Myocardial Contractile Dysfunction Is Associated With Impaired Mitochondrial Function and Dynamics in Type 2 Diabetic but Not in Obese PatientsCLINICAL PERSPECTIVE
Background—Obesity and diabetes mellitus are independently associated with the development of heart failure. In this study, we determined the respective effects of obesity, insulin resistance, and diabetes mellitus on the intrinsic contraction and mitochondrial function of the human myocardium before the onset of cardiomyopathy.
Methods and Results—Right atrial myocardium was obtained from 141 consecutive patients presenting no sign of cardiomyopathy. We investigated ex vivo isometric contraction, mitochondrial respiration and calcium retention capacity, and respiratory chain complex activities and oxidative stress status. Diabetes mellitus was associated with a pronounced impairment of intrinsic contraction, mitochondrial dysfunction, and increased myocardial oxidative stress, regardless of weight status. In contrast, obesity was associated with less pronounced contractile dysfunction without any significant perturbation of mitochondrial function or oxidative stress status. Tested as continuous variables, glycated hemoglobin A1C, but neither body mass index nor the insulin resistance index (homeostasis model assessment–insulin resistance), was independently associated with cardiac mitochondrial function. Furthermore, diabetes mellitus was associated with cardiac mitochondrial network fragmentation and significantly decreased expression of the mitochondrial fusion related protein MFN1. Myocardial MFN1 content was inversely proportional to hemoglobin A1C.
Conclusion—Worsening of intrinsic myocardial contraction in the transition from obesity to diabetes mellitus is likely related to worsening of cardiac mitochondrial function because impaired mitochondrial function and dynamics and contractile dysfunction are observed in diabetic patients but not in “metabolically healthy” obese patients at early stage in insulin resistance.
Obesity and type 2 diabetes mellitus (DM) have reached epidemic levels worldwide. These 2 metabolic disorders are independent risk factors for the development of heart failure.1–3 Epidemiological and clinical studies strongly support the existence of obesity and diabetic cardiomyopathies unrelated to coronary artery disease, hypertension, and other comorbidities.4,5
Clinical Perspective on p 564
Studies in rodent models of obesity and DM have identified intrinsic cardiomyocyte dysfunctions secondary to alterations in energy substrate utilization, mitochondrial dysfunction, increased oxidative stress, and intracellular accumulation of lipotoxic byproducts. Similarly, human studies have shown that dysregulation of the energy conversion process is one of the major characteristics of the failing heart of subjects with cardiomyopathy related to DM or obesity.6,7
Exploration of human DM atrial myocardium clearly showed reduced mitochondrial respiration and increased oxidative stress, thus identifying a role for mitochondrial dysfunction in the pathophysiology of heart failure in DM patients.8 More recently, obesity has been linked to cardiac mitochondrial dysfunction and oxidative damage in patients.9 However, because obesity and hyperglycemia overlapped in these studies,8,9 it was not possible to distinguish whether mitochondrial alterations were strictly related to obesity, insulin resistance, or DM.10 Furthermore, no evidence was provided in these studies for a potential link between mitochondrial dysfunction and poor myocardial contractile performance in patients with metabolic disorders.
We investigated ex vivo isometric contraction of myocardium trabeculae, mitochondrial respiration and calcium retention capacity in cardiac permeabilized fibers, and electron transport chain complex activities and myocardial oxidative stress status in atrial myocardium to determine the effects of obesity, insulin resistance, and DM on intrinsic myocardial contraction and cardiac mitochondrial function in a large population without overt signs of cardiomyopathy. Finally, given the impact of DM on mitochondrial function and the important link between mitochondrial function and dynamics, we explored the cardiac mitochondrial network organization in diabetic patients.
Patient Selection and Atrial Appendage Collection
From March 2012 to August 2013, all patients undergoing coronary artery bypass graft surgery in the heart surgery ward at the Lille University Regional Hospital were prospectively included in the study if they presented none of the following exclusion criteria: (1) type 1 diabetes mellitus; (2) overt signs of cardiomyopathy, that is, presenting with clinical signs of heart failure, left ventricular ejection fraction <50%, left atrial dilatation (>40 mm), systolic pulmonary artery pressure >40 mm Hg, or brain natriuretic peptide level >100 ng/L; (3) history of atrial fibrillation/atrial flutter; or (4) stenosis >50% of the right coronary artery. Patients were also excluded when the amount of tissue was too small to assess mitochondrial respiration, that is, a biopsy weight <25 mg.
