In vivo binding to dopamine receptors: a correlate of potential antipsychotic activity

European Journal of Pharmacology, 215 (1992) 29-34

29

g> 1992 Elsevier Science Publishers B.V. All rights reserved 0014-2999/92/$05.00

EJP 52390

In vivo binding to dopamine receptors: a correlate of potential antipsychotic activity R o b e r t D. M c Q u a d e , R u t h A. Duffy, Vicki L. Coffin a n d A l l e n B a r n e t t Schering-Plough Research, 60 Orange Street, Bloomfield. NJ 07003, U.S.A.

Received 24 October 1991, revised MS received 14 January. 1902, accepted 21 January 1992

Antagonists of dopamine receptors (especially those of the D e subtype) have long been recognized as effective antipsychotics. SCH 39166, a dopamine D I selective antagonist, is now also being evaluated for its clinical antipsychotic properties. The studies described herein determine the binding affinity of a variety of dopamine receptor antagonists (both dopamine D I and D e selective compounds) for the dopamine D 1 and D z receptors, in vivo, and correlate this affinity with their behavioral activity in the rat conditioned avoidance response (CAR) test. The in viw) binding affinities of the D i selective compounds at the dopaminc D 1 site exhibited a high correlation (r = 0.97) with their activities in the rat CAR test. Likewise, D 2 selective compounds' inhibition of in vivo binding to dopaminc D 2 receptors correlated with their behavioral potencies (r = 0.98). Conversely, any binding of selective agents to their non-targeted receptor did not correlate with their behavioral activity. These data suggest that in vivo binding to either dopamine D 1 and/or D e receptors is predictive of potential antipsychotic efficacy. Conditioned avoidance response; Dopamine D~ receptors; Dopamine D 2 receptors; (In vivo binding)

1. Introduction It is well recognized that the standard antipsychotic medications used in the clinic today are all antagonists of the dopamine D 2 receptor. As such, the correlation between binding to dopamine D 2 receptors and clinical efficacy has been described previously (Seeman, 1987). However, new potential antipsychotic therapies are being developed which are not antagonists of the dopamine D 2 receptor, but rather antagonize the dopamine D~ receptor (Chipkin et al., 1988; Kerkman et al., 1988; Andersen et al., 19881. Among these compounds is the dopamine D 1 selective antagonist, SCH 39166, which is currently undergoing clinical testing (Chipkin, 1990). All dopamine receptor antagonists which are either available clinically or are undergoing clinical testing, exhibit specific preclinical behavioral activities. The primary relevant activity is the ability of the compound to inhibit the rat conditioned avoidance response (CAR). The antipsychotic class of therapeutics inhibit this behavior at doses which do not result in a complete inhibition of motor activity and this behavioral

Correspondence to: R.D. McQuade, Schering-Plough Research, 611 Orange Street, Bloomfield, NJ 07003, U.S.A. Tel. 1.201.429 3840, fax 1,201.429 44111.

paradigm is considered to be predictive of potential antipsychotic activity in humans (Davidson and Weidley, 1976; Cook and Davidson, 1978). SCH 39166 exhibited this activity in rats with a minimum effective dose of 10 m g / k g p.o. (Chipkin et al., 1988). Further, this compound has been demonstrated to be dopamine D~ selective both in vitro (Chipkin et al., 1988; McQuade et al., 1991b) and in vivo (McQuade et al., 1991a,b) and appears less likely to produce the abnormal movements associated with dopamine D 2 antagonists (Coffin et al., 1989, in press; McHugh and Coffin, 1991). The current study examines in greater detail the relationship between antagonism of dopamine D~ receptors in vivo and activity in the CAR test. Similar studies were also conducted with dopamine D 2 and mixed D I / D 2 antagonists.

