Heterogeneity within cell populations allows the heart to function as an integrated unit with distinct tasks allocated to sub-specialized cells. During aging, some cells might acquire an overt disordered state, while others still remain normal or are caught in an intermediary state. However, the significance of cellular plasticity within the heart remains unclear. Our current understanding of cellular states mostly originates from population experiments, in which the molecular biological status of individual cells is lost. Thus, little is known about the true relationships between subsets of normal or deregulated cell populations. Breaking down superficially homogenous cell populations of the heart into subtypes of cells that together determine cardiovascular function enables us to understand better how this complex organ operates in physiological and pathological conditions. To shed light on the heterogeneity and intercommunication of cardiac cells in the adult and aged mouse heart, we applied single-nucleus RNA sequencing on young (12 weeks) and aged (18 months) mouse hearts. By including 3 mice per group, we received a total number of 14,827 nuclei from young and 12,981 nuclei from aged hearts that were bioinformatically assessed using a t-distributed sto-chastic neighbor embedding plot (short: t-SNE plot). This procedure allowed us to visualize clusters of nuclei that share similar transcriptomic signatures. Unsupervised clustering resulted in a total of 15 distinct nuclei clusters that were annotated to specific cell types using the expression of cell-specific genes and by using al-ready published scRNA / snRNA seq data sets179,205. In this way, we determined 7 major cell types among of which fibroblasts, cardiomyocytes, endothelial cells, pericytes, immune cells, epicardial cells and adipocytes were present, whereas fibroblasts contained 2 sub-clusters and endothelial cells as well as cardiomyocytes contained 3 sub-clusters each. In total 128 uniquely differentially expressed (non-redundant) genes (DEGs) were identified in young versus aged nuclei. Among of the DEGs, 107 genes were uniquely up- and 21 genes were uniquely down-regulated in aged hearts. Surprisingly, the majority of all DEGs were associated with fibroblasts, suggesting that cardiac aging mainly influences this cell population. A ligand-receptor analysis furthermore predicted that fibroblasts could mostly interact with endothelial cells. Thus, we applied conditioned media from fibroblasts derived from young and aged hearts on endothelial cells. Functional analyses indicated deterioration of paracrine signatures between fibroblasts and endothelial cells in old hearts. Aged heart-derived fibroblasts impaired endothelial cell angiogenesis and autophagy but augmented the pro-inflammatory response. In particular, expression of Serpine1 (PAI-1) and Serpine2 (PN-1) were significantly increased and secreted by old fibroblasts to exert anti-angiogenic on endothelial cells, an effect that could be significantly prevented by using neutralizing antibodies. Moreover, we found an enlarged subpopulation of aged fibroblasts expressing osteoblast genes in the epicardial layer associated with increased calcification. Taken together this study provides system-wide insights and identifies molecular changes of aging cardiac fibroblasts, which may contribute to declined heart function.