Diabetic cardiomyopathy-associated dysfunction in spatially distinct mitochondrial subpopulations

Abstract

Diabetic cardiomyopathy is the leading cause of heart failure among diabetic patients, and mitochondrial dysfunction has been implicated as an underlying cause in the pathogenesis. Cardiac mitochondria consist of two spatially, functionally, and morphologically distinct subpopulations, termed subsarcolemmal mitochondria (SSM) and interfibrillar mitochondria (IFM). SSM are situated beneath the plasma membrane, whereas IFM are embedded between myofibrils. The goal of this study was to determine whether spatially distinct cardiac mitochondrial subpopulations respond differently to a diabetic phenotype. Swiss-Webster mice were subjected to intraperitoneal injections of streptozotocin or citrate saline vehicle. Five weeks after injections, diabetic hearts displayed decreased rates of contraction, relaxation, and left ventricular developed pressures (P < 0.05 for all three). Both mitochondrial size (forward scatter, P < 0.01) and complexity (side scatter, P < 0.01) were decreased in diabetic IFM but not diabetic SSM. Electron transport chain complex II respiration was decreased in diabetic SSM (P < 0.05) and diabetic IFM (P < 0.01), with the decrease being greater in IFM. Furthermore, IFM complex I respiration and complex III activity were decreased with diabetes (P < 0.01) but were unchanged in SSM. Superoxide production was increased only in diabetic IFM (P < 0.01). Oxidative damage to proteins and lipids, indexed through nitrotyrosine residues and lipid peroxidation, were higher in diabetic IFM (P < 0.05 and P < 0.01, respectively). The mitochondria-specific phospholipid cardiolipin was decreased in diabetic IFM (P < 0.01) but not SSM. These results indicate that diabetes mellitus imposes a greater stress on the IFM subpopulation, which is associated, in part, with increased superoxide generation and oxidative damage, resulting in morphological and functional abnormalities that may contribute to the pathogenesis of diabetic cardiomyopathy.

diabetic cardiomyopathy has been implicated as a primary cause of heart failure among diabetic patients, and it is thought to occur independent of vascular pathogenesis (22, 27, 39, 52, 63). The hyperglycemic environment presented by diabetes mellitus leads to enhanced ROS generation, and, although many potential sources of ROS exist, the mitochondrion is considered to be the primary site (6, 8, 15, 28, 50). Several sites in the electron transport chain (ETC) are particularly prone to the formation of ROS and include oxidizable electron carriers in the inner mitochondrial membrane (12, 32). This has implications for ETC proteins because a major constituent of these structures is their iron-sulphur centers (39), which can react with ROS such as superoxide (O₂^•−) or H₂O₂ and produce the highly reactive hydroxyl radical (·OH) (14). Increased mitochondrial ROS production has been linked to mitochondrial dysfunction (13, 50, 51, 53, 54), which can, in turn, alter the integrity of the inner mitochondrial membrane, facilitating further dysfunction in mitochondria. In particular, proteins and lipids within mitochondria are critical targets of elevated ROS production, and their oxidative modification potentiates mitochondrial dysfunction by limiting adequate production of ATP. One potential target of mitochondrial dysfunction is the mitochondria-specific phospholipid cardiolipin. Cardiolipin is a diphosphatidylglycerol enriched in the inner membrane that contains oxidatively sensitive acyl groups that may act as selective targets for ROS (33, 34, 46). Cardiolipin interacts with a number of mitochondrial proteins, including F₀F₁-ATPase, adenine nucleotide translocase, cytochrome c, and ETC complexes I, III, and IV, and its oxidative modification may be a critical event for apoptosis initiation (23).

The cardiac myocyte contains two distinct mitochondrial subpopulations that are characterized by their spatial arrangement within the cell. These two disparate populations have been termed subsarcolemmal mitochondria (SSM) and interfibrillar mitochondria (IFM) due to their subcellular locations, which either abut the sarcolemma or situate between the contractile apparatus (44, 49). In addition to spatial differences, mitochondrial subpopulations differ in structure, size, ATP levels, protein import rates, substrate utilization, and other biochemical properties (1, 21, 26, 28, 31, 34, 38, 44). Morphologically, IFM are smaller, more compact, and possess functionally greater respiratory rates, supplying ATP mainly for contractile function. SSM are larger, more variable in shape, and produce ATP primarily for electrolyte and protein transport across the plasma membrane (44). The two mitochondrial subpopulations respond differently to physiological stimuli, including exercise, aging, obesity, fasting, apoptotic initiators, and ischemia-reperfusion (I/R) injury (1, 31, 34, 38, 48, 57). Decreased ETC function and elevated oxidative stress have been reported in SSM after myocardial I/R, with no differences observed in IFM (33). Using electron microscopy, Kelley et al. (28) observed decreased IFM size in skeletal muscle of type 2 diabetic patients, which was not observed in SSM. Others (21, 26) have observed decreased IFM ETC function with aging. These findings indicate that although mitochondria are similar in their central role in cellular function, spatially distinct populations are influenced by pathological states differently, requiring careful examination of individual mitochondrial subpopulations.

Because much of the previous work examining the impact of diabetes mellitus on mitochondrial disposition has been performed on total mitochondria, it is difficult to assess the impact of the pathology on spatially distinct mitochondrial populations. The goal of this study was to determine the differential response of individual mitochondrial subpopulations subjected to a diabetic phenotype to further our understanding of their specific contribution to the pathogenesis of diabetic cardiomyopathy. Our results indicate that the IFM subpopulation is affected by diabetes mellitus to a greater extent than SSM, as reflected by greater morphological changes, elevated ROS, and enhanced oxidative damage. This study is the first to determine the spatial influence of the diabetic phenotype on cardiac mitochondrial dysfunction.

