Proc. Nat. Acad. Sci. USA Vol. 72, No. 9, pp. 3472-3476, September 1975 Biochemistry Receptor-mediated shifts in cGMP and cAMP levels in neuroblastoma cells {acetylcholine receptor/prostaglandin Ey receptor /adenosine receptor /synapse /neurotransmitter) HIROSHI MATSUZAWA AND MARSHALL NIRENBERG Laboratory of Biochemical Genetics, National Heart and Lung Institute, National Institutes of Health, Bethesda, Maryland 20014 Contributed by Marshall Nirenberg, July 10, 1975 ABSTRACT 1,5'-eGMP levels of neuroblastoma NIE-115 cells increase as much as 200-fold upon activation of muscar- inic acetylcholine receptors, resulting in intracellular cGMP concentrations >600 pmol/m of protein. The cells also have receptors for adenosine which mediate an increase in 3/:5'- cAMP levels. Unexpectedly, prostaglandin E, was found to increase the concentrations of both cGMP and cAMP. Car- bamylcholine, adenosine, and PGE) were added to cells sep- arately and in pairs to determine the effect of one compound on cell responses to another. Reciprocal inhibition, unilateral inhibition, additive, and nonadditive responses were ob- served with respect to cGMP and cAMP levels when differ- ent pairs of receptors were activated simultaneously. The response of neurons to transmitters frequently can be modified by other species of [transmitter-receptor] interac- tions; thus, two types of synaptic stimuli may potentiate or inhibit one another. The pharmacology and electrophysiolo- gy of such phenomena have been studied extensively but much remains to be learned about the mechanisms which couple the activities of different species of receptors. In part, the difficulty in defining the reactions associated with re- ceptor activation stems from the heterogeneity of neural tissues with regard to cell type. One approach to this prob- lem is to define receptor-mediated responses of relatively homogeneous, clonal neuroblastoma cells which have excit- able membranes and respond to chemical and electrical stimuli. In this report the effects of activators of inhibitory mus- carinic acetylcholine receptors, adenosine receptors, and prostaglandin E; (PGE) receptors (1-5) on 37:5’-cGMP and 3/-5’-cAMP levels of neuroblastoma cells are described. Mus- carinic acetylcholine receptors frequently mediate slow, prolonged responses which may be excitatory or inhibitory; in contrast to the nicotinic acetylcholine receptor, which in higher organisms mediates fast excitatory responses. Activa- tion of the inhibitory or excitatory muscarinic acetylcholine receptors of cardiac (6, 7) and smooth muscle (7) results in an increase in cGMP concentration. We find that cGMP lev- els of neuroblastoma NIE-115 cells are elevated as much as 200-fold upon activation of muscarinic acetylcholine recep- tors. When different pairs of receptors were activated simul- taneously, reciprocal inhibition, unilateral inhibition, addi- tive, and nonadditive responses were observed with respect to cGMP and cAMP levels. MATERIALS AND METHODS Chemicals were obtained from the follawing sources: car- bamylcholine chloride, atropine sulfate, and d-tubocurarine Oe Abbreviations: cGMP, guanosine 3/:5'-cyclic monophosphate; cAMP, adenosine 3':5’-cyclic monophosphate; PGE, prostaglandin E,; IBMX, 3-isobutyl-1-methy]-xanthine, Hepes, N-2-hydroxy- ethylpiperazine-N 1.9.ethanesulfonic acid; DMEM, Dulbecco-Vogt modification of Eagle’s minimal essential medium. chloride from Sigma; N -2-hydroxyethylpiperazine-N’-2- ethanesulfonic acid (Hepes), Calbiochem; 3-isobuty]-1- methyl-xanthine (IBMX) from Aldrich Chem. Co.; PGE), a gift