SYNAPSE 11:35-46 (1992)

Neonatal 6-OHDA Lesions Differentially Affect Striatal D1 and D2 Receptors BETHANY S. NEAL AND JEFFREY N. JOYCE Departments of Psychiatry and Pharmacology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 191 04

KEY WORDS

Neonatal dopamine depletion, Patchlmatrix, Striatum, Dopamine receptors, Quantitative autoradiography

ABSTRACT Lesions to the dopamine (DA) system in early postnatal development have different behavioral consequences compared to lesions made in adulthood. Intrastriatal injections of the neurotoxin 6-hydroxydopamine (6-OHDA)on the day of birth (PO)or postnatal day 1 (Pl) produce a selective supersensitivity to D1 receptor agonists and a subsensitivity to D1 antagonists (Neal and Joyce, 1991a). In this paper, we describe the long-term effects of early DA loss on DA receptor regulation. Pups received bilateral intrastriatal injections of the neurotoxin 6-OHDA (4 kg per striatum) on PO or P1. Adult rats were killed at 90 days of age and the brains were processed for quantitative autoradiography (QAR) or tyrosine hydroxylase (TH) immunocytochemistry. Cohorts were tested for the behavioral responses to the selective D1 receptor agonist SKF38393 (10 mgkg). Neonatally lesioned rats exhibited increases in abnormal perioral movements in response to D1 receptor stimulation. There was a heterogenous and patchy loss ( 4 6 5 0 % ) of [3H]mazindol binding to high-affinity DA uptake sites (a marker of DA terminal density) and a similar loss of TH-like immunoreactivity within the striata of the neonatally lesioned rats. There was also a reduction in the number of popioid receptor patches (labelled with [3H]naloxone),a marker for the striatal patch compartment, and a similar patchy loss of D1 binding sites (labelled with [3H]SCH23390).The binding of [3H]spiroperidol to D2 sites was not altered. This is in contrast to the changes observed following adult 6-OHDA lesions, wherein there is a significant increase in the number of D2 binding sites (Joyce, 1991a,b). The results are discussed with respect to the behavioral consequences of neonatal lesions and the differences between neonatal and adult lesions. 0 1992 Wiley-Liss, Inc.

sioned rats do exhibit some long-lasting deficits in cogINTRODUCTION nitive functioning and skilled motor control (Whishaw Insults to the central nervous system early in develet al., 19871, as well as problems responding to acute opment have different neurochemical andlor behavioral homeostatic imbalances (Bruno et al., 1986; Rogers and consequences compared to adult lesions. The mechaDunnett, 1989a). Most importantly, neonatal lesions nisms responsible for the differential plasticity of the produce animals that are highly sensitive to the admindeveloping vs. the mature nervous system are not well of the DA precursor L-dihydroxyphenylalaistration understood. Over the past 15 years, a behavioral model nine (L-DOPA) or the D1 receptor agonist SKF’38393. for examining the plasticity of dopamine (DA) systems, L-DOPA was reported to induce self-mutilatory behavinvolving the administration of the neurotoxin 6-hydroxydopamine (6-OHDA)to damage the mesostriatal ior (SMB) in adult rats that received extensive DA leDA system, has been well characterized. In adult rats, sions on postnatal day 5 (P5) (Breese et al., 1984a,b). extensive lesions of DA neurons produce profound be- The SMB can be blocked by the D1 antagonist havioral impairments, such as aphagia, adipsia, akine- SCH23390, but only partially by the D2 antagonist sia, and sensory neglect, which mimic changes observed haloperidol. Adult-lesioned rats are more sensitive to in Parkinson’s disease (Marshall et al., 1974; Unger- the mixed D1/D2 agonist apomorphine and the D2 agostedt, 1971; Zigmond and Stricker, 1973). In contrast, nist quinpirole (Breese et al., 198413)and do not engage extensive neonatal 6-OHDA lesions do not induce a Parkinsonian-like syndrome (Bruno et al., 1984, 1987; Received August 8,1991; accepted in revised form September 10,1991 Lytle et al., 1972; Smith et al., 1973). Neonatally le0 1992 WILEY-LISS, INC.

36

B.S. NEAL AND J.N. JOYCE

in SMB. Thus, neonatally lesioned rats are more sensitive to D1 receptor activation, whereas adult-lesioned rats have a greater selective sensitivity for D2 receptor activation (Breese et al., 1985a,b). To explore the issue of selective behavioral sensitivity to D1 receptor-selective drugs, we gave bilateral intrastriatal6-OHDA injections to rat pups on the day of birth (PO) or P1 (Neal and Joyce, 1991a).At 90 days of age, the long-term functional consequences of these lesions were examined. Neonatally lesioned rats exhibited self-biting behavior following treatment with L-DOPA. The rats also exhibited increases in abnormal perioral movements, as well as some self-biting, following treatment with SKF38393; however, their responses to apomorphine and quinpirole were no different from the controls. The lesioned rats were also subsensitive to the cataleptic effects of the D1 antagonist SCH23390, but not of the D2 antagonist haloperidol. Analysis of the striatum with quantitative autoradiography (QAR) showed that these rats had sustained a patchy loss of 13Hlmazindol-labelledDA uptake sites. Intrastriatal 6-OHDA injections on PO or P1 cause a selective lesion of the DA system innervating the patch compartment of the striatum (Gerfen et al., 1987b). This suggests that our early neonatally lesioned rats received a lesion limited to the patch-directed DA system, but exhibited functional changes similar to those reported for rats with more extensive neonatal lesions. However, there appear to be more selective effects on D1 responsitivity with the intrastriatal lesions (e.g., rats with extensive neonatal lesions are subsensitive to both D1 and D2 antagonists (Bruno et al., 1985; Duncan et al., 1987;Johnson and Bruno, 1990). The mechanisms that underlie the differential behavioral effects of neonatal and adult 6-OHDA lesions remain unclear. Past work has focused on the possibility that residual DA in the striatum of the neonatally lesioned rats may contribute to the behavioral sparing or that another transmitter system [e.g., serotonin (5-HT)Iis substituting for the lost DA system, whereas few investigators have explored the contribution of differential DA receptor regulation. There is considerable information available regarding the development (Rao et al., 1991; Murrin et al., 1985; Murrin and Zeng, 1989) and organization of the pre- and postsynaptic components of the DA system (Boyson et al., 1986; Fallon and Moore, 1978;Gerfen et al., 1987a;Joyce, 1991a),as well as their regulation following lesions of the mature DA system (Marshall et al., 1990; Joyce, 1991a,b). Regulation of DA receptors following neonatal lesions has received less attention, and those few reports have not reached a consensus. Breese and associates (Breese et al., 1986, 1987; Duncan et al., 1987) found no changes in the binding of either [3Hlspiroperidol (D2 binding sites) or [3HlSCH23390 (D1 binding sites) to striatal membranes from adult rats that had received intracisternal(1C) 6-OHDA injections as neonates. Re-

