Nitro-oleic acid enhances mitochondrial metabolism and ameliorates heart failure with preserved ejection fraction in mice

Abstract

The prevalence of heart failure with preserved ejection fraction (HFpEF) is increasing, while treatment options are inadequate. Hypertension and obesity-related metabolic dysfunction contribute to HFpEF. Nitro-oleic acid (NO₂-OA) impacts metabolic syndromes by improving glucose tolerance and adipocyte function. Here we show that treatment with NO₂-OA ameliorates diastolic dysfunction and heart failure symptoms in a HFpEF mouse model induced by high-fat diet and inhibition of the endothelial nitric oxide synthase. Proteomic analysis of left ventricular tissue reveals that one-third of identified proteins, predominantly mitochondrial, are upregulated in hearts of NO₂-OA-treated HFpEF mice compared to naïve and vehicle-treated HFpEF mice. Increased mitochondrial mass and numbers, and enhanced mitochondrial respiration are linked with this response, as assessed by transmission electron microscopy and high-resolution respirometry. Activation of the 5'-adenosine-monophosphate-activated-protein-kinase (AMPK) signaling pathway mediates the enhancement of mitochondrial dynamics in hearts of NO₂-OA-treated HFpEF mice. These findings suggest that targeting mitochondrial function with NO₂-OA may represent a promising therapeutic strategy for HFpEF.

Fig. 1. Nitro-oleic acid mitigates heart failure in mice with HFpEF.

Mice received a high-fat diet (HFD) and the endothelial nitric oxide synthase inhibitor L-NAME (HFD + L-NAME) or a standard chow diet (Chow) for 15 weeks (wk). They were treated with vehicle or nitro-oleic acid (NO₂-OA) during the last 4 wk. Echocardiography (Echo), glucose tolerance test (GTT), hemodynamics (PV-loop), exercise, transmission electron microscopy (TEM), proteomics of left ventricular tissue (LV proteome) and respirometry of isolated cardiomyocytes was conducted at the final time point. a Experimental protocol. b Left ventricular ejection fraction (LVEF) and left ventricular posterior wall thickness (LVPW, d) (N = 19/14/18). c Representative ultrasound images of 4-chamber B-Mode (upper panel) and LV tissue doppler at mitral annulus (lower panel) to assess diastolic function. d Ratio of early diastolic mitral inflow velocity to early diastolic mitral annulus velocity (MV E/e´) and isovolumetric relaxation time (IVRT) (N = 19/14/18). e LV global longitudinal strain (GLS) (N = 9/9/9). f Compliance index shown as E/e´ in ratio to the individual end diastolic volume (EDV) (N = 19/14/18). g, h LV end-diastolic pressure (LV EDP) (g) and chamber stiffness index ß_W (h) derived from hemodynamic mesurements (N = 11/7/6). i Work rate calculated from maximum exhaustion exercise test (N = 18/14/17). j Left atrial area (LA area) (N = 19/14/18). k Representative images of muscularization of pulmonary arteries by staining α-smooth muscle actin (αSMA) and von Willebrand factor (vWF) (left) and quantification of αSMA-positive area related to vessel circumference of arterioles (right) (N = 9/8/9). l LV arterial elastance (LV E_a) (N = 11/7/6). Data are presented as mean ± standard deviation. Statistical significance was calculated by One-way ANOVA followed by Bonferroni´s post-hoc test for b, d, e, g-l. Statistical significance for f was calculated by Kruskal-Wallis-test followed by Dunn´s multiple comparison test. Only statistically significant differences are indicated. p*<0.05, **<0.01, ***<0.001, ****<0.0001.

Fig. 2. Nitro-oleic acid treatment alters myocardial proteome and glucose metabolism in HFpEF mice.

