Distribution of Serotonin 5-HT2A Receptorsin Afferents of the Rat Striatum

MICHAEL BUBSER,1 JON R. BACKSTROM,1 ELAINE SANDERS-BUSH,1 BRYAN L. ROTH,2AND ARIEL Y. DEUTCH1*

1Departments of Psychiatry and Pharmacology and Center for Molecular Neuroscience, Vanderbilt UniversitySchool of Medicine, Nashville, Tennessee

2Departments of Psychiatry, Biochemistry, and Neuroscience, Case Western Reserve University Schoolof Medicine, Cleveland Ohio

KEY WORDS antipsychotic drug; cortex; dopamine; extrapyramidal side effects; glo-bus pallidus; Parkinson’s disease

ABSTRACT Treatment with conventional antipsychotic drugs (APDs) is accompa-nied by extrapyramidal side effects (EPS), which are thought to be due to striataldopamine D2 receptor blockade. In contrast, treatment with atypical APDs is marked bya low incidence or absence of EPS. The reduced motor side effect liability of atypicalAPDs has been attributed to a high serotonin 5-HT2A receptor affinity coupled with arelatively low D2 affinity. Despite the high density of 5-HT2A binding sites in thestriatum, there are few detectable 5-HT2A mRNA-expressing neurons in the striatum.This suggests that most striatal 5-HT2A receptors are heteroceptors located on afferentaxons. A combined retrograde tracer-immunohistochemistry method was used to deter-mine the sites of origin of striatal 5-HT2A-like immunoreactive axons. 5-HT2A-likeimmunoreactive neurons in both the cortex and globus pallidus were retrogradelylabeled from the striatum; very few nigrostriatal or thalamostriatal neurons expressed5-HT2A-like immunoreactivity. Within the striatum, parvalbumin-containing interneu-rons displayed 5-HT2A immunolabeling; these neurons are the targets of cortical andpallidal projections. Our data indicate that cortico- and pallido-striatal neurons are themajor source of 5-HT2A receptor binding in the striatum, and suggest that cortico- andpallido-striatal neurons are strategically positioned to reduce the motor side effects thataccompany striatal D2 receptor blockade or are seen in parkinsonism. Synapse 39:297–304, 2001. © 2001 Wiley-Liss, Inc.

INTRODUCTION

Antipsychotic drugs (APDs) are the mainstay in thetreatment of schizophrenia. Treatment with typicalAPDs is accompanied by a high incidence of extrapy-ramidal side effects (EPS). These parkinsonian-likeside effects are correlated with high in vivo occupancyof D2 receptors in the striatum (Farde et al., 1986,1988; Kapur et al., 1996; Heintz et al., 1996). So-calledatypical APDs either do not induce EPS or have asubstantially lower EPS liability than typical APDs,and have therefore become the first line treatment forschizophrenia. The lower EPS liability of atypicalAPDs has been suggested to be due to their high affin-ity for serotonin 5-HT2A receptors coupled with low-to-moderate D2 receptor affinity (Altar et al., 1986; Melt-zer et al., 1989; Matsubara et al, 1993; Nyberg et al.,1993; Schotte et al., 1993; Kapur et al., 1996; Roth etal., 1998).

Receptor binding and in situ hybridization histo-chemistry studies have revealed a widespread distri-

bution of 5-HT2A receptors in the brain (Pompeiano etal., 1994; Wright et al., 1995; Mengod et al., 1996; Wardand Dorsa, 1996; Mijnster et al., 1997; Roth et al.,1998). 5-HT2A receptor binding is moderately high but5-HT2A mRNA levels are very low in the striatum,suggesting that 5-HT2A receptors are localized primar-ily to striatal afferents.

Since 5-HT2A receptors appear to play a crucial rolein the actions of atypical APDs, we examined the dis-tribution of 5-HT2A receptors in striatal afferents byimmunohistochemical methods. We focused on striatalafferents in view of a large body of literature that hasimplicated the striatum in EPS (Costall et al., 1972;

Contract grant sponsors: the National Parkinson Foundation Center of Excel-lence at Vanderbilt University, NARSAD Young Investigator awards, and NIH;Contract grant numbers: MH45124, MH57995, MH 34001, MH57635, KO2MH001366.

*Correspondence to: Ariel Y. Deutch, Psychiatric Hospital at Vanderbilt, Suite313, 1601 23rd Avenue South, Nashville, TN 37212.E-mail: ariel.deutch@mcmail.vanderbilt.edu

Received 19 May 2000; Accepted 28 July 2000

SYNAPSE 39:297–304 (2001)

© 2001 WILEY-LISS, INC.

Ossowka et al., 1990; Deutch et al., 1992; Kapur et al.,1996; Kaur et al., 1997) and previous findings indicat-ing that lesions of striatal afferents reduce the catalep-togenic actions of D2 antagonists (Sanberg, 1980; Scat-ton et al. 1982; Worms et al., 1985; Yoshida et al., 1991;Hauber et al., 1998).

MATERIALS AND METHODSSurgery

Adult male Sprague-Dawley rats (Harlan, Birming-ham, AL) weighing 275–300 g at the start of the studywere housed in groups of four with food and wateravailable ad libitum. All experimental procedures werecarried out in accord with the Guide for the Care andUse of Laboratory Animals of the U.S. National Insti-tutes of Health. Under pentobarbital anesthesia theretrograde tracer Fluoro-gold (FG; Fluorochrome, Den-ver, CO) was deposited into the dorsal striatum byiontophoresis. Glass micropipettes with a diameter of20–30 mm were used to eject FG (3% in 0.1 M sodiumcacodylate) by means of a pulsed positive current (2.5mA, 7 sec on/off) for 7–10 min.

