Proc. Natl. Acad. Sci. USA Vol. 83, pp. 942-946, February 1986 Biochemistry Bradykinin-activated transmembrane signals are coupled via N, or N; to production of inositol 1,4,5-trisphosphate, a second messenger in NG108-15 neuroblastoma—glioma hybrid cells (GTP-binding proteins / pertussis toxin /adenylate cyclase/ion channels /phosphatidylinositol turnover) HARUHIRO HIGASHIDA*, RICHARD A. STREATY!, WERNER KLEE‘, AND MARSHALL NIRENBERG* *Laboratory of Biochemical Genetics, National Heart, Lung and Blood Institute and +Laboratory of Molecular Biology, National Institute of Mental Health, National Institutes of Health, Bethesda, MD 20892 Contributed by Marshall Nirenberg, October 7, 1985 ABSTRACT The addition of bradykinin to NG108-15 cells results in a transient hyperpolarization followed by prolonged cell depolarization. Injection of inositol 1,4,5-trisphosphate or Ca?* into the cytoplasm of NG108-15 cells also elicits cell hyperpolarization followed by depolarization. Tetraethylam- monium ions inhibit the hyperpolarizing response of cells to bradykinin or inositol 1,4,5-trisphosphate. Thus, the hyper- polarizing phase of the cell response may be due to inositol 1,4,5-trisphosphate-dependent release of stored Ca** into the cytoplasm, which activates Ca’*+-dependent K* channels. The depolarizing phase of the cell response to bradykinin is due largely to inhibition of M channels, thereby decreasing the rate of K* efflux from cells and, to a lesser extent, to activation of Ca”*-dependent ion channels and Ca?* channels. In contrast, injection of inositol 1,4,5-trisphosphate or Ca”* into the cytosol did not alter M channel activity. Incubation of NG108-15 cells with pertussis toxin inhibits bradykinin-dependent cell hyperpolarization and depolarization. Bradykinin stimulates low K,, GTPase activity and inhibits adenylate cyclase in NG108-15 membrane preparations but not in membranes prepared from cells treated with pertussis toxin. Reconstitution of NG108-15 membranes from cells treated with pertussis toxin with nanomolar concentrations of a mixture of highly purified N, and N; [guanine nucleotide-binding proteins that have no known function (N,) or inhibit adenylate cyclase (N,)] restores bradykinin-dependent activation of GTPase and inhibition of adenylate cyclase. These results show that [bradykinin-recep- tor] complexes interact with N, or Nj, and suggest that N, and/or N; mediate the transduction of signals from bradykinin receptors to phospholipase C and adenylate cyclase. Bradykinin (BK) is a nonapeptide derived by proteolysis of higher molecular weight precursor proteins synthesized in the liver and hypothalamus (1, 2). The pharmacological actions of BK or precursor proteins include muscle contrac- tion, hypertension, pain generation, increase in vascular permeability, and blood coagulation 3-5). However, rela- tively little is known about the function of BK in the nervous system. NG108-15 neuroblastoma—glioma hybrid cells have BK receptors (6, 7) and respond to BK by a transient cell hyperpolarization followed by a long-lived depolarization (6, 8) accompanied by an increase in secretion of acetylcholine from NG108-15 cells at synapses with myotubes (8). BK also stimulates phosphatidylinositol turnover in NG108-15 cells (8) and increases cellular levels of inositol 1,4,5-trisphosphate (InsP;) (9) and the release of Ca2* from intracellular stores into the cytoplasm (10). The 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. 