A selective inhibitor of mitofusin 1-βIIPKC association improves heart failure outcome in rats

Abstract

We previously demonstrated that beta II protein kinase C (βIIPKC) activity is elevated in failing hearts and contributes to this pathology. Here we report that βIIPKC accumulates on the mitochondrial outer membrane and phosphorylates mitofusin 1 (Mfn1) at serine 86. Mfn1 phosphorylation results in partial loss of its GTPase activity and in a buildup of fragmented and dysfunctional mitochondria in heart failure. βIIPKC siRNA or a βIIPKC inhibitor mitigates mitochondrial fragmentation and cell death. We confirm that Mfn1-βIIPKC interaction alone is critical in inhibiting mitochondrial function and cardiac myocyte viability using SAMβA, a rationally-designed peptide that selectively antagonizes Mfn1-βIIPKC association. SAMβA treatment protects cultured neonatal and adult cardiac myocytes, but not Mfn1 knockout cells, from stress-induced death. Importantly, SAMβA treatment re-establishes mitochondrial morphology and function and improves cardiac contractility in rats with heart failure, suggesting that SAMβA may be a potential treatment for patients with heart failure.

Fig. 1

βIIPKC inhibition reduces mitochondrial fragmentation in heart failure. a Schematic panel of heart failure induction by myocardial infarction (MI) and the treatment protocol. Twelve-week-old male rats were subjected to MI-induced heart failure by left anterior descending coronary artery permanent ligation. Four weeks after MI induction, rats were treated with either the global βIIPKC-specific inhibitor (βII_V5-3) or with a control peptide (TAT, used to deliver βII_V5-3 into the heart). Peptide treatment was continuous (for 6 weeks) using an Alzet pump, delivering at a rate of 3 mg per Kg per day. b Sarcomeric shortening in isolated ventricular cardiomyocytes, c left ventricular ejection fraction and d LVEDd [left ventricular end-diastolic dimension] measured by echocardiography at the end of the experimental protocol, input: delta of measurements performed before and after treatment; e representative cardiac transmission electron micrographs (scale bar: 0.5 μm); f quantification of intermyofibrillar mitochondrial number and area in the transmission electron micrographs; g PKCs and RACK1 protein levels in cardiac mitochondrial fraction and h representative western blots; and i representative western blots of mitochondrial βIIPKC location in sham (white bars, n = 8), TAT-treated heart failure (HF-Ctr, gray bars, n = 5) and βII_V5-3-treated heart failure (HF-βII_V5-3, red bars, n = 7). Biochemical measurements were performed in the cardiac remote (viable) zone. Data are means ± SEM. *P < 0.05 vs. Sham rats. #P < 0.05 vs. HF-Ctr rats. One-way analysis of variance (ANOVA) with post-hoc testing by Duncan. For all the cardiac function studies, the observer was blinded to the experimental groups

Fig. 2

βIIPKC inhibition reduces mitochondrial fragmentation in cultured cardiomyocytes. a Representative transmission electron micrographs (white arrows indicate small mitochondria, scale bar: 1 μm) and b quantification of mitochondrial number and size in the transmission electron micrographs of neonatal rat cardiomyocytes in culture treated with TAT (carrier) or βII_V5-3 peptides for 30 min followed by incubation with angiotensin II (0.5 µM for 4 h, n = 5 per group). c Cells were then stained with anti-Tom20 antibody (green) and Hoechst stain and counted. Mitochondrial morphology was analyzed using a 63x oil immersion lens (scale bar: 5 μm). The analysis was done in a blinded fashion. The boxed area in each upper panel is enlarged under each micrograph. d Neonatal rat cardiomyocytes in culture were treated with TAT or βII_V5-3 peptides for 30 min followed by incubation with either C2-ceramide (40 μM) or angiotensin II (0.5 µM) for 4 h. After incubation, the mitochondrial levels of βIIPKC and Mfn1 were detected using specific antibodies (representative blot of three independent experiments). e Neonatal rat cardiomyocytes in culture were transfected with βIIPKC silence RNA (βIIPKC-siRNA) or control (NC-siRNA). f 48 h later cells were treated with TAT or βII_V5-3 peptides for 30 min followed by incubation with H₂O₂ (100 µM, 24 h). After incubation, cell viability and toxicity were measured by MTT assay and LDH release, respectively. g Adult rat cardiomyocytes in culture were transfected with βIIPKC-siRNA or NC-siRNA (n = 6 per group). h 24 h later cells were treated with TAT or βII_V5-3 peptides for 30 min followed by incubation with H₂O₂ (100 µM, 1 h). Cell toxicity was measured by LDH release. Data are means ± SEM. *P < 0.05 vs. control cells. #P < 0.05 vs. Ang II- or H₂O₂-treated cells. One-way analysis of variance (ANOVA) with post-hoc testing by Duncan

