Abstract Body (Do not enter title and authors here): Background: Cachexia is a common complication of heart failure (HF) and is associated with high mortality. Growth differentiation factor 15 (GDF15) regulates food intake and can cause cancer cachexia. GDF15 is a sensitive biomarker in humans, though its biologic function in HF is unknown.
Objective: We sought to define the role of GDF15 in HF.
Methods: We utilized a genetic mouse model of dilated cardiomyopathy (DCM) caused by a mutation in the phospholamban gene (PLN^R9C). PLN^R9C mice have dysregulated cardiac calcium handling (a common feature of nearly all forms of HF) and develop progressive DCM that leads to HF and premature death. Expression analyses were performed on tissue and isolated myocardial cells via flow cytometry and single nucleus RNA-sequencing (snRNA-seq) in PLN^R9C and age-matched wild type (WT) mice. A double transgenic mouse was created by crossing PLN^R9C with a constitutive Gdf15 knock-out (KO). Cardiac function and fibrosis were assessed using echocardiography and histochemistry, respectively. We measured food intake in individually housed mice, and tissue composition was quantified by weight at necropsy, dual-energy X-ray absorptiometry (DXA) and histochemistry. Survival was assessed by Kaplan-Meier.
Results
GDF15 mRNA (43-fold; p<0.01) and protein (54-fold; p<0.01) were increased in LV tissue, and circulating GDF15 was elevated (8.3-fold; p=0.03) in PLN^R9C mice. In other organs, Gdf15 expression was minimal and did not increase with DCM. Single nucleus sequencing revealed that Gdf15 was specifically increased in cardiomyocytes with progression to DCM. The GDF15-specific receptor GDNF-family receptor a-like (GFRAL) was selectively expressed in the hindbrain of these mice. PLN^R9C mice consumed less food and developed cachexia (reduced fat and lean mass at necropsy, reduced fat mass by DXA and reduced skeletal muscle cross-sectional area; p<0.01 for all vs. WT), findings that were reversed in PLN^R9C-Gdf15 KO mice. Gdf15 KO had no effect on cardiac structure or function by echocardiography and PLN^R9C/Gdf15 KO mice displayed only a small reduction in cardiac fibrosis relative to PLN^R9C mice (-3%; p<0.01). Despite this, Gdf15 KO prolonged survival in PLN^R9C (29±3 vs. 25±3 weeks; p<0.01).
Conclusions
Selective upregulation of GDF15 in failing cardiomyocytes triggers cardiac cachexia and worsens survival by an extra-cardiac mechanism. We therefore propose that GDF15 is a novel cardiac hormone produced in HF.