PIEZO1 gain-of-function mutation drives cardiomyopathy by disrupting myocardial lipid homeostasis besides iron overload

Abstract

As a mechanosensitive channel, PIEZO1 translates mechanical stretching of cardiomyocytes into Ca²⁺ signaling, underpinning the Frank-Starling law. This mechanism contributes to compensatory responses in heart failure. However, the relationship between PIEZO1 mutations and the development of cardiomyopathy remains unclear. Acute heart failure complicated with severe myocardial iron deposition was identified in the 31-year-old male proband of PIEZO1^D669Y variant. However, PIEZO1 gain-of-function (GOF) mutation D674Y mice and cardiomyocyte-specific Piezo1 overexpression disrupted cardiac function besides iron overload. Using single-cell RNA sequencing, we observed suppression of lipid metabolism pathways in cardiomyocytes with the PIEZO1 GOF mutation, with forkhead box O3 (FOXO3) as a key mediator in lipid metabolism pathways. Specifically, the PIEZO1 GOF mutation increased Ca²⁺ levels, leading to calcium- and calmodulin-dependent protein kinase II (CaMKII) activation and subsequent FOXO3 down-regulation. Together, we demonstrate that PIEZO1 GOF mutation contributes to cardiomyopathy by disrupting myocardial lipid metabolism. This study challenges the current clinical focus on iron-related mechanisms in cardiomyopathy and supports PIEZO1 as a potential candidate for future genetic screening for cardiomyopathy.

Fig. 1.. Identification of PIEZO1^D669Y variant carriers with DCM and iron overload.

(A) Timeline depicting the proband’s (III-1) medical history, key diagnoses, and etiological analysis. (B) Pedigrees and Sanger validation of the PIEZO1^D669Y variants of the patients in the family. Upper section (pedigree chart): adapted from (14), with permission. (C) Representative four-chamber and long-axis view echocardiographic images of the proband patient (III-1) at two time points. (D) Cardiac MRI in four-chamber and short-axis views of the patient (III-1). Upper section (0 month): reprinted from (14), with permission. (E), The duration of medication administration (indicated by gray bars) for the proband patient (III-1) and dynamic changes in LVEF, left ventricular diameter at end-diastole (LV diastolic diameter), and serum ferritin levels over time are shown. Medications, including deferasirox (30 mg/kg/day) for iron overload, metoprolol (starting from 23.75 mg, titrated based on heart rate, reaching 190 mg by month 30 with heart rate maintained between 55 and 60 bpm), enalapril [5 mg daily (qd)], and, later, Angiotensin receptor-neprilysin inhibitor (ARNI) [25 mg twice daily (bid)] for heart failure management, and dapagliflozin (10 mg/day) as standard heart failure treatment. Avg, average.

Fig. 2.. Piezo1^D674Y mutation impairs heart function.

(A) HW/BW ratio of male WT and Piezo1^D674Y (MW) mice at 3 to 12 months (n = 8). (B) HW/TL ratio of male WT and MW mice at 3 to 12 months (n = 8). (C) Representative M-mode echocardiographic images of male WT and MW mice at 3 to 12 months. (D) Echocardiographic analysis of the ejection fraction, FS, left ventricular internal dimension at diastole (LVID;d) and left ventricular internal dimension at systole (LVID;s) of male WT and MW mice at 3 to 12 months (n = 10). (E and F) Representative images (E) and quantification (F) of Masson’s trichrome staining of cardiac tissues from male WT and MW mice at 3 to 12 months (n = 6). Scale bars, 20 μm. Two-tailed nonparametric Mann-Whitney test (B) or unpaired Student’s t test [(A), (B), (D), and (F)] was used. The number of samples in each group is indicated by n. The data are presented as the means ± SEM. M, months; ns, not significant.

Fig. 3.. Piezo1^D674Y is a GOF mutation.

