Intrastriatal cerebellar grafts: differentiation of cerebellar anlage and sprouting of Purkinje cell axons

30

Detelopmental Brain Re,searctt, 74 ( 19931 30 -l() ,t~ 1993 Elsevier Science Publishers B.V. All rights reserved )165-3806;93/$06. )tl

BRESD 51648

Intrastriatal cerebellar grafts: differentiation of cerebellar anlage and sprouting of Purkinje cell axons C. Gerloff, U.J.K. Knappe, U. Hettmannsperger, T.K. Duffner and B. Volk Neuropathologische Abteilung, Pathologisches Institut, Albert-Ludwigs-Unit,ersitiit Freiburg, Freiburg i.Br. (Germany) (Accepted 19 January 1993)

Key words. Fetal cerebellar graft; Stereotaxic implantation; Axonal sprouting; Blood-brain barrier; Purkinje celt; Anti-Leu-4 (CD3); lmmunohistochemistry

Pieces of cerebellar primordia were obtained from Gt6 (day 16 of gestation) rat fetuses and stereotaxically injected into the striatum of adult Wistar rats. The transplants were allowed to integrate with the host brain for 2 h up to 6 months after implantation. Ninety four out of 105 transplants perfectly integrated with the host brain (90%) and established the typical trilaminar histoarchitecture of cerebellar cortex. The transplants were sufficiently vascularized. Vessels seen within the grafts provided all ultrastructural elements of a blood-brain barrier. Light microscopic evaluation of graft development showed no considerable retardation of cerebellar histogenesis. Electron microscopic examination disclosed normal ultrastructure of cerebellar neurons, as well as elements of regular synaptic organization. The topic of efferent graft-to-host projections was investigated 2.5 months after transplantation using the monoclonal Purkinje cell marker anti-Leu-4 (CD3). This method allowed us to detect immunoreactive, morphologically intact axons of grafted Purkinje cells running over long distances (at least 500/zm) within the host striatum. Whilst afferent but in no case efferent connections of heterotopic cerebeltar transplants had been demonstrated elsewhere, we could now prove the reciprocal modus of graft-host interaction with heterotopic cerebellar grafts.

INTRODUCTION T h e t e c h n i q u e of i n t r a c e r e b r a l t r a n s p l a n t a t i o n of fetal b r a i n tissue to a d u l t r e c i p i e n t s can p o t e n t i a l l y be very useful in r e s t o r a t i v e surgery in s o m e b r a i n diso r d e r s 42'45 as well as in analyzing d e v e l o p m e n t a l a n d r e g e n e r a t i v e m e c h a n i s m s 13'24'33'36'37'59'65'73. L i m i t i n g factor for t h e f u n c t i o n a l p e r s p e c t i v e o f i n t r a c e r e b r a l grafts is p e r f e c t a n a t o m i c a l i n t e g r a t i o n . This t e r m refers to sufficient v a s c u l a r i z a t i o n , e s t a b l i s h m e n t o f a tight g r a f t - h o s t i n t e r f a c e ( c o n t a c t o f g r e a t e r t h a n 20% o f graft surface with the host b r a i n 21), r e g u l a r graft differe n t i a t i o n (cyto- a n d h i s t o a r c h i t e c t u r e ) , f o r m a t i o n o f n o r m a l synaptic e l e m e n t s inside t h e graft, g r a f t - h o s t signal i n t e r a c t i o n a n d a b s e n c e o f i n f l a m m a t o r y or deg e n e r a t i v e host tissue r e a c t i o n s 21. R e s t o r a t i o n o f normal b r a i n functions p r o b a b l y n o t r e q u i r i n g cell-to-cell synaptic c o n t a c t s is b e i n g widely investigated, typically

in context with P a r k i n s o n ' s d i s e a s e 1e'1~'45. Still, m u c h less is k n o w n a b o u t the r e s t o r a t i v e c a p a c i t y of axonal s p r o u t i n g in d i s e a s e s r e q u i r i n g a r e - f o r m a t i o n of synaptic circuitries 13,19,73. T o a d d r e s s the topic o f axonal s p r o u t i n g a n d to investigate basic d e v e l o p m e n t a l m e c h a n i s m s , the m o d e l of h e t e r o t o p i c s t e r e o t a x i c i m p l a n t a t i o n o f fetal c e r e b e l lar tissue into a d u l t rat s t r i a t u m has b e e n i n t r o d u c e d at o u r l a b o r a t o r i e s . T h e suitability of fetal d o n o r tissue has b e e n p r o v e d previously 22"43. F e t a l c e r e b e l l a r tissue is p a r t i c u l a r l y suitable for i m p l a n t a t i o n a n d post-grafting analysis d u e to the distinctive h i s t o a r c h i t e c t u r e a n d well-known d e v e l o p m e n t of the c e r e b e l l u m in situ. M o r e o v e r , a serious i n t e r e s t in the plasticity of c e r e b e l lar n e u r o n s e m e r g e s from t h e p o o r t h e r a p e u t i c a l options in c h r o n i c d e g e n e r a t i v e c e r e b e l l a r diseases. H e t e r o t o p i c t r a n s p l a n t a t i o n has t u r n e d out to be especially s u i t e d to investigate f u n d a m e n t a l aspects of

Correspondence." B. Volk and/or C. Gerloff, Neuropathologische Abteilung am Pathologischen Institut der Albert-Ludwigs-Universit~it Freiburg, Albertstr. 19, D-7800 Freiburg i.Br., Germany.

31 neuronal plasticity and regenerative connectivity (in terms of axonal sprouting), since the inserted tissue is in vivo separated from its physiological environment. The choice of the (anterior) striatum as target area was based on the following considerations: First, intrastriatal grafts are strictly intraparenchymal. Second, cerebellar intrastriatal grafts are not likely to be influenced by physiological host-cerebellar circuits. Third, myelihated areas and neuropil of the striatum provide clearly distinct environmental conditions for axonal sprouting activities, which is relevant to the issue of dominating guidance mechanisms 15'24'5s. Moreover, the striatum is well accessible by stereotaxic means and host animals tolerate implants into the anterior striatum without disabling neurological or behavioral deficits. The stereotaxic approach assures minimal lesioning of adjacent host brain tissue and enables reproducible positioning of the graft. Heterotopic transplantation of fetal cerebellar tissue to adult rats has been initially reported by Kromer et al. 4° and, in more detail, by Alvarado-Mallart and Sotelo s. Target areas of these and further comparable studies were different neocortical regions 8'4°'41'6°'7~ as well as the anterior eye chamber 32'35'7~'v~. Hine 34 and Wells and McAllister 72 implanted cerebellar primordia to neocortical areas of neonate rats. Briefly summarizing former investigations, we know that heterotopically transplanted cerebellar primordia can develop normal lamination and foliation, containing the five categories of neurons which characterize normal cerebellar cortex; regarding cyto- and histogenesis, there was evidence of a developmental retardation up to 5 days compared to cerebellar development in s~tu "; the synaptic pattern of normal cerebellar cortex was grossly established within the implants (electron microscopic results), also the corticonucleocortical loop inside the graft could be traced by H R P injection methodsS; some studies gave evidence of non-specified or afferent fibers penetrating the graft-host interface 24'~'72. To our knowledge, in no case efferent fibers originating from heterotopic cerebellar grafts and verifiably sprouting into the host brain could be disclosed• In contrary, the few Purkinje cell (PC) axons observed within the graft-host interface and within adjacent host tissue definitely, after a short course, re-entered the graft or degenerated providing the typical retraction bulbs 6°. The prospective value of heterotopically grafted cerebellar tissue not establishing efferent contacts with the host brain would be decidedly poor. Thus, the present study was performed to validate a new model of heterotopic cerebellar grafting by evaluation of the anatomical integration and differentiation of cerebellar anlage within the striatum of adult host •

