is not myristoylated (3). However, when synthesized in Escherichia coli, a, has reduced affinities for f8y, adenylyl cyclase, and Ca2+ channels (4-6). Hypothetically, the differences between native and recombinant a, are due to the lack of unknown posttranslational modifications of the recombinant protein (4). Furthermore, the structural features of a, necessary for association with membranes have not been fully characterized (7-9). Other a subunits, including members of the Gq family (activators of phospholipase C-*), lack the requisite glycine residue at position 2 (10) and are also presumably not myristoylated. Some membrane-associated proteins, including certain forms of p2lms and receptors such as rhodopsin, are palmitoylated (11). Palmitate is almost always linked to cysteine residues through a thioester bond. The function of proteinbound palmitate is poorly defined. In an attempt to identify posttranslational modifications of as and aq, we examined a subunits for incorporation of radioactive palmitate. We report here that tritiated palmitate is incorporated into a, and aq and, in addition, into a subunits that are also myristoylated (ao, ai, and az). tHowever, simple interpretation of these data is hampered by the fact that myristoylation is a stable modification, whereas palmitoylation is dynamic. The specific activity of protein-bound myristate would reflect that of the precursor pool over the time course of the experiment. The specific activity of protein-bound palmitate might reflect that of the precursor pool only at the end of the experiment if tumover were sufficiently rapid. MATERIALS AND METHODS 3675The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.

The action of aldosterone to increase apical membrane permeability in responsive epithelia is thought to be due to activation of sodium channels. Aldosterone stimulates methylation of a 95-kDa protein in apical membrane of A6 cells, and we have previously shown that methylation of a 95-kDa protein in the immunopurified Na ؉ channel complex increases open probability of these channels in planar lipid bilayers. We report here that aldosterone stimulates carboxylmethylation of the ␤ subunit of xENaC in A6 cells. In vitro translated ␤ subunit, but not ␣ or ␥, serves as a substrate for carboxylmethylation. Carboxylmethylation of ENaC reconstituted in planar lipid bilayers leads to an increase in open probability only when ␤ subunit is present. When the channel complex is immunoprecipitated from A6 cells and analyzed by Western blot with antibodies to the three subunits of xENaC, all three subunits are recognized as constituents of the complex. The results suggest that Na ؉ channel activity in A6 cells is regulated, in part, by carboxylmethylation of the ␤ subunit of xENaC.Aldosterone-stimulated sodium channel activity has been shown to involve methylation of membrane proteins (1, 2). Moreover, the aldosterone-induced increase in the activity of xENaC is blockable by the methylation inhibitor, 3-deazaadenosine (3). Sariban-Sohraby has demonstrated that aldosterone stimulates the methylation of a 90 -95-kDa protein in the apical membrane of A6 cells (4). We have previously shown that methylation of the 95-kDa subunit of an immunopurified renal sodium channel complex reconstituted in lipid bilayers results in increased sodium channel activity (5). In both studies the identity of the methylated protein is unknown.

