Neuron, Vol. 38, 291–303, April 24, 2003, Copyright 2003 by Cell Press
Dopaminergic Supersensitivity in G Protein-Coupled Receptor Kinase 6-Deficient Mice Raul R. Gainetdinov,1,2 Laura M. Bohn,1,2 Tatyana D. Sotnikova,1,2 Michel Cyr,1,2 Aki Laakso,1,2 Alexander D. Macrae,1,3,5 Gonzalo E. Torres,1,2 Kyeong-Man Kim,1,2 Robert J. Lefkowitz,1,3,4,* Marc G. Caron,1,2,3,* and Richard T. Premont3 1 Howard Hughes Medical Institute Laboratories 2 Department of Cell Biology 3 Department of Medicine 4 Department of Biochemistry Duke University Medical Center Durham, North Carolina 27710
Summary Brain dopaminergic transmission is a critical component in numerous vital functions, and its dysfunction is involved in several disorders, including addiction and Parkinson’s disease. Responses to dopamine are mediated via G protein-coupled dopamine receptors (D1-D5). Desensitization of G protein-coupled receptors is mediated via phosphorylation by members of the family of G protein-coupled receptor kinases (GRK1-GRK7). Here we show that GRK6-deficient mice are supersensitive to the locomotor-stimulating effect of psychostimulants, including cocaine and amphetamine. In addition, these mice demonstrate an enhanced coupling of striatal D2-like dopamine receptors to G proteins and augmented locomotor response to direct dopamine agonists both in intact and in dopamine-depleted animals. The present study indicates that postsynaptic D2-like dopamine receptors are physiological targets for GRK6 and suggests that this regulatory mechanism contributes to central dopaminergic supersensitivity. Introduction G protein-coupled receptors (GPCRs) are regulated via activation-dependent phosphorylation by a family of G protein-coupled receptor kinases (GRKs) (Premont et al., 1995; Pitcher et al., 1998; Penn et al., 2000). This process, which modulates the coupling state of GPCRs to their G proteins, is critically involved in GPCR desensitization (Pitcher et al., 1998; Penn et al., 2000). GRKphosphorylated receptors bind to an arrestin protein, which prevents the receptor from activating more G proteins despite the continued binding of agonist (Pitcher et al., 1998). Seven distinct GRK genes are known, named GRK1 through GRK7, that are classified into three distinct groups (Pitcher et al., 1998; Premont et al., 1999; Penn et al., 2000). GRK6 is a member of the GRK4 subfamily of GRKs, which also contains GRK4 *Correspondence:
[email protected] (M.G.C.), lefko001@ receptor-biol.duke.edu (R.J.L.) 5 Current address: Clinical Research and Development, GlaxoSmithKline, Harlow, Essex, United Kingdom.
and GRK5 (Benovic and Gomez, 1993; Premont et al., 1999). Multiple GRK enzymes (GRK2, GRK3, GRK5, and GRK6) are found throughout the brain (Gainetdinov et al., 2000; Erdtmann-Vourliotis et al., 2001), but the relative physiological importance of each GRK to the regulation of any given neurotransmitter receptor is unclear. The catecholamine dopamine plays a critical role in several brain functions, including movement control, emotion, and affect (Carlsson, 2001; Greengard, 2001). Physiological responses to dopamine are controlled by a family of five G protein-coupled dopamine receptors (D1-D5), which are classified into D1-like (D1, D5) and D2-like (D2, D3, D4) subfamilies based on functional and pharmacological properties (Grandy and Civelli, 1992; Schwartz et al., 1993; Seeman and Van Tol, 1994; Picetti et al., 1997; Missale et al., 1998; Sibley, 1999). Sensitivity of dopamine receptors to endogenous and exogenous ligands is known to be an important modulator of dopamine-related functions in physiology and pathology. Supersensitivity of dopamine signaling has been postulated in several brain disorders (Pandey et al., 1977; Jenner and Marsden, 1987; Kirkpatrick et al., 1992; Singer, 1994; Willner, 1997), including addiction (Nestler and Aghajanian, 1997; Nestler, 2001; Hyman and Malenka, 2001; Laakso et al., 2002). Particularly, it is believed that chronic psychostimulant-induced sensitization, an early biochemical and behavioral manifestation of cellular plasticity leading to addiction, is associated with long-term changes in dopamine receptor responsiveness (Nestler, 2001; Hyman and Malenka, 2001; Laakso et al., 2002). Previous in vitro studies have shown that the D1, D2, and D3 dopamine receptors can be phosphorylated by some GRKs (Tiberi et al., 1996; Ito et al., 1999; Kim et al., 2001), but the role of GRK6 in dopamine receptor regulation has never been investigated. Furthermore, no data on physiological significance of receptor regulation or the in vivo specificity of dopamine receptor/GRK interaction are currently available. To understand the role of GPCR desensitization mechanisms in the regulation of dopamine function, we have begun to examine mice bearing inactivated GRK genes for alterations in cocaine responses and dopamine receptor properties. In a previous investigation, we found no evidence for dopaminergic supersensitivity in mice lacking GRK5 (Gainetdinov et al., 1999). Here we report that mice lacking GRK6 are supersensitive to locomotor stimulating effects of several psychostimulants, display enhanced coupling of D2-like receptors to their G proteins, and show enhanced behavioral effects of direct dopamine agonists. Results Generation of GRK6-KO Mice To address the functions of GRK6 in vivo, the GRK6 gene was targeted by homologous recombination in ES cells, as shown in Figures 1A–1C. The targeting construct introduces into the mouse GRK6 gene two loxP sites, flanking exons 3 through 9, as well as a selection
Neuron 292
Figure 1. Targeted Inactivation of the Mouse GRK6 Gene (A) Schematic diagram of the wild-type GRK6 locus, the GRK6/lox targeting vector, the integrated targeting construct, and the Cre recombinasedeleted GRK6 locus (GRK6-KO). GRK6 exons are shown as open boxes and numbered from the first coding exon (Premont et al., 1999). LoxP sites are shown as filled triangles, and the location of the Southern blot probe as a hatched box. Relevant NheI restriction sites are indicated. (B) Genotyping of targeted GRK6-KO mice. The wild-type and GRK6-KO loci were distinguished by triplex PCR amplification as described in Experimental Procedures. The wt GRK6 locus gives a 460 bp band while the GRK6-KO locus gives a 610 bp band, as indicated. (C) GRK6 protein expression by Western blotting. Membrane proteins from brainstem and striatum (data not shown) of wild-type and GRK6KO animals were subjected to immunoblotting using an anti-GRK6 antiserum (Stoffel et al., 1994). GRK6-KO homozygote animals exhibit a loss of the 68 kDa immunoreactive band compared to wild-type animals (arrow). The 69 kDa band is a nonspecific interaction, since it is present in GRK5 and GRK4 knockout animals and is not recognized by another GRK6 antiserum (data not shown). (D) GRK6 is present in striatal neurons expressing DARPP-32. Top left: Immunofluorescence analysis reveals GRK6 immunoreactivity in the striatal neurons of wt mouse (⫹0.74 from bregma). Top right: Lack of GRK6 immunoreactivity in the striatal neurons of GRK6-KO mouse. Bottom left: DARPP-32 immunoreactivity in the striatal neurons of wt mouse. Bottom right: GRK6 and DARPP-32 are colocalized in the same neuronal population in the striatum of wt mouse. GRK6 immunoreactivity was detected using a commercially available anti-GRK6 antibody (rabbit, 1:50, sc-566, Santa Cruz Biotechnologies, Inc.). Similar observations (data not shown) were made using another anti-GRK6 antiserum (Stoffel et al., 1994). Note that GRK6 is also expressed in large cholinergic striatal cells, which do not express DARPP-32, but can be labeled with anti-choline acetyltransferase antibody (data not shown). Scale bar is equal to 50 m.
