INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

THE VITAL IMPORTANCE of the n-3 and n-6 polyunsaturated fatty acids (PUFAs), so called because of the location of the first double bond (at carbon no. 3 or 6, respectively, from the methyl end of the carbon chain), is becoming widely recognized in human physiology. Many studies suggest that n-3 fatty acids have beneficial effects on human health. In contrast, a diet enriched in n-6 fatty acids provokes atherosclerosis, carcinogenesis, and heart disease (see Ref. 19 for review). Although the underlying mechanisms of these dramatic differences are presently unclear, a different sensitivity of ionic channels to n-3 and n-6 types of fatty acids might contribute to this phenomenon.

Voltage-gated K⁺ channels control many aspects of cardiac performance in health and disease. In particular, these channels determine ventricular repolarization and modulate antiarrhythmic drug effects. It is known that the properties of K⁺ channels are influenced by their lipid environment (23, 25), and recently a large body of evidence has evolved indicating that changes in cardiac K⁺-channel properties are induced by fatty acids (8, 13, 17, 18). However, these studies did not differentiate the effects of n-3 versus n-6 PUFAs on K⁺ currents.

The purpose of the present study was to compare the effects of the n-3 and n-6 PUFAs on the transient K⁺ outward current (I_to) and other K⁺ currents. It has been shown that with the notable exception of guinea pigs, the I_to plays an important role in the cardiac action potential repolarization of several mammalian tissues, including human atrium (26) and ventricular myocytes from humans (30), dogs (20), rabbits (11), ferrets (3), and rats (1). In the present study to characterize the effects of PUFAs, we used rat ventricular myocytes in which K⁺ currents were recorded by the whole cell patch-clamp technique.

	METHODS

Top Abstract Introduction Methods Results Discussion References

All the experiments were conducted at room temperature (21-23°C) on cardiac myocytes enzymatically isolated from 2- to 3-mo-old Sprague-Dawley rats as previously described (4). A whole cell voltage clamp was employed using an Axopatch 1D (, or gain, = 0.1; Axon Instruments). The recording pipette resistance ranged between 1.5 and 3 M. To isolate K⁺ currents, Na⁺ and Ca²⁺ currents were blocked by means of replacing external Na⁺ with N-methyl-D-glucamine and by adding 0.2 mM CdCl₂. The bath solution contained (in mM) 137 N-methyl-D-glucamine, 3.7 KCl, 1.2 KH₂PO₄, 1 CaCl₂, 1.2 MgSO₄, 15 glucose, and 20 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), with pH adjusted to 7.4 using KOH. The solution in the recording pipette contained (in mM) 110 KCl, 10 ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid, 10 HEPES, 3 MgATP, 0.1 Na₂GTP, 5 glucose, and 10 NaCl, with pH adjusted to 7.2 using KOH. When the action potentials of myocytes were recorded in current-clamp mode using 1.5× threshold pulses with a duration of 5 ms, the bath solution contained 137 mM NaCl (instead of N-methyl-D-glucamine) and no Cd²⁺ was used.

The peak I_to transient amplitude was measured as the difference between peak current and the steady current at the end of a 200-ms voltage step. The inactivation time constant for I_to was quantified by fitting with a single exponential function. To demonstrate the currents affected by the drug, the difference currents were obtained by subtraction of currents before the drug application from those after the drug. Note that in traces of difference currents upward deflections represent an increase in net outward membrane current and downward deflections represent a decrease in outward current with application of the drug.

In some studies, 4-aminopyridine (4-AP), indomethacin, and eicosatetraynoic acid (ETYA) were also used. Fatty acids and ETYA were diluted with nitrogen-saturated ethanol, and the stock solution was added to the superfusion solution to achieve the final desired concentration immediately before each experiment. These solutions were used within 60 min of preparation to minimize oxidation of fatty acids and were protected from light. 4-AP was prepared as a 5 M stock solution with pH adjusted to 7.4 with HCl. Indomethacin was made fresh at 0.1 M in dimethyl sulfoxide.