cently, Dewar and associates (1990) reported that intracerebroventricular (ICV) injections of 6-OHDA on P3 cause a small increase in D2 receptor density, with no change in D1 binding in adulthood. However, another laboratory (Gelbard et al., 19901,measuring binding on P21, found a large loss of D1 binding sites (up to 82 percent and no change in D2 binding, following a similar type of neonatal lesion. The loss of D1 receptors was reversed by a subsequent treatment with SKF38393 from P6-Pl8. Caboche and associates (Caboche et al., 1991) reported that intrastriatal injections of 6-OHDA on P2 alter the expression of p-opioid receptor patches, but not D1 receptors, when examined on P7 or P21. On the other hand, another laboratory (Broaddus and Bennett, 1990)reported that D1 receptor number was increased following bilateral intrastriatal injections of 6-OHDA on P2 or P3; however, the authors administered a large dose of 6-OHDAin a large volume (20 pg in 20 pl per striaturn). The purpose of this study was to examine the effects of the neonatal lesions on the pre- and postsynaptic components of the dopamine system in animals that show selective changes in D1 behavioral sensitivity. The use of quantitative autoradiography (QAR)to examine D1 and D2 receptors, p-opioid receptors (the prototypical patch marker), and DA uptake sites (as a measure of terminal density), provided a high level of anatomical resolution to examine striatal organization following the loss of DA input to the patches and allowed for the quantification of receptor numbers within discrete regions. Tyrosine hydroxylase (TH) immunocytochemistry was used to examine the extent of the DA lesions, as well as marking the patch/matrix boundaries in the neonatally lesioned rats (Gerfen et al., 1987bl

MATERIALS AND METHODS Subjects Timed-pregnant Sprague-Dawley rats (Charles River) were received on day 14 or 15 of gestation and housed individually in plastic cages with food and water freely available. Animals were maintained on a 12 h light-dark cycle (lights on at 07.00 h) in a colony room at 22-24°C. Beginning on day 19, cages were checked twice daily at approximately 10.00 and 17.00 h for the presence of pups. On PO or P1, all pups were removed from their mothers, randomized and redistributed to the nursing dams with a litter size of 10 (equal numbers of males and females). Litters remained with their foster mothers until weaned at 23 days of age. After weaning, rats were group housed by sex ( 3 4 per cage) and maintained with food and water freely available. On PO o r P1, 1 male and 1 female from each litter were left alone (noninjected controls), while the remaining pups received intraperitoneal (IP) injections of 25 mgkg desmethylimipramine (DMI, Sigma) to protect noradrenergic neurons. Thirty minutes later, rats were

EARLY DOPAMINE LOSS AND RECEPTOR REGULATION

anesthetized on ice and then received bilateral intrastriatal injections of 6-OHDA HBr (4 pghtriatum, as the free base, Sigma) or vehicle (2 ~1 per striatum of 0.9% NaCl containing 0.1% ascorbate). Intrastriatal injections were made with a 30-gauge needle attached to a Hamilton syringe, according to the method of Gerfen and coworkers (1987b). Animals were then numbered by toe clip, warmed, and returned to their dams. Weights were recorded every other day until weaning and then at weekly intervals until testing began. Quantitative receptor autoradiography At 90 days of age, rats were decapitated and the brains were quickly removed. The brains were frozen intact in isopentane for 2 min at -20°C. Coronal sections 20 p thick were cut in a cryostat at - 12 to - 15°C through the striatum [plates 9-23 from the atlas of Paxinos and Watson (198613. Sections were thawmounted onto gelatin-coated slides and then dehydrated for 2 h a t 0 4 ° C under vacuum. After dehydration, sections were frozen and stored at -20°C until binding assays were performed. Serial sections were used to determine both total and nonspecific binding for p-opioid receptors, D1 and D2 receptors, and high-affinity DA uptake sites. For TH immunocytochemistry, sections were thaw-mounted onto the slides, quickly dried under a stream of cool air, and frozen in the bottom of the cryostat. They were then stored a t -20°C until further processing. p-Opioid receptors were labelled with [3H]naloxone according to a procedure adapted from Herkenham and Pert (1981). Slide-mounted tissue sections were brought t o 0°C 30 min prior to incubation. After a 5-min preincubation a t 4°C in 50 mM Tris-HC1buffer (pH 7.4) plus 100 nM NaCl (Tris-saline), tissue sections were incubated for 60 rnin at 4°C in Tris-saline buffer containing 2.5 nM [3Hlnaloxone (Dupont; 45.1 Ci/mmol). Binding in the presence of 10 pM naloxone was used to define nonspecific binding. The incubation was followed by two 30-sec rinses in ice-cold phosphate-buffered saline. For labelling of D1 binding sites with [3H]SCH23390, a method previously published by this group was utilized (Joyce, 1991a,b; Marshall et al., 1989). Slidemounted tissue sections were brought to 0°C 30 min prior to incubation, and to room temperature just prior to incubation. Sections were preincubated for 5 min in 50 mM Tris-HC1 buffer (pH 7.41, containing 120 mM NaC1, 50 mM KC1, 20 mM CaCl,, and 10 mM MgClz (TPI), followed by a 60-min incubation at room temperature in TPI buffer with 1mM ascorbic acid (TPIA).The incubation media included 1.0 nM [3H]SCH23390(Dupont; 70.7 Cilmmol) and 40 nM ketanserin. Binding in the presence of 5 pM (+ )-butaclamol defined nonspecific binding. The incubation was followed by two 20-sec rinses in ice-cold TPIA buffer.

