sub-chronic treatment with classical but not atypical antipsychotics produces morphological changes...

Sub-chronic treatment with classical but not atypicalantipsychotics produces morphological changes in ratnigro-striatal dopaminergic neurons directly related to`early onset' vacuous chewing

Giorgio Marchese,1 Maria Antonietta Casu,1 Francesco Bartholini,1 Stefania Ruiu,1 Pierluigi Saba,3

Gian Luigi Gessa1,2,3 and Luca Pani1,2

1Neuroscienze S.c.a.r.l.,2Institute of Neurogenetics, Neuropharmacology, C.N.R. and3`B.B. Brodie' Department of Neuroscience, University of Cagliari, Italy

Keywords: neurotoxicity, Parkinson, tardive dyskinesia

Abstract

In the present work, we investigated if an impairment of dopaminergic neurons after subchronic haloperidol treatment might be a

possible physiopathologic substrate of the `early onset' vacuous chewing movements (VCMs) in rats. For this purpose, different

antipsychotics were used to analyse a possible relationship between VCMs development and morphological alterations of

tyrosine-hydroxylase-immunostained (TH-IM) neurons.Rats treated twice a day with haloperidol displayed a signi®cant increase of VCMs that was both time- (2±4 weeks) and dose

(0.1±1 mg/kg) dependent. Immunocytochemical analysis showed a shrinkage of TH-IM cell bodies in substantia nigra parscompacta and reticulata and a reduction of TH-immunostaining in the striatum of haloperidol treated rats with the arising of

VCMs. No differences were observed in TH-IM neurons of ventral tegmental area and nucleus accumbens vs. control rats. The

atypical antipsychotics risperidone (2 mg/kg, twice a day), amisulpride (20 mg/kg, twice a day) and clozapine (10 mg/kg, twice aday) did not produce any nigro-striatal morphological changes or VCMs. TH-IM nigro-striatal neuron morphological alterations and

VCMs were still present after three days of withdrawal in rats treated for four weeks with haloperidol (1 mg/kg). Both the main

morphological changes and the behavioural correlate disappeared after three weeks of withdrawal. These results suggest thathaloperidol induces a morphological impairment of the dopaminergic nigro-striatal neurons which is directly associated with the

arising, permanency and disappearance of VCMs in rats.

Introduction

Neuroleptic drugs are chronically used in the treatment of schizo-

phrenia and other psychotic disorders but, unfortunately, their use is

often associated with acute and delayed motor side-effects, including

parkinsonism, akathisia and tardive dyskinesia (TD). TD, character-

ized by involuntary movements predominantly in the orofacial

region, develops in up to 20% of patients chronically treated with

classical neuroleptics such as haloperidol (Klawans & Rubovits,

1972; Tarsy & Baldessarini, 1974, 1977; Jeste & Wyatt, 1979).

Although the relationship between TD and long-term haloperidol

treatment has been established, the pathophysiology of this motor

disturbance is still unknown. A peculiar characteristic of this

syndrome is that TD generally persists after haloperidol withdrawal

and occasionally becomes irreversible, indicating that haloperidol has

produced long lasting changes in brain function that are no longer

related to the presence of the drug (Meshul et al., 1992, 1994). Rats

treated with repeated high doses of haloperidol develop vacuous

chewing movements (VCMs), a syndrome similar to human TD for

time course and signs (Ellison & See, 1989; Tamminga et al., 1990;

Waddington, 1990, 1997; Egan et al, 1996). Enhanced oral activity

has been reported after subchronic (3±4 weeks) exposure to

haloperidol (early onset VCMs) (Egan et al., 1996; Diana et al,

1992) or only after chronic (6±12 months) drug administration

(tardive VCMs) (Waddington, 1990; Egan et al., 1996). Differences

in these reports led to different interpretations of the results (Egan

et al., 1996; Ikeda et al., 1999) and several hypotheses have been

formulated to explain the development of VCMs, including the

existence of a dopaminergic receptor supersensitivity (Klawans &

Rubovits, 1972, 1977; Tarsy & Baldessarini, 1974; Burt et al., 1977),

a dopamine D1/D2 receptor imbalance (Casey, 1995; Peacock &

Gerlach, 1997; Waddington, 1997) and a g-aminobutyric acid

(GABA) de®ciency (Gale, 1980; Fibiger & Lloyd, 1984). All these

studies considered VCMs as a consequence of the haloperidol

mediated receptor blockade and focused their attention on basal

ganglia circuits and on the altered neurotransmitter-receptor equilib-

rium in these areas. Recently, the emphasis on VCMs studies has

shifted from neurotransmitter models to an approach that considers

haloperidol-induced VCMs as a consequence of a direct neurotoxic

effect mediated by haloperidol itself (De Keyser, 1991). Ex vivo

studies have reported neuropathological alterations in basal ganglia of

Correspondence: Dr Luca Pani, as above.E-mail: [email protected]

Received 22 August 2001, revised 3 January 2002, accepted 8 February 2002

European Journal of Neuroscience, Vol. 15, pp. 1187±1196, 2002 ã Federation of European Neuroscience Societies

rats chronically treated with haloperidol, such as an increase of

perforated postsynaptic density (Meshul et al., 1992) and a spine

density alteration (Kelley et al., 1997). Moreover, signi®cant

neuronal loss and glial reactions have been shown in the caudate

nucleus of patients who had received chronic treatment with classical

neuroleptic (Nielsen & Lyon, 1978; Andreassen et al., 1998).

Accordingly, several studies demonstrated that an haloperidol

metabolite, haloperidol-pyridinium (HPP+), shared structural simil-

arity and toxic effect with 1-methyl-4-phenylpyridinium (MPP+), the

metabolite that mediates the toxicity on nigro-striatal dopaminergic

neurons induced by the administration of 1-methyl-4-phenyl-1,2,3,6-

tetrahydropyridine (MPTP) (Fang et al, 1995a; Wright et al., 1998;

Petzer et al., 2000). Moreover, some researchers pointed out that the

increase in cathecolamines turnover mediated by haloperidol may

result in a particular vulnerability of basal ganglia dopaminergic

neurons due to dopamine auto-oxidation (Cadet, 1988; Lohr, 1991).

