Effects of graft-derived dopaminergic innervation on the target neurons of patch and matrix compartments of the striatum

Pergamon

PII:

Neuroscience Vol. 76, No. 4, pp. 1173–1185, 1997 Copyright ? 1996 IBRO. Published by Elsevier Science Ltd Printed in Great Britain 0306–4522/97 $17.00+0.00 S0306-4522(96)00379-X

EFFECTS OF GRAFT-DERIVED DOPAMINERGIC INNERVATION ON THE TARGET NEURONS OF PATCH AND MATRIX COMPARTMENTS OF THE STRIATUM N. RAJAKUMAR,*‡ W. RUSHLOW,* B. RAJAKUMAR,* C. C. G. NAUS,* A. J. STOESSL*† and B. A. FLUMERFELT*§ *Department of Anatomy and Cell Biology, University of Western Ontario, London, Ontario, Canada, N6A 5C1 †Department of Clinical Neurological Sciences, University of Western Ontario, London, Ontario, Canada, N6A 5C1 ‡Department of Psychiatry, University of Western Ontario, London, Ontario, Canada, N6A 5C1 Abstract––Fetal dopaminergic neurons grafted into the dopamine-depleted striatum have previously been shown to normalize neurochemical and behavioural abnormalities. However, the extent of graft-induced recovery of striatal compartments, which differ in their ontogeny, neurochemical properties and function, is still not clear. The striosome and matrix compartments of the striatum provide a segregated projection to somatostatin-containing GABAergic neurons of the rostral part of the entopeduncular nucleus and somatostatin-negative GABAergic neurons of the caudal part of the entopeduncular nucleus, respectively. In the present study, preprosomatostatin and glutamate decarboxylase messenger RNA levels in the rostral and caudal parts of the entopeduncular nucleus were determined six and 18 months postgrafting in rats with complete recovery of rotational behaviour following apomorphine challenge, and in rats with unilateral 6-hydroxydopamine lesions or sham lesions and no grafts. Sections were processed for in situ hybridization using 35S-labelled cRNA probes for glutamate decarboxylase (67,000 mol. wt isoform; GAD67) and preprosomatostatin. Autoradiographs showed a marked increase in preprosomatostatin messenger RNA within the ipsilateral entopeduncular nucleus in 6-hydroxydopamine-lesioned rats, and a substantially lower increase six months postgrafting. At 18 months postgrafting, the preprosomatostatin messenger RNA levels were symmetrical within the entopeduncular nucleus. Unilateral depletion of striatal dopamine resulted in a moderate increase in GAD67 messenger RNA levels within the ipsilateral entopeduncular nucleus, along with a substantial decrease in GAD67 levels within the contralateral nucleus. By six months postgrafting, the GAD67 levels had decreased considerably within the ipsilateral entopeduncular nucleus, while the messenger RNA levels had returned to normal within the contralateral nucleus. Interestingly, at 18 months postgrafting, the GAD67 levels remained decreased within the ipsilateral entopeduncular nucleus and were significantly lower than the normal value. The results indicate that fetal nigral grafts placed within the dopamine-depleted striatum can restore the neurochemical alterations seen in striatal target areas such as the entopeduncular nucleus. This may form the neurochemical basis of graft-induced behavioural recovery, as the normalization of neurotransmitter messenger RNA levels in the entopeduncular nucleus reflects the restoration of overall activity in both direct and indirect striatal output pathways. The results also indicate that the graft-derived dopaminergic innervation restores the output of both striosome and matrix compartments of the striatum. The present results also showed a progressive recovery leading to over-compensation of neurotransmitter messenger RNA levels following grafting, perhaps indicating the importance of feedback regulation of grafted dopaminergic neurons by the host. Copyright ? 1996 IBRO. Published by Elsevier Science Ltd. Key words: entopeduncular nucleus, glutamate decarboxylase, somatostatin, in situ hybridization, 6-hydroxydopamine, Parkinson’s disease.

Fetal dopaminergic neurons grafted into the previously dopamine-depleted adult striatum survive, reinnervate the host striatum1,8–10,14,45,60,61 and reverse a number of behavioural abnormalities.3,23–27,43,44,56,59,61 Electron microscopic studies §To whom correspondence should be addressed. Abbreviations: DMEM, Dulbecco’s modified Eagle medium; EPN, entopeduncular nucleus; GAD, glutamate decarboxylase; 6-OHDA, 6-hydroxydopamine; TH, tyrosine hydroxylase.

have shown that the graft-derived dopaminergic fibres establish homotypic synaptic connections with striatal projection neurons.12,31,50,52,71,78 In addition, these grafts normalize the striatal dopamine receptor supersensitivity resulting from previous depletion of dopamine2,9,15,30 and restore a number of neurochemical indices of striatal neurons,5,15,16,18–20,30,51,54,70,73,74 as well as their basal firing rate.29,77 The mammalian striatum is a heterogeneous structure, in which at least two neurochemically and

1173

1174

N. Rajakumar et al.

functionally distinct compartments can be identified, namely, the striosome (or patch) and the matrix.36,40,41,62,65 The striosome is involved in the control of dopaminergic and serotoninergic neurons, while the matrix is involved in a loop that includes the thalamus and the cerebral cortex.33,34,38,47,67 The striosome–matrix compartmentalization is respected by all major striatal afferent projections, including nigrostriatal dopaminergic inputs.36,37,46 Dopaminergic neurons of the substantia nigra pars reticulata and the ventral tier of the substantia nigra pars compacta project to the striosomes, whereas those in the dorsal tier of the substantia nigra pars compacta, retrorubral area and the medial part of the ventral tegmental area project to the matrix.36,37,46 The density of dopaminergic terminals is high in striosomes, particularly during early postnatal periods.39,62 Dopaminergic regulation is also believed to be different between these two compartments.48 However, there are no reports on the pattern of graft-derived dopaminergic innervation and functional restoration of striatal compartments. In the present study, we determined the levels of mRNA involved in the biosynthesis of neurotransmitters in the target neurons of striosome and matrix compartments of the striatum in order to assess the functional recovery of the striatal compartments. We have shown previously that striosome neurons project to somatostatin-containing GABAergic neurons of the rostral entopeduncular nucleus (EPN), whereas matrix neurons project to GABAergic neurons of the caudal EPN.67 Following 6hydroxydopamine (6-OHDA)-induced unilateral depletion of striatal dopamine, preprosomatostatin mRNA levels increase significantly within the ipsilateral EPN,75 while glutamate decarboxylase (GAD) mRNA levels remain unchanged within the ipsilateral EPN and decrease in the contralateral nucleus.76 Therefore, quantitative in situ hybridization histochemistry was employed to measure the mRNA levels of preprosomatostatin and GAD in the EPN of 6-OHDA-lesioned rats with intrastriatal dopaminergic grafts to assess the role that grafts play in the functional recovery of striatal targets. EXPERIMENTAL PROCEDURES