marker cassette also flanked by loxP sites (Figure 1A). The floxed selection marker genes were removed in vivo by breeding a heterozygote mouse bearing the targeted allele with a mouse carrying a CMV-Cre transgene. A pup bearing a deletion of both cassettes, leaving an inactivated GRK6 gene and no selection marker genes, was identified and used to generate the colony of GRK6 knockout (GRK6-KO) mice used here. Deletion of exons 3 through 9 leads to a GRK6 that lacks most of the amino-terminal RGS-like domain as well as half of the conserved catalytic domain elements (Figure 1A; Hanks et al., 1988). Beyond the deletion, exon 10 is out of frame relative to exon 2, so that there is little chance of any active protein fragments being produced. Animal genotyping was performed using triplex PCR (Figure 1B) and was confirmed by Southern blotting (data not shown) and by analysis of GRK6 protein levels in membranes prepared from brainstem (Figure 1C) and striatum (data not shown). Immunohistochemical Localization of GRK6 in Striatal Cells Previous histological examinations have shown that GRK6 mRNA is expressed in many brain regions, including primary dopaminergic areas, such as substantia ni-
gra as well as dorsal and ventral striatum (ErdtmannVourliotis et al., 2001). The expression level of GRK6 mRNA in the striatum was found to be higher than that of other GRKs (GRK2, GRK3, and GRK5), suggesting that GRK6 might be a predominant receptor kinase in this brain area (Erdtmann-Vourliotis et al., 2001). A detailed investigation of the expression pattern of GRK6 protein using immunohistochemistry revealed expression of this kinase in the majority of cells in both dorsal and ventral striatum (Figure 1D). Particularly, GRK6 protein was found in the same neuronal population that expresses DARPP-32 (dopamine- and cyclic AMP-regulated phosphoprotein, apparent molecular weight of 32,000 Da), a molecule involved in dopaminergic signaling mediated by both D1-like and D2-like dopamine receptors, and a phenotypic marker of the medium-size spiny GABA neurons of the mammalian striatum (Figure 1D; Greengard, 2001). These neurons represent a major striatal cell group that receive dopaminergic input, express varying levels of both D1-like and D2-like dopamine receptors (Le Moine et al., 1991; Aizman et al., 2000; Aubert et al., 2000), and are believed to be critical to cellular mechanisms of addiction (Nestler, 2001; Hyman and Malenka, 2001; Laakso et al., 2002). In addition, dense expression of GRK6 protein was detected in a
Role of GRKs in the Responsiveness to Cocaine 293
Figure 2. Cocaine Supersensitivity in GRK6 Mutant Mice (A) Locomotor response of GRK6 mutant (wt, n ⫽ 24; GRK6 heterozygous, n ⫽ 21; GRK6-KO, n ⫽ 15) mice to cocaine (20 mg/kg, i.p.) administration. GRK6 heterozygous and GRK6-KO mice are significantly different from wt controls in responses to cocaine. p ⬍ 0.001, twoway analysis of variance (ANOVA). Cocaine-induced locomotion at the time of the peak effect (10 min after injection) was 184% and 185% in GRK6 heterozygous and GRK6-KO mice versus wt controls, respectively. (B) Dose-response curve of the effect of cocaine (10–30 mg/kg, i.p.) on horizontal activity of GRK6-KO, heterozygous, and wt mice (n ⫽ 8–24 per group). Both GRK6 heterozygous and GRK6-KO mice are significantly different from wt controls in responses to cocaine. p ⬍ 0.001, twoway ANOVA. Symbols are the same as in (A). (C) Cocaine sensitization in GRK6-KO mice. Mice (wt, n ⫽ 16; GRK6-KO, n ⫽ 14) were injected daily with cocaine (20 mg/kg, i.p.) for 5 days, and 48 hr after the last injection animals were challenged with the same dose of the drug. Locomotor activity measurements were performed on days 1 (top left) and 7 (top right). Two-way ANOVA revealed a significant difference (p ⬍ 0.001) between responses of wt mice in Day 7 versus Day 1, but no such difference was observed in GRK6-KO mice. In addition, responses in sensitized wt mice (Day 7) were not different from that of GRK6-KO mice either in Day 1 or Day 7. The accumulated distance traveled by mice in the 90 min period after cocaine administration on days 1 and 7 are shown in the lower panel. **p ⬍ 0.01; ***p ⬍ 0.001 versus wt littermates for the first day group, Student’s t test. Analysis of accumulated distances over 15 min, 30 min, or 60 min after cocaine administration reveals a significant difference (p ⬍ 0.001) between wt and GRK6-KO mice in Day 1 at any period analyzed, but no such differences were observed between sensitized wt and GRK6KO mice in Days 1 or 7. In sensitized GRK6-KO mice (Day 7), locomotor responses to cocaine were not enhanced versus that in Day 1 when 30 min, 60 min, or 90 min periods after injection were analyzed. However, analysis of first 15 min period after cocaine revealed a moderate increase in total distance traveled by GRK6-KO mice in Day 7 versus Day 1 (GRK6-KO, Day 1, 3786 ⫾ 459 cm/15 min; Day 7, 5386 ⫾ 571 cm/15 min, p ⬍ 0.05, Student’s t test; for comparison, distance traveled by wt mice, Day 1, 1686 ⫾ 252 cm/15 min; Day 7, 4077 ⫾ 443 cm/ 15 min, p ⬍ 0.001, Student’s t test).
population of large-sized aspiny cholinergic interneurons, which represent another major group of dopaminoceptive striatal cells expressing predominantly D2 dopamine receptors (Le Moine et al., 1991; Di Chiara et al., 1994; Aubert et al., 2000). Thus, the prominent expression of GRK6 in the major cell groups known to express both D1-like and D2-like receptors suggests that this kinase could be involved in the regulation of dopamine receptor signaling. Behavioral Supersensitivity to Psychostimulants in GRK6 Mutant Mice Heterozygote and homozygote GRK6-KO mice are viable and present no gross anatomical or behavioral abnormalities, although GRK6-KO mice demonstrate re-
duced lymphocyte chemotaxis (Fong et al., 2002). In locomotor activity tests, unchallenged knockout mice were not different from wild-type (wt) littermates either in horizontal (Figures 2A and 2B) or vertical activities, or stereotypy (data not shown). However, acute administration of cocaine (20 mg/kg, i.p.) resulted in a markedly enhanced locomotor response in GRK6 mutant mice (Figures 2A and 2B). In this paradigm, the GRK6-KO mice exhibited a more pronounced and longer lasting locomotor activation, as measured by horizontal (Figures 2A and 2B) activity in response to cocaine (10–30 mg/kg, i.p.) than did wild-type littermate mice. Similarly, analysis of vertical activity and stereotypy score showed the same dose-dependent enhancement of responses to cocaine in GRK6 mutant mice (data not shown). Inter-
Neuron 294
estingly, the mice heterozygous for GRK6 deletion were as responsive to cocaine as GRK6 “null” mice (Figures 2A and 2B), suggesting that even minor changes in GRK6 levels or activity may result in significant behavioral alterations. The degree of activation induced by cocaine (20 mg/kg, i.p.) in both heterozygous and knockout mice was substantially higher, not only relative to littermate wild-type mice, but also relative to mice of both parental strains (C57BL/6J and 129/SvJ) used to generate the mutants (data not shown). Repeated administration of cocaine is known to result in a progressive enhancement of psychomotor responses. This phenomenon, termed “behavioral sensitization” or “reverse tolerance,” is believed to relate to neuronal adaptations associated with drug addiction (Robinson and Berridge, 1993; Nestler and Aghajanian, 1997; Hyman and Malenka, 2001). In experimental animals, it is often modeled by analyzing locomotor responses to repeated intermittent treatments with the same dose of cocaine (Wise and Rompre, 1989; Robinson and Berridge, 1993; Nestler and Aghajanian, 1997; Wang et al., 1997; Xu et al., 2000; Nestler, 2001; Hyman and Malenka, 2001). To test if the development of sensitization is affected by the lack of GRK6, such a cocainesensitization paradigm was employed in wild-type and GRK6-KO mice (Wang et al., 1997; Xu et al., 2000). Mice received daily injections of cocaine for 5 consecutive days and were tested for their responses to this drug on the seventh day. Compared to the first day of treatment, wild-type animals exhibited enhanced locomotor responses (ⵑ2-fold) to cocaine on day 7, indicative of sensitization (Figure 2C). By contrast, GRK6-KO animals were as responsive to cocaine on the first day as were wild-type mice following the sensitization protocol (Figure 2C). As might be expected from their initial exaggerated response to cocaine, GRK6-KO mice showed little further sensitization of the locomotor response following this sensitization regimen. In fact, analysis of total distance traveled for 90 min did not reveal significant differences between day 1 and day 7 (Figure 2C). Nonetheless, in the first 15 min after cocaine administration, sensitized GRK6-KO mice did exhibit a slightly enhanced response (Figure 2C, legend), suggesting that a further sensitization to cocaine can still be developed in these mice. It is well established that the locomotor stimulating action of cocaine is mediated by the blockade of the dopamine transporter (DAT), resulting in elevation of extracellular dopamine in the striatum and related brain areas (Wise and Rompre 1989; Robinson and Berridge, 1993; Gainetdinov et al., 2002). Another psychostimulant known to markedly enhance central dopaminergic transmission via complex interaction with the DAT is amphetamine (Jones et al., 1998; Gainetdinov et al., 2002). Similarly to cocaine, amphetamine-induced locomotor activation was significantly enhanced in both GRK6 heterozygous and “null” mice (Figure 3A). Furthermore, enhanced locomotor responses were observed in both heterozygous and homozygous GRK6 mutant mice when the endogenous “trace amine” -phenylethylamine (Janssen et al., 1999; Borowsky et al., 2001) was administered (Figure 3B). While the functions and mechanism of the stimulant action of -phenylethylamine have not been fully determined (Janssen et al., 1999; Borowsky et al., 2001; Premont et al., 2001),
Figure 3. Enhanced Locomotor Effects of d-Amphetamine and -Phenylethylamine in GRK6 Mutant Mice (A) Time course of horizontal locomotor response of wt (n ⫽ 10) and GRK6 mutant (GRK6 heterozygous, n ⫽ 15; GRK6-KO, n ⫽ 9) in response to d-amphetamine (3 mg/kg, i.p.). GRK6 heterozygous and GRK6-KO mice are significantly different from wt controls in responses to d-amphetamine. p ⬍ 0.001, two-way ANOVA. (B) Time course of horizontal locomotor response of wt (n ⫽ 6) and GRK6 mutant (GRK6 heterozygous, n ⫽ 11; GRK6-KO, n ⫽ 6) mice in response to -phenylethylamine (50 mg/kg, i.p.). GRK6 heterozygous and GRK6-KO mice are significantly different from wt controls in responses to -phenylethylamine. p ⬍ 0.001, two-way ANOVA.
