Differential sensitivity of inward rectifier K+ channels to metabolic inhibitors.

Inhibition of inward rectifier K(+) channels under ischemic conditions may contribute to electrophysiological consequences of ischemia such as cardiac arrhythmia. Ischemia causes metabolic inhibition, and the use of metabolic inhibitors is one experimental method of simulating ischemia. The effects of metabolic inhibitors on the activity of inward rectifier K(+) channels K(ir)2.1, K(ir)2.2, and K(ir)2.3 were studied by heterologous expression in Xenopus oocytes and two-electrode voltage clamp. 10 microm carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) inhibited K(ir)2.2 and K(ir)2.3 currents but was without effect on K(ir)2.1 currents. The rate of decline of current in FCCP was faster for K(ir)2.3 than for K(ir)2.2. K(ir)2.3 was inhibited by 3 mm sodium azide (NaN(3)), whereas K(ir)2.1 and K(ir)2.2 were not. K(ir)2.2 was inhibited by 10 mm NaN(3). All three of these inward rectifiers were inhibited by lowering the pH of the solution perfusing inside-out membrane patches. K(ir)2.3 was most sensitive to pH (pK = 6.9), whereas K(ir)2.1 was least sensitive (pK = 5.9). For K(ir)2.2 the pK was 6.2. These results demonstrate the differential sensitivity of these inward rectifiers to metabolic inhibition and internal pH. The electrophysiological response of a particular cell type to ischemia may depend on the relative expression levels of different inward rectifier genes.

Inward rectifier K ϩ channels are known to be involved in the electrophysiology of several different tissues, including the heart (1). In fast response myocardial tissue they are responsible for the relatively negative value of the resting membrane potential and the late rapid repolarization phase of the action potential (2). Their characteristic property of inward rectification allows the myocardial cell to generate long lasting action potentials while minimizing ion fluxes (3).
Ischemia produces electrophysiological changes that can lead to the generation of cardiac arrhythmias (4) or cerebral excitotoxicity (5). Simulated ischemia has been shown to inhibit an inwardly rectifying current in isolated cardiac myocytes (6). This effect may contribute to arrhythmogenesis and is therefore worthy of further study. Three closely related inward rectifier genes, K ir 2.1, K ir 2.2, and K ir 2.3, are expressed in myocardium (7,8), so each of these could potentially be involved in the aforementioned ischemic response in the heart. These genes are also expressed in the brain (9 -11). In this report we have simulated ischemia by applying metabolic inhibitors to Xenopus oocytes expressing these three inward rectifier K ϩ channels. We show that K ir 2.1, K ir 2.2, and K ir 2.3 are affected differently by metabolic inhibition. We also show that the sensitivity of K ir 2.2 to internal pH is intermediate between that of K ir 2.1 and K ir 2.3.
Oocyte Isolation-Stage V-VI oocytes were surgically removed from Xenopus laevis frogs (Nasco) under anesthesia (0.03% benzocaine for 10 -15 min) and incubated with 1.3 mg/ml collagenase (Worthington, type CLS3) for 2 h at 22°C in 96 mM NaCl, 2 mM KCl, 1 mM MgCl 2 , 5 mM HEPES, pH 7.4, with agitation to remove connective tissue. Oocytes were then washed several times with 96 mM NaCl, 2 mM KCl, 1 mM MgCl 2 , 1.8 mM CaCl 2 , 5 mM HEPES, pH 7.4, and stored in the same solution at 18°C. Both solutions also contained 100 g/ml streptomycin and 60 g/ml ampicillin. Between 16 and 24 h later, oocytes were injected (Nanoliter; World Precision Instruments) with amounts of RNA that gave approximately equal expression levels (0.2 ng of K ir 2.1, 4 ng of K ir 2.2, and 4 ng of K ir 2.3) in 50 nl of nuclease-free water.
Two-electrode Voltage Clamp-Inward rectifier currents were recorded 1-2 days after injection using a TEC-03 amplifier (npi, Germany) controlled by Pulse 8.4 software (Heka, Germany) via an ITC-16 computer interface (Instrutech, Long Island, NY). Currents were filtered at 500 Hz and digitized at 1 kHz. Data were analyzed using Pulse 8.4 and Prism 3.02 (GraphPad). Microelectrode pipettes had resistances of 0.5-1.5 megohms. Oocytes were placed in a 100-l volume recording chamber that was continuously perfused at a rate of ϳ1.2 ml/min. Different solutions entered the chamber via a common outlet with a dead space of ϳ50 l. Under control conditions oocytes were perfused with "90K" solution: 90 mM KCl/KOH, 3 mM MgCl 2 , 5 mM HEPES, pH 7.4. Sodium azide (NaN 3 ) and BaCl 2 were dissolved in aliquots of this solution as solid. Carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) 1 (ICN, Costa Mesa, CA) was dissolved in dimethyl sulfoxide (Me 2 SO) to 100 mM. This stock solution was diluted into 90K.
Patch Clamp Recording-Inward rectifier currents were recorded in "giant" inside-out patches from Xenopus oocytes (13) 3-9 days after injection with K ir 2.x RNA. The patch clamp amplifier was an Axopatch 200B (Axon Instruments). Currents were filtered at 1 kHz, and data were acquired at 4.76 kHz with a Digidata 1320A computer interface and pClamp 8 software (Axon Instruments). Data were analyzed using pClamp 8 and this solution that had been further adjusted to different pH values with HCl. Switching between different perfusion solutions was achieved with a BPS-8 valve control system and a QMM micromanifold with a dead space of ϳ1 l (ALA Scientific Instruments, Westbury, NY). The perfusion rate was 85 l/min.

