Sage Journals: Discover world-class research

Abstract

According to the ternary complex model of G-protein linkage to receptors, agonists increase the affinity of the receptors for the G protein. The model predicts that an endogenous agonist's constant of inhibition toward an agonist radioligand is lower than that toward an antagonistic radioligand. The authors hypothesized that competition from endogenous dopamine in striatum of living mice should have a greater effect on the binding of the D_2,3 partial agonist N-[³H]propylnorapomorphine than on the binding of the D_2,3 antagonist [¹¹C]raclopride. The baseline binding potential (p_B(0)), defined as the ratio of bound-to-unbound ligand in the absence of competition from endogenous dopamine, was simultaneously measured in mouse striatum for [¹¹C]raclopride (p_B(0) = 8.5) and N-[³H]propylnorapomorphine (p′_B(0) = 5.3). The baseline was established by treatment with α-methyl-p-tyrosine and reserpine. Relative to these baseline values in saline-treated mice, the p_B of N-[³H]propylnorapomorphine decreased 52% whereas the p_B of [¹¹C]raclopride decreased only 30%, indicating greater sensitivity of the former compound to inhibition by synaptic dopamine. Furthermore, amphetamine decreased the p_B of N-[³H]propylnorapomorphine to a greater extent (73%) than that of [¹¹C]raclopride (43%) relative to the reserpine condition. For both radioligands, the occupancy of the dopamine receptors by endogenous agonist obeyed Michaelis-Menten kinetics over a wide range of agonist concentrations established by the pharmacologic treatments. The apparent inhibition constant of endogenous dopamine depended on the dopamine occupancy and decreased to a value 1.66 times greater for N-[³H]propylnorapomorphine than for [¹¹C]raclopride at its highest occupancies. The results are consistent with the hypothesis that agonist binding is more sensitive than antagonist binding to competition from endogenous dopamine. Therefore, dopamine agonist ligands may be superior to benzamide antagonist ligands for the estimation of dopamine receptor occupancy by endogenous synaptic dopamine. The analysis of the effect of dopamine occupancy on the inhibition of N-[³H]propylnorapomorphine binding indicated a limited supply of G protein with a maximum ternary complex fraction of 40% of maximum agonist binding capacity.

Keywords

Dopamine Receptors Affinity Agonist Amphetamine Reserpine

Dopamine D_2,3 antagonist binding sites in living brain can be labeled with [¹¹C]raclopride and other benzamide radioligands for positron emission tomography (PET) or single photon emission computed tomography (SPECT) studies of dopamine receptors. The binding of exogenous ligands to dopamine receptors in vivo normally occurs in the presence of competition from endogenous synaptic dopamine. Consequently, pharmacologic depletion of dopamine increases the binding of benzamide radioligands in the brains of rodents (Young et al., 1991; Inoue et al., 1991), nonhuman primates (Ginovart et al., 1997), and humans (Abi-Dargham et al., 2000) in proportion to the prevailing basal occupancy. Conversely, the stimulation of dopamine release with amphetamine or the blockade of dopamine transport with cocaine results in further displacement of benzamide radioligands from specific D₂ binding sites in the striatum of living rats (Young et al., 1991), baboons (Dewey et al., 1993; Carson et al., 1997), and humans (Schlaepfer et al., 1997; Laruelle et al., 1997). Thus, changes in radioligand binding in the brain can be indicative of altered synaptic dopamine concentrations.

The pharmacodynamic response to G-protein–mediated neurotransmission is generally believed to occur in proportion to the quantity of ternary agonist receptor–G-protein complexes (Burstein et al., 1997), as defined by DeLean et al. (1980) and Samama et al. (1993). The affinity of G-protein–coupled receptors for the activating agonist is low when the receptor is dissociated from its G protein and high during an interaction with the guanosine triphosphate (GTP)-free form of the protein (GTP-shift) (Jiang et al., 2001). This property of G-protein–coupled agonist-binding sites affects the interpretation of studies of dopamine release and can be modeled. An extended model proposed by Gjedde and Wong (2001) was based on the original ternary complex model of del Castillo and Katz (1957), according to which the quantity of bound agonist can be expressed as

B_{a} = B_{max} (\frac{C_{a}}{K_{a}^{'}}) / (1 + \frac{C_{a}}{K_{a}^{'}})

(1)

where a is the specific agonist, B_a represents the quantity of the receptor–agonist complex that is not G-protein coupled, K_a is the basal affinity to the agonist, and K′_a represents receptor affinity as modified by G-protein binding as follows:

K_{a}^{'} = K_{a} / (1 + \frac{C_{g}}{K_{g}}) = \frac{K_{a}}{1 + χ_{g}}

(2)

where C_g is the G-protein concentration, χ_g is the concentration of G protein relative to its affinity for the ternary complex, and K_g is the receptors' half-saturation concentration toward the GTP-free G proteins (for definitions see Table 1).

TABLE 1.

