Making changes: m⁶A-mediated decay drives the endothelial-to-haematopoietic transition

Summary

Zhang et al. (2017) report critical functions of m⁶A during the endothelial-to-haematopoietic transition in zebrafish embryogenesis. m⁶A affects Notch signaling, likely though m⁶A-dependent and YTHDF2-mediated mRNA decay.

Main text

Haematopoietic stem/progenitor cells (HSPCs) are the progenitors of all blood cell types, replenishing old blood cells throughout life. In vertebrates, HSPCs are formed during embryogenesis through the process of endothelial-to-haematopoietic transition (EHT). Cells of the haemogenic endothelium undergo drastic changes in morphology, separating from the ventral wall of the dorsal aorta and entering the sub-aortic space to become HSPCs. HSPCs are phenotypically distinct from their precursors, and previous studies have shown that stem cell differentiation and fate transition rely on precise post-transcriptional gene regulation. A new study by Zhang et al. demonstrates that epitranscriptomic modifications are essential for proper gene regulation during EHT ¹.

Over 150 different types of chemical modifications make up an epitranscriptome that plays key roles in gene expression regulation. N⁶-methyladenosine (m⁶A) is the most abundant post-transcriptional mRNA modification in eukaryotes, and has recently been shown to influence a broad range of biological processes across multiple species ². It is installed by a “writer” complex containing the methyltransferases METTL3 and METTL14, both of which are required for proper methylation. Loss of METTL3 causes defects in processes including spermatogenesis, T cell differentiation, and stem cell differentiation. The demethylases FTO and ALKBH5 remove m⁶A and impact processes such as metabolism, neuronal function, spermatogenesis and cancer.

m⁶A carries out many of its functions through “reader” proteins that bind specifically to m⁶A-marked transcripts. The reader YTHDF2 accelerates the decay of m⁶A-containing transcripts by interacting with the CCR4-NOT deadenylase complex ³. This was the first mechanistically characterized function of m⁶A in mRNA, which led us to propose that m⁶A plays a crucial role by regulating transcriptome switching; it coordinates decay of a set of transcripts after they have been translated, allowing expression of a different set of transcripts. Indeed, YTHDF2 promotes the decay of methylated maternal mRNAs during the maternal-to-zygotic transition in zebrafish ⁴. YTHDF2 also regulates dosage of maternal transcripts during oocyte development in mice ⁵.

In their recent paper, Zhang et al. (2017) hypothesized that m⁶A affects cell differentiation during EHT by affecting the decay of certain transcripts (Figure 1). The authors used a mettl3 morpholino (MO) to reduce protein levels of METTL3 in zebrafish embryos, thus reducing m⁶A levels. By using antibody based immunoprecipitation coupled with high-throughput sequencing (MeRIP-seq/m⁶A-seq) to profile the m⁶A methylome in control or mettl3-depleted zebrafish embryos at 28 hours post fertilization (hpf), the authors found that targets of METTL3 include many embryonic development transcripts. Prominently, reduction of METTL3 impaired generation of HSPCs. Instead of differentiating into HSPCs, endothelial cells in mettl3 morphants retained their endothelial phenotype and demonstrated increased expression of several arterial endothelial marker transcripts, including deltaC, tbx20, hey2, and ephrinB2a. Levels of HSPC markers runx1 and cmyb were reduced in mettl3 morphants, and HPSCs demonstrated reduced differentiation toward erythroid, myeloid, and lymphoid cells. These findings demonstrated that the m⁶A mRNA methylation is critical for EHT.

Figure 1.

m⁶A mediates proper endothelial-to-haematopoietic transition by regulating notch1a transcript decay.

To investigate the mechanism underlying the defect in EHT, the authors used RNA-seq to evaluate the expression of METTL3 targets in mettl3 morphants. Notably notch1a, a key regulator of cell fate decisions that represses HSPC programming during EHT, demonstrated increased transcript expression and substantially decreased m⁶A methylation in mettl3 morphants. Inhibition of notch1a expression using a Notch inhibitor or notch1a MO rescued the expression of HSPC markers runx1 and cmyb.

Because YTHDF2 decreases the stability of many m⁶A-marked transcripts, the authors then hypothesized that notch1a expression increases in mettl3 mutants due to loss of the YTHDF2-mediated RNA decay. Depletion of ythdf2 caused defects in EHT similar to those in mettl3 mutants, supporting the authors’ hypothesis. Similar to loss of mettl3, loss of ythdf2 increased the lifetime of notch1a mRNA. Overexpression of YTHDF2 rescued the HSPC phenotype, and this rescue was not possible using a YTHDF2 mutant without m⁶A binding activity, demonstrating that m⁶A is necessary for proper YTHDF2-mediated regulation of EHT. The authors also constructed a notch1a-EGFP reporter with either a wild-type or mutated m⁶A modification site near the stop codon, determined using single-nucleotide resolution m⁶A profiling (miCLIP-seq). Both reporters decreased runx1 expression. However, rescue by overexpression of mettl3 and ythdf2 was achievable only with the wild-type reporter, demonstrating the necessity of m⁶A for proper notch1 clearance.

Finally, the authors showed that m⁶A also regulates EHT in mammals by using dissected mouse aorta-gonad-mesonephros (AGM) to study EHT. siRNA-mediated Mettl3 depletion reduced colony formation and increased the expression of Notch1, which also contains m⁶A in mouse AGM. Altogether, this study demonstrated that m⁶A plays a critical role in EHT by regulating the YTHDF2-mediated decay (Figure 1).

The study by Zhang et al. adds to the quickly growing body of work demonstrating involvement of m⁶A in turnover of certain transcripts or groups of transcripts in stem cells or progenitor cells during cell differentiation. Three studies by Wang et al. (2014), Batista et al. (2014), and Geula et al. (2015) demonstrate the importance of m⁶A in transcript turnover (reviewed in ²). In mouse embryonic stem cells (mESCs), levels of m⁶A in pluripotency transcripts are high, thus leading to timely decay of these pluripotency transcripts upon differentiation. Loss of m⁶A through depletion of METTL3 leads to prolonged expression of the ESC master regulator NANOG, consequently impairing ESC exit from self-renewal toward differentiation. Mice without Mettl3 do not survive, and cells from early embryos remain at a hyper-pluripotent state and contain higher levels of pluripotent transcripts. The study by Zhang et al. reveals that this role of m⁶A applies also to the formation of stem cells from endothelium.

Requiring further investigation is the mechanism by which transcripts are chosen for methylation, as well as the downstream consequences of this selection. Although the above studies on ESCs -- as well as studies on T cells, zebrafish maternal-to-zygotic transition, and murine oocyte maturation -- demonstrate that m⁶A marks transcripts that must be degraded for a cell to differentiate, methylation could have complex roles in coordinating translation and in some case may stabilize modified transcripts. These diverse roles of m⁶A may depend on the regional specificity of methylation or the interplay between writers, readers, and erasers with other cellular factors. Future studies focusing on these questions will guide our understanding of these processes.