Endurance exercise training increases cardiac energy metabolism through poorly understood mechanisms. Nitric oxide (NO) produced by endothelial NO synthase (eNOS) in cardiomyocytes contributes to cardiac adaptation. Here we demonstrate that the NO donor diethylenetriamine-NO (DETA-NO) activated mitochondrial biogenesis and function, as assessed by upregulated peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α), nuclear respiratory factor 1, and mitochondrial transcription factor A (Tfam) expression, and by increased mitochondrial DNA content and citrate synthase activity in primary mouse cardiomyocytes. DETA-NO also induced mitochondrial biogenesis and function and enhanced both basal and insulin-stimulated glucose uptake in HL-1 cardiomyocytes. The DETA-NO-mediated effects were suppressed by either PGC-1α or Tfam small-interference RNA in HL-1 cardiomyocytes. Wild-type and eNOS−/− mice were subjected to 6 wk graduated swim training. We found that eNOS expression, mitochondrial biogenesis, mitochondrial volume density and number, and both basal and insulin-stimulated glucose uptake were increased in left ventricles of swim-trained wild-type mice. On the contrary, the genetic deletion of eNOS prevented all these adaptive phenomena. Our findings demonstrate that exercise training promotes eNOS-dependent mitochondrial biogenesis in heart, which behaves as an essential step in cardiac glucose transport.
- endothelial nitric oxide synthase
- exercise training
- nitric oxide
- mitochondrial biogenesis
- glucose uptake
moderate endurance exercise training fosters cardiovascular health and is prescribed for the prevention and the treatment of heart diseases (13). Heart adaptation to chronic exercise implies increased glucose transport and mitochondrial respiratory capacity, resulting in improved efficiency of cardiac muscle contraction, but our understanding of the underlying mechanisms is very limited and currently debated (1, 34). Metabolic remodeling of skeletal muscle to exercise training is the consequence of a coordinated genetic response that boosts peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α)-dependent mitochondrial biogenesis, increases the size and number of mitochondria, and induces the GLUT4 (insulin-sensitive) isoform of the glucose transporters (15). Wu et al. has additionally shown that the transcriptional coactivator PGC-1α, that is induced by exercise training in skeletal and cardiac muscle, differently contributes to the modulation of endoplasmic reticulum homeostasis in these tissues (48). Thus the distinctive molecular mechanisms governing cardiac adaptation to exercise training, including the exercise-dependent increase of myocardial substrate utilization and enhanced metabolic capacity, are largely unresolved.
Nitric oxide (NO), tonically synthesized in cardiomyocytes by the constitutively expressed endothelial NO synthase (eNOS) (5, 27), mediates key aspects of the cardiac adaptive responses to exercise (13). We previously demonstrated that NO increases PGC-1α, nuclear respiratory factor-1 (NRF-1), and mitochondrial transcription factor A (Tfam) expression, triggering mitochondrial biogenesis in a variety of cells (28, 29), and found reduced PGC-1α levels in the heart of eNOS null mutant (eNOS−/−) mice together with reduced mitochondrial DNA (mtDNA) content and O2 consumption compared with wild-type (WT) animals (29).
Exercise training induces eNOS mRNA and protein expression in mouse left ventricular tissue (18). Further studies have observed increased eNOS activation in hearts from trained mice and suggested that exercise-stimulated GLUT4 translocation to the cell membrane and glucose uptake in the heart could be mediated via a phosphatidylinositol 3-kinase-protein kinase B (Akt)-NO-dependent mechanism (49). In spite of this evidence, the contribution of the NO-mediated mitochondrial biogenesis in heart adaptations to exercise and its possible role in adaptive changes of glucose metabolism remain to be investigated. Here we show that the activation of a complete eNOS-dependent transcriptional program increasing cardiomyocyte mitochondrial mass and function is crucial for cardiac metabolic adaptation to exercise.
MATERIALS AND METHODS
Mice and exercise protocol.
