A single dose of methamphetamine leads to a long term reversal of the blunted dopamine D1 receptor-mediated neocortical c-fos responses in mice deficient for D2 and D3 receptors.

Dopamine D(1) receptors play an essential role in the induction of expression of the immediate-early gene c-fos in response to pharmacological stimuli. In the forebrain of wild-type mice, administration of a D(1) receptor agonist leads to c-fos mRNA expression levels that are substantially higher than corresponding levels expressed after indirect stimulation of dopamine receptors with methamphetamine. In mice deficient for D(2) and D(3) receptors, c-fos mRNA levels expressed in response to D(1) agonist administration are significantly blunted. However, a single dose of methamphetamine (5 mg/kg) leads to a long lasting reversal of the blunted c-fos responses in these mutants. In the forebrain, this reversal is restricted to the neocortex. Moreover, methamphetamine also enhances c-fos expression levels in preadolescent wild-type mice that normally express low c-fos mRNA in response to D(1) agonist stimulation. Thus, a single dose of methamphetamine leads to a long term increase in D(1) receptor-dependent c-fos responses in brains with either low (preadolescent mice) or blunted (adult D(2) and D(3) mutant mice) c-fos expression levels. A similar long term reversal of the blunted c-fos responses is achieved with a single dose of a full D(1) agonist. These results indicate that the constitutive inactivation of D(2) and D(3) receptors leads to a decrease in agonist-promoted D(1) receptor activity that can be reversed by intermittent agonist stimulation.

Dopamine D 1 receptors play an essential role in the induction of expression of the immediate-early gene cfos in response to pharmacological stimuli. In the forebrain of wild-type mice, administration of a D 1 receptor agonist leads to c-fos mRNA expression levels that are substantially higher than corresponding levels expressed after indirect stimulation of dopamine receptors with methamphetamine. In mice deficient for D 2 and D 3 receptors, c-fos mRNA levels expressed in response to D 1 agonist administration are significantly blunted. However, a single dose of methamphetamine (5 mg/kg) leads to a long lasting reversal of the blunted c-fos responses in these mutants. In the forebrain, this reversal is restricted to the neocortex. Moreover, methamphetamine also enhances c-fos expression levels in preadolescent wild-type mice that normally express low c-fos mRNA in response to D 1 agonist stimulation. Thus, a single dose of methamphetamine leads to a long term increase in D 1 receptor-dependent c-fos responses in brains with either low (preadolescent mice) or blunted (adult D 2 and D 3 mutant mice) c-fos expression levels. A similar long term reversal of the blunted c-fos responses is achieved with a single dose of a full D 1 agonist. These results indicate that the constitutive inactivation of D 2 and D 3 receptors leads to a decrease in agonist-promoted D 1 receptor activity that can be reversed by intermittent agonist stimulation.
The induction of expression of the immediate-early gene c-fos, a gene with low base-line levels of expression in brain, is a well established and powerful tool for examining neuronal circuits that are activated biochemically in response to a variety of different stimuli. For example, studies on the induction of c-fos in response to an acute administration of drugs of abuse identified a common neuroanatomical pattern of expression (for review see Ref. 1), and it has been shown that the subchronic administration of such drugs (which is associated with a progressive sensitization of neuronal systems) leads to distinct alterations in the anatomic pattern of c-fos expression (2). In addition, studies on mutant mice have demonstrated that the expression of dopamine D 1 receptors is essential for the control of immediate-early gene expression by psychomotor stimulants, such as cocaine and amphetamine (3). Although D 1 receptors are essential for the induction of c-fos expression in response to psychostimulants, the magnitude of the D 1 -dependent c-fos expression levels appears to be modulated by both D 2 and D 3 receptors. For example, the study of Moratalla et al. (3) identified anatomically restricted alterations in c-fos responses to haloperidol, a neuroleptic drug that blocks the D 2 -like dopamine receptor subtypes D 2 and D 3 , and other studies on mice deficient for D 3 receptors revealed blunted c-fos responses to D 1 agonist stimulation. These responses were even further reduced when D 3 mutants were pretreated with the D 2 -like antagonist eticlopride (4).