from Dr. John Pike, the Upjohn Co.; adenosine, Nutri- tional Biochem. Corp.; cAMP, Schwarz/Mann; cGMP, JEM Research Products; [G-7H]cAMP (22.1 Ci/mmol), New En- gland Nuclear; [8-3H|eGMP (15 Ci/mmol), Amersham/ Searle Corp.; cAMP-dependent protein kinase, Sigma; cGMP antiserum and succinyl-cGMP 125]-tyrosine methy] ester, Collaborative Research, Inc.; fetal bovine serum, Colo- rado Serum Co.; and DMEM (the Dulbecco-Vogt modifica- tion of Eagle’s minimal essential medium), GIBCO, Cat. no. H_-21. Other chemicals were of reagent grade purity. Cell Culture. The following cell lines were used: mouse neuroblastoma C-1300 clones NI1E-115 and N18 (8), mouse neuroblastoma X mouse L cell hybrid clone NLIF (9), and rat glioma clone C6BU-1 (10). Neuroblastoma cells were grown in 100 mm petri dishes (Falcon, 55 cm? surface area) in 15 ml of 90% DMEM-10% fetal-bovine serum (340 mOsm/kg) in a humidified atmosphere of 10% CO2-90% air at 37°. Hybrid cells were grown in the same medium sup- plemented with 0.1 mM hypoxanthine, 1 ~M aminopterin, and 16 uM thymidine (HAT). Cells usually were grown to confluency (1 to 3 X 10° cells, 0.1 to 0.3 mg of protein per cm2) prior to use. Incubation of Cells. Each plate was washed three times with 7 ml portions of solution A (DMEM minus NaHCOs with 25 mM Hepes adjusted to pH 7.4 with NaOH and to 340 mOsm/kg with NaCl). Unless stated otherwise, 10 ml of solution A supplemented with 0.5 mM IBMX were added to each plate and cells were equilibrated for 30 min at 37°. Re- actions were initiated by the addition of 0.1 ml of a solution of the compound to be tested dissolved in 0.17 M NaCl, or 0.1 ml of a PGE; solution (1 mM PGE), 0.153 M NaCl, and 10% ethanol). The final concentration of ethanol in the petri dish, 0.1%, did not alter cGMP or cAMP levels of cells. Cells were incubated at 37° for the times specified, then the me- dium was removed by aspiration (discarded unless specified otherwise) and 4 ml of cold 5% trichloroacetic acid was added to the monolayer to terminate reactions and extract nucleotides. A solution (0.05 ml) containing 10,000 cpm of (SH|cGMP (0.67 pmol) and/or [SH|cAMP (0.45 pmol) was added to each dish and cells were detached by scraping. The suspension was transferred to a tube, the dish was washed with 1 ml of cold 5% trichloroacetic acid, and the combined suspension and wash was centrifuged for 15 min at 20,000 x gat 3°. Purification of cGMP and cAMP. A rapid method was devised for separating cGMP from cAMP and removing tri- chloroacetic acid and endogenous interfering compounds from cyclic nucleotide fractions. The supernatant solution © the deproteinized sample in 5% trichloroacetic acid was ap plied to an ion exchange column (0.7 X 8 cm) of AG50W-X4 3472 Biochemistry: Matsuzawa and Nirenberg (200-400 mesh, hydrogen form, Bio-Rad) previously washed and equilibrated with H2O. The column was washed with 2 ml of H2O (eluates discarded) and then with 2 additional ml of HzO. The latter eluate (2.0 ml) which contains cGMP was collected on a neutral alumina (WN-3, Sigma) column (0.7 X 3 cm) previously washed with 20 ml of 200 mM sodium acetate, pH 6.2, and then with 10 ml of 5 mM sodium ace- tate, pH 6.2. The eluate was discarded, and the column was washed with 3 ml of 5 mM sodium acetate buffer, pH 6.2, and then with 1 ml of 200 mM sodium acetate buffer, pH 6.2 (eluates discarded). Then cGMP was eluted with 2 ml of 200 mM sodium acetate, pH 6.2, and portions of the eluate were assayed for cGMP concentration and for recovery. The average recovery of cGMP was about 50%; <1% of the cAMP of the