The possibility that haloperidol might damage the dopaminergic

neurons, led us to investigate the presence of morphological

alterations in these neurons after subchronic haloperidol treatment

and, to verify whether a correlation exists between any morphological

changes and the development of early onset VCMs in rats. In addition

we wanted to compare the effects of haloperidol with those of three

atypical antipsychotics: clozapine, risperidone and amisulpride,

which possess low propensity to produce acute and delayed

extrapyramidal side-effects in the clinical studies.

Materials and methods

Animals

Male Sprague±Dawley albino rats (Charles River, Como, Italy)

weighting 100±125 g were kept on a 12-h light : 12 h dark cycle with

food and tap water available ad libitum.

Animals (12 rats for each experimental group) were intraperitone-

ally treated twice a day (at 09.00 h and 21.00 h) with the different

drugs (doses are always refereed to the drug base form).

All experimental protocols were approved by the Ethical

Committee at the University of Cagliari and performed in strict

accordance with the E.C. regulation for care and use of experimental

animals (CEE N°86/609).

Drugs and chemicals

Haloperidol hydrochloride and clozapine (free base) were purchased

from Tocris Cookson Ltd. (Avonmouth Bristol, UK). Risperidone

(free base), 3,3¢-diaminobenzidine and Entellan were from Sigma Co.

(St. Louis, MO, USA). Amisulpride mesylate was generously

supplied by Sano®-SyntheÂlabo (Bagneaux, France).

Immunocytochemistry

Rats (9 animals for each experimental group randomly chosen among

the 12 treated rats) were anaesthetized with Equithesin (2.5 mg/kg,

i.p.) and perfused transcardially with 4% paraformaldehyde in 0.1%

phosphate buffer (PB), pH 7.4. The brains were subsequently post-

®xed in the same ®xative for two hours and cryoprotected overnight

with a solution of 30% sucrose in 0.1 M PB at 4 °C. Forty micrometer

(in substantia nigra and VTA studies) or 20 mm (in striatum and

nucleus accumbens studies) thick coronal sections were cut using a

vibratome (Leica VT1000S, Leica Microsystems, Nussloch, Germany)

or a cryostat (Leica CM3050, Leica Microsystems), respectively. Free-

¯oating cytochemical stainings were performed in strict accordance

with the general methodological procedures indicated by CoteÁ et al,

1993). Controls were performed by subtracting the primary antibody in

the immunocytochemistry procedure.

Immunostaining for tyrosine hydroxylase (TH)

Brie¯y, after rinsing in phosphate-buffered saline with 0.2%

Triton X-100 (PBS + T), sections were incubated with 0.3% of

H2O2 in PBS and, after extensive washing, with a blocking solution

containing 1% BSA and 20% normal goat serum in PBS + T to

reduce background. Sections were then incubated overnight at 4 °C

with a mouse anti-TH antibody (1 : 400; from Chemicon, Temecula,

CA, USA). After rinsing, sections were incubated with a goat anti-

mouse biotinylated IgG (1 : 600; Vector, Burlingame, CA, USA) for

one hour, followed by an avidin-biotin complex (1 : 500; Vectastain

ABC kit from Vector) for an additional hour.

Immunostaining for dopamine transporter (DAT)

Sections were extensively washed in Tris-buffered saline (TBS) and

then incubated in 0.3% H2O2 in TBS for 15 min. Subsequently,

sections were incubated for 1 h in 1% BSA and 20% normal goat

serum in TBS + 0.3% Triton X100 (TBS + T) and then incubated

with a rat anti-DAT antibody (1 : 5000; Chemicon) for two days at

4 °C. After rinsing, a goat anti-rat biotinylated IgG (1 : 800; Vector)

was added for 1 h. Sections were incubated with ABC (1 : 500) for

1 h.

After ABC incubation and washing, sections for both TH and DAT

staining were exposed to 3,3¢-diaminobenzidine (0.06% in PBS)

containing 1% cobalt chloride and 1% nickel ammonium sulphate for

15 min. Immunostaining was developed by adding 5 mL H2O2 (0.1%

in PBS) to each 500 mL of 3,3¢-diaminobenzidine. After washing in

PBS + T, all sections were mounted on gelatin-coated glass slides,

air-dried, dehydrated in ascending concentrations of ethanol, cleared

with xylene, and coverslipped with Entellan. The mounted sections

were examined under a BX-60 Olympus light-microscope (Olympus

Optical Co. GmbH, Hamburg, Germany).

Quantitative analysis of immunostaining in the striatum andnucleus accumbens

The region of analysis corresponded to bregma coordinates +2.2 to

+0.7 mm as indicated in the Paxinos & Watson atlas (1986) The

immunostaining in the striatum and nucleus accumbens, was

quanti®ed using an image analysis system (KS 300; Karl Zeiss

Vision GmbH, Hallbergmoos, Germany). The analysis was carried

out by taking, for each section, ®ve randomly chosen ®elds and at

least ten alternate sections from each animal (n = 9). The quanti-

®cation was carried out by measuring the percentage of the area

occupied by TH-positive ®bers with respect to a 2000 mm2

standardized area (IM area percentage), using a 3100 objective

lens and a Zeiss Sound Vision digital microscope video camera (total

magni®cation on computer screen 75003). Gray values, measured in

immunostaining negative areas of the sections, were subtracted as

background from the resulting binary picture (Bjorklund &

Stromberg, 1997). The analysis procedure was strictly carried out

as a blind study.

Cell body size measurement

The region of analysis corresponded to bregma coordinates ±4.8 to

±6.04 mm, in order to evaluate the cell body size of substantia nigra

(pars compacta and reticulata) and VTA, known to be populated by

dopaminergic neurons. For each section (eight sections for each rat),

cell body cross-sectional area (mm2) was determined, using a 340

objective lens (total magni®cation on computer screen 30003), in

those neurons where nuclei could be observed. An average of 60 cells

in substantia nigra pars compacta, 25 cells in pars reticulata and 40 in

VTA were quanti®ed for each rat.