To further investigate the role of D 2 and D 3 receptors in the modulation of c-fos responses to pharmacological stimuli, the present study used mice deficient for D 2 and D 3 receptors to analyze their levels of forebrain c-fos mRNA expressed in response to (a) direct stimulation of D 1 receptors with a full D 1 agonist, (b) indirect stimulation of dopamine receptors via methamphetamine-induced dopamine release, and (c) agonist stimulation of D 1 receptors 1-3 weeks after an application of either methamphetamine or the D 1 agonist. This study revealed that the constitutive inactivation of D 2 and D 3 receptors leads to a decrease in agonist-promoted D 1 receptor activity that can be reversed in a long term manner by a single dose of either methamphetamine or a D 1 agonist.

MATERIALS AND METHODS
Animals-The generation of D 2 and D 3 receptor knockout mice was described previously (5). The present study used the fifth generation of congenic C57Bl/6 mutants and their wild-type littermates. Experiments were carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at Columbia University. All mice were housed in groups of 4 -5 animals/cage with free access to food and water. Animals housed in the same cage received the same drug treatment (see below) and were returned to their home cage after drug injection until they were killed by decapitation.
Drug Treatments-All drugs were dissolved in saline and administered intraperitoneally. The D 1 agonist SKF82958, the D 1 antagonist SCH23390, and S-(ϩ)-methamphetamine hydrochloride were purchased from Research Biochemicals, Inc. (Natick, MA). The doses of methamphetamine (2-8 mg/kg) administered to the animals were calculated based on the molecular weight of the salt compound (C 10 H 15 N⅐HCl).
RNA Extraction and Northern Blotting-After decapitation, the brain was rapidly removed, and the forebrain was dissected. For this dissection, the mesodiencephalic junction was used as the anatomic landmark for the caudal border of the forebrain. In some experiments, the forebrain neocortex was further dissected from the extraneocortical structures containing the striatum, hypothalamus, thalamus, and epithalamus. RNA was extracted using the guanidine/cesium chloride ultracentrifugation method. 20 g of total RNA (extracted from tissues pooled from 2 to 4 animals/genotype) was loaded onto each lane of * This work was supported by National Institutes of Health Grant MH56123 and National Science Foundation Grant IBN 9808567. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
‡ To whom correspondence should be addressed: Columbia University/NYSPI 1051 Riverside Dr., Box 42, New York, NY 10032. Tel.: 212-543-6505; Fax: 212-543-6017; E-mail: schmauss@neuron.cpmc. columbia.edu. formaldehyde/agarose gels. Northern blots of these gels were probed with a 32 P-labeled random-primed cDNA corresponding to nucleotides 2160 -2690 of the mouse c-fos gene (6). To confirm equal gel loading and membrane transfer of RNA, all blots were reprobed with a 32 P-labeled cDNA encoding the entire coding region of the human small nuclear ribonucleoprotein N gene that is expressed exclusively and at high levels in neurons (7). 32 P signals on autoradiograms were assessed densitometrically using the NIH Image Analysis software. Optical density measurements of standards on the film were made to construct a standard calibration curve. Relative optical densities were determined for optical densities of signals located in equal size sample areas, and the ratio of c-fos/N mRNA was determined for the signals on each lane. For animals at postnatal age 60 (P60) 1 , the mean differences of mRNA levels between genotypes were determined with 4 -5 independent Northern blotting experiments (each performed with RNAs pooled from two animals/genotype). For each series of these experiments, Northern blots were probed with equal aliquots of the same 32 P-radiolabeled c-fos cDNA and exposed to the same film. Multiple means of optical densities were compared with a one-way analysis of variance (ANOVA), and the significance of differences was assessed by Duncan's Studentized Range Test for comparisons of multiple means (threshold of significance, p Ͻ 0.05).

RESULTS
A first series of experiments determined the basal, D 1 agonist, and methamphetamine-induced c-fos mRNA levels in the forebrains of D 2 and D 3 mutant mice and their wild-type littermates. Compared with wild-type mice (in which c-fos mRNA levels are undetectable), the basal c-fos mRNA levels are higher in the forebrains of both D 2 and D 3 mutant mice (Fig. 1A).
The induction of c-fos mRNA expression was determined 60 min after application of the full D 1 agonist SKF82958 (1 mg/ kg). This induction is robust in wild type, but by comparison it is drastically blunted in both D 2 and D 3 mutants (Fig. 1A). A comparison of optical densities (OD) determined for equal size fields of the autoradiogram shown in Fig. 1A (OD wild type, 6.0; D 2 mutants, 0.5; D 3 mutants, 0.8) indicates that the c-fos responses of D 2 and D 3 mutants are reduced to 8.3 and 13.3%, respectively, of the corresponding wild-type level. In five independent experiments (see "Materials and Methods"), c-fos mRNA levels expressed in D 2 mutants were only 12.2 Ϯ 3.7% of the corresponding wild-type levels (p Ͻ 0.001), and c-fos mRNA levels of D 3 mutants reached only 15.7 Ϯ 9.4% of levels expressed in wild type (p Ͻ 0.001). c-fos mRNA levels expressed in D 2 and D 3 mutants did not differ significantly (Table I).
