EXPERIMENTALNEUROLOGY 118,1&23(1%2)
Magnetic Resonance Imaging to Monitor Pathology of Caudate-Putamen after Excitotoxin-Induced Neuronal Loss in the Nonhuman Primate Brain P.HANTRAYE,*
A.LEROY-WILLIG,A.
DENYS, D. RICHE,?
O.ISACSON,$
M. MAZIERE,*
AND A. SYROTA
Service F&&c Joliot, CEA, 4 place du G&&-al Leclerc, 91406 Orsay, France; *URA CNRS 1285, 4 place du G&&al Leckrc, 91406 Orsay, France; TLaboratoire de Physiologic nerveuse, Equipe de Neuroanatomie fonctionnelle, CNRS, 91198 GifiYvette, France; and $Neuroregeneration Laboratory, Harvard Medical School, McLean Hospital, MRC 119, Belmont, MA 02178
of the typical abnormal movements observed in the human disease such as orofacial dyskinesias, dyskinesias of extremities, chorea, and dystonia can be observed in this primate model (7, 12). As these symptoms essentially develop during the first 2-4 weeks following the injection of the excitotoxin (6, 7) and then remain unmodified for more than 1; years we monitored the progression of the striatal lesions during the critical first 4 weeks using the magnetic resonance imaging (MRI) technique. In an attempt to correlate indices of neuronal cell loss in the striatum with behavioral and biochemical alterations, these MRI studies were conducted on three IA-lesioned baboons in parallel with positron emission tomography (PET) investigation of the Da-receptors and dopamine metabolism (8) and behavioral studies aimed at detecting striatal functional impairments (6).
We used MR imaging to locate and monitor in viva the pathological events taking place 2 to 4 weeks after unilateral striatal injections of ibotenic acid (IA) in the Pupio papio baboon. As early as 2 weeks after IA injections, excitotoxic lesions in the caudate and the putamen were directly visualized on Tl-weighted images as small areas of low signal intensity. On T2-weighted images, the lesion sites were visualized as areas of highintensity signal, spreading over larger areas than the corresponding regions in T l-weighted images. These alterations of T2-values in the lesioned striatum persisted 4 weeks after surgery. However, as the striatal degeneration progressed from 2 to 4 weeks after lesion, the size of the areas of high signal intensity on T2weighted images decreased, whereas the same regions appeared essentially unmodified on Tl-weighted images. A marked enlargement of the ipsilateral lateral ventricle (a characteristic of excitotoxic striatal lesions) could be detected 4 weeks after surgery, on both axial Tl- and T2-weighted images. Comparisons of MR images with postmortem anatomical data indicated that areas of increased T 1 values corresponded to regions of severe neuronal depletion (a direct result of the excitotoxic lesion), whereas areas of increased T2 values corresponded to regions of increased content in astrocytes and ferritin and probably in the early period following lesion (2 weeks) to a superimposed edema. o 1982 Academic
Press,
METHODS Excitotoxic Striatal Lesion Three baboons (Papio papio, one male and two females) were studied (body weight, lo-20 kg at the beginning of the study). They received a total of 700 pg (baboons 1 and 2) or 350 pg (baboon 3) of IA (concentration, 10 PgIpl), distributed at seven injection sites in two different sessionsseparated by 3 days. Three injections (10 ~1 each) were made into the dorsolateral part of the putamen (session 1) and four injections in the dorsomedial part of the caudate nucleus (session 2) of the right hemisphere using an experimental protocol previously described (6). The stereotaxic coordinates for the striatal injections were determined according to the stereotaxic atlas of the baboon brain by Riche et al. (21). MR examinations were performed 2 weeks (baboons 1 and 2) and 4 weeks (baboons 1,2,3) after the second surgical session.
Inc.
INTRODUCTION Rat models of Huntington’s disease (HD) mimicking many neurochemical and neuropathological aspects of this neurological disorder can be obtained by intrastriatal injections of “excitotoxic” glutamate receptor agonists (3, 5, 10, 11, 15). Recently, we have developed a similar model of HD in the primate using ibotenate (IA) or quinolinate, as excitotoxins. This model of excitotoxic striatal lesions replicates the striatal degeneration and astrocytosis observed in HD patients (6, 7, 12). In addition, after dopamine agonist administration, many 0014-4686/92 Copyright AI1 rights
$5.00 0 1992 by Academic Press, of reproduction in any form
Apparatus and Experimental Protocol MR examination was done with a 0.5 T MR Max (G.E.) system. We used a homemade Helmholtz re18
Inc. reserved.
MR
IMAGING
OF
STRIATAL
EXCITOTOXIC
FIG. 1. Sagittal Tl-weighted image of the lesioned brain of baboon 1 at 4 weeks after lesion. This plane allows the localization of the lesion (small arrows) 25 mm anterior to the ear bar plane. The frontal hyposignal surrounded by an hypersignal (arrowhead) is probably a small traumatic lesion due to the passage of the needle used to perform the injections (20-gauge needle).
