- Email: [email protected]
Striatal transplants prevent AF64A-induced retention deficits
PII SOO24-3205(98)00473-l
STRIATAL TRANSPLANTS Magda Giordano’,
Life Sciences, Vol. 63, No. 22, pp. 19531%1, 1998 Copyright0 1998 Ekvier Science Inc. Printed in the USA. All rights mewed ocm-3205/98 $19.00 t .oo
PREVENT AF64A-INDUCED
RETENTION DEFICITS
Rigoberto Salado-Castillo’, Manuel Sgnchez-Alavez3, Roberto A. Prado-AlcalB’
and
‘Centro de Neurobiologia, Campus UNAM-UAQ, Juriquilla, Qro. 76230, ‘Departamento de Psicologia, IJniversidad de Panamli, Panam& and ‘Departamento de Fisiologia, Facultad de Medicina, UNAM, Mexico, D. F. 04510, Mexico.
(Received
in final form September 14, 1998) Summary
The relevance of the cholinergic system in mnemonic processes has been repeatedly In addition to the cholinergic systems that project to the demonstrated. telencephalon, there are subcortical nuclei with intrinsic cholinergic cells which appear to be involved in memory consolidation; among these is the striatum. Intrastriatal administration of anticholinergic drugs, as well as excitotoxic and electrolytic lesions have been shown to disrupt the acquisition and retention of instrumentally conditioned behaviors. In the present study male Wistar rats were used to confirm the reported detrimental effects of striatal lesions produced by the cholinotoxin AF64A on long-term retention (LTR) of inhibitory avoidance and spontaneous locomotor activity, to determine its effects on short-term retention (STR) and to investigate whether intrastriatal homotopic transplants can reverse the AF64A-striatal lesions did not interfere with AF64A-induced behavioral deficits. STR but disrupted LTR of the inhibitory avoidance task, and striatal transplants prevented this deficit. Spontaneous locomotor activity increased after the lesion but promptly returned to baseline levels. These results support previous findings showing striatal involvement in long-term but not short-term retention and indicate that homotopic transplants induce behavioral recovery of a learning task in striatal lesioned rats. Key Words: ethylcholine mustard aziridinium, caudate-putamen, inhibitory avoidance, locomotor activity
long-term
retention,
striatal transplants,
The relevance of the cholinergic system in mnemonic processes has been repeatedly demonstrated in a variety of animal models (1). Clinical studies with aged and Alzheimer’s patients have also indicated that loss of cholinergic function may be related to the memory impairments observed in these patients; indeed systemic injections of anticholinergics result in memory loss (1). Cholinergic cells thought to be involved in memory can be found in several brain regions, including intrinsic cells within the caudate-putamen or striatum (2). Lesions of the striatum in rats have been shown to disrupt the acquisition and retention of many conditioned behaviors (3), and intrastriatal injections of anticholinergic drugs impair long-term memory of inhibitory avoidance (4) but leave short-term memory intact (5). Consistent with these data are results showing that choline administered to the striatum has a facilitating effect on positively and. negatively motivated conditioned behaviors (4) and that intrastriatal infusion of AF64A, an aziridinium analog of choline whose neurotoxic effects on cholinergic neurons is well documented (6, 7), impairs acquisition and retention of inhibitory avoidance conditioning (8). Send correspondence to: Magda Giordano, Ph.D., Centro de Neurobiologia, P.O. Box 70228, UNAM, 04510 M6xico’ D.F., Mexico, Telephone: (525) 623-4061, 623-4005, Fax: (524) 234-03-44, E-mail [email protected].
1054
AF64A and Striatal Transplants
Vol. 63, No. 22, 1998
Striatal transplants reverse behavioral deficits induced after excitotoxic or electrolytic lesions of the striatum (9-15). This compensatory effect is probably mediated by some of the many functional which include recovery of transmitter release and of consequences of the transplant electrophysiological activity, establishment of afferent and efferent connections and of synaptic relationships with the host brain (I 6-2 I ). The purpose of the present study was to test the reliability of the reported detrimental effects of striatal lesions produced by AF64A on retention of inhibitory avoidance and on spontaneous locomotor activity, to explore whether this lesion interferes with short-term memory, and to determine whether intrastriatal homotopic transplants reverse AF64A-induced behavioral deficits. Methods Animals. Naive male Wistar rats (200-300 g) were maintained in individual cages under continuous light and with free access to food and water. They were kept in these conditions at least live days before the experiments were initiated and were randomly assigned to one of five groups, three of which were lesioned and tested in the inhibitory avoidance task and whose locomotor activity was recorded on three occasions, and two naive groups which served as controls for the inhibitory avoidance task. One of the naive groups did not receive a foot-shock during avoidance training (NAWO), while the other naive group was trained as all the experimental animals (NAWS). Locomotor activity of some of the naive animals was not recorded. Apparafus. Black acrylic locomotor activity chambers (43 x 43 x 25.5 cm) with sixteen infrared sensors on the horizontal plane were used to record spontaneous locomotor activity for I5 min. Activity was recorded as counts per minute. The inhibitory avoidance chamber was a wooden box with two compartments of the same size (30 x 30 x 30 cm each), separated by a guillotine door. The lid of each compartment and the guillotine door were made of orange-colored lucite. The floor of one of the compartments was a grid made of aluminum bars (6 mm in diameter), separated I.5 cm center to center. The V-shaped lateral walls of the second compartment were stainless steel, and each was continuous with half the floor; there was a 1.5 cm slot separating each half-floor. Thus, when in this compartment, the rats were in contact with both walls and floor, which could be electrified using a square-pulse stimulator (Grass Instrument Co., model S-44) connected in series with a constant current unit (Grass Instrument Co., model CCUI). Illumination was provided by a IO-W light bulb located in the center of the lid of the compartment with the grid. Foot-shock delivery and measurement of latencies to cross from one compartment to the other one were accomplished by use of automated equipment. The conditioning box was located inside a dark, sound-deadening room, provided with background masking noise (BSR/LVE, model AU-902). Training and testing. On the day of training each animal was put inside the safe compartment and IO set later the door dividing the two compartments was opened and the latency to cross to the darker (electrifiable) compartment with all four