Another series of experiments measured c-fos mRNA levels expressed in response to a single dose of methamphetamine (8 mg/kg). In the forebrain of wild-type mice, c-fos responses are substantially lower than corresponding responses to the D 1 agonist (Fig. 1A). A comparison of optical densities on the autoradiogram shown in Fig. 1A indicates that the magnitude of these c-fos responses (OD, 1.0) is only 16.7% of the responses detected in wild-type animals after D 1 agonist application (OD, 6.0). In four independent experiments, c-fos levels expressed in methamphetamine-treated wild-type mice reached only 23.7 Ϯ 12.0% of the corresponding levels expressed in SKF-treated wild-type mice (Student's t test, p Ͻ 0.001). c-fos mRNA levels induced with only 2 mg/kg methamphetamine also do not differ from c-fos mRNA levels induced with 8 mg/kg methamphetamine (data not shown). Moreover, in contrast to the results obtained with the D 1 agonist, methamphetamine-induced c-fos responses of D 2 or D 3 mutant mice do not differ from wild type (Fig. 1A). Nevertheless, as shown in Fig. 1B in wild type and D 2 and D 3 mutants, c-fos responses to methamphetamine treatment are markedly reduced by pretreatment with the D 1 receptor antagonist SCH23390 (0.3 mg/kg), but they are unaf-fected by a pretreatment with saline. These data confirm that the induction of c-fos expression by amphetamine-like drugs is dependent upon D 1 receptor activation (3).
Additional studies compared c-fos responses to D 1 agonist stimulation in drug-naive mice and mice that received a single dose of methamphetamine (5 mg/kg) 1, 2, or 3 weeks before SKF treatment. In wild-type animals, the c-fos responses to D 1 agonist stimulation are not significantly altered 1 and 2 weeks after methamphetamine pretreatment (Fig. 1C). Interestingly however, as many as 2 weeks after a single dose of methamphetamine, the levels of c-fos induced in both D 2 and D 3 mutants are similar to the levels expressed in either drug-naive or methamphetamine-pretreated wild-type mice (Fig. 1C). Thus, in contrast to the blunted c-fos responses to D 1 agonist stimulation of drug-naive mutants (Fig. 1A), methamphetamine- were determined 60 min after an intraperitoneal saline injection. c-fos responses to an intraperitoneal injection of the D 1 agonist SKF82958 (1 mg/kg) and methamphetamine (8 mg/kg) in wild type (lanes 1), D 2 (lanes 2), and D 3 mutants (lanes 3) were also measured 60 min after drug administration. A 0.24 -9.5-kilobase RNA ladder (Life Technologies, Inc.) indicates a ϳ2.3-kilobase c-fos mRNA species that is known to encode the c-fos protein. B, c-fos responses to methamphetamine of wild type (wt), homozygous D 2 (D2Ϫ/Ϫ), and D 3 (D3Ϫ/Ϫ) mutants (lanes 1) are unaffected by pretreatment with saline (lanes 2), but they are markedly reduced by pretreatment with the D 1 -selective antagonist SCH23390 (0.3 mg/kg intraperitoneal) (lanes 3). C, c-fos responses to SKF (1 mg/kg) administered 1 and 2 weeks after a single injection of methamphetamine (5 mg/kg intraperitoneal) (lanes 1 and 2, respectively) in wild type, homozygous D 2 , and D 3 mutants compared with corresponding responses measured in drug-naive animals (lanes 0). D, when SKF is administered 3 weeks after methamphetamine pretreatment, blunted c-fos responses of D 2 and D 3 mutants are again apparent when compared with wild type. E, to test for possible differences in the amounts of mRNAs loaded onto and/or transferred from each lane of the gels, all blots were reprobed with c-DNA encoding the brain-specific small nuclear ribonucleoprotein N (7). Representative examples shown for; blots A, B, and D illustrate that equal levels of the 1.6-kilobase N-encoded mRNA are detected on all lanes. pretreated D 2 and D 3 mutants show substantially more robust c-fos responses to D 1 agonists (see below). However, 3 weeks after methamphetamine administration, the blunted c-fos responses of the mutants to D 1 agonist stimulation are again apparent, and a comparison of the optical densities of signals on the autoradiogram shown in Fig. 1D (OD wild type, 4.08; D 2 mutants, 0.8; D 3 mutants, 1.0) indicates that D 2 and D 3 mutants express only 19.5 and 24.4% of the corresponding wildtype c-fos levels (these levels are similar to the levels determined above for SKF-treated drug-naive mutants).