ceive-only probe designed to facilitate the positioning of the head and to provide a high sensitivity. In our experiments, the probe was located very close to the brain to achieve a higher sensitivity. This particular arrangement results in an higher signal intensity at the periphery compared to more central regions of the brain (see Fig. 1). To ensure stereotaxic and reproducible head positioning between weekly scanning sessions, a nonmagnetic head holder equipped with ear bars was constructed. The anesthetized animal (im, ketamine-xylazine, 15-1.5 mg/kg) (2), equipped with the head holder, was positioned ventrally into the MR apparatus. Symmetrical placement of the head holder into the magnet was obtained by using the geometrical center of the ear bars as reference and superimposing these marks with a standard light beam figuring the axial reference plane of the machine. The same stereotaxic plane of reference used to perform the lesions (frontal plane 0, stereotaxic atlas of Riche et al. (21)) was chosen as the reference plane for the MR examinations. MR-Imaging
Protocol
Tl and T2 values of white and grey matter were first measured in one control baboon, which provided values very similar to those published at 0.5 T for the human brain (14,17). Therefore, we used the same sequences as in human brain exploration (in particular, sequences with very long TR and TE values recommended for visu-
LESIONS
IN
BABOON
19
alization of cerebrospinal fluid and edema in T2weighted images) for these nonhuman primate studies. To control the accuracy of the positioning into the magnet, 10 sagittal Tl-weighted sections (TR = 360 ma, TE = 12 ma, field of view = 20 cm, 5 mm thickness, two excitations) of the brain were first performed. These images allowed us to localize the frontal planes containing the ear bars (frontal plane 0) and the needle tracks (lesion plane) (Fig. 1). Then, a set of 16 Tl-weighted images were performed in the frontal plane using a gradient echo sequence (TR = 360-460 ma, TE = lo-12 ma, flip angle = 90°, four excitations). To optimize image quality, these slices were obtained in two “interleaved” acquisitions, yielding contiguous 3-mm thick slices, with an in-plane resolution of 600-780 pm. In one case, supplementary slices interspaced by 1 mm were acquired to check the position of the lesions plane. To obtain further information about the lesions, a heavily TB-weighted spin echo sequence was performed in a single frontal plane corresponding to the center of the lesion, using pulse gating and a TR value between 2300 and 2600 ma, TE = 25 and 120 ma, one excitation. In some experiments, a Tl-weighted image performed at the level of the lesion was obtained following 0.15 mmol/kg gadolinium (Gd-DOTA, Guerbet, France) intravenous injection, to explore possible blood-brainbarrier alterations following excitotoxic lesion. On average, each MR experiment lasted less than 1 h. Measurements Ventricle areas were measured on Tl-weighted images (four frontal planes: the plane containing the lesions sites, one rostral, and two caudal planes), by selecting the pixel values with a signal lower than the average brain signal. The ratio between the right to the left volume was taken as an index of the ventricular enlargement and therefore of the shrinkage of the lesioned caudate-putamen complex. Anatomy In order to analyze the modifications of MR signals seen in the lesioned areas, histological and immunocytochemical studies were performed in one of the baboons (baboon 1) 2 months after the last MRI examination. Even if the excitotoxic striatal lesion cannot be considered as static, at 4 weeks after excitotoxin infusion, the acute phase of striatal atrophy is over and there is no problem in comparing the pathology obtained 2 months after the last MR image. Under deep anesthesia, the animal received an intracardiac perfusion of physiological saline with heparin followed by fixation with a solution of 1% paraformaldehyde and 1% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4). The brain was removed from the skull and postfixed overnight in the same solution. Coronal sections at the level of the stria-
20
HANTRAYE
ET
AL.
FIG. 2. Frontal Tl-weighted images of the lesions plane obtained 2 and 4 weeks after lesion on baboon 1. (A) On first examination, 2 weeks after IA injections, lesions were clearly visible in the right putamen and in the right caudate nucleus (appearing on the left of the image) as areas of low signal intensity (arrows). (B) Two weeks later, the right ventricle (white cross, appearing on the left side of the image) is markedly enlarged and the putaminal lesion is surrounded by an hypersignal of unknown origin.
turn were obtained. Frozen sections were cut from the most anterior part of the lesioned striatum containing the most frontal part of the lesion and processed for cresyl violet, acetyl choline esterase reaction (AChE), glial fibrillary acidic protein immunocytochemistry (GFAP, Dakopat), and ferritin immunoreactivity (Realef/biodesign). In addition, some selected sections were processed for Gomori reaction (9) to visualize the presence of iron in the lesioned and nonlesioned striatum. RESULTS
Anatomical
Localization
of Striatal
Lesions
When comparing MR images obtained on successive sessionsin baboons 1 (Figs. 2A and 2B) and 2, the positioning procedure provided reproducible localization of the head into the magnet (scan to scan variation cl mm). This protocol allowed localization of the lesion sites in the striatum and the monitoring of the alterations in the MR signals with time. The lesion sites were observed as small (2 mm) tetrahedral or triangular structures (explored in 3-mm thick slices), located approximately within 1 mm from the aimed stereotaxic coordinates (Fig. 1).
Tl- Weighted Images On Tl-weighted images, lesion sites appeared as small regions (2 mm diameter) of low signal intensity, observable in the caudate and putamen in a frontal plane located 26 mm anterior to the ear bars. On sagittal sections, we could visualize the needle tracks going
through the overlying cerebral cortex to the caudateputamen (Fig. 1). Two weeks after lesion, no obvious enlargement of the ipsilateral ventricle was noted in the two baboons studied (Fig. 2A and Table 1, baboons 1 and 2). At 4 weeks after lesion, a significant enlargement of the ipsilateral ventricle was noticeable on the two baboons receiving the largest dose of IA (Fig. 2B and Table l), extending from frontal plane + 23 mm to frontal plane + 32 mm, whereas no obvious enlargement was noted in the baboon receiving 350 pg IA. In comparing this enlargement of the ventricle to the progressive atrophy of the ipsilateral caudate nucleus, we observed a significant atrophy of the injected caudate nucleus at TABLE
1
Time Course of the Lateral Ventricular Enlargement Unilateral Excitotoxic Striatal Lesions in the Baboon Following
Baboon 1 (700 /.a) Weeks Mean
after R/L
lesion ratio
Baboon 2 (700 Pd
Baboon 3 (350 rd
2
4
2
4
4
1.5
2.2
1.0
1.5
1.1
Note. Results are expressed as the ratio between the volumes of the lateral ventricle in the lesioned hemisphere (R) and in the contralatera1 hemisphere (L). Volumes of the lateral ventricles were estimated by measuring their corresponding areas on four adjacent planes (the lesion plane, one anterior, and two posterior; thickness, 3 mm). Areas corresponding to lateral ventricles were summated and the ratios between right (lesioned hemisphere) and left (control side) ventricle areas (R/L ratio) computed.
MR
FIG. 3. and caudate Four weeks
IMAGING
OF
STRIATAL
EXCITOTOXIC
LESIONS
IN
BABOON
21
TB-weighted image of the lesion plane in baboon 2 (TR = 2500 ma, TE = 120 ms). (A) Two weeks after lesion, the right putamen nucleus excitotoxic lesions are seen as heterogeneous hypersignal areas. The right lateral ventricle (star) is not yet dilated. (B) after lesion, the right lateral ventricle (star) is slightly enlarged.