2. Materials and methods 2.1. Materials

The following compounds were synthesized at Schering-Plough Research, under the supervision of Dr. Joel Berger: SCH 39166 ( ( - ) - t r a n s - l l - c h l o r o 6,6a,7,8,9,13b-hexahydro-7-methyl-5H-benzo[d]naphtho[2,1-b]-azepin-12-ol), SCH 23390 (R-(+)-8-chloro2,3,4,5-tetrahydro-3-methyl-5-phenyl-1 H-3-benzazepine

30

7-ol), SCH 15198 (R-(+)-2,3,4,5-tetrahydro-8-methoxy-3-methyl-5-phenyl-1H-3-benzazepine-7-ol), SCH 23389(R-( + )-2,3,4,5-tetl'ahydro-3,8-dimethyl-5-phenyl1H-3-benzazepine-7-ol), SCH 38840 (R-(+)-2,3,4,5tetrahydro-8-iodo-3-methyl-5-phenyl- 1H-3-benzazepine -7-ol), SCH 40853 ((-)-trans-ll-chloro-6,6a,7,8,9,13bhexahydro-5 H-benzo[d]naphtho[2,1-b]-azepin- 12-ol) and SCH 12679 (R-(+)-2,3,4,5-tetrahydro-7,8-dimethoxy-3-methyl-5-phenyl-lH-3-benzazepine) and perphenazine. A-66359 was the gift of Dr. John Kebabian (Abbott Laboratories, Chicago IL). Raclopride was a gift from Astra Research Centre (Sodertalje, Sweden); haloperidol, chlorpromazine and thioridazine were purchased from Research Biochemicals, Inc. (Natick, MA). 2.2. In vi~'o binding to dopamine D 1 and D 2 receptors

In vivo binding studies were conducted as described previously (McQuade et al., 1988a, 1991a). The radioligands [125I]SCH 38840 and [3H]raclopride were used to label dopamine D 1 and D 2 receptors, respectively. Briefly, the radioligand was administered to male Sprague-Dawley rats concommitantly with unlabeled test compound via s.c. injection. One h later, the animals were sacrificed and striatal tissue was excised. Previous studies have demonstrated that a 1 h pretreatment period was optimal for in vivo binding of the radioligands (McQuade et al., 1988a, 1991a). All unlabeled drugs were tested at a minimum of three doses, over a 3 log unit range; each dose was tested in a minimum of three animals. Bound radioligand was quantified as described previously and data were expressed as a percent of the binding in the absence of test compound (% control). The ED50 was defined as

TABLE

the dose of compound required to inhibit binding by 50% and was determined by the E B D A computer program (McPherson, 1985). 2.3. Rat conditioned acoidance response (CAR) test

The rat C A R test was performed as previously described (Iorio et al., 1983). Briefly, rats were trained to jump onto a platform in response to a tone and prior to a foot shock. After meeting criteria, the rats received test drug s.c. and were tested 30 rain later. This pretreatment time has been previously determined to exhibit maximal activity of the dopamine D I antagonists (Iorio et al., 1983). If the rat jumped onto the platform after the tone, but before the shock, an avoidance was registered. If the rat jumped onto the platform during the shock, it was termed an escape, while if the rat did not jump at all, it was termed a failure. Data were expressed as the number of avoidances and the minimal effective dose was defined as the lowest dose of compound which resulted in a significant decrease in the number of avoidances. All drugs were tested at a minimum of three doses with an n of at least six animals/dose.

3. Results

The compounds that were tested included seven dopamine D 1 antagonists, of at least three distinct molecular types, a purported 'pro-drug' of a D I antagonist, two selective dopamine D 2 antagonists and three mixed D I / D 2 antagonists, with selectivity ratios ranging from 6 to 23 (in favor of their D 2 affinities). The data for these compounds are summarized in table 1.

1

C o m p a r i s o n o f the c a r a n d in v i t r o a n d in vivo b i n d i n g activities o f s e l e c t e d c o m p o u n d s at d o p a m i n e D I a n d D e r e c e p t o r s . Compound

S C H 23390 SCH 39166 S C H 23389 S C H 15198 S C H 3884{I S C H 40853 A-66359 S C H 12679 Thioridazine Chlorpromazine Perphenazine Haloperidol Raclopride

C a r activity

In v i t r o b i n d i n g activity

M E D (s.c.) (mg/kg)

K i (nM)

0.01 0.03 b 0.10 0.10 1.00 f 1.00 3.00 5.60 30.0 3.00 0.10 0.10 0.03

E D s o ( m g / k g , s.c.)