The results summarized above were obtained from adult mice that received the first drug injection at P60. To determine whether similar results were obtained in preadolescent mice, experiments were also performed in mice at postnatal age 30. As shown in Fig. 2A, P30 mice express only marginally increased levels of c-fos in response to the D 1 agonist (1 mg/kg), and these levels do not differ from corresponding levels induced by methamphetamine (8 mg/kg). Moreover, at these low expression levels, no significant differences are found between wild type and D 2 mutants, and D 3 mutants express only slightly reduced levels of c-fos in response to both SKF and methamphetamine ( Fig. 2A, see lanes marked 1, 2, and 3). However, when P30 mice (both wild type and mutants) are treated with a single dose of methamphetamine (5 mg/kg) and challenged with SKF at P47, c-fos responses are robust in all genotypes, and no differences are found between wild-type and mutant mice. Furthermore, D 2 single mutants that received methamphetamine at postnatal day 24 also express high levels of c-fos in response to SKF administered at P30, and the c-fos mRNA levels of these mutants do not differ from the levels expressed in mice treated with methamphetamine at P30 and challenged with SKF at P47 (Fig. 2B). Altogether, these results indicate that a single dose of methamphetamine leads to a long term increase in c-fos responses to D 1 agonist stimulation in brains with either low (preadolescent mice) or blunted (adult D 2 and D 3 mutant mice) c-fos expression levels.
To further determine whether the D 1 agonist and methamphetamine induce different levels of c-fos in different anatomic areas of the forebrain, additional experiments compared c-fos mRNA levels in the neocortex and remaining forebrain of adult wild-type and mutant mice. These results revealed that the levels of c-fos mRNA expressed in response to a single dose of methamphetamine are higher in the forebrain neocortex compared with the inner (extraneocortical) mass of the forebrain (Fig. 3). This finding is in contrast to the D 1 agonist, which induces similar levels of c-fos in the neocortex and in extraneocortical forebrain structures (Fig. 4). Moreover, in both D 2 and D 3 mutants that were pretreated with methamphetamine, drastically enhanced c-fos responses to SKF treatment are detected in the neocortex, but the c-fos mRNA levels in the extraneocortical forebrain remain low (Fig. 3). In fact, on the autoradiogram shown in Fig. 3, neocortical c-fos levels of D 2 and D 3 mutants are 3.4-and 2.1-fold higher, respectively, compared with wild-type c-fos levels (OD wild type, 1.37; D 2 mutants, 4.70; D 3 mutants, 2.87). By comparison, the optical densities of c-fos signals obtained from extraneocortical forebrain mRNA are 1.56, 0.5, and 0.5 for wild type, D 2 mutants, and D 3 mutants, respectively. Thus, the methamphetamine-induced reversal of the blunted c-fos expression of both mutants in  2. c-fos mRNA responses to a D 1 agonist and methamphetamine in the forebrains of wild-type, D 2 , and D 3 mutant mice at postnatal age P30. A, the c-fos mRNA levels expressed in P30 wild type (lanes 1), homozygous D 2 mutants (lanes 2), and homozygous D 3 mutants (lanes 3) in response to SKF82958 (SKF, 1 mg/kg) and methamphetamine (METH, 8 mg/kg) are compared with c-fos mRNA levels expressed in SKF-treated D 3 mutants at P60 (lane P60). B, c-fos mRNA levels of SKF-treated P30 wild-type mice (wt P30) are compared with corresponding mRNA levels expressed in wild-type (wt), homozygous D 2 , and homozygous D 3 mutant mice that received a single injection of methamphetamine (5 mg/kg) at P30 followed by a single dose of SKF (1 mg/kg) at P47, and D 2 mutants that received an equal dose of methamphetamine at P15 followed by SKF treatment at P30. For each genotype, brain tissues of four animals/genotype were pooled for RNA extraction, and 20 g of total RNA was loaded onto each lane. The blots shown in A and B were exposed to film for 16 and 6 h, respectively.