Clinical, biological, and echocardiographic data were obtained at the preoperative consult (details are given in the online-only Data Supplement). The study was approved by the local Patient Protection Committee (Lille, France), and patients gave informed consent.
Fasting plasma glucose (>7 mmol/L) or the use of diabetes medications was used for DM diagnosis. Insulin resistance was assessed by the homeostasis model assessment–insulin resistance (HOMA-IR), calculated as fasting insulinemia (mUI/L) times fasting glycemia (mmol/L) divided by 22.5. Obesity was defined as body mass index (BMI) >30 kg/m2, overweight as BMI between 25 and 30 kg/m2, and normal weight as BMI <25 kg/m2.
Right atrial tissue was obtained during cannulation of the right atrium in preparation for cardiopulmonary bypass. Samples of the appendage were immediately prepared for functional studies, and the remaining tissue was frozen in liquid N2.
Contractile Function of Atrial Myocardium
Trabeculae were isolated from the atrial appendage (length ranging between 5 and 10 mm; diameter ranging between 400 and 600 μm) and attached to a force transducer in organ baths superfused with Krebs-Henseleit buffer (95% O2, 5% CO2, 37°C). After stabilization, the twitch force generated by trabeculae was assessed during isometric contractions after gradual stretching to the length at which force generation was optimal. Contraction (twitch force and time to peak tension) and relaxation (times to 50% and 90% maximal relaxation and exponential relaxation constant [τ]) parameters were defined as previously described11 and measured in steady-state conditions at stimulation rates of 60 and 180 bpm.
Mitochondrial Functional Studies
Mitochondrial respiration rates and the mitochondrial calcium retention capacity (mCRC) were explored on permeabilized atrial fibers as previously described.12
Thin bundles were separated from atrial trabeculae in ice-cold relaxing solution. Bundles were then permeabilized for 30 minutes with saponin solution and washed 3 times for 5 minutes in respiration medium. The relaxing solution contained 2.77 mmol/L CaK2EGTA, 7.23 mmol/L K2EGTA, 6.56 mmol/L MgCl2, 0.5 mmol/L dithiothreitol, 50 mmol/L K-MES, 20 mmol/L imidazole, 20 mmol/L taurine, 5.3 mmol/L Na2ATP, and 15 mmol/L phosphocreatine, pH 7.1. The respiration medium contained 110 mmol/L sucrose, 20 mmol/L HEPES, 10 mmol/L KH2PO4, 20 mmol/L taurine, 3 mmol/L MgCl2, 60 mmol/L MES-K, 0.5 mmol/L EGTA, and 0.1% BSA, pH 7.1.
Respiration rates were measured in permeabilized fibers at 37°C with different substrates and inhibitors. Respiration rates were expressed in picomoles of O2 per second per 1 mg dry weight. Data acquisition and analysis were performed with Datlab4 software (Oroboros, Innsbruck, Austria). Substrate or inhibitors for respiratory experiments were added in the respiration medium in a step-by-step manner with the use of microsyringes. Chemical agents were administered sequentially. At the end of each experiment, respirometer chambers were calibrated for zero O2 content with dithionite.
Two mitochondrial substrate/inhibitor protocols were applied. In the carbohydrate protocol, substrate combinations were used for electron flow through complexes I and II. Complex I–dependent state 2 respiration (Vpyr+mal) was determined in the presence of 10 mmol/L pyruvate and 2 mmol/L malate. Complex I–dependent state 3 respiration (Vpyr+ADP) was determined as a phosphorylation-stimulated respiration rate in the presence of 5 mmol/L ADP. In all groups, the lack of a significant increase in respiration after the addition of cytochrome c (10 μmol/L) confirmed the integrity of the outer mitochondrial membrane. Subsequently, complex I was inhibited by rotenone (0.5 μmol/L), and complex II–dependent state 3 respiration (Vsucc+ADP) was determined in the presence of 10 mmol/L succinate. Complex IV–dependent uncoupled state of respiration (VCox) was determined in presence of 2 mmol/L ascorbate, 0.5 mmol/L TMPD, the complex III inhibitor antimycin-A (2.5 μmol/L), and the oxidative phosphorylation uncoupler FCCP (2 μmol/L; VTMPD-asc). TMPD-ascorbate auto-oxidation (VKCN) was determined as the O2 consumption in presence of 1 mmol/L KCN added to the VTMPD-asc medium. VCox was calculated as VTMPD-asc minus VKCN. Because antimycin-A inhibits complex III, VCox estimated complex IV–related maximum respiration rate.