sample was present in the cGMP fraction. Cyclic AMP fractions were obtained as follows: after cGMP was eluted from the AG50W-X4 column, the column was washed with 2 ml of HgO (eluate discarded) and cAMP was eluted with 3 additional ml of H2O. Portions of the el- uate were assayed for cAMP concentration and recovery. The cAMP fraction contained <0.2% of the cGMP of the sample prior to fractionation. The average recovery of (3HIcAMP was 70%. Each AG50W-X4 column was regenerated by washing with 8 ml of 1 N HCl and, prior to use, with 30 ml of water. Assay of cGMP and cAMP. The cGMP concentrations of purified alumina column eluates were determined by the ra- dioimmune method of Steiner et al. (11) modified as fol- lows: each 0.5 ml reaction mixture contained 200 mM (in some experiments 140 mM) sodium acetate, pH 6.2 (solution B); 17 fmol of succinyl-eGMP !**I-tyrosine methyl ester {10,000-20,000 cpm); cGMP antibody; and sample (0-20 pmol of cGMP). After incubation (14-18 hr, 3°) each reac- tion mixture was diluted with 2 ml of solution B at 3°; 10 min later the sample was passed through a Millipore filter (0.45 um pore size, 25 mm diameter, HAWP 02500) pre- viously rinsed with solution B and washed three times with cold solution B (4 ml each wash). Each filter was placed in a scintillation vial with 1 ml of 2-methoxyethanol; after filters dissolved (0.5-1 hr) 15 ml of Aquasol (New England Nucle- ar) were added. Radioactivity was determined 12-24 hr later in a Beckman liquid scintillation counter (3H plus !4C isoset). Cyclic AMP concentrations of samples purified as de- scribed above were determined by the method of Gilman (12). Each cGMP or cAMP sample was assayed at three or four concentrations. Each value shown corresponds to 100% recovery of cyclic nucleotide and is the mean of duplicate dishes. Protein was determined by a modification of the method of Lowry et al. (13). RESULTS A rapid method was devised for separating cGMP from cAMP and removing trichloroacetic acid and endogenous factors which interfere with the cGMP assay. The proce- dure, based upon the Materials and Methods section, is based upon the methods of Salomon, Londos, and Rodbell (14) and White and Zenser (15); CGMP and cAMP are recov- ered in separate fractions in high yield and in sufficiently high concentration to assay directly without lyophilization. One hundred samples can be purified in 2-3 hr. The effect of carbamylcholine, a relatively stable analog of acetylcholine, upon intracellular cGMP levels of neuro- blastoma and other cell lines in the absence of a 3/:5’-cyclic nucleotide phosphodiesterase inhibitor is shown in Fig. 1. Cyclic GMP levels of neuroblastoma clones NI8 and N1E- Proc. Nat. Acad. Sci. USA 72 (1975) 3473 TOOR — pmol cGMP/mg protein MINUTES Fic. 1. Effect of 1 mM carbamylcholine on intracellular levels of cGMP in neuroblastoma clones NIE-115 and N18, neuroblasto- ma X L cell hybrid clone NL1F, and glioma clone C6BU-1. Cul- tures were confluent; the average protein content was: 11.5, 16.4, 7.6 and 6.9 mg of protein per 100 mm petri dish for N1E-115, N18, NLIF, and C6BU-1, respectively. A 3’:5’-cyclic nucleotide phos- phodiesterase inhibitor was not present. 