(A and B) Representative Western blot images (A) and quantitative analyses (B) of Piezo1 in cardiac tissues from female and male mice at 3 months (n = 5). (C) Piezo1 concentration in the hearts of male mice at 3 months (n = 6). (D and E) Representative fluorescence surface plot (D) and quantification (E) of Ca²⁺ sparks in the indicated cardiomyocytes from male WT and MW mice (n = 10 cells). (F) Representative images and quantifications of peak h and Tau of cardiomyocytes Ca²⁺ transient traces recorded from male WT and MW mice (n = 16 cells). (G) Representative images and quantifications of peak h and Tau of cardiomyocytes Ca²⁺ transient traces recorded from male WT mice with or without Yoda1 stimulation (n = 16 cells). (H) Snapshots of the WT and MW trimers embedded in the phospholipid membrane at (a) 0 ns, (b) 120 ns, (c) 240 ns, (d) 360 ns, (e) 420 ns, and (f) 450 ns during the MD simulations. (I) Global view and close-up view of the structures of the WT and MW trimers after 450-ns MD simulations. Two-tailed unpaired Student’s t test was used [(B), (C), and (E) to (G)]. The number of samples in each group is indicated by n. The data are presented as the means ± SEM. a.u., arbitrary unit.

Fig. 4.. Piezo1 GOF causes cardiac dysfunction without iron overload.

(A and B) Representative images (A) and quantitative (B) of Perls Prussian blue staining of liver tissues from male WT and MW mice at 3 to 12 months (n = 5). Scale bars, 1 mm and 10 μm. (C and D) Representative images (C) and quantitative (D) of Perls Prussian blue staining of cardiac tissues from male WT and MW mice at 3 to 12 months (n = 5). Scale bars, 1 mm and 10 μm. (E) Iron content in the livers from male WT and MW mice (n = 4). (F) Iron content in the serum of male WT and MW mice (n = 8). (G) Transferrin saturation in the serum of male WT and MW mice (n = 8). (H) Ferritin concentrations in the serum of male WT and MW mice (n = 6). Two-tailed unpaired Student’s t test was used [(B) and (D) to (H)]. The number of samples in each group is indicated by n. The data are presented as the means ± SEM.

Fig. 5.. Piezo1 GOF drives cardiac dysfunction by promoting myocardial lipid accumulation.

(A) Uniform manifold approximation and projection (UMAP) visualization of snRNA-seq profiles from mouse hearts. Cardiomyocyte (CM), smooth muscle cell (SMC), endothelial cell (EC),Lymphatic endothelial cells (LEC). (B) Dotplot showing suppressed pathways in mut3m samples compared with wt3m samples. x axis represents the normalized enrichment score (NES), dot size reflects the gene count within each pathway, and color indicates the P value. (C and D) GSEA of the “fatty acid metabolic process” (C) and “fatty acid beta-oxidation” (D) pathway conducted in cardiomyocytes. (E and F) Representative images (E) and quantification (F) of ORO staining of cardiac tissues from male WT and MW mice at 3 months (n = 5). Scale bar, 20 μm. (G) Representative TEM images of the myocardium of male WT and MW mice at 3 months. Scale bars, 2 and 1 μm. (H) Quantification of the lipid droplet (LD) number and diameter in male WT and MW mice at 3 months (n = 6). (I and J) Representative Western blot images (I) and quantitative analyses (J) of carnitine palmitoyltransferase 1 (Cpt1), Cpt2, CD36, and peroxisome proliferator-activated receptor-gamma coactivator 1 alpha (Pgc-1α) in cardiac tissues from male WT and MW mice at 3 months (n = 6). (K and L) Representative Western blot images (K) and quantitative analyses (L) of pyruvate dehydrogenase kinase 4 (Pdk4), Pdh, and P-Pdh in cardiac tissues from male WT and MW mice at 3 months (n = 6). (M) Pdh activity in cardiac tissues from male WT and MW mice at 3 months (n = 5). P values were determined by GSEA using permutation tests and adjusted using the Benjamini-Hochberg correction method [(B) to (D)]. Two-tailed unpaired Student’s t test was used [(F), (H), (J), (L), and (M)]. The number of samples in each group is indicated by n. The data are presented as the means ± SEM.