In the fatty acid protocol, respiration was measured with palmitoyl-L-carnitine (80 μmol/L) and malate (2 mmol/L; Vpalm), followed by the addition of ADP (5 mmol/L; Vpalm+ADP) feeding electrons into electron transferring flavoprotein and complex I.
The respiratory control ratios (RCRpyr and RCRpalm) were calculated by dividing state 3 respiration by state 2 respiration in the respective carbohydrate and fatty acid protocols.
Mitochondrial Calcium Retention Capacity
The mCRC was determined as the capacity of mitochondria to uptake calcium before permeability transition, which leads to massive calcium release. Maximum calcium uptake capacity of permeabilized fibers (ADP-deprived buffer) was quantified by monitoring calcium-green (1 μmol/L) fluorescence at 506 to 535 nm. mCRC was expressed as total mitochondrial calcium retention in nanomoles per 1 mg dry weight.
Mitochondrial Complex Activity and Myocardial Oxidative Stress
Citrate synthase, electron transport chain complex activities, and reactive oxygen species (ROS) levels were measured as previously described12 and detailed in the online-only Data Supplement.
Transmission electron microscopy was used to examine the cardiac mitochondrial network ultrastructure as previously described.13 Blocks of atrial trabeculae from diabetic and nondiabetic nonobese patients (n=3 per group) were fixed with glutaraldehyde, embedded, and oriented for longitudinal sectional views. Four thin sections per block were prepared and stained with uranyl acetate and lead citrate and examined with the use of a Philips CM10 transmission electron microscope (Philips, Eindhoven, the Netherlands). Technicians who were blinded to the groups performed the electron microscopic examination of sections and photographed 3 representative images per section (standardized magnification ×12 500). Interfibrillar mitochondrial density and interfibrillar mitochondrial length distribution were measured and analyzed with Metamorph software (Molecular Devices, Sunnyvale, CA).
Western Blot Analysis
Atrial biopsies were homogenized with a Polytron homogenizer in ice-cold radioimmunoprecipitation assay buffer (10 mmol/L Tris-HCl, pH 7.4, 150 mmol/L NaCl, 1 mmol/L EDTA, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS). After centrifugation at 10 000 rpm for 30 minutes at 4°C, supernatants were collected, and the amount of protein was evaluated with the BCA assay (Pierce). Supernatants were heat-denatured in Laemmli sample buffer (Bio-Rad), and proteins were separated on 10% NuPAGE Novex gels (Invitrogen). Proteins were transferred to nitrocellulose membranes, which were then saturated with 5% nonfat dry milk in 15 mmol/L Tris-HCl, pH 8.0, 140 mmol/L NaCl, and 0.05% Tween. Membranes were incubated with primary antibodies directed against OPA1, MFN1, DRP1, ATG5, PINK1 (Novus Biologicals), MFN2 (Boster Biologicals), FIS1 (Enzo Life Sciences), PARK2 (Millipore), LC3 (Map1lc3a; Abcam), and PGC1-α, with β-actin (Santa Cruz) as a loading control. Secondary antibodies coupled to horseradish peroxidase were used to detect immune complexes by chemiluminescence (Pierce). Band intensities were normalized to β-actin.
Continuous variables with a gaussian distribution are given as mean±SD or mean±SEM as specified. Continuous variables with no gaussian distribution are given as median (25th–75th) percentiles. Categorical variables are given as the number (percentage) of patients with the respective attribute. Bivariate comparisons were performed by use of the t test for normally distributed continuous variables or the Mann-Whitney U test for variables not normally distributed. Bivariate comparisons of categorical variables were performed with the χ2 test.
Two-way ANOVA (diabetic status times obese status) was performed to assess the respective effects of diabetes status, obese status, and their interaction on mitochondrial and contractile functional parameters. Post hoc t tests were used with Bonferroni corrections.
