Phosphodiesterase 1B differentially modulates ... - Wiley Online Library

7 downloads 0 Views 169KB Size Report
of methamphetamine on locomotor activity and spatial learning through. DARPP32-dependent pathways: evidence from PDE1B-DARPP32 double-knockout ...
Genes, Brain and Behavior (2006) 5: 540–551

# 2006 The Authors Journal compilation Ó 2006 Blackwell Munksgaard

Phosphodiesterase 1B differentially modulates the effects of methamphetamine on locomotor activity and spatial learning through DARPP32-dependent pathways: evidence from PDE1B-DARPP32 double-knockout mice L. A. Ehrman†, M. T. Williams‡, T. L. Schaefer‡, G. A. Gudelsky§, T. M. Reed{, A. A. Fienberg**, P. Greengard†† and C. V. Vorhees*,‡ † Division of Developmental Biology, ‡ Division of Neurology, Cincinnati Children’s Research Foundation and University of Cincinnati College of Medicine, §College of Pharmacy, University of Cincinnati, ¶ Department of Biology, College of Mount St. Joseph, Cincinnati, OH, †† Intra-Cellular Therapies, Inc., and §§Laboratory of Molecular and Cellular Neuroscience, Rockefeller University, New York, NY, USA *Corresponding author: C. V. Vorhees, Division of Neurology (MLC 7044), Cincinnati Children’s Research Foundation, 3333 Burnet Ave., Cincinnati, OH 45229-3039, USA. E-mail: [email protected]

Mice lacking phosphodiesterase 1B (PDE1B) exhibit an exaggerated locomotor response to D-methamphetamine and increased in vitro phosphorylation of DARPP32 (dopamine- and cAMP-regulated phosphoprotein, Mr 32 kDa) at Thr34 in striatal brain slices treated with the D1 receptor agonist, SKF81297. These results indicated a possible regulatory role for PDE1B in pathways involving DARPP32. Here, we generated PDE1B x DARPP32 doubleknockout (double-KO) mice to test the role of PDE1B in DARPP32-dependent pathways in vivo. Analysis of the response to D-methamphetamine on locomotor activity showed that the hyperactivity experienced by PDE1B mutant mice was blocked in PDE1B–/– x DARPP32–/– double-KO mice, consistent with participation of PDE1B and DARPP32 in the same pathway. Further behavioral testing in the elevated zero-maze revealed that DARPP32–/– mice showed a less anxious phenotype that was nullified in double-mutant mice. In contrast, in the Morris water maze, double-KO mice showed deficits in spatial reversal learning not observed in either single mutant compared with wild-type mice. The data suggest a role for PDE1B in locomotor responses to psychostimulants through modulation of DARPP32-dependent pathways; however, this modulation does not necessarily impact other behaviors, such as anxiety or learning. Instead, the phenotype of double-KOs observed in these latter tasks may be mediated through independent pathways. Keywords: DARPP32, dopamine signaling, locomotor activity, methamphetamine, phosphodiesterase, spatial learning

540

Received 14 September 2005, revised 18 October 2005, accepted for publication 28 December 2005

Phosphodiesterase 1B (PDE1B) is a member of the PDE superfamily that is responsible for catalyzing the hydrolysis of cyclic nucleotides to their corresponding 50 monophosphates. The superfamily of PDEs comprises 11 families with PDE1B belonging to the Ca2þ/calmodulin-dependent PDE1 family (Erneux et al. 1985; Francis et al. 2001). The striatum, dentate gyrus, and olfactory tubercles express high levels of PDE1B, but the particular signaling pathways within these brain regions that require PDE1B for normal neurotransmission are not known (Polli & Kincaid 1992, 1994; Yan et al. 1994). Mice with a targeted inactivation of PDE1B exhibit increased baseline locomotor activity, exaggerated locomotor hyperactivity in response to D-methamphetamine, and impaired spatial learning in the Morris water maze (Reed et al. 2002). Methamphetamine increases locomotor activity by indirectly stimulating dopaminergic and serotonergic pathways which both use cyclic nucleotides as their second messengers (Barnes & Sharp 1999; Vallone et al. 2000). It is likely that PDE1B down-regulates these pathways by hydrolyzing cyclic nucleotides to their 50 monophosphates. The exaggerated locomotor response to D-methamphetamine in mice without PDE1B may therefore be the product of prolonged or increased activity of downstream effectors of these pathways. DARPP32 [dopamine and cyclic-adenosine 50 -phosphate (cAMP)-regulated phosphoprotein, Mr 32 kDa] is a shared downstream component of both dopaminergic and serotonergic pathways (Fienberg et al. 1998; Svenningsson et al. 2000, 2002a, 2003, 2004). Upon stimulation of dopaminergic D1 receptors or serotonergic receptors (5-HT4, 5-HT6, 5-HT7), adenylyl cyclase activity increases and causes cAMP and cGMP concentrations to rise (Altar et al. 1990; Barnes & Sharp 1999; Monsma et al. 1990). The cyclic nucleotides in turn activate their corresponding protein kinase (PKA or PKG) to phosphorylate DARPP32 at Thr-34 and to activate/amplify the downstream response (Hemmings & Greengard 1986; Svenningsson et al. 2002a, 2002b, 2003, 2004). In vitro evidence, using PDE1B–/– mouse striatal slice preparations

doi: 10.1111/j.1601-183X.2006.00209.x

PDE1B x DARPP32 double knockout mice

exposed to the dopamine D1 receptor agonist, SKF81297, indicates that the downstream effector, DARPP32, is dysregulated and becomes hyperphosphorylated at Thr-34 (Reed et al. 2002). These data indicate that PDE1B and DARPP32 may participate in the same pathways. Additionally, in contrast to PDE1B–/– mice, a targeted deletion of DARPP32 causes a reduction in the locomotor response to psychostimulants such as cocaine or amphetamine at moderate doses (Fienberg et al. 1998; Svenningsson et al. 2003). Although in vitro data suggest a role for PDE1B in pathways involving DARPP32, in vivo evidence supporting such a connection does not exist. To test this connection in vivo, we generated double-mutant mice for DARPP32 and PDE1B and analyzed their response to the same behavioral tests where we had shown differences in PDE1B knockouts. We specifically predicted that DARPP32 deletion would block the exaggerated locomotor hyperactivity effect induced by methamphetamine in PDE1B–/– mice (Reed et al. 2002). A test of anxiety was added to determine whether changes on this measure might be related to changes in other behaviors we measured. The results show that whereas the DARPP32 loss of function produced the predicted effect on locomotor responses to methamphetamine in PDE1B–/– mice, the effect of the double deletion was different on other behaviors tested (anxiety and spatial learning).

