Dopamine Receptors: Novel Insights from ... - SAGE Journals

2 downloads 0 Views 349KB Size Report
Dopamine receptors are targets for drugs with antipsychotic potency, and they ... as the D1 class of receptors (D1 and D5) and the D2 class of receptors (D2, D3, ...
eceptor Expression and Function Volume 6, Number 2, 2000THE NEUROSCIENTIST

REVIEW n

Dopamine Receptors: Novel Insights from Biochemical and Genetic Studies CLAUDIA SCHMAUSS Department of Psychiatry/Neuroscience Columbia University New York, NY

Dopamine receptors are targets for drugs with antipsychotic potency, and they are also the primary target in the treatment of Parkinson’s disease. Molecular cloning has identified five genes that code for dopamine receptors. These receptors belong in two functionally distinct classes of G-protein-coupled receptors, known as the D1 class of receptors (D1 and D5) and the D2 class of receptors (D2, D3, and D4). The diversity of dopamine receptor subtypes that belong to the same functional class, their high degree of structural similarity, and the lack of antagonists with selectivity for each of the individual subtypes have challenged studies on the function of the individual receptor subtypes. This review focuses on the recent progress made with studies on the expression and function of D1, D2, and D3 receptors. It summarizes results of studies that suggest that these receptor proteins are expressed in monomeric and oligomeric forms and reviews results of a growing number of gene-targeting studies that begin to illustrate major differences in the phenotypes of D1-, D2-, and D3-mutant mice. KEY WORDS Dopamine receptors, Oligomeric assembly, Gene targeting, Homologous recombination

Four major neuronal systems of the brain use dopamine as the principal neurotransmitter to modulate locomotor behavior (nigrostriatal system), motivated behavior (mes olimbic system ), learning and m em or y (mesocortical system), and the release of prolactin (tuberoinfundibular system). Dopamine receptors are expressed in the targets of these pathways as well as in dopamine-synthesizing neurons, and they are targets for drugs employed in the treatment of psychotic disorders and Parkinson’s disease. It is also well established that psychostimulants such as cocaine and amphetamine indirectly activate these receptors. To date, molecular cloning has identified five genes that code for dopamine receptor subtypes. These receptors fall into two functionally distinct classes. The D1 class of receptors (composed of the subtypes D1 and D5) couple to stimulatory subsets of heterotrimeric G proteins, and the D2 class of receptors (composed of the subtypes D2, D3, and D4) couple to inhibitory G proteins (1). Since the discovery of these five receptor subtypes, a wealth of information has been obtained about their anatomic distribution, the structure of their genes, and the alternative posttranscriptional processing of some of their encoded pre-mRNAs. The pharmacological properties of recombinant and natively expressed receptors have been determined, and some of the signal transduction pathways activated by these receptors have been Address reprint requests to: Claudia Schmauss, Dept. Psychiatry/Neuroscience, Columbia University and New York State Psychiatric Institute, Box 42, 1051 Riverside Drive, New York, NY 10032 (e-mail: [email protected]).

Volume 6, Number 2, 2000 Copyright © 2000 Sage Publications, Inc. ISSN 1073-8584

identified. Comprehensive reviews of the results of these studies can be found in Refs. 1 to 3. Although some differential distribution of the various subtypes of dopamine receptors is thought to exist, the anatomic distribution of these receptors also shows a substantial regional overlap. Moreover, efforts to identify distinct functional properties of receptor subtypes that belong to the same functional class, and that share a significant degree of primary-structure similarity, have been hampered by the lack of antagonists selective for each of the receptor subtypes. This has complicated the precise assignment of distinct physiological and/or behavioral functions of individual dopamine receptor subtypes. A number of investigators, therefore, either have turned to biochemical approaches to study the expression of dopamine receptor proteins or have employed gene targeting via homologous recombination to study the phenotype of mutant mice that are deficient for the individual dopamine receptors. The gene targeting studies are also complemented by studies that have used antisense oligoribonucleotides to achieve a knockdown of expression of individual dopamine receptors in vivo (for review, see Ref. 4). The present review focuses on the expression and function of dopamine D1, D2, and D3 receptors. It provides an overview of results of studies that are relevant to the unexpected experimental observation of apparently oligomeric forms of these receptor proteins. Furthermore, it discusses results of studies on D1-, D2-, and D3-receptor mutant mice that have aimed at defining their behavioral phenotypes, the neurochemical differentiation of the striatum, and their responses to drugs of

THE NEUROSCIENTIST

127

Fig. 1. The arrangement of transmembrane α-helices of frog rhodopsin determined by Unger et al. (6). The authors used electron cryo-microscopy to study frozen-hydrated two-dimensional crystals of frog rhodopsin. A three-dimensional map resolved all seven α helices and the approximate tilts of these helices could be determined. The results indicate a three-layered structure in which helices 2 and 3 keep helices 6 and 7 separate from helix 4. The figure shows a solid model of the frog rhodopsin map seen from four views that are approximately perpendicular to each other. A, view from helix 2 toward helix 6; B, from helix 5 toward helix 6; C, from helix 6 toward helix 3; D, from helix 1 toward helix 5. The central sections of the seven transmembrane helices are marked with lines. The cytoplasmic side is on the top, and the extracellular side is on the bottom. Reprinted with permission from Nature (UngerVM, Hargrave PA, Baldwin JM, Schertler GFX. Arrangement of rhodopsin transmembrane alpha-helices. Nature 1997;389:203–6), copyright 1997 Macmillan Magazines Ltd.

abuse. (A summary of recent results of studies on mice deficient for D4 and D5 receptors can be found in Ref. 4.)

Dopamine Receptor Proteins: Are They Oligomers? Hydrophobicity analysis of the primary peptide sequences of all cloned dopamine receptors revealed that the most probable structure of these receptors places them in the family of G-protein-coupled receptors (GPCRs). Although the crystallographic structure of a GPCR has yet to be determined, high- and low-resolution electron cryo-microscopy of the structurally similar bacteriorhodopsin (5) and frog rhodopsin (6), respectively, have revealed a tertiary organization of GPCR molecules with seven transmembrane α helices (see Figs. 1 and 2). The current model of the structure of GPCRs places the N-terminus extracellularly. Ligandbinding sites are predicted to be located either between transmembrane helices or within the extracellular domain. The intracellular loops i2 and i3 and the cytoplasmic C-terminus are thought to be crucial for the interaction with G proteins (for review, see Ref. 1). The N-termini of all dopamine receptors are of similar length. The lengths of the i3 loop and the C-terminus, however, distinguish D1-like receptors from D2-like receptors. Receptors of the D1 class characteristically have short i3 loops and long C-termini, whereas members of the D2 class have long i3 loops and short C-termini. These are structural characteristics of receptors that couple, respectively, to stimulatory and inhibitory