All averaged and normalized results are presented as means ± SE. Statistical significance of differences in the calculated mean values was evaluated using Student's t-test, and P values are provided (in parentheses) in the text. The level of statistical significance was considered at P < 0.05.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Under control conditions in the presence of Ca²⁺- and Na⁺-channel blockade, depolarization steps of 200 ms from a holding potential of 70 mV evoked an outward K⁺ current. This outward current rose to a peak and then slowly decayed (Fig. 1, left). The outward K⁺ current is thought to consist of two components that can be separated both kinetically and pharmacologically (1, 14). The rapidly activating (and inactivating) transient outward current, I_to, is defined by its sensitivity to 4-AP. The second current component remaining after I_to inactivation was named as the sustained depolarization-induced current (I_sus) because the delayed, time-dependent current (I_K) is either absent or insignificant in rat ventricular myocytes (5, 27). I_sus activates more slowly than I_to and is not sensitive to 4-AP. Original tracings of K⁺ currents before and after addition of 4-AP (5 mM) are shown in Fig. 1, right. 4-AP clearly abolishes I_to, whereas no significant changes in the steady current at the end of the test step can be detected. These characteristics of the outward current indicate that the current being studied under these conditions is I_to (1, 14). We have thus taken the difference between a peak I_to and the current at the end of the 200-ms voltage step (I_sus) as the amplitude of I_to.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Characterization of transient outward current (Ito) utilizing 4-aminopyridine (4-AP; 5 mM). Ito was elicited by voltage pulses of 200-ms duration from 70 to +60 mV in 20-mV steps. Holding potential (HP) was 70 mV. Left, control recordings obtained in Na+-free external solution. Right, 4-AP abolished Ito at every potential. Current at end of pulse (Isus) was not changed by 4-AP.

Effects of n-3 PUFAs on I_to current. The time course of docosahexaenoic (C22:5n-3) acid (DHA) effect on outward current amplitude is shown in Fig. 2. Gradually increasing downward deflections of difference currents (decreased net outward current) with time after DHA addition appear to indicate a reduction only of I_to, because the difference currents have kinetic properties similar to those of I_to. These effects of DHA were not observed when 4-AP (5 mM) was present in the bath (not shown).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Time course of docosahexaenoic acid (DHA; 5 µM) effect on outward current elicited by voltage steps from 70 to +60 mV (top traces). Bottom traces: difference (Diff) currents obtained by subtraction of currents before application of DHA from those after 5 µM DHA; increased net outward current is an upward deflection and decreased net outward current is a downward deflection. Time (in seconds) after DHA addition is marked above traces.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Effects of 5 µM DHA on Ito. A: Ito was recorded during 200-ms depolarization steps to +60 mV from 70 mV before (control) and 3 min after addition of 5 µM DHA; original traces (bold lines) with superimposed monoexponential fits. B: time course of DHA effect on inactivation time constant () of Ito and a reversal of this effect after addition of 0.1% bovine serum albumin (BSA).

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Normalized current-voltage relations for Ito amplitude before and after addition of 50 µM DHA. Outward currents were evoked by depolarizations to potentials between 30 and +60 mV from HP of 70 mV. Difference between peak current and current at end of 200-ms step was taken as amplitude of Ito. For each cell, amplitudes of Ito were measured at each test potential and normalized to its control amplitude evoked at +30 mV. Mean (±SE) normalized values (n = 5) are plotted.

Concentration-dependent effects of n-3 PUFAs on I_sus. The effects of PUFAs on K⁺ currents were concentration dependent between 5 and 50 µM (Fig. 5). Bath application of 5 and 10 µM EPA caused a decrease in I_to without affecting other outward K⁺ currents. At the higher concentrations of EPA tested (20 and 50 µM, n = 4) the inhibition of I_sus (measured at the end of the 300-ms voltage step) was significant and concentration dependent (16 ± 4 and 56 ± 11%, respectively). Application of 50 µM DHA suppressed I_sus by 32 ± 9% (data not shown).

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Effects of 10, 20, and 50 µM eicosapentaenoic acid (EPA) on K+ currents. Superimposition of K+ current traces in response to voltage steps from 70 to +50 mV in control (C) and 3 (1) and 10 (2) min after EPA addition is shown. At low concentrations, EPA inhibits only Ito; higher concentrations further inhibit Ito and also inhibit Isus.

Effects of n-3 PUFAs on I_K1. Although the present study focused mainly on the effects of PUFA on the outward K⁺ currents, we also investigated PUFA effects on inward rectifier K⁺ current (I_K1). Hyperpolarization from 70 to 120 mV produced a time-independent inward current, I_K1. Figure 6, left, shows I_K1 tracings in control and 10 min after the addition of 50 µM EPA. I_K1 amplitude was measured at the end of the 200-ms voltage step. Superfusion of cells (n = 5) for up to 12 min did not have any significant effect on I_K1. EPA failed to induce any significant changes in I_K1 at all voltage steps studied (120 to 80 mV). Figure 6, right, demonstrates the recordings of outward K⁺ currents in control and 10 min after application of 50 µM EPA in the same cell. An absence of effects on I_K1 was also observed for DHA (data not shown).

