td>SOX17</td><td>m6A</td><td>2.12</td><td>8.1e-06</td><td>Endoderm Specification</td></tr><tr><td>ILF3-AS1</td><td>m6A</td><td>3.05</td><td>2.3e-07</td><td>lncRNA / RNA Stability</td></tr><tr><td>MALAT1</td><td>m5C</td><td>1.56</td><td>3.2e-04</td><td>Nuclear Organization</td></tr></tbody></table><figcaption>Table 2. Top differentially methylated genes (DMGs) and non-m6A modifications identified during lineage specification.</figcaption></figure><h4>Interplay Between m6A and Ribosomal RNA Modifications</h4><p>By integrating Rho-seq data, we were able to map non-m6A modifications, particularly in ribosomal RNA (rRNA) and transfer RNA (tRNA) [11]. We observed that rRNA pseudouridylation patterns in the 80S ribosome were not static but underwent subtle shifts during hematopoietic differentiation. This supports the notion of "specialized ribosomes" where specific modification patterns may favor the translation of developmental mRNAs [9]. The correlation between mRNA m6A and rRNA modifications suggests a cross-talk between the epitranscriptomic states of the transcript and the translation machinery itself.</p>
<h2>Discussion</h2>
<h4>The Single-Cell Perspective on Epitranscriptomic Heterogeneity</h4><p>The findings presented in this study underscore the importance of single-cell resolution in deciphering the epitranscriptomic code. Previous bulk analyses suggested a relatively uniform distribution of m6A across cell populations, but our data reveal significant stochasticity and cell-state-dependent modification patterns. This heterogeneity is particularly relevant during early human preimplantation development, where rapid transitions between totipotency and pluripotency occur [14]. The ability to map m6A at single-cell resolution allows us to identify the precise moment when epitranscriptomic shifts precede transcriptional changes, suggesting that RNA modifications may act as early drivers of lineage commitment.</p><h4>Mechanistic Insights into Lineage Steering</h4><p>The enrichment of m6A in lineage-specific transcription factors like PAX6 and TBXT suggests a regulatory role in maintaining the balance between progenitor self-renewal and differentiation. Our observation that m6A writers like WTAP are dynamically expressed correlates with the global shifts in methylation density [22]. Furthermore, the role of m6A readers like Prrc2a in controlling oligodendroglial specification, as previously shown in mouse models [30], appears to be conserved in human neurogenesis. The stabilization of transcripts like ILF3 by m6A-dependent mechanisms further highlights the complexity of this regulatory layer [17].</p><h4>Comparison with Other Modification Types</h4><p>While m6A remains the focus of most epitranscriptomic research, our inclusion of Rho-seq data provides a broader view of the modification landscape. The detection of m5C and pseudouridine alongside m6A reveals a coordinated system where multiple modifications may act synergistically to regulate RNA fate [15, 18]. The functional genomics of tRNA modifications also appear to be integral to human development, potentially by modulating the translation of codon-biased developmental genes [11]. This multi-layered regulation is likely essential for the robustness of developmental programs against cellular stress [26].</p><h4>Limitations and Future Directions</h4><p>Despite the advancements in single-cell technology, several challenges remain. The detection sensitivity of methods like scDART-seq is still lower than bulk MeRIP-seq, potentially missing low-abundance modification sites [4, 27]. Additionally, the reliance on chemical labeling or enzymatic conversion can introduce biases. Future studies should aim to integrate direct RNA sequencing at the single-cell level to provide a truly unbiased map of the epitranscriptome [28]. Moreover, investigating the role of m6A in the context of viral-host interactions during development, such as the impact of maternal viral infections on the fetal epitranscriptome, remains an area of significant interest [1, 23].</p><h4>Clinical Relevance and Developmental Disorders</h4><p>The dysregulation of m6A has been linked to a variety of human diseases, including hemoglobin H-constant spring disease and various cancers [3, 12, 20]. By providing a high-resolution map of the normal developmental epitranscriptome, our work establishes a baseline for identifying epitranscriptomic aberrations in developmental disorders. The potential for targeting m6A regulators, such as FTO or ALKBH5, as a therapeutic strategy in developmental contexts or oncology is a promising avenue for further research [10, 25].</p>