1188 G. Marchese et al.

ã 2002 Federation of European Neuroscience Societies, European Journal of Neuroscience, 15, 1187±1196

VCMs quanti®cation

VCMs were measured 12 h, 3 days and 3 weeks after the last dose of

neuroleptic. Animals (12 rats for each experimental group) at least

12 h withdrawn from the different drugs used in this study, were

placed individually in a small (30 3 20 3 30 cm) plexiglass cage,

and were allowed to adapt to the observation cage for a period of 1 h.

In the morning (between 09.00 h and 11.00 h) VCMs were scored

FIG. 1. Micrographs showing TH-IM ®bers in the striatum of rats treated twice a day with vehicle (A) and haloperidol 1 mg/kg (B). (C and D) Micrographsshowing TH positive neurons (arrows) from substantia nigra pars compacta of rats treated with vehicle and haloperidol 1 mg/kg, respectively. (E and F) TH-IM cell bodies from substantia nigra pars reticulata (arrows) of rats treated twice a day with vehicle and haloperidol 1 mg/kg, respectively. Scale bars, 10 mm(A and B); 20 mm (C±F).

Morphological changes in dopaminergic neurons 1189


during a 30-min observation period, and counting of VCMs was

stopped whenever the animal began grooming and resumed when

grooming stopped (Gunne et al., 1982). For calculation purposes,

each burst of purposeless chewing or of jaw tremor was counted as

one, if its duration was at least of 3 s. A VCM consisted of a rapid

movement of the jaw, which resembled chewing but did not appear to

be directed at any particular stimulus. In all the experiments, the

observer was blind to the treatment given to the rats.

Statistical analysis

The statistical signi®cance of the effect of the different treatments

were evaluated by Student's t-test, one-way or two-ways analysis of

variance (ANOVA). When a signi®cant interaction (P < 0.05) was

demonstrated, the Newman±Keuls post hoc test was applied to

compare the effect of different treatments.

Results

VCMs and morphological changes in rats following 4 weeks oftreatment with haloperidol (0.1 mg/kg and 1 mg/kg)

A dense TH-immunoreactivity was observed in the striatum (Fig. 1A

and B) with an apparent compartmental organization and granular

texture. This pattern of labelling was similar in the DAT-

immunostaining (Ciliax et al., 1995). TH-IM neurons were seen

throughout the entire substantia nigra pars compacta (Fig. 1C and D)

and were rarely distributed in pars reticulata (Fig. 1E and F). The

somatodendritic pro®le was characterized by large perikarya with

dendritic arborization. Control sections did not display any staining.

The comparison of striatal TH-IM area percentage between rats

treated with vehicle, haloperidol 0.1 mg/kg and haloperidol 1 mg/kg,

displayed a signi®cant difference (one-way ANOVA F2,24 = 23.2,

P < 0.01). A signi®cant lower TH-IM area percentage was found in

the striatum of rats treated for 4 weeks with haloperidol at the dose of

1 mg/kg vs. vehicle treated rats (P < 0.01) (Fig. 2A). Conversely,

haloperidol at the dose of 0.1 mg/kg produced only a small reduction

of TH-IM area percentage that did not reach statistical signi®cance.

The evaluation of cell body size in substantia nigra pars compacta

and reticulata, showed signi®cant differences between the three

different groups of treatment (one-way ANOVA F2,24 = 13.72,

P < 0.01, pars compacta; F2,24 = 6.07, P < 0.05, pars reticulata)

(Fig. 2B). A signi®cant reduction of cell body size only in rats treated

with haloperidol 1 mg/kg vs. vehicle was found both in pars

compacta (P < 0.01) and pars reticulata (P < 0.05) (Fig. 2B).

After 4 weeks of treatment with vehicle or haloperidol (0.1 or

1 mg/kg), signi®cant differences in VCMs were observed (one-way

ANOVA, F2,33 = 39.63, P < 0.01). At the dose of 1 mg/kg of

haloperidol, rats showed signi®cantly more VCMs with respect to

rats treated with vehicle (P < 0.01) (Fig. 2C), in contrast no

signi®cant differences in VCMs were found after haloperidol at the

dose of 0.1 mg/kg (Fig. 2C).

A dense TH staining was also found in the core (Fig. 3A and B)

and shell of the nucleus accumbens (Fig. 3C and D). Moreover, a vast

number of small and rounded oligodendritic TH-IM dopamine

neurons were observed in VTA (Fig. 3E and F). No differences

were found between the treatments in the TH-IM area percentage

both in core (one-way ANOVA, F2,24 = 0.21, P > 0.05) and shell of

nucleus accumbens (one-way ANOVA, F2,24 = 0.38, P > 0.05)

(Fig. 4A). The analysis of cell body area also revealed no differences

between the treatments in the VTA (one-way ANOVA, F2,24 = 0.42,

P > 0.05) (Fig. 4B).

The use of the antibody against DAT con®rmed the striatal

reduction of IM area percentage observed using TH-antibody, in

haloperidol treated rats vs. vehicle (Student's t-test, P < 0.05)

(Fig. 5).

VCMs and morphological changes in rats following 2, 3,4 weeks of treatment with haloperidol and withdrawal

Rats treated with haloperidol (1 mg/kg, twice a day) for 2, 3, 4 weeks

and in withdrawal for 3 days and 3 weeks (after 4 weeks haloperidol

treatment) showed a signi®cant alterations both in striatum (one-way

ANOVA, F5,48 = 11.4, P < 0.01) and in nigra pars compacta (one-way

ANOVA, F5,48 = 24.3, P < 0.01) and reticulata (one-way ANOVA,

F5,48 = 21.16, P < 0.01). In the same experimental groups signi®cant

differences in VCMs were observed (two-ways ANOVA test,

FIG. 2. (A) Histogram showing the TH-IMarea percentage in the striatum of rats treatedtwice a day with vehicle, haloperidol 0.1 and1 mg/kg. A signi®cant reduction in the TH-IMarea percentage was found with haloperidol atthe dose of 1 mg/kg. (B) Cell body area ofTH-IM neurons in the substantia nigra parscompacta and reticulata. Haloperidol, at thedose of 1 mg/kg, determined a signi®cantreduction of cell body area both in parscompacta and reticulata with respect tovehicle. (C) Histogram illustrating the numberof VCMs in rats treated twice a day withvehicle and haloperidol at the dose of 0.1 and1 mg/kg. This last dose caused a signi®cantincrease of VCMs. Statistical signi®cance vs.respective control treated rats has beenanalyzed using one-way ANOVA test followedby Newman-Keuls post hoc test (*P < 0.05;**P < 0.01, n = 9 in morphological andn = 12 in behavioural studies). In all theexperiments, rats received the last injection ofhaloperidol or vehicle 12 h before themorphological and behavioural experimentalprocedure.