View larger version (10K): [in this window] [in a new window]   Fig. 6.   Representative effects of 50 µM EPA on inward rectifier (IK1; left) and outward (right) K+ currents. Superimposition of K+ current traces in response to voltage steps in control (C) and 10 min after EPA addition is shown. EPA inhibits Ito and Isus without affecting IK1.

Effects of n-3 PUFAs on action potential duration. Figure 7 illustrates a concentration-dependent effect of EPA on the action potential in rat ventricular myocytes. At low concentrations of EPA (5 and 10 µM), the action potential duration was gradually increased without change in amplitude. At a higher concentration (20 µM), a lengthening in the action potential duration was accompanied by a significant reduction in amplitude and decrease of the maximal rate of depolarization. In some experiments, the high concentration of EPA elicited a block of excitation. The decrease in the maximal rate of depolarization is in accordance with blocking effects of PUFAs on Na⁺ channels observed earlier (15, 31).

View larger version (10K): [in this window] [in a new window]   Fig. 7.   Effects of 5, 10, and 20 µM EPA on action potential in rat cardiomyocytes. Superimposition of action potential traces in control (C); 3, 5, and 10 min after EPA addition; and after washout with 0.1% BSA (bold line) is shown. At low concentrations, action potential duration (APD) was increased; higher concentrations depressed plateau level and APD.

Effects of n-6 PUFA on K⁺ currents and action potential. The effects of arachidonic (C20:4n-6) acid (AA), which belongs to the n-6 class of PUFA, are shown in Fig. 8. Like n-3 PUFA, AA had no effect on I_K1 (Fig. 8A, left). In contrast to n-3 PUFA, AA markedly increased the steady-state current, I_sus (Fig. 8A, right). Values of the current at the end of the 200-ms step were normalized to the value of the current in control in response to a voltage step to +30 mV for each cell (n = 5) and are plotted as a function of clamp voltage in Fig. 8B. AA (50 µM) increased I_sus by ~80% within 10 min. There was an 80% increase in I_sus regardless of the test potential (30 to +60 mV).

View larger version (9K): [in this window] [in a new window]   View larger version (13K): [in this window] [in a new window]   Fig. 8.   Effects of 50 µM arachidonic acid (AA) on K+ currents. A: superimposition of K+ current traces in response to voltage steps from 70 to 120 mV (left) and from 70 to +60 mV (right) in control and 10 min after AA addition. B: current-voltage relations for steady level of K+ current (Isus) in control and in presence of AA. HP, 70 mV. Values of currents at end of 200-ms step were normalized to value of current in control in response to voltage step to +30 mV for each cell (n = 5).

View larger version (9K): [in this window] [in a new window]   View larger version (9K): [in this window] [in a new window]   Fig. 9.   Time course of AA effect on outward current elicited by voltage steps from 70 to +60 mV (top traces). Bottom traces: Diff currents obtained by subtraction of currents before application of AA (arrows, B) from those after AA; increased net outward current is an upward deflection and decreased net outward current is a downward deflection. Time (in seconds) after AA addition is marked above traces. A: AA, 5 µM. B: AA, 50 µM.

View larger version (18K): [in this window] [in a new window]   Fig. 10.   Inhibitory effect of 4-AP (5 mM) on noninactivating outward K+ current increased earlier by AA (5 µM). K+ current traces in response to voltage steps from 70 to +60 mV are superimposed.

View larger version (8K): [in this window] [in a new window]   Fig. 11.   Effects of 10 µM eicosatetraynoic acid (ETYA), a nonmetabolizable AA analog, on K+ currents. Superimposition of K+ current traces in response to voltage steps from 70 to 120 mV (left) and from 70 to +60 mV (right) in control and 10 min after ETYA addition is shown.

View larger version (11K): [in this window] [in a new window]   Fig. 12.   Effects of 10 µM linoleic acid (LLA) on outward K+ currents (left) and action potential (right).

