Effects of Palmitate on Ca²⁺ Handling in Adult Control and ob/ob Cardiomyocytes : Impact of Mitochondrial Reactive Oxygen Species

Fluo-3 AM, tetra-methyl rhodamine-ethyl ester (TMRE), and MitoSOX Red were from Molecular Probes/Invitrogen. Carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP), N-acetylcysteine (NAC), ebselen, palmitate, and laminin were from Sigma. All compounds were prepared as stock solutions in appropriate solvents. On the day of the experiment, stock solutions were diluted to the desired final concentration in the bath solution; when required, the same concentration of solvent was added to the control solution.

We used young (aged 3–5 months) C57BL leptin-deficient, obese male mice (ob/ob) and their wild-type counterparts (Taconic, Lille Skensved, Denmark). We recently measured body weights of ∼50 and 27 g, respectively, in ob/ob and wild-type mice of this age range (14). Mice were killed by rapid neck disarticulation, and the heart was excised. All experiments were approved by the Stockholm North local ethical committee. Single cardiomyocytes were isolated from the ventricles following the protocols developed by the Alliance for Cellular Signaling (AfCS Procedure Protocol ID PP00000 125) (15). After being loaded with fluorescent indicators (see below), cardiomyocytes were put on laminin-coated coverslips that made up the bottom of the perfusion chamber. Cells were allowed to attach to the coverslip for ∼5 min before the experiment started. They were then superfused with standard Tyrode solution (in mmol/l): 121 NaCl, 5.0 KCl, 1.8 CaCl₂, 0.5 MgCl₂, 0.4 NaH₂PO₄, 24 NaHCO₃, 0.1 EDTA, and 5.5 glucose. Palmitate (1.2 mmol/l), bound to 0.06% BSA (3), was added in some experiments. This palmitate concentration is similar to the serum concentration of fed control mice and postabsorptive ob/ob mice (3). The Tyrode solution was bubbled with 5% CO₂/95% O₂, which gives a bath pH of 7.4, and experiments were performed at room temperature (∼24°C). Cells were continuously stimulated at 1 Hz with 1-ms current pulses delivered via two platinum electrodes, one on each side of the perfusion chamber. Measurements were only performed on rod-shaped cells that displayed a uniform contraction in response to each stimulation pulse and showed no spontaneous activity.

The free cytosolic [Ca²⁺] was measured with the fluorescent Ca²⁺ indicator fluo-3 and confocal microscopy using a Bio-Rad MRC 1024 unit attached to a Nikon Diaphot inverted microscope with a Nikon Plan Apo 60× oil immersion objective (NA 1.3) (14). Confocal images were obtained by line scanning along the long axis, with focus in the middle of the cell. Stored confocal images were analyzed with ImageJ (National Institutes of Health [available at http://rsb.info.nih.gov/ij]). To enable comparisons between cells, changes in the fluo-3 fluorescence signal (ΔF) were divided by the fluorescence immediately before a stimulation pulse was given under control conditions (F₀). The time course of Ca²⁺ transients was assessed by measuring the half-width (D_1/2 [i.e., the duration at 50% of ΔF]) and the time constant (τ) of the exponential part of the decay phase, ignoring the initial phase that often diverges from a monoexponential function. The edges of the fluorescence images of the cells were followed during contractions, which allowed determination of cell shortening simultaneously with measurements of Ca²⁺. The cell length was measured in the rested state and the maximally contracted state, and shortening is expressed as a percentage of the resting cell length. Measurements of Ca²⁺ transients and cell shortening were performed immediately before and 20 min after palmitate exposure.