F1,22 = 698, P < 0.001). Rats treated with haloperidol (1 mg/kg) for

2 weeks, showed no difference in striatal TH-IM area percentage

(Fig. 6A) or in cell body size in substantia nigra pars compacta and

reticulata (Fig. 6B), but displayed signi®cant VCMs increase with

respect to vehicle treated rats (P < 0.05) (Fig. 6C). After three weeks

of treatment, there was a signi®cant decrease of TH-IM area

percentage in striatum (P < 0.05) (Fig. 6A), while a minor, although

not signi®cant, decrease was found in cell body area of TH-IM

FIG. 3. Micrographs of TH-IM ®bers in the core of the nucleus accumbens of rats treated twice a day with vehicle (A) and haloperidol 1 mg/kg (B). (C andD) TH-IM ®bers in the shell of the nucleus accumbens of rats treated with vehicle and haloperidol 1 mg/kg, respectively. (E and F) Micrographs showing THpositive neurons (arrows) from VTA of rats treated with vehicle and haloperidol, respectively. Scale bars, 10 mm (A±D); 20 mm, (E and F).



neurons in substantia nigra pars compacta and reticulata (Fig. 6B).

After 3 weeks of haloperidol treatment, rats showed a high signi®cant

increase of VCMs compared to vehicle treated rats (P < 0.01)

(Fig. 6C).

In accordance to the results showed above, rats treated with

haloperidol for 4 weeks, displayed a decrease of TH-IM area

percentage in striatum and a cell body shrinkage of TH-IM neurons

in substantia nigra pars compacta and reticulata, associated with an

increase of VCMs (Fig. 6A±C).

The morphological alterations, observed after 4 weeks of halo-

peridol treatment, persisted in the different areas after 3 days of

withdrawal (P < 0.01 in each condition vs. vehicle treated rats)

(Fig. 6A and B). Moreover rats still exhibited signi®cantly more

VCMs than vehicle treated rats (Fig. 6C). After 3 weeks of

withdrawal, both VCMs and striatal TH-immunostaining were

completely normalized and no differences could be observed with

respect to vehicle treated rats (Fig. 6A±C). However in substantia

nigra pars reticulata (P < 0.05) but not in compacta the shrinkage of

the cell body area still persisted after 3 weeks of drug withdrawal

(Fig. 6B).

VCMs and morphological changes in rats following 4 weeks oftreatment with haloperidol and different atypical neuroleptics

The comparison of the effect after 4 weeks treatment with vehicle,

haloperidol (1 mg/kg, twice a day) or atypical neuroleptics such as

risperidone (2 mg/kg, twice a day), clozapine (10 mg/kg, twice a day)

FIG. 4. (A) Histogram showing the TH-IM area percentage in the nucleus accumbens (core and shell) of rats treated twice a day with vehicle, haloperidol 0.1and 1 mg/kg. No differences were found in TH-IM nerve ®bre between treatments (n = 9). (B) Cell body area of TH-IM neurons in the VTA. No differenceswere found between treatments.

FIG. 5. DAT-IM area percentage in the striatum of rats treated twice a day with vehicle and haloperidol 1 mg/kg. A signi®cant reduction in the DAT-IM areapercentage was found with haloperidol at the dose of 1 mg/kg. Statistical signi®cance vs. respective control treated rats has been analysed using Student's t-test (*P < 0.05, n = 9). Rats received the last injection of haloperidol or vehicle 12 h before morphological experimental procedure.



and amisulpride (20 mg/kg, twice a day), displayed signi®cant

morphological differences between groups in striatum (one-way

ANOVA, F4,40 = 18.27, P < 0.01), nigra pars compacta (one-way

ANOVA, F4,40 = 4.69, P < 0.01) and nigra pars reticulata (one-way

ANOVA, F4,40 = 4.16, P < 0.01) (Fig. 7A and B). Signi®cant staining

reductions were con®rmed in rats treated with haloperidol, but no

signi®cant differences were found after atypical neuroleptics treat-

ments vs. vehicle treated rats. In the same manner VCMs increased in

haloperidol treated rats but did not change with atypical neuroleptic

treatments compared to vehicle treated rats (Fig. 7C).

Discussion

The present results indicated that a four-week treatment with high

doses of haloperidol can profoundly affect the morphology of

nigrostriatal dopaminergic neurons in rats. The immunocytochemical

study showed that haloperidol decreased (by approximately 30%) the

normal TH-immunostaining in striatum and produced a signi®cant

reduction of cell body size of dopaminergic neurons in the substantia

nigra. The use of two antibodies directed against two different

markers (TH and DAT) for dopaminergic ®bers indicated that the

staining reduction observed in striatum was likely due to a

morphological alteration of dopaminergic ®bers rather than to the

marker expression levels. The quanti®cation method applied in the

present work does not allow us to state that the observed striatal TH-

immunostaining reduction should be uniquely attributed to a

FIG. 7. (A) Histogram showing the TH-IM area percentage in the striatum of rats treated with vehicle, haloperidol 1 mg/kg (twice a day) and atypicalneuroleptics risperidone (2 mg/kg, twice a day), clozapine (10 mg/kg, twice a day) and amisulpride (20 mg/kg, twice a day) for four weeks. A signi®cantreduction in the TH-IM area percentage was found with haloperidol at the dose of 1 mg/kg but not with the atypical neuroleptics. (B) Cell body area of TH-IM neurons in the substantia nigra pars compacta and reticulata. Compare with vehicle, haloperidol determined a signi®cant reduction of cell body area bothin pars compacta and reticulata, while atypical neuroleptics had no effect. (C) Histogram illustrating the number of VCMs in rats treated twice a day withvehicle, haloperidol 1 mg/kg and atypical neuroleptics risperidone (2 mg/kg), clozapine (10 mg/kg) and amisulpride (20 mg/kg). Haloperidol caused asigni®cant increase of VCMs, while the atypical neuroleptics did not cause any change in VCMs compared to vehicle. Statistical signi®cance vs. respectivecontrol treated rats has been analyzed using ANOVA test followed by Newman-Keuls post hoc-test (*P < 0.05; **P < 0.01, n = 9 in morphological and n = 12in behavioural studies). In all the experiments and for each drug, rats received the last injection of 12 h before the morphological and behaviouralexperimental procedure.