MATERIALS AND METHODS

Experimental animals and the induction of diabetes.

The animal experiments in this study conformed to National Institutes of Health (NIH) Guidelines for the Care and Use of Laboratory Animals and were approved by the West Virginia University Animal Care and Use Committee. Female Swiss-Webster mice (Harlan, Indianapolis, IN) were housed in the West Virginia University Health Sciences Center animal facility. Mice were given unlimited access to a rodent diet and water. Diabetes was induced in 8-wk-old mice following the protocol of the Animal Models of Diabetic Complications Consortium using multiple low-dose streptozotocin (STZ; Sigma, St. Louis, MO) injections. A multiple low-dose STZ protocol was chosen because previous reports (24, 29) have indicated that this model limits the body weight losses associated with diabetic protocols. Injections of 50 mg/kg body wt STZ dissolved in sodium citrate buffer (pH 4.5) were performed daily for 5 consecutive days after 6 h of fasting. Mice that served as vehicle controls were given the same volume per body weight of sodium citrate buffer. One week postinjection, hyperglycemia was confirmed by measuring urinary glucose (Chemstrip 2GP Urine test strips, Roche Diagnostics, Indianapolis, IN), where >2,000 mg/dl was considered diabetic. To confirm the diabetic phenotype, urinary glucose levels were monitored weekly using a DiaScreen 50 Urine Chemistry Analyzer (Arkray, Edina, MN) and DiaScreen 10 reagent strips. Five weeks after the onset of hyperglycemia, animals were killed for further experimentation.

Cardiac contractile function.

Hearts from diabetic and control animals were isolated and transferred to a Langendorff setup for contractile experiments as previously described (19, 20, 59). In brief, hearts were removed from anesthetized mice and immersed in cold cardioplegic solution. After cannulation of the aorta on a luer stub, hearts were perfused retrograde at 37°C with a modified Krebs-Henseleit buffer [containing (in mmol/l) 0.5 pyruvate, 0.4 caprylic acid, 118 NaCl, 4.7 KCl, 2.25 CaCl₂, 1.2 MgSO₄, 1.2 KH₂PO₄, 25 NaHCO₃, 0.5 Na₂EDTA, and 5.5 glucose] that was subjected to oxygenation. A small fluid-filled balloon was inserted into the left ventricular (LV) cavity and coupled to a pressure transducer (Argon Medical Devices, Athens, TX). The balloon was inflated until the end-diastolic pressure reached 10 mmHg. Platinum wires were placed on the surface of the right atrium and used to pace the heart at 420 beats/min (7 Hz) with increasing voltage until the heart rate was consistent (captured). During this time period, hearts remained submerged in a 37°C jacketed chamber, and a heat lamp attached to a variable alternating current transformer was used to maintain temperature, which was monitored by a needle temperature probe inserted just before the luer cannula (Physitemp, Clifton, NJ). Hearts were perfused for 15 min to achieve stable cardiac function and then followed for an additional 45 min, yielding a total protocol time of 1 h. During this time, digitized recordings of the LV pressure and its first derivative as well as heart rate were captured on an IBM-compatible PC using ChartPro software (AD Instruments, Colorado Springs, CO). LV peak systolic pressure, developed pressure (DP), maximum speed of contraction (+dP/dt), and maximum speed of relaxation (−dP/dt) were analyzed from the recordings using ChartPro software.

Preparation of individual mitochondrial subpopulations.

Five weeks after the onset of hyperglycemia, animals were killed, and hearts were removed. Hearts were rinsed in PBS (pH 7.4), blotted dry, and then weighed. SSM and IFM were isolated on ice following the methods of Palmer et al. (44) with minor modifications. Briefly, the ventricles were minced and homogenized at 1:10 (wt/vol) in cold Chappel-Perry buffer [containing (in mmol/l) 100 KCl, 40 Tris·HCl, 10 Tris base, 5 MgCl₂, 1 ATP, and 1 EDTA; pH 7.4]. Homogenates were then centrifuged at 800 g for 10 min. The supernatant was extracted and centrifuged again at 9,000 g to isolate SSM. The SSM pellet was washed and centrifuged two more times at 9,000 g and once more at 5,000 g to obtain a clean SSM fraction. The remaining pellet from the 800-g spin was resuspended in buffer 2 [containing (in mmol/l) 100 KCl, 5 MgSO₄, 5 EGTA, and 50 Tris·HCl; pH 7.4] and exposed to 5 mg/kg trypsin for 10 min (49). After 10 min, the IFM pellet was diluted twofold with buffer and spun down at 5,000 g for 5 min. The supernatant was discarded, and the pellet was resuspended in buffer and spun down at 800 g for 10 min. The supernatant was saved, and the pellet was resuspended and spun down again at 800 g for 10 min to maximize the IFM yield. Next, supernatants were combined and spun down at 9,000 g to yield IFM. IFM were washed several times and spun down at a final spin of 5,000 g for 10 min. Pellets were resuspended in a sucrose buffer [containing (in mmol/l) 220 sucrose, 70 mannitol, 10 Tris·HCl, and 1 EDTA; pH 7.4], and protein concentrations were determined using the Bradford method with BSA as a standard (4). O₂^•− production, oxygen consumption, nitrotyrosine protein, and lipid peroxidation were assessed on freshly isolated subfractions.

Mitochondria size and internal complexity.