6-Hydroxydopamine lesions Twenty adult female Wistar rats (Charles River, St Constant, Quebec, Canada), with body weights ranging from 260 to 280 g, were injected with desipramine hydrochloride (25 mg/kg, i.p.; Sigma, St Louis, MO, U.S.A.) 30 min prior to receiving intracerebral injections of 6-OHDA or its vehicle (0.05% ascorbic acid in saline) in order to protect noradrenergic neurons. Rats were then anaesthetized with an intraperitoneal injection of sodium pentobarbital (35 mg/kg; MTC Pharmaceuticals, Cambridge, Ontario, Canada) and placed in a Kopf stereotaxic frame. A freshly prepared solution of 6-OHDA hydrobromide (8 µg/4 µl; Sigma) was stereotaxically injected into the left medial forebrain bundle with a glass micropipette using the coordinates of Paxinos and Watson64 (AP 4.5 mm, DV 1.5 mm and ML 1.5 mm from the interaural

point). Sham-lesioned animals received a similar injection of 0.05% ascorbic acid without 6-OHDA. Injections were made slowly over 8 min and the pipette was left in situ for a further 5 min. Postoperatively, rats received analgesia (Pentazocine; 10 mg/kg, s.c.) and were kept warm until recovery. Two weeks following the lesion, the rats were injected subcutaneously in the neck with apomorphine (0.5 mg/kg) and their rotational behaviour was assessed with a computerized video activity monitor (Videomex, Columbus Instruments, OH, U.S.A.). Animals showing 10 or more clockwise (contralateral) rotations per minute were included in the study. Fetal nigral cell grafting Three to four weeks after the initial 6-OHDA lesion, suspensions of fetal ventral mesencephalic cells were placed into the lesioned striatum according to a protocol modified from Bjorklund et al.10 Briefly, the ventral mesencephalon of 14-day-old rat fetuses from the same strain was dissected in Hank’s balanced salt solution and collected in Dulbecco’s modified Eagle medium (DMEM; Gibco, Burlington, Ontario, Canada). Tissue was incubated in DMEM containing 0.2% trypsin (Sigma) at 37)C for 20 min, and then rinsed in five to six changes of fresh DMEM. Following this, the pieces of tissue were dissociated in 50–60 µl of DMEM (3 µl for each piece of ventral mesencephalon) with a sterile pipette tip (i.d. 0.5 mm) by repeated trituration until the suspension became milky. Once dissociated, the suspension was kept at room temperature and grafted within 2 h. Cell viability was assessed at the beginning of each grafting session using a 1:1 mixture of Acridine Orange and ethidium bromide.53 Suspensions with a cell viability of more than 90% were used in all grafts. A sterile 10-µl Hamilton syringe attached to a stereotaxic tower was used for grafting. Host rats were anaesthetized as described above, and 4 µl of ventral mesencephalic cell suspension was placed stereotaxically within the dopaminedepleted striatum using the coordinates of Paxinos and Watson64 (AP 7 mm, DV 5.5–6.5 mm, ML 3 mm). All grafted animals were tested for rotational behaviour as described above following apomorphine challenge, at three and five months postgrafting, and only those showing complete recovery (i.e.
Dopaminergic graft-induced recovery of striatal target neurons containing 2.4 kbp of cat GAD (67,000 mol. wt isoform; GAD67) cDNA28 (generously provided by Dr A. J. Tobin) in the sense and antisense orientations (SP65-16 and SP6513, respectively) were used to generate the cDNA template necessary to produce the riboprobes. Isolation and purification of plasmid DNA were performed according to the procedures outlined by Sambrook et al.69 Once the DNA was isolated, the plasmids were linearized with Pharmacia restriction endonucleases (Hind III and Eco R1 for somatostatin antisense and sense, respectively, while Sal-I was used for GAD67 sense and antisense). Production of riboprobes and the in situ hybridization procedure were performed as described previously.68 Following stringency washes (0.2 M standard saline citrate at 50)C for 1 h, 0.02 M at 55)C for 1 h and 0.02 M at 60)C for 1 h), sections were dehydrated in an ascending series of alcohol (50%, 70%, 90%, 95%, 100%) containing 300 mM ammonium acetate. Following this, slides were dipped in NBT-2 nuclear emulsion (50%; Kodak) at 45)C, allowed to air dry for 2 h and placed in light-tight boxes containing dry rite (BDH Inc.) for 15–20 days at 4)C. After the exposure, slides were processed with D19 developer (Kodak) and fixer (Kodak), washed, counterstained with Neutral Red, dehydrated in an ascending series of alcohol, cleared in xylene and coverslipped with Entelan (Merck). Data analysis All sections included in the quantitative analysis were processed simultaneously in order to minimize technical variability. For every section, the density of labelling in areas known to be devoid of specific labelling (e.g., the thalamus in the case of GAD in situ hybridization) was taken as the background intensity to extrapolate the absolute value of the signal. The intensity and distribution of silver grains in the autoradiographs of in situ hybridization material were quantified using the Mocha image analysis program (Jandel Scientific, San Rafael, CA, U.S.A.). The preprosomatostatin and GAD mRNA levels were determined in transverse sections taken through the rostral EPN at 6.45 and 6.9 mm, and the caudal EPN at 5.9 and 6.2 mm from the interaural plane, respectively. The EPN was examined with #10 and #40 objective lenses under bright-field illumination and captured with a frame grabber to determine the density of silver grains over individual cells. At each level, 50 neuronal cell bodies were selected randomly for the analysis of average grain density. RESULTS

Specificity of probes used in in situ hybridization In the present study, the labelling obtained with GAD67 and preprosomatostatin probes was concentrated over neuronal cell bodies whose distribution in the striatum, globus pallidus, EPN and reticulothalamic nucleus of sham-lesioned animals was identical to that described in normal rats in previous in situ hybridization studies.28,68,75 Sections hybridized without probe or with a sense probe were devoid of specific labelling, confirming the specificity of the technique. Extent of the initial lesion and the graft-derived innervation Injection of 6-OHDA into the medial forebrain bundle resulted in a selective and complete loss of TH-immunoreactive neurons within the ipsilateral

