ELSEVIER REVIEW Atypical Neuroleptics Have Low Affinity for Dopamine D2 Receptors or Are Selective for D 4 Receptors Philip Seeman, M.D., Ph.D., Roy Corbett, Ph.D., and Hubert H. M. Van Toi, Ph.D. This review examines the possible receptor basis of the For example, clozapine revealed a radioligand-independent atypical action of those atypical antipsychotic drugs that value of 1.6 nM at the dopamine D4 receptor, agreeing with elicit low levels of Parkinsonism. Such an examination the value directly measured with [3H]-clozapine at D4. requires consistent and accurate dissociation constants for However, because clozapine competes with endogenous the antipsyclwtic drugs at the relevant dopamine and dopamine, the in vivo concentration of clozapine (to occupy serotonin receptors. It has long been known, however, that dopamine D4 receptors) can be derived to be about 13 nM, the dissociation constant of a given antipsychotic drug at agreeing with the value of 12 to 20 nM in the plasma water the dopamine D2 receptor varies between laboratories. or spinal fluid observed in treated patients. The atypical Although such variation depends on several factors, it has neuroleptics remoxipride, clozapine, perlapine, seroquel, recently been recognized that the radioligand used to and melperone had low affinity for the dopamine D2 measure the competition between the antipsychotic drug receptor (radioligand-independent dissociation constants of and the radioligand is an important variable. The present 30 to 90 nM). Such low affinity makes these latter five review summarizes information on this radioligand drugs readily displaceable by high levels of endogenous dependence. In general, a radioligand of low solubility in dopamine in the caudate or putamen. Most typical the membrane (i.e., low tissue:buffer partition) results in a neuroleptics have radioligand-independent values of 0.3 to low value for the antipsychotic dissociation constant when 5 nM at dopamine D2 receptors, making them more the drug competes with the radioligand. Hence, by first resistant to displacement by endogenous dopamine. Finally, obtaining the antipsychotic dissociation constants using a relation was found between the neuroleptic doses for rat different radioligands of different solubility in the catalepsy and the D2:D4 ratio of the radioligand­ membrane, one can then extrapolate the data to low or independent K values for these two receptors. Thus, the "zero" ligand solubility. The extrapolated value represents atypical neuroleptics appear to fall into two groups, those the radioligand-independent dissociation constant of the that have a low affinity for dopamine D2 receptors and those antipsychotic. These values are here given for dopamine D2 that are selective for dopamine D4 receptors. © 1997 and D4 receptors, as well as for serotonin 5-HT2A receptors. American College of Neuropsychopharmacology These values, rnoreover, agree with the dissociation constant [Neuropsychopharmacology 16:93-110, 1997] directly obtained with the radioactive antipsychotic itself. ------ KEY WORDS: Dopamine D4 receptors; Neuroleptics; Serotonin 2 receptors; Parkinsonism; Clozapine; Catalepsy ------------ From the Departments of Pharmacology (PS, HHMVT) and of Address correspondence to: Philip Seeman, M.D., Ph.D., Depart­ Psychiatry, and the Molecular Neurobiology Laboratory (HHMVT) ment of Pharmacology, Medical Science Building, Room 4344, Uni­ University of Toronto, Toronto, Canada; and the Department of Bio­ versity of Toronto, Toronto, Ontario, Canada M5S 1A8. logical Research (RC), Hoechst-Roussel Pharmaceuticals Inc., Som­ Re~eived January 30, 1996; revised May 19, 1996; accepted May erville, NJ. 27, 1996. NEUROPSYCHOPHARMACOLOGY 1997-VOL. 16, NO. 2 © 1997 American College of Neuropsychopharmacology Published by Elsevier Science Inc. 0893-133X/97 /$17.00 655 Avenue of the Americas, New York, NY 10010 PI! S0893-133X(96)00187-9 94 P. Seeman et al. NEUROPSYCHOPHARMACOLOGY 1997-VOL. 16, NO. 2 RECEPTOR HYPOTHESES FOR ATYPICAL membrane also determines the membrane:buffer parti­ NEUROLEPTIC ACTION tion coefficient of neuroleptics (Kwant and Seeman 1969), as does the membrane lipid composition (Kwant and The blockade of dopamine D2 receptors alleviates psy­ Seeman 1971). Many of these physicochemical factors, chosis but also produces Parkinsonism (Seeman et al. including temperature, that affect the partition of the 1974, 1975; Creese et al. 1976; Seeman 1992a, 1995a). An­ neuroleptic in the membrane have been previously re­ tipsychotic drugs (neuroleptics) that elicit few or none viewed (Seeman 1972). As for the role of physicochemi­ of the extrapyramidal signs of Parkinsonism are re­ cal factors on the dissociation constant of agonists and ferred to as atypical antipsychotic drugs (Meltzer and antagonists, the sodium and magnesium ion concentra­ Nash 1991). What is the receptor basis for this atypical tions determine the proportion of the dopamine D2 re­ action of these particular neuroleptics? There are sev­ ceptors that are in the high-affinity state (Grigoriadis eral current