37

For labelling of D2 sites with [3Hlspiroperidol(Joyce, 1991a,b; Joyce and Marshall, 19871, tissue sections were brought to 0°C 30 min prior to incubation, and to room temperature just prior to incubation. Sections were preincubated for 5 rnin at room temperature, followed by a 55-min incubation at 37°C in 50 mM TPIA buffer (pH 7.11, containing 0.7 nM [3H]spiroperidol (Amersham; 94.6 Ci/mmol). Ketanserin (40 nM) was added to prevent spiroperidol binding to 5-HT2binding sites. Nonspecific binding was defined by the addition of 1 pM (+I-butaclamol to the incubation media. Following incubation, tissue sections were rinsed twice for 20 sec in ice-cold TPIA buffer. Presynaptic high-affinity DA uptake sites were labelled with [3Hlmazindol, as described in detail elsewhere (Joyce, 1991a,b; Marshall et al., 1989). Slidemounted tissue sections were brought to 0°C 30 min prior to incubation. After preincubation for 5 min in ice-cold buffer (50 mM Tris-HC1plus 300 mh4 NaCl and 5 mM KC1) plus 300 nM DMI, tissue sections were incubated for 40 min at 0°C in buffer plus DMI, containing 15 nM [3Hlmazindol (Dupont; 15.8 Ci/mmol). Binding in the presence of 30 pM benztropine defined nonspecific binding. Following incubation, tissue sections were rinsed twice for 3 min in ice-cold buffer. After the incubation and buffer rinse procedures, the tissue sections were rinsed for 10 sec in ice-cold distilled water to remove buffer salts. The tissue sections were then dried on a 55°C hot plate after the aspiration of excess fluid around the sections, followed by further drying overnight in a sealed plastic bag containing Drierite. The tissue sections and standards (ARC, Inc.) were loaded into light-proof X-ray cassettes and apposed to tritium-sensitive film (Amersham Hyperfilm) for the appropriate length of time (2 months[3Hlnaloxone; 15 day~-[~HISCH23390; 19 days[3Hlspiroperidol;24 day~-[~HImazindol).The film was then developed with cold Kodak GBX developer (5 min) and fixed with Kodak GBX fixer (5 min). Autoradiographic images were digitized and linearized with the aid of a DUMAS (Drexel Unix-based Microcomputer Analysis System, Drexel University, Philadelphia, PA) image analyzer. The mean optical density of each brain region of interest was converted to a value of fmol/mg protein. A standard curve for 13H]based on calibrations of the low-activity L3H1-plastic standards to 13Hlisoleucine-tissue mash standards was used. This allowed for the transformation of the autoradiograph so that the gray values of each pixel were a linear function of the quantity of radioligand bound per milligram of protein. The value of the specific binding for each region was obtained by subtracting the value obtained for nonspecific binding from that of total binding. The caudate putamen (CPu) and the nucleus accumbens septi (NAS) were analyzed for radioligand binding. The CPu was divided into quadrants for more sensitive regional analysis: dorsolateral (DL), dorsomedial

38

B.S. NEAL AND J.N. JOYCE

(DM), ventrolateral (VL), and ventromedial (VM) CPu (see Figs. 4 and 5). The mean density of binding sites (+SEM) was calculated from determinations on 4 noninjected rats, 10 vehicle-treated rats, and 10 neonatally lesioned rats. Only rats with a successful lesion (at least 1 striatal quadrant with a significant loss of [3H]mazindolbinding (Neal and Joyce, 1991a) were included in the analyses. TH immunocytochemistry Frozen sections were fixed on the slides in a solution of 0.1 M phosphate buffer containing 4% paraformaldehyde. TH-like immunoreactivity was detected by the avidin-biotin immunoperoxidase technique on slide, according to a modification of a protocol already described (Schneider and Markham, 1986). Briefly, sections were rinsed twice for 10 min in KF'BS and then reacted overnight at 4°C with 2% normal goat serum and 0.3% Triton-X. Sections were then rinsed several times in KF'BS and reacted for 48 h at 4°C with anti-TH antibody, diluted 1:500. Sections were rinsed, reacted with a biotinylated secondary antibody, rinsed, and then reacted with an avidin and biotinylated horseradish peroxidase complex. After rinsing, peroxidase reaction product was visualized with 0.05% DAB and 0.04% H,Oz.

100

**

1

LESION

9-10

19-20 29-30 39-40 49-50 59-60

TIME INTERVAL (MIN) Fig. 1. The incidence of abnormal perioral movements in adult rats following treatment with the D1 agonist SKF38393. Rats were observed during 1-min periods at 10-min intervals for 60 min immediately following drug treatment. Data represent the mean (?SEM) incidence for each 1-min period. Neonatally lesioned rats (bilateral intrastriatal 6-OHDA on PO or P1) exhibited increased abnormal perioral movements following SKF38393 (10 mgkg). *P < 0.05, **P s 0.01, significantly different from the nonlesioned rats.

Behavioral testing Data analysis To ensure that the lesioning paradigm was having For locomotor activity and the observational behavthe effects on D1 behavioral sensitivity as previously reported (Neal and Joyce, 1991a), cohorts of the rats iors, data were analyzed by repeated measures analyused for the QAR experiments were injected with the ses of variance (ANOVA), with time interval as the D1 agonist SKF38393 (10 mgkg, sc) a t 90 days of age. repeated measure. Differences in [3H]radioligandbindRats were immediately placed into large plexiglass ob- ing were determined by a repeated measures ANOVA, servation chambers (51 x 51 x 38 cm) for a 60-min ses- with the region of the striatum as the repeated measion. Rats were observed for 1-min periods at 10-min sure. Individual comparisons were made by the Newintervals throughout the session by an observer un- man-Kuels range test. Significance was defined as aware of which rats were lesioned. Each 1-min period P 0.05. was divided into 15-sec blocks. The presence of the folRESULTS lowing behaviors during each 15-sec block was noted: Neonatal lesioning had no effect on weight gain, nor sniffing, rearing, self-biting, grooming, licking, gnawing, paw treading, yawning, head nodding, jumping, did it affect the time of eye opening or have any other and abnormal perioral movements (jaw tremor, tongue profound effects (data not shown). As adults, the neonaprotrusions, vacuous chewing). The percent incidence tally lesioned rats (n = 10) engaged in more abnormal of each behavior during every 1-min period was then perioral movements following treatment with determined according to a method derived from Breese SKF38393, as compared to the controls (F = 7.1; and associates (Breese et al., 1984a,b, 1985,1987;Dun- P < 0.05; Fig. 1). The data for the noninjected rats can et al., 1987).For example, if a behavior was present (n = 5) were combined with data for the vehicle-induring a single 15-secblock out of the 1-min period, this jected rats (n = 91, as there were no significant differrepresented an incidence of 25%, if it occurred during 2 ences between the 2 control groups in their responses to of the blocks it represented an incidence of 50%, etc. SKF38393. There were significant differences between During a separate 2-min period out of every 10 min, nonlesioned and lesioned rats at 20 and 30 min postinlocomotor activity was also noted. The wire-mesh floor jection (P < 0.05). At 50 min postinjection, lesioned rats of the observation chamber was divided into 4 equal had a 90% incidence of oral dyskinesias, whereas only quadrants. The number of times the rat moved at least 48% of the nonlesioned rats exhibited oral dyskinesias 2 paws into a different quadrant was recorded as a line at that time (P < 0.01). Locomotor activity, or the incicrossing, and the total number of line crosses for each dence of any other behavior, did not differ between non2-min period was determined. lesioned and lesioned rats in response t o SKF38393

EARLY DOPAMINE LOSS AND RECEPTOR REGUJATION

39

A Fig. 2. The effect of neonatal 6-OHDA lesions on TH-like immunoreactivity in the striatum. Rats received bilateral intrastriatal injections (4 kg per striatum) on PO or P1,and were killed a t 90 days of age. There was homogeneous staining for TH-like immunoreactivity in the striata of the nonlesioned rats (A), whereas there was a patchy loss of TH-like immunoreactivity in the striata of the neonatally lesioned rats (B). The NAS was less affected by the 6-OHDA, and the olfactory tubercles appeared to be unaffected.