the earliest examination, whereas ipsilateral ventricular enlargements were only noted on later examinations (4 weeks after lesion). In one case (baboon l), a hypersignal surrounding the lesion area was observed 4 weeks after IA injections (Fig. 2B). Except for a hypersignal in the frontal cortex corresponding to the needle track in one case (Fig. 1, baboon l), no other regions of noticeable abnormal signal intensity could be visualized. Four weeks after IA injections, in the experiment with gadolinium intravenous injection there was no apparent penetration of the Gd-DOTA into the brain at the level of the excitotoxic lesion (data not shown). T2- Weighted Images On T2-weighted images, large areas of high signal intensity surrounding the small regions of decreased Tl values were observed in the lesioned striatum of all animals at 2 (Fig. 3A) and 4 weeks after lesion (Fig. 3B). However, the area of those regions of T2 hyperintensity decreased between the 2- and 4-week examinations. Analysis of the cerebral ventricles (clearly visible in T2 images as regions of high signal intensity), confirmed the presence of an enlargement of the ipsilateral ventricle 4 weeks after lesion in the animals receiving the largest dose of IA (baboons 1 and 2), whereas no obvious ventricular enlargement could be observed in the baboon receiving the lowest 350 pg IA dose (baboon 3). Anatomical Results Two months after the last MRI study (i.e., 4 months after the excitotoxin injections), the presence of an ex-
citotoxic striatal lesion in baboon 1 was apparent by neuronal loss in Nissl stain and decreased AChE histochemical reaction in the injected caudate nucleus and putamen (Fig. 4a). In agreement with previous observations in IA-lesioned baboons (6), there was also a marked increase in GFAP immunoreactivity in the lesioned area (not shown). On an adjacent brain section immunostained to reveal ferritine, a marked and homogeneous increase in ferritine immunoreactivity was observed in the entire lesioned striatum (Fig. 4b), matching the GFAP-positive region. In contrast, discrete iron deposits in the lesioned striatum at the level of the injection sites were observed using the technique of Gomori (designed for iron or iron-derivative deposits staining). DISCUSSION
Previous studies conducted in rodents using similar lesion models have demonstrated the feasibility of studying in viuo excitotoxic lesions using MRI (18-20, 22, 24). In general, the intrastriatal injection of excitotoxin results in alterations (increases) of the signal intensity in T2-weighted images, in the injected areas. However, observations in rodents are greatly limited by the size of the rat brain which renders it difficult to discriminate between different brain structures (striaturn, thalamus, cerebral ventricle) in both Tl- or T2weighted images. Despite a relatively low magnetic field (0.5 T), the use of nonhuman primates allowed a better discrimination between different structures of the basal ganglia and localization of the lesion sites and cerebral ventricles within the brain. As our purpose was to follow
22
HANTRAYE
FIG. 4. (a) Section of the primate brain at the level of the caudate-putamen (CP) complex stained for acetylcholinesterase (AChE) reaction. The normal side on the left shows an intense reaction in the caudate nucleus (NC) and in the putamen (P). The injected side, on the right, shows a severe decrease in staining accompanied by a severe shrinkage of the CP complex. (b) Adjacent section immunostained with an antibody raised against ferritin, a potential marker for reactive microglia and macrophages. The area of increased reaction observed in the injected CP corresponds to a region devoid of AChE activity and of decreased T2-signal. Scale bar, 5 mm.
alterations of signal in striatal excitotoxic lesions, it was of crucial importance to minimize scan-to-scan variation in the repositioning of the monkey into the gantry. A MRI-guided stereotaxic technique has recently been described to allow accurate placement of microdialysis probes in the striatum in nonhuman primates (23). In the present study, we used a similar approach based on the use of a MR compatible “stereotaxic-like” head holder. One important feature of our protocol is that we rely on the reference plane defined by the surgical stereotaxic head holder to locate the head of the baboon into the magnet. This plane corresponds to the frontal plane 0 of the stereotaxic atlas of the baboon’s brain by Riche et al. (21). The same positioning device was used to perform the positron emission tomography studies (8), providing anatomical support for the interpretation
ET
AL.
of the PET scans. As assessedby comparison of MR images obtained at different time points after lesion, the repositioning in the magnet of each particular baboon was highly reproducible (~1 mm variation in the anteroposterior coordinates). From a biological point of view, the MR technique provides a simple and nontraumatic assessment of the tissue loss and modifications of signals following excitotoxin-induced striatal lesions. Previous studies in the rat have shown that excitotoxic lesions are normally associated with progressive neuronal loss, astrocytosis, microgliosis, edema, and increased T2-signals in the lesioned areas (16, 18-20, 22, 24). The comparisons between MR images and anatomical sections in the present study indicate that the increased T2 values observed in the lesioned primate striatum corresponds to astrocytosis (probably with a superimposed edema at 2 weeks after lesion). The hyposignals observed on Tl-weighted images are associated with regions of severe neuronal depletion (core of the lesion, as defined by Beal et al. (4)). Between 2 and 4 weeks after lesion, the area of increased T2 values decreases in size, probably as a result of regression of the edema. The histological observations obtained in baboon 1,4 weeks after the last MR examination, indicate a severe neuronal loss in the injected striatum on Nissi-stained sections. A marked astrocytosis was also observed in the lesioned caudate and putamen using GFAP immunocytochemistry as well as a marked increase in ferritin content. The neuronal cell loss as well as the glial response are typical features of excitotoxic lesions. On the other hand, the increase in ferritin content merits some comment, as the presence of ferritin in a form complexed to iron is believed to decrease T2 signal in the striatum of HD patients (1). Ferritin plays a central role in recycling iron for synthesis of heme and other proteins and acts as an iron-binding protein in the central nervous