The Pearson test was performed to evaluate potential correlations between mitochondrial functional parameters and metabolic variables on one hand and contractile function parameters on the other hand.
The effect of metabolic parameters on cardiac mitochondrial function was assessed for both Vpalm+ADP and mCRC with a multivariate linear regression model using backward variable selection including the continuous covariates with a value of P<0.10 on bivariate analysis as predictors, that is, age, systolic blood pressure, log hemoglobin A1C (HbA1C), and log HOMA-IR. BMI was also forced into the model because of its clinical relevance. Coefficients and standard errors are given for continuous variables.
Electron microscopy data were analyzed with a linear mixed model. A value of P<0.05 was considered statistically significant. All analyses were conducted with SAS version 9.3 (SAS Institute, Cary, NC). Figures were drawn with GraphPad Prism version 5.0.
Study Flow Chart and Patient Characteristics by Diabetic Status
During the 18-month period, 153 consecutive patients were eligible for the study. Atrial biopsies could be obtained from 141 patients. As a result of different biopsy amount or quality, contraction, mitochondrial respiration, and mCRC explorations could be performed on 67, 139 and 87 patient samples, respectively. The flow chart of the study population and functional explorations is presented in Figure 1.
The characteristics of the patients for whom mitochondrial respiration assessment was performed are presented in Table 1 and Table I in the online-only Data Supplement. Ninety-four patients (68%) were nondiabetic, and 45 (32%) suffered from DM. The 2 groups of patients were well matched for most of the clinical variables except hypertension, which was significantly more frequent in diabetic patients. As expected, diabetic patients showed higher BMI, fasting glucose, HOMA-IR, and glycohemoglobin and lower high-density lipoprotein cholesterol compared with nondiabetic subjects. No difference in the preoperative plasma brain natriuretic peptide levels was observed between groups. As a direct consequence of the strict exclusion criteria used to study patients without overt cardiomyopathy, no significant differences in echocardiographic parameters were observed between the groups (Table 2 and Table I in the online-only Data Supplement).
Association Between Myocardial Contractile Performance, Mitochondrial Function, and Metabolic Status
We first assessed the potential link between ex vivo contractile performance and mitochondrial function parameters in patients without signs of cardiomyopathy (as detailed in Methods). As shown in Figure 2A and 2B, a weak positive correlation was found between the 2 mitochondrial function parameters Vpalm+ADP and mCRC and the twitch force developed by trabeculae.
Given the potential interaction of metabolic status with mitochondrial function, we next assessed the relative impact of DM and weight status on intrinsic contraction. The twitch force developed ex vivo in isometric conditions by atrial trabeculae from diabetic patients was significantly lower than in nondiabetic patients at stimulation rates of 60 and 180 bpm (Table 3). The trend toward prolonged contraction and relaxation time intervals in the diabetic myocardium (versus nondiabetic) did not reach statistical significance, even at the high stimulation rate (Table 3). As shown in Figure 2C, DM was associated with pronounced contractile dysfunction of atrial trabeculae regardless of weight status (Table II in the online-only Data Supplement). In contrast, obesity was associated with only a slight decrease in the twitch force developed at a stimulation rate of 60 bpm (Figure 2C). Thus, we provide evidence for a marked impact of DM, whereas obesity has only a slight effect on intrinsic myocardial contraction.
Disturbed Mitochondrial Function and Increased Myocardial Oxidative Stress in Diabetic but Not in Obese Patients
To determine whether DM and obesity alter cardiac mitochondrial function, we performed a detailed characterization of the cardiac mitochondrial function and myocardial oxidative stress.
We first explored respiratory chain complex activities in atrial tissue homogenates. Citrate synthase activity, an index of mitochondrial content, was not different between the groups, nor was respiratory chain complex IV activity (Figure 3A and 3D). Complex I activity was significantly reduced in obese patients (Figure 3B and Table II in the online-only Data Supplement). Complex II+III activity was reduced in diabetic patients (Figure 3C) but not in obese patients.