(data not shown), although the lesioned rats did show some evidence of self-biting behavior. As shown in Figure 2, the intact adult striatum shows a dense, homogenous staining for TH-like immunoreactivity. Following neonatal intrastriatal6-OHDA lesions, there was a patchy loss of immunoreactivity in the adult striatum. This loss of immunoreactivity appeared to be greatest in the dorsal striatum, although lack of staining was also apparent in the ventral striatum. The olfactory tubercles were not affected. The degree of loss of DA afferents can also be measured by quantifying the loss of [3Hlmazindol-labelled high-affinity DA uptake sites (Joyce, 1991a,b; Marshall et al., 1989). In the nonlesioned animals (noninjected and vehicle-treated), the density of [3Hlmazindol sites was higher in the lateral CPu than in the medial CPu or NAS, although binding was fairly dense throughout the entire CPu (Figs. 4-6). The rostral CPu also had somewhat higher levels of [3Hlmazindol binding (Fig. 41, as compared to the more caudal CPu (Fig. 5). There were

no differences between the noninjected and the vehicletreated rats. Following intrastriatal6-OHDA administration, there were significant decreases in [3H]mazindo1 binding in all quadrants of the rostral CPu. A repeated measures ANOVA revealed a significant treatment effect (F = 24.0; P < 0.01) and a significant interaction between treatment and the CPu quadrant (P < 0.01). The loss appeared “patchy” in nature, with the largest decrease (43% loss) in the dorsomedial quadrant of the left striatum (Fig. 4). The ventrolateral quadrants were the least affected. There was also about a 15% loss of DA uptake sites in the NAS. Overall [3H]mazindolbinding was decreased about 25% in the striatum of neonatally lesioned rats as compared to the nonlesioned rats. A similar pattern of loss was seen in the caudal CPu (Fig. 5), although the degree of loss was less. For example, there was a 29% decrease in [3H]mazindol binding in the dorsomedial quadrant of the left striatum, with 5-15% decreases in the lateral quadrants, and with an overall loss of about 13%.A

B.S. NEAL AND J.N. JOYCE

40 6

5

Ei X

TI

0

3000 7,

NON-LESION

gc 4 a

83 M w

!42

z

1

n

8

2500

4

Fig. 3. The effect of neonatal 6-OHDA lesions on the number of I*-opioidreceptor patches, the prototypical marker of the striatal patch compartment. There was a significant decrease in the number of I*-opioid patches (labelled with [3Hlnaloxone, 2.5 nM) in the rostral CPu of the neonatally lesioned rats. *P 5 0.05, significantly different from the nonlesioned rats. L = left; R = right. DL = dorsolateral; DM = dorsomedial; VL = ventrolateral.

repeated measures ANOVA did reveal a significant treatment effect (F = 6.43, P < 0.051, and a significant treatment by quadrant interaction (P < 0.05). Distinct patches of popioid receptor binding were observed throughout the rostral striatum of the nonlesioned animals (Figs. 3 and 6). The number of patches was fairly homogeneous throughout the striatum, except that there were fewer patches in the ventromedial striatum. The number of patches decreased and were less distinct in the caudal striatum; therefore, quantification is only reported for the rostral striatum. High levels of binding were also apparent in the subcallosal streak, the medial NAS, and the olfactory tubercles. The density of binding was approximately 3-4 times higher in the patches than in the surrounding matrix. Although the density of [3Hlnaloxone labelling within the patches and throughout each striatal quadrant was determined, only the data for the number of p-opioid receptor patches within each quadrant are reported, as this appears to be the most sensitive measure of a neonatal lesion effect (personal observations). As has been previously reported (Gerfen et al., 1987b), successful neonatal 6-OHDA lesions caused a significant decrease (50-60%) in the number of p-opioid receptor patches in the dorsolateral and dorsomedial striatum (Fig. 3). The density of binding in the remaining patches was also somewhat less than that of the controls (data not shown).

Fig. 4. The effect of neonatal 6-OHDA lesions on DA systems within the rostral striatum. The mean specific binding (*SEMI of the following radioligands in various quadrants of the rostral CPu and the nucleus accumbens septi (NAS)is depicted: (A) 13HIMazindol(15 nM), demonstrating a heterogeneous loss of DA uptake sites in the CPu of neonatally lesioned rats; (B) L’HHISCH23390 (1.0 nM), demonstrating a heterogeneous loss of D1 binding sites in the CPu of neonatally lesioned rats; and (C) [3Hlspiroperidol(0.7 nM) in the presence of ketanserin to block 5-HT, sites, demonstrating no effect of neonatal 6-OHDA administration on D2 binding sites. L = left hemisphere, R = right hemisphere, DL = dorsolateral, DM = dorsomedial, VL = ventrolateral, and VM = ventromedial. *P s 0.05, significantly different from the nonlesioned rats.

EARLY DOPAMINE LOSS AND RECEPTOR REGULATION

3000

2

1

2500

Fig. 5 . The effect of neonatal 6-OHDA lesions on DA systems within the caudal striatum. The mean specific binding ( 2 SEMI of L3H1mazindo1 to high-affinity DA uptake sites (A),F'HISCH23390 to D1 binding sites (B), and [3H]spiroperidol to D2 binding sites (C) in various quadrants of the caudal CPu are illustrated. For abbreviations, see Fig. 4. *P s 0.05, significantly different from the nonlesioned rats.

In nonlesioned adults, [3H]SCH23390binding to D1 sites was fairly homogenous throughout the CPu, with lower levels of binding in the NAS (Figs. 4-6). As with DA uptake sites, D1 binding was higher in the rostral CPu (Fig. 41, as compared to the caudal CPu (Fig. 5). There was a significant decrease in D1 binding in the rostral CPu of animals having received neonatal 6-OHDA lesions (Fig. 4). A repeated measures ANOVA

41

revealed a significant treatment effect (F = 6.42, P < 0.05), with no significant interactions. Decreases of l0-15% were observed in the medial quadrants (Fig. 41, with decreases of 5-10% in the lateral quadrants. There was also a 9% decrease in D1 binding in the NAS; however, this loss did not reach statistical significance. In the rostral CPu (Fig. 5), the only significant decrease in D1 binding was found in the dorsomedial quadrants (13%and 16% for the left and right striatum, respectively). As can be seen in Figs. 4-43, D2 binding in the striata of nonlesioned rats showed the typical lateral to medial gradient, with the highest levels in the ventrolateral quadrants (500-600 fmol/mg protein), as compared to the ventromedial quadrants (400-450 fmoYmg protein). A similar pattern was observed in the dorsal quadrants. Neonatal 6-OHDA lesions had no significant effect on D2 binding in the striatum (F = 2.76, p = . l l for the rostral CPu, and F = .06, p = .81 for the caudal CPu).

