Characterization of bone marrow-derived mesenchymal stromal cells (MSC) based on gene expression profiling of functionally defined MSC subsets

Abstract

Background aims: Human mesenchymal stromal cells (MSC) are promising candidates for cell therapy because of their intriguing properties (high proliferation and differentiation capacity, microenvironmental function and immune modulation). However, MSC are heterogeneous and a better understanding of the heterogeneity of the cells that form the MSC cultures is critical.

Methods: Human MSC were generated in standard cultures and stained with carboxyfluorescein succinimidyl ester (CFSE) for cell division tracking. Gene expression profiling of MSC that were sorted based on functional parameters (i.e. proliferation characteristics) was utilized to characterize potential MSC subpopulations (progenitor content and differentiation capacity) and identify potential MSC subpopulation markers.

Results: The majority of MSC had undergone more than two cell divisions (79.7+/-2.0%) after 10 days of culture, whereas 3.5+/-0.9% of MSC had not divided. MSC were then sorted into rapidly dividing cells (RDC) and slowly/non-dividing cells (SDC/NDC). Colony-forming unit-fibroblast (CFU-F) frequencies were lowest in NDC and highest in RDC with low forward-/side-scatter properties (RDC(lolo)). Comparative microarray analysis of NDC versus RDC identified 102 differentially expressed genes. Two of these genes (FMOD and VCAM1) corresponded to cell-surface molecules that enabled the prospective identification of a VCAM1(+)/FMOD(+) MSC subpopulation, which increased with passage and showed very low progenitor activity and limited differentiation potential.

Conclusions: These data clearly demonstrate functional differences within MSC cultures. Furthermore, this study shows that cell sorting based on proliferation characteristics and gene expression profiling can be utilized to identify surface markers for the characterization of MSC subpopulations.