1. The presence and properties of K+ channels activated by arachidonic acid were studied in neuronal cells cultured from the mesencephalic and hypothalamic areas of rat brain.2. Arachidonic acid produced a concentration-dependent (5-50 #UM) and reversible activation of whole-cell currents. 3. In excised membrane patches, arachidonic acid applied to the cytoplasmic or extracellular side of the membrane caused opening of three types of channels whose current-voltage relationships were slightly outwardly rectifying, inwardly rectifying and linear, and whose single channel slope conductances at +60 mV were 143, 45 and 52 pS, respectively.4. All three currents were K+ selective and blocked by 2 mm Ba2+ but not by other K+ channel blockers such as tetraethylammonium chloride, 4-aminopyridine and quinidine. The outwardly and inwardly rectifying currents were slightly voltage dependent with higher channel activity at more depolarized potentials.5. Arachidonic acid activated the K+ channels in cells treated with cyclo-oxygenase and lipoxygenase inhibitors (indomethacin and nordihydroguaiaretic acid), indicating that arachidonic acid itself can directly activate the channels. Alcohol and methyl ester derivatives of arachidonic acid failed to activate the K+ channels, indicating that the charged carboxyl group is important for activation. 6. Certain unsaturated fatty acids (linoleic, linolenic and docosahexaenoic acids), but not saturated fatty acids (myristic, palmitic, stearic acids), also reversibly activated all three types of K+ channel. 7. All three K+ channels were activated by pressure applied to the membrane (i.e. channels were stretch sensitive) with a half-maximal pressure of -18 mmHg. The K+ channels were not blocked by 100 ,UM GdCl3. 8. A decrease in intracellular pH (over the range 5 6-7 2) caused a reversible, pH-dependent increase in channel activity whether the channel was initially activated by arachidonic acid or stretch.9. Glutamate, a neurotransmitter reported to generate arachidonic acid in striatal neurons, did not cause activation of the K+ channels when applied extracellularly in cell-attached patches. 10. It is suggested that the K+ channels described here belong to a distinct family of ion channels that are activated by either fatty acids or membrane stretch. Although the physiological roles of these K+ channels are not yet known, they may be involved in cellular processes such as cell volume regulation and ischaemia-induced elevation of K+ loss.

Various tissues including heart express specific binding sites for endothelin. Endothelins have been reported to increase the force of contraction of cardiac muscle, presumably via specific receptors. Specific binding of endothelin to atrial tissue is particularly high. In spontaneously contracting rat atrial cells used in this study, all three isoforms of endothelin (endothelin-1, endothelin-2, and endothelin-3) decreased the rate of beating and caused an increase in inwardly rectifying K+ current in voltage-clamped whole cells. Endothelin-3 was the most potent isoform, and its effects on beating rate and K+ current were present at a concentration as low as 100 pM (Kd, approximately 1 nM). the atrial cells did not have the hyperpolarization-activated current (the pacemaker current), If. In excised inside-out patches, all three isoforms of endothelin activated a population of K+ channels with kinetic properties identical to those of acetylcholine (muscarinic)-activated K+ channels, and this was GTP dependent. Endothelin failed to decrease the beating rate or to elicit the K+ current in pertussis toxin-treated cells. These results indicate that endothelin has a potent negatively chronotropic effect by activation of the inwardly rectifying, muscarinic K+ channel and therefore could be an important regulator of heart function.

Arachidonic acid has been shown to activate K(+)-selective, mechanosensitive ion channels in cardiac, neuronal and smooth muscle cells. Since the cardiac G protein (GK)-gated, muscarinic K+ (KACh) channel can also be activated by arachidonic acid, we investigated whether the KACh channel was also sensitive to membrane stretch. In the absence of acetylcholine (ACh), KACh channels were not active, and negative pressure failed to activate these channels. With ACh (10 microM) in the pipette, applying negative pressure (0 to -80 mm Hg) to the membrane caused a reversible, pressure-dependent increase in channel activity in cell-attached and inside-out patches (100 microM GTP in bath). Membrane stretch did not alter the sensitivity of the KACh channel to GTP. When GK was maximally activated with 100 microM GTP gamma S in inside-out patches, the KACh channel activity could be further increased by negative pressure. Trypsin (0.5 mg/ ml) applied to the membrane caused activation of the KACh channel in the absence of ACh and GTP; KACh channel activity was further increased by stretch. These results indicate that the atrial muscarinic K+ channels are modulated by stretch independently of receptor/G protein, probably via a direct effect on the channel protein/ lipid bilayer.