14 Biological Chemistry Part (v) Neurochemistry Receptors for Drugs and Neurotransmitters ~~ By J. F. COLLINS Department of Chemistry City of London Polytechnic London EC3N 2EY 1 Introduction In an interview' published in 1974 Sir Robert Robinson questioned about the future direction of organic chemistry replied that he thought that among areas which would develop rapidly would be the field associated with the transmission of nerve impulses. This review aims to outline some of the advances made in this field during the past five years. In particular emphasis will be placed on how drugs and neurotransmitters interact with their post-synaptic receptors. It has long been thought that neurotransmitters exert their effects on target tissues by combining with specific receptor sites and in recent years it has been possible to demonstrate this resulting from the development receptor-specific ligands radiolabelled at high specific activities.However a number of basic criteria must be fulfilled* before it can be suggested that binding of a radiolabelled ligand in intact tissue or to a subcellular fraction represents selective binding to the receptor in question. These are (i) Specificity concentrations of drugs which are pharmacologically effective at a particular receptor should displace the saturable component of binding while pharmacologically effective concentrations of drugs acting at different receptors should be ineffective. Known structure-activity relationships must also be complied with and the receptor must have a high affinity for the radiolabelled ligand.(ii) Saturability a component of the binding site must saturate with increasing concentrations of the radiolabelled ligand since the number of receptors is finite. (iii) Comparability the apparent binding affinity of the radiolabelled ligand should be close to that necessary for it to act as an antagonist/agonist of the physiological response. (iv) Localization the saturable component of the specific receptor binding should be localized only to those tissues which are known to show the appropriate pharmacological response. A number of receptors have been characterized resulting in some very significant biochemical advances. 1 R. Robinson Chem. in Britain 1974 57. 2 N.M.Birdsall and E.C. Hulme J. Neurochem. 1976,27,7; S. H. Snyder Biochem Pharmacol. 1975,24 1371. 416 Biological Chemistry -Part (v ) Neurochemistry 2 The Opiate Receptor For many years it has been known that the analgesic potency of opiates is highly stereospecific with virtually all the pharmacological activity residing in those isomers with a configuration analogous to that of (-)-morphine. The (+)-isomers have been shown to be pharmacologically inert having neither agonist nor antagonist proper- ties. Goldstein3 suggested that similarly to the opiate analgesics the pharmacologi- cally relevant opiate receptor binding should be stereospecific. It was assumed that there are three kinds of interaction between an opiate and membranes containing opiate receptors (i) a non-saturable interaction associated with the physical solution of the lipophilic opiate molecules in the lipidic membranes; (ii) a non-specific saturable binding associated with ionic interaction between the protonated nitrogen of the opiate and anionic groups on the membrane; and (iii) the stereospecific interaction between the (-)-opiate and the receptor.If the membrane containing the receptors is incubated with a radiolabelled ligand (expt 1) then the bound radioactivity is a measure of the sum of all three types of binding. If initially the membrane is incubated with an excess of unlabelled opiate of the inactive configuration and the radiolabelled ligand added (expt 2) and the bound radioactivity measured then the difference 1-2 measures the non-specific saturable binding as the unlabelled opiate excludes radiolabelled ligand from the non-specific saturable sites.In the third experiment the membrane is initially incubated with a non-radioactive opiate of the (-)-configuration and the radiolabel- led ligand is then added and the bound radioactivity measured (expt 3). The radiolabelled ligand is thus excluded from both the non-specific and stereospecific sites and consequently the difference 2-3 in the bound radioactivity will be a measure of the stereospecific binding. Using this conceptualization it has been ~hown,~ using [3H]naloxone (an opiate antagonist) of high specific activity that stereospecific opiate receptor binding could be observed in homogenates of mouse brain.Levorphanol a potent opiate agonist was shown to reduce the total binding of [3H]naloxone by 70°/0 whereas at similar concentrations its analgesically inactive( +)-enantiomer dextrorphan was without effect on the binding. More re~ently,~ using opiate ligands of higher specific activity >90% of the binding has been shown to be stereospecific. This stereospecific binding affinity has been shown to parallel the pharmacological potency of a wide variety of both opiate agonists and antagonists.6 Similar results have been obtained by Terenius’ and Simon,’ using a number of different radiolabelled antagonists and agonists. Within the brain it has been shown that stereospecific opiate receptor binding is localized to the synaptic membranes.’ Investigations have also been carried out into the regional distribution of the specific opiate receptor within the brain and very considerable variations have been detected in both monkey and human brain with the anterior amygdala containing the highest concentration of binding sites.’ No correlation of opiate A.Goldstein L. Lowney and B. K. Pal Proc. Nut. Acad. Sci. U.S.A. 1971 68 1742. C. B. Pert and S. H. Snyder Science 1973,179 1011. S. H. Snyder ‘Handbook of Psychopharmacology’ Plenum New York 1976 Vol. 5 p. 335. R. S. Wilson M.E. Rogers and S. H. Snyder J. Medicin. Chem. 1975 18 240; C. B. Pert and S. H. Snyder Proc. Nat. Acad. Sci. U.S.A.,1973,70 2243. L. Terenius Acra Pharmacol. Toxicol. 1973,33 377; 1974 34 88. E. J. Simon J. M. Hiller and I. Edelman Proc.Nar. Acad. Sci. U.S.A.,1973 70 1947. 9 M. J. Kuhar C. B. Pert and S. H. Snyder Nature 1973,245,447. 418 J. F. Collins binding with the distribution of acetylcholine catecholamines y -aminobutyric acid or serotonin could be detected while lesions of well-known nerve tracts specific for acetylcholine noradrenaline and serotonin produced no change in specific opiate binding in the affected areas.9 Sodium ions have been shown to be important in mediating the specific binding." Sodium ions considerably enhance the binding of antagonists such as naloxone whereas the binding of agonists such as morphine is markedly decreased Mixed agonist-antagonist drugs e.g. pentazocine are affected by sodium ions in an intermediate fashion. The increase in antagonist binding found on the addition of sodium ions has been shown to be caused not by a change in the receptor affinity but by an increase in the number of receptor sites available.Iodoacetamide," a protein-modifying reagent has been shown to markedly reduce the receptor binding of a series of agonists but does not affect the binding of antagonists. Proteolytic enzymes1* have also been shown to inhibit agonist binding more than antagonist binding whereas bivalent cations especially manganese ions have been shown to have the opposite effect.13 It has therefore been suggested that the opiate receptor can assume two different conformation^,'^ one favouring antagonists the other favouring agonists. The two conformations can be allosterically transformed by sodium ions and it would appear that the agonist conformation is the more labile.Snyder15 has proposed a model of the opiate receptor explaining the structure-activity relationship of opiates and suggesting molecular mechanisms for the interconversion of the receptor between the agonist binding state and the antagonist binding state. The agonist conformation at some sites of action has been shown to inhibit a linked adenylate cyclase enzyme.16 As the opiate analgesics have been shown to interact with stereospecific receptors in certain discrete regions of brain and as it has been shown that none of the established or putative neurotransmitter substances appeared to interact with the opiate receptors it was suggested that the mode of action of the opiates might well involve an unknown mechanism in brain.As many neurotransmitters are derived from amino-acids it was suggested that the endogeneous ligand of the opiate receptor might be a peptide. This theory was substantiated in 1975." In subsequent work it was found that the substance which had been named 'enkephalin' could be extracted from pig brain'' and it was found to have a molecular weight of approxi- mately 1000. Biological activity was found to be very rapidly destroyed if the substance was exposed to proteolytic enzymes and a similar substance was shown to be present in extracts of cat rabbit guinea-pig and cow brains. An important 10 C. B. Pert and S. H. Snyder Mol. Pharmacol. 1974 10 868. C. B. Pert G. W. Pasternak and S. H. Snyder Science 1973,182 1359.11 H. A. Wilson G. W. Pasternak and S. H. Snyder Nature 1975,253,448. ** G. W. Pasternak and S. Snyder Mol. Pharmacol. 1975,11,478. l3 G. W. Pasternak and S. H. Snyder Mol. Pharmacol. 1975 11,735. 14 C. B. Pert and S. H. Snyder Mol. Pharmacol. 1974 10 868; Neurosci. Res. Prog. Bull. 1975 73. 15 A. P. Feinberg I. Creese and S. H. Snyder Proc. Nar. Acad. Sci. U.S.A.,1976,73 4215. 16 S. K. Sharma M. Birenberg and W. A. Klee Proc. Nar. Acad. Sci. U.S.A. 1975,72 590; 0.J. Collier and A. C. Roy Nature 1974,248 24. 17 J. Hughes BrainRes. 1975,88,295;L. Terenius and A. Wahlstrom Acta Physiol. Scand. 1975,94,74. 18 J. Hughes T.W.Smith H. W. Kosterlitz L. A. Fothergill B. A. Morgan and H. R.Morris Nafure,1975 258 577. Biological Chemistry -Part (u)Neurochemistry 419 similarity between morphine and enkephalin was that the effects of both substances in blocking electrically evoked contractions in either the guinea-pig ileum or mouse vas deferens could be reversed by low concentrations of specific morphine antagon- ists such as naloxone.Enkephalin” also displaced radioactively labelled opiates in the binding assay and it was found that the ability of enkephalin to displace the radiolabelled ligands was decreased by the addition of sodium ions suggesting that the compound is an agonist at opiate receptors. [3H]Met-enkephalin has been shown to bind with high affinity to the opiate receptor in the absence of sodium ions. It has also been demonstrated that displacement by morphine or Met-enkephalin of a