Population Without Reciprocal Antagonist/Agonist States. *?Susan R. George ... Abstract: Although dopamine agonists can recognize two states of the D2 ...
J(iurrial of Neurot lirtnistiy Raven Press. New York 5 1985 International Society for Neurochemistry
Dopamine D, Receptors in the Anterior Pituitary: A Single Population Without Reciprocal Antagonist/Agonist States *?Susan R. George, “Masayuki Watanabe, and “Philip Seeman Departments of *Phurmucology irnd ?Medicine, University of Toronto, Toronto, Ontario, Ctrntrdu
Abstract: Although dopamine agonists can recognize two states of the D2 dopamine receptor in the anterior pituitary (Dyghand Dil’w),we examined whether the dopamine antagonists such as [’Hlspiperone could recognize these two sites with different affinities. Using up to 30 concentrations of [‘Hlspiperone, however, we could only detect a single population of binding sites (porcine anterior pituitary honiogenates) with a dissociation constant (K,) of 130 pM. When specific [’Hlspiperone binding was defined by a low concentration of (+)-butaclamol (100 nM), the apparent density was low. When defined by a high concentration of (+)-butaclamol (10 pM), nonspecific sites became detectable, thus revealing two apparent populations of sites for [ 3 H ] ~ p i p e r ~ nonly e , one of which was specific for dopamine. Sodium chloride reduced the K,, of the single population of specific D, sites to 64 pM. Guanine nucleotide by itself had no effect on the K,, but enhanced the density by 25%. Since the density-enhancement could be eliminated by extensive washing of membranes, and could be restored by preincubation with dopamine, the nucleotide-induced elevation of D, density appeared to be a result of the release of tightly bound endogenous dopamine. Thus, monovalent cations and guanine nucleotides appear to have separate regulatory effects on the anterior pituitary D, receptor that modulate antagonist-receptor interactions. Several maneuvers were used to test whether [3H]spiperone could differentiate between the two agonist-detected subpopulations of
sites. Twentyfold different concentrations of [‘HIspiperone (47 pM and 1000 pM) were found to label identical proportions of receptors in the DYgh and DPwstates as detected by the agonist 6,7-dihydroxyaminotetralin (ADTN), suggesting that spiperone labelled equal proportions of Dg :h and DPwsites without differential affinity for them. In addition, competition of spiperone for D P h sites selectively labelled by the agonist [’Hln-propylnorapomorphine (NPA) had a virtually identical K,, for spiperone as did the total D2 receptor population as determined by direct binding studies (75 pM versus 64 pM). [’HISpiperone also bound to a uniform population of DP” sites induced by preincubation with guanine nucleotide with identical affinity as to the total D2 population. Thus, these data do not support a “reciprocal model” for the D, receptor (i.e., antagonist having low affinity for DYgh and high affinity for DPwin a manner reciprocal to agonists). It is therefore concluded from these studies that spiperone recognizes the DYgh and DP“ states of the receptor with equal affinity and no evidence is provided in support for reciprocal modulation of antagonistiagonist affinities. Key Words: Dopamine receptor- Anterior pituitary-Spiperone-Sodium ions-Guanine nucleotides-Antagonist-Agonist. George S. R. et al. Dopamine D, receptors in the anterior pituitary: A single population without reciprocal antagonistiagonist states. J . Neurochem. 44, 1168-1177 (1985).
Although dopamine agonists recognize two states of the D, dopamine receptor (Sibley et al., 1982; George et al., 1983), it has not yet been clearly tested whether these two states can be discriminated by dopamine antagonists. A “reciprocal model” for the D, dopamine receptor has been postulated (De Lean et al., 1982a,b) wherein one state of the receptor would have high affinity for agonist and low affinity for antagonist, whereas the other state of the receptor would have low affinity for the agonist but high affinity for the antagonist. The lit-
erature is not consistent with such a model (Sibley et al., 1982; Lazareno, 1983). The present studies were designed to test this postulate of reciprocal antagonist/agonist affinities for the two states of the D, receptor in anterior pituitary. Agonist and antagonist affinities were examined under a variety of conditions that modulate the relative proportions of the D, receptor states from 0 to 100%; the antagonist spiperone recognized both agonist-discriminated D, states with equal affinity. Antagonist interaction with the entire
Received May 30, 1984; accepted October 2, 1984. Address correspondence and reprint requests to Dr. S. R . George, Department of Pharmacology, University of Toronto, Toronto, Ontario, Canada M5S 1A8.
