Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The role of fatty acids in signaling processes is becoming increasingly important. Various fatty acids have been shown to alter several ion currents, including Na⁺ (Xiao et al., 1995; Kang and Leaf, 1996), Ca²⁺ (Huang et al., 1992; Hashimoto et al., 1999), K⁺ (Bogdanov et al., 1998; Crumb et al., 1999), and Cl (Ordway et al., 1991). Arachidonic acid (C20:4) has been widely studied for the actions of the parent compound as well as the extensive metabolic pathways, including the prostaglandins and leukotrienes. Although less widely studied, linoleic acid (C18:2) is a major component of cell membrane phospholipids and is subject to some of the same metabolic pathways. A linoleic acid metabolite (LAM) of interest is 9,10-epoxy-12-octadecenoic acid (EOA), also known as leukotoxin. EOA is formed by lipid autoxidation in the lungs (Sevanian et al., 1979) and by spontaneous reaction of oxygen radicals with linoleic acid in neutrophil membranes (Hayakawa et al., 1996) (Fig. 1). Although undetectable in normal patients (Hayakawa et al., 1990), EOA has been associated with acute respiratory distress syndrome in burn patients where it can reach plasma concentrations up to 300 µM (mean peak plasma concentration of 99 ± 25 µM) (Kosaka et al., 1994). EOA has also been shown to cause cardiac arrest in dogs (Fukushima et al., 1988) and relax pulmonary artery smooth muscle (Takahashi et al., 1992). In pig experimentally induced cardiac ischemia caused EOA to increase from undetectable levels to on the order of 1 µg/g tissue (Dudda et al., 1996).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Metabolism of linoleic acid. The structures of the tested compounds and a pathway by which they may be produced endogenously are illustrated.

Recent studies have questioned whether EOA is the toxic agent involved, because it may be further metabolized by epoxide hydrolases to form 9,10-dihydroxy-12-octadecenoic acid (DHOA) (Moghaddam et al., 1997). Both EOA and DHOA have been reported to be toxic in humans and various animal preparations (Fukushima et al., 1988; Kosaka et al., 1994). During myocardial ischemia in pig hydroxy-metabolites of linoleic acid were found to increase from 1 to 3 µg/g tissue up to 4 to 22 µg/g tissue, depending on the metabolite formed (Dudda et al., 1996). Studies have suggested that EOA must be metabolized to DHOA before producing toxic effects (Moghaddam et al., 1997; Greene et al., 2000). In accordance with this evidence, in isolated rat cardiac myocytes 9,10-dihydroxy-12-octadecenoic methyl ester (DHOM) (a methyl ester of DHOA) has been shown to block both Na⁺ current (I_Na) and transient outward K⁺ current (I_to), whereas 9,10-epoxy-12-octadecenoic methyl ester (EOM) (a methyl ester of EOA) had no significant effect (Stimers et al., 1999).

This study further investigated the effects of EOA, DHOA, and their corresponding methyl esters (EOM and DHOM) on sodium current in adult rat ventricular myocytes. This study was designed to investigate the role of each of these four LAM in blocking Na⁺ channels and identify the possible mechanism of action. A kinetic analysis of the actions of each compound on I_Na in adult rat cardiac myocytes using whole-cell patch-clamp techniques was performed. Although it has been suggested that these compounds, being lipid soluble, may have actions on cell membranes that indirectly alter I_Na, we show herein for the first time that each compound has direct actions on Na⁺ channels that suggest selective interaction with the channels. A preliminary report of this work was published previously (Harrell and Stimers, 2002).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. EOA, EOM, DHOA, and DHOM were all synthesized and purified from linoleic acid as described previously (Moran et al., 2000). Ethanol stock solutions were prepared before use and stored at 40°C. Unless otherwise noted, all other chemicals were purchased from either Sigma-Aldrich (St. Louis, MO) or Fisher Scientific (Fair Lawn, NJ).