We next assessed whether these alterations of chain complex activities in in vitro assays translate into in situ mitochondrial dysfunction in permeabilized atrial fibers. As a functional reflection of impaired mitochondrial electron transport chain activity, state 3 respiration supported by fatty acid–like, pyruvate, or succinate substrates (Vpalm+ADP, Vpyr+ADP, and Vsucc+ADP, respectively) was reduced in diabetic patients (Figure 4A–4C). Complex IV–related maximum respiration rate (VCox) was not affected by DM (Figure 4D). DM was associated with a poor coupling between substrate oxidation and phosphorylation, as demonstrated by the low RCRpalm and RCRpyr (Figure 4E and 4F). In contrast, obese status was not associated with altered mitochondrial respiration rates regardless of the respiratory chain substrate (Figure 4A–4F and Table II in the online-only Data Supplement). Along the same lines, the functional parameter mCRC, assessed in permeabilized fibers, was significantly reduced in diabetic patients (Figure 4G). A tendency for an impact of obesity on mCRC was observed, but it did not reach statistical significance (P=0.16).
No significant interaction between DM and obese status was observed on the mitochondrial parameters (P=NS by 2-way ANOVA).
Considering the important role of mitochondria in oxidative stress, we next assessed ROS production and antioxidative systems in atrial tissue homogenates. Diabetic patients showed increased myocardial oxidative stress, that is, elevated atrial ROS production, along with increased activity of the antioxidant enzymes, that is, mitochondrion-located superoxide dismutase and cytosolic catalase activities (Figure 5A–5D and Table II in the online-only Data Supplement). In contrast, obesity did not influence oxidative stress markers in atrial tissue, nor was an interaction between obese and diabetes statuses observed for these parameters.
Association of Cardiac Mitochondrial Function With HbA1C but Not HOMA-IR or BMI
Using glycemia, insulin resistance, and weight as continuous variables, we plotted HbA1C, HOMA-IR, and BMI against functional mitochondrial parameters to further assess the link between those parameters and metabolic status. Of note, as a result of different biopsy amounts or qualities, the entire functional characterization could not be performed on tissues from all the patients: mitochondrial respiration data were obtained for 139 patients and mCRC data for 87 patients.
In simple regression analysis, the associations between mitochondrial function parameters and HbA1C (log-transformed) were weak but significant: Vpalm+ADP and mCRC were negatively associated with log(HbA1C) (Figure 6A and 6B). This association was stronger after adjustment by multilinear regression analysis (Table 4). In contrast, despite a weak association in simple regression, HOMA-IR was not individually predictive of mitochondrial function in multivariate analysis (Figure 6C and 6D and Table 3). Finally, no association with BMI as a continuous variable was found for any of the mitochondrial parameters (Figure 6E and 6F and Table 4).Together, these data suggest an adverse impact of DM on cardiac mitochondrial function likely mediated by chronic glucotoxicity. In contrast, early insulin resistance in “metabolically healthy” obese patients does not seem to affect mitochondrial function.
Cardiac Mitochondrial Network Fragmentation in Diabetic Patients
Given the important link between mitochondrial function and dynamics and diabetic status, we explored the cardiac mitochondrial network organization in diabetic patients.
Electron microscopy imaging of atrial tissue sections showed no difference in interfibrillar mitochondrial density between diabetic and nondiabetic patients (Figure 7A). This was in accordance with the absence of a difference in citrate synthase activities and PGC1-α protein expression (Figure 7E), indicating that the diabetic myocardium does not display a defect in mitochondrial biogenesis. However, a shift toward smaller cardiac mitochondria was observed in DM with a significantly lower mean mitochondrial length (1.14 μm [SE, 0.021 μm] in diabetic versus 1.41 μm [SE, 0.022 μm] in nondiabetic patients; P=0.001; Figure 7B–7D).
We next explored whether the fragmentation of the cardiac mitochondrial network observed in DM can be the result of an imbalance in mitochondrial dynamics (ie, fusion and fission) and autophagy. The expression of the fusion-related MFN1 protein was significantly lower in atrial tissue from DM patients (Figure 7E). Conversely, no alteration in the expression of other dynamic-related proteins, that is, OPA1, MFN2, FIS1, and DRP1, was observed in DM. Interestingly, the content in myocardial MFN1 protein correlated negatively with HbA1C concentrations (Pearson coefficient r2=0.36, P=0.02).
Autophagy is a self-digestion process involved in the selective clearance of damaged and dysfunctional mitochondria. Although the expression of the ATG5 protein, a key actor in the mitophagy process, was significantly decreased in diabetic myocardium (Figure 7E), expression of the autophagy-related PARK2 and PINK1 proteins (Figure 7E) and the ratio of LC3-II to LC3-I (Figure I in the online-only Data Supplement) were not altered, suggesting normal autophagy in the diabetic myocardium.