7 ~

animals. Subject of main interest was the possible outgrowth of efferent axonal projections from the graft into the surrounding host tissue. MATERIALS

AND METHODS

Animals Male and female Wistar rats were housed individually in plastic cages, receiving a standard dry diet ad libitum. Day I of gestation (G1) was determined by checking vaginal smears. P1 (postnatal day 1): fictitious day of birth after 22 days of gestation.

Preparation and transplantation qf f?'tal tissues On day 16 of pregnancy (GI6) mothers were anesthetized with ether and sodium thiopental and the uterine horns were surgically exposed. Individual fetuses were removed from the uterus and immersed in physiologic saline. The calvarium was removed to expose the cerebellar primordia. These were pinched off individually with a pair of fine forceps and placed in physiologic saline. Meninges and accompanying vasculature were gently teased away under a dissecting microscope. Cerebella of two fetuses were pooled together and mechanically homogenized with a razor blade. Four p,I of tissue suspension were aspirated into a glass needle made out of a finecaliber glass tube, cut and sharpened at its tip (outer diameter 1 ram), attached to a 2-ml syringe containing physiologic saline. For the surgical procedure, a stereotaxic fixation and target device (David Kopf Instruments) was used. After exposure of the skull, bore-hole trepanation and incision of the dura, 4 #1 of donor tissue suspension were slowly injected into the left rostral striatum of anesthetized ( e t h e r / s o d i u m thiopental) male adult Wistar rals weighing 200-250 g. Target coordinates were obtained from Pellegrino's stereotaxic atlas~5:1.8 mm ventrally and 2.5 mm to the left of the bregma (i.e., the intersection of coronal and sagittal suture), 4.5 mm perpendicularly into the parenchyma, measuring from the level of dura mater. The transplants were allowed to differentiate and integrate with the host tissue for periods ranging from 2 h up to 6 months. The survival time for more detailed studies on neuronal connectivity was 21 days, 35 days and 2.5 months, respectively. The tolal n u m b e r of host rats was 105.

Com'entional light and electron microscopy For conventional stainings, animals were anesthetized with e t h e r / s o d i u m thiopental and perfused via left cardiac ventricle with 5% formaldehyde in phosphate buffered saline (pH 7.35) or with 3.5% glutardialdehyde in S6rensen's phosphate buffer (pH 7.35). Brains were removed immediately, stored in the perfusion fixative for another 24 h and processed for standard light and electron microscopy (paraffin and Araldit M embeddings). Paraffin sections were stained using haematoxyline/eosine, Bodian's neurofilamenl impregnation, Kluever Barrera's myelin stain and Tibor Pap impregnation according to standard methods ~7. Semi-thin sections (1.0 # m ) were stained with toluidine blue. Ultrathin sections (0.07 # m ) were stained with uranyl acetate and lead acetate and investigated with an EM 95-2 electron microscope (Zeiss, Oberkochen, Germany).

Anti-Leu-4 (CD3) immunohistochemistry For immunohistochemistry, animals were sacrificed in deep ether anesthesia. Brains were immediately removed en bloc and cryofixed (mcthylbutane, liquid nitrogen). 20 /_tm cryostat sections were m o u n t e d on glass slides and fixed in - 2 0 ° C cold 100% acetone for 5 rain. Rehydrated sections were incubated with biotinylated anti-Leu-4 (CD3) (monoclonal antibody; Becton Dickinson), diluted 1:200, washed and incubated with A B C peroxidase reagent (Vector Laboratories). Bound peroxidase was visualized with 3,3'-diaminobenzidinetetrahydrochloride and 3-amino-9-ethylcarbazole, respectively. Eventually, washed sections were contrasted for 30 s in (I.1% OsO 4 (only in D A B preparations) and counterstained with Mayer's hemalaun or, exceptionally, with Kluever Barrera's myelin stain or Bodian's neurofilament impregnation.

32 RESULTS

Ninety-four out of 105 transplants perfectly integrated 21 with the host brain (success rate: 90%), 11 remained in an extraparenchymal situation or became necrotic. The latter were contaminated with leptomeningeal tissue. In 10% of the surviving 94 transplants small necrotic areas were seen. Twenty two of the vital transplants (23%) were displaced to regions others than the striatum (e.g., along the corpus callosum or to more basally located brain regions). Local edema due to the implantation procedure was seen until postoperative day 5 (5DAG). Leukocytes within the grafts were present only at 3DAG and 5DAG. Haemosiderin deposits and some siderophages occurred within the graft-host interface and in the adjacent host tissue in all preparations. No behavioral abnormalities of host animals were noticed. Neither wound infections nor other inflammatory changes were seen.

Development of cerebellar anlage (light microscopic results') Early postoperatiue period (1-2.5 h, ODAG which corresponds to G16) and early development (3DAG-5DAG which corresponds to G19-G21). Immediately after injection, grafts appeared as clusters of proliferating cells with mitotic activity and immature postmitotic neurons. On G19-G21, the external granular cell layer (EGL) could be identified consisting of 1-2 rows of cells. Immature PCs (9-13 ~ m in diameter) were irregularly grouped. Molecular layer (ML) and internal granular layer (IGL) were still absent on G21.