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Apkon and Nerbonne (1) reported that there are two K⁺ current components that contribute to the total depolarization-activated outward currents in adult rat ventricular myocytes: 1) a 4-AP-sensitive current that activates and inactivates rapidly on depolarization (I_to) and 2) a TEA-sensitive component that, after a delay, activates slowly to a steady level (I_sus). The results presented here in adult rat ventricular myocytes reveal that at low concentrations (5 and 10 µM), I_to is suppressed by both n-3 and n-6 PUFAs and that this depression is mediated by both a decrease in I_to amplitude and an apparent acceleration of I_to decay. This suppression cannot be attributed to the action of cyclooxygenase metabolites, because a cyclooxygenase inhibitor, indomethacin, did not prevent this effect. We also observed similar results in experiments with cardiomyocytes isolated from dog ventricle, where 10 µM DHA completely abolished the I_to current (Bogdanov, unpublished observation). The application of PUFAs (5-50 µM) to cells pretreated with 4-AP (I_to blocker) had no effect on K⁺ currents. This further supports the interpretation that PUFAs at micromolar concentrations affect I_to current in rat ventricular myocytes.

At high concentrations (20 and 50 µM) all PUFAs except AA inhibited the slowly activating K⁺ current, I_sus, in whole cell recordings of rat ventricular cells. This result is in accordance with a report by Honore et al. (13) that DHA (30 µM) blocks the cardiac delayed rectifier K⁺ channel (Kv1.5) and whole cell K⁺ currents in neonatal mouse and rat cardiomyocytes. To determine whether fatty acid blockade of the Kv1.5 channel occurred from the intracellular or the extracellular side of the membrane, Honore et al. (13) filled pipettes with the PUFAs and found no effect of fatty acids applied from inside. However, this null effect in their experiments might be the result of limited diffusion from the pipette to the cell interior because of sticking of the PUFA to the glass and high light sensitivity of the PUFAs. In contrast to our results, Honore et al. (13) also observed an inhibition of I_sus by 30 µM AA. The reason for this discrepancy is not clear. Molecular studies (9) reveal at least five subunit messages in adult rat heart (Kv1.2, Kv1.4, Kv1.5, Kv2.1, and Kv4.2). However, the relationship between the expression of these subunits and the functional properties of K⁺ channels in vivo is unknown, although it has been suggested (9) that the Kv4.2 subunit contributes to the I_to in adult rat ventricular myocytes.

However, the significance of our data obtained after direct exposure of free PUFAs to cells, as in the present study, is unclear because PUFAs incorporated into phospholipids of the membrane bilayer may have different effects on ionic channels (10, 12) than do free fatty acids. Also, concentrations of free fatty acids >20 µM can contain micelles and can cause various nonspecific effects, such as detergent effects, on ion channels (32). Therefore, the inhibition of I_sus observed after application of 20 and 50 µM PUFA (see Figs. 5 and 6), in principle, could be caused by the detergent effects of the PUFAs.

The present study demonstrates similarities and also significant differences in the effects of AA and other PUFAs tested on outward K⁺ currents. Like all PUFAs, AA initially depressed I_to. However, after 3 min, micromolar concentrations of AA began to increase the steady-state outward K⁺ current, I_sus. The results suggest that the AA-induced noninactivating K⁺ current seems to be I_to with no or very slow inactivation kinetics. In adult rat ventricular cells, the AA-activated K⁺ current, I_K,AA, was found by Kim and Duff (18). They proposed that AA in micromolar concentrations can cause activation of I_K,AA in heart cells, altering the gating properties of an existing K⁺ channel. Our data suggest that the K⁺ channel affected by AA could be the I_to channel. Pronounced voltage dependence of the AA-induced noninactivating K⁺ current (see Fig. 8B) can be considered as evidence that the current is not the voltage-independent K⁺ current found by Kim and Clapham (17) in neonatal atrial cells.

Damron et al. (8) also observed an inhibition of I_to in rat cardiomyocytes after application of 50 µM AA. However, they did not report a delayed increase in I_sus as described in the present study. This apparent discrepancy could be explained by differences in voltage clamp technique (perforated patch vs. classic whole cell method in our experiments). The effect of AA to increase I_sus is likely caused by an action of AA metabolites, because the nonmetabolizable analog of AA, ETYA (see Fig. 11), inhibited both I_to and I_sus. A faster rate of biosynthesis of eicosanoids from the AA and their stronger activity than those formed from other PUFAs (19) could be possible causes of the AA effect found in present study. It is also possible that the conditions of Damron et al. (8) did not permit the metabolism of AA or that in our whole cell experiments soluble factors, such as protein kinases A and C, that regulate the cardiac K⁺ channel (29) were washed out.