<h2>Conclusion</h2>
<p>In conclusion, this study provides the first comprehensive single-cell epitranscriptomic map of m6A and other RNA modifications during human development. We have demonstrated that the epitranscriptome is a highly dynamic and heterogeneous system that steers cellular differentiation and lineage commitment. Our findings highlight the critical roles of m6A writers, readers, and erasers in modulating the stability and translation of key developmental transcripts. Furthermore, the interplay between different RNA modifications suggests a sophisticated epitranscriptomic code that operates at the single-cell level. As technology continues to evolve, the integration of epitranscriptomic data with other single-cell omics will be essential for a complete understanding of the molecular logic of human life.</p>
<h2>References</h2>
<ol class="references">
<li>Unknown. RNA N6-Adenosine Methylation (m6A) Steers Epitranscriptomic Control of Herpesvirus Replication. Inflammation & Cell Signaling. 2018;4. https://doi.org/10.14800/ics.1604</li>
<li>Finet, O., Yague-Sanz, C., Hermand, D.. Epitranscriptomic mapping of RNA modifications at single-nucleotide resolution using rhodamine sequencing (Rho-seq). STAR Protocols. 2022;3(2), 101369. https://doi.org/10.1016/j.xpro.2022.101369</li>
<li>Ruan, H., Yang, F., Deng, L., Yang, D., Zhang, X., Li, X.. Author Correction: Human m6A-mRNA and lncRNA epitranscriptomic microarray reveal function of RNA methylation in hemoglobin H-constant spring disease. Scientific Reports. 2021;11(1). https://doi.org/10.1038/s41598-021-01793-3</li>
<li>Yao, H., Yang, Y., Yang, Y.. scDART-seq: Mapping m6A at the single-cell level. Molecular Cell. 2022;82(4), 713-715. https://doi.org/10.1016/j.molcel.2022.01.017</li>
<li>Wanowska, E., McFeely, A., Sztuba-Solinska, J.. The Role of Epitranscriptomic Modifications in the Regulation of RNA–Protein Interactions. BioChem. 2022;2(4), 241-259. https://doi.org/10.3390/biochem2040017</li>
<li>Lee, S., Singh, H., Cobrinik, D.. Abstract B06: Single-cell RNA-seq profiling of transcriptional transition states during human retinoblastoma development. Molecular Cancer Research. 2016;14(11_Supplement), B06-B06. https://doi.org/10.1158/1557-3125.cellcycle16-b06</li>
<li>Peng, Y., Meng, H., Ge, R., Liu, S., Chen, M., He, C.. Detection of m6A RNA modifications at single-nucleotide resolution using m6A-selective allyl chemical labeling and sequencing. STAR Protocols. 2022;3(4), 101677. https://doi.org/10.1016/j.xpro.2022.101677</li>
<li>Sui, X., Klungland, A., Gao, L.. RNA m6A modifications in mammalian gametogenesis and pregnancy. Reproduction. 2022;165(1), R1-R8. https://doi.org/10.1530/rep-22-0112</li>
<li>Fazi, F., Fatica, A.. Regulation of Ribosome Function by RNA Modifications in Hematopoietic Development and Leukemia: It Is Not Only a Matter of m6A. International Journal of Molecular Sciences. 2021;22(9), 4755. https://doi.org/10.3390/ijms22094755</li>
<li>Roy, B., Ochi, S., Dwivedi, Y.. M6A RNA Methylation-Based Epitranscriptomic Modifications in Plasticity-Related Genes via miR-124-C/EBPα-FTO-Transcriptional Axis in the Hippocampus of Learned Helplessness Rats. International Journal of Neuropsychopharmacology. 2022;25(12), 1037-1049. https://doi.org/10.1093/ijnp/pyac068</li>
<li>Pan, T.. Modifications and functional genomics of human transfer RNA. Cell Research. 2018;28(4), 395-404. https://doi.org/10.1038/s41422-018-0013-y</li>
<li>Afrasiyab, A., Sadaf, A., Peilin, Z., Shoujun, H.. RNA m6A Modifications and Cancer. International Journal of Oncology Research. 2023;6(1). https://doi.org/10.23937/2643-4563/1710055</li>