Solutions. The techniques used to isolate and measure I_Na were similar to those described previously (Stimers et al., 1999). All intracellular (pipette) solutions contained 120 mM DL-aspartic acid, 120 mM CsOH, 10 mM CsCl, 10 mM NaCl, 1 mM CaCl₂, 5 mM MgCl₂, 11 mM EGTA, and 10 mM HEPES, pH 7.2. A control extracellular solution containing 145 mM NaCl, 1 mM MgCl₂, 5.1 mM HEPES, 4.9 mM Na HEPES, 5.55 mM dextrose, 2 mM BaCl₂, 0.2 mM CdCl₂, 5 mM KCl, and 11.69 mM sucrose, pH 7.4, was used to achieve a giga-ohm seal and whole-cell configuration. Once whole-cell configuration of the patch clamp was achieved, the external solution was changed to 30 Na⁺ that contained 30 mM NaCl, 115 mM N-methyl-d-glucamine, 1 mM MgCl₂, 0.2 mM CdCl₂, 2 mM BaCl₂, 10 mM HEPES, 5.55 mM dextrose, and 11.69 mM sucrose, pH 7.4. These solutions block all major ion currents other than I_Na and reduce the magnitude of I_Na so that good voltage control could be maintained in each cell. Linoleic acid metabolites tested were dissolved in ethanol and added to the 30 Na⁺ extracellular solutions. Ethanol concentration did not exceed 0.5%, which had no effect on I_Na.

Cell Isolation and Culture. Cardiac myocytes from adult male Sprague-Dawley rats were isolated by Langendorff perfusion of 0.05% collagenase (type II; Worthington Biochemicals, Freehold, NJ), as described previously (Dobretsov et al., 1998). Myocytes were then plated in a KB buffer that contained the following: 85 mM KCl, 5 mM HEPES, 5 mM Tris, 25 mM KH₂PO₄, 0.5 mM EGTA, 5 mM MgSO₄, 5 mM K-pyruvate, 5 mM Na₂-creatine phosphate, 20 mM taurine, 0.1 mM CaCl₂, and 20 mM dextrose, pH 7.4. After allowing cells to attach to the culture dishes, KB buffer was replaced with culture media of M199 (Invitrogen, Carlsbad, CA) with 4% fetal bovine serum and 1% penicillin-streptomycin. Cells were then incubated at 37°C under 95% air, 5% CO₂ for 1 to 4 days until use. As shown in Fig. 2 there was no significant change in the magnitude or voltage dependence of peak I_Na over this time in culture.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Effect of time in culture on voltage dependence of peak INa. The data were collected as described in Fig. 3. Peak INa measured for voltage pulses from a holding potential of 100 mV to potentials between 80 and +60 mV in 10-mV increments. Data were collected from two cultures that were examined on all 4 days. There are no differences between peak currents on any day (n = 7, 7, 10, and 11 on days 1-4, respectively).

Patch Clamp. Individual rat myocytes were patch clamped in the whole-cell configuration with an integrating patch clamp (3900A; Dagan, Minneapolis, MN) using standard techniques (Dobretsov and Stimers, 1996; Stimers and Dobretsov, 1998; Dobretsov et al., 1998). Culture dishes were placed on the stage of an inverted microscope (Diaphot; Nikon, Melville, NY) and perfused with solutions at a rate of 5 ml/min. Patch pipettes were produced from borosilicate glass using a Flaming-Brown P77 puller (Sutter Instrument Co., Novato, CA). The pipettes were then fire polished to produce pipettes with a final resistance of 1 to 3 M. After forming a seal, the holding potential was set at 100 mV. Upon formation of a giga-ohm seal, the electrode capacitance was compensated. A brief electrical pulse was used to achieve the whole-cell configuration. Cell capacitance, input resistance, and series resistance were measured in every cell by applying a 10-mV pulse. Cell capacitance and series resistance compensation were used before a voltage-clamp protocol to improve the quality of the voltage clamp. Pulses were generated and collected using pClamp software and a Digidata 1200 computer acquisition system (Axon Instruments, Union City, CA). In all pulse protocols used in this study, a P/4 protocol was used to subtract capacity transients and leak. All experiments were performed at room temperature (20-22°C).