radiolabelled opiate agonist followed different kinetics from their displacement of labelled antagonists.20 Met-enkephalin has been shown to be much weaker in competing for binding sites for radiolabelled naloxone than in competing for its own binding sites and morphine was much less effective in competing for [3H]enkephalin binding than in the displacement of 3H-opiates.21As a consequence it has been suggested either that there are different binding sites for enkephalins and opiates or that the opiate receptor can exist in multiple receptor conformations.l9 Hughes et al. have purified enkephalin and sequenced the peptide using both the dansyl-Edman technique and mass spectrometric methods. Porcine enkephalin has been found to consist of a mixture of two pentapeptides H-Tyr-Gly-Gly-Phe-Met- OH (Met-enkephalin) and H-Tyr-Gly-Gly-Phe-Leu-OH (Leu-enkephalin) in a ratio of about 3 1.Both compounds have been synthesized,22 and the synthetic peptides have been shown to mimic the actions of naturally occurring enkephalin in the biological test systems. On theoretical grounds G~ldstein~~ has designed and synthesized a heptapeptide H-Try-Gly-Gly-Gly-Lys-Met-Gly-OH, which posses- sed low opiate activity. A large number of synthetic peptides have now been evaluated for their opiate activity and some important structure-activity relationships have emerged (i) when the free amino-group of the tyro~ine~~ residue is blocked25 or removed the opiate activity completely disappears; (ii) phenylalanine cannot replace tyro~ine,~~ hence the phenolic group is also necessary for activity; (iii) a tetrapeptide H-Tyr-Gly-Gly- Phe-OH appears to be the smallest peptide which will interact with the opiate receptor;25 (iv) amidation of the C-terminal of Met-enkephalin considerably increases the and the duration of action; (v) potency is lost when tyrosine is substituted for 4-Phe;24 (vi) both Met-enkephalin and Leu-enkephalin have been found to be very susceptible to hydrolysis by non-specific proteolytic enzymes in 19 R.Simontov and S. H. Snyder in ‘Opiates and Endogeneous Opiate Peptides’ ed. H. W. Kosterlitz North Holland Amsterdam 1976 p. 41; R. Simontov and S. H. Snyder J. Neurochem. 1977,28 13. 20 J. A. H. Lord A. A. Waterfield J. Hughes and H.W. Kosterlitz in ‘Opiates and Endogeneous Opiate Peptides’ ed. H. W. Kosterlitz North Holland Amsterdam 1976 p. 275. 21 N. Birdsall in ‘Opiates and Endogeneous Opiate Peptides’ ed. H. W. Kosterlitz North Holland Amsterdam 1976 p. 19. 22 J. D. Bower K. P. Guest and B. A. Morgan J.C.S. PerkinI 1976,2488; W. Voelter E. Pietrzik and H. Kalbacher Tetrahedron Letters 1976 2 119. 23 A. Goldstein J. S. Goldstein and B. M. Cox Life Sci. 1976 17 1643. 24 N. J. M. Birdsall A. F. Bradbury A. S. V. Burgen E. C. Hulme D. G. Smyth and C. R. Snell Brit. J. Pharmacol. 1976,58 460P. 25 N. Ling and R. Guillemin Prm.Nat. Acad. Sci. U.S.A.,1976 73 3308. 420 J. F. Collins brain and as a consequence analgesia induced by enkephalin is of a very short duration and high doses of enkephalin are required.26 Deactivation has been shown to proceed via cleavage of the Try-Gly amide bond.27 However when 2-Gly is replaced2’ by 2-D-Ala in Met-enkephalin the product after amidation binds almost as tightly as Met-enkephalin to opiate recep- tors.As the compound is not susceptible to enzymic degradation low doses cause long-lasting morphine-like analgesia when microinjected into mouse brain. Similarly when 2-Gly is replaced by 2-D-Ala and 5-Leu by 5-D-LeU in Leu- enkephalin a potent opiate peptide is obtained.28 Other active pentapeptides have been prepared by replacement of 2-Gly by D-proline and D-sarcosine. As the opiate receptor is highly stereospecific for morphine derivatives but clearly not so for enkephalin derivatives a number of investigators have attempted to determine the conformation of enkephalin in solution in an attempt to relate this to the structure of morphine-derived anLigesics.On theoretical grounds it has been postulated that in Met-enke~halin,~~ a P-bend exists involving the two glycine residues in positions 2 and 3 with a hydrogen bond between the carbonyl group of the tyrosine and the amino-group of the phenylalanine. Attention has been drawn to the close similarity of this conformation of enkephalin to the structure of oripavine a potent opiate agonist. The conformation of enkephalin in solution has been investigated by several groups using n.m.r. spectroscopy. One series of experiments using the Karplus- Bystrov curves suggests that for Met-enkephalin in DMSO the methionine amino- proton is involved in a hydrogen bond most probably within a Gly-Gly-Phe-Met type 10 t~x-n.~”~~ Leu-enkephalin has been found to have basically the same conformation in DMS0.31 However other have interpreted relaxation times to suggest that there are no intramolecular hydrogen bonds and that the tyrosine side-chain of Met-enkephalin exhibits restricted motion with respect to the main peptide backbone of the molecule unlike the phenylalanyl and methionyl side-chains which undergo intramolecular reorientation with relatively high fre- quency.It is important to note however that whatever the conformation of enkephalin in solution there may be no correlation between this and the conformation adopted on the receptor.It has also become apparent that there are other large peptides present in brain which are also potent opiate agonists and the generic title ‘endorphin’ has been proposed for them. A number of endorphins have been isolated and most appear to 26 J. B. Chang B. W. T. Fong A. Pert and A. C. Pert LifeSci. 