Search

Human alveolar macrophages were obtained from macroscopically normal lung tissue obtained at surgical resections, isolated by adherence, and identified by morphology. Whole cell recordings were made from cells 1-3 h in culture, using electrodes containing potassium chloride. From a holding potential of -100 mV, depolarizing pulses to -40 mV or greater activated an outward current. Tail current reversals showed that this current was potassium selective. Margatoxin completely blocked the current; the concentration giving half-maximal block was 160 pM. In current clamp recordings, the resting membrane potential was -34 mV; margatoxin depolarized cells to close to 0 mV. A pure macrophage population was isolated by fluorescence-activated cell sorting, using the phagocytosis of BODIPY-labeled zymosan particles. Reverse transcription-polymerase chain reaction showed that, of 13 voltage-gated K+ (Kv) potassium channels sought, only Kv1.3 mRNA was present. Margatoxin (1 nM) did not affect the percentage of cells showing phagocytosis sorted from the total population. Under these experimental conditions Kv1.3 sets the resting potential of the cells, but it is not required for Fc receptor-mediated phagocytosis.

Structural determinants of permeation in large unit conductance calcium-activated potassium channels (BK channels) were investigated. Y293 and F294 in the P-region of dSlo were substituted by tryptophans. Compared to wild-type channels, Y293W channels displayed reduced inward unitary currents while F294W channels exhibited normal inward current amplitudes but flickery kinetics. Both mutations produced changes in current/voltage relations under bi-ionic conditions. Sensitivity to block by external tetraethylammonium (TEA) was affected in both channels, and the voltage dependence of TEA block was increased in F294W channels. Both mutations also affected gating by shifting the half-maximal activation voltage of macroscopic conductance/voltage relations to more positive potentials, and eliminating a slow component of deactivation. The double mutant did not produce ionic currents. These data are consistent with a model in which Y293 contributes to a potassium-binding site close to the outer mouth of the dSlo pore, while F294 contributes to an energy barrier near this site.

1. The effects of multivalent cations on membrane currents recorded from Xenopus oocytes were studied. 2. The hyperpolarization-activated chloride current was reversibly blocked by lanthanum; half-maximal block occurred at a concentration of 8 microM. Zinc, cadmium, cobalt and nickel were less potent than lanthanum, and gadolinium, manganese, barium and strontium had no effect at a concentration of 100 microM. 3. The calcium-activated chloride current was blocked by gadolinium (50 microM), and lanthanum, cadmium, cobalt, nickel and manganese were equally effective. The actions of gadolinium and lanthanum were almost irreversible, while partial (30-80%) recovery was observed with the other cations. Zinc (100 microM) had no effect. 4. In lanthanum (100 microM), membrane depolarizations from -70 mV activated an outward potassium current that was partially blocked by barium (0.1-2 mM). The barium-sensitive current was confined to potentials less negative than -70 mV. The current consisted of a time-independent as well as a time-dependent component, the latter of which had voltage dependence similar to the M-current. 5. It is proposed that lanthanum, gadolinium and barium can usefully separate these endogenous membrane currents in Xenopus oocytes.

An opioid-like receptor has been cloned by several groups of researchers and recently shown to be activated by an endogenous heptadecapeptide termed orphanin FQ (or nociceptin). We isolated the corresponding mouse cDNA and coexpressed it in Xenopus laevis oocytes with the potassium channel subunits Kir3.1 (GIRK1) and Kir3.4 (CIR, rcKATP). Orphanin FQ evoked potassium currents, with 50% of the maximal effect at approximately 1 nM; [Tyr1]orphanin FQ was equally effective, and des-pheorphanin FQ was without activity. Dynorphin A, dynorphin(1-9), dynorphin(1-13), and alpha-neoendorphin were > 100 times less potent, and other agonists active at mu-, delta-, and kappa-opioid receptors had no effect. Naloxone (1 microM) and norbinaltorphimine (1 microM) had no antagonist action. Conversely, oocytes expressing kappa receptors responded to dynorphin (half-maximal concentration, 0.3 nM) but not to orphanin FQ. Thus, both kappa and orphanin FQ receptors readily couple to potassium channels, but the highly selective activation by dynorphin and orphanin FQ is consistent with distinct functional pathways in vivo.