Abbreviutions u s e d : A D T N , 6,7-DihydroxyaminotetraIin; Gpp(NH)p, 5’-Guanylylimidodiphosphate;NPA, n-Propylnorapomorphine.
DA RECEPTOR STATES population of D, receptors was uniformly regulated by monovalent cations and guanine nucleotides.
EXPERIMENTAL PROCEDURES Membrane preparation Porcine pituitaries (frozen on dry ice; Bocknek Organic Materials, Toronto) were stored at -70°C. On each assay day, pituitaries were thawed on ice and the neurointermediate lobe and pituitary stalk were dissected away. Anterior pituitaries (-200 mg each) were minced finely with scissors, suspended in chilled buffer (pH 7.4 at 0”C, containing Tris-HC150 mM, EDTA acid 1 mM, MgCI, 4 mM, CaCI, 1.2 mM, KCI 5 mM), disrupted with a Brinkmann Polytron (20 s at setting 7), and passed through two layers of cheesecloth. The homogenate was centrifuged at 480 x g for 10 min at 4°C. The supernatant was kept aside, and the pellet was resuspended and centrifuged again similarly in an equal volume of buffer. The combined supernatants were then centrifuged at 48,000 x g for 30 min at 4°C. The resulting pellet was redispersed in buffer (200 mg original wet weightiml) containing the same ingredients as above (pH 7.4 at 37°C). This membrane suspension was preincubated for 10 min at 3 7 T , then kept chilled until addition to the incubation tubes in 1OO-kl aliquots to give a final concentration of approximately 100 kg proteiniml of incubation medium. Binding of 3H-ligands to homogenates The binding of 13H]spiperone or [3H]n-propylnorapomorphine (NPA) was done at equilibrium with incubation initiated by the addition of anterior pituitary membrane suspension to tubes containing buffer (pH 7.4 at 20°C) as described, with nialamide 12.5 F M , ascorbic acid 0.1%, 3H-ligand, and drug(s) in a final assay volume of 5.0 ml. At the end of the incubation period, bound ligand was separated from free by rapid filtration (5- 10 s), through glass fiber filter mats using a cell harvester (Skatron). Following two 10-s washes (approximately 10 ml), the filters were air-dried, punched into minivials, shaken overnight in 4-ml of scintillation fluid, and counted in a Packard 460C counter with an efficiency of 50-60%. All experiments were carried out in triplicate; all values reported are means ? SEM. Materials L3H1Spiperone (25-35 Ciimmol) and [3H]NPA (40-50 Ciimmol) were from New England Nuclear (Boston, MA). 5’-Guanylylimidodiphosphate[Gpp(NH)p] and dopamine hydrochloride were from Sigma (St. Louis, MO). ( +)-Butaclamol was donated by Ayerst Research Laboratories (Montreal); (?)-sulpiride was donated by Delagrange International (Paris, France). Data analysis Data were analyzed by a weighted nonlinear leastsquares curve-fitting program LIGAND (Munson and Rodbard, 1980) with minor modifications, using a Digital Systems DEC-10 computer. Data were fit successively to one-site and two-site models, based on the assumption that deviations of the data from that expected for a onesite bimolecular reaction (law of mass action) resulted from different affinities of the ligand to more than one class of binding site present. A two-site model was retained as appropriate only if a statistically significant improvement of fit was obtained. Parameter estimates
1169
within the selected m d e l were obtained with iterative fits, with no constraints imposed, such that the best-fitted simulated line for a given data set did not significantly differ from the experimental data points. Data were plotted by a Nicolet Zeta-8 plotter using our own program (D. E. Grigoriadis, S. R. George, and P. Seeman, unpublished).