TMRE was used to measure mitochondrial membrane potential (ΔΨ_m) (16). Isolated cardiomyocytes were loaded with TMRE (1 μmol/l) in Dulbecco's modified Eagle's medium for 15 min at room temperature, followed by washout in medium without TMRE. Confocal images of TMRE fluorescence were obtained by excitation at 568 nm while measuring the emitted light at 585 nm. To minimize the impact of subcellular variability in ΔΨ_m, TMRE fluorescence was measured in five different areas in each cell (16). Images were taken every 5 min, and fluorescence signals were normalized to the fluorescence measured in each cell at the start of the experiment, which was set to 100%. At the end of each experiment, cells were exposed to the mitochondrial uncoupler FCCP (1 μmol/l) to determine the dynamic range of the dye.

MitoSOX Red was used to measure mitochondrial ROS production (17). Isolated cardiomyocytes were loaded with MitoSOX Red (5 μmol/l) in Dulbecco's modified Eagle's medium for 15 min at room temperature, followed by washout. Confocal images were obtained at 5-min intervals by excitation at 488 nm and measuring the emitted light at 585 nm. MitoSOX Red fluorescence was measured at each time point as the mean of five measurements performed at different areas in each cell. The signal from each cell was normalized to that at the start of the experiment, which was set to 100%. As a positive control, 1 mmol/l of H₂O₂ was added at the end of each experiment, and this resulted in a marked increase in the fluorescence signal in all cells (data not shown).

Immediately after killing the mouse, the heart was excised and the ventricles were dissected and frozen in liquid nitrogen. The ventricles were homogenized (50 μl/mg wet wt) with ground-glass homogenizers in ice-cold buffer consisting of (in mmol/l): 50 Tris-HCl, 5 sodium citrate, 0.6 MnCl₂, 1 cysteine, and 0.05% (vol/vol) Triton X-100 (pH 7.4). The homogenate was centrifuged at 4°C at 1,400g for 1 min, and an aliquot of the supernatant immediately was assayed for aconitase activity following the conversion of citrate to isocitrate coupled with isocitrate dehydrogenase (measuring the production of NADPH at 340 nm) (18). The remaining supernatant was stored at −80°C until assayed for citrate synthase activity (19) and protein (Bio-Rad). The protein concentration was not different between wild-type and ob/ob hearts (88.1 ± 0.7 vs. 87.4 ± 2.4 μg/mg wet wt; n = 4). The enzyme activities of each heart were corrected for protein concentration.

Frozen ventricles were homogenized in lysis buffer consisting of 50 mmol/l KH₂PO₄ (pH 7.5), 1 mmol/l EDTA, 0.05% Triton X-100 (vol/vol), and protease inhibitor cocktail (Roche). Lysates were cleared by centrifugation at 1,400g for 30 s at 4°C. The protein content was determined using the Bradford assay (Bio-Rad). Equal amounts of total protein (30 μg) were loaded into each well and separated by electrophoresis using NuPAGE Novex 4–12% Bis-Tris Gels (Invitrogen) and transferred onto a polyvinylidine fluoride membrane (Bio-Rad). Membranes were blocked in 5% (wt/vol) nonfat milk in Tris-buffered saline containing 0.05% Tween 20, followed by incubation with primary antibody (anti-SOD2, Biosite; 1:1,000 dilution). Membranes then were incubated with horseradish peroxidase–conjugated antibody (anti-goat Ig; 1:2,000 dilution), and immunoreactive bands were visualized using chemiluminescence (Super Signal; Pierce). Membranes then were stripped and reblotted against the dihydropyridine receptor (anti-DHPR, Abcam; 1:500 dilution), which served as a loading control, followed by incubation with horseradish peroxidase–conjugated antibody (anti-mouse Ig; 1:1,000 dilution).

Data are presented as means ± SE, and the number of cardiomyocytes (n) is given. For each experimental condition, cells were isolated from at least three age-matched mice, except in the ebselen experiments, where only two mice were used in each condition. Statistically significant differences were assessed with Student's t test (for paired or unpaired samples) or, when three or more groups were compared, one-way ANOVA with a Newman-Keuls post hoc test. P < 0.05 was considered significant.