DISCUSSION Following intrastriatal injections of 6-OHDA early in postnatal development, rats are supersensitive to selective effects of the D1 agonist SKF38393. As previously reported (Neal and Joyce, 1991a) and as confirmed in this study, these early neonatally lesioned rats have a higher incidence of abnormal perioral movements compared to controls; however, the incidence of various other behaviors is not altered. These rats also engage in self-biting behavior following treatment with DA agonists, but do not show any differences in the degree of locomotor activity. Thus, intrastriatal6-OHDA administration on PO or P1 appears to have more selective effects than does 6-OHDA administration IC or ICV during the first week of life, as those rats exhibit a much broader range of behavioral changes (Breese et al., 1984a,b, 1985a,b, 1987). Widespread DA depletion during a more prolonged or later (after P5) neonatal period also causes rats to have a long-term subsensitivity to both D1 and D2 antagonists, as the rats display decreased cataleptic responses to both SCH23390 and haloperidol (Bruno et al., 1985; Duncan et al., 1987; Johnson and Bruno, 1990). However, early neonatal intrastriatal lesions result in a subsensitivity to the cataleptic effects of the D1 antagonist SCH23390, with no altered response to the D2 antagonist haloperidol (Neal and Joyce, 1991a). Thus, the early neonatally lesioned rats exhibit functional changes similar to those reported for rats with more extensive neonatal lesions (vide infra); however, our lesioning paradigm appears t o have more selective effects on D1 responsitivity. Intrastriatal lesions of DA neurons following 6-OHDA administration on PO or PI induced permanent changes in DA systems within the striatum. The neonatal 6-OHDA lesions caused a significant decrease

B.S. NEAL AND J.N. JOYCE

42

VEHICLE

6-OHDA

Fig. 6. Photomicrographs of the autoradiographic distribution of the binding of various radioligands to sections through the striatum of a vehicle-treated (nonlesion) rat and a rat that received bilateral intrastriatal 6-OHDA injections (4 pg per striatum) on PO or PI. The distribution of the following radioligands is illustrated: 13Hlmazindol to high-affinity DA uptake sites (A,B);L'HJnaloxone to p-opioid receptors (C,D); I"HlSCH23390 to D1 receptors (E,F); and L3Hlspiroperidolto D2 receptors (G,H).

EARLY DOPAMINE LOSS AND RECEPTOR REGULATION

in the number of p-opioid receptor patches in the dorsolateral and dorsomedial striatum. A patchy loss of DA uptake sites (labelled with [3Hlmazindol)was also observed. As seen with p-opioid receptor patches, the greatest loss of DA uptake sites (a measure of DA terminal density) was evident in the medial CPu. Although DA levels were not directly measured, it is likely that there was a similar decrease, as decreases in number of DA uptake sites and the concentration of DA are simlarly affected in the mature straitum following 6-OHDA lesions (Joyce, 1991a,b; Marshall et al., 1989). There was also a significant loss of TH-like immunoreactivity in the lesioned striata, which was heterogeneous in nature (patchy). When receptor numbers were determined with QAR, a small but significant loss of D1 receptors (10-15%) was found within the striatum. The greatest loss was observed in the medial CPu, particularly in the dorsomedial quadrant. Neither the number nor the topography of D2 binding sites were affected by DA lesions at this very early postnatal time point. Although the mechanisms underlying the differential behavioral effects of neonatal as compared to adult 6-OHDA lesions remain unclear, differential regulation of DA receptors appears to provide a reasonable explanation for the differences. Following adult 6-OHDA lesions, there is an increase in D2 receptor binding in the lateral striatum (Marshall et al., 1989; Savasta et al., 1986),which is thought to underlie the supersensitivity to D2 agonists and the eventual behavioral recovery observed in these animals (Creese et al., 1977; Marshall, 1984). Increases in D1 receptor density following adult 6-OHDA lesions have not been observed (Ariano, 1987; Marshall et al., 1989; Savasta et al., 1987); instead, there have been reports of small decreases, which may reflect other events besides or instead of the loss of DA (Joyce, 1991a,b; Marshall et al., 1989). On the other hand, there are no consistent reports of an increase in D1 or D2 receptors following neonatal 6-OHDA lesions to explain the increased sensitivity to D1 agonists observed in these animals. In fact, with our protocol there is a small, but significant loss of D1 receptors that appears patchy in nature. We believe that this alteration in receptor density reflects the developmental organization of the rat striatum. Gerfen and associates (1987b) have reported that intrastriatal 6-OHDA lesions t o the rat on PO or P1 should selectively affect the patch-directed DA system, while leaving the matrix-directed system relatively intact. Our data suggest that the development of the presynaptic DA system is important for the expression of DA receptors in the different compartments of the striatum. The striatum of the rat, once thought to be homogeneous in nature, is clearly heterogeneous and highly compartmentalized. The neurochemical organization of the striatum appears to obey principles based on an intrinsic organization of neurons, known as the patch (striosome)/matrix organization (Graybiel and Rags-

43

dale, 1978; Herkenham and Pert, 1981; Pert et al., 1976). The distribution of many neurotransmitters, afferent projections, and cell bodies of striatal efferents obey this organization. DA afferents t o the patch compartment develop prior to those innervating the matrix (Gerfen et al., 1987b; Graybiel, 1984; Olson et al., 1972) and arise from separate cell populations (Gerfen et al., 1987a). The neurons of the patch comparment migrate, mature, and give rise to their efferents prior to the neurons in the matrix compartment (Fishell and van der Kooy, 1987; Van der Kooy and Fishell, 1987). The neurons in the patch compartment also express DA receptors, preferentially of the D1 subtype, prior to the expression of DA receptors in the matrix (Murrin and Zeng, 1989; Rao et al., 1991). This early-maturing patch-directed DA system is more sensitive to neonatal 6-OHDA administration than to the DA system innervating the matrix (Gerfen et al., 1987b); thus, damage to the patch system may subserve the behavioral effects of the neonatal lesions. There are temporal and spatial differences in the postnatal development of the DA receptor subtypes, which are related to the development of the patch- and matrix-directed DA innervation (Murrin et al., 1985; Murrin and Zeng, 1989; Rao et al., 1991). D1 receptors are visible as distinct patches a t P1, with the highest density in the ventrolateral CPu. The expression of receptors in these patches is coincident with the maturation of the DA input to that compartment, as measured by the concentration of DA and the binding of [3H]mazindol to DA uptake sites. Beginning at P3, there is a subsequent development of D2 receptors and [3H]hemicholinium-3binding to high-affinity transport sites for choline, with a topography that is largely distinct from that of the D1 receptor. Between P7 and P10, the topography of D1 receptors and [3H]mazindolbinding become distinctly less patchy due to a marked increase in the density of both systems in the matrix. By P15, D1 receptors and [3H]mazindolbinding show the fairly homogenous pattern of distribution throughout the striatum, as is seen in the adult. Therefore, the results of the ontogeny studies suggest that DA systems develop within the striatum in three distinct stages: 1) an increase in DA input to the patch compartment with a concurrent increase in the number of D1 receptors within this compartment; 2) starting at P5, an increase in D2 receptors that show a different developmental pattern compared to D1 receptors; and 3) a maturation of D1 receptors within the matrix compartment (tripling between P10 and P16), which is correlated with an increase in the DA input to the matrix. It is evident that separate patch- and matrix-related D1 receptor systems develop at disparate times (associated with DA input), and their developmental organization is separable from that of the D2 receptor system (associated with cholinergic input); thus, DA may not be acting in the same manner to affect the development of the 2 receptor subtypes.