paws was measured (acquisition latency). When the animals had crossed to this compartment the guillotine door was closed and a I mA foot-shock was delivered through the stainless steel plates. After five set the door was reopened, thus allowing the animal to escape to the first compartment and to remain there for 30 set before being put back in its home cage. The escape latency was also recorded; upon the rat’s escape the door was closed and the footshock was turned-off. Thirty minutes and twenty-four h later tests of retention were programmed exactly as the training session, except that the foot-shock was not delivered. If a rat did not cross within 600 set to the compartment where the foot-shock had been given the session was ended and a score of 600 was assigned. Opening and closing of the guillotine door, which was manually operated, activated the programming equipment that controlled shock delivery and timing of latencies. Lesion surgery: Animals were anesthetized with ketamine (50 mg/ml/kg i.p.) and received atropine sulfate (I ml/kg i.p.) and penicillin (1,200,OOO UI, i.m.). They were placed on a stereotaxic frame, an incision made and the skull exposed. Using an automatic infusion pump 1 pl of AF64A was delivered bilaterally over 5 minutes through a 27 gauge stainless steel injector whose tip was aimed at the dorsal striatum (I.5 mm anterior to bregma, +2.6 mm lateral and 4.5 mm ventral from dttt-a; incisor bar at -2.5 mm) (22); after an additional minute the injection needle, which was positioned perpendicular to the skull, was retracted, the wound sutured and the animals returned to their home cages. AF64A was activated according to the manufacturer’s specifications (RBI, Natlick MA,
Vol. 63, No. 22, 1998
AF64A and Striatal Transplants
1955
method I). Briefly, after dissolving the compound in distilled water, the pH was brought to 11.7 and maintained for 30 min at room temperature. Then the pH was lowered to 7.0 with concentrated HCI and adjusted to 7.4 with solid NaI-ICOS. The solution was used immediately while kept on ice, or it was frozen immediately and placed in a -8O’C freezer; final concentration was I I nmoles/pl. Animals were allowed fifteen days to recover after lesion surgery. Transplant surges: Following the same general procedure as used for lesion surgery, blocks of striatal or cerebellar fetal tissue (4 ~1) were bilaterally delivered into the same lesioned striatal site. Tissue was obtained from fetuses of 17 to 19 days of gestation (average CRL=I 9fO.5 mm). The brain was placed on its ventral surface, the cortex was removed and peeled laterally to expose the striatal primordium, the entire striatal primordium (medial and lateral ganglionic eminences) was clipped off and sectioned into tissue blocks. With the brain on the same position, the cerebellar anlagen located immediately underneath the colliculi, were cut and divided into tissue blocks. AAer dissection the tissue blocks were aspirated using a Hamilton syringe (50 ~1) fitted to a glass capillary attached to a standard needle, the syringe was then placed on the stereotaxic frame and the fetal tissue delivered into the lesi(aned striatum. Procedure: Before recording baseline activity levels, all animals were habituated to the activity chambers for 15 min on two non-consecutive days. After baseline activity levels were recorded, experimental! animals were lesioned with AF64A. Fifteen days later, locomotor activity was recorded again and 2 days later lesioned animals received intrastriatal transplants of striatal tissue (STR), cerebellar tissue (CEREB) or vehicle only (SH-TR). Six weeks after transplantation (60 days after lesion), their locomotor activity was recorded for a third time and 5-7 days afterwards the animals were trained and tested in inhibitory avoidance. All these behavioral variables were also measured in naive rats which belonged to one of two groups, a naive group trained with foot-shock (NAWS) or a naive group trained without foot-shock (NAWO). Following the last retention test, all lesioned and a few naive animals were anesthetized and intracardially perfused. Histology: .4Rer behavioral tests were completed, all lesioned animals, were intracardially perfused with 4% paraformaldehyde in PBA 0.1 M, their brains removed and kept in 20% sucrose until sectioned in a freezing microtome. Alternate sections (40 urn) were processed for Nissl staining and acetylcholinesterase (AChE) histochemistry. The protocol for AChE histochemistry was based on that of Van Ooteghem and Shipley (23). To make volume determinations transplants were measured using Cavalieri’s principle (24). Briefly, areas were measured by means of an image analysis system (IP-LAB) which corrected for the magnification used so that results were expressed in pm2, areas for each transplant were added up, multiplied by the distance between sections and converted to mm’. Statistical analysis: Locomotor activity data were analyzed using a two-way analysis of variance. Inhibitory avoidance data were analyzed using the Kruskal-Wallis analysis of variance followed by comparisons between the naive foot-shocked group and each of the other groups using MannWhitney U tests. Transplant volume for striatal and cerebellar grafts were compared using a t-test for independent samples, scores during the 24-hour retention test and transplant volumes were correlated using Pearson’s correlation coefficient. Results AF64A striatal lesions produced significant enlargement of the lateral Histological analysis. ventricles, reactive gliosis along the cannula track, and decreased AChE staining in the area immediately adjacent to the ventricles; no evidence of necrosis was evident in the remaining striatal tissue. In some animals (56.5%) lesions included portions of the anterior pole of the frontal cortex, while in a few others (17.4%) the caudal portion of the lateral ventricles appeared somewhat enlarged and/or septal damage was apparent (Fig. I). Striatal and cerebellar transplants were visible in all cases, were located bilaterally within the striatal parenchyma and in some cases they extended into the cortex, and /or the lateral ventricles. Size and location of the transplants varied, generally striatal transplants were larger than cerebellar ones, although some relatively large cerebellar transplants we;e found (Fig. I). No significant differences between average volume of striatal (4.69f3.45 mm ; minimum 1.75, maximum 10.85 mm’) and cerebellar (1.94f2.07 mm3; minimum 0.19, maximum 5.07 mm’) transplants were found. Both cerebellar and striatal transplants were positive for AChE staining, and striatal transplants showed the characteristic patchy appearance described previously
1956
AF64A and Striatal Transplants
Vol. 63, No. 22, 1998
(25, 26). All lesioned animals with evidence of bilateral lesions and all transplanted animals were used for analysis of behavioral data.