system. However, in the primate model, most of the ferritin present in the injected caudate-putamen was not associated with iron 4 months after lesion. The absence of iron in the lesioned CP could explain the discrepancy between our increased T2 values and the decreased T2 values observed in the HD striatum. However, although excitotoxic lesions have been used to model HD, it is likely that these acute injections are not able to mimic all the features of a slowly progressive degenerative illness. In this respect, it would be interesting to study the effect of a more chronic excitotoxic lesion on MR signals and to determine if with time, iron can be stored in the ferritin accumulated in the lesioned striatum, therefore lowering the T2 signal in this area. The significance of the increase in ferritin immunoreactivity following excitotoxic lesion requires further research. Recent findings suggest that ferritin immunoreactivity may be a marker for microglia and macrophages (13). Activated microglia/macrophagic cells, together
MR IMAGING
OF STRIATAL
EXCITOTOXIC
with reactive astrocytes, are the predominant glial cell types migrating into the lesioned area as a consequence of the inflamatory response (16). One hypothesis to explain the increased ferritin content in the insulted striatal regions is through the proliferation of ferritin-containing cells such as the microglial/macrophage cells associated with inflammatory response such as the one induced by acute excitotoxic lesions. REFERENCES 1. ACHNER, F.TH., S. FEUZER, W. POEWE, G. R&mMAyR, AND F. GERSTENBRAND. 1990. Magnetic resonance imaging of extrapyramidal diseases. In Aging Brain and Dementia: New Trends in Diagnosis and Therapy. pp. 483-495. Lisa, New York. 2. BANKNIEDER, A. R., J. M. PHILLIPS, K. T. JACKSON, AND S. I. VINAL. 1978. Comparison of ketamine with the combination of ketamine and xylazine for effective anesthesia in the rhesus monkey (Macaca mu&a). Lab. Anim. Sci. 28: 742-745. 3. BEAL, M. F., N. W. KOWALL, D. W. ELLISON, M. F. MAZUREK, K. J. SWARTZ, AND J. B. MARTIN. 1986. Replication of the neurochemical characteristics of Huntington’s disease by quinolinic acid. Nature 321: 168-171. 4. BEAL, M. F., N. W. KOWALL, K. J. Swmm, R. J. FERRANTE, AND J. B. bfARTIN. 1989. Differential sparing of somatostatin-neuropeptide Y and cholinergic neurons following striatal excitotoxin lesions. Synapse 31: 38-47. 5. COYLE, J. T., AND R. SCHWARCZ. 1976. Lesion of striatal neurons with kainic acid provides a model for Huntington’s chorea. Nature 263: 244-246. 6. HANTRAYE, P., D. RICHE, M. MAZIERE, B. MAZIERE, C. LOC’H, AND 0. ISACSON. 1989. Anatomical, behavioral and positron emission tomography studies of unilateral excitotoxic lesions of the baboon caudate-putamen as a primate model of Huntington’s disease. In Neural Mechanisms in Disorders of Movement (A. R. Crossman and M. A. Sambrook, Eds.), pp. 183-193. Libbey, London. 7. HANTFUYJZ, P., D. RICHE, M. MAZIERE, AND 0. ISACSON. 1990. A primate model of Huntington’s disease: Behavioral and anatomical studies of unilateral excitotoxic lesions of the caudate-putamen in the baboon. Exp. Neurol. 108: 91-105. 8. HANTRAYE, P., C. LOC’H, M. KHALILI-VARASTEH, C. CROUZEL, D. FOURNIER, J-C. Yom, 0. STULZAFT, D. RICHE, 0. ISACSON, AND M. tinzmz. 1992. 6(18F)-Fluoro-L-Dopa uptake and (76Br)bromolisuride binding in the excitotoxically lesioned caudate-putamen of non-human primates studied using positron emission tomography. Exp. Neural. 115: 218-227. 9. HUMASON, G. L. 1962. Animal Tissue Techniques, pp l-233. Freeman, San Francisco and London. 10. ISACSON, O., S. B. DUNNETT, AND A. BJORKLUND. 1985. Graftinduced behavioral recovery in an animal model of Huntington’s disease. Proc. Natl. Acad. Sci. USA 83: 2728-2732.
LESIONS
IN BABOON
23
11. ISACSON, O., W. FISCHER, K. WICTOFUN, D. DAWBARN, AND A. BJORLIJND. 1987. Astroglial response in the excitotoxically lesioned neostriatum and its projection areas in the rat. Neuroscience 20: 1043-1056. 12. ISACSON, O.,P. HANTRAYE, M. MAZIERE, M. V. SOFRONIEW,AM) D. RICHE. 1990. Apomorphine-induced dyskinesias after excitotoxic caudate-putamen lesions and the effects of neural transplantation in non-human primates. Prog. Brain Res. 82: 523533. 13. KANEKO, Y., T. KITAMOTO, J. TATEISHI, AND K. YAMAGUCHI. 1989. Ferritin immunohistochemistry as a marker for microglia. Acta Neuropathal. 79: 129-136. 14. KOENIG, S. H., D. G. BROWN, III, M. SPILLER, AND N. LUNDBOM. 1990. Relaxometry of Brain: Why white matter appears bright in MRI. Magn. Reson. Med. 14: 482-495. 15. MCGEER, E. G., AND P. L. MCGEER. 1976. Duplication of biochemical changes of Huntington’s chorea by intrastriatal injections of glutamic and kainic acids. Nature 263: 517-519. 16. IMARTY, S., I. DUSART, AM) M. PESCHANSKI. 1991. Glial changes following an excitotoxic lesion in the CNS-I. Microglia/Macrophages. Neuroscience 45: 529-539. 17. MILLER, D. H., G. JOHNSON, P. S. TOFTS, D. IMAcms, AND W. I. MCDONALD. 1989. Precise relaxation times measurements of normal-appearing white matter in inflammatory central nervous system disease. Magn. Reson. Med. 11: 331-336. 18. NORMAN, A. B., S. R. THOMAS, R. J. PRAY, R. C. SAMARATUNGA, AND P. R. S-ERG. 1990. Tl and T2 weighted magnetic resonance imaging of excitotoxic lesions and neural transplants in rat brain in vivo. Exp. Neurol. 109: X4-170. 19. NORMAN,A. B., S. R. THOMAS, R. J. PRAY, R. C. SAMARATUNGA, AND P. R. SANBERG. 1989. Magnetic resonance imaging of the rat brain following kainic lesions and fetal striatal tissue transplant. Brain Rex 483: 188-191. 20. PESCHANSKI, M., M. RUDIN, 0. ISACSON, M. DELEPIERRE, AND B. P. ROCQVES. 1988. Magnetic resonance imaging of intracerebral neural grafts. In Transplantation into the Mammalian CNS: Progress in Brain Research (D. M. Gash and J. R. Sladek, Jr., Eds.), Vol. 78, pp. 619-624. Elsevier, Amsterdam. 21. RICHE, D., A. CHRISTOLOMME, J. BERT, AND R. NAQIJET. 1968. Atlas stereotaxique du cerveau de babouin (Papio papio). Edition du CNRS, Paris. pp 7-207. 22. VAN BRUGGEN, N., M. DORAN, R. G. AHIER, M. D. KING, A. L. Busa, D. J. BROOKS, AND S. R. WILLIAMS. 1990. Imaging of kainate lesions in the rat brain. Proc. Meet. Sot. Magn. Reson. Med. 9th 2: 634. 23. WANG, J., S. SKIRBOLL, T. G. AIGNER, R. C. SAUNDERS, J. HSIAO, AND K. S. BANKIEWICZ. 1990. Methodology of microdialysis of neostriatum in hemiparkinsonian nonhuman primates. Exp. Neurol. 110: 181-186. 24. WANG, P. C., C. WAMBAFIE, A. GUIDOTTI, AND A. MURAKI. 1988. High resolution, high field (4,7 T) MRI of Huntington’s disease in an experimental rat model. Proc. Meet. Sot. Magn. Reson. Med. 7th 1: 486.
