Alterations in cerebellar glutamic acid decarboxylase (GAD) activity in a genetic model of torsion dystonia (rat)

EXPERIMENTAL

NEUROLOGY

85,216-222

(1984)

RESEARCH

NOTE

Alterations in Cerebellar Glutamic Acid Decarboxylase (GAD) Activity in a Genetic Model of Torsion Dystonia (Rat) G. A. OLTMANS,

M. BEALES, J. F. LGRDEN,

AND J. H. GORDON’

Department of Pharmacology, University of Health Sciences/The Chicago Medical School, North Chicago, Illinois 60064, and Department of Psychology, University of Alabama in Birmingham, Birmingham, Alabama 35294 Received October Il. 1983; revision received February 6, 1984 Glutamic acid decarboxylase (GAD) activity was studied in specific brain regions of a newly identified genetic (rat) model of human torsion dystonia. GAD activity was found to be significantly increased in the deep cerebellar nuclei of dystonic rats at 16, 20, and 24 days of age. GAD activity in the other regions examined (vermis, cerebellar hemispheres, caudate nucleus, and globus pallidus) did not differ from that of age-matched normal littermate controls. Diazepam treatment significantly reduced the frequency of dystonic movements in the mutant.

A genetically determined rat model for human torsion dystonia has been described recently (genotype dt) (10). The dystonic symptoms appear at about 10 days of age after a period of normal development. The early symptoms include falling to the side, excessive pivoting, and alternating torticollis. The symptoms progress during subsequent days to include advanced signs such as inappropriate limb placement during locomotion, self-clasping of limbs, and hyperflexion of the trunk. Preliminary histological analyses have failed to identify either central or peripheral nervous system pathology in the mutant. In this respect, no difference was found between dystonic and normal animals in neural or nonneural tissue stained with hematoxylin and eosin, or in neural tissue stained with 1~x01 fast blue, periodic acid-Schiff, or cresyl violet (10). In those studies Abbreviations: NE-nompinephrine, DA-dopamine, GAD-glutamic acid decarboxylase, GABA-gamma-aminobutyric acid. ’ Supported in part by National Institutes of Health grant NS 18062. Please send reprint requests to Dr. Oltmans, Dept. of Pharmacology, UHS/CMS, 3333 Green Bay Rd, N. Chicago, IL 60064. 216 0014-4886/84 $3.00 Copyright Q 1984 by Academic Press, Inc. All ri&ts of reproduction in any form reserved.

CEREBELLAR

GAD

ACTIVITY

IN

TORSION

DYSTONIA

217

particular attention was focused on spinal and cerebellar regions, as well as the red nucleus and striatum. These latter two regions were shown to be abnormal in a mutant mouse model of dystonia (13). A Golgi study of the striatum showed similar cell types in the dystonic and normal rats (12). In contrast to the anatomic findings, both neurochemical and psychopharmacologic abnormalities were found in the dystonic mutant. Cerebellar norepinephrine (NE) concentrations were significantly higher in dt rats than in age-matched normal littermates, although cortical and hippocampal NE concentrations did not differ between the two groups ( 10). There also were no apparent differences between dystonic and normal animals in either striatal dopamine (DA) or DA receptor concentrations. The dystonic animal did, however, show a decreased response to the catalepsy-inducing effects of the DA-receptor blocker haloperidol, suggesting other possible abnormalities in the striatum or its efferent pathways (12). The elevated cerebellar NE and the decreased behavioral response to DAreceptor blockade suggests altered neuronal function in two brain regions known to be involved in the control of movement. To further explore the integrity of the cerebellum and basal ganglia in the dt rat, glutamic acid decarboxylase (GAD) activity was measured in specific regions of those structures. Gamma-aminobutyric acid (GABA)-containing neurons are present in both the cerebellum and basal ganglia (l), and GAD activity is believed to reflect neuronal activity in GABA systems (4). Movement disorders resulting from cerebellar or striatal dysfunction might, therefore, be detected as an abnormality in GAD activity. In addition to the analysis of GAD activity, dystonic animals were administered diazepam to determine whether or not this GABA-facilitating agent (5), which was found to alleviate dystonic symp toms in some humans (1 l), would have similar effects in the animal model. The subjects were dystonic rat pups and their age-matched normal littermate controls obtained from the colony maintained at the University of Alabama in Birmingham (Birmingham, Alabama, U.S.A.). They were killed by decapitation at either 16, 20, or 24 days of age and the brains immediately removed, frozen on dry ice, and stored at -70°C until the time of dissection. At these ages the dystonic symptoms are clearly present and increasing in severity. Three cerebellar and two basal ganglia regions were obtained using the following procedure: The frozen brains were removed from the -70°C freezer and gradually warmed to -4°C to -5°C in a cryostat. For dissection of the cerebellum each brain was placed on its ventral surface on an ice-cold metal dissection block. Two angled cuts bordering the cerebellar peduncles were made starting at the dorsal surface of the cerebellum and extending through the cerebellum to the ventral surface of the brain as shown in Fig. 1A. This piece of tissue contained the deep cerebellar nuclei and samples of the vermis and cerebellar hemispheres. These three areas were removed as

218

OLTMANS

ET AL.

