Cardiac Slc25a49-Mediated Energy Reprogramming Governs Doxorubicin-Induced Cardiomyopathy through the G6P-AP-1-Sln Axis

Abstract

Doxorubicin (Dox), a potent antitumor drug, is linked to cardiac toxicity. Few mechanism-based therapies against cardiotoxicity are available. Dysfunction in mitochondrial energy metabolism contributes to Dox-induced cardiomyopathy. It is aimed at exploring the association between specific mechanism of energy reprogramming and Dox-induced cardiomyopathy. Cardiac-specific ablation of Slc25a49 mice are generated by crossing Slc25a49^flox/flox mice with Myh6-Cre mice. Slc25a49^HKO mice or SLC25A49^KD cardiomyocytes is treated with Dox. Echocardiography, histological analysis, transmission electron microscopy, bulk RNA sequencing, cell bioenergetic profiling, metabolomics test, chromatin immunoprecipitation, and dual-luciferase reporter assay are conducted to delineate the phenotype and elucidate the molecular mechanisms. Specific ablation of Slc25a49 in cardiomyocytes leads to exacerbated Dox-induced cardiomyopathy, characterized by compromised mitochondrial respiration enhanced glycolysis and increased glycolytic metabolite glucose-6-phosphate (G6P) levels, subsequently activating the activator protein-1 (AP-1) complex. The stimulation of the G6P-AP-1 axis intensifies myocardial damage via transcriptionally regulating Sarcolipin (Sln) expression. Strikingly, targeting of this axis with the AP-1 inhibitor T-5224 effectively improves survival and enhances cardiac function in Dox-induced cardiomyopathy. This study provides mechanistic insights into energy reprogramming that permits myocardial dysfunction, and thus provides a proof of concept for antienergy reprogramming therapy for Dox-induced cardiomyopathy through directly modulating G6P-AP-1-Sln axis.

Keywords: Slc25a49–G6P–AP‐1–Sln axis; cardiomyopathy; doxorubicin; energy reprogramming; mitochondria.

Figure 1

Slc25a49^HKO exacerbates Dox‐induced cardiomyopathy. A) Experiment timeline in vivo. B) Kaplan–Meier plot showing increased mortality in Dox‐treated Slc25a49^HKO mice (n = 12 for each group). C) Serum levels of Ck, Ldh, Ast, Ck‐mb, and Ldh‐1 were measured in Slc25a49^flox/flox and Slc25a49^HKO mice treated with Corn oil or Dox at 3 months (n = 6 for each group). D) Real‐time quantitative PCR analysis of Anp and Bnp mRNA expression in four groups of mice at 3 months (n = 5 for each group). E) Echocardiography of four groups of mice at 3 months. F) Echocardiography parameters of (E) were calculated (n = 6 for each group). G) Horizontal morphology of Slc25a49^flox/flox and Slc25a49^HKO hearts with or without Dox‐treatment (scale bar = 1 mm). H) Comparison of heart weight among four groups of mice (n = 6 for each group). I) Wheat germ agglutinin (WGA) staining to compare sectional cardiomyocyte size in four groups of mice (Scale bar = 100 µm). J) Quantification of cardiomyocyte area in (I) using Image J (n = 10 for each group). K) Representative Sirus Red staining among four groups of mice (Scale bar = 50 µm). L) Quantification of K) (n = 10 for each group). M) Real‐time quantitative PCR analysis of Col1a1, Col3a1, and Mmp2 mRNA expression in four groups of mice at 3 months (n = 5 for each group). Ck, creatine kinase; Ldh, lactic dehydrogenase; Ast, aspartate aminotransferase; Anp, atrial natriuretic peptide; Bnp, brain natriuretic peptide; LVEF, left ventricular ejection fraction; LVFS, left ventricular fraction shortening; LVID;s, left ventricular internal dimension at systole; LVESV, left ventricular end‐systolic volume; CO, cardiac output; and CI, cardiac input. Data were presented as means ± SD, and analyzed by one‐way ANOVA test. ns, not significant, **P < 0.01, and ***P < 0.001.