Postnatal development (7DAG-34DAG P1-P28). P1P5: grafted neurons were preferentially arranged in a concentric manner around blood vessels. The E G L contained 3 - 6 cell rows on P1, eight cell rows on P5. On P1, PCs were aligned in 2-3 rows (Fig. 1). On P4, the typical PC monolayer appeared. Initially on P4, an IGL of 2 cell rows was seen. Correspondingly, on P4 the transitory ML could be identified containing few cells which were specified as migratory granule cells on the basis of their characteristic morphology and location (Fig. 2). P6-P12: the maximum extension of the E G L was reached on P12 with up to 13 cell rows and an extension of 40-50 ~m. The PC monolayer consisted of PCs having apical cones and primary dendrites with varying (including few atypical) adjustments: mostly towards the E G L as regular, partly in opposite direction, partly in a parallel manner to the ML. Some minor clusters of large cells were still present, most likely resembling deep cerebellar nuclei (DCN) neurons. The width of the IGL was 2 or 3 cell rows on P8, 4 or 5 cell rows on P12. P14-P28: the

Fig, 1. Pl (which correponds to 7DAG). Grafted PCs larmws) with apical cones ( * 1: external granule cells (EGC). Toluidine blue (semithin section, 1 ,am), scale bar 50 p-m. Fig. 2. P4 (which corresponds to 10DAG). Monolayer of grafted PC: transitory molecular layer (ML); migrating granule cells (arrows). iGL, internal granular layer. Toluidine blue (semithin section, 1 ,am), scale bar 50 ~rn. Fig 3. Normal trilaminar histoarchitecture of differentiated cerebellar anlage, 2 months after grafting: IGL; PC monolayer (arrows); molecular layer (ML). Note the numerous axonal structures within the graft medullary layer (*); H, host striatum. Bodian (paraffin section, 5/xm), scale bar 50 #m. Fig. 4.2.5 months after grafting. Well differentiated, isolated mature PC within the host striatum (see text). Note the affinity of the dendritic system to the myelinated parts of the host striatum fiber tracts (SM). Anti-Leu-4 (CD3)/hemalaun (cryosection, 20 ~m), scale bar 50 txm.

extension of the E G L decreased constantly (nine cell rows on P14, two or three on P21, one, two to none on P24, one or none on P28). On P24, the extension of the ML was 100 ixm with still some migratory granule cells. PCs, aligned in monolayers, were characterized by typical pear-shaped cell bodies of up to 28/xm in diameter (P20). The adjustment of the dendritic systems was variable (cf. P12). The IGL on P24 was built of 5-8 rows of granule cells. During this period, the cortex of the grafted minicerebellum reached its determinate structure. Long-term results. Preparations of host animals 2-6 months after transplantation disclosed the typical trilaminar architecture of cerebellar cortex, comparable to

33 chromatin was thinly and uniformly distributed with few sites where it was collected into small masses. The cytoplasm of grafted PCs (P14, P30) appeared striated (known as PlasmastraBen 54) with regular equipment of organelles, including the characteristic hypolemmal cisternae (Figs. 5, 6). Also granule cells, closely packed in clusters with their somatic surfaces in juxtaposition, did not show any deviations from their ultrastructure in situ (P14, P30). A massive amount of spine synapses (Gray type 1) between ultrastructurally regular PC dendritic thorns and parallel fibers was present in the ML (P30) (Fig. 5). Synapses 'en passant' (Fig. 6) between basket cell axons and the proximal PC dendrite were established (P14). Gray type 2-synapses between basket cell axons and PC somata (P14, P30) provided indirect evidence for the existence of faint pericellular baskets. Typical pinceaux formations (i.e., periaxonal plexus of basket cell axon terminals surrounding the initial segment of the PC axon) were not identified. Within the granular layer, in few cases cerebellar glomeruli with a central mossy fiber could be seen (Fig. 7).

the in situ situation (Fig. 3). The E G L had disappeared, the ML measured approximately 100/zm. PCs were arranged in monolayers, the average diameters of PC somata were 2 0 - 2 5 / z m . The IGL consisted of 4 - 8 cell rows. Although in most cases some large neurons were located within the medullary layer, a definite DCN zone was absent. Unexpectedly, we found a single differentiated PC in an isolated position within the graft-host interface on P29 (which corresponds to 35DAG) and a completely isolated mature PC within the host striatum 2.5 months after implantation (anti-Leu-4 (CD3) preparations). These PCs had well differentiated, showing the typical PC cytoarchitecture with pear-shaped cell soma and extensive dendritic tree. On the level of light microscopy, the 2.5 month PC was lacking any contact with the actual graft. Considerable deviations from the in situ morphology were quantitative in nature, mainly consisting in a reduced dendritic extension (Fig. 4).

Cellular ultrastructure (electron microscopic results) The ultrastructural cytoarchitecture of differentiated cerebellar neurons was regular. In particular, PCs and granule cells including the expected synaptic contacts were studied (20DAG which corresponds to P14 and 36DAG which corresponds to P30). In PCs, the nucleus displayed its typical structure with invaginations and secondary folds (P14, P30). The nuclear

Vascularization (electron microscopic results) Already 3 days after implantation capillary bridges penetrating the graft-host interface were established (Fig. 8). Electron microscopic evaluation of the ultrastructure of capillaries inside the graft (two prepara-

Fig. 5. P30 (which corresponds to 36DAG). Spine synapses (Gray type 1) between PC dendritic thorns (arrowheads) and parallel fibers ( * ) within the molecular layer. Note the characteristic hypolemmal cisternae (ttC) underlying the plasma m e m b r a n e of the Purkinje cell dendrite (PD). EM (ultrathin section, 0.07/~m), scale bar 1 /_tin. Fig. 6. P14 (which corresponds to 20DAG). Synapse en passant between a basket cell axon (BAX, arrowheads) and a proximal Purkinje cell dendrite (PD). PS, Purkinje cell soma; GL, glial processes ensheething the PC soma and PC dendrite. EM (ultrathin section, 0.07/zm), scale bar 1 /xm.

Fig. 7. P30 (which corresponds to 36DAG). Cerebellar glomerulus with a central mossy fiber (MF) contacted by granule cell dendrites ( * ). EM (ultrathin section. 0.07 g m ) , scale bar 1 /zm.

34

Bodian preparations. On G21, neither nerve fibers

Fig. 8. G19 (which correponds to 3DAG). Capillary vessel (*) penetrating the graft-host interface. T, transplant; H, host striatum. Toluidine blue (semithin section, 1/zm), scale bar 50 ~zm. Fig. 9. P14 (which corresponds to 20DAG). Capillary inside a graft: endothelial cell (EC); basement membrane (BM); pericyte (PE); astroglial processes (AG). EM (ultrathin section, 0.07 #,m), scale bar 1/zm.

tions) disclosed all ultrastructural elements of a blood-brain barrier on P14 (which correponds to 20DAG): (i) endothelial cells sealed together in an overlapping fashion; (ii) oblique tight junctions connecting the endothelial cells; (iii) a basement membrane; and (iv) astroglial processes ('pericapillary end feet') having extremely narrow intracellular gaps (Fig. 9).