Data Analysis. Data were collected from only one cell in each culture dish to ensure there was no residual contamination from previous drug applications. Each cell is considered as an independent measure (n) because control data were recorded from each cell before drug application, so each cell serves as its own control. To ensure we were not investigating a unique animal, all experiments were repeated on cells from at least two isolations. It is our experience for inbred animals that have not been treated in any way that the variability between cells from a single heart is greater than the variability between animals. For example, in four animals used in this study, the standard deviation for measurements of peak I_Na at 30 mV in each animal were 9.8, 10.9, 10.7, and 25.6 pA/pF (n = 25, 9, 8, and 6, respectively); however, the standard deviation of all four animals was only 8.8 pA/pF (n = 4). This leads us to conclude that it is appropriate for these studies to consider each cell as an independent measurement.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Activation. To examine the effect of LAMs on I_Na activation, I_Na was measured using whole-cell patch-clamp techniques. Membrane potential was held at 100 mV and pulsed to potentials between 80 and 60 mV in 10-mV increments for 25 ms (Fig. 3). Typical records of Na⁺ currents from single cells are shown after exposure to control solution and 50 µM DHOA or DHOM (Fig. 3). In this cell, DHOA had no effect on I_Na, whereas DHOM blocked about 50% of the inward current. Other experiments applied 50 µM EOA or EOM (data not shown). Chemicals were applied for 3 to 5 min before data were collected. Control currents displayed typical activation and inactivation kinetics and voltage dependence for I_Na. Peak current measurements were made in each cell, normalized for cell capacitance, averaged between cells with like treatments, and plotted in Fig. 4. Figure 4A shows that on average in these experiments DHOA had no significant effect on I_Na at any of the tested potentials; however, 50 µM DHOM significantly inhibited I_Na at potentials between 40 and +30 mV (P < 0.05). In contrast, both 50 µM EOA and EOM significantly inhibited I_Na between 30 and +30 mV (P < 0.05; data not shown for clarity). Figure 4B shows the percentage of inhibition caused by each compound with respect to control measured at 30 mV, the peak of the current-voltage relation. DHOM showed the greatest inhibition of all tested compounds. When DHOM was included in the patch pipette solution, there was no inhibition of I_Na (data not shown). There was no apparent shift in the voltage dependence for I_Na activation. Such a shift would indicate that LAMs affected the potentials necessary to activate the sodium channels in rat ventricular myocytes. However, no such shift occurred; therefore, LAMs have no effect on steady-state activation.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Voltage dependence of INa. Top, protocol used to measure INa in cardiac myocytes. Membrane potential was held at 100 mV and pulsed to potentials between 80 and +60 mV in 10-mV increments, for 20 ms. Bottom (three panels), resulting current recordings from a single myocyte under control conditions and subsequently exposed to 50 µM DHOA and 50 µM DHOM. For clarity pulses are shown in steps of 20 mV; however, data were collected every 10 mV. Cell capacitance was 94 pF. Time scale indicates from the start of data acquisition.

View larger version (32K): [in this window] [in a new window]   Fig. 4.   Voltage dependence of INa and effect of linoleic acid metabolites. Peak inward current measured at each potential from data in Fig. 3 and other cells were averaged and plotted versus membrane potential. A, cells that were exposed to 50 µM DHOA and 50 µM DHOM. Symbols represent mean ± S.E. Data points are connected by straight lines. B, effect of each chemical on peak INa measured at 30 mV, the peak of the current-voltage relation. Number of cells used was six to seven for each chemical. DHOM, EOA, and EOM significantly inhibited peak INa between 30 and +30 mV. DHOA has no significant effect on INa in these cells.