1976,18,1473; J. D. Beluzzi N. Grant V. Garsky D. Sarantakis C. D. Wise and D. Sarantakis Nature 1976 260 604. 27 J. M. Hambrook B. A. Morgan M. J. Rance and C. F. C. Smith Nature 1976,262,782. 28 C.P. Pert A. Pert J. K. Chang and B. T. Fong Science 1976,194,330; M. G. Baxter D. Goff A. A. Miller and I. A. Saunders Brit. J. Pharmacol 1977,59,455P. 29 A. F. Bradbury D. G. Smyth and C. R. Snell Nature 1976,260 165. 30 B. P. Roques C. Garbay-Jaurguiberry R.Oberlin M. Antenius and A. K. Lala Nature 1976,262,778. 31 C. R. Jones W. A. Gibbons and V. Garsky Nature 1976,262,779. 32 H. E. Bleich J. D. Cutnell A. R. Day R. J. Freer J. A. Glasel and J. F. McKelvy Proc. Nat. Acad. Sci. U.S.A.,1976 73 2589. Biological Chemistry -Part (v ) Neurochemistry 42 1 be fragments of p-lipotropin a pituitary ~eptide~~ containing 9 1 amino-acids (see Figure). These fragments are (i) C-fragment of @ -1ipotropin (or p-endorphin) residues 6 1-9 1of p -1ipotropin; (ii) a-endorphin corresponding to residues 61- 76; and (iii) y-endorphin corresponding to residues 61-87. All three exhibit potent opiate agonist activity though -endorphin is much the most active;33 in the opiate binding assay employing rat brain @ -endorphin was found to be 30-fold more potent in displacing [3H]naloxone than Met-enke~halin.~~ @ -Endorphin has been syn- the~ized~~ and the product has been shown to possess activity comparable with that of naturally occurring /3 -endorphin.1 5 10 16 Try -Gly-Gly -Phe-Met -Thr-Ser-Glu-Lys-Ser-Gln-Thr-Pro- Leu-Val-Thr-20 27 31 Leu-Phe-Lys-Asn-Ala-lle-Val-Lys-Asn-Ala-His-Lys-Lys-Gly-Gln-OH Residues 1-5 of p-endorphin Met-enkephalin Residues 1-16 of p -endorphin a-endorphin Residues 1-27 of p -endorphin y-endorphin Figure C-Fragment of p-1ipotropin or @-endorphin The amino-acid sequence of Met-enkephalin corresponds to residues 6 1-65 of p-lipotropin and a-,p- and y-endorphins share this sequence at the N-terminal end. p -Lipotropin however is itself devoid of opiate activity.It has been suggested that one naturally occurring pituitary endorphin is @ -endorphin while the smaller endorphins including Met-enkephalin are formed from this ~eptide.~~ No candidate precursor for Leu-enkephalin has yet been found. It has been that removal of amino-acid residues 30 and 3 1(Gly-Gln) from @-endorphin has little effect on the binding properties but further removal of residues 28 and 29 (Lys-Lys) (leaving y-endorphin) reduces the binding affinity by a factor of 20. It has been suggested that interaction of C-terminal residues including lysines 28 and 29 with the receptor may make an important contribution to the binding of P-endorphin both augmenting the affinity and modifying the binding properties of the N-terminal pentapeptide sequence.Birdsall and co-workers have shown that the tridecapeptide composed of residues 19-3 1 inhibits [3H]naloxone binding despite the absence of the Met-enkephalin sequence thus suggesting that there is a binding site for the C-terminu~.~~ The enkephalins have been suggested to be endogeneous neurotransmitters l9 but this role has been questioned33 and the longer endorphins have been suggested as more plausible candidates. However the existence of enkephalin nerve terminals has been demonstrated (using immumohistochemistry) and their distribution corre- A. Goidstein Science 1976,193 1081; A. F. Bradbury D. G. Smyth C. R. Snel1,N. J. Birdsall andE. C. Hulme Nature 1976 260 793; B. M. Cox A. Goldstein and C. H.Lio Proc. Nut. Acad. Sci. U.S.A. 1976,73 1821; R. Guillemin N. Ling and R. Burgus Compt. rend. 1976,282 C,783; N. J. Birdsall A. F. Bradbury A. S. V. Burgen E. C. Huime D. G. Smyth and C. R. Snell Brit J. Pharmacof. 1976,58 460P. A. A. Waterfield J. Hughes and H. W. Kosterlitz Nature 1976,260,624; L. F. Tseng H. H. Loh and C. H. Li Proc. Nut. Acad. Sci. U.S.A.,1976.73 4187. C. H. Li S. Lemaire D. Yamashiro and B. A. Doneen Biochem. Biophys. Res. Comm. 1976 71 19. 422 J. F. Collins lates well with the opiate receptors as determined using a~toradiography.~~ P -Endorphin has been found to occur in mammalian brain but the concentrations found are very much lower than those of the enkephalin~,~~ and it has been suggested that whereas the pituitary contains high levels of the larger peptides and has low enkephalin levels the brain contains high levels of enkephalins but low levels of larger endorphins.l9 Both the enkephalins and P -endorphin have been shown to produce tolerance and eventually dependence when given intra~erebrally.~~ However P -endorphin has been shown to induce much more profound analgesis than the enkephalins when given intraventricularly but this may be due to the rapid inactivation of enkepha- 1i1-1.~~ Cross-tolerance has also been observed both between enkephalin and mor- phine and P -endorphin and Snyder39 has shown that increased concen- trations of enkephalin can be detected in morphine-dependent rats. This increase may be due to pre-synaptic synthesis of enkephalin continuing in the absence of release.Endorphins other than those derived from P -endorphin have been dete~ted.