Immunohistochemistry

One week after surgery, rats were deeply anesthe-tized and transcardially perfused with PBS followed by250 ml of 4% paraformaldehyde in 0.1 M sodium phos-phate (pH 7.4). The brains were removed from thecranium, postfixed, cryoprotected, and cut on a freezingmicrotome.

Three different 5-HT2A antibodies were used: amouse monoclonal antibody directed against amino ac-ids 1–71 of the amino terminus (Pharmingen, San Di-ego, CA; 1:1000), a rabbit polyclonal antiserum gener-ated against amino acids 22–41 in the N-terminus(Berry et al., 1996; 1:2,000), and an affinity-purifiedrabbit antiserum generated against the carboxy termi-nus (residues 330–345) of the receptor (Backstrom andSanders-Bush, 1997; 1:1,500). In most double-labelingimmunohistochemical studies the mouse monoclonalanti-5-HT2A antibody was used together with rabbitanti-FG (Chemicon, Temecula, CA; 1:3,000) or anti-parv-albumin (SWANT, Bellinzona, Switzerland; 1:2,000)antibodies.

Sections were washed in 50 mM Tris-buffered saline(TBS, pH 7.4) and incubated in methanolic peroxide toinhibit endogenous peroxidase activity. Sections werethen incubated for '36 h at 4°C in the primary anti-body diluted in TBS containing 4% normal horse serumand 0.2% Triton X-100 (TBS1). Subsequently, sectionswere washed in TBS1 and incubated for 90 min at roomtemperature in biotinylated antimouse IgG (JacksonImmunochemicals, West Grove, PA; 1:1,500), washed,and then incubated for 1 h in horseradish peroxidase-conjugated streptavidin (Vector, Burlingame, CA;1:200). 5-HT2A-like immunoreactivity (-li) was visual-

ized by incubating sections in TBS containing 0.025%3,39-diaminobenzidine, 0.001% hydrogen peroxide,1.5% nickel ammonium sulfate, and 0.15% cobalt chlo-ride to yield a blue/black precipitate.

In double-labeling experiments, sections were pro-cessed for 5-HT2A-li using the mouse monoclonal anti-body and then taken through an ascending–descend-ing series of methanol in TBS followed by methanolicperoxide to eliminate residual peroxidase activity (Ax-elson et al., 1992; Bubser and Deutch, 1998). Sectionswere then processed to reveal FG- or parvalbumin-li.Briefly, sections were incubated in primary antibodyfor '36 h and subsequently incubated in affinity-puri-fied donkey-antirabbit IgG (Jackson; 1:60 in TBS1)followed by rabbit PAP (Sternberger Immunochemi-cals, Lutherville, MD; 1:200). Sections were then re-acted in 0.25 M acetate buffer (pH 5.6) containing0.033% 3-amino-9-ethyl-carbazole, 6.7% dimethylform-amide, and 0.025% hydrogen peroxide to obtain a redprecipitate (see Bubser and Deutch, 1998).

RESULTS

The regional distribution of neurons as revealed bythe three different 5-HT2A antibodies was identical.However, some differences in the cellular distributionof 5-HT2a-li were seen: the mouse monoclonal antibodystrongly labeled neuronal processes and the perimeterof the soma, while the other two antibodies labeled thesoma strongly but did not stain processes as robustly(see Fig. 1). In addition, one of the antibodies (Back-strom and Sanders-Bush, 1997) revealed diffuse stria-tal staining suggestive of the presence of 5-HT2A-im-munolabeling of afferents to the striatum.

Striatum

Only a few intensely labeled 5-HT2A-li cells wereseen in the striatum (Fig. 1). The great majority($90%) of these 5-HT2A-li cells were also parvalbumin(PV)-li (Fig. 2C). Moderately strong staining of thestriatal neuropil was seen with the antibody of Back-strom and Sanders-Bush (1997) (see Fig. 1B); less neu-ropil staining was seen with the other two 5-HT2A

antibodies. We occasionally saw medium spiny neuronsthat appeared to be 5-HT2A immunoreactive; however,this labeling was inconsistent across animals and didnot appear to be related to perfusion characteristics,postperfusion handling, or duration of immuoperoxi-dase development.

Cerebral cortex

All three 5-HT2A receptor antibodies labeled pyrami-dal cells in the cerebral cortex. Pyramidal neuronsexpressing 5-HT2A-li were concentrated in superficiallayer V, with intense immunoreactivity of the apicaldendrites and light to moderate staining of the perim-eter of the soma (see Fig. 2B). 5-HT2A-li was also seenin some nonpyramidal cells.

298 M. BUBSER ET AL.

FG deposits into the dorsal striatum (Fig. 3) retro-gradely labeled cortical neurons in accord with theexpected topography of corticostriatal neurons; greaterthan 90% of cortical neurons retrogradely labeled fromthe striatum were 5-HT2A-li (Fig. 2B). Most of theseneurons were in superficial layer V (Fig 2B). FG wasalso seen in a much smaller number of layer Vb cells;these cells often did not express 5-HT2A-li (Fig. 2B).

Globus pallidus

Each of the three 5-HT2A receptor antibodies labeledpallidal neurons (Fig. 1). Injections of FG into the stri-atum labeled cells in the GP in a topographically de-fined manner. Not all pallidal neurons were 5-HT2A

immunoreactive; about half of the retrogradely labeledpallidal cells were 5-HT2A-li (Fig. 2A), and of the palli-dal neurons that expressed the 5-HT2A receptor, ap-proximately 70% were retrogradely labeled from thestriatum.