128

THE NEUROSCIENTIST

subsets of heterotrimeric G proteins (for review, see Ref. 7). The individual members of the dopamine receptor family display significant sequence similarity within clusters of hydrophobic residues that are thought to constitute the seven transmembrane-spanning domains (TMDs). This sequence similarity is highest for members of the same functional group. D1 and D5 receptors share 80% identity, D2 and D3 receptors share ~75% identity, and D2 and D4 receptors share ~53% identity of transmembrane-spanning peptide sequences (1). The C-termini of all dopamine receptors are rich in serine and threonine residues and contain a cysteine residue that is conserved in all GPCRs and is thought to anchor their cytoplasmic tails to the membrane (8, 9). Until most recently, the transmembrane “allosteric” interactions of GPCRs (ligand binding and G-protein coupling) have been thought to be mediated by a single receptor unit. Recent experimental evidence suggests, however, that GPCR molecules can also form transmembrane oligomers. The first evidence for intramolecular interactions between N- and C-terminally truncated GPCR molecules (“split” receptors) came from studies of Kobilka et al. (10). They found that two independent peptides of the β2-adrenergic receptors, one comprising TMDs I to V, the other comprising TMDs VI and VII, displayed no β2-adrenergic binding when expressed alone in Xenopus oocytes. However, when both peptides were co-expressed, the β2-adrenergic binding profile was restored and agonist stimulation led to activation of adenylyl cyclase, albeit less efficient compared to the corresponding effect mediated by the wildtype receptor.

Dopamine Receptor Expression and Function

Fig. 2. Schematic diagram of the proposed transmembrane topology of D1-like and D2-like dopamine receptors. The shaded horizontal box represents the plasma membrane, and the open vertical boxes represent the putative seven transmembrane-spanning domains. The amino terminus of the receptor protein (N) is localized extracellularly, and the dark lines connecting the transmembrane domains represent the extracellular and cytoplasmic loops. The different lengths of the third cytoplasmic loop (i3) and the cytoplasmic carboxyl terminal tail (C) distinguish D1-like from D2-like receptors.

Evidence for an intermolecular interaction between GPCRs, that is, the molecular basis for receptor dimerization , cam e from studies on chi m er i c muscarinic/adrenergic receptors (11). In these chimera, the C-terminal receptor portion of the m3 muscarinic receptor (which contains TMD VI and VII) was replaced w ith a corresponding region of t he α2-adrenergic receptor and vice versa. When both chimera were expressed alone in transfected COS-7 cells, no specific binding was detected for either a muscarinic or an adrenergic antagonist. Co-transfection of both chimera, however, restored specific binding for both ligands with ligand-binding properties similar to the wildtype receptors. Furthermore, stimulation of these chimeric receptors with a muscarinic receptor agonist led to stimulation of phosphatidylinositol hydrolysis, the

Volume 6, Number 2, 2000

magnitude of which reached ~50% of the corresponding wildtype m3-receptor-mediated effect. Further support for the possible existence of oligomeric forms of GPCRs came from studies of Wregget and Wells (12) who purified cardiac muscarinic M2 receptor proteins that had electrophoretic mobilities suggesting monomeric, trimeric, and tetrameric configurations. Indeed, the authors conclude that the characteristics of ligand binding to the purified receptors indicate cooperativity within a receptor that is, most likely, in a tetrameric configuration. Oligomeric forms of GPCR receptors have been directly observed in several GPCR systems (see 13 and references therein), and they have been frequently observed for dopamine receptor proteins. For example, when a c-myc epitope-tagged human D1 receptor was

THE NEUROSCIENTIST

129

expressed in infected Spodoptera frugiperda (Sf9) insect cells, it bound agonists and antagonists with affinities similar to those known for the D1 receptor. Furthermore, agonist-stimulation of the receptor resulted in activation of adenylyl cyclase, indicating intact G-protein coupling (14). However, when immunoblots of proteins extracted from membranes of these cells were probed with a monoclonal anti-myc antibody, three immunoreactive bands were detected, which migrated at 43 to 50 kDa, 86 to 100 kDa, and above 200 kDa. Similar observations were made for human D2L receptors expressed in Sf9 cells (15). These receptors were found to bind ligands with a pharmacological profile similar to that reported for neuronal and cloned D2 receptors. Furthermore, agonist binding was sensitive to guanyl nucleotide, indicating intact coupling to endogeneous G proteins. A polyclonal anti-D2 antibody recognized at least two D2-like immunoreactivities of 44 and 96 kDa on immunoblots (15). Subsequent studies revealed additional D2-like immunoreactivities of ~160, ~220, and ~350 kDa in D2-infected Sf9 cells (16). Another study used two monoclonal anti-D3 antibodies (IgG/D3 and IgM/D3) to immunoprecipitate D3 receptor proteins expressed in the rodent, monkey, and human brains (17). In rat and monkey brains, three distinct D3 proteins with apparent molecular masses of 45, 90, and 180 kDa were immunoprecipitated (see Fig. 3) and D3 proteins detected in membrane preparations of human motor cortical tissue migrated at 50 and 200 kDa. The above studies describe two phenomena that are characteristic of all higher-molecular weight GPCRs. (1) They are resistant to SDS and reducing agents, and (2) they are always integral multiples of the low molecular weight proteins. One would expect that a random intermolecular interaction between solubilized hydrophopic molecules would result in the formation of large protein aggregates. However, such aggregates would likely be of varying size and intermediate molecular mass and thus not always a multiple of the lowest molecular mass. Furthermore, the higher-molecular-mass proteins do not appear to result from random formations of disulfide bonds as they are resistant to reducing agents. In addition, the large difference in the mass between the low- and high-molecular-weight protein isoforms cannot solely be explained by extensive posttranslational modifications of the core protein. Nimchinsky et al. (17) showed, for example, that, although dopamine receptors are thought to be extensively N-glycosylated, treatment of D 3 immunoprecipitates with N-glycosidase F does not abolish the high-molecular-weight D3 proteins. Thus, it appears that the intermolecular interactions between dopamine receptor molecules result in the formation of stable dimers and tetramers that are integral membrane proteins. If a G PCR m olecule can oligom eriz e, s uch oligomerization should, in principle, not be restricted to homoligomers. Indeed, Nimchinsky et al. (17) found