RESULTS Saturation of binding by [3H]spiperone T h e binding of [3H]spiperone t o anterior pituitary membranes was saturable, attaining equilibrium in 60-70 min a t room temperature. An incubation time of 75 min, therefore, was used routinely. Under all conditions studied, less than 1% of [3H]spiperone w a s bound (directly measured by centrifugation studies, d a t a not shown), t h u s validating t h e assumptions that the concentration of receptors was infinitely dilute a n d that t h e total [3H]spiperone concentration closely approximated the true “free” [3H]spiperone concentration (cf. Seeman e t al., 1982, 1984). Saturation c u r v e s of [’Hlspiperone binding t o porcine anterior pituitary membranes were best fit by a model suggestive of a single class of noncooperative binding sites. In a series of eight detailed saturation experiments, some of which included up t o 30 experimental data points per isotherm, each curve was best-fitted by a model suggesting a single uniform population o f D, s i t e s d e t e c t e d by [3H]spiperone. Simultaneous coanalysis of the entire series of saturation curves yielded parameters for a single class of binding sites with dissociation constants (KD) for [3H]spiperone of 130.6 i 5.7 pM and a density of 5.35 +- 0.15 pM, equivalent t o 1.34 x sites per gram wet weight of porcine anterior pituitary tissue. Analysis of these isotherms for more than one class of binding site led t o either a deterioration of fit, no significant improvement in fit, or a complete inability of the program t o fit the data t o a model incorporating two sites, thus allowing us t o conclude that the best fit statistically was t o a single binding site. A composite of data from three saturation isotherms is shown in Fig. 1 (only three are shown for clarity). Effect of NaCl on [3H]spiperonebinding In order t o preclude the possibility that differences in the NaCl composition of the incubation media from various laboratories accounted for differences in the results, NaCl 100 m M was added a n d f o u n d t o d e c r e a s e significantly t h e KD of [3H]spiperone for the single class of D, binding sites in anterior pituitary t o 63.6 2 2.6 p M (n = 8, p < 0.01) with n o significant difference in t h e total number of sites detected. A representative pair of [ 3 H ] ~ p i p e r ~saturation ne isotherms (in the presence a n d absence of NaCl 100 mM) is shown as Scatchard analyses in Fig. 1B. J . Neurochem., Vol. 44, N o . 4, 1985
I170
S . R . GEORGE ET AL.
A
1
t .oo
I
I
1
I
cl
I
0.0
I
I
I
2.0
1.0
3.0
BOUND
I
I
(bb)
I
7.0
6.0
I
I
B
I
\
5.0
I
. 00
FIG. 1. A: Binding isotherm of [3H]spiperone in anterior pituitary showing saturability (inset) and Scatchard analysis representing binding to a single uniform class of dopamine receptor sites. Data shown is a composite of three separate studies, with each point representing the mean of triplicate or quadruplicate observations. B:A representative Scatchard analysis of [3H]spiperone binding in the absence and presence of 100 mM sodium chloride. In the absence of sodium, K, = 107.8 ? 17.7 pM, B,, = 4.62 -t 0.57 pM; in the presence of sodium, K, = 58.1 i 3.2 pM, B,, = 4.84
W W
. 06
sodi'um
t Y LL
\
a
.04
7
3
0
m .02
'6 no sodi'um
. 00
1 0.0
1
1.0
1
2.0
BOUND
I
(bb)
I
4.0
\
I
5.0
5
0.10 PM.
DA RECEPTOR STATES
a
7 0
6 0
-
0
-
1
3
a
v
1
0 0,
1171
-
4 L L 'H-sp/perone
p
1
lo-lo
58 pM
,o-9
I O - ~
(+) B U T A C L A M O L
lo-'
I O - ~
(M)
FIG. 2. (+)-Butaclamol displacement of [3H]spiperone binding in anterior pituitary. High (575 pM) and low (58 pM) concentrations of [3H]spiperone were used. Arrows denote KD values that are identical for the two curves, 1.3 nM and 1.1 nM, respectively.
To resolve whether the effects of NaCl on [ 3 H ] ~ p i p e r ~ binding ne were specific for either the cation or the anion, the experiments were repeated with sodium sulfate 50 mM, and identical effects were obtained as with sodium chloride 100 mM ( K , 56 pM cf. 64 pM). Similar effects were demonstrated with equimolar concentrations of potassium chloride ( K , 78 pM) and lithium chloride (52 pM).
