Generation of Hematopoietic Stem and Progenitor Cells from Human iPSCs

National Heart, Lung, And Blood Institute

Investigators

Linked publications & trials

Abstract

Objective 1: Develop a scalable culture system for hematopoietic differentiation of hiPSCs To address the limitations of current iPSC differentiation approaches, we initially sought to develop a chemically-defined and scalable hematopoietic differentiation protocol requiring no replating, EB formation or co-culture on stromal elements. Through systematic optimization steps, we established a simple 21-day monolayer-based culture system that recapitulates ex vivo the emergence of human hematopoiesis. Under culture conditions that favored mesodermal specification of human iPSCs (Day 0 to 3), an adherent monolayer rapidly formed. With the subsequent addition of hematopoietic cytokines (Day 3 to 21), hematopoietic clusters emerged from the monolayer before their eventual release in the supernatant fraction. We systematically characterized cells arising from this system by harvesting supernatant and monolayer populations at regular intervals between day 5 and 21 of differentiation. Within the supernatant, hematopoietic cells (CD43+CD45+/-) underwent sequential development with features of primitive wave hematopoiesis (peak at day 7), definitive multilineage HSPCs with potent colony formation activity in vitro but limited engraftment potential in vivo (peak at day 12), and definitive erythroid-committed progenitors expressing adult-type globin chains (peak at day 17 to 21). To understand the possible causes underpinning the absence of engraftable HSPCs in this system, we examined the cellular constituents of the CD43-CD45- non-hematopoietic fraction that supports hematopoiesis during differentiation. We first identified a prevalent population of mesenchymal cells throughout differentiation. Perivascular mesenchymal cells are known to interact with HSPCs and maintain their activity in the adult BM niche, but their role in promoting definitive HSPC development during ontogeny has not been demonstrated. Importantly, arterial HE was largely absent within the supportive monolayer. Thus, in keeping with prevailing arterial-specification models, we deduced that the adherent monolayer was likely inadequate to support the generation of bona fide engrafting HSCs in culture. Objective 2. Optimize iPSC differentiation conditions to promote arterial HE formation To augment the production of arterial HE during iPSC differentiation, we postulated that simultaneous activation of key signaling pathways independently shown to control arterial fate during vascular development, including WNT/-catenin, activin/nodal/TGF, and mitogen-activated protein kinase (MAPK)/ERK pathways, might provide a synergy sufficient to further instruct definitive hematopoiesis with engraftment potential from human iPSCs. To test this possibility, we supplemented the culture medium with the WNT/-catenin agonist CHIR99021 (CHIR) and nodal/activin/TGF inhibitor SB431542 (SB) at day 2 of iPSC differentiation and activated MAPK/ERK signaling with LY294002 (LY) from day 3 to day 6 of culture. Control cultures contained no CHIR/SB/LY (non-treated), or were supplemented with CHIR/SB or LY only. Compared to control cultures, addition of CHIR/SB/LY led to a marked increase in percentages and numbers of CD144+CD34hiCD73-CD184+ arterial HE, peaking at day 5 of differentiation. We next investigated whether this early increase in arterial HE formation observed in the presence of CHIR/SB/LY influenced hematopoietic development. Addition of CHIR/SB/LY decreased overall CD43+/-CD45+ hematopoietic cell numbers but a notable rise in percentages of phenotypically defined definitive HSPCs (CD34+CD45RA-CD90+) was observed within the hematopoietic population at day 12 of differentiation compared to controls. In CFU assays, the frequency of progenitors with multilineage differentiation capacity was similar between control groups but significantly increased in the CHIR/SB/LY group. To further assess the self-renewal and differentiation capacity of iPSC-derived hematopoietic CD34+ cells, colonies derived from the first round of CFU plating were pooled and replated in secondary CFU assays. Notably, we observed a 3-fold increase in total CFU numbers from CD34+ cells derived from CHIR/SB/LY cultures compared to control groups. However, these cells did not sustain long-term hematopoietic engraftment after transplantation into NSG mouse recipients, indicating that additional revisions to this system are required. Objective 3. Identify HSC-specific superenhancers and associated master transcription factors While activation of an arterial program is required for HSC induction ex vivo, our data suggest that current culture conditions fail to modulate other critical molecular programs uncoupled from arterial development. We hypothesized that inadequate activation of core TF networks defining HSC identify could explain the lack of engraftment potential in HSPCs generated ex vivo. Because SEs represent dense binding platforms for cell-type specific master TFs and transcriptional co-activators, we first sought to delineate the SE landscape and the TFs they regulate in HSC-enriched populations. We conducted genome-wide integrative ChIP-seq and ATAC-seq analyses on phenotypically defined CD34+CD38- HSCs purified from mobilized PB samples obtained from 3 independent healthy volunteers. A total of 873 SEs were identified by elevated H3K27ac signal density; they represented 3.8% of total enhancers and their median size was 10-fold larger than genomic regions encompassed by typical-enhancers. By filtering ATAC-seq data for conserved domains, and applying binding motif enrichment and proximity algorithms, we identified 594 TFs encoded by genes regulated by SEs in human CD34+CD38- cells. The curated TF gene set was validated by interrogating published gene expression datasets in human CD34+CD38- cells and via unbiased gene ontology (GO) analysis. To identify putative master regulators of HSC identity, we narrowed our list of 594 SE-regulated TFs to include only transcription factors with binding motifs within their own SEs, denoting a distinctive auto-regulatory property of master TFs. A total of 34 master regulators were uncovered. Objective 4. Enhance cell engineering through manipulation of master transcriptional regulators To evaluate a potential role of the uncovered HSC-specific master TFs for the generation of functional HSPCs ex vivo, we independently cloned the assembled library of 34 TFs into doxycycline-inducible lentiviral vectors and transduced human iPSCs with the complete library or a subset of 13 TFs deemed more promising based on preliminary testing in CFU assays and absence of expression in iPSC-derived HSPCs. Transduced iPSCs were subjected to hematopoietic differentiation using our previously optimized protocol. Addition of doxycycline from day 5 of differentiation, when arterial HEs have formed and EHT begins, allowed for a stage-appropriate transient expression of the hematopoietic master regulators. To evaluate functionality of HSPCs obtained at day 12 of culture, cells were plated in CFU assays or transplanted intrafemorally into NBSGW immune-deficient murine recipients. Notably, we observed 2-fold increase in numbers of CFUs from cultures with enforced expression of master TFs compared to mock-transduced control experiments. Colony size was also enhanced and a preponderance of the most primitive CFU types (CFU-GM and CFU-GEMM) was noted with the introduction of master TFs during hematopoietic differentiation of human iPSCs. Efficient short-term engraftment of human iPSC-derived HSPCs (15% in PB) was observed two months after transplantation, but long-term engraftment was not sustained.