Thirty-six adult (8-wk-old) male WT (F2 Hybrid B6.129S2 obtained from crossing C57BL/6J and 129S1/SvImJ mice) and male B6.129P2-Nos3tm1Unc/J (eNOS−/−) mice (38) (Jackson Laboratory) were treated according to the European Union guidelines and with the approval of the Institutional Ethical Committee. Body weight and food consumption were monitored throughout the experimental period. Swim training was chosen as an endurance training model since it is widely adopted for the study of physiological cardiac adaptation phenomena (1, 6, 13, 49). Both WT and eNOS−/− mice (n = 18 mice/group) were assigned randomly to either swim training or to have no lifestyle modifications. Mice swam one time a day for 5 days/wk in a graduated protocol (6) starting at 10 min daily, with a 10-min increase each day until 90 min daily at the end of the 2nd wk. Next, mice swam 30 days on the full training regimen (90 min daily, 5 days/wk). Swimming sessions were supervised to prevent drowning.
Insulin tolerance test.
Insulin tolerance test (ITT) was performed 2 days after the last bout of exercise in 8-h-fasted animals (n = 6 mice/group) by an intrperitoneal injection of insulin (0.5 U/kg body wt diluted in 0.9% NaCl) (Actrapid HM; Novo Nordisk). Blood samples were obtained from the tail vein at 0, 5, 10, 15, 20, 30, 40, 60, and 90 min postinjection and used to determine plasma glucose levels.
Tissue glucose utilization index.
At the end of the training period and 2 days before the clamp studies, a catheter was inserted in the right femoral vein under general anesthesia with pentobarbital sodium. Tissue glucose uptake studies were performed on mice under conscious and unstressed conditions after an 8-h fast. The nonmetabolizable glucose analog 2-deoxy-d-[1-3H]glucose ([3H]DG; 10 μCi) (Amersham Biosciences) was injected as an intravenous bolus in the basal condition or after hyperinsulinemic euglycemic clamp as previously described with minor modifications (44). One hundred twenty minutes after tracer injection, animals were killed, and tissues were quickly collected in liquid nitrogen and kept at −80°C for subsequent analysis. The glucose utilization index was derived from the amount of [3H]DG-6-phosphate ([3H]DGP) measured in the various tissues as previously described (14) thus using the accumulation of [3H]DGP as an index of the glucose metabolic rate.
Blood glucose was measured using the One Touch Blood Glucose Monitoring System (LifeScan Italia; Johnson & Johnson Medical). Plasma insulin was measured by radioimmunoassay (Linco Research). Plasma free fatty acids (FFA) were quantified using the NEFA-HR(2) kit (Wako Chemicals).
Cell cultures and treatments.
Adult mouse cardiomyocytes were isolated and cultured as previously described (10). Hearts were excised from heparinized and anesthetized mice, mounted on a steel cannula, and perfused for 3 min with a Ca2+-free buffer, previously gassed with 95% O2-5% CO2. Enzymatic digestion was performed for 10 min by adding liberase blendzyme 4 (Hoffmann-La Roche). The left ventricle was removed, cut into chunks, and further digested for 10 min at 37°C. The dispersed myocytes were resuspended in buffers containing progressively increased concentration of Ca2+. Cardiomyocytes were plated in laminin-coated dishes with MEM containing 2.5% FBS and antibiotics. After 1 h of culture, the medium was changed to FBS-free MEM. Primary cardiomyocytes were exposed to test drugs the day after seeding. HL-1 cells (a gift from W. C. Claycomb) (9) were also used in the present study since they are easily amenable to genetic manipulation and can be cultured for prolonged time while keeping a differentiated cardiomyocyte phenotype and contracting in vitro (47). HL-1 cardiomyocytes were plated in fibronectin/gelatin-coated flasks and grown to 70–80% confluence in Claycomb medium (JRH Biosciences) supplemented with 100 μM norepinephrine, 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, and 10% FBS (JRH Biosciences) as described (10). Cells were exposed to diethylenetriamine-NO (DETA-NO), S-nitroso-N-acetylpenicillamine (SNAP) (both from Sigma-Aldrich), or vehicle as described in results. NO donor treatment left unchanged cardiomyocyte survival in culture (data not shown). For PGC-1α and Tfam knockdown experiments, HL-1 cells were transfected 24 h after seeding with 50 nM PGC-1α or Tfam small-interference RNA (siRNA) SMARTpool (Dharmacon) or siGENOME nontargeting (NT) siRNA following the manufacturer's instruction. At the end of the treatments, HL-1 cells were harvested for further analysis.