Thalamus and substantia nigra

Few 5-HT2A-li neurons were seen in the thalamicintralaminar nuclei, including the centrolateral andparacentral nuclei, and a moderate number of5-HT2A-li cells were present in the parafascicular nu-cleus. However, thalamic 5-HT2A-li neurons wererarely retrogradely labeled from the striatum (data notshown).

A moderate number of 5-HT2A-li neurons were seenin the pars reticulata of the substantia nigra (SN), withfew 5-HT2A-li cells seen in the pars compacta. Retro-gradely labeled nigral cells rarely displayed 5-HT2A

immunoreactivity.

DISCUSSION

We found that cortical and pallidal 5-HT2A-li neu-rons were retrogradely labeled from the striatum.These data suggest that most of the striatal 5-HT2A

binding sites reflects the presence of the receptor onterminals of corticostriatal and pallidostriatal neurons.

5-HT2A-li cells in the striatum

Striatal neurons expressing 5-HT2A-li were scatteredand few in number. The myelinated fibers of corticos-triatal axons were densely stained. Our observationthat immunolabeling was restricted to relatively fewscattered large 5-HT2A-li neurons agrees with most insitu hybridization studies (Pompeiano et al., 1994;Wright et al., 1995; Laprade et al., 1996; Mijnster et al.,1997), which have found few cells expressing 5-HT2A

mRNA in the dorsal striatum. In the ventrolateralstriatum and fundus striatum, however, 5-HT2A

mRNA-expressing medium spiny neurons are more fre-quently encountered (Mijnster et al., 1997).

We found that striatal 5-HT2A-li cells also containedthe calcium-binding protein parvalbumin, indicatingthat the 5-HT2A receptor is expressed in a subset ofstriatal interneurons (Cowan et al., 1990; Kita et al.,1990; Kubota et al., 1993; Augood et al., 1995)

Ward and Dorsa (1996) reported that 5-HT2A mRNAis relatively abundant in the striatum and that thetranscript is present in striatofugal neurons. In addi-tion, Cornea-Hebert et al. (1999) reported that mediumspiny neurons express 5-HT2A immunoreactivity, al-though the labeling intensity of these cells was much

Fig. 1. The distribution of 5-HT2A-li cells in the basal ganglia asrevealed by three different antibodies (A: Pharmingen; B: Backstromand Sanders-Bush, 1997; C: Berry et al., 1996). All three antiserarevealed numerous 5-HT2A-li perikarya in the globus pallidus (GP),but only scattered 5-HT2A-li cells (arrow) in the striatum (CP).

5-HT2A RECEPTORS IN STRIATAL AFFERENTS 299

less than seen in large interneurons. On some occa-sions we also observed lightly labeled striatal cells thatcorrespond in size to medium spiny neurons, but in ourhands such labeling was inconsistent. It is possiblethat 5-HT2A transcript and protein is present in verylow abundance in striatofugal (medium spiny) neurons,usually below the threshold for detection. This sugges-tion is consistent with the finding that lesions disrupt-ing the striatal dopamine (DA) innervation increase5-HT2A mRNA in striatal neurons (Numan et al.,1995).

Cortical 5-HT2A-li neurons

Layer V pyramidal cells exhibited dense 5-HT2A im-munolabeling. Previous studies have reported thatlayer V pyramidal cells exhibit strong 5-HT2A-li as wellas express 5-HT2A mRNA (Wright et al., 1995; Burnetet al., 1996; Mijnster et al., 1997; Willins et al., 1997;Hamada et al., 1998; Wu et al., 1998). Some PV- andcalbindin-containing cortical interneurons also express5-HT2A-li (Willins et al., 1997; Jakab and Goldman-Rakic, 1998). It was our impression that 5-HT2A-li in-terneurons were more frequent in deep layers of thecortex, consistent with the suggestion of Jakab andGoldman-Rakic (1998), and more often encountered inlayer Vb than Va.

The distribution of cortical neurons that were retro-gradely labeled from the striatum followed the topo-graphical organization of corticostriatal projections(McGeorge and Faull, 1989; Kincaid and Wilson, 1996).In the cortex, double-labeled (FG- and 5-HT2A-li posi-tive) cells were frequently encountered in superficiallayer V. In contrast, cells deeper in the cortex that wereretrogradely labeled typically did not express 5-HT2A-li. This is consistent with the observation that5-HT2A-li cells in layer Vb are more often interneuronsthan projection (pyramidal) neurons.

Since striatal projections arising from superficial layerV neurons have been reported to preferentially target thestriatal matrix compartment (Gerfen, 1989; Kincaid andWilson, 1996), it is likely that cortical 5-HT2A-containingafferents mainly innervate the striatal matrix. The DAinnervation of the matrix preferentially degenerates inanimals treated with the parkinsonian neurotoxin1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (Moratallaet al., 1992), suggesting that the matrix compartment isinvolved in extrapyramidal motor function. We also foundthat PV striatal interneurons, which have been suggestedto interface between patch and matrix compartments,express 5-HT2A-li. Since cortical projection neurons syn-apse with both medium-spiny neurons and PV-li inter-neurons (Smith and Bolam, 1990; Bennett and Bolam,

Fig. 2. Localization of 5-HT2A-li neurons in cortico-striato-pallidalsites. A: 5-HT2A-li neurons (black reaction product) in the globuspallidus are retrogradely labeled (red) from the striatum (A). B: In thecortex, most 5-HT2A-li neurons (black reaction product) in superficiallayer V are retrogradely labeled from the striatum (red product); an

example is marked by an arrow. One of the few retrogradely labeledcells that is not 5-HT2A-li (arrowhead) can be seen in deep layer V. C:Shows a double-labeled 5-HT2A-li striatal interneuron (black) thatalso expresses parvalbumin (red). The scale bar seen in A is the samefor all panels.