Standard definitions of terms used*

Symbol	Definition	Description (unit)
α	≈1/(χ_ap_B(0))	Isotope correction factor (ratio)
B _a	B_max σ_a	Mass of bound agonist (μmol/g)
B _g	B_max σ_a σ_g	Mass of bound GTP-free G protein (μmol/g)
B _max		Maximum binding capacity in unit of mass (μmol/g)
C		Aqueous concentration of receptor ligand (μmol/mL)
C _a		Aqueous concentration of agonist receptor ligand (μmol/mL)
C _g		Aqueous concentration of GTP-free G protein (μmol/mL)
G		Mass of all GTP-free G protein (μmol/g)
K		Michaelis half-saturation concentration of ligand (μmol/mL)
K _a		Half-saturation concentration of agonist measured with antagonist tracer (μmol/mL)
K ^′ _a	K_a/(1 + [C_g/K_g])	half-saturation concentration of agonist measured with agonist tracer (μmol/mL)
K_d		Half-saturation concentration of antagonist radioligand (μmol/mL⁻¹)
K ^′ _d		Half-saturation concentration of agonist radioligand (μmol/mL⁻¹)
K_g		Receptor half-saturation concentration towards GTP-free G protein (μmol/mL)
χ	C/K	Normalized concentration of receptor ligand (ratio)
χ_a	C_a/K_a	Normalized concentration of agonist measured with antagonist tracer (ratio)
χ^′_a	C_a/K^′_a	Normalized concentration of agonist measured with agonist tracer (ratio)
χ_g	C_g/K_g	Normalized concentration of unbound GTP-free G protein (ratio)
χ_G	G/(K_gV^″_d)	Normalized mass of all GTP-free G protein (ratio)
p _B	B_max/(KV_d)	Binding potential of receptor ligand (ratio)
p _B(0)	B_max/(K_dV_d)	Binding potential of antagonist radioligand in absence of competition (ratio)
p ^′ _B(0)	B_max//(K^′_dV^′_d)	Binding potential of agonist radioligand in absence of competition (ratio)
p ^″ _B(0)	B_max/(K_gV^″_d)	Binding potential of GTP-free G protein in absence of competition (ratio)
p ^′ _B(I)	B_max/(K_dV_d[1 + χ_a])	Binding potential of antagonist radioligand in presence of bound agonist (ratio)
p ^′ _B(I)	B_max/(K^′_dV^′_d[1 + χ^′_a])	Binding potential of agonist radioligand in presence of bound agonist (ratio)
σ	χ/(1 + χ)	Receptor occupancy of receptor ligand (ratio)
σ_a	χ_a/(1 + χ_a)	Receptor occupancy of agonist measured with antagonist tracer (ratio)
σ^′_a	χ^′_a/(1 + χ^′_a)	Receptor occupancy of agonist measured with agonist tracer (ratio)
σ_g	χ_g/(1 + χ_g)	Receptor occupancy of GTP-free G protein (ratio)
V _d		Physical volume of distribution of antagonist (mL/g)
V ^′ _d		Physical volume of distribution of agonist (mL/g)
V ^″ _d		Physical volume of distribution of GTP-free G protein mL/g)

Definitions are those used in publications from Wong et al. (1986) to Gjedde and Wong (2001). Primes refer to properties of agonist radioligand or to results of measurements with agonist radioligand. Double primes refer to properties of GTP-free G proteins. GTP, guanosine triphosphate.

The sensitivity of agonist binding to the presence of the ternary complex results in a biphasic displacement of antagonist radioligands by dopamine, with distinct components at nanomolar (D^High₂) and micromolar concentrations (D^Low₂) (Cresse and Sibley, 1979; Battaglia and Titeler, 1982). Because dopamine receptor agonists interact with the receptors and their G proteins in such a way that the affinity of the receptors toward the agonist increases when the agonist concentration increases, the interaction leads to the prediction that the inhibitory constant of dopamine toward an agonist ligand is lower than the inhibitory constant toward an antagonistic ligand. We tested the prediction by determining the dopamine inhibitory constant toward the agonist ligand N-[³H]prophylapomorphine ([³H]NPA) relative to dopamine's inhibitory constant toward the antagonist ligand [¹¹C]raclopride in the striatum of living mice.

THEORY

Definitions

The treatment assumes that the binding of radioligands is at equilibrium, but substantial changes in competition from an endogenous or exogenous agonist render this assumption invalid. When such changes occur, the calculated binding potentials represent some weighted average of the true binding potential for each of the changing levels of competition. This problem affects all in vivo studies of drug challenges, except those in which the steady state of competitor binding is established (e.g., by protracted infusion of large quantities of an innocuous competitor). No correction for the nonsteady state was possible in the present study.

The theory arises from the common principles of receptor-ligand interaction in vivo described by, among many others, Gjedde and Wong (1990), and the key points are briefly iterated here. The receptor occupancy of a ligand subject to simple competitive binding is a hyperbolic function of the ligand concentration according to the equation (Michaelis and Menten, 1913)

σ \equiv \frac{C}{K + C}

where σ is the occupancy, C the ligand concentration, and K the Michaelis half-saturation concentration of the ligand. Normalization reduces the number of variables from two to one as follows:

σ = \frac{χ}{1 + χ}

where χ is the normalized ligand concentration. The binding potential of a ligand is the mass ratio of bound to unbound ligand (Gjedde et al., 1986)

p_{B} \equiv \frac{B_{max}}{(K + C) V_{d}}

which normalizes to

p_{B} = \frac{B_{max}}{K V_{d} (1 + χ)}

where p_B is the binding potential and V_d is the volume of distribution of unbound and nonspecifically adsorbed ligand. By combining the equations for χ = 0 and χ > 0, the binding potential becomes a function of the normalized concentration of the ligand and the binding potential in the limiting case of χ = 0

p_{B} = \frac{p_{B (0)}}{1 + χ}

where p_B(0) is the binding potential in the limiting case of zero occupancy.

Agonist occupancy

The occupancy of the endogenous or exogenous agonist was calculated from the decline of the binding potential relative to the dopamine-depleted baseline obtained with reserpine plus α-methyl-p-tyrosine methyl ester (AMPT), according to a modification of the previously described equation of Gjedde and Wong (2001)

p_{B (I)} = (1 - σ_{a}) p_{B (0)}

(3)

where σ_a is the occupancy of an agonist, p_B(I) is the binding potential of a tracer radioligand in the presence of the agonist, and p_B(0) is the binding potential in the absence of the competing agonist. This linear equation assumes a constant agonist concentration and a constant affinity of the receptors toward the radioligand, regardless of brain region or agonist concentration. However, for mixed antagonist and agonist radioligands and occupying competitors, the decline of the binding potential also varies with a variable affinity due to the agonist binding and with the variable receptor density and affinity toward different radioligands in the agonist-depleted state.

