INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

CALCIUM IONS SERVE a dual-messenger role in the heart in that they not only mediate excitation-contraction coupling during cardiac work and ATP hydrolysis but also activate dehydrogenases and the generation of ATP by oxidative phosphorylation, allowing maintenance of a balance of energy supply and demand (14, 16, 18, 22, 31). There are three Ca²⁺-activated dehydrogenases, namely, NAD-isocitrate, 2-oxoglutarate, and pyruvate dehydrogenase (PDH), which catalyze near-irreversible reactions in the terminal oxidation of carbohydrate and lipid substrate (16). These enzymes reside in the mitochondrial (Mito) matrix, within the permeability barrier imposed by the inner Mito membrane, and their activation requires the transmission of changes in cytosolic Ca²⁺ into the mitochondria, a process that involves separate electrophoretic uptake (uniporter) and electroneutral 2 Na⁺/Ca²⁺ release (antiporter) pathways (10, 51). Ruthenium red blocks Mito net uptake of Ca²⁺ by impeding uniporter function (11) and thus prevents the activation of PDH (17, 30, 51). Ca²⁺ activates a phosphatase that dephosphorylates three serine residues of the E₁ subunit of PDH to form catalytically active PDH_A (53). This action confers stable interconversion and thus permits preservation of the in situ fractional activation of PDH_A/PDH_Total during tissue extraction under appropriate conditions (21).

Although Ca²⁺ play a positive role in control of cardiac Mito metabolism under physiological conditions, massive Mito accumulation of Ca²⁺ may occur following ischemia and can lead to Mito damage and cell death. The final step in this series of events has been proposed to be the Ca²⁺-dependent opening of a nonselective, high-conductance channel in the Mito inner membrane (3) known as the "mitochondrial permeability transition" (MPT) pore, which causes the loss of proton motive force and failure of ATP generation (10).

With increased age, the n-6 PUFA content of rat cardiac cell and inner Mito membrane has been shown to be elevated in contrast to a decline in n-3 PUFA, and these changes are associated with abnormal cell Ca²⁺ balance leading to enhanced Ca²⁺-dependent arrhythmogenesis during reperfusion following ischemia (32) and disparities in the specific activity and thermotropic kinetics of Mito membrane-bound enzymes (39). Consumption of a diet rich in "-3" or n-3 polyunsaturated fatty acids (PUFA) has been shown to increase the ratio of n-3 to n-6 PUFA content in cardiac cell membranes (23, 33, 45, 46). Compared with rats fed n-6 PUFA-rich diet, normoxic isolated working hearts from n-3 PUFA-treated rats exhibited a higher efficiency in terms of work performed per unit O₂ consumed (45). These differences were augmented following ischemia and reperfusion (I/R) such that O₂ utilization efficiency was decreased further in n-6 PUFA-treated hearts compared with n-3; however, this difference was abolished following ruthenium red-induced inhibition of Mito Ca²⁺ uptake (45).

The goal of the present study was to test whether direct modulation of the n-3-to-n-6 PUFA ratio in cardiac Mito membranes by dietary intervention could abolish age-associated changes in Mito membranes and rescue function in aged hearts by increasing their efficiency in Ca²⁺ handling and energy utilization. We postulated that a higher Mito Ca²⁺ content in postischemic n-6 PUFA hearts may be associated with higher Ca²⁺ cycling rates across the Mito membrane compared with n-3 hearts. Although this alone would cause a loss of efficiency, this might be compounded by excessive dehydrogenase activation, high values of proton motive force, and high rates of proton leakage across the Mito membrane. This would constitute a degree of uncoupling and a direct loss of efficiency according to the chemiosmotic mechanism of oxidative phosphorylation. We measured the fractional activation of Ca²⁺-dependent PDH (PDH_A/PDH_Total) under conditions of increased work or ischemic stress in isolated perfused hearts from 6-mo and 24-mo rats that were fed a n-3 PUFA- or n-6 PUFA-rich diet for 6 wk. Total Mito Ca²⁺ concentration ([Ca²⁺]) was measured in Mito preparations obtained using a rapid isolation technique that minimizes loss or gain of Ca²⁺ (29). Proton leak in mitochondria was inferred by measuring state 4 respiration (absence of ADP) with a variety of oxidizable substrates. The effects of age and dietary lipid on membrane PUFA, cardiolipin, and other phospholipids were measured in mitochondria isolated from ventricular myocardium.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Animals and dietary lipid supplementation. Male Wistar rats from the Gerontology Research Center Aging Colony were utilized in strict accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. Isocaloric diets were formulated by an 11.7% (wt/wt) addition of either animal fat (rich in n-6 PUFA) or marine fish oil (rich in n-3 PUFA) to a standard low-fat reference diet (Ref) that fulfilled essential fatty acid requirements for normal growth (46). The fat-supplemented diets contained 15.6% total fat, whereas the low-fat Ref diet contained 3.9% total fat (wt/wt). Purified fish oil was obtained from the NIH/National Oceanic and Atmospheric Administration Biomedical Test Material Program, National Marine and Fisheries Service, Charleston Laboratory, Charleston, SC.