FIG. I, A cerebellar section containing the deep cerebellar nuclei and samples of the cerebellar hemispheres and vermis was removed by making two angled cuts as shown by the heavy lines in A. The specifk sections were removed from this section as shown in B. a-Cerebellar hemisphere, b-vermis, c-deep nuclei. Figures are modified from Paxinos and Watson (15).

CEREBELLAR

GAD

ACTIVITY

IN

TORSION

DYSTONIA

219

depicted in Fig. 1B. Basal ganglia sections were obtained by making three coronal cuts in a rostral to caudal direction and removing caudate and globus pallidus tissue using the atlas of K&rig and Klippel(8) as a guide. The specific sections were refrozen on dry ice, weighed, immediately placed in 6 or 12 vol of 100 mM KC1 containing 0.5% Triton X-100 and 3.3 mM EDTA, and sonicated 30 s in tubes bathed in ice water. The samples were then assayed for GAD activity or refrozen and stored at -70°C until the time of assay. Tissue from normal and dystonic animals of the same age was dissected and assayed at the same time. GAD activity for each section was determined as described by Gordon et al. (6). Each assay consisted of the same tissue sample (e.g., 16day-old vermis) from at least seven dystonic and seven normal rats. GAD activity is expressed as nanomoles GABA formed per milligram tissue-hour. The effects of diazepam on the dystonic movements were studied in 18to 20-day-old pups as described elsewhere (10) with minor modifications. In brief, subjects were removed from their cages during the light part of the lightdark cycle, placed in an open field, and dystonic movements were recorded for a 3-min period. This treatment elicits substantial activity and a high incidence of dystonic movements which continues as long as the animal is in the open field. The 3-min observation was found to reliably produce an adequate number of dystonic movements against which drug effects could be compared, yet was brief enough to avoid major problems with temperature regulation and maternal separation in the pups. Diazepam (Roche) was then administered (i.p.) and subjects returned to the home cage. At 15 and 30 min after drug treatment the animals were again placed in the open field and the number of dystonic movements in a 3-min period recorded. Thus, each animal was observed on three occasions, once before and twice after drug treatment. The movements recorded were paw clasps, falls to the side with limb extension, and axial twisting. Three doses of diazepam (O-25,0.37, and 0.50 mg/kg) were administered to three different groups of rats (N = S/group). All values are reported as the M f SD. GAD activity in specific regions of dystonic and normal rats were compared by a t test, Effects of diazepam on the dystonic movements were determined by a paired t test comparing pre- and postdrug frequencies (9). The results of the GAD analyses are presented in Table 1. In the 16-dayold animals, GAD activity in the deep cerebellar nuclei of the dystonic rats was significantly higher (+23%) than that in the normal littermate controls. In the other cerebellar regions examined (vermis, hemispheres) the normal and dystonic animals did not differ significantly. Similarly, no differences were found between 16-day-old dystonic and normal rats in GAD activity in the caudate nucleus or globus pallidus. The results in the older animals parallelled those found in the 16-day-old rats, as GAD activity in the deep

220

OLTMANS

ET

TABLE Glutamic

Mqtation

Age (days)

(~-9

I

Acid Decarboxylase (GAD) Regions of the Genetically GAD Deep cerebellar nuclei

AL.

activity

Activity Dystonic (nmol

Cerebellar hemispheres

in Specific Rat” GABA/mg

Brain

tissue-h)

Caudate Vermis

nUdeUS

Globus pallidus

Dystonic Normal

(8) (8)

16 16

17.0 k 2.9* 13.8 k I.1

9.8 k 2.7 10.2 + 1.0

9.6 k 2.1 9.1 + 1.4

2.4 * 0.7 2.5 f 0.6

7.6 f 0.8 8.1 f 1.1

Dystonic Normal

(8) (8)

