Proof of principle for transfusion of in vitro-generated red blood cells

Abstract

In vitro RBC production from stem cells could represent an alternative to classic transfusion products. Until now the clinical feasibility of this concept has not been demonstrated. We addressed the question of the capacity of cultured RBCs (cRBCs) to survive in humans. By using a culture protocol permitting erythroid differentiation from peripheral CD34(+) HSC, we generated a homogeneous population of cRBC functional in terms of their deformability, enzyme content, capacity of their hemoglobin to fix/release oxygen, and expression of blood group antigens. We then demonstrated in the nonobese diabetes/severe combined immunodeficiency mouse that cRBC encountered in vivo the conditions necessary for their complete maturation. These data provided the rationale for injecting into one human a homogeneous sample of 10(10) cRBCs generated under good manufacturing practice conditions and labeled with (51)Cr. The level of these cells in the circulation 26 days after injection was between 41% and 63%, which compares favorably with the reported half-life of 28 ± 2 days for native RBCs. Their survival in vivo testifies globally to their quality and functionality. These data establish the proof of principle for transfusion of in vitro-generated RBCs and path the way toward new developments in transfusion medicine. This study is registered at http://www.clinicaltrials.gov as NCT0929266.

Trial registration: ClinicalTrials.gov NCT00929266 NCT00929266.

Figure 1

Expansion and differentiation of cRBCs. (A) Amplification of human CD34⁺ cells obtained by G-CSF–mobilized LK. Results are expressed as the mean ± SD (n = 9). (B) NMB staining from cRBC and (C) from native reticulocytes (original magnification ×50). (D) Flow cytometric analysis of one representative experiment from 9 independent experiences: expression of glycophorin A-PE, CD71-PE, and CD36-FITC from cRBC and (E) from native reticulocytes. Solid histograms represent relevant mAbs and open ones negative controls with irrelevant mAbs.

Figure 2

Deformability of cRBCs studied by LORCA. Deformability of cRBCs was studied by LORCA. The deformability profile of day 18 cRBCs (n = 2, red curves) was compared with those of (1) mature peripheral RBCs from healthy donors (native RBC) as the control (n = 10, black circles) and (2) native reticulocytes (n = 3, gray crosses). The EI at 3 and 30 Pa is given in the table for each group and results are expressed as the mean ± SEM.

Figure 3

Functionality of cRBC hemoglobin. (A) Hemoglobin status of cRBCs determined by HPLC (Bio-Rad Variant II). The percentage of hemoglobin in the elution peak is indicated for the Hb0, HbF, HbA1c, and HbA2 fractions. One representative graph from 9 independent experiments is shown. (B) Tonometric oxygen binding curves at 37°C for a cRBC (triangles) and a control RBC suspension (dotted line) at different DPG/Hb₄ ratios in 10mM HEPES buffer (pH 7.4) containing 150mM NaCl. The RBC isotherms were simulated from the average MWC parameters for 10 different blood samples (black line).

Figure 4

In vivo maturation of cRBCs in the NOD/SCID mouse model. (A) Kinetics of CD71⁺ cells among CFSE⁺ cells in mouse blood after injection of purified CFSE⁺ cRBCs. Results are expressed as the mean ± SEM (n = 4). (B) CD71 expression on CFSE⁺ cells in 1 representative experiment. Quadrant statistics are given in each dot plot. The percentage of CFSE⁺ cells on days 1, 2, 3, and 5 was 2.65%, 19.2%, 20.5%, and 1.2%, respectively. (C) Kinetics of LDS expression on CFSE⁺ cells in 1 experiment. (D) Confocal microscopy images of CFSE⁺ cells before (top left) and after injection into mice (bottom) compared with native RBCs (top middle) and native reticulocytes (top right). On the bottom are front, profile, and side views of cells recovered 3 days after their injection. The cell diameter before and after 3 days injection was 11 μm and 7.5 μm, respectively. Magnification ×500. On day 3, sorted CFSE⁺ cells were colabeled with an anti-RhD antibody (solid histogram) or its isotype control (open histogram).

Figure 5

Long-term storage of cRBCs. Purified cRBCs and native reticulocytes were stored at 4°C for up to 4 weeks in a Sag-Mannitol–based preservative solution. (A) Comparative kinetics of cell recovery during long-term storage of cRBCs and native reticulocytes (native Ret). Mean ± SEM of 9 and 6 experiments, respectively. (B) Evolution of the hemoglobin content during storage of cRBCs. Mean ± SEM and the number of experiments is indicated on each bar. (C-D) CD71 expression, reticulocyte content (C) and deformability (D) were evaluated before (fresh cRBCs) and after 4 weeks of storage (stored cRBC). Mean of 2 experiments. (E) Comparative kinetics of CD71⁺ cells among CFSE⁺ cells in mouse blood after injection of fresh (n = 4 mice) or stored (n = 2 mice) purified cRBC. Data for fresh cRBC are from Figure 4A, and the results are expressed as the mean ± SEM. Confocal microscopy image shows CFSE-labeled stored cells 3 days after injection into mice.