All diets contained the same amount of antioxidant vitamins, including 0.05% butylated hydroxytoluene to prevent fat peroxidation during storage (4°C). The final ratio of n-3 to n-6 PUFA in the diets was 4, 0.2, and 0.2 for n-3 PUFA, n-6 PUFA, and Ref diets, respectively. The n-3 PUFA present in the Ref and n-6 PUFA diets (0.25%) was 18-carbon chain-length terrestrial origin -linolenic acid (18:3). Longer-chain n-3 PUFAs, such as eicosapentaenoic (20:5) and docosahexaenoic acids (22:6), were only present in the purified fish oil-enriched diet. Rats were fed for a minimum of 6 wk and were 6 and 24 mo old at the time of isolated heart perfusions.

Isolated heart preparation. After rapid cannulation of the aorta, coronary perfusion of the isolated isovolumic hearts was commenced at a constant pressure of 75 mmHg with a filtered (0.45 µm) bicarbonate buffer that consisted of (in mmol/l): 3.48 KCl, 116.4 NaCl, 26.2 NaHCO₃, 1.67 NaH₂PO₄, 0.69 MgSO₄, 1.5 CaCl₂, 11.1 glucose, and 0.2 octanoate. The buffer was gassed with 95% O₂-5% CO₂ and equilibrated at pH 7.38-7.42, 35°C. The coronary flow rate, which was monitored during all experimental protocols, was maintained at 19 ± 0.5 ml/min, but this was decreased to 1 ml/min to produce low-flow global ischemia. Four perfusion protocols were implemented: 3-Hz electrical pacing, with hearts stabilized for 15 min (Control); 5 Hz + 10⁷ M norepinephrine (NE) (stimulated increased work); 15-min low-flow ischemia + 5-min reperfusion, with 3-Hz electrical pacing throughout (I/R); and perfusion with 3.4 µM ruthenium red during control period, 15-min low-flow ischemia, and 5-min reperfusion, with 3 Hz throughout (I/R + RR).

Rapid Mito isolation. Mitochondria were prepared from these hearts by a rapid isolation method (29) that was designed to minimize Ca²⁺ redistribution. At the end of each protocol the ventricles were cut free, a portion was freeze-clamped in liquid N₂, and the rest was dropped directly into ice-cold medium containing (in mmol/l) 250 sucrose, 0.03 diltiazem, 0.0032 ruthenium red, 2 K-EGTA, and 10 K-HEPES, pH 7.4, and was homogenized for 20 s with a Brinkman Polytron (12,000-15,000 rpm). The homogenate was subjected to rapid centrifugation to isolate the mitochondria, as described previously (21).

Atomic absorption spectroscopy. A portion of mitochondria isolated from each heart subjected to one of the four perfusion protocols was stored at 80°C until Ca²⁺ content could be determined. Mito suspensions were digested in high-purity nitric acid. Mito Ca²⁺ was determined in samples containing 0.1% lanthanum oxide (to prevent phosphate interference) by atomic absorption spectrophotometric analysis (Perkin-Elmer Zeeman 5000) with CaCO₃ as the standard. Absorbance readings were corrected for lanthanum, and final values were expressed as nanomoles per milligram protein.

