Journal of Neurochemistry Raven Press, Ltd., New York 0 1990 International Society for Neurochemistry

Homologous Desensitization of the D1 Dopamine Receptor Anthony J. Balmforth, Philip Warburton, and Stephen G. Ball Department of Cardiovascular Studies, University of Leeds, Leeds, England

Abstract: Preincubation of D384 cells, derived from the human astrocytoma cell line G-CCM, with dopamine resulted in a time-dependent attenuation of cyclic AMP responsiveness to subsequent dopamine stimulation. This effect was agonist specific because the prostaglandin El (PGEI) stimulation of cyclic AMP of similarly treated cells remained unchanged. The attenuation by dopamine was concentration dependent with a maximum observed at 100 ph4. A comparison of dopamine concentration-response curves of control and dopamine-preincubatedcells revealed no change in the K, apparent value, but a marked attenuation of the maximal response. Preincubation of cells with dopamine in the presence of DI but not D2 selective antagonists partially prevented the observed attenuation. Attenuations in dopamine responsiveness were also obtained when D384 cells were preincubated with DI but not D2receptor agonists. The level of attenuation attained related to agonist efficiency in stim-

ulating cyclic AMP SKF38393 < 3,4-dihydroxynomifensine < fenoldopam < 2-amino-6,7-dihydroxy-1,2,3,4-tetrahydronaphthalene = dopamine. However, increasing the efficiency of 3,4-dihydroxynomifensine stimulation of cyclic AMP, using the synergisticeffect of adding a low concentration of forskolin, produced no further change in the attenuation of the subsequent response to dopamine. Thus, the D1dopamine receptors expressed by D384 cells undergo homologous desensitization. Uncoupling of the D1 dopamine receptor appears to be independent of cyclic AMP formation, analogous to a mechanism proposed for the 8-adrenergic receptor. Key Words: DI dopamine receptor-P-Adrenergic receptor-Prostaglandin El-Human astrocytoma-Cyclic AMP formation-Homologous desensitization. Balmforth A. J. et al. Homologous desensitization of the DI dopamine receptor. J. Neurochem. 55, 21 11-21 16 (1990).

Prolonged exposure of cells and tissues to hormones or neurotransmitters attenuates their responsiveness to further stimulation. This process of desensitization has been widely studied in various cells with receptors linked to the adenylate cyclase (EC 4.6.1.1) signal transduction pathway (Sibley and Lefkowitz, 1985). Receptor desensitization has been divided into two broad categories. Agonist-specific or homologous desensitization occurs when, after stimulation of a cell by an agonist, sensitivity is lost only to that agonist while responsiveness to other different agonists operating through other distinct receptors remains unchanged. Conversely, agonist-nonspecific or heterologous desensitization describes the situation where exposure to one agonist attenuates the responsiveness of a number of different receptors to their individual agonists. Homologous desensitization of the 0-adrenergic receptor consists of a number of sequential events (Be. initial step involves a rapid novic et al., 1 9 8 8 ~ )The uncoupling of receptors from adenylate cyclase fol-

lowed by a slower removal of receptors from the outer membrane to an internal store for processing. The uncoupling of the receptor has recently been attributed to phosphorylation of the agonist-occupied receptor by a unique cyclic AMP-independent protein kinase, P-adrenergic receptor kinase (Benovic et al., 1986). Following sequestration the receptors are thought to be dephosphorylated and prepared for recycling to the cell surface. In contrast to the P-adrenergic receptor, little is known about the regulation of the D1 dopamine receptor which is similarly linked to adenylate cyclase. Prolonged stimulation of striatal slices with dopamine has been observed to attenuate subsequent stimulation of adenylate cyclase by dopamine, suggesting homologous desensitization (Memo et al., 1982). In addition, chronic treatment of rats with the D, dopamine receptor antagonist SCH23390 results in an increase in the number of receptor binding sites (Hess et al., 1988). Furthermore, amphetamine treatment, presumed to release dopamine, has recently been shown to desen-

Received March 29, 1990; revised manuscript received May 15, 1990; accepted June 11, 1990. Address correspondenceand reprint requeststo Dr. A. J. Balmforth at Department of Cardiovascular Studies, University of Leeds, Leeds LS2 9JT, U.K.

