Erythroid differentiation of human induced pluripotent stem cells is independent of donor cell type of origin

Authors: Isabel Dorn, Katharina Klich, Marcos J Arauzo-Bravo, Martina Radstaak, Simeon Santourlidis, Foued Ghanjati, Teja F Radke, Olympia E Psathaki, Gunnar Hargus, Jan Kramer, Martin Einhaus, Jeong Beom Kim, Gesine Kogler, Peter Wernet, Hans R Scholer, Peter Schlenke, Holm Zaehres

Abstract

This study tested whether induced pluripotent stem cells (iPSC) remember the type of adult cell they came from and whether that memory affects their ability to turn into red blood cells in the lab. iPSC made from cord blood CD34+ hematopoietic stem cells and from neural stem cells showed differences in DNA methylation patterns that reflect their original cell type. Despite this epigenetic memory, both types of iPSC behaved similarly when pushed to become blood cells and mature red blood cells. All iPSC lines produced mostly fetal hemoglobin and reached late stages of red blood cell maturation, including enucleated reticulocytes. The main difference was a modestly higher growth rate of erythroid cells from CD34+ iPSC. Detailed methylation of blood-related gene promoters was similar between the groups, matching their similar differentiation performance. These results are relevant for efforts to make blood and other cell types from iPSC and for choosing source cells for stem cell banking.

Introduction

What are iPSC and why this matters

• Induced pluripotent stem cells, or iPSC, are adult cells reprogrammed back into a flexible, stem-like state. They can self-renew and become many cell types in the lab.
• A concern in the field has been epigenetic memory. That means reprogrammed cells sometimes retain molecular marks from their original tissue, which might bias how well they become other cell types.

Why study red blood cell production

• There is strong interest in manufacturing red blood cells (RBC) outside the body for transfusions and personalized medicine. Cord blood or adult hematopoietic stem cells (HSC) are efficient sources, but supplies are limited. iPSC offer a theoretically unlimited source if they can be reliably driven into RBC.
• The study asks whether iPSC made from blood stem cells (CD34+ cells from cord blood) are better at making RBC than iPSC made from neural stem cells or fibroblasts because of epigenetic memory.

Methods in plain language

• Source cells: human cord blood CD34+ hematopoietic stem cells (CB CD34+), neural stem cells (NSC), and fibroblasts.
• Reprogramming: cells were converted into iPSC by introducing pluripotency factors (OCT4, SOX2, KLF4, c-MYC in some lines; fewer factors in others). Established human embryonic stem cells (hESC H1) were used as a reference.
• Epigenetic analysis: genome-wide DNA methylation of promoter regions was measured. DNA methylation is a chemical tag on DNA that often reduces gene activity. Patterns were compared between iPSC and their cells of origin.
• Differentiation to blood cells: iPSC were made into embryoid bodies, then treated with growth factors that promote blood and red blood cell development. Cultures were monitored by markers measured with flow cytometry, colony assays, microscopy, and hemoglobin analysis by HPLC.

Key results

• Reprogramming and pluripotency

• CD34+ cord blood cells were reprogrammed successfully into iPSC that resemble embryonic stem cells in gene expression and function, including forming teratomas that contain tissues from all three germ layers when tested in mice.

• Epigenetic memory is present but limited in effect

• Genome-wide analysis of promoter CpG methylation showed that iPSC retain an epigenetic signature related to their cell of origin. In other words, epigenetic memory exists even at passage numbers above 15.
• However, when the authors looked specifically at promoters of blood and erythroid genes (for example, CD34, GATA1, EPO, hemoglobin gene clusters), methylation levels were similar between CD34+-derived iPSC and NSC-derived iPSC. This suggests the epigenetic differences did not strongly affect the blood gene promoters tested.

• Hematopoietic induction and erythroid differentiation

• After 21 days of embryoid body differentiation, about 20% of cells in many iPSC lines expressed CD43, a marker of early blood commitment. Most CD43+ cells were already committed precursors, not long-term HSC.
• In the erythroid maturation phase (up to 25 days):
• iPSC from cord blood CD34+ cells and from neural stem cells both produced high percentages of erythroid cells (greater than 96% GPA+ at terminal stages).
• Enucleation rates (percentage of cells that expelled their nucleus to become reticulocytes) were similar across iPSC lines, around 21% to 29%.
• Scanning electron microscopy showed reticulocytes and occasional biconcave-shaped red blood cells.