Figure 2

Slc25a49^HKO inhibits OXPHOS with Dox‐treatment. A) Heatmap presenting differentially expressed mitochondrial respiratory ETC (Complex I‐V) genes in Slc25a49^flox/flox and Slc25a49^HKO hearts with or without Dox‐treatment at 3 months of age (n = 3 for each group). B) Real‐time quantitative PCR analysis for determining the mRNA expression of selected ETC complex genes in four groups of mice (n = 5 for each group). C) Representative transmission electron microscopy images of four groups of mice at 3 months (Top: 7392 scale bar = 1 µm; bottom: scale bar = 2 µm). D) Quantification of mitochondria‐related parameters from (C). Mitochondria/µm² refers to the average number of mitochondria (n = 7 for each group); ‰ mitochondrial area refers to the ratio of mitochondrial area to image area (n = 7 for each group); % abnormal mitochondria refers to the percentage of abnormal mitochondria in individual samples (n = 7 for each group); cristae/mitochondira refers to the surface area of the inner mitochondrial membrane (n = 7 for each group). E) Measurement of mitochondrial membrane potential in AC16 clones of SLC25A49^WT , SLC25A49^KD , and SLC25A49^KD + rescue treated with or without Dox for 24 h (Dox, 500 nm; n = 3 for each group). F) ATP content in four groups of mice at 3 months (n = 6 for each group). G) ATP content in six groups of AC16 clones SLC25A49^WT , SLC25A49^KD , SLC25A49^KD + rescue treated with or without Dox for 24 h (Dox, 500 nm; n = 4 for each group). H) Real‐time monitoring the OCR in AC16 clones of SLC25A49^WT , SLC25A49^KD , and SLC25A49^KD + rescue treated with or without Dox for 24 h (Dox, 500 nm; n = 6 for each group). I) OCR was measured in AC16 clones of SLC25A49^WT , SLC25A49^KD , and SLC25A49^KD + rescue treated with or without Dox for 24 h (Dox, 500 nm; n = 6 for each group). OXPHOS, oxidative phosphorylation; ETC, electron transport chain; ATP, adenosine triphosphate; OCR, oxygen consumption rates. Data were presented as means ± SD, and analyzed by one‐way ANOVA test. ns, not significant, *P < 0.05, **P < 0.01, and ***P < 0.001.

Figure 3

Slc25a49 ^HKO upregulates glycolysis and accumulates G6P. A) Real‐time monitoring the ECAR in AC16 clones of SLC25A49^WT , SLC25A49^KD , and SLC25A49^KD + rescue treated with or without Dox for 24 h (Dox, 500 nm; n = 6 for each group). B) ECAR was measured in AC16 clones of SLC25A49^WT , SLC25A49^KD , and SLC25A49^KD + rescue treated with or without Dox for 24 h (Dox, 500 nm; n = 6 for each group). C) Heatmap showing differential metabolites in Slc25a49^flox/flox and Slc25a49^HKO hearts with or without Dox‐treatment at 3 months of age (n = 3 for each group). D) G6P levels in four groups of mice (n = 3 for each group). E) G6P levels in SLC25A49^WT , SLC25A49^KD , and SLC25A49^KD + rescue hearts treated with or without Dox for 24 h (Dox, 500 nm; n = 3 for each group). F) Lactic acid levels in AC16 clones of SLC25A49^WT , SLC25A49^KD , and SLC25A49^KD + rescue treated with or without Dox for 24 h (Dox, 500 nm; n = 3 for each group). G) G6P levels in AC16 clones of SLC25A49^WT , SLC25A49^KD , and SLC25A49^KD + rescue treated with or without rotenone for 24 h (Rotenone, 1 µm; n = 3 for each group). H) Lactic acid levels in AC16 clones of SLC25A49^WT , SLC25A49^KD , and SLC25A49^KD + rescue treated with or without rotenone for 24 h (Rotenone, 1 µm; n = 3 for each group). I) Schematic representation of G6P‐related changes in glycolysis. J). Real‐time quantitative PCR determining the mRNA expression of G6P‐related genes during glycolysis in four groups of mice (n = 5 for each group). K) Representative immunoblotting images showing G6P‐related protein during glycolysis in four groups of mice (n = 3 for each group). L) Quantification of (K). ECAR, extracellular acidification rate; HK1, hexokinase 1; G6P, glucose‐6‐phosphate; GPI, glucose‐6‐phosphate isomerase; F‐6‐P, fructose‐6‐phosphate. Data were presented as means ± SD, and analyzed by one‐way ANOVA test. ns, not significant, **P < 0.01, and ***P < 0.001.

Figure 4

G6P enhances AP‐1 phosphorylation and nuclear translocation. A) KEGG enrichment analysis of RNA‐sequencing data from 3‐month‐old hearts of Dox‐treated Slc25a49^flox/flox and Slc25a49^HKO mice. B) Heatmap presenting differentially expressed AP‐1 family genes in Slc25a49^flox/flox and Slc25a49^HKO hearts with or without Dox‐treatment at 3 months of age (n = 3 for each group). C) Real‐time quantitative PCR analysis determining the mRNA expression of AP‐1 family genes in four groups of mice (n = 6 for each group). D) Representative immunoblotting images showing AP‐1 family protein in four groups of mice (n = 3 for each group). E) Quantification of (D). F–H) AC16 clones transfected with shRNA‐targeting SLC25A49 or negative control were treated with Dox for 24 h. AP‐1 family proteins were measured by immunoblotting F), nucleocytoplasmic separation G), and quantified H) (n = 3 for each group). I) Real‐time quantitative PCR analysis determining the mRNA expression of AP‐1 family genes in AC16 clones treated with 0.2 mm G6P or control for 24 h (n = 3 for each group). J–M) AC16 clones were treated with 0.2 mm G6P or control for 24 h. AP‐1 family proteins were measured by immunoblotting J), nucleocytoplasmic separation L), and quantified K,M) (n = 4 for each group). AP‐1, activator protein‐1; G6P, glucose‐6‐phosphate. Data were presented as means ± SD, and analyzed by one‐way ANOVA test. ns, not significant, *P < 0.05, and ***P < 0.001.