Abbreviations used: 6,7-ADTN, 2-amino-6,7-dihydroxy-1,2,3,4tetrahydronaphthalene;IBMX, 3-isobutyl-I-methylxanthine; N0434, (+)-2-(N-phenylethyl-N-propyl)amino-5-hydroxytetralin;PGEl, prostaglandin El ;TCA, trichloroacetic acid.

2111

A. J. BALMFORTH ET AL.

2112

sitize both rat striatal D1 and D2 dopamine receptors, again suggesting homologous desensitization (Rosenboom and Gnegy, 1989). However, the heterogeneity of cell types present in the CNS complicates attempts to define and localize the biochemical mechanisms that regulate receptors. To reduce the complexity of this system, we and others have turned to tissue culture. For example, considerable advances in understanding the regulation of P-adrenergic receptors have been made using glial cell lines (Sibley and Lefkowitz, 1985). Previously we reported the isolation of a clone D384 from the human astrocytoma cell line G-CCM which expressed a dopamine receptor with a pharmacological profile typical of a mammalian striatal D1 dopamine receptor (Balmforth et al., 1986, 1988). The human astrocytoma cell line D384 therefore offers another approach to studying the regulation of DI dopamine receptors which has some advantages over the use of rat striatal homogenates for biochemical studies. It provides a homogeneous cell population, expressing D1 dopamine receptors, which can be grown in large numbers. In addition, high sensitivity and reproducibility are obtained because dopamine-induced stimulation of cyclic AMP formation in intact D384 cells is up to 50-fold above basal, compared to only two- to threefold observed in striatal homogenates. In the present study we have demonstrated that the DI dopamine receptor expressed by D384 cells is subject to homologous or agonist-specific desensitization. In addition, we have examined whether the mechanism of desensitization of the D1 dopamine receptor is similar to that established for the P-adrenergic receptor. Preliminary results of this study have been reported previously in abstract form (Balmforth et al., 1989).

MATERIALS AND METHODS Materials Dopamine, isoprenaline, adrenaline, prostaglandin El (PGE,), forskolin, and 3-isobutyl-1-methylxanthine (IBMX) were purchased from Sigma Chemical (Poole, Dorset, England); SCH23388, SKF38393, 2-amino-6,7-dihydroxy1,2,3,4-tetrahydronapthalene (6,7-ADTN), (?)-2-(N-phenylethyl-N-propyl)amino-5-hydroxytetralin (N0434), (+)sulpiride, and (-)-sulpiride from Semat (St. Albans, England); [8-3H]adenine and adenine [U-14C]cyclicAMP from Amersham International P.L.C. (Amersham, Bucks, England); and all tissue culture reagents and plastics from GIBCO (Paisley, Scotland). The following reagents were generously donated SCH23390 from Schering (Bloomfield, NJ, U.S.A.); cisand trans-flupenthixol from H. Lundbeck A/S (Copenhagen, Denmark); domperidone from Janssen Pharmaceuticals (Wantage, England); 3,4-dihydroxynomifensine from Hoechst U.K. (Hounslow, Middlesex, England); pergolide and quinpirole (LY171555) from Dr. H. Herdon (Department of Pharmacology, Leeds University, England); and fenoldopam from Smith, Kline and French Research (Welwyn, Hertfordshire, England).

Cell culture Human astrocytoma clone D384 cells were subcultured at a density of 8 X lo4 cells/well in six-well multidishes conJ. Neurochem., Vol. 55, No. 6, 1990

taining 4 ml of 1: 1 Ham's FlO/Dulbecco's modified Eagle's medium supplemented with 10%fetal calf serum and 2 mM L-glutamine. The cells were maintained at 37°C in an atmosphere of 2% COz and 98% humidified air. Cells were used 8 days after subculture in all experiments.