~~.~’ These are different in that (i) their biological activity is destroyed by trypsin which is not the case for those derived from p-lipotropin; (ii) even in impure preparations they are considerably more potent than the P-lipotropin endorphins; (iii) the biological activity of these endorphins is insensitive to cyanogen bromide treatment indicating the absence of an essential methionine residue. Wahl~trom~’ has also isolated a peptide with opiate activity from human cerebrospinal fluid which is chemically different from all known fragments of P -1ipotropin. A blood-extracted substance of unknown structure but of low molecular weight named an~dynin,~~ both competes for opiate receptor binding and is a potent analgesic when injected directly into cat brain.Anodynin has been shown to be absent from the blood of hypophysectomized rats and consequently is thought to derive from the pituitary. 3 The Cholinergic Receptor Acetylcholine an excitatory neurotransmitter has been shown to act at two different types of post-synaptic receptor namely nicotinic and muscarinic receptor^.^^ Although no successful attempts have yet been made to characterize the nicotinic receptor in mammalian brain tissue the nicotinic receptors present at the neuromus- cular junction have been successfully identified44 using elapid toxins. In particular a-neurotoxins and polypeptide toxins such as a-bungarotoxin isolated from the venom of Bungarus multicinctus have been extensively used in the receptor isolation studies from various tissues e.g.the electric organs of the Torpedo and electric eel 36 R. Elde T. Kokfelt 0.Johansson and L. Terenius Neuroscience 1976 1,349. 37 A. Lampert M. Nirenberg and W. A. Klee Proc. Nut. Acad. Sci. U.S.A.,1976,73,3165; J. M. Van Ree D. De Wied A. F. Bradbury E. C. Hulme D. G. Smyth and C. R. Snell Nature 1976,264 792. 38 H. H. Loh L. F. Tseng E. Wei and C. H. Li Proc. Nut. Acad. Sci. U.S.A.1976 73 2895. 39 R. Simantov and S. H. Snyder Nature 1976,262 505. 40 B. M. Cox S. Gentleman T. P. Su and A. Goldstein Bruin Res. 1976 115 285. 41 A. Wahlstrom in ‘Opiates and Endogeneous Opiate Peptides’ ed. H. W. Kosterlitz North Holland Amsterdam 1976 p.49. 42 C. B. Pert A. Pert and J. H. Tallman Proc. Nut. Acud. Sci. U.S.A.1976,73 2226. 43 R. B. Barlow ‘Introduction to Chemical Pharmacology’ Methuen London 1964. 44 J. P. Changeux in ‘Proceedings of the Sixth International Congress of Pharmacology’ ed. F. Klinge Pergamon Press Oxford 1976 p. 165. Biological Chemistry -Part (u)Neurochemistry 423 and striated muscle. The successful use of these toxins as essential tools in the identification localization and isolation of the nicotinic acetylcholine receptor is due to two factors (i) the action of the toxin is very specific and (ii) the snake toxin associates in a very tight but reversible manner with the receptor so that the rate of dissociation is very low. The polypeptide toxin can be labelled in a number of different ways the most important employing 1251for labelling.The radiolabelling is of course essential for localization studies and also for receptor purification. The solubilized receptor has been purified using affinity chromatography in which the a-neurotoxin was conjugated to an agarose The receptor has been characterized as a and the molecular weight has been estimated at 360 000. It has been solubilized using Triton X-100.46 Extensive reviews have been published on the biochemical characterization of the nicotinic re~eptor.~’ Myas-thenia gravis has been shown to be caused by an autoimmune response to the nicotinic acetylcholine Unlike nicotinic receptors muscarinic receptors have been successfully studied in the mammalian brain.High affinity binding sites have been demonstrated to exit using a large number of radiolabelled muscarinic antagonists. These include the potent antagonist 3-quinuclidinyl benzilate (QNB),49 atr~pine,~’ benzetimide,” and propylbenzilylcholine (PrBCh).” The specific binding of C3H]QNB was found to be almost completely blocked by excess cold QNB indicating saturable binding. The specific binding defined as the total binding minus the binding in the presence of 0.01 pmol I-’ unlabelled QNB is saturable with respect to [3H]QNB and tissue concentration and is dependent on time temperature and pH.49 Other muscarinic antagonists and agonists were shown to displace specific [3H]QNB binding,49 but nicotinic antagonists showed no affinity for the QNB binding sites.For a given 3H-antagonist the maximum degree of displacement of binding is the same for all competing muscarinic antagonists or agonists. The level of binding in synaptosomal fractions derived from rat brain has been shown to be similar for all the 3H-antagonists so far studied and to be in the range 1.6-2.2 nmol per g pr~tein.’~*~* A close correlation has also been observed between the pharmacological potency of a large number of cholinergic agents and their ability to inhibit radiolabelled antagonist binding in guinea-pig ileum.52 It has been shown that the antagonist binding sites in very diverse tissues such as smooth muscle the parotid gland and brain show very similar affinity constants which do not display variations between different mammalian species.52 The binding 45 J.P. Changeux M. Kasai and C. Y. Lee Proc. Nat. Acad. Sci. U.S.A.,1970 67 1241; R. Miledi P. Molinoff and L. T. Potter Nature 1971,229,554; R. P. Klett B. W. Fulpius D. Cooper and E. Reich in ‘Synaptic Transmission and Neuronal Interaction’ Raven Press New York 1974 pp. 179. 