In the present application of the distinct affinities and binding potentials of one specific antagonist and one specific agonist radioligand, for all magnitudes of agonist binding, p_B(I) as a function of p_B(0) was approximated by the arbitrary relation

p_{B (I)} = (1 - e^{- α p_{B (0)}}) p_{B (0)}

(4)

where α is a scaling factor that depends on the apparent inhibitory constant of the inhibiting agonist relative to its concentration and on the maximum binding x capacity and affinity of the receptors toward the specific radioligand in the agonist-depleted state. Therefore, the product α x p_B(0) does not depend on the properties of the radioligand used to determine the binding potentials. When the magnitude of α had been determined experimentally, the occupancy of the inhibitor was calculated as prescribed by combining Eqs. 3 and 4 as follows:

σ_{a} = e^{- α p_{B (0)}} .

(5)

Again, the occupancy of the competing agonist is a function of the agonist's normalized concentration

σ_{a} = \frac{χ_{a}}{1 + χ_{a}}

(6)

where χ_a is the concentration of the agonist relative to its apparent affinity (χ_a = C_a/K_a or C_a/K′_a, depending on the radioligand).

The combination of Eqs. 5 and 6 yielded the solution for the normalized concentration of the competing agonist

χ_{a} = \frac{1}{e^{α p_{B (0)}} - 1}

(7)

where the apparent magnitude of the normalized concentration of the competing agonist reflects the properties of the radioligand used to study the competition. The ratio between the apparent affinities of the receptors implied by the results from two different radioligands was computed as

\frac{χ_{a}^{'}}{χ_{a}} = \frac{e^{α p_{B (0)}} - 1}{e^{α p_{B (0)}^{'}} - 1} = \frac{K_{a}}{K_{a}^{'}} = 1 + \frac{C_{g}}{K_{g}} = 1 + χ_{g}

(8)

where C_g/K_g is the concentration of unbound GTP-free G protein relative to its affinity for the receptors, and χ′_a and χ_a are the apparent normalized concentrations of the endogenous agonist inhibitor measured with an agonist and an antagonist radioligand, respectively. Equation 8 yields estimates of the normalized concentration of unbound GTP-free G protein, χ_g, which in turn yields the occupancy of the GTP-free G protein bound to agonist-primed receptors

σ_{g} = \frac{χ_{g}}{1 + χ_{g}}

(9)

where σ_g is the G-protein occupancy.

Guanosine triphospate–shift of affinity

The theory of the GTP-shift of affinity yields the prediction that the ratio between the antagonist and agonist affinities depends on the ratio between the concentration of the GTP-free G protein, which forms the ternary complex together with the receptor and its bound agonist. If this ratio declines with increasing agonist concentrations, the explanation may be that the G-protein concentration is too small to render the unbound G-protein concentration essentially independent of the binding. Calculating the quantity of ternary complexes as follows tested this hypothetical explanation:

B_{g} = B_{max} σ_{g} σ_{a}^{'}

(10)

where the bound GTP-free G protein (B_g) is the difference between the total mass of GTP-free G protein (G) and the mass of unbound GTP-free G protein (C_g V“_d), and σ_g and σ′_a are the occupancies of the GTP-free G protein and agonist, respectively, the latter of which is determined with the agonist radioligand. It follows that

B_{g} = G - C_{g} V_{d}^{'}

(11)

where V“_d is the volume of distribution of dissolved and nonspecifically adsorbed (if any) GTP-free G protein, and G is the mass of all GTP-free G protein, such that

G - C_{g} V_{d}^{'} = B_{max} σ_{g} σ_{a}^{'}

(12)

which rearranges to form the equation

\frac{G}{K_{g} V_{d}^{'}} - χ_{g} = \frac{B_{max}}{K_{g} V_{d}^{'}} σ_{g} σ_{a}^{'}

(13)

where χ_g is the normalized concentration of GTP-free G protein available for binding, and V“_d is the volume of distribution of dissolved and nonspecifically bound GTP-free G protein. The equation further rearranges to

χ_{G} - \frac{σ_{g}}{1 - σ_{g}} = p_{B (0)}^{″} σ_{g} σ_{a}^{'}

(14)

where χ_G is the normalized mass (G/[K_g V“_d]) of all GTP-free G protein, and p“_B(0) is the baseline binding potential of the G protein. This equation was solved for the agonist occupancy to form

σ_{a}^{'} = \frac{χ_{G}}{p_{B (0)}^{″} σ_{g}} - \frac{1}{p_{B (0)}^{″} (1 - σ_{g})}

(15)

and subsequently fitted to the experimentally observed relation between σ′_a and σ_g.

MATERIALS AND METHODS

Male mice (n = 126) of the CD1 strain (Charles River Laboratories, Wilmington, MA, U.S.A.) weighing 25 to 28 g were in experiments approved by the research ethics committee of Johns Hopkins Medical Institutions. In experiment 1, groups of 18 mice were pretreated with intraperitoneal injections of 200 μL saline vehicle, 1 or 10 mg/kg quinagolide (CV 205,502, Novartis), or 10 mg/kg amphetamine sulfate (Sigma Chemicals, St. Louis, MO, U.S.A.) administered 10 minutes before the radiotracer was injected. In experiment 2, groups of 18 mice were pretreated with intraperitoneal saline or reserpine (5 mg/kg) 20 hours before tracer injection, supplemented with 100 mg/kg intraperitoneal α-methyl-p-tyrosine methyl ester administered 1 hour before tracer injection, or with 100 μg/kg subcutaneous nicotine 3 minutes before tracer injection.

Mice were briefly immobilized for bolus injection to a tail vein of 200 μL saline containing a mixture of 0.1 MBq [³H]NPA (New England Nuclear; specific activity, 2 GBq/μmol) and 7 MBq [¹¹C]raclopride (specific activity 200 GBq/μmol at time of delivery) synthesized by the method of Ehrin et al. (1987) using a radioisotope generated by the Johns Hopkins GE PET Trace cyclotron. The calculated doses per injection were 35 pmol raclopride and 50 pmol [³H]NPA. Mice were killed by cervical dislocation 5, 10, 20, 30, 45, and 60 minutes after tracer injection, with triplicate determinations at each circulation time.