Role of (+)-butaclamol baseline in definition of nonspecific [3H]spiperone binding On occasion it was noted that the Scatchard plots displayed slight curvilinearity at high [3H]spiperone concentrations, and that the data at times appeared to fit a two-site model, although only rarely was there a significant improvement in fit from a onesite model. To clarify whether these findings represented evidence for two antagonist-detectable sites or evidence for one site with residual nonspecific binding, the particular ( + )-butaclamol concentration chosen to define specific [3H]spiperone binding was studied. As shown in Fig. 2, in the presence of 100 mM NaCl, (+)-butaclamol displacement of 58 pM L3H1spiperone binding was monophasic with 10% of total binding being nonspecific. In contrast, ( + )-butaclamol displacement of an approximately 10-fold higher concentration of
0.0
,.o
2.0
8.6
I. 0
e.0
'.'( p M i n FIG. 3. Scatchard analysis of [3H]spiperone binding isotherms in anterior pituitary with specific binding defined by 0.1 (A), 1.0 (B), and 10.0 p.M (C) (+)-butaclamol, respectively. The data are seen to model best to one-site (A,B) or two-site (C) fits. E%ND
[3H]spiperone (575 pM) was also steep, but no welldefined plateau of maximal ( + )-butaclamol activity was seen. Instead, there was a slow drift downward with increasing displacement of ['Hlspiperone binding by increasing (+)-butaclamol concentrations. (+)-Butacfamol (1 p M ) displaced 75% of total 575 pM [3H]spiperonebinding and 90% of total 58 pM ['Hlspiperone binding. In order to illustrate the vital importance of the appropriate selection of the baseline that defines the total receptor population, saturation isotherms of [ 3 H ] ~ p i p e r ~binding ne were repeated with a variety of baselines. As shown in Fig. 3, the use of ( + ) butaclamol 0.1 p M , I p M , and 10 KM to define nonspecific binding of ['Hlspiperone had dramatically different results. The binding isotherm with J . Neurochem., V o l . 44, No. 4 , 1985
1172
S. R. GEORGE ET A L .
5 120 t VI
z 0
z 0 t +.
80
40 0 VI +. VI
"
0
-
6
3
-
Na
GN
Na
+
GN
w m
L
4
0 L
z
2
n
o
,VI + W z
FIG. 4. Effect of 100 m M sodium chloride (Na+) and 100 Gpp(NH)p (GN) on the K, (dissociation constant) and B,,, (density of sites) of specific [3H]spiperone binding i n anterior
pituitary, defined by 1 F M (+)-butaclamol. Significant difference from control at p < 0.01 is denoted by *. Values shown are the means 2 SEM of four to eight independent observations.
+
specific [3H]spiperone defined by 0.1 p M ( )-butaclamol was best fit by a one-site model with deterioration of fit when a two-site fit was attempted; the isotherm defined by 1 pM (+)-butaclamol was also best fit by a one-site model, with no significant improvement in fit by the two-site model; and the isotherm defined by 10 p M (+)-butaclamol was best fit by a model suggesting two sites. Saturation curves obtained with the specific D, receptor antagonist ti)-sulpiride 10 F M (in the presence of NaCl 100 mM) to define [3H]spiperone specific binding consistently yielded a single class of sites with an affinity of 52.5 k 8.4 pM and a density of 5.34 ? 0.26 pM. Effect of guanine nucleotide on [ 3H]spiperone binding The effect of the guanine nucleotide Gpp(NH)p on the binding of ['Hlspiperone to D2 receptors was examined. Under all conditions examined, the binding isotherm of ['Hlspiperone was best fit by a single, uniform class of sites. The presence of 100 p M Gpp(NH)p did not change the affinity of ('Hlspiperone for the D, receptor but resulted in a 25% increase in the density of sites detected (Fig. 4). The addition of 100 mM NaCl together with 100 p M Gpp(NH)p resulted in an additive effect equivalent to the sum of each of NaCl and Gpp(NH)p effccts separately, i .e., to increase the affinity and the number of sites detected by ['Hlspiperone, respectively (Fig. 4). To investigate the origin of the increased density of receptors seen with the inclusion of Gpp(NH)p, binding isotherms were repeated in membranes that had been extensively washed three times by resuspension and recentrifugation after the preincubation
