Early Erythroid Development Is Enhanced with Hypoxia and Terminal Maturation with Normoxia in a 3D Ex Vivo Physiologic Eythropoiesis Model

Reproduction of dynamic physiologic erythropoiesis in vitro requires a three-dimensional (3D) architecture, erythroblast-macrophage interactions and cytokines such as erythropoietin (EPO). The role of oxygen concentration gradients in this process is unclear. We have created a 3D bone marrow (BM) biomimicry using collagen-coated polyurethane scaffolds (5mm³) to expand cord blood mononuclear cells (CBMNCs) in a cytokine-free environment for 28 days (D). Addition of EPO to this system induces mature erythropoiesis. We hypothesised that physiologic concentrations of cytokines - stem cell factor (SCF) / EPO - and a hypoxia (H)/normoxia (N) schedule to mimic BM oxygen gradients would enhance erythropoiesis. CBMNCs were seeded (4x10⁶ cells/scaffold) in 3D serum-free cultures supplemented with 10ng/mL SCF (D0-D28), and 100mU/mL EPO (D7-D28), with medium exchange every 3D. Three conditions were compared: N (20%), H (5%) and 2-step oxygenation HN (H D0-D7 and N thereafter). Erythroid maturation was monitored weekly by flow cytometry (CD45/CD71/CD235a) both in situ (i.e., in scaffolds) and in supernatant (S/N) cells. D0-7 H was more efficient in early induction of CD235a in the absence of exogenous EPO (H 13% vs N 8% CD45^loCD71⁺CD235a^lo cells, p<0.05). This maturation profile was also observed in D10 S/N cells, in which CD45^loCD71⁺CD235a⁺ cells were proportionately more in H (30%) and HN (27%) than in N (16%, p<0.05). By D14, N and HN stimulated the appearance of CD45^-CD71⁺CD235a⁺ cells, whereas H maintained the CD45^loCD71⁺CD235a^-/lo phenotype. By D21, a CD45^-CD71⁺CD235a⁺ mature population was clearly distinguished in all conditions, most notably in N (16%) and HN (21%) vs H (9%). At D28, more mature CD45^-CD71^loCD235a⁺ cells were observed in normoxia conditions, N 3% and HN 4%, vs H 0.3%. A renewed population of erythroid progenitors was also evident at this time (H 62%, N 51% and HN 46% CD45^loCD71^lo/+CD235a^- cells).

In order to assess the impact of H and N on erythroid gene transcription, we evaluated erythroid signatures by qRT-PCR. GATA-1 expression was detected from D7, highest for H at D14 (p<0.05), and decreased thereafter. GATA-2 expression was up-regulated only at D28, in particular in N (p<0.05), and correlated with emerging erythroid progenitors identified at this stage. At D14, EPOR expression was maximal, especially in HN (p<0.05), simultaneous with high pSTAT5 levels, suggesting activation of EPOR signalling. Also at D14, H upregulated γ-globin (p<0.05). By Western Blot, only H and HN still produced γ-globin whereas β-globin expression was clearly detected in all conditions by D28.

In situ production of cytokines was evaluated by cytometric bead array in the exhausted media. IL-6, G-CSF, GM-CSF, IL-1, TNF-α and IL-17 were detected at higher concentrations during the first 7 days, declining to undetectable thereafter. IL-21 was not detected at any point. IL-3 was detected from D13, with highest expression in H (p<0.05, D22). VEGF was also expressed after D7, highest in H (p<0.05, D16 & D19), concurrent with HIF-1α up-regulation observed at D7 and D14. TNF-α was produced with variable intensity from D4. These data suggested that D7-D14 was a crucial period for culture dynamics, in particular for H and HN, with up-regulation of erythroid transcription factors, EPOR signalling, and endogenous cytokine production. BFU-E and CFU-E also dominated the first 14 days of culture.

Scanning electron microscopy at D17 and D25 revealed niche-like structures in situ, which expressed STRO-1, osteopontin and vimentin at D19 by confocal immunofluorescent microscopy, indicative of an endogenous stromal cell microenvironment. CD68⁺ cells were also detected at D19 in proximity to CD71⁺ cells suggesting formation of erythroblastic islands.

In this 3D ex vivo biomimicry using near-physiologic cytokine and oxygen conditions, H induced initial erythroid commitment and established an early erythroid progenitor population. N was required at later maturational stages and enhanced the γ-globin to β-globin switch. We identified D7-D14 as a crucial timeframe in this system wherein endogenous cytokine production as well as up-regulation of GATA-1, EPOR and HIF-1α was observed. We propose that a combined HN schedule in this 3D BM biomimicy may enable a more robust and physiologic culture platform to study normal and abnormal erythroid differentiation.