D-1

D-2

D-I

D-2

0.3 ~ 1.3 1.1 c 2.4 c 1.5 " 3.2 24.7 ~ 491 c 59.0 ~' 74.0 ~ 29.9 a 835 ~' 22720 ~

760 ~' 570 1062 ~" 3770 " 2377 c 4930 2580 5218 ~" 9.1 a 8.2 ~' 1.3 ~' 1.8 ~ 7.6

0.005 b 0.016 ~ 0.022 0.021 d {).176 0.282 0.468 0.600 d 3.35 1.30 4.79 > 100 f > 100 b

> 100 d > 100 ND 13.65 a ND ND ND 90.76 d 8.74 0.651 0.041 0.013 0.002 b

D a t a f r o m : a B i l l a r d et al., 1984; b M c Q u a d e et al., 1991a; c M c Q u a d e et al., 1988b; d B r e e s e et al., 1990; ~ M c Q u a d e et al., 1991b; ~ M c Q u a d e et al., 1988a.

31

tively. The affinities of the other drugs, in order of decreasing affinity, were: perphenazine > chlorpromazine > thioridazine > SCH 15198 > SCH 12679 > SCH 39166 and SCH 23390. These latter two compounds exhibited EDs0s in excess of 100 m g / k g . All of the drugs were also tested for their behavioral activity in the rat C A R test. SCH 23390 appeared to be the most potent compound with an M E D of 0.01 m g / k g s.c. This was followed in order of decreasing potency by raclopride = SCH 39166 > haloperidol = SCH 23389 = SCH 15198 perphenazine > SCH 38840 = SCH 40853 = > chlorpromazine = A-68359 > SCH 12679 > thioridazine. Correlation coefficients were then determined for the compounds in vivo affinities at dopamine D I and D 2 receptors and their activities in the C A R test. For the sake of this exercise, compounds with EDs0s greater than 100 m g / k g were assigned a value of 100. When

The seven dopamine D l antagonists all inhibited the binding of [125I]SCH 38840 to dopamine D l receptors in the rat striatum. SCH 23390 was the most potent compound, with an EDs0 of 0.005 m g / k g s.c. SCH 39166 was the next most potent molecule, followed in order, by SCH 15198, SCH 23389, SCH 38840, SCH 40853 and A-66359. SCH 12679, the purported D l pro-drug (Breese et al., 1990), was the next most potent compound at dopamine D1 receptors in vivo followed by chlorpromazine, thioridazine, perphenazine, haloperidol and raclopride; these latter two compounds exhibited EDs0's greater than 100 m g / k g s.c. Many of the drugs were then tested for their ability to inhibit the in vivo binding of [3H]raclopride to rat striatum. In contrast to their lack of inhibition of in vivo dopamine D l binding, raclopride and haloperidol were the most potent inhibitors of dopamine D 2 binding with EDs0s of 0.002 and 0.013 m g / k g s.c., respec-

1000A

>I,.--

m

$

100-

I

10-

Z

z_~

Ok

•

i

•

1

•

I

IIIQ "

&

1

e d

|

0 01

b

c

a

Z 001 001

.

.

.

.

.

.

.

.

,

.

.

.

.

.

.

.

.

,

01

. . . . . . . .

1

,

.

.

.

.

.

.

.

.

CAR

,

.

.

.

.

.

.

.

.