44

B.S. NEAL AND J.N. JOYCE

The developmental study provides a new framework for understanding why there is a selective loss of D1 receptors within the patch compartment following very early 6-OHDA administration, but why later lesions may affect both D1 and D2 receptor systems. When the 6-OHDA is given prior to P3, the D1 receptors within the patch compartment undergo a period of rapid development, whereas D2 receptors and D1 receptors within the matrix are not yet expressed. Thus, the coincidental maturation in the early postnatal period of the early “islandic”DA terminals and D1 receptors on the patchlocated cells may make that system particularly vulnerable to the effects of early 6-OHDA administration. The degree of DA depletion, per se, does not play a major role in either the neurochemical or functional outcomes of these lesions, as there are differences even when both groups have almost complete DA depletion. Extensive lesions (>95% DA depletion) in adults cause severe behavioral deficits, along with increases in D2 receptor sensitivity and D2 receptor number. Similar DA depletions in neonates do not cause these severe behavioral deficits. This may be because the neuronal system mediating these behaviors are not functional early in the first postnatal week. Rogers and Dunnett (198913) have shown that 80% of neonatally lesioned rats will show the full Parkinsonian-like syndrome if they are given a second 6-OHDA injection in adulthood. This suggests that the DA system underlying the behavioral deficits following adult lesions (e.g., the D2 system) is spared in the neonatally lesioned rat, even though DA depletion in the CPu is >95% following their neonatal lesioning technique. The pattern of D1 receptor loss following neonatal or adult 6-OHDA lesions is not overlapping; the neonatal lesions seemingly induce a loss of D1 receptors within the patch compartment, whereas adult lesions affect those receptors residing in the matrix compartment (Joyce, 1991a). These differential results on the location of the receptor loss could also be involved in the differential behavioral consequences of the lesions. As previously discussed, neonatal lesions on PO or P1 cause selective neurochemical and behavioral changes in the D1 receptor system. If the lesion is given a few days later in development (e.g., P3 or P5), a different pattern of alterations is observed. Behavioral supersensitivity t o both D1 and D2 agonists and antagonists is seen following these later lesions, although the D1 alterations are more prominent (Breese et al., 1984a,b, 1985a,b, 1987; Duncan et al., 1987). There is reportedly no long-term change in the density of D1 receptors measured by homogenate binding with [3H]SCH23390 (Breese et al., 1987; Duncan et al., 1987; Dewar et al., 1991). However, Dewar and associates (1991) did find an approximate 10% loss in D1 binding in the rostral striatum 90 days postlesioning, but it was not statistically significant. On the other hand, the authors did report a significant increase in D2 receptor density (32%)in the rostral striatum, contrary to other reports

(Breese et al., 1987; Duncan et al., 1987). Lesions given at this slightly later point in postnatal development may be able to affect D2 receptor expression, because the receptors are beginning to proliferate; however, lesions given on PO or P1 would not affect D2 receptor expression, as the receptors are not yet present in any appreciable number at that time (Rao et al., 1991). An alternative hypothesis is that the increase in D2 receptor density is somehow dependent upon the 5-HT hyperinnervation observed in the striatum following extensive depletions of DA during the neonatal period (Dewar et al., 1990; Stachowiak et al., 1984). Dewar and associates (1990) reported that the increase in both D2 receptors and 5-HT content occurred in the rostral striatum and that the increase in receptor number was closely correlated with 5-HT turnover. With the intrastriatal6-OHDA injections, the degree of DA depletion (40-50%) is not high enough to cause the 5-HT hyperinnervation; thus, one would not expect an increase in D2 receptors. However, we have found that a dose of 20 p,g of 6-OHDA per striatum on PO or P1 causes a complete loss of DA uptake sites within the striatum (suggesting a severe loss of DA), with no increase in the number of D2 receptors (Neal and Joyce, 1991b). These data support our hypothesis that it is the timing of the lesion that is important, not the extent of the lesion. The behavioral effects due to the loss of D 1 receptors can be understood if these receptors are part of a neuronal system functioning as a negative feedback system to the substantia nigra (SN). For example, there are DA neurons in the SN that receive synaptic input from terminals exhibiting substance P (SP)-like immunoreactivity (Kawai et al., 1987; Mahalik, 1988). These striatonigral SP neurons are located in the rostral striatum and project ipsalaterally to the SN (Krause et al., 1984). Dense patches of SP-like immunoreactivity are present in the developing striatum as early as PO and coincide with the early patches of TH-like immunoreactivity (“DA islands”) (Zahm et al., 1990). As mentioned earlier, it appears that these patches also express D1 receptors early in development (Murrin and Zeng, 1989; Rao et al., 1991). Sivam and associates (1987) found that SP in the striatum is significantly decreased following neonatal 6-OHDA lesions, an effect not seen with adult lesions. Furthermore, Sivam (1989) determined that there was a functional link between D1 receptors and the SP in the striatonigral neurons in the neonatally lesioned rats. Also, Gerfen and associates (1990) recently reported that the majority of striatonigral output neurons express D1 receptors, as well as SP and dynorphin. Thus, it is possible that neonatal 6-OHDA lesions are selectively affecting this Dl/SP system, and alterations in this system underlie the behavioral alterations observed with neonatal 6-OHDA lesions. In summary, intrastriatal 6-OHDA administration on PO or P1 caused a patchy loss of DA uptake sites and TH-like immunoreactivity within the striatum. There