46 J. Lindstrom and J. Patrick in ‘Synaptic Transmission and Neuronal Interaction’ Raven Press New York 1974 p. 191. 47 M. A. Raftery. J. Deutch K. Reed R. Vandlen and T. Lee in ‘Proceedings of the Sixth International Congress of Pharmacology’ ed. F. Klinge Pergamon Press Oxford 1976 p. 87; Z. Vogel and M. P. Daniels ibid. p. 59; F. A. Barnard and J. 0. Dolly ibid. p. 77; C. Y. Lee in ‘Neuropoisons; their Pathophysiological Actions’ ed.L. L. Simpson Plenum New York 1971 Vol. 1 p. 21. 48 A. Aharonov R. Tarrab-Hazdai 0.Abramsky and S. Fuchs Proc. Nut. Acad. Sci. U.S.A. 1975 72 1456. 49 H. I. Yamamura and S. H. Snyder Proc. Nut. Acad. Sci. U.S.A.,1974,71 1725. J. T. Farrow and R. D. O’Brien Mol. Pharmacol. 1973,9 33; A. J. Beld and E. J. Ariens European J. Phannacol. 1974,30 360. 51 N. J. Birdsall A. S. Burgen C. R. Hiley and E. C. Hulme J. Suprumol. Structure 1976 4 367. s2 N. J. Birdsall and E. C. Hulme J. Neurochem. 1976 27 7. 424 J. F. Collins curves observed by a number of the radiolabelled antagonists are exactly of the form expected when the law of mass action is applied to the interaction of a ligand with a single uniform set of binding sites. The regional distribution of muscarinic antagonist sites in brain has been examineds3 using [3H]QNB and the caudate nucleus and the cerebral cortex have been found to contain high concentrations of muscarinic receptor sites.In contrast to the simple antagonist binding studies have revealed that muscarinic agonist binding is complex and cannot be described by a single affinity constant,52 and it has been suggested that a heterogeneous population of agonist binding sites exists. It has been postulated that the heterogeneity of the agonist binding sites can be explained on the basis that (i) two classes of agonist binding site exist one having a high affinity constant and the other a low affinity constant for agonist binding; (ii) the antagonist binding sites linked to the two types of agonist site have identical affinities for a given antagonist; and (iii) binding of agonists and antagonists is competitive and mutually exclusive.Similar suggestions have been made for a number of other receptor systems including the opiate receptor and the glycine re~eptor.’~ The muscarinic receptor has not yet been solubilized using detergents as binding activity is de~troyed.’~ However solubilization using digitonin has been reported.” 4 Catecholamine Receptors Although specific P -adrenoceptor binding in tissues has proved difficult to observe several groups55 have now managed to demonstrate its existence. It was necessary to use potent high specific activity P-adrenoceptor antagonists as ligands. Typical antagonists used were D~-[~H]proprananol (-)-[3H]alprenolol and [‘251]hydroxybenzylpindolol.All have very high binding constants ca.lo9 1mol-’. Initially avian or amphibian red blood cell membranes were used as the source tissue for binding studies but more recently 0-adrenergic receptors have been charac- terized in rat brain using (-)-[3H]alprenolol.56 The binding was found to be specific very rapid saturable and reversible. It was found that the (-)-isomers of the unlabelled antagonists propranolol and isoprenaline were highly potent in compet- ing for the binding sites in brain the (+)-inactive-isomers being about two orders of magnitude less potent. Binding could also be inhibited by other P-adrenoceptor antagonists and by catecholamines such as (-)-adrenaline and (-)-noradrenalhe.At high concentrations phentolamine an a -adrenoceptor antagonist displaced (-)-[’H]alprenolol but this is thought to be similar to the non-specific inhibitory effect of high concentrations of phentolamine on 0-adrenergic physiological responses. The highest number of [3H]alprenolol binding sites was found to be in the hippocampus and limbic forebrain areas of the brain (0.26 and 0.28 pmol per mg protein respectively) but there appeared to be no correlation between levels of noradrenaline and the number of P-adrenergic receptors in various regions of the 53 S. Yamamura M. Kuhar and S. Snyder Bruin Res. 1974,66 541; 1974,80 170. 54 A. Young and S. H. Snyder Roc.Nut. Acad. Sci. U.S.A.,1973,70,2832. 55 G.D.Aurbach S. A. Fedak C. J. Woodard J.S. Palmer D. Hauser and F. Troxler Science 1974,186 1223;R.W. Alexander L. T. Williams and R. J. Lefkowitz,Proc. Nut. Acud. Sci. U.S.A.,1975,72,1564; A. Levitzki D.Atlas and M. J. Steer ibid. 1974 71 2773 4246. 56 R. W. Alexander J. N. Davis and R. J. Lefkowitz Nature 1975,258 437. Biological Chemistry -Part (v)Neurochemistry 425 brain. Similar results have been obtained using DL-[3H]propranolol in chick brain.57 Stereospecific binding of [3H]dopamine to membranes from the corpus striatum of rat brains has been dem~nstrated.