Human D2, D3, D4 and dopamine receptors were individually coexpressed in Xenopus oocytes with a G protein-regulated inwardly rectifying potassium channel (GIRK1). At -100 mV in 96 mM potassium, dopamine (0.1-100 nM) evoked an inward current; the current showed inward rectification, reversed polarity at 0 mV, and was blocked by barium (50% inhibition by 10 microM). The concentrations of dopamine activating 50% of the maximal current (EC50) were not different (2-4 nM) for D2, D3, and D4 receptors, but the maximal current was 3-fold larger for D2 and D4 than for D3 receptors. Dopamine evoked reproducible inward currents at D2 and D4 receptors when applied repeatedly, but second responses could not be observed in oocytes expressing D3 receptors. 7-Hydroxy-N,N-di-n-propyl-2-aminotetralin mimicked the effect of dopamine (EC50 of approximately 2, approximately 3, and approximately 19 nM at D2, D3, and D4, respectively). (-) Sulpiride reversibly blocked the dopamine-induced current with IC50 values of 5, 300, and 2000 nM for D2, D3, and D4 receptors, respectively. Dopamine was ineffective in oocytes injected 2 hr previously with pertussis toxin. We concluded that all three D2-like dopamine receptors share the potential to activate inwardly rectifying potassium channels.

In Parkinson's disease, there is evidence of impaired mitochondrial function which reduces the capacity to synthesize ATP in dopamine neurons. This would be expected to reduce the activity of the sodium pump (Na+/K+ ATPase), causing increased intracellular levels of Na+. Patch pipettes were used to introduce Na+ (40 mM in pipette solutions) into dopamine neurons in the rat midbrain slice in order to study the electrophysiological effects of increased intracellular Na+. We found that intracellular Na+ loading evoked 100-300 pA of outward current (at -60 mV) and increased whole-cell conductance; these effects developed gradually during the first 10 min after rupture of the membrane patch. Extracellular Ba2+ reduced most of the outward current evoked by Na+ loading; this Ba(2+)-sensitive current reversed direction at the expected reversal potential for K+ (EK), and was also blocked by extracellular tetraethylammonium (30 mM) and intracellular Cs+ (which replaced K+ in pipette solutions). The sulfonylurea drugs glipizide (IC50 = 4.9 nM), tolbutamide (IC50 = 23 microM) and glibenclamide (1 microM) were as effective as 300 microM Ba2+ in reducing the K+ current evoked by Na+ loading. When recording with "control" pipettes containing 15 mM Na+, diazoxide (300 microM) increased chord conductance and evoked outward current at -60 mV, which also reversed direction near EK. Effects of diazoxide were blocked by glibenclamide (1 microM) or glipizide (300 nM). Diazoxide (300 microM) and baclofen (3 microM), which also evoked K(+)-mediated outward currents recorded with control pipettes, caused little additional increases in outward currents during Na+ loading. Raising ATP concentrations to 10 mM in pipette solutions failed to significantly reduce currents evoked by diazoxide or Na+ loading, suggesting that these currents may not be mediated by ATP-sensitive K+ channels. Finally, Na+ loading using pipettes containing Cs+ in place of K+ evoked a relatively small outward current (50-150 pA at -60 mV), which developed gradually over the first 10 min after rupturing the membrane patch. This current was reduced by dihydro-ouabain (3 microM) and a low extracellular concentration of K+ (0.5 mM instead of 2.5 mM), but was not affected by Ba2+. We conclude that intracellular Na+ loading evokes a current generated by Na+/K+ ATPase in addition to sulfonylurea-sensitive K+ current. This Na(+)-dependent K+ current is unusual in its sensitivity to sulfonylureas, and could protect dopamine neurons against toxic effects of intracellular Na+ accumulation.