FIG. 6. (A) Histogram showing the TH-IM area percentage in the striatumof rats treated twice a day with vehicle, haloperidol 1 mg/kg for 2, 3,4 weeks and after 3 days and 3 weeks of withdrawal. A signi®cantreduction in the TH-IM area percentage was observed at 3 and 4 weeks oftreatment and this reduction was still present at 3 days of withdrawal. (B)Cell body area of TH-IM neurons in the substantia nigra pars compacta andreticulata. Haloperidol (1 mg/kg) administered for 4 weeks and after awithdrawal of 3 days determined a signi®cant reduction of cell body areaboth in pars compacta and reticulata with respect to vehicle. In parsreticulata the reduction of nerve ®bre density persisted at 3 weeks ofwithdrawal. (C) Histogram illustrating the number of VCMs in rats treatedtwice a day with vehicle and haloperidol at the dose of 1 mg/kg in the samegroups described above. Statistic analysis for morphological study (n = 9)was carried out using one-way ANOVA while vacuous chewing quanti®cation(n = 12) was followed by two-way ANOVA; in both cases Newman-Keulspost hoc test was applied to compare haloperidol treated rats with respect tocontrols (*P < 0.05; **P < 0.01). (Rats treated with haloperidol 1 mg/kg for2, 3, 4 weeks rat received the last administration 12 h before morphologicaland behavioural experimental procedure.)



dopamine ®bre depletion. However, Cottingham et al. (1990) found

no changes in nigral TH message following chronic haloperidol

treatment and at 16 h withdrawal period. Moreover, Lerner et al.

(1977) reported changes of TH Vmax after chronic haloperidol

administration and suggested an elevated TH protein expression.

Similarly, different authors showed that DAT binding density (Allard

et al., 1990; Rivest et al., 1995; Ase et al., 1999) and RNA level

(Rivest et al., 1995) remained unaltered in the nigro-striatal neurons

after chronic haloperidol treatment. Hence, both the overlapping of

results obtained by using DAT- and TH-antibody and the observed

shrinkage of dopaminergic cell bodies in substantia nigra supports the

hypothesis that haloperidol alters the morphology of dopamine nigro-

striatal neurons. Consequently, we suggest that the observed TH-

immunostaining reduction might be related to a decrease of

dopaminergic ®bers in striatum or alternatively to a reduced caliber

of such ®bers, instead of a TH down regulation.

The reversible alterations in the striatum associated with the

shrinkage of dopaminergic neuron cell bodies in the substantia nigra

after chronic treatment with haloperidol, could be interpreted as a

sign of functional impairment of the nigro-striatal dopaminergic

pathway. Different studies, in fact, have reported cell body shrinkage

and axonal density reduction in nigro-striatal dopaminergic neurons

in neuro-degenerative processes induced by toxic drugs, like

6-hydroxydopamine (6-OHDA), MPTP and by high concentrations

of dopamine (Sherman & Moody, 1995; Simantov et al., 1996;

Cochiolo et al, 2000), or in pathological states such as Parkinson's

and Huntington's diseases (Waters et al., 1988). Interestingly,

haloperidol affected the nigrostriatal but not the meso-limbic

dopaminergic neurons, resembling what has been observed in

MPTP treated animals (Hung & Lee, 1998; Nomoto et al, 2000),

suggesting that the present results might be related with several

studies showing that haloperidol could produce a neuronal toxicity

similar to that exerted by MPP + (Bloomquist et al., 1994). In

addition, haloperidol and its metabolites have been shown to inhibit

monoamino oxidase type B (MAO-B) and by this inhibition to

possibly increase dopamine auto-oxidation and the production of free

radicals. (Fang & Yu, 1995b). Free radicals possess many disruptive

toxic effects that depend on the detoxifying enzymes' pro®le of the

different cells. In this regard, basal ganglia are supposed to be

particularly vulnerable to lipid peroxidation induced by free radicals

resulting from catecholamine oxidation (Cadet, 1988; Lohr, 1991).

Haloperidol treatment has also been shown to increase glutamate

release (See & Lynch, 1995; De Souza et al., 1999) and glutamatergic

synapses in striatum (Meshul et al., 1994), suggesting a possible

toxicity induced by calcium.

All these hypotheses on haloperidol toxicity are not con¯icting but

rather represent different attempts to identify a ®nal common

pathway leading to neuronal abnormalities that might underlie the

development of VCMs in rodents.

In the present study, we found a strong dose-dependent and, after

three weeks of treatment, a temporal correlation between the

development of rat èarly onset' VCMs and the appearance of

morphological de®cits in the nigro-striatal dopaminergic neurons.

This correlation was maintained in striatum and substantia nigra pars

compacta when the haloperidol treatment was terminated and normal

behaviour re-established, suggesting a possible direct impairment of

the nigro-striatal dopaminergic neurons in the physiopathology of

VCMs in rats. On the other hand, the maintenance of cell body

shrinkage in substantia nigra pars reticulata, after behaviour recovery,

leads us to wonder about the relevance of these dopaminergic neurons

on the development of VCMs. In agreement with the hypothesis that a

nigro-striatal dopaminergic neurons impairment induces VCMs in

rats, it has been shown that striatal lesions produced by intracerebral

injection of 6-OHDA exerted purposeless chewing in rats (Jicha &

Salamone, 1991). Moreover an increase in VCMs in neonatally

lesioned rats after haloperidol treatment was observed (Huang et al.,

1997). Of course, as the source of neuronal damage is different, a

direct relation between our results and those of these studies should

be considered a daring attempt. However, it is remarkable that a

speci®c nigro-striatal neuronal damage might lead to a high count of

VCMs.