Assay for cyclic AMP formation in intact cells The measurement of intracellular cyclic AMP formation was based on a modification of the prelabelling technique (Shimizu et al., 1969) as described previously by Balmforth et al. (1988). In brief, cell adenine nucleotide pools were labelled by replacing the medium with fresh medium (2 ml), containing 4 pCi of [3H]adenine, and incubating at 37°C for 2 h. To determine levels of desensitization, drugs (see legends to Figs. 1-6) were added during the final hour of incubation. The cells were then washed three times with serum-free medium (2 ml). Cyclic AMP formation was measured by addition of agonists as specified in the figures, in serum-free medium containing 0.5 mM IBMX. Propranolol (10 p M ) was present in all dopamine experiments to block 6-adrenergic receptors. After 12 min at 37"C, incubations were terminated by aspiration of the medium and subsequent addition of 1.5 ml of ice-cold 5% trichloroacetic acid (TCA) containing 1.25 nCi of ['4C]cyclic AMP/ml as an internal standard. Cyclic AMP was isolated from the TCA extracts by sequential chromatography on Dowex AG50.X4 anion-exchange and neutral alumina columns (Salomon et al., 1974). The [3H]cyclicAMP content of each sample was corrected for quench and [14C]cyclicAMP recovery (60-80%) and expressed as dpm using a Packard 2000CA scintillation counter.

RESULTS Effects of preincubating cells with dopamine, isoprenaline, and adrenaline Preincubation of D384 cells with dopamine (1 mM) resulted in a time-dependent attenuation of cyclic AMP responsiveness to subsequent dopamine (1 mM) stimulation (Fig. 1A). This effect was agonist specific because the PGEl (10 p M ) stimulation of cyclic AMP of identically treated cells remained unchanged. A similar agonist specific attenuation of P2-adrenergicreceptor stimulation of cyclic AMP was observed when cells were preincubated with isoprenaline (1 p M ) or adrenaline (0.1 mM, Fig. 1 B). However, although dopamine, isoprenaline, and adrenaline attenuated their respective responses to similar levels (reduced to 37 f 2, 37 k 3, and 39 k 3% of control respectively, n = 3) by a 1-h preincubation, there were substantial differences at earlier time points. Compared to dopamine, isoprenaline and adrenaline attenuation was rapid, with the majority of the effect occurring within 2 min. In contrast a 20-min preincubation of dopamine was required to achieve the same level of attenuation. The desensitizationby dopamine was concentration dependent (ECSo,1 p M ) , with a maximal attenuation observed at 100 pM (Fig. 2). A comparison of dopamine concentration-response curves of control and dopamine-preincubated cells revealed no change in the K, apparent value (agonist concentration producing half-maximal stimulation) for dopamine ( 1.6 pM and

DI DOPAMINE RECEPTOR DESENSITIZATION

2113

1 PGE 1

s

3 2

9 9 s

60

-

40

-

2o n "

t

Dopamine

2o "

I

0

20

10

30

40

50

t

60

~

I /

0

-7

'*Or

s

3b

ao

9

9 s

6o 40

0

I

a

;

I

tt

-4

1

-3

M

FIG. 2. Concentration-dependent attenuationof the D, doparnine receptor response. Cells were preincubated with various concentrations of dopamine for 1 h prior to measurementof dopamine (1 mM)-stimulated cyclic AMP formation. Each point represent the mean SEM of three independent experiments each performed in triplicate. Data are expressed as a percentage of the response produced by dopamine (1 mM) on untreated cells (37,393 dpm/ well).

+

PGE1

-

~

-5

Lop [ Doparnine

Time lrnin)

Ioo~+€---r---

~

-6

i

Adrenal ine lsoprenaline

cubated with D1 but not D2 receptor agonists (Fig. 5). The level of attenuation attained appeared to be related to the agonists' efficiency in stimulating cyclic AMP: SKF38393 < 3,4-dihydroxynomifensine < fenoldopam < 6,7-ADTN = dopamine (Table 1).