Brains were rapidly removed and the cerebellar hemispheres and corpora striata were dissected. The left striatum and cerebellar hemispheres were weighed and transferred to plastic vials for immediate measurement of γ emissions using a γ-counter (LKB 1285). Right striata and cerebellar hemispheres were transferred to 20-mL glass liquid scintillation vials to which 1 mL organic base (Solvable;, Packard Instruments, Downers Grove, IL, U.S.A.) was added. After dissolving overnight at 40°C, 10 mL liquid scintillation cocktail (Formula 989, Packard Instruments) was added to each vial and the tritium concentration was measured using a β-radiation counter (Tricarb, Packard Instruments).

Radioactivities were calculated as percentage injected dose per gram tissue. Using the cerebellum radioactivities as the inputs, the binding potentials of [¹¹C]raclopride and [³H]NPA in mouse striatum were calculated for the baseline and drug-treatment conditions in each of the three mice using the reference-region method of Lammertsma et al. (1996). The parameter estimates were subsequently averaged for each group of three mice.

RESULTS

Single time–radioactivity curves for [³H]NPA and [¹¹C]raclopride in the striatum of randomly selected single mice in the groups of three identically treated mice are illustrated in Fig. 1. Random results from a mouse pretreated with reserpine and AMPT are shown in Fig. 1A, and random results from mice pretreated with saline (Fig. 1B), amphetamine (Fig. 1C), and the high dose of quinagolide (Fig. 1D) are shown in Figs. 1B to 1D. The lines connect points predicted by the deconvolution of the striatal uptake from the cerebellar uptake according to the reference region model of Lammertsma et al. (1996) for the estimation of binding potential.

FIG. 1.

Radioactivity concentrations for [³H]NPA and [¹¹C]raclopride in a single mouse striatum during a 1-hour period after intravenous injection of a single bolus containing 0.1 MBq [³H]NPA and 7 MBq [¹¹C]raclopride to groups of 18 mice pretreated with (A) 5 mg/kg intraperitoneal reserpine 20 hours before tracer followed by 100 mg/kg AMPT 1 hour before tracer injection, (B) 200 μL intraperitoneal saline vehicle, (C) 10 mg/kg intraperitoneal amphetamine10 minutes before tracer injection, and (D) 10 mg/kg intraperitoneal quinagolide 10 minutes before tracer administration. Radioactivity concentrations are calculated as the percentage of total injected dose per milligram tissue. Lines connect the points predicted by the results of fitting a single-compartment model to the radioactivities measured in the striatum and cerebellum, assuming reversible specific binding in the striatum and no specific binding in the reference tissue. The resulting binding potentials, which were averaged for each group of three mice, are shown in Fig. 2.

The group means of the estimated binding potentials (p_B) under the several conditions are summarized in bar-graph form in Fig. 2A. The results of the fits of equation 4 to the data are shown in Table 2. In the baseline condition established by administration of reserpine and AMPT, the mean (± SD) of three estimates of the agonist-depleted binding potentials (p_B(0)) were 5.30 ± 0.77 for [³H]NPA and 8.42 ± 1.62 for [¹¹C]raclopride. In two separate estimations of the group mean of the saline-treated condition relative to the baseline (p_B(0)), the binding potential of [³H]NPA declined by an average of 52% (p_B(I)), whereas the binding potential of [¹¹C]raclopride declined by 30% (p_B(I)). Amphetamine reduced the binding potential of [³H]NPA by 73% and the binding potential of [¹¹C]raclopride by 47%. The low dose of quinagolide decreased the binding potential of [³H]NPA by 77% and the binding potential of [¹¹C]raclopride by 63%, and the high dose decreased both binding potentials by approximately 94%. Relative to the saline conditions, the nicotine challenge actually increased the binding potential of [³H]NPA by 10% and did not alter the binding potential of [¹¹C]raclopride.

TABLE 2.

Parameter estimates from regression of Eq. 4 to measured bindings potentials shown in Fig. 2

Treatment	R ²	α ± SE
20 h P̅ reserpine/AMPT	1.00	5.2 ± 0.2
Nicotine (0.1 mg/kg)	0.89	0.17 ± 0.03
Saline 1 (vehicle)	1.00	0.15 ± 0.01
Saline 2 (vehicle)	0.95	0.11 ± 0.02
Amphetamine (1 mg/kg)	0.94	0.089 ± 0.014
Quinagolide (1 mg/kg)	1.00	0.053 ± 0.002
Quinagolide (10 mg/kg)	0.96	0.0076 ± 0.0007

SE, standard error; AMPT, α-methyl-p-tyrosine methyl ester.

FIG. 2.

Binding potentials of [¹¹C]raclopride and [³H]NPA in the striatum of mice after the administration of several pharmacologic treatments, and the estimation of occupancy by dopamine, a putative endogenous agonist inhibitor, or quinagolide, a known exogenous agonist inhibitor, under the same condition. (A) Binding potentials of (left) [¹¹C]raclopride and (right) [³H]NPA in mouse striatum 20 hours after reserpine treatment, and after acute treatments with 0.1 mg/kg subcutaneous nicotine, 200 μL intraperitoneal saline (two separate experiments), 1 mg/kg intraperitoneal amphetamine, or 1 or 10 mg/kg quinagolide. (B) Binding potentials in the presence of endogenous or exogenous agonist inhibitor versus agonist-depleted binding potentials measured after reserpine and AMPT treatment. Curves represent regression by Eq. 4 with the corresponding estimates of α plus or minus standard error listed in Table 2. Standard deviations of binding potential estimates are hidden by size of symbols. (C) Occupancies of endogenous agonist inhibitor determined with antagonist (RAC, open circles) and agonist (NPA, filled circles) radioligand according to Eqs. 6 and 7, indicating greater estimates of occupancy with agonist (NPA) ligand. (D) Antagonist-determined occupancies of endogenous agonist inhibitor versus agonist-determined occupancies, indicating second-order polynomial relation. (E) Lower occupancies plotted at higher magnification.