:’
.:
-.
Fig. 1 Photomicrographs of coronal sections through the striatum stained for AChE (a, c, e) and Nissl staining (b, d, f). Representative examples of AF64-A lesions restricted to the striatum (a) and including portions of the overlaying cortex (b). Appearance of one of the largest and best integrated cerebellar transplant (c, d). Also shown are two representative examples of striatal transplants (e) and (f). Note patchy appearance of AChE staining within transplanted tissue. S, lesioned striatum; T, transplant. Bar=O.lmm.
Vol. 63, No. 22, 1998
AF64A and Striatal Transplants
1957
Spontaneous locomotor activity. There was a significant session effect on spontaneous locomotor activity (F[2,126]=4.24, ~~0.05). Post hoc comparisons (Tukey’s HSD test) revealed that for all lesioned groups there was a significant increase in locomotor activity after the lesion (p
3 5
12000
g
10000
0
I F
8000
; 0
6000
5 5
4000
a
: i 2 4 z
2000
0 BASELINE
POST-TRANSPLANT
POST-LESION
Fig. 2 Mean r.pontaneous locomotor activity f SEM of AF64A-lesioned and naive animals. All lesioned groups showed increased activity during the post-lesion session when compared to baseline. This increase returned to baseline on the post-transplant session. See text for details. Bar fills are as follows: Striatal transplants, empty bars (n=14); cerebellar transplants, left-hatched bars (n=14); sham-transplants, right-hatched bars (n=23); naive group trained with foot-shock during inhibitory avoidance training (n=lO), cross-hatched bars; naive group trained without foot-shock during inhibitory avoidance training, horizontal lined bars (n=lO). *p
AF64A and Striatai Transplants
1958
Vol. 63, No. 22, 1998
700 600 500 -
c ii -
400 -
B d
200 -
ts
100 -
300 -
#
O-
w
a
-_
b
p z
700
P Y
600 -
z
500
g
400
g
300 200 100 O-
STR CEREB
SH-TR
NAWS
NAWO
Fig. 3 Median retention scores for AF64A-lesioned and naive animals 30 minutes (a) and 24 hours (b) after training on an inhibitory avoidance task. Abbreviations are as follows: STR, striatal transplants (n=l4); CEREB, cerebellar transplants (n=14); SH-TR, sham transplants (n=23); NAWS, naive group trained with foot-shock during inhibitory avoidance training (n=20); NAWO, naive group trained without foot-shock during inhibitory avoidance training (n=l9). The boundary of the box closest to zero indicates the 25’h percentile, a line within the box marks the median, and the boundary of the box farthest from zero indicates the 75’h percentile. Whiskers above and below the box indicate the 90th and 10th percentiles. *p
Vol. 63, No. 22, 1998
AF64A and Striatal Transplants
1959
It is important to note, however, that under certain training conditions (e.g., increased intensity of foot-shock)’ cholinergic, KCI or lidocaine blockade of the striatum has no effect on long-term retention (3), thus suggesting the participation of other brain regions in this process. Indeed, in the present study, although the lesioned groups that had cerebellar or sham transplants were significantly impaired during long-term retention testing, they had significantly higher retention scores than the naive group trained without a foot-shock. Lower retention scores in an inhibitory avoidance task have been reported after large electrolytic striatal lesions (13). There are two possible explanations for this difference in outcome: either the AF64A lesions were not large enough and the remaining portion of the striatum was still functional, or the electrolytic lesions by destroying passing fibers affected the! integrity of a circuit involved in memory consolidation in which other brain areas, such as the subsiantia nigra and amygdala, seem to participate (3). AF64A is considered mainly as a cholinotoxin, although there is evidence that it may have nonspecific effects depending on the dose and mode of preparation (27). In the present study, AF64A produced significant lesions of the striatal parenchyma as evidenced by the ventricular enlargement observed. Adjacent to the cavities, a narrow band of tissue devoid of AChE staining was visible, while the rest of the striatum appeared almost normal, with no evidence of extensive necrosis, just as previously described (6). A similar ventricular enlargement has been observed after lesioning the striatum with other neurotoxins such as ibotenic and kainic acid when delivered into the striatum (I L, 26). Damage to extra-striatal areas was also apparent in some animals, however, the behavioral measures of these animals were not appreciably different from those without extra-striatal damage. The dose of the AF64A used in the present study is similar to that employed by Sandberg et al. (23), in fheir study 8 nmoles per pl were delivered into the striatum and similar behavioral