To index mitochondrial subpopulation size and complexity, we performed flow cytometry analyses using a FACS Calibur equipped with a 15 MW 488-nm argon laser and 633 red diode laser (Becton Dickinson, San Jose, CA) as previously described (8, 9). Each individual parameter (gating, size, and complexity) was performed using specific light sources (laser and photomultiplier tube) and specific detectors. MitoTracker deep red 633 (Invitrogen, Carlsbad, CA), which moves into intact mitochondria due to membrane potential, was used to selectively stain intact mitochondria (emission wavelength: 633 nm, fluorescent 633 red diode laser) and exclude debris, which contains no membrane potential, enabling accurate gating (R1) of the mitochondria. Once the gating parameters were established, gated events (20,000 events/sample) were subsequently examined using the forward scatter detector (FSC; 488-nm argon laser and diode detector) and side scatter detector (SSC; photomultiplier tube and 90° collection lens) and represented in FSC vs. SSC density plots. The geometric mean (arbitrary units), representing FSC (logarithmic scale), was used as an indicator of size, whereas values from SSC (logarithmic scale) were used to indicate complexity in the subpopulations. Although the FSC is proportional to the individual mitochondria particle size, the absolute value still remains an arbitrary unit. Thus, to confirm the absolute mitochondria size, we used a flow cytometry size calibration kit (Invitrogen), which uses a set of microsphere suspensions (0.5–6 μm) to serve as reliable size references for flow cytometric analyses. All flow cytometric measurements were performed under the supervision of the West Virginia University Flow Cytometry Core Facility.

Mitochondrial cardiolipin content.

Cardiolipin was determined as previously described (16). Briefly, isolated mitochondrial subpopulations (200 μg) were incubated with the synthetic phospholipid 1,1′,2,2′- tetramyristoylcardiolipin (T14:0 CL; Avanti Polar Lipids, Alabaster, AL), which served as an internal standard. Total lipids, including the internal standard, were extracted using the method of Bligh and Dyer (3). Briefly, MeOH-H₂O and chloroform were added to the sample, mixed, and then centrifuged at ∼200 g for 10 min. The chloroform layer was extracted from each sample and then dried under a nitrogen stream. Each sample was resuspended in 4 ml chloroform-MeOH (1:1) and washed with 1.8 ml of 20 mM LiCl aqueous solution. The organic layer was dried under a nitrogen stream, and the residue was resuspended in 1 ml chloroform and then filtered through a 0.2-μm poly(tetrafluoroethylene) (PTFE) syringe filter into a 5-ml glass centrifuge tube. The filtrate was filtered a second time with a 0.2-μm PTFE syringe filter and then dried under a nitrogen stream. Samples were resuspended in 500 μl/mg protein in chloroform-MeOH (1:1) and diluted further before injection into the mass spectrometer. LiOH (50 nmol/mg protein) was added to each sample just before analysis.

Cardiolipin detection was performed by electrospray ionization mass spectrometry (ESI-MS) using a quadrupole ion trap Finnigan LCQ DECA (ThermoFisher, Waltham, MA) operated in the negative ion mode. Mass spectrometer instrument conditions included a spray voltage of 5.2 kV, capillary voltage of −4 V, heated capillary temperature of 300°C, and a sheath gas (N₂) flow rate of 40 arbitrary units. The cardiolipin solution was infused using a 100-μl syringe at a flow rate of 5 μl/min. Mitochondrial cardiolipin and internal standard spectra were identified in the full scan mode. Peak intensities of cardiolipin versus the internal standard were estimated using single ion monitoring on doubly charged m/z 723 and m/z 619 in both diabetic and control mitochondria.

ETC respiration.

State 4 respiratory capacity was assessed in isolated mitochondrial subpopulations following the method of Hofhaus et al. (18) with slight modifications (19). After the mitochondrial subpopulation isolation, samples were resuspended in respiration buffer [containing (in mmol/l) 20 HEPES, 10 MgCl₂, and 250 sucrose], and equal volumes were loaded into a Gilson chamber (Gilson, Middleton, WI) attached to a Yellow Springs Instruments (YSI) 5300 biological oxygen monitor (YSI, Yellow Springs, OH). Respiration through complex I, complex II, and complex IV were determined by measuring the rate of oxygen consumed in the presence of specific substrates. These substrates included glutamate and malate for complex I, succinate for complex II, and ascorbate and N,N,N′,N′-tetramethyl-p-phenylenediamine for complex IV, and respiration was calculated as the fraction that was sensitive to the specific inhibitors rotenone (complex I), antimycin (complex II), and sodium cyanide (complex IV). Assessment of complex III activity was performed spectrophotometrically as previously described (60) by following the reduction of cytochrome c in the presence of reduced decylubiquinone. Protein content was determined following the Bradford method as described above, and values were expressed as nanomoles of O₂ consumed per minute per milligram of protein (complexes I, II, and IV) or the reduction of cytochrome c in nanomoles (activity) per minute per milligram of protein (complex III).

Electron paramagnetic resonance spectroscopy.

