Cardiac fibrosis is a pathological process that contributes to adverse cardiac remodeling. It is a consequence of tissue repair processes driven mainly by cardiac fibroblasts (CFbs). In response to stress, CFbs proliferate and secrete extracellular matrix components which, if excessive, leads to scar formation. Scar tissue can interrupt the connections between cardiomyocytes, ultimately compromising the structural integrity and function of the heart. Functional recovery of the myocardium is not only hindered by the formation of fibrotic tissue but also by the irreversible loss of cardiomyocytes. In addition to the key role of CFbs in scar formation, it has been suggested that a subset of CFbs may be the optimal cell source to generate cardiomyocytes through direct reprogramming. Direct cardiac reprogramming of CFbs represents a promising approach that could lead to regeneration of cardiomyocytes from the endogenous fibroblasts while reducing scar tissue formation. Several studies have demonstrated in vivo direct reprogramming of CFbs leads to an improvement in cardiac function and has been shown to be exceedingly more efficient in the context of recent cardiac injury. Despite the prominent role of CFbs in both scar formation, and in the potential generation of new cardiomyocytes through reprogramming, characterization of these cells is still limited. This is mainly due to lack of reliable markers to identify cardiac fibroblasts, their heterogeneity, and the effects of genetic variation when studying these cells in a diverse population. These constraints prompted us to first identify a panel of surface markers to prospectively identify CFbs. We further performed a comprehensive investigation to identify the developmental heterogeneity of CFbs. We then sought to determine whether developmental origin of CFbs may influence their contribution to formation of scar as well as its effect on their direct reprogramming into iCMs. Finally, by studying CFbs from multiple inbred mouse strains and their response to cardiac insult we aimed to investigate the effect of genetic variation in pathogenesis of cardiac fibrosis.
To undertake a comprehensive study of CFbs, we established a panel of surface markers that can efficiently isolate the majority of CFbs from the adult mouse heart. We employed lineage tracing, transplantation studies, and parabiosis to show that most adult CFbs are derived from the epicardium, a minority arises from endothelial cells, with no contribution from bone marrow or circulating cells. Intriguingly, developmentally distinct CFbs showed similar proliferation rates, and similar gene expression profiles in response to pressure overload injury. We next sought to determine whether this heterogeneity of CFbs may affect their efficiency to generate cardiomyocytes via direct reprogramming, mainly in the context of injury. Using genetic fate-mapping techniques, transplantation studies and gene expression profiling, we showed that the majority of CFbs originate from a shared mesodermal ancestor as cardiomyocytes while a minority of the CFb population originates from neural crest-derived precursors. We provide compelling evidence that, regardless of their developmental origin, CFbs are able to be successfully converted to functional iCMs through in vitro direct reprogramming. However, CFbs generated iCMs with higher efficiency compared to fibroblasts of extra-cardiac organs of identical developmental origin, emphasizing the importance of the physiological microenvironment on cell fate. Remarkably, cardiac injury induced unique re-expression of early developmental genes in CFbs that corresponded to their developmental origin. Finally, we studied the contribution of CFbs from multiple inbred mouse strains following insult to the heart. Our data showed that despite similar increases in proliferation within the different strains, fibroblast activation is a response that correlates with the extent of scar formation. Additionally, by comparing CFbs from multiple strains, we were able to identify potential pathways as therapeutic targets with latent TGF-b binding protein-2 (LTBP2) as a promising diagnostic marker for fibrosis, with relevance to patients with underlying myocardial fibrosis.
Together, our findings suggest that common signaling mechanisms stimulate the pathological response of different CFb populations. However, in the context of direct cardiac reprogramming after injury, the developmental heterogeneity of CFbs may be an essential contributing factor. Our findings also highlight the importance of genetic variation in cardiac fibrosis. Therefore, therapeutic strategies for reducing pathogenic CFbs should target these common pathways instead of targeting fibroblasts of other sources. It may be crucial to study the effects of injury on different CFb subsets for the development of targeted therapies to promote cardiac repair.