PDH and citrate synthase assays. We measured the fractional activation of PDH (PDH_A/PDH_Total) in freeze-clamped portions of ventricles according to the methods of Hansford et al. (21). Briefly, the frozen ventricular tissue was finely ground in a mortar under liquid N₂ with 10 volumes of a frozen "stop" solution (to prevent PDH_A-PDH_I interconversion) composed of (in mmol/l) 50 KP_i (pH 7.2), 3 EDTA, 25 NaF, 1 K-dichloroacetate, 1 1,4-dithiothreitol, 1 ADP, and 0.05 leupeptin. The ultrafine tissue powder was eventually permitted to reach 0°C on ice and thus thawed into a suspension that was centrifuged at 15,000 g for 5 min at 0°C. The supernatant was removed and stored on ice. PDH_A was assayed spectrophotometrically at 340 nm and 30°C by adding 100 µl of extract to 900 µl of the following medium (in mmol/l): 50 K-HEPES (pH 7.2), 1 MgCl₂, 4 1,4-dithiothreitol, 0.004 rotenone, 1.67 NAD, 0.2 thiamine pyrophosphate, 0.15 coenzyme A (CoA), 2 ADP, 0.08 EGTA, 16.7 L-lactate (pH 7.2), and 2 U/ml lactate dehydrogenase. Citrate synthase activity was assayed in a reaction mixture composed of 0.1 M Tris-Cl (pH 8), 0.1 mM DTNB, 100 µM acetyl CoA, and 0.05% Triton X-100. Extract (20 µl) was added, followed 1 min later by 0.5 µmol of oxaloacetate. The total volume was 2 ml, and the reaction was spectrophotometrically followed by an increase in absorbance at 412 nm.

Substrate oxidation by isolated mitochondria. Mitochondria were isolated from ventricular tissue using the Nagarse digestion method (29). Oxygen uptake of Mito suspensions was measured using an O₂ electrode (Yellow Springs Instruments) and a thermostat-regulated chamber of 2 ml. Temperature was maintained at 25 or 37°C as specified. The basal composition of the medium was (in mmol/l) 120 KCl, 20 K-HEPES (pH 7.2), 10 NaCl, 2 KP_i, and 1 mg/ml of fatty acid-poor bovine serum albumin. The respiratory substrates are specified below.

Mito membrane lipid analysis. Lipids were extracted by the method of Bligh and Dyer (5). The antioxidant butylated hydroxytoluene was included in the extract (0.1% of the lipid dry weight). Phospholipids were separated from the total lipid extract by thin-layer chromatography on silica gel H plates, which were developed in petroleum ether-diethyl ether-acetic acid (90:15:1). The phospholipids remaining at the origin were eluted from the silica gel and methylated in 1% (vol/vol) H₂SO₄ in methanol by heating at 70°C for 3 h. The fatty acid profiles of the diets and isolated Mito membrane phospholipids were determined by gas-liquid chromatography (5710A Hewlett-Packard) using columns packed with equal parts of SP 2310 and SP 2330 (Supelco).

Chemicals. Reagents were purchased from Sigma, St. Louis, MO, except acetyl CoA and oxaloacetate, which were from Calbiochem (LaJolla, CA). Lactate dehydrogenase and NAD were from Boehringer-Mannheim (Indianapolis, IN).

Statistical analyses. Data are presented as means ± SE. Group contrasts were tested where indicated by three-way ANOVA or by two-way ANOVA (SPSS for Windows). Individual comparisons were determined by Student-Newman-Keuls post hoc test. Significant differences were accepted at P < 0.05.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

PDH_A and Mito Ca²⁺ during -adrenergic receptor stimulation. Table 1 shows that under control perfusion conditions peak systolic pressure did not differ among any of the dietary or age groups. PDH_A levels tended to be higher in 24-mo compared with 6-mo hearts and this was significant after n-6 PUFA-rich diet (P < 0.05 vs. 6 mo). Although NE raised peak systolic pressure from control levels by more than twofold, an equivalent level of contractile performance was observed among all of the groups. NE markedly raised PDH_A in all groups compared with respective control perfused hearts. In 6-mo rat hearts, the highest level of PDH_A after control perfusion or NE was measured in the n-6 PUFA group and the lowest in the n-3 PUFA group. In Ref hearts, the effect of NE stimulation was significantly greater when they were aged 24 mo compared with 6 mo (P < 0.05). However, in 6-mo n-6 PUFA hearts, NE treatment elevated PDH_A to a level equal to that measured in 24-mo n-6 PUFA hearts. Values of PDH_A for NE-stimulated hearts were lower in the n-3 PUFA group than in the n-6 PUFA group, and no significant difference in PDH_A occurred among 6- and 24-mo n-3 PUFA hearts. A three-way ANOVA indicated significant effects of age, dietary lipid, and treatment with 10⁷ M NE (F = 30.8, P < 0.001) on the activation of PDH (PDH_A/PDH_Total). There were no significant three-way or two-way interactions.