it is believed that -phenylethylamine primarily acts as an “endogenous amphetamine” via DAT-mediated efflux of dopamine from intraneuronal stores to extracellular spaces (Janssen et al., 1999). Accordingly, in vivo microdialysis experiments revealed that -phenylethylamine (50 mg/kg, i.p.) induced potent but transient elevation in striatal extracellular dopamine to the same degree (6-fold) in both GRK6-KO and wild-type mice (data not shown). Thus, an enhanced locomotor response of GRK6 mutants to -phenylethylamine, as well as cocaine and amphetamine, is consistent with an enhanced responsiveness to dopaminergic activation. Neurochemical Characterization of Dopamine Function in GRK6-KO Mice Behavioral supersensitivity to psychostimulants could be explained either by alterations in presynaptic dopaminergic function in these mice leading to augmented
Role of GRKs in the Responsiveness to Cocaine 295
drug-evoked extracellular dopamine levels or by altered postsynaptic receptor responsiveness to the same level of dopamine. To test the status of striatal presynaptic dopaminergic transmission in mutant mice, we assessed the function of the dopamine system using a set of neurochemical approaches (Figure 4; Wang et al., 1997; Xu et al., 2000). GRK6-KO mice were not different from wild-type controls in any of the neurochemical parameters examined. In particular, dopamine turnover and storage (tissue dopamine and dopamine metabolite content), dopamine re-uptake by DAT (synaptosomal dopamine uptake rates), and dopamine release (basal and cocaine-stimulated extracellular dopamine levels as assessed by in vivo microdialysis) were not affected in mutant mice (Figures 4A–4D). Thus, pronounced locomotor supersensitivity to psychostimulants in the GRK6 mutant mice occurs without measurable alterations in presynaptic dopamine function.
Figure 4. Analyses of Presynaptic Dopamine Function in Wild-Type and GRK6-KO Mice (A) Striatal tissue levels of dopamine, 5-HT, and their metabolites in GRK6-KO and wt littermate mice measured by HPLC-EC (wt, n ⫽ 5; GRK6-KO, n ⫽ 7). (B) [3H]dopamine uptake in striatal synaptosomes from GRK6-KO and wt mice (wt, n ⫽ 4; GRK6-KO, n ⫽ 4). (C) Extracellular dopamine levels in the striatum of freely moving mice measured using quantitative low perfusion rate microdialysis (wt, n ⫽ 6; GRK6-KO, n ⫽ 9). (D) Effect of saline and cocaine (20 mg/kg, i.p.) on extracellular
Dopamine Receptor Supersensitivity in GRK6-KO Mice Behavioral supersensitivity to psychostimulants could arise from increased responsiveness of postsynaptic dopamine receptors. In the case of GRK6-deficient mice, this could result from the upregulation of receptors, increased affinity for endogenous agonist, and/or enhanced coupling to G proteins. To assess the receptor levels and agonist binding parameters of striatal D1like and D2-like dopamine receptors, specific antagonist radioligand binding to striatal membranes from wildtype and GRK6 mutant animals was performed, and the ability of dopamine to compete with antagonist binding was assessed. Data were fitted according to a two-site model (Figure 5). It has been appreciated for many years that G protein-coupled receptors, including dopamine receptors, exist in two affinity states, and that the proportion of receptors in the high-affinity state versus lowaffinity state, or the relative affinity of these states for agonists, reflect the coupling of the receptor to downstream G proteins (Kent et al., 1980; De Lean et al., 1982). No changes were found in the total levels of D1 dopamine receptor, proportions in high and low agonist affinity states, or the affinity for dopamine between wildtype and GRK6-deficient mice striatal membranes (Figure 5A). In contrast, the affinity of dopamine for D2 receptors in the high agonist affinity state was increased in heterozygous (3-fold) and knockout (4-fold) mice (Khigh [nM]: wt, 15.0 ⫾ 4.4; heterozygote, 4.9 ⫾ 1.1*; KO, 4.1 ⫾ 1.2*; *p ⬍ 0.05, one-way ANOVA followed by NewmanKeuls test), without changes in total receptor levels (Figure 5B). These data suggest that the absence or decrease in GRK6 levels leads to enhanced coupling of striatal D2-like receptors, but little or no change in D1like receptors (Kent et al., 1980; De Lean et al., 1982). To assess potential changes in basal dopamine receptor sensitivity in the absense of GRK6, an analysis of dopamine receptor coupling to G proteins was per-
dopamine level in the striatum of freely moving mice measured using conventional microdialysis. Data are presented as a percentage of the average level of dopamine measured in at least three samples collected before the drug administration. (Saline, wt, n ⫽ 5; GRK6KO, n ⫽ 4; Cocaine, wt, n ⫽ 7; GRK6-KO, n ⫽ 6).
Neuron 296
Figure 5. Binding Parameters of D1 and D2 Dopamine Receptors in the Striatum of GRK6 Mutant Mice Measured in Competition Binding Assay D1 receptor binding (A) was assessed using [3H]SCH23390 (n ⫽ 4 per genotype) and D2 receptor binding (B) using [3H]raclopride (n ⫽ 6 per genotype) as antagonist radioligands, respectively. Dopamine was used as a competing agonist at 24 concentrations. Under the present conditions no alterations in D1 dopamine receptor characteristics were observed (Bound [fmol/mg]: wt, 688 ⫾ 18; heterozygote, 730 ⫾ 42; KO, 741 ⫾ 46; Khigh [nM]: wt, 235 ⫾ 70; heterozygote, 334 ⫾ 69; KO, 303 ⫾ 55; Klow [M]: wt, 15.8 ⫾ 1.5; heterozygote, 17.2 ⫾ 1.1; KO, 16.7 ⫾ 0.2; Khigh/Klow [%]: wt, 32/68; heterozygote, 37/63; KO, 36/64). Properties of D2-like dopamine receptors in high agonist affinity state were significantly altered in GRK6 mutant mice (Bound [fmol/mg]: wt, 200 ⫾ 15; heterozygote, 188 ⫾ 11; KO, 191 ⫾ 7; Khigh [nM]: wt, 15.0 ⫾ 4.4; heterozygote, 4.9 ⫾ 1.1*; KO, 4.1 ⫾ 1.2*; Klow [M]: wt, 1.47 ⫾ 0.36; heterozygote, 0.70 ⫾ 0.07; KO, 0.85 ⫾ 0.22; Khigh/Klow [%]: wt, 49/51; heterozygote, 43/57; KO, 45/55). *p ⬍ 0.05, one-way ANOVA followed by Newman-Keuls test.