An electron paramagnetic resonance (EPR) spin trapping technique was used to detect short-lived free radicals such as ·OH and O₂^•− (62). This method is based on the reaction of a short-lived radical binding with a paramagnetic compound to form a relatively long-lived free radical product (spin adduct). This adduct can then be observed using conventional EPR, whereby the intensity of the signal is used to measure the amount of short-lived radicals trapped. The hyperfine couplings of the spin adduct are generally characteristics of the original trapped radicals. This method is ideal for the detection and identification of free radicals because of its specificity and sensitivity. All EPR measurements were conducted using a Bruker EMX spectrometer (Bruker Instruments, Billerica, MA) and a flat cell assembly. Hyperfine couplings were measured (to 0.1 G) directly from magnetic field separation using potassium tetraperoxochromate (K₃CrO₈) and 1,1-diphenyl-2-picrylhydrazyl (DPPH) as reference standards (5, 25). The Acquisit program was used for data acquisitions and analyses (Bruker Instruments). Isolated heart mitochondria subpopulations were resuspended in 500 μl of 1× PBS (pH 7.4). Each sample (100 μg) was brought up to 200 μl with PBS and incubated with the spin trap 5,5-dimethyl-1-pyrroline-N-oxide (200 mM) in the presence or absence of the excess complex I respiratory substrates glutamate and malate, reaching an end volume of 500 μl. Samples were incubated for 3 min at 37°C and then transferred to an EPR flat cell for measurement at room temperature with instrument settings of 63.6 mW, modulation amplitude 1 G, receiver gain 1.00 × 10⁵, conversion time 40.960 ms, and time constant 40.960 ms. All spectra shown are an accumulation of three scans. The reaction of xanthine and xanthine oxidase was used as a reference. The relative radical concentration was estimated by measuring the peak to peak height (mm) of the observed spectra.

Protein nitrotyrosine content.

Oxidatively modified proteins were examined by measuring nitrosylated groups introduced into protein side chains using a commercially available kit (Cell Sciences, Canton, MA). Nitrotyrosine-containing proteins were measured using a solid-phase enzyme-linked immunosorbent assay based on the sandwich principle. Samples were incubated in microtiter wells coated with antibodies recognizing nitrotyrosine residues. After an incubation and a wash, a biotinylated secondary antibody (tracer) was added, followed by a wash and the addition of a streptavidin-peroxidase conjugate. Color development was measured spectrophotometrically at 450 nm after the addition of tetramethylbenzidine using a Biotek Synergy HT plate reader (Biotek, Winooski, VT), and values were compared against known nitrotyrosine standards.

Lipid peroxidation products.

Peroxidation of lipids was assessed by measurements of malondialdehyde (MDA) and 4-hydroxyalkenal (4-HAE), stable end products formed from the oxidation of polyunsaturated fatty acids and esters. Equal volumes of freshly isolated mitochondrial subpopulations were analyzed for MDA and 4-HAE using a colorimetric assay kit (Oxford Biomedical Research, Oxford, MI). This assay is based on the reaction of a chromogenic reagent, N-methyl-2-phenylindole, with MDA and 4-HAE at 45°C. One molecule of either MDA or 4-HAE reacts with two molecules of the reagent to yield a stable chromophore with a maximal absorbance at 586 nm. Absorbance was measured on a Biotek Synergy HT plate reader (Biotek, Winooski, VT), and protein content was assessed as described above with final values expressed per milligram of protein.

Statistics.

Means and SEs were calculated for all data sets. Data were analyzed with a one-way ANOVA method to evaluate the main treatment effect, diabetes induction (Systat version 5.03, Evanston, IL). Fisher's least-significant-difference post hoc tests were performed to determine significant differences among means. When appropriate, a Student's t-test was employed. P < 0.05 was considered significant.

RESULTS

Cardiac contractile function in the diabetic heart.

Heart weight, body weight, and heart weight-to-body weight ratios were not significantly altered after STZ treatment (Table 1), which is in agreement with other studies (24, 29) using multiple low-dose STZ protocols. Analyses of nonfasting urinary ketone levels 5 wk after the onset of diabetes were negative for diabetic animals (data not shown). Five weeks after the onset of diabetes and hyperglycemia, animals were killed, and the heart were removed for measurements of contractile function using a Langendorff perfusion apparatus. +dP/dt and −dP/dt as well as DP were significantly decreased in diabetic versus control hearts (P < 0.05 for all three; Table 1), demonstrating decreased contractile function. These data are in agreement with others examining the influence of the diabetic phenotype using STZ injection as a model (15, 52, 56, 58).

Table 1.

Heart weight, body weight, contractile function, and mitochondrial yields

Group	Body Weight, g	Heart Weight, mg	Heart Weight-to-Body Weight Ratio, mg/g	+dP/dt, mmHg/s	−dP/dt, mmHg/s	Developed Pressure, mmHg	Subsarcolemmal Mitochondrial Protein, mg/g tissue	Interfibrillar Mitochondrial Protein, mg/g tissue
Control	27.4±1.3	102±4	3.7±0.1	3,398±128	2,310±114	96±2	18.9±2.1	10.3±0.8
Diabetic	28.4±0.8	108±6	3.8±0.2	1,934±116*	1,680±144*	57±3*	19.5±2.2	9.5±1.5

Values are expressed as means ± SE; n = 8 animals/group. Heart weight, body weight, and heart weight-to-body weight ratios were determined 5 weeks after streptozotocin treatment (diabetic group) or citrate saline injection (control group). Functional measurements were performed in Langendorff-perfused hearts and are expressed as the rate of contraction (+dP/dt), rate of relaxation (−dP/dt), and developed pressure. Individual mitochondrial subpopulation protein yields are also shown.

^*P < 0.05 for diabetic vs. control groups.

Mitochondrial subpopulation morphology.