20 20

20.3 + 5.0. 15.4 ” 2.2

7.4 k 0.8 8.1 + 0.8

5.9 f 0.4 6.2 * 0.4

8.8 f 1.3 9.5 + 2.2

4.9 f 0.7 5.0 f 0.9

Dystonic Normal

(7) (7)

24 24

24.4 -t 4.6** 18.4 f 2.2

9.2 k 1.0 10.2 + 0.4

9.4 * 1.2 8.8 k 0.8

13.7 * 3.3 10.8 f 6.0

8.0 f 0.6 7.3 f 0.8

@Values are M f SD. * Differs significantly from normal littermate control, P < 0.05. l * Differs significantly from normal littermate control, P < 0.01.

cerebellar nuclei of dystonic rats was significantly increased at both 20 (+32%) and 24 (+33%) days of age. No differences were found in the other sections of the older animals. Prior to drug administration the dystonic animals averaged 45 f 8 dystonic movements per 3-min period. Diazepam treatment produced a dose-dependent decrease in the frequency of these movements. Although the lowest dose of diazepam (0.25 mg/kg) did not have a significant effect at either 15 or 30 min after drug treatment (x = 107 f 25% and 98 * 11% of baseline, respectively), the middle dose (0.35 mg/kg) produced a significant decrease in the frequency of the dystonic movements at both times (x = 39 f 34% of baseline at 15 min; x = 72 + 13% at 30 min; P < 0.05 for both comparisons). The highest dose of diazepam (0.50 mg/kg) also decreased the frequency of the dystonic symptoms at both the 15- and 30-min time points (x = 27 f 28% and 10 f 15% of baseline, respectively, P < 0.05 for both comparisons). Qualitatively, the medium and high doses of diazepam either eliminated the dystonic movements or produced an attenuation in the severity of the movements when they did occur. In this respect: (i) The paw clasping movements were eliminated by both the medium and high drug dose; (ii) When falling occurred, although the limbs were extended, the digits no longer extended as in a nontreated animal; (iii) The strong axial twisting became more of a rocking, side to side oscillation. The righting reflex remained intact, however, indicating that the animals had not been sedated to the point of unconsciousness. The results indicate a localized increase in GAD activity in the deep cerebellar nuclei of the dystonic mutant. Studies of cerebrospinal fluid in humans suffering from dystonia have not, however, revealed any remarkable abnor-

CEREBELLAR

GAD ACTIVITY

IN TORSION

DYSTONIA

221

mality in GABA concentration ( 16). As not all human dystonias are genetically determined, the failure to see a change in GABA may represent mixing of patients with syndromes of differing etiologies. The current results, however, suggest another possibility. A significant difference in GAD activity was found in only one of the several subsystems examined. Analysis of larger brain regions, or a measurement which includes contributions from multiple sources (such as cerebrospinal fluid), could easily miss such a difference. Thus, it may be necessary to more precisely assessGABA function in human dystonia before a disorder in GABA systems is ruled out. In fact, indirect support for a GABA dysfunction can be inferred from the observation that in some cases of human dystonia limited therapeutic success has been seen to follow the administration of benzodiazepines ( 11, 18). Because benzodiazepines are believed to facilitate neurotransmission in GABA systems (5), the therapeutic response may reflect the attenuation of a GABA disorder such as that reported here. The finding that diazepam treatment attenuated the severity of the dystonia in the rats provides some support for this hypothesis. In addition to increased GAD activity, the dystonic rat has significantly elevated cerebellar norepinephrine (10). Cerebellar NE input arises principally from the locus ceruleus and projects to the Purkinje cells (2), where it is believed NE release inhibits Purkinje cell activity (7, 17). In addition to its direct effect, NE enhances the inhibitory effects of GABA on Purkinje cells ( 14, 19). The Purkinje cells in turn send projections believed to be GABAergic to the deep nuclei (3). The increased GAD activity in the deep nuclei may, therefore, reflect a failure of both the direct inhibitory NE input and NEpotentiated GABA inhibition of the Purkinje cells. The results indicate that the increase in GAD activity in the deep nuclei persists for the time span studied (16 to 24 days of age) and suggest this change is permanent. Because the dystonic pups cannot be identified prior to the onset of the symptoms, it is not currently possible to determine if the GAD change precedes, occurs concomitantly with, or follows the onset of the dystonic movements. Because our measurements were made at a time when the dystonia was clearly developed, the changes observed may represent compensatory attempts to attenuate the dystonia. Alternatively, the cerebellar NE-GAD sequence of abnormalities may be causative for the dystonia. Additional work will be required to clarify these possibilities. REFERENCES I. BAXTER, C. F., AND E. ROGERS. 1960. Demonstration of thiosemicarbazide-induced convulsions in rats with elevated brain levels of gamma-aminobutyric acid. Proc. Sot. Exp. Biol. Med. 104: 426-421. 2. BWM, F. E., B. J. HOF’FER,AND G. R. SKKXNS. 197 1. Studies on norepinephrine-containing afferents to Purkinje cells of rat cerebellum: I. Localization of the fibers and their synapses. Brain Rex 25: 501-521.