formed using a [35S]GTP␥S binding assay (Gainetdinov et al., 1999; Lazareno, 1999; Newman-Tancredi et al., 1999). The ability of D2/D3 dopamine receptor agonist, quinpirole, to stimulate binding of [35S]GTP␥S to the Gi/Go proteins in the striatal membranes prepared from wild-type and GRK6-KO mice was determined (Figure 6A). Striatal D2/D3 dopamine receptors are able to stimulate a higher level of [35S]GTP␥S binding in the GRK6KO mice, indicating that the lack of GRK6 leads to D2/ D3 dopamine receptors that are more efficiently coupled to their downstream G proteins. G␣ protein (Gi/o/z) levels in GRK6-KO mice were not altered, as assessed by immunoblotting assay (data not shown). This suggests that under basal conditions in the intact animals, dopamine receptors are tonically inhibited by GRK6, and the loss or reduction of this inhibition in GRK6 mutant animals leads directly to receptor supersensitivity (higher coupling). To confirm the ability of GRK6 to regulate D2-like dopamine receptors in a cellular system, GRK6 was coexpressed with either D2 or D3 dopamine receptors (D2R or D3R) in HEK293 cells, and dopamine-stimulated [35S]GTP␥S binding to membranes was determined. In agreement with in vivo observations on loss of GRK6, coexpression of GRK6 substantially impairs D2 and D3 dopamine receptor coupling to G proteins (Figures 6B and 6C). Thus, the absence and overexpression of GRK6 both have profound effects on D2-like dopamine receptor function. Interestingly, in the cell culture system, the effect of GRK6 on the coupling of D2-like dopamine receptors is observed in membrane preparations from cells with no prior exposure to dopamine. Because these
assays contain no ATP, any membrane-associated GRKs are presumed to be inactive during the assay, so that any differences observed reflect the coupling state of the receptors established in the intact cells. GRK6 is known to be membrane associated and to increase the basal (agonist-independent) phosphorylation of several receptors (Pitcher et al., 1998; Penn et al., 2000). This suggests that the wild-type level of GRK6 may provide a basal level of D2/D3 receptor phosphorylation, as well as agonist-stimulated phosphorylation, that is lost in the GRK6 mutant animals, leading to higher receptor sensitivity. Enhanced Locomotor Responses to Combined Stimulation of D1-like and D2-like Dopamine Receptors in GRK6-KO Mice It is well established that locomotor activation induced by dopaminergic drugs requires stimulation of both D1and D2-like receptors (Walters et al., 1987; White et al., 1988; Xu et al., 1997). In order to assess the impact of GRK6 deletion on the responsiveness of dopamine receptors in vivo, we compared the locomotor effects of selective D1 and D2 dopamine receptor agonists injected individually as well as in combination (Figure 7). Wild-type and GRK6-KO mice received injections of saline, D1-like dopamine receptor agonist SKF 81297 (3 mg/kg, i.p.), D2-like dopamine receptor agonist quinpirole (2 mg/kg, i.p.), and the combination of these drugs (Xu et al., 1997). While the locomotor-stimulating effect of SKF 81297 was not significantly different between genotypes, a significant difference was found after quinpirole administration (Figures 7A and 7B). It is well known
Role of GRKs in the Responsiveness to Cocaine 297
that the locomotor effects of D2-like agonists is biphasic, with pronounced hypoactivity induced by lower (presynaptic) doses of the drugs (Usiello et al., 2000) and modest activation induced by higher (postsynaptic) doses of the drugs (Van Hartesveldt, 1997; Zhuang et al., 2001). However, in mice, unlike in rats, the stimulatory action of D2-like agonists is barely evident (Halberda et al., 1997; Usiello et al., 2000; Xu et al., 2000; Zhuang et al., 2001). Accordingly, we observed a rapid and potent inhibitory action of quinpirole on the activity of wt mice, followed by only a slight activation (15–35 min after injection) that is far below the activity level of saline-treated mice (Figures 7A and 7B). In GRK6-KO mice, quinpirole (2 mg/kg, i.p.) potently inhibited locomotion in the first 5 min period after drug injection (Figure 7B), as well as at any time period when administered at lower doses (data not shown). However, at 2 mg/kg it was also significantly more effective in inducing locomotor activation (10–45 min after injection), suggesting enhanced sensitivity of postsynaptic D2-like receptors (Figure 7B). Finally, mice were treated with a combination of D1 and D2 dopamine receptor agonists (SKF 81297, 3 mg/kg, i.p. plus quinpirole, 2 mg/kg, i.p.), and locomotor activity was assessed (Figures 7A and 7C). It is known that in intact mice a tonic stimulation of dopamine receptors by endogenous dopamine and the relatively weak postsynaptic activity of D2 dopamine agonists complicate the clear demonstration of D1/D2 dopamine receptor synergism in the control of locomotion (Walters et al., 1987; Xu et al., 1997). Accordingly, we did not see a significant enhancement of D1 agonist-induced hyperlocomotion by additional stimulation of D2 receptors in both genotypes, but, importantly, a significant difference between wild-type and GRK6-KO mice was found (Figures 7A and 7C). Thus, the lack of GRK6 in mice results in exaggerated locomotor responses to combined stimulation of D1 and D2 dopamine receptors, the difference most likely determined by the altered responsiveness of postsynaptic D2-like receptors. Furthermore, in a separate experiment, pretreatment (30 min) of mice by either D1 dopamine receptor antagonist SCH 23390 (0.1 mg/kg, i.p.) or D2 dopamine receptor antagonist raclopride (2 mg/kg, i.p.) was similarly effective in both genotypes in blocking activation induced by a combination of D1 and D2 agonists (data not shown), thereby supporting the requirement for the cooperative action of D1 and D2 receptors in the control of locomotion (White et al., 1988; Xu et al., 1997). Figure 6. Alterations in GRK6 Level Modulate Dopamine Receptor Coupling to G Proteins (A) [35S]GTP␥S binding to striatal membranes from mutant and wildtype mice. [35S]GTP␥S binding to striatal membranes was determined after stimulation with quinpirole. Percent stimulated [35S]GTP␥S binding was calculated by dividing unstimulated [35S]GTP␥S binding into each agonist-stimulated point. Nonlinear regressions were used to calculate the EC50 parameters (wt, 2.0 ⫾ 0.5 M; GRK6-KO, 1.9 ⫾ 0.6 M). In the absence of agonist stimulation, basal [35S]GTP␥S binding did not differ between genotypes. Experiments were performed in triplicate in which wt and GRK6KO striatal tissue were analyzed simultaneously (n ⫽ 8 per group). p ⬍ 0.001, two-way ANOVA, GRK6-KO versus wt controls. (B) [35S]GTP␥S binding to HEK-293 cell membranes expressing D2R was determined after stimulation with dopamine. At least two independent experiments were performed in triplicate. The same procedure was employed for data treatments. p ⬍ 0.001, two-way ANOVA.
Enhanced Postsynaptic Responsiveness to Dopamine Agonist in GRK6-KO Mice Striatal D2/D3 dopamine receptors have been localized both on presynaptic dopamine nerve terminals and postsynaptic striatal cells (Schwartz et al., 1993; Seeman and Van Tol, 1994; Picetti et al., 1997; Missale et al., 1998; Sibley, 1999). To assess directly postsynaptic
(C) [35S]GTP␥S binding to HEK-293 cell membranes expressing D3R/ Go␣ was determined after stimulation with dopamine. At least two independent experiments were performed in triplicate. The same procedure was employed for data treatments. p ⬍ 0.001, two-way ANOVA.
Neuron 298
Figure 7. Effects of D1- and D2-like Dopamine Receptor Agonists on Locomotor Activity of GRK6-KO and Wild-Type Mice GRK6-KO and wild-type mice (n ⫽ 10) received injections of saline, SKF 81297 (3 mg/ kg, i.p.), quinpirole (2 mg/kg, i.p.), and the combination of these two drugs in consecutive trials separated by 7 days. (A) Total distance traveled by the animals for 90 min after drugs. p ⬍ 0.05 versus effect of the same treatment in wild-type mice, Student’s t test. (B) Time course of the effect of quinpirole (2 mg/kg, i.p.) on locomotion in GRK6-KO and wild-type mice. p ⬍ 0.001 versus wild-type group, two-way ANOVA. (C) Time course of the effect of combination of D1 and D2 agonists (SKF 81297 [3 mg/kg, i.p.] plus quinpirole [2 mg/kg, i.p.]) on the locomotor activation in GRK6-KO and wildtype mice. p ⬍ 0.001 versus wild-type group, two-way ANOVA.
dopamine receptor responsiveness in GRK6-KO mice, the effect of the nonselective dopamine agonist apomorphine was tested in dopamine-depleted mice, where no endogenous dopamine neurotransmission was present (Figure 8). The paradigm used was originally developed to test drugs effective in Parkinson’s disease, where profound loss of dopamine innervation occurs (Carlsson et al., 1991; Carlsson, 2001). Wild-type and GRK6-KO mice were treated with reserpine (5 mg/kg, i.p.) to deplete intraneuronal storage of monoamines including dopamine, and with ␣-methyl-p-tyrosine (250 mg/kg, i.p.) to inhibit dopamine synthesis. Both GRK6 mutant and control mice were completely immobilized by this treatment. Locomotion in dopamine-depleted wild-type and mutant mice was restored by administration of the nonselective D1/D2 dopamine receptor agonist apomorphine (0.2–1 mg/kg, s.c.) (Figures 8A–8C). GRK6-KO mice showed a markedly enhanced locomotor response to apomorphine in comparison to wild-type littermates, directly demonstrating that postsynaptic dopamine receptor responsiveness is enhanced in GRK6 mutant mice. Furthermore, these data suggest that a decrease in GRK6 levels or activity could enhance the behavioral effects of dopamine agonists in situations where the dopaminergic stimulus is limiting, such as Parkinson’s disease (Carlsson, 2001). Discussion An important mechanism for GPCR desensitization is the uncoupling of the activated receptors from further stimulation of their G proteins. This form of desensitization is mediated by the phosphorylation of the activated receptor by members of the family of GRKs (Premont
et al., 1995; Pitcher et al., 1998; Penn et al., 2000). GRKphosphorylated receptors bind to an arrestin protein, which prevents the receptor from activating more G proteins, despite the continued binding of agonist (Pitcher et al., 1998). Several GRK enzymes (GRK2, GRK3, GRK5, and GRK6) are found in the brain (Gainetdinov et al., 2000; Erdtmann-Vourliotis et al., 2001), but the physiological importance and receptor specificity of these enzymes remain unclear. It is evident that specificity of GPCR regulation by GRKs requires cellular colocalization of the proteins but may also involve specificity of receptors for particular GRKs. Further, GRK activity is a highly regulated process and can be modulated by subcellular compartmentalization, alterations in intrinsic activity, or expression levels of the kinase (Penn et al., 2000). Since there are no known general or subtype-specific GRK antagonists, we have utilized a gene “knockout” strategy to begin to address the physiological role of individual GRKs (Jaber et al., 1996; Peppel et al., 1997; Gainetdinov et al., 1999). Here we report that one of the G protein-coupled receptor kinases, GRK6, is localized in striatal neurons receiving dopaminergic input, and that postsynaptic D2/D3 dopamine receptors are physiological targets of this kinase. Furthermore, supersensitivity to dopamine agonist stimulation that occurs in striatal neurons in the absence or reduction of GRK6 suggests that this direct regulation of D2/D3 dopamine receptor by GRK6 represents an important determinant by which responses to psychostimulants or other dopaminergic drugs can be controlled. Interestingly, cellular experiments with coexpressed D2-like dopamine receptors and GRK6 indicate that elevated GRK6 levels can reduce dopamine receptor sig-
Role of GRKs in the Responsiveness to Cocaine 299
Figure 8. Locomotor Responses to Direct Dopamine Receptor Activation Are Enhanced in Dopamine-Depleted GRK6-KO Mice To deplete brain dopamine, animals were treated with a combination of reserpine (5 mg/kg, i.p.) and ␣-methyl-p-tyrosine (250 mg/kg, i.p.) as described in Experimental Procedures. (A) Time course of effect of apomorphine (0.5 mg/kg, s.c.) on the horizontal activity (counts) of dopamine-depleted wild-type (n ⫽ 11) and GRK6-KO (n ⫽ 7) mice. GRK6-KO mice are significantly different from wt controls. p ⬍ 0.001, two-way ANOVA. (B and C) Dose-response of the effect of apomorphine (0.2–1 mg/ kg, s.c.) on the locomotion of dopamine-depleted wild-type and mutant mice (n ⫽ 6–11 per group). Note that GRK6-KO mice were more affected by apomorphine both in terms of horizontal activity counts (B) and total distance traveled (C). p ⬍ 0.001 versus wildtype group for horizontal activity counts (B) and p ⬍ 0.05 for total distance traveled (C) measurements, two-way ANOVA.