1175

substantia nigra, and a corresponding degeneration of fibres throughout the striatum. The TH immunolabelling appeared symmetrical within the striatum and the midbrain in sham-lesioned animals. The ventral mesencephalic grafts consisted of numerous densely immunoreactive TH-labelled neurons, confined to the middle part of the striatum. A robust fibre outgrowth radiated from the graft into the surrounding striatum. In animals six months postgrafting, a gradient of TH immunoreactivity was evident, such that the area surrounding the graft (400–500 µm) was densely labelled, whereas more distal regions of the striatum showed only sparse TH immunolabelling. However, at 18 months postgrafting, the distribution of TH-immunoreactive fibres was virtually homogeneous throughout the striatum. Double-immunolabelling for Calbindin-D28k and TH indicated that adjacent striosome and matrix compartments possessed a similar density of dopaminergic fibres at both six and 18 months postgrafting (Fig. 1). Preprosomatostatin mRNA in the entopeduncular nucleus In sham-lesioned animals, neurons containing preprosomatostatin mRNA were distributed symmetrically on both sides. In addition to EPN neurons, preprosomatostatin mRNA was seen over neurons of the lateral and medial aspects of the hypothalamus. In the present study, only those neurons in the EPN and the adjacent lateral hypothalamic area were analysed. The most conspicuous feature of 6-OHDAlesioned brains was the marked increase in the mRNA content of somatostatin-containing neurons within the ipsilateral EPN and the adjacent lateral hypothalamus (Fig. 2A, B). This was the result of an increase in silver grains per cell and there was no change in the number of labelled neurons. These changes were similar in rats seven and 19 months postlesion. The increase in the density of silver grains per cell was not significantly different among neurons situated in the medial, lateral, dorsal or ventral aspects of the EPN, nor those situated in different rostrocaudal levels of the nucleus, indicating a uniform regulation of somatostatin neurons of the EPN by striatal dopamine activity. The number of labelled neurons and the density of silver grains in the contralateral EPN appeared similar to those of sham-lesioned animals. At six months postgrafting, there was a considerable decrease in the preprosomatostatin mRNA levels in EPN neurons on the ipsilateral side (Fig. 2C). However, the density of silver grains per cell was still markedly higher than that seen in the EPN of sham-lesioned rats (Table 1). Interestingly, a few clusters of cells in the medial aspect of the EPN did not show any improvement (Fig. 2C). The contralateral EPN, however, remained comparable to that

1176

N. Rajakumar et al.

Fig. 1. Fluorescence photomicrographs of a section double-labelled for calbindin and TH, illustrating the reinnervation of striosomes by graft-derived dopaminergic fibres. (A) Grafted striatum at lower magnification, showing the calbindin-rich matrix (M) and a calbindin-poor striosome (S). (B, C) Double-exposure of the area outlined in A shown at high magnification. Note that the striosome (arrows) and the matrix (arrowheads) seen in B are richly innervated by TH-labelled fibres, shown in C. Identical fibre bundles are indicated by ‘‘x’’. Scale bars=100 µm (A), 40 µm (B, C).

Fig. 2. Dark-field photomicrographs showing the expression of preprosomatostatin mRNA in individual neurons (arrows) within the ipsilateral EPN of sham-lesioned rats (A) and 6-OHDA-lesioned rats (B), six months postgrafting (C) and 18 months postgrafting (D). Note the progressive decrease in mRNA levels in C and D. The distribution of silver grains over individual cells in D is comparable to that in A. Scale bar=25 µm (A–D).

Dopaminergic graft-induced recovery of striatal target neurons 1177

1178

N. Rajakumar et al. Table 1. Average silver grain density determined from 100 randomly selected neurons at two transverse levels of the entopeduncular nucleus in each animal Treatment

Preprosomatostatin, ipsilateral

GAD, ipsilateral

50.24 (49.47) 52.77 49.25 46.81

30.47 (28.28) 28.72 26.34 27.58

Lesion (7 months)

112.65 (116.31) 129.74 116.25 112.12 110.80

45.74 (45.92) 47.82 49.98 44.58 41.49

12.45 (12.28) 12.15 10.18 13.85 12.78

Lesion (19 months)

115.75 (117.3) 118.85

47.15 (47.76) 48.36

12.45 (11.67) 10.90

Graft (6 months)

80.15 (76.45) 88.45 70.68 61.08 76.82 81.55

35.48 (33.5) 31.17 32.96 34.15 37.05 30.18

30.45 (28.74) 26.12 30.05 25.78 31.95 28.10

Graft (18 months)

44.70 (45.01) 41.65 48.68

20.72 (22.02) 21.85 23.48

28.65 (30.33) 30.58 31.75

Sham

GAD, contralateral

The mean density of each group is given in parentheses; for entries where no value appears, the mean is the same as that given in the row(s) above. Optical units: 0=background, 250=black.

following sham lesioning. In animals 18 months postgrafting, there was a dramatic decrease in the density of silver grains throughout the ipsilateral EPN compared to that in ungrafted rats 19 months postlesion (Fig. 2D). The labelling was virtually symmetrical, and the density of silver grains per cell was comparable to that of sham-lesioned rats. Glutamate decarboxylase (67,000 mol. wt isoform) mRNA in the entopeduncular nucleus In sham-lesioned animals, virtually all neurons in the EPN showed labelling indicating the presence of GAD67 mRNA. However, the intensity of labelling per cell was much lower than that seen in the globus pallidus or the reticulothalamic nucleus. The density of silver grains appeared uniform in the mediolateral and dorsoventral directions, but was higher in the caudal part of the nucleus, which corresponds to areas of the EPN receiving projections from the matrix compartment of the striatum (cf. Rajakumar et al.67). Therefore, transverse sections through the caudal EPN at 6.2 and 5.9 mm from the interaural plane were examined. Following 6-OHDA lesion, there was a significant increase in the intensity of labelling in the ipsilateral EPN and a marked decrease of labelling within the contralateral nucleus (Fig. 3). The labelling seen in the reticulothalamic nucleus appeared symmetrical, indicating that the changes observed in the contralateral EPN were not artifactual. The increase in the density of silver grains appeared uniform throughout

the EPN. The changes remained similar in rats 19 months postlesion (Table 1). In animals six months postgrafting, the intensity of labelling in the ipsilateral EPN decreased considerably, and appeared comparable to that of shamlesioned rats (Fig. 4A). The labelling seen within the contralateral EPN increased significantly compared to that of ungrafted animals seven months postlesion, and appeared similar to sham-lesioned cases (Fig. 4B). Eighteen months postgrafting, the intensity of labelling within the ipsilateral EPN decreased markedly, and was significantly lower than that of shamlesioned rats (Fig. 4C, Table 1). The labelling within the contralateral nucleus, however, appeared normal. DISCUSSION