Effects of FCCP on Inward Rectifier Currents in Intact
Oocytes-Inward rectifier currents were monitored by means of a continuous two-electrode voltage clamp protocol. Membrane potential was held at 0 mV and at 20-s intervals was stepped to 50 mV for 75 ms and then to Ϫ50 mV for 350 ms. Expression of inward rectifier channels was confirmed by brief application of 20 mM Ba 2ϩ , which completely blocks K ir 2.1, K ir 2.2, and K ir 2.3 (9 -11). Oocytes were first perfused with 90K to ensure a steady current, and then the perfusion solution was switched to 90K plus 10 M FCCP, which inhibits ATP production by dissipating the H ϩ gradient across the inner mitochondrial membrane (15). (This solution also contained 0.01% Me 2 SO. Me 2 SO alone had no effect on inward rectifier currents at this concentration.) In oocytes expressing K ir 2.1, inward rectifier current remained steady in the presence of 10 M FCCP. In contrast, 10 M FCCP inhibited inward rectifier current in oocytes expressing K ir 2.2 or K ir 2.3. Examples of these experiments are shown in Fig. 1. To rule out the unlikely possibility that the decline of K ir 2.2 and K ir 2.3 currents was because of spontaneous "rundown," we perfused oocytes expressing these clones with 90K (plus 0.01% Me 2 SO) for prolonged periods without switching to FCCP. As shown by the examples in Fig. 2, spontaneous rundown did not occur. Indeed, in some cases (as in Fig. 2) the current had a tendency to slowly increase. This was not because of an increase in leak current, as demonstrated by applying 20 mM Ba 2ϩ .
Because the rate of decrease of inward rectifier currents in FCCP was quite slow, we chose an arbitrary time point of 15 The voltage clamp protocol was the same as in Fig. 1. The data in panel A represent the current recorded at Ϫ50 mV. The oocytes were briefly exposed to 20 mM Ba 2ϩ at the times indicated. Panels B and C show current traces obtained at the times indicated by the letters a, b, and c, for K ir 2.2 and K ir 2.3, respectively. The horizontal line indicates zero current level. Figs. 1 and 2. The current recorded at Ϫ50 mV was used for this analysis. Oocytes were perfused with 90K plus 10 M FCCP or 90K alone (plus 0.01% Me 2 SO) for 15 min. The perfusate was then switched to 90K plus 20 mM BaCl 2 . The current remaining in the presence of Ba 2ϩ was subtracted from currents recorded at the beginning and end of the 15-min interval. The Ba 2ϩ -sensitive current at the end of the 15-min interval was normalized to the Ba 2ϩ -sensitive current at the beginning of the 15-min interval. A standard t test was used to compare pairs of data groups. p values are as follows: p Ͻ 0.005 for column 1 versus column 2, column 2 versus column 3, and column 2 versus column 4; p Ͻ 0.0001 for column 3 versus column 5. Bars represent means Ϯ S.E. from 6 oocytes. min to compare the response of K ir 2.1, K ir 2.2, and K ir 2.3 to FCCP. Fig. 3 shows that there was a significant difference between K ir 2.2 and K ir 2.3 in terms of the fractional current remaining after 15 min of exposure to FCCP, demonstrating that the rate of K ir 2.2 current decline was slower than that of K ir 2.3. The fractional K ir 2.2 current remaining after 15 min was significantly different from that of K ir 2.1, which was unaffected by FCCP. In excised membrane patches, 10 M FCCP was without effect on K ir 2.2 and K ir 2.3 (data not shown).