An additional aspect that might correlate VCMs and a dopamin-

ergic neuronal alteration, is the down-regulation of GABA receptor in

globus pallidus after chronic haloperidol treatment and after 6-OHDA

striatal lesion (Pan et al., 1985; Sasaki et al., 1997). Reduced

GABAergic activity has been correlated with enhanced oral activity

in neuroleptic-treated animals (Gunne et al., 1984, 1986) and

proposed as a pathophysiological hypothesis for TD (Fibiger &

Lloyd, 1984). Very recently, the impairment of the striato-nigral

GABAergic system induced by haloperidol was not found after

atypical antipsychotic treatment (Sakai et al., 2001).

In contrast to that observed with haloperidol, we did not observe

any morphological changes or development of VCMs in rats treated

with high doses of three unrelated classes of atypical antipsychotics:

the dibenzodiazepine clozapine, the benzo-iso-oxazol derivative

risperidone and the substituted benzamide amisulpride, con®rming

the correlation between the nigro-striatal morphological impairments

and the development of èarly onset' VCMs in rats.

The nigro-striatal dopamine neuron impairment observed in

the present study, could add further explanations to the

dopamine receptor alterations induced by haloperidol treatments,

because 6-OHDA-mediated lesions of the nigrostriatal neurons

produce not only a D2 supersensitivity to apomorphine but

also a D1/D2 dopamine receptor density imbalance (Kelly et al.,

1975; Lau & Fung, 1986; Lange, 1990; Narang & Wamsley, 1995),

recently proposed as a substrate for VCMs (Peacock & Gerlach,

1997).

In conclusion, the data presented here raise the possibility that

structural alterations and damages of dopaminergic neuronal archi-

tecture induced by haloperidol, may be involved in the development

of èarly onset' oral dyskinesia in rats. Further studies are needed to

identify which neurons are primarily involved in the morphological

changes observed after chronic haloperidol treatment and to address

the clinical relevant question whether longer treatment (i.e. 6±

12 months) with haloperidol doses more similar to that used in

human therapy might produce the same damage, and if this damage

could permanently affect the morphology of the nigro-striatal

dopaminergic system and the behaviours related to its proper

function. Moreover, a possible correlation between the dopaminergic

neuron morphological impairments and the different levels of

`tardive' VCMs (high and low), described in rats treated with

haloperidol for a longer period of time (Tamminga et al., 1990;

Meshul et al., 1996) cannot be excluded.

Abbreviations

DAT, dopamine transporter; GABA, g-aminobutyric acid; HPP+, haloperidolpyridinium; MPP+, 1-methyl-4-phenylpyridinium; MPTP, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine; 6-OHDA, 6-hydroxydopamine; PB, phosphatebuffer; PBS + T, phosphate buffered-saline + 0.2% Triton X-100; TBS, Tris-buffered saline; TBS + T, Tris-buffered saline + 0.3% Triton X-100; TD,tardive dyskinesia; TH, tyrosine hydroxylase; TH-IM, tyrosine hydroxylase-immunostained; VCMs, vacuous chewing movements; VTA, ventral teg-mental area.



References

Allard, P., Eriksson, K., Ross, S.B. & Marcusson, J.O. (1990) Unaltered [3H]-GBR-12935 binding after chronic treatment with dopamine active drugs.Psychopharmacology (Berl.), 102, 291±294.

Andreassen, O.A., Ferrante, R.J., Beal, M.F. & Jorgensen, H.A. (1998) Oraldyskinesias and striatal lesions in rats after long-term co-treatment withhaloperidol and 3-nitropropionic acid. Neuroscience, 87, 639±648.

Ase, A.R., Amdiss, F., Hebert, C., Huang, N., van Gelder, N.M. & Reader,T.A. (1999) Effects of antipsychotic drugs on dopamine and serotonincontents and metabolites, dopamine and serotonin transporters, andserotonin1A receptors. J. Neural. Transm., 106, 75±105.

Bjorklund, L. & Stromberg, I. (1997) Dopaminergic innervation of striatalgrafts placed into different sites of normal striatum: Differences in thetyrosine hydroxylase immunoreactive growth pattern. Exp. Brain Res., 113,13±23.

Bloomquist, J., King, E., Wright, A., Mytilineou, C., Kimura, K., Castagnoli &K.Castagnoli, N. Jr (1994) 1-Methyl-4-phenylpyridinium-like neurotoxicityof a pyridinium metabolite derived from haloperidol: cell culture andneurotransmitter uptake studies. J. Pharmacol. Exp. Ther., 270, 822±830.

Burt, D.R., Creese, I. & Snyder, S.H. (1977) Antischizophrenic drugs: chronictreatment elevates dopamine receptor binding in brain. Science, 196, 326±328.

Cadet, J.L. (1988) Free radical mechanism in the central nervous system: anoverview. Int. J. Neurosci., 40, 13±18.

Casey, D.E. (1995) Tardive dyskinesya: pathophysiology. In Bloom FE,Kupfer DJ, (eds) Psychopharmacology: The Fourth Generation of Progress,Raven, New York, pp. 1497±1502.

Ciliax, B.J., Heilman, C., Demchyshyn, L.L., Pristupa, Z.B., Ince, E., Hersch,S.M., Niznik, H.B. & Levey, A.I. (1995) The dopamine transporter:immunochemical characterization and localization in brain. J. Neurosci., 15,1714±1723.

Cochiolo, J.A., Ehsanian, R. & Bruck, D.K. (2000) Acute ultrastructuraleffects of MPTP on the nigrostriatal pathway of the C57BL/6 adult mouse:evidence of compensatory plasticity in nigrostriatal neurons. J. Neurosci.Res., 59, 126±135.