Magnetic Resonance Imaging to Monitor Pathology of Caudate-Putamen after Excitotoxin-Induced Neuronal Loss in the Nonhuman Primate Brain P.HANTRAYE,*
A.LEROY-WILLIG,A.
DENYS, D. RICHE,?
O.ISACSON,$
M. MAZIERE,*
AND A. SYROTA
Service F&&c Joliot, CEA, 4 place du G&&-al Leclerc, 91406 Orsay, France; *URA CNRS 1285, 4 place du G&&al Leckrc, 91406 Orsay, France; TLaboratoire de Physiologic nerveuse, Equipe de Neuroanatomie fonctionnelle, CNRS, 91198 GifiYvette, France; and $Neuroregeneration Laboratory, Harvard Medical School, McLean Hospital, MRC 119, Belmont, MA 02178
of the typical abnormal movements observed in the human disease such as orofacial dyskinesias, dyskinesias of extremities, chorea, and dystonia can be observed in this primate model (7, 12). As these symptoms essentially develop during the first 2-4 weeks following the injection of the excitotoxin (6, 7) and then remain unmodified for more than 1; years we monitored the progression of the striatal lesions during the critical first 4 weeks using the magnetic resonance imaging (MRI) technique. In an attempt to correlate indices of neuronal cell loss in the striatum with behavioral and biochemical alterations, these MRI studies were conducted on three IA-lesioned baboons in parallel with positron emission tomography (PET) investigation of the Da-receptors and dopamine metabolism (8) and behavioral studies aimed at detecting striatal functional impairments (6).
We used MR imaging to locate and monitor in viva the pathological events taking place 2 to 4 weeks after unilateral striatal injections of ibotenic acid (IA) in the Pupio papio baboon. As early as 2 weeks after IA injections, excitotoxic lesions in the caudate and the putamen were directly visualized on Tl-weighted images as small areas of low signal intensity. On T2-weighted images, the lesion sites were visualized as areas of highintensity signal, spreading over larger areas than the corresponding regions in T l-weighted images. These alterations of T2-values in the lesioned striatum persisted 4 weeks after surgery. However, as the striatal degeneration progressed from 2 to 4 weeks after lesion, the size of the areas of high signal intensity on T2weighted images decreased, whereas the same regions appeared essentially unmodified on Tl-weighted images. A marked enlargement of the ipsilateral lateral ventricle (a characteristic of excitotoxic striatal lesions) could be detected 4 weeks after surgery, on both axial Tl- and T2-weighted images. Comparisons of MR images with postmortem anatomical data indicated that areas of increased T 1 values corresponded to regions of severe neuronal depletion (a direct result of the excitotoxic lesion), whereas areas of increased T2 values corresponded to regions of increased content in astrocytes and ferritin and probably in the early period following lesion (2 weeks) to a superimposed edema. o 1982 Academic
Press,
METHODS Excitotoxic Striatal Lesion Three baboons (Papio papio, one male and two females) were studied (body weight, lo-20 kg at the beginning of the study). They received a total of 700 pg (baboons 1 and 2) or 350 pg (baboon 3) of IA (concentration, 10 PgIpl), distributed at seven injection sites in two different sessionsseparated by 3 days. Three injections (10 ~1 each) were made into the dorsolateral part of the putamen (session 1) and four injections in the dorsomedial part of the caudate nucleus (session 2) of the right hemisphere using an experimental protocol previously described (6). The stereotaxic coordinates for the striatal injections were determined according to the stereotaxic atlas of the baboon brain by Riche et al. (21). MR examinations were performed 2 weeks (baboons 1 and 2) and 4 weeks (baboons 1,2,3) after the second surgical session.
Inc.
INTRODUCTION Rat models of Huntington’s disease (HD) mimicking many neurochemical and neuropathological aspects of this neurological disorder can be obtained by intrastriatal injections of “excitotoxic” glutamate receptor agonists (3, 5, 10, 11, 15). Recently, we have developed a similar model of HD in the primate using ibotenate (IA) or quinolinate, as excitotoxins. This model of excitotoxic striatal lesions replicates the striatal degeneration and astrocytosis observed in HD patients (6, 7, 12). In addition, after dopamine agonist administration, many 0014-4686/92 Copyright AI1 rights
$5.00 0 1992 by Academic Press, of reproduction in any form
Apparatus and Experimental Protocol MR examination was done with a 0.5 T MR Max (G.E.) system. We used a homemade Helmholtz re18
Inc. reserved.
MR
IMAGING
OF
STRIATAL
EXCITOTOXIC
FIG. 1. Sagittal Tl-weighted image of the lesioned brain of baboon 1 at 4 weeks after lesion. This plane allows the localization of the lesion (small arrows) 25 mm anterior to the ear bar plane. The frontal hyposignal surrounded by an hypersignal (arrowhead) is probably a small traumatic lesion due to the passage of the needle used to perform the injections (20-gauge needle).