Under control conditions, Ca²⁺ transients were smaller and slower in ob/ob compared with wild-type cardiomyocytes (Fig. 1; Table 1) (14). In wild-type cells, application of palmitate significantly decreased the Ca²⁺ transient amplitude by ∼20% and increased half-width as well as the decay time constant (Table 1). In ob/ob cardiomyocytes, on the other hand, palmitate did not have any significant effect on the Ca²⁺ transient amplitude, but the transients became significantly faster (Fig. 1; Table 1).

In the presence of palmitate, relative cell shortening was decreased by ∼20% in wild-type cardiomyocytes (Fig. 2Aa), whereas it was increased by ∼40% in ob/ob cells (Fig. 2Ba). Palmitate significantly increased the ECC efficiency (i.e., the ratio between the relative cell shortening and the amplitude of the Ca²⁺ transient) by ∼25% in ob/ob cells (Fig. 2Bb). In wild-type cardiomyocytes, on the other hand, palmitate had no significance in the ECC efficiency (Fig. 2Ab) due to the concomitant decreases in cell shortening and Ca²⁺ transient amplitude.

The SR Ca²⁺ content was assessed by measuring the amplitude of cytosolic Ca²⁺ transients induced by rapid application of caffeine (20). In these experiments, the steady-state Ca²⁺ transients were established during 25- min continuous stimulation in the absence or presence of palmitate. The stimulation was then stopped and the superfusate was rapidly switched to a solution containing caffeine (10 mmol/l). In the presence of palmitate, the amplitude of the caffeine-induced Ca²⁺ transient was decreased by ∼25% in wild-type cardiomyocytes (Fig. 3A), whereas it was not affected in ob/ob cells (Fig. 3B). To sum up, in wild-type cells, palmitate decreased the amplitude of electrically evoked Ca²⁺ transients and the SR Ca²⁺ content, and this led to a reduced cell shortening. In ob/ob cardiomyocytes, on the other hand, palmitate made the Ca²⁺ transient faster without affecting the SR Ca²⁺ content and cell shortening was enhanced.

Increased mitochondrial ROS production and oxidative stress are known to affect Ca²⁺ handling in cardiac cells (12,13). Therefore, we hypothesized that the impaired cellular Ca²⁺ handling induced by palmitate in wild-type cells was caused by increased mitochondrial ROS production. To test this hypothesis, we measured cytosolic Ca²⁺ transients and cell shortening in the presence of the nonspecific antioxidant NAC (20 mmol/l). In wild-type cardiomyocytes, palmitate application in the presence of NAC did not affect the Ca²⁺ transient characteristics (Fig. 1Ab; Table 2), the cell shortening or the ECC efficiency (Fig. 2A), or the SR Ca²⁺ content (Fig. 3A). In ob/ob cardiomyocytes, on the other hand, NAC had no effect on the palmitate-induced increases in Ca²⁺ transient kinetics, shortening amplitude, and ECC efficiency (Figs. 1Bb and 2B; Table 2), whereas it increased the SR Ca²⁺ content (Fig. 3B).

We also studied Ca²⁺ transients in the presence of another antioxidant, ebselen (5 μmol/l), which acts as a glutathione peroxidase mimetic that removes hydrogen peroxide in the presence of reduced glutathione (21). The effects of ebselen were similar to those obtained with NAC (Table 2). Thus, in the presence of ebselen, palmitate had no effect on Ca²⁺ transient characteristics in wild-type cells, whereas palmitate still made the Ca²⁺ transients faster in ob/ob cells (although the decrease in τ did not reach statistical significance). In summary, these results with antioxidants indicate that the palmitate-induced changes in cellular Ca²⁺ handling in wild-type cardiomyocytes were mediated by increased ROS production, whereas the palmitate effects in ob/ob cells mostly were ROS independent.