Fig. 3

βIIPKC inhibition improves bioenergetics in failing hearts. a Mitochondrial respiratory control ratio, b State 3-dependent oxygen control rate, c absolute H₂O₂ release and d H₂O_:O₂ in heart samples from sham (white bars, n = 10), TAT-treated heart failure (HF-Ctr, gray bars, n = 10) and βII_V5-3-treated heart failure (HF-βII_V5-3, red bars, n = 10). e Mfn1, Mfn2, Opa1, Drp1, and Fis1 protein levels and representative western blots; and f Mfn1 and Mfn2 GTPase activity in cardiac mitochondrial fraction (n = 5 per group) from sham, HF-Ctr and HF-βII_V5-3. g Cardiac mitochondrial Mfn1 and βIIPKC immunoprecipitate probed against anti-βIIPKC, Mfn1 and phosphorylated serine/threonine antibodies (representative blot of three independent experiments) from sham, HF-Ctr and HF-βII_V5-3 groups. Biochemical measurements were performed in the cardiac remote (viable) zone. These measurements were performed at the end of the experimental protocol. h Mfn1 wild type (WT, n = 5) and knockout (KO, n = 5) MEFs were treated with PMA [10 nM phorbol ester 12-myristate 13-acetate, for 30 min, an activator of most PKC isozymes] and accumulation of mitochondrial βIIPKC was analyzed by western blot (representative blots of three independent experiments). Data are means ± SEM. *P < 0.05 vs. Sham rats. One-way analysis of variance (ANOVA) with post-hoc testing by Duncan

Fig. 4

Characterization of Mfn1-βIIPKC protein–protein interaction. Recombinant Mfn1 (GST-Mfn1, 25 ng) was incubated with recombinant βIIPKC (GST-βIIPKC, 25 ng) in the presence and absence of phosphatidylserine and diacylglycerol (PS/DAG/Ca²⁺, PKC activators) for 30 min at 37 °C. a Co-immunoprecipitates with anti-Mfn1 and βIIPKC were analyzed by western blot. b Mfn1 phosphorylation was evaluated using radioactive [³²P] ³²P-ATP incorporation after incubation with either βIIPKC or its alternative splicing βIPKC in the presence and absence of classic PKC activators. Representative blots of three independent experiments. c GTPase activity of Mfn1 determined in the presence of active βIIPKC or its alternative splicing βIPKC (n = 6 per group). d Conservation of the βIIPKC phosphorylation in human orthologs of Mfn1 in rat, mouse, and zebrafish and Mfn2 in mouse. Ribbon presentation of the 3D structure of Mfn1 phosphorylation at the βIIPKC site (PDB: 5GOM). e Individual replacement of serine by alanine at residues 86, 284, and 290 in Mfn1 knockout MEFs. Cells were transfected with the following constructs: WT, S86A, S284A and S290A. 48 h after transfection cells were incubated with H₂O₂ (100 μM, 24 h) and measured cytotoxicity (LDH release, n = 6 per group). Data are means ± SEM. *P < 0.05 vs. control. One-way analysis of variance (ANOVA) with post-hoc testing by Duncan