From these changes of the binding potential and relative to the those measured in the agonist depleted condition, the receptor occupancies (σ_a) were calculated according to Eq. 6 and plotted as a function of the concentration of dopamine relative to its affinity, whereas χ_a (Fig. 2C) was calculated from Eq. 7. By definition the relation obeyed the Michaelis-Menten formulation for both ligands. As predicted, the occupancy was higher for [³H]NPA than for [¹¹C]raclopride in each pharmacologic condition. The resulting pairs of σ_a versus χ_a were plotted in Fig. 2C, and values of χ′_a were higher than values of χ_a. Figures 2D and 2E show the ratio between the antagonist and agonist affinities fitted to a second-order polynomial, which approaches a slope of 1.66 ± 0.11 (standard error) at the highest values of χ′_a. The slope indicates that the affinity of dopamine receptors toward an agonist approaches a value 1.66-fold that of the receptors toward an antagonist, assuming that the total concentration of available binding sites is the same for agonists and antagonist ligands. Another way of expressing the same result is to state that the C_g/K_g ratio (χ_g) declines to a ratio close to 0.66 for large values of χ′_a.

Figures 3A and 3B show the relation between the normalized concentration of the GTP-free G protein and the concentration and occupancy of the endogenous agonist assumed to be dopamine, showing the decline of the estimated GTP-free G-protein concentration at elevated concentrations (Fig. 3A) and occupancies (Fig. 3B) of the endogenous agonist competitor. Figure 3C yields estimates for χ_G and p“_B(0) listed in Table 3, which show that the inherent affinities of the receptors toward G protein and agonist are similar (4.0 vs. 5.3) and that the relative mass of available G protein is only half of the mass of receptors (2.1 vs. 4.0). The product of the occupancies of G protein and agonist defines the fraction of the total number of receptors in the ternary configuration. Figure 3D shows that this fraction reaches only 40% of the receptors at total agonist occupancy, indicating that no more than 40% of the receptors can be fully activated.

TABLE 3.

Baseline variables

Variable	Symbol	Mean ± SD
Raclopride binding potential	p _B(0)	8.5 ± 1.62
NPA binding potential	p′_B(0)	5.3 ± 0.77
G-protein binding potential	p“_B(0)	4.0 ± 0.26
Total GTP-free G protein concentration (normalized)	χ_G	2.1 ± 0.08

NPA, N-[³H]-prophylapomorphine; GTP, guanosine triphosphate.

FIG. 3.

Binding potentials of G protein versus binding potentials of endogenous agonist inhibitor. (A) Normalized concentration of G protein (χ_g) versus normalized concentration of endogenous agonist inhibitor (χ′_a) determined with agonist radioligand (NPA) according to Eq. 7 (modified from Fig. 2D). (B) Normalized concentration of G protein (χ_g) versus binding potential of endogenous agonist inhibitor (σ′_a) determined with agonist radioligand. (C) Binding potential of GTP-free G protein (σ_g) versus binding potential of endogenous agonist inhibitor (σ′_a) determined with agonist radioligand. Curve represents regression by Eq. 15, yielding estimates of total normalized concentration (χ_G) and binding potential (p_B) of G protein listed in Table 3. (D) Ternary complex fraction of total binding capacity (χ_g χ′_a) versus occupancy of endogenous agonist inhibitor (χ′_a), indicating maximum achievable receptor activation (40%).

DISCUSSION

Partial depletion of dopamine increases the availability of binding sites for [¹¹C]raclopride and other benzamide radioligands in the human brain. The extent of this increase is greater in patients with schizophrenia than in healthy volunteers (Abi-Dargham et al., 2000), a finding that was interpreted to reveal the presence of elevated basal occupancy in schizophrenia. However, the partial dopamine depletion increased the availability of D₂ antagonist binding sites by less than 20%, even in schizophrenic subjects. In contrast, the present study and earlier results of experimental animals (Ross and Jackson, 1989a,b; Ginovart et al., 1987) indicate that basal occupancy of antagonist bindings sites in the mouse is approximately 40%. If complete dopamine depletions are not obtained in the human brain, differences in p_B could be attributed to differential availability of the extracellular dopamine in addition to differences in occupancy.

The present results show that a radiolabeled dopamine agonist is highly sensitive to changes in extracellular dopamine concentrations, and may thus be more suited than antagonists for detecting pathophysiologic changes in occupancy of human brain receptors. Several radiolabeled agonist ligands have been tested as in vivo tracers for dopamine receptors (Closse et al., 1988; Nobrega and Seeman, 1994; Mukherjee et al., 2000). The present findings with [³H]NPA in living mouse suggest that ¹¹C-labeled NPA ([¹¹C]NPA; Hwang et al., 2000; Gillings and Cumming, 2000) is a good candidate for PET investigations of basal occupancy of dopamine receptors by endogenous dopamine in living human subjects.

The differences between the properties of the radioligands [¹¹C]raclopride and [³H]NPA are difficult to unravel because their effects are additive. However, the present study revealed two important differences. First, the maximally obtainable binding potential of the radioligands (i.e., the potential measured in the absence of competition from any competitor other than the tracer itself) is greater for raclopride than for NPA. This difference is presumed to reflect different volumes of distribution, receptor affinities, or maximum binding capacities of the two ligands. Second, the decline of the binding potential in the presumed presence of an agonist competitor is relatively greater for NPA than for raclopride. Therefore, the greater binding potential of one ligand in the baseline will be magnified during competition from an agonist by the greater responsiveness to competition of the other ligand. This difference is shown in Fig. 4A, in which the binding potentials are plotted against the normalized concentrations of the agonist, as calculated from the agonist-derived binding potentials in the present study.

FIG. 4.

Binding potentials (A) and ratio of binding potentials (B) of antagonist (raclopride [RAC]) and agonist (NPA) radioligand in competition with endogenous or exogenous agonist, fulfilling prediction of greater sensitivity of agonist radioligand to competition from agonist, quantified either as natural logarithm of normalized agonist concentration determined with agonist radioligand (abscissa, A), or as agonist occupancy determined with antagonist radioligand (abscissa, B).