130

THE NEUROSCIENTIST

Fig. 3. Detection of multimeric dopamine D3-receptor proteins in monkey and rat brain. D3 immunoprecipitate (IP) obtained from monkey basal ganglia (lane 1) and rat prefrontal cortex (lane 2). Proteins were immunoprecipitated with a monoclonal antibody (IgG) raised against a peptide sequence constituting the putative third intracellular domain of the D3 protein. Western blots of these IPs were probed with a monoclonal antibody (IgM) that recognizes the peptide sequence that is also recognized by the IgG antibody. The highest molecular mass of the D3 immunoreactivity (~180 kDa) is thought to be a tetramer, and the intermediate mass of ~85 kDa is thought to be a dimer. Reprinted with permission from the Journal of Biological Chemistry (Nimchinsky EA, Hof PR, Janssen WGM, Morrison JH, Schmauss C. Expression of dopamine D3 receptor dimers and tetramers in brain and in transfected cells. J Biol Chem 1997;272:29229–37), copyright 1997 of the American Society for Biochemistry and Molecular Biology.

that human D3 receptors expressed in rat GH3 cells do not form oligomeric proteins when they are expressed alone. However, when D3-expressing GH3 cells were cotransfected with cDNA encoding the truncated D3-like mRNA D3nf, multimeric D3 immunoreactivities with electrophoretic mobilities similar to those expressed in brain were detected in immunoprecipitates. (In contrast, when D2- and D3-encoded cDNAs were cotransfected, only monomeric D3 immunoreactivity was detected.) D 3nf mRNA results from alternative splicing of D 3 pre-mRNA (18) and encodes a protein that is identical to the D3 protein in the amino-terminal sequence that extends into the amino terminus of the predicted third intracellular loop. D3nf differs from D3 receptors in its length and amino acid sequence of the C-terminus (19), and the D3nf-specific C-terminus is unlikely to have the additional TMDs VI and VII. Immunoprecipitation experiments, performed in conjunction with D3- and D 3nf -specific antibodies, showed that the higher-molecular-weight D3 immunoreactivities found in brain and in D2/D3 double transfectants also contain D3nf immunoreactivity, suggesting that D3 and D3nf proteins form heteroligomers (17).

Dopamine Receptor Expression and Function

Fig. 4. Cellular distribution of D3 and D3nf-immunoreactivities in the rat sensorimotor cortex. In these confocal micrographs, D3 immunoreactivity is pseudocolored green (A and D) and D3nf-immunoreactivity is pseudocolored red (B and E ). An overlay of these images (C and F ) illustrates the region of greatest co-localization (pseudocolored yellow) in the proximal apical dendrite (arrows). The arrowhead indicates a portion of the proximal apical dendrite that is out of the confocal plane. Scale bar = 30 µm. Reprinted with permission from the Journal of Biological Chemistry (Nimchinsky EA, Hof PR, Janssen WGM, Morrison JH, Schmauss C. Expression of dopamine D3 receptor dimers and tetramers in brain and in transfected cells. J Biol Chem 1997;272:29229–37), copyright 1997 of the American Society for Biochemistry and Molecular Biology.

A prerequisite for an intermolecular interaction between D3 and D3nf in vivo is that the expression of both proteins overlaps at the cellular level. Therefore, Nimchinsky et al. (17) examined with confocal microscopy neurons of the rat and monkey neocortex that were stained with D3 and D3nf-specific antibodies. Indeed, a substantial overlap was found in the expression of both proteins, which was highest in proximal regions of apical dendrites of pyramidal-like neurons (see Fig. 4). The observation that D3 and D3nf, but not D2 and D3, proteins form heteroligomers suggests that one requirement for heteroligomeric assembly are intact and identical amino termini (17). This is similar to the results of Wang et al. (20) who showed that, to maintain the ability of an amino-terminal, truncated protein to participate in heteroligomerization with full-length subunits of the nicotinic acetylcholine receptor, it must have an intact amino terminus with (at least) one full TMD. These findings suggested the possibility that the expression of a short amino-terminal, truncated mutant D3 peptide would compete with wildtype D 3 and other native D3-like molecules in the oligomerization process. To test this, Jung et al. (21) immunoprecipitated D3 proteins

Volume 6, Number 2, 2000

extracted from brain of wildtype mice and heterozygous D3 mutants. In these D3-mutant mice, gene targeting via homologous recombination disrupted the coding region of D3 mRNA at sequences constituting the amino terminal part of the second intracellular domain. Hence, in heterozygous D3 mutants, a truncated, short D3 protein constituting peptide sequences of TMDs I to III is coexpressed with the wildtype D3 receptor peptide (21). (Note that this mutant D3 peptide comprises only 130 amino acids, whereas D3nf has 382 amino acids.) Interestingly, the heterozygous D3 mutants do not express the oligomeric D 3 proteins that are typically found in wildtype brains (see Fig. 5). Although small amounts of D3 proteins of higher molecular weight can also be detected in brains of heterozygous D3 mutants, the mass of these molecules is smaller than that of wildtype dimeric (~100 kDa) D3 proteins, suggesting that these higher-molecular-weight species are heteroligomers of full-length and truncated D3 proteins (21; see Fig. 5). These results provide strong support for the existence of D3 receptor oligomers in vivo. For the β2-adrenergic receptor, a putative dimerization motive, located at the C-terminus of TMD VI and

THE NEUROSCIENTIST

131

Fig. 5. D3-receptor protein expression in brains of wild type (+/+) and heterozygous D 3 mutants (+/–). Proteins were extracted from the dorsal striatum, ventral midbrain (VMB), and the remaining brain tissue (-Str.; -VMB) and immunoprecipitated as described in Figure 3. Note that both wild type and +/– D3 mutants do not express D3 oligomers in the VMB. In the other brain regions of +/– D3 mutants, however, wildtype D3 oligomers are absent and the position of the band representing abnormal D3 oligomers is indicated by an arrowhead. Reprinted from Neurocience (Jung M-Y, Skryabin BV, Arai M, Abbondanzo S, Fu D, Brosius J, et al. Potentiation of the D2 mutant motor phenotype in mice lacking dopamine D2 and D3 receptors. Neuroscience 1999;91:911–24); copyright 1999, with permission from Elsevier Science.