deficits were observed in long-term retention. Along this line, decrements in performance of passive avoidance and radial maze tasks have been reported in other studies in rats, using doses of AF64A ranging frotn 3 to 32 nmoles (27). These studies compared groups of animals with vehicle infUsions to groups with AF64A infusions finding significant differences in retention levels among them. Other studies in our laboratory (4, 5) have shown that implanted rats with vehicle infusions do not significantly differ in their retention scores in inhibitory avoidance tasks from those of intact animals. Thus. in the present study the vehicle-control group was obviated and two intact groups, a positive and a negative control were added. The relevance of striatal involvement in long-term memory was made evident by the fact that only striatal transplants were able to prevent the amnestic effects of the AF64A lesion. Animals with cerebellar and sham transplants showed lower scores than those of the shocked naive (NAWS) group. Similar results have been obtained after intrastriatal electrolytic lesions followed by striatal or mesencephalic transplants. Only striatal transplants reversed the initial memory impairment induced by the electrolytic lesion and such effect was correlated with the acetylcholine synthetic enzyme, choline acefyltransferase (I 3). In contrast, the presence of AChE, which hydrolyzes acetylcholine, appears not to be correlated with behavioral recovery (28). The present results support this notion since both cerebellar and striatal transplants showed AChE positive staining but only striatal transplants resulted in a significant recovery of retention deficits. It has been suggested that functional efticacy of striatal transplants is related to the presence of striatal specific P-zone regions (29) and that this zones are more abundant when the donor tissue is obtained from the lateral ganglionic eminence (30). In the present study, striatal transplant derived from the entire ganglionic Further studies are needed to assess the eminence, prevented AF64A-induced retention deficits. separate contribution of medial versus lateral ganglionic eminence transplants in this behavioral model. Intracerebral fetal transplants are believed to exert their effect through the establishment of afferent and efferent connections with the host brain and through indirect mechanisms such as unregulated release of trophic factors or neurotransmitters (3 I). In the present study, the testing and training sessions were scheduled 6 weeks after the transplant since at this time striatal transplants can be expected to have established afferent and efferent connections with the host brain (19). At the embryonic a.ges selected, cholinergic cells are already present in the developing striatum (32), and since both the striatum and the cerebellum continue to develop postnatally (33, 34), the tissue had potential for continued growth. The fact that only striatal and not cerebellar transplants prevented long-term memory loss, indicates that non-specific factors are insufficient to induce complete The finding that behavioral recovery in a learning task, and that homotopic transplants are required.
1960
AF64A and Striatal Transplants
Vol. 63, No. 22, 1998
volume of transplant did not significantly correlate with retention latencies, suggests that other factors besides volume, such as type of tissue transplanted, are more relevant for functional recovery in this behavioral paradigm. With regard to locomotor behavior, AF64A striatal lesions turned out to be innocuous inasmuch as all lesioned groups (SH-TR, STR, CEREB) showed only a transient increment in this behavior. Other studies (35) have shown significant increases in spontaneous locomotor behavior five weeks after similar lesions, in animals kept on a light/dark cycle; and striatal lesioned animals have been shown to increase their spontaneous activity levels during the dark phase (15). Since the animals used in the present study were kept under continuous light, it is possible that their waking-rest cycle was different and that this fact may be related to the short-lived effect of the lesion on spontaneous activity levels. In conclusion, striatal AF64A lesions do not interfere with short-term memory but disrupt long-term memory of an inhibitory avoidance task, and such deficit is prevented by transplantation of fetal striatal transplants. Acknowledgments
This work was supported by DGAPA-UNAM. We wish to thank Carlos Medina, Leopold0 GonzalezSantos and Rafael Favila Humara for technical support, and the staff at the Animal Laboratory Facility of the Faculty of Medicine, UNAM, for providing the timed-pregnant female and male rats. References
1. 2. 3.