Purkinje cell axons as graft-to-host projections (light microscopic results) Axonal elements in general were evaluated throughout graft development in Bodian preparations (49 grafts). To address the direction (efferent vs. afferent) of axons crossing the graft-host interface, serial cryosections of 20 grafts on 21DAG, 35DAG and 2.5 months after transplantation, were incubated with the specific monoclonal PC marker anti-Leu-4 (CD3).

crossing the graft-host interface nor fibers within the graft were seen. Initially on P1, axonal structures had entered the interface and were present inside the graft. P12 preparations already showed a dense network of axonal structures occupying the graft-host interface (Fig. 10). Two, 3, 4 and 6 months after transplantation, networks of axonal structures crossing the interface as well as many axons within the grafts were present in all preparations. Anti-Leu-4 (CD3) immunohistochemistry. Grafted PCs displayed a pattern of Leu-4 (CD3) antigenicity comparable to PCs in situ (Fig. 11, cf. also Fig. 4). On 21DAG, 35DAG and 2.5 months after implantation, immunopreparations counterstained with Bodian's neurofibril impregnation and hemalaun, respectively, disclosed anti-Leu-4 (CD3) positive axons, i.e., PC axons, among the mass of nerve fibers crossing the graft-host interface (Figs. 11, 12). This was observed in all grafts examined on 21DAG, in five of the six grafts on 35DAG and in eight of the ten grafts after 2.5 months. Within all grafts numerous immunoreactive axons were running their typical courses throughout IGL and medullary layer. In two grafts (21DAG, 35DAG), PC axons inside the minicerebellar structure converged toward few larger anti-Leu-4 (CD3) negative neurons, possibly resembling DCN cells. While the used 20 /zm cryostat sections and very slight counterstaining provided excellent results regarding the detection of PC axons within interface and host tissue, an additional clear analysis of single immunonegative cells (such as DCN neurons) inside the graft was not possible. Eventually, the 2.5 months preparations disclosed immunoreactive PC axons running over distances of 5 0 0 / z m and more within the host striatum, thus proving the existence of efferent graft-to-host projections originating from heterotopic fetal cerebellar grafts (Fig. 13). In seven of the ten grafts examined at 2.5 months after implantation, a total of 23 axons could be identified within host striatums, all of them emerging from PCs at the PC layers of the grafted minicerebellar structures. In a remarkable fashion, these axons showed a wave-like course, whenever possible placed at the boundaries between myelinated areas and neuropil of the host striatum (Fig. 13). PC axons in the host brain and inside the graft displayed their typical morphology as known from investigations using anti-Leu-4 (CD3) immunohistochemistry in normal rat cerebella in situ 2s. Small immunopositive aggregations led to intermittent bead-like dilatations throughout the whole course of each PC axon (Fig. 13). Additionally, larger focal swellings were

35

SM

|

~4

12

H

13

,,,

Fig. 10. P12 (which corresponds to 18DAG). Dense network of axons occupying the graft-host interface. T, transplant; H, host striatum. Bodian (paraffin section, 5 gm), scale bar 50 ~m, Fig. 11. P15 (which corresponds to 21DAG). Interface between a well differentiated and integrated cerebellar graft (T) and the host striatum (H). Note the orientation of PCs with their basal (axonal) pole towards the host tissue; single axon hillocks can be seen (arrows), igl, internal granular layer; ml, molecular layer. Anti-Leu-4 (CD3)/Bodian (cryosection, 20 izm), Nomarski technique, scale bar 50 ~m. Fig. 12.2.5 months after grafting. PC axons penetrating the graft-host interface (arrowheads). PC, Purkinje cell somata inside the graft; H, host striatum. Anti-Leu-4 (CD3)/hemalaun (cryosection, 20 ~m), scale bar 50 gm. Fig. 13. 2.5 months after grafting. PC axon (arrowheads) within the host striatum. Note the wave-like course along the boundaries between the neuropil of the host striatum (SN) and myelinated parts of the host striatum (SM). Anti-Leu-4 (CD3)/hemalaun (cryosection, 20 #m), scale bar

50/xm. present at various localizations of PC axons within the IGL. Except for one immunolabelled axon with a slightly reduced diameter and a terminal large retraction bulb, PC axons within the host striatum showed no signs of axonal degeneration. All immunoreactive axons which could be followed in serial sections terminated within the host striatum. Most of them ended forming a small terminal varicosity. However, since small bead-like dilatations were seen throughout all stretches of the PC axons, the method used did not allow to reliably distinguish synaptic varicosities from, for instance, rapid changes of axon orientations which could imitate axonal terminations on the lightmicroscopic level.

DISCUSSION The present results provide evidence that cerebellar primordia heterotopically grafted to the striatum of adult recipients survive and differentiate regularly. Long-term observations up to 6 months after transplantation showed no signs of rejection or degeneration in the well-developed cerebellar implants. In particular, we could prove the capacity of heterotopically grafted PCs to sprout axons into the host brain. Regular graft differentiation is a basic requirement for successful neural transplantation using fetal donor tissues 21. C o m p a r e d to in situ data as published by Altman and co-workers I 7, no considerable develop-

36 nections have been described in various models o f homotopic and heterotopic grafting: (i) organo-typicat afferent fibers to homotopic and heterotopic neural grafts~°'47's='6~; (ii)afferent fibers from atypical sources to heterotopic grafts~8"3~'34'5e; (iii) efferent fibers to target areas normally (in situ) innervated by the grafted tissue (reported with homotopic grafts)Z~Y'~'°~;~: and (iv) efferent fibers to atypical target areas (reported with heterotopic grafts) 19'5z63. Focusing on heterotopic cerebellar grafts, few investigations gave evidence of non-specified or afferent fibers penetrating the host-graft interface (degeneration methods, 5-HT immunohistochemistry) 34J'°'7~. Concerning efferent fiber connections, previous reports on transplantation of fetal cerebellar tissue to cavities in the occipital and parietal cortex of young rats indicated that axons of heterotopically grafted PCs fail to definitely invade the surrounding host tissue s'~'°. Using anti-Leu-4 (CD3) immunohistochemistry 2~,, the present study gave first evidence that axons of heterotopically grafted PCs can penetrate the graft-host interface and invade the host striatum, thus establishing efferent projections. As argued by Oblinger and Das :~~'5~, possibly the very large interface region between host and graft created by the completely intraparenchymal

mental retardation was noted. In particular, there was no abnormality in the time course of PC differentiation as has been described with superficial cerebellar grafts to the neocortex of neonate rats (5 days delay as evident by monolayer formation; see Table 1)72. A comparative synopsis of developmental events of cerebellar G16 primordia grafted to the striatum of adult rats (strictly intraparenchymal; present results), cerebellar GI8 primordia grafted to the neocortical surface of neonate rats (predominantly extraparenchymal) 72 and cerebellar anlage in situ is given in Table I. Since published data about the development of cerebellar elements are abundant and the transplantation versus in situ data have been summarized in Table I, we will basically confine the following discussion to the observation of graft-to-host projections and to peculiar aspects of graft differentiation. Neuronal connectivity The establishment of reciprocal graft-host interactions is an essential criterion of successful graft integration ~2'2~'52'61. With a view to clinical relevance, it is a crucial requirement for the potential restoration of complex brain functions. In general, the following categories of neural con-

TABLE l Development of cerebellar anlage after heterotopic transplantation and in situ Comparative summary of developmental events of cerebellar GI6 primordia grafted to the striatum of adult rats (strictly intraparenchymal localization; present results), cerebellar G18 primordia grafted to the neocortical surface of neonate rats (predominantly extraparenchymal localization)7~ and cerebellar anlage in situ~7. CA, cerebellar anlage; preop., preoperative period; postop, postoperative period, black bar: present results; white bar: cerebellar anlage grafted close to the neocorticat surface7Z;crossed bar: cerebellar anlage development in situ~-7. EGL: formation maximum

exteneion

cessation MOLECULAR

LAYER:

...............