containing a leucine and 2 glycine residues, has been shown to be an essential part of an interface for receptor dimerization. A peptide derived from this region inhibits both receptor dimerization and receptor activation (13). Dopamine D 1 receptors, however, lack such a dimerization motif, and incubation of D1 receptors with a peptide derived from TMD VI did not inhibit dimerization (22). However, this peptide completely abolished D1-receptor function in an irreversible and dose-dependent manner. What is the functional significance of dopamine receptor oligomerization? Hebert et al. (13) suggested a dynamic equilibrium of monomers and dimers of the β2-adrenergic receptor, which would imply that, if there is a role for dimers, the monomer:dimer equilibrium should be affected by ligand binding. Indeed, agonist binding stabilizes the dimeric form of the β2-adrenergic receptor whereas inverse agonists favor the monomeric receptor isoform (13). For dopamine receptors, a detailed study on the ability of monomers and oligomers to bind agonists and antagonists is still missing. However, Zawarynski et al. (16) observed a different binding pattern of spiperone and benzamine congeners to D2 receptors of lower and higher molecular weight, suggesting that different ligands have different affinities to monomeric and oligomeric forms of receptors. Most

132

THE NEUROSCIENTIST

intriguing, mice lacking D3 oligomers (i.e., heterozygous mutants as described above) have lost almost all [125I]iodosulpiride binding to D3 receptors, suggesting that wildtype oligomers provide most of the D3-receptor [125I]iodosulpiride binding sites (23). In summary, intra- and intermolecular interactions between GPCR molecules have been observed in transfected cells. Moreover, results from immunoblots and immunoprecipitation experiments suggest the existence of homo- and heteroligomeric forms of dopamine receptors and other GPCRs in vivo. A short aminoterminal peptide derived from the D3 receptor was found to compete with other full-length D3 molecules in the dimerization process and thus to prevent the formation of wildtype oligomers. Another peptide, derived from sequences constituting TMD VI of the D1 receptor, does not alter the ability of the receptor molecule to dimerize but inhibits the activation of the receptor. The interactions between two or more hydrophobic GPCR molecules appear to be resistant to SDS and reducing agents. The precise nature of their intermolecular interaction, however, is still unknown. Although oligomeric forms of GPCRs have now been observed in many different GPCR systems, the monomeric form of all GPCR contains both ligand-binding and G-protein coupling domains. The functional significance of oligomeric GPCRs is therefore unclear. Information about the crystal structure of a ligand-bound GPCR is clearly needed but will be difficult to obtain within the immediate future. However, other methods to probe single molecules have now been developed that should allow us to study the intermolecular interaction between GPCR molecules. The recent advances of single molecule detection by laser-induced fluorescence enable high co-localization accuracy, in particular, when two fluorophores interact by fluorescence resonance energy transfer (FRET; for review, see 24). Thus, if two potentially interacting molecules (D3 and D3nf, for example) are tagged with two such flurophores and expressed in cells of a primary neuronal culture, one should be able to measure the distance between the two fluorophores on a 2- to 8-nm scale and also detect dynamic changes in the distance between them in the presence and absence of different ligands. If such studies verify the existence of oligomeric assemblies of dopamine receptors, it will be challenging to determine the functional significance of this oligomerization and this will require a detailed understanding of the distinct functional properties of the individual dopamine receptor subtypes. As outlined above, molecular, anatomical, and pharmacological studies have already provided a wealth of information about the expression and function of these receptors (1–3). The following section discusses results of recent studies on knockout mice that have built upon these findings. In view of the present section, it is also emphasized that, in some cases, the gene-targeting strategies resulted in the generation of mutant mice that express N-terminal (truncated) dopamine receptor proteins. Although such mutant proteins

Dopamine Receptor Expression and Function

Table 1. Locomotor phenotypes of mice lacking dopamine D1, D2, and D3 receptors D1 (–/–)

D2 (–/–)

D3 (–/–)

D2/D3 (–/–)

Postnatal development

Retarded postnatal growth (25, 26)

Retarded postnatal growth Normal postnatal growth More severely retarded (21, 34); normal postnatal (21, 23, 45) postnatal growth when development (35) compared to D2 single mutants (21)

Body size

Decreased (25, 26)

Decreased (21, 34); normal (35)

normal (21, 23, 45)

Decrease in basal locomotor activity (bradykinesia) and postural abnormalities (21, 34)

Rapidly habituating Significant potentiation of increase in locomotor the D2-mutant locomotor activity in response to a phenotype (21) novel environment (23, 45)

Locomotor activity Increased basal locomotor activity and increased locomotor response to a novel environment (25, 28) Normal basal locomotor activity (26) and impairment in the initiation of spontaneous locomotor activity and/or locomotor responses to external stimuli (29)

Reduced spontaneous Normal locomotor activity locomotion and when evaluated during decreased initiation of longer test periods (21) movements, no postural abnormalities (37)

are likely to be nonfunctional when expressed alone, they may nevertheless be capable of forming heteroligomers in vivo. Thus, the question arises whether one of the reasons for the phenotypic differences between these mutants and null-mutants is the expression of truncated dopamine receptors. Studies on Dopamine Receptor Function in Knockout Mice

D1-Receptor Knockout The first successful generation of homologously recombinant mice lacking a dopamine receptor was reported by Xu et al. (25) who generated D1 mutants in which the majority of the intronless part of the gene comprising the entire open reading frame of the D1 receptor was removed. Shortly thereafter, another successful generation of D1 mutant mice was reported by Drago et al. (26) who used a targeting vector that, upon homologous recombination, disrupted the D1 gene at sequences encoding the putative TMD V. D1-deficient mice are on average 30% smaller than their wildtype littermates, and their postnatal development is retarded (25, 26). Adult D1 mutants appear to be healthy. The analysis of Nissl-stained brain sections confirmed that, despite the smaller size of the brain (26), the general anatomy of brain structures is maintained (25). Moreover, the expression of the rate-limiting enzyme for dopamine synthesis, tyrosine hydroxylase (TH), as well as the dense nigrostriatal dopaminergic innervation are unaltered (25). However, a neurochemical hallmark

Volume 6, Number 2, 2000

Further decrease in body size compared to D2 single mutants (21)

of the striatal phenotype of these mutants is an almost complete lack of dynorphin-positive cell clusters in striosomes and a down-regulation of dynorphin elsewhere in the striatum (25, 27). These findings confirm that D1 receptors, which are primarily expressed in dynorphin-expressing striatal neurons, tightly control the expression of the dynorphin gene in these neurons. Furthermore, D1 mutants express lower levels of the neuropeptide substance P, which is coexpressed with dynorphin in striatal neurons (25, 26). The expression of striatal enkephalin mRNA, however, which is thought to be tightly regulated by D2 receptors, is normal in D1 mutants (26). The D1 mutants generated by Xu et al. (25, 28) show an increased basal locomotor activity and an increased locomotor activity in response to a novel environment. In contrast, D1 mutants generated by Drago et al. (26) are somewhat less active than their wildtype littermates when examined in the open field test (see Table 1). A more detailed behavioral analysis of these mutants even suggests an impairment in the initiation of spontaneous locomotor activity and/or response to external stimuli (29). However, following an injection of saline and placement into a novel test environment, these mice have also been reported to display an increased locomotor activity (30). Cromwell et al. (31) also found that, although these mutants show no impairment in the generation of movement, they show an impairment of the sequential integrity of serially ordered pattern of movements that occur during grooming. An inherent problem with the interpretation of complex behavioral phenotypes of mutants is the possibly different contribution of background genes derived from