300 M. BUBSER ET AL.

1994), 5-HT2A antagonists may coordinately influence ac-tivity in both patch and matrix striatal compartments.

5-HT2A-li cells in the globus pallidus

We observed 5-HT2A-li cells in the globus pallidus(GP). These 5-HT2A-li neurons were somewhat morefrequently seen in the caudal GP. Pallidal 5-HT2A neu-rons were typically bipolar and found interspersedamong unlabeled cells.

Two previous studies have reported the presence ofpallidal 5-HT2A-li cells (Hamada et al., 1998; Cornea-Hebert et al., 1999), but in situ hybridization studieshave failed to detect or found very low levels of 5-HT2A

mRNA. It is unlikely that this mismatch between thereceptor transcript and the protein is due to artifactualimmunolabeling. We used three different 5-HT2A re-ceptor antibodies, all of which revealed the same dis-tribution of 5-HT2A-li neurons. In a previous study(Willins et al., 1997) we showed that two of the anti-bodies labeled the same pyramidal neuron. Moreover,Cornea-Hebert et al. (1999) used yet another antibodyto reveal pallidal 5-HT2A-li neurons with a morphologythat appears identical to the 5-HT2A-li pallidal cellsthat we observed. These observations strongly suggestthat the staining we observed was specific labeling of5-HT2A receptors. Discrepancies between mRNAs andtheir proteins as revealed by immunohistochemistryare not uncommon, and mismatches between 5-HT2A

mRNA and protein have been previously reported (Wuet al., 1998).

Consistent with earlier studies, injections of FG intothe striatum retrogradely labeled pallidal neurons in atopographically organized manner (Staines et al., 1981;Staines and Fibiger, 1984; Shu and Peterson, 1988;

Rajakumar et al., 1994). Some pallidal neurons weredouble-labeled for FG and 5-HT2A-li. Pallidostriatalcells, which comprise about 25% of GP neurons, termi-nate selectively on striatal PV interneurons (Ruskinand Marshall, 1997; Bevan et al., 1998). Pallidostriatal5-HT2A-li neurons may also have additional targets,since some of these cells send axon collaterals to otherbasal ganglia sites, including the subthalamic andento-peduncular nuclei and the SN (Staines andFibiger, 1984; Kita and Kitai, 1994; Bevan et al., 1998).

5-HT2A in other striatal afferent neurons

The presence of 5-HT2A-li cells in the SN agrees withearlier reports indicating moderate numbers of5-HT2A-li and 5-HT2A mRNA-containing cells in the SNpars reticulata (Wright et al., 1995). Consistent withprevious studies, many nigral neurons were retro-gradely labeled from the striatum (Fallon et al., 78;Deutch et al., 1985), but these cells were rarely immu-noreactive for the 5-HT2A receptor. Similarly, althoughcertain thalamic nuclei have widespread striatal pro-jections (Beckstead, 1984; Berendse and Groenewegen,1990), we found thalamic 5-HT2A-li neurons rarelyproject to the striatum.

Functional implications

Treatment with conventional APDs is often accom-panied by EPS, while treatment with atypical APDshas a lower incidence of EPS. The concept that APD-elicited EPS can be reduced by coadministration ofserotonin antagonists was proposed a generation ago(Waldmeier and Delini-Stula, 1979). Subsequent stud-ies revealed that atypical APDs exhibit a high5-HT2A:D2 receptor affinities ratio (Altar et al., 1986;Meltzer et al., 1989; Matsubara et al., 1993; Nyberg etal., 1993). Most animal studies have found that halo-peridol-induced catalepsy is reduced by 5-HT2 antago-nists (including clozapine), but potentiated by 5-HT2

agonists or serotonin reuptake blockers (Balsara et al.,1979; Lucas et al., 1997; Bartoszyk et al., 1996; Neal-Beliveau et al., 1993; Young et al., 1999; see, however,Kalkman et al., 1998).

Clinical studies have similarly suggested that 5-HT2

antagonists may be useful in the treatment of Parkin-son disease (PD) (see Miyawaki et al., 1997). Thus,5-HT2 antagonists reduce symptoms in PD patients(Henderson et al., 1992) and improve neuroleptic-in-duced parkinsonism (Bersani et al., 1990; Ikeguchi andKuroda, 1995). Moreover, low-dose clozapine treatmentof PD patients with iatrogenic psychosis reduces motorsymptoms (Bonuccelli et al., 1997; Durif et al., 1997;Friedman et al., 1999; Pirelli et al., 1998). Since DAantagonists exacerbate parkinsonian symptoms, it islikely that the high 5-HT2A affinity of clozapine con-tributes to the motoric improvement.

The striatum is thought to be a major site where D2

receptor blockade induces EPS. Intrastriatal injections

Fig. 3. Photomicrograph of an iontophoretic Fluoro-gold depositinto the dorsolateral striatum.

5-HT2A RECEPTORS IN STRIATAL AFFERENTS 301

of typical APDs induce catalepsy (Costall et al., 1972;Dunstan et al., 1981; Kaur et al., 1997) and haloperi-dol-induced motor deficits are abolished by excitotoxiclesions of the striatum (Sanberg, 1980; Calderon et al.,1988; Hauber and Schmidt, 1993). Indeed, lesions atseveral striatopallido-thalamocortical sites, includingthe cortex and GP, reduce DA receptor antagonist-elicited catalepsy (Costall and Olney, 1971; Scatton etal., 1982; Worms et al., 1985; Yoshida et al., 1991;Hauber et al., 1998). In particular, several studies havesuggested that the GP is critical for the development ofAPD-elicited motor deficits and for parkinsonian signs(Delfs et al., 1995a,b; Chesselet and Delfs, 1996;Hauber, 1988; Hauber et al., 1998).

