Hematopoietic development occurs in defined niches that ensure specific interactions and cross-talk with the surrounding stromal cells and different hematopoietic cells themselves. For instance, erythropoiesis occurs on the macrophage island within the bone marrow and the central macrophage is believed to regulate pro-erythroblast differentiation, the final stages of enucleation and reticulocyte maturation. We have observed that the expansion of erythroblasts from total peripheral blood mononuclear cells is increased compared to CD34+ Hematopoietic Stem/Progenitor Cells (HS(P)C) isolated from the same amount of blood [van den Akker, Haematologica, 2010]. This suggests i) the presence of CD34-cells that contribute to erythropoiesis and/or ii) that cell-cell contact or specific secreted growth factors by “helper” cells in these cultures can regulate hematopoiesis/erythropoiesis to increase erythroblast yield. Identifying the specific population(s) underlying the increased erythroid yield and understanding their way of action and regulatory mechanism during HSC differentiation and erythropoiesis is not only important to improve erythroblast culture conditions but may also provide clues to the function of hematopoietic effector cells in the various HS(P)C/erythroblast niches.

Using specific lineage depletion (among which CD3 and CD14) we have identified and quantified various human erythroid and non-erythroid CD34+ and CD34- populations on the basis of CD36 co-expression in peripheral blood mononuclear cells (PBMC). Erythroid outgrowth from these CD34- populations and CD34+ populations and their contribution to the total erythroid yield from PBMC was assessed. Interestingly, total erythroid yield from the individual sorted populations did not reach the erythroid yield obtained from total PBMC. We hypothesized that support/feeder cells present in total PBMC are positively influencing in vitro erythropoiesis. In agreement with this, PBMC immuno-depletion of specific hematopoietic cell types identified CD14 cells (monocyte/macrophages) and to a lesser extend CD3 cells (lymphocytes) to be also partly responsible for the increased erythroblast yield. Compared to HS(P)C alone, co-culture of CD14 cells and HS(P)C isolated from PBMC resulted in a 5-10 times increase in CD71high/CD235med erythroblasts. Conditioned medium of CD14 cells as well as transwell experiments reconstituted the effect of the HS(P)C-CD14 co-cultures to 70%-80%, indicating that cell-cell contact plays a minor role. CD14 cells could elicit their effect at different stages during HSPC/HSC differentiation to erythroblasts. Co-culture of CD14 cells with pro-erythroblasts did not increase the cellular yield or proliferation rate. In contrast, two days of CD14 co-culture with CD34+ cells results in a 5 fold increase of total colony forming units without altering the colony lineage dynamics. In agreement with this a 5 fold increase in CD34+ cells was observed. These results indicate that CD14 cells elicit their effect on early hematopoietic progenitors but not on the erythroblast population. The results predict that depletion of CD14+ cells from PBMC should result in a decrease in the total number of CD34+cells. Indeed, we observed a 2 fold decrease of specifically HS(P)Cs and MEPs after two days of culture in PBMCs depleted for CD14 cells.

Taken together our data i) identify previously unrecognized erythroid and non erythroid CD34- and CD34+ populations in peripheral blood that contribute to erythroid yield from total PBMC and ii) indicate modulation of HS(P)C outgrowth by specific hematopioietic effector cells present in peripheral blood that can also be found near specific hematopoietic niches in the bone marrow. The involvement of CD3 and CD14 immune cells suggests that HS(P)C and erythropoiesis may be modulated by immune-responses.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.