Materials and methods Subjects DARPP32 and PDE1B knockout mice were generated as described previously (Fienberg et al. 1998; Reed et al. 2002). PDE1B–/– mice were selected from the fifth generation backcross onto a C57BL/6 background (Charles River, Wilmington, MA). DARPP32–/– mice were selected from the tenth generation backcross onto a C57BL/6 background (The Jackson Laboratory, Bar Harbor, ME). PDE1B–/– mice were mated with DARPP32–/– mice to generate double heterozygous-breeding pairs. PDE1B–/– and DARPP32–/– single-mutant, double-mutant, and wild-type (WT) offspring from double heterozygous mating were used for testing. Genotyping of the offspring was carried out using polymerase chain reaction (PCR) analysis. For PDE1B, the PCR primers for the WT band (1420 bp) were 50 -cacacaggcacaaccaac-30 and 50 -ttgggtggctgatgtcagca-30 , and for the mutant band (1183 bp) were 50 -ctgctaaagcgcatgctccagactgccttg-30 and 50 -accttcacccagcagacccga-30 . For DARPP32, the PCR primers for the WT band (150 bp) were 50 agagaactgaatcttcttgcg30 and 50 gcgggattttcctgg30 ; and for the mutant band (300 bp) were neo primers, 50 gcaaggtgagatgacaggagatc30 and 50 cgcttgggtggagaggctattc30 . Mice were earmarked on postnatal day 7 (P7) and weighed weekly from P7 to P105. Tail biopsies for PCR genotyping analysis were obtained on P21 and after P50. Litters were weaned on P28. Same gender mice were housed one to four per cage. Genes, Brain and Behavior (2006) 5: 540–551

Locomotor activity On P50, P51, or P52, locomotor activity was measured in a 41  41  30-cm Digiscan activity monitor equipped with 16 pairs of photodetector-LED beams along the x and y axis (Accuscan Electronics with VersaMax software, Columbus, OH). Mice were acclimated to the chamber for 1 h immediately prior to being challenged with a subcutaneous injection of either 0.8 mg/kg D-methamphetamine HCl (free base) or saline, both administered in a volume of 5 ml/kg of body weight. Postchallenge activity was measured for 3 h. Horizontal activity and total distance were recorded in 3-min intervals during the pre- and postchallenge periods and analyzed in 6 and 30-min intervals.

Zero-maze On P85 (>30 days after activity assessment), animals were tested in an elevated zero maze as a test of anxiety (Shepherd et al. 1994). Briefly animals were placed in the center of one of the closed areas of the ring-shaped apparatus, and behavior was recorded for 5 min with an overhead camera connected to a video recorder as described previously (Williams et al. 2003). Overhead fluorescent lighting was turned off, and a single halogen light was used. Between animals the maze was cleaned with 70% ethanol. The dependent measures for this task were scored from the video recordings and included the number of head dips, stretch-attends, and time in the open area. Time in the open was considered when animals had both front paws past the boundary of the closed area and extending into an open area.

Morris water maze – cued learning A Morris water maze was used with modifications for mice (Upchurch & Wehner 1988). The maze consists of a circular stainless steel perimeter (122 cm in diameter). To facilitate tracking, we painted the maze interior white and colored the water with white tempera paint. Water temperature was 20–22  C. On the Monday after P85, mice were subjected to 6 days of cued Morris water maze to evaluate proximal cued learning. Black curtains were drawn around the maze to minimize distal cues. The platform (10  10 cm, submerged 1 cm below the water) was marked using a black styrofoam cylinder (7 cm in diameter, 5 cm in height) mounted 12 cm above the surface of the water on a 20-cm rod. Four consecutive trials were administered each day, and for each trial, the platform and start positions were rotated to different locations. The time limit per trial was 1 min, and the intertrial interval (ITI) was 15 seconds on the platform plus an additional 30 seconds in a polycarbonate cage while the platform was relocated. Animals not finding the platform within the 1-min time limit were placed on the platform for 15 seconds. Mice were observed on closed-circuit television, and latencies were recorded. No animals were excluded from cued

541

Ehrman et al.

learning because of failure to search for the platform (floating). Mice began this test 30 days after activity testing.

Morris water maze – place learning On the Monday following cued learning, mice were subjected to 6 days of hidden platform trials to assess spatial acquisition. The curtains were open during this phase, and the wall of the room had a variety of prominent cues, including geometric shapes that were visible from the maze. A 10  10-cm platform covered in white felt was submerged 1 cm below the surface of the water in the SW quadrant. Daily sessions consisted of four trials with a 1-min time limit and a 15-second ITI. Animals not finding the platform within 1 min were placed on the platform for 15 seconds. Start positions (N, E, SE, and NW) were randomized daily such that each position was used once per day. On the seventh day, the platform was removed and mice were started from a novel position (NE) during a 1-min probe trial. On the Monday following the acquisition trials, a reduced 5  5-cm platform covered in white felt was placed in the NE quadrant for reversal learning. The same procedure used for acquisition was used for reversal (6 days, four trials/day, 1-min time limit, 15-second ITI). Start positions were again randomized daily (SE, S, W, and NW). On the seventh day, the platform was removed and a 1-min probe trial was administered beginning from the SW position. For acquisition and reversal phases of the Morris water maze, animals were tracked with the Polytrack video-tracking system (San Diego Instruments, San Diego, CA). For the learning trials, path length, cumulative distance from the platform, latency, and first bearing were analyzed. For the probe trials, measures recorded were time and distance in the target quadrant, number of target site crossings, average distance from the platform site, first bearing, and percent time in each of three annuli. The annuli dimensions (with maze radius of 61 cm) were target annulus ¼ 16.2, outer annulus ¼ 20.1, and inner annulus ¼ 24.7 cm. The tracker recorded the animal’s position every 55 ms. First bearing was determined based on the animal’s average heading during the first 13 cm of tracking of each trial relative to a direct line to the goal. No mice were excluded from the analysis of acquisition because of floating, but six mice were excluded from the more difficult reversal phase because of floating. The genotype of the excluded mice were four WT (two females and two males) and two PDE1B–/– (both males) mice. No DARPP32–/– or double-knockout (doubleKO) mouse were excluded because of floating.

remaining tissue was bisected midsagittally and the hippocampus removed from each section. Each region was pooled bilaterally and stored at 70  C until assayed. Tissues were thawed, weighed, and diluted with 50 volumes of 0.2 N perchloric acid, homogenized, and centrifuged at 10 000  g for 5 min. Twenty-microlitre samples were injected onto a C18column (5 m, 100  2 mm) connected to an LC-4B amperometric detector (Bioanalytical Systems, West Lafayette, IN) with a reference electrode maintained at an oxidation potential of þ0.60 V. The mobile phase (pH 4.2) consisted of 35 mM citric acid, 54 mM sodium acetate, 50 mg/l disodium ethylenediamine tetraacetate, 50 mg/l octane sulfonic acid sodium salt, 3% methanol, and 3% acetonitrile. The flow rate was 0.4 ml/min. Chromatograms were recorded and integrated and neurotransmitter concentrations calculated from standard curves generated for each analyte.

Statistics Locomotor activity was analyzed in 30-min intervals using split-plot ANOVA (SAS PROC GLM) as a 3-between (PDE1B  DARPP32  sex), 1-within (interval) design. Because genotype had a significant effect on baseline activity, postchallenge activity was analyzed as percent change relative to the last prechallenge interval. Because the purpose of the experiment was to test whether the DARPP32 allele blocked the effects of methamphetamine on the locomotor activity of PDE1B–/– mice, postchallenge comparison of the PDE1B–/– to the double-KO group was performed separately. For the Morris water maze, zero-maze, and neurotransmitters, the ANOVA between factors were genotype (either PDE1B or DARPP32), sex, and test history, and for the Morris water maze, the additional within factors of day and trial were included. Test history was not significant on any neurotransmitter, zero-maze, or Morris water maze findings, therefore, this factor was combined for all subsequent analyses. Split-plot ANOVA F-ratios involving within factors were corrected in cases where the covariance matrix was significantly non-spherical using the Greenhouse–Geisser method. Interaction effects were localized using simple-effect ANOVAs prior to individual group comparisons by the step-down Ftest method (Kirk 1995) that controls for multiple comparisons. Thus, most of the ANOVAs were 3-between, 1- or 2within designs, in which there were two levels of the PDE1B genotype (PDE1B allele present or absent), two levels of the DARPP32 genotype (DARPP32 allele present or absent), two levels of sex, and multiple levels of the remaining within factor(s). Effects were considered significant at P < 0.05 and trends at P < 0.10.