FIG. 3. c-fos mRNA responses to methamphetamine and SFK82958 in the neocortex and extraneocortical forebrain structures of wild-type, D 2 , and D 3 mutant mice.
Top left, c-fos mRNA levels detected in the neocortex (CTX) and extraneocortical forebrain (fb) 60 min after methamphetamine (8 mg/kg) treatment of P60 wild-type (wt) mice. Top middle, the blot shown on the left was reprobed with radiolabeled N-encoded cDNA. Top right, comparison of c-fos mRNA levels expressed in response to SKF treatment in the neocortex and in extraneocortical forebrains of wild type, homozygous D 2 (D2), and D 3 (D3) mutants that were treated with a single dose of methamphetamine (5 mg/kg) 1 week before SKF administration. Bottom, the first three lanes show basal levels of c-fos mRNA detected 1 week after the administration of a single dose of methamphetamine. The following lanes show c-fos responses to methamphetamine (8 mg/ kg) given to either methamphetamine-pretreated animals (Meth/Meth) or drug-naive animals (Meth-naive). 20 g of total RNA was loaded onto each lane, and all blots were exposed to film for 6 h. response to D 1 agonist treatment is only due to an increase in their neocortical c-fos mRNA responses. In four independent experiments (see Table I), neocortical c-fos mRNA levels of D 2 mutants were 310 Ϯ 40% of the corresponding wild-type levels, and this difference is significant (p Ͻ 0.01). Mean neocortical c-fos mRNA levels of D 3 mutants were also significantly increased (229 Ϯ 39.4% of corresponding wild-type levels, p Ͻ 0.05).
Pretreatment with methamphetamine, however, does not affect c-fos responses to subsequent stimulation with methamphetamine. As shown in Fig. 3, methamphetamine-induced c-fos responses are indistinguishable between drug-naive and methamphetamine-pretreated wild-type and mutant mice. Interestingly however, as also shown in Fig. 3, the basal levels of c-fos mRNA measured 1 week after methamphetamine administration no longer differ between wild-type and mutant mice (compare with Fig. 1A).
Another experiment tested whether pretreatment with a single dose of the D 1 agonist (1 mg/kg) also affects c-fos responses to subsequent D 1 agonist stimulation. These results are shown in Fig. 4. In both neocortex and extraneocortical forebrains, c-fos induction by SKF treatment is robust in wild type and in D 2 and D 3 mutants that were pretreated with SKF 1 week earlier. In fact, compared with wild type, optical densities of the signals on the autoradiogram shown in Fig. 4 (top right) revealed a 1.8-and 1.3-fold increase in neocortical c-fos mRNA levels expressed in D 2 and D 3 mutants, respectively. In four independent experiments (see Table I), neocortical c-fos mRNA levels of D 2 and D 3 mutants were significantly increased to 201 Ϯ 10.2% and 151 Ϯ 30.6%, respectively, of the corresponding wild-type levels (D 2 mutants/wild type, p Ͻ 0.001; D 3 mutants/wild type, p Ͻ 0.01; D 2 mutants/D 3 mutants, p Ͻ 0.01).
A further comparison between c-fos responses of SKF-naive and SKF-pretreated animals revealed that SKF pretreatment, similar to methamphetamine pretreatment, markedly increased c-fos responses only in the mutants (Fig. 4, bottom). However, in contrast to methamphetamine, SKF does not significantly alter the increased basal levels of c-fos in D 2 mutants. Also, D 3 mutants continue to express detectable c-fos mRNA, although to a lesser extent (Fig. 4, bottom). DISCUSSION This study shows that compared with methamphetamine, higher c-fos mRNA levels are expressed in response to D 1 agonist stimulation and that adult mice lacking either the D 2 or D 3 receptor show blunted c-fos responses to the D 1 agonist. A single dose of methamphetamine induces a long lasting enhancement of c-fos responses in brains with either low (preadolescent wild-type and mutant mice) or blunted (adult D 2 and D 3 mutants) c-fos expression levels. Moreover, the enhanced c-fos responses to the D 1 agonist seen in methamphetaminepretreated adult mutants are only detected in the neocortex, a brain region in which the acute administration of methamphetamine itself induces the largest c-fos responses. Furthermore, a single dose of a full D 1 agonist elicits similar long lasting enhancement of c-fos responses to subsequent D 1 agonist stimulation. These data suggest that despite the unaltered expression of D 1 ligand-binding sites in D 2 and D 3 mutants (8,9), the chronic inactivation of D 2 and D 3 receptors leads to a decreased responsiveness of D 1 receptors to agonist stimulation, which can be reversed by a single dose of either methamphetamine or a D 1 agonist.