a Principal component analysis (Chow: N = 9; HFD + L-NAME vehicle: N = 8; HFD + L-NAME NO₂-OA: N = 6). Volcano-plots comparing high-fat diet (HFD) + L-NAME vehicle with chow mice (b) and NO₂-OA-treated with vehicle-treated HFD + L-NAME mice (d). Significantly changed proteins are marked (significance ≥ -log₁₀(0.05); red: fold change > log₂(1.5); blue: fold change < -log₂(1.5)). Protein names, log fold changes and p-values are listed in the source data file. c KEGG pathway enrichment analysis of significantly increased proteins of HFD + L-NAME vehicle compared to chow mice (protein count ≥ 2; significance ≥ -log₁₀(0.05)). e Venn diagram of proteins significantly decreased in the HFD + L-NAME vehicle compared to the chow group and proteins significantly increased in NO₂-OA-treated compared to vehicle-treated HFD + L-NAME mice. f Body weight (N = 12/12/12) and glucose tolerance test (N = 7/7/7). g Protein abundances of solute carrier family 2, member 4 (GLUT4) and pyruvate dehydrogenase kinase 4 (PDK4) relative to cytochrome c oxidase 4 (COX4) derived from liquid chromatography–mass spectrometry (N = 8/8/6). h Quantification of immunoblots (IB) for PDK4 relative to COX4 and phosphorylated pyruvate dehydrogenase (p-PDH) relative to COX4 (N = 9/8/6). Protein levels were normalized to the total protein amount assessed on IB by fluorescence labeling (N = 9/8/6) (Fig. S12 and S13). Data are presented as mean ± standard deviation for f (left), g and h. f (right) shows mean ± standard error of the mean. Data in g and h are shown relative to the mean of the chow group. Statistical significance for f (left), g and h (left) was calculated by One-way ANOVA followed by Bonferroni´s post-hoc test. Kruskal-Wallis-test followed by Dunn´s multiple comparison test was used for h (right). Two-way ANOVA followed by Bonferroni´s multiple comparisons test was used for f (right). * indicates differences between HFD + L-NAME vehicle and HFD + L-NAME NO₂-OA. Statistical significance was calculated with Perseus (MPI Biochemistry) for b and d and with two-tailed Fisher’s exact test for c. Only statistically significant differences are indicated. p *<0.05, **<0.01, ****<0.0001. Outliers in g were identified using ROUT method and excluded from analysis.

Fig. 3. Mitochondrial dynamics is stimulated by nitro-oleic acid in LV of mice with HFpEF.

Identification of mitochondrial proteins (a) and localization (b) by mitoCarta. a Heatmap of mitochondrial proteins detected by liquid chromatography–mass spectrometry in left ventricular tissue of mice after chow diet, 15 weeks of high-fat diet (HFD) and L-NAME with 4 weeks of vehicle (HFD + L-NAME vehicle) or NO₂-OA treatment (HFD + L-NAME NO₂-OA). Protein names and z-scores are in the source data file. b Shown is the percentage of identified mitochondrial proteins (grey) and the percentage of mitochondrial proteins with significantly enhanced abundance after administration of NO₂-OA in HFD + L-NAME mice compared to vehicle-treated HFD + L-NAME mice (red). c Protein levels of highly abundant mitochondrial proteins (-log₁₀(p-value) > 4.5; log₂(fold change) > 1.9) (N = 9/8/6), ABCD3: ATP-binding cassette subfamily D member 3; COX: cytochrome c oxidase; TOM70: translocase of outer mitochondrial membrane 70; TM177: transmembrane protein 177. d Representative immunoblots (IB) and quantification of COX4 and TOM70 (N = 9/8/6). Protein levels were normalized to the total protein amount assessed on IB by fluorescence labeling (Figs. S12 and S13). e Representative TEM images (N = 5 animals per group) at magnifications of 5 K (scale bar 5 µm), 10 K (scale bar 2 µm), and 31.5 K (scale bar 1 µm), sc: sarcomere; ld: lipid droplet; mt: mitochondria with intact cristae; white arrow: early cristae fragmentation; white asterisk: cristae adhesion; black asterisk: swollen, ruptured mitochondria. f Number of mitochondria normalized by mitochondrial area (24 images per group, N = 4/5/4). g Correlation of f with isovolumetric relaxation time (IVRT). h Relative mitochondrial DNA (mtDNA) copy number and protein abundance of the mitochondrial transcription factor (TFAM). Data are mean ± standard deviation for c, d, f and h. Data in c, d and h are shown relative to the mean of the chow group. Statistical significance was calculated with One-way ANOVA followed by Bonferroni´s post-hoc test for c, d, f and h. Kruskal-Wallis-test followed by Dunn´s multiple comparison test was used for COX1 in c and TOM70 in d. Pearson correlation was used for g. Only statistically significant differences are indicated. p: *<0.05, **<0.01, ***<0.001, ****<0.0001.