Organization of 5-HT2A immunoreactiveneurons in corticostriatopallidal circuitry

The current findings on the organization of 5-HT2A-containing afferents to the striatum offer an anatomi-cal substrate for the ability of 5-HT2A antagonists tomodify parkinsonian symptoms and EPS (see Fig. 4).Corticostriatal neurons terminate on both PV interneu-rons and medium spiny neurons (Bennett and Bolam,1994), while pallidostriatal neurons terminate on PVinterneurons (Bevan et al., 1998). Stimulation of corti-costriatal projections or disinhibition of corticostriatalneurons from tonic GABAergic regulation results inthe activation of enkephalin-containing medium spinyneurons that project to the pallidum and PV-containinginterneurons (Berretta et al., 1997; Parthasarathy andGraybiel, 1997). This anatomical organization offers ameans whereby feed-forward inhibition of select striat-

ofugal neurons may occur. Moreover, the pallidal5-HT2A-containing projection to striatal PV-containinginterneurons would similarly regulate striatal projec-tions. The 5-HT2A receptor, localized to three elementsof the striatum (corticostriatal, pallidostriatal, and PV-containing intrinsic striatal neurons), is therefore ide-ally situated to regulate cortiostriatopallidal activityand thereby modulate the proposed dysfunction inthese circuits that occurs in parkinsonism. In particu-lar, the PV interneuron appears likely to play a keyrole in striatal information processing.

CONCLUSIONS

Our data indicate that cortico-striatal and pallido-striatal neurons express the 5-HT2A receptor. This ob-servation is consistent with the suggestion that manystriatal 5-HT2A receptors are heteroceptors on striatalafferents rather than intrinsic striatal neurons. Futurestudies will be required to determine the relative con-tribution of 5-HT2A-expressing corticostriatal, pal-lidostriatal, and intrinsic striatal neurons to the totalpopulation of available 5-HT2A binding sites in thestriatum. Since we found that few striatal neuronsexpress detectable 5-HT2A receptor, it appears unlikelythe low incidence of EPS associated with atypical APDtreatment occurs solely by blocking postsynaptic5-HT2A receptors on intrinsic striatal neurons. Thepresence of 5-HT2A receptors in striatal afferents sug-gests that atypical APDs may exhibit less motor sideeffect liability because they modulate cortical and pal-lidal regulation of striatal function. Future studies ofthe actions of APDs and pathophysiology of movementdisorders, especially those using in vivo imaging ap-proaches, should focus attention on the cortex and glo-bus pallidus as well as the striatum.

ACKNOWLEDGMENT

We thank Sachin Patel for expert technical assis-tance.

REFERENCES

Altar CA, Waley AM, Neale RF, Stone GA. 1986. Typical and atypicalantipsychotic occupancy of D2 and S2 receptors: an autoradio-graphic analysis in rat brain. Brain Res Bull 16:517–525.

Augood SJ, Herbison AE, Emson PC. 1995. Localization of GAT-1GABA transporter mRNA in rat striatum: cellular coexpressionwith GAD67 mRNA, GAD67 immunoreactivity, and parvalbuminmRNA. J Neurosci 15:865–874.

Axelson JF, Shannon W, van Leeuwen FW. 1992. Immunocytochem-ical localization of estrogen receptors within neurotensin cells in therostral preoptic area of the rat hypothalamus. Neurosci Lett 136:5–9.

Backstrom JR, Sanders-Bush E. 1997. Generation of anti-peptideantibodies against serotonin 5-HT2a and 5-HT2c receptors. J Neu-rosci Methods 77:109–117.

Balsara JJ, Jadhav JH, Chandorkar AG. 1979. Effect of drugs influ-encing central serotonergic mechanisms on haloperidol-induced cat-alepsy. Psychopharmacology 62:67–69.

Bartoszyk GD, Roos C, Ziegler H. 1996. 5-HT1a receptors are notinvolved in clozapine’s lack of cataleptogenic potential. Neurophar-macology 35:1645–1646.

Beckstead RM. 1984. The thalamostriatal projection in the cat.J Comp Neurol 223:313–346.

Fig. 4. Schematic representation of the distribution of 5-HT2Areceptors in cortico-striato-pallidal circuitry. The 5-HT2A receptor islocalized to three key sites. Certain pyramidal cells of layer V give riseto projections to PV-containing interneurons, which also express the5-HT2A receptor. The pyramidal cells also innervate striatal medium-spiny neurons, which either do not express the serotonin receptor ordo so at very low levels. 5-HT2A-li pallidostriatal neurons selectivelyinnervate the PV-containing interneuron.

302 M. BUBSER ET AL.

Bennett BD, Bolam JP. 1994. Synaptic input and output of parvalbu-min-immunoreactive neurones in the neostriatum of the rat. Neu-roscience 62:707–719.

Berendse HW, Groenewegen HJ. 1990. Organization of the thalam-ostriatal projections in the rat, with special emphasis on the ventralstriatum. J Comp Neurol 299:187–228.

Berretta S, Parthasarathy HB, Graybiel AM. 1997. Local release ofGABAergic inhibition in the motor cortex induces immediate-earlygene expression in indirect pathway neurons of the striatum. J Neu-rosci 17:4752–4763.

Berry SA, Shah MC, Khan N, Roth BL. 1996. Rapid agonist-inducedinternalization of the 5-hydroxytryptamine2A receptors occurs viathe endosome pathway in vivo. Mol Pharmacol 50:306–313.

Bersani G, Grispini A, Marini S, Pasini A, Valducci M, Ciani N. 1990.5-HT2 antagonist ritanserin in neuroleptic-induced parkinsonism:a double-blind comparison with orphenadrine and placebo. ClinNeuropharmacol 13:500–506.