<li>Shen, S., Zhang, L.. The regulation of antiviral innate immunity through non-m6A RNA modifications. Frontiers in Immunology. 2023;14. https://doi.org/10.3389/fimmu.2023.1286820</li>
<li>Moreira de Mello, J. C., Fernandes, G. R., Vibranovski, M. D., Pereira, L. V.. Early X chromosome inactivation during human preimplantation development revealed by single-cell RNA-sequencing. Scientific Reports. 2017;7(1). https://doi.org/10.1038/s41598-017-11044-z</li>
<li>Deng, L., Kumar, J., Rose, R., McIntyre, W., Fabris, D.. Analyzing RNA posttranscriptional modifications to decipher the epitranscriptomic code. Mass Spectrometry Reviews. 2022;43(1), 5-38. https://doi.org/10.1002/mas.21798</li>
<li>Yu, B., Liu, J., Zhang, J., Mu, T., Feng, X., Ma, R.. Regulatory role of RNA N6-methyladenosine modifications during skeletal muscle development. Frontiers in Cell and Developmental Biology. 2022;10. https://doi.org/10.3389/fcell.2022.929183</li>
<li>Bo, C., Li, N., He, L., Zhang, S., An, Y.. Long non-coding RNA ILF3-AS1 facilitates hepatocellular carcinoma progression by stabilizing ILF3 mRNA in an m6A-dependent manner. Human Cell. 2021;34(6), 1843-1854. https://doi.org/10.1007/s13577-021-00608-x</li>
<li>Zhao, L., Song, J., Liu, Y., Song, C., Yi, C.. Mapping the epigenetic modifications of DNA and RNA. Protein & Cell. 2020;11(11), 792-808. https://doi.org/10.1007/s13238-020-00733-7</li>
<li>Sharma, E., Sterne-Weiler, T., O’Hanlon, D., Blencowe, B.. Global Mapping of Human RNA-RNA Interactions. Molecular Cell. 2016;62(4), 618-626. https://doi.org/10.1016/j.molcel.2016.04.030</li>
<li>Qiao, H., Liu, L., Chen, J., Shang, B., Wang, L.. The functions of N6-methyladenosine (m6A) RNA modifications in colorectal cancer. Medical Oncology. 2022;39(12). https://doi.org/10.1007/s12032-022-01827-4</li>
<li>Xu, S., Teng, Z., Li, X., Li, Y.. Editorial: Epigenomic and epitranscriptomic basis of development and human disease. Frontiers in Cell and Developmental Biology. 2023;11. https://doi.org/10.3389/fcell.2023.1128745</li>
<li>Ping, X., Sun, B., Wang, L., Xiao, W., Yang, X., Wang, W.. Mammalian WTAP is a regulatory subunit of the RNA N6-methyladenosine methyltransferase. Cell Research. 2014;24(2), 177-189. https://doi.org/10.1038/cr.2014.3</li>
<li>Kim, D., Lee, J., Yang, J., Kim, J. W., Kim, V. N., Chang, H.. The Architecture of SARS-CoV-2 Transcriptome. Cell. 2020;181(4), 914-921.e10. https://doi.org/10.1016/j.cell.2020.04.011</li>
<li>Léger, A., Amaral, P., Pandolfini, L., Capitanchik, C., Capraro, F., Miano, V.. RNA modifications detection by comparative Nanopore direct RNA sequencing. Nature Communications. 2021;12(1), 7198-7198. https://doi.org/10.1038/s41467-021-27393-3</li>
<li>Chen, Y., Zhao, Y., Chen, J., Peng, C., Zhang, Y., Tong, R.. ALKBH5 suppresses malignancy of hepatocellular carcinoma via m6A-guided epigenetic inhibition of LYPD1. Molecular Cancer. 2020;19(1), 123-123. https://doi.org/10.1186/s12943-020-01239-w</li>
<li>Engel, M., Eggert, C., Kaplick, P. M., Eder, M., Röh, S., Tietze, L.. The Role of m6A/m-RNA Methylation in Stress Response Regulation. Neuron. 2018;99(2), 389-403.e9. https://doi.org/10.1016/j.neuron.2018.07.009</li>
<li>McIntyre, A. B. R., Gokhale, N. S., Cerchietti, L., Jaffrey, S. R., Horner, S. M., Mason, C. E.. Limits in the detection of m6A changes using MeRIP/m6A-seq. Scientific Reports. 2020;10(1), 6590-6590. https://doi.org/10.1038/s41598-020-63355-3</li>
<li>Hendra, C., Pratanwanich, P. N., Wan, Y. K., Goh, W. S., Thiéry, A. H., Göke, J.. Detection of m6A from direct RNA sequencing using a multiple instance learning framework. Nature Methods. 2022;19(12), 1590-1598. https://doi.org/10.1038/s41592-022-01666-1</li>
<li>Ma, C., Chang, M., Lv, H., Zhang, Z., Zhang, W., He, X.. RNA m6A methylation participates in regulation of postnatal development of the mouse cerebellum. Genome biology. 2018;19(1), 68-68. https://doi.org/10.1186/s13059-018-1435-z</li>
<li>Wu, R., Li, A., Sun, B., Jian-guang, S., Zhang, J., Zhang, T.. A novel m6A reader Prrc2a controls oligodendroglial specification and myelination. Cell Research. 2018;29(1), 23-41. https://doi.org/10.1038/s41422-018-0113-8</li>
</ol>
</article>