'

. . . . . . . . . . . . . . . . . . . . . . . . . .

"

"

formation •

maximum

extension

_ P U R _K_I N J E

CELLS:

..li::':.

'

"

:::: ' :

............ :'::: I " " : " ' . • :

•

~ •

I

~. . . .

....... : .... ~ :

,

-

......

......

:

) ~

:..5

: : ...... :::

.......

i

: .r

)

)

' .

• : •

~

"

i

:

:

I

prollferatlon mlgratlon aggregaUon beneath formation monolayer maximum soma

t

............................... !:i": •:.: ::ii: ::'.: : .: . . : . . . . . : : ,.. . . . . . . . . . . . . . . .

EGL of

i.

.

.

:. :

~

.

,

,

,

,

, . ~

;.

.

:~l/' ~: •

i

. •

size

I G16

I

: :: :

I

~, ~

:

:

.

formation

:~:-~:.:;:.

,,

i

:::

: :

,

I

IGL:

:

:

i..

i :

. . . . . *

.

:

t

.. ,,:.:.:.:~,;:~.,:~.

:

:<:

::::

:

i

..:.:

:

i

• :

....

P1

:

I

,

I I I I r t G21

. ,

.!~'~"I,'T~'!~'~'~'-'~:~

~'

.-

•

.

~

....

•

,~. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

I

]

~5

:.. ~

....

. . . . P10

~

P 2 0

:

:

,

I I T I I [ I I I l l l P15

. :

Ill P 2 5

37 nature of the analyzed implants provides the necessary access to and from the graft for sprouting afferent and efferent fibers. In fact, in the studies cited above s'6° the transplants were predominantly extraparenchymal. Furthermore, in cerebellar transplants obtained with the cavity technique 6° most of the PC axons end in the graft zone containing the DCN. In none of our grafts a clear DCN zone was developed. This is in accordance with the results of Sotelo and Alvarado-Mallart 6° who found that DCN neuron survival is poor when the pipette-injection technique is used for the grafting procedure. Whether or not a deficit of D C N neurons in heterotopic cerebellar grafts is crucial for or at least facilitating the formation of PC axon graft-to-host projections, requires further investigation. The immunomorphology of PC axons seen in the host striatum was perfectly comparable to their morphology as demonstrated by anti-Leu-4 (CD3) immunohistochemistry in situ 2~. Characteristic features of normal anti-Leu-4 positive PC axons are intermittent small dilatations providing a beaded appearance of the nerve fiber as well as some large focal swellings occurring in the granular layer. Both phenomenons are known to occur under non-degenerative conditions in normal developing and mature cerebella 11'2°'28'3°'5° Unfortunately, this morphology made it impossible to closer define the endpoints of PC axons within the host striatum, since small varicosities could not simply be interpreted as synaptic boutons. A typical degenerative retraction bulb 6°, however, was only seen in one serial section. Thus, we propose that the majority of axons documented within the host brain were morphologically intact. As suggested with reference to the concept of excess connections in the CNS, target cells maintain their supplying neurons by means of the same retrograde trophic factors that are molecules required for neuronal survival during development 15. Consequently, the integrity of an axon 2.5 months after transplantation would, at least, indicate some functional contact to neurons or humeral trophic factors of the host brain. In frontal sections, the PC axons were seen to run a wave-like course closely apposed to the boundaries between myelinated areas and neuropil of the host striatum. The significance of boundaries between different types of tissues as non-specific guidance pathways in axonal growth has been stressed by various authors (reviewed in ref. 15, Ch. 6). Except for one case, these axons never penetrated the myelinated areas, thus lending further support to the efficacy of CNS myelin as non-permissive substrate for neurite growth 5s. However, to explain the point of inflection where the PC axon leaves one boundary and crosses the neuropil in search of the next one, always preserving a main

orientation away from the graft, other positive guidance m e c h a n i s m s must be considered such as chemotropism and cell surface or extracellular matrix adhesion molecules z4'15. The striatum verified to be an excellent target area to further investigate guidance mechanisms of axonal outgrowth due to the clearly distinctive growth-manipulating qualities of myelinated areas and neuropil. Comments on axogenesis. Formerly, developing PC axons in situ were shown to trail behind the PC as the body and dendrites migrate up to the cerebellar cortex 4'as'v°. The axons remain in position, ready to establish synaptic contacts with the deep nuclear cells 49. According to these findings, PC axons would be primarily oriented toward their target cells and finally have to elongate by means of intercalated growth. Contrarily, in cerebellar slices of P0-3 rats, PC axons end in growth cones which are still in the subcortical medullary zone 4~'. In vitro axons, moreover, elongate on substrates with high adhesive properties 44. It was therefore hypothesized that in situ some post-migratory, cortical PCs may send their axons to the deep nuclear neurons along the same pathways (of high adhesive properties) as migrating PCs reach the cortex 7°. PC axons then would again be primarily oriented toward their main target cells, but would extend by axonal sprouting. More recently, Altman and Bayer ~ and Bourrat and Sotelo u described an early ventral translocation of the differentiating DCN neurons, i.e., a change of position of these neurons from a rostrodorsal ( G 1 6 - G 1 7 ) to a ventrocaudal cerebellar location (G19-G20). Their results clearly show that, when PCs initiate their axogenesis in situ, DCN neurons have a superficial position in the rostral half of the cerebellar plate. Thus, the PC axons obviously do not have to be primarily oriented toward their main target cells. Proving that PC axons establish graft-to-host projections by axonal sprouting along boundaries between myelinated areas and neuropil of the host striatum, our results support and extend the more recent findings14'7°: in PC axogenesis (i) axons apparently do not have to be primarily oriented toward any typical target cells; (ii) axonal sprouting (i.e., distal apposition) positively is one optional mechanism of PC axon elongation in vivo; and (iii) heterotopic PC axons can elongate on substrates with high adhesive properties under in vivo conditions.