Bevan MD, Booth PAC, Eaton SA, Bolam JP. 1998. Selective inner-vation of neostriatal interneurons by a subclass of neuron in theglobus pallidus of the rat. J Neurosci 18:9438–9452.

Bonuccelli U, Ceravolo R, Salvetti S, D’Avino C, Del Dotto P, Rossi G,Murri L. 1997. Clozapine in Parkinson’s disease tremor—effects ofacute and chronic administration. Neurology 49:1587–1590.

Bruggeman R, Westerink BHC, Timmerman W. 1997. Effects of ris-peridone, clozapine and haloperidol on extracellular recordings ofsubstantia nigra reticulata neurons of the rat brain. Eur J Phar-macol 324:49–56.

Bubser M, Deutch AY. 1998. Thalamic paraventricular nucleus neu-rons colateralize to innervate the prefrontal cortex and nucleusaccumbens. Brain Res 787:304–310.

Burnet PWJ, Chen CPL-H, McGowan S, Franklin M, Harrison PJ.1996. The effects of clozapine and haloperidol on serotonin-1A, -2Aand -2C receptor gene expression and serotonin metabolism in therat forebrain. Neuroscience 73:531–540.

Calderon SF, Sanberg PR, Norman AB. 1988. Quinolinic acid lesionsof the rat striatum abolish D1- and D2-dopamine receptor mediatedcatalepsy. Brain Res 450:403–407.

Chesselet MF, Delfs JM. 1996. Basal ganglia and movement disor-ders: an update. Trends Neurosci 19:417–422.

Cornea-Hebert V, Riad M, Wu C, Singh SK, Descarries L. 1999.Cellular and subcellular distribution of the serotonin 5-HT2A re-ceptor in the central nervous system of adult rat. J Comp Neurol409:187–209.

Costall B, Olley JE. 1971. Cholinergic and neuroleptic induced cata-lepsy: modification by lesions in the globus pallidus and substantianigra. Neuropharmacology 10:581–594.

Costall B, Naylor RJ, Olley JE. 1972. Catalepsy and circling behav-iour after intracerebral injections of neuroleptic, cholinergic andanticholinergic agents into the caudate-putamen, globus pallidusand substantia nigra of rat brain. Neuropharmacology 11:645–663.

Cowan RL, Wilson CJ, Emson PC, Heizmann CW. 1990. Parvalbu-min-containing GABAergic interneurons in the rat neostriatum.J Comp Neurol 302:197–205.

Delfs JM, Anegawa NJ, Chesselet MF. 1995a. Glutamate decarbox-ylate messenger RNA in rat pallidum: comparison of the effects ofhaloperidol, clozapine and combined haloperidol-scopolamine treat-ments. Neuroscience 66:67–80.

Delfs JM, Ellison GD, Mercugliano M, Chesselet MF. 1995b. Expres-sion of glutamic acid decarboxylase mRNA in striatum and palli-dum in an animal model of tardive dyskinesia. Exp Neurol 133:175–188.

Deutch AY, Goldstein M, Roth RH. 1985. The ascending projections ofthe dopamine neurons of the substantia nigra, zona reticulata: acombined retrograde tracer-immunohistochemical study. NeurosciLett 71:257–263.

Deutch AY, Lee MC, Iadarola MJ. 1992. Regionally specific effects ofantipsychotic drugs on Fos expression: the nucleus accumbens shellas a locus of antipsychotic drug action. Mol Cell Neurosci 3:332–342.

Dunstan R, Broekkamp CL, Lloyd KG. 1981. Involvement of caudatenucleus, amygdala or reticular formation in neuroleptic and nar-cotic catalepsy. Pharmacol Biochem Behav 14:169–174.

Durif F, Vidailhet M, Assal F, Roche C, Bonnet AM, Agid Y. 1997.Low-dose clozapine improves dyskinesias in Parkinson’s disease.Neurology 48:658–662.

Fallon JH, Moore RY. 1978. Catecholamine innervation of the basalforebrain. IV. Topography of the dopamine projection to the basalforebrain and neostriatum. J Comp Neurol 180:545–580.

Farde L, Hall H, Ehrin E, Sedvall G. 1986. Quantitative analysis ofD2 dopamine receptor binding in the living human brain by PET.Science 231:258–261.

Farde L, Wiesel FA, Halldin C, Sedvall G. 1988. Central D2-dopaminereceptor occupancy in schizophrenic patients treated with antipsy-chotic drugs. Arch Gen Psychiatry 45:71–76.

Friedman J, Lannon M, Comella C, Factor S, Kurlan R, Richard I,Parsa M, Pfeiffer R, Davies R, Janko K, Brown D, Gardner I,Pearson N, Lare K, Rast S, Oakes D, Goetz C, Paulson G, MarshallF, Kieburtz K, Rudolph A, Bourgeois K, Casacelli C, Freimuth A,Guthrie B, Pelusio R, Watts A, Tariot P, Ruabertas, Greenamyre T.1999. Low-dose clozapine for the treatment of drug-induced psycho-sis in Parkinson’s disease. N Engl J Med 340:757–763.

Gerfen CR. 1989. The neostriatal mosaic: patch-matrix organizationis related to cortical lamination. Science 246:385–388.

Hamada S, Senzaki K, Hamaguchi-Hamada K, Tabuchi K, YamamotoH, Yamamoto T, Yoshikawa S, Okano H, Okado N. 1998. Localiza-tion of 5-HT2a receptor in rat cerebral cortex and olfactory systemrevrealed by immunohistochemistry using two antibodies raised inrabbit and chicken. Mol Brain Res 54:199–211.