Thus, we provide data arguing for a link between chronic hyperglycemia and decreased expression of the mitofusion protein MFN1, which may contribute to the fragmentation of the mitochondrial network in DM.
The aim of this study was to determine whether obesity, insulin resistance, and DM influence cardiac mitochondrial function in a distinct manner and to establish a potential link between mitochondrial network dysfunction and contractile alterations of the myocardium in these metabolic disorders. Our findings clearly show an association of mitochondrial dysfunction with decreased contractile performance in heart tissue of diabetic patients before the onset of clinical cardiomyopathy. These effects in diabetic hearts are associated with the fragmentation of the mitochondrial network and decreased expression of the mitofusion protein MFN1. In contrast, although ex vivo contractile performance was also decreased in obesity, albeit to a lesser extent than in DM, this was not associated with any major perturbation of mitochondrial function. On the basis of our findings, chronic hyperglycemia can be postulated as a major driver of mitochondrial dysfunction in the diabetic myocardium, whereas early-stage alterations in glucose homeostasis (eg, insulin resistance) per se do not alter cardiac mitochondrial function.
Cardiac Mitochondria in DM and Obesity
Patients with cardiomyopathy related to DM or obesity exert abnormal myocardial energy homeostasis, particularly in conditions of high cardiac workload.6,7 Cardiac mitochondrial dysfunction has been observed in rodent models of DM and obesity, highlighting cardiac mitochondria as key players in the bioenergetic impairment in the associated cardiomyopathies.4,5 Recently, small studies have been performed on human atrial tissue to characterize cardiac mitochondrial function in DM and obesity.8,9 Anderson et al8 reported mitochondrial dysfunction in the atrial myocardium of 11 patients suffering from DM. Although cardiac mitochondrial dysfunction was convincingly illustrated, these authors included diabetic patients regardless of their body weight status, mixing obese and nonobese patients. Thus, they were not able to discriminate the relative contributions of DM and obesity to mitochondrial dysfunction. Moreover, no functional data were reported, thus precluding conclusions on the potential contribution to diabetic cardiomyopathy. Our data obtained with an ex vivo contractility assay show an association between cardiac mitochondrial dysfunction and infraclinical contractile dysfunction in diabetic patients.
Our data challenge the results recently reported by Niemann et al9 on cardiac mitochondrial function in 30 obese patients. These authors concluded that obesity resulted in disturbed mitochondrial function and increased oxidative stress in cardiomyocytes. However, this conclusion was based on the sole observation that in vitro respiratory chain complex activities were abnormal in obese patients. Although we also found such an alteration in in vitro complex activities as a result of the obese status, we demonstrate that this does not translate into any significant mitochondrial dysfunction, as shown by unaltered mitochondrial respiration and mCRC. A second study from the same group failed to show any effect of weight status on mitochondrial function, as assessed by mitochondrial respiration on permeabilized cardiac fibers.14 Second, we did not find evidence of increased oxidative stress in the myocardium of obese patients, although it was clearly increased in DM patients. A possible explanation for this discrepancy may stem from the fact that Niemann et al9 enrolled obese patients who had already transitioned from early insulin resistance to DM, as indicated by the large variation in HbA1C (5.9±6.2%). Thus, our results question the potential impact of obesity per se on cardiac mitochondria. Because mitochondrial function is not the sole factor controlling cardiac bioenergetics, we cannot exclude a possible abnormal energy homeostasis in obese hearts, as previously reported by Rider et al.7
Mechanisms of Mitochondrial Dysfunction in the “Metabolic” Heart: Hyperglycemia and Insulin Resistance
Hyperglycemia and insulin resistance are believed to be responsible for cardiac mitochondrial dysfunction in diabetic and obese patients.15
Insulin resistance can lead to mitochondrial dysfunction in several ways. First, systemic insulin resistance alters systemic substrate availability, which in turn leads to early changes in myocardial substrate utilization in both DM and obesity, that is, an increase in fatty acid utilization and a decrease in glucose utilization. Therefore, systemic insulin resistance in rodents drives increased fluxes of free fatty acids in cardiomyocytes, which display reduced oxidative phosphorylation, mitochondrial uncoupling, and oxidative stress.15 Second, direct cardiac impairment of insulin signaling can precipitate mitochondrial dysfunction, as observed in hearts of mice with cardiomyocyte-restricted deletion of the insulin receptor.16 A reduction in mitochondrial oxidative capacity and biogenesis has been described as driving mitochondrial dysfunction associated with impaired myocardial insulin signaling in these mice.17