To our knowledge, this is the first study demonstrating an effect on neurotransmitter expression outside the striatum following dopaminergic grafts placed in the striatum. Earlier studies by Dawson et al.20 and Manier et al.51 reported graft-induced restoration of D1 receptor levels in the substantia nigra and enkephalin levels in the globus pallidus, respectively. However, as these are presynaptic indices related to striatal terminals, the results simply reflected the functional restoration of striatal projection neurons. The present results indicate that grafted dopaminergic neurons innervate both striosome and matrix compartments of the striatum, and restore their function as shown by the recovery of neurotransmitters in

Fig. 3. Dark-field photomicrographs illustrating the expression of GAD67 mRNA in the EPN. Note the individual cells with silver grains (arrowheads) representing the mRNA expression. (A) Low-magnification view showing the EPN of a sham-lesioned rat in relation to the reticulothalamic nucleus (RT). (B) High-magnification view of EPN neurons expressing GAD67 mRNA (arrowheads) in a sham-lesioned animal. (C) The ipsilateral EPN following 6-OHDA lesions. Note the moderately elevated mRNA levels in individual cells (arrowheads). (D) The contralateral EPN showing markedly decreased mRNA levels following 6-OHDA lesions. Scale bars=60 µm (A), 25 µm (B–D).

Dopaminergic graft-induced recovery of striatal target neurons 1179

1180

N. Rajakumar et al.

Fig. 4. Dark-field photomicrographs showing the expression of GAD67 mRNA in the ipsilateral (A) and contralateral (B) EPN six months postgrafting. Note that the silver grains over individual cells (arrowheads) are comparable to those in sham-lesioned rats (Fig. 3B). (C) The ipsilateral EPN 18 months postgrafting. Note the markedly decreased levels of silver grains over individual cells (arrowheads). Scale bar=25 µm (A–C).

Dopaminergic graft-induced recovery of striatal target neurons

1181

Table 2. Summary of graft-induced restorative changes of preprosomatostatin and glutamate decarboxylase (67000 mol. wt isoform) mRNA levels in the entopeduncular nucleus Ipsilateral EPN

Lesion (7 months) Lesion (19 months) Graft (6 months) Graft (18 months)

Contralateral EPN

Preprosomatostatin

GAD67

Preprosomatostatin

GAD67

__ __ _ Normal

_ _ Normal `

Normal Normal Normal Normal

`` `` Normal Normal

Arrows: _, moderately increased; =, markedly increased; `, moderately decreased; >, markedly decreased.

their target neurons within the EPN. The results also suggest a temporally different pattern of recovery in that the EPN neurons receiving matrix projections (i.e. the caudal EPN) may recover earlier than those receiving striosome inputs (i.e. the rostral EPN). This could be due partly to a temporally different pattern of restoration of dopaminergic reinnervation of striosome and matrix neurons of the striatum. However, these somatostatin-containing neurons are also GABAergic, and therefore it is possible that GAD levels may have recovered earlier than the preprosomatostatin levels in these neurons, and the observed difference may be due to differential recovery of GABA versus peptide transmitter systems. Further studies are required to resolve this issue. It is important to note that unilateral depletion of striatal dopamine results in bilateral neurochemical changes both within the striatum and its target centres.57,75,76 Therefore, comparison was made with sham-lesioned animals which were otherwise identically treated. Graft-derived innervation of striatal compartments The present results obtained in animals six months postgrafting agree with previous data demonstrating a gradient of reinnervation of the striatum by grafted nigral cells.1,9,10,22,32 Previous studies have found that the graft-derived innervation can extend up to 1.5 mm into the surrounding striatum, leaving distal areas virtually denervated (for review, see Herman and Abrous42). Recently, however, Nikkhah et al.60 have obtained an extensive reinnervation of the striatum by means of a microtransplantation procedure. The present results clearly show that a single placement of graft can reinnervate the entire striatum. To date, no studies have directly examined the pattern of graft-derived dopaminergic innervation of striatal compartments. Sirinathsinghji and Dunnett73 have observed that the depletion of striatal µ-opiate receptor binding following nigrostriatal dopamine lesions can be restored by fetal nigral grafts. These authors argued that µ-opiate receptors are located presynaptically on dopaminergic terminals within striosomes, and therefore the restoration of these receptors indicates reinnervation of striosomes by dopaminergic fibres from the graft. Nevertheless,

recent in situ hybridization data21 indicating high levels of µ-opiate receptor mRNA in striosomal neurons favour possible recovery of those neurons. Restoration of mRNA levels in entopeduncular nucleus neurons by the graft Recovery of GAD and preprosomatostatin mRNA levels in the EPN observed in the present study was undoubtedly due to fetal nigral grafting, as nongrafted animals did not show any improvement of these mRNA levels, even at 19 months postlesion (Table 2). In addition to the substance P-containing GABAergic striatal projection, the EPN also receives a direct dopaminergic input from the substantia nigra.63 Direct dopaminergic action is probably mediated via D1 dopamine receptors located on the terminals of striato-EPN fibres.6 Moreover, DeLong and co-workers7 have recently suggested that loss of dopamine in the internal segment of the globus pallidus (EPN in the rat) is responsible for the development of tremor in Parkinson’s disease. Therefore, it is possible that the effect seen in the EPN following 6-OHDA lesions in the present series is due to a loss of direct dopaminergic modulation on EPN neurons, as well as the indirect effects mediated via the striatum. Since the graft-derived dopaminergic fibres did not extend into the EPN, the reversal of mRNA levels in EPN neurons following grafting suggests that most of the effects seen are secondary to striatal changes. According to the functional model of the basal ganglia proposed by Albin et al.,4 depletion of striatal dopamine leads to decreased activity in the striato-EPN (direct) projection and increased activity in the striato-pallido-subthalamo-EPN (indirect) projection. Consequently, the activity of EPN neurons will be augmented, resulting in inhibition of thalamocortical neurons.4,7,55 Recent evidence indicates that alterations of substance P, dynorphin, enkephalin and GABA levels of striatal projection neurons following dopamine depletion are normalized by fetal nigral grafting.5,16,51,54,72 This indicates a possible normalization of striatal activity in direct as well as indirect pathways, which has very likely contributed to the recovery of preprosomatostatin and GAD mRNA levels in EPN neurons seen in the present