Mitochondrial subpopulations were isolated, and yields were similar between control and diabetic hearts (Table 1). To determine morphological differences between control and diabetic mitochondrial subpopulations, we used a novel flow cytometry approach in which gating was performed to make an accurate estimation of mitochondrial size and complexity. Assessment of forward-scattered light (FSC) was used to estimate size, whereas assessment of side-scattered light (SSC) was used to estimate mitochondrial complexity, both of which were based on a logarithmic scale. Individual mitochondrial subpopulations were stained with MitoTracker deep red 633, which incorporates into intact mitochondria. In Fig. 1A, a typical dot plot showing MitoTracker deep red 633-stained mitochondria is shown, with intact mitochondria indicated in red and unstained debris indicated in black. Using this information, we gated the mitochondria (R1) to exclude unstained debris and applied the R1 gate to analyses on individual subpopulations. These analyses lend insight into relative morphological differences between the two subpopulations and help to confirm the success of the isolation procedure. To confirm differences in absolute mitochondria size, we included size calibration beads composed of microsphere suspensions ranging in size from 0.5 to 6 μm to serve as reliable size references. Using this approach, SSM were larger in size (FSC) and possessed greater internal complexity (SSC) compared with IFM, which were smaller and more compact (Fig. 1, B and C). These results are consistent with previously published reports (28, 48, 49, 55). Mitochondrial size was significantly decreased by 35% (P < 0.01) in diabetic IFM compared with control IFM, whereas SSM showed no significant changes (Fig. 1D). Mitochondrial complexity in diabetic IFM was also significantly decreased, by 40% (P < 0.01), compared with control IFM (Fig. 1E). No significant differences in SSC were observed in the SSM population (Fig. 1E). These results indicate that only IFM morphology is impacted as a result of the STZ-induced diabetic insult.

Fig. 1.

Mitochondria subpopulation morphological assessment. Determination of the relative size and internal complexity of distinct mitochondrial subpopulations [subsarcolemmal mitochondria (SSM) and interfibrillar mitochondria (IFM)] using flow cytometric analyses is shown. A: representative density plot of isolated mitochondria stained with MitoTracker deep red 633, which incorporates into intact mitochondria. Intact mitochondria that take up MitoTracker deep red 633 dye are gated (R1) and shown in red. Noise and debris, which does not take up MitoTracker deep red 633, is excluded from measurements (outside the R1 gate) and shown in black. B: representative gated density plot indicating size [forward scatter (FSC)] versus internal complexity [side scatter (SSC)] of isolated SSM. C: representative gated density plot indicating size (FSC) versus internal complexity (side scatter) of isolated IFM. Analysis of FSC and SSC were calculated per 20,000 gated events for all mitochondrial subpopulations. Red lines on the x-axes of B and C indicate microsphere size references from left to right: 0.5, 1, 2, 4, and 6 μM, respectively. D: analysis of cardiac SSM and IFM size (FSC) in control and diabetic groups. E: analysis of cardiac SSM and IFM complexity (SSC) in control and diabetic groups. Values for size and complexity are expressed as arbitrary units (AU) ± SE; n = 3 for each group. †P < 0.01 for diabetic vs. control groups.

ETC respiration.

Mitochondrial ETC function is altered with hyperglycemia, but it is unclear whether these alterations are uniform between mitochondrial subpopulations. We used polarography to assess changes in ETC complex I, II, and IV respiration and spectrometry to measure complex III activity within individual mitochondrial subpopulations from diabetic and control hearts. Complex I, II, and III were significantly decreased in diabetic IFM relative to control (P < 0.01 for both, Fig. 2, A and B; and P < 0.05, Fig. 2C), whereas only complex II was significantly decreased in SSM (P < 0.05, Fig. 2B). No significant changes were observed with complex IV in either subpopulation (Fig. 2D). These data indicate that STZ-induced diabetic insult impacts both SSM and IFM, but these effects are greater in IFM.

Fig. 2.

Mitochondria subpopulation respiration. Individual mitochondrial subpopulations were isolated from diabetic and control hearts, and polarographic measurements were performed to index oxygen consumption under state 4 respiration conditions. Respiration of individual electron transport chain (ETC) complexes was defined as the rate of oxygen consumed in the presence of specific substrates [glutamate and malate (complex I), succinate (complex II), and ascorbate and N,N,N′,N′-tetramethyl-p-phenylenediamine (complex IV)] and was calculated as the fraction that was sensitive to the specific inhibitors rotenone (complex I), antimycin (complex II), and sodium cyanide (complex IV). Measurement of complex III was performed spectrophotometrically by the assessment of the reduction of cytochrome c in the presence of reduced decylubiquinone. A: complex I. B: complex II. C: complex III. D: complex IV. Values are expressed as means ± SE; n = 7 for each group. *P < 0.05 for diabetic vs. control groups; †P < 0.01 for diabetic vs. control groups.

ROS production.

Because the mitochondrion is centrally involved in the formation of ROS, we determined whether the diabetic phenotype enhanced ROS formation in a subpopulation-specific manner. Mitochondrial subpopulations were incubated with the complex I substrates glutamate and malate to fuel the ETC, and ROS generation was analyzed using EPR spectroscopy. We observed increases in diabetic IFM spin trapping peaks relative to control IFM (P < 0.01; Fig. 3, C, D, and H), and these differences were not observed in SSM (Fig. 3, A, B, and H). The spin trapping pattern observed was indicative of O₂^•−, and the addition of SOD confirmed the observed spectra as O₂^•− (Fig. 3, E and F). A spectrum for xanthine oxidase, a known O₂^•− generator, is included for spin trapping pattern comparison (Fig. 3G). These data suggest that ROS generation is significantly increased in the IFM, which may be the result of enhanced O₂^•− production.

Fig. 3.