942 In this report we show that pertussis toxin partially inhibits the effects of BK on NG108-15 cells and that activation of BK receptors by BK results in activation of GTPase of Nj (11) or N, (12) [guanine nucleotide-binding proteins that have no known function (N,) or inhibit adenylate cyclase (Nj)], which are substrates for pertussis toxin (13, 14). The results suggest that N; or N, are involved in signal transduction initiated by [BK-receptor] complexes that result in activation of phos- pholipase C and inhibition of adenylate cyclase. METHODS AND MATERIALS BK was obtained from Sigma or Calbiochem; InsP; and inositol 2-phosphate were obtained from Sigma; [p- Ala”,Met*]enkephalinamide was from Calbiochem. Pertussis toxin was a gift from R. Sekura (National Institute of Child Health). “CaCl, was obtained from Amersham, and [y- 32P|GTP was from New England Nuclear. NG108-15 cells used for electrophysiological studies were cultured in polyornithine-coated 35-mm Petri dishes and were treated with 10 uM prostaglandin E, and 1 mM theophylline for 1-3 weeks before use (15). RESULTS Cell Responses to BK, InsP3, or Ca?*+. Addition of BK by iontophoresis to the external surface of an NG108-15 cell resulted in cell membrane hyperpolarization followed by depolarization (Fig. 1A) (8). Injection of InsP; (Fig. 1B) or Ca** (Fig. 1D) into the cytoplasm of a cell also resulted in cell hyperpolarization followed by depolarization. Injection of inositol 2-phosphate into the cytoplasm had little or no effect on the cell membrane potential (Fig. 1C). These results suggest that the responses of NG108-15 cells to BK are dependent on an increase in cellular levels of InsP3, presum- ably due to activation of the phospholipase C that catalyzes the formation of InsP; and diacylglycerol from phosphati- dylinositol 4,5-bisphosphate followed by an InsP3-dependent release of stored Ca”+ into the cytoplasm. The relation between the amount of iontophoretic current used for intracellular InsP, injection and the magnitude of the hyperpolarizing and depolarizing cell responses is shown in Fig. 2A. Both responses of cells increased as the quantity of InsP; iontophoretic currents (A:sec) was increased up to about 65 nC. With some cells, injection of InsP; with relatively high amounts of iontophoretic current, such as —220 nC, elicited a depolarizing cell response but little or no hyperpolarizing response (Fig. 2 B and C). The mean hyper- Abbreviations: BK, bradykinin; InsP3, inositol 1,4,5-trisphosphate; Et,NCI, tetraethylammonium chloride; N, or N;, guanine nucleotide- binding proteins that couple some species of activated receptors with adenylate cyclase stimulation or inhibition, respectively; N,, an N protein of unknown function. Biochemistry: Higashida e¢ al. Membrane potential, mV l i ] l j 20 40 60 80 =: 100 —20 Time, sec Fic. 1. Typical membrane potential changes of NG108-15 cells evoked by extracellular BK or intracellular injection of InsP; or Ca?*. (A) Cell hyperpolarization followed by depolarization elicited by extracellular application of BK. The potential changes were recorded with an intracellular microelectrode filled with 3 M KCI. A second micropipette filled with 0.1 mM BK dissolved in 0.1 mM HCl] was located extracellularly close to the cell surface. BK was applied at zero time by iontophoresis (50 nA for 1 sec). (B) Cell hyperpolar- izing and depolarizing responses induced by intracellular injection of InsP;. The recording electrode (3 M KCl) and a micropipette filled with 1 mM InsP, dissolved in H,O were inserted into an NG108-15 cell. At zero time —100 nA was passed through the InsP; pipette for 0.5 sec. (C) One mM inositol 2-phosphate dissolved in H,O was injected intracellularly (—100 nA for 1 sec). (D) Ca?