10

1

100

ACTIVITY (MED)

1000 >I--

a b c d e f g h i j k I m

SCH 23390 SCH 39166 SCH 15198 SCH 23389 SCH 38840 SCH 40853 A-66459 SCH 12679 CHLORPROMAZINE THIORIDAZINE PER P H E N ~ . Z I N E HALOPERIDOL RACLOPRIDE

100" p0 10 (3 Z ,..,., ~

OI

j

1

~a

a

•

1

•

d

g e

0 Ol

Z 0ol ool

Ol

1 CAR

1

lo

lOO

ACTIVITY (MED)

Correlation between the in vivo dopamine D 1 binding potencies and anti-CAR activities of a series of dopamine receptor antagonists. I n vivo binding potency is plotted as EDs0, while anti-CAR activity is plotted as M E D . ( A ) Correlation for all compounds tested; correlation coefficient (r) = 0.400, P = 0.175 (Spearman correlation coefficient). (B) Correlation for all compounds exhibiting D I selectivity, in vitro, and for compounds exhibiting less than 10-fold selectivity for the dopamine D 2 receptor; correlation coefficient ( r ) = 0.973, P < 0.001 (Spearman correlation coefficient). F i g . 1.

32 the d o p a m i n e D l affinity of all tested c o m p o u n d s was plotted as a function of C A R activity (fig. 1), the resulting data exhibited a correlation coefficient (r) of 0.400, which was not statistically significant ( S p e a r m a n correlation, P > 0.10). If, however, the three most D 2 selective compounds, i.e. perphenazine, haloperidol and raclopride, were d r o p p e d from this comparison, a statistically significant correlation was observed (r = 0.973, P < 0.001). W h e n the in vivo d o p a m i n e D 2 binding data was plotted againist C A R activity (fig. 2), there was no correlation between the two ( r = - 0 . 1 5 0 , P>0.70), Following the deletion, however, of the D~ antagonists S C H 12679, S C H 15198, S C H 39166 and S C H 23390, the in vivo d o p a m i n e D 2 affinities of the remaining c o m p o u n d s showed good correlation with their behavioral activity (r = 0.975, P < 0.005).

1000

4. D i s c u s s i o n

The overall conclusion from the data presented herein indicates that occupancy of either d o p a m i n e D~ or D 2 receptors is correlated with behavioral activity which is predictive of clinical antipsychotic efficacy. T h e observation of a correlation between in vivo d o p a m i n e D 2 binding to rat striatum and activity in the rat C A R test is not surprising, as the predictive nature of this behavioral paradigm was established using d o p a m i n e D 2 antagonists. The correlation of C A R activity with in vivo striatal d o p a m i n e D I binding was not so obvious. This observation confirms previous data and strengthens the contention that d o p a m i n e D~ antagonists will have antipsychotic efficacy in a clinical environment. T h e correlation, however, is not meant to imply that the behavioral or clinical activity of D~

A o

o

0

h

I

[

0 o g

0 Z Z

&

to r~ LU v

o d 1

oC

0 0

01

b

a

Z

0 001

.

.

.

.

.

.

.

,

.

001

.

.

.

.

.

.

.

,

.

01

.

.

.

.

.

.

.

,

.

1 CAR

.

.

.

.

.

.

.

,

.

1

.

.

.

.

.

.

.

10

.

100

ACTIVITY (MED)

1000"

a b c d e f g h i

RACLOPRIDE HALOPERIDOL PERPHENAZINE CHLORPROMAZINE THIORIDAZINE SCH 12679 SCH 15198 SCH 39166 SCH 23390

I100-

F0 I0"

0 Z

Z to m c3 iii v D

I

1

/b

0 01 •

Z 001

.

O01

.

.

.

.

.

.

.

i

.

.

.

.

.

.

.

.

Ol

i

.

.

.

.

.

1 CAR

.

.

.

,

1

.

.

.

.

.

.

.

.

, lO

........ lOO

ACTIVITY (MED)

Fig. 2. Correlation between the in vivo dopamine D 2 binding potencies and anti-CAR activities of a series of dopamine receptor antagonists. In vivo binding potency is plotted as EDs0, while anti-CAR activity is plotted as MED. (A) Correlation for all compounds tested; correlation coefficient (r)= 0.150, P= 0.701 (Spearman correlation coefficient). (B) Correlation for all compounds exhibiting D 2 selectivity, in vitro; correlation coefficient (r) = 0.975, P < 0.005 (Spearman correlation coefficient).

