In vitro erythrocyte production using human-induced pluripotent stem cells: determining the best hematopoietic stem cell sources

Abstract

Background: Blood transfusion is an essential part of medicine. However, many countries have been facing a national blood crisis. To address this ongoing blood shortage issue, there have been efforts to generate red blood cells (RBCs) in vitro, especially from human-induced pluripotent stem cells (hiPSCs). However, the best source of hiPSCs for this purpose is yet to be determined.

Methods: In this study, hiPSCs were established from three different hematopoietic stem cell sources-peripheral blood (PB), cord blood (CB) and bone marrow (BM) aspirates (n = 3 for each source)-using episomal reprogramming vectors and differentiated into functional RBCs. Various time-course studies including immunofluorescence assay, quantitative real-time PCR, flow cytometry, karyotyping, morphological analysis, oxygen binding capacity analysis, and RNA sequencing were performed to examine and compare the characteristics of hiPSCs and hiPSC-differentiated erythroid cells.

Results: hiPSC lines were established from each of the three sources and were found to be pluripotent and have comparable characteristics. All hiPSCs differentiated into erythroid cells, but there were discrepancies in differentiation and maturation efficiencies: CB-derived hiPSCs matured into erythroid cells the fastest while PB-derived hiPSCs required a longer time for maturation but showed the highest degree of reproducibility. BM-derived hiPSCs gave rise to diverse types of cells and exhibited poor differentiation efficiency. Nonetheless, erythroid cells differentiated from all hiPSC lines mainly expressed fetal and/or embryonic hemoglobin, indicating that primitive erythropoiesis occurred. Their oxygen equilibrium curves were all left-shifted.

Conclusions: Collectively, both PB- and CB-derived hiPSCs were favorably reliable sources for the clinical production of RBCs in vitro, despite several challenges that need to be overcome. However, owing to the limited availability and the large amount of CB required to produce hiPSCs, and the results of this study, the advantages of using PB-derived hiPSCs for RBC production in vitro may outweigh those of using CB-derived hiPSCs. We believe that our findings will facilitate the selection of optimal hiPSC lines for RBC production in vitro in the near future.

Keywords: Erythropoiesis; Hematopoietic stem cells; Human-induced pluripotent stem cell; Red blood cell.

Fig. 1

Diagrammatic representation of the hematopoietic and erythroid differentiation of hiPSCs. Abbreviations: KOSR, Knock Out Serum Replacement; BMP4, bone morphogenetic proteins 4; VEGF, vascular endothelial growth factor; b-FGF, fibroblast growth factor-basic; SCF, stem cell factor; IGF2, insulin-like growth factor 2; IBMX, 3-isobutyl-1-methylxanthine; SR1, StemReagenin1; HC, hydrocortisone; IL-3, interleukin-3; EPO, erythropoietin

Fig. 2

Characterization of human-induced pluripotent stem cells (hiPSCs) derived from three different sources. A Quantitative real-time polymerase chain reaction (qRT-PCR) results of the reprogramming factors OCT4, SOX2, c-MYC, KLF4, and NANOG. B Immunofluorescence assay results of the pluripotency markers, TRA1-60, SSEA4, SOX4, SOX2, and NANOG. C Flow cytometric analysis results of TRA-1-60 and SSEA4. D qRT-PCR results of the three germ layer markers, Brachyury (mesoderm), SOX17 (endoderm), and Nestin (ectoderm). E Karyotyping results. Abbreviations: PB, peripheral blood; CB, cord blood; BM, bone marrow

Fig. 3

Morphological analysis results and differential counts of hiPSC-differentiated cells during differentiation. A Wright–Giemsa staining images of enucleating reticulocytes with an extruded nucleus (red arrow), enucleated reticulocytes (blue arrows), and histiocytes (green arrow) observed in hiPSC-differentiated erythroid cells over time (1000× magnification, scale bar = 10 μm). B Differential counts of erythroid cell subpopulation ratios during differentiation. Abbreviations: PB, peripheral blood; CB, cord blood; BM, bone marrow

Fig. 4

Flow cytometric analysis of hematopoietic and erythroid markers during differentiation. A Chronological shift of CD markers over time. B Comparison of CD235a+/CD71+ and CD235a+/CD71− cell counts during differentiation presented as mean ± SD [*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (n = 3)]. Abbreviations: PB, peripheral blood; CB, cord blood; BM, bone marrow

Fig. 5

Scanning electron microscopy images of erythroid cells on day 34 (scale bar = 10 μm). Abbreviations: PB, peripheral blood; CB, cord blood; BM, bone marrow

Fig. 6

hiPSC-derived erythroid cells at the end of the differentiation process indicating primitive erythropoiesis. A Hemoglobin type qRT-PCR analysis results for erythroid cells on day 31 as determined using ANOVA [*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001 (n = 3)]. B Oxygen equilibrium curves of erythroid cells on day 34. Abbreviations: ANOVA, analysis of variance; HB, hemoglobin; PB, peripheral blood; CB, cord blood; BM, bone marrow