~~ The stereospecific component of the binding was defined as the amount of [3H]dopamine bound in the presence of (-)-butaclamol (an inactive compound) minus that bound in the presence of (+)-butaclamol (a potent neuroleptic compound).A number of antipsychotic drugs the most potent being spiroperidol inhibited the stereospecific binding. The inhibitory potencies correlated well with clinical doses used in the treatment of schizophrenia. Snyder” and co-workers have also shown using as a definition of specific dopamine binding the total minus the binding in the presence of a large excess of non-radioactively labelled dopamine that the binding is saturable and that apomorphine is about twice as potent as dopamine in competing for binding sites. [3H]Apomorphine has recently been shown to be a better radioligand than [3H]dopamine or [3H]haloperidol.60 Specific binding of the a -noradrenergic agonist [3H]clonidine and the antagonist [3H]-2-(2’6’-dimethoxyphenoxyethylamino)methylbenzodioxan(WB 4101) has been shown to occur in rat brain.61 Both bind with high affinity and selectivity and the binding was found to be saturable.In exchange experiments a-agonists have much higher affinities for the [3H]clonidine sites than for 3H-WB 4101 sites whereas the reverse has been shown to hold for a -antagonists. Mixed agonist-antagonist drugs have been shown to have a similar affinity for binding for the two radiolabelled ligands. It has been suggested that [3H]clonidine and [3H]WB 4101 respectively label separate agonist and antagonist states of the receptor and that the receptor exists in two conformations with selective affinities for agonists and antagonists.61 5 Amino-acid Receptors A number of amino-acids have been suggested to act as neurotransmitters.Amongst these are L-glutamate glycine and y -aminobutyric acid (GABA). L-Glutamate has been suggested to be an excitatory neurotransmitter in the mammalian central nervous system62 as it has been found to be a potent excitant when applied microionophoretically to single neurones in certain brain areas. Certain neurones in the CNS also appear to be enriched in gl~tamate.~~ The postsynaptic glutamate receptor has been studied using [3H]kainic Kainic acid a conformationally restricted analogue of L-glutamate is reported to be a powerful excitant which potentiates the excitatory action of glutamate.6s It has S7 S. R. Nahorski Nature 1976 259 488. s8 P. Seeman M. Chau-Wong J. Tedesco and K. Wong Proc. Nut. Acad. Sci. U.S.A. 1975,72,4376.s9 D. R. Burt S. J. Enna I. Creese and S. H. Snyder Proc. Nut. Acad. Sci. U.S.A. 1975,72 4655. 60 P. Seeman T. Lee M. Chau-Wong J. Tedesco and K. Wong Proc. Nut. Acad. Sci. U.S.A. 1976 73 4354. 6l D. A. Greenberg D. C. U. Prichard and S. H. Snyder Life Sci. 1976,19,69. 62 D. R. Curtis and G. A. R. Johnston Ergebn. Physiol. 1974,69,98; K. Krnjevic Physiol. Rev. 1974,54 419. 63 J. A. Harvey C. N. Scholfield L. T. Graham and M. H. Aprison J. Neurochem. 1975,24,445. 64 J. R. Simon J. F. Contrera and M. J. Kuhar J. Neurochem. 1976 26 141. 65 H. Shinozaki and K. Konishi Brian Res. 1970 24 368. 426 J. F. Collins been shown to be very much more potent than L-glutamate in both frog and cat spinal neurones. Synaptosomal uptake of kainic acid has been shown to be minimal and it is not a potent inhibitor of high afFinity L-glutamate uptake.For these reasons it appeared to be a suitable agonist to use in labelling the glutamate receptor. The specific binding of C3H]kainic acid to synaptic membranes was shown to be saturable with a dissociation constant of about 60 nmol l-'.64 Specific C3H]kainic acid binding was obtained by subtracting from the total bound radioactivity the amount not displaced by high concentrations of kainic acid or glutamate. The maximum number of binding sites was found to be about 1pmol per mg protein. Quisqualic acid a structural analogue ofL-glutamate was found to be the most potent displacer of specific r3H]kainic acid binding with a potency of about one-third that of kainic acid itself.L-Glutamate was found to be less potent in binding displacement than either kainic acid or quisqualic acid with a potency of about 1/25 that of kainic acid. Other compounds tested such as L-aspartate a putative excitatory neurotransmit- ter were found to be very much less potent. The specific binding appears to be localized to grey matter. On this evidence it has been suggested that [3H]kainic acid is binding to the glutamate receptor. Glycine66 has been shown to be a major inhibitory neurotransmitter in the mammalian CNS and is thought to be especially important as a transmitter in the spinal cord. Strychnine acts as a potent and selective antagonist for the synaptic actions of gly~ine.~~ The glycine receptor has been studied68 using [3H]strychnine rather than [3H]glycine as use of [3H]strychnine avoids binding of glycine to presynaptic membrane fragments associated with the glycine uptake system and also since pharmacological studies had suggested that strychnine had a much greater affinity for the glycine receptor than glycine itself.Specific strychnine binding to synaptic membrane preparations of spinal cord and brain stem of the rat and monkey has been demonstrated using high specific activity strychnine. The binding has been found to be maximal in the synaptic membrane fraction containing both pre- and post-synaptic membrane fragments. Bound [3H]strychnine has been shown to be displaced by glycine and the specific binding shown to be saturable. In addition [3H]strychnine has been shown to bind to a single population of receptor sites.The number