Mitochondria subpopulation ROS generation. Individual mitochondrial subpopulations were isolated from diabetic and control hearts, and electron paramagnetic resonance (EPR) spectroscopy was performed to index ROS generation. Isolated mitochondrial subpopulations were incubated with 5,5-dimethyl-1-pyrroline-N-oxide in the presence or absence of excess complex I respiratory substrates glutamate and malate, incubated for 3 min at 37°C, and then transferred to an EPR flat cell for measurement at room temperature with instrument settings of 63.6 mW, modulation amplitude 1 G, receiver gain 1.00 × 10⁵, conversion time 40.960 ms, and time constant 40.960 ms. All spectra shown are an accumulation of 3 scans. A: representative spectra for control SSM. B: representative spectra for diabetic SSM. C: representative spectra for control IFM. D: representative spectra for diabetic IFM. E: representative spectra for SSM with the inclusion of SOD into the assay medium to confirm the superoxide (O₂^•−) signal. F: representative spectra for IFM with the inclusion of SOD into the assay medium to confirm the O₂^•− signal. G: xanthine oxidase (XO) spectra were included as a representative of the O₂^•− spectra pattern. H: ROS formation was significantly increased in diabetic IFM compared with control IFM. Values are expressed as means ± SE; n = 4 for each group. †P < 0.01 for diabetic vs. control groups.

Oxidative damage to proteins and lipids.

To determine whether oxidative stress levels are greater in a specific mitochondrial subpopulation, we examined protein nitrotyrosine contents as well as the levels of lipid peroxidation breakdown products MDA and 4-HAE in both mitochondrial subpopulations after diabetic insult. Nitrotyrosine content was significantly greater in diabetic IFM relative to control, and no significant differences were observed in SSM (P < 0.05; Fig. 4). MDA and 4-HAE were significantly higher in both diabetic subpopulations compared with control (P < 0.01 for SSM and P < 0.005 for IFM; Fig. 5). These data indicate that STZ-induced diabetic insult enhances protein modification as a result of nitrosylation in diabetic IFM, with no significant impact on SSM. Furthermore, lipid peroxidation is enhanced in both mitochondrial subpopulations, but this increase may be greater in diabetic IFM.

Fig. 4.

Mitochondrial subpopulation protein nitrotyrosine content. Individual mitochondrial subpopulations were isolated from diabetic and control hearts. Assessment of protein nitrotyrosine residues was performed using a sandwich ELISA, and results were compared against a known protein nitrotyrosine standard curve. Results are expressed as nanomoles of nitrotyrosine protein per milligram of protein. Values are expressed as mean ± SE; n = 5 for each group. *P < 0.05 for diabetic vs. control groups.

Fig. 5.

Mitochondrial subpopulation lipid peroxidation content. Individual mitochondrial subpopulations were isolated from diabetic and control hearts. Assessment of the lipid peroxidation byproducts malondialdehyde (MDA) and 4-hydroxyalkenal (4-HAE) was performed using a colorimetric assay and spectrophotometric measurement at 586 nm. Sample values were compared against a standard curve of known 4-HAE and MDA, and data are expressed as micromoles of MDA + 4-HAE per milligram of protein. Values are expressed as means ± SE; n = 5 for each group. †P < 0.01 for diabetic vs. control groups; ††P < 0.005 for diabetic vs. control groups.

Mitochondrial cardiolipin content.

Cardiolipin is a phospholipid unique to the mitochondrial inner membrane that is a diphosphatidylglycerol, containing oxidatively sensitive acyl groups that may act as selective targets for ROS. Depending on the pathological state, cardiolipin content within mitochondrial subpopulations has been shown to be affected differently (34). Because cardiolipin is susceptible to ROS (46), we examined whether diabetic cardiomyopathy is associated with mitochondrial subpopulation-specific alteration and, thus, subcellular spatial influence. MS analyses were performed to detect alterations in concentrations of the most abundant cardiolipin molecular species (T18:2 cardiolipin, m/z 723) (15). Cardiolipin and its internal standard gave both singly and doubly charged molecules (representative spectra for IFM are shown in Fig. 6, A and B). Cardiolipin was detected based on its doubly charged signal at m/z 723 and its singly charged ion at m/z 1447. The internal standard (T14:0 CL) was detected based on doubly and singly charged signals at m/z 619 and m/z 1238, respectively. A ratio of cardiolipin to the internal standard was calculated by single ion monitoring of both m/z 723 and m/z 619 to determine the change in cardiolipin levels between control and diabetic samples. Using this approach, we observed a significant decrease of ∼60% in diabetic IFM compared with controls (P < 0.01; Fig. 6, A–C). In contrast, no significant differences were observed in diabetic SSM compared with controls (Fig. 6C). These data indicate that in response to the diabetic phenotype, cardiolipin content is significantly decreased in the IFM subpopulation, with no significant change in the SSM subpopulation.

Fig. 6.

Mitochondrial subpopulation cardiolipin content. Individual mitochondrial subpopulations were isolated from diabetic and control hearts, and lipids were extracted. Assessment of cardiolipin content was performed by electrospray ionization mass spectrometry (ESI-MS) on diluted samples from control and diabetic mitochondrial subpopulations. The ions of the predominant molecular species of cardiolipin (T18:2 CL, m/z 723) and the internal standard (T14:0 CL, m/z 619) are shown by arrows. A: representative negative ion ESI-MS spectra from control IFM. B: representative negative ion ESI-MS spectra from diabetic IFM. C: summary data of cardiolipin expressed as the ratio of m/z 723 to m/z 619. Values are expressed as means ± SE; n = 3 for each group. †P < 0.01 for diabetic vs. control groups.

DISCUSSION

Diabetes mellitus is associated with a cardiomyopathy that is independent of atherosclerosis and characterized by abnormal ventricular contractile function (2, 15, 22, 27, 52, 63). Diabetic cardiomyopathy has been shown to progress to heart failure in both type 1 and 2 diabetic models, although at a much higher rate in the type 1 setting (15, 52). Using a well-described and -utilized model of diabetes induction, STZ injection, we sought to determine the effect of diabetic insult on cardiac contractile function. STZ treatment had no significant impact on heart weight, body weight, or heart weight-to-body weight ratios (Table 1). In contrast, STZ treatment significantly decreased +dP/dt, −dP/dt, and DP compared with citrate saline-injected controls (Table 1). Our observations are in agreement with others who have observed similar cardiac contractile deficits in the diabetic context, characteristic of diabetic cardiomyopathy (2, 15, 22, 52, 63).