4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
BARTUS X., R.L DEAN., and C. FLICKER Psychopharmacology: The Third Generation Of Progress, H.Y. Meltzer (Ed), 219-232, Raven Press, New York (1987). P.E. PHELPS, C.R. HOUSER and J E. VAUGH, J. Comp. Neurol. 238 286-307 (1985). R.A. PRADO-ALCALA, Plasticity in the central nervous system: Learning and memory, J.L.McGaugh, F. Bermtidez-Rattoni and R.A. Prado-Alcala (Eds.), 57-65, Erlbaum, New Jersey (1995). R.A. PRADO-ALCALA, J. FERNANDEZ-RUIZ, and G.L. QUIRARTE, Synaptic Transmission 2, T.W. Stone (Ed.), 57-69, Taylor and Francis, London (1993). R.A. PRADO-ALCALA, L. SIGNORET-EDWARD, M. FIGUEROA and M.A. BARRIENTOS, Behav. Neural Biol. 42 81-84 (1984). K. SANDBERG, I. HANIN, A. FISHER and J.T.COYLE, Brain Res 293 49-55 (1984). Z. Q. XIONG, Y.F. HAN, and X.C. TANG, NeuroReport 6 2221-2224 (1995). K. SANDBERG, P.R. SANBERG, I. HANIN, A. FISHER and J.T. COYLE, Behav. Neurosci. 98 162-165 (1984). DUNNETT, 0. ISACSON, D.J.S. SIRINATHSINGHJI, D.J. CLARKE and A. BJORKLUND, Neuroscience 24 813-820 (1988). R.A. FRICKER, L.E. ANNETT., E.M. TORRES and S.B. DUNNET, Brain Res. Bull. 41 409416 (1996). M.GIORDANO, S. HAGENMEYER-HOUSER and P.R. SANBERG, Brain Res. 446 183-188 (1988). 0. ISACSON, S.B. DUNNETT and A. BJGRKLUND, P. Natl. Acad. Sci. USA 83 2728-2732 (1986). A.L. PINA, C.E. ORMSBY, M.E. MIRANDA, N. JIMENEZ, R. TAPIA, and F. BERMUDEZRATTONI, J. Neural Transplantation and Plasticity 5 1l-16 (1994). L.L. PUNDT, T. KONDOH, J.A. CONRAD and W.C. LOW, Neurosci. Res. 24 415-420 (1996). P.R. SANBERG, M.A. HENAULT, and A.W. DECKEL, Pharmacol. Biochem and Behav. 25 297-300 (1986). CAMPBELL, P. KALEN, K. WICTORIN, C. LUNDBERG, R.J. MANDEL and A. BJGRKLUND, Neuroscience 53 403-415 ( 1993). D.J CLARKE and S.B DUNNETT,, J Chem Neuroanat. 6 147-58 (1993). A. RUTHERFORD, M. GARCIA-MUNOZ, S.B. DUNNETT and G.W. ARBUTHNOTT, Neurosci. Lett. 83 275-281 (1987). K. WICTORIN, Prog. Neurobiol. 38 611-639 (1992). Z.C. XU, C.J. WILSON and P.C. Emson, Neuroscience 29 539-550 (1989).
Vol. 63, No. 22, 1998
AF64A and Striatal Transplants
l%l
21. Z.C. XU, C.J. WILSON and P.C. Emson, J. Comp. Neural. 303 2-14 (1990). 22. G. PAXINOS and C. WATSON, The Rat Brain in Stereotaxic Coordinates, Academic Press, San Diego ( 1986). 23. S.A. VAN OOTEGHEM and M.T SHIPLEY, Brain Res. Bull. 12 :543-553 (1984). 24. R.P. MlCHEL and L.M. CRUZ-ORIVE, J. Microscopy, 150:117-136 (1988). 25. A.M. GRAYBIEL, F.-C. LB-J and S.B. DUNNETT, J. Neurosci. 9 3250-3271 (1989). Neuroscience 16 799-817 26. 0. ISACSON, P. BRUNDIN, F.H. GAGE and A. BJGRKLUND, (1985). 27. I. HANIN, A. FISHER, H. HGRTNAGL H., S.M. LEVENTER, P.E. POTTER and T.J. WALSH, Psychopharmacology: The Third Generation Of Progress, H.Y. Meltzer (Ed), 341-349 (1987). M. GIORDANO and K.H. 28. P.R. SANBERG, M.A. HENAULT, S.H. HAGENMEYER-HOUSER, RUSSELL, Cell and Tissue Transplantation into the Adult Brain, E.C. Azmitia and A. Bjiirkhmd (Eds.), 781-785, Annals of the New York Academy of Science 495 (1987). N. NAKAO, 0. LINDVALL and P. BRUNDIN, Neuroscience 29. E.M GRASBON-FRODL., 73 :171-183, (1996). 30. T.B FFEEMAN., P.R SANBERG and 0. ISACSON, Cell Transplantation 4 :539-545 (1995). 31. BJGRKLUND, 0. LINDVALL, 0. ISACSON, P. BRUNDIN, K. WICTORIN, R.E STRECKER,., D.J. CLARKE and S.B. DUNNETT. Trends Neurosci. 10 509-516 (1987). 32. U.B. SCHAMBRA, K.K SULIK., P. PETRUSZ and J.M LAUDER, J. Comp. Neurol. 288 : 1Ol122 (1989). 33. M. BEFRY and P. BRADLEY, Brain Res.112 :l-35, (1976). COWAN, Anat. Embryol. 163 : 275-298, 34. J.C. FENTRESS, B.B. STANFIELD and W.M. (1981). 35. K. SANDBERG, P.R. SANBERG and J.T. COYLE, Brain Res. 299 339-343 (1984).
STRIATAL TRANSPLANTS Magda Giordano’,
Life Sciences, Vol. 63, No. 22, pp. 19531%1, 1998 Copyright0 1998 Ekvier Science Inc. Printed in the USA. All rights mewed ocm-3205/98 $19.00 t .oo
PREVENT AF64A-INDUCED
RETENTION DEFICITS
Rigoberto Salado-Castillo’, Manuel Sgnchez-Alavez3, Roberto A. Prado-AlcalB’
and
‘Centro de Neurobiologia, Campus UNAM-UAQ, Juriquilla, Qro. 76230, ‘Departamento de Psicologia, IJniversidad de Panamli, Panam& and ‘Departamento de Fisiologia, Facultad de Medicina, UNAM, Mexico, D. F. 04510, Mexico.
