Fig. 6

Mouse epicardial adipocytes do not share the age-related metabolic remodeling trait of human eAT (A) Schematic of RNA-seq experimental setup for young and aged mouse eAT. Arrowheads indicate the ventricular area covered by eAT. Scale bar, 1 mm. (B) Heatmap of the top 30 DEGs in young compared with aged mouse eAT. (C) PCA score plot showing clustering of adipocyte populations from zebrafish, mouse, and human based on the ranked expression of one-to-one orthologs. (D) Correlation matrix of PCs with known variables. The indicated values and color gradient correspond to the Pearson correlation coefficient (r) for each pair of variables. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001. (E) Clustered heatmap depicting the row-normalized ranking scores of all DEGs shared by human and zebrafish eAT compared with white adipose tissue in zebrafish, human, and mouse samples. Green boxes indicate clusters where the transcription profile is most similar between human and zebrafish and yellow boxes between human and mouse. (F) Pathway enrichment analysis of gene clusters with similar expression in human and zebrafish (cluster 3) and similar expression in human and mouse (cluster 5). All pathway analyses were done using the KEGG and Reactome databases. Enrichment is expressed as the −log[P] adjusted for multiple comparisons.