As the primary source of energy for the cardiac myocyte, mitochondria play a central role in cellular homeostasis. Not surprisingly, disruption of this critical organelle is regarded as a key contributor to the development of pathological states, including diabetic cardiomyopathy (50, 53, 54). Nevertheless, examination of mitochondria is complicated by the fact that two mitochondrial subpopulations are present in the cardiac myocyte: IFM, which situate between the contractile apparatus, and SSM, which exist beneath the plasma membrane. These two disparate mitochondrial subpopulations are distinguished by specific spatial arrangements, distinct functional properties, and differential responses to pathological conditions. In general, SSM have been reported to be larger and more complex internally, whereas IFM tend to be smaller, elongated, and densely packed between myofibrils (21, 31, 38, 44). Using flow cytometric analyses, we observed morphological differences between SSM and IFM that were similar to other reports (44, 49) using electron microscopy in cells and isolated mitochondria (Fig. 1). The examination of mitochondrial subpopulations from hearts subjected to STZ treatment revealed decreases in both size and complexity of the IFM population, with no significant effect on the SSM population (Fig. 1). Our findings are in agreement with those of Kelley et al. (28), who observed decreases in the size of mitochondria located along the Z line (presumably reflective of IFM) in the vastus lateralis of type 2 diabetic patients. In contrast, Shen et al. (54) observed mitochondrial swelling in a novel model of type 1 diabetes (the OVE26 mouse), suggestive of an enhanced apoptotic program. These conflicting results may be due to a number of differences between the studies. The OVE26 mouse model uses transgenic modification to alter calmodulin levels in the pancreatic β-cell, precipitating a profound early-onset diabetic phenotype (10). Furthermore, OVE26 transgenic mice were examined at ∼4 mo of age, a substantially longer time period of diabetes exposure than our present study of 5 wk. It is possible that mitochondria display a biphasic morphological response during diabetic insult, characterized by an initial decrease in size followed by an increase in size. In addition, examination of the total mitochondria fraction may be incapable of resolving the differences between individual mitochondrial subpopulations. In either case, these data indicate that mitochondria display changes in overall morphology as a result of diabetic insult and that these changes may be specific for a distinct subpopulation.

Because of its enhanced propensity for ROS-mediated damage, mitochondrial ETC function can be profoundly affected by oxidative stress-associated pathologies, including diabetes mellitus. We (8) have previously observed changes in ETC respiration during diabetic insult, which is in agreement with other reports (28, 30). Specifically, we observed significant decreases in oxygen consumption at both complex I and complex II but not at complex IV (8). However, examination of the effect of diabetes on individual mitochondrial subpopulations revealed a differential response with decreased respiration at complex I and II and decreased activity at complex III in IFM, whereas only complex II respiration was decreased in SSM (Fig. 2). Although not significant, we observed a decrease in complex IV respiration of diabetic IFM (Fig. 2C). These results are similar to others (11, 22) that have found decreased complex III and complex IV respiration in the IFM population with no effect in the SSM population in response to aging. Taken together, our findings indicate that ETC complex II may be a specific locus for the deleterious effects associated with STZ insult, imparting damage to both IFM and SSM subpopulations, whereas the effects on complex I and III may be specific to IFM. Our results suggest that with type 1 diabetes mellitus, ETC function is compromised, impacting both subpopulations, with the effects being greatest in IFM. Because the IFM population provides ATP for the contractile process, the compromise in ETC function of this mitochondrial population may be particularly detrimental to cardiac contractile function, contributing to the contractile deficits associated with diabetic cardiomyopathy.

It has been suggested that an increase in ROS in response to hyperglycemia is the proximal defect that leads to many of the subsequent pathological consequences resulting from diabetes mellitus (6, 13, 28, 50). Studies have indicated that various ROS are increased during diabetes mellitus in the heart, including O₂^•− (7, 35), peroxynitrite (43), ·OH (41, 45), and H₂O₂ (36). Using the nonspecific fluorometric probe dichlorofluorescein (DCF), we (8) have previously observed an increase in ROS production in total diabetic mitochondria. Our results are in agreement with others (50) observing similar increases in ROS presence resulting from diabetic insult using DCF as a probe. However, because the DCF probe detects ROS in a nonspecific manner, we extended our findings by determining the specific reactants that are increased in cardiac diabetic mitochondria using an EPR spin trapping methodology. Our results indicate that diabetic insult significantly increases O₂^•− only in IFM, suggesting that increased ROS occurs in a subpopulation-specific manner (Fig. 3). Verification of the presence of O₂^•− was accomplished by the addition of exogenous SOD, which essentially abolished the signal (Fig. 3, E and F). Comparison of the observed spectra to the spectra of a known O₂^•− generator, xanthine oxidase (Fig. 3G), provided further confirmation to the nature of the reactant. These interesting results indicate that the enhanced ROS presence as a result of diabetes mellitus is particularly relevant in the IFM subpopulation and that O₂^•− appears to be a primary ROS generated in diabetic mitochondria. Our findings are similar to others (26, 57) examining different cardiac pathologies using DCF detection or an amplex red assay. Moghaddas et al. (37) found an increase in H₂O₂ from complex III in IFM with aging and no differences in SSM. Indirectly, our results indicating an enhanced O₂^•− presence are also in agreement with Shen et al. (53), who observed a restoration of mitochondrial function in MnSOD transgenic mice crossed with OVE26 mice. It should be pointed out that the primary sites for ROS generation in the mitochondrion are at complex I and III; thus, decreases in respiration of these complexes, as observed in IFM, should limit electron flux into complex III, attenuating ROS generation. Our results do not support this scenario but rather may be the function of enhanced oxidative damage to proteins in these complexes such that, although electron flux through these complexes is decreased, electron leakage is greater, resulting in the amplification of ROS generation. The results warrant further investigation.