Steady-State Inactivation. Steady-state inactivation was measured to examine the effect of EOA, EOM, DHOA, and DHOM. Myocytes were voltage clamped at 100 mV and a two-pulse protocol was applied in control and in the presence of 50 µM of each compound (Fig. 5). Similar to the activation protocol used above, a prepulse was applied from the holding potential to potentials between 80 and 10 mV in 5-mV increments for 20 ms to induce Na⁺ channel inactivation. After a 2.5-ms repolarization to 100 mV, a second depolarizing pulse to 10 mV was applied for 10 ms to measure the amount of I_Na that was available for activation. This protocol is designed to measure the number of Na⁺ channels not inactivated by the prepulse. Results from a typical experiment are shown in Fig. 5 where the protocol was applied in control conditions and after exposure to 50 µM EOA. In this experiment about 40% of I_Na was blocked by EOA; however, inactivation can still be easily measured by this protocol. The peaks of the I_Na elicited by the second pulse were measured, fit to a Boltzmann equation, and the minimum and maximum fit parameters were used to normalize the data that were plotted versus the prepulse potential (Fig. 6). The curves in Fig. 6 represent the best fit of a Boltzmann equation to each data set. Fifty micromolar DHOM, EOA, and EOM all significantly hyperpolarized the steady-state inactivation curve. EOM also significantly increased the slope of the steady-state inactivation curve (Fig. 6B). Parameter values are given in Table 1.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Steady-state fast inactivation of INa. Top, double pulse protocol used to measure the voltage dependence of inactivation. Membrane potential was held at 100 mV and pulsed to potentials between 80 and 60 mV in 5-mV increments for 20 ms. After a 2-ms return to 100 ms a test pulse to 10 mV was applied for 10 ms. Bottom, current records from a single cell in control conditions and exposed to 50 µM EOA. During the prepulse INa activates and inactivates in a voltage-dependent manner. The subsequent test pulse elicits INa that is still available for activation. Peak inward current during the test pulse is measured and normalized to the maximum current. Cell capacitance was 94 pF.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Voltage dependence of inactivation. A, mean ± S.E. values obtained from experiments like that shown in Fig. 5. Data are normalized to the maximum current obtained in each condition. This shows that DHOM, EOA, and EOM all cause a significant shift of the steady-state inactivation curve to hyperpolarized potentials by about 6 mV. The data points for DHOM, EOA, and EOM significantly differ from control values between 60 and 40 mV (P < 0.05). B, isolates the same data for control and EOM to more clearly show the significant increase in slope caused by EOM. Lines are best-fit Boltzmann relations to each data set. Fit parameters and number of cells are given in Table 1.

Recovery from Steady-State Inactivation. Fig. 7 shows the protocol used and typical results obtained from a single myocyte to measure the rate of recovery from inactivation. A two-pulse protocol was used with both pulses being identical steps from the holding potential of 100 to 0 mV for 10 ms. The interpulse interval was varied from 2 to 142 ms in 10-ms increments to determine the rate at which the Na⁺ channels recover from inactivation induced by the first pulse. In the figure, the first trial of the protocol is bolded. In subsequent trials, the first pulse is repeated followed by the second pulse with a greater delay. Peak currents measured during the test (second) pulse were normalized by dividing by the peak current during the prepulse. The figure shows that under control conditions the test pulse currents returned to control levels rapidly. DHOM reduced the size of the currents but recovery from inactivation was clearly seen in this cell. Results for all cells were averaged for each treatment and plotted versus interpulse interval in Fig. 8. Although control records (half-filled squares) showed a rapid recovery of current, recovery was only 91% complete in these experiments, suggesting that some slower inactivation process is not fully measured in this time scale. The time between trials in these experiments was set to 10 s to allow full recovery between each trial. Data were plotted versus interpulse interval and fit to a single exponential function. Parameters are given in Table 2. Although results similar to control were obtained in cells treated with 50 µM DHOA, those cells treated with DHOM, EOA, and EOM all showed a significant slowing in the rate of recovery. EOM, although significantly different from control, was also significantly different from both DHOM and EOA. In addition, significant differences in the extent of recovery were found between control and DHOM-, EOA-, and EOM-treated cells. This suggests that these compounds are also affecting this slower inactivation process.