(Received
in final form September 14, 1998) Summary
The relevance of the cholinergic system in mnemonic processes has been repeatedly In addition to the cholinergic systems that project to the demonstrated. telencephalon, there are subcortical nuclei with intrinsic cholinergic cells which appear to be involved in memory consolidation; among these is the striatum. Intrastriatal administration of anticholinergic drugs, as well as excitotoxic and electrolytic lesions have been shown to disrupt the acquisition and retention of instrumentally conditioned behaviors. In the present study male Wistar rats were used to confirm the reported detrimental effects of striatal lesions produced by the cholinotoxin AF64A on long-term retention (LTR) of inhibitory avoidance and spontaneous locomotor activity, to determine its effects on short-term retention (STR) and to investigate whether intrastriatal homotopic transplants can reverse the AF64A-striatal lesions did not interfere with AF64A-induced behavioral deficits. STR but disrupted LTR of the inhibitory avoidance task, and striatal transplants prevented this deficit. Spontaneous locomotor activity increased after the lesion but promptly returned to baseline levels. These results support previous findings showing striatal involvement in long-term but not short-term retention and indicate that homotopic transplants induce behavioral recovery of a learning task in striatal lesioned rats. Key Words: ethylcholine mustard aziridinium, caudate-putamen, inhibitory avoidance, locomotor activity
long-term
retention,
striatal transplants,
The relevance of the cholinergic system in mnemonic processes has been repeatedly demonstrated in a variety of animal models (1). Clinical studies with aged and Alzheimer’s patients have also indicated that loss of cholinergic function may be related to the memory impairments observed in these patients; indeed systemic injections of anticholinergics result in memory loss (1). Cholinergic cells thought to be involved in memory can be found in several brain regions, including intrinsic cells within the caudate-putamen or striatum (2). Lesions of the striatum in rats have been shown to disrupt the acquisition and retention of many conditioned behaviors (3), and intrastriatal injections of anticholinergic drugs impair long-term memory of inhibitory avoidance (4) but leave short-term memory intact (5). Consistent with these data are results showing that choline administered to the striatum has a facilitating effect on positively and. negatively motivated conditioned behaviors (4) and that intrastriatal infusion of AF64A, an aziridinium analog of choline whose neurotoxic effects on cholinergic neurons is well documented (6, 7), impairs acquisition and retention of inhibitory avoidance conditioning (8). Send correspondence to: Magda Giordano, Ph.D., Centro de Neurobiologia, P.O. Box 70228, UNAM, 04510 M6xico’ D.F., Mexico, Telephone: (525) 623-4061, 623-4005, Fax: (524) 234-03-44, E-mail [email protected].
1054
AF64A and Striatal Transplants
Vol. 63, No. 22, 1998
Striatal transplants reverse behavioral deficits induced after excitotoxic or electrolytic lesions of the striatum (9-15). This compensatory effect is probably mediated by some of the many functional which include recovery of transmitter release and of consequences of the transplant electrophysiological activity, establishment of afferent and efferent connections and of synaptic relationships with the host brain (I 6-2 I ). The purpose of the present study was to test the reliability of the reported detrimental effects of striatal lesions produced by AF64A on retention of inhibitory avoidance and on spontaneous locomotor activity, to explore whether this lesion interferes with short-term memory, and to determine whether intrastriatal homotopic transplants reverse AF64A-induced behavioral deficits. Methods Animals. Naive male Wistar rats (200-300 g) were maintained in individual cages under continuous light and with free access to food and water. They were kept in these conditions at least live days before the experiments were initiated and were randomly assigned to one of five groups, three of which were lesioned and tested in the inhibitory avoidance task and whose locomotor activity was recorded on three occasions, and two naive groups which served as controls for the inhibitory avoidance task. One of the naive groups did not receive a foot-shock during avoidance training (NAWO), while the other naive group was trained as all the experimental animals (NAWS). Locomotor activity of some of the naive animals was not recorded. Apparafus. Black acrylic locomotor activity chambers (43 x 43 x 25.5 cm) with sixteen infrared sensors on the horizontal plane were used to record spontaneous locomotor activity for I5 min. Activity was recorded as counts per minute. The inhibitory avoidance chamber was a wooden box with two compartments of the same size (30 x 30 x 30 cm each), separated by a guillotine door. The lid of each compartment and the guillotine door were made of orange-colored lucite. The floor of one of the compartments was a grid made of aluminum bars (6 mm in diameter), separated I.5 cm center to center. The V-shaped lateral walls of the second compartment were stainless steel, and each was continuous with half the floor; there was a 1.5 cm slot separating each half-floor. Thus, when in this compartment, the rats were in contact with both walls and floor, which could be electrified using a square-pulse stimulator (Grass Instrument Co., model S-44) connected in series with a constant current unit (Grass Instrument Co., model CCUI). Illumination was provided by a IO-W light bulb located in the center of the lid of the compartment with the grid. Foot-shock delivery and measurement of latencies to cross from one compartment to the other one were accomplished by use of automated equipment. The conditioning box was located inside a dark, sound-deadening room, provided with background masking noise (BSR/LVE, model AU-902). Training and testing. On the day of training each animal was put inside the safe compartment and IO set later the door dividing the two compartments was opened and the latency to cross to the darker (electrifiable) compartment with all four paws was measured (acquisition latency). When the animals had crossed to this compartment the guillotine door was closed and a I mA foot-shock was delivered through the stainless steel plates. After five set the door was reopened, thus allowing the animal to escape to the first compartment and to remain there for 30 set