Xenopus oocytes were injected with RNAs for the two inward-rectifier potassium channel subunits Kir3.1 (GIRK1) and Kir3.4 (rcKATP or CIR) in addition to RNA from the neuroblastoma cell line KAN-TS. Potassium currents were evoked by neuropeptide Y in oocytes injected with polyadenylated RNA or with cRNA from pools of a neuroblastoma (KAN-TS) cDNA library, and progressive subdivision of responding pools yielded a single cDNA. The encoded protein contains 381 amino acids, has the seven hydrophobic domains characteristic of G protein-coupled receptors, and is 31% identical to the Y1 receptor: potassium currents were induced by neuropeptide Y (EC50=60pm) and Y2-selective analogues. Coexpression with potassium channel subunits will be a generally useful method for the cloning of G protein-coupled receptors.

Potassium channel subunits have six or two transmembrane segments in addition to a conserved pore-forming (P) domain; four subunits come together to form a channel. A gene was identified in S. cerevisiae (J0911) encoding a protein with eight probable membrane-spanning segments and two such P regions. This protein (Duk1p) is a potassium channel because Xenopus oocytes injected with the corresponding RNA express potassium currents activated by depolarization that are not seen in control oocytes. Similar potassium currents were recorded from wildtype S. cerevisiae spheroplasts, but not from those in which the DUK1 locus had been disrupted. Cells carrying the duk1 delta 1::HIS disruption in addition to a chimeric gene comprising DUK1 behind the GAL1 promoter showed outward currents when grown in galactose, but not when grown in glucose. Additional sequences with the duplicate pore motif were found in C. elegans, suggesting that these proteins represent a novel structural family of potassium channel proteins.

A concatenated cDNA was made that comprised four contiguous copies of the rat brain potassium channel subunit Kv1.1. Currents were measured in oocytes that had been injected with in vitro transcribed RNA. A Pro residue was introduced by site-directed mutagenesis into the S4 transmembrane region of one of the four domains, at a position corresponding to that in which a Pro is found in voltage-dependent sodium and calcium channels. This substitution (L305P in any one domain) led to currents having a shallow activation curve, and an additional fast component of deactivation from strongly positive potentials. cDNAs with L305P substitution in two domains formed functional channels only if the domains were non-adjacent; the properties of the currents were similar to the wild-type concatemer. In both cases, 100-1000 x more RNA was injected to obtain maximal currents similar to those seen with the wild-type concantemer. The point mutation Y379V is known to reduce the blocking potency of extracellular tetraethylammonium; this residue from each of the four domains is exposed at the outer mouth of the pore because the progressive introduction of 1, 2, 3 and 4 such mutations causes progressive reductions in tetraethylammonium sensitivity. The Y379V mutation was introduced into concatenated cDNAs with two non-adjacent Pro-containing domains; the sensitivity to TEA of the resulting currents showed that the Pro-containing subunits did not contribute to the pore-forming part of the channel. The results suggest that channels can form with a Pro substitution in a single S4 domain, but they require strong depolarization to open and deactivate rapidly upon repolarization. With Pro substitutions in two domains, channels appear to be formed as a multimerized concatemer, in which the Pro-containing domains are excluded from pore formation.

The Slowpoke locus of Drosophila melanogaster encodes a family of alternatively spliced mRNAs which encode large conductance calcium-activated potassium channels. Variability residues in blocks of amino acids designated boxes A, C, E, G, and I. Oocytes were injected with cRNAs that had been chosen for direct functional comparison of single box differences. Single channel records from inside-out patches of oocyte membranes expressing A1 or A3 forms, E1 or E2 forms, and G2-G5 forms were analyzed and compared. The main functional difference between A1 and A3 was in unitary conductance, whereas the main difference in properties between E1 and E2 was in calcium sensitivity. Activation kinetics were different between G3 and G5, but not consistently in different A and E box backgrounds. The results indicate that alternative splicing of a common RNA precursor contributes to the functional diversity of the expressed channel. Our findings suggest that the variable region of the Slowpoke channel subunit comprises modular, yet interactive functional domains which influence the essential features of unit conductance, calcium sensitivity, and gating.