115 rose 40- and 210-fold, respectively, upon incubation with carbamylcholine for 15 sec, and then fell rapidly to normal levels (50% decrease in cGMP at 45 sec). Carbamyl- choline also elicited a 6-fold increase in the cGMP concen- tration of the neuroblastoma X L cell hybrid, NL-1F, but had no effect upon cGMP levels of rat glioma clone C6BU-1. Thus, two types of cell lines were found with respect to ace- tylcholine-receptor-mediated reactions which increase cGMP levels: cell lines which are sensitive to carbamylcho- line and a line which is insensitive. The effect of IBMX, a 3’:5’-cyclic nucleotide phosphodies- terase inhibitor, upon the rate of disappearance of intracel- lular cGMP after stimulation with carbamylcholine is shown in Fig. 2A. In the absence of IBMX, maximum cGMP levels were attained 15 sec after the addition of carbamylcholine, and then rapidly decreased (50% decrease at 45 sec; Figs. 1 and 2A). In the presence of 0.5 mM IBMX, maximum cGMP levels were attained at 30 sec and cGMP levels then de- creased at a slower rate (50% decrease at 3 min). The effect of IBMX upon carbamylcholine dependent cGMP accumulation also was studied with NIE-115 cells adapted to grow in suspension culture. The assay was modi- fied so that both intra- and extracellular cGMP was deter- mined. As shown in Fig. 2B, carbamylcholine elevated cGMP levels of suspension cells and in the presence of IBMX cGMP levels remained constant during the 10 min incuba- tion period. These results suggest that part of the cGMP formed by cells may be excreted into the medium. Other 3':5’-cyclic nucleotide phosphodiesterase inhibitors such as 4-(3-butoxy-4-methoxybenzyl)-2-imidazolidinone — (Ro-20- 1724) and papaverine were not as effective as IBMX in maintaining cGMP levels of N18 cells. The relation between carbamylcholine concentration and eGMP accumulation in N1E-115 cells is shown in Fig. 3. The maximum increase in cGMP concentration was ob- tained with 1073 M carbamylcholine; 50% of the maximum increase in cGMP concentration was evoked by 1074 M car- bamylcholine. The effects of atropine and d-tubocurarine, selective in- hibitors of muscarinic and nicotinic acetylcholine receptors, 3474 Biochemistry: Matsuzawa and Nirenberg pmo! cGMP or cAMP/mg protein ett SUSPENSION CELLS AND MEDIUM 200k c [ B 100 9 , & 50 ao £ = 20 8 2 woke ° & ef a 5 a_i 4 1 1 1 1 L L 1 oT 23 4 5 6 7 8 9 10 MINUTES Fic. 2. Effect of 1 mM carbamylcholine in the presence of 0.5 mM IBMX, a 3:5’-cyclic nucleotide phosphodiesterase inhibitor, on cGMP levels of neuroblastoma NI1E-115 cells. (A) Cells were grown for 9 days to confluence; the average cellular protein was 12.2 mg/100 mm dish. The broken line corresponds to the data for N1E-115 shown in Fig. 1 (cells were incubated in the absence of IBMX). (B) Cells were grown in suspension culture in DMEM with 10 mM NaHCOs, 25 mM Hepes, and 2% fetal bovine serum (pH 7.3, 340 mOsm/kg). Cells were washed and 1 ml portions (4 X 10 cells) were preincubated in the presence of 0.5 mM IBMX for 15 min with gentle shaking. Two experiments are shown. Cells were incubated for 3 min or less (aA) with 1 mM carbamylcholine; or (4) without this compound. In another experiment (m) cells were incubated for 0 to 10 min with 1 mM carbamylcholine. In both experiments reactions were terminated by the addition of 4 ml of cold 6% trichloroacetic acid to each tube. Hence the com- bined intra- and extracellular cGMP concentration was deter- mined. respectively, upon carbamylcholine-dependent accumula- tion of cGMP in N1E-115 cells are shown in Fig. 4. Atro- pine, at 1 X 1077 M, inhibited the carbamylcholine-depen- dent increase in cGMP concentration by 50%; 97% inhibi- tion was obtained with 1 X 1076 M atropine. In contrast, d- tubocurarine did not inhibit appreciably even at 1 X 1074 M: in fact, in other experiments d-tubocurarine potentiated the carbamylcholine-dependent elevation of cGMP. The in- hibition by atropine suggests that the increase