Erythropoietin modulation is associated with improved homing and engraftment after umbilical cord blood transplantation

Abstract

Umbilical cord blood (UCB) engraftment is in part limited by graft cell dose, generally one log less than that of bone marrow (BM)/peripheral blood (PB) cell grafts. Strategies toward increasing hematopoietic stem/progenitor cell (HSPC) homing to BM have been assessed to improve UCB engraftment. Despite recent progress, a complete understanding of how HSPC homing and engraftment are regulated is still elusive. We provide evidence that blocking erythropoietin (EPO)-EPO receptor (R) signaling promotes homing to BM and early engraftment of UCB CD34+ cells. A significant population of UCB CD34+ HSPC expresses cell surface EPOR. Exposure of UCB CD34+ HSPC to EPO inhibits their migration and enhances erythroid differentiation. This migratory inhibitory effect was reversed by depleting EPOR expression on HSPC. Moreover, systemic reduction in EPO levels by hyperbaric oxygen (HBO) used in a preclinical mouse model and in a pilot clinical trial promoted homing of transplanted UCB CD34+ HSPC to BM. Such a systemic reduction of EPO in the host enhanced myeloid differentiation and improved BM homing of UCB CD34+ cells, an effect that was overcome with exogenous EPO administration. Of clinical relevance, HBO therapy before human UCB transplantation was well-tolerated and resulted in transient reduction in EPO with encouraging engraftment rates and kinetics. Our studies indicate that systemic reduction of EPO levels in the host or blocking EPO-EPOR signaling may be an effective strategy to improve BM homing and engraftment after allogeneic UCB transplantation. This clinical trial was registered at www.ClinicalTrials.gov (#NCT02099266).

© 2016 by The American Society of Hematology.

Figures

Figure 1.

EPOR is expressed on a subset of UCB CD34+cells and is enriched in the CD34+CD38–population. (A) Representative flow cytometry data showing EPOR and CD34 expression on UCB CD34+-enriched cells with gating strategy. (B) The percentage of CD34+, EPOR+, and CD34+EPOR+ expression on enriched UCB cells from 5 units with high purity (90% or higher CD34+ cells) by flow cytometry. (C) EPOR expression in enriched cells from 3 UCB units by western blot. (D) Representative flow cytometry data showing EPOR expression on UCB CD34+ cell subsets CD34+CD38– and CD34+CD38–. (E) EPOR expression by cell percentage (left) and MFI (right) on UCB CD34+ subsets. (F) Verification of EPOR shRNA transfection efficiency of UCB CD34+ cells by determining the percentage of GFP+ cells. (G) Micrograph showing GFP-expressing UCB CD34+ cells post lentiviral infection, confirming expression of EPOR shRNAs. Fluorescent images were taken at room temperature using an Olympus IX71 inverted microscope and Olympus DP71 camera. DP controller was used for software acquisition and Adobe Photoshop was used for image processing (scale bar represents 50 µm). (H) mRNA in EPOR knocked down UCB CD34+ relative to control confirming depletion of EPOR by RNAi. (I) Micrographs show formation of hematopoietic cell colonies. Erythroid colonies are shown on the right side with EPOR-depleted UCB CD34+ colonies shown in the right lower corner. Bright field images were taken at room temperature using an Olympus IX71 inverted microscope, 10x objective lens, and Olympus DP71 camera. DP controller was used for software acquisition and Adobe Photoshop was used for image processing (scale bar represents 200 µm). (J) Quantitation of CFU-G/M vs BFU-E colonies with or without EPOR-depletion. Single (*) and double (**) asterisks indicate statistical significance. CFU-G/M, colony-forming units–granulocytes and macrophages.

Figure 2.

