Integrative single-cell transcriptomic to identify novel mediator of human blood progenitor proliferation

Background

Hematopoietic stem cells (HSCs) are the most widely used cells for cell therapy because of their unique ability to reconstitute the entire blood and immune system upon transplantation. Despite being the only means of cure for several blood disorders, their application faces limited availability, resulting in an unmet demand. HSCs are obtained from two main sources: cord blood (CB) and bone marrow. Bone marrow HSCs are the most used as they can be collected in larger numbers from peripheral blood upon mobilisation, but compared to CB, they are less potent when transplanted and also more immune-reactive as they have been exposed to infections. Another source of potential HSCs investigated in recent years is pluripotent stem cell (PSC) differentiation, where we identified HSCs-like cells using single-cell transcriptomics. Both PSCs- and CB-derived stem and progenitor cells would be a significant advance for a tailored approach to cell therapy, allowing for precise matching of the haplotypes. Yet, this is hindered by their limited numbers; thus, defining the condition for successful expansion exvivo or in-vitro will establish a novel precision medicine approach to transplantation.

Fetal HSCs, such as those found in the CB, massively expand exclusively in the fetal liver soon after they develop during embryonic life. Hence, understanding the molecular mechanisms that control the proliferation of these cells within the fetal liver will aid their ex-vivo expansion in the laboratory. Furthermore, we have compared iPSC-derived progenitor cells with those present in the human fetal liver and showed that they are transcriptionally very close1, despite expressing a lower level of transcription factors associated with cell proliferation, such as EGR1, JUN, JUNB and FOS, making them ideal candidate genes for manipulation.

Altogether, this supports the hypothesis that defective proliferation capacity ex-vivo and invitro limits the clinical application of iPSC- and CB-derived HSCs.

Aims

This multidisciplinary project aims to:

1) identify secreted molecules that could be used to expand HSPCs. To do this, we will screen published human whole (blood and niche cells) fetal liver single-cell transcriptomic datasets2,3 for putative secretory molecules using single cell-to-cell interaction prediction models. We have developed an analytical pipeline based on a neural network that identifies corresponding cell types between datasets. Using this, we will integrate the available fetal liver atlases and identify consistent putative cell-to-cell interactions between blood progenitors and their niche.

2) Validate cell-cell interactions with candidate secreted factors (e.g. cytokines) or small molecules that modulate associated signalling pathways using our iPSCs model and CB-derived CD34+ cells. Furthermore, relevant molecule-receptor interactions will be validated in-vivo by immunocytochemistry or RNAscope on human fetal liver.

3) Assess the function of the transcription factor EGR1 in the context of HSPC expansion. To do this, tamoxifen-inducible EGR1-ERT2 iPSC lines will be generated by genome editing. These cell lines will be differentiated into HSPCs to assess proliferative capacity. If significant effects are observed, we will identify downstream targets of EGR1 using CHiP-seq.

Training Outcomes

Altogether, this project will explore the fetal liver niche and identify transcription factors and novel signalling molecules that support HSPC production and proliferation. This interdisciplinary research project will ensure training in using quantitative methods in bioinformatics and machine learning, as well as life science methods that will ultimately impact healthcare strategies by improving the culture system for cell therapy. The student will be supervised by three teams with complementary skills spanning from bioinformatics and machine learning, to cell models of human development and in vivo hematopoiesis, providing a robust approach to niche modelling for medical applications. At the end of the training, the student will have gained fundamental skills in experimental design, as well as writing, presentation and critical interpretation of scientific findings, and will also participate in public engagement events organised by CRM.

References

1. Fidanza, A. et al. Single cell analyses and machine learning define hematopoietic progenitor and HSC-like cells derived from human PSCs. Blood 136, 2893–2904 (2020).

2. Popescu, D.-M. et al. Decoding human fetal liver haematopoiesis. Nature 1–7 (2019). doi:10.1038/s41586-019-1652-y

3. Calvanese, V. et al. Mapping human haematopoietic stem cells from haemogenic endothelium to birth. Nat. 2022 6047906 604, 534–540 (2022).