* was injected into the cytoplasm of an NG108-15 cell from a micropipette filled with a solution containing 0.5 M CaCl, by iontophoresis with 100 nA for 1 sec at zero time. The upward and/or downward deflections of the traces between 0 and 2 sec are due to the iontophoretic current rather than the compound applied. polarizing and depolarizing responses of many cells to injected InsP; are shown in Fig. 2C as a function of the amount of iontophoretic current applied. Tetraethylammonium ions (Et,N*) completely blocked BK- or InsP3-dependent cell hyperpolarization but had little or no effect on cell depolarization (Fig. 3A and C). EtyN* is known to inhibit some Ca?*-dependent K* channels (16), which suggests that BK- or InsP3-dependent hyperpolariza- tion may be due to activation of K* channels by Ca**, thereby increasing the rate of K* efflux from the cells. In contrast, Co*t had little effect on the hyperpolarizing re- sponse of NG108-15 cells to BK or InsP3 but inhibited most of the BK- or InsP3-dependent cell depolarization (Figs. 3 B and D, respectively). When NG108-15 cells were incubated without Ca** and with or without 0.1 mM EGTA, the amplitudes of BK-dependent hyperpolarizations and depo- larizations were reduced greatly (not shown). Incubation of cells in the absence of Na* or in the presence of 1 uM tetrodotoxin had little or no effect on cell responses to BK. The Ca?* channel blocker nifedipine (0.1 pM) had no effect on BK- or InsP3-dependent hyperpolarizing or depolarizing responses of NG108-15 cells (not shown). The mean input membrane resistance, determined from the amplitudes of hyperpolarizations induced by repetitive or constant currents applied through the intracellular recording electrode, was reduced to 42 + 9% (+ SEM, n = 5) of the mean control value (12.9 + 1.1 MQ, n = 8) during the hyperpolarizing phase of the responses of NG108-15 cells to BK (Fig. 4A). The membrane resistance increased to 131 + 10% (16.8 MQ) of the control value during the BK-dependent depolarization phase. BK-dependent increases in membrane resistance during the depolarizing phase were found over a Proc. Natl. Acad. Sci. USA 83 (1986) 943 T T T T T T T TT" $e tf ao) ) ~ —100 —200 Membrane potential, mV Time, sec Fic.2, The relation between the amount of iontophoretic current used for intracellular InsP; injection and the magnitudes of the hyperpolarizing and depolarizing cell responses. (A) Typical exam- ples of hyperpolarizing and depolarizing responses of NG108-15 cells evoked by intracellular injection of InsP; for 2 sec; the amount of current in nC is indicated in each panel. (B) An example of InsP; intracellular injection by iontophoresis using a large amount of current (—220 nC), which elicited primarily cell depolarization. The discontinuities in the trace at approximately 10~15 sec suggest some cell hyperpolarization. At 30-60 sec passive hyperpolarizing cur- rents were injected to test input membrane resistance. (C) The mean changes in membrane potential during the hyperpolarizing and depolarizing phases of NG108-15 cell responses to intracellular injection of InsP; with different amounts of iontophoretic current (nC). Each point represents the mean of values obtained from 5-19 cells; each cell was used for only one determination. The hyper- polarization and depolarization values shown are the maximum membrane potential deflections at 5-20 and 30—45 sec, respectively, after injection of InsP3. wide range of membrane potentials (—80 to —10 mV) (not shown). These results suggest that the addition of BK to NG108-15 cells results in the closure of some ion channels that are open in the absence of BK. In contrast, the average membrane resistance values during the hyperpolarizing and depolarizing phases of the cell response to injection of InsP; decreased to 40 + 6% (n = 6) and 58 + 7% (n = 18), respectively, of the control value (27.1 + 2.5 MQ, n = 14) (Fig. 4 B and C). The reversal potential for hyperpolarization induced by extracellular application of BK or intracellular injection of Ca2+ was —70 to —80 mV, which is close to the K* equilibrium potential and suggests that the hyperpolar- izing phase is due to the activation of Ca?