ceive-only probe designed to facilitate the positioning of the head and to provide a high sensitivity. In our experiments, the probe was located very close to the brain to achieve a higher sensitivity. This particular arrangement results in an higher signal intensity at the periphery compared to more central regions of the brain (see Fig. 1). To ensure stereotaxic and reproducible head positioning between weekly scanning sessions, a nonmagnetic head holder equipped with ear bars was constructed. The anesthetized animal (im, ketamine-xylazine, 15-1.5 mg/kg) (2), equipped with the head holder, was positioned ventrally into the MR apparatus. Symmetrical placement of the head holder into the magnet was obtained by using the geometrical center of the ear bars as reference and superimposing these marks with a standard light beam figuring the axial reference plane of the machine. The same stereotaxic plane of reference used to perform the lesions (frontal plane 0, stereotaxic atlas of Riche et al. (21)) was chosen as the reference plane for the MR examinations. MR-Imaging
Protocol
Tl and T2 values of white and grey matter were first measured in one control baboon, which provided values very similar to those published at 0.5 T for the human brain (14,17). Therefore, we used the same sequences as in human brain exploration (in particular, sequences with very long TR and TE values recommended for visu-
LESIONS
IN
BABOON
19
alization of cerebrospinal fluid and edema in T2weighted images) for these nonhuman primate studies. To control the accuracy of the positioning into the magnet, 10 sagittal Tl-weighted sections (TR = 360 ma, TE = 12 ma, field of view = 20 cm, 5 mm thickness, two excitations) of the brain were first performed. These images allowed us to localize the frontal planes containing the ear bars (frontal plane 0) and the needle tracks (lesion plane) (Fig. 1). Then, a set of 16 Tl-weighted images were performed in the frontal plane using a gradient echo sequence (TR = 360-460 ma, TE = lo-12 ma, flip angle = 90°, four excitations). To optimize image quality, these slices were obtained in two “interleaved” acquisitions, yielding contiguous 3-mm thick slices, with an in-plane resolution of 600-780 pm. In one case, supplementary slices interspaced by 1 mm were acquired to check the position of the lesions plane. To obtain further information about the lesions, a heavily TB-weighted spin echo sequence was performed in a single frontal plane corresponding to the center of the lesion, using pulse gating and a TR value between 2300 and 2600 ma, TE = 25 and 120 ma, one excitation. In some experiments, a Tl-weighted image performed at the level of the lesion was obtained following 0.15 mmol/kg gadolinium (Gd-DOTA, Guerbet, France) intravenous injection, to explore possible blood-brainbarrier alterations following excitotoxic lesion. On average, each MR experiment lasted less than 1 h. Measurements Ventricle areas were measured on Tl-weighted images (four frontal planes: the plane containing the lesions sites, one rostral, and two caudal planes), by selecting the pixel values with a signal lower than the average brain signal. The ratio between the right to the left volume was taken as an index of the ventricular enlargement and therefore of the shrinkage of the lesioned caudate-putamen complex. Anatomy In order to analyze the modifications of MR signals seen in the lesioned areas, histological and immunocytochemical studies were performed in one of the baboons (baboon 1) 2 months after the last MRI examination. Even if the excitotoxic striatal lesion cannot be considered as static, at 4 weeks after excitotoxin infusion, the acute phase of striatal atrophy is over and there is no problem in comparing the pathology obtained 2 months after the last MR image. Under deep anesthesia, the animal received an intracardiac perfusion of physiological saline with heparin followed by fixation with a solution of 1% paraformaldehyde and 1% glutaraldehyde in 0.1 M phosphate buffer (pH 7.4). The brain was removed from the skull and postfixed overnight in the same solution. Coronal sections at the level of the stria-
20
HANTRAYE
ET
AL.
FIG. 2. Frontal Tl-weighted images of the lesions plane obtained 2 and 4 weeks after lesion on baboon 1. (A) On first examination, 2 weeks after IA injections, lesions were clearly visible in the right putamen and in the right caudate nucleus (appearing on the left of the image) as areas of low signal intensity (arrows). (B) Two weeks later, the right ventricle (white cross, appearing on the left side of the image) is markedly enlarged and the putaminal lesion is surrounded by an hypersignal of unknown origin.
turn were obtained. Frozen sections were cut from the most anterior part of the lesioned striatum containing the most frontal part of the lesion and processed for cresyl violet, acetyl choline esterase reaction (AChE), glial fibrillary acidic protein immunocytochemistry (GFAP, Dakopat), and ferritin immunoreactivity (Realef/biodesign). In addition, some selected sections were processed for Gomori reaction (9) to visualize the presence of iron in the lesioned and nonlesioned striatum. RESULTS
Anatomical
Localization
of Striatal
Lesions
When comparing MR images obtained on successive sessionsin baboons 1 (Figs. 2A and 2B) and 2, the positioning procedure provided reproducible localization of the head into the magnet (scan to scan variation cl mm). This protocol allowed localization of the lesion sites in the striatum and the monitoring of the alterations in the MR signals with time. The lesion sites were observed as small (2 mm) tetrahedral or triangular structures (explored in 3-mm thick slices), located approximately within 1 mm from the aimed stereotaxic coordinates (Fig. 1).
Tl- Weighted Images On Tl-weighted images, lesion sites appeared as small regions (2 mm diameter) of low signal intensity, observable in the caudate and putamen in a frontal plane located 26 mm anterior to the ear bars. On sagittal sections, we could visualize the needle tracks going
through the overlying cerebral cortex to the caudateputamen (Fig. 1). Two weeks after lesion, no obvious enlargement of the ipsilateral ventricle was noted in the two baboons studied (Fig. 2A and Table 1, baboons 1 and 2). At 4 weeks after lesion, a significant enlargement of the ipsilateral ventricle was noticeable on the two baboons receiving the largest dose of IA (Fig. 2B and Table l), extending from frontal plane + 23 mm to frontal plane + 32 mm, whereas no obvious enlargement was noted in the baboon receiving 350 pg IA. In comparing this enlargement of the ventricle to the progressive atrophy of the ipsilateral caudate nucleus, we observed a significant atrophy of the injected caudate nucleus at TABLE
1
Time Course of the Lateral Ventricular Enlargement Unilateral Excitotoxic Striatal Lesions in the Baboon Following
Baboon 1 (700 /.a) Weeks Mean
after R/L
lesion ratio
Baboon 2 (700 Pd
Baboon 3 (350 rd
2
4
2
4
4
1.5
2.2
1.0
1.5
1.1
Note. Results are expressed as the ratio between the volumes of the lateral ventricle in the lesioned hemisphere (R) and in the contralatera1 hemisphere (L). Volumes of the lateral ventricles were estimated by measuring their corresponding areas on four adjacent planes (the lesion plane, one anterior, and two posterior; thickness, 3 mm). Areas corresponding to lateral ventricles were summated and the ratios between right (lesioned hemisphere) and left (control side) ventricle areas (R/L ratio) computed.
MR
FIG. 3. and caudate Four weeks
IMAGING
OF
STRIATAL
EXCITOTOXIC
LESIONS
IN
BABOON
21
TB-weighted image of the lesion plane in baboon 2 (TR = 2500 ma, TE = 120 ms). (A) Two weeks after lesion, the right putamen nucleus excitotoxic lesions are seen as heterogeneous hypersignal areas. The right lateral ventricle (star) is not yet dilated. (B) after lesion, the right lateral ventricle (star) is slightly enlarged.