before being put back in its home cage. The escape latency was also recorded; upon the rat’s escape the door was closed and the footshock was turned-off. Thirty minutes and twenty-four h later tests of retention were programmed exactly as the training session, except that the foot-shock was not delivered. If a rat did not cross within 600 set to the compartment where the foot-shock had been given the session was ended and a score of 600 was assigned. Opening and closing of the guillotine door, which was manually operated, activated the programming equipment that controlled shock delivery and timing of latencies. Lesion surgery: Animals were anesthetized with ketamine (50 mg/ml/kg i.p.) and received atropine sulfate (I ml/kg i.p.) and penicillin (1,200,OOO UI, i.m.). They were placed on a stereotaxic frame, an incision made and the skull exposed. Using an automatic infusion pump 1 pl of AF64A was delivered bilaterally over 5 minutes through a 27 gauge stainless steel injector whose tip was aimed at the dorsal striatum (I.5 mm anterior to bregma, +2.6 mm lateral and 4.5 mm ventral from dttt-a; incisor bar at -2.5 mm) (22); after an additional minute the injection needle, which was positioned perpendicular to the skull, was retracted, the wound sutured and the animals returned to their home cages. AF64A was activated according to the manufacturer’s specifications (RBI, Natlick MA,
Vol. 63, No. 22, 1998
AF64A and Striatal Transplants
1955
method I). Briefly, after dissolving the compound in distilled water, the pH was brought to 11.7 and maintained for 30 min at room temperature. Then the pH was lowered to 7.0 with concentrated HCI and adjusted to 7.4 with solid NaI-ICOS. The solution was used immediately while kept on ice, or it was frozen immediately and placed in a -8O’C freezer; final concentration was I I nmoles/pl. Animals were allowed fifteen days to recover after lesion surgery. Transplant surges: Following the same general procedure as used for lesion surgery, blocks of striatal or cerebellar fetal tissue (4 ~1) were bilaterally delivered into the same lesioned striatal site. Tissue was obtained from fetuses of 17 to 19 days of gestation (average CRL=I 9fO.5 mm). The brain was placed on its ventral surface, the cortex was removed and peeled laterally to expose the striatal primordium, the entire striatal primordium (medial and lateral ganglionic eminences) was clipped off and sectioned into tissue blocks. With the brain on the same position, the cerebellar anlagen located immediately underneath the colliculi, were cut and divided into tissue blocks. AAer dissection the tissue blocks were aspirated using a Hamilton syringe (50 ~1) fitted to a glass capillary attached to a standard needle, the syringe was then placed on the stereotaxic frame and the fetal tissue delivered into the lesi(aned striatum. Procedure: Before recording baseline activity levels, all animals were habituated to the activity chambers for 15 min on two non-consecutive days. After baseline activity levels were recorded, experimental! animals were lesioned with AF64A. Fifteen days later, locomotor activity was recorded again and 2 days later lesioned animals received intrastriatal transplants of striatal tissue (STR), cerebellar tissue (CEREB) or vehicle only (SH-TR). Six weeks after transplantation (60 days after lesion), their locomotor activity was recorded for a third time and 5-7 days afterwards the animals were trained and tested in inhibitory avoidance. All these behavioral variables were also measured in naive rats which belonged to one of two groups, a naive group trained with foot-shock (NAWS) or a naive group trained without foot-shock (NAWO). Following the last retention test, all lesioned and a few naive animals were anesthetized and intracardially perfused. Histology: .4Rer behavioral tests were completed, all lesioned animals, were intracardially perfused with 4% paraformaldehyde in PBA 0.1 M, their brains removed and kept in 20% sucrose until sectioned in a freezing microtome. Alternate sections (40 urn) were processed for Nissl staining and acetylcholinesterase (AChE) histochemistry. The protocol for AChE histochemistry was based on that of Van Ooteghem and Shipley (23). To make volume determinations transplants were measured using Cavalieri’s principle (24). Briefly, areas were measured by means of an image analysis system (IP-LAB) which corrected for the magnification used so that results were expressed in pm2, areas for each transplant were added up, multiplied by the distance between sections and converted to mm’. Statistical analysis: Locomotor activity data were analyzed using a two-way analysis of variance. Inhibitory avoidance data were analyzed using the Kruskal-Wallis analysis of variance followed by comparisons between the naive foot-shocked group and each of the other groups using MannWhitney U tests. Transplant volume for striatal and cerebellar grafts were compared using a t-test for independent samples, scores during the 24-hour retention test and transplant volumes were correlated using Pearson’s correlation coefficient. Results AF64A striatal lesions produced significant enlargement of the lateral Histological analysis. ventricles, reactive gliosis along the cannula track, and decreased AChE staining in the area immediately adjacent to the ventricles; no evidence of necrosis was evident in the remaining striatal tissue. In some animals (56.5%) lesions included portions of the anterior pole of the frontal cortex, while in a few others (17.4%) the caudal portion of the lateral ventricles appeared somewhat enlarged and/or septal damage was apparent (Fig. I). Striatal and cerebellar transplants were visible in all cases, were located bilaterally within the striatal parenchyma and in some cases they extended into the cortex, and /or the lateral ventricles. Size and location of the transplants varied, generally striatal transplants were larger than cerebellar ones, although some relatively large cerebellar transplants we;e found (Fig. I). No significant differences between average volume of striatal (4.69f3.45 mm ; minimum 1.75, maximum 10.85 mm’) and cerebellar (1.94f2.07 mm3; minimum 0.19, maximum 5.07 mm’) transplants were found. Both cerebellar and striatal transplants were positive for AChE staining, and striatal transplants showed the characteristic patchy appearance described previously
1956
AF64A and Striatal Transplants
Vol. 63, No. 22, 1998
(25, 26). All lesioned animals with evidence of bilateral lesions and all transplanted animals were used for analysis of behavioral data.