1 0

10

20

30

40

50

60

Time (mid

FIG. 1. Time courses of (A) and (B) . , D,. doDamine . . &adreneraic receptor desensitization. Cells were preincubated with either dopamine (1 mM), isoprenaline (1 pM), or adrenaline (0.1 mnn) for different periods of time during prelabelling of the cells with f'H]adenine. Dopamine (1 mM, O), PGEl (10 pM, m), isoprenaline (1 pM, A), and adrenaline (0.1 mM, 0) stimulation of cyclic AMP (CAMP)formation was subsequently measured in the presence of 0.5 mM IBMX for 12 min. Each point represent the means f SEM of three independentexperiments each performed in triplicate. Data are expressed as a percentage of responses produced by these agents on untreatedcells, which were (A) dopamine (36,892 dpm/ well) and PGEl (80,919 dpm/well) and (B) adrenaline (52,310 dpm/ well), isoprenaline(59,546 dpm/well), and PGE, (87,891 dpm/well).

Effect of conversion of a partial to "full agonist" A low concentration of forskolin (0.25 p M ) produced a modest stimulation of cyclic AMP in D384 cells, similar to that of 3,4-dihydroxynomifensine (Fig. 6). In contrast, when added in combination with 3,4dihydroxynomifensine, the resulting stimulation of cyclic AMP was greater than additive and 160%of that achieved by dopamine (10 pM, Fig. 6). The preincu-

loo -

1.9 pM, respectively) but a marked attenuation of the maximal response of the dopamine-preincubated cells (reduced to 43 & 4% of control, n = 3, Fig. 3). Effects of selective dopamine antagonists and agonists Antagonists. Preincubation of cells with dopamine in the presence of the DI receptor antagonists SCH23390 and cis-flupenthixol partially prevented the observed attenuation of responsiveness to subsequent dopamine stimulation (Fig. 4). In contrast, at equimolar concentrations, the inactive isomers of the D1 selective antagonists SCH23390 and cis-flupenthixol (SCH23388 and trans-flupenthixol, respectively) had no effect. A similar lack of effect was observed with the D2 receptor antagonists domperidone and sulpiride. Agonists. Attenuations in dopamine responsiveness were also observed when D384 cells were prein-

80

9

Q,

-

60-

40i+Lk Desensit ized

20 0

-7

-6

Log

-5

DowmiNt

-4

1

-3

M

FIG. 3. Dopamine concentration-relatedeffects on intracellular

cyclic AMP formation of control and desensitized cells. Cells were preincubated in the presence or absence of dopamine (10 ~JV) for 1 h prior to measurement of dopamine (1 m#)-stimulated cyclic AMP formation. Each point represent the mean f SEM of three independent experiments each performed in triplicate. Data are expressed as a percentage of the response produced by dopamine (1 M ) on untreated cells (29,703 dpm/well).

J. Neurvehem., Vol. 55. No. 6, 1990

2114

A. J. BALMFORTH ET AL.

i

TABLE 1. Eflects of dopamine agonists on stimulating intracellulur cvclic AMP formation

loo 80

Agonist

-

No WSCH (S)SCH Cis Antamst c- 2.5xiO-BM

m

D1-antapmists

Trans

m

Dopamine (0.1 mM) SKF38393 (10 jtM) 3,4-Dihydroxynomifensine (1 M) Fenoldopam (0.1 mM) 6,7-ADTN (0.1 mM) Dan

[3H]CyclicAMP dpm/well

Percentage of dopamine response

25,493 k 74 1,827 f 161

100 7

4,268 f 103 19,961 522 29,317 202

17 78 115

* *

(+)Sul (-)%I

c 10-7M-

Cells were exposed to dopamine agonists in the presence of 0.5 mMIBMX for 12 min. The results represent the means of triplicate

*

D1-antagcntsts

FIG. 4. Effect of preincubationof cells with dopamine in the presence of dopamine receptor antagonists. Cells were preincubated in the absence (untreated) or presence of 10 pM dopamine (no antagonist) or 10 pM dopamine plus either (R)-SCH23390, (S)SCH23388, cis-flupenthixol,trans-flupenthixol,domperidone, (+)sulpiride, or (-)-sulpiride, for 1 h prior to measurementof dopamine (1 mM)-stimulated cyclic AMP formation. Each point represents the mean & SEM of three independentexperiments each performed in triplicate. Data are expressed as a percentage of the response produced by dopamine (1 mM) on untreated cells (34,632 dpm/ well).

bation of cells with this combination produced no further change in the attenuation of the subsequent response to dopamine with respect to 3,4-dihydroxynomifensine alone (Fig. 6).

determinations SEM with the mean basal level subtracted (2,084 f 126 dpm/well) from a single experiment which was replicated three times.