Figure 5

G6P–AP‐1 axis aggravates myocardial injury by targeting and upregulating Sln expression. A) Volcano plot of differentially expressed genes between 3‐month‐old Dox‐treated Slc25a49^flox/flox and Slc25a49^HKO hearts. B) Real‐time quantitative PCR analysis determining the mRNA expression of Sln in Slc25a49^flox/flox and Slc25a49^HKO hearts with or without Dox‐treatment at 3 months of age (n = 5 for each group). C) Representative immunoblotting images showing Sln protein expression in four groups of mice (n = 3 for each group). D) Quantification of (C). E) Representative immunoblotting images showing SLN protein expression in AC16 clones of SLC25A49^WT and SLC25A49^KD with or without Dox‐treatment (n = 3 for each group). F) Real‐time quantitative PCR analysis determining the mRNA expression of SLN in AC16 clones added with 0.2 mm G6P or a control for 24 h (n = 3 for each group). G,H) Representative immunoblotting images showing SLN protein expression in AC16 clones added with 0.2 mm G6P or a control for 24 h, and quantified H) (n = 3 for each group). I) In the murine Sln promoter, a scheme summarizing the two predicted AP‐1 binding sites and their corresponding mutant forms. WT, wild‐type Sln promoter; Mut, Sln promoter with mutated AP‐1 binding site (red). J,K) ChIP‐qPCR analysis of the interaction between AP‐1 and two AP‐1 binding sites in the murine Sln promoter region under basal condition J) or Dox‐treated condition K). L) As indicated in panel (I), Sln promoter luciferase reporter plasmids containing wild‐type (WT) or mutated (Mut) AP‐1 binding sites were transfected into AC16 cell lines, and their luciferase reporter activities were assayed. M) Real‐time quantitative PCR analysis determining the mRNA expression of ANP and BNP in AC16 clones treated with SLN siRNA for 24 h (n = 3 for each group). N) Real‐time quantitative PCR analysis determining the mRNA expression of AP‐1 family and SLN genes in Dox‐treated AC16 clones added with 40 µm T‐5224 or a control for 24 h (Dox, 500 nm; n = 3 for each group). O) Dox‐treated AC16 clones were added with 40 µm T‐5224 or a control for 24 h. AP‐1 family and SLN proteins were measured by immunoblotting and quantified (n = 3 for each group). AP‐1, activator protein‐1; Sln, sarcolipin; G6P, glucose‐6‐phosphate. Data were presented as means ± SD, and analyzed by one‐way ANOVA test. ns, not significant, *P < 0.05, **P < 0.01, and ***P < 0.001.

Figure 6

Inhibiting AP‐1 rescues Dox‐induced cardiomyopathy. A) Schematic diagram illustrating the experimental strategy for AP‐1 inhibition. B) Kaplan–Meier plot demonstrating decreased mortality in heart failure mice after AP‐1 inhibition (n = 12 for each group). C) Serum levels of Ck, Ldh, Ast, Ck‐mb, and Ldh‐1 were measured in four groups of mice (n = 6 for each group). D) Real‐time quantitative PCR analysis of Anp and Bnp mRNA expression in four groups of mice (n = 5 for each group). E) Echocardiography of Dox‐treated Slc25a49^flox/flox and Slc25a49^HKO mice with or without AP‐1 inhibition at 3 months. F) Echocardiography parameters of (E) were calculated (n = 6 for each group). G) Horizontal morphology of four groups of mice at 3 months (scale bar = 1 mm). H) Comparison of heart weight in four groups of mice at 3 months (n = 6 for each group). I) Wheat germ agglutinin (WGA) staining to compare the sectional cardiomyocyte size in four groups of mice (Scale bar = 100 µm). J) Quantification of cardiomyocyte area in (I) using Image J (n = 10 for each group). K) Comparison of Sirus Red staining in four groups of mice (Scale bar = 50 µm). L) Quantification of (K) (n = 10 for each group). M) Real‐time quantitative PCR analysis of Col1a1, Col3a1, and Mmp2 mRNA expression in four groups of mice (n = 5 for each group). Ck, creatine kinase; Ldh, lactic dehydrogenase; Ast, aspartate aminotransferase; Anp, atrial natriuretic peptide; Bnp, brain natriuretic peptide; LVEF, left ventricular ejection fraction; LVFS, left ventricular fraction shortening; LVID;s, left ventricular internal dimension at systole; LVESV, left ventricular end‐systolic volume; CO, cardiac output; and CI, cardiac input. Data were presented as means ± SD, and analyzed by one‐way ANOVA test. ns, not significant, *P < 0.05, **P < 0.01, and ***P < 0.001.