*-dependent K* channels, thereby increasing the rate of K* efflux from cells. BK elicited cell hyperpolarization after a delay of 1-2 sec. The delay in cell response was shorter when InsP; was injected into the cytosol, and no delay in response was detected when Ca?* was injected. The decreasing delay in cell responses to BK, InsP3, or Ca2* suggests the following sequential reactions; BK receptor occupancy results in activation of phospholipase C; phosphatidylinositol 4,5- bisphosphate is converted to InsP; and diacylglycerol; InsP; induces release of stored Ca?* into the cytosol; Ca?*-depen- dent K* channels are opened; and K* efflux hyperpolarizes the cell. Addition of 10-1,000 nM phorbol 12-tetradecanoate 13- acetate or 1-oleoyl-2-acetylglycerol had no effect on the resting membrane potential or on cell responses to BK 0-60 min after addition (not shown). Also, 10 4M A23187 did not mimic the effects of BK on cell membrane potential in control or phorbol! ester-treated cells (not shown). 944 Biochemistry: Higashida et al. TAT ELN" 7“ Te LAAT. rc tt bit do ry PD oe T it L cot J 59 —20 O 20 40 60 80 100 120 Time, sec Fic, 3. Effects of Et,N* (A and C) or Co?* (B and D) on the responses of NG108-15 cells to extracellular BK (A and B) or intracellular InsP; (C and D). (A) NG108-15 cells were incubated for 20-30 min in a 35-mm Petri dish containing 2 ml of Dulbecco’s modified Eagle’s medium supplemented with 5 mM Et,N*. At zero time 2 ul of a solution containing 10 »M BK dissolved in 150 mM NaCl was added to the surface of the medium close to the cell being tested. Action potentials were evoked by passing depolarizing pulses of current (0.2 nA for 60 msec) through the intracellular recording microelectrode. (B) NG108-15 cells were incubated in 2 mM Co?* /20 mM Tris-HCl, pH 7.2/150 mM NaCl/5.4 mM KC1/0.8 mM Mg- Cl,/1.8 mM CaCl,/20 mM glucose for approximately 10-20 min. Hyperpolarizing currents (0.5 nA for 60 msec) were passed through the intracellular recording microelectrode but failed to evoke off- spikes. BK was applied at zero time in a 2-yl drop as described above. (C) NG108-15 cells were incubated in the presence of 5 mM Et,Nt as described above. InsP; (0.1 mM) was injected into the cytoplasm by iontophoresis (—100 nA for 1 sec). (D) Cells were incubated in the presence of 2 mM Co** in the medium as described above for 10-20 min. InsP; was injected intracellularly at zero time (—200 nA for 1 sec). BK-Dependent Efflux of “Ca?+. As reported (8), BK stimulates “Ca?+ influx into NG108-15 cells. The effect of BK on “Ca?* efflux from NG108-15 cells is shown in Fig. 5. Cells were incubated in the presence of 1.8 mM “CaCl, for 8 min to promote *Ca?* uptake and washed for 3 min, and then dishes were perfused with control medium or medium with BK. BK stimulated the rate of ®Ca?*+ efflux approxi- mately 3-fold during the first 20 sec, but by 80 sec the rate of 45Ca?* efflux had returned to the control value. Effects of Pertussis Toxin on Cell Responses to BK. Both the hyperpolarizing and depolarizing responses of NG108-15 cells to BK were inhibited by treatment of the cells with pertussis toxin (200 ng per ml, 15 hr) (Fig. 6). The maximum inhibition by pertussis toxin was obtained with 1 ~M BK; higher concentrations of BK resulted in less inhibition by pertussis toxin. The maximum inhibition of hyperpolariza- tion (72%) was obtained with 1000 ng of pertussis toxin per ml, the highest concentration tested; 50% of the maximum observed inhibition was obtained with 35-100 ng of pertussis toxin per ml of medium. These results suggest that GTP- binding