Significantly reduced c-fos protein responses to D 1 -agonist stimulation were previously reported (4) for the same D 3 mutants in this study. The present analysis of c-fos mRNA expression levels revealed similarly blunted c-fos responses in mice deficient for D 2 receptors, an effect that could only marginally be detected in the previous protein study (4). This analysis suggests that D 2 but not D 3 mutants develop compensatory mechanisms that operate either at the translational or posttranslational level to maintain wild-type-like c-fos protein responses to D 1 agonist stimulation. The present study also found increased basal levels of c-fos mRNA in the forebrain of mice deficient for D 2 and D 3 receptors. This result is similar to the results of previous pharmacological studies showing increased c-fos expression levels in rats that were treated acutely or chronically with the D 2 /D 3 receptor blocker haloperidol (10,11).
The c-fos responses to methamphetamine differ both quantitatively and qualitatively from the c-fos responses induced by the D 1 agonist. In contrast to the widespread c-fos expression induced by the D 1 agonist, the effects of methamphetamine are delimited to the neocortex. These effects suggest that the two types of pharmacological stimuli activate different neuronal populations/circuitries that express c-fos, and future and more detailed investigations of the anatomic distribution of c-fos mRNA expression will need to test investigation. The quantitative differences of c-fos expression levels induced by methamphetamine and D 1 agonists may also reflect differences in the activation of the two principal transcriptional activators of the FOS gene, pMAP kinase and cAMP-response element-binding protein (pCREB) (12). In any case, a main finding of the present study is that a single dose of methamphetamine (5 mg/kg) leads to a long term (as many as 2 weeks) reversal of the blunted c-fos responses to D 1 agonist stimulation in the forebrain of mice deficient for D 2 and D 3 receptors. Interestingly, the same dose of amphetamine has previously also been shown to induce a long-lasting behavioral and neuroendocrine sensitization in rats that is accompanied by an increase in electrically evoked dopamine release in the forebrain (13). Moreover, FIG. 4. c-fos mRNA responses to SFK82958 in the neocortex and extraneocortical forebrain structures of SKF-naive and SKF-pretreated wild-type, D 2 , and D 3 mutant mice. Top left, c-fos mRNA levels expressed 60 min after SKF injection in the neocortex (CTX) and in the extraneocortical forebrain (fb) of SKF-naive wild types (wt). Top right, c-fos mRNA expressed in wild-type and mutant mice that were treated with a single dose of SKF (1 mg/kg) 1 week before a subsequent dose of SKF. Middle, the first three lanes show basal levels of c-fos mRNA detected 1 week after the administration of a single dose of SKF. The following lanes show c-fos responses to SKF given either to SKF-pretreated animals (SKF/SKF) or to SKF-naive animals (SKFnaive). Bottom, the blot shown in the middle was reprobed with Nencoded cDNA. 20 g of total RNA was loaded onto each lane, and all blots were exposed to film for 3 h. as shown here one long term consequence of this dose of methamphetamine is a decrease of the (abnormally) high basal c-fos levels in D 2 and D 3 mutants. This consequence may perhaps be one of the mechanisms by which methamphetamine (but not the D 1 agonist) increases the responsiveness of neurons to subsequent D 1 agonist stimulation.
It will be of great interest to further elucidate the molecular mechanisms that lead to the methamphetamine-induced and D 1 agonist-induced long term enhancement of cortical c-fos responses to D 1 agonist stimulation in brains with low or abnormally blunted c-fos responses. The decreased agonist-promoted D 1 receptor activity detected in D 2 and D 3 mutants suggest that the chronic treatment with neuroleptic drugs that block D 2 and D 3 receptors (a common therapeutic intervention of schizophrenia) could impair the function of D 1 receptors. Several other studies provided evidence for a reduced cortical D 1 receptor activity during chronic neuroleptic treatment (14 -16), and results of a most recent study (14) suggested for the first time that a short term co-administration of a D 1 -selective agonist to monkeys chronically treated with neuroleptics can improve behavioral deficits associated with a decreased cortical D 1 receptor activity. The present study now shows that the decreased response of neocortical D 1 receptors to agonist stimulation in mice deficient for D 2 and D 3 receptors is not irreversible and suggests that an intermittent stimulation of dopamine release by amphetamine-like drugs during treatment with typical neuroleptics can result in a long term increase in agonist-promoted D 1 receptor activity.