View this table: [in this window] [in a new window]   Table 1.   Effect of 107 M NE in perfused hearts on PSP, PDHA, and Mito [Ca2+]

PDH activation and Mito [Ca²⁺] following I/R. When hearts were subjected to 15 min of low-flow (1 ml/ml) ischemia followed by 5 min of reperfusion (I/R), there was a striking increase in PDH_A (Fig. 1). Equally striking was the prevention of this increase by the Mito Ca²⁺ uptake blocker, ruthenium red. Highly significant effects of age, diet, and I/R treatment on PDH_A were observed (3-way ANOVA, F = 40.5, P < 0.0001; no significant 3-way interactions). PDH_A was significantly greater in the postischemic n-6 PUFA group than in the n-3 PUFA group, with the Ref group displaying intermediate values (diet × I/R, F = 17, P < 0.0001). PDH_A values tended to be higher in 24-mo Ref and n-6 PUFA hearts compared with 6-mo hearts (NS). I/R made little impact on PDH_A levels in n-3 PUFA hearts compared with 24-mo Ref and n-6 PUFA hearts.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Effect of ischemia and reperfusion (I/R) and I/R with 3.4 µM ruthenium red (RR) following steady-state control perfusion (C) on the activation of pyruvate dehydrogenase (PDHA) in perfused hearts from each diet and age group. PUFA, polyunsaturated fatty acid. Values are means ± SE; n = 6. * P < 0.05 vs. C.  P < 0.05 vs. reference diet (Ref).  P < 0.05 vs. n-6 PUFA.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Effect of C, I/R, and I/R with 3.4 µM RR on total mitochondrial calcium content (nmol/mg protein) as measured by atomic absorption spectroscopy. [Ca2+], Ca2+ concentration. Values are means ± SE; n = 6. * P < 0.05 vs. C.  P < 0.05 vs. Ref.  P < 0.05 vs. n-6 PUFA.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Activation of cardiac Mito dehydrogenases, including PDH, by Ca²⁺ is considered to lead to the maintenance of NADH/NAD⁺ and the proton-motive force required for oxidative phosphorylation in the face of increased energy demand (6, 18). The present study, by directly manipulating the cardiac Mito membrane ratio of n-3 to n-6 PUFA, has demonstrated that the activation of Ca²⁺-dependent PDH can be augmented when the n-3-to-n-6 PUFA ratio is low (n-6 PUFA-rich diet; 24-mo hearts) or attenuated when this ratio is relatively high (n-3 PUFA-rich diet). An important finding was that n-3 PUFA-rich diet in 24-mo rats could raise the n-3-to-n-6 PUFA ratio of Mito membrane phospholipid and prevent the significantly greater levels of Mito [Ca²⁺] and PDH_A that occurred with age in Ref or n-6 PUFA-rich dietary groups, particularly after I/R or positive inotropic stimulation by NE.