CoteÁ, S.L., Ribeiro-da-Silva, A. & Cuello, A.C. (1993) Current protocols forlight microscopy immunocytochemistry. In Cuello AC (ed). Immuno-histochemistry II John Wiley, West Sussex, pp. 147±168.

Cottingham, S.L., Pickar, D., Shimotake, T.K., Montpied, P., Paul, S.M. &Crawley, J.N. (1990) Tyrosine hydroxylase and cholecystochinin mRNAlevels in the substantia nigra, ventral tegmental area, and locus coeruleus areunaffected by acute and chronic haloperidol administration. Cell. Mol.Neurobiol., 10, 41±50.

De Keyser, J. (1991) Excytotoxic mechanisms may be involved in thepathophysiology of tardive dyskinesia. Clin. Neuropharmacol., 4, 562±565.

De Souza, I.E., McBean, G.J. & Meredith, G.E. (1999) Chronic haloperidoltreatment impairs glutamate transport in the rat striatum. Eur. J.Pharmacol., 382, 139±142.

Diana, M., Collu, M., Mura, A. & Gessa, G.L. (1992) Haloperidol-inducedvacuous chewing in rats: suppression by alpha-methyl-tyrosine. Eur. J.Pharmacol., 211, 415±419.

Egan, M.F., Hurd, Y., Ferguson, J., Bachus, S.E., Hamid, E.H. & Hyde, T.M.(1996) Pharmacological and neurochemical differences between acute andtardive vacuous chewing movements induced by haloperidol.Psychopharmacology (Berl), 127, 337±345.

Ellison, G. & See, R.E. (1989) Rats administered chronic neuroleptics developoral movements which are similar in form to those in humans with tardivedyskinesia. Psychopharmacology (Berl), 98, 564±566.

Fang, J., Zuo, D. & Yu, P.H. (1995a) Comparison of cytotoxicity of aquaternary pyridinium metabolite of haloperidol (HP+) with neurotoxin N-methyl-4-phenylpyridinium (MPP+) towards cultured dopaminergicneuroblastoma cells. Psychopharmacology (Berl), 121, 373±378.

Fang, J. & Yu, P.H. (1995b) Effect of haloperidol and its metabolites ondopamine and noradrenaline uptake in rat brain slices. Psychopharmacology(Berl), 121, 379±384.

Fibiger, H.C. & Lloyd, K.G. (1984) Neurobiological substrates of tardivedyskinesia: the GABA hypothesis. Trends Neurosci., 7, 462±464.

Gale, K. (1980) Chronic blockade of dopamine receptors by antischizophrenicdrugs enhances GABA binding in substantia nigra. Nature, 283, 569±570.

Gunne, L.M., Andersson, U., Bondesson, U. & Johansson, P. (1986)Spontaneous chewing movements in rats during acute and chronicantipsychotic drug administration. Pharmacol. Biochem. Behav., 25, 897±901.

Gunne, L.M., Growdon, J. & Glaeser, B. (1982) Oral dyskinesia in rats

following brain lesions and neuroleptic drug administration.Psychopharmacology (Berl), 77, 134±139.

Gunne, L.M., Haggstrom, J.E. & Sjoquist, B. (1984) Association withpersistent neuroleptic-induced dyskinesia of regional changes in brainGABA synthesis. Nature, 309, 347±349.

Huang, N.Y., Kostrzeva, R.M., Li, C., Perry, K.W. & Fuller, R.W. (1997)Persistent spontaneous oral dyskinesias in haloperidol-withdrawn ratsneonatally lesioned with 6-hydroxydopamine: absence of an associationwith the Bmax for [3H] Raclopride binding to neostriatal homogenates. J.Pharmacol. Exp. Ther., 280, 268±276.

Hung, H.C. & Lee, E.H. (1998) MPTP produces differential oxidative stressand antioxidative respiration in the nigrostriatal and mesolimbicdopaminergic pathways. Free Radic. Biol. Med., 24, 76±84.

Ikeda, H., Adachi, K., Hasegawa, M., Sato, M., Hirose, N., Koshikawa, N. &Cools, A.R. (1999) Effects of chronic haloperidol and clozapine on vacuouschewing and dopamine-mediated jaw movements in rats: evaluation of arevised animal model of tardive dyskinesia. J. Neural. Transm., 106, 1205±1216.

Jeste, D.V. & Wyatt, R.J. (1979) In search of a treatment for tardivedyskinesia: review of the letterature. Skizphr. Bull., 5, 251±292.

Jicha, G.A. & Salamone, J.D. (1991) Vacuous jaw movements and feelingde®cits in rats with ventrolateral striatal dopamine depletion: possiblerelation to parkinsonian symptoms. J. Neurosci., 11, 3822±3829.

Kelley, J.J., Gao, X.M., Tamminga, C.A. & Roberts, R.C. (1997) The effect ofchronic haloperidol treatment on dendritic spines in the rat striatum. Exp.Neurol., 146, 471±478.

Kelly, P.H., Seviour, P.W. & Iversen, S.D. (1975) Amphetamine andapomorphine responses in the rat following 6-OHDA lesions of thenucleus accumbens septum and corpus striatum. Brain Res., 94, 507±522.

Klawans, H.L., D'amico, D.J., Nausieda, P.A. & Weiner, W.L. (1977) Thespeci®city of neuroleptic- and methysergide-induced behaviouralhypersensitivity. Psychopharmacology (Berl), 55, 49±52.

Klawans, H.L. & Rubovits, R. (1972) An experimental model of tardivedyskinesia. J. Neural. Transm., 33, 235±246.

Lange, K.W. (1990) Behavioural effects and supersensitivity in the ratfollowing intranigral MPTP and MPP+ administration. Eur. J. Pharmacol.,175, 57±61.

Lau, Y.S. & Fung, Y.K. (1986) Pharmacological effects of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) on striatal dopamine receptor system.Brain Res., 369, 311±315.

Lerner, P., Nose, P., Gordon, E.K. & Lovenberg, W. (1977) Haloperidol:effect of long-term treatment on rat striatal dopamine synthesis andturnover. Science, 197, 181±183.

