Fig 2: Schematic diagram of the proangiogenic mechanisms of CD44 in MI. CD44 mediates the activation of the proangiogenic FGFR2 signaling pathway in endothelial cells and regulates the CAV1-dependent uptake of plasma exosomes, of which the contents miR-125b-5p and miR-223-3p enhance the FGFR2 signaling pathway to promote angiogenesis post-MI

Fig 3: CD44 is associated with FGFR2 in ECs. A The STRING database was used to predict the protein-protein interaction network between CD44 and FGFR2. Pink edges represent the known protein-protein associations that are experimentally determined. Yellow edges are meant to be the predicated association via text mining. B Representative confocal microscopy images of CD44 (green) and FGFR2 (red) in HUVECs. Nuclei were stained with DAPI (blue). White arrowheads denote the colocalization of CD44 and FGFR2. Scale bars: 10 µm (left panel); 3 µm (right panel). C CD44 and FGFR2 were co-expressed in HUVECs, when the CD44 protein was immunoprecipitated by an anti-CD44 antibody, the indicated proteins, including FGFR2, were detected by immunoblotting. The same was true for the immunoprecipitation of the FGFR2 protein via an anti-FGFR2 antibody. Normal rabbit IgG acted as a negative control. D Representative confocal microscopy images of proximity ligation assay (PLA) signals (green) in HUVECs. CD44 served as the blank control (n = 32), CD44 + Lamin A/C served as the negative control (n = 22), and CD44 + CD9 served as the positive control (n = 14). CD44 + FGFR2 (n = 32). Scale bar: 20 µm. E The statistical data of D. F The FGFR2 distribution and clustering on the HUVEC cell surface were imaged at a nanoscale by N-STORM super-resolution microscopy. Scale bars: 1 µm (upper panel); 40 nm (lower panel). G Ripley’s H functions were acquired by NIS-Elements AR software. H The statistical data of the rmax of H(r) between the siControl and siCD44 groups. n = 8 per group. ***P < 0.001

Fig 4: CD44 removal weakens angiogenic function of plasma exosomes by regulating miRNA-FGFR2 signaling. A Venn diagram of miRNAs regulating FGFR signaling pathway putatively overlapping with highly expressed in KOMI mouse plasma exosomes. B Real-time quantitative polymerase chain reaction (RT-qPCR) analysis of the two candidate miRNAs, miR-125b-5p and miR-223-3p, in border zone of CD44 WT and KO infarcted hearts. Data were normalized to U6 (n = 4–7). C TargetScan database predicated interaction between miR-125b-5p, miR-223-3p and FGFR2 3’-untranslated region (3’-UTR). D, E Western blot analysis of target genes of miR-125b-5p and miR-223-3p in the FGFR2 pathway (n = 3). *P < 0.05, **P < 0.01