Monoamine neurotransmitters Approximately 48 h following behavioral testing, mice were decapitated and brains rapidly removed and the neostriatum (caudate) and hippocampus dissected over ice using a brain block (Zivic-Miller, Pittsburgh, PA). First, a coronal cut was made at the optic chiasm, and a 2-cm slice obtained rostral to this cut was where the neostriatum was obtained. The

542

Results General characteristics There was no mortality during the course of the experiment. Animals of all four genotypes appeared normal. An ANOVA Genes, Brain and Behavior (2006) 5: 540–551

PDE1B x DARPP32 double knockout mice

PDE1B–/– and DARPP32–/– single-KO groups were significantly more active than the WT group (P’s < 0.0001; Fig. 1a, left). This interaction also revealed that the PDE1B–/– group differed from WT from the outset, whereas the DARPP32–/– group started out similar to WT then diverged (not shown). However, the two single-KO groups’ activity converged with WT during the last two intervals (Fig. 1a, inset). The double-KO group was significantly more active at all prechallenge intervals except the first. Because of the prechallenge baseline differences, postchallenge activity was analyzed as percent change relative to the last prechallenge interval (as the best reflection of each groups’ habituated activity). Figure 1a shows the preand postchallenge activity in absolute terms, and Fig. 1(b) shows the postchallenge activity as percent change relative to the last prechallenge interval. As can be seen, all groups given saline showed comparable levels of postchallenge activity (Fig. 1b, Saline panel). There was no significant effect of either genotype or interactions of either genotype with other factors among the saline-challenged groups. Furthermore, activity of all groups after saline challenge was below the prechallenge level, reflecting habituation during the 3 h postchallenge test interval.

performed on body weight on P49 (the day before behavioral testing began) showed no significant effects of genotype (P > 0.28). Females were lighter than males (P < 0.0001), but there was no significant genotype by sex interaction (P > 0.74).

Locomotor activity D-Methamphetamine-stimulated

(0.8 mg/kg, free base) activity was recorded for WT, PDE1B–/–, DARPP32–/–, and doubleKO mice. Separate sets of animals of each genotype were treated with saline at the end of the prechallenge interval to determine whether there were persistent group differences in activity. Prechallenge baseline activity, analyzed using 3-between  1-within ANOVAs for mice challenged with methamphetamine, revealed effects of PDE1B (F1,71 ¼ 37.16, P < 0.0001) and DARPP32 (F1,71 ¼ 24.63, P < 0.0001) but no PDE1B  DARPP32 interaction. However, there was a DARPP32  interval interaction (F9,639 ¼ 3.96, P < 0.001) and a PDE1B  DARPP32  interval interaction (F9,639 ¼ 2.12, P < 0.05). Further analyses of the latter interaction revealed that the double-KO group was significantly more active than WT (P < 0.0001) and either single-KO group (P’s < 0.01). In addition, 2500

(a)

(b)

2000 1500

Meth

200

1000

le al Ch

Horizontal activity (counts)

0

16 000

48

54

12 000

8000

4000

0 30 60

30 60 90 120150180

60

Post horizontal activity-percentage change

Meth

500

ng

e

20 000

150

100

Saline WT PDEIB–/– DARPP32–/– Double-KO–/–

50

0

–50

–100 Interval

Figure 1: (a) Mean ± SEM horizontal activity expressed in 30-min intervals during a 1-h prechallenge baseline and a 3-h postchallenge period. Inset, prechallenge activity in 6-min intervals shown for the last three intervals of the prechallenge 60-min session showing the convergence of activity for all groups except the double-knockout (double-KO) group. (b) Horizontal activity expressed in 30-min intervals during the 3-h postchallenge interval in the groups receiving D-methamphetamine (0.8 mg/kg) (left panel) or saline (right panel) as a function of genotype expressed as a percentage of prechallenge activity. Percent change was calculated by normalizing mean horizontal activity over the postchallenge intervals relative to the last prechallenge interval. Horizontal activity is the total number of photobeam interruptions per interval. The hyperactivity to D-methamphetamine in PDE1B–/– mice (P < 0.023) was blocked in double-KO mice. No differences in postchallenge activity were observed in separate groups of mice challenged with saline. *P < 0.05 vs. WT. Males and females combined. Group sizes: WT ¼ 49 (male ¼ 31, female ¼ 18), PDE1–/–¼ 45 (male ¼ 22, female ¼ 23), DARPP32–/–¼ 29 (male ¼ 16, female ¼ 13), Double-KO ¼ 26 (male ¼ 14, female ¼ 12). Genes, Brain and Behavior (2006) 5: 540–551

543

Ehrman et al.

The post-D-methamphetamine challenge percent change analyses using 3-between, 1-within ANOVAs showed that there were PDE1B  interval (F29,2059 ¼ 3.66, P < 0.003) and DARPP32  interval (F29,2059 ¼ 1.98, P < 0.067) effects on horizontal activity (Fig. 1b, Meth panel); however, there was no significant PDE1B  DARPP32 interaction nor PDE1B  DARPP32 higher order interactions with interval, sex or sex and interval. Separate group comparisons based on the premise of the experiment that DARPP32 would block the D-methamphetamine-induced hyperactivity seen in PDE1B–/– mice showed a hyperactivity response in the PDE1B–/– mice (P < 0.05) as previously reported (Reed et al. 2002) compared with WT controls during the first three 30-min intervals (Fig. 1b). However, the hyperactivity response of PDE1B–/– mice was blocked in the double-KO mice during these same three 30min intervals because the double-KO group differed from the PDE1B–/– group (P < 0.05) and did not differ from WT (Fig. 1b). DARPP32–/– single-mutant mice showed a response to D-methamphetamine similar to that of WT mice.

DARPP32–/– mice spent more time in the open than WT mice (Fig. 2a). The PDE1B–/– mice showed a similar but non-significant trend in the same direction (P < 0.08).

Learning and memory Morris water maze – cued learning Analysis of latency for cued learning using a 3-between, 2within ANOVA, during which proximal cues were present and prominent distal cues were minimized by curtains, showed a significant main effect of the PDE1B genotype (F1,134 ¼ 30.39, P < 0.0001), a PDE1B–DARPP32 interaction (F1,134 ¼ 5.49, P < 0.02), and a PDE1B–day interaction (F5,670 ¼ 11.61, P < 0.0001). On day 1, PDE1B–/– and double-KO mice took significantly longer (both P < 0.001) than WT mice to reach the visible platform. On days 3 and 4, only double-KO mice took significantly longer than WT mice (both P < 0.01; Fig. 2b). On days 5 and 6, no significant group differences in performance remained. Therefore, when mice entered the place-learning phase, they were all performing at comparable levels.