naling. Further, when D3 dopamine receptor was coexpressed with GRK2 or GRK3, only a minor effect (less than 15% reduction in dopamine-stimulated [35S]GTP␥S binding) was observed (data not shown). This suggests that specificity of D3 dopamine receptor regulation resides in a preference for GRK6 over other GRKs. Coexpression of GRK6 also enhanced the basal (unstimulated) translocation of -arrestin2 to D2R or D3R, reflecting increased basal receptor phosphorylation (data not shown). Thus, in an in vitro system, GRK6 appears to induce a basal level of desensitization of D2/D3 dopamine receptors. This is in agreement with many previous studies that have shown in other receptor systems that membrane-associated GRK6 induces basal (activationindependent) receptor phosphorylation (Pitcher et al., 1998; Penn et al., 2000), and suggests that this basal receptor phosphorylation tone is physiologically important, at least for some dopamine receptors. Importantly, mice lacking another member of this kinase family, GRK5, do not demonstrate such dopaminergic supersensitivity, but do display exaggeration of central muscarinic cholinergic responses (Gainetdinov et al., 1999). Another GRK widely expressed in the brain,
GRK3, is unlikely to be involved in the regulation of dopaminergic receptors, since mice lacking GRK3 (Peppel et al., 1997) do not demonstrate enhanced responses to either cocaine or apomorphine (unpublished data). GRK4 has very limited expression in the brain (Pitcher et al., 1998; Gainetdinov et al., 2000), precluding its involvement in dopamine receptor regulation in the brain. Deletion of the GRK2 gene results in embryonic lethality (Jaber et al., 1996), so the role of this GRK on dopamine-mediated behaviors has not been fully examined. Thus, GRK6 may play a unique role in D2-like dopamine receptor regulation. Dopamine receptors belong to a large family of receptors that are linked to their signal transduction pathways through G proteins (Grandy and Civelli, 1992; Schwartz et al., 1993; Seeman and Van Tol, 1994; Picetti et al., 1997; Missale et al., 1998; Sibley, 1999). A variety of signaling events can be regulated by dopamine receptors, including adenylyl cyclase and phospholipase C activities and various ion channels. Based on physiological and pharmacological properties, dopamine receptors are defined as D1-like (D1 and D5) and D2-like (D2, D3, D4) subfamilies of dopamine receptors (Missale et al., 1998). Many of these subtypes (particularly belonging to the same subfamily) respond similarly to pharmacological agents, making it difficult to selectively stimulate or block a specific receptor subtype using pharmacological approaches (Missale et al., 1998; Sibley, 1999). In the brain, the various receptor subtypes display specific distributions, with D1-like receptors being mainly postsynaptic and D2-like receptors being both pre- and postsynaptic (Picetti et al., 1997; Missale et al., 1998; Sibley, 1999). An important question arising from the present observations is what subtypes of dopamine receptors are primarily regulated by GRK6. Assessment of binding characteristics of striatal dopamine receptors and locomotor responses to selective agonists demonstrated that properties of D2-like dopamine receptors, but not those of D1-like, are significantly affected in GRK6 mutant mice. Furthermore, D2-like dopamine receptors appear to be more efficiently coupled to their G proteins and thereby contribute to the enhanced locomotor responses to dopaminergic stimulation in GRK6-KO mice. Due to lack of selectivity of the pharmacological approaches, it is impossible at present to discriminate between individual subtypes of D2-like dopamine receptors responsible for these effects. Nevertheless, since D4 dopamine receptor has a limited expression in the striatum (Rivera et al., 2002), the most likely candidates would be D2 and D3 dopamine receptors. Direct challenge of postsynaptic dopamine receptors using dopamine depletion approach (Carlsson et al., 1991) revealed that postsynaptic receptors are primarily affected. However, the present data do not exclude the possibility that presynaptic D2/D3 dopamine “autoreceptors,” low levels of striatal D4 dopamine receptors, and/or various populations of dopamine receptors in the other brain areas might also be affected in GRK6-KO mice. In addition, the widespread expression of GRK6 in brain suggests that nondopaminergic receptor types may also be physiological targets for this kinase; detailed investigation will be required to establish the portfolio of receptors affected in GRK6 mutant mice. It has been previously reported that supersensitivity
Neuron 300
to light is present in GRK1-KO animals (Chen et al., 1999), cardiac supersensitivity to -adrenergic stimulation is present in heterozygous GRK2-KO mice (Rockman et al., 1998), supersensitivity to olfactory stimuli (Peppel et al., 1997) and enhanced muscarinic cholinergic airway responsiveness (Walker et al., 1999) are present in GRK3-KO animals, and central muscarinic receptor supersensitivity is observed in GRK5-KO mice (Gainetdinov et al., 1999). These observations as well as present findings strongly suggest a generalized principle that abrogation of GRK-mediated receptor desensitization can enhance responses to both endogenous and exogenous ligands and thus may represent a potential mechanism for behavioral or physiological supersensitivity. One example of behavioral supersensitivity is chronic psychostimulant-induced enhancement of dopaminergic responses (Wise and Rompre, 1989; Robinson and Berridge, 1993), which is believed to be associated with changes in several signaling molecules downstream of DA receptors (Nestler and Aghajanian, 1997; Hyman and Malenka, 2001; Nestler, 2001; Laakso et al., 2002). The present observatons raise the possibility that GRK6regulated desensitization of dopamine receptors might represent a novel component in the development of such sensitization. The observation that single-allele inactivation of GRK6 in mice produces a phenotype identical to the complete knockout of the gene raises the possibility that even subtle allelic variations in the human GRK6 gene or drug-induced alterations in GRK6 activity might contribute to individual sensitivity to psychostimulants and/or other drugs affecting dopaminergic function. Furthermore, since supersensitivity of DA receptors has been implicated in the pathogenesis or adverse reactions associated with treatment of schizophrenia, Tourette Syndrome, and Parkinson’s disease (Pandey et al., 1977; Kirkpatrick et al., 1992; Jenner and Marsden, 1987; Singer, 1994), a role for GRK6-mediated DA receptor regulation in these conditions would be of interest to consider. Particularly, a pharmacological strategy based on blockade of GRK6-mediated desensitization may be effective in enhancing the potency of the low levels of residual dopamine found in parkinsonian patients, or augmentating the efficacy of L-DOPA replacement therapy. Experimental Procedures Targeted Deletion of the Mouse GRK6 Locus The three Triple-Lox vectors described for the GRK5 knockout (Gainetdinov et al., 1999) were modified as follows. The LoxL vector was modified to have a new multiple cloning site containing NotI and NheI sites, a loxP element, and SpeI, XhoI, and Sse8387I sites. The LoxR/DT vector was modified to contain a multiple cloning region containing NotI, AscI, XbaI, XhoI, EcoRI, and PmeI sites. Phage carrying fragments of the mouse GRK6 gene from the 129/SVJ strain were obtained and sequenced as described (Premont et al., 1999). The 2.75 kb XbaI-NheI fragment containing exons 3 through 9 (the gene fragment to be flanked by loxP sites, or floxed) and the 6.8 kb NheI-NotI fragment containing exons 10 through 15 (the long recombination arm) were ligated into the SpeI and NotINheI sites of the LoxL vector, respectively. The 1.3 kb XbaI gene fragment containing exon 2 (the short recombination arm) was prepared by partial digestion and ligated into the XbaI site in the LoxR/ DT vector. The cassette containing the long recombination arm, loxP site, and gene fragment to be floxed were excised from the