We have studied the function of acetylcholine (AcCho) receptors (AcChoRs) in rat soleus endplates before and after exposing the muscles to forskolin, a potent activator of adenylate cyclase. AcChoR function was tested by recording the membrane depolarization evoked by pulses of ionophoretically applied AcCho. Brief (2 msec) AcCho pulses delivered at 7 Hz evoked constant responses at untreated endplates. In contrast, after 10-100 ,LM forskolin was added to the bath, responses to similar pulse trains fell by as much as 80% within 1 sec. AcCho sensitivity recovered completely in <1 min after the pulses were stopped but fell again when the pulses were resumed. Similarly, longer (1 sec) ionophoretic AcCho pulses evoked roughly constant responses at control endplates, but after forskolin treatment the depolarization fell by one-half within <200 msec. These results indicate that forskolin increases the rate at which AcChoRs desensitize when exposed to agonist. Focal extracellular recordings showed that 20-100 ,AM forskolin also increased the decay rate of miniature endplate currents, indicating that forskolin may decrease AcChoR channel open time. Inhibitors of cAMP phosphodiesterase increased the potency of forskolin. When used alone, these inhibitors had effects similar to those of forskolin but smaller. Patch-clamp experiments indicated that forskolin at 100 ,uM may also interact with AcChoR channels directly, but at 20 ,M this effect is negligible. Therefore, it is likely that the forskolin effects were mediated primarily by increased levels of intracellular cAMP.Protein phosphorylation is an important mechanism for modulating the gating properties of many species of ionic channels (reviewed in refs. 1 and 2). Examples of ionic currents modulated by cAMP-dependent protein phosphorylation include voltage-sensitive calcium currents (3, 4), voltage-dependent potassium currents (5-7), serotonin-sensitive potassium currents (8), and calcium-dependent potassium currents (9). Other species of protein kinase also modulate gating (1,2).In virtually all cases, the physiological consequences of kinase activation are better defined than the underlying biochemical mechanisms. In particular, in most systems it is not known if the critical phosphoprotein is the channel itself or a separate regulatory protein. The situation is just the reverse for the nicotinic acetylcholine (AcCho) receptor (AcChoR), currently the best-characterized chemically gated ionic channel (10). In this case, it is clear that the AcChoR itself can be phosphorylated by three different species of protein kinase, but the functional significance of phosphorylation and dephosphorylation is poorly understood (11)(12)(13)(14).We have begun to explore the possible effects of AcChoR phosphorylation by cAMP-dependent protein kinase by studying the function of receptors in rat soleus endplates before and after treating the muscles with forskolin, a potent activator of adenylate cyclase (15). We found that forskolin effects a striking but reversible increase i...

To test the hypothesis that intracellular Ca2+activation of large-conductance Ca2+-activated K+ (BK) channels involves the cytosolic form of phospholipase A2(cPLA2), we first inhibited the expression of cPLA2 by treating GH3 cells with antisense oligonucleotides directed at the two possible translation start sites on cPLA2. Western blot analysis and a biochemical assay of cPLA2activity showed marked inhibition of the expression of cPLA2 in antisense-treated cells. We then examined the effects of intracellular Ca2+ concentration ([Ca2+]i) on single BK channels from these cells. Open channel probability ( P o) for the cells exposed to cPLA2 antisense oligonucleotides in 0.1 μM intracellular Ca2+ was significantly lower than in untreated or sense oligonucleotide-treated cells, but the voltage sensitivity did not change (measured as the slope of the P o-voltage relationship). In fact, a 1,000-fold increase in [Ca2+]ifrom 0.1 to 100 μM did not significantly increase P oin these cells, whereas BK channels from cells in the other treatment groups showed a normal P o-[Ca2+]iresponse. Finally, we examined the effect of exogenous arachidonic acid on the P oof BK channels from antisense-treated cells. Although arachidonic acid did significantly increase P o, it did so without restoring the [Ca2+]isensitivity observed in untreated cells. We conclude that although [Ca2+]idoes impart some basal activity to BK channels in GH3 cells, the steep P o-[Ca2+]irelationship that is characteristic of these channels involves cPLA2.

Incubating skeletal muscle fibers with forskolin, an activator of adenylate cyclase, increases the rate at which nicotinic acetylcholine receptors (AChRs) desensitize when exposed to ACh. Several reports indicate that this is due to the phosphorylation of AChRs by cAMP-dependent protein kinase, but other studies suggest that forskolin interacts with AChRs directly and that second-messenger systems are not required. To help clarify this issue, we studied the effects of forskolin and several other drugs on AChR function in embryonic rat myotubes. AChR function was studied by recording ACh-induced membrane depolarizations and ACh-induced single-channel currents. Our results indicate that forskolin at low concentrations enhances AChR desensitization through the action of a second messenger, most likely cAMP. An analog of forskolin that is much less effective in activating adenylate cyclase (1,9-dideoxyforskolin) is also much less potent in enhancing desensitization. Forskolin at low concentrations does not alter single-channel conductance or mean channel open time. However, when used at concentrations above 20 microM, forskolin may also exert direct drug effects on AChRs.