Zero-maze

Morris water maze – place learning: acquisition

Analysis of time in open using a 3-between ANOVA showed a significant PDE1B  DARPP32 genotype effect (F1,131 ¼ 6.83, P < 0.01), no effect of sex, and no genotype–sex interaction. Group comparisons showed that the

Analyses of path length, cumulative distance, and first bearing to the platform for the acquisition phase of the Morris water maze by 3-between, 2-within ANOVAs showed significant main effects of genotype for multiple measures

(a)

WT PDEIB–/– DARPP32–/– Double-KO

(b) 40

90

35

WT PDEIB–/– DARPP32–/– Double-KO

80 70 60 50 40

Latency (seconds)

Time in open (seconds)

100

30 25 20 15

30 10 20 10 0

5 0 1

2

3

4

5

6

Day

Figure 2: (a) Time in open zones in the zero-maze test of anxiety (mean ± SEM) during the 5-min test. **P < 0.01 vs. wild-type (WT). (b) latency (mean  SEM) to reach the visible platform during the cued trials in the Morris water maze (four trials/day) by genotype. On day 1, PDE1B–/– and double-KO mice took significantly longer to find the platform (both P < 0.001). On days 3 and 4, PDE1B–/– mice showed only a trend toward longer latencies (day 3, P < 0.066; day 4, P < 0.055), whereas double-KO mice continued to exhibit longer latencies (day 3, P < 0.001; day 4, P < 0.008). On days 5 and 6, there were no significant differences (double-KO mice vs. WT latencies on day 5, P < 0.08; day 6, P < 0.06). Males and females combined. Group sizes for zero-maze: WT ¼ 47 (male ¼ 30, female ¼ 17), PDE1B–/–¼ 45 (male ¼ 22, female ¼ 23), DARPP32–/–¼ 29 (male ¼ 16, female ¼ 13), Double-KO ¼ 26 (male ¼ 13, female ¼ 13). Group sizes of cued MWM: WT ¼ 48 (male ¼ 31, female ¼ 17), PDE1B–/–¼ 42 (male ¼ 22, female ¼ 20), DARPP32–/–¼ 27 (male ¼ 16, female ¼ 11), Double-KO ¼ 25 (male ¼ 14, female ¼ 11). **P < 0.01, ***P < 0.001 vs. WT.

544

Genes, Brain and Behavior (2006) 5: 540–551

PDE1B x DARPP32 double knockout mice

[DARPP32 for path length (F1,134 ¼ 10.97, P < 0.001; cumulative distance, F1,134 ¼ 8.56, P < 0.004; first bearing, F1,134 ¼ 8.01, P < 0.005; and for PDE1B for first bearing (F1,134 ¼ 13.64, P < 0.001)], but not for latency. In addition, there were higher order interactions for path length, cumulative distance, and first bearing involving PDE1B  DARPP32  other factors (e.g. day and/or trial); P ’s < 0.05 or beyond). Group comparisons indicated that for each of the significantly affected measures, the doubleKO mice had more difficulty finding the platform than WT controls (Fig. 3b, panels 1–4; P’s < 0.01). Compared with WT mice, DARPP32–/– mice showed a trend toward increased path length (P < 0.07) and first bearing (P < 0.06), and PDE1B–/– mice showed an impairment in first bearing (P < 0.004). Learning curves for path length showed that there were no significant genotype differences on day 1 (Fig. 3a, panel 1), indicating that mutant mice did not begin the test with preexisting differences. Analyses of probe trial performance showed no significant impairment as a function of either genotype for average distance from the platform (Fig. 4a, panel 1), platform site crossings, time in each annulus, latency to the target site, or percent time in the target quadrant. For percent time in the target quadrant where 25% represents chance, the groups performed as follows (mean  SEM): WT ¼ 48.6  2.0, PDE1B–/–¼ 49.8  2.0, DARPP32–/–¼ 48.5  3.0, and double-KO ¼ 44.2  2.9), indicating that all groups remembered the target quadrant equally well.

Morris water maze – place learning: reversal Analyses of path length, cumulative distance from the target, latency, and first bearing for the reversal trials (Fig. 3b, panels 5–8) by 3-between, 2-within ANOVAs revealed significant main effects of both genotypes for path length [PDE1B (F1,128 ¼ 12.19, P < 0.001; DARPP32 (F1,128 ¼ 13.35, P < 0.001)]. In addition, the PDE1B–DARPP32 interaction was significant (F1,128 ¼ 7.14, P < 0.01). Similarly, for cumulative distance the significant effects were PDE1B (F1,128 ¼ 12.30, P < 0.001), DARPP32 (F1,128 ¼ 8.33, P < 0.01), PDE1B  DARPP32 (F1,128 ¼ 5.59, P < 0.02). For latency, the effects were PDE1B (F1,128 ¼ 5.64, P < 0.02) and PDE1B  DARPP32 (F1,128 ¼ 5.27, P < 0.03). For first bearing, the significant effects were PDE1B (1128) ¼ 12.34, P < 0.001), PDE1B  trial (F3,384 ¼ 2.79, P < 0.04) and DARPP32  day  trial (F15, 1920 ¼ 2.47, P < 0.001). Group comparisons of the PDE1B–DARPP32 interactions showed that double-KO mice were impaired in finding the smaller platform when it was in the opposite quadrant compared with WT mice on all measures. No significant differences between DARPP32–/– mice and WT controls were found. PDE1B–/– mice were impaired compared with WT on first bearing (P < 0.02; Fig. 3b, panel 8). Reversal learning curves for path length (Fig. 3a, panel 2) showed that double-KO mice had considerably more difficulty finding the smaller platform in its new Genes, Brain and Behavior (2006) 5: 540–551

location and never reached asymptotic performance with WT or single-KO mice. Analyses of probe trial performance at the end of reversal learning by 3-between, 1-within ANOVA showed a significant main effect of DARPP32 for average distance from the target (F1,127 ¼ 5.11, P < 0.03) and a significant interaction of PDE1B  DARPP32 (F1,127 ¼ 3.94, P < 0.05). A posteriori group comparison (Fig. 4a, panel 2) showed that double-KO mice were farther from the target site than WT mice (P < 0.05). ANOVAs of percent time and percent distance in the target quadrant were similar, but not significant. Percent times in the target quadrant by genotypes (where 25% represents chance performance) were as follows (mean  SEM): WT ¼ 35.1  3.1, PDE1B–/–¼ 38.0  2.7, DARPP32–/–¼ 32.5  3.4, and double-KO ¼ 28.2  4.6. Dividing the maze into three annuli and examining percent time in each annulus revealed a significant PDE1B  DARPP32 effect in the target annulus (F1,117 ¼ 5.85, P < 0.02) and the outer annulus (F1/117 ¼ 6.47, P < 0.01). Group comparisons showed that Double-KO mice tended to spend more time in the outer annulus (P < 0.07) and spent significantly less time in the target annulus (P < 0.05) compared with WT controls (Fig. 4b). PDE1B– DARPP32 interactions were found for percent distance in the target (F1,117 ¼ 5.29, P < 0.03) and outer annuli (F1,117 ¼ 6.29, P < 0.02). To determine whether learning the original platform position on acquisition interfered with learning the new platform position during reversal, we analyzed percent time spent in each of the four quadrants during the 24 learning trials of reversal. The results are shown in Fig. 5. As can be seen (Fig. 5a), percent time spent in the new target quadrant increased progressively whereas time in the former target quadrant, where the platform had been during acquisition (Fig. 5b), decreased progressively, but the double-KO group improved at a much slower rate and extinguished their visits to the previous target quadrant more gradually. However, perseveration, as measured by percent time spent in the previous target quadrant, was not the major determinant of the poorer performance of the double-KO mice; instead, these mice were distinguished from the other groups by their failure to search the new platform location as efficiently as did mice in the other groups. Because of this, the doubleKO group never reached even 30% preference for the reversal target quadrant. Because differences among genotypes were found on cued learning, it is possible that genotype differences on place learning were due to differences in swimming ability rather than differences in spatial navigation, although by the end of cued learning, all groups were performing comparably. To address this, we reanalyzed the acquisition and reversal path length data for place learning using each animal’s cued performance as a covariate in an analysis of covariance (ANCOVA). The results are shown in Fig. 6. The same pattern of effects was obtained as seen in Fig. 3

545

Ehrman et al.