LoxL vector with NotI and Sse8387I and ligated into those sites of the LoxC/TK-NEO vector. The long recombination arm, floxed gene fragment, and floxed TK-NEO marker cassette were excised from this vector with NotI and AscI and inserted into those sites in the LoxR/DT vector to create the final targeting construct (Figure 1A). Growth and selection of targeted ES cells and creation of chimeric mice was performed essentially as described (Hogan et al., 1994). The targeting DNA was linearized by digestion with NotI and electroporated into AK7 ES cells. Cells were selected for growth in media containing 200 g/ml G418, and surviving clones were tested for proper integration by amplification of a DNA band from the PGK promoter of the NEO marker gene to a region near exon 2 adjacent to the targeting construct. The identity of positive clones was confirmed by Southern blotting of genomic DNA isolated from the cells and digested with SpeI, using a probe from adjacent to exon 2 (Figure 1A). A targeted cell clone with a normal karyotype was expanded and microinjected into day 3.5 C57BL/6J mouse blastocysts, which were then injected into the uterus of a day 2.5 pseudopregnant B6SJLF1/J mouse. Chimeric offspring were crossed with C57BL/6J mice to generate agouti pups that carried the targeted “lox” GRK6 gene. F1 heterozygote animals were bred with transgenic mice bearing CMV-Cre (backcrossed to a C57Bl/6J genetic background; Nagy et al., 1998) to induce deletion of the floxed cassettes. From offspring of these crosses, GRK6-KO animals were obtained, in which both the exon 3–9 cassette and the TK-NEO marker gene cassettes were deleted (i.e., the gene is inactive). Genotyping was routinely performed on tail tip DNA using a PCR method utilizing three primers to simultaneously detect the wild-type and mutant loci (Figure 1B). Western Blots Mouse brain regions were dissected on ice and immediately frozen in liquid nitrogen. Crude membranes were prepared from mouse brain regions by polytron homogenization in buffer (20 mM Tris, 1 mM EDTA, 100 mM NaCl [pH 7.4]) followed by centrifugation at 200 ⫻ g for 2 min and then at 21,000 ⫻ g for 30 min. Aliquots (60 g) of each sample were solubilized by addition of SDS-PAGE sample buffer and separated by 10% SDS-PAGE. Transferred proteins were blotted with polyclonal anti-GRK6 (Stoffel et al., 1994) and visualized using enhanced chemiluminescent development (ECL, Amersham, Piscataway, NJ). Immunohistochemistry Wild-type or GRK6-KO mice (n ⫽ 3) were anesthetized with chloral hydrate (400 mg/kg, i.p.) and perfused transcardially with 0.9% saline, followed by 4% paraformaldehyde in 0.1 M borax buffer (pH 9.5 at 4⬚C). Brains were post-fixed for 1–3 days and cryoprotected in 10% sucrose. Free floating coronal or saggital sections of 25 m were incubated 32 hr at 4⬚C with a mixture of antibodies against DARPP-32 (mouse, 1:5000, D97520, BD Transduction Laboratories) and GRK6 (rabbit, 1:50, sc-566, Santa Cruz Biotechnologies, Inc.). The second immunoreaction step was performed by incubation (1 hr at room temperature) with each of the following antibodies: fluorescein-isothiocyanate-labeled goat anti-rabbit IgG and Texas Redlabeled goat anti-mouse IgG (Vector Laboratories). Slides were viewed on a laser scanning Zeiss confocal microscope (LSM-510) using the Roper Scientific Cooled CCD digital camera (CoolsnapFX, BioVision Technologies, Inc., PA) and the IPLab software for Windows v3.0 for image processing (BioVision Technologies, Inc). Images were acquired separately in each channel (dual scan mode) to eliminate the possibility of signal bleed-over from one channel to the other. Animal Treatment/Drugs/Behavior 3- to 4-month-old littermate wild-type and GRK6 mutant mice (C57BL/6J ⫻ 129/SvJ) were used in these experiments. In all experiments, wild-type littermates served as controls for mutant mice, and all the genotypes were evaluated concurrently. Horizontal, vertical, and stereotypical activities of littermate wild-type, heretozygote, and knockout mice of both genders were measured in an Omnitech Digiscan activity monitor (42 cm2). Locomotor activity was measured at 5 min intervals and cumulative counts were taken for data analysis (Wang et al., 1997; Xu et al., 2000). To evaluate the effects of cocaine,
Role of GRKs in the Responsiveness to Cocaine 301
amphetamine, and -phenylethylamine on locomotor behavior, mice were placed in activity monitor, 30–60 min later they were injected with drugs or vehicle i.p., and locomotor activity was monitored for the following 90 min. In cocaine sensitization (Wang et al., 1997; Xu et al., 2000) experiments, mice were treated chronically with cocaine (20 mg/kg, i.p.) for 5 days and their responses to challenging dose of cocaine were analyzed at day 7. To analyze the effect of direct dopamine agonist in dopamine-depleted mice (Carlsson et al., 1991), animals were pretreated with a combination of reserpine (5 mg/kg, i.p. 20 hr before the experiment) and ␣-methyl-p-tyrosine (250 mg/kg, i.p., 1 hr before the experiment). This treatment resulted in depletion of striatal dopamine to less than 0.75% in both wildtype and GRK6-KO mice (data not shown). Mice were completely immobilized by this treatment. Dopamine-depleted mice were treated with vehicle or D1/D2 dopamine receptor agonist apomorphine (0.2–1 mg/kg, s.c.), and locomotor activity was immediately analyzed as described above. A nonrandomized repeated measures design was employed to assess the locomotor effects of selective and combined stimulation of D1-like and D2-like dopamine receptors (Xu et al., 1997). Once a week each mouse was habituated to locomotor activity chambers (30 min) and received saline, R-(⫹)SKF-81297 (SKF 81297, 3 mg/kg, i.p.), quinpirole (2 mg/kg, i.p.), or combination of SKF 81297 (3 mg/kg, i.p.) plus quinpirole (2 mg/kg, i.p.), consecutively. In all other experiments (with exception of the sensitization paradigm), each animal received only a single injection with a tested drug. All the data presented in this study are expressed as means ⫾ SEM. Neurochemical Assessments For monoamine analyses, brain regions were dissected and monoamines extracted and analyzed for levels of dopamine, serotonin (5-hydroxytryptamine, 5-HT), and metabolites 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), and 5-hydroxyindoleacetic acid (5-HIAA) using high performance liquid chromatography with electrochemical detection as described (Wang et al., 1997; Xu et al., 2000). To perform in vivo microdialysis experiments (Wang et al., 1997; Xu et al., 2000), mice were anesthetized and dialysis probes were implanted into the right striatum. Twenty-four hours after surgery, the dialysis probe was connected to a syringe pump and perfused with artificial CSF. Quantitative “low perfusion” rate (70 nl/ min) microdialysis experiments (Wang et al., 1997; Xu et al., 2000) were conducted in freely moving mice for determination of basal extracellular dopamine levels in striatum. To analyze the effects of cocaine on the extracellular dopamine levels in striatum, “conventional” microdialysis method (perfusion flow rate 1 l/min) in freely moving animals was employed (Wang et al., 1997; Xu et al., 2000). To measure [3H]dopamine uptake in striatal synaptosomes, striatal tissue from wild-type and GRK6-KO mice were analyzed as described (Sandoval et al., 2001). Analyses of Dopamine Receptor Coupling by [35S]GTP␥S Binding In Vivo and In Vitro In in vivo experiments, D2/D3 dopamine receptor agonist quinpirolestimulated [35S]GTP␥S binding to striatal membranes from GRK6KO and wild-type mice was assessed as previously described (Gainetdinov et al., 1999). To directly assess in vitro the role of GRK6 in D2 and D3 dopamine receptor regulation, dopamine-stimulated [35S]GTP␥S binding to cultured cell membranes was used (NewmanTancredi et al., 1999). HEK-293 cells were transfected with D2R or D3R/Go␣ with and without GRK6. For D2R, 20 g of cell membrane proteins were incubated in a buffer containing 20 mM HEPES (pH 7.4), 10 mM MgCl2, 150 mM NaCl, 3 M GDP, and 0.1 nM [35S]GTP␥S for 1 hr at room temperature. For D3R, 20 g of cell membrane proteins were incubated in a buffer containing 25 mM HEPES (pH 7.4), 120 mM NaCl, 1.8 mM KCl, 20 mM MgCl2, 20 M GDP, 0.2 nM [35S]GTP␥S, and 1 mM sodium deoxycholate for 2 hr at 30⬚C. Incubation mixtures were filtered with GF/B filter and washed with 10 mM sodium phosphate buffer. Agonist Competition Binding Assay Striata from GRK6 wild-type, heterozygote, and knockout mice were rapidly dissected on ice and frozen in liquid nitrogen. Tissue was homogenized in an ice-cold buffer containing 50 mM Tris-HCl (pH
