OPIATE BECEPTGES AS REGULATGBS OF ADENYLATE CYCLASE Werner A. Klee, Shail K. Sharma* and Marshall Nirenberg Laboratory of General and Comparative Biochemistry, NIMB, and the Laboratory of Biochemical Genetics, IBILI, NIH, Bethesda, Maryland 20014 (Received in final form day 24, 1975) We have reported the presence of opiate receptors in some neuroblastoma derived cell lines cultured in vitro and that a neuroblastoma x glioma hybrid cell line, NG108-15, contains a particularly large number of morphine receptors (1). Table 1 shows that whereas the hybrid cell line has opiate receptors II-dihydromorphine H naloxone HYBRIDS -15 fmoleslmg protein 17 37 (19) PARENTS -2 0 11 (6) c6BU-1 1 1 (1) The concentration of radioactive narcotic was 1 n&l in each case. In neither case is this close to a saturating amount, naloxone one has twice the affinity of dihydromorphine and so to be comparable the ,naloxone data should be divided by 2 (numbers in parenthesis);, NG108-15 (also called lOSCCl5) was obtained by B. Siamprecht, T. Amano and M. Nirenberg (in preparation), NlOTG-2 by Minna et al. (2), CLBU-1 by Amano et al. (3). which are rea dilydemonstrntod gands those of N18TG2 were only detected with=II-naloxone binding whereas the C6BU-1 line does not have a detectable number of opiate receptors by either assay method. This group of cell lines with no, few and an abundance of opiate receptors has provided us with material with which to study the biochemical consequences of the interaction of morphine with its receptor. Collier and Boy reported that morphine and related drugs inhibit the PGE. stimulated conversion of3B ATP into CAMP by rat brain homogenates in a way that correlates with agonist potency and receptor affinity (4.5). These exneriments promoted us to examine the effect of morphine on adenylate cyclase activity and on the CAMP levels of 1?G108-15 hvbrid cells and their oarental cell lines (6). We found that morphine inhibits the adenyiate cvclase activity of NG108-15 cells and lowers cellular cAMP levels in the presence and in the absence of added PGFJ (fig 1). '*Fogarty InternationalFellow, on leave from the Department of Diochemistry, All India Institute of Medical Sciences, Dew Delhi, India. 117 Figure 1. Inhibition by morphine (lOdM) of the rate of CAMP accumulation in intact NG108-15 hybrid cells(part A) and of adeny- late cyclase activity in homogenates (part B). Basal and PGQ (lO-SM) stimulated results are shown (6). There is also a dramatic reduction in the adenosine stimulated rise in cellular cAMP levels in the presence of morphine (17). Thus, in this cell, morphine inhibits both stimulated ,- and unstimulated adonylate cyclase. \- Morphine inhibits the adenylate cyclase of the neuro- blastoma parent somewhat, but does not affect the activity the enzyme found in the glioma parent (Table II). Thus, the de- Tablemxcx of morphine on adenylate cyclase activity of neuroblastoma and glioma parents N18'IG-2 C6BU-1 Addition* pmole/min/mg protein None 6 20 Morphine 4 19 Naloxone 20 Morphine + naloxone 55 20 PGC, 24 PGQ + morphine 67: PGQ + naloxone 72 ii PGQ morphine + naloxone 70 23 o lQrm of each component gree of inhibition or adenylate cyclase by morphine is correlated with the number of opiate receptors. There are a number of other properties of the enzyme of NG108-15 cells which show that the inhibition by opiates is mediated by their receptors. Thus, naloxone, an apparently pure antagonist of narcotic drugs reverses morphine inhibition of adenylate cyclase (6). Furthermore, the inhibition is sterospecific in that levorphanol, but not its inactive isomer, dextrorphan, inhibits adenylate cyclase (6). Traber et al. (7-9) have also reported that morphine reduces PGE .clevation of cAt!P levels and Dlosser et al. (10) have reported that morphine inhibits the PGI$ dependent activation of adenylate cylase in neuroblastoma or hybrid cell lines derived from neuro- blastoma cells. There is a good correlation between the concentrations at which opiate agonists displace JIl-naloxone from the receptors and and those required for inhibition of adenylate cyclase. This agreement is readily apparent in the data presented in Table III. 118 Table III Comparison of narcotic affinity for the opiate receptor and ability to inhibit adenylate