Enhancement of mitochondrial calcium uptake is cardioprotective against maladaptive hypertrophy by retrograde signaling uptuning Akt

Abstract

Regulation of mitochondrial Ca²⁺ uptake is critical in cardiac adaptation to chronic stressors. Abnormalities in Ca²⁺ handling, including mitochondrial uptake mechanisms, have been implicated in pathological heart hypertrophy. Enhancing mitochondrial Ca²⁺ uniporter (MCU) expression has been suggested to interfere with maladaptive development of heart failure. Here, we addressed whether MCU modulation affects the cardiac response to pressure overload. MCU content was quantified in human and murine hearts at different phases of myocardial hypertrophy. Cardiac function/structure were analyzed after Transverse Aortic Constriction (TAC) in mice undergone viral-assisted overexpression or downregulation of MCU. In vitro and ex vivo assays determined the effect of MCU modulation on mitochondrial Ca²⁺ uptake, cellular phenotype and hypertrophic signaling. In human and murine hearts MCU levels increased in the adaptive phase of myocardial hypertrophy and declined in the failing stage. Consistently, modulation of MCU had a cell-autonomous effect in cardiomyocyte/heart adaptation to chronic overload. Indeed, upon TAC MCU-downregulation accelerated development of contractile dysfunction, interstitial fibrosis and heart failure. Conversely, MCU-overexpression prolonged the adaptive phase of hypertrophic response, as, in advanced stages upon TAC, hearts showed preserved contractility, absence of fibrosis and intact vascularization. In vitro and ex vivo analyses indicated that enhancement in mitochondrial Ca²⁺ uptake in cardiomyocytes entails "mitochondrion-to-cytoplasm" signals leading to ROS-mediated activation of Akt, which may explain the protective effects towards heart response to TAC. Enhanced mitochondrial Ca²⁺ uptake affects the compensatory response to pressure overload via retrograde mitochondrial-Ca²⁺/ROS/Akt signaling, thus uncovering a potentially targetable mechanism against maladaptive myocardial hypertrophy.

Keywords: Akt; beta adrenergic receptor; heart failure; mitochondrial calcium uniporter; myocardial hypertrophy.

Fig. 1.

Cardiomyocyte MCU protein content decreases in murine and human failing hearts. (A and B) Hematoxylin-eosin staining of ventricular sections from 14-wk sham-operated and TAC mice. RV, right ventricle; IVS, interventricular septum; LV, left ventricle. (C) MCU protein content in protein extracts of the LV from 14-wk sham-operated and TAC mice. Immunoreactivity for actin was used as loading control. (D) Densitometry of (C). Differences among groups were determined using two-independent-sample t test or Welch t test. (***P < 0.001; Each point represents the pool of 2 hearts. A total of n = 4 sham-operated and n = 6 TAC hearts were analyzed). (E) MCU protein content in cardiomyocytes isolated from the LV of 14-wk sham-operated and TAC mice. Immunoreactivity for actin and TOM20 were used as loading control. (F) Densitometry of (E). Differences among groups were determined using either two-independent-sample t test or Welch t test (*P < 0.05; n = 4 sham-operated; n = 4 TAC hearts). (G) MCU protein content in human cardiac biopsies as acquired from patients during aortic valve replacement. Patients were stratified in two groups based on the Ejection Fraction (EF) into respectively group#1 EF > 50% and group#2, EF < 30%. (H) Densitometry of (G). Differences among groups were determined using a two-independent-sample t test (***P < 0.001; n = 8 patients for group#1; n = 5 patients for group#2).

Fig. 2.

MCU modulation affects heart adaptation to pressure overload. (A) Echocardiographic long-axis view of Empty, MCU-OE, and MCU-KD hearts 4 wk after TAC. (B) Linear regression analysis of Ejection Fraction in Empty, MCU-OE, and MCU-KD mice at different time points after TAC. For comparison between groups, two-independent-sample t test was used (*P < 0.05; **P < 0.01; ***P < 0.001; n = 6 mice/group). (C) Picro-Sirius red staining of ventricular sections of Empty, MCU-OE, and MCU-KD hearts, at baseline, 4 and 8 wk after TAC. RV, right ventricle; IVS, interventricular septum; LV, left ventricle. (D) Confocal immunofluorescence of ventricular sections from Empty, MCU-OE, and MCU-KD hearts, at baseline and 4 and 8 wk after TAC. Sections were costained with antidystrophin (red signal) and AlexaFluor™488-conjugated isolectin (green signal). Nuclei were counterstained with DAPI (blue signal). (E) Evaluation of capillary/CM ratio. Differences among groups were determined using One-way ANOVA with Holm’s correction for multiple comparisons (i.e., 0, 4 wk) or either two-independent-sample t test (i.e., 8 wk) (***P < 0.001; Values represent the average capillary/CM density calculated in at least 3 nonconsecutive cryosections/heart. A total of 4 hearts/group was analyzed). (F) RTqPCR on heart extracts from Empty, MCU-OE, and MCU-KD hearts at baseline and 4 and 8 wk after TAC. Differences among groups were determined using ANOVA tests with Holm’s correction for multiple comparisons. (*P < 0.05; n = 4 hearts/group). (G) Confocal immunofluorescence of ventricular sections from Empty, MCU-OE, and MCU-KD hearts, at baseline and 4 wk after TAC, stained with antidystrophin (red signal). (H) Relative CM growth in 4 wk Empty, MCU-OE, and MCU-KD TAC hearts, compared to baseline controls. Differences among groups were determined using ANOVA tests. (***P < 0.001. Each individual value in the plot represents the average CM cross-sectional area, calculated in at least three nonconsecutive cryosections/heart, normalized to the same measurement in the corresponding sham-operated groups. More than 350 CMs/heart were evaluated, in a total of 4 hearts/group).