In our study there was a close correspondence between values of total Mito [Ca²⁺] and PDH_A/PDH_Total. These results provide valuable support for the model of dehydrogenase control by Mito free Ca²⁺ (18, 31). PDH interconversion responds to changes in Mito [Ca²⁺], and its sensitivity has been defined in experiments using indo 1- or fura 2-loaded mitochondria (14, 36). Although total Mito Ca²⁺ is measured in the present work, this is a good surrogate for Mito matrix free [Ca²⁺], because Mito matrix free [Ca²⁺] has been shown to be a linear function of total Mito [Ca²⁺] over this range of Mito loading (19). The range over which dietary modification affects total Mito [Ca²⁺] in these experiments is exactly the range over which PDH interconversion has been shown to be sensitive (21). Although it is still controversial whether Mito matrix free [Ca²⁺] follows changes in cytosol free Ca²⁺ on a beat-to-beat basis (27), it is clear that the larger systolic Ca²⁺ transients that occur in the cytosol when cardiac myocytes are stimulated with NE result in a significant increase in Mito matrix free [Ca²⁺] over the course of a few seconds (35). Furthermore, the blockade of excessive Mito Ca²⁺ uptake by ruthenium red before I/R in this study completely prevented the I/R-augmented activation of PDH. The effect of ruthenium red to abrogate a stimulated increase in PDH_A, which has been shown to result in decreased tissue ATP/(ADP × P_i) (51), confirms earlier work (17, 30). A caution in making this sort of relation between Mito total Ca²⁺ content and either PDH_A content or respiratory activity is that the former is measured in a preparation that is derived largely from subsarcolemmal mitochondria, whereas the latter parameters derive from both subsarcolemmal and intermyofibrillar mitochondria. These two populations of cardiac myocyte mitochondria have been shown to differ significantly in both Ca²⁺ transport and respiratory properties. There is no reason to believe that the relative size of these two populations is changed by our dietary manipulations, although it could be so.

Although PDH is emphasized in this report, findings of altered activation by Ca²⁺ are thought to relate equally to the Ca²⁺-sensitive dehydrogenases of the tricarboxylate cycle (31). This is important, because the majority of the reducing equivalents available for oxidation by the respiratory chain are generated in the tricarboxylate cycle, whether the substrate is carbohydrate or fat. Thus there is no suggestion that increased values of PDH_A change the partition of flux between glucose and fatty acid oxidation, but rather that substrate oxidation overall is activated. In this context, we used both glucose and octanoate as substrate for the perfusions to mimic the availability of both glucose and fatty acids in vivo, while avoiding problems with solubility of long-chain fatty acids. In addition, the presence of fatty acid depresses PDH_A levels due to the activation of PDH kinase by acetyl CoA and NADH (20, 53) and allows more scope for increase in PDH_A due to Ca²⁺. Inclusion of fatty acid in our perfusions does not permit direct comparison with the work of Kobayashi and Neely (28), in which they utilized glucose as the sole oxidizable substrate and did not observe elevated PDH_A after reperfusion.

Total Mito Ca²⁺ remained within the physiological range, regardless of the diet or age group, most likely because of the brevity of the ischemic period. However, the mitochondria from n-6 PUFA hearts clearly showed a greater net uptake of Ca²⁺ than those in the n-3 PUFA group. On this basis, we may predict that n-6 PUFA-rich hearts are likely to reach Ca²⁺ overload before n-3 PUFA hearts, following longer periods of ischemia, with the consequent opening of the MPT channel, decrease in proton-motive force, and cellular de-energization (10). Thus the high proportion of n-3 PUFA in cardiac Mito membranes after n-3 PUFA-rich diet may confer protection against I/R injury by minimizing the impact of Mito Ca²⁺ overload. We may also predict an impact of altered n-3 PUFA in Mito membranes on both necrotic and apoptotic cell death. By affecting the tendency of the mitochondria to overload with Ca²⁺ after I/R, the n-3-to-n-6 PUFA ratio may determine whether the MPT channel opens or not. This has been implicated in the process of apoptosis, which can be opposed by Ca²⁺ chelators, as well as by oxygen radical scavengers and the protein BCL-2 (48, 49). However, it remains controversial whether Mito membrane depolarization precedes MPT pore opening, or vice versa, in apoptosis.

In response to stimulation by NE, the hearts from animals fed n-6 PUFA-rich diet appear to have an advantage, in that they showed a more powerful activation of PDH and presumably also of the tricarboxylate cycle. However, the workload sustained in n-6 PUFA hearts with NE was not significantly different compared with n-3 PUFA hearts because left ventricular systolic pressure (Table 1), coronary flow rates, coronary perfusion pressure, and electrical stimulation rates were similar in both dietary groups regardless of age. This indicates that activation of PDH, and by inference activation of the Krebs cycle, was inefficiently high in n-6 PUFA hearts for an equivalent amount of mechanical work, compared with n-3 PUFA.