cyclase Kd Ki Narcotic receptor Adenylate cyclase Narcotic nM nM Btorphine 5 10 Levorphanol 200 200 Morphine 4,000 2,000 3-Nlylprodine 10,000 50,000 Dextrorphan 10,000 -- Naloxone 20 -- However, we found th the opiate binding and enzyme inhibition curves not superimpo:ible (6) Enzyme inhibition takes place over a nmch narrower range of-drug concentrations than does dis- placement of %I-naloxone from the receptors. This behavior im- plies cooperativity among liganded receptors in their interaction with the adenylate cyclase complex. Analysis of the data by means of Hill plots shaws that the slope for narcotic binding (reaction1 of scheme 1) is close to 1 indicating little or no cooperativity a ? ??? ?? receptor4 receptor*M o ?o i enzyme*receptor.M 1 in the formation of Garcotic.receptod complex but that the maximum slopes of the cumes for adenylate cyclase inhibition by narcotics (reaction 2) are between 2 and 3, indicat' g strong positiv cooperativity in the reactions that couple K arcotic- recepto 3 wmplexes with adenylate cyclase. Coupling of opiate-receptor complexes to adenylate cyclase may occur by any of 3 general mechanisms: 1) Direct interaction of receptor and enzyme, by analogy with enzyme systems composed of catalytic and regulatory subunits,or linkage via a modulator (15). 2) Opiate-receptor complexes may elicit the production of chem- ical messages, suggested as a rather unlikely possibility by 1I.O.J. Collier (personal wmmunication). 3) Indirect coupling mediated by conformational transitions of the membrane. The membrane conformation may reflect either the proportions of receptors in states A and B or may change in re- ponse to the process of transition of the conformation of the receptors betweenhtates A and B induced by the association and dissociation'of agonists (but not of antagonists, since Na', main- tains receptors in the B state). The indirect coupling mechanisms, which involve membrane chanqes , may allow opiate receptors to wpress their interaction with narcotics in more than one way. Thus, the inhibition of adenylate cyclase and the electro- physiologocal effects of narcotics on NG108-15 cells found by Traber et al. (9), Myers and Livengood (16) and in our own laboratories can be different manifestations of the same fundamental effects on membrane structure. An important property of many of the narcotic analgesics is that of mixed agonist-antagonist behavior. Perhaps the best studied example of such a compound is nalorphine which is a potent antagonist of morphine, but also is a goad analgesic in its cwn right (11). fIow may this dualism of action be understood 119 in the context of opiate action as an inhibitor of adenylate cyclase? Figure 2 shows the effects of nalorphine upon adenylate l-l- Figure 2. Effects of nalorphine on the adenylate cyclase activity of homogenates of NG108-15 cells. The curves, reading from top to bottom, represent experiments performed at the following wncent ations 5 of morphine: none, 2 x 10 MI lo-*M, and 2 x 10' M. The data have been normalized so that uninhibited adenylate cyclase activity is wnstant. cyclase activity at several wncentrations of morphine. In the absence of morphine, nalorphine inhibits the enzyme but only partially when compared with the degree of inhibition produced by morphine. In the presence of morphine, on the other hand, the effect of nalorphine is to reverse the inhibition produced by morphine. !rhe reversal of morphine inhibition by nalorphine is also not complete but only restores enzyme activity to the level seen in the presence of nalorphine alone. There is evidence that opiate receptors exist in two wnfor- mational states (12, 13) as shown below: Na+ A .d Receptor B inhibition of adenylate cyclase uncoupled Receptor form A has a high affinity for agonists and form B for antagonists as shown. Binding of agonists to the receptor will shi the equilibrium o the left and convert most receptors to the fL onisteReceptor A wmplex which will result in inhibition of aden late cyclase. Ii i Conv sely, when receptors are in the form of the Antagonist*Receptor B complex, as the result of interaction with a pure antagonist, adenylate cyclase is not inhibited. Mixed agonist-antagonists, such as nalorphine, may have a comparable af- finity for