View larger version (35K): [in this window] [in a new window]   Fig. 7.   Recovery from inactivation. Top, protocol used to measure the rate of recovery of INa from fast inactivation. From a holding potential of 100 mV a 10-ms prepulse to 0 mV was followed 2 to 140 ms later by an identical test pulse. Bottom, current records from a single cell under control conditions and after exposure to 50 µM DHOM. Peak currents were measured during each test pulse and normalized to the current magnitude during the prepulse. Cell capacitance was 136 pF.

View larger version (27K): [in this window] [in a new window]   Fig. 8.   Effect of linoleic acid metabolites on recovery from inactivation. Data from experiments like that shown in Fig. 7 were collected from multiple cells (n given in Table 2). Symbols represent mean ± S.E. for each treatment. Cells exposed to 50 µM DHOM and EOA show a recovery time that was significantly slower than control, and 50 µM EOM was significantly slower than control but significantly faster than DHOM and EOA. Data points for DHOM, EOA, and EOM are significantly different from control values for recovery times greater than 10 ms (P < 0.05). Lines represent best fit of a single exponential function to each data set. Fit parameters are given in Table 2.

Dose-Response Relationship. The dose dependence of the I_Na inhibition displayed by LAMs was determined by applying a single repeating pulse from the holding potential of 100 to 10 mV for 10 ms to a whole-cell patch-clamped myocyte. After the I_Na reached equilibrium, the solution was changed to a higher concentration of drug and allowed to reach a new equilibrium. This process was completed for concentrations of 0.1, 1, 3, 10, 20, and 50 µM LAMs. Higher concentrations were not used because the myocytes either became leaky or contracted into balls in such conditions. This may be reflective of our previous observation that these compounds have multiple effects (Stimers et al., 1999; Moran et al., 2001; Ha et al., 2002). The peak I_Na from each pulse was plotted against time (data not shown). This plot was used to identify the average peak I_Na of the various concentrations. These peak I_Na values were divided by the control peak I_Na of each cell and expressed as percentage of control in Fig. 9. The curves represent the best fit of the Hill equation to each data set assuming that at high enough concentrations all of I_Na would be blocked and a Hill slope of 1. The half-inhibitory concentration (IC₅₀) values obtained from these fits are given in Fig. 9. As can be seen, DHOM and EOA showed lower IC₅₀ values than did DHOA or EOM. It should be noted that if different assumptions are made to generate the fit curve (e.g., less than 100% inhibition) then different parameter values would be obtained. Due to the uncertainty of these fit parameters, no statistical inferences are being made herein.

View larger version (25K): [in this window] [in a new window]   Fig. 9.   Dose-response curves for DHOA, DHOM, EOA, and EOM. To determine the dose dependence of INa inhibition caused by these four compounds, a single repeating voltage pulse (from 100 to 10 mV for 20 ms at 0.1 Hz) was applied to a whole-cell patch-clamped myocyte. After INa reached equilibrium in control solution, each compound was applied in sequentially increasing concentrations (without washout) from 0.1 to 50 µM. The average peak INa for each concentration was then normalized to the average control peak INa measured in the same cell, yielding the percentage of inhibition. The curves shown are the best fit of a Hill equation to each data set. Because it was not possible to use higher concentrations of these compounds because of cell death, two assumptions were made in fitting these data. First, it was assumed that all four compounds could completely inhibit INa. Second, it was assumed the Hill coefficient was equal to 1. IC50 values are given in the Figure. Six to seven cells were used in each dose-response curve.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Linoleic acid metabolites have been shown to be very toxic in a variety of cell types, in several species of animals, and most importantly in humans. Effects include cell death, cardiac arrhythmias, multiple organ failure, and cardiac arrest. In a clinical study, it was found that severely burned patients' mean peak plasma concentrations of EOA were 99 ± 25 µM and that the level was correlated with mortality (Kosaka et al., 1994). This study was undertaken to identify the mechanism by which these compounds affect electrical activity in the heart. Previously, we have shown that DHOM prolongs the cardiac action potential and slows its rate of rise (Stimers et al., 1999). This suggested that Na⁺ channels were involved. In this study, we have investigated in detail the effects of four structurally similar linoleic acid metabolites on I_Na. Significant new findings in this study have shown in adult rat ventricular myocytes a pronounced dose-dependent reduction in Na⁺ current density caused by DHOM and EOA especially. Current inhibition was not due to shifts in activation kinetics; however, a significant shift in steady-state inactivation was measured. Interestingly, of the four tested compounds only EOM caused a significant change in the slope of the steady-state inactivation curve. Recovery from fast inactivation was slowed by EOA, EOM and DHOM, suggesting an interaction with the inactivation gates or a stabilization of the inactivated state.