(a) 1300

1

1200

2

1100 WT

1000

–/–

PDEIB DARPP32–/– Double-KO

Path length (cm)

900 800 700

* * * *

600 500 400 300 200 100

1

300 200 100 0

*

4000

5

2

3000 2000 1000

First bearing (degrees)

Latency (seconds)

10

1

* *

WT –/– PDEIB DARPP32–/– Double-KO 1000

40

3

* * * *

600 400 200 0

4

* * * *

* * *

30 20 10

0

5

2

800

0

50

3

6

5000

20 15

5

0

Acquistion

50 40 30 20 10 0

7

* *

4

Cummulative distance (cm)

Path length (cm)

400

* * *

4

5

6

* * * *

15 000

6 10 000

5000

0

70 First bearing (degrees)

(b) 500

3

Path length (cm)

2

Latency (seconds)

1

Cummulative distance (cm)

0 Day:

60

8

* *

* * * *

50 40 30 20 10 0

Reversal

Figure 3: (a) Learning curves for path length in the Morris water maze place learning during acquisition (panel 1) and reversal (panel 2) for each genotype. Data are averaged across trials for each day (four trials/day, means  SEM). For the acquisition phase, double-knockout (double-KO) mice had significantly more difficulty finding the platform (overall group differences, see below) but eventually reached wild-type (WT) performance. For the reversal phase, double-KO mice had more difficulty finding the platform than WT mice on all days and never reached WT performance levels. ****P < 0.001 vs. WT. (b) Morris water maze place (hidden platform) performance during acquisition and reversal phases of learning for each genotype (mean  SEM) averaged across trials, days, and sex. Panels are for acquisition: path length (1), cumulative distance from the platform (2), latency (3), and first bearing (4) and for reversal: path length (5), cumulative distance (6), latency (7), and first bearing (8). Double-KO mice had more difficulty finding the platform than WT controls (path length, P < 0.01; cumulative distance, P < 0.01; first bearing, P < 0.0001). PDE1B–/– mice had a significantly increased first bearing to the platform (P < 0.004) compared with WT. Group sizes during acquisition: WT ¼ 48 (male ¼ 31, female ¼ 17), PDE1B–/– ¼ 42 (male ¼ 22, female ¼ 20), DARPP32–/– ¼ 27 (male ¼ 16, female ¼ 11), Double-KO ¼ 25 (male ¼ 14, female ¼ 11), Group sizes during reversal ¼ WT ¼ 44 (male ¼ 29, female ¼ 15), PDE1B–/–¼ 40 (male ¼ 20, female ¼ 20), DARPP32–/–¼ 27 (male ¼ 16, female ¼ 11), Double-KO ¼ 25 (male ¼ 14, female ¼ 11). *P < 0.10, **P < 0.05, ***P < 0.01, ****P < 0.001 vs. WT.

without the covariate, except that the DARPP32–/– path length effect was significant rather than being a trend as seen without the covariate included. This suggests that whatever the source of the difference on cued trials, it did not account for the place-learning deficits seen during spatial

546

navigation in the double-KO mice. Because the PDE1B–/– group showed effects on first bearing, a comparable ANCOVA was performed on these data. For acquisition, all three mutant genotype groups were significantly more off-course than WT (PDE1B–/– vs. WT, P < 0.01; DARPP32–/– vs. WT, Genes, Brain and Behavior (2006) 5: 540–551

PDE1B x DARPP32 double knockout mice

(a) 1

50

WT PDEIB–/– DARPP32–/– Double-KO–/–

(a) 2

WT PDEIB–/– DARPP32–/– Double-KO–/–

40 Percent time target quadrant

Average distance from target (cm)

60

40 30 20 10

30

20

0 60

10

(b)

(b) Percent time previous target quadrant

Percent time in annulus

50

40

30

20

10

0

Outer

Target

Inner

Figure 4: (a) Morris water maze place learning probe trial performance in the acquisition (panel 1) and reversal (panel 2) phases for each genotype. Data are average distance from the target (means  SEM). No significant differences among the genotypes were observed for the acquisition phase. For the reversal phase, double-knockout (double-KO) mice were significantly impaired relative to wild-type (WT) mice (P < 0.05). *P < 0.05 vs. WT. (b) Analysis of Morris water maze hiddenplatform probe trial performance on reversal by time in each of three annuli. Shown is the percent time in each annulus [outer, middle (target), and inner]. Compared with WT mice, the doubleKO mice spent a significantly larger proportion of their time in the outer annulus (P < 0.10) of the maze and significantly smaller proportion in the target annulus (P < 0.04). *P < 0.10, **P < 0.05, vs. WT.

P ¼ 0.05; double-KO vs. WT, P < 0.001; not shown). On reversal trials, only the double-KO group differed significantly from WT after covariate adjustment (P < 0.05).

40

30

20

10 1

2

3

4

5

6

Day

Figure 5: Percent time in target and previous target quadrants of the Morris water maze for each day of reversal learning (4 trials/day) with the smaller platform. (a) Percent time in the reversal target quadrant (NE). (b) Percent time in the previous (acquisition) target quadrant (SW). Time in the SW quadrant reveals the extent to which animals persisted in swimming to the original target quadrant after the platform was moved. Chance performance in each quadrant (25%) is shown by a horizontal line.

genotype–sex interaction was found on 5-HT concentrations in the hippocampus; however, a significant sex effect was found in which females averaged across all groups had lower 5-HT levels than males. No effect of either genotype, sex, or any interactions were found for 5-HIAA in the hippocampus.

Discussion Brain monoamine concentrations Monoamine concentrations in the neostriatum and hippocampus are shown in Table 1. A 3-between (PDE1B  DARPP32  sex) ANOVA showed no effects of either genotype, sex, or any interactions among these factors for concentrations of DA, DOPAC, 5-HT, and 5-HIAA in the neostriatum. Similarly, no effect of either genotype or Genes, Brain and Behavior (2006) 5: 540–551

We generated mice deficient for both PDE1B and DARPP32 to test the hypothesis that PDE1B plays a role in DARPP32dependent signaling. We previously showed that PDE1B–/– mice are initially hyperactive but return to WT levels, exhibit an exaggerated locomotor response to D-methamphetamine, and show an impairment of spatial Morris water maze acquisition (Reed et al. 2002). Here, we demonstrate that lack of

547

Ehrman et al. WT PDEIB–/– DARPP32–/– Double-KO–/–

Path length (cm)

500

1000

(a)

400

800

300

600

200

400

100

200

(b)

0

0 ACQ

REV

Figure 6: Path length during place learning for the acquisition and reversal phases of the Morris water maze adjusted (by ANCOVA) using performance on cued trials as the covariate (adjusted mean ± SEM; method of least squares by planned comparisons). *P < 0.05, **P < 0.01 vs. WT.