In the context of the lowered thermodynamic efficiency of the n-6 PUFA hearts, three possibilities can be envisaged from the present results. 1) The elevated Mito matrix free [Ca²⁺], which can be directly inferred from the elevated total Mito [Ca²⁺], will result in increased rates of Ca²⁺ cycling across the Mito membrane. Because the uniporter is driven by and the antiporter by pH (10), this will lead to dissipation of proton-motive force (the energy currency of the mitochondrion). Under physiological conditions, the energy usage of this futile cycle is usually regarded as minimal (2), but it may become substantial after I/R. 2) The greater degree of activation of PDH, and presumably of the tricarboxylate cycle, after n-6 PUFA-rich diet or with increased age would be expected to generate high proton-motive forces and high rates of proton leak across the inner Mito membrane. Such an effect is seen in studies where high, nonphysiological concentrations of substrate are used in experimental work with isolated mitochondria. Under such conditions, the conductance of the membrane to protons becomes "nonohmic" and proton leak increases rapidly with proton-motive force (38). It is indeed possible to generate very high Mito NADH-to-NAD⁺ ratios, and presumably proton- motive forces, by exposing heart mitochondria to buffered 1 µM free Ca²⁺, pyruvate, and 2-oxoglutarate as substrates (20, 37). Although we suspect that such a mechanism is a plausible source of inefficiency, it remains somewhat speculative in the absence of a direct measurement of Mito protonmotive force in situ in the perfused isolated beating heart. 3) We have presently shown that state 4 rates of respiration are raised in mitochondria from the n-6 PUFA hearts, even when Ca²⁺ ions are excluded from the incubation. This indicates an intrinsic increase in proton permeability of the membrane, which is likely a direct consequence of the changed lipid composition of these organelles following dietary treatment.

The degree to which the feeding of n-3 PUFA-rich diet prevents the age-associated decrease in the n-3-to-n-6 PUFA ratio that occurs with n-6 PUFA and Ref diets is striking. Our present finding of dietary lipid modulation of cardiac membrane ratio of n-3 to n-6 PUFA is supported by other studies that have used a similar dietary model with younger adult rats, monkeys, or dogs and have shown that following n-3 PUFA-rich diet, postischemic recovery of Mito energy metabolism and myocardial pump function is improved and that, in addition, there is a reduced incidence of reperfusion arrhythmias compared with n-6 PUFA (9, 23, 33, 34, 45, 46, 57, 58). The Ref diet is a standard rat chow (except that all long-carbon chain n-3 PUFAs are excluded), and thus is relevant to previous studies of heart mitochondria in aging. Nohl (39) reported that the ratio of unsaturated to saturated fatty acids of lipids from the inner membrane of rat heart mitochondria was significantly decreased and n-6 PUFA content rose in contrast to a significant decline in n-3 PUFA in senescent rats. A higher n-3-to-n-6 PUFA ratio (longer carbon chain vs. shorter carbon chain PUFA) implies a greater degree of membrane fluidity; thus the n-3 PUFA-rich diet may be expected to prevent a decrease in membrane fluidity otherwise seen in aging. Increased cardiac Mito membrane local lipid viscosity may alter lipid-protein relationships and thus the activity of intrinsic membrane proteins (25, 26). For example, it has been shown that following increased dietary incorporation of n-3 PUFA into cardiac sarcoplasmic reticulum membranes, there is a lower relative activity of Ca²⁺-Mg²⁺-ATPase and a 30% decrease in ATP-dependent Ca²⁺-uptake (50). Na⁺/Ca²⁺ exchange in cardiac sarcolemmal vesicles has been shown to be stimulated by acute exposure to PUFA (47). Single cardiac myocyte studies have demonstrated that the activities of dihydropyridine-sensitive L-type Ca²⁺ channels (44), transient outward K⁺ channels (8, 24, 52), and Na⁺ channels (54) are specifically attenuated following acute exposure to nanomolar concentrations of free n-3 PUFA. Although the mechanism of action underlying acute exposure may not be the same as that of dietary-induced incorporation of PUFA into membrane phospholipids, the ability to modulate the activity of such protein complexes, according to the type of PUFA content in related phospholipids, indicates that n-3 PUFA-rich membranes may minimize the extent of I/R-induced perturbation of intracellular ion homeostasis. The impact of dietary lipid modulation of the n-3-to-n-6 PUFA ratio is far reaching and multifarious because of effects on Ca²⁺-homeostasis, altered Ca²⁺-dependent phospholipase A₂ activity, the release of free PUFAs from membranes, and their subsequent interaction with cyclooxygenases, lipoxygenases, and other enzymes in the formation of prostanoids, leukotrienes, cytokines, fatty acid hydroperoxides, and 4-hydroxyalkenals (1, 42, 49). A detailed review of these important "cascade" effects and their feedback is, however, beyond the scope of this discussion.