Decelopment of cerebellar anlage and cascular&ation Light microscopic results. In agreement with the here presented results, trilaminar organization of cerebellar cortex has been reported previously for heterotopic

38 grafts placed in the anterior eye chamber 32'35'~'4'71 and in neocortical areas 8'4~'71'72. In marked contrast to the in situ situation, the grafted folia-like structures fail to establish one preferential three-dimensional orientation (present results; cf. refs. 8,41,72). The lack of extensive foliation in our transplants is most likely secondary to physical hindrance, since the grafts were not placed in a preformed cavity (cf. refs. 8,41), but stereotaxically pipette-injected. The variable adjustment of some PC dendritic trees is most likely a consequence of: (i) physical forces limiting the extension of the grafted 'minicerebella'; (ii) the organization of parallel fibers in bundles which were not always parallel to each other; and (iii) a numeral mismatch between granule cells and PCs, leading to a reduction of parallel fiber synapses with PC dendritic spines s'41. Similar modifications of dendrite orientation have been shown in mutant mice in situ due to the absence of parallel fibers ('weaver' and 'reeler' cerebella) 62 and with homotopic cerebellar transplants, again as a consequence of abnormal parallel fiber input 38. A numeral mismatch among the grafted cerebellar neurons is plausible, since the mechanical dissection of two cerebellar anlagen into small pieces probably disturbs the local numeral proportions within the donor tissue. The capacity of grafted PCs to establish normal synapses with parallel fibers is well preserved. Therefore, the dendritic disorientation of some grafted PCs represents an adaptation to a quantitatively modified input pattern rather than a transplantation-induced deficiency on an intracellular level. On the other hand, the observation of two PCs localized outside the actual graft with almost orderly adjusted dendrites substantiates, that a lack of the parallel fiber guidance matrix can be compensated by other guidance mechanisms. As to the localization of the isolated PCs, we propose that these ceils had migrated through the graft-host interface during differentiation, since most immature PCs on G16 (day of implantation) are premigratory and since we found PCs within the graft-host interface already on G21 (which corresponds to 5DAG). Similar invasive behavior has been observed for immature PCs in earlier studies on heterotopic 6° and homotopic17'25'38"61 cerebellar grafting. Electron microscopic results. In accordance with previously published data on cerebellar grafting 8'25'6~ we found ultrastructurally regular Gray type 1-synapses, synapses en passant, as well as Gray type 2-synapses. Additionally, in few cases cerebellar glomeruli with a central mossy fiber were documented. Although the vast majority of mossy fibers originates from extracerebellar sources, some mossy fibers have been shown to

arise from neurons of the cerebellar nuclear zone (in rat, cat and monkey cerebellum) 23'29'56,6~'. The establishment of cerebellar glomeruli indicates that, beside the standard corticonuclear projection (PCs to nuclear neurons), this return nucleocortical projection (nuclear neurons via mossy fibers to granule cells) is also achieved in transplants. In conclusion, the present electron microscopic findings reveal normal basic synaptic elements in intrastriatal cerebellar grafts. Comments on vascularization. Capillaries seen within the grafts provided all ultrastructural elements of a normal blood-brain barrier. Adding support to the functional significance of the present ultrastructural findings, Brundin et all 6 and more recently Geist et al. 27 demonstrated the establishment of a functionally intact blood-brain barrier at the graft site after implantation of fetal neural tissue suspensions (Evans Blue and H R P injection methods).

Concluding remarks The used stereotaxic approach to the insertion of heterotopic cerebellar grafts into a subcortical brain area such as the striatum with its myelinated areas and neuropil turned out to be a suitable tool to investigate basic questions in neuronal development and regenerative plasticity. High survival rates, regular differentiation of cerebellar anlage, establishment of typical synaptic structures, sufficient vascularization 9 and sprouting of efferent PC axons into the host striatum support the effectiveness of this grafting technique. The transplants are spatially separated from the host cerebellum and can be specifically labelled and studied, not interfering with regular cerebellar elements (as opposed to homotopic cerebellar grafts). Due to their standardized position relative to bregma and dura mater, the transplants can be manipulated in vivo in a variety of ways (e.g., application of trophic factors, neurotoxins, immunomodulators, drugs with known toxicity to cerebellar neurons67-69; induction of toxic metabolic states39; bioptic monitoring), permitting a level of investigation not possible in intact animals. New therapeutical approaches to degenerative cerebellar diseases could be based on the concept of neural grafting 25'61. To delineate the preconditions for successful transplantation and to estimate the achievable functional significance of cerebellar grafts, more detailed knowledge of the plasticity of cerebellar neurons is mandatory. Proving the capability of grafted PCs to sprout axons into a non-physiological environment along well defined boundaries, the presented results add elementary information to this topic. The method used offers a solid basis to further explore fundamental

39 qualities of developing and differentiated cerebellar neurons. Acknowledgements. This work was supported by the Deutsche Forschungsgemeinschaft (D.F.G.), VO: 272/5-2, and the BMFT.

ABBREVIATIONS

CNS DAB xDAG DCN EGL Gx HRP IGL Px PC

central nervous system diaminobenzidine x days after grafting deep cerebellar nuclei external granular cell layer gestational day x horseradish peroxidase internal granular cell layer postnatal day x Purkinje cell

REFERENCES 1 Altman, J. and Bayer, S.A., Embryonic development of the rat cerebellum. I. Delineation of the cerebellar primordium and early cell movements, J. Comp. Neurol., 231 (1985) 1-26. 2 Altman, J. and Bayer, S.A., Embryonic development of the rat cerebellum. If. Translocation and regional distribution of the deep neurons, J. Comp. Neurol., 231 (1985) 27-41. 3 AItman, J. and Bayer, S.A., Embryonic development of the rat cerebellum. III. Regional differences in the time of origin, migration and settling of Purkinje cells, J. Comp. NeuroL, 179 (19851 42-65. 4 Airman, J., Morphological development of the rat cerebellum and some of its mechanisms, Exp. Brain Res. 6 (1982) 8-49. In S.L. Palay and V. Chan-Palay (Eds.), The cerebellum - new t~istas, 1st edn., Springer, Berlin Heidelberg New York, 1982. 5 Altman, J., Postnatal development of the cerebellar cortex in the rat. I. The external germinal layer and the transitional molecular layer, J. Comp. Neurol., 145 (1972) 353-398. 6 Altman, J., Postnatal development of the cerebellar cortex in the rat. II. Phases in the maturation of Purkinje cells and of the molecular layer, J. Comp. Neurol., 145 (1972) 399-464. 7 Altman, J., Postnatal development of the cerebellar cortex in the rat. III. Maturation of the components of the granular layer, J. Comp. Neurol., 145 (1972) 465-514. 8 Alvarado-Mallart, R.M. and Sotelo, C., Differentiation of cerebellar anlage beterotopically transplanted to adult rat brain: a light and electron microscopic study, J. Comp. Neurol., 212 (1982) 247-267. 9 Anagnostopoulos, J., Knoth, R., Duffner, T., KIein, P J . and Volk, B., Vascularization of fetal cerebellar tissue transplanted into the striatum of adult rats. In J. Cervos-Navarro and R. Ferszt (Eds.), Stroke and Microcirculation, Raven Press, New York, 1987, pp. 10 Armengol J.A., Sotelo C., Angaut, P., Alvarado-Mallart, R.M., Organization of host afferents to cerebellar grafts implanted into kainate lesioned cerebellum in adult rats, Eur. J. Neurosci., 1 (19891 75-93. 11 Bishop, G., The pattern of distribution of the local axonal collaterals of Purkinje cells in the intermediate cortex of the anterior lobe and paramedian lobule of the cat cerebellum, J. Comp. Neurol., 210 (1982) 1 9. 12 Bj6rklund. A., Dunnett, S.B., Stenevi, U., Lewis, M.E. and Iversen, S.D., Reinnervation of the denervated striatum by substantia nigra transplants: functional consequences as revealed by