33

antagonists resides solely in the striatum; rather, the correlation suggests that in vivo binding to dopamine receptors in striatum reflects the binding in other areas which may be more directly involved in the behavioral activities of the dopamine receptor antagonists. Recently, a similar correlation has also been observed in primates. Bergman et al. (1991) have reported that both dopamine D~ and D 2 antagonists decrease schedule-controlled responding in squirrel monkeys, and that these activities correlate with the respective receptor affinities of the compounds. Several other conclusions can also be drawn from these correlation data. The first involves the actions of so-called non-selective dopamine antagonists. In vitro studies (Billard et al., 1984) indicated that chlorpromazine (D~ "D 2 ratio of 9), thioridazine (ratio of 6.5) and perphenazine (ratio of 23) all exhibited some preference for dopamine D 2 receptors, but yet exhibited reasonable dopamine D~ affinity. The current study demonstrates that chlorpromazine and thioridazine bind to both dopamine D1 and D 2 receptors, in vivo, at doses which are behaviorally active. This observation suggests that the behavioral activities of these two drugs are mediated by both receptors. However, perphenazine in vivo appears to be selective for the dopamine D 2 receptor. While the number of drugs analyzed is small, one could hypothesize that in vitro selectivity ratios in excess of 20 impart in vivo selectivity on the actions of a dopaminergic antagonist. Another interesting observation from this work is the relationship between 50% receptor occupancy and the behavioral MED. These data clearly indicate that greater than 50% of the dopamine receptors must be occupied by drug prior to the emergence of behavioral activity. This is true for D t selective, D 2 selective and non-selective antagonists. This preclinical observation mirrors what occurs in patients currently undergoing antipsychotic drug therapy. Sedvall and colleagues (1988) have demonstrated, using positron emission tomography (PET) that greater than 60% of dopamine D 2 receptors are occupied in patients receiving antipsychotic medications. The correlations described above also can be useful in determining the mechanism of action of pro-drugs. This is illustrated by the activities of SCH 12679. This compound exhibits poor affinity for both dopamine D~ and D 2 receptors in vitro, but demonstrated behavioral activity at 5.6 mg/kg. Analysis of in vivo binding, however, indicated that SCH 12679 was binding to dopamine D I receptors following s.c. administration. These data strongly suggested, as has been previously hypothesized (Breese et al., 1990), that SCH 12679 was metabolized to an active, dopamine D~ selective antagonist (most likely, SCH 15198). In summary, in vivo binding to both dopamine D I and D 2 receptors correlates with activity in the rat

CAR test, a paradigm considered predictive of clinical antipsychotic activity. The relationship between these two measures also has applications in the identification of active metabolites, determination of in vivo selectivity and in the measurement of behaviorally relevant receptor occupancy.