1182

N. Rajakumar et al.

study. However, this requires normalization of activity in the globus pallidus and subthalamic nucleus. Studies are currently underway to verify this notion. Soghomonian and Chesselet75,76 reported that unilateral depletion of striatal dopamine results in an ipsilateral effect on preprosomatostatin mRNA levels and a contralateral effect on GAD mRNA levels in the EPN. The bilateral effects seen on GAD mRNA levels in the present study may be due to differences in the time-points and/or the rostrocaudal levels examined. Since somatostatin-containing neurons receive projections from striosomes,67 the asymmetry seen in preprosomatostatin mRNA levels may reflect a functional difference between striosome and matrix output systems. As the matrix projection is a link in the basal ganglia–thalamocortical loop, the observed contralateral decrease of GAD mRNA may be a compensatory mechanism mediated by the motor cortex. This is further supported by the parallel recovery of GAD mRNA levels within the contralateral EPN following grafting. Further studies are required to elucidate the mechanism underlying contralateral changes. Graft-induced neurochemical over-compensation The present study indicates an over-compensation of GAD mRNA levels in the ipsilateral EPN following the dopaminergic grafts. However, the statistical significance of this effect cannot be tested from the present results, as sham-lesioned animals were killed seven months after surgery, and they were significantly younger than 18-month postgrafted animals. Although a tendency for a progressive recovery in both GAD and preprosomatostatin mRNA levels in grafted animals was seen in this study, the 18-month postgrafting period may not be sufficient to examine this phenomenon in preprosomatostatin expression. Dopaminergic graft-induced over-compensation has been described for changes in D2 receptors,11 enkephalin mRNA levels16,74 and amphetamineinduced c-fos activation.2,15,17 Interestingly, Abrous et al.2 and Cenci and Bjorklund17 have observed that the over-compensation of striatal c-fos expression was associated with contralateral rotation, indicating a behavioural correlate for this neurochemical overshoot. The over-compensation in striatal target regions seen in the present study may be a reflection of over-compensated basal activity of the striatum. It is interesting to note that the striatal substance P level is normalized, but never elevated above normal levels,16,54 while enkephalin is over-compensated by the graft.16 Therefore, it is possible that the negative imbalance caused by the graft in the activity of direct and indirect projections from the striatum to the EPN may be partly responsible for the progressive reduction of GAD levels in the EPN. Nevertheless, this over-compensation of GAD levels in the EPN following grafting could also be due to a loss

of dopaminergic modulation of striatal activity involving the D1 receptors in the EPN itself. Functional correlates of striatal target recovery The present study clearly shows that the graftinduced normalization of striatal function leads to neurochemical recovery of its target neurons. This may in turn contribute to functional recovery.8,23– 27,66 Information reaching the striatum via corticostriatal projections is modulated by a number of neural circuits, including cholinergic, peptidergic and GABAergic interneurons, recurrent collateral branches of projection neurons, and dopaminergic, serotoninergic, thalamic, pallidal and amygdaloid inputs. The processed information is then sent to striatal target centres.35 Therefore, it is not surprising that striatal function is remarkably compensated during an insult and presumably also following recovery. Hence, it is important to ascertain that striatal recovery is indeed reflected in functional normalization of its target centres. Moreover, the basal ganglia are characterized by a massive reduction in nuclear volume such that large areas of the striatum converge on to the much smaller EPN, the recovery of which therefore provides a better index of functional normalization of striatal output. Following placement of dopaminergic grafts in the striatum, recovery of preprosomatostatin and GAD mRNA levels was observed in ipsilateral EPN neurons that receive striatal projections from striosome and matrix compartments, respectively. The somatostatin-containing EPN neurons project to the lateral habenular nucleus and are involved in the regulation of midbrain serotoninergic neurons.67 Therefore, the recovery of transmitter levels in these EPN neurons may indicate a restoration of striatal control on serotoninergic neurons following grafting. The GABAergic EPN neurons that receive projections from the matrix, on the other hand, are responsible for the disinhibitory influence of basal ganglia over thalamocortical neurons.67 Increased inhibitory activity of these EPN (globus pallidus internal segment in human) neurons is believed to be responsible for the majority of parkinsonian signs.80 Improvement of akinesia, rigidity and tremor following stereotaxic lesions of the globus pallidus in Parkinson’s patients further substantiates this notion.49,79 Therefore, the recovery of GAD levels observed in the present study may form the substrate for functional recovery reported in Parkinson’s patients following intrastriatal grafting of dopaminergic neurons.13

CONCLUSIONS

The present results indicate for the first time that graft-derived dopaminergic innervation of the striatum can lead to neurochemical restoration in

Dopaminergic graft-induced recovery of striatal target neurons

its target regions, such as the EPN. Our findings also indicate that the graft-derived dopaminergic innervation restores the output of both striosome and matrix compartments of the striatum, as shown by the recovery of neurotransmitter expression by their target neurons in the EPN. The demonstration of neurochemical over-compensation of striatal target centres following grafting represents an important issue in terms of its potentially deleterious effects on

1183

the host, and further studies are required to determine the underlying mechanism. Acknowledgements—The assistance of Mrs Judy Sholdice in the preparation of photomicrographs is greatly appreciated. This investigation was supported by the Medical Research Council of Canada and the Parkinson’s Foundation of Canada. N.R. and W.R. were also supported by the Ontario Mental Health Foundation and the Huntington’s Society of Canada.

REFERENCES

1. 2. 3. 4. 5.

6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23.

