No evidence for αGal epitope transfer from media containing FCS onto human endothelial cells in culture

Current clinical applications of cell therapies and tissue engineered (TE) constructs aim to generate non-immunogenic cells in the best-case scenario of autologous origin. As the cells are cultured, it is theoretically possible that immunoreactive molecules present in xenogenic cell culture media components, such as fetal calf serum (FCS), are transmitted in the culturing process. This problem has propelled the search for xeno-free culture media; however, in vitro culturing of many cell types, especially TE constructs which consist of several cell types, still relies to a great extent on FCS. In this study, we investigated the degree to which xenoantigens are transmitted to human endothelial cells (EC) cultured in medium containing FCS.

Human EC were isolated from pulmonary artery fragments and atrial appendage tissue samples by enzymatic digestion followed by magnetic-activated cell separation (MACS) utilizing CD31 antibodies. The cells were cultured in EGM-2 medium containing 10% FCS for several passages. Griffonia Simplicifolia Lectin I - Isolectin B4 (GSL I-B4) was used to detect cell surface-bound αGal epitopes either microscopically or flow cytometrically. Antibody binding to cells exposed to human sera prepared from healthy blood donors was investigated to detect surface-located xenoantigens. An antibody-dependent cytotoxicity assay was conducted with heat-inactivated human serum supplemented with rabbit complement and analyzed by flow cytometry after staining for living and dead cells (LIVE/DEAD assay kit). In all experiments, cells cultured in EGM-2 supplemented with 10% human serum (HS) served as controls.

Human EC were isolated and cultured successfully for ≥6 passages. GSL I-B4 staining showed no difference between human EC cultured in FCS and in HS. In contrast to porcine EC which showed strong staining with GSL I-B4 and binding of preformed human serum antibodies, human EC cultured in FCS media did not bind human antibodies from high titer anti-αGal and anti-Neu5GC antibody serum. Along these lines, the antibody-dependent cytotoxicity assay showed that human EC cultures independent of FCS or HS usage were not affected, whereas about 40% of porcine EC did not survive.

Despite culturing cells in an environment containing xenoantigens, we were unable to demonstrate the translocation of xenogenic epitopes onto the surface of human EC or find an increased sensitivity in preformed human xenoantibody-dependent complement activity. Therefore, our results suggest that the use of human cells for TE or cell therapy grown in cell culture systems complemented with FCS does not necessarily lead to an acute rejection reaction upon implantation.