DISCUSSION Prolonged incubation of D384 cells with dopamine resulted in a specific attenuation of cyclic AMP responsiveness to subsequent dopamine stimulation. In contrast, there was no loss of PGEl responsiveness in similarly treated cells. This suggests a mechanism involving the uncoupling of the D1 dopamine receptor from the hormone-sensitive adenylate cyclase signal transduction pathway, rather than alterations in either the activity of the guanine nucleotide binding protein G, or adenylate cyclase. The &adrenergic receptor expressed by D384 cells (Balmforth et al., 1988) also ex-

r 'Desensitization

DA

=

10-4M

SKF 10-5M

W

FEN

10-34 10-4M

D1-amists

ADTN E R G 10-4M

W I N NO434

f-

10-7M-

D2-amists

FIG. 5. Effect of preincubating cells in the presence of dopamine agonists on subsequent dopamine-stimulated cyclic AMP formation. Cells were preincubated for 1 h in the presence of either dopamine (DA), SKF38393 (SKF), 3,4-dihydroxynomifensine (DHNF), fenoldopam (FEN), 6,7-ADTN, pergolide(PERG), quinpirole (QUIN), or NO434 prior to measurement of dopamine (1 mM)-stimulated cyclic AMP formation. Each point representthe mean k SEM of three independent experimentseach performed in triplicate. Data are expressed as a percentage of the response produced by doparnine (1 rnM) on untreated cells (31,432 dpm/well). The concentration of drugs used for desensitizationwere chosen as follows. The D, receptor agonist concentrations were maximal for stimulation of cyclic AMP, whereas the D2receptor agonist concentrations were sufficient to activate D2 receptors without appreciable D, receptor activation.

J. Neurochem., Vol. 55, No. 6, 1990

DA

oH\F

Fors

Fys

ctN=

DA

DHN Fys

DHN

FIG. 6. Relationship between cyclic AMP formation and level of desensitizationattained. A Cyclic AMP formation was measured in cells incubated with either dopamine (1 mM), forskolin (0.25 &f), 3,4-dihydroxynomifensine(DHNF, 1 mM),or 3,4-dihydroxynomifensine (1 mM) plus forskolin (0.25 phf) for 12 rnin in the presence of 0.5 mM IBMX. Each point represent the mean & SEM of three independent experiments each performed in triplicate. Data are expressed as a percentageof the response produced by dopamine (42,153 dpm/well). B Cells were preincubated with the above for 1 h prior to measurement of dopamine (1 mM)-stimulated cyclic AMP formation. Each point represent the mean ? SEM of three independent experiments each performed in triplicate. Data are expressed as a percentageof the response produced by dopamine (1 mM) on untreated cells (42,153 dpm/well).