the earliest examination, whereas ipsilateral ventricular enlargements were only noted on later examinations (4 weeks after lesion). In one case (baboon l), a hypersignal surrounding the lesion area was observed 4 weeks after IA injections (Fig. 2B). Except for a hypersignal in the frontal cortex corresponding to the needle track in one case (Fig. 1, baboon l), no other regions of noticeable abnormal signal intensity could be visualized. Four weeks after IA injections, in the experiment with gadolinium intravenous injection there was no apparent penetration of the Gd-DOTA into the brain at the level of the excitotoxic lesion (data not shown). T2- Weighted Images On T2-weighted images, large areas of high signal intensity surrounding the small regions of decreased Tl values were observed in the lesioned striatum of all animals at 2 (Fig. 3A) and 4 weeks after lesion (Fig. 3B). However, the area of those regions of T2 hyperintensity decreased between the 2- and 4-week examinations. Analysis of the cerebral ventricles (clearly visible in T2 images as regions of high signal intensity), confirmed the presence of an enlargement of the ipsilateral ventricle 4 weeks after lesion in the animals receiving the largest dose of IA (baboons 1 and 2), whereas no obvious ventricular enlargement could be observed in the baboon receiving the lowest 350 pg IA dose (baboon 3). Anatomical Results Two months after the last MRI study (i.e., 4 months after the excitotoxin injections), the presence of an ex-
citotoxic striatal lesion in baboon 1 was apparent by neuronal loss in Nissl stain and decreased AChE histochemical reaction in the injected caudate nucleus and putamen (Fig. 4a). In agreement with previous observations in IA-lesioned baboons (6), there was also a marked increase in GFAP immunoreactivity in the lesioned area (not shown). On an adjacent brain section immunostained to reveal ferritine, a marked and homogeneous increase in ferritine immunoreactivity was observed in the entire lesioned striatum (Fig. 4b), matching the GFAP-positive region. In contrast, discrete iron deposits in the lesioned striatum at the level of the injection sites were observed using the technique of Gomori (designed for iron or iron-derivative deposits staining). DISCUSSION
Previous studies conducted in rodents using similar lesion models have demonstrated the feasibility of studying in viuo excitotoxic lesions using MRI (18-20, 22, 24). In general, the intrastriatal injection of excitotoxin results in alterations (increases) of the signal intensity in T2-weighted images, in the injected areas. However, observations in rodents are greatly limited by the size of the rat brain which renders it difficult to discriminate between different brain structures (striaturn, thalamus, cerebral ventricle) in both Tl- or T2weighted images. Despite a relatively low magnetic field (0.5 T), the use of nonhuman primates allowed a better discrimination between different structures of the basal ganglia and localization of the lesion sites and cerebral ventricles within the brain. As our purpose was to follow
22
HANTRAYE
FIG. 4. (a) Section of the primate brain at the level of the caudate-putamen (CP) complex stained for acetylcholinesterase (AChE) reaction. The normal side on the left shows an intense reaction in the caudate nucleus (NC) and in the putamen (P). The injected side, on the right, shows a severe decrease in staining accompanied by a severe shrinkage of the CP complex. (b) Adjacent section immunostained with an antibody raised against ferritin, a potential marker for reactive microglia and macrophages. The area of increased reaction observed in the injected CP corresponds to a region devoid of AChE activity and of decreased T2-signal. Scale bar, 5 mm.
alterations of signal in striatal excitotoxic lesions, it was of crucial importance to minimize scan-to-scan variation in the repositioning of the monkey into the gantry. A MRI-guided stereotaxic technique has recently been described to allow accurate placement of microdialysis probes in the striatum in nonhuman primates (23). In the present study, we used a similar approach based on the use of a MR compatible “stereotaxic-like” head holder. One important feature of our protocol is that we rely on the reference plane defined by the surgical stereotaxic head holder to locate the head of the baboon into the magnet. This plane corresponds to the frontal plane 0 of the stereotaxic atlas of the baboon’s brain by Riche et al. (21). The same positioning device was used to perform the positron emission tomography studies (8), providing anatomical support for the interpretation
ET
AL.
of the PET scans. As assessedby comparison of MR images obtained at different time points after lesion, the repositioning in the magnet of each particular baboon was highly reproducible (~1 mm variation in the anteroposterior coordinates). From a biological point of view, the MR technique provides a simple and nontraumatic assessment of the tissue loss and modifications of signals following excitotoxin-induced striatal lesions. Previous studies in the rat have shown that excitotoxic lesions are normally associated with progressive neuronal loss, astrocytosis, microgliosis, edema, and increased T2-signals in the lesioned areas (16, 18-20, 22, 24). The comparisons between MR images and anatomical sections in the present study indicate that the increased T2 values observed in the lesioned primate striatum corresponds to astrocytosis (probably with a superimposed edema at 2 weeks after lesion). The hyposignals observed on Tl-weighted images are associated with regions of severe neuronal depletion (core of the lesion, as defined by Beal et al. (4)). Between 2 and 4 weeks after lesion, the area of increased T2 values decreases in size, probably as a result of regression of the edema. The histological observations obtained in baboon 1,4 weeks after the last MR examination, indicate a severe neuronal loss in the injected striatum on Nissi-stained sections. A marked astrocytosis was also observed in the lesioned caudate and putamen using GFAP immunocytochemistry as well as a marked increase in ferritin content. The neuronal cell loss as well as the glial response are typical features of excitotoxic lesions. On the other hand, the increase in ferritin content merits some comment, as the presence of ferritin in a form complexed to iron is believed to decrease T2 signal in the striatum of HD patients (1). Ferritin plays a central role in recycling iron for synthesis of heme and other proteins and acts as an iron-binding protein in the central nervous