Although the function of I_to is not fully understood, in most mammalian tissues including human, dog, rabbit, ferret, and rat ventricle (1, 3, 11, 20, 30) it is one of the currents that underlie repolarization of the action potential. In addition, I_to, because of its nonuniform distribution across the myocardium, can play an important role contributing to the heterogeneity of myocardial repolarization. Previous data had shown that I_to is significantly greater in myocytes from epicardial regions than in endocardial myocytes (20, 30). This electrical heterogeneity is intensified during ischemia and may facilitate the development of reentrant arrhythmias in epicardium (20). The accumulation of AA occurs early during myocardial ischemia and has been estimated to be between 20 and 40 µM in the dog after 1 h of ischemia (6). Thus the concentrations of AA used in this study (5-50 µM) may be similar to those observed in ischemic myocardial tissue. It has been suggested that selective blockade of I_to (e.g., by 4-AP or all PUFAs except AA) may be a useful antiarrhythmic intervention. Experimental results obtained by Tsuchihashi and Curtis (28) also support this hypothesis.

A beneficial effect of long-chain n-3 PUFAs from fish oil on cardiac performance has been well recognized (see Ref. 19 for review). Recent studies by McLennan et al. (21) demonstrate an important role for fish oil PUFAs in the prevention of fatal ventricular arrhythmias. Billman et al. (2) also observed that infusion of -3 PUFA in dogs could prevent ventricular fibrillations occurring in response to acute occlusion of a coronary artery. The present results in rat and those in dog (Bogdanov, unpublished observation) ventricular cells support the hypothesis that the antiarrhythmic effects of the fish oil PUFAs in situ may be related in part to the inhibition of I_to and I_sus.

Kang et al. (15) found that the n-3 PUFAs (DHA and EPA) decrease the action potential duration of neonatal rat cardiac myocytes. This result conflicts with data obtained by us in experiments on adult rat cardiomyocytes (see Fig. 7). However, Kilborn and Fedida (16) reported that the contribution of the transient versus sustained currents to the total outward K⁺ current in rat ventricle was age dependent. They observed a fourfold increase in I_to from neonatal to adult age in rat cells, indicating that I_to appears to play a major role in repolarization of adult but not neonatal ventricular action potentials. No differences were observed in the kinetic properties of I_to (16). Thus the shortening of the action potential in neonatal rat cardiomyocytes caused by 10 µM EPA (15) could be explained by the very small amplitude of the I_to in these cells and by an inhibitory effect of the PUFAs on I_Ca (Bogdanov, unpublished observation).

Large-conductance channels predominantly permeable to Cl ions are present in cardiac plasma membranes of the newborn rat heart cells (7). Therefore, in principle, a part of the effect of PUFAs on outward currents observed could be caused by an action on Cl current. However, a reversal potential (E_rev) for a putative Cl current in our experiments in adult rat cells should be positive (as a result of 110 mM Cl inside vs. 6 mM outside), about +75 mV. In contrast, as indicated by Figs. 4 and 8B, the current-voltage relationships show a negative E_rev for the currents affected by PUFAs, suggesting that PUFAs affected K⁺ currents rather than Cl currents.

As shown in Fig. 7, 20 µM EPA blocked excitation of the cell and this result is in agreement with data by Xiao et al. (31), in which the blocking effect of PUFAs on Na⁺ channels was demonstrated in neonatal rat ventricular cells. A prior study (24) failed to observe a significant effect of 5 µM AA or DHA on Ca²⁺ current (I_Ca) amplitude in adult rat ventricular cells. However, other recent studies (Y.-F. Xiao, J. X. Kang, J. P. Morgan, and A. Leaf, personal communication) have shown effects of low concentrations of PUFAs on both I_Ca and I_sus. Further study is required to determine why the potency of the PUFA effect on specific ion channels differs among these studies.

In summary, we conclude that at low concentrations (5 and 10 µM) both n-3 and n-6 PUFAs inhibit I_to current in adult rat cardiac myocytes and this lengthens the action potential. Higher concentrations of PUFAs (20 and 50 µM) further inhibited I_to and also inhibited I_sus and I_Ca. However, our results indicate that noninactivating outward K⁺ current, I_sus, is also augmented by AA metabolites. Thus the overall effect of AA on action potential repolarization is determined by the overall net effect, which likely varies with the cellular metabolic state. It is presently unknown whether results obtained in experiments with direct application of free fatty acids can be used to study the beneficial effects of diet enriched with PUFAs.