in cGMP elic- ited by carbamylcholine is mediated by the muscarinic ace- tylcholine receptors. The effects of PGE; upon cAMP and cGMP levels of NIE-LI5 cells are shown in Fig. 5. Cyclic AMP levels were found to increase markedly upon the addition of 10 4M PGE). Unexpectedly, the level of cGMP also increased 6- fold in the presence of PGE). The maximum concentration of cAMP attained in this experiment was 9-fold higher than that of cGMP; however, in other experiments, the concen- tration of cGMP dependent on PGE, increased to 250 pmol Proc. Nat. Acad. Sci. USA 72 (1975) 250 as 70 . 200 s rr a i S & 50 -TUBOCURARINE © g 6 g 5 150 os js £ Ww g & 30 ATROPINE | x & 100 6 © Fi e 2 1 9 3 20 - € E z 4 | t \° ° ie oo ie MOLARITY io io Ww 1 IC MOLARITY CARBAMYLCHOUINE Fic. 3 (left). Relation between carbamylcholine concentration and cGMP levels of neuroblastoma N1E-115 cells. Cultures were grown for 9 days to confluence; the average cellular protein was 13.5 mg/100 mm dish. Cells were incubated for 30 sec in the pres- ence of the indicated concentrations of carbamylcholine and IBMX. Fic. 4 (right). Effect of atropine and d-tubocurarine concen- tration on the carbamylcholine dependent increase in cGMP level of N1E-115 cells. Cultures were grown for 9 days to confluence; the average cellular protein was 14.6 mg/100 mm dish. Cells were incu- bated for 30 sec in the presence of 1 mM carbamylcholine and the indicated concentrations of atropine or d-tubocurarine. of cGMP per mg of protein, approximately 60% of the con- centration attained by cAMP in the presence of PGE. In other experiments not shown here, the increase in cGMP concentration elicited by PGE; was shown to be insensitive to atropine. Thus, the effect of PGE; upon eGMP is not me- diated by the muscarinic acetylcholine receptor. The maxi- mum cGMP concentration was achieved 30 sec after the ad- dition of PGE,; cGMP levels decreased during subsequent incubation. In contrast, the maximum cAMP level was achieved 4 min after the addition of PGE). The effects of carbamylcholine, PGE), and adenosine upon cGMP and cAMP levels of N1E-115 cells plotted on a logarithmic scale to illustrate inhibitory as well as stimulato- ry effects are shown in Fig. 6A and B, respectively. In the presence of carbamylcholine or PGE;, cGMP levels in- creased and reached maximum levels within 1 min. The rates of disappearance of cGMP during the subsequent incu- bation period were linear. In contrast, the basal level of cGMP was reduced approximately 50% in the presence of 0.1 mM adenosine. As shown in Fig. 6B, cAMP levels in- creased markedly in the presence of PGE; or adenosine, but decreased 30% in the presence of carbamy|choline. Various combinations of carbamylcholine, PGE), and adenosine were added simultaneously to determine the ef- fect of one compound upon cell responses to another (Fig. b OQ pmol cGMP “mg protein nN Oo NM ° 3 pmol cAMP/mg protein Oo-Tss4567 8 MINUTES Fic. 5. Effect of PGE, on intracellular levels of cGMP and cAMP in neuroblastoma N1E-115 cells. Cultures were grown for © days; the average cellular protein was 6.8 mg/100 mm dish. Close and open circles correspond to cGMP levels with and without 10 uM PGE), respectively; closed and open triangles correspond t¢ cAMP levels with and without 10 uM PGE), respectively. Biochemistry: Matsuzawa and Nirenberg 200 pea FO cGMP 1 ® cAMP loo E, 4 F CARBACHOL 4 50 + pce, 4500 Cc a & 6 2 a 9 F 20 Poe, 4200 § a t 1 £ S + swe 4 a Qo 10 + ADENOSINE 11909 5S 2 ¥ contro f { s ° + 4 —_ 4 3 B 7 7°90 CONTROL a oO ‘ ¥ 7 ADENOSINE 2b ai 4 20 E rt ° CARBACHOL J I 1 4 1 1 C 4 1 4 4 10 Oo t 23 4 Oo |! 