EPO exposure impedes stromal-derived factor-1–induced migration of UCB CD34+cells. (A)The percentage of freshly enriched UCB CD34+ cell subsets CD34+CD38– and CD34+CD38– that migrated toward SDF-1 gradient after culture in erythropoietin at 100 and 200 ng/mL concentrations. (B) UCB CD34+CD38– and CD34+CD38+ populations cell migration toward SDF-1 after incubation with neutralizing EPO (αEPO) and EPOR (αEPOR) antibodies and culture in erythropoietin at 200 ng/mL concentration. (C) The percentage of EPOR-depleted UCB CD34+ cells migrated toward SDF-1 in the presence of erythropoietin at 200 ng/mL concentration relative to control (nontransfected UCB CD34+ cells) normalized to 100%. Two separate UCB units were examined with 3 experimental replicates per cord. The data are presented as mean ± SEM. Single (*), double (**), and triple (***) asterisks indicate statistical significance when compared with untreated. Single (#) and double (##) hatchmarks indicate statistical significance when compared with isotype control.

Figure 3.

Transient reduction of systemic EPO levels in hosts using HBO increases BM homing of UCB CD34+cells. (A) Serum EPO levels in pg/mL in HBO-treated mice (n = 6) and in non-HBO mice (n = 6) 3 hours after UCB CD34+ cell infusion. (B) Percent of human CD34+ cells in BM and spleen of HBO mice (n = 8) and in non-HBO mice (n = 8) 3 hours after UCB CD34+ cell infusion. (C) EPO rescue effects on UCB CD34+EPOR+ cell homing and engraftment in HBO-treated mice (n = 8). (C, left) Representative flow cytometry data showing dual expression of human CD34+ and EPOR+ in EPO-treated (n = 4) and control mice (n = 4) BM-gated populations. (C, right) Bar graph showing the relative percentage of human CD34+ (hCD34+) and separately human CD34+EPOR+ (hCD34+EPOR+) cells in BM of HBO-treated mice with either EPO treatment (+EPO) or without EPO treatment (–EPO) normalized to 100%. The data are presented as mean ± SEM. Single (*) and triple (***) asterisks indicate statistical significance.

Figure 4.

Low EPO environment enhances myeloid differentiation of UCB CD34+cells in vitro and in vivo. (A) Experimental design of in vitro experiment in which human UCB CD34+ cells migrate toward SDF-1 gradient in the presence of murine BM. The migrated human UCB and murine BM cells are collected and plated for CFU assay. (B) Number of migrated human UCB CD34+ cells in the presence or absence of EPO. (C) Images of types of CFU formed after plating migrated human UCB CD34+ cells and murine BM. Bright field images were taken at room temperature using an Olympus IX71 inverted microscope and Olympus DP71 camera. DP controller was used for software acquisition and Adobe Photoshop was used for image processing (scale bar represents 200 µm). (D) Genomic DNA polymerase chain reaction of 3 types of colonies appeared in methylcellulose medium confirming their human origin. (E) Relative percentage of total colonies formed in the presence of EPO compared with control conditions in the absence of EPO normalized to 100%. (F) Relative percentage of CFU-G/M and BFU-E colonies formed in the presence of EPO compared with control conditions in the absence of EPO normalized to 100%. (G) HBO, as an inducer of low EPO environment, effects on differentiation of transplanted UCB CD34+ cells. BM-isolated mononuclear cells from HBO (n = 3, done in duplicate) and non-HBO mice (n = 3, done in duplicate) 1 week after UCB CD34+ cell infusion were plated and counted. Ratio of BFU-E to total CFU and CFU-G/M to total CFU and in HBO and control mice. The asterisk (*) indicates statistical significance.

Figure 5.

Day +100 Kaplan-Meier survival curves of HBO-treated patients (blue) compared with historic controls (red).

Figure 6.

Recipient pretreatment with HBO significantly, but transiently, reduces EPO serum levels at the time of UCB transplantation. (A) Serum EPO levels in mU/mL from individual patients who received myeloablative (blue line) and reduced-intensity preparative regimen (red line) plotted over time (in hours) from the start of HBO therapy. (B) The median serum EPO levels (mU/mL) in myeloablative (blue line) and reduced-intensity preparative regimen (red line) are plotted over time (in hours) from the start of HBO therapy. Baseline value (0-hour) was drawn just before HBO therapy. Blood levels were also drawn 6 hours, 8 hours, 24 hours, and 48 hours from the start of HBO.