222

OLTMANS

ET AL.

3. CHAN-PALAY, V. 1982. Gamma-aminobutyric acid pathways in the cerebellum studied by retrograde and anterograde transport of glutamic acid decarboxylase (GAD) antibody after in vivo injections. Prog. Brain Res. 55: 51-76. 4. COLLINS, G. G. S. 1972. GABA- oxoghrtarate transaminase, glutamate decarboxylase and the half-life of GABA in different areas of rat brain. Biochem. Pharmacol. 21: 2849-2858. 5. COSTA, E., AND A. GUIDOIX. 1979. Molecular mechanisms in the receptor action of benzodiazepines. Annu. Rev. Pharrnacol. Toxicol. 19: 531-545. 6. GORDON, J. H., D. M. NANCE, J. E. SHRYNE, AND R. A. GORSKI. 1977. Effects of estrogen on dopamine turnover, glutamic acid decarboxylase, and lordosis behavior. Brain Res. Bull. 2: 431-436. 7. HOFFER,B. J., G. R. SIGGINS,AND F. E. BLOOM. I97 1. Studies on norepinephrine containing afIerentsto Purkinje cells of rat cerebelhrm: II. Sensitivity of Purkinje cells to norepinephrine and related substances administered by microiontophoresis. Brain Res. 25: 523-534. 8. K~NIG, J. F. R., AND R. A. KLIPPEL. 1963. The Rat Brain: a Stereotaxic Atlas of the Forebrain and Lower Parts of the Brain Stem. Williams & Wilkins, Baltimore. 9. KURTZ, K. H. 1965. Foundations of Psychological Research. Allyn and Bacon, Boston. 10. LORDEN, J. F., T. W. MCKEON, H. J. BAKER, N. Cox, AND S. U. WALKLEY. 1984. Characterization of the rat mutant dystonic (dt): a new animal model of dystonia musculorum deformans. J. Neurosci., in press. 11. MARSDEN, C. D. 1981. Treatment of torsion dystonia. Pages 81-104 in A. BARBEAU, Ed., Current Status of Modem Therapy Vol. 8: Disorders of Movement. Lippincott, Philadelphia/ Toronto. 12. MCKEON, T. W., J. F. LQRDEN, G. A. OLTMANS, M. BEALES, AND S. WALKLEY. 1984. Decreased catalepsy response to haloperidol in the genetically dystonic (dt) rat. Brain Res.. in press. 13. MESSER, A., AM) N. L. STROMINGER. 1980. An allele of the mouse mutant dystonia musculorum exhibits lesions in red nucleus and striatum. Neuroscience 5: 543-549. 14. MOISES, H. C., D. J. WOODWARD, B. J. HOFFER, AND R. FREEDMAN. 1979. Interactions of norepinephrine with Purkinje cell responses to putative amino acid neurotransmitters applied by microiontophoresis. Exp. Neural. 64: 493-5 15. 15. PAXINOS, G., AND C. WATSON. 1982. The Rat Brain in Stereotaxic Coordinates. Academic Press, New York/London. 16. PERRY,T. L., S. HANSEN, N. QUINN, AND C. D. MARSDEN. 1982. Concentrations ofGABA and other amino acids in CSF from torsion dystonia patients. J. Neurochem. 39: 11881191. 17. SIGGINS,G. R., B. J. HOFFER,AND F. E. BLOOM. 197 1. Studies on norepinephrine-containing afIerents to Purkinje cells of rat cerebellum: III. Evidence for mediation of norepinephrine effectsby cyclic 3’,S-adenosine monophosphate. Brain Res. 25: 535-553. 18. WACHTEL, R. C., M. L. BATSHAW, R. ELDRIDGE, W. JANKEL, AND M. CATALDO. 1982. Torsion dystonia. Johns Hopkins Med. J. 151: 355-361. 19. YEH, H. H., H. C. MOISES, B. D. WATERHOUSE, AND D. J. WOODWARD. 1981. Modulatory interactions between norepinephrine and taurine, beta-alanine, gamma-amino butyric acid and muscimol, applied iontophoretically to cerebehar Purkinje cells. Neuropharmacology

20: 549-560.