DARPP32 expression in PDE1B–/– mice (double-KO mice) normalizes the change in locomotor activity in response to –/– D-methamphetamine compared with PDE1B single-KO mice. The double-deletion effect was task specific, however, with different effects on anxiety and learning. Specifically, DARPP32–/– mice showed a less-anxious phenotype compared with double-KO mice in the elevated zero maze, and double-KO and PDE1B–/– mice exhibit different degrees of impairment of spatial learning and memory in the Morris water maze compared with WT or DARPP32–/– mutants. Drugs such as D-methamphetamine are known to increase locomotor activity by acting on dopaminergic and

serotonergic pathways. DARPP32 is a central component of dopaminergic and serotonergic signaling pathways in the striatum (Fienberg et al. 1998; Svenningsson et al. 2000, 2002a, 2003, 2004). Dopamine and serotonin-mediated adenylyl cyclase activation causes increased cAMP, leading to activation of PKA that phosphorylates DARPP32 at Thr-34 (Hemmings et al. 1984; Svenningsson et al. 2000, 2002a). When this occurs, activated DARPP32 inhibits protein phosphatase-1 (PP-1) which amplifies downstream cellular responses (Svenningsson et al. 2003). PDEs, such as PDE1B, by contrast, down-regulate these pathways by hydrolyzing cAMP to its corresponding 50 -AMP. In vitro evidence suggested a role for PDE1B in DARPP32-signaling cascades, as striatal brain slices from PDE1B–/– mice show increased phosphorylation of DARPP32 at Thr-34 in response to the D1 agonist, SKF-81297 (Reed et al. 2002). In the absence of PDE1B, cAMP levels presumably remain elevated after stimulation of dopaminergic and/or serotonergic pathways and therefore lead to prolonged phosphorylation of DARPP32 and activation of downstream cellular responses. This prolonged activated state in DARPP32dependent pathways may contribute to the hyperactive locomotor response to D-methamphetamine exhibited in PDE1B–/– mice. To test for a role for PDE1B in DARPP32 pathways in vivo, we investigated the effects of D-methamphetamine on locomotor behavior in DARPP32 and PDE1B double-KO mice. Deletion of DARPP32 prevented the hyperactive response of PDE1B–/– mice to D-methamphetamine. This provides some evidence of a role for PDE1B in DARPP32-dependent pathways. The data are limited by the fact that only one dose of methamphetamine could be assessed due to the limited number of double-KO mice that were available, and mice could not be retested at different methamphetamine doses

Table 1: Monoamine concentrations in selected brain regions in PDE1B and DARPP32 single- and double-knockout (Double-KO) mice vs. wild-type (WT) mice (ng/mg tissue) Group genotype PDE1B–/–

DARPP32–/–

Brain region

Sex

Mono-amine

n

WT

Neostriatum

Male Male Male Male

DA DOPAC 5-HT 5-HIAA

20 20 20 20

10.78 0.73 0.58 0.31

   

0.45 0.05 0.03 0.02

11.27 0.73 0.49 0.27

   

0.68 0.04 0.03 0.02

10.55 0.60 0.63 0.34

   

0.47 0.02 0.05 0.03

10.21 0.64 0.56 0.31

   

0.40 0.02 0.04 0.03

Neostriatum

Female Female Female Female

DA DOPAC 5-HT 5-HIAA

16 16 16 16

11.39 0.71 0.59 0.41

   

0.56 0.04 0.03 0.03

11.26 0.84 0.55 0.34

   

0.60 0.07 0.04 0.03

9.62 0.64 0.66 0.38

   

0.49 0.04 0.03 0.03

11.64 0.75 0.48 0.32

   

0.74 0.05 0.03 0.02

Hippocampus

Male Male Female* Female

5-HT 5-HIAA 5-HT 5-HIAA

10 10 10 10

0.91 0.37 0.86 0.41

   

0.04 0.01 0.03 0.02

0.85 0.38 0.77 0.36

   

0.03 0.02 0.05 0.01

0.91 0.40 0.82 0.41

   

0.04 0.02 0.03 0.02

0.85 0.37 0.85 0.38

   

0.04 0.03 0.03 0.02

Double-KO

*P < 0.05, mean across groups for males ¼ 0.88 vs. females ¼ 0.82.

548

Genes, Brain and Behavior (2006) 5: 540–551

PDE1B x DARPP32 double knockout mice

without inducing sensitization. Interestingly, the effect of the double deletion was to increase baseline locomotor activity to higher levels than seen in the PDE1B single-KO mice. The response of the double-KO mice to methamphetamine therefore represents a reduction in the relative response to the drug, i.e. both double-KO and PDE1B–/– mice exhibited similar locomotor activity peaks after methamphetamine, but the double-KO response was a smaller change relative to its prechallenge baseline than that seen in the PDE1B–/– group. The DARPP32–/– mice showed an apparent inhibition in locomotor activity in response to D-methamphetamine. It has previously been reported that DARPP32–/– mice exhibit an inhibited locomotor response following lower doses (10 mg/ kg) but not higher doses (20 mg/kg) of cocaine (Fienberg et al. 1998), but a clearly lower response to methamphetamine in the present study in DARPP32–/– mice was not seen. Cocaine mediates its locomotor-stimulating effects by inhibiting dopamine reuptake (Xu et al. 1994a, 1994b, 2000; Giros et al. 1996). D-Methamphetamine, however, not only inhibits reuptake but also causes release of dopamine, serotonin, and norepinephrine (Rothman et al. 2001), making it unclear whether the DARPP32–/– response was due to effects on these other neurotransmitters or some yet unidentified factor. Because anxiety/fearfulness may impact spatial learning, animals were tested in the elevated zero-maze. There were no differences in the number of open-area entries, but increases in time in open were seen. Had basal activity levels in this test been affected, it might be predicted that the double-KO mice would have had more open-area entries than mice of the other genotypes. However, this was not the case even though on the principle measure of anxiety, time in open, the DARPP32–/– mice spent more time in the open, suggesting reduced anxiety. No other group was significantly different from WT, including the double-KO mice, suggesting that the PDE1B deletion in the double-KO mice normalized the anxiety state of these mice in contrast to the DARPP32–/– single-KO mice. While limited by the use of only one test of anxiety, the data provide a preliminary indication that DARPP32–/– mice may be less anxious. The effect of the PDE1B deletion on the reduced anxiety of the DARPP32–/– mice may be the result of complex differences in expression of these genes in the amygdala. DARPP32 shows varied expression levels in different nuclei of the amygdala (Perez & Lewis 1992), whereas PDE1B shows low levels of expression in the central nucleus and higher expression in surrounding amygdaloid regions (Furuyama et al. 1994). How these pathways interact with respect to PDE1B and DARPP32 is not currently known, but the data in the zero-maze hint that PDE1B may be expressed in pathways within the amygdala that have different effects to those affected by the DARPP32 deletion. Dopaminergic and serotonergic signaling pathways have also been implicated in spatial learning and memory (Brandeis et al. 1989; D’Hooge & De Deyn 2001; McNamara & Skelton 1993). We used the Morris water Genes, Brain and Behavior (2006) 5: 540–551