7.4), 120 mM NaCl, and 1 mM EDTA, and centrifuged for 20 min at 40,000 ⫻ g at 4⬚C. Pellets were resuspended in the same buffer and incubated for 15 min at 37⬚C to facilitate the removal of endogenous DA. After incubation, membranes were centrifuged for 10 min at 40,000 ⫻ g at 4⬚C, resuspended in ice-cold buffer, and centrifuged again. The resulting pellet was then homogenized in the binding buffer containing 50 mM Tris-HCl (pH 7.4), 120 mM NaCl, 5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, and 0.1% ascorbic acid. D1 receptor binding was assessed using 1 nM [3H]SCH23390 in the presence of 100 nM ketanserin (to prevent binding to 5-HT2A receptors), and D2 receptor binding was assessed using 3 nM [3H]raclopride (New England Nuclear/Perkin Elmer Life Sciences). Dopamine was used as the competing agonist (24 concentration points in duplicate, from 10⫺8.5 to 10⫺2 M for D1 receptor, and from 10⫺9.5 to 10⫺3 M for D2 receptor). Protein concentration was measured using the Lowry assay, and 40 g and 60 g of membrane protein in 200 l reaction volume was used for each point for D1 and D2 receptor binding, respectively. Binding reaction was incubated for 45 min at room temperature, and bound ligand was separated from free by rapid filtration onto Whatman GF/C filters washed with ice-cold Tris buffer. Two-site competition curves were fitted individually for every animal (striata from two animals were pooled for each “n” in D2 binding assay) by nonlinear regression using GraphPad Prism software (GraphPad Software Inc., San Diego, CA). Values of KHigh, KLow, fraction of receptors in KHigh, and the amount of specifically bound radioligand generated by nonlinear regression algorithm were subjected to statistical analysis. Statistical significances were assessed with one-way ANOVA followed by Newman-Keuls test for post hoc analyses. Acknowledgments This work was supported in part by grants from the National Institutes of Health (DA-14600 to L.M.B., NS-19576 and MH-40159 to M.G.C., and HL-16037 to R.J.L.) and unrestricted grants from BristolMyers Squibb (Neuroscience [M.G.C.]; Cardiovascular [R.J.L.]). M.C. was supported by FRSQ fellowship. A.L. was supported in part by Academy of Finland. We thank Cheryl Bock and the Transgenic Core Facility of the Duke Comprehensive Cancer Center for assistance in generating the GRK6-KO mice; Dr. Mark Bender (Fred Hutchison Cancer Research Center) for the C57Bl/6J-backcrossed CMV-Cre transgenic mouse strain; and Sandy Duncan, Rachel McAdams, Kristina Riebe, and Susan Suter for animal care. M.G.C. and R.J.L. are Investigators of the Howard Hughes Medical Institute. Received: September 17, 2002 Revised: February 13, 2003 Accepted: March 8, 2003 Published: April 23, 2003 References Aizman, O., Brismar, H., Uhlen, P., Zettergren, E., Levey, A.I., Forssberg, H., Greengard, P., and Aperia, A. (2000). Anatomical and physiological evidence for D1 and D2 dopamine receptor colocalization in neostriatal neurons. Nat. Neurosci. 3, 226–230. Aubert, I., Ghorayeb, I., Normand, E., and Bloch, B. (2000). Phenotypical characterization of the neurons expressing the D1 and D2 dopamine receptors in the monkey striatum. J. Comp. Neurol. 418, 22–32. Benovic, J.L., and Gomez, J. (1993). Molecular cloning and expression of GRK6. A new member of the G protein-coupled receptor kinase family. J. Biol. Chem. 268, 19521–19527. Borowsky, B., Adham, N., Jones, K.A., Raddatz, R., Artymyshyn, R., Ogozalek, K.L., Durkin, M.M., Lakhlani, P.P., Bonini, J.A., Pathirana, S., et al. (2001). Trace amines: identification of a family of mammalian G protein-coupled receptors. Proc. Natl. Acad. Sci. USA 98, 8966– 8971. Carlsson, A. (2001). A paradigm shift in brain research. Science 294, 1021–1024. Carlsson, M., Svensson, A., and Carlsson, A. (1991). Synergistic interactions between muscarinic antagonists, adrenergic agonists and NMDA antagonists with respect to locomotor stimulatory effects
Neuron 302
in monoamine-depleted mice. Naunyn Schmiedebergs Arch. Pharmacol. 343, 568–573. Chen, C.K., Burns, M.E., Spencer, M., Niemi, G.A., Chen, J., Hurley, J.B., Baylor, D.A., and Simon, M.I. (1999). Abnormal photoresponses and light-induced apoptosis in rods lacking rhodopsin kinase. Proc. Natl. Acad. Sci. USA 96, 3718–3722. De Lean, A., Kilpatrick, B.F., and Caron, M.G. (1982). Dopamine receptor of the porcine anterior pituitary gland. Evidence for two affinity states discriminated by both agonists and antagonists. Mol. Pharmacol. 22, 290–297. Di Chiara, G., Morelli, M., and Consolo, S. (1994). Modulatory functions of neurotransmitters in the striatum: ACh/dopamine/NMDA interactions. Trends Neurosci. 17, 228–233. Erdtmann-Vourliotis, M., Mayer, P., Ammon, S., Riechert, U., and Hollt, V. (2001). Distribution of G-protein-coupled receptor kinase (GRK) isoforms 2, 3, 5 and 6 mRNA in the rat brain. Brain Res. Mol. Brain Res. 95, 129–137. Fong, A.M., Premont, R.T., Richardson, R.M., Yu, Y.R., Lefkowitz, R.J., and Patel, D.D. (2002). Defective lymphocyte chemotaxis in beta-arrestin2- and GRK6-deficient mice. Proc. Natl. Acad. Sci. USA 99, 7478–7483. Gainetdinov, R.R., Bohn, L.M., Walker, J.K., Laporte, S.A., Macrae, A.D., Caron, M.G., Lefkowitz, R.J., and Premont, R.T. (1999). Muscarinic supersensitivity and impaired receptor desensitization in G protein-coupled receptor kinase 5-deficient mice. Neuron 24, 1029– 1036. Gainetdinov, R.R., Premont, R.T., Caron, M.G., and Lefkowitz, R.J. (2000). Reply: receptor specificity of G-protein-coupled receptor kinases. Trends Pharmacol. Sci. 21, 366–367. Gainetdinov, R.R., Sotnikova, T.D., and Caron, M.G. (2002). Monoamine transporter pharamacology and mutant mice. Trends Pharmacol. Sci. 23, 367–373. Grandy, D.K., and Civelli, O. (1992). G-protein-coupled receptors: the new dopamine receptor subtypes. Curr. Opin. Neurobiol. 2, 275–281. Greengard, P. (2001). The neurobiology of slow synaptic transmission. Science 294, 1024–1030. Hanks, S.K., Quinn, A.M., and Hunter, T. (1988). The protein kinase family: conserved features and deduced phylogeny of the catalytic domains. Science 241, 42–52. Halberda, J.P., Middaugh, L.D., Gard, B.E., and Jackson, B.P. (1997). DAD1- and DAD2-like agonist effects on motor activity of C57 mice: differences compared to rats. Synapse 26, 81–92. Hogan, B., Beddington, R., Constantini, F., and Lacy, E. (1994). Manipulation of the Mouse Embryo: A Laboratory Manual, Second Edition (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press).
Kim, K.M., Valenzano, K.J., Robinson, S.R., Yao, W.D., Barak, L.S., and Caron, M.G. (2001). Differential regulation of the dopamine D2 and D3 receptors by G protein-coupled receptor kinases and betaarrestins. J. Biol. Chem. 276, 37409–37414. Kirkpatrick, B., Alphs, L., and Buchanan, R.W. (1992). The concept of supersensitivity psychosis. J. Nerv. Ment. Dis. 180, 265–270. Laakso, A., Mohn, A.R., Gainetdinov, R.R., and Caron, M.G. (2002). Experimental genetic approaches to addiction. Neuron 36, 213–228. Lazareno, S. (1999). Measurement of agonist-stimulated [35S]GTP gamma S binding to cell membranes. In Receptor Binding Techniques, Volume 106, M. Keen, ed. (Totowa, NJ: Humana Press), pp. 231–247. Le Moine, C., Normand, E., and Bloch, B. (1991). Phenotypical characterization of the rat striatal neurons expressing the D1 dopamine receptor gene. Proc. Natl. Acad. Sci. USA 88, 4205–4209. Missale, C., Nash, S.R., Robinson, S.W., Jaber, M., and Caron, M.G. (1998). Dopamine receptors: from structure to function. Physiol. Rev. 78, 189–225. Nagy, A., Moens, C., Ivanyi, E., Pawling, J., Gertsenstein, M., Hadjantonakis, A.K., Pirity, M., and Rossant, J. (1998). Dissecting the role of N-myc in development using a single targeting vector to generate a series of alleles. Curr. Biol. 8, 661–664. Nestler, E.J. (2001). Molecular basis of long-term plasticity underlying addiction. Nat. Rev. Neurosci. 2, 119–128. Nestler, E.J., and Aghajanian, G.K. (1997). Molecular and cellular basis of addiction. Science 278, 58–63. Newman-Tancredi, A., Cussac, D., Audinot, V., Pasteau, V., Gavaudan, S., and Millan, M.J. (1999). G protein activation by human dopamine D3 receptors in high-expressing Chinese hamster ovary cells: a guanosine-5⬘-O-(3-[35S]thio)- triphosphate binding and antibody study. Mol. Pharmacol. 55, 564–574. Pandey, G.N., Garver, D.L., Tamminga, C., Ericksen, S., Ali, S.I., and Davis, J.M. (1977). Postsynaptic supersensitivity in schizophrenia. Am. J. Psychiatry 134, 518–522. Penn, R.B., Pronin, A.N., and Benovic, J.L. (2000). Regulation of G protein-coupled receptor kinases. Trends Cardiovasc. Med. 10, 81–89. Peppel, K., Boekhoff, I., McDonald, P., Breer, H., Caron, M.G., and Lefkowitz, R.J. (1997). G protein-coupled receptor kinase 3 (GRK3) gene disruption leads to loss of odorant receptor desensitization. J. Biol. Chem. 272, 25425–25428. Picetti, R., Saiardi, A., Abdel Samad, T., Bozzi, Y., Baik, J.H., and Borrelli, E. (1997). Dopamine D2 receptors in signal transduction and behavior. Crit. Rev. Neurobiol. 11, 121–142. Pitcher, J.A., Freedman, N.J., and Lefkowitz, R.J. (1998). G proteincoupled receptor kinases. Annu. Rev. Biochem. 67, 653–692.
