2026 Theses Doctoral

The Role of the Cardiac Fibroblast in the Development and Treatment of BAG3-mediated Dilated Cardiomyopathy

Morsink, Margaretha A.

Dilated cardiomyopathy (DCM) has long been understood through the lens of the cardiomyocyte, with pathogenic variants in sarcomeric genes driving contractile dysfunction as the prevailing paradigm. Yet many disease-associated genes, including BAG3, are expressed ubiquitously across cell types, raising the critical question of whether non-myocyte populations make independent contributions to disease.

The cardiac fibroblast (CF) is a compelling candidate in this regard: far beyond its canonical role in extracellular matrix production and structural maintenance, the CF actively participates in mechano-sensing, paracrine signaling, and electrical coupling through gap junctions, making it a central orchestrator of cardiac homeostasis.

To investigate whether BAG3 plays a role in CFs and whether that role contributes to DCM, I first validated a mature, patient-derived human induced pluripotent stem cell (hiPSC)-based engineered heart tissue (EHT) model, which recapitulated key aspects of BAG3-mediated DCM. Then, I leveraged an isogenic pair of wild-type and BAG3 knockout hiPSCs, differentiated towards cardiomyocytes and CFs, to dissect cell-type-specific contributions to cardiac function.

Using a combinatorial mixing strategy in EHTs, I demonstrated that loss of BAG3 in CFs intrinsically impairs tissue contractility and drives a fibrotic phenotype, recapitulating hallmarks of DCM independent of cardiomyocyte dysfunction. Mechanistically, BAG3 regulates TGFBR2 degradation through the BAG3-HSP70-CHIP axis, controlling TGF-β signaling and downstream fibrosis. BAG3-null CFs further exhibited hyperproliferation and dysregulated extracellular matrix deposition. This fibrotic transcriptional signature was recapitulated in CFs from BAG3-variant DCM patients, as revealed by single-nuclei RNA sequencing of human cardiac tissue.

To translate these findings into therapeutic discovery, we developed a machine-learning classifier capable of distinguishing CF genotypes and deployed it in a high-throughput small-molecule screen, identifying a compound that improved contractility in fibroblast-specific knockout tissues and partially rescued force production in full BAG3 knockout EHTs.

These results fundamentally reframe our understanding of BAG3-mediated DCM, establishing the cardiac fibroblast as an active and independent driver of disease pathogenesis. More broadly, this thesis provides a framework for interrogating the contributions of non-myocyte cell types to genetic cardiomyopathies, demonstrating that the path toward effective therapy may require moving beyond the cardiomyocyte-centric paradigm to embrace the full multicellular complexity of the heart.