Application of palmitate resulted in a highly significant decrease in the TMRE fluorescence within 5 min in wild-type cardiomyocytes, whereas there was no change in ob/ob cells over the 25-min observation period (Fig. 4). Addition of FCCP (1 μmol/l), a mitochondrial uncoupler that dissipates ΔΨ_m, resulted in a prompt decrease of the TMRE fluorescence in both wild-type and ob/ob cardiomyocytes. After 5 min exposure to FCCP, the TMRE signal was similar in wild-type and ob/ob cells (Fig. 4A).

We next tested the effect of palmitate on mitochondrial ROS production using MitoSOX Red, which is targeted to the mitochondria and increases its fluorescence when oxidized. Palmitate application caused a rapid increase in the MitoSOX Red signal in wild-type cardiomyocytes but not in ob/ob cardiomyocytes (Fig. 5). Thus, these results show that palmitate exposure induced a marked depolarization of ΔΨ_m and increased the mitochondrial ROS production in wild-type but not in ob/ob cardiomyocytes.

While MitoSOX Red readily detects changes in mitochondrial ROS production during an experiment, it is less suitable for detecting differences in basal mitochondrial ROS production between different cell populations. This is because it is a nonratiometric indicator, and, hence, the basal fluorescence depends on, for instance, the concentration of the dye. Thus, the MitoSOX Red results presented in Fig. 5 do not provide any information regarding possible differences in mitochondrial ROS production in wild-type and ob/ob hearts. One way to assess this is to measure the aconitase activity because the function of this mitochondrial enzyme critically depends on an iron-sulfur cluster, and proteins with such clusters may lose activity in response to increased mitochondrial oxidative stress (18). The aconitase activity was similar in wild-type and ob/ob hearts (Fig. 6A). We also measured the citrate synthase activity, which is frequently used to assess the cellular mitochondrial content, and found no difference between wild-type and ob/ob hearts (127 ± 3 vs. 131 ± 3 μmol · min⁻¹ · g⁻¹ wet wt; n = 4).

Mitochondrial manganese-dependent superoxide dismutase (MnSOD or SOD2) plays a central role in the defense against superoxide ions produced by respiration. The expression of SOD2 generally is upregulated in response to an increased mitochondrial ROS production (22). We used Western blots to measure SOD2 protein expression and found no difference between ob/ob and wild-type hearts (Fig. 6B).

Inhibition of glycolysis and altered glucose oxidation may cause Ca²⁺ alternans, which is defined as Ca²⁺ transients of alternating amplitude on a beat-to-beat basis (23–25). In the normal Tyrode solution, no wild-type cells displayed Ca²⁺ alternans, whereas it was present in 19% of the ob/ob cells, i.e., 7 of 37 cells. (These cells were not included in the Ca²⁺ transient analyses described above because the amplitude and kinetics varied on a beat-to-beat basis.) After application of palmitate, Ca²⁺ alternans were abolished in the ob/ob cells that displayed this behavior (Fig. 7). Thus, no Ca²⁺ alternans were observed in either wild-type or ob/ob cardiomyocytes in the presence of palmitate.

The heart is metabolically versatile, switching its preferred substrate depending on the environmental and physiological or pathological conditions (1). In line with this, recent studies (3,4) showed markedly different responses when hearts from wild-type and insulin-resistant ob/ob mice were exposed to the saturated fatty acid palmitate. The aim of the present study was to reveal cellular mechanisms underlying these differences. Our results show that application of palmitate had adverse effects in wild-type cardiomyocytes (i.e., it impaired intracellular Ca²⁺ handling and contraction, depolarized ΔΨ_m, and increased mitochondrial ROS production). In ob/ob cardiomyocytes, on the other hand, palmitate exposure improved cellular Ca²⁺ handling and contraction and prevented the occurrence of Ca²⁺ alternans, whereas it had no effect on ΔΨ_m or ROS production.