maze as a test for spatial learning and memory and found that inactivation of DARPP32 and PDE1B in the double-KO mutant mice caused spatial navigation deficits in acquisition and reversal learning. Inactivation of only DARPP32 did not produce deficits in either phase of the Morris water maze, except in first bearing when adjusted for cued performance. Double-KO mice also showed deficits in cued learning. However, further analyses of spatial learning adjusted for cued performance by ANCOVA showed that the cued deficits did not account for the differences in spatial learning found in the double-KO mice or in first bearing differences seen in PDE1B–/– mice during acquisition. Analysis of the acquisition trials for the Morris water maze task indicates that double-KO mice had difficulty finding the platform because they swam in more circuitous paths to the target than WT and single-mutant mice. Although double-KO mice had more difficulty learning the location of the platform, they eventually were able to learn the task to levels of performance not different from WT or single mutants by day 6 of testing. This was confirmed by comparable probe trial performance. Shifting the platform to the opposite quadrant and reducing its size by 75% for the reversal phase revealed a more pronounced learning impairment in doubleKO mice but did not result in a deficit in DARPP32–/– mice. However, a deficit in PDE1B–/– mice on first bearing was again seen. A reversal learning deficit has been reported in DARPP32–/– mice on a different task, appetitive FR15 operant responding (Heyser et al. 2000), but the relationship of this effect to the present finding in the Morris water maze is unclear. The double-KO mice had a more difficult time learning the new location of the platform during reversal as indicated by the increase in path length, cumulative distance, and first bearing to the target compared with WT and singlemutant mice. Analysis of the learning curves for the reversal phase indicated that the double-KO mice were never able to learn the task as well as WT. The performance difference between the acquisition and reversal phases suggests that the double-KO mice had profound difficulty acquiring new associations. It should be noted that the reversal deficit involved two simultaneous platform changes (change of location and size). Therefore, it cannot be determined which change or whether the combination of both contributed most to the reversal phase group differences. Previously, we reported that the PDE1B–/– mice exhibit a spatial learning deficit in the acquisition phase of the hidden platform procedure in the Morris water maze (Reed et al. 2002), but not in reversal. In the current study, a significant difference in first bearing on acquisition and reversal and a trend on path length on acquisition was found in PDE1B–/– mice, indicating that PDE1B–/– mice did not know the direction of the platform as well as WT mice. The PDE1B–/– mice did not show acquisition differences, however, on other measures of Morris water maze performance as found previously. This may be due to procedural differences between experiments. For the present study, 6

549

Ehrman et al.

days of cued learning were administered before place learning was begun, whereas in the previous study, cued learning was administered after place learning. Cued learning prior to place learning may have provided a significant amount of additional ‘strategic’ information that facilitated performance during acquisition of the hidden platform condition in the present experiment as has been suggested previously (Whishaw 2004), i.e. that non-spatial pretraining allows animals to learn necessary response skills and general strategies that are prerequisites to spatial learning. Such prior training is sometimes sufficient to eliminate impairments in acquisition when a deficit is due to non-spatial factors (Bannerman et al. 1995; Cain et al. 1996; Hoh & Cain 1997; Saucier et al. 1996; Williams et al. 2002). Additional procedural differences between experiments for the PDE1B–/– mice in the Morris water maze were inclusion of straight channel swimming trials prior to Morris water maze previously, different ITIs (30 vs. 15 seconds), different number of probe trials (2 vs. 1), timing of the probe trial(s) (given on days 3 and 6 in the previous experiment vs. on day 7 herein), use of different start positions (use of cardinal vs. distal start positions), and different quadrants used for platform placement. Integration of the findings suggests that convergent expression of PDE1B and DARPP32 in the striatum may account for the double-KO block of the locomotor effects of –/– D-methamphetamine in PDE1B mice. For spatial learning, the situation is more complex. DARPP32 mRNA expression is relatively uniform through CA1-CA3 and the dentate gyrus of the hippocampus (Perez & Lewis 1992). By contrast, PDE1B mRNA expression is non-uniform in the hippocampus, showing high expression in the dentate gyrus and lower expression in CA1-CA2 (Furuyama et al. 1994). Spatial learning is strongly influenced by dentate gyrus disruption but also has significant contributions from other structures (Brandeis et al. 1989; Eichenbaum et al. 1990; McNamara & Skelton 1993), including striatum (Devan & White 1999; Devan et al. 1996, 1999; Furtado & Mazurek 1996), and prefrontal cortex (D’Hooge & De Deyn 2001). This may explain why the combined deletion of PDE1B and DARPP32 had a less predictable effect on spatial learning than was seen on locomotor behavior and suggests that simultaneous deletion of genes with different regional expression patterns may result in novel effects as seen here in the double-KO mice. The observation that PDE1B regulates the locomotor response to an indirect sympathomimetic stimulant through a DARPP32-dependent mechanism indicates that PDE1B plays an important role in striatal function and supports the possibility that PDE1B may be a useful target for the development of drugs that modulate basal ganglia function.

References Altar, C., Boyar, W. & Kim, H. (1990) Discriminatory roles for D1 and D1 dopamine receptor subtypes in the in vivo control of neostriatal cyclic GMP. Eur J Pharmacol 181, 17–21.

550

Bannerman, D.M., Good, M.A., Butcher, S.P., Ramsay, M. & Morris, R.G.M. (1995) Distinct components of spatial learning revealed by prior training and NMDA receptor blockade. Nature 378, 182–186. Barnes, N.M. & Sharp, T. (1999) A review of central 5-HT receptors and their function. Neuropharmacology 38, 1083–1152. Brandeis, R., Brandys, Y. & Yehuda, S. (1989) The use of the Morris water maze in the study of memory and learning. Int J Neurosci 48, 29–69. Cain, D.P., Saucier, D., Hall, J., Hargreaves, E.L. & Boon, F. (1996) Detailed behavioral analysis of water maze acquisition under APV or CNQX: contribution of sensorimotor disturbances to drug-induced acquisition deficits. Behav Neurosci 110, 86–102. D’Hooge, R. & De Deyn, P.P. (2001) Applications of the Morris water maze in the study of learning and memory. Brain Res Rev 36, 60–90. Devan, B.D., Goad, E.H. & Petri, H.L. (1996) Dissociation of hippocampal and striatal contributions to spatial navigation in the water maze. Neurobiol Learn Mem 66, 305–323. Devan, B.D., McDonald, R.J. & White, N.M. (1999) Effects of medial and lateral caudate-putamen lesions on place- and cueguided behaviors in the water maze: relation to thigmotaxis. Behav Brain Res 100, 5–14. Devan, B.D. & White, N.M. (1999) Parallel information processing in the dorsal striatum; Relation to hippocampal function. J Neurosci 19, 2789–2798. Eichenbaum, H., Stewart, C. & Morris, R.G.M. (1990) Hippocampal representation in place learning. J Neurosci 10, 3531–3542. Erneux, C., VanSande, J., Miot, F., Cochaux, P., Decoster, C. & Dumont, J. (1985) A mechanism in the control of intracellular cAMP level: the activation of a calmodulin-sensitive phosphodiesterase by a rise of intracellular free calcium. Cell Endocrinol 43, 123–134. Fienberg, A.A., Hiroi, N., Mermelstein, P.G., Song, W.J., Snyder, G.L., Nishi, A., Cheramy, A., O’Callaghan, J.P., Miller, D.B., Cole, D.G., Corbett, R., Haile, C.N., Cooper, D.C., Onn, S.P., Grace, A.A., Ouimet, C.C., White, F.J., Hyman, S.E., Surmeier, D.J., Girault, J.-A., Nestler, E.J. & Greengard, P. (1998) DARPP-32: regulator of the efficacy of dopaminergic neurotransmission. Science 281, 838–842. Francis, S.H., Turko, I. & Corbin, J.D. (2001) Cyclic nucleotide phosphodiesterases: relating structure and function. Prog Nucl Acid Res Mol Biol 65, 1–52. Furtado, J.C.S. & Mazurek, M.F. (1996) Behavioral characterization of quinolinate-induced lesions of the medial striatum: relevance for Huntington’s disease. Exp Neurol 138, 158–168. Furuyama, T., Iwashashi, Y., Tano, Y., Takagi, H. & Inahaki, S. (1994) Localization of 63-kDa calmodulin-stimulated phosphodiesterase mRNA in the rat brain by in situ hybridization histochemistry. Mol Brain Res 26, 331–336. Giros, B., Jaber, M., Jones, S.R., Wightman, R.M. & Caron, M.G. (1996) Hyperlocomotion and indifference to cocaine and amphetamine in mice lacking the dopamine transporter. Nature 379, 606–612. Hemmings, H.C. & Greengard, P. (1986) DARPP-32, a dopamineand adenosine 30 :50 -monophosphate-regulated phosphoprotein: regional, tissue, and phylogenetic distribution. J Neurosci 6, 1469–1481. Hemmings, H.C., Williams, K.R., Konigsberg, W.H. & Greengard, P. (1984) DARPP-32, a dopamine- and adenosine 30 :50 -monophosphate-regulated neuronal phosphoprotein. J Biol Chem 259, 14486–14490. Genes, Brain and Behavior (2006) 5: 540–551

