Analyses of erythropoiesis from embryonic stem cell-CD34+ and cord blood-CD34+ cells reveal mechanisms for defective expansion and enucleation of embryomic stem cell-erythroid cells

Abstract

Red blood cells (RBCs) generated ex vivo have the potential to be used for transfusion. Human embryonic stem cells (ES) and induced pluripotent stem cells (iPS) possess unlimited self-renewal capacity and are the preferred cell sources to be used for ex vivo RBC generation. However, their applications are hindered by the facts that the expansion of ES/iPS-derived erythroid cells is limited and the enucleation of ES/iPS-derived erythroblasts is low compared to that derived from cord blood (CB) or peripheral blood (PB). To address this, we sought to investigate the underlying mechanisms by comparing the in vitro erythropoiesis profiles of CB CD34⁺ and ES CD34⁺ cells. We found that the limited expansion of ES CD34⁺ cell-derived erythroid cells was associated with defective cell cycle of erythroid progenitors. In exploring the cellular and molecular mechanisms for the impaired enucleation of ES CD34⁺ cell-derived orthochromatic erythroblasts (ES-ortho), we found the chromatin of ES-ortho was less condensed than that of CB CD34⁺ cell-derived orthochromatic erythroblasts (CB-ortho). At the molecular level, both RNA-seq and ATAC-seq analyses revealed that pathways involved in chromatin modification were down-regulated in ES-ortho. Additionally, the expression levels of molecules known to play important role in chromatin condensation or/and enucleation were significantly lower in ES-ortho compared to that in CB-ortho. Together, our findings have uncovered mechanisms for the limited expansion and impaired enucleation of ES CD34⁺ cell-derived erythroid cells and may help to improve ex vivo RBC production from stem cells.

Keywords: ATAC-Seq; RNA-Seq; enucleation; erythropoiesis.

FIGURE 1

Proliferation potential of erythroid cells derived from CB CD34⁺ or ES CD34⁺ cells. (A) Growth curve of erythroid cells derived CB CD34⁺ or and ES CD34⁺ cells in 3 phase erythroid culture medium (N = 6). (B) Representative flow cytometry profiles of apoptosis as assessed by Annexin V and 7AAD staining of Day 7 erythroid cells. (C) Quantitative analyses of apoptosis. N = 3. (D) Representative flow cytometry profiles of cell cycle as assessed by Edu and 7AAD staining of Day 7 erythroid cells. (E) Quantitative analyses of cell cycle. N = 3. (F) Colony forming ability of Day 4 erythroid cells derived from CB CD34⁺ or ES CD34⁺ cells in BFU‐E colony medium or CFU‐E colony medium. Scale bar, 200 μm. (G) mRNA levels (normalized to actin) of cyclin E, CDK2, CDK4 and p57, as assessed by qRT‐PCR. (H) Representative Western blot analysis of cyclin E, CDK2, CDK4 and p57 (left panel) and quantitative analysis of relative protein expression levels from three independent experiments (right panel). *p < 0.05, **p < 0.01, ***p < 0.001

FIGURE 2

Terminal erythroid differentiation of CB CD34⁺ or ES CD34⁺ cells. (A) Flow cytometry analysis showing the percentage of GPA⁺ cells of Day 9 erythroid cells. (B) Quantitative analysis showing the mean fluorescent intensity (MFI) of GPA. (C) Representative flow cytometry analysis of band3 and α4‐integrin expression on GPA⁺ erythroblasts. (D) Quantitative analyses of erythroblasts at distinct developmental stages based on the expression of band3 and α4‐integrin shown in Figure 2C. N = 3. *p < 0.05, **p < 0.01

FIGURE 3

Impaired enucleation and chromatin condensation of orthochromatic erythroblasts derived from ES CD34⁺ cells. (A) Representative flow cytometry analyses of enucleation as assessed by Syto16 staining. (B) Quantitative analyses of enucleation. (C) Representative cytospin images of erythroblasts. Scale bar, 20 μm. (D) Representative cytospin images of polychromatic and orthochromatic erythroblasts. Scale bar, 10 μm. (E) Quantitative analyses of nuclear area (μm²) of poly and ortho erythroblasts by ImageJ. (F) Representative ImageStream analysis of band 3 and α4‐integrin expression on Day 15 of GPA⁺ erythroblasts. (G) Representative ImageStream images of poly and ortho erythroblasts. (H) Quantitative analyses of nuclear area (μm²) of poly and ortho erythroblasts by ImageStream. N = 3. ***p < 0.001

FIGURE 4

RNA‐seq analyses of CB and ES orthochromatic erythroblasts. (A) Principal component analyses. (B) Heat map of differentially expressed genes. (C) Bar plot of differentially expressed gene numbers. (D) Top 10 down‐regulated pathways in ES‐ortho. (E) Top 10 up‐regulated pathways in ES‐ortho. The colour represents log‐transformed adjusted q‐value; the width indicates the number of differentially expressed genes in the category

FIGURE 5

ATAC‐seq analyses of CB‐ortho and ES‐ortho. (A) Principal component analyses. (B) Heat map of differential chromatin accessibility. (C) Numbers of increased and decreased accessible peaks. (D) The distribution of peaks relative to gene features for ES‐ortho increased and decreased accessible peaks. (E) Enriched GO terms of decreased accessible peaks in ES‐ortho. (F) Enriched GO terms of increased accessible peaks in ES‐ortho. The colour represents adjusted binomial q‐value, and the bar width indicates the number of differentially accessible peaks in the category

FIGURE 6

Correlation of open chromatin and gene expression. (A) The log₂(fold change) of differentially accessible peaks (DAPs) plotted against the log₂(fold change) of differentially expressed genes (DEGs) with DAPs at promoter regions. (B) The log₂(fold change) of DAPs plotted against the log₂(fold change) of DEGs with DAPs at non‐promoter regions. (C) Up‐regulated and down‐regulated differentially expressed gene numbers with or without DAPs at promoter regions. (D) Bar plot of enriched GO terms of down‐regulated DEGs with DAPs at promoter regions. (E) Bar plot of enriched GO terms of up‐regulated DEGs with DAPs at promoter regions. The colour represents adjusted binomial q‐value, and the bar width indicates the number of differentially accessible peaks in the category

FIGURE 7

Expression of molecules known to be involved in enucleation. (A) The normalized count value of differentially expressed genes (DEGs) in ES‐ortho and CB‐ortho from RNA‐seq as indicated. (B) Western blot analysis showing protein level differences between ES‐ortho and CB‐ortho. Quantitative analysis of protein expression levels from three independent experiments is shown (bottom panel). (C) The normalized count value of differentially accessible peaks (DAPs) in ES‐ortho and CB‐ortho from ATAC‐seq as indicated. (D) The reads distribution of differentially accessible peaks at promoter regions of genes as indicated. *p < 0.05, **p < 0.01, ***p < 0.001