THE NEUROSCIENTIST

133

Table 2. Responses of D1-, D2-, and D3-receptor-deficient mice to drugs of abuse D1 (–/–) Locomotor responses

Lack of cocaineinduced hyperlocomotion (28, 30, 33)

D2 (–/–)

D3 (–/–)

Unaltered morphineinduced hyperlocomotion (42)

Increased locomotor activity in response to low doses of cocaine (45)

Decreased sensitivity to locomotor depressant and ataxic effects of alcohol (43) Motivational behavior

Normal rewarding/ reinforcing effects of cocaine (30)

Blunted responses to rewarding/reinforcing effects of opioids (42)

134

THE NEUROSCIENTIST

?

?

Increased responsiveness to the rewarding actions of low doses of amphetamine (45)

Reduced intake and Reduced intake and preference for alcohol preference for alcohol (44) (43)

the parental strain that are either linked to the mutation or located at other chromosomal loci (32). Different sources of 129Sv-derived embryonic stem (ES) cells used to generate chimeric mice, genetic background differences of hybrid 129Sv × C57Bl/6 mice, and the different targeting strategies used (genetic null-mutation vs. disruption of genes at positions closer to the 3′ end of the open reading frame) might influence the phenotype of the resulting knockout mice. Thus, it is possible that such experimental differences contribute to the above-described differences in the motor phenotypes of the two different types of D1 mutants. Dopamine receptors have also been implicated in the behavioral responses to drugs of abuse. Stimulation of D1 receptors is thought to affect addictive behavior and movement control, effects that depend on the striatal dopaminergic system. Despite the somewhat different findings regarding the spontaneous locomotor phenotype of D1 mutants, the different laboratories clearly found that the absence of D1 receptors leads to a complete lack of the increased locomotor responses following cocaine administration (28, 30, 33) (see Table 2). These results confirmed that D1 receptors play a significant role in the locomotor activating effects of psychostimulants. In place-preference test experiments, however, it was found that the expression of D1 receptors does not affect the rewarding and reinforcing effects of cocaine (30). Another well-studied molecular response to the acute administration of psychostimulants is the induction of expression of immediate-early genes (IEG) that lead to transcriptional activation of AP1-promotor-containing (late-response) genes whose expression, in turn, results in long-term adaptive changes in mature neurons. In D1 mutants, psychostimulants fail to induce the expression of the IEGs c-fos, JunB, and zif 268 (27, 33). Altogether, these data confirmed that expression of D1 receptors is essential for the induction of expression

D2/D3 (–/–)

?

?

of IEGs and the increased locomotor activity that occur i n r es pons e t o t he acut e adm i ni s t r at i o n o f psychostimulants. However, D1 receptors apparently do not mediate the rewarding and reinforcing effects of psychostimulants.

D2-Receptor Knockout The first successful generation of mutant mice carrying a null mutation of the D2 gene was reported by Baik et al. (34). Subsequently, two additional groups generated D2 mutant mice. One group used the same targeting strategy to generate a null mutation of the D2 gene (21), and the other deleted the C-terminal region of the open reading frame of the D2 gene beginning at sequences that code for the predicted third intracellular loop of the receptor (35). Homozygous mutants generated with the targeting vectors that resulted in a null mutation of the D2 gene are viable, their body weight is 15% lower than that of wildtype mice, their postnatal growth is delayed, and their fertility is greatly reduced (21, 34). Analysis of Nissl-stained sections demonstrated that brain structures develop normally. Furthermore, the staining for TH immunoreactivity and acetylcholine esterase activity are normal in the dorsal striatum of these mutants (36). However, a significantly increased expression of enkephalin mRNA was found in the dorsal striatum (21, 34). In contrast to D1 mutants, however, the levels of dynorphin and substance P mRNA are unaltered (34). These results illustrate the crucial role of D2 receptors in the regulation of transcription of the enkephalin, but not the dynorphin, gene. D2 mutant mice generated with a targeting vector that results in a null-mutation of the D2 gene have a very characteristic locomotor phenotype, which includes a severe bradykinesia and postural abnormalities such as paw flattening and sprawling of hindlegs (see Fig. 6). Open-field experiments (34) and experiments using

Dopamine Receptor Expression and Function

Fig. 6. Wild type (WT), homozygous D2 mutants (D2), and homozygous D2/D3 double mutants (DM) at postnatal age P30. Note the smaller body size and the postural abnormalities of D2 and D2/D3 mutants that are described by Jung et al. (21). Details of the mutant phenotypes are summarized in Table 1.

locomotor activity boxes (21) revealed a significant reduction in the spontaneous locomotor activity. Rotarod and ring test experiments further illustrated severe differences in movement coordination and times spent in immobility (catalepsy) (34). Altogether, the motor phenotypes of these mutants closely resemble those precipitated by D2-receptor antagonists. Interestingly, D2 mutant mice generated by Kelly et al. (35) with a targeting vector that eliminates the C-terminus of the D2 receptor do not develop the severe locomotor phenotypes described above (see Table 1). In addition, whereas Jung et al. (21) found significantly higher levels of the dopamine metabolites DOPAC and HVA in the striatum of homozygous D2 mutants, the corresponding levels of these metabolites are not significantly altered in the mutants generated by Kelly et al. (37). In fact, the mice generated by Kelly et al. develop normally and are fertile, do not show postural abnormalities, and the net reduction of locomotor activity measured for these mutants on either a 129Sv or a C57Bl/6 background (37) is much less than the ones reported by Baik et al. (34) and Jung et al. (21) for mice with a hybrid 129Sv × C57Bl/6 background. This finding makes unlikely the possibility that differences in the genetic backgrounds between the various D2 mutants