step. The Gpp(NH)p effect was lost in well-washed membranes and, as shown in Fig. SA, there was no significant increase in total receptor density C6.32 +- 0.29 pM in the absence and 6.52 k 0.38 pM in the presence of Gpp(NH)p]. Attempts were made to identify the substance responsible for the observed effects of guanine nucleotides. Membrane homogenates were preincubated with dopamine at a concentration of 100 p M for 10 min at 37°C and then aliquots were added to incubation tubes. After preincubation with dopamine, the effects of NaCl and Gpp(NH)p were greatly exaggerated. As shown in Fig. 5B, in the absence of sodium chloride or guanine nucleotide, the dissociation constant (K,) of ['Hlspiperone for the D, sites was 221.72 +- 51.77 pM and the detected density of sites 2.96 i 1.49 pM, changing in the presence of Gpp(NH)p to a K,of 164.37 16.98 p M and a density of 6.78 t 0.41 pM, respectively, i.e., no significant change in affinity, but a 129% increase in total density of sites. Sodium chloride addition, as shown in Fig. 5C, resulted in a twofold increase in the affinity of D, sites for [ 3 H ] ~ p i p e r ~ nas e , was similarly detected in the absence of exogenously added dopamine; it also resulted in a 63% increase in the density of sites detected. The addition of sodium chloride together with Gpp(NH)p did not further alter the affinity appreciably but resulted in a further 47% increase in the detected density of sites. +_
Agonist/[3HIspiperone competition Since we could only detect one population of D, receptors by saturation isotherms with [3H]spiperone under a variety of conditions, other methods were devised to test whether the antagonist could indeed distinguish between the two states of the receptor and whether these states had reciprocally high and low affinities for agonistb and antagonists, respectively, as predicted by the "reciprocal model. ' ' Shown in Fig. 6A are typical data obtained from agonist competition curves of [3H]spiperone binding in anterior pituitary membranes. The binding was dopaminergic, as confirmed by the rank order of overall potencies: 6,7-dihydroxyaminotetralin (ADTN) > dopamine > norepinephrine > serotonin. ADTN and dopamine were able to distinguish between the two states of the D, receptor with the curves being analyzed to identify agonist high-affinity sites and agonist low-affinity sites in approximately equal proportions as shown (Fig. 6A). Membranes were also preincubated with 500 p M unlabelled spiperone, which theoretically should have saturated the antagonist high-affinity sites if present so that the majority of detected sites should demonstrate low-affinity for antagonists but high-affinity for agonists. Fig. 6B shows the results of agonist competition curves performed in mem-
D A RECEPTOR STATES
1173
B 05
FIG. 5. Attenuation of the Gpp(NH)p effect on the density of [3H]spiperone-detected sites (B,,, from Scatchard analysis) in anterior pituitary by extensive washing of membranes (A). K, and &,,,values of the two lines are not significantly different. Preincubation of membranes with 100 p.M dopamine enhanced the effect of Gpp(NH)p on Bmaxof[3H]spiperone binding in the absence of sodium (B),and in the presence of 100 m M sodium chloride (C).
dopamine preincubation -
-
W
.
W
c!L
-k 0
03
Z
3 02
0
m
01
00
C 10
-
dopamine preincubation -
n
5
.04
0
m
02
--
nn 0.0
1.0
2.0
3.0
BOUND
('pk)
5.0
6.0
7.0
J . Neurorhem., Vol. 44, N o . 4, 1985
S . R . GEORGE ET A L .
1174
presence o f
AGONIST
(M)
10"
lo-'
500 pM s p l p e r o n e
lo6
AGONIST
lo-'
(M)
to-'
(+)arc
FIG. 6. Competition of specific [3H]spiperone binding [as defined by 1 IJ.M (+)-butaclamol, (BTC)] by (t)-6,7-dihydroxyaminotetralin (ADTN), dopamine (DA), (-)-norepinephrine (NE), and serotonin (5-HT), before (A) and after (6)partial occlusion of dopaminergic sites by 500 pM unlabelled spiperone. [3H]Spiperone concentrations of 100-200 pM were used for the studies in A and 900-1000 p M for the studies in 9.Arrows denote the K,values of the agonist-detected D, sites. The proportion ("A) of sites in the two states detected by ADTN and DA are indicated.
branes containing receptors partially occluded with 500 pM spiperone. Again, the [3H]spiperonebinding studied was dopaminergic as evidenced by the rankorder potencies of competing agonists. However, unlike the results predicted by the reciprocal model, the agonists ( K , shifted to the right because high ['H]spiperone concentrations had to be used) still detected D2 receptors in high- and low-affinity states in the same proportions as detected previously. The relative proportions of receptors present in the agonist high- and low-affinity states as detected by ADTN competing for 47 pM ['Hlspiperone and 1 nM ['Hlspiperone are shown in Fig. 7. Contrary to what might be expected if the 20-fold different concentrations of [ 3 H ] ~ p i p e r ~ were n e labelling different proportions of the subpopulations of D, sites,
ADTN
(M)
FIG. 7 . ADTN competition of dopamine receptor binding by 47 pM and 1 nM [3H]spiperone, showing virtually identical K, values (arrows) and proportions of sites (%). The K, values of ADTN obtained for the upper line were 9 t 2 and 220 ? 30 nM, and 8 5 3 and 160 t 36 n M for the lower line.