regulatory proteins such as Nj; (13, 14) or N, (12) mediate the hyperpolarizing cell response to BK. The results also are consistent with the suggestion that pertussis toxin acts primarily to decrease receptor affinity for ligands (17). Because all known GTP-binding regulatory proteins also display hormone-stimulated GTPase activities, we measured the effect of BK on GTPase activity of NG108-15 cell membranes. The data shown in Table 1 demonstrate that BK or the opioid peptide [D-Ala?,Met*]enkephalinamide stimu- lated the activity of a low K,, GTPase in NG108-15 mem- branes and that the stimulatory effects of BK or [p- Proc. Natl. Acad. Sci. USA 83 (1986) TI00 nAT T T T T A 5 sec Membrane potential, mV | | Time, sec Fic. 4. Changes in membrane resistance elicited by BK iontophoresis (A), InsP; injection (B), or Ca** injection (C). (A) Membrane resistance was monitored by passing repetitive hyperpo- larizing pulses of current (0.5 nA for 400 msec) through the intra- cellular recording microelectrode. BK was applied extracellularly by iontophoresis (100 nA for 5 sec). The amplitudes of the hyperpolar- izing pulses are measures of membrane resistance. (B) InsP; was injected intracellularly by iontophoresis (—100 nA for 400 msec). Membrane resistance was measured by passing hyperpolarizing pulses (0.2 nA for 200 msec) through the intracellular recording microelectrode. (C) Ca?* was injected intracellularly by iontopho- resis (100 nA for 120 msec). Repetitive hyperpolarizing pulses of current (0.2 nA for 300 msec) were passed through the recording intracellular microelectrode. Ala’,Met*}Jenkephalinamide were inhibited by treating the cells with pertussis toxin before the membranes were pre- pared. BK-dependent stimulation of low K,, GTPase activity was restored to membranes of pertussis toxin-treated 160-77 ToT 0 140 4 120;- 4 e 8 Qo oS T T 1 1 Ca** released, nmol per dish na Ss T 1 40 | I I ! 1 1 I I | -60 —-40-20 0 20 40 60 80 100 120 140 Time, sec Fic. $. BK-stimulated “Ca?* efflux from NG108-15 cells. The cells were incubated in the medium described in the legend to Fig. 3B without Co?* for 2 min and then for 8 min at room temperature in medium supplemented with 4 «Ci (1 Ci = 37 GBq) of *CaCl,, washed three times with medium without *°Ca?*, and then perfused for 2.9 min. At zero time cells were perfused with medium that contained 1 uM BK (e) or medium without BK (0). Fractions (1 ml) were collected at 20-sec intervals. Each point represents the mean of 3-4 values + SEM. Each Petri dish contained approximately 5 x 10° cells. Biochemistry: Higashida et al. 10 5 o an. -10 Membrane potential, mV I a -15 2 i i ! i 5 7 6 5 4 3 Bradykinin, — log M Fic. 6. The effects of pertussis toxin on the hyperpolarizing (@, ©) and depolarizing (@, 0) responses of NG108-15 cells grown in dishes that were not coated with polyornithine. Droplets (2 1 each) of a solution containing the indicated concentration of BK dissolved in 150 mM NaCl were applied to the surface of the medium near the cell that was being tested. Each point represents the mean value of responses of nine cells. Cells were incubated for 15 hr with 0.34 mM ammonium sulfate (control cells) or with 200 ng of pertussis toxin/ml of medium and 0.34 mM ammonium sulfate before being tested. o, Hyperpolarizing responses of control cells to BK; e, BK-dependent hyperpolarizing responses of NG108-15 cells that had been treated with pertussis toxin; 0, depolarizing responses of control cells to BK; mg, BK-dependent depolarizing responses of cells that had been treated with pertussis toxin. NG108-15 cells by the addition of a mixture of bovine brain N, and N; (approximately 70% N, and 30% Nj) estimated to be >95% pure (19) (Fig. 7A); half-maximal stimulation of