• Cell expansion and hemoglobin type

• Growth: erythroid expansion peaked around days 15 to 18. CD34+-derived iPSC showed higher expansion than NSC-derived iPSC (average 1,368-fold vs 635-fold from initial hematopoietic cells). When calculated from unselected single cells after embryoid body digestion, CD34+-iPSC produced on average about 33,000 erythroid cells per 1,000 seeded cells with about 9,000 enucleated; NSC-iPSC produced about 18,000 with 5,000 enucleated.
• Hemoglobin: iPSC-derived RBC mainly expressed fetal hemoglobin (HbF). Typical results: CD34+-iPSC 89% HbF, NSC-iPSC 91% HbF, fibroblast-iPSC ~85% HbF. Embryonic hemoglobin made up 6%-7%; adult hemoglobin was low (roughly 2%-7%). By contrast, adult CD34+ HSC-derived cultures produce mainly adult hemoglobin and much larger expansion and enucleation rates (for example, unpublished comparison numbers in the study reported ~89,000-fold expansion and ~85% enucleation for adult HSC under the same conditions).

What this means for stem cell banking and regenerative medicine

• Source choice: although iPSC retain some epigenetic memory tied to their donor cell type, that memory did not prevent robust hematopoietic induction and erythroid maturation when iPSC were derived from either cord blood CD34+ cells or neural stem cells. CD34+-iPSC had a modest advantage in expansion, suggesting that blood-derived starting cells may help increase yield when making blood cells.
• Cord blood advantages: cord blood CD34+ cells are young, have fewer accumulated mutations, are already banked worldwide, and are well characterized for immune and infectious markers. For these reasons, cord blood is attractive both for direct therapeutic use and as a starting material to create iPSC banks that could generate blood products or histocompatible cell therapies on demand.
• Clinical translation challenges: iPSC-derived red blood cells currently fall short of adult HSC in two key ways: expansion scale and adult hemoglobin expression. Large-scale production for transfusion remains a major technological hurdle. Strategies to improve yields could include transient expression of proliferation-promoting factors, targeted gene regulation to promote adult hemoglobin switching, or improved culture niches.
• Broader regenerative relevance: preserving young, well-characterized cells in cord blood banks supports future options. If a person has their cord blood stored, that material could later be reprogrammed or used directly to make personalized cell therapies with lower risk of accumulated mutations, which relates to healthy aging and longevity of cell function.

Discussion and practical takeaways

• Main conclusion: iPSC retain some epigenetic memory of their tissue of origin, but for erythroid differentiation under these lab conditions, iPSC made from cord blood CD34+ cells and neural stem cells performed similarly in producing mature red blood cells. The main practical difference was a higher expansion from CD34+-derived iPSC.
• For stem cell banking customers: cord blood banking preserves a source of youthful hematopoietic cells that can be used directly for transplantation or as a starting point to create iPSC lines. That may offer advantages for generating blood products and personalized regenerative therapies later in life.
• Remaining research needs: improving expansion scale, achieving adult hemoglobin expression, and ensuring safety of any genetically manipulated or reprogrammed cells are critical steps before iPSC-derived blood products become clinically routine.

Methods and data notes (brief)

• Reprogramming of CB CD34+ cells used lentiviral vectors expressing OCT4, SOX2, KLF4 and c-MYC or only OCT4 and SOX2. iPSC lines were expanded to passage 15 or higher before analysis.
• Hematopoietic induction used embryoid bodies and a cytokine cocktail including SCF, TPO, FLT3-L, IL-3, IL-6, VEGF, BMP4 and EPO. Erythroid maturation was performed in liquid culture with staged cytokine support for up to 25 days.
• Key readouts: flow cytometry markers (CD43, CD34, CD45, GPA, CD36, CD71), colony-forming assays, cell counts, microscopy (light and electron), HPLC for hemoglobin composition, and genome-wide promoter CpG methylation profiling.

Selected references and data access

• Full article: Haematologica 2015;100(1):32-41. DOI: 10.3324/haematol.2014.108068. PMCID: PMC4281310.
• Proof-of-principle transfusion of lab-made RBC: Giarratana et al., Blood 2011;118:5071-9. (See ref 7 in the original paper.)
• Papers on epigenetic memory in iPSC: Kim et al., Nature 2010;467:285-90; Polo et al., Nat Biotechnol 2010;28:848-55. (See refs 23, 24 in the original paper.)
• Data from this study are available at GEO accession GSE55109.

Acknowledgments

• The work was supported by multiple foundations and government grants. The authors acknowledge technical support and contributions from collaborating labs. Data are deposited at NCBI GEO under accession GSE55109.