J . Nerrrochetn., Vol. 44, No. 4, 1985
ADTN competition of the high and low concentration of bound [3H]spiperone yielded similar results with respect to the proportion of D, receptors detected in the agonist high- or low-affinity states, i.e., S0:SO. Antagonist affinity for agonist-detected high and low D, affinity states The labelled agonist, [3H]NPA was used to label selectively the agonist high-affinity state of the receptor (Sibley and Creese, 1983). As seen in Fig. 8, ADTN competed for ['HINPA binding in the presence of 100 mM NaCl in a monophasic fashion, for a uniform population of labelled sites having K , 5.4 k 1.5 nM for ADTN, which was equivalent to the affinity for the agonist high-affinity state obtained
D A RECEPTOR STATES
100% ACONIST HIGH A F F I N I T Y
100% ACONIST LOW A F F I N I T Y
MIXED H I G H & LOW A F F I N I T Y
7 0 0-
60 0 h
2 a
-
v
c
5 0 0-
2
-
< t-
m40 0 -
z
0
0
-
2 3 0 0-
0
H
tQ
H 2 0 0-
0
0 ln
-
Ln
H 10 0-
a
00-
FIG. 9. Comparison of the dissociation constants (K,) of [3H]spiperone for D, dopamine receptor populations present 100% in the agonist high-affinity form ([3H]NPA-detected), 100% in the agonist low-affinity form (GTP preincubated) or in a mixed proportion of both forms. Values shown are the means of parameters obtained from Scatchard analyses or competition curves.
by ADTN competition of [3H]spiperone binding (2.2 -+ 0.4 nM). Also seen in Fig. 8 is spiperone competition of [3H]NPA binding, also monophasic, the sites displaying an affinity of 79.0 k 7.4 pM for spiperone, equivalent to the overall affinity of [3H]spiperone for the total population of D, sites (63.6 t 2.6 pM). Membranes were also preincubated with guanine nucleotide (GTP 1 mM for 10 min at 37°C) to convert all the receptors to the agonist low-affinity state (Sibley et al., 1982). The affinity for [‘HIspiperone in the presence of 100 mM NaCl for a homogeneous population of D, receptors in the agonist low-affinity state (by GTP preincubation; 66 pM) was identical to that of a homogeneous population of D, receptors in the agonist high-affinity state ([3H]NPA-detected; 79 pM) and a mixed population of D, receptor sites (64 pM), as shown in Fig. 9. DISCUSSION The present study indicates that spiperone failed to discriminate between the two agonist-detected forms of D, in anterior pituitary. Spiperone detected a single, homogeneous population of dopaminergic binding sites. Moreover it appears critical that the concentration of (t)-butaclamol chosen to define
1175
specific binding be done on the basis of competition curves. At concentrations of [ 3 H l ~ p i p e r ~ nine the range of the affinity constant of the D, sites detected (58 pM), 90% of drug bound is to receptors and (+)-butaclamol is able to compete specifically for these sites (Fig. 2). However, at approximately 10-fold higher [3H]spiperone concentrations (575 pM), there is more drug nonspecifically bound that can also be displaced by high concentrations of ( + )butaclamol (Fig. 2). Thus, the inappropriate choice of drug concentrations to define specific binding can artifactually give rise to a second class of binding sites (Fig. 3), apparent only at the higher radioligand concentrations. In studies where D2specific drugs were used, such as sulpiride (our studies; Theodorou et al., 1980) or domperidone (Lazareno, 1983), only a single population of antagonist-detected D, sites was found. Thus, before evidence can be accepted for a second antagonist-detected D, site, the specific dopaminergic nature of such a site would have to be conclusively demonstrated since nonspecific saturable binding sites may be present (Seeman, 1980). D, dopamine receptors in brain are regulated by sodium ions and guanine nucleotides (Hamblin and Creese, 1982). Effects of these substances on D, receptor-antagonist interactions have not been fully investigated previously. In our system, sodium chloride significantly increased the affinity of D, receptors for [3H]spiperone by a factor of 2. This effect was specific for monovalent cations but not for the anion. Guanine nucleotide addition did not alter the affinity of the receptors for [3H]spiperone,but caused a significant increase in the maximal binding capacity (Fig. 4). These additional [’Hlspiperone binding sites apparent in the presence of Gpp(NH)p could potentially have several explanations. The most plausible hypotheses considered were: ( 1 ) the conversion of previously low antagonist-affinity sites into high antagonist-affinity sites b y Gpp(NH)p; (2) the “clearing” or unmasking of agonist-occupied sites so that they were now available for [3Hlspiperone binding. In keeping with the latter hypothesis, it was possible to reduce the Gpp(NH)p effect by extensive washing of membranes (Fig. 5A) and to enhance the Gpp(NH)p effect by preincubation with dopamine (Fig. 5B,C). It was also possible to restore the phenomenon in extensively washed membranes by subsequent preincubation with dopamine (data not shown but identical to Fig. 5). It was also apparent from these studies that the effect of sodium to enhance D, receptor affinity for [ 3 H l ~ p i p e r ~ nwas e independent of the presence of dopamine, since identical effects were seen in wellwashed membranes as after preincubation with dopamine. The inconsistently observed effects of sodium to increase the maximal binding capacity of
J . Nrururhem., V d . 44, No.4 , 1985
1176
S.