Baseline ratio of binding potentials

In the striatum of living mouse, the apparent B_max and K_d of [³H]NPA (Ross and Jackson, 1989a) and [³H]raclopride (Ross and Jackson, 1989b) are similar. For both radioligands, dopamine depletion with reserpine decreased and psychostimulants increased the magnitude of the apparent affinity without altering B_max. The p_B of [¹¹C]raclopride in the striatum of reserpine-treated or saline-treated mice was close to the ratio of radioactivity in the striatum to that in cerebellum (< 1) 30 minutes after [³H]raclopride injection (Ross and Jackson, 1989b). Likewise, the estimates of p_B for [³H]NPA in the mouse striatum were consistent with earlier reports of [³H]NPA binding in the rodent striatum during 1-hour tracer circulation (Hall et al., 1983; van der Werf et al., 1983; Benedetti et al., 1990).

The somewhat lower p_B reported by Zijlstra et al. (1992) for (N-[¹¹C]methyl)-norapomorphine (p_B = 1.5) may have been due to lower affinity of that tracer for dopamine D_2,3-receptors. The differential sensitivity of NPA and raclopride to competition from dopamine is not evident from the earlier analyses, which were performed in separate groups of mice. In the present study, we used the dynamic time–radioactivity curves measured in the brain during 1-hour dual-tracer circulation to calculate the values of p_B in the striatum at equilibrium.

The apomorphines, like raclopride, do not distinguish between D₂ and D₃ receptors (Levant, 1997). In the present study, the p_B(0) of [¹¹C]raclopride exceeded that of [³H]NPA after reserpine treatment and in other conditions (Fig. 2A). A single class of binding sites for [¹¹C]raclopride and [³H]NPA and no endogenous competition would yield the following ratio of the baseline binding potentials at equilibrium:

\frac{p_{B (0)}}{p_{B (0)}^{'}} = \frac{K_{d}^{'} V_{d}^{'} B_{max}}{K_{d} V_{d} B_{max}^{'}}

(16)

where V_d refers to the volumes of distribution of dissolved and non-specifically bound radioligand in brain tissue, and the primes refer to the agonist. In contrast to the present findings, the 10-fold–higher in vitro affinity of [³H]NPA (300 pmol/L) (Camus et al., 1986) than of [³H]raclopride (3 nmol/L) (Lahti et al., 1995) predicts a higher p_B(0) for the agonist. This discrepancy might be attributed to differences in B_max (Battaglia and Titeler, 1982; van der Werf et al., 1986), though the additional binding of [³H]spiperone in some studies could be attributed to dopamine D₄ or serotonin 5HT_2A receptors. Alternatively, the tissue distribution volumes of the two tracers may favor the binding of [¹¹C]raclopride in vivo.

The 6-hydroxybenzoquinoline compound [³H]quinagolide is an agonist that binds to dopamine D₂ receptors in rat brain sections with an apparent affinity of 0.6 nmol/L (Charuchinda et al., 1987). In the present study, the low dose of quinagolide displaced [³H]NPA slightly more effectively than [¹¹C]raclopride, but the two ligands were not discriminated by the high quinagolide dose. Likewise, it was noted previously that nanomolar concentrations of NPA completely inhibited dopamine D₂ antagonist binding in the rodent brain (Andersen and Nielsen, 1986). Partial agonism of dopamine D₂ sites may underlie the neuroleptic property of NPA (Campbell et al., 1991; Tarazi et al., 1997). Like the apomorphine derivatives, quinagolide may also be a partial agonist, accounting for the ability of a low dose of quinagolide to differentiate only partially between the two radioligands.

Interaction with G protein

The findings are consistent with the hypothesis that dopamine or exogenous agonist competitors must appear to occupy a greater fraction of dopamine receptors when the occupancy is measured with an agonist radioligand than when it measured with an antagonist radioligand. We based the hypothesis on the claim that an agonist radioligand undergoes the same GTP-shift of affinity exhibited by dopamine and exogenous agonist competitors, and hence exhibits binding that reflects accurately the ratio of bound-to-free agonist. The antagonist radioligand undergoes no GTP-shift of affinity, and hence underestimates the bound-to-free ratio at all concentrations of dopamine or an exogenous agonist competitor. The ratio between the binding potentials determined with an antagonist and an agonist radioligand, as defined previously, has the simple solution

\frac{p_{B (I)}}{p_{B (I)}^{'}} = (1 + σ_{a} χ_{g}) [\frac{p_{B (0)}}{p_{B (0)}^{'}}]

(17)

where σ_a is the agonist occupancy measured in the traditional way with an antagonist radioligand. The relation approaches a straight line, the slope of which depends on the concentration of unbound GTP-free G protein. This relation is shown in Fig. 4B.

The long-term activation of G-protein–linked receptors can result in the phenomena of homologous and heterologous sensitization in the brain (Morelli and Di Chiara, 1987; Watts and Neve, 1996; Zhang et al., 2000). The several sensitization phenomena can be attributed to diverse mechanisms, including receptor recruitment (Holtback et al., 1999) and enhanced coupling of receptors to signal transduction pathways (Tenn and Niles, 1997). In addition, altered expression of G proteins has been specifically implicated in the sensitization of dopamine receptors by chronic reserpine treatment (Butkerait and Friedman, 1993) or morphine (Van Vliet et al., 1993). The present finding that G-protein availability limits the binding and, hence, the efficacy of dopamine D₂ ₃ agonists in the mouse striatum suggests that altering the expression of G_i/o in the striatum should modulate the efficacy of dopamine receptor agonists.

Footnotes

Acknowledgments:

The authors thank Bisma El-Humadi and Paige Rauseo for expert technical assistance, and acknowledge the generous gift of quinagolide from Novartis.

References

Abi-Dargham

Rodenhiser

Printz

Zea-Ponce

Gil

Kegeles

L.S.

Weiss

Cooper

T.B.

Mann

J.J.

van Heertum

R.L.

Gorman

J.M.

Laruelle

(2000) From the cover: Increased basal occupancy of D2 receptors by dopamine in schizophrenia. Proc Natl Acad Sci U S A 97:8104–8109.

Andersen

P.H.