GTPase by BK was obtained with <1 nM N,/Nj. As reported earlier (19) and as shown in Fig. 7A, the N,/N; mixture also reconstituted a [D-Ala?,Met°]enkephalinamide-dependent, opiate receptor-mediated stimulation of GTPase. Interesting- ly, BK-stimulated GTPase activity was reconstituted at appreciably lower concentrations of the N,/Nj mixture than those needed for opiate-stimulated GTPase. We found also that BK stimulation of GTPase activity was diminished only slightly in the presence of a saturating concentration (10 4M) of [p-Ala?,Met*]enkephalinamide (Fig. 7B). Such additivity of effects suggests that BK receptors may interact preferen- tially with Ny, and opiate receptors may interact preferen- tially with N;. Additivity of the effects of BK and opiates is confirmed by the observation that BK-dependent inhibition of adenylate cyclase still was observed in the presence of saturating concentrations of opiates (Fig. 7C). Ligands for other species of NG108-15 receptors that mediate inhibition of adenylate cyclase, such as norepinephrine, which acti- vates a-adrenergic receptors (21), or somatostatin (22), no Table 1. Attenuation of peptide-stimulated GTPase by pertussis toxin P; formed, pmol/min per mg of protein Pertussis Control toxin-treated Addition cells cells Water 26.0 + 1.5 19.9 + 1.6 BK (10 4M) 32.6 + 1.8 18.9 + 1.8 [p-Ala?,Met*]- enkephalinamide (10 4M) 38.0 + 2.3 15.7 + 1.6 Attenuation of peptide-stimulated low K,, GTPase activity by pertussis toxin. Membranes prepared from NG108-15 cells were incubated with or without 20 ng of pertussis toxin per ml of medium for 22 hr. Low K,, GTPase activity was assayed as described (18). Proc. Natl. Acad. Sci. USA 83 (1986) 945 < 36 B TT 6 C ne 86 3 34F PS) 4782.5 oS fo \ E E 3210 0 {Bs a i) af oo go 30 7 44 ES 5 ‘eo a = 285 ON 9.06/70 se Bu . 2o oy 26 < 0 966 2 as 24 Pix OW 162 & = E a OO A 8 > 2r Of 453 = > z 20k , . S Cad 54 1 1 4 4 1 1 J 1 i i % 2.5 5 7.5 1012.5 By -7 -6 -5 -4 -7 -6 -5 =4° N protein, nM BK, log M Fic. 7. (A) Reconstitution of BK-stimulated (@) or opiate- stimulated (A) low K,, GTPase activities of membranes prepared from pertussis toxin-treated NG108-15 cells. Membranes were incu- bated with the indicated concentrations of a mixture of bovine brain N; and N, estimated to be >95% pure for 15 min at 30°C in the standard assay mixture (18) without radioactive GTP and then were incubated for an additional 15 min in the presence of approximately 60,000 cpm (1 4M) [y-*P]GTP. *2P; released in the absence of BK or [p-Ala?,Met*}Jenkephalinamide has been subtracted. Mean values (three experiments) are shown. (B) Low K,, GTPase activity of membranes prepared from untreated NG108-15 cells is shown as a function of BK concentration. o, BK with 10 wM [p-Ala?,Met*]en- kephalinamide; @, BK without [D-Ala?,Met*]enkephalinamide. The ordinate scale is pmols of P; released per min/mg of protein. (C) Adenylate cyclase activity of membranes prepared from untreated NG108-15 cells, assayed as described (20), is shown as a function of BK concentration in the absence (@) or presence (©) of 10 uM [p-Ala?,Met*]Jenkephalinamide. longer inhibit adenylate cyclase or stimulate low K,, GTPase in the presence of saturating concentrations of opiates. Thus, the mechanism of coupling BK receptors to inhibition of adenylate cyclase differs from the mechanism coupling a2- receptors or somatostatin receptors to inhibition of adenylate cyclase. DISCUSSION Exposure of NG108-15 cells to BK elicits increases in InsP; and diacylglycerol levels (8); in turn, InsP3 stimulates the release of stored Ca?‘ into the cytoplasm (10), thereby activating Ca*+-dependent K* channels and increasing the rate of K* efflux. Cell hyperpolarization results from the increase in K* efflux. The observations that support these conclusions are: (i) injection of InsP; or Ca?