Calcium-dependent potassium (KCa) channels carry ionic currents that regulate important cellular functions. Like some other ion channels, KCa channels are modulated by protein phosphorylation. The recent cloning of complementary DNAs encoding Slo KCa channels has enabled KCa channel modulation to be investigated. We report here that protein phosphorylation modulates the activity of Drosophila Slo KCa channels expressed in Xenopus oocytes. Application of ATP-gamma S to detached membrane patches increases Slo channel activity by shifting channel voltage sensitivity. This modulation is blocked by a specific inhibitor of cyclic AMP-dependent protein kinase (PKA). Mutation of a single serine residue in the channel protein also blocks modulation by ATP-gamma S, demonstrating that phosphorylation of the Slo channel protein itself modulates channel activity. The results also indicate that KCa channels in oocyte membrane patches can be modulated by an endogenous PKA-like protein kinase which remains functionally associated with the channels in the detached patch.

Unitary currents were recorded from inside-out membrane patches pulled from Xenopus oocytes that had been injected with RNA transcribed from a cDNA encoding the Drosophila maxi-K channel (Slowpoke). Site-directed mutagenesis was used to make cDNAs encoding channel subunits with single amino acid substitutions (Y308V and C309P). The extracellular side of the patch was exposed to tetraethylammonium (TEA) in the pipette solution; unitary currents in the presence of TEA were compared with currents in the absence of TEA to compute the inhibition. Amplitude distributions were fit by beta functions to estimate the blocking and unblocking rate constants. For wild-type channels, TEA blocked with an apparent Kd of 80 microM at 0 mV and sensed 0.18 of the membrane electric field; the voltage dependence lay entirely in the blocking rate constant. TEA blocked currents through C309P channels with a similar affinity to wild-type at 0 mV, but this was not voltage-dependent. Currents through Y308V channels were very insensitive to any block by TEA; the apparent Kd at 0 mV was 26 mM and the blockade sensed 0.18 of the electric field. Oocytes injected with a mixture of RNAs encoding wild-type and Y308V channels showed unitary currents of four discrete amplitudes in the presence of 3 mM TEA; at 40 mV these corresponded to inhibitions of approximately 80%, 55%, 25% and 10%.(ABSTRACT TRUNCATED AT 250 WORDS)

Four copies of the coding sequence for a voltage-dependent potassium channel (RBK1, rat Kv1.1) were ligated contiguously and transcribed in vitro. The resulting RNA encodes four covalently linked subunit domains ([4]RBK1). Injection of this RNA into Xenopus oocytes resulted in the expression of voltage-dependent potassium currents. A single amino acid substitution, Tyr-->Val, located within the outer mouth of the pore, introduced into the equivalent position of any of the four domains, reduced affinity for external tetraethylammonium by approximately the same amount. In constructs containing 0, 1, 2, 3, or 4 Tyr residues the free energy of binding tetraethylammonium was linearly related to the number of Tyr residues. A different amino acid substitution, Leu-->Ile, located in the S4 region, was made in the equivalent position of one, two, three, or four domains. The depolarization required for channel activation increased approximately linearly with the number of Ile residues, whereas models of independent gating of each domain predict marked nonlinearity. Expression of this concatenated channel provides direct evidence that voltage-dependent potassium channels have four subunits positioned symmetrically around a central permeation pathway and that these subunits interact cooperatively during channel activation.

Whole-cell recordings were made from neurons of the rat locus coeruleus in a tissue slice removed from rat brain. Substance P caused an inward current in cells voltage-clamped at -60 mV. The effect of substance P was concentration-dependent (30 nM-3 microM) and was mimicked by similar concentrations of substance K and neuromedin K. The inward current resulted predominantly from an increase in membrane cation conductance; in potassium-free solutions it reversed polarity at about 12 mV. Substance P also reduced the conductance of an inwardly rectifying potassium current; this action was studied with low external sodium concentration. It is concluded that substance P excites rat locus coeruleus neurons by activating an intracellular transduction pathway leading to both cation conductance increase and potassium conductance decrease.