Fig. 3.

MCU overexpression in cultured cardiomyocytes enhances hypertrophic response to noradrenaline (NE) via Akt. (A) Western blotting on protein extracts from cultured rat neonatal CMs infected with either Ad-Empty (Empty) or Ad-MCU-flag (MCU-OE) viral vectors. Analysis was used to assess MCU protein content. Immunoreactivity for TOM20 was used to ensure equal protein loading. (B) Confocal immunofluorescence of cultured CMs infected with an Ad-MCU-flag viral vector. Cells were costained with antibodies to a-actinin (red signal) and antiflag (green signal). (C) Relative mitochondrial Ca²⁺ content, measured in MCU-OE and Empty CMs, transfected with mito-CaMeleon. Differences among groups were determined using a Mann–Whitney test. (*P < 0.05; Empty: n = 52 CMs; MCU-OE: n = 56 CMs). (D and E) Live imaging of mitochondrial Ca²⁺ uptake measured, during electrical stimulation, in MCU-OE and Empty CMs, transfected with mito-CaMeleon. (D) Representative trace of mtCa²⁺ dynamics during paced contractions. Panel (E) shows average steady-state Ca²⁺ levels estimated in Empty and MCU-OE CMs. Differences among groups were determined using Mann–Whitney test. (***P < 0.001; Empty: n = 36 CMs; MCU-OE: n = 33 CMs). (F) Confocal immunofluorescence of Empty vs. MCU-OE neonatal CMs treated with either saline solution or NE for 72 h. Cells were costained with antibodies to a-actinin (red signal) and flag (green signal). Nuclei were counterstained with DAPI (blue signal). (G) Effect of MCU-OE ± NE, on cell growth, relative to Empty—NE. Differences among groups were determined using Mann–Whitney test. (*P < 0.05; ***P < 0.001. n = 75 to 100 cells for group, from three independent cultures). (H) Western blotting on protein extracts from cultured Empty and MCU-OE CMs, in either absence or presence of NE. (I–K) Densitometry of the western blotting in (H). Differences among groups were determined using a Two-way ANOVA test with Holm’s correction for multiple comparisons. (*P < 0.05; **P < 0.01; ***P < 0.001. n = 4/6 samples from three independent cultures). (L) Quantification of NFATc3 nuclear translocation assessed in Empty and MCU-OE CMs, transfected with a plasmid encoding eGFP-NFATc3, in either absence (−) or presence (+) of NE. Differences among groups were determined using a Two-way ANOVA test with Holm’s correction for multiple comparisons. (**P <0.01; ***P < 0.001. n = from 400 to 500 CMs for group, from three independent cell cultures). (M) Confocal immunofluorescence on Empty and MCU-OE CMs, expressing a dominant negative Akt (DN-Akt), in the presence (+) or the absence (−) of NE. Cells were stained with an antibody to a-actinin (red signal). (N) Effect of DN-Akt expression on MCU-OE CM size, normalized to correspondingly treated Empty controls. Differences among groups were determined using Mann–Whitney tests. (***P < 0.001; ****P < 0.0001, n = 200 CMs for group, from three independent cell cultures).

Fig. 4.

In vivo MCU-OE enhances Akt phosphorilation via ROS. (A–E) In vitro experiments on cultured cardiomyocytes (CMs). (A) Quantification of mitochondrial ROS production by MTR fluorescence and effects of the mitochondrial ROS scavenger MitoTEMPO (mitoT). Differences among groups were determined using a Two-way ANOVA test with Holm’s correction for multiple comparisons. (***P < 0.001. Each value in the graph represents the average quantitation in 100 CMs, in three independent cell preparations). (B) Western blotting on protein extracts from cultured Empty and MCU-OE CMs ± the antioxidant MitoTEMPO. a-tubulin was used as loading control. (C and D) Densitometry of the western blotting in (B). Differences among groups were determined using a Two-way ANOVA test with Holm’s correction for multiple comparisons. (*P < 0.05; **P < 0.01; n = 6 samples from three independent cultures). (E) Relative CM growth in cultured Empty and MCU-OE cells, in the absence (−) or the presence (+) of NE or the antioxidant MitoTEMPO. Differences among groups were determined using a Two-way ANOVA test with Holm’s correction for multiple comparisons (***P < 0.001. Each individual value in the plot represents the average CM cross-sectional area calculated in at least 120 CMs/cell preparation, normalized to Empty—NE—mitoT group). (F–J) Ex vivo experiments on isolated hearts. (F) Fluorescent microscopy–based assessment of fluorescence intensity emitted by heart sections from Empty and MCU-OE hearts processed with HNE staining. Differences among groups were determined using a two-independent-sample Student t test (****P < 0.0001. n = 6 nonconsecutive sections from n = 4 hearts/group). (G) Western blotting on protein extracts from hearts of Empty and MCU-OE mice, in basal conditions. Actin was used to ensure equal protein loading. (H) Densitometry of the western blotting in (G). Differences among groups were determined using a two-independent-sample Welch test (****P < 0.0001; n = 5 hearts/group, in three independent experiments). (I) Western blotting on protein extracts from Empty vs. MCU-OE hearts, treated with the antioxidant MPG. Protein levels of phospho-Akt were normalized to those of total Akt. (J) Densitometry of the western blotting in (I). Differences among groups were determined using a two-independent-sample Student t test (n = 6 replicates/group, in three independent experiments).