* into the cytoplasm of NG108-15 cells results in cell hyperpolarization followed by depolar- ization; (ii) hyperpolarization of cells elicited by BK, InsP3, or Ca?* is accompanied by a decrease in membrane resist- ance—i.e., an increase in cell permeability to ions; (iii), the reversal potential for BK-dependent hyperpolarization is approximately —80 mV, which is close to the equilibrium potential for K*; (iv) the amplitudes of the hyperpolarizing responses of 108CC-25 (23) and NG108-15 cells (unpublished results) to BK are dependent on the concentration of extra- cellular K*; and (v) BK-dependent hyperpolarization is blocked by EtsN*, which is known to inhibit some Ca?*- dependent K* channels (16). The long-lasting depolarization evoked by BK is associated with an increase in membrane resistance—i.e., cell perme- ability to ions decreases. Other species of receptors have been shown to mediate cell depolarization by decreasing cell permeability, such as muscarinic acetylcholine receptors (24) and receptors for luteinizing hormone-releasing hormone (25—27), substance P (28, 29), thyrotropin-releasing hor- mone (30, 31), neurotensin (32), somatostatin (33), angioten- sin (34), prostaglandin D2 (35), and the small cardioactive 946 Biochemistry: Higashida et al. peptide (36). Cell depolarization mediated by muscarinic acetylcholine receptors has been attributed to inhibition of M channels for K*, which decreases K* efflux from cells (24). Recent results obtained with voltage-clamp conditions show that NG108-15 cells have M channels that are inhibited by BK (Higashida and Brown, unpublished data). In contrast, injection of InsP; or Ca?* into the cytoplasm of NG108-15 cells results in cell hyperpolarization followed by a long-lasting depolarization that is accompanied by a decrease in membrane resistance—i.e., an increase in cell permeability to ions. Injection of InsP; or Ca2* into the cytoplasm did not affect the activity of M channels; therefore, elevation of cytosolic levels of InsP; or Ca2* elicited most, but not all, of the cell responses to extracellular BK. Because BK stimulates the formation of InsP3 in NG108-15 cells, part of the BK-induced depolarization probably also is due to an increase in cell permeability to ions. The most likely candi- dates for the ion channels that are activated during the depolarizing phase are the Ca?*-dependent cation channel (37) and a Ca?* channel. We find that BK-dependent cell hyperpolarization and depolarization are inhibited by pertussis toxin, which sug- gests that a GTP-binding regulatory protein such as N, or N; plays a role in the signal transduction process. BK was shown to stimulate a low K,, GTPase and to inhibit adenylate cyclase in NG108-15 membrane preparations, and these effects of BK were blocked by pertussis toxin. Reconstitution of mem- branes from pertussis toxin-treated NG108-15 cells with nanomolar concentrations of a mixture of highly purified bovine brain N, and Nj; restored BK-dependent activation of low K,, GTPase and inhibition of adenylate cyclase. BK- and [D-Ala*, Met>Jenkephalinamide-dependent inhibitions of ade- nylate cyclase were additive, whereas norepinephrine- and [D-Ala”,Met*]enkephalinamide-dependent inhibitions were not additive, which suggest that BK and opiates inhibit adenylate cyclase by different mechanisms. Because the [By] subunits of the N proteins are functionally interchangeable (38), activation of any N protein, with dissociation of [a-B-y] complexes to a and [6-y] will result in inhibition of adenylate cyclase by mass action, since the released [f-y] subunits will combine with the free a subunit of activated N,, the N protein that stimulates adenylate cyclase. In addition, our results suggest that N, or Nj are involved in the transduction of signals from BK receptors to Ca?