Overproduction of ROS represents an initial event in the pathogenesis of diabetic cardiomyopathy creating an oxidative imbalance in the cell. The resulting oxidative milieu promotes damage to subcellular components such as membranes and proteins. Submitochondrial targets are particularly prone to damage from enhanced ROS due to their close proximity to the source of generation. Because we observed increases in mitochondrial ROS generation as a result of STZ exposure, we examined oxidative modification to both protein and lipid components in individual mitochondrial subpopulations to determine whether oxidative damage occurs in a subpopulation-specific manner. Assessment of protein modification was performed by analysis of nitrotyrosine adducts. Our results indicate that only proteins from the IFM population display enhanced nitrosylated proteins in response to STZ exposure (Fig. 4). These findings are similar with others (26) who have demonstrated enhanced protein carbonyl groups in cardiac IFM as a result of aging. Interestingly, these authors (26) also observed an increase in protein carbonyls in the SSM population as a result of aging, although this increase was not as great as that observed in IFM. The reason for these differences may be related to the pathologies studied (aging vs. diabetes mellitus) and/or the specific protein modification examined (nitrotyrosine vs. carbonyl). Lipid peroxidation was determined by the combined assessment of both MDA and 4-HAE (Fig. 5). Our results indicate that STZ exposure increases lipid peroxidation in both SSM and IFM subpopulations, although the increase may be greater in IFM (Fig. 5). Our findings are somewhat in agreement with others (26, 57) who have observed increases in lipid peroxidation primarily in cardiac IFM as a result of aging. Of particular interest in the present study was our observation that STZ exposure enhanced lipid peroxidation in both subpopulations, although the increase appeared to be larger in IFM. Our findings are in contrast to previous reports (26, 57) and may be a function of the different pathologies examined and/or assay methodology. These studies examined lipid peroxidation in aging cardiac mitochondrial subpopulations by assessing 4-HAE via Western blot analyses and/or MDA via spectrophotometric measurement of thiobarbituric acid-reactive substances. In our present study, we assessed lipid peroxidation by assessing a combination of MDA and 4-HAE spectrophotometrically using the chromogenic reagent N-methyl-2-phenylindole. Additionally, oxidative stress to mitochondrial lipid components may be different in the aging context compared with diabetic insult. It is important to point out that while STZ effects are believed to be primarily associated with the pancreas, its short-term administration has been associated with ROS generation and oxidative damage to other tissues such as the kidney and liver (47). Thus, one must consider the potential for a direct effect of STZ on the heart independent of enhanced glucose content resulting from pancreatic β-cell loss.

Diabetic mitochondria display increases in mitochondrial permeability transition pore (mPTP) opening (42) as well as changes in mitochondrial membrane fluidity, which may contribute to changes in mitochondrial membrane potential (ΔΨ_m) (61). These effects may be influenced by phospholipid content and, in particular, cardiolipin, which would have a profound impact on mitochondrial function (16). Cardiolipin is a diphosphatidylglycerol enriched in the inner membrane containing oxidatively sensitive acyl groups that may act as selective targets for ROS (17, 40). Cardiolipin interacts with a number of mitochondrial proteins, including components of the mPTP, cytochrome c, and ETC complexes; thus, its oxidative modification may be a critical event for the initiation of pathological states (23). We determined whether cardiolipin was a specific lipid target of STZ-induced diabetic insult. Our findings indicate that loss of cardiolipin occurs as a result of STZ exposure only in the IFM subpopulation (Fig. 6C). Our results are in contrast to those of Lesnefsky et al. (34), who observed a selective decrease in cardiolipin content in SSM after I/R insult. These authors (34) suggested that SSM possess a decreased capacity relative to IFM for Ca²⁺ accumulation, which is a hallmark of I/R insult. As a result, SSM sustain a faster onset for ischemic damage compared with IFM (34). Although increased Ca²⁺ content may be involved in the pathogenesis of diabetes mellitus, its involvement has much greater effects during myocardial I/R, which may account for the difference in the results observed in the studies.

In conclusion, we report here, for the first time, that STZ-induced diabetic insult differentially affects morphological, functional, and oxidative properties of spatially distinct mitochondrial subpopulations. Type 1 diabetic insult is associated with a greater stress on the IFM subpopulation, as indicated by increased ROS and oxidative damage, impacting ETC function and cardiolipin content, all of which may contribute to cardiac contractile dysfunction. The heterogeneous response displayed by individual mitochondrial subpopulations during diabetic insult emphasizes the importance of incorporating spatial influence into the study of mitochondria in disease states.

GRANTS

This work was supported by National Institutes of Health (NIH) Award DP2-DK-083095 (to J. M. Hollander). This work was also supported by American Heart Association (AHA) Beginning Grant-In-Aid 0665237B (to J. M. Hollander) and Grant-In-Aid 0855484D (to J. M. Hollander). E. Dabkowski was a recipient of AHA Predoctoral Fellowship 0815406D. Flow cytometry experiments were supported in part by NIH Grants RR-020866 and RR-16440.

DISCLAIMER

The findings and conclusions in this report are those of the author(s) and do not necessarily represent views of the National Institute for Occupational Safety and Health.