Fig. 5

Rational design of peptide that selectively inhibits Mfn1 and βIIPKC interaction. a Sequence alignment of human βIIPKC and Mfn1 identified a short sequence of homology, RNAENF/NELENF. b NELENF in Mfn1 (PDB: 5GOM) and RNAENF in the V5 domain of βIIPKC (PDB: 3PFQ) are exposed and available for protein−protein interactions (see colored structures). Note that the alternatively spliced βIPKC varies from βIIPKC only in the V5 domain. c Conservation of RNAENFDRF sequence in βIIPKC in a variety of species. d RNAENFDRF sequence is not present in the V5 domain of βIPKC, a βIIPKC alternative spliced form. e Conservation of NELENFTKQ in Mfn1 in a variety of species. f NELENFTKQ sequence is not present in Mfn2. g RNAENFDRF sequence is found in 6 human proteins. Heat-map of the RNAENFDRF conservation in orthologous of these proteins shows RNAENFDRF conservation only in βIIPKC. (Further information about all the proteins is given in the Supplementary Table 2). *Denotes identity, and ^ denotes homology. h Mfn1 wild type (WT, n = 4–6) and i Mfn1 knockout (KO, n = 4–6) MEFs were treated with TAT, p251–255 or SAMβA peptides for 30 min followed by incubation with either H₂O₂ (100 μM, 24 h) or Antimycin A (10 μM, 24 h). After incubation, cytotoxicity was measured by LDH release. j Neonatal rat cardiomyocytes in culture were treated with TAT or SAMβA for 30 min followed by incubation with angiotensin II (0.5 µM for 4 h, n = 5 per group). Cells were then stained with anti-Tom20 antibody (green) and Hoechst stain and counted. Mitochondrial morphology was analyzed using a ×63 oil immersion lens (scale bar: 10 μm). k Co-immunoprecipitates with anti-βIIPKC were analyzed in the mitochondrial fraction by western blot (representative blots of three independent experiments). l Neonatal rat cardiomyocytes in culture were treated with TAT, P251–255, SAMβA or βII_V5-3 peptides for 30 min followed by incubation with H₂O₂ (100 μM for 24 h, n = 4–5 per group). After incubation, cell viability was measured by MTT assay. Data are means ± SEM. *P < 0.05 vs. Ctr. #P < 0.05 vs. SAMβA-treated cells. One-way analysis of variance (ANOVA) with post-hoc testing by Duncan

Fig. 6

SAMβA treatment reduces mitochondrial fragmentation in heart failure. a Schematic panel of heart failure induction by myocardial infarction (MI) and the treatment protocol. Twelve-week-old male rats were subjected to MI-induced heart failure by left anterior descending coronary artery permanent ligation. Four weeks after MI induction, the rats were treated with the either SAMβA peptide (a selective antagonist of Mfn1-βIIPKC association) or with a control peptide (TAT, used to deliver SAMβA into the heart). Peptide treatment was continuous for 6 weeks using an Alzet pump delivery at 3 mg per Kg per day. b Left ventricular ejection fraction and c LVEDd [left ventricular end-diastolic dimension] measured by echocardiography at the end of the experimental protocol, input: delta of measurements performed before and after treatment; d representative cardiac transmission electron micrographs (scale bar: 1 μm); e quantification of mitochondrial number and area in the transmission electron micrographs, the analysis was done in a blinded fashion; f mitochondrial respiratory control ratio, g state 3-dependent oxygen control rate, h absolute H₂O₂ release and i H₂O₂:O₂ in heart samples from sham (white bars, n = 7), TAT-treated heart failure (HF-Ctr, gray bars, n = 8), and SAMβA-treated heart failure (HF-SAMβA, blue bars, n = 11). j Cardiac mitochondrial βIIPKC immunoprecipitate probed against anti-βIIPKC, Mfn1 and phosphorylated serine/threonine antibodies; and βIIPKC immunoprecipitate from heart lysate probed with anti-βIIPKC, troponin I and phosphorylated troponin I antibodies (representative blot of three independent experiments) from sham, HF-Ctr and HF-SAMβA groups. Biochemical measurements were performed in the cardiac remote (viable) zone. These measurements were performed at the end of the experimental protocol. Data are means ± SEM. *P < 0.05 vs. Sham rats. #P < 0.05 vs. HF-Ctr rats. One-way or two-way analyses of variance (ANOVA) with post-hoc testing by Duncan. For all the cardiac function studies, the observer was blinded to the experimental groups

Fig. 7

Sustained SAMβA treatment results in better heart failure outcome. a Left ventricular ejection fraction and b LVEDd (left ventricular end-diastolic dimension) measured by echocardiography before and after treatment; input: delta of measurements performed before and after treatment, in animals from sham (white bars, n = 7), TAT-treated heart failure (HF-Ctr, gray bars, n = 9), βII_V5-3-treated heart failure (HF-βII_V5-3, red bars, n = 7), and SAMβA-treated heart failure groups (HF-SAMβA, blue bars, n = 11). Data are means ± SEM. *P < 0.05 vs. Sham rats. #P < 0.05 vs. HF-Ctr rats. &P < 0.05 vs. before treatment. One-way or two-way analyses of variance (ANOVA) with post-hoc testing by Duncan. For all the cardiac function studies, the observer was blinded to the experimental groups. c Proposed model for the SAMβA-mediated cardioprotection in post-myocardial infarction-induced heart failure