Volume 6, Number 2, 2000

solely account for the rather striking differences in their motor phenotypes. Perhaps one should also consider that the different D2 mutant mice that are compared here resulted from different gene-targeting strategies. As noted above, for D1 knockout mice, similarly different targeting strategies have also resulted in similar differences in the locomotor phenotypes. In fact, given the very minor locomotor phenotype of the mutants generated by Kelly et al. (35) the possibility must be considered that some truncated GPCR proteins can participate in functional oligomeric assemblies. As outlined in the previous section, inter- and intramolecular interactions between GPCR molecules have been demonstrated. Furthermore, N-terminal peptides of GPCRs that contain 2 to 5 TMDs have been shown to represent independent folding units, and they are targeted to the plasma membrane where they can interact with C-terminal folding units to form functional protein complexes (38). Thus, it would be interesting to know whether a truncated D2 protein comprising TMDs I through V is expressed in brains of the mutants generated by Kelly et al. (35) and whether it interacts with other dopamine receptor– derived molecules. Furthermore, it would be of interest to stest in more detail the radioligand binding profile of a variety of different D2-receptor agonists and antagonists

THE NEUROSCIENTIST

135

in brain tissues from these mutants. These D2 mutants could potentially provide a suitable tool for in vivo studies on oligomeric assemblies of dopamine receptors. Other studies have used D2 mutant mice to study the role of dopamine in the control of pituitary function. D2 receptors are highly expressed in lactotrophs and melanotrophs of the anterior and intermediate lobes, respectively, of the pituitary. Studies on D2 mutant mice demonstrated that dopamine has antiproliferative effects on the developing pituitary gland and a negative control over prolactin synthesis. The absence of D2 receptors leads to a progressive increase in the number of lactotrophs (35, 39), which can lead to tumors in the aged animal (39), and prolactin levels are chronically elevated in D2 mutant mice (35, 39). Despite the suggested presence of D3 and D4 receptors in the pituitary, these effects are mediated solely via D2 receptors (39). Similarly, results of studies on D2 mutant mice suggest that D2-like autoreceptor functions are also solely mediated by D2 receptors. Although both D2 and D3 receptors are expressed in the substantia nigra and the ventral tegmental area (VTA), electrophysiological studies on D2 mutants detected a complete loss of inhibition of the spontaneous firing of dopaminergic neurons in these regions upon application of dopamine and quinpirole (40). Furthermore, studies on [3H]-dopamine-loaded striatal synaptosomes showed that D2 receptors, but not D3 receptors, are involved in the autoreceptor-mediated inhibition of evoked dopamine release (41). Finally, how do D2 mutants respond to drugs of abuse? Whereas D1 receptors have been shown to be essential mediators of locomotor responses, but appear not to mediate responses to the rewarding and reinforcing properties of psychostimulants, the role of D2 receptors in mediating behavioral responses to opioid drugs seems to be the opposite (42) (see Table 2). Acute administration of morphine elicits locomotor hyperactivity in both wildtype and D2 mutant mice, and their naloxone-precipitated opioid-withdrawal symptoms are indistinguishable. However, in place preference tests, D2 mutants did not show a preference for the drug-associated compartment. This motivational deficit indicates that the expression of responses to the rewarding properties of opioid drugs requires expression of D 2 receptors. Mice lacking D2 receptors also show a reduced preference for alcohol in a free-choice test design, and they are less sensitive to the locomotor depressant and ataxic effects of ethanol (43). However, alcohol intake and preference are also reduced in mice lacking D1 receptors and the study of El-Ghundi et al. (44) suggests that D1 receptors play a greater role in alcohol seeking behavior than D2 receptors. In any case, it appears that D1 and D2 receptors modulate both the locomotor response and the motivational response to alcohol (see Table 2).

D3-Receptor Knockout The first report on a successful knockout of the D3 receptor gene was published by Accili et al. (23). The au-

136

THE NEUROSCIENTIST

thors disrupted the D3 gene at sequences that code for the putative second intracellular loop of the receptor. Two other D3 knockouts were subsequently reported by Xu et al. (45) who generated a null-mutation of the D3 gene, and by Jung et al. (21) who, similar to the approach of Accili et al., disrupted the D3 gene at sequences coding for the second intracellular loop. Compared to D1- and D2-receptor-deficient mice, D3 mutants show the most normal phenotype. Homozygous D3 mutants develop normally and are fertile (see Table 1). In Nissl-stained sections, the structure of the striatum appears normal (36), and the expression of TH immunoreactivity, enkephalin mRNA, D 1 -receptor ligand-binding sites, and D 2 receptor mRNA and ligand-binding sites are unaltered in these mutants (21, 45). Furthermore, tissue levels of striatal dopamine, DOPAC and HVA are indistinguishable from corresponding wildtype levels (21). Extracellular single unit recordings from neurons of the ventral midbrain of D3 mutants, as well as measurements of striatal dopamine synthesis (46), confirmed that D3 receptors do not significantly contribute to D2-like autoreceptor functions (see 40, 41). In in vivo microdialysis studies, however, higher basal levels of extracellular dopamine were detected in the ventral striatum of D3 mutant mice, and it is suggested that D3 receptors may participate in regulating dopamine release via postsynaptically activated short-loop feedback modulation (46). Accili et al. (23) and Xu et al. (45) detected locomotor hyperactivity of their mutants during the first 15 min of exposure to the test environment. However, this hyperactivity habituates rapidly, and locomotor activity measured 15 to 20 min after placement into locomotor activity boxes is indistinguishable from wildtype (45). Indeed, when Jung et al. (21) compared 30-min values for the horizontal locomotor activity and the total distance traveled, no significant difference was found between wildtype and D3 mutants (see Table 1). Further experiments on the mutants generated by Accili et al. (23) revealed a reduced thigmotaxis in the open field and an altered behavior in the elevated maze, suggesting that some of the initial hyperactivity seen in response to the exposure to a novel environment can be attributed to a reduced level of anxiety of D3 mutants (47). However, similar experiments performed by Xu et al. (45) did not confirm these findings. Their studies revealed, however, a distinct role for D3 receptors in the modulation of locomotor activity elicited by coactivation of D1 and D2 receptors. The authors found that stimulation of D3 receptors leads to a suppression of the locomotor activity induced by simultaneous D1- and D2-receptor activation and also blunts the magnitude of the behavioral sensitization to low (positively rewarding) doses of psychostimulants (see Table 2). Thus, one role of D3 receptors is to inhibit the cooperative locomotor and motivational effects of postsynaptic D1- and D2-receptor stimulation. Furthermore, in the modulation of c-fos responses to either direct D1-agonist stimulation or indirect stimulation of dopamine receptors by amphetamine,