FIG. 3. Comparison of the effects of FCCP on K ir 2.1, K ir 2.2, and K ir 2.3. Currents were recorded by two-electrode voltage clamp as in
Effects of Sodium Azide on Inward Rectifier Currents in Intact Oocytes-To gain further evidence that the effect of FCCP was because of metabolic inhibition, we tested another metabolic inhibitor, sodium azide (NaN 3 ) (16,17). Voltage clamp experiments were carried out using the same protocol as that used for the FCCP experiments. As shown in Fig. 4, K ir 2.3 currents decreased significantly during 15 min of exposure to 3 mM NaN 3 , whereas K ir 2.1 and K ir 2.2 currents were unaffected by this treatment. However, K ir 2.2 currents were inhibited by 10 mM NaN 3 .
Effects of Internal pH on Inward Rectifier Currents in Insideout Patches-It has been reported that metabolic inhibitors decrease intracellular pH (18,19). It has also been reported that K ir 2.3 is more sensitive than K ir 2.1 to decreased intracellular pH (14). Therefore the sensitivity of K ir 2.3 to metabolic inhibitors compared with K ir 2.1 is consistent with a decrease in intracellular pH caused by metabolic inhibition. To investigate whether the intermediate sensitivity of K ir 2.2 to metabolic inhibitors could be explained in terms of pH sensitivity, we compared the dose response of the three inward rectifiers to internal pH in inside-out membrane patches. As shown in Fig.  5, the sensitivity of K ir 2.2 to decreasing pH was intermediate between that of K ir 2.1 and K ir 2.3.

DISCUSSION
Acute ischemia produces many changes in the intracellular environment. These changes affect the activity of several different ion channels, leading to cellular electrophysiological alterations that can produce arrhythmias in the heart (4) or excitotoxicity in the brain (5). Inward rectifier K ϩ channels play an important role in the electrophysiology of the cardiac myocyte (2,20), and inhibition of their activity will have a significant and possibly arrhythmogenic effect (4). Acute ischemia produces a shortening of the action potential because of activation of ATP-sensitive K ϩ channels preceded in some cases by a lengthening. The lengthening has been attributed to inhibition of the transient outward current accompanied by inhibition of the inward rectifier current in some preparations (6) but not others (21). Further studies are required to determine the contribution of inward rectifier channels to changes in action potential duration during ischemia relative to other channel types.
The results presented here suggest that the effect of ischemia on inward rectifier current will depend on the molecular identity of the channels. We have demonstrated that the inward rectifiers K ir 2.1, K ir 2.2, and K ir 2.3 are differentially sensitive to inhibition of cellular metabolism. Because cellular metabolism is inhibited in ischemia (4), our experimental conditions simulate some of the conditions prevailing during ischemia. Therefore it seems reasonable to propose that of these three related inward rectifiers K ir 2.3 will be inhibited the most under ischemic conditions, whereas K ir 2.1 will be inhibited the least.
One of the intracellular changes that occurs during ischemia and metabolic inhibition is a decrease in pH (4,18,19). In this ] at which the current is 50% inhibited and n is the Hill coefficient. pK values are 5.9, 6.2, and 6.9 for K ir 2.1, K ir 2.2, and K ir 2.3, respectively. Hill coefficients are 2.2, 2.8, and 3.0 for K ir 2.1, K ir 2.2, and K ir 2.3, respectively. study, the order of sensitivity of K ir 2.1, K ir 2.2, and K ir 2.3 to metabolic inhibition was the same as the order of sensitivity to lowering internal pH. However, the relative sensitivity to metabolic inhibition cannot be accounted for by a change in internal pH alone. For example, the degree of inhibition of K ir 2.2 after 15 min of exposure to 10 M FCCP (15%; Fig. 3, column 2) corresponds to an internal pH of 6.5 (Fig. 5D), whereas the degree of inhibition of K ir 2.3 after 15 min of exposure to 10 M FCCP (38%; Fig. 3, column 3) corresponds to an internal pH of 7.0 (Fig. 5D). Therefore our results suggest the influence of other intracellular consequences of metabolic inhibition. For example, metabolic inhibitors have been shown to deplete intracellular ATP in Xenopus oocytes (22).
Experimental data and simulations indicate that a change of a few nanosiemens in the K ϩ conductance of a cardiac myocyte can significantly affect action potential duration (23). Therefore, inhibition of only a fraction of the Ϸ150 nanosiemens inward rectifier conductance in the cardiac myocyte (24) may have a significant effect in ischemia. In the heart the inward rectifier current (I K1 ) is probably because of expression of both K ir 2.1 and K ir 2.2 with a small or no contribution from K ir 2.3, depending on the species (7)(8)(9)(10)(11)25). However, the relative functional expression levels of these different channels are not known accurately. Inward rectifier mRNA levels have been measured in human heart, but they may not reflect expression at the functional level (8). Differences in the relative expression levels of these genes in different anatomical locations in the heart may produce spatial disparities in the refractory period in response to ischemia. Such situations are proarrhythmic (4). Details of the spatial distributions of inward rectifier gene expression in the heart are unknown at present. As an added complication, it is possible that K ir 2.2 exists as a heteromultimer with K ir 2.1 (25). Although K ir 2.3 may not contribute much to I K1 in the heart, its location in neurons (26,27) raises the possibility of a role in ischemia-induced excitotoxicity.