system. However, in the primate model, most of the ferritin present in the injected caudate-putamen was not associated with iron 4 months after lesion. The absence of iron in the lesioned CP could explain the discrepancy between our increased T2 values and the decreased T2 values observed in the HD striatum. However, although excitotoxic lesions have been used to model HD, it is likely that these acute injections are not able to mimic all the features of a slowly progressive degenerative illness. In this respect, it would be interesting to study the effect of a more chronic excitotoxic lesion on MR signals and to determine if with time, iron can be stored in the ferritin accumulated in the lesioned striatum, therefore lowering the T2 signal in this area. The significance of the increase in ferritin immunoreactivity following excitotoxic lesion requires further research. Recent findings suggest that ferritin immunoreactivity may be a marker for microglia and macrophages (13). Activated microglia/macrophagic cells, together
MR IMAGING
OF STRIATAL
EXCITOTOXIC
with reactive astrocytes, are the predominant glial cell types migrating into the lesioned area as a consequence of the inflamatory response (16). One hypothesis to explain the increased ferritin content in the insulted striatal regions is through the proliferation of ferritin-containing cells such as the microglial/macrophage cells associated with inflammatory response such as the one induced by acute excitotoxic lesions. REFERENCES 1. ACHNER, F.TH., S. FEUZER, W. POEWE, G. R&mMAyR, AND F. GERSTENBRAND. 1990. Magnetic resonance imaging of extrapyramidal diseases. In Aging Brain and Dementia: New Trends in Diagnosis and Therapy. pp. 483-495. Lisa, New York. 2. BANKNIEDER, A. R., J. M. PHILLIPS, K. T. JACKSON, AND S. I. VINAL. 1978. Comparison of ketamine with the combination of ketamine and xylazine for effective anesthesia in the rhesus monkey (Macaca mu&a). Lab. Anim. Sci. 28: 742-745. 3. BEAL, M. F., N. W. KOWALL, D. W. ELLISON, M. F. MAZUREK, K. J. SWARTZ, AND J. B. MARTIN. 1986. Replication of the neurochemical characteristics of Huntington’s disease by quinolinic acid. Nature 321: 168-171. 4. BEAL, M. F., N. W. KOWALL, K. J. Swmm, R. J. FERRANTE, AND J. B. bfARTIN. 1989. Differential sparing of somatostatin-neuropeptide Y and cholinergic neurons following striatal excitotoxin lesions. Synapse 31: 38-47. 5. COYLE, J. T., AND R. SCHWARCZ. 1976. Lesion of striatal neurons with kainic acid provides a model for Huntington’s chorea. Nature 263: 244-246. 6. HANTRAYE, P., D. RICHE, M. MAZIERE, B. MAZIERE, C. LOC’H, AND 0. ISACSON. 1989. Anatomical, behavioral and positron emission tomography studies of unilateral excitotoxic lesions of the baboon caudate-putamen as a primate model of Huntington’s disease. In Neural Mechanisms in Disorders of Movement (A. R. Crossman and M. A. Sambrook, Eds.), pp. 183-193. Libbey, London. 7. HANTFUYJZ, P., D. RICHE, M. MAZIERE, AND 0. ISACSON. 1990. A primate model of Huntington’s disease: Behavioral and anatomical studies of unilateral excitotoxic lesions of the caudate-putamen in the baboon. Exp. Neurol. 108: 91-105. 8. HANTRAYE, P., C. LOC’H, M. KHALILI-VARASTEH, C. CROUZEL, D. FOURNIER, J-C. Yom, 0. STULZAFT, D. RICHE, 0. ISACSON, AND M. tinzmz. 1992. 6(18F)-Fluoro-L-Dopa uptake and (76Br)bromolisuride binding in the excitotoxically lesioned caudate-putamen of non-human primates studied using positron emission tomography. Exp. Neural. 115: 218-227. 9. HUMASON, G. L. 1962. Animal Tissue Techniques, pp l-233. Freeman, San Francisco and London. 10. ISACSON, O., S. B. DUNNETT, AND A. BJORKLUND. 1985. Graftinduced behavioral recovery in an animal model of Huntington’s disease. Proc. Natl. Acad. Sci. USA 83: 2728-2732.
LESIONS
IN BABOON
23
11. ISACSON, O., W. FISCHER, K. WICTOFUN, D. DAWBARN, AND A. BJORLIJND. 1987. Astroglial response in the excitotoxically lesioned neostriatum and its projection areas in the rat. Neuroscience 20: 1043-1056. 12. ISACSON, O.,P. HANTRAYE, M. MAZIERE, M. V. SOFRONIEW,AM) D. RICHE. 1990. Apomorphine-induced dyskinesias after excitotoxic caudate-putamen lesions and the effects of neural transplantation in non-human primates. Prog. Brain Res. 82: 523533. 13. KANEKO, Y., T. KITAMOTO, J. TATEISHI, AND K. YAMAGUCHI. 1989. Ferritin immunohistochemistry as a marker for microglia. Acta Neuropathal. 79: 129-136. 14. KOENIG, S. H., D. G. BROWN, III, M. SPILLER, AND N. LUNDBOM. 1990. Relaxometry of Brain: Why white matter appears bright in MRI. Magn. Reson. Med. 14: 482-495. 15. MCGEER, E. G., AND P. L. MCGEER. 1976. Duplication of biochemical changes of Huntington’s chorea by intrastriatal injections of glutamic and kainic acids. Nature 263: 517-519. 16. IMARTY, S., I. DUSART, AM) M. PESCHANSKI. 1991. Glial changes following an excitotoxic lesion in the CNS-I. Microglia/Macrophages. Neuroscience 45: 529-539. 17. MILLER, D. H., G. JOHNSON, P. S. TOFTS, D. IMAcms, AND W. I. MCDONALD. 1989. Precise relaxation times measurements of normal-appearing white matter in inflammatory central nervous system disease. Magn. Reson. Med. 11: 331-336. 18. NORMAN, A. B., S. R. THOMAS, R. J. PRAY, R. C. SAMARATUNGA, AND P. R. S-ERG. 1990. Tl and T2 weighted magnetic resonance imaging of excitotoxic lesions and neural transplants in rat brain in vivo. Exp. Neurol. 109: X4-170. 19. NORMAN,A. B., S. R. THOMAS, R. J. PRAY, R. C. SAMARATUNGA, AND P. R. SANBERG. 1989. Magnetic resonance imaging of the rat brain following kainic lesions and fetal striatal tissue transplant. Brain Rex 483: 188-191. 20. PESCHANSKI, M., M. RUDIN, 0. ISACSON, M. DELEPIERRE, AND B. P. ROCQVES. 1988. Magnetic resonance imaging of intracerebral neural grafts. In Transplantation into the Mammalian CNS: Progress in Brain Research (D. M. Gash and J. R. Sladek, Jr., Eds.), Vol. 78, pp. 619-624. Elsevier, Amsterdam. 21. RICHE, D., A. CHRISTOLOMME, J. BERT, AND R. NAQIJET. 1968. Atlas stereotaxique du cerveau de babouin (Papio papio). Edition du CNRS, Paris. pp 7-207. 22. VAN BRUGGEN, N., M. DORAN, R. G. AHIER, M. D. KING, A. L. Busa, D. J. BROOKS, AND S. R. WILLIAMS. 1990. Imaging of kainate lesions in the rat brain. Proc. Meet. Sot. Magn. Reson. Med. 9th 2: 634. 23. WANG, J., S. SKIRBOLL, T. G. AIGNER, R. C. SAUNDERS, J. HSIAO, AND K. S. BANKIEWICZ. 1990. Methodology of microdialysis of neostriatum in hemiparkinsonian nonhuman primates. Exp. Neurol. 110: 181-186. 24. WANG, P. C., C. WAMBAFIE, A. GUIDOTTI, AND A. MURAKI. 1988. High resolution, high field (4,7 T) MRI of Huntington’s disease in an experimental rat model. Proc. Meet. Sot. Magn. Reson. Med. 7th 1: 486.