We also measured, in hearts treated to n-3 PUFA-rich diet, a significant increase in the presence of cardiolipin (diphosphatidylglycerol), a phospholipid present only in the inner Mito membrane that is required for the activity of cytochrome-c oxidase (the terminal member of the respiratory electron-transfer chain), carnitine translocase, and adenine nucleotide translocase (4, 41). Our present finding of an n-3 PUFA-induced increase in cardiolipin content is in agreement with a previous study of fish oil-fed dogs (34). In contrast, Yamaoka et al. (55, 56), using an essential fatty acid-deficient diet supplemented with long-chain n-3 PUFAs, found no alteration in cardiolipin but instead a significant decrease in cytochrome-c oxidase activity and oxidative phosphorylation. However, in our study, we avoided essential fatty acid deficiency. Previous work has shown that the rates of transport of adenine nucleotides, pyruvate, and carnitine are decreased with increased age (15, 40, 43). Substrate transport and electron transport chain function are dependent on membrane viscosity (12). Thus the decreased cardiolipin content found with increased age (43) may alter the activity of inner Mito membrane proteins. Generally, the lipids of metabolically active intracellular membranes such as mitochondria are considerably unsaturated (7); thus specific proportions of long-chain highly unsaturated n-3 PUFAs may be essential in metabolic and ionic homeostasis. It is important to note that we observed no major difference in Mito membrane n-6 PUFAs or n-3 PUFAs between the n-6 PUFA and the Ref dietary groups. These two diets were devoid of all n-3 PUFAs with carbon chains longer than -linolenic acid (18:3); i.e., n-3 PUFAs of marine origin were excluded. However, Ref, a comparatively low-fat version of the n-6 PUFA-rich diet, had the same n-3-to-n-6 PUFA ratio of 0.2 as the n-6 PUFA diet. Thus these Mito differences following n-3 PUFA-rich diet that we presently report may be ascribed to increased incorporation of long-carbon chain n-3 PUFAs into Mito membrane phospholipids.

In conclusion, this study has shown the following. 1) Direct manipulation of the cardiac Mito membrane content ratio of n-3 to n-6 PUFA results in augmented activation of Ca²⁺-dependent PDH when the n-3-to-n-6 PUFA ratio is very low (n-6 PUFA-rich diet; 24-mo hearts) or attenuated activation when this ratio is relatively high (n-3 PUFA-rich diet). 2) Aged Ref or n-6 PUFA-rich dietary groups had significantly elevated Mito [Ca²⁺] and PDH_A (particularly after I/R or positive inotropic stimulation by NE). 3) Notably, this could be prevented in 24-mo rats by raising the n-3-to-n-6 PUFA ratio with n-3 PUFA-rich diet. 4) Increased state 4 rates of respiration (coupled, minus ADP), in the absence of Ca²⁺, contributed to a loss of efficiency intrinsic to the Mito membranes from n-6 relative to n-3 PUFA hearts. Age-linked increases in state 4 respiration observed in n-6 PUFA hearts were not apparent in n-3 PUFA-rich hearts. 5) Mito membrane phosphatidylcholine and n-6 PUFAs were increased in Ref hearts from aged animals, whereas cardiolipin and n-3 PUFA content were decreased. These age-associated effects were augmented by n-6 PUFA and prevented by n-3 PUFA-rich diet. Thus enhanced PDH activation with aging or increased n-6 PUFA content is considered to be mediated through increased net Mito Ca²⁺ uptake. The relative contributions to this effect of changes in cytosol Ca²⁺ and the intrinsic alterations in Mito membrane phospholipid composition documented here remain to be elucidated. The key role of membrane lipids and constituent fatty acids in defining and modulating membrane permeability, membrane protein function, and intercompartment signaling, especially in aging, where specific PUFA accumulation and deficits occur, requires extensive study.