Abrous D. N., Guy J., Vigny A., Calas A., LeMoal M. and Herman J. F. (1988) Development of intracerebral dopaminergic grafts: a combined immunohistochemical and autoradiographic study of its time course and environmental influences. J. comp. Neurol. 273, 26–41. Abrous D. N., Torres E. M., Annett L. E., Reading P. J. and Dunnett S. B. (1992) Intrastriatal dopamine-rich grafts induce a hyperexpression of Fos protein when challenged with amphetamine. Expl Brain Res. 91, 181–190. Abrous D. N., Torres E. M. and Dunnett S. B. (1993) Dopaminergic grafts implanted into the neonatal or adult striatum: comparative effects on rotation and paw reaching deficits induced by subsequent unilateral nigrostriatal lesions in adulthood. Neuroscience 54, 657–668. Albin R. L., Young A. B. and Penney J. B. (1989) The functional anatomy of the basal ganglia disorders. Trends Neurosci. 12, 366–374. Bal A., Savasta M., Chritin M., Mennicken F., Abrous D. N., LeMoal M., Feuerstein C. and Herman J. P. (1993) Transplantation of fetal nigral cells reverses the increase of preproenkephalin mRNA levels in the rat striatum caused by 6-OHDA lesion of the dopaminergic nigrostriatal pathway: a quantitative in situ hybridization study. Molec. Brain Res. 18, 221–227. Barone P., Tucci I., Parashos S. A. and Chase T. N. (1987) D-1 dopamine receptor changes after striatal quinolinic acid lesion. Eur. J. Pharmac. 138, 141–145. Bergman H., Wichmann T., Karmon B. and DeLong M. R. (1994) The primate subthalamic nucleus. II. Neuronal activity in the MPTP model of Parkinsonism. J. Neurophysiol. 72, 507–520. Bjorklund A. and Stenevi U. (1979) Reconstruction of the nigrostriatal dopamine pathway by intracerebral nigral transplants. Brain Res. 177, 555–560. Bjorklund A., Schmidt R. H. and Stenevi U. (1980) Functional re-innervation of the neostriatum in the adult rat by use of intraparanchymal grafting of dissociated cell suspensions from the substantia nigra. Cell Tiss. Res. 212, 39–45. Bjorklund A., Stenevi U., Schmidt R. H., Dunnett S. B. and Gage F. H. (1983) Intracerebral grafting of neuronal cell suspensions. II. Survival and growth of nigral cell suspensions implanted in different brain sites. Acta physiol. scand. 522, Suppl., 9–18. Blunt S. B., Jenner P. and Marsden C. D. (1992) Autoradiographic study of striatal D1 and D2 dopamine receptors in 6-OHDA-lesioned rats receiving foetal ventral mesencephalic grafts and chronic treatment with -DOPA and carbidopa. Brain Res. 582, 299–311. Bolam J. P., Freund T. F., Bjorklund A., Dunnett S. B. and Smith A. D. (1987) Synaptic input and local output of dopaminergic neurons in grafts that functionally reinnervate the host neostriatum. Expl Brain Res. 68, 131–146. Brundin P. and Lindvall O. (1993) Transplantation in Parkinson’s disease. In Basic and Clinical Aspects of Neuroscience (eds Weil C., Muller E. E. and Thorner M. O.), Vol. 5, pp. 5–16. Springer, Berlin. Brundin P., Isacson O. and Bjorklund A. (1985) Monitoring cell viability in suspensions of embryonic CNS tissue and its use as a criterion for intracerebral graft survival. Brain Res. 331, 251–259. Cenci M. A., Kalen P., Mandel R. J., Wictorin K. and Bjorklund A. (1992) Dopaminergic transplants normalize amphetamine- and apomorphine-induced Fos expression in the 6-hydroxydopamine-lesioned striatum. Neuroscience 46, 943–957. Cenci M. A., Campbell K. and Bjorklund A. (1993) Neuropeptide messenger RNA expression in the 6-hydroxydopamine-lesioned rat striatum reinnervated by fetal dopaminergic transplants: differential effects of the grafts on preproenkephalin, preprotachykinin and prodynorphin messenger RNA levels. Neuroscience 57, 275–296. Cenci M. A. and Bjorklund A. (1994) Transection of corticostriatal afferents abolishes the hyperexpression of Fos and counteracts the development of rotational overcompensation induced by intrastriatal dopamine-rich grafts when challenged with amphetamine. Brain Res. 665, 167–174. Chritin M., Savasta M., Mennicken F., Bal A., Abrous D. N., LeMoal M., Feuerstein C. and Herman J. P. (1992) Intrastriatal dopamine-rich implants reverse the increase of dopamine D2 receptor mRNA levels caused by lesion of the nigrostriatal pathway: a quantitative in situ hybridization study. Eur. J. Neurosci. 4, 663–672. Dawson T. M., Dawson V. L., Gage F. H., Fisher L. J., Hunt M. A. and Wamsley J. K. (1991) Downregulation of muscarinic receptors in the rat caudate–putamen after lesioning of the ipsilateral nigrostriatal dopamine pathway with 6-hydroxydopamine (6-OHDA): normalization by fetal mesencephalic transplants. Brain Res. 540, 145–152. Dawson T. M., Dawson V. L., Gage F. H., Fisher L. J., Hunt M. A. and Wamsley J. K. (1991) Functional recovery of supersensitive dopamine receptors after intrastriatal grafts of fetal substantia nigra. Expl Neurol. 111, 282–292. Delfs J. M., Kong H., Mestek A., Chen Y., Yu L., Reisine T. and Chesselet M.-F. (1994) Expression of mu opioid receptor mRNA in rat brain: an in situ hybridization study at the single cell level. J. comp. Neurol. 345, 46–68. Doucet G., Brundin P., Descarries L. and Bjorklund A. (1990) Effect of prior dopamine denervation on survival and fiber outgrowth from intrastriatal fetal mesencephalic grafts. Eur. J. Neurosci. 2, 279–290. Dunnett S. B., Bjorklund A., Stenevi U. and Iversen S. D. (1981) Grafts of embryonic substantia nigra reinnervating the ventrolateral striatum ameliorate sensorimotor impairments and akinesia in rats with 6-OHDA lesions of the nigrostriatal pathway. Brain Res. 229, 209–217.

1184 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51.

52. 53. 54.

