Isolation of a viable, highly purified cardiac fibroblast fraction devoid of epicardial cells from the infarcted murine heart

Abstract

Cardiac fibroblasts (CF) migrate and differentiate into myofibroblasts (activated cardiac fibroblasts ‐ aCF) in response to myocardial infarction (MI), thereby promoting structural and functional repair. However, the CF population is believed to be heterogeneous and difficult to capture by conventionally used markers. An additional problem concerns the potential contamination in CF preparations by endothelial and immune cells. Furthermore, contamination with epicardial mesenchymal progenitor cells (epicardium‐derived cells ‐ EPDC), formed on the epicardial surface after MI, is another critically factor, which so far has not been evaluated.We have established a novel CF isolation protocol using enzymatic digestion of the murine heart via the coronaries (Langendorff) and the simultaneous enzymatic removal of the EPDC layer from the outside in a single procedure (within 12 min). Subsequently, the cell suspension derived from the EPDC‐depleted heart is separated from cardiomyocytes by centrifugation and endothelial cells (CD31+) as well as immune cells (CD45+) are removed by magnetic bead purification. Analysis of cellular composition of the non‐cardiomyocyte fraction in the control heart by flow cytometry revealed that endothelial cells (CD31+) are the most abundant cell type (45.3 ± 1.4%) within the non‐cardiomyocyte fraction, followed by CF (CD31−/CD45−) and immune cells (CD45+) accounting for 40.4 ± 1.9% and 12.1 ± 0.3%, respectively. Purified CF (CD31−/CD45−) were further analyzed for common CF markers: Pdgfrα and the MEFSK4 epitope was expressed on the majority of isolated CF (72.2 ± 2.9% and 52.1 ± 4.6%, respectively), while CD90+, Sca1+ and Vimentin+ cells amounted to 23.3 ± 2%, 12.4 ± 0.7% and 35.5 ± 2.1%, respectively, suggesting a heterogeneous CF population. Expression of non‐fibroblast markers revealed that the fraction of pericytes (Ng‐2: 1.1 ± 0.3 %) and smooth muscle expressing cells (α‐SMA: 12.3 ± 0.5%) was rather small. In addition we studied the expression of 23 genes encoding for activation markers (a‐Sma, Tnc, Postn, Col1a1, Col3a1, Fn), cytokines and growth factors (Il‐6, Il‐1b, Il‐10, Vegfa, Vegfc, Tgfb1, Fgf1) as well as cardiac genes and transcription factors (Wt‐1, Tbx‐18, Mef2c, Nkx2.5, Gata‐4, Tnnt2, Hand2, Tbx‐5, Tbx‐20, Tcf‐21) between CF of the unstressed heart, aCF and EPDC isolated 5 days after MI (n=4). Analysis revealed that expression of WT‐1, an established epicardial marker, was 25‐fold/86‐fold higher in EPDC compared to aCF/CF, respectively. Expression of activation markers in aCF and EPDC were in a similar range and higher as compared to quiescent CF. Cardiac genes and transcription factors were expressed in CF, aCF and EPDC, however, with large individual differences: troponin T was 21‐fold higher expressed in CF/aCF when compared to EPDC, while Tbx‐5 was 4‐fold higher in EPDC when compared to CF/aCF. These profound differences between EPDC and aCF/CF highlight that the removal of EPDC is an essential prerequisite for studies of the CF phenotype after MI.In summary, we report a rapid isolation technique for cardiac fibroblasts pre and post MI, which are largely free of contaminating immune cell, endothelial cells, pericytes and EPDC. This protocol will permit the separate analysis of infarct‐induced phenotype changes in freshly isolated as well as cultured CF and EPDC.Support or Funding InformationFunded by DFG (IRTG1902)This abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.