Calcium-activated potassium channels were expressed in Xenopus oocytes by injection of RNA transcribed in vitro from complementary DNAs derived from the slo locus of Drosophila melanogaster. Many cDNAs were found that encode closely related proteins of about 1200 aa. The predicted sequences of these proteins differ by the substitution of blocks of amino acids at five identified positions within the putative intracellular region between residues 327 and 797. Excised inside-out membrane patches showed potassium channel openings only with micromolar calcium present at the cytoplasmic side; activity increased steeply both with depolarization and with increasing calcium concentration. The single-channel conductance was 126 pS with symmetrical potassium concentrations. The mean open time of the channels was clearly different for channels having different substituent blocks of amino acids. The results suggest that alternative splicing gives rise to a large family of functionally diverse, calcium-activated potassium channels.

1. Voltage-clamp recordings of membrane current were made from Xenopus oocytes that had been injected with RNA which had been transcribed in vitro from a cloned complementary DNA. 2. Depolarization from -80 mV evoked outward potassium currents that developed very slowly. At -20 mV the time constant for activation was about 50 s, and at +40 mV about 6 s. 3. The potassium current was increased by the calcium ionophore A23187 or by intracellular injection of inositol 1,4,5-trisphosphate (IP3), each of which should increase the intracellular calcium concentration ([Ca2+]i). The current was decreased by injection of BAPTA (1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid). The current was also reduced by phorbol esters; this effect was blocked by staurosporine. 4. In oocytes that had also been injected with RNA encoding the 5-hydroxytryptamine (5-HT2) receptor, 5-HT increased the potassium current. After caffeine pretreatment, to block the release of intracellular calcium, 5-HT decreased the current; this decrease was prevented by staurosporine. 5. It is concluded that the slowly activating, voltage-dependent potassium current expressed in Xenopus oocytes is increased by increases in [Ca2+]i and is decreased by activation of protein kinase C. Stimulation of 5-HT2 receptors can have both these effects, but the former normally predominates.

A slowly activating potassium current was expressed in Xenopus oocytes by injection of RNA transcribed from a rat kidney cDNA clone. Hypotonic solutions (160 mOsmol/l; control was 220 mOsmol/l) increased the current by increasing the rate of activation and by decreasing the depolarization needed to activate the current. This effect of hypotonicity was not observed in calcium-free solution, but was unaffected by staurosporine or the calmodulin antagonist W7. Cytochalasin D reduced the current and prevented the increase by hypotonic solution. The results suggest that the increase in this potassium current by hypotonic solution might result from calcium entry and changes in the actin network.

A slowly activating, voltage-dependent potassium channel protein cloned from rat kidney was expressed in Xenopus oocytes. Two activators of protein kinase C, 1-oleoyl-2-acetyl-rac-glycerol and phorbol 12,13-didecanoate, inhibited the current. This inhibition was blocked by the kinase inhibitor staurosporine. Inhibition of the current was not seen in channels in which Ser103 was replaced by Ala, although other properties of the current were unchanged. These results indicate that inhibition of the potassium current results from direct phosphorylation of the channel subunit protein at Ser103.

RNAs encoding a wild-type (RBK1) and a mutant (RBK1(Y379V,V381T); RBK1*) subunit of voltage-dependent potassium channels were injected into Xenopus oocytes. When expressed separately, they made homotetrameric channels that differed about 100-fold in sensitivity to tetraethylammonium (TEA). Mixtures of channels having one, two, or three low affinity subunits were expressed by injecting various proportions of RBK1 and RBK1* RNAs. The affinity for TEA of these three channel species was deduced by fitting concentration-response curves for the inhibition of potassium currents. DNAs were also concatenated to construct a sequence that encoded two connected subunits, and channels that contained four, two, or no TEA-sensitive subunits were expressed. The results suggest that bound TEA interacts simultaneously with all four subunits.