:’
.:
-.
Fig. 1 Photomicrographs of coronal sections through the striatum stained for AChE (a, c, e) and Nissl staining (b, d, f). Representative examples of AF64-A lesions restricted to the striatum (a) and including portions of the overlaying cortex (b). Appearance of one of the largest and best integrated cerebellar transplant (c, d). Also shown are two representative examples of striatal transplants (e) and (f). Note patchy appearance of AChE staining within transplanted tissue. S, lesioned striatum; T, transplant. Bar=O.lmm.
Vol. 63, No. 22, 1998
AF64A and Striatal Transplants
1957
Spontaneous locomotor activity. There was a significant session effect on spontaneous locomotor activity (F[2,126]=4.24, ~~0.05). Post hoc comparisons (Tukey’s HSD test) revealed that for all lesioned groups there was a significant increase in locomotor activity after the lesion (p
3 5
12000
g
10000
0
I F
8000
; 0
6000
5 5
4000
a
: i 2 4 z
2000
0 BASELINE
POST-TRANSPLANT
POST-LESION
Fig. 2 Mean r.pontaneous locomotor activity f SEM of AF64A-lesioned and naive animals. All lesioned groups showed increased activity during the post-lesion session when compared to baseline. This increase returned to baseline on the post-transplant session. See text for details. Bar fills are as follows: Striatal transplants, empty bars (n=14); cerebellar transplants, left-hatched bars (n=14); sham-transplants, right-hatched bars (n=23); naive group trained with foot-shock during inhibitory avoidance training (n=lO), cross-hatched bars; naive group trained without foot-shock during inhibitory avoidance training, horizontal lined bars (n=lO). *p
AF64A and Striatai Transplants
1958
Vol. 63, No. 22, 1998
700 600 500 -
c ii -
400 -
B d
200 -
ts
100 -
300 -
#
O-
w
a
-_
b
p z
700
P Y
600 -
z
500
g
400
g
300 200 100 O-
STR CEREB
SH-TR
NAWS
NAWO
Fig. 3 Median retention scores for AF64A-lesioned and naive animals 30 minutes (a) and 24 hours (b) after training on an inhibitory avoidance task. Abbreviations are as follows: STR, striatal transplants (n=l4); CEREB, cerebellar transplants (n=14); SH-TR, sham transplants (n=23); NAWS, naive group trained with foot-shock during inhibitory avoidance training (n=20); NAWO, naive group trained without foot-shock during inhibitory avoidance training (n=l9). The boundary of the box closest to zero indicates the 25’h percentile, a line within the box marks the median, and the boundary of the box farthest from zero indicates the 75’h percentile. Whiskers above and below the box indicate the 90th and 10th percentiles. *p
Vol. 63, No. 22, 1998
AF64A and Striatal Transplants
1959
It is important to note, however, that under certain training conditions (e.g., increased intensity of foot-shock)’ cholinergic, KCI or lidocaine blockade of the striatum has no effect on long-term retention (3), thus suggesting the participation of other brain regions in this process. Indeed, in the present study, although the lesioned groups that had cerebellar or sham transplants were significantly impaired during long-term retention testing, they had significantly higher retention scores than the naive group trained without a foot-shock. Lower retention scores in an inhibitory avoidance task have been reported after large electrolytic striatal lesions (13). There are two possible explanations for this difference in outcome: either the AF64A lesions were not large enough and the remaining portion of the striatum was still functional, or the electrolytic lesions by destroying passing fibers affected the! integrity of a circuit involved in memory consolidation in which other brain areas, such as the subsiantia nigra and amygdala, seem to participate (3). AF64A is considered mainly as a cholinotoxin, although there is evidence that it may have nonspecific effects depending on the dose and mode of preparation (27). In the present study, AF64A produced significant lesions of the striatal parenchyma as evidenced by the ventricular enlargement observed. Adjacent to the cavities, a narrow band of tissue devoid of AChE staining was visible, while the rest of the striatum appeared almost normal, with no evidence of extensive necrosis, just as previously described (6). A similar ventricular enlargement has been observed after lesioning the striatum with other neurotoxins such as ibotenic and kainic acid when delivered into the striatum (I L, 26). Damage to extra-striatal areas was also apparent in some animals, however, the behavioral measures of these animals were not appreciably different from those without extra-striatal damage. The dose of the AF64A used in the present study is similar to that employed by Sandberg et al. (23), in fheir study 8 nmoles per pl were delivered into the striatum and similar behavioral deficits were observed in long-term retention. Along this line, decrements in performance of passive avoidance and radial maze tasks have been reported in other studies in rats, using doses of AF64A ranging frotn 3 to 32 nmoles (27). These studies compared groups of animals with vehicle infUsions to groