Transcriptomic Analysis of the m6A Reader YTHDF2 in the Maintenance and Differentiation of Human Embryonic Stem Cells

Neurology Bioengineering Cell Biology Biochemistry Genetics Life Sciences Medicine
m6A modification YTHDF2 human embryonic stem cells neuroectoderm differentiation transcriptomics ROBO1

1. Research Background and Significance

Over the past decade, the role of epigenetics in regulating cell fate decisions and disease development has become increasingly prominent. As a crucial part of epigenetic regulation, modifications at the RNA level—especially N6-methyladenosine (m6A)—have been confirmed to be widely present within eukaryotic mRNAs, playing key roles in regulating mRNA stability, splicing, export, degradation, and translation. While many m6A “writers,” “erasers,” and “readers” have been gradually identified, the specific roles and mechanisms of the m6A reader YTHDF2 in human embryonic stem cell (hESC) self-renewal and differentiation remain largely unknown.

Stem cells are fundamental to regenerative medicine, as their pluripotency and differentiation potential provide scientific support for the “in vitro directed generation of functional cells and tissues.” Understanding and elucidating the molecular mechanisms underlying stem cell fate transitions is of great importance for developing new strategies for cell therapy, disease modeling, and drug screening. Although YTHDF2, as an m6A recognition protein, has been reported in mouse embryonic development and neurogenesis, its roles and targets in human stem cell fate regulation have yet to be systematically revealed. Therefore, this paper takes YTHDF2 as a breakthrough point to systematically explore its molecular mechanisms in hESC self-renewal and multi-lineage differentiation (especially towards neuroectoderm/neural progenitor cells), aiming to fill this gap in the field and provide a theoretical foundation for regenerative medicine.

2. Introduction to the Authors and Source

This study was collaboratively carried out by Boshi Feng, Yanxi Chen, Huanchang Tu, Jin Zhang, Lingling Tong, Xiaohan Lyu, Aaron Trent Irving, Di Chen, and others, spanning the Center for Reproductive Medicine and the Center for Regeneration and Cell Therapy of Zhejiang University-University of Edinburgh Institute (ZJU-UoE Institute), the Center of Infection Immunity and Cancer, and the Edinburgh Medical School in the UK—among other internationally renowned medical research institutions. The corresponding author is Professor Di Chen (dichen@intl.zju.edu.cn). The paper was published in 2025 by Oxford University Press in Science and Technology of Molecular and Cellular Life Sciences and has already been released as an open access publication for the global academic community.

3. Detailed Description of Research Workflow and Innovative Methods

1. Overall Research Workflow

The study follows the main thread of “targeted gene knockout—transcriptome profiling—functional validation—mechanistic exploration,” including the following key steps:

  1. Construction of hESCs with complete YTHDF2 knockout
  2. Analysis of maintenance and proliferation abilities in knockout versus control cells
  3. Induction of multi-lineage (endoderm/mesoderm/ectoderm) differentiation and molecular phenotype assessment
  4. Large-scale transcriptome sequencing and bioinformatics analysis
  5. Directed differentiation towards neural progenitors and specific expression profiling
  6. Identification of YTHDF2 binding targets using iTRIBE combined with high-throughput sequencing
  7. Target gene functional validation and dissection of m6A-dependent mechanisms

2. Key Experimental Methods and Technical Details

2.1 Establishment of YTHDF2 Complete Knockout Cell Lines

  • Method: The CRISPR/Cas9 system was used to design two sgRNAs targeting the start and stop codons of YTHDF2, combined with donor plasmids (carrying antibiotic resistance cassettes) for homologous recombination replacement, achieving a full ~32 kb genomic segment knockout.
  • Procedure: Four recombinant plasmids were introduced into hESCs by electroporation, followed by antibiotic selection, single clone picking, genomic PCR, and Western blot, among other assays to verify knockout efficiency.
  • Innovation: Employing a large fragment “gene replacement-style knockout” approach ensured complete inactivation of YTHDF2 with no residual functional activity.

