Controlling the traffic flow in Endothelial-to-Hematopoietic Transition (EHT)

By Yee Hoon Foong

In the developing embryos, definitive hematopoiesis arises from precursor endothelial cells in a highly conserved process termed endothelial-to-hematopoietic transition (EHT). During this conversion, a subset of hemogenic endothelial (HE) cells in the aortic region undergo waves of transcriptional, epigenetic, and pathway remodeling to adopt a hematopoietic lineage. This is accompanied by a morphological shift in which the HEs adopt a rounded shape and break away from the tight junction of neighboring cells to form an intra-arterial cluster (IAC). The IACs are endowed with hematopoietic stem cell (HSC)-potential and upon migration to the fetal liver niche, they mature and expand to sustain the adult HSC pool.

Dr. Qin Zhu is a recent graduate from the Genomics and Computational Biology graduate program who lately published his seminal thesis work on elucidating the molecular underpinnings of EHT in the journal Blood1. Zhu’s work revolves around the emerging trend in the field of regenerative medicine: using patient-derived HSCs for the treatment of hematologic disorders. “It remains a critical challenge to find matched healthy bone marrow donors for hematologic malignancies, and this issue is accentuated in patient minorities of mixed ethnicity. One of the ultimate goals of regenerative medicine is to derive patient-specific HSCs from a patient’s own cells, such as blood vessel cells and induced pluripotent stem cells (iPSCs). Even though significant efforts have been made, there are so far no robust methods for production of HSCs outside the human body,” explains Qin. Indeed, although the ontogeny of the hematopoietic system has been well mapped out, the spatiotemporal dynamics of key transcription factors and signaling pathways driving EHT remain understudied, largely hampered by low cell numbers in developing embryos that precludes bulk sequencing technologies.

To tackle this, Zhu et al. leveraged single cell RNA-sequencing (scRNA-Seq) and single-cell assay for transposable-accessible chromatin sequencing (scATAC-Seq) on endothelial (E), HE, and IAC cells. By profiling 40, 000 cells from the caudal arteries of mouse embryos aged 9.5 days post coitus (dpc) to 11.5dpc, they delineated the developmental trajectory from E to HE to IAC cells. Through this analysis, they identified the rate-limiting step of developmental trajectory (“developmental bottleneck”) which is regulated by a key regulator Runx1, providing an entry point for potential therapeutic interventions.

By applying uniform manifold approximation and projection (UMAP) on their scRNA-seq datasets, Zhu et al. showed a continuous trajectory from E to IAC cells that could be further categorized into seven distinct subpopulations based on their transcriptomic signatures. At the base of the UMAP were two streams of arterial E cells with distinct Wnt signaling levels that eventually converge towards the HE and IAC lineages. Interestingly, E cells from vitelline and umbilical arteries contribute to one of the streams, whereas cells from dorsal aorta contributes to both streams, suggesting E cells with different molecular signature and loci may converge towards a common developmental fate.

Layering their UMAP with pseudotime trajectories and RNA velocity, Zhu et al. discovered that the RNA velocity was markedly reduced before the HE stage, potentially suggesting a differentiation barrier that limits a cell’s progression towards HE. This was mirrored by an accumulation of a distinct cluster of endothelial termed as pre-HE at the developmental bottleneck. Analysis of key marker genes suggested Runx1 as the predominant factor that regulates the passage of cells through the developmental bottleneck. To test the effects of Runx1 dosage level on cell fate determination, Zhu et al. performed scRNA-Seq of AE, pre-HE, HE, and IAC derived from Runx1 haploinsufficient embryos (Runx1 +/-) and their wildtype littermates (Runx+/+). They found that Runx1 haploinsufficiency increased the proportion of pre-HE while reducing the fraction of HE and IAC cells. The opposite trend was observed when they ectopically overexpressed Runx1. The findings validated their speculation of Runx1 as a gatekeeper that controls the flux of pre-HE cells through the bottleneck to become HE cells. Along with studies from other labs, Zhu et al. highlighted how the spatiotemporal expression of Runx1 are intimately tied to the EHT process.

To unbiasedly identify the transcription factors that modulate Runx1 expression, the authors performed tandem scRNA-Seq and scATAC-Seq on 10.5dpc CD44+ E, HE, and IAC cells. By assessing the differential transcription factor motif patterns on open chromatin, they discovered that the binding sites of HSC-specific transcription factors, including Runx1, become more accessible in pre-HE stage, indicating the concerted action of HSC-specific transcription factors in driving pre-HE to HE transition.

By integrating their scATAC-Seq data with a novel computational approach, the authors further mapped out enhancers whose chromatin accessibility significantly correlated with Runx1 expression. They identified one candidate enhancer located 371kb upstream that may potentially drive Runx1 expression during the pre-HE to HE transition. Interestingly, this candidate enhancer harbored GATA, STAT, and JUN transcription factor motifs, suggesting that these factors may orchestrate Runx1 expression.

Lastly, Zhu et al. investigated the derivation of IAC from HE cells, and the composition of IAC cells. Principal component analysis (PCA) overlayed with scRNA-seq expression showed that the cells repress the arterial E gene signature and activate hematopoietic genes during the HE to IAC transition. The authors further discovered that IACs are composed of at least two distinct hematopoietic stem and progenitor cell (HSPC) subtypes, committed lymphomyeloid-biased progenitors (LMPs) and pre-HSCs. These two subtypes arise sequentially, with LMPs enriched on 10.5dpc and pre-HSCs enriched on 11.5dpc. However, the LMPs appear to be more developmentally “mature” as compared to the type II pre-HSCs, suggesting that they are more driven towards terminal differentiation.

“This work would not be possible without the joint efforts of Dr. Kai Tan’s team, who pioneers in systems biology and single-cell genomics, and Dr. Nancy Speck’s team, who specializes in HSC formation and function,” Qin remarks.

Commenting on future experiments, Qin states, “One future direction is to figure out the exact function of the distal enhancer of Runx1. Previous literature suggests that this enhancer is capable of driving reporter gene expression in the intermediate cell mass and posterior blood island of zebrafish embryos, but it remains to be shown that the enhancer directly regulates Runx1 expression in pre-HE and HE cells.”

Together, this groundbreaking work by Zhu et al. provides new insights into the process by which endothelial cells differentiate into pre-HSCs and pinpoints the rate-limiting step, which could be exploited to potentially render autologous HSC transplant a more viable and scalable therapeutic plan.