Nielsen

E.B.

(1986) The benzazepine, SCH 23390 inhibits 3H-NPA binding in mouse brain in vivo. Acta Pharmacol Toxicol 59:315–318.

Battaglia

Titeler

(1982) [3H]N-Propylapomorphine and [3H]spiperone binding in brain indicate two states of the D2-dopamine receptor. Eur J Pharmacol 81:493–498.

Benedetti

M.S.

Dostert

Barone

Efthymiopoulos

Peretti

Roncucci

(1990) In vivo interaction of cabergoline with rat brain dopamine receptors labeled with [3H]N-n-propylnorapomorphine. Eur J Pharmacol 197:399–408.

Burstein

E.S.

Spalding

T.A.

Brann

M.R.

(1997) Pharmacology of muscarinic receptor subtypes constitutively activated by G proteins. Mol Pharmacol 51:312–319.

Butkerait

Friedman

(1993) Repeated reserpine increases striatal dopamine receptor and guanine nucleotide binding protein RNA. J Neurochem 60:566–571.

Cambell

Yeghiayan

S.A.

Baldessarini

R.J.

Neumeyer

J.L.

(1991) Selective antidopaminergic effects of S(+)N-m-propylnorapomorphines in limbic versus extrapyramidal sites in rat brain: Comparisons with typical and atypical antipsychotic agents. Psychopharmacology 103:323–329.

Camus

Javoy-Agid

Dubois

Scatton

(1986) Autoradiographic localization and quantification of dopamine D2 receptors in normal human brain with [3H]N,n-propylnorapomorphine. Brain Res 375:135–149.

Carson

R.E.

Breier

de Bartolomeis

Saunders

R.C.

T.P.

Schmall

Der

M.G.

Pickar

Eckelman

W.C.

(1997) Quantification of amphetamine-induced changes in [11C]raclopride binding with continuous infusion. J Cereb Blood Flow Metab. 17:437–447.

10.

Charuchinda

Supavilai

Karobath

Palacios

J.M.

(1987) Dopamine D2 receptors in the rat brain: Autoradiographic visualization using a high affinity selective agonist ligand. J Neurosci 7:1352–1360.

11.

Closse

Camps

Wanner

Palacios

J.M.

(1988) In vivo labeling of brain dopamine D2 receptors using the high-affinity specific D2 agonist [3H]CV 205–502. Brain Res 440:123–132.

12.

Cresse

Sibley

D.R.

(1979) Radioligand binding studies: Evidence for multiple dopamine receptors. Commun Psychopharmacol 3:385.

13.

del Castillo

Katz

(1957) Interaction at the endplate receptors between different choline derivatives. Proc R Soc Lond B Biol Sci 146:369–381.

14.

De Lean

Stadel

J.M.

Lefkowitz

R.J.

(1980) A ternary complex model explains the agonist-specific binding properties of the adenylate cyclase-coupled beta-adrenergic receptor. J Biol Chem 255:7108–7117.

15.

Dewey

S.L.

Smith

G.S.

Logan

Brodie

J.D.

Fowler

J.S.

Wolf

A.P.

(1993) Striatal binding of the PET ligand 11C-raclopride is altered by drugs that modify synaptic dopamine levels. Synapse 13:350–356.

16.

Ehrin

Gawell

Hogberg

de Paulis

Strom

(1987) Synthesis of [methoxy-3H]- and [methoxy-11C]-labeled raclopride, a specific dopamine-D2 receptor ligand. J Labeled Comp Radiopharm 24:931–939.

17.

Gillings

Cumming

(2000) Synthesis of (R)-[1–11C]propylnorapomorphine. Paper presented at the 7th International Symposium on the Synthesis of Isotopes and Isotopically Labeled Compounds. Dresden, June 18–22.

18.

Ginovart

Farde

Halldin

Swahn

C.G.

(1997) Effect of reserpine-induced depletion of synaptic dopamine on [11C]raclopride binding to D2-dopamine receptors in the monkey brain. Synapse 25:321–325.

19.

Gjedde

Wong

D.F.

(1990) Modeling neuroreceptor binding of radioligands in vivo. Quantitative imaging of neuroreceptors ( Frost

Wagner

H.N.

Jr , eds), New York, NY: Raven Press, pp 51–79.

20.

Gjedde

Wong

D.F.

(2001) Quantification of neuroreceptors in living human brain. Part V: Endogenous neurotransmitter inhibition of haloperidol binding in psychosis. J Cereb Blood Flow Metab 21:982–992.

21.

Gjedde

Wong

D.F.

Wagner

H.N.

Jr (1986) Transient analysis of irreversible and reversible tracer binding in human brain in vivo. In: PET and NMR: new perspectives in neuroimaging and clinical neurochemistry ( Battistin

, ed), New York, NY: Alan R Liss, pp 223–235.

22.

Hall

M.D.

Jenner

Marsden

C.D.

(1983) Differential labeling of dopamine receptors in rat brain in vivo: Comparison of [3H]piribedil, [3H]S 3608, and [3H]N,n-propylnorapomorphine. Eur J Pharmacol 87:85–94.

23.

Holtback

Brismar

DiBona

G.F.

Greengard

Aperia

(1999) Receptor recruitment: A mechanism for interactions between G protein-coupled receptors. Proc Natl Acad Sci U S A 96:7271–7275.

24.

Hwang

D.R.

Kegeles

L.S.

Laruelle

(2000) (−)-N-[(11)C]propylnorapomorphine: A positron-labeled dopamine agonist for PET imaging of D(2) receptors. Nucl Med Biol 27:533–539.

25.

Inoue

Kobayashi

Tsukada

Itoh

Langstrom

(1991) Difference in in vivo receptor binding between [3H]N-methylspiperone and [3H]raclopride in reserpine-treated mouse brain. J Neural Transm 85:1–10.

26.

Jiang

Spicher

Boulay

Wang

Birnbaumer

(2001) Most central nervous system D2 dopamine receptors are coupled to their effectors by Go. Proc Natl Acad Sci U S A 98:3577–3582.