Fig. 4. Activation of the AMPK signaling pathway is induced by nitro-oleic acid in LV of HFpEF mice.

a KEGG pathway enrichment analysis of proteins derived from liquid chromatography followed by mass spectrometry (LC-MS) with significantly enhanced abundance after administration of NO₂-OA in high-fat diet (HFD) + L-NAME treated mouse hearts compared to vehicle-treated mouse hearts (HFD + L-NAME vehicle) (inclusion criteria: protein count ≥ 10; pathway enrichment factor ≥ 30%; significance ≥ -log₁₀(0.05)). The most significant KEGG pathway is marked with a red arrow. b Heatmap of all proteins associated with the AMP-activated protein kinase (AMPK) signaling pathway and identified by LC-MS analysis. Protein names and z-scores are in the source data file. c Representative immunoblots (IB) and quantification of AMPK and phosphorylation level of AMPK at threonine residue 172 (p-AMPK), sirtuin 1 (SIRT1) and sirtuin 3 (SIRT3) (N = 9/8/6). Protein levels were normalized to the total protein amount assessed on IB by fluorescence labeling (Figs. S12 and S14). Data in c are shown relative to the mean of the chow group as mean ± standard deviation. Statistical significance was calculated by two-tailed Fisher’s exact test for a and One-way ANOVA followed by Bonferroni´s post-hoc test for c. Only statistically significant differences are indicated. p: **<0.01, ***<0.001, ****<0.0001.

Fig. 5. Nitro-oleic acid enhances cardiac mitochondrial respiration.

Cardiomyocytes of mice after chow diet, 15 weeks (wk) of high-fat diet (HFD) and L-NAME with 4 wk of vehicle (HFD + L-NAME vehicle) or NO₂-OA treatment (HFD + L-NAME NO₂-OA) were isolated and subjected to high resolution respirometry. a Representative trace of oxygen flux (upper panel). Addition of components are marked with blue lines, CarnPCoA: carnitine palmitoyl-CoA; ADP: adenosine diphosphate; PGM: pyruvate glutamate malate; CytC: cytochrome C; Succ: succinate; Omy: oligomycin; FCCP: mitochondrial uncoupler; Rot: rotenone; AMA: antimycin A. Quantification of complex I- and complex II-linked respiration, oxidative phosphorylation (OXPHOS), and carnitine palmitoyltransferase 1 (CPT1B)-linked respiration (lower panel) (N = 8/6/5). b Isolated adult murine cardiomyocytes were cultivated for 48 h under control conditions (Ctrl) with normal glucose levels (low Glu: low glucose levels), under metabolic stress conditions induced by high glucose levels (high Glu), endothelin-1 (ET1) and hydrocortisone (HC) with treatment of methanol (Vehicle) or nitro-oleic acid (NO₂-OA). c Representative immunoblot (IB) and phosphorylation level of AMP-activated protein kinase at threonine residue 172 (p-AMPK). Protein levels were normalized to the total protein amount assessed on IB by fluorescence labeling (Fig. S15) and shown relative to the individual Ctrl group (N = 4). d Quantification of fluorescence intensity of translocase of outer mitochondrial membrane 70 (TOM70), shown as arbitrary units (a.u.), for all analyzed individual cardiomyocytes of three experiments (N = 3). e High resolution respirometry was performed with primary murine cardiomyocytes after cultivation and treatment. Representative trace of oxygen flux. The addition of components are marked with blue lines. Oxidative phosphorylation (OXPHOS) values were extracted from oxygen flux traces (N = 9). Data are presented as mean ± standard deviation. Statistical significance was calculated by One-way ANOVA followed by Bonferroni´s post-hoc test for a and d and unpaired, two-sided Student’s t-test for e. Only statistically significant differences are indicated. p: *<0.05, ***<0.001, ****<0.0001.

Fig. 6. Nitro-oleic acid enhances cardiac fatty acid metabolism in HFpEF mice.

a Heatmap of all proteins associated with fatty acid metabolism and identified by liquid chromatography mass spectrometry (LC-MS) after administration of NO₂-OA in high-fat diet (HFD) + L-NAME treated mouse hearts compared to vehicle-treated mouse hearts. Protein names and z-scores are in the source data file. b Protein abundances of carnitine palmitoyltransferase 1B (CPT1B) relative to cytochrome c oxidase 4 (COX4) and long chain fatty acid transporter (FAT/CD36) derived from LC-MS. Data are shown relative to the mean of the chow group (N = 9/8/6). c Representative images of adult murine cardiomyocytes stained for neutral lipids (left). Quantification of lipid droplet area normalized to cardiomyocyte area shown for all individual analyzed images (right) (9 to 10 images at a magnification of 63x per animal, N = 6/5/6). Data are presented as mean ± standard deviation. Statistical significance was calculated by One-way ANOVA followed by Bonferroni´s post-hoc test. Only statistically significant differences are indicated. p: *<0.05, **<0.01, ***<0.001, ****<0.0001.