Voltage-gated potassium channels play important functional roles in the development and maintenance of human lymphocyte functions. One such channel, known as the type n channel, has been well defined in human T cells and exhibits unique functional properties that distinguish it from other species of potassium channels. We report the characterization of a human genomic DNA clone, HGK5, encoding a 523-amino-acid potassium channel protein encoded by an open reading frame on a single exon. RNA transcribed in vitro from HGK5 genomic DNA directs expression of functional voltage-dependent potassium currents in Xenopus oocytes. The functional characteristics of the expressed channels are strikingly similar to those of the type n channel on human T lymphocytes. This, together with the presence of significant levels of HGK5 mRNA in human T lymphocytes, supports the notion that HGK5 encodes the human type n voltage-gated potassium channel. The effects of concanavalin A treatment on HGK5 mRNA levels in cultured human T lymphocytes was also examined. Mitogenic concentrations of concanavalin A induced a time-dependent decrease in HGK5 mRNA levels, suggesting that previously observed increases in potassium current density following concanavalin A treatment of human T lymphocytes are not due to increased transcriptional activity of the type n potassium channel gene.

1. Unitary potassium currents were recorded in outside-out patches of membrane from guinea-pig submucosal neurones. The actions of alpha 2-adrenoceptor agonists, somatostatin and [Met5]enkephalin were studied. 2. Three main groups of background potassium channels were active. At -70 mV with 160 mM-potassium on both sides of the membrane, they had conductances of 30-65 (small), 120-160 (intermediate) and 220-260 pS (large). 3. The open channel current-voltage relation showed only constant-field rectification. Extracellular barium (2 mM) and caesium (2 mM) decreased inward but not outward currents. Tetraethylammonium (10 mM) had no effect. 4. Noradrenaline, somatostatin and [Met5]enkephalin each increased the open probability of all three classes of channel when two or more unitary amplitude channels were active in the membrane patch. Agonists were ineffective when no channel, or a single channel, was discernible in the patch. Agonists did not cause the appearance of unitary currents distinct from those seen prior to their application. 5. The effect of the agonists required intracellular guanosine 5'-triphosphate. 6. The results show that the hyperpolarization of submucosal plexus neurones by noradrenaline, somatostatin and [Met5]enkephalin results from the increased opening of at least three types of background potassium channel, and that the coupling from the receptors to the channels is maintained in excised membrane patches.

alpha-Dendrotoxin (DTX) is a 60-amino acid peptide belonging to the family of mamba snake neurotoxins; it is a potent blocker of some but not all voltage-gated potassium currents. Potassium currents recorded from oocytes injected with cloned potassium channel RNAs also vary in sensitivity to DTX. Expression of channels that were chimeras of the DTX-sensitive channel RBK2 and the DTX-insensitive channel RGK5 showed that the putative extracellular loop between transmembrane domains S5 and S6 contributes strongly to DTX sensitivity. Mutation of two residues (Ala352Glu353) in this region of RBK1 to conform to those at equivalent positions in RGK5 (Pro374Ser375) reduced the potency of DTX about 70-fold, and the substitution of Tyr379 in RBK1 by its counterpart in RGK5 (His401) caused an additional 2.5-fold decrease in sensitivity. Converse substitutions in RGK5 significantly increased sensitivity to DTX. The results suggest that these residues contribute significantly to the channel-toxin interaction, providing further evidence that the S5-S6 loop lies at or near the external mouth of the channel, where DTX binding leads to channel occlusion. They offer a molecular explanation for the differences in DTX sensitivity observed among native potassium channels.