Fig. 6.. Foxo3 ameliorates cardiolipotoxicity and cardiac dysfunction in mice with Piezo1 GOF mutation.

(A) The activity of the top five TFs in male WT and MW mice at 3 months. (B) Representative M-mode echocardiographic images of male WT and MW mice injected with indicated AAV9 vector. (C) Echocardiographic analysis of the EF, FS, LVID;d, and LVID;s of male WT and MW mice injected with indicated AAV9 vector (n = 6). (D and E) Representative images (D) and quantification (E) of Masson’s trichrome staining of cardiac tissues from male WT and MW mice injected with indicated AAV9 vector (n = 6). Scale bars, 20 μm. (F) Representative images of myocardial TEM images from male WT and MW mice injected with indicated AAV9 vector. Scale bars, 2 and 1 μm. (G) Quantification of the lipid droplet number and diameter in male WT and MW mice injected with indicated AAV9 vector (n = 5). (H) Representative images and quantitative analysis of C11-BODIPY–stained cardiomyocytes from male WT and MW mice injected with indicated AAV9 vector (n = 6). Scale bar, 20 μm. (I) Representative Western blot images and quantitative analyses of cardiac Cpt1, Cpt2, CD36, and Pgc-1α protein expression in male WT and MW mice injected with indicated AAV9 vector (n = 6). (J) Representative Western blot images and quantitative analyses of cardiac Pdk4, Pdh, and P-Pdh protein expression in male WT and MW mice injected with indicated AAV9 vector (n = 6). One-way ANOVA with Tukey’s test [(C), (E), and (G) to (J)], or Welch test (C) were used for the comparison of multiple groups. The number of samples in each group is indicated by n. The data are presented as the means ± SEM. Ctrl, control.

Fig. 7.. Cardiomyocyte-specific Piezo1 overexpression impairs cardiac function and cardiac lipid metabolism.

(A) Representative M-mode echocardiographic images of male WT and Piezo1-TG^Myl2 (TG^Myl2) mice at 4 weeks. (B) Echocardiographic analysis of the EF, FS, LVID;d, and LVID;s in male WT and TG^Myl2 mice at 6 weeks (n = 10). (C) HW/BW ratios of male WT and TG^Myl2 mice at 6 weeks (n = 6). (D and E) Representative images (D) and quantification (E) of Masson’s trichrome staining of cardiac tissues from male WT and TG^Myl2 mice at 6 weeks (n = 6). Scale bar, 20 μm. (F) Representative images of ORO staining and TEM of cardiac tissues from male WT and TG^Myl2 mice at 6 weeks. Scale bars, 20 μm. (G) Quantification of ORO staining, lipid droplet number, and diameter in male WT and TG^Myl2 mice at 6 weeks (n = 5). (H and I) Representative Western blot images (H) and quantitative (I) of Cpt1, Cpt2, CD36, Pgc-1α, Pdk4, Pdh, and P-Pdh in cardiac tissues from male WT and TG^Myl2 mice (n = 6). (J) Pdh activity in cardiac tissues from male WT and TG^Myl2 mice at 6 weeks (n = 5). (K) Concentrations of adenosine 5′-triphosphate (ATP) in cardiac tissues from male WT and TG^Myl2 mice at 6 weeks (n = 6). (L and M) Representative Western blot images (L) and quantitative (M) of P-CaMKII, Foxo3, and P-Foxo3 in cardiac tissues from male WT and TG^Myl2 mice (n = 6). Two-tailed unpaired Student’s t test was used [(B), (C), (E), (G), (I), (J), (K), and (M)]. The number of samples in each group is indicated by n. The data are presented as the means ± SEM.