Hyman, S.E., and Malenka, R.C. (2001). Addiction and the brain: the neurobiology of compulsion and its persistence. Nat. Rev. Neurosci. 2, 695–703.
Premont, R.T., Inglese, J., and Lefkowitz, R.J. (1995). Protein kinases that phosphorylate activated G protein-coupled receptors. FASEB J. 9, 175–182.
Ito, K., Haga, T., Lameh, J., and Sadee, W. (1999). Sequestration of dopamine D2 receptors depends on coexpression of G-proteincoupled receptor kinases 2 or 5. Eur. J. Biochem. 260, 112–119.
Premont, R.T., Macrae, A.D., Aparicio, S.A., Kendall, H.E., Welch, J., and Lefkowitz, R.J. (1999). The GRK4 subfamily of G proteincoupled receptor kinases: alternative splicing, gene organization and sequence conservation. J. Biol. Chem. 274, 29381–29389.
Jaber, M., Koch, W.J., Rockman, H., Smith, B., Bond, R.A., Sulik, K.K., Ross, J., Jr., Lefkowitz, R.J., Caron, M.G., and Giros, B. (1996). Essential role of beta-adrenergic receptor kinase 1 in cardiac development and function. Proc. Natl. Acad. Sci. USA 93, 12974–12979. Janssen, P.A., Leysen, J.E., Megens, A.A., and Awouters, F.H. (1999). Does phenylethylamine act as an endogenous amphetamine in some patients? Int. J. Neuropsychopharmacol. 2, 229–240. Jenner, P., and Marsden, C.D. (1987). Chronic pharmacological manipulation of dopamine receptors in brain. Neuropharmacology 26, 931–940. Jones, S.R., Gainetdinov, R.R., Wightman, R.M., and Caron, M.G. (1998). Mechanisms of amphetamine action revealed in mice lacking the dopamine transporter. J. Neurosci. 18, 1979–1986. Kent, R.S., De Lean, A., and Lefkowitz, R.J. (1980). A quantitative analysis of beta-adrenergic receptor interactions: resolution of high and low affinity states of the receptor by computer modeling of ligand binding data. Mol. Pharmacol. 17, 14–23.
Premont, R.T., Gainetdinov, R.R., and Caron, M.G. (2001). Following the trace of elusive amines. Proc. Natl. Acad. Sci. USA 98, 9474– 9475. Rivera, A., Cuellar, B., Giron, F.J., Grandy, D.K., de la Calle, A., and Moratalla, R. (2002). Dopamine D4 receptors are heterogeneously distributed in the striosomes/matrix compartments of the striatum. J. Neurochem. 80, 219–229. Robinson, T.E., and Berridge, K.C. (1993). The neural basis of drug craving: an incentive-sensitization theory of addiction. Brain Res. Brain Res. Rev. 18, 247–291. Rockman, H.A., Choi, D.J., Akhter, S.A., Jaber, M., Giros, B., Lefkowitz, R.J., Caron, M.G., and Koch, W.J. (1998). Control of myocardial contractile function by the level of beta-adrenergic receptor kinase 1 in gene-targeted mice. J. Biol. Chem. 273, 18180–18184. Sandoval, V., Riddle, E.L., Ugarte, Y.V., Hanson, G.R., and Fleckenstein, A.E. (2001). Methamphetamine-induced rapid and revers-
Role of GRKs in the Responsiveness to Cocaine 303
ible changes in dopamine transporter function: an in vitro model. J. Neurosci. 21, 1413–1419. Schwartz, J.C., Levesque, D., Martres, M.P., and Sokoloff, P. (1993). Dopamine D3 receptor: basic and clinical aspects. Clin. Neuropharmacol. 16, 295–314. Seeman, P., and Van Tol, H.H. (1994). Dopamine receptor pharmacology. Trends Pharmacol. Sci. 15, 264–270. Sibley, D.R. (1999). New insights into dopaminergic receptor function using antisense and genetically altered animals. Annu. Rev. Pharmacol. Toxicol. 39, 313–341. Singer, H.S. (1994). Neurobiological issues in Tourette syndrome. Brain Dev. 16, 353–364. Stoffel, R.H., Randall, R.R., Premont, R.T., Lefkowitz, R.J., and Inglese, J. (1994). Palmitoylation of G protein-coupled receptor kinase, GRK6. Lipid modification diversity in the GRK family. J. Biol. Chem. 269, 27791–27794. Tiberi, M., Nash, S.R., Bertrand, L., Lefkowitz, R.J., and Caron, M.G. (1996). Differential regulation of dopamine D1A receptor responsiveness by various G protein-coupled receptor kinases. J. Biol. Chem. 271, 3771–3778. Usiello, A., Baik, J.H., Rouge-Pont, F., Picetti, R., Dierich, A., LeMeur, M., Piazza, P.V., and Borrelli, E. (2000). Distinct functions of the two isoforms of dopamine D2 receptors. Nature 408, 199–203. Van Hartesveldt, C. (1997). Temporal and environmental effects on quinpirole-induced biphasic locomotion in rats. Pharmacol. Biochem. Behav. 58, 955–960. Walters, J.R., Bergstrom, D.A., Carlson, J.H., Chase, T.N., and Braun, A.R. (1987). D1 dopamine receptor activation required for postsynaptic expression of D2 agonist effects. Science 236, 719–722. Walker, J.K., Peppel, K., Lefkowitz, R.J., Caron, M.G., and Fisher, J.T. (1999). Altered airway and cardiac responses in mice lacking G protein-coupled receptor kinase 3. Am. J. Physiol. 276, R1214– R1221. Wang, Y.M., Gainetdinov, R.R., Fumagalli, F., Xu, F., Jones, S.R., Bock, C.B., Miller, G.W., Wightman, R.M., and Caron, M.G. (1997). Knockout of the vesicular monoamine transporter 2 gene results in neonatal death and supersensitivity to cocaine and amphetamine. Neuron 19, 1285–1296. Willner, P. (1997). The mesolimbic dopamine system as a target for rapid antidepressant action. Int. Clin. Psychopharmacol. 12 (Suppl 3), S7–S14. Wise, R.A., and Rompre, P.P. (1989). Brain dopamine and reward. Annu. Rev. Psychol. 40, 191–225. White, F.J., Bednarz, L.M., Wachtel, S.R., Hjorth, S., and Brooderson, R.J. (1988). Is stimulation of both D1 and D2 receptors necessary for the expression of dopamine-mediated behaviors? Pharmacol. Biochem. Behav. 30, 189–193. Xu, M., Koeltzow, T.E., Santiago, G.T., Moratalla, R., Cooper, D.C., Hu, X.T., White, N.M., Graybiel, A.M., White, F.J., and Tonegawa, S. (1997). Dopamine D3 receptor mutant mice exhibit increased behavioral sensitivity to concurrent stimulation of D1 and D2 receptors. Neuron 19, 837–848. Xu, F., Gainetdinov, R.R., Wetsel, W.C., Jones, S.R., Bohn, L.M., Miller, G.W., Wang, Y.M., and Caron, M.G. (2000). Mice lacking the norepinephrine transporter are supersensitive to psychostimulants. Nat. Neurosci. 3, 465–471. Zhuang, X., Oosting, R.S., Jones, S.R., Gainetdinov, R.R., Miller, G.W., Caron, M.G., and Hen, R. (2001). Hyperactivity and impaired response habituation in hyperdopaminergic mice. Proc. Natl. Acad. Sci. USA 98, 1982–1987.