EARLY DOPAMINE LOSS AND RECEPTOR REGULATION

was also a loss of y-opioid receptor patches, suggestive of a loss of DA input to the striatal patch compartment. This loss of DA input to the patches was associated with a loss of D1 receptors. Contrary to what is observed following adult 6-OHDA lesions, there were no changes in D2 binding. The neonatal lesions also resulted in an increased incidence of abnormal perioral movements and self-biting behavior following the administration of the D1 agonist SKF’38393 in adulthood. Thus, the lesioned rats exhibited an increased sensitivity to a D1 agonist concomitant with a loss of D1 receptors. These changes in D1 behavioral sensitivity may be due to the loss of an inhibitory patch-associated D1 receptor system. ACKNOWLEDGMENTS The authors wish to thank Ms. Anne Blood for her assistance in completing these studies and Dr. Jay Schneider (Hahnemann University) for his help with the TH immunocytochemistry. This work was supported in part by USPHS grants MH43852, F32 MH09888, and P30 DH26979, and a grant from the Tourette Syndrome Association. REFERENCES Ariano, M.A. (1987) Comparison of dopamine binding sites in the rat superior cervical ganglion and caudate nucleus. Brain Res., 421 ~245-254. Boyson, S.J., McGonigle, P., and Molinoff, P.B. (1986) Quantitative autoradiographic localization of the D1 and D2 subtypes of dopamine receptors in rat brain. J. Neurosci., 6:31177-3188. Breese, G.R., Baumeister, A., Napier, T.C., Frye, G.D., and Mueller, R.A. (1985a)Evidence that D-1 dopamine receptors contribute to the supersensitive behavioral responses induced by L-dihydroxyphenylalanine in rats treated neonatally with 6-hydroxydopamine. J. Pharmacol. Exp. Ther., 235:287-295. Breese, G.R., Baumeister, A.A., McCown, T.J., Emerick, S.G., Frye, G.D., Crotty, K., and Mueller, R.A. (1984a) Behavioral differences between neonatal and adult 6-hydroxy-dopamine-treatedrats to dopamine agonists: Relevance to neurological symptoms in clinical syndromes with reduced brain dopamine. J . Pharmacol. Expt. Ther., 231 :343-354. Breese, G.R., Baumeister, A.A., McCown, T.J., Emerick, S.G., Frye, G.R., and Mueller, R.A. (1984b)Neonatal-6-hydroxydopamine treatment: Model of susceptibility for self-mutilation in the Lesch-Nyhan syndrome. Pharmacol. Biochem. Behav., 21:459-461. Breese, G.R., Duncan, G.R., Napier, T.C., Bondy, S.C., Iorio, L.C., and Mueller, R.A. (1987) 6-Hydroxydopamine treatments enhance behavioral responses to intracerebral microinjection of D, and D,dopamine agonists into nucleus accumbens and striatum without changing dopamine antagonist binding. J . Pharmacol. Exp. Ther., 240:167-176. Breese, G.R., Napier, T.C., and Mueller, R.A. (198b) Dopamine agonist-induced locomotor activity in rats treated with 6-hydroxydopamine at differing ages: Functional supersensitivity of D-1 dopamine receptors in neonatally-lesioned rats. J. Pharmacol. Exp. Ther., 234:447455. Broaddus, W.C. and Bennett, J.P., Jr. (1990)Postnatal development of striatal dopamine function. 11. Effects of neonatal 6-hydroxydopamine treatments on D1 and D2 receptors, adenylate cyclase activity and presynaptic dopamine function. Dev. Brain Res., 52:273-277. Bruno, J.P., Jackson, D., Zigmond, M.J., and Stricker, E.M. (1987) Effect of dopamine-depleting lesions in rat pups: Role of striatal serotonergic neurons in behavior. Behav. Neurosci., 101:80EL811. Bruno, J.P., Snyder, A.M., and Stricker, E.M. (1984) Effect of dopamine-depleting brain lesions on suckling and weaning in rats. Behav. Neurosci., 98:156-161. Bruno, J.P., Stricker, E.M., and Zigmond, M.J. (1985) Rats given dopamine-depleting brain lesions as neonates are subsensitive to dopaminergic antagonists as adults. Behav. Neurosci., 99:771-775.

45

Bruno, J.P., Zigmond, M.J., and Stricker, E.M. (1986) Rats given dopamine-depleting brain lesions as neonates do not respond to acute homeostatic imbalances as adults. Behav. Neurosci., 100:125128. Caboche, J., Rogard, M., and Besson, M.-J. (1991) Comparative development of D,-dopamine and p opiate receptors in normal and in 6-hydroxydopamine-lesionedneonatal rat striatum: Dopaminergic fibers regulated p but not D, receptor distribution. Dev. Brain Res., 58:111-122. Creese, I., Schneider, R., and Snyder, S.H. (1977) 3H-spiroperidol labels dopamine receptors in pituitary and brain. Eur. J. Pharmacol., 46~377-381. Dewar, K.M., Soghomonian, J . J . , Bruno, J.P., Descarries, L., and Reader, T.A. (1990) Elevation of dopamine D, but not D, receptors in adult rat striatum after neonatal 6-hydroxydopamine denervation. Brain Res., 536:287-296. Duncan, G.E., Criswell, H.E., McCown, T.J., Paul, LA., Mueller, R.A., and Breese, G.R. (1987)Behavioral and neurochemical responses to haloperidol and SCH-23390 in rats treated neonatally or as adults with 6-hydroxydopamine.J. Pharmacol. Exp. Ther., 243:1027-1033. Fallon, J.H. and Moore, R.Y. (1978) Catecholamine innervation of the basal forebrain. IV. Topography of the dopamine projection to the basal forebrain and neostriatum. J . Comp. Neurol., 180545-580. Fishell, G. and van der Kooy, D. (1987) Pattern formation in the striatum: Developmental changes in the distribution of striatonigral neurons. J. Neurosci., 7:1969-1978. Gelbard, H.A., Teicher, M.H., Baldessarini, R.J., Gallitano, A.L., Marsh, E.R., Zorc, J., and Faedda, G. (1989) Dopamine D, receptor development depends on endogenous dopamine. Dev. Brain Res., 56:137-140. Gerfen, C.R. (1984) The neostriatal mosiac: Compartmentalization of corticostriatal input and striatonigral output systems. Nature, 311:461-464. Gerfen, C.R., Herkenham, M., and Thibault, J . (1987a)The neostriatal mosiac: 11. Patch- and matrix-directed mesostriatal dopaminergic and non-dopaminergic systems. J . Neurosci., 7:3915-3934. Gerfen, C.R., Baimbridge, K.G., and Thibault, J . (1987b) The neostriatal mosiac: 111. Biochemical and developmental dissociation of patch-matrix mesostriatal systems. J. Neurosci., 7:3935-3944. Gerfen, C.R., Engber, T.M., Mahan, L.C., Susel, Z., Chase, T.N., Monsma, F.J., Jr., and Sibley, D.R. (19901D, and D, dopamine receptorregulated gene expression of striatonigral and striatopallidal neurons. Science, 250:1429-1432. Graybiel, A.M. (1984) Correspondence between the dopamine islands and striosomes of the mammalian striatum. Neuroscience, 13:11571187. Graybiel, A.M. and Ragsdale, C.W. (1978) Histochemically distinct compartments in the striatum of human, monkey and cat demonstrated by acetylcholinesterase staining. Proc. Natl. Acad. Sci. USA, 7557236726, Herkenham, M. and Pert, C.B. (1981) Mosiac distribution of opiate receptors, parafascicular projections and acetylcholinesterase in rat striatum. Nature (London), 291:415-417. Johnson, B.J. and Bruno, J.P. (1990)D, and D, receptor contributions to ingestive and locomotor behavior are altered after dopamine depletions in neonatal rats. Neurosci. Lett., 118:2@123. Joyce, J.N. and Marshall, J.F. (1987) Quantitative autoradiography of dopamine D-2 sites in rat caudate-putamen: Localization to intrinsic neurons and not to neocortical afferents. Neuroscience, 20:773795. Joyce, J.N. and Marshall, J.F. (1985) Striatal topography of D-2 receptors correlates with indexes of cholinergic neuron localization. Neurosci. Lett., 53:127-131. Joyce, J.N. (1991a)Differential response of striatal dopamine receptor subtypes to the loss of dopamine following 6-hydroxydopamine or reserpine treatment. Exp. Neurol., 113261-276, Joyce, J.N. (1991b) Relationship of striatal dopamine receptor subtypes and cholinergic muscarinic receptors to their uptake sites and regulation following 6-hydroxydopamine lesions of the mesostriatal dopamine system. Exp. Neurol., 113:277-290. Kawai, Y., Takagi, H., Kumoi, Y., Shiosaka, S., and Tohyama, M. (1987) Nigrostriatal dopamine neurons receive substance P-ergic inputs in the substantia nigra: Application of the immunoelectron microscopic technique to fluorescent double-staining for transmitter-specific projections. Brain Res., 401:371-376. Krause, J.E., Reiner, A.J., Advis, J.P., and McKelvy,J.F. (1984)Invivo biosynthesis of L3%1-and l3H1substanceP in the striatum of the rat and their axonal transport to the substantia nigra. J . Neurosci., 4:775-785. Lytle, L.D., Shoemaker, W.J., Cottman, K., and Wurtman, R.J. (1972) Long-term effects of postnatal 6-hydroxydopaminetreatment on tissue catecholamine levels. J . Pharmacol. Exp. Ther., 18856-64. Mahalik, T.J. (1988) Direct demonstration of interactions between