of such binding sites in the rat spinal cord is ca. 39 pmol g-'. The regional distribution of strychnine binding has been examined and has been found to be highest in the spinal cord and the medulla oblongata-pons. Negligible binding was observed in the cerebellum hippocampus and cerebral cortex. The distribution has been found to parallel closely that of endogenous glycine and the high affinity glycine uptake system. In further confirma- tion of the specificity of the strychnine binding the ability of a series of amino-acids to displace bound [3H]strychnine was compared with their ability to mimic the strychnine-antagonized neurophysiological actions of glycine. p -Alanine and glycine have been found to be the most potent amino-acids both neurophysiologi- cally and in the displacement of [3H]strychnine.As in the case of the opiate receptor the agonist-glycine and the antagonist- strychnine appear to bind to distinct but mutually interacting portions of the 66 S. H. Snyder Brit. J. Pharmacol. 1975 53 473. 67 D. R. Curtis A. Duggan and J. Johnston Exp. Brain Res. 1971 12 547. 68 A. B. Young and S. H. Snyder Proc. Nat. Acad. Sci. U.S.A.,1973,70 2832. Biological Chemistry -Part (v)Neurochemistry 427 receptor.69 It has been demonstrated that protein-modifying reagents such as diazonium tetrazole and acetic anhydride while having little affect on the total amount of [3H]strychnine bound to synaptic membranes interfere with the ability of glycine to displace [3H]strychnine but not with the ability of non-radioactive strychnine to displace the binding.Other protein-modifying reagents such as tetranitromethane and dinitrofluorobenzene have been shown to inhibit the total amount of binding of [3H]strychnine but unlabelled glycine and strychnine still displace the remaining bound [3H]strychnine to similar extents. The differential influence of these reagents suggests that glycine displaces strychnine by acting at a site other than the strychnine binding site. The ability of a series of anions to inhibit [3H]strychnine binding in spinal cord synaptic membranes has been e~amined,~’ and it has been suggested that the observed inhibition is closely correlated with the ionic conductance mechanism for the chloride ion in the glycine receptor.Young and Snyder have postulated that glycine binds to its recognition site. This site is composed of regulatory subunits associated with the conductance-modulating site where it is suggested strychnine may bind. A correlation between the clinical efficacy of a series of benz~diazepines~’ and the ability of the compounds to reduce the specific strychnine binding to glycine receptors has given rise to the suggestion that benzodiazepines exert their phar- macological activities by interacting with the glycine receptor. However from in vivo studies with the cat,72 no interaction could be demonstrated between diazepam strychnine and glycine. y-Aminobutyric acid (GABA) has been shown to be a major inhibitory neuro- transmitter in the mammalian CNS.73 Its inhibitory actions are selectively antagonized by the phthaleido-isoquinoline alkaloid (+)-bic~culline,~~ in contrast to the other inhibitory amino-acid neurotransmitter glycine.The inactivity of (-)-bicuculline shows that antagonism to be stereospe~ific.’~ Specific [3H]GABA binding has been examined76 using synaptosomal fractions from rat brain. The binding measured in the absence of sodium ions is selective and saturable. Various amino-acids have been shown to inhibit the [3H]GABA binding and their ability to do this parallels the synaptic inhibitory actions of GABA and does not correlate with their relative affinity for presynaptic high-affinity GABA uptake system. Bicuclline has been shown to inhibit [3H]GABA binding with half maximal effects at 5 pmol 1-’.Thus it would appear that bicuculline has ca. 1/50 the affinity of GABA for the GABA binding sites in synaptic membranes in contrast to strychnine the glycine antagonist which has 10 000 times the affinity of glycine for the glycine receptor. Specific [3H]GABA binding has been shown to be highest in the cerebellum and least in the spinal cord. 69 A. B. Young and S. H. Snyder Mol. Pharmacol. 1974,10,790. 70 A.B.Young and S. H. Snyder Proc. Nut. Acad. Sci. U.S.A.,1974,71,4002. 71 A.B.Young S. R. Zukin and S. H. Snyder Proc. Nut. Acad. Sci. U.S.A. 1974.71 2246. 72 D. R.Curtis C. J. A. Game and D. Lodge Brit. J. Pharmacol. 1976,56,307. 73 D.R.Curtis and G. A. R. Johnston in ‘Handbook of Neurochemistry’ ed.A. Lajtha Plenum Press New York 1970,Vol. 4,p. 115;K. Krnjevic Physiof.Rev. 1974,54,418. 74 D. R. Curtis A. W. Duggan D. Felix and G. A. R. Johnston Brain Res. 1971,33 57. 75 J. F. Collins and R. G. Hill Nature 1974,249 845. 76 S.R.Zukin A. B. Young and S. H. Snyder Proc. Nut. Acad. Sci. U.S.A. 1974,71,4802;S.J. Enna and S. H. Snyder Brain Res. 1975,100,81;S.J. Enna M. J. Kuhar and S. H. Snyder ibid. 1975,93,168. 428 J. F. Collins It has been suggested” that two different binding sites exist on the GABA receptor as the effects of GABA antagonists in the superior cervical ganglion in the rat could be reversed by the application of hypnotic barbiturates whereas pentobar- bitone had no effect on r3H]GABA binding in rat brain synaptosomal preparation^.^^ 77 N.G.Bowery and A.Dray Nature 1976,264,276.