pharmacological and sensorimotor testing. Brain Res., 199 (1980) 3(17-333. 13 Bj6rklund, A., Nilsson, O.O. and Kalen, P., Reafferentation of the subcortically denervated hippocampus as a model for transplant-induced functional recovery in the CNS, Progress in Brain Research, 83 (1990) 411-426. 14 Bourrat, F. and Sotelo, C., Neuronal migration and dendritic maturation of the medial cerebellar nucleus in rat embryos: an HRP in vitro study using cerebellar slabs, Brain Res., 378 (1986) 69 85. 15 Brown, M.C., Hopkins, W.G. and Keynes, R.J., Essentials of neural det,elopment, 1st edn., Cambridge University Press, Cambridge, 1991, 46 pp. 16 Brundin, P., Widner, H., Nilsson, O.G., Strecker, R.E. and Bj6rklund, A., Intracerebral xenografts of dopamine neurons: the role of immunosuppression and the blood brain barrier, Exp. Brain Res., 75 (1989) 195-207. 17 Chang, A.C., Triarhou, L.C., Alyea, C.J., Low, W.C. and Ghetti, B., Developmental expression of polypeptide PEP-10 in cerebellar cell suspensions transplanted to the cerebellum of pcd mutant mice, Exp. Brain Res., 76 (1989) 639-645. 18 Chang, F., Steedman, J. and Lund, R., The lamination and connectivity of embryonic cerebral cortex transplanted into newborn rat cortex, J. Comp. Neurol., 244 (19861 401-411. 19 Clarke, D.J., Nilsson, O.G., Brundin, P. and Bj6rklund, A., Synaptic connections formed by grafts of different types of cholinergic neurons in the host hippocampus, Exp. NeuroL, 107 (1990) 11-22. 20 Crepel, F., Delhaye-Bouchaud, N., Dupont, J.L. and Sotelo, C., Dendritic and axonic fields of Purkinje cells in developing and X-irradiated rat cerebellum. A comparative study using intracellular staining with horseradish peroxidase, Neuroscience. 5 (1980) 333-347. 21 Das, G.D., Neural transplantation in mammalian brain: some conceptual and technical considerations, in R.B. Wallace and G.D. Das (Eds.), Neural tissue transplantation Research, 1st edn., Springer, New York, 1983, 1-64. 22 Das, G.D., Transplantation of embryonic tissue in the mammalian brain. I. Growth and differentiation of neuroblasts from various regions of the embryonic brain in the cerebellum of neonate rats, Life Sci., 4 (1974) 93-124. 23 Dietrichs, E., The cerebellar corticonuclear and nucleocortical projections in the cat as studied with anterograde and retrograde transport of horseradish peroxidase, ili. The anterior lobe, Anat. Embryol., 162 (1981) 223-247. 24 Gage, F.H. and Fisher, L.J., Intracerebral grafting: a tool for the neurobiologist, Neuron, 6 (1991) 1-12. 25 Gardene, R., Crepel, F., Alvarado-Mallart, R.M. and Sotelo, C., Fate of grafted embryonic Purkinje cells in the cerebellum of the adult "Purkinje cell degeneration' mutant mouse. II. Development of synaptic responses: an in vitro study, J. Cornp. Neurol., 295 (1990) 188-196. 26 Garson, J.A., Beverley, P.C.L., Coakham, H.B. and Harper, E.I., Monoclonal antibodies against human T lymphocytes label Purkinje neurons of many species, Nature, 298 (1982) 375-377. 27 Geist, M.J., Marls, D.O. and Grady, M.S., Blood-brain barrier permeability is not altered by allograft or xenograft fetal neural cell suspension grafts, Exp. Neurol., 111 (1991) 166-74. 28 Gerloff, C., Knoth, R. and Volk, B., Cytoplasmic expression of the Leu-4 (CD3) antigen in developing Purkinje cells in the rat cerebellum, Neuropath. Appl. Neurobiol., in press. 29 Gould, B.B., The organization of afferents to the cerebellar cortex in the cat: projections from the deep cerebellar nuclei, J. Comp. Neurol., 184 (1979) 27-42. 30 Gravel, C., Leclerc, N., Plioplys, A. and Hawkes, R.B., Focal axon swellings in rat cerebellar Purkinje cells during normal development, Brain Res., 363 (1986) 325-332. 31 Hallas, B.H., Brian, H. and Das, G.D., Afferents of embryonic tectal, diencephal and neocortical transplants in the cerebellum of the rat. Anat. Rec., 196 (19801 70A. 32 Hamori, J. and Takacs, J., Morphological study of cerebellar transplant cocultivated with cerebral cortical graft in the anterior

