Elisa Messina, Mark A. Sussman, Faiz Ramjankhan, Alessandro Giacomello, Elvira Forte, Dries A. M. Feyen, Pieter A. Doevendans, Joost O. Fledderus, Joost P.G. Sluijter, Roberto Gaetani, and Hendrik Gremmels
The discovery and isolation of different resident cardiac progenitor cells (CPCs) a decade ago, as described by several research groups, stimulated the use of these cells for cardiac regeneration. Human CPCs are moving towards the clinic as one of the most promising cell types for cardiac repair, but the extent to which their molecular profiles vary as a result of donor heterogeneity or different isolation methods remain unclear. Defining a common molecular profile that defines CPC’s is therefore an important goal. Similarly, identifying robust and multilaboratory isolation and culture protocols that generate reproducible cell populations from genetically diverse donors is critical for their translational success. In this respect, we collected human auricle biopsy samples anonymously from 20 different adult patients that underwent bypass surgery and generated a total number of 33 different cardiac derived progenitor cell (CPC) lines (Table S1). Human CPCs were isolated according the original published protocol, based on c-kit 1 or Sca-1 2 expression or auricles were cut in 1 mm3 parts and cultured as explants to obtain Cardiospheres (CSps) 3 and Cardiosphere Derived Cells (CDCs) 4. CPCs were subsequently propagated in a panel of different media formulations, either in their originally described culture media or switched to media and culture coatings of the other CPC subsets (Figure S1, Table S7). When comparing individual CPC cell-lines, isolated with different methodologies, they shared a high degree of similarities and correlation in gene expression patterns (Fig. (Fig.1B).1B). By averaging expression profiles of individual CPC conditions, thereby reducing donor variability, similarities increased even more, ranging from 0.92 to 0.96 (Fig. (Fig.1C;1C; Table S2). These results suggest that individual donor differences were larger than influences of isolation and medium conditions. Moreover, the strongest correlations between the different CPC lines were observed when cells were isolated and cultured in the same conditions. Among the different CPCs, spheres-growing CSps showed the least correlation (0.91–0.96), while monolayer-growing CPCs shared higher correlations among them (0.96–0.98). We performed a moderated t-test to evaluate significant differentially expressed genes between the individual samples (Tables S3 and S4). Out of the 13,073 analysd genes, we found only few genes differentially expressed in 5 of 20 different monolayer-cultured CPC cell-lines comparisons. Only when the 3D-cultured CSps were compared with the other CPCs more differently expressed genes could be identified. Although only limited genes were different, we further explored if we could identify differences in gene patterns between the different CPC populations, based on selected genes important for stem cell-maintenance, their growth and biology. In particular, we evaluated genes involved in the regulation of different stem cell pathways like TGF-β, Wnt, NFkB, p53, JAK/STAT, Notch and Hedgehog (Fig. S2A), cell cycle (Fig. S2B), stem cell transcription factors (Fig. S2C), and growth factors, cytokines and chemokines (Fig. S2D). Detailed heat map analysis showed again; however, a very similar profile among all samples, with small differences mainly related to individual donors and not to different cell types or conditions (Fig. S2). Since CSps and monolayer growing CPCs have differently expressed patterns, we selected all the significantly differentially expressed genes that displayed a two fold or more difference and compared them with CDCs, and c-Kit and Sca-1+ CPCs monolayer-cultures (Table S6). Ingenuity pathway analysis identified a gene network in CSps that is enriched in genes encoding for growth factor production and signalling molecules involved in the development of cardiac muscle, vasculogenesis and angiogenesis (Fig. (Fig.2).2). Among them BMP-2, HGF, LIF, PTGS-2, VEGFA and PDGFRB are known to play an important role during cardiac development. Moreover, having a protective effect on a developing heart failure. Figure 1 Experimental design of the project (A) and hierarchical clustering of CPCs samples (B and C). Sca-1+ cells were isolated from human auricle biopsy and cultured in gelatin coated flask and Sca-1 medium (Sca GEL S-MED) (2) (d). After expansion cells were ... Figure 2 Ingenuity molecular networks analysis of the differentially expressed genes. Fold difference ≥2; p