Lohr, J.B. (1991) Oxygen radicals and neuropsychiatric illness. Somespeculations. Arch. Gen. Psychiatry, 48, 1097±1106.

Meshul, C.K., Andreassen, O.A., Allen, C. & Jorgensen, H.A. (1996)Correlation of vacuous chewing movements with morphological changesin rats following 1-year treatment with haloperidol. Psychopharmacology,125, 238±247.

Meshul, C.K., Janowsky, A., Casey, D.E., Stalbaumer, R.K. & Taylor, B.(1992) Effect of haloperidol and clozapine on the density of `perforated'synapses in caudate, nucleus accumbens, and medial prefrontal cortex.Psychopharmacology (Berl), 106, 45±52.

Meshul, C.K., Stallbaumer, R.K., Taylor, B. & Janowsky, A. (1994)Haloperidol-induced morphological changes in striatum are associatedwith glutamate synapses. Brain Res., 648, 181±195.

Narang, N. & Wamsley, J.K. (1995) Time dependent changes in DA uptakesites, D1 and D2 receptor binding and mRNA after 6-OHDA lesions of themedial forebrain bundle in the rat brain. J. Chem. Neuroanat., 9, 41±53.

Nielsen, E.B. & Lyon, M. (1978) Evidence for cell loss in corpus striatum afterlong-term treatment with a neuroleptic drug (¯upenithixol) in rats.Psychopharmacology (Berl), 59, 85±89.

Nomoto, M., Kaseda, . S., Iwata, S., Shimizu, T., Fukuda, T. & Nakagawa, S.(2000) The metabolic rate and vulnerability of dopaminergic neurons, andadenosine dynamics in the cerebral cortex, nucleus accumbens, caudatenucleus, and putamen of the common marmoset. J. Neurol., 247, 16±22.

Pan, H.S., Penny, J.B. & Young, A.B. (1985) Gamma-aminobutyric acid andbenzodiazepine receptor changes induced by unilateral 6-hydroxydopaminelesions of the medial forebrain bundle. J. Neurochem., 45, 1396±1404.

Paxinos, G. & Watson, C. (1986) The Rat Brain in Stereotaxic Coordinates,2nd edn. Academic Press, New York.

Peacock, L. & Gerlach, J. (1997) A reanalysis of the dopamine theory oftardive dyskinesia the hypothesis of dopamine D1/D2 imbalance. In: YassaR, Nair NPV, Jeste DV, (eds) Neuroleptic-Induced Movements Disorders.,Cambridge University Press, New York, pp. 141±160.



Petzer, J.P., Bergh, J.J., Mienie, L.J., Castagnoli, N. Jr & Van der Schyf, C.J.(2000) Metabolic defects caused by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) and by HPTP (the tetrahydropyridinylanalogue of haloperidol), in rats. Life Sci., 66, 1949±1954.

Rivest, R., Falardeau, P. & Di Paolo, T. (1995) Brain dopamine transporter:gender differences and effect of chronic haloperidol. Brain Res., 692, 269±272.

Sakai, K., Gao, X.M., Hashimoto, T. & Tammiga, C.A. (2001) Traditional andnew antipsychotic drugs differentially alter neurotransmission markers inbasal ganglia-thalamocortical neural pathways. Synapse, 39, 2±16.

Sasaki, K., Kennedy, J.L. & Nobrega, J.N. (1997) Localized changes in GABAreceptor-gated chloride channel in rat brain after long-term haloperidol:relation to vacuous chewing movements. Synapse, 25, 73±79.

See, R.E. & Lynch, A.M. (1995) Chronic haloperidol potentiates stimulatedglutamate release in caudate putamen, but not prefrontal cortex.Neuroreport, 6, 1795±1798.

Sherman, T.G. & Moody, C.A. (1995) Alterations in tyrosine hydroxylaseexpression following partial lesions of the nigrostriatal bundle. Mol. BrainRes., 29, 285±296.

Simantov, R., Blinder, E., Ratovitski, T., Tauber, M., Gabbay, M. & Porat, S.(1996) Dopamine-induced apoptosis in human neuronal cells: inhibition bynucleic acids antisense to the dopamine transporter. Neuroscience, 74, 39±50.

Tamminga, C.A., Dale, J.M., Goodman, L., Kaneda, H. & Kaneda, N. (1990)Neuroleptic-induced vacuous chewing movements as an animal model of

tardive dyskinesia: a study in three rat strains. Psychopharmacology (Berl),102, 474±478.

Tarsy, D. & Baldessarini, R.J. (1974) Behavioural supersensitivity toapomorphine following chronic treatment with drugs which interfere withthe synaptic function of catecholamines. Neuropharmacology, 13,927±940.

Tarsy, D. & Baldessarini, R.J. (1977) The pathophysiological basis of tardivedyskinesia. Biol. Psychiat., 12, 431±450.

Waddington, J.L. (1990) Spontaneous orofacial movements induced in rodentsby very long-term neuroleptic drug administration: phenomenology,pathophysiology and putative relationship to tardive dyskinesia.Psychopharmacology (Berl), 101, 431±447.

Waddington, J.L. (1997) Rodent and other animals models of tardivedyskinesia during long-term neuroleptic drug administration.Controversies and Implications for the Clinical Syndrome. In Yassa R,Nair NPV, Jeste DV (eds) Neuroleptic-Induced Movements Disorder.,Cambridge University Press, New York, pp. 225±237.

Waters, C.M., Peck, R., Rossor, M., Reynolds, G.P. & Hunt, S.P. (1988)Immunocytochemical studies on the basal ganglia and substantia nigra inthe Parkinson's disease and Huntington's chorea. Neuroscience, 25, 419±438.

Wright, A.M., Bempong, J., Kirby, M.L., Barlow, R.L. & Bloomquist, J.R.(1998) Effects of haloperidol metabolites on neurotransmitter uptake andrelease: possible role in neurotoxicity and tardive dyskinesia. Brain Res.,788, 215±222.



sub-chronic treatment with classical but not atypical antipsychotics produces morphological changes...

Documents