As reviewed in detail elsewhere,15 although the pathogeneses of type 1 DM and type 2 DM are distinct, alterations in cardiac mitochondria observed in these 2 pathologies share numerous aspects, suggesting that hyperglycemia is a potential driver of mitochondrial dysfunction. Taken together, our findings argue for a link among hyperglycemia, decreased expression of MFN1, fragmentation of the cardiac mitochondrial network, and its dysfunction. The quality of the mitochondrial population is controlled by mitochondrial network dynamics and the balance between biogenesis of new mitochondria and autophagy of damaged organelles. We have previously shown that mitochondrial homeostasis is altered in a murine model of diabetic cardiomyopathy.18 A growing body of evidence indicates that mitochondrial network morphology and function are closely connected in active metabolic cells, particularly in cardiomyocytes.19 Yu et al20 showed that hyperglycemia leads to Drp1-induced mitochondrial fission in neonatal rat ventricular myocytes, which in turn induced mitochondrial dysfunction and proapoptotic pathway activation. In agreement, Shenouda et al21 showed that hyperglycemia alters the expression of proteins implicated in mitochondrial dynamics, that is, increased expression of FIS1 and DRP1 proteins, and induces mitochondrial fragmentation in human endothelial cells. In these 2 studies, however, MFN1 protein expression was not explored. In contrast to results on rodent cardiomyocytes and human endothelial cells, we report here a decreased expression of the fusion-related protein MFN1 in the diabetic myocardium but no significant alteration of the fission-related actors FIS1 and DRP1. Although not explored in our study, posttranslational modifications of proteins involved in mitochondrial dynamics such as glycation and O- Glc-NAcylation have been observed in high-glucose conditions and associated with mitochondrial dysfunction and fragmentation.22,23 Further studies are required to test whether modulation of the mitochondrial dynamics process would improve mitochondrial function and thus contractile performance in the diabetic human myocardium.
There are several limitations to the present study directly linked to the exploration of human myocardial tissue samples.
As a result of ethical issues, we could explore only the atrial myocardium. However, studies in rodent models of the metabolic syndrome indicate that the mitochondrial alterations in atrial tissue closely reflect the situation in the ventricular myocardium,9 suggesting that atrial tissue represents a valuable and available source to assess mitochondrial function in the myocardium of patients suffering from metabolic disorders.
We explored myocardial tissues from patients with significant coronaropathy. Although we excluded patients with significant right coronary artery stenosis to decrease the impact of confounding ischemia on the studied tissue, existing microvascular pathologies could have influenced our results.
Although significant, the correlation coefficients between the metabolic and mitochondrial parameters were low (<0.25) in simple regression analysis, meaning that HbA1C, taken alone, has a low predictive power for mitochondrial function parameters.24
Finally, because of limitations in tissue availability, we were not able to assess all functional parameters in all patients or to directly measure mitochondrial ROS production, in complement to the measurement of ROS by electron paramagnetic resonance.
Based on the present findings, worsening of intrinsic myocardial contraction in the transition from obesity to DM is likely related to worsening of cardiac mitochondrial function because impaired mitochondrial function and dynamics and contractile dysfunction are observed in diabetic patients but not in obese patients with early-stage insulin resistance. Moreover, hyperglycemia can be postulated as a major driver of both mitochondrial dysfunction and mitochondrial network fragmentation in the human diabetic heart.
We thank Mathilde Coevoet, Florian Delguste, and Anne Loyens (University Lille 2), as well as Claude Cachera and Isabelle Kim (CHRU Lille), for their technical assistance. Prof. Staels is a member of the Institut Universitaire de France.
Sources of Funding
This work was supported by grants from Program Projet Emergeant Région Nord-Pas-de-Calais, Fédération Française de Cardiologie and CHRU Lille. Dr Sebti was supported by a grant from Lille Métropole Communauté Urbaine.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/lookup/suppl/doi:10.1161/CIRCULATIONAHA.113.008476/-/DC1.