DI DOPAMINE RECEPTOR DESENSITIZATION hibited a similar desensitization following prolonged incubation with isoprenaline or adrenaline. Interestingly, the time course of the &adrenergic receptor uncoupling from adenylate cyclase was rapid when compared to the DI dopamine receptor, but in keeping with previous reports for P-adrenergic receptors using human astrocytoma, rat astrocytoma, and HeLa cells (Su et al., 1980; Kassis and Fishman, 1984). The desensitization of the D1dopamine receptor by prolonged exposure to dopamine was concentration and time dependent, requiring at least 5 min of preincubation and concentrations of dopamine between 0.1 and 100 pM. Interestingly, the concentration of dopamine required to produce half-maximal desensitization (1 p M ) was very similar to the K, apparent value (2 p M ) of the D1 dopamine receptor for dopamine stimulation of cyclic AMP (Balmforth et al., 1988). That desensitization requires activation of the receptor site by an agonist is supported by the finding that dopamine desensitization could be partially prevented by the coincubation of dopamine with the selective D1 dopamine receptor antagonists SCH23390 and cis-flupenthixol (Billard et al., 1984; Hess and Creese, 1987), but not the selective D2 dopamine receptor antagonists domperidone and sulpiride (Hess and Creese, 1987), nor the inactive isomers of the D1-selectiveantagonists, SCH23388 and trans-flupenthixol (Billard et al., 1984). Complete prevention of desensitization by coincubation of dopamine with higher concentrations of selective D1 dopamine receptor antagonists could not be achieved due to difficulty in washing out the antagonists prior to measurement of dopamine-stimulated cyclic AMP formation. Desensitization was also achieved by preincubating cells with the D1 receptor agonists SKF38393, 3,4-dihydroxynomifensine, fenoldopam, and 6,7-ADTN (Watling and Dowling, 1981; Flaim et al., 1986; Hess and Creese, 1987), but not the D2 receptor agonists pergolide, quinpirole and NO434 (Arnt and Hyttel, 1985; Seeman et al., 1985). Previous reports concerning homologous desensitization of the P-adrenergic receptor in cell cultures have suggested that the efficacy of an agonist to stimulate cyclic AMP formation is related to the level of desensitization attained during short term incubation (Su et al., 1980; Pittman et al., 1984). However, homologous desensitization is unlikely to be mediated through cyclic AMP itself because such desensitization has been observed in S-49 cyc- lymphoma cells which lack G, and hence are unable to produce cyclic AMP through receptor stimulation (Green and Clark, 1981). More recently, a clear relationship between the ability of a partial agonist to stimulate adenylate cyclase activity and to promote P-adrenergic receptor phosphorylation by P-adrenergic receptor kinase, thought to be the first step in uncoupling the receptor, has been reported (Benovic et al., 198%). Thus it has been proposed that the agonist-induced conformational changes required for receptor activation are related to those required for /3-adrenergic receptor kinase phosphorylation which

2115

leads to receptor uncoupling. The similar relationship between DI agonist efficacy in stimulating cyclic AMP and degree of desensitization observed in our study suggests a similar mechanism for homologous desensitization of the D I receptor. The possibility that the D, receptor is a substrate for P-adrenergic receptor kinase is supported by the observations that this enzyme’s substrates also include the agonist-occupied forms of somatostatin, PGEl, and a2-adrenergicand muscarinic receptors which are also coupled to adenylate cyclase (Benovic et al., 1988a; Kwatra et al., 1989). Forskolin is widely used as a direct activator of adenylate cyclase to study the regulation and physiological effects of cyclic AMP in tissues and cell cultures. In addition to its direct action on adenylate cyclase, forskolin also increases the efficiency of coupling between G, and adenylate cyclase (Yamashita et al., 1986).The latter effect is believed to be responsible for the synergistic action of forskolin on the stimulation of adenylate cyclase by partial and full dopamine agonists (Battaglia et al., 1986). Thus, using a low concentration of forskolin we were able to increase the efficiency of cyclic AMP stimulation by the partial agonist 3,4-dihydroxynomifensine to a level in excess of the natural agonist dopamine. The lack of effect of this intervention on the level of desensitization attained by the partial agonist 3,4-dihydroxynomifensine again suggests that homologous desensitization of the DI receptor is not cyclic AMP mediated and thus further demonstrates the similarities between the mechanisms of homologous desensitization of D1dopamine and P-adrenergic receptors. Prolonged incubation of striatal slices with dopamine has previously been shown to attenuate subsequent stimulation of cyclic AMP by dopamine (Memo et al., 1982). A reduction in the affinity of N-[3H]propylnorapomorphine binding suggested the possibility of a mechanism involving the uncoupling of the D1receptor from G,. Furthermore, amphetamine treatment, which is known to stimulate endogenous dopamine release in rat striatum (Kelly and Nahorski, 1987), has recently been shown to desensitize D1dopamine receptors coupled to adenylate cyclase (Rosenbloom and Gnegy, 1989). In contrast, adenosine and muscarinic receptor cyclic AMP responses in the striatum were unaffected by amphetamine treatment, suggesting a mechanism that did not involve alterations in G-protein or adenylate cyclase activity. Combined with the observation of a 30% decrease in the number of high-affinity [3H]dopamine binding sites, thought to represent receptors coupled to G,, but without any decrease in total D1receptor number measured by [3H]SCH23390 binding, a mechanism involving the uncoupling of the D, dopamine receptor from G, seems likely. Because of the low number of D1dopamine receptors expressed by untreated D384 cells (14-20 fmol/mg), we were unable to pursue adequately the effect of prolonged incubation of dopamine on receptor number and affinity states. J. Neuroehem., Vol. 55, No. 6, 1990