Hauber W. 1998. Involvement of basal ganglia transmitter systems inmovement initiation. Prog Neurobiol 56:507–540.

Hauber W, Schmidt WJ. 1993. Discrete quinolinic acid lesions of thelateral, but not the medial caudate-putamen reversed haloperidol-induced catalepsy in rats. J Neural Transm [Gen Sect] 94:103–114.

Hauber W, Lutz S, Munkle M. 1998. The effects of globus palliduslesions on dopamine-dependent motor behaviour in rats. Neuro-science 86:147–157.

Heinz A, Knable MB, Weinberger DR. 1996. Dopamine D2 receptorimaging and neuroleptic drug response. J Clin Psychiatry 57(Suppl.11):84–88.

Henderson J, Yiannikas C, Graham JS. 1992. Effects of ritanserin, ahighly selective 5-HT2a antagonist, on Parkinson’s disease. ClinExp Neurol 29:277–282.

Ikeguchi K, Kuroda A. 1995. Mianserin treatment of patients withpsychosis induced by antiparkinsonian drugs. Eur Arch PsychiatryClin Neurosci 244:320–324.

Jakab RL, Goldman-Rakic PS. 1998. 5-Hydroxytryptamine2A seroto-nin receptors in the primate cerebral cortex: possible site of actionof hallucinogenic and antipsychotic drugs in pyramidal cell apicaldendrites. Proc Natl Acad Sci USA 95:735–740.

Kalkman HO, Neumann V, Nozulak J, Tricklebank MD. 1998. Cata-leptogenic effect of subtype selective 5-HT receptor antagonists inthe rat. Eur J Pharmacol 343:201–207.

Kapur S, Remington G, Jones C, Wilson A, DaSilva J, Houle S,Zipursky R. 1996. High levels of dopamine D2 receptor occupancywith low-dose haloperidol treatment: a PET study. Am J Psychiatry153:948–950.

Kaur S, Ozer H, Starr M. 1997. MK 801 reverses haloperidol-inducedcatalepsy from both striatal and extrastriatal sites in the rat brain.Eur J Pharmacol 332:153–160.

Kincaid AE, Wilson CJ. 1996. Corticostriatal innervation of the patchand matrix in the rat neostriatum. J Comp Neurol 374:578–592.

Kita H, Kitai ST. 1994. The morphology of globus pallidus projectionneurons in the rat: an intracellular staining study. Brain Res 636:308–319.

Kita H, Kosaka T, Heizmann CW. 1990. Parvalbumin-immunoreac-tive neurons in the rat neostriatum: a light and electronmicroscopicstudy. Brain Res 536:1–15.

Kubota Y, Mikawa S, Kawaguchi Y. 1993. Neostriatal GABAergicinterneurones contain NOS, calretinin or parvalbumin. Neurore-port 5:205–208.

Laprade N, Radja F, Reader TA, Soghomonian J-J. 1996. Dopaminereceptor agonists regulate levels of the serotonin 5-HT2a receptorand its mRNA in a subpopulation of rat striatal neurons. J Neurosci16:3727–3736.

Lucas G, Bonhomme N, De Deutwaerdere P, Le Moal M, SpampinatoU. 1997. 8-OH-DPAT, a 5-HT1a agonist and ritanserin, a 5-HT2a/2cantagonist reverse catalepsy independently of striatal dopaminerelease. Psychopharmacology 131:57–63.

Matsubura S, Matsubura R, Kusumi I, Koyama T, Yamashita I. 1993.Dopamine D1, D2 and serotonin2 receptor occupation by typical andatypical antipsychotic drugs in vivo. J Pharmacol Exp Ther 265:498–508.

McGeorge AJ, Faull RLM. 1989. The organization of the projectionfrom the cerebral cortex to the striatum in the rat. Neuroscience29:503–537.

Meltzer HY, Matsubura S, Lee JC. 1989. The ratios of serotonin2 anddopamine2 affinities differentiate atypical and typical antipsychoticdrugs. Psychopharmacol Bull 25:390–392.

Mengod G, Vilaro MT, Raurich A, Lopez-Gimenez JF, Cortes R, Pala-cios JM. 1996. 5-HT receptors in mammalian brain: receptor auto-radiography and in situ hybridization studies of new ligands andnewly identified receptors. Histochem J 28:747–758.

5-HT2A RECEPTORS IN STRIATAL AFFERENTS 303

Mijnster MJ, Raimundo AGV, Koskuba K, Klop H, Docter GJ, Groe-newegen HJ, Voorn P. 1997. Regional and cellular distribution ofserotonin 5-hydroxytryptamine2a receptor mRNA in the nucleusaccumbens, olfactory tubercle, and caudate-putamen of the rat.J Comp Neurol 389:1–11.

Miller CH, Fleischhacker WW, Ehrmann H, Kane JM. 1990. Treat-ment of neuroleptic induced akathisia with the 5-HT2 antagonistritanserin. Psychopharmacol Bull 26:373–376.

Miyawaki E, Meah Y, Koller WC. 1997. Serotonin, dopamine, andmotor effects in Parkinson’s disease. Clin Neuropharmacol 20:300–310.

Moratalla R, Quinn B, DeLanney LE, Irwin I, Langston JW, GraybielAM. 1992. Differential vulnerability of primate caudate-putamenand striosome-matrix dopamine systems to the neurotoxic effects of1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine. Proc Natl Acad SciUSA 89:3859–3863.

Morelli M, Porceddu ML, Imperato A, Di Chiara G. 1981. Role ofsubstantia nigra pars reticulata neurons in the expression of neu-roleptic-induced catalepsy. Brain Res 217:375–379.