2 3 4 MINUTES FIG, 6. Effect of 1 mM carbamylcholine (carbachol), 0.1 mM adenosine, or 10 uM PGE, on intracellular levels of cGMP (A) and cAMP (B) of neuroblastoma N1E-115 cells. Culture conditions were identical to those described in the legend to Fig. 5. 7). The effect of exposing cells to both carbamylcholine and adenosine upon cGMP and cAMP levels is shown in Fig. 7A and B, respectively. Adenosine inhibited the carbamylcho- line-dependent increase in cGMP concentration 25 and 60% when incubations were 30 and 60 sec, respectively. In Fig. 7B, the effect of carbamylcholine upon the adeno- sine-dependent increase in cAMP concentration is shown. Carbamylcholine did not inhibit the adenosine-stimulated increase in cAMP during the first 30 sec of incubation even though the maximum increase in cGMP concentration evoked by carbamylcholine was achieved at 30 sec. How- ever, during further incubation, carbamylcholine inhibited almost completely the adenosine-dependent elevation in cAMP concentration. Carbamylcholine and PGE, added separately increased cGMP levels as indicated by the broken lines in Fig. 7C. The effects of carbamylcholine and PGE, added simultaneously to cultures were additive or more than additive. However, carbamylcholine inhibited the PGE,-induced increase in cAMP concentration, particularly during the latter part of the incubation period (Fig. 7D). Thus, carbamylcholine may have a delayed inhibitory effect upon both adenosine- and PGE\-dependent increases in cAMP concentrations. Atro- pine completely blocked the inhibitory effect of carbamyl- choline upon adenosine- or PGE)-dependent increases in cAMP levels (data not shown). The effect of activating receptors for adenosine and PGE; separately and simultaneously upon cGMP and cAMP levels is shown in Fig. 7E and F, respectively. Adenosine inhibited slightly the rise in CGMP levels elicited by PGE). The results show that the increases in cAMP levels elicited either by adenosine or PGE, are not additive. In other experiments, the sequence of addition of carbamylcholine and either adenosine or PGE, was varied, but the results did not differ significantly from those found when two compounds were added simultaneously. DISCUSSION Two types of cells were found with respect to muscarinic acetylcholine receptor activity: cells sensitive to carbamyl- choline and cells insensitive to this compound. The degree of response also varied widely from one clone to another. Acti- vation of the muscarinic acetylcholine receptors of neuro- Proc. Nat. Acad. Sci. USA 72 (1975) 3475 150 Ff T r T ‘TT v ¥ T (@) cGMP {® cAMP oO Pry 4 l0Ors \\ CARBACHOL + ADENOSINE ew & c 2 3 ° 3 a a £ 2 a a = = 20 g : 6 3 & 0) E =--- ===>. A bo Oo O-1 234 01234 MINUTES Fic. 7. Effect of various combinations of 1 mM carbamylcho- line, 0.1 mM adenosine, and 10 uM PGE, on intracellular levels of cGMP (panels A, C, and E) and cAMP (panels B, D, and F) of neu- roblastoma N1E-115 cells. Culture conditions were identical to those described in the legend to Fig. 5. The broken lines corre- spond to the effect of each compound added separately (data shown in Fig. 6); C, A, and P represent carbamylcholine, adeno- sine, and PGE, respectively. blastoma N1E-115 cells evoked up to a 200-fold increase in cGMP concentration, yielding levels in excess of 600 pmol of cGMP per mg of protein. These cells have high tyrosine hy- droxylase activity (8), generate action potentials in response to electrical stimuli (8), have neurosecretory vesicles, and also possess PGE, receptors and adenosine