References Andersen, P.H., F.C. Gronvald, R. Hohlweg, L.B. Hansen, E. Guddal and C. Braestrup, 1988, NO-112, NO-756 New dopamine DI selective antagonists, Abstr, Soc. Neurosci. Ann. Meeting, 935. Bergman, J., B.K. Madras and R.D. Spealman, 1991, Behavioral effects of D t and D 2 dopamine receptor antagonists in squirrel monkeys, J. Pharmacol. Exp. Ther. 258, 911). Billard, W., V. Ruperto, G. Crosby, L.C. lorio and A. Barnett, 1984, Characterization of the binding of 3H-SCH 233911, a selective D-1 receptor antagonist ligand, in rat striatum, Life Sci. 35, 1885. Breese, G.R., H.E. Criswell, R.D. McQuade, L.C. lorio and R.A. Mueller, 1990, Pharmacological evaluation of SCH-12679: Evidence for an in vivo antagonism of Dl-dopamine receptors, J. Pharmacol. Exp. Ther. 252, 558. Chipkin, R.E., 1990, DI antagonist in clinical trial, Trends Pharmacol. Sci. 11, 185. Chipkin, R.E., L.C. Iorio, V.L. Coffin, R.D. McQuade, J.G. Berger and A. Barnett, 1988, Pharmacological profile of SCH 39166: A dopamine D1 selective benzonaphthazepine with potenial antipsychotic activity, J. Pharmacol. Exp. Ther. 247,1093. Coffin, V.L., M.B. Latranyi and R.E. Chipkin, 1989, Acute extrapyramidal syndrome in Cebus monkeys: Development mediated by dopamine D 2 but not D 1 receptors, J. Pharmacol. Exp. Ther. 249, 769. Coffin, V.L., D.T. McHugh, R.E. Chipkin and A. Barnett, SCH 39166, a potential antipsychotic drug, does not evoke movement disorders in Cebus monkeys, Neurochem. Int. (in press). Cook, L. and A.B. Davidson, 1978, Behavioral pharmacology: Animal models involving aversive control of behavior, in: Psychopharmacology: A Generation of Progress, eds. M.A. Lipton, A. Dimascio and K. Killam (Raven Press, New York) p. 563. Davidson, A.B. and E. Weidley, 1976, Differential effects of neuroleptics and other psychotropic agents on acquisition of avoidance in rats, Life Sci. 18, 1279. Iorio, L.C., A. Barnett, F.H. Leitz, V.P. Houser and C.A. Korduba, 1983, SCH 23390, a potential benzazepine antipsychotic with unique interactions on dopaminergic systems, J. Pharmacol. Exp. Ther. 226, 462. Kerkman, D.J., M. Ackerman, L.D. Artman, K. Asin, L. Bednarz, M.C. Johnson, R.G. MacKenzie, W. Montana, H. Stampfli and J.W. Kebabian, 1988, A-69024 a non-benzazepine D I selective antagonist, Abstr. Soc. Neurosci. Ann. Meeting, 934. McHugh, D. and V. Coffin, 1-)91, The reversal of extrapyramidal side effects with SCH 39166, a dopamine D1 receptor antagonist, Eur. J. Pharmacol. 202, 133. McPherson, G.A., 1985, Kinetic, EBDA, Ligand, Lowry: A Collection of Radioligand Binding Analysis Programs (Elsevier Science Publishers BV, Amsterdam) p. 14. McQuadc, R.D., N. Amlaiky, M. Caron, R. Chipkin, L. Iorio and A. Barnett, 1988a, Characterization of the radioiodinated analogue of SCH 23390: in vitro and in vivo D I dopamine receptor binding studies, Life Sci. 43, 1151. McQuade, R.D., D. Ford, R.A. Duffy, R.E. Chipkin, L.C. Iorio and A. Barnett, 1988b, Serotonergic component of SCH 23390: In vitro and in vivo binding analyses, Life Sci. 43, 1861.

34 McQuade, R.D., R.A. Duffy, Y.L. Coffin, R.E. Chipkin and A. Barnett, 1991a, In vivo binding of SCH 39166: a D-I selective antagonist, J. Pharmacol. Exp. Ther. 257, 42. McQuade, R.D., R.A. Duffy, C.C. Anderson, G. Crosby, Y.L. Coffin, R.E. Chipkin and A. Barnett, 1991b, [3H]SCH 39166, a new Dl-selective radioligand: in vitro and in vivo binding analyses, J. Neurochem. 57, 2(101.

Sedvall, G., L. Farde and H. Hall, 1988, D1 and D2 dopamine receptors in the human brain as studied by in vitro analysis and PET. in: Central and Peripheral Dopamine Receptors: Biochemistry and Pharmacology, eds. P.F. Spano, G. Biggio, G. Toffano and G.L. Gessa (Liviana Press, Padova, Italy) p. 81. Seeman, P., 1987, Dopamine receptors and the dopamine hypothesis of schizophrenia, Synapse 1, 133.