2116

A. J. BALMFORTH ET AL.

Our results suggest that the DI dopamine receptor is subject to homologous or agonist-specific desensitization. The facts that desensitization was (a) related to the efficacy of the dopamine agonist, (b) independent of intracellular cyclic AMP formation, and (c) recently shown to involve a loss of high-affinity agonist binding sites (Rosenboom and Gnegy, 1989) strongly suggest a mechanism similar to that of the P-adrenergic receptor, involving an initial uncoupling of the receptor from G, by P-adrenergic receptor kinase. However, if P-adrenergic receptor kinase is involved in homologous desensitization of both Pz-adrenergic and D1 dopamine receptors expressed by D384 cells, the slower uncoupling of the DI dopamine receptor might indicate that it is a poorer substrate for the kinase. An alternative explanation would be that there are spare D1receptors, but the lack of change in the Ka apparent value of cyclic AMP formation by dopamine of control versus desensitized cells argues strongly against this. Thus, the cell line D384 expresses D1dopamine receptors which undergo homologous desensitization through a non-cyclic AMP-dependent process. Acknowledgment: This study was supported by the Medical Research Council (Project Grant G8820 156N),the British Heart Foundation, and the Wellcome Trust.

REFERENCES Arnt J. and Hyttel J. (1985) Differential involvement of dopamine D1 and D2 receptors in the circling behaviour induced by apomorphine, SK and F 38393, pergolide and LY171555 in 6-hydroxydopamine-lesionedrats. PsychopharmacologV 85,346-352. Balmforth A. J., Ball S. G., Freshney R. I., Graham D. I., McNamee H. B., and Vaughan P. F. T. (1986) D-1 dopaminergic and 0adrenergic stimulation of adenylate cyclase in a clone derived from the human astrocytoma cell line G-CCM. J. Neurochem. 41,715-719. Balmforth A. J., Yasunari K., Vaughan P. F. T., and Ball S. G. (1988) Characterization of dopamine and 0-adrenergic receptors linked to cyclic AMP formation in intact cells of the clone D384 derived from a human astrocytoma. J. Neurochem. 51, 1510-1515. Balmforth A. J., Warburton P., and Ball S. G. (1989)Agonist-induced desensitization of the D1 dopamine receptor expressed by a human astrocytoma cell line. Br. J. Pharmacol. 98, 704P. Battagha G., Norman A. B., Hess E. J., and Creese I. ( 1986) Forskolin potentiates the stimulation of rat striatal adenylate cyclase mediated by D- 1 dopamine receptors, guanine nucleotides, and sodium fluoride. J. Neurochem. 46, 1 180- I 185. Benovic J. L., Strasser R. H., Caron M. G., and Lefkowitz R. J. (1986) 0-Adrenergic receptor kinase: identification of a novel protein kinase which phosphorylates the agonist-occupied form of the receptor. Proc. Natl. Acad. Sci. USA 83, 2797-2801. Benovic J. L., Bouvier M., Caron M. G., and Lefkowitz R. J. (1988~) Regulation of adenylyl cyclase-coupled (3-adrenergicreceptors. Annu. Rev. Cell Biol. 4,405-428. Benovic J. L., Staniszewski C., Mayor F., Caron M. G., and Lefkowitz R. J. (1988b) (3-Adrenergic receptor kinase: activity of partial agonists for stimulation of adenylate cyclase correlates withability to promote receptor phosphorylation. J. Biol. Chem. 263,38393897.