Hyperglycemia and high serum free fatty acid levels are important contributors to the pathological adaptations in diabetes, and they share at least one pathogenic mechanism: overproduction of ROS by the mitochondrial electron-transport chain (26–28). Moreover, several studies (27,29–31) have demonstrated that long-chain saturated fatty acids, such as palmitate, can promote apoptosis in a variety of cell types, including adult cardiomyocytes. Palmitate-induced proapoptotic signaling has been associated with ΔΨm dissipation and increased ROS production (27,29,30). The mitochondrial ROS production is likely to occur mainly at the level of complexes I and III of the mitochondrial respiratory chain, and the function of both these complexes has been shown to be altered by palmitate (29,32). Increased ROS production may impair cellular Ca²⁺ handling by decreasing the l-type Ca²⁺ current amplitude, increasing the open probability of the SR Ca²⁺ release channels, slowing SR Ca²⁺ reuptake, and increasing the sarcolemmal Na⁺/Ca²⁺ exchange activity, eventually leading to reduced SR Ca²⁺ content (11,33–35). Thus, the present results in wild-type cardiomyocytes all can be explained by a model in which palmitate dissipates ΔΨm and increases mitochondrial ROS production (Figs. 4 and 5), which results in impaired SR Ca²⁺ handling and, consequently, decreased shortening (Figs. 1–3). In accordance with this model, palmitate had no effect on SR Ca²⁺ handling and contraction in wild-type cells in the presence of antioxidants.

In contrast to the situation in wild-type cardiomyocytes, palmitate application did not affect the ΔΨm or mitochondrial ROS production in ob/ob cardiomyocytes. Accordingly, antioxidants had no effect on palmitate-induced changes in Ca²⁺ handling and contraction in ob/ob cells. Furthermore, aconitase activity and SOD2 protein expression were not significantly different between wild-type and ob/ob cardiomyocytes (Fig. 6), which indicates no major difference in mitochondrial ROS production in vivo between the two groups despite the fact that ob/ob hearts are exposed to a markedly higher concentration of saturated fatty acids (3).

A number of studies have reported an increased fatty acid oxidation associated with a marked reduction of glycolysis and glucose oxidation in obesity and type 2 diabetes (2–4,36–38). This metabolic switch suggests a faster ATP production with palmitate than with glucose in ob/ob cardiomyocytes. Accordingly, palmitate induced a NAC- and ebselen-independent increase in the rate of SR Ca²⁺ reuptake, the extent of cell shortening, and the ECC efficiency in ob/ob cells. These processes depend on ATPase activity and would be accelerated by an improved mitochondrial ATP production, resulting in an increased ATP-to-ADP ratio and decreased inorganic phosphate ion concentration (39–41). Furthermore, inhibition of energy metabolism leading to decreased ATP formation has been shown to impair SR Ca²⁺ release and reuptake (39,42,43), and this can lead to unstable Ca²⁺ cycling in a beat-to-beat basis and favor the occurrence of Ca²⁺ alternans (44). Although the exact mechanisms of Ca²⁺ alternans still is unclear, an alternating SR Ca²⁺ load seems to be a major event in triggering Ca²⁺alternans, and faster SR Ca²⁺ uptake may stabilize the SR Ca²⁺ load and prevent Ca²⁺ alternans (25,45). Accordingly, Ca²⁺ alternans frequently were observed in ob/ob cells in the presence of glucose, but never after application of palmitate, where the rate of SR Ca²⁺ uptake was increased.

Based on the present results, we propose a model in which high fatty acid levels initially induce increased mitochondrial ROS production, which has serious negative effects on cardiomyocyte function. During prolonged exposure to increased fatty acid levels, as in ob/ob mice, cardiomyocytes adapt so that fatty acids are preferred to glucose as a substrate for energy metabolism, and, hence, fatty acid exposure results in improved cellular Ca²⁺ handling and contraction.

The present study was supported by the Swedish Research Council (project no. 10842 and 14453), the Swedish Heart and Lung Foundation, the Swedish Diabetes Foundation, and Funds at the Karolinska Institutet.