2.2 Assessment of Cell Self-Renewal Capacity and Proliferation

  • Experimental content:

    • Morphological observation, immunofluorescence and Western blot analysis of core pluripotency factors (OCT4, NANOG, SOX2, etc.).
    • SSEA4 flow cytometric sorting.
    • EdU incorporation for proliferation rate assessment.

  • Results analysis: According to multiple criteria, YTHDF2 knockout did not significantly affect hESC self-renewal ability or proliferation rate, and was similar to the control group.

2.3 Multi-lineage Differentiation and Molecular Phenotyping

  • Differentiation protocol: Induction of endoderm, ectoderm, and mesoderm differentiation was performed on both knockout and control hESCs using commercial differentiation media, with cells harvested at key time points for RNA extraction.

  • Molecular detection:

    • Immunofluorescence staining for germ-layer–specific markers (SOX17, TBXT, SOX1, etc.).
    • Real-time quantitative PCR for pluripotency genes and lineage-specific gene expression profiles.

  • Sample size: Multiple biological replicates were set up in each group to ensure reliable results.

2.4 Transcriptome Sequencing and Bioinformatics Analysis

  • Sequencing procedures: RNA was extracted from cells at each stage, libraries were constructed, and paired-end sequencing was performed on the NovaSeq 6000 high-throughput platform.
  • Analysis workflow:
    • FastQC for quality control, Cutadapt for adapter trimming, STAR for mapping to the hg38 human genome.
    • Quantification with featureCounts, differential expression analysis (DEGs) with DESeq2 (log2FC > 2, adj p < 0.05).
    • Downstream functional enrichments: KEGG, GO, and GSEA.

2.5 iTRIBE Technology for YTHDF2 Target Identification

  • Method principle: Fusing YTHDF2 with the ADAR catalytic domain, constructing a Tet-on inducible expression system, then tracing downstream mRNA binding targets via A-to-I editing signals after overexpression.
  • Bioinformatics: Using the TRIBE algorithm to detect A-to-G base conversions (threshold: editing ratio ≥ 1%, reads ≥ 20), and cross-filtering with known m6A-modified gene sets (external GEO datasets).

2.6 Targeted Functional Damage Modeling and Mechanistic Validation

  • New strategy: Developing a dCas13b-ALKHB5 + crRNA-Robo1 system to simulate specific site m6A demethylation on Robo1 mRNA, to assess the actual phenotypic impact of this modification.

3. Main Analytical Conclusions

3.1 The Role of YTHDF2 in hESC Self-Renewal

  • Conclusion: Complete knockout of YTHDF2 did not cause significant changes to hESC morphology, pluripotency gene expression, cell proliferation, or SSEA4 positivity. YTHDF2 is therefore not essential for hESC self-renewal and maintenance.

3.2 The Specific Role of YTHDF2 in Multi-lineage Differentiation

  • Endoderm/Mesoderm/Ectoderm Differentiation: Overall transcriptomes and lineage-specific gene expression showed no large differences after knockout, but transcriptomic differences were most prominent in ectoderm.
  • In Ectoderm Differentiation: 3,027 upregulated and 2,106 downregulated genes (ectoderm) with significant differences were identified; pathway analysis indicated strong impact on neural development pathways, including neurogenesis and neuronal migration.
  • Neural Progenitor Differentiation: YTHDF2-deficient cells could still initiate expression of some neural progenitor cell (NPC) markers under NPC-directed differentiation conditions, but displayed pronounced transcriptomic differences, especially enriched in neural-related pathways such as GABAergic synapse and EGFR tyrosine kinase inhibitor resistance.

3.3 Identification and Mechanistic Dissection of YTHDF2 Target mRNAs

  • iTRIBE combined with m6A modification screening: Found ~4,332 YTHDF2-bound and m6A-modified target genes; intersecting with differentially expressed genes refined this to 86 upregulated and 224 downregulated candidates.
  • Discovery and Validation of Robo1:
    • Robo1 is a crucial receptor for axon guidance and migration, and its expression is regulated by YTHDF2-m6A recognition.
    • Using dCas13b-ALKHB5 to specifically demethylate Robo1 mRNA, simulating the knockout state, caused Robo1 levels to drop, with downstream neural-specific genes such as NEUROG1 and EOMES also downregulated, validating the functional linkage.
  • Mechanistic model: YTHDF2 exerts positive regulation over neural development genes via m6A-dependent binding and stabilization of target mRNAs—not just mRNA degradation, but with evidence of positive regulation via stabilizing effects.