J. Neurochem., Vol. 55, No. 6. 1990

Billard W., Rupert0 V., Crosby G., Iorio L. C., and Barnett A. (1984) Characterization of the binding of 3H-SCH23390,a selective D1 receptor antagonist ligand, in rat striatum. Life Sci. 35, 18851893. Flaim K. E., Gessner G. W., Crooke S. T., Heys J. R., and Weinstock J. (1986) Regulation of agonist and antagonist binding to striatal D-1 dopamine receptors: studies using the selective D-1 antagonist [’H]SK&F R-83566. Li$e Sci. 38, 2087-2096. Green D. A. and Clark R. B. ( I 98 1) Adenylate cyclase coupling proteins are not essential for agonist-specificdesensitization of lymphoma cells. J. Biol. Chem. 256, 2105-2108. Hess E. J. and Creese I. (1987) Biochemical characterization of dopamine receptors, in Receptor Biochemistry and Methodology, Vol. 8: Dopamine Receptors (Creese I. and Fraser C. L., eds), pp. 1-27. Alan R. Liss, New York. Hess E. J., Norman A. B., and Creese I. (1988) Chronic treatment with dopamine receptor antagonists: behavioral and pharmacological effects on D 1 and D2 dopamine receptors. J. Neurosci. 8,2361-2370. Kassis S. and Fishman P. H. (1984) Functional alteration of the 0adrenergic receptor during desensitization of mammalian adenylate cyclase by 0-agonists.Proc. Natl. Acad. Sci. USA 81,66866690. Kelly E. and Nahorski S. R. (1987) Endogeneous dopamine functionally activates D-1 and D-2 receptors in striaturn. J. Neurochem. 49, 115-120. Kwatra M. M., Benovic J. L., Caron M. G., Lefkowitz R. J., and Hosey M. M. (1989) Phosphorylation of chick heart muscarinic cholinergic receptors by the (3-adrenergic receptor kinase. Biochemistry 28,4543-4547. Memo M., Lovenberg W., and Hanbauer I. (1982) Agonist-induced subsensitivity of adenylate cyclase coupled with a dopamine receptor in slices from rat corpus striatum. Proc. Natl. Acad. Sci. USA 79,4456-4460. Pittman R. N., Reynolds E. E., and Molinoff P. B. (1984) Relationship between intrinsic activities of agonists in normal and desensitized tissue and agonist-induced loss of 0-adrenergic receptors. J. Pharmacol. Exp. Ther. 230,6 14-6 18. Rosenboom P. H. and Gnegy M. E. (1989) Acute in vivo amphetamine produces a homologous desensitization of dopamine receptor-coupled adenylate cyclase activities and decreased agonist binding to the D1 site. Mol. Pharmacol. 34, 148-156. Salomon Y., Londos C., and Rodbell M. (1974) A highly sensitive adenylate cyclase assay. Anal. Biochem. 58, 541-548. Seeman P., Watanabe M., Grigoriadis D., Tedesco J. L., George S. R., Svensson U., Nilsson J. G., and Neumeyer J. L. (1985) Dopamine D2 receptor binding sites for agonists: a tetrahedral model. Mol. Pharmacol. 28, 391-399. Shimizu H., Daly J. W., and Creveling C. R. (1969) A radioisotopic method for measuring the formation of adenosine 3’,5’-cyclic monophosphate in incubated slices of brain. J. Neurochem. 16, 1609-16 19. Sibley D. R. and Lefkowitz R. J. (1985) Molecular mechanisms of receptor desensitization using the (3-adrenergic receptor-coupled adenylate cyclase system as a model. Nature 317, 124-129. Su Y.-F., Harden T. K., and Perkins J. P. (1980) Catecholaminespecific desensitization of adenylate cyclase. J. Biol. Chem. 255, 7410-7419. Watling K. J. and Dowling J. E. (198 I ) Dopaminergic mechanisms in the teleost retina: I. Dopamine-sensitive adenylate cyclase in homogenates of carp retina; effects of agonists, antagonists and ergots. J. Neurochem. 36, 559-568. Yamashita A., Kurokawa T., Higashi K., Dan’ura T., and Ishibashi S. (1986) Forskolin stabilizesa functionally coupled state between activated guanine nucleotide-binding stimulatory regulatory protein, N,, and catalytic protein of adenylate cyclase system in rat erythrocytes. Biochem. Biophys. Res. Commun. 137, 190194.