- Received December 28, 2013.
- Accepted June 6, 2014.
- © 2014 American Heart Association, Inc.
- Bertoni AG,
- Tsai A,
- Kasper EK,
- Brancati FL
- Boudina S,
- Abel ED
- Scheuermann-Freestone M,
- Madsen PL,
- Manners D,
- Blamire AM,
- Buckingham RE,
- Styles P,
- Radda GK,
- Neubauer S,
- Clarke K
- Rider OJ,
- Francis JM,
- Ali MK,
- Holloway C,
- Pegg T,
- Robson MD,
- Tyler D,
- Byrne J,
- Clarke K,
- Neubauer S
- Montaigne D,
- Marechal X,
- Lefebvre P,
- Modine T,
- Fayad G,
- Dehondt H,
- Hurt C,
- Coisne A,
- Koussa M,
- Remy-Jouet I,
- Zerimech F,
- Boulanger E,
- Lacroix D,
- Staels B,
- Neviere R
- Bugger H,
- Abel ED
- Boudina S,
- Bugger H,
- Sena S,
- O’Neill BT,
- Zaha VG,
- Ilkun O,
- Wright JJ,
- Mazumder PK,
- Palfreyman E,
- Tidwell TJ,
- Theobald H,
- Khalimonchuk O,
- Wayment B,
- Sheng X,
- Rodnick KJ,
- Centini R,
- Chen D,
- Litwin SE,
- Weimer BE,
- Abel ED
- Belke DD,
- Betuing S,
- Tuttle MJ,
- Graveleau C,
- Young ME,
- Pham M,
- Zhang D,
- Cooksey RC,
- McClain DA,
- Litwin SE,
- Taegtmeyer H,
- Severson D,
- Kahn CR,
- Abel ED
- Lancel S,
- Montaigne D,
- Marechal X,
- Marciniak C,
- Hassoun SM,
- Decoster B,
- Ballot C,
- Blazejewski C,
- Corseaux D,
- Lescure B,
- Motterlini R,
- Neviere R
- Yu T,
- Sheu SS,
- Robotham JL,
- Yoon Y
- Shenouda SM,
- Widlansky ME,
- Chen K,
- Xu G,
- Holbrook M,
- Tabit CE,
- Hamburg NM,
- Frame AA,
- Caiano TL,
- Kluge MA,
- Duess MA,
- Levit A,
- Kim B,
- Hartman ML,
- Joseph L,
- Shirihai OS,
- Vita JA
- Gawlowski T,
- Suarez J,
- Scott B,
- Torres-Gonzalez M,
- Wang H,
- Schwappacher R,
- Han X,
- Yates JR 3rd.,
- Hoshijima M,
- Dillmann W
- Hinkle DE,
- Wiersma W,
- Jurs SG
Obesity and diabetes mellitus are independently associated with the development of heart failure. In this study, we determined the respective effects of obesity, insulin resistance, and diabetes mellitus on intrinsic contraction and mitochondrial function of the human myocardium before the onset of cardiomyopathy. Exploring right atrial human samples, we show that diabetes mellitus is associated with a pronounced impairment of intrinsic contraction, mitochondrial dysfunction, and increased myocardial oxidative stress, regardless of weight status. These effects in diabetic hearts are associated with a fragmentation of the mitochondrial network and decreased expression of the MFN1 protein involved in mitochondrial fusion. In contrast, obesity was associated with less pronounced contractile dysfunction without any significant perturbation of mitochondrial function or oxidative stress status. Tested as continuous variables, glycated hemoglobin, but not body mass index or insulin resistance (homeostasis model assessment–insulin resistance), was independently associated with cardiac mitochondrial function. On the basis of the present findings, worsening of intrinsic myocardial contraction in the transition from obesity to diabetes mellitus is likely related to worsening of cardiac mitochondrial function because impaired mitochondrial function and dynamics and contractile dysfunction are observed in diabetic patients but not in “metabolically healthy” obese patients at early stages of insulin resistance. Moreover, chronic hyperglycemia is postulated to be a major driver of both mitochondrial dysfunction and mitochondrial network fragmentation in the human diabetic heart. This study opens a research field for potential therapies to prevent heart failure in diabetic patients by targeting the mitochondria.