4. Research Highlights and Innovations

  • Methodological innovation: The combination of large-fragment gene knockout with “iTRIBE fusion editing + high-throughput sequencing” systematically mapped the human YTHDF2 target gene network and function for the first time; point-specific m6A demethylation with dCas13b-ALKHB5 used to precisely simulate loss of m6A, exemplifying methodological originality and scalability.
  • Focused scientific breakthrough: This study is the first to explicitly identify YTHDF2’s unique role in steering hESCs towards neural lineages, and reveals a positive regulatory effect on migration-related genes such as Robo1, overturning the conventional notion that YTHDF2 predominantly mediates mRNA degradation and attributing to it new positive regulatory functions.
  • Clear mechanistic pathway: The m6A-YTHDF2-Robo1 axis is validated as a key regulatory link for neuroectoderm differentiation, clarifying the “reader” role of YTHDF2 in stem cell fate determination.
  • Clinical and theoretical significance: Provides a theoretical model for the mechanisms of human neurodevelopmental diseases. The findings can be referenced in stem cell neural differentiation, regenerative medicine, and neuro-mitochondrial disease research, and advance studies on the YTHDF m6A reader family’s functional diversity.

4. Discussion and Outlook

Combining transcriptomics and diverse functional experiments, this study explores compensatory effects from upregulation of other YTHDF family members (especially YTHDF3), given their high homology and functional redundancy; and provides a comprehensive overview of the relationship between the breadth of m6A modification and the specificity of YTHDF2 targets. Notably, YTHDF2’s function is not limited to m6A-dependent degradation, but also includes regulation of mRNA stability and multi-level fate intervention. The authors also note that different techniques (e.g., siRNA knockdown vs. CRISPR knockout) and different cell models (hESC vs. hiPSC) may yield different YTHDF2 phenotypes in neural differentiation, underscoring the need to consider the impact of technical and experimental system differences.

From a disease association perspective, Robo1 has been linked to developmental dyslexia, autism, and other human neurological disorders, indicating broad application prospects of the “m6A-YTHDF2-Robo1” axis in brain development pathologies and regenerative medicine. In the future, use of organoid models or three-dimensional brain tissues derived from pluripotent stem cells will provide new avenues to further study the role of YTHDF2 in neurogenesis and disease emergence.

5. Conclusions and Academic Value

In summary, this research fills a knowledge gap regarding YTHDF2’s role in regulating the neural differentiation of human embryonic stem cells, expanding the multifaceted model of m6A modification in cell fate regulation. Its original experimental strategy, panoramic transcriptome analysis, and precise mechanism-based disease validation greatly enrich understanding of the RNA epigenetic network in stem cell fate, and provide foundational theoretical and technical models for efficient preparation and application of functional neural cells as well as for analyzing neural-related diseases.

6. Other Important Information

  • Data Sharing: All sequencing data has been publicly uploaded to NCBI GEO (e.g., GSE268806), and bioinformatics codes released on Zenodo, facilitating reproducibility and further research.
  • Ethics Statement: The hESCs used were sourced from established cell banks, and all experiments were conducted in vitro with no new subject recruitment, conforming to international ethical standards.

Conclusion: As research into RNA epigenetic regulation heats up, the functional diversity and specific targeting mechanisms of m6A reader family members such as YTHDF2 are certain to become a major frontier in the intersectional fields of stem cell fate, neurodevelopment, and regenerative medicine. This work not only sets a paradigm for basic biology but also lays innovative theoretical and technological foundations for the diagnosis and treatment of human neurological diseases and for the industrialization of stem cell applications.