RGK5 is a rat genomic DNA clone that encodes the n-type potassium channel found in T-lymphocytes and other cells. Current through this channel declines (inactivates) over a period of hundreds of milliseconds during a maintained depolarizing pulse, whether in lymphocytes or when expressed in Xenopus oocytes. Here we demonstrate that an amino acid residue near the outer pore of the channel, histidine401, is involved in the inactivation process. Replacement of this residue by tyrosine, the amino acid found in the equivalent position of the homologous but non-inactivating channel RBK1, reduced inactivation of RGK5 over a 5 s depolarizing pulse from 84.3 +/- 1.9% to 18.3 +/- 1.1%. Conversely, replacement of this tyrosine in RBK1 (Tyr379) by histidine increased its inactivation from 21.6 +/- 1.1% to 42.3 +/- 1.5%. These results suggest a mechanism of channel inactivation distinct from that previously described for the A-type potassium channel.

Voltage-dependent K+ channels (RBK1, RBK2 and RGK5) were co-expressed in Xenopus oocytes with 5-hydroxytryptamine (5-HT2) receptors. K+ currents measured 2-4 days later were inhibited by 5-HT (100 nM-10 microM, 20-30 s application) by up to 90%. The effect of 5-HT was mimicked by intracellular injection of Ins(1,4,5)P3. Increasing the Ca2+ concentration at the inner surface of excised membrane patches did not decrease the K+ current.

Extracellular tetraethylammonium (TEA) inhibits currents in Xenopus oocytes that have been injected with mRNAs encoding voltage-dependent potassium channels. Concentration-response curves were used to measure the affinity of TEA; this differed up to 700-fold among channels RBK1 (KD 0.3 mM), RGK5 (KD 11 mM), and RBK2 (KD greater than 200 mM). Studies in which chimeric channels were expressed localized TEA binding to the putative extracellular loop between trans-membrane domains S5 and S6. Site-directed mutagenesis of residues in this region identified the residue Tyr379 of RBK1 as a crucial determinant of TEA sensitivity; substitution of Tyr in the equivalent positions of RBK2 (Val381) and RGK5 (His401) made these channels as sensitive to TEA as RBK1. Nonionic forces are involved in TEA binding because (i) substitution of the Phe for Tyr379 in RBK1 increased its affinity, (ii) protonation of His401 in RGK5 selectively reduced its affinity, and (iii) the affinity of TEA was unaffected by changes in ionic strength. The results suggest an explanation for the marked differences in TEA sensitivity that have been observed among naturally occurring and cloned potassium channels and indicate that the amino acid corresponding to residue 379 in RBK1 lies within the external mouth of the ion channel.

Voltage-dependent potassium currents were measured in Xenopus oocytes previously injected with RNAs generated in vitro from each of three cloned cDNAs (RBK1, RBK2, and RGK5). The currents differed in their sensitivities to blockade by tetraethylammonium (TEA; respective KDs 0.3, greater than 100, and 10 mM) and in their inactivation during a depolarizing pulse. Injections of RNA combinations (RBK1/RBK2 and RBK1/RGK5) caused currents that had TEA sensitivities different from those expected from the sum, in any proportion, of the two native channels. It is concluded that novel potassium channels are formed by the oocytes injected with two RNAs, presumably by heteropolymerization of subunits; such heteropolymerization would contribute functional diversity to voltage-dependent potassium channels in addition to that provided by a large gene family.

1. Intracellular recordings were made from neurones in the nucleus accumbens in slices from the rat brain maintained in vitro. 2. 5-Hydroxytryptamine (5-HT.1-100 microM) depolarized 170 of 203 (84%) neurones and caused them to discharge action potentials. The depolarization was associated with an increase in the input resistance, and was reversed in polarity by conditioning hyperpolarization; this reversal potential was linearly related to the logarithm of the extracellular potassium concentration. 3. Application of 5-HT to neurones voltage-clamped near their resting potential (typically about -80 mV) caused an inward current and a decrease in the slope conductance. The current caused by 5-HT reversed polarity at the potassium equilibrium potential. Analysis with an equivalent circuit model of the neurone at steady state indicated that 5-HT selectively reduced the inward rectifier potassium conductance. 4. The depolarization caused by 5-HT persisted in tetrodotoxin (1 microM). It was reduced but not abolished by a solution that contained lower levels of calcium (0.24 instead of 2.4 mM), higher levels of magnesium (5 instead of 1.2 mM), and cobalt (2 mM). 5. The depolarization caused by 5-HT was competitively antagonized by the 5-HT2 antagonists ketanserin and mianserin with dissociation equilibrium constants of 3 and 45 nM respectively: spiperone (300 nM) also blocked the action of 5-HT. The depolarization was not mimicked or blocked by a number of other agonists and antagonists selective for the 5-HT1 and 5-HT3 receptor types.

