Media matters: culture medium-dependent hypervariable phenotype of mesenchymal stromal cells

Abstract

Abstract Background Despite a long history of investigation and sustained efforts in clinical testing, the number of market authorisations for mesenchymal stromal cell (MSC) therapies remains limited, with none approved by the United States Food and Drug Administration. Several barriers are impeding the clinical progression of MSC therapies, to the forefront of these is a lack of standardised manufacturing protocols which is further compounded by an absence of biologically meaningful characterisation and release assays. A look at clinical trial registries demonstrates the diversity of MSC expansion protocols with variabilities in cell source, isolation method and expansion medium, among other culture variables, making it extraordinarily difficult to compare study outcomes. Current identification and characterisation standards are insufficient; they are not specific to MSCs and do not indicate cell function or therapeutic action. Methods This work analysed the influence of five widely used culture media formulations on the colony-forming potential, proliferation kinetics, trilineage differentiation potential and immunomodulatory potential of human bone marrow-derived MSCs (BM-MSCs). The surface marker expression profiles were also characterised using a high-content flow cytometry screening panel of 243 markers. Results Significant differences in the biological attributes of BM-MSCs including clonogenicity, proliferation, differentiation propensity and immunomodulatory capacity were revealed in response to the composition of the culture medium. Despite their biological differences, all cell preparations uniformly and strongly expressed the standard positive markers proposed for BM-MSCs: CD73, CD90 and CD105. Immunophenotypic profiling revealed that the culture medium also had a significant influence on the surface proteome, with one-third of tested markers exhibiting variable expression profiles. Principal component analysis demonstrated that BM-MSCs isolated and expanded in a proprietary xeno- and serum-free medium displayed the most consistent cell phenotypes with little variability between donors compared to platelet lysate and foetal bovine serum-containing media. Conclusions These data suggest that media composition has a highly significant impact on the biological attributes of MSCs, but standard surface marker tests conceal these differences. The results indicate a need for (1) standardised approaches to manufacturing, with an essential focus on defined media and (2) new biologically relevant tests for MSC characterisation and product release.