views: and Seeman 1985; Watanabe et al. 1985). Sodium also is essential for the high-affinity binding of benzamide 1. Some atypical neuroleptics may have a low affinity neuroleptics to dopamine receptors (Jarvie et al. 1987). for dopamine D2 receptors and may thus be readily Nevertheless, despite standard laboratory experi­ displaced by high endogenous concentrations of mental conditions internationally (e.g., pH of 7.4 and dopamine in the caudate or putamen. physiological concentrations of Na+, Mg++, and other 2. Atypical neuroleptics may block both dopamine D 2 ions), it has long been known that the dissociation con­ receptors and muscarinic receptors. stant of a particular neuroleptic may vary considerably 3. Atypical neuroleptics may have a balanced block of between laboratories, particularly when different radio­ dopamine D2 receptors and serotonin2A (5-HT2A) re­ ligands are used. For example, the dissociation constant ceptors (Meltzer 1989, 1995; Leysen et al. 1994; Hut­ for clozapine at the dopamine D receptor is approxi­ tunen 1995). 2 mately 150 nM (range 70-400 nM), when using [3H]­ 4. Atypical neuroleptics may selectively block dopa­ spiperone as a radioligand (Seeman 1992a). However, mine D receptors. 4 when [3H]-raclopride is used as the radioligand, the dis­ We examined these various hypotheses after first sociation constant for clozapine at D2 is between 35 and considering the values for the neuroleptic dissociation 60 nM (Malmberg et al. 1993). constants at the dopamine D2, D4, and 5-HT2A receptors. Recently, therefore, this dependence of the neurolep­ tic dissociation constant on the radioligand was studied in more detail (Seeman and Van Tol 1995; Seeman 1995b). It was found that the neuroleptic dissociation THE NEUROLEPTIC DISSOCIATION constant depended on the tissue:buffer partition coeffi­ CONSTANT VARIES WITH THE TISSUE: cient of the radioligand. A similar finding has recently BUFFER PARTITION OF THE RADIOLIGAND been made by Durcan et al. (1995). For example, clozapine at the D2 receptor revealed a To derive the therapeutic receptor-blocking concentra­ dissociation constant of 390 nM with [3H]-nemonapride, tions of neuroleptics and to assess the different hypoth­ 186 nM with [3H]-spiperone, and 83 nM with [3H]-raclo­ eses for atypical neuroleptic action, it is essential to use pride. Haloperidol also had a dissociation constant of 3 neuroleptic dissociation constants that have been mea­ 9.6 nM at dopamine D2 receptors using [ H]-nem­ sured with the minimum experimental artifacts. onapride, 2.7 nM using [3H]-spiperone, and 0.67 nM us­ For example, it has been found that the dissociation ing [3H]-raclopride. These neuroleptic dissociation con­ constant of a neuroleptic (e.g. [3H]-spiperone) can range stants were related to the tissue:buffer partition coef­ from 30 pM up to 1,600 pM as the final concentration of ficients of the radioligands, as shown in Figures 1 and 2. tissue is increased (Seeman et al. 1984). Currently, how­ ever, the final concentrations of tissue in most laborato­ ries is kept very low to minimize tissue dependence. Although many other factors may also determine the THE RADIOLIGAND-INDEPENDENT dissociation constant of an antipsychotic drug at various DISSOCIATION CONSTANT receptors, the literature on this topic is not extensive. The pH, for example, determines not only the surface It is possible to eliminate the dependence of the neuro­ activities (Seeman and Bialy 1963) and the membrane: leptic dislocation constant on the radioligand and thereby buffer partition coefficients of neuroleptics (Seeman obtain the dissociation constant of the neuroleptic in the 1966a, 1966b; Seeman and Weinstein 1966; Seeman and absence of any competing radioligand. This may be Kwant 1969), but also the dissociation constant of the done by extrapolating the relation shown (Figure 2) neuroleptic at the dopamine D2 receptor (e.g., sulpiride; down to either unity or zero partition, yielding an inter­ Presland and Strange 1991). The surface charge of the cept. This intercept represents the dissociation constant NEUROPSYCHOPHARMACOLOGY 1997-VOL. 16, NO. 2 Atypical Neuroleptics and DA D2 and D4 Receptors 95 3000 2500 /~ .... ,.--~ Q) K = 1.2 nM - 2000 <;>)('I - 'o c.. '°~- K = 7.3 nM Cl 1500 1000 HUMAN D2 LONG 0.01 0.1 10 100 1000 10µM nM Haloperidol S-Sulpiride 15 10 P = 14 pmol/g 5 nM El 0 E C. 0 "C 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 C :J [3H]Spiperone, nM 0 Ill 30 0 20 pmol/g 10 • e~\\\C nM 1 2 3 4 5 6 7 [3H]Raclopride, nM Figure 1. Top, a representative experiment showing that the haloperidol dissociation constant (or inhibition constant) was high (7.3 nM) when competing versus PH]-nemonapride, but lower when competing versus PH]-spiperone (2.8 nM) and even lower when using [3H]-raclopride (1.2 nM). These K values were derived by the Cheng-Prusoff (1973) equation, using the appropriate KJ value for each ['HJ-ligand (adapted from Seeman and Van Tol 1995). (The Hill coefficient for the competi­ tion of haloperidol with each radioligand was not significantly different from unity).