46

B.S. NEAL AND J.N. JOYCE

substance P immunoreactive terminals and tyrosine hydroxylase immunoreactive neurons in the substantia nigra of the rat: An ultrastructural study. Synapse, 2:508-515. Marshall, J.F. (1984) Behavioral consequences of neuronal plasticity following injury to nigrostriatal dopaminergic neurons. In: Aging and Recovery of Function in the Central Nervous System. S.W. Scheff, ed. Plenum Press, New York, pp. 101-127. Marshall, J.F., Navarrete, R., and Joyce, J.N. (1989) Decreased striatal D1 binding density following mesotelencephalic 6-hydroxydopamine injections: An autoradiographic analysis. Brain Res., 493~247-257. Marshall, J.F., Richardson, J.S., and Teitelbaum, P. (1974) Nigrostriatal bundle damage and the lateral hypothalamic syndrome. J . Comp. Physiol. Psych., 87:80%330. Murrin, L.C. (1982) In vivo studies of dopamine receptor ontogeny. Life Sci., 31:971-980. Murrin, L.C. and Zeng, W. (1989) Dopamine D, receptor development in the rat striatum: Early localization in striosomes. Brain Res., 480:170-177. Neal, B.S. and Joyce, J.N. (1991a) Dopamine D1 receptor behavioral responsitivity following selective lesions of the striatal patch compartment during development. Dev. Brain Res., 60:105-113. Neal, B.S. and Joyce, J.N. (1991b) D1 and D2 receptors do not upregulate in response to neonatal intrastriatal 6-hydroxydopamine (6-OHDA)lesions. SOC.Neurosci. Abst., 17:610. Olson, L., Seiger, A,, and Fuxe, K. (1972)Heterogeneity of striatal and limbic dopamine innervation: Highly fluorescent islands in developing and adult rats. Brain Res., 44:283-288. Paxinos, G. and Watson, C. (1986)The Rat Brain in Stereotaxic Coordinates. Academic Press, Sydney. Rao, P.A., Molinoff, P.B., and Joyce, J.N. (1991)Ontogeny of dopamine D1 and D2 receptor subtypes in rat basal ganglia: A quantitative autoradiographic study. Dev. Brain Res., 60:167-177. Rogers, D.C. and Dunnett, S.B. (1989a) Hypersensitivity to alphamethyl-p-tyrosine suggests that behavioural recovery of rats receiving neonatal 6-OHDA lesions is mediated by residual catecholamine neurones. Neurosci. Lett., 102:108-113. Rogers, D.C. and Dunnett, S.B. (1989b)Neonatal dopamine-rich grafts and 6-OHDA lesions independently provide partial protection from the adult nigrostriatal lesion syndrome. Behav. Brain Res., 34:131146. Savasta, M., Dubois, A,, and Scatton, B. (1986)Autoradiographic local-

ization of D1 receptors in the rat brain with L’HISCH23390. Brain Res., 375:291-301. Savasta, M., Dubois, A,, Benavides, J., and Scatton, B. (1988) Different plasticity changes in D1 and D2 receptors in rat striatal subregions following impairment of dopaminergic transmission. Neurosci Lett., 85119-124. Schneider, J.S. and Markham, C.H. (1986)Neurotoxic effects of N-methyl-4-phenyl-l,2,3,6-tetrahydropyridine (MPTP) in the cat. Tyrosine hydroxylase immunohistochemistry. Brain Res., 373:258267. Sivam, S.P. (1989)D1 dopamine receptor-mediated substance P depletion in the striatonigral neurons of rats subjected to neonatal dopaminergic denervation: Implications for self-injurious behavior. Brain Res., 500:119-130. Sivam, S.P., Breese, G.R., Krause, J.E., Napier, T.C., Mueller, R.A., and Hong, J.3. (1987) Neonatal and adult 6-hydroxydopamine-induced lesions differentially alter tachykinin and enkephaline gene expression. J. Neurochem., 49:1623-1633. Smith, R.D., Cooper, B.R., and Breese, G.R. (1973)Growth and behavioral changes in developing rats treated intracisternally with 6-hydroxydopamine: Evidence for involvement of brain dopamine. J. Pharmacol. Exp. Ther., 185:609-619. Stachowiak, M.K., Bruno, J.P., Snyder, A.M., Stricker, E.M., and Zigmond, M.J. (1984)Apparent sprouting of striatal serotonergic terminals after dopamine-depleting brain lesions in neonatal rats. Brain Res., 291:164-167. Ungerstedt, U. (1971) Adipsia and aphagia after 6-hydroxydopamineinduced degeneration of the nigro-striatal dopamine system. Acta Physiol. Scand. Suppl., 367:95-122. Whishaw, I.Q.,Funk, D.R., Haeryluk, S.J., and Karbashewski, E.D. (1987)Absence of sparing of spatial navigation, skilled forelimb and tongue use and limb posture in the rat after neonatal dopamine depletion. Physiol. Behav., 40:247-253. Zahm, D.S., Eggerman, K.W., Sprung, R.F., Wesche, D.E., and Payne, E. (1990) Postnatal development of striatal neurotensin immunoreactivity in relation to clusters of substance P immunoreactive neurons and the “dopamine islands” in the rat. J . Comp. Neurol., 296:403-414. Zigmond, M.J. and Stricker, E.M. (1973) Recovery of feeding and drinking by rats after intraventricular 6-hydroxydopamine or lateral hypothalamic lesions. Science, 182:717-720.