Dopamine Receptor Expression and Function

both D2 and D3 receptors were found to participate cooperatively to enable maximum c-fos responses (Jung and Schmauss, unpublished observation). The available results, therefore, suggest that D3 receptors act as modulators of motor responses, c-fos responses, and reward-related behavior resulting either from the concurrent stimulation of D1 and D2 receptors or from the stimulation of D1 receptors alone. Finally, Jung et al. (21) generated mice lacking both D2 and D3 receptors. Interestingly, the double mutants show a motor phenotype that qualitatively resembles that of D2 mutants, but that is significantly more severe (see Fig. 6). Furthermore, the increased levels of striatal DOPAC and HVA found in D2 mutants are further increased in D2/D3 double mutants in which they reach levels previously shown to result from chronic blockade of D2-like receptors by neuroleptics (see Ref. 21). These results suggest that some of the functional properties of D3 receptors resemble those of D2 receptors but that these functions of D3 receptors remain masked when the abundant D2 receptor is expressed. They also suggest that D3 receptors can compensate for some of the functional consequences of a D2 knockout. Indeed, Jung et al. found that in wildtype mice, the expression of the tetrameric form of the D3 protein is gradually downregulated during later stages of postnatal development. In brains of D2 mutants, however, this downregulation does not occur, thus resulting in higher levels of D3 proteins in maturing D2 mutants compared to their wildtype littermates (21). In summary, studies on D1, D2, and D3 mutants have begun to provide new insights into the in vivo function of these receptors with regard to their role in locomotion and autoreceptor function and their role in mediating the locomotor and rewarding effects of drugs of abuse. Because all these mutants were generated with the constitutive knockout technique, it remains to be tested which of the identified phenotypes truly relate to the lack of the receptor and which have resulted from developmental adaptations to the loss of receptors. Future studies on inducible and tissue-specific knockouts may clarify this issue. The current knockout mice are also of enormous practical value as they allow the first stringent tests of the selectivity of ligands for the various subtypes of dopamine receptors. Two studies (46, 48) already have used D3 mutant mice to test whether any of the currently available D3-“selective” ligands do indeed possess a selectivity for D3 receptors in vivo. Both studies found that none of the ligands selectively activated or blocked D3 receptors. Thus, although the genetic studies have built on, rather than overturned, previous pharmacological findings, the increasing availability of mutant mice should continue to aid investigations into the function of dopamine receptors in the central nervous system.

Volume 6, Number 2, 2000

References 1. 2. 3. 4. 5.

6. 7. 8.

9. 10.

11.

12. 13.

14.

15. 16. 17.

18. 19.

20. 21.

Missale C, Nash SR, Robinson SW, Jaber M, Caron MG. Dopamine receptors: from structure to function. Physiol Rev 1998;78:189–225. Levant B. The D3 dopamine receptor: neurobiology and potential clinical relevance. Pharmacol Rev 1997;49:231–52. Shafer RA, Levant B. The D3 dopamine receptor in cellular and organismal function. Psychopharmacology 1998;135:1–16. Sibley DR. New insight into dopaminergic receptor function using antisense and genetically altered animals. Ann Rev Pharmacol Toxicol 1999;39:313–41. Henderson R, Baldwin JM, Ceska TA, Zemlin F, Beckmann E, Downing KH. Model for the structure of bacteriorhodopsin based on high-resolution electron cryo-microscopy. J Mol Biol 1990;213:899–929. Unger VM, Hargrave PA, Baldwin LM, Schertler GF. Arrangement of rhodopsin transmembrane alpha-helices. Nature 1997;389:203–6. O’Dowd BF. Structure of dopamine receptors. J Neurochem 1993;60:804–16. O’Dowd BF, Hnatowich M, Caron MG, Lefkowitz RJ, Bovier RJ. Palmytoylation of the human β2 adrenergic receptor: mutation of CYS341 in the carboxy tail leads to an uncoupled, non-palmitoylated form of the receptor. J Biol Chem 1989;264:7564–9. Ovchinnikov V, Abdulaev N, Bogachuk A. Two adjacent cysteine residues in the C-terminal cytoplasmic fragment of bovine rhodopsin are palmitoylated. FEBS Lett 1988;230:1–5. Kobilka BK, Kobilka TS, Daniel K, Regan JW, Caron MG, Lefkowitz RJ. Chimeric α2-,β2-adrenergic receptors: delineation of domains involved in effector coupling and ligand binding specificity. Science 1988;240:1310–16. Maggio R, Vogel Z, Wess J. Coexpression studies with mutant muscarinic / adrenergic receptors provide evidence for intermolecular “cross-talk” between G-protein-coupled receptors. Proc Natl Acad Sci U S A 1993;90:3103–7. Wreggett KA, Wells W. Cooperativity manifest in the binding properties of purified cardia muscarinic receptors. J Biol Chem 1995;270:22488–99. Hebert TE, Moffett S, Morello J-P, Loisel TP, Bichet DG, Barret C, et al. A peptide derived from a β 2 -adrenergic receptor transmembrane domain inhibits both receptor dimerization and activation. J Biol Chem 1996;271:16384–92. Ng GYK, Mouillac B, George SR, Caron M, Dennis M, Bovier M, et al. Desensitization, phosphorylation and palmitoylation of t he hum a n dopa m i ne D 1 r e c e pt or. E ur J P h a r m a c o l 1994;267:7–19. Ng GYK, O’Dowd BF, Caron M, Dennis M, Brann MR, George SR. Phosphorylation and palmitoylation of the human D2l dopamine receptor in Sf9 cells. J Neurochem 1994;63:1589–95. Zawarynski P, Tallerico T, Seeman P, Lee SP, O’Dowd BF, George SR. Dopamine D2 receptor dimers in human and rat brain. FEBS Lett 1998;441:383–6. Nimchinsky EA, Hof PR, Janssen WGM, Morrison JH, Schmauss C. Expression of dopamine D3 receptor dimers and tetramers in brain and in transfected cells. J Biol Chem 1997;272:29229–37. Schmauss C. Enhanced cleavage of an atypical intron of dopamine D3-receptor pre-mRNA in chronic schizophrenia. J Neurosci 1996;16:7902–9. Liu K, Bergson C, Levenson R, Schmauss C. On the origin of mRNA encoding the truncated dopamine D3-type receptor D3nf and detection of D3nf-like immunoreactivity in human brain. J Biol Chem 1994;269:29220–6. Wang Z-Z, Hardy SF, Hall ZW. Assembly of the nicotinic acetylcholine receptor. J Biol Chem 1996;271:27575–84. Jung M-Y, Skryabin BV, Arai M, Abbondanzo S, Fu D, Brosius J, et al. Potentiation of the D2-mutant motor phenotype in mice

THE NEUROSCIENTIST

137

22.

23.

24. 25.

26.

27.

28.

29. 30. 31. 32.