Neal-Beliveau BS, Joyce JN, Lucki I. 1993. Serotonergic involvementin haloperidol-induced catalepsy. J Pharmacol Exp Ther 265:207–217.

Numan S, Lundgren KH, Wright DE, Herman JP, Seroogy KB. 1995.Increased expression of 5HT2 receptor mRNA in rat striatum fol-lowing 6-OHDA lesions of the adult nigrostriatal pathway. MolBrain Res 29:391–396.

Nyberg S, Farde L, Eriksson L, Halldin C, Eriksson B. 1993. 5-HT2and D2 dopamine receptor occupancy in the living human brain. APET study with risperidone. Psychopharmacology 110:265–272.

Parthasarathy HB, Graybiel AM. 1997. Cortically driven immediate-early gene expression reflects modular influence of sensorimotorcortex on identified striatal neurons in the squirrel monkey. J Neu-rosci 17:2477–2491.

Pirelli F, Adipietro A, Soldati G, Fattapposta F, Pozzessere G, Scope-tta C. 1998. Low dosage clozapine effects on L-dopa induced dyski-nesias in parkinsonian patients. Acta Neurol Scand 97:295–299.

Pompeiano M, Palacios JM, Mengod G. 1994. Distribution of theserotonin 5-HT2 receptors family mRNAs: comparison between5-HT2A and 5-HT2C receptors. Mol Brain Res 23:163–178.

Rajakumar N, Elisevich K, Flumerfelt BA. 1994. The pallidostriatalprojection in the rat: a recurrent inhibitory loop? Brain Res 651:332–336.

Roth BL, McLean S, Zhu X-Z, Chuang D-M. 1987. Characterization oftwo [3H]-ketanserin recognition sites in rat striatum. J Neurochem49:1833–1838.

Roth BL, Meltzer HY, Khan N. 1998. Binding of typical and atypicalantipsychotic drugs to multiple neurotransmitter receptors. AdvPharmacol 42:482–485.

Ruskin DN, Marshall JF. 1997. Differing influences of dopamineagonists and antagonists on Fos expression in identified popula-tions of globus pallidus neurons. Neuroscience 81:79–92.

Sanberg PR. 1980. Haloperidol-induced catalepsy is mediated bypostsynaptic dopamine receptors. Nature 284:472–473.

Scatton B, Worms P, Lloyd KG, Bartholini G. 1982. Cortical modula-tion of striatal function. Brain Res 232:331–343.

Schotte A, Janssen PF, Megens AA, Leysen JE. 1993. Occupancy ofcentral neurotransmitter receptors by risperidone, clozapine andhaloperiodol, measured ex vivo by quantitative autoradiography.Brain Res 631:191–202.

Shu SY, Peterson GM. 1988. Anterograde and retrograde axonaltransport of Phaseolus vulgaris leucoagglutinin PHA-L from theglobus pallidus to the striatum of the rat. J Neurosci Methods25:175–180.

Smith AD, Bolam JP. 1990. The neural network of basal ganglia asrevealed by the study of synaptic connections of identified neurones.Trends Neurosci 13:259–265.

Staines WA, Fibiger HC. 1984. Collateral projections of neurons of therat globus pallidus to the striatum and substantia nigra. Exp BrainRes 56:217–220.

Staines WA, Atmadja S, Fibiger HC. 1981. Demonstration of a pal-lidostriatal pathway by retrograde transport of HRP-labeled lectin.Brain Res 206:446–450.

Wadenberg ML, Ahlenius S. 1995. Antagonism by the 5-HT2a/c re-ceptor antagonist DOI of raclopride-induced catalepsy in the rat.Eur J Pharmacol 294:247–251.

Waldmeier PC, Delini-Stula AA. 1979. Serotonin-dopamine interac-tions in the nigrostriatal system. Eur J Pharmacol 55:363–373.

Ward RP, Dorsa DM. 1996. Colocalization of serotonin receptor sub-types 5-HT2A, 5-HT2C, and 5-HT6 with neuropeptides in rat stri-atum. J Comp Neurol 370:405–414.

Willins DL, Deutch AY, Roth BL. 1997. Serotonin 5-HT2a receptorsare expressed on pyramidal cells and interneurons in the rat cortex.Synapse 27:79–82.

Worms P, Willigens M-T, Continsouza-Blanc D, Lloyd KG. 1985. Theeffects of different types of cortical lesions on drug-induced cata-lepsy in rats: a pharmacological analysis. Eur J Pharmacol 113:53–59.

Wright DE, Seroogy KB, Lundgren KH, Davis BM, Jennes L. 1995.Comparative localization of serotonin1A, 1C, and 2 receptor sub-type mRNAs in rat brain. J Comp Neurol 351:357–373.

Wu C, Yoder EJ, Shih J, Chen K, Dias P, Shi L, Ji -D, Wei J, ConnerJM, Kumar S, Ellisman MH, Singh SK. 1998. Development andcharacterization of monoclonal antibodies specific to the serotonin5-HT2a receptor. J Histochem Cytochem 46:811–824.

Yoshida Y, Ono T, Kizu A, Fukushima R, Miyagishi T. 1991. StriatalN-methyl-D-aspartate receptors in haloperidol-induced catalepsy.Eur J Pharmacol 203:173–180.

Young CD, Bubser M, Meltzer HY, Deutch AY. 1999. Clozapine pre-treatment modifies haloperidol-elicited forebrain fos induction: aregionally specific double dissociation. Psychopharmacology 144:255–263.

Zadow B, Bubser M, Schmidt WJ. 1994. Quinolinic acid-induced le-sion of the substantia nigra pars reticulata: behavioural and neu-rochemical effects. Soc Neurosci Abstr 20:1365.

304 M. BUBSER ET AL.