receptors which upon activation lead to elevated cAMP levels (2, 3). Adenosine, carbamylcholine, and PGE, were added to cells separately and in pairs to determine the effect of one compound upon cell responses to another. The results, sum- marized in Table 1, show that in the presence of adenosine cAMP levels increase and cGMP levels decrease; whereas, in Table 1. Summary of receptor-mediated shifts in cGMP and cAMP levels in neuroblastoma cells cGMP cAMP Addition percent None 100 100 Adenosine 40 500 Carbamylcholine 1500 70 PGE, 500 1250 Adenosine + carbamylcholine 650 250 PGE, + carbamylcholine 2050 1000 PGE, + adenosine 300 1100 100% corresponds to 7.5 pmol of cGMP/mg of protein or 27.5 pmol of cAMP/mg of protein. The data are trom the experiments shown in Figs. 6 and 7. 3476 Biochemistry: Matsuzawa and Nirenberg the presence of carbamylcholine, cGMP levels increase and cAMP levels decrease. In addition to the inverse relationship between cGMP and cAMP concentrations, a novel phenom- enon was observed; that is, in the presence of PGE; the lev- els of both cGMP and cAMP increase. The increase in cGMP elicited by carbamylcholine was not inhibited by PGE; in spite of the fact that PGE; elicits a marked increase in cAMP levels. Prostaglandin-E)- and carbamylcholine-de- pendent increases in cGMP were fully additive, and in some experiments, were greater than additive. In contrast, the ele- vated cAMP concentrations induced either by PGE or by adenosine were not additive when both compounds were added simultaneously to cells. We conclude that activation of a species of muscarinic acetylcholine receptor inhibits adenosine- and PGE)-recep- tor-mediated reactions which elevate cAMP levels, whereas, activation of adenosine receptors inhibits reactions mediated by muscarinic acetylcholine receptors and PGE, receptors which elevate cGMP levels. In contrast to these results, acti- vation of PGE, receptors did not affect acetylcholine- or adenosine-receptor-mediated reactions which affect cGMP or cAMP levels. Thus, reciprocal inhibition, unilateral inhi- bition, additive, and nonadditive responses were observed when different pairs of receptors were activated simulta- neously. Relatively little is known at the molecular level about the mechanisms which couple one receptor with another. The observed phenomena are obviously complex and may reflect changes in membrane permeability, metabolism of cyclic nucleotides, and so forth. Other examples of reciprocal cou- pling of cGMP and cAMP levels and possible mechanisms of coupling have been discussed by Goldberg et al. (22, 23). Blume et al. (5) have reported that acetylcholine inhibits the increase in cAMP levels induced by PGE, or adenosine in mouse neuroblastoma cells. Acetylcholine also has been shown to inhibit the PGE)-dependent elevations of cAMP levels, the inhibition mediated by excitatory muscarinic ace- tylcholine receptors in neuroblastoma X glioma hybrid cells (16). Acetylcholine and carbamylcholine also have been re- ported to stimulate mouse neuroblastoma adenylate cyclase activity (17). . The inhibitory effect of carbamylcholine on reactions me- diated by other species of receptors shows that some N1E- 115 cells with muscarinic acetylcholine receptors also have adenosine or PGE) receptors. Some cells also have both adenosine receptors and PGE receptors. Although N1E-115 is a clonal cell line, single cell analysis by electrophysiologic methods reveals two acetylcholine receptor phenotypes; cells sensitive and cells insensitive to carbamylcholine. All NIE- 115 cells sensitive to