Production and stability of cultured red blood cells depends on the concentration of cholesterol in culture medium

Abstract

The production of cultured red blood cells (cRBC) for transfusion purposes requires large scale cultures and downstream processes to purify enucleated cRBC. The membrane composition, and cholesterol content in particular, are important during proliferation of (pro)erythroblasts and for cRBC quality. Therefore, we tested the requirement for cholesterol in the culture medium during expansion and differentiation of erythroid cultures with respect to proliferation, enucleation and purification by filtration. The low cholesterol level (22 µg/dl) in serum free medium was sufficient to expand (pro)erythroblast cultures. Addition of 2.0 or 5.0 mg/dL of free cholesterol at the start of differentiation induction inhibited enucleation compared to the default condition containing 3.3 mg/dl total cholesterol derived from the addition of Omniplasma to serum free medium. Addition of 5.0 mg/dl cholesterol at day 5 of differentiation did not affect the enucleation process but significantly increased recovery of enucleated cRBC following filtration over leukodepletion filters. The addition of cholesterol at day 5 increased the osmotic resistance of cRBC. In conclusion, cholesterol supplementation after the onset of enucleation improved the robustness of cRBC and increased the yield of enucleated cRBC in the purification process.

Figure 1

Cholesterol supplementation does not affect expansion of erythroblast cultures. Erythroblasts were cultured from PBMC in presence of Epo (1 U/ml), SCF and dexamethasone. Cholesterol was added on day 7 at 2.0 or 5.0 mg/dl. (A) Cells were counted at indicated time intervals and cell density was maintained between 0.5 × 10⁶/ml and 1.5 × 10⁶/ml through dilution with fresh medium (plus or minus cholesterol). Cumulative cell numbers were calculated (n = 3). (B) Expression of transferrin receptor (CD71) and Glycophorin A (CD235a) in presence and absence of cholesterol was evaluated by flow cytometry on day 11. Percentage of cells in each quadrant is shown: Q1 (stage 1) CD71^neg/CD235a^neg ; Q2 (stage 2) CD71^pos/CD235a^neg ; Q3 (stage 3) CD71^pos/CD235a^pos ; Q4 (stage 4) CD71^neg/CD235a^pos. Nine independent experiments; error bars indicate standard deviation.

Figure 2

Cholesterol supplementation does not affect erythroblast differentiation. (A–C) Day 11 erythroblast cultures (Fig. 1B) were transferred to terminal differentiation conditions (plasma, heparin, Epo (5 U/ml)) for another 11 days with or without supplementation of additional cholesterol (2 or 5 mg/dl) at the start of differentiation (t = 0) or at day 5 of differentiation induction. Cell count (A) and cell volume (fl, femtoliter) (B) were measured on a CASY cell counter. The cell count of distinct cultures was normalized to the cell count of cultures without cholesterol supplementation. Hemoglobin was measured with a colorimetric reaction as arbitrary units (a.u.) and corrected for the cell volume (C). n = 3, Error bars indicate standard deviation, *p < 0.05.

Figure 3

Cholesterol supplementation increases the yield of enucleated cRBC following filtration over a leukodepletion filter. Enucleated cells in cRBC cultures differentiated for 11 days in absence or presence of cholesterol (2 or 5 mg/dl; see Fig. 2) were analyzed by DRAQ5 staining and flow cytometry. (A, B) Enucleation efficiency was measured as the percentage of FSC-high DRAQ5 negative cells versus all FSC-high cells (supplemental Fig. 2A). Percentage enucleation pre-filtration (A) and post filtration (B). (C, D) The recovery of enucleated cRBC was calculated as percentage of the number of total cells prefiltration (C) and as percentage of the enucleated fraction prefiltration (D). n = 5, *p < 0.05, **p < 0.01. (E, F) Cholesterol (5 mg/dl) was added at day 5 and day 10 to differentiating erythroblast cultures. Enucleation (E) and Recovery from enucleated fraction (F) was measured similar to panels (A) and (D), respectively (n = 2).

Figure 4

Cholesterol supplementation increases deformability and isotonic stability of cRBC. (A–C) Enucleated cells in cRBC cultures differentiated for 11 days in absence or presence of cholesterol (2 or 5 mg/dl, added at the start (d0) or at day 5 or day 10 after induction of differentiation; see Fig. 2) were purified. (A, B) Analysis by automated rheoscope and cell analyzer (ARCA) at a shear stress of 3 Pa. The elongation under shear stress was compared to freshly isolated erythrocytes and reticulocytes from peripheral blood. (C, D) Hemolysis assay at NaCl concentrations of 0.5% w/v (B) and 0.6% w/v (C). n = 6, error bars indicate standard deviation.

Figure 5

LDL-receptor expression during erythroblast differentiation. Erythroblast cultures in expansion medium (day 11 from seeding PBMC; left panel) and erythroblasts subsequently differentiated for 3 days (right panel) were analyzed by flow cytometry for LDL-receptor expression. (A–C) Gating strategy: (A) selected live cells based on forward scatter (FSC) versus side scatter (SSC). (B) Within live cells the selection of cells expressing Glycophorin A (CD235a) or integrin subunit α4 (CD49d). (C) On this erythroblast selection FSC versus CD44 were visualized and divided in 6 subpopulations (P1 to P6) with increasing maturity (CD44 decreases from proerythroblast (P1) to reticulocyte (P6). (D) Histogram overlay of Mean Fluorescence Intensity (MFI) of LDL Receptor during developmental stages (P1 through P6). (E) MFI of LDL-R expression in P1–P6 as in (D) for cultures without and with supplementation of cholesterol (2 and 5 mg/dl) Error bars indicate standard deviation (n = 3).

Figure 6

Rapid time and concentration dependent uptake of cholesterol in differentiating cRBC cultures. Cultures were exposed to differentiation conditions in presence of 5 mg/dl cholesterol for 5 days. Bodipy-labeled cholesterol was added at 0.5, 2.0 and 5.0 µM for increasing incubation times 0, 2, 4, 8, 16, and 24 h). (B) 5 days differentiated cRBC cultures were incubated overnight in presence of Cholesterol 2 mg/dL plus Bodipy-Cholesterol 2 µM. After overnight incubation, cells were washed and resuspended in medium containing Cholesterol 2 mg/dL (without Bodipy). Bodipy-Cholesterol Wash-Out Samples were taken at 0, 2, 4, 16, 24 h. Enucleated cRBC, Nucleated cRBC, Pyrenocytes distinguished on the basis of DRAQ5 and Forward Scatter.