1. Intracellular recordings of membrane potential and current were made from neurones in the lateral parabrachial nucleus in slices of rat brain in vitro. 2. The membrane was hyperpolarized by the opioid peptides Tyr-D-Ala-Gly-MePhe-Gly-ol (DAGOL, 0.01-1 microM) and [Met5]enkephalin (3-30 microM), though not by Tyr-D-Pen-Gly-Phe-D-Pen and U50488. In two experiments, naloxone competitively antagonized the effects of DAGOL and [Met]enkephalin with equilibrium dissociation constants of 0.8 and 3.2 nM, respectively. 3. Baclofen (0.3-30 microM) also hyperpolarized the neurones; this action was unaffected by naloxone. 4. DAGOL, [Met5]enkephalin and baclofen caused outward currents at the resting potential. These currents reversed polarity at a membrane potential which changed with the logarithm of the extracellular potassium concentration. 5. Muscarine has been shown previously to increase the potassium conductance by an action at M2-receptors: the potassium currents induced by maximal concentrations of muscarine, baclofen and [Met5]enkephalin were non-additive, indicating that these agonists opened the same population of potassium channels. 6. Noradrenaline, UK14304, carboxamidotryptamine, dopamine, adenosine and somatostatin had little or no effect on membrane potential. 7. It is concluded that rat lateral parabrachial neurones express mu-opioid, gamma-aminobutyric acidB (GABAB), and M2-muscarinic receptors: activation of any of these receptors increases the potassium conductance of the membrane and inhibits the neurones through hyperpolarization.

Potassium channels are ubiquitous membrane proteins with essential roles in nervous tissue, but little is known about the relation between their function and their molecular structure. A complementary DNA library was made from rat hippocampus, and a complementary DNA clone (RBK-1) was isolated. The predicted sequence of the 495-amino acid protein is homologous to potassium channel proteins encoded by the Shaker locus of Drosophila and differs by only three amino acids from the expected product of a mouse clone MBK-1. Messenger RNA transcribed from RBK-1 in vitro directed the expression of potassium channels when it was injected into Xenopus oocytes. The potassium current through the expressed channels resembles both the transient (or A) and the delayed rectifier currents reported in mammalian neurons and is sensitive to both 4-aminopyridine and tetraethylammonium.

Currents through single-ion channels were recorded in the cell-attached configuration from locus ceruleus neurons enzymatically dissociated from newborn rats. When the selective mu opioid receptor agonist Tyr-D-Ala-Gly-MePhe-Gly-ol was in the patch-clamp electrode, unitary inward currents were observed with conductance of approximately 45 pS (measured at zero pipette potential, with 150 mM potassium in the recording electrode). Long silences, lasting many seconds to minutes, separated periods of activity of similar durations. Within such activity periods the distribution of closed times of the channels was best fitted by the sum of two exponential functions (time constants approximately 1 and 30 ms), and the durations of channel openings were fit by a single exponential function; mean open time increased from 2 to 120 ms as agonist concentration increased. Channel activity was not seen when high concentrations of opioids were applied to the neuron outside the patch-clamp recording electrode, indicating intimate coupling between receptor and potassium channel. Unitary currents with similar properties were also seen when pipettes contained alpha 2 adrenoceptor agonists or somatostatin. Taken with previous findings, the results indicate that mu opioid receptors, alpha 2 adrenoceptors, and somatostatin receptors can couple directly to membrane potassium channels through the local intermediary action of a GTP binding protein.