R. GEORGE ET AL.
[3Hlspiperone was clearly dependent upon dopamine concentrations in the preparation, as there was a pronounced increase in total binding sites with addition of sodium after dopamine preincubation (Fig. 5). The concomitant addition of sodium and Gpp(NH)p merely resulted in the additive effects of each seen separately. These findings suggest that the effect of sodium ions to increase the affinity and the effect of Gpp(NH)p to unmask agonist-occupied receptors not otherwise available for ['Hlspiperone binding are mediated at different loci on the receptor-effector complex of D, receptors. Similar findings have been observed for aipha,-adrenergic receptors (Woodcock and Murley, 1982; Limbird et al., 1982; Snavely and Insel, 1982). In spite of the inability to demonstrate [3H]spiperone differentiation of the two agonist-affinity states of D2, attempts were made to test the hypothesis of reciprocal modulation of agonist and antagonist affinities. Since the reported affinities for the two antagonist sites are of the order of 40 pM and 400 pM for ['Hjspiperone (De Lean et al., 1982a), membranes were preexposed to 500 pM spiperone, so that 90% of the high-affinity sites (by a concentration of 10 x K,) and 50% of the lowaffinity sites (by a concentration equal to K,) should be ocduded. According to the hypothesis, the remaining unoccupied sites should largely (over 80%) have low-affinity for antagonists and highaffinity for agonists. Agonist competition curves of these sites (Fig. 6 ) revealed proportions of agonist high- and low-affinity (-5050) that were identical to control membranes, suggesting that 500 pM spiperone had occluded an equal number of agonist high- and low-affinity sites. By the same rationale, 47 pM [3H]spiperone should be expected to label 50% of antagonist high-affinity sites (at a concentration equal to K,) and 10% of antagonist lowaffinity sites (at a concentration of 1/10 KD), but ADTN detected approximately 50% of [3H]spiperone-labelled sites with high-affinity and approximately 50% with low-affinity (Fig. 7). Simifarly, I nM [3H]spiperone should have labelled very different proportions of receptor subpopulations (at least 90% of antagonist high-affinity and greater than 50% of antagonist low-affinity) but ADTN competition for 1 nM ['Hlspiperone indicated that the receptors were present in the agonist high- and low-affinity forms in equal numbers (Fig. 7). Radioligand binding studies with the agonist [?H]NPA were performed to label preferentially the agonist high-affinity state of the D2 receptor. Agonist (ADTN) competition of L3H]NPA binding (Fig. 8) revealed affinity identical to that obtained for the agonist high-affinity state obtained by indirect binding studies. This form, postulated to have lowaffinity for antagonist, is shown by competition of ['HINPA binding, to have an affinity for spiperone, equivalent to that obtained by direct [3H]spiperoneJ . Nerrroclwm., Vol. 44, No. 4 , 198s
binding isotherms. Similarly, the affinity of recep-
tors in the agonist low-affinity state (converted by
GTP preincubation) for [3H]spiperone was shown to be similar to that of receptors in the agonist highaffinity state, or for a mixed population of sites in both high- and low-affinity states (Fig. 9). Thus, we are unable to provide any evidence in support of a reciprocal model of antagonistiagonist affinities for the D2 recepfor of anterior pituitary. [3H1~piperonedetects a single, uniform population of sites in porcine anterior pituitary membranes under a variety of conditions. In addition, we have shown that modulation of the receptor by monovalent cations and guanine nucleotide alters receptorantagonist interactions. Acknowledgments: We thank D. Grigoriadis, Dr. J . Wells, and Dr. L. E n d r e n yi for excellent advice and assistance in computer-assisted analysis. This research was supported by the Medical Research Council of C anada and t h e Ontario Mental Health Foundation. M.W. is a postdoctoral Fellow of the Canadian Friends of Schizop h re n ic s . S.R.G. was a p o s td oc t or a l Fellow of M R C , Canada.