N. Rajakumar et al. Dunnett S. B., Bjorklund A., Stenevi U. and Iversen S. D. (1981) Behavioural recovery following transplantation of substantia nigra in rats subjected to 6-OHDA lesions of the nigrostriatal pathway. I. Unilateral lesions. Brain Res. 215, 147–161. Dunnett S. B., Bjorklund A., Schmidt R. H., Stenevi U. and Iversen S. D. (1983) Intracerebral grafting of neuronal cell suspensions. IV. Behavioural recovery in rats with unilateral 6-OHDA lesions following implantation of nigral cell suspension in different brain sites. Acta physiol. scand. 522, Suppl., 29–37. Dunnett S. B., Bjorklund A., Schmidt R. H., Stenevi U. and Iversen S. D. (1983) Intracerebral grafting of neuronal cell suspensions. V. Behavioural recovery in rats with bilateral 6-OHDA lesions following implantation of nigral cell suspensions. Acta physiol. scand. 522, Suppl., 39–47. Dunnett S. B., Whishaw I. Q., Rogers D. C. and Jones G. H. (1987) Dopamine-rich grafts ameliorate whole body asymmetry and sensory neglect but not independent limb use in rats with 6-OHDA lesions. Brain Res. 415, 63–78. Feldblum S., Erlander M. G. and Tobin A. J. (1993) Different distributions of GAD65 and GAD67 mRNAs suggest that the two glutamate decarboxylases play distinctive functional roles. J. Neurosci. Res. 34, 689–706. Fisher L. J., Young S. J., Tepper J. M., Groves P. M. and Gage F. H. (1991) Electrophysiological characteristics of cells within mesencephalon suspension grafts. Neuroscience 40, 109–122. Freed W. J., Ko G. N., Niehoff D. L., Kuhar M. J., Hoffer B. J., Olson L., Cannon-Spoor H. E., Morihisa J. M. and Wyatt R. J. (1983) Normalization of spiroperidol binding in the denervated rat striatum by homologous grafts of substantia nigra. Science 222, 937–939. Freund T. F., Powell J. F. and Smith A. D. (1984) Tyrosine hydroxylase-immunoreactive boutons in synaptic contact with identified striatonigral neurons, with particular reference to dendritic spines. Neuroscience 13, 1189–1215. Freund T. F., Bolam J. P., Bjorklund A., Stenevi U., Dunnett S. B., Powell J. F. and Smith A. D. (1985) Efferent synaptic connections of grafted dopaminergic neurons reinnervating the host striatum: a tyrosine hydroxylase immunocytochemical study. J. Neurosci. 5, 603–616. Gerfen C. R. (1984) The neostriatal mosaic: compartmentalization of corticostriatal input and striatonigral output systems. Nature 311, 461–464. Gerfen C. R. (1985) The neostriatal mosaic. I. Compartmental organization of projections from the striatum to the substantia nigra in the rat. J. comp. Neurol. 236, 154–176. Gerfen C. R. (1992) The neostriatal mosaic: multiple levels of compartmental organization in the basal ganglia. A. Rev. Neurosci. 15, 285–320. Gerfen C. R., Baimbridge K. G. and Miller J. J. (1985) The neostriatal mosaic: compartmental distribution of calcium-binding protein and parvalbumin in the basal ganglia of the rat and monkey. Proc. natn. Acad. Sci. U.S.A. 82, 8780–8784. Gerfen C. R., Herkenham M. and Thibault J. (1987) The neostriatal mosaic. II. Patch- and matrix-directed mesostriatal dopaminergic systems. J. Neurosci. 7, 3915–3934. Gimenez-Amaya J. M. and Graybiel A. M. (1990) Compartmental origins of the striatopallidal projection in the primate. Neuroscience 34, 111–126. Graybiel A. M. (1984) Correspondence between the dopamine islands and striosomes of the mammalian striatum. Neuroscience 13, 1157–1187. Graybiel A. M. and Ragsdale C. W. (1978) Histochemically distinct compartments in the striatum of human, monkey, and cat demonstrated by acetylthiocholinesterase staining. Proc. natn. Acad. Sci. U.S.A. 75, 5723–5726. Herkenham M. and Pert C. B. (1981) Mosaic distribution of opiate receptors, parafascicular projections and acetylcholinesterase in rat striatum. Nature 291, 415–418. Herman J. P. and Abrous N. D. (1994) Dopaminergic neural grafts after fifteen years: results and perspectives. Prog. Neurobiol. 44, 1–35. Herman J. P., Abrous D. N. and LeMoal M. (1991) Anatomical and behavioral comparison of unilateral dopamine-rich grafts implanted into the striatum of neonatal and adult rats. Neuroscience 40, 465–475. Herman J. P., Rouge-Pont F., LeMoal M. and Abrous D. N. (1993) Mechanisms of amphetamine-induced rotation in rats with unilateral intrastriatal grafts of embryonic dopaminergic neurons: a pharmacological and biochemical analysis. Neuroscience 53, 1083–1095. Hudson J. L., Bickford P., Johansson M., Hoffer B. J. and Stromberg I. (1994) Target and neurotransmitter specificity of fetal central nervous system transplants: importance for functional reinnervation. J. Neurosci. 14, 283–290. Jimenez-Castellanos J. and Graybiel A. M. (1987) Subdivisions of the dopamine-containing A8–A9–A10 complex identified by their differential mesostriatal innervation of striosomes and extrastriosomal matrix. Neuroscience 23, 223–242. Jimenez-Castellanos J. and Graybiel A. M. (1989) Compartmental origins of striatal efferent projections in the cat. Neuroscience 32, 297–321. Krebs M.-O., Trovero F., Desban M., Gauchy C., Glowinski J. and Kemel M. L. (1991) Distinct presynaptic regulation of dopamine release through NMDA receptors in striosome- and matrix-enriched areas of the rat striatum. J. Neurosci. 11, 1256–1262. Laitinen L. V., Bergenheim A. T. and Hariz M. I. (1992) Leksell’s posteroventral pallidotomy in the treatment of Parkinson’s disease. J. Neurosurg. 76, 53–61. Mahalik T. J., Finger T. E., Stromberg I. and Olson L. (1985) Substantia nigra transplants into denervated striatum of the rat: ultrastructure of grafts and host interconnections. J. comp. Neurol. 240, 60–70. Manier M., Abrous D. N., Feuerstein C., LeMoal M. and Herman J. P. (1991) Increase of striatal methionin enkephalin content following lesion of the nigrostriatal dopaminergic pathway in adult rats and reversal following the implantation of embryonic dopaminergic neurons: a quantitative immunohistochemical analysis. Neuroscience 42, 427–439. Mendez I., Elisevich K. and Flumerfelt B. (1991) Dopaminergic innervation of substance P-containing striatal neurons by fetal nigral grafts: an ultrastructural double-labeling immunocytochemical study. J. comp. Neurol. 308, 66–78. Mendez I., Elisevich K., Naus C. and Flumerfelt B. (1992) Restoration of nigrostriatal synaptic circuitry, striatal mRNA expression, and motor symmetry following embryonic substantia nigra grafts. Clin. Neurosurg. 38, 180–209. Mendez I. M., Naus C. C., Elisevich K. and Flumerfelt B. A. (1993) Normalization of striatal proenkephalin and preprotachykinin mRNA expression by fetal substantia nigra grafts. Expl Neurol. 119, 1–10.