33.

34. 35.

138

lacking dopamine D2 and D3 receptors. Neuroscience 1999;91:911–24. George SR, Lee SP, Varghese G, Zeman PR, Seeman P, Ng GYK, et al. A transmembrane domain-derived peptide inhibits D1 dopamine receptor function without affecting receptor oligomerization. J Biol Chem 1998;273:30244–8. Accili D, Fishburn CS, Drago J, Steiner H, Lachowicz JE, Park B-H, et al. A targeted mutation of the D3 dopamine receptor gene is associated with hyperactivity in mice. Proc Natl Acad Sci U S A 1996;93:1945–9. Weiss S. Fluorescence spectroscopy of single biomolecules. Science 1999;283:1676–83. Xu M, Moratalla R, Gold LH, Hiroi N, Koob GF, Graybiel AM, et al. Dopamine D1 receptor mutant mice are deficient in striatal expression of dynorphin and in dopamine-mediated behavioral responses. Cell 1994;79:729–42. Drago J, Gerfen CR, Lachowicz JE, Steiner H, Hollon TR, Love PE, et al. Altered striatal function in a mutant mouse lacking D1A dopamine receptors. Proc Natl Acad Sci U S A 1994;91:12564–8. Moratalla R, Xu M, Tonegawa S, Graybiel AM. Cellular responses to psychostimulant and neuroleptic drugs are abnormal in mice lacking the D1 dopamine receptor. Proc Natl Acad Sci U S A 1996;93:14928–33. Xu M, Hu X-T, Cooper DC, Moratalla R, Graybiel AM, White FJ, et al. Elimination of cocaine-induced hyperactivity and dopamine-mediated neurophysiological effects in dopamine D1 receptor mutant mice. Cell 1994;79:945–55. Smith DR, Striplin CD, Geller AM, Mailman RB, Drago J, Lawler CP, et al. Behavioral assessment of mice lacking D1A dopamine receptors. Neuroscience 1998;86:135–46. Miner L, Drago J, Chamberlain PM, Donovan D, Uhl GR. Retained cocaine conditioned place preference in D1 receptor deficient mice. Neuroreport 1995;6:2314–16. Cromwell HC, Berridge KC, Drago J, Levine MS. Action sequencing is impaired in D 1 A-deficient mutant mice. Eur J Neurosci 1998;10:2426–32. Crawley JN, Belknap JK, Collins A, Crabbe JC, Frankel W, Henderson N, et al. Behavioral phenotypes of inbred strains: implications and recommendations for molecular studies. Neuropharmacology 1997;132:107–24. Drago J, Gerfen CR, Westphal H, Steiner H. D1-dopamine receptor-deficient mouse: cocaine-induced regulation of immediate-early gene and substance P expression in the striatum. Neuroscience 1996;74:813–23. Baik J-H, Picetti R, Dalardi A, Thirlet G, Dierich A, Depaulis A, et al. Parkinsonian-like locomotor impairment in mice lacking dopamine D2 receptors. Nature 1995;377:424–8. Kelly MA, Rubinstein M, Asa SL, Zhang G, Saez C, Bunzow JR, et al. Pituitary lactotroph hyperplasia and chronic

THE NEUROSCIENTIST

hyperprolactinemia in dopamine D2 receptor-deficient mice. Neuron 1997;19:103–13. 36. Jung M-Y, Hof PR, Schmauss C. Expression of calbindin-D28K immunoreactivity in the dorsal striatum of mice lacking dopamine D2, D3, and D2/D3 receptors. Soc Neurosci Abstr 1999;25:151–8. 37. Ke l l y M A , R ubi ns t e i n M , P hi l l i ps T J , L e s s ov C N , Burkhart-Kasch S, Zhang, et al. Locomotor activity in D2 dopamine receptor-deficient mice is determined by gene dosage, genetic background, and developmental adaptations. J Neurosci 1998;18:3470–9. 38. Schöneberg T, Liu J, Wess J. Plasma membrane localization and functional rescue of truncated forms of a G protein-coupled receptors. J Biol Chem 1995;270:18000–6. 39. Saiardi A, Bozzi Y, Baik J-H, Borelli E. Antiproliferative role of dopamine: loss of D2 receptors causes hormonal dysfunction and pituitary hyperplasia. Neuron 1997;19:115–26. 40. Mercuri NB, Saiardi A, Bonci A, Picetti R, Calabresi P, Bernadi G, et al. Loss of autoreceptor function in dopaminergic neurons from dopamine D2 receptor deficient mice. Neuroscience 1997;79:323–7. 41. L’hirondel ML, Cheramy A, Godeheu G, Artaud F, Saiardi A, Borreli E, et al. Lack of autoreceptor-mediated inhibitory control of dopamine release in striatal synaptosomes of D2 receptor-deficient mice. Brain Res 1998;792:253–62. 42. Maldonado R, Saiardi A, Valverde O, Samad TA, Roques BP, Borreli E. Absence of opiate rewarding effects in mice lacking dopamine D2 and D3 receptors. Nature 1997;388:586–9. 43. Phillips T, Brown KJ, Burkhart-Kasch S, Wenger CD, Kelly MA, Rubinstein M, et al. Alcohol preference and sensitivity are markedly reduced in mice lacking dopamine D2 receptors. Nature Neurosci 1998;1:610–15. 44. El-Ghundi M, George SR, Drago J, Fletcher PJ, Fan T, Nguyen T, et al. Disruption of dopamine D1 receptor gene expression att e nua t e s a l c ohol - s e e ki ng be havi or. E ur J P h a r m a c o l 1998;353:149–58. 45. Xu M, Koeltzow TE, Santiago GT, Moratalla R, Cooper DC, Hu X-T, et al. Dopamine D3 receptor mutant mice exhibit increased behavioral sensitivity to concurrent stimulation of D1 and D2 receptors. Neuron 1997;19:837–48. 46. Koeltzow TE, Xu M, Cooper DC, Hu X-T, Tonegawa S, Wolf ME, et al. Alterations in dopamine release but not dopamine autoreceptor function in dopamine D3 receptor mutant mice. J Neurosci 1998;18:2231–8. 47. Steiner H, Fuchs S, Accili D. D3 dopamine receptor-deficient mouse: evidence for reduced anxiety. Physiol Behav 1998;63: 137–41. 48. Xu MX, Koeltzow TE, Cooper DC, Tonegawa S, White FJ. Dopamine D3 receptor mutant and wild-type mice exhibit identical responses to putative D3 receptor-selective agonists and antagonists. Synapse 1999;31:210–15.

Dopamine Receptor Expression and Function