* mobilization, presumably via activation of phospholipase C, as has been suggested for several other species of receptors (39-42). 1. Takagaki, Y., Kitamura, N. & Nakanishi, S. (1985) J. Biol. Chem. 260, 8601-8609. 2. Perry, D.C. & Snyder, S.H. (1984) J. Neurochem. 43, 1072-1080. Regoli, D. & Barabe, J. (1980) Pharmacol. Rev. 32, 1-46. Correa, F. M. A. & Graeff, F. G. (1975) J. Pharmacol. Exp. Ther. 192, 670-676. 5. Rocha e Silva, M. (1976) in Chemistry and Biology of the Kallikrein-Kinin System in Health and Disease, Publ. No. 76-791, Ed. Pisano, J.J. & Austen, K. F, (Department of Health, Education and Welfare, Washington, DC), pp. 7-35. 6. Higashida, H., Wilson, S. P., Adler, M. & Nirenberg, M. (1978) Soc. Neurosci. Abstr. 4, 591, AY 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22, 23. 24. 25. 26. 27. 28. 29, 30. 31. 32. 33. 34, 35. 36. 37. 38. 39, 4l. 42. Proc. Natl. Acad. Sci. USA 83 (1986) Reiser, G., Walter, U. & Hamprecht, B. (1984) Brain Res. 290, 367-371. Yano, K., Higashida, H., Inoue, R. & Nozawa, Y. (1984) J. Biol. Chem. 259, 10201-10207. Yano, K., Higashida, H., Hattori, H. & Nozawa, Y. (1985) FEBS Lett. 181, 403-406. Ohsugi, T., Uchida, S. & Yoshida, H. (1985) Folia Pharmacol. Jpn., in press. Katada, T. & Ui, M. (1982) Proc. Natl. Acad. Sci. USA 79, 3129-3133. Sternweis, P. C. & Robinshaw, J. D. (1984) J. Biol. Chem. 259, 13806-13813. Burns, D. L., Hewlett, E. L., Moss, J. & Vaughan, M. (1983) J. Biol. Chem. 258, 1435-1438. Kurose, H., Katada, T., Amano, T. & Ui, M. (1983) J. Biol. Chem. 258, 4870-4875. Nirenberg, M., Wilson, S., Higashida, H., Rotter, A., Krueger, K., Busis, N., Ray, R., Kenimer, J. G. & Adler, M. (1983) Science 222, 794-799. Findlay, I., Dunne, M. J., Ullrich, S., Wollheim, C. B. & Petersen, O. H. (1985) FEBS Lett. 185, 4-8. Hsia, J. A., Moss, J., Hewlett, E. L. & Vaughan, M. (1984) J. Biol. Chem. 259, 1086-1090. Koski, G. & Klee, W. A. (1981) Proc. Natl. Acad. Sci. USA 78, 4185-4189. Milligan, G. & Klee, W. (1985) J. Biol. Chem. 260, 2057-2063. Sharma, S., Nirenberg, M. & Klee, W. A. (1975) Proc. Natl. Acad. Sci. USA 72, 590-594. Sabol, S. L. & Nirenberg, M. (1979) J. Biol. Chem. 254, 1913-1920. Traber, J., Glaser, T., Brandt, M., Klebensberger, W. & Hamprecht, B. (1977) FEBS Lett. 81, 351-354. Reiser, G. & Hamprecht, B. (1982) Brain Res. 239, 191-199, Adams, P. R., Brown, D. A. & Constanti, A. (1982) J. Physiol. (London) 332, 223-262. Jan, L. Y. & Jan, Y.N. (1982) J. Physiol. (London) 327, 219-246. Katayama, Y. & Nishi, S. (1982) J. Physiol. (London) 333, 305-313. Kuffler, S. W. & Sejnowski, T. J. (1983) 7. Physiol. (London) 341, 257-278. Murase, K. & Randic, M. (1984) J. Physiol. (London) 346, 203-217. Surprenant, A. (1984) J. Physiol. (London) 351, 343-364. Ozawa, S. (1981) Brain Res. 209, 240-244. Dubinsky, J. M. & Oxford, G. S. (1985) Proc. Natl. Acad. Sci. USA 82, 4282-4286. Nakagawa, Y., Higashida, H. & Miki, N. (1984) J. Neurosci. 4, 1653-1661. Delfs, J. R. & Dichter, M. A. (1983) J. Neurosci. 3, 1176-1188. Brown, D. A., Constanti, A. & Marsh, S. (1980) Brain Res. 193, 614-619. Higashida, H., Nakagawa, Y. & Miki, N. (1984) Brain Res. 295, 113-119. Abrams, T. W., Castellucci, V. F., Camardo, J. S., Kandel, E. R. & Lloyd, P. E. (1984) Proc. Natl. Acad. Sci. USA 81, 7956-7960. Yellen, G. (1982) Nature (London) 296, 357-359. Gilman, A. G. (1984) Cell 36, 577-579, Berridge, M. J. & Irvine, R. F. (1984) Nature (London) 312, 315-321. Okajima, F. & Ui, M. (1984) J. Biol. Chem. 259, 13863-13871. Bokoch, G. M. & Gilman, A. G. (1984) Cell 39, 301-308. Cockcroft, S. & Gomperts, B. D. (1985) Nature (London) 314, 534-536.