PDE1B x DARPP32 double knockout mice Heyser, C.J., Fienberg, A.A., Greengard, P. & Gold, L.H. (2000) DARPP-32 knockout mice exhibit impaired reversal learning in a discriminated operant task. Brain Res 867, 122–130. Hoh, T.E. & Cain, D.P. (1997) Fractionating the nonspatial pretraining effect in the water maze task. Behav Neurosci 111, 1285–1291. Kirk, R.E. (1995) Experimental Design: Procedures for the Behavioral Sciences. Brooks/Cole Publishing Company, Pacific Grove. McNamara, R.K. & Skelton, R.W. (1993) The neuropharmacological and neurochemical basis of place learning in the Morris water maze. Brain Res Rev 18, 33–49. Monsma, F., Mahan, L., McVittie, L., Gerfen, C. & Sibley, D. (1990) Molecular cloning and expression of a D1 dopamine receptor linked to adenylyl cyclase activation. Proc Natl Acad Sci 87, 6723–6727. Perez, R.G. & Lewis, R.M. (1992) Regional distribution of DARPP-32 (dopamine- and adenosine 30 ,50 -monophosphateregulated phosphoprotein of Mr ¼ 32,000) mRNA in mouse brain. J Comp Neurol 318, 304–315. Polli, J.W. & Kincaid, R.L. (1992) Expression of a calmodulindependent phosphodiesterase isoform (PDE1B1) correlates with brain regions having extensive dopaminergic innervation. J Neurosci 14, 1251–1261. Polli, J.W. & Kincaid, R.L. (1994) Molecular cloning of DNA encoding a calmodulin-dependent phosphodiesterase enriched in striatum. Proc Natl Acad Sci 89, 11079–11083. Reed, T.M., Repaske, D.R., Snyder, G.L., Greengard, P. & Vorhees, C.V. (2002) Phosphodiesterase 1B knock-out mice exhibit exaggerated locomotor hyperactivity and DARPP-32 phosphorylation in response to dopamine agonists and display impaired spatial learning. J Neurosci 22, 5188–5197. Rothman, R.B., Baumann, M.H., Dersch, C.M., Romero, D.V., Rice, K.C., Carroll, F.I. & Partilla, J.S. (2001) Amphetaminetype central nervous system stimulants release norepinephrine more potently than they release dopamine and serotonin. Synapse 39, 32–41. Saucier, D., Hargreaves, E.L., Boon, F., Vanderwolf, C.H. & Cain, D.P. (1996) Detailed behavioral analysis of water maze acquisition under systemic NMDA or muscarinic antagonism: nonspatial pretraining eliminates spatial learning deficits. Behav Neurosci 110, 103–116. Shepherd, J.K., Grewal, S.S., Fletcher, A., Bill, D.J. & Dourish, C.T. (1994) Behavioural and pharmacological characterization of the elevated ‘zero-maze’ as an animal model of anxiety. Psychopharmacology 116, 56–64. Svenningsson, P., Fienberg, A.A., Allen, P.B., Le Moine, C., Lindskog, M., Fisone, G., Greengard, P. & Fredholm, B.B. (2000) Dopamine D1 receptor-induced gene transcription is modulated by DARPP-32. J Neurochem 75, 248–257. Svenningsson, P., Nishi, A., Fisone, G., Girault, J.-A., Nairn, A.C. & Greengard, P. (2004) DARPP-32 – an integrator of neurotransmission. Annu Rev Pharmacol Toxicol 44, 269–296.

Genes, Brain and Behavior (2006) 5: 540–551

Svenningsson, P., Tzavara, E.T., Carruthers, R., Rachleff, I., Wattler, S., Nehls, M., McKinzie, D.L., Fienberg, A.A., Nomikos, G.G. & Greengard, P. (2003) Diverse psychotomimetics act through a common signaling pathway. Science 302, 1412–1415. Svenningsson, P., Tzavara, E.T., Liu, F., Fienberg, A.A., Nomikos, G.G. & Greengard, P. (2002a) DARPP-32 mediates serotonergic neurotransmission in the forebrain. Proc Natl Acad Sci 99, 3188–3193. Svenningsson, P., Tzavara, E.T., Witkin, J.M., Fienberg, A.A., Nomikos, G.G. & Greengard, P. (2002b) Involvement of striatal and extrastriatal DARPP-32 in biochemical and behavioral effects of fluoxetine (Prozac). Proc Natl Acad Sci 99, 3182–3187. Upchurch, M. & Wehner, J.M. (1988) Differences between inbred strains of mice in Morris water maze performance. Behav Genet 18, 55–68. Vallone, D., Picetti, R. & Borrelli, E. (2000) Structure and function of dopamine receptors. Neurosci Biobehav Res 24, 125–132. Whishaw, I.Q. (2004) Posterior neocortical (visual cortex) lesions in the rat impair matching-to-place navigation in a swimming pool: a reevaluation of cortical contributions to spatial behavior using a new assessment of spatial versus non-spatial behavior. Behav Brain Res 155, 177–184. Williams, M.T., Moran, M.S. & Vorhees, C.V. (2003) Refining the critical period for methamphetamine-induced spatial deficits in the Morris water maze. Psychopharmacology 168, 329–338. Williams, M.T., Vorhees, C.V., Boon, F., Saber, A.J. & Cain, D.P. (2002) Methamphetamine exposure from postnatal days 11–20 causes impairments in both behavioral strategies and spatial learning in adult rats. Brain Res 958, 312–321. Xu, M., Guo, Y., Vorhees, C.V. & Zhang, J. (2000) Behavioral responses to cocaine and amphetamine administration in mice lacking the dopamine D1 receptor. Brain Res 852, 198–207. Xu, M., Moratalla, R., Gold, L., Hiroi, N., Koob, G., Graybiel, A. & Tonegawa, S. (1994a) Dopamine D1 receptor mutant mice are deficient in striatal expression of dynorphin and in dopaminemediated behavioral responses. Cell 79, 729–742. Xu, M., Moratalla, R., Gold, L., Hiroi, N., Koob, G., Graybiel, A. & Tonegawa, S. (1994b) Elimination of cocaine-induced hyperactivity and dopamine-mediated neurophysiological effects in dopamine D1 receptor mutant mice. Cell 79, 945–955. Yan, C., Bentley, J.K., Sonnenburg, W.K. & Beavo, J.A. (1994) Differential expression of the 61 kDa and 63 kDa calmodulindependent phosphodiesterases in the mouse brain. J Neurosci 14, 973–984.

Acknowledgments This work was supported by NIH-training grant ES07051 (LAE) and research grants DA014269 (MTW), DA006733 (CVV), DA007427 (GAG), and MH40899 (PG), DA10044 (PG), DOD/ USAMRAA (DAMD17-02-1-0705 (PG), The Michael Stern Parkinson’s Research Foundation (PG), The Peter Jay Sharp Foundation (PG), and The Simons Foundation (PG).

551