Cell glucose uptake.
HL-1 cells were serum-starved for 2 h and incubated in the presence or absence of 300 nM insulin (Novo Nordisk) for 30 min and then with 1.5 μCi/ml [3H]DG (Amersham Biosciences) for 15 min. Cells were washed with ice-cold phosphate-buffered saline and lysed in 0.5 M NaOH. Radioactivity was measured by liquid scintillation counting (Wallac, Turku, Finland). Each condition was assayed in three independent experiments in triplicate.
Cell GLUT4myc translocation.
HL-1 cells were seeded on fibronectin-coated cover slips and transfected with 1.5 μg pCMV-GLUT4myc plasmid or empty vector (a gift from J. E. Pessin, Albert Einstein College of Medicine, New York, NY) in Opti-MEM I Reduced Serum medium with Lipofectamine-2000 (Invitrogen). Cells were treated with vehicle or 100 μM DETA-NO for 72 h and subsequently incubated with or without 300 nM insulin for 30 min. Cover slips were fixed in 4% paraformaldehyde, saturated with 5% horse serum-5% bovine serum albumin in PBS, incubated with Myc-Tag 9B11 mouse monoclonal antibody (Cell Signaling Technology) then with Alexa Fluor 546 anti-mouse IgG, counterstained with Hoechst 33342, mounted with ProLong Gold antifade reagent (all from Invitrogen), and analyzed with the confocal laser-scanning microscope Leica DMI6000 CS SP8 (Leica Microsystem). Image analysis was performed with the NIH ImageJ software.
Quantitative reverse transcription-PCR reactions were performed as previously described (10, 42) and run with the iQ SybrGreenI SuperMix (Bio-Rad Laboratories) on an iCycler iQ Real Time PCR detection system (Bio-Rad Laboratories, Hercules, CA). Calculations were performed by a comparative method (2−ΔΔCt) using 18S rRNA as an internal control. Primers (Table 1) were designed using Beacon Designer 2.6 software (Premier Biosoft International).
Western blotting was performed on protein extracts as previously described (42). Primary antibodies were as follows: anti-cytochrome c (Cyt c) (BD Pharmingen), anti-cytochrome c oxidase IV (COX IV) (Molecular Probes), anti-eNOS (Transduction Labs), anti-phospho-Akt (Ser473) and anti-Akt (Cell Signaling Technology), anti-GLUT4, anti-PGC-1α (both Santa Cruz Biotechnology), anti-Tfam (Aviva Systems Biology), anti-GAPDH (Histoline Laboratories), and anti-β-actin (Sigma-Aldrich). Myocardial membrane proteins were extracted using a membrane protein extraction kit (Biovision). The membrane protein was subsequently used for Western blot analysis of GLUT4.
mtDNA copy number was measured by means of quantitative PCR from the Cyt b mtDNA gene compared with the large ribosomal protein p0 (36B4) nuclear gene as previously described (10).
Electron microscopy studies were conducted on heart tissue processed as previously described (10). Ultrathin sections were doubly stained with uranyl acetate and lead citrate and examined under a Philips CM10 transmission electron microscope (Philips, Eindhoven, the Netherlands). For morphometric analysis of mitochondria, randomly selected areas of tissue derived from three animals per group were photographed at a ×11,500 magnification and analyzed with the NIH Image software.
Citrate synthase assay.