Structure-Related Effects. Our previous study used only the methyl esters of these compounds and showed that Na⁺ current was sensitive to DHOM but not to EOM (Stimers et al., 1999). It was assumed that the cell's intrinsic esterase activity would convert these compounds to their free acid counterparts (DHOA and EOA). Results in this study suggest that this is not the case. Because DHOA was found to be almost completely without activity against I_Na, it is very unlikely that DHOM was converted to DHOA in either the previous study or the present one. In addition, in this study EOA was found to be more effective than EOM in its effects on most of the parameters measured. This further supports the idea that these compounds are not being de-esterified during these experiments.

Voltage Dependence of Inactivation. Inhibition of I_Na by any compound can occur by a single mechanism or by a combination of effects. Peak I_Na can be inhibited by blocking the conductance pathway, by shifting the voltage dependence of activation to more depolarized potentials, or by shifting the voltage dependence of inactivation to more hyperpolarized potentials. In the data shown in Figs. 3 and 4 there is no evidence of a shift in the voltage dependence of activation as the current first activates between 60 and 50 mV and it reaches its maximum at 30 mV under all tested conditions. However, the same is not true for inactivation. As shown in Figs. 5 and 6, steady-state inactivation is shifted about 6 mV in the hyperpolarizing direction by 50 µM DHOM, EOA, and EOM. Furthermore, EOM caused a significant increase in the slope of the steady-state inactivation curve. This suggests that the charge on this compound may be interacting with the voltage sensor or the membrane field in these experiments. Despite the shift in inactivation to more hyperpolarized potentials, this is not sufficient to explain the decreased magnitude, because I_Na was inhibited even with a steady-state holding potential of 100 mV, at which there was no detectable steady-state inactivation. Thus, although the shift in inactivation could contribute to the block of I_Na, it is likely that there is also a significant block of the channel pore by these compounds. Of course, this could be due to a block of channel gating rather than a physical block of the pore.

Kinetics of Recovery from Inactivation. One mechanism for inhibiting Na⁺ current that is exhibited by local anesthetics is a stabilization of the inactivated state (Jia et al., 1993). This is characterized by a slowing of the rate of recovery of I_Na from inactivation. The data shown in Figs. 7 and 8 confirm that three of the compounds do significantly slow the rate of recovery from inactivation. In addition, there was a slower component (not slow inactivation) not measured by these experiments that was also slowed, or at least its extent was enhanced, because the fractional recovery over this time period (140 ms) was significantly reduced. Note that this slowing of recovery was in addition to the steady-state block produced by these agents, because these data were normalized to the prepulse current amplitude, not control amplitude. Together, the shift of the steady-state inactivation curve and the increased recovery time suggest that these compounds inhibit I_Na by stabilizing the inactivated state of the cardiac sodium channel.

Physiological Significance. These data are of physiological significance because slight changes in cardiac sodium current can, over time, develop into arrhythmias and possibly lead to cardiac arrest. Furthermore, the characteristics of these LAMs on I_Na are similar to the effects of class I antiarrhythmic agents such as lidocaine and others. This suggests caution should be used in patients with elevated circulating fatty acids (e.g., severely burned or ischemic patients) needing treatment for arrhythmias. Further studies in this area could test the interaction of these compounds with lidocaine effects on I_Na.