acetylcholine or carbamylcholine exhib- it hyperpolarizing responses to these compounds, frequently —25 mV in magnitude and 10-15 sec in duration. Hence, N1E-115 cells have inhibitory muscarinic acetylcholine re- ceptors. The activity of PGE, in eliciting increases in both cGMP and cAMP levels should be considered in context with the different actions of PGE; on various tissues which have been reported. For example, PGE, elevates cAMP levels in some tissues and reduces hormone-dependent elevation of cAMP levels in others (18), activates adenylate cyclase activity in Proc. Nat. Acad. Sci. USA 72 (1975} homogenates (19), including those prepared from neuroblas- toma X glioma hybrid cells (20), and inhibits the vasopres- sin-dependent activation of adenylate cyclase (21). Prostaglandin E, elicits a transitory, 30 sec, increase in cGMP accumulation and a prolonged increase in cAMP ac- cumulation. The difference in the duration of cGMP and cAMP accumulations may reflect the activities of different enzymes which catalyze the synthesis and perhaps the hy- drolysis of the nucleotides, or alternatively, may indicate the presence of two species of PGE, receptor, one mediating an increase in cAMP concentration, the other, an increase in cGMP. Preliminary results which support the latter hypoth- esis will be described elsewhere. We thank Doyle Mullinax, Deborah Carper, and Linda Lee for their excellent assistance, and Mary Ellen Miller for typing the manuscript. 1. Gilman, A. G. & Nirenberg, M. (1971) Nature 234, 356-358. Blume, A. J., Dalton, C. & Sheppard, H. (1973) Proc. Nat. Acad. Sci. USA 70, 3099-3102. Hamprecht, B. & Schultz, J. (1973) FEBS Lett. 34, 85-89. 4. Schultz, J. & Hamprecht, B. (1973) Naunyn-Schmiedeberg's Arch. Pharmacol. 278, 215-225. 5. Blume, A. J., Foster, C. & Karp, G. (1974) Trans. Amer. Soc. Neurochem. 5, 150. 6. George, W. J., Polson, J. B., O'Toole, A. G. & Goldberg, N. D. (1970) Proc. Nat. Acad. Sci. USA 66, 398-403. 7. Lee, T., Kuo, J. F. & Greengard, P. (1972) Proc. Nat. Acad. Sci. USA 69, 3287-3291. 8. Amano, T., Richelson, E. & Nirenberg, M. (1972) Proc. Nat. Acad. Sci. USA 69, 258-263. : 9. Minna, J., Glazer, D. & Nirenberg, M. (1972) Nature New Biol. 235, 225-231. 10. Amano, T., Hamprecht, B. & Kemper, W. (1974) Exp. Cell Res. 85, 399-408. 11. Steiner, A. L., Parker, C. W. & Kipnis, D. M. (1972) J. Biol Chem. 247, 1106-1113. 12. Gilman, A. G. (1970) Proc. Nat. Acad. Sci. USA 67, 305-312. 13. Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J (1951) J. Biol. Chem. 193, 265-275. 14. Salomon, Y., Londos, C. & Rodbell, M. (1974) Anal. Biochem. 58, 541-548. 15. White, A. A. & Zenser, T. V. (1971) Anal. Biochem. 41, 372- 396. 16. Traber, J., Fischer, K., Buchen, C. & Hamprecht, B. (1975) Nature 255, 558-560. 17. Prasad, K. N., Gilmer, Kk. N. & Sahu, S. K. (1974) Nature 249, 765-767. 18. Butcher, R. W. & Baird, C. E. (1968) J. Biol. Chem. 243, 1718-1717. 19. Wolfe, S. M. & Shulman, N. R. (1969) Biochem. Biophys. Res. Commun. 35, 265-272. 20. Sharma, S. K., Nirenberg, M. & Klee, W. A. (1975) Proc. Nat. Acad. Sci. USA 72, 590-594. 21. Marumo, F. & Edelman, I. S. (1971) J. Clin. Invest. 50, 1613-1620. 22, Goldberg, N. D., O'Dea, R. F. & Haddox, M. K. (1973) in Ad- vances in Cyclic Nucleotide Research, eds. Greengard, P. & Robison, G. A. (Raven Press, Publishers, New York), Vol. 3. pp. 155-223. 23. Goldberg, N. D., Haddox, M. K., Dunham, E., Lopez, Gc & Hadden, J. W. (1974) in Control of Proliferation in Animal Cells, eds. Clarkson, B. & Baserga, R. (Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.), pp. 609-625. e