Citrate synthase activity was measured in either tissue or whole cell extracts and expressed as nanomole citrate produced per minute per milligram protein as described (42).
Results were expressed as means ± SE unless otherwise specified. Data were analyzed by GraphPad Prism 5.0 software using either unpaired Student's t-test or one-way ANOVA or two-way ANOVA with Newman-Keuls' post hoc test as appropriate.
NO donors induce mitochondrial biogenesis in mouse cardiomyocytes.
NO is a crucial mediator of mitochondrial biogenesis in various cell types of mammals (28, 29). To investigate its contribution to mitochondrial biogenesis and function in cardiac cells, we exposed primary mouse cardiomyocytes to the slow-acting NO donor DETA-NO (half-time of NO release, 20 h at 37°C) (17). DETA-NO (100 μM) treatment for 24 h induced PGC-1α, NRF-1, and Tfam mRNA expression and increased mtDNA content and citrate synthase activity (i.e., indexes of mitochondrial mass and function) in primary cardiomyocytes (Fig. 1A). This concentration of DETA-NO is expected to result in ∼100 nM NO in vitro, which is within the wide range of NO concentrations that is known to mediate discrete biological responses in vivo (43). Experiments in HL-1 cardiomyocytes (a murine cardiac-derived cell line that displays features similar to those of adult cardiomyocytes in terms of organized structure and ability to contract in culture) (47) were preliminary conducted to assess dose and time dependency of the DETA-NO-mediated effect on gene expression (data not shown). Treatment with 100 μM DETA-NO for 72 h increased PGC-1α, NRF-1, Tfam, COX IV, Cyt c, F1F0-ATP synthase, and mitofusin (Mfn)-1 and -2 mRNA levels in HL-1 cells (Fig. 1B). Moreover, treatment with DETA-NO dose-dependently increased PGC-1α, COX IV, and Cyt c protein levels (Fig. 1C) as well as mtDNA amount and citrate synthase activity (Fig. 1D). Similar results were obtained treating HL-1 cells for 72 h with a structurally unrelated NO donor (SNAP, 1–100 μM) (2) (data not shown).
The NO donor DETA-NO induces cardiomyocyte glucose uptake.
We then investigated whether DETA-NO treatment affects glucose metabolism in HL-1 cardiomyocytes. Both basal and insulin-dependent [3H]DG uptake was increased by DETA-NO (100 μM for 72 h) (Fig. 2A). DETA-NO also promoted GLUT4 recruitment to the cell membrane in unstimulated HL-1 cells (Fig. 2C) and further enhanced the insulin-mediated GLUT4 translocation (Fig. 2C). The serine/threonine protein kinase Akt/PKB is a crucial regulator of insulin-mediated GLUT4 trafficking (46). We observed that DETA-NO significantly increased Akt phosphorylation over basal values and enhanced Akt phosphorylation in insulin-stimulated HL-1 cells (Fig. 2B). These data demonstrate that NO regulates multiple steps of the glucose uptake pathway in cardiac cells.
Mitochondrial biogenesis is necessary for DETA-NO-mediated glucose uptake in HL-1 cardiomyocytes.
To assess whether the NO-mediated induction of mitochondrial biogenesis is relevant for cardiac glucose uptake, HL-1 cells were transfected with siRNA against PGC-1α. The effectiveness of PGC-1α knockdown by siRNA oligos was evaluated by measuring both mRNA and protein levels. Compared with cells transfected with NT siRNA, PGC-1α knockdown cells displayed an ∼60% reduction in PGC-1α mRNA (Fig. 3A). PGC-1α protein levels were concomitantly reduced by ∼55% (Fig. 3A). Furthermore, PGC-1α siRNA fully prevented the DETA-NO-mediated increase of PGC-1α expression (Fig. 3A). PGC-1α siRNA also prevented the insulin-mediated increase of glucose uptake and fully blocked the DETA-NO-increased glucose uptake both in basal conditions and under insulin (Fig. 3C).