REFERENCES De Lean A., Kilpatrick B. K . , and Caron M. G. ( 1 9 82 ~) Guanine nucleotides regulate both dopaminergic agonist and antagonist binding in porcine anterior pituitary. E t i d ~ ~ r i ~ l ~ g p 110, 1064-1066. De Lean A,. Kilpatrick B. K.. and Caron M . G. (1982h) Dopamine receptor of the porcine anterior pituitary gland: evidence for two affinity states of the receptor discriminated by both agonists and antagonist%.M o l . P h r > ~ m 1 c ~22, ) l . 290297. George S. R.. Watanabe M . , and Seeman P. (1983) The dopamine receptor of the anterior pituitary gland, in L)opamine Rc7cepiors, ACS Sympo.\cum Series N o . 224 (Kaiser c. and Kebabian J . W., eds). p. 93-99, American Chemical Society. Hamblin M. W. and Creese I. (1982) %-Dopamine binding to rat striatal D-2 and D-3 sites: enhancement by magnesium and inhibition by guanine nucleotides and sodium. Life Sci. 30, 1587- 1.595. Lazareno S. (1983) Effects of GTP on >H-domperidone binding and its displacement by dopamine in rat striatal membranes. J . Recep. k e s . 3, 1631175. Limhird LA.~E.., Stxck . . ~ . J. L . . and Smith S. K. (1Y82) Sodium ion modulates agonist and antagonist interactions with the human platelet alpha2-adrenergic receptor in membrane and solubilized preparations. Mol. Phiirmacol. 22, 609-617. Munson P. J. and Rodbard D. (1980) LIGAND: a versatile computerized approach for characterization of ligand-binding systems. Anal. Biochem. 107,220-239. Seeman P. ( 1980) Brain dopamine receptors. Pharmucoi. Rcw. 32, 229-313. Seeman P., Ulpian C., and Wells J. (19821 Dopamine receptor parameters (detected by 'H-spiperone) depend on tissue concentration. S O C . Neurosci. Absrr. 8 , 718. Seeman P., Ulpian C., Wreggett K. A . , and Wells J. W. (1984) Dopamine receptor parameters detected by ['Hlspiperone depend on tissue concentration: analysis and examples. J . NeLrrochem. 43, 221 -235. Sibley D. R. and Creese I . (1983) Regulation of ligand binding to pituitary D-2 dopaminergic receptors: effects of divalent cations and functional group modification. J. B i d . Chem. 258. 4957-4965. ~
DA RECEPTOR STATES Sibley D. R., De Lean A,, and Creese 1. (1982) Anterior pituitary dopamine receptors: demonstration of interconvertible high and low affinity states of the D2 dopamine receptor. J . Biol. Chern. 257, 6351-6361. Snavely M. D. and lnsel P. A. (1982) Characterization of alphaadrenergic receptor subtypes in the rat renal cortex: differential regulation of alphal- and alpha2-adrenergic receptors by guanyl nucleotides and N a + . Mol. Phuvmacol. 22, 532546.
1177
Theodorou A. E., Hall M. D., Jenner P., and Marsden C . D. (1980) Cation regulation differentiates specific binding of 3Hsulpiride and 'H-spiperone to rat striatal preparations. J . Phurm. Phurmacd. 32, 441-444. Woodcock E. A . and Murley B . (1982) Increased central alpha?adrenergic receptors measured with ['HI-yohimbine in the presence of sodium ion and guanyl nucleotides. Biochem. Biophys. Res. Commun. 105, 252-258.
J . Nrurochem., Vol. 44, N o . 4 , 1985