27.

Lahti

R.A.

Roberts

R.C.

Tamminga

C.A.

(1995) D2-family receptor distribution in human post mortem tissue: An autoradiographic study. NeuroReport 6:2505–2512.

28.

Lammertsma

A.A.

Bench

C.J.

Hume

S.P.

Osman

Gunn

Brooks

D.J.

Frackowiak

R.S.J.

(1996) Comparison of methods for analysis of clinical [11C]raclopride studies. J Cereb Blood Flow Metab 16:42–52.

29.

Laruelle

Iyer

R.N.

al-Tikriti

M.S.

Zea-Ponce

Malinson

Zoghbi

S.S.

Baldwin

R.M.

(1997) Microdialysis and SPECT measurement of amphetamine-induced dopamine release in non-human primates. Synapse 25:1–14.

30.

Levant

(1997) The dopamine D3 receptor: Neurobiology and potential clinical relevance. Pharmacol Rev 49:221–252.

31.

Michaelis

Menten

M.L.

(1913) Die kinetik der invertinwirkung. Biochem Z 49:333–369.

32.

Mukherjee

Narayanan

T.K.

Christian

B.T.

Shi

Dunigan

K.A.

Mantil

(2000) In vitro and in vivo evaluation of the binding of the dopamine D2 receptor agonist (11)C-(R,S)-5-hydroxy-2-(di-n-proplyamino)tetralin in rodents and nonhuman primate. Synapse 37:64–70.

33.

Morelli

Di Chiara

(1987) Agonist-induced homologous and heterologous sensitization to D-1- and D-2-dependent contraversive turning. Eur J Pharmacol 141:101–107.

34.

Nobrega

J.N.

Seeman

(1994) Dopamine D2 receptors mapped in rat brain with [3H](+)PHNO. Synapse 17:167–172.

35.

Ross

S.B.

Jackson

D.M.

(1989a) Kinetic properties of the in vivo accumulation of 3H-(–)-N-n-propylapomorphine in mouse brain. Naunyn Schmiedebergs Arch Pharmacol 340:13–20.

36.

Ross

S.B.

Jackson

D.M.

(1989b) Kinetic properties of the accumulation of 3H-raclopride in the mouse brain in vivo. Naunyn Schmiedebergs Arch Pharmacol 340:6–12.

37.

Samama

Cotecchia

Costa

Lefkowitz

R.J.

(1993) A mutation-induced activated state of the beta 2-adrenergic receptor. Extending the ternary complex model. J Biol Chem 268:4625–4636.

38.

Schlaepfer

T.E.

Pearlson

G.D.

Wong

D.F.

Marenco

Dannals

R.F.

(1997) PET study of competition between intravenous cocaine and [11C]raclopride at dopamine receptors in human brain. Am J Psychiatry 154:1209–1213.

39.

Tarazi

F.I.

Yeghiayan

S.K.

Baldessarini

R.J.

Kula

N.S.

Neumaeyer

J.D.

(1997) Long-term effects of S(+)N-n-propylnorapomorphine compared with typical and atypical antipsychotics: Differential increases of cerebrocortical D2-like and striatolimbic D4-like dopamine receptors. Psychopharmacology 17:186–196.

40.

Tenn

C.C.

Niles

L.P.

(1997) Sensitization of G protein-coupled benzodiazepine receptors in the striatum of 6-hydroxydopamine-lesioned rats. J Neurochem 69:1920–1926.

41.

van der Werf

J.F.

Sebens

J.B.

Vaalburg

Korf

(1983) In vivo binding of N,n-propylnorapomorphine in the rat brain: Regional localization, quantification in striatum and lack of correlation with dopamine metabolism. Eur J Pharmacol 87:259–270.

42.

van der Werf

J.F.

Sebens

J.B.

Korf

(1986) Tracer and maximal specific binding of tritiated spiperone and N-n-propylnorapomorphine to quantify dopamine receptors in rat brain regions in vivo. Life Sci 39:155–160.

43.

Van Vliet

B.J.

Van Rijswijk

A.L.

Wardeh

Mulder

A.H.

Schoffelmeer

A.N.

(1993) Adaptive changes in the number of Gs- and Gi-proteins underlie adenylyl cyclase sensitization in morphine-treated rat striatal neurons. Eur J Pharmacol 245:23–29.

44.

Watts

V.J.

Neve

K.A.

(1996) Sensitization of endogenous and recombinant adenylate cyclase by activation of D2 dopamine receptors. Mol Pharmacol 50:966–976.

45.

Wong

D.F.

Gjedde

Wagner

H.N.

Jr (1986) Quantification of neuroreceptors in the living human brain. Part I: Irreversible binding of ligands. J Cereb Blood Flow Metab 6:137–146.

46.

Young

L.T.

Wong

D.F.

Goldman

Minkin

Chen

Matsumura

Scheffel

Wagner

H.N.

Jr (1991) Effects of endogenous dopamine on kinetics of [3H]N-methylspiperone and [3H]raclopride binding in the rat brain. Synapse 9:188–194.

47.

Zhang

Tarazi

F.I.

Campbell

Baldessarini

R.J.

(2000) GABA(B) receptors: Altered coupling to G proteins in rats sensitized to amphetamine. Neuroscience 101:5–10.

48.

Zijlstra

van der Worp

Wiegman

Visser

G.M.

Korf

Vaalburg

(1993) Synthesis and in vivo distribution in the rat of a dopamine agonist: N-([11C]methyl)norapomorphine. Nucl Med Biol 20:7–12.

Specific Binding of [ 11 C]Raclopride and N -[ 3 H]Propyl-Norapomorphine to Dopamine Receptors in Living Mouse Striatum: Occupancy by Endogenous Dopamine and Guanosine Triphosphate–Free G Protein

Abstract

Keywords

THEORY

Definitions

Agonist occupancy

Guanosine triphospate–shift of affinity

MATERIALS AND METHODS

RESULTS

DISCUSSION

Baseline ratio of binding potentials

Interaction with G protein

Footnotes

Acknowledgments:

References