Dopaminergic graft-induced recovery of striatal target neurons 55. 56. 57. 58. 59. 60. 61.

62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80.

1185

Mitchell I. J., Clarke C. E., Boyce S., Robertson R. G., Peggs D., Sambrook M. A. and Crossman A. R. (1989) Neural mechanisms underlying parkinsonian symptoms based upon regional uptake of 2-deoxyglucose in monkeys exposed to MPTP. Neuroscience 32, 213–226. Nadaud D., Herman J. F., Simon H. and LeMoal M. (1984) Functional recovery following transplantation of ventral mesencephalic cells in rat subjected to 6-OHDA lesions of the mesencephalic dopaminergic system. Brain Res. 304, 137–141. Najlerahim A. and Pearson R. C. A. (1992) Changes in glutamic acid decarboxylase mRNA in the pallidum of the rat following unilateral damage of the striatum and overlying cortex. Expl Neurol. 118, 352–356. Naus C. C. G., Miller F. D., Morrison J. H. and Bloom F. E. (1988) Immunohistochemical and in situ hybridization analysis of the development of the rat somatostatin-containing neocortical neuronal system. J. comp. Neurol. 269, 448–463. Nikkhah G., Duan W. M., Knappe U., Jodicke A. and Bjorklund A. (1993) Restoration of complex sensorimotor behavior and skilled forelimb use by a modified nigral cell suspension transplantation approach in the rat Parkinson model. Neuroscience 56, 33–43. Nikkhah G., Olsson M., Eberhard J., Bentlage C., Cunningham M. G. and Bjorklund A. (1994) A microtransplantation approach for cell suspension grafting in the rat Parkinson model: a detailed account of the methodology. Neuroscience 63, 57–72. Nishino H., Hashitani T., Kumazaki M., Sato H., Furuyama F., Isobe Y., Watari N., Kanai M. and Shiosaka S. (1990) Long-term survival of grafted cells, dopamine synthesis/release, synaptic connections, and functional recovery after transplantation of fetal nigral cells in rats with unilateral 6-OHDA lesions in the nigrostriatal pathway. Brain Res. 534, 83–93. Olson L., Seiger A. and Fuxe K. (1972) Heterogeneity of striatal and limbic dopamine innervation: highly fluorescent islands in developing and adult rats. Brain Res. 44, 283–288. Parent A. and Smith Y. (1987) Differential dopaminergic innervation of the two pallidal segments in the squirrel monkey. Brain Res. 426, 397–400. Paxinos G. and Watson C. (1987) The Rat Brain in Stereotaxic Coordinates. Academic, London. Penny G. R., Wilson C. J. and Kitai S. T. (1988) Relationship of the axonal and dendritic geometry of spiny projection neurons to the compartmental organization of the neostriatum. J. comp. Neurol. 269, 275–289. Perlow M. I., Freed W. J., Hoffer B. J., Seiger A., Olson L. and Wyatt R. J. (1979) Brain grafts reduce motor abnormalities produced by destruction of nigrostriatal dopamine system. Science 204, 643–647. Rajakumar N., Elisevich K. and Flumerfelt B. A. (1993) Compartmental origin of the striato-entopeduncular projection in the rat. J. comp. Neurol. 331, 286–296. Rushlow W., Naus C. G. and Flumerfelt B. A. (1994) Colocalization of prosomatostatin-derived peptides in the caudate–putamen of the rat. J. comp. Neurol. 349, 583–595. Sambrook J., Fritsch E. F. and Maniatis T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Savasta M., Mennicken F., Chritin M., Abrous D. N., Feuerstein C., LeMoal M. and Herman J. P. (1992) Intrastriatal dopamine-rich implants reverse the changes in dopamine D2 receptor densities caused by 6-hydroxydopamine lesion of the nigrostriatal pathway in rats: an autoradiographic study. Neuroscience 46, 729–738. Schultzberg M., Dunnett S. B., Bjorklund A., Stenevi U., Hokfelt T., Dockray G. J. and Goldstein M. (1984) Dopamine and cholecystokinin immunoreactive neurons in mesencephalic grafts reinnervating the neostriatum: evidence for selective growth regulation. Neuroscience 12, 17–32. Segovia J., Meloni R. and Gale K. (1989) Effect of dopaminergic denervation and transplant-derived reinnervation on a marker of striatal GABAergic function. Brain Res. 493, 185–189. Sirinathsinghji D. J. S. and Dunnett S. B. (1989) Disappearance of the µ-opiate receptor patches in the rat neostriatum following lesioning of the ipsilateral nigrostriatal dopamine pathway with MPP+: restoration by embryonic nigral dopamine grafts. Brain Res. 504, 115–120. Sirinathsinghji D. J. S. and Dunnett S. B. (1991) Increased proenkephalin mRNA levels in the rat neostriatum following lesion of the ipsilateral nigrostriatal dopamine pathway with MPP+: reversal by embryonic nigral dopamine grafts. Molec. Brain Res. 9, 263–269. Soghomonian J. J. and Chesselet M. F. (1991) Lesions of the dopaminergic nigrostriatal pathway alter preprosomatostatin messenger RNA levels in the striatum, the entopeduncular nucleus and the lateral hypothalamus of the rat. Neuroscience 42, 49–59. Soghomonian J. J. and Chesselet M. F. (1992) Effects of dopamine nigrostriatal lesions on the levels of messenger RNAs encoding two isoforms of glutamate decarboxylase in the globus pallidus and entopeduncular nucleus of the rat. Synapse 11, 124–133. Stromberg I., Johnson S., Hoffer B. and Olson L. (1985) Reinnervation of dopamine-denervated striatum by substantia nigra transplants: immunohistochemical and electrophysiological correlates. Neuroscience 14, 781–790. Stromberg I., Bygdeman M. and Almqvist P. (1992) Target-specific outgrowth from human mesencephalic tissue grafted to cortex or ventricle of immunosuppressed rats. J. comp. Neurol. 315, 445–456. Vitek J. L., Baron M. and Kaneoke Y. (1993) Microelectrode-guided pallidotomy is an effective treatment for medically intractable Parkinson’s disease. Neurology 44, A304. Wichmann T., Bergman H. and DeLong M. R. (1994) The primate subthalamic nucleus. III. Changes in motor behaviour and neuronal activity in the internal pallidum induced by subthalamic inactivation in the MPTP model of Parkinsonism. J. Neurophysiol. 72, 521–530. (Accepted 3 July 1996)