PGC-1α is known to affect GLUT4 expression (23). To exclude that the inhibitory effects of PGC-1α knockdown were the result of reduced GLUT4 expression, HL-1 cells were transfected with Tfam siRNA. Tfam is critical for mitochondrial biogenesis, being required for mtDNA replication (37). Furthermore, Shi et al. (39) have reported that GLUT4 translocation, but not GLUT4 expression, is impaired in Tfam siRNA-transfected adipocytes. Tfam siRNA reduced Tfam mRNA and protein levels by ∼80 and 70%, respectively, compared with NT siRNA and prevented the DETA-NO-mediated increase of Tfam expression in HL-1 cells (Fig. 3A). Tfam knockdown also led to an ∼30% decrease in mtDNA copy number (Fig. 3B) and abolished the DETA-NO-mediated increase of mtDNA amount (Fig. 3B). Moreover, a significant impairment in citrate synthase activity was seen in Tfam knockdown cells (data not shown). Interestingly, Tfam depletion, although not significantly affecting basal glucose uptake, was able to reduce the insulin-stimulated glucose uptake and impair the stimulatory effect of DETA-NO (Fig. 3C). These data demonstrate that efficient mitochondrial biogenesis is essential for NO to induce glucose uptake in HL-1 cardiomyocytes.
Exercise training increases mitochondrial biogenesis in heart of WT but not eNOS−/− mice.
To investigate the contribution of the NO-mediated mitochondrial biogenesis in cardiometabolic adaptations in vivo, we randomly assigned WT and eNOS−/− mice to sedentary conditions or to 6 wk of graduated swim training. Aged eNOS−/− mice (from 12.5 mo onward) have reduced maximal work capacity in acute bouts of treadmill exercise (30). The graduated swim training protocol we used did not lead to exhaustion of the mice, and we did not observe an obvious difference between the swimming performance of our 2-mo-old eNOS−/− mice compared with WT mice (data not shown). The body weight was comparable between WT and eNOS−/− mice and did not significantly change after endurance exercise training (Table 2). As expected (12, 27), the fasting plasma concentrations of both insulin and FFA were statistically higher in eNOS−/− than in WT mice (Tables 2 and 3). Food consumption and fasting concentrations of glucose, insulin, and FFA were unaffected by endurance exercise in mice of both genotypes (Tables 2 and 3). Insulin injection significantly reduced plasma FFA levels in trained WT but not eNOS−/− mice (Table 3).
We found that eNOS mRNA and protein levels are significantly higher in left ventricle of swim-trained than in sedentary WT mice (Fig. 4A). Increased eNOS expression was accompanied by increased expression of PGC-1α, NRF-1, Tfam, and Mfn-2 and by increased mtDNA content (Fig. 4, B and C). Mitochondrial biogenesis parameters were lower in left ventricle of sedentary eNOS−/− than sedentary WT animals (Fig. 4, B and C) as described in other tissues (28, 29). Interestingly, the endurance exercise training failed to increase mitochondrial biogenesis in left ventricle of eNOS−/− mice (Fig. 4, B and C). Electron microscopy showed smaller and fewer mitochondria with altered cristae organization in left ventricles of eNOS−/− mice compared with WT mice. Moreover, lipid droplets were present among mitochondria in knockout mice (Fig. 4D). The morphometric analysis confirmed that, in the cardiomyocytes of the left ventricle, mitochondrial mass (mitochondrial volume density and number) was lower in sedentary eNOS−/− than in sedentary WT animals. Although mitochondrial volume density and number were increased by exercise training in sedentary WT mice (Fig. 4D), these parameters did not differ between exercise-trained and sedentary eNOS−/− mice (Fig. 4D). These results further support our previous finding that 4 wk treadmill exercise training increases mitochondrial biogenesis and function in left ventricles of WT but not eNOS−/− middle-aged mice (10), demonstrating the necessary role of eNOS-NO signals in the heart mitochondrial biogenic response to different types of endurance training.
Exercise training does not improve basal and insulin-stimulated glucose uptake in the heart of eNOS−/− mice.
Because we found that mitochondrial biogenesis is an essential process for the cardiomyocyte glucose uptake adaptation in vitro, and eNOS-derived NO is a critical mediator of cardiac-adaptive mitochondrial biogenesis, we assessed glucose metabolism in left ventricles of sedentary or trained WT and eNOS−/− mice. Basal glucose uptake was greatly increased after exercise training in WT mice (Fig. 5A). On the contrary, it was not affected by chronic exercise in knockout mice (Fig. 5A). To confirm the relevance of eNOS in basal glucose uptake, we examined the content of the GLUT1 (insulin-independent) glucose transporter in left ventricle. Unlike WT mice, which displayed a significant increase of cardiac GLUT1 expression after exercise training, eNOS−/− mice showed unmodified cardiac GLUT1 content (Fig. 5B).
The whole body glucose uptake was then measured after insulin perfusion. The ITT showed higher insulin sensitivity, resulting in lower plasma glucose levels, in sedentary WT than in sedentary eNOS−/− mice (Fig. 5C). Exercise training induced a marked increase in whole body insulin sensitivity in WT mice, with a higher drop in plasma glucose levels compared with sedentary WT mice (Fig. 5C). The insulin-sensitizing effect of exercise training was not found in eNOS−/− mice (Fig. 5C). Next, we measured the insulin-dependent glucose uptake in the left ventricle. This was markedly increased by exercise training in WT mice but remained unchanged in trained eNOS−/− mice (Fig. 5D) Consistently, GLUT4 mRNA levels were increased by exercise training in left ventricle of WT unlike eNOS−/− mice (Fig. 5E). GLUT4 translocation, measured as membrane fraction of GLUT4, behaves similarly (data not shown). These results demonstrate that signals generated by eNOS are necessary for the adaptive modifications of glucose transport in the heart of exercise-trained mice.
Cellular and molecular mechanisms underlying the cardiometabolic benefits associated with regular exercise remain largely unresolved. Exercise training improves cardiac insulin sensitivity by modulating GLUT4 translocation to the plasma membrane and increasing glucose uptake in the myocardium (49). NO has been suggested to play important roles in these processes (49). The novelty of our study lies in three main aspects. First, we demonstrate that mitochondrial biogenesis is key in the regulation of cardiomyocyte glucose uptake by NO both at baseline and under insulin-stimulated conditions. Second, the eNOS-derived NO is necessary to improve mitochondrial biogenesis and glucose mobilization in trained hearts. Third, the demonstration that mitochondrial volume density and number is increased in cardiac myocytes from trained mice undermines the general perception that heart cannot augment its mitochondrial content because of the large share of cardiomyocyte volume occupied by the organelles to maintain its continuous mechanical work.
Only a few studies have so far addressed the effects of exercise training on rodent cardiac mitochondrial ultrastructure/density, with controversial results probably depending on exercise intensity (3, 4, 32). Applying a morphometric analysis on electron microscopy specimens, we found evidence of increased mitochondrial size and number in cardiac myocytes of swim-trained WT but not eNOS−/− mice. Biphasic changes have been described in PGC-1α, Tfam, and Cyt c proteins, with decreased levels by 5-day and increased levels by 10-day treadmill running in left ventricles of female rats (45). A 14-day swim training with a ramp protocol recently demonstrated the activation of a metabolic gene set, including PGC-1α and its downstream targets in the heart (6), including the respiratory chain NADH dehydrogenase (Ndufs2 and Ndufv2 subunits) and the F1F0-ATP synthase (ATP5o subunit) that interestingly plays a role in maturation of mitochondrial cristae (8). Furthermore, a proteomics analysis investigating the rat cardiac muscle adaptation under different intensities of swimming exercise (8 wk after graduated overload) recently showed that the majority (27%) of regulated proteins were mitochondrial, including the mitochondrial import receptor subunit TOM34 (35), which plays a role in the import of cytosolically synthesized preproteins in mitochondria. Because the assembly of the mitochondrial network is an intricate process, proper assessment of mitochondrial biogenesis requires the comprehensive evaluation of multiple parameters, including the mitochondrial biogenic transcriptional machinery, the expression of proteins of the respiratory chain and/or mitochondrial transport proteins, and indexes of mitochondrial mass (mtDNA content or morphological evidence of mitochondrial size and number) and functional activity (22, 40). The rate of mitochondrial protein synthesis is proposed by some authors as a more appropriate strategy to assess mitochondrial biogenesis that circumvents issues resulting from mitochondrial remodeling and mitophagy (25). However, the latter approach acquires significance only in the context of the assessment of multiple parameters as reported above (22, 40).