40

33 34 35

36

37

38

39

40

41

42

43 44 45

46

47

48

49

50

51

52

53

eye chamber. II. Purkinje cells and molecular layer, Anat. Embryol. (Berl.), 177 (1988) 557-569. Hankin, M. and Lund, R., How do retinal axons find their targets in the developing brain? TINS, 14 (1991) 224-228. Hine, R.J., Transplanted cerebellar tissue in the rat: its growth and its afferents, Anat. Rec., 187 (1977) 605. Hoffer, B., Seiger, A., Ljungberg, T. and Olson, L., Electrophysiological and cytological studies of brain homografts in the anterior chamber of the eye: maturation of cerebellar cortex in oculo, Brain Res., 79 (1974) 165-184. Ikegami, S., Nihonmatsu, I. and Kawamura, H., Transplantation of ventral forebrain cholinergic neurons to the hippocampus ameliorates impairment of radial-arm maze learning in rats with AF64A treatment, Brain ICes., 548 (1991) 187-195. Kawamura, K., Murase, S. and Yuasa, S., Development of the rodent cerebellum and synaptic reformation of donor climbing terminals on spines of the host Purkinje dendrites after chemical deafferentation, J. Exp. Biol., 153 (1990) 289-303. Kawamura, K., Nanami, T., Kikuchi, Y. and Kitakami, A., Grafted granule and Purkinje cells can migrate into the mature cerebellum of normal adult rats, Exp. Brain Res., 70 (1988) 477-484. Kleihues, P., Kiessling, M., Thilmann, R., Xie, Y., Uozumi, A and Volk, B., Resistance to hypoglycemia of cerebellar transplants in the rat forebrain, Acta Neuropathol. (Berl.), 72 (1986) 23-28. Kromer, L.F., Bj6rklund, A and Stenevi, U., Intracephalic implants: a technique for studying neuronal interactions, Science, 204 (1979) 1117-1119. Kromer, L.F., Bj6rklund, A and Stenevi, U., Intracephalic embryonic neural implants in the adult rat brain. I. Growth and mature organization of brainstem, cerebellar and hippocampal implants, J. Comp. Neurol., 218 (1983) 433-459. Labandeira-Garcia, J.L., Wictorin, K., Cunningham, E.T, and Bj6rklund, A., Development of intrastriatal striatal grafts and their afferent innervation from the host, Neuroscience, 42 (1991) 407-426. LeGros Clark, W.E., Neuronal differentiation in implanted foetal cortical tissue, J. Neurol. Psychiatr., 3 (1940) 263-272. Letourneau, P., Cell-to-substratum adhesion and guidance of axonal elongation, Deal. BioL, 44 (1975) 92-101. Lindvall, O., Widner, H., Rehncrona, S., Brundin, P., Odin, P., Gustavii, B., Frackowiak, R., Leenders, K., Sawle, G., Rothwell, J.C., Bj6rklund, A. and Marsden, C.D., Transplantation of fetal dopamine neurons in Parkinson's disease: one-year clinical and neurophysiological observations in two patients with putaminal implants, Ann. Neurol., 31 (1992) 155-165. Llinfis, R. and Sugimori, M., Calcium conductance in Purkinje cell dendrites: their role in development and integration, Progress in Brain Research, 51 (1979) 323-334. Lund, R.D. and Hauschka, S.D., Transplanted neural tissue develops connections with host rat brain, Science, 193 (1976) 582-584. McGeer, P.L., Eccles, J. and McGeer, E.G., Molecular neurobiology of the mammalian brain, 2nd edn., Plenum Press, New York, 1987, pp. Miale, I. and Sidman, R.L., An autoradiographic analysis of histogenesis in the mouse cerebellum, Exp Neurol.. 4 (1961) 277-296. Nagy, J.l., Yamamoto, T., Dewar, K., Geiger, J.D. and Daddona, P.E., Adenosine deaminase-'like' immunoreactivity in cerebellar Purkinje cells of rat, Brain Res., 457 (1988) 21-28. Nonaka, T., Araki, M., Kimura, H., Nagatsu, I., Satoh, F. and Masuzawa, T., Transplantation of the rat pineal organ to the brain: pinealocyte differentiation and innervation, Cell Tissue Res., 260 (1990) 273-278. Oblinger, M.M. and Das, G.D., Connectivity of neural transplants in adult rats: analysis of afferents and efferents of neocortical transplants in the cerebellar hemisphere, Brain Res., 249 (1982) 31-49. Oblinger, M.M., Hallas, B.H. and Das, G.D., Neocortical transplants in the cerebellum of the rat: their afferents and efferents, Brain Res., 189 (1980) 228-232.

54 Palay, S.L. and Chan-Palay, V., Cerebellar Corte~t. (_'ytology anti Organization, 1st edn., Springer, Berlin Heidelberg, 1974. 55 Pellegrino, L.J., Pellegrino, A.S and Cushman, A.J., A stereotaxic atlas of the rat brain, 2nd edn., Plenum Press, New York London, 198l. 56 Ramon y Cajal, S., Histologie du ,~vst~me Nert,eua" de I'Homme et des Vert~br&', Maloine, Paris, 191 I. 57 Romeis, B., Mikroskopische Technik, 2. Aufl.. Vcrlag M. Oldenburg, 1988. 58 Schwab, M.E. and Caroni, P., Oligodendrocytes and CNS myelin are nonpermissive substrates for neurite growth and fibroblast spreading in vitro, J. Neurosci., 8 (1988) 2381-2393. 59 Silver, R., Lehman, M.N., Gibson, M., Gladstone, W.R. and Bittman, E.L., Dispersed cell suspensions of fetal SCN restore circadian rhythmicity in SCN-lesioned adult hamsters, Brain Res.. 525 (1990) 45-58. 60 Sotelo, C. and Alvarado-Mallart, R.M., Cerebellar transplants: immunocytochemical study of the specificity of Purkinje cells inputs and outputs. In A. Bj6rklund and U: Stenevi (Eds.), Neural grafting in the mammalian CNS, 1st edn., Elsevier, Amsterdam, 1985, pp. 205-215. 61 Sotelo, C., Alvarado-Mallart, R.M., Gardette, R. and Crepel, F., Fate of grafted embryonic Purkinje cells in the cerebellum of the adult 'Purkinje cell degeneration' mutant mouse. 1. Development of reciprocal graft-host interactions, Z Comp. Neurol., 295 (t990) 165-187. 62 Sotelo, C., Mutant mice and the formation of cerebel[ar circuitry, TINS, 3 (1980) 33-36. 63 Stenevi, U., Bj6rklund, A and Svendgaard, N.A., Transplantation of central and peripheral monoamine neurons to the adult rat brain: techniques and conditions for survival, Brain Res., 114 (1976) 1-20. 64 Takacs, J. and Hamori, J., Morphological study of cerebellar transplant cocultivated with cerebral cortical graft in the anterior eye chamber. I. Granular layer, Anat. Embryol. (Berl.), 177 (1988) 543-556. 65 Tan, K., Le Douarin, N.M., Development of the nuclei and cell migration in the medulla oblongata. Application of the quail-chick chimera system, Anat. Embryol Berl., 183 (1991) 321-343. 66 Tolbert, D.L., Bantli, H. and Bloedel, J.R., Organizational features of the cat and monkey cerebellar nucleocortical projection. J. Comp Neurol., 182 (1978) 39-56. 67 Volk, B., Hettmannsperger, U., Papp, T., Amelizad, Z., Oesch, F. and Knoth, R., Mapping of phenytoin-inducible cytochrome P450 immunoreactivity in the mouse central nervous system, Neuroscience, 42 (1991) 215-235. 68 Volk, B. and Kirchg~issner, N., Damage of Purkinje cell axons following chronic phenytoin administration, an animal model of distal axonopathy, Acta Neuropathol. (Berl.), 67 (1985) 67-74. 69 Volk, B., Kirchg~issner, N. and Dettmar, M., Degeneration of granule ceils following chronic phenytoin administration: an electron microscopic investigation of the mouse cerebellum, Exp. Neurol., 91 (1986) 60-70. 70 Wassef, M. and Sotelo, C., Asynchrony in the expression of guanosine 3',5'-phosphate-dependent protein kinase by clusters of Purkinje cells during the perinatal development of rat cerebellum, Neuroscience, 13 (1984) 1217-1241. 71 Wassef, M., Sotelo, C., Thomasset, M., Granholm, A.C., Leclerc, N., Raftafi, J. and Hawkes, R., Expression of compartmentation antigen zebrin I in cerebellar transplants, J. Comp. NeuroL, 294 (1990) 223-234. 72 Wells, J. and McAllister, J.P., The development of cerebellar primordia transplanted to the neocortex of the rat, Det,. Brain Res., 4 (1982) 167-179. 73 Wictorin, K., Brundin, P., Gustavii, B., Lindvall, O. and Bj6rklund, A., Reformation of long axon pathways in adult rat central nervous system by human forebrain neuroblasts, Nature, 347 (1990) 556-558. 74 Wille, W., Goldowitz, D., Seiger, A. and Olson, L., The neurological mutation staggerer is expressed in embryonic cerebellar transplants matured in the anterior eye chamber of normal mice, Neurosci Lett., 42 (1983) 1-6.