receptors in both states. The interaction of opiate receptors with agonist-antmonist narcotics will then result in the receptor complexes being partitioned between states A and B in comparable amounts. Inhibition of adenylate cyclase will thus be only partial as is observed. When NGlO8-15 cells are cultured in the preselre of morphine for a number of days, the level of adenylate cyclase activity in- creases by approximately 50-1008. An experiment which dcmon- strate, this phenomenon is shown in fig 3. The cells after 2 or more days of exposure to morphine are tolerant in the sense that adenylate cyclase activity is nearly normal when assayed in the 120 adenylate cyclase activity is nearly normal when assayed in the 34YS CELLS TREATED WITH MORPHINE Figure 3. activity of nasal and PGQ stinulatcd adenylate cyclase homogenates of 16108-15 cells cultured in the presence of 10 bl morphine for the tines shown (17). presence of morphine. They are dependent upon it in the sense that adenylate cyclasa activity measured in its absence is abnormally high. The dependence phenomenon is dranatically seen when cNIP levels of cells cultured in the presence of morphine for 48 hours are measured after a brief exposure to naloxone. Table IV CAMP levels of nornal and addicted cells (11) CELLS Conditions of Assay basal pnolcs cNJP/ng protein 20 23 naloxone 21 37 PGC 264 81 PGE + naloxqne 241 1183 adenoske ' 103 65 adenosine + naloxone 72 217 naloxone e of IlQxlal 113 175 4:: 3:: Addicted cells show as much as a 4 to 5 fold increase in ci%P levels Over the control, in the presence but not in the absence of naloxonc precipitated withdra%lal. There ia no change in the number of opiate receptors in tolerant cells (17). Pigure 4 summarizes the general conclusions which we have reached wncerning tolerance and dependence on the basis of these and other (17), related, experiments. Ne find that morphine inhibits adenylate cyclase activity and thus de- creases cAflP levels. On continued exposure tc morphine the cells adapt by an increase in adonylate cyclase activity which results in tolerance and dependence. The fully toler- ant cells have CAMP levels close to normal in the presence of morphine. When the opiate is withdratm on addition of an antagonist, CAM? levels rise to abnormally high values. This abrupt increase in cA!lP indicates that the cells are depen- 121 ADD WITHDRAW MORPHINE MORPHINE ii ~200- 1 f J 0. 5 ISO- 0 0 Figure 4. TIME dent upon morphine and is the biochemical counterpart of the abstinence syndrome. Recovery of the cells from the'addictcd state requires the return of adenylate cyclase activity to its normal levels. These results support the suqgestions of Caldstein and Goldstein (14) and Shuster (19), made many years ago, that drugs may act as enzyme inducers. Increases in CAMP levels in the abstinence syndrome of animals has recently been demonstrated by Collier and Francis (18). 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. E: 16. 17. it: W.A. ELEE, M. NIRENBERG, Proc. Nat. Acad. SCi. U.S.A., 71, 3474-3477 (1974). T MINNA, 0. GLAZER, M. NIRENBERG, Nature New Biol., 235 225-231 (1972). -- T. AMANO, B. IIAMPRECRT, W. KEMPER, Exp. Cell Res., 85 399-408 (1974). H.O.J. tiLL1F.k; A.C. ROY, 1l.O.J. COLLIER, A.C. ROY, S.K. SIIARMA, 14. NIRENBERG, W.A. U.S.A., 72 590-594 (1975). -TRAIIEr K. FISCRER, S. LATZIN, B. HANPRECRT, FEDS LettS., 49 260-263 (1974). J.TRABER, K. FISCHER, S. LATZIM, B. IIAMPRECIIT, Nature, Land., 253 120-122 (1975). J.TRABER, G. REISER, K. FISCHER, 8. RNE'RECRT, FEBS Letts., 52 327-332 (1975). rC. BLGSSER, J.R. ABBOTT, W. SRAIB, Fed. Proc., 34 713 (1975). R.I. TABER, D.D. GREENHOUSE, J.K. RCNhELL S. IR!-m, J. Pharmacol. Exp. Ther., 169 29-38 (1969;. C.B. PERT, G. PASTERIIAR, ST SEYDER, Science, 182 1359-1361 (1973). E.J. SIMON, J.M. BILLER, J. GROTR, I. EDELMAN, J. Pharnacol. Exp. Ther., in press (1975). 0.~. GOLDSTEIN, A. GOLDSTEIN, Biochem. Pharmacol., 8 48 (1961). M. RODBELL, H.C. LIN, Y. SALOMm C LO'IDOS J P B~RWUOD D.R. MARTIN, H. RENDELI., M. BBR&,'Adv~ in'Cyk!c NucleoLide Research, 5 3-29 (1975). P MYERS 5. LIVENGODD, Fed. Proc., 43 359 (1955). Sk. SIIAhA, W.A. IiLER, M . .RC*Tin preparation. U.O.J. COLLIER, D.L. FRAN&IS, Nature, Lond., 255 159-162 (1975). L. SIIUSTER, Nature, Lond., 189 314-315 (1961).