We found that prolonged exposure of primary cardiomyocytes or HL-1 cells to NO donors promotes mitochondrial biogenesis and function, as assessed measuring the expression of nuclear-encoded transcription factors and coactivators (NRF-1, Tfam, PGC-1α), mitochondrial inner membrane proteins (i.e., ATP synthase and respiratory proteins like COX IV and Cyt c), and mtDNA amount and citrate synthase activity. In parallel, glucose uptake and GLUT4 translocation are increased by NO. Notably, specific silencing of either PGC-1α (that co-ordinately activates the mitochondrial biogenic program) or Tfam (which is crucial for mtDNA transcription and maintenance) completely blocks the NO-dependent glucose uptake in HL-1 cells. Our findings are in line with observations from diverse cell-based models showing that mtDNA replication is an essential process for glucose transport. In fact, depletion of mtDNA in rat L6 skeletal myocytes by low doses of ethidium bromide, which inhibits mtDNA replication without affecting nuclear DNA (50), causes a drastic decrease in basal and insulin-stimulated glucose uptake and GLUT4 translocation (33). Murine C2C12 myotubes with defective mtDNA replication or oligomycin-induced mitochondrial dysfunction show impaired glucose utilization, with a marked reduction of GLUT4 expression and translocation (20). Finally, knockdown of Tfam impairs both mitochondrial respiratory function and the insulin-stimulated GLUT4 trafficking and glucose uptake in mouse adipocytes (39). Previous studies in isolated rat left ventricular papillary muscles showed that NO donors and cGMP analogs increase cell surface GLUT4 and glucose uptake (19) and that eNOS activation contributes to cardiomyocyte glucose uptake (19). Our in vitro evidences demonstrate that the induction of cardiomyocyte mitochondrial biogenesis is critical for the modulation of glucose uptake mediated by NO.
By examining the myocardium of swim-trained mice, we found increased eNOS expression together with: 1) increased mitochondrial biogenesis; 2) increased mitochondrial volume and number in cardiac myocytes; and 3) increased GLUT1 and GLUT4 expression and both basal and insulin-stimulated glucose uptake. Of note, the genetic ablation of eNOS (eNOS−/− mice) abrogates all of these metabolic adaptive events. The NO generated by eNOS is known to mediate the length-dependent increase in cardiac contraction force (5) and takes part in multiple aspects of cardiovascular remodeling evoked by endurance exercise (13). Zhang et al. (49) reported that graduated swim training increases eNOS expression and phosphorylation in rat heart and sensitizes myocardial response to insulin by enhancing glucose uptake via Akt-eNOS-dependent mechanisms. Our findings confirm and extend those results. The observation that the NO donor increases Akt phosphorylation in cardiac myocytes suggests that a positive feedback effect would arise favoring training-induced myocardial insulin sensitization. Interestingly, we further demonstrate that eNOS-derived NO plays an obligatory role in metabolic adaptation of cardiac muscle to exercise.
We recognize that our study has some limits, since the global eNOS knockout mouse model does not allow to dissect the contribution of eNOS in the different cell types present in heart. Therefore, we cannot rule out that eNOS induction in heart endothelial cells of exercised mice could favor NO release, that acting in a paracrine manner may promote the cardiomyocyte mitochondrial biogenesis, or even that mitochondrial adaptations to chronic exercise could occur in cardiac endothelial cells besides cardiomyocytes.
A recent study revealed that autophagy, the intracellular recycling system that intervenes in organelle turnover and cell quality control, is required for metabolic adaptations (i.e., mitochondrial biogenesis and angiogenesis) to endurance exercise training in skeletal muscle (21). Autophagy and mitochondrial biogenesis are strictly interconnected cellular processes (24) and are coregulated to replace old inefficient mitochondria with new, fuel-efficient ones. The possible effects of regular exercise on heart autophagy and its cross talk with NO-dependent mitochondrial biogenesis could be the matter of future investigation.
Challenging the long-standing view that adult heart mitochondria are relatively static, most recent reports show in turn that they retain their ability to undergo fission and fusion, suggesting the relevance of mitochondrial dynamics in heart pathophysiology and cardioprotection (31). Accordingly, together with evidence of ultrastructural changes in cardiomyocyte mitochondria, we found that exercise increases cardiac expression of Mfn-2, an essential regulator of fusion and biogenesis in the mitochondrial network, in an eNOS-dependent manner. Thus eNOS-derived NO might take part in multiple PGC-1α-dependent mitochondrial adaptations to exercise (i.e., mitochondrial biogenesis and fusion) in cardiac myocytes, in analogy to what has been observed in the aorta of trained mice (26).
Most recently, Kasahara et al. (16) reported that mitochondrial fusion directs cardiomyocyte differentiation. Accordingly, mitochondrial activity is required for the differentiation and maturation in cardiomyocytes of a cardiac stem cell population, and this phenomenon was increased by NO (36). Ongoing studies suggest that eNOS-mediated mechanisms are implicated in swimming exercise-evoked activation of endogenous cardiac stem/progenitor cells (13). The role of eNOS-mediated mitochondrial biogenesis in these phenomena taking part in the beneficial adaptation of heart to exercise remains to be investigated.
It is worth to note at this point that, different from physiological cardiac hypertrophy in response to exercise training, pathological cardiac hypertrophy resulting from transaortic constriction is associated with reduced PGC-1α expression (6). This suggests that increasing the cardiomyocyte energetics via exercise-induced PGC-1α cascade could be investigated as a therapeutic approach to improve the functional resilience to cardiac dysfunction. Interestingly, exercise training in mice can reverse the contractile abnormalities associated with diabetic cardiomyopathy (41), a condition characterized by mitochondrial dysfunction in cardiomyocytes (11) and derangements in fuel utilization because of GLUT4 depletion (7).
In summary, our results provide novel insights into the mechanisms governing heart glucose homeostasis after exercise training, further supporting the health effects of the NO-dependent mitochondrial biogenesis (10, 42). The effects of exercise training on metabolic cardiac adaptation mediated by the eNOS signaling system are worth research also in pathological conditions and in clinical settings.
This work was supported by the Ministero dell'Istruzione, dell'Università e della Ricerca (Italy), Grants 2009E48P9M_001 (E. Nisoli), 2009E48P9M_003 (A. Valerio), and 2007BRR57M_004 (R. Vettor) and the CARIPARO Foundation (R. Vettor). L. Tedesco received a research contract cofinanced by NicOx Research Institute and Regione Lombardia.
No conflicts of interest, financial or otherwise, are declared by the authors.
R.V. and E.N. contributed to the conception and design of the research; R.V., M.R., E.T., and E.N. analyzed the data; R.V., A.V., and E.N. interpreted the results of the experiments; R.V., A.V., E.T., and E.N. drafted the manuscript; R.V., A.V., M.O.C., and E.N. edited and revised the manuscript; R.V., A.V., M.O.C., and E.N. approved the final version of the manuscript; A.V. and E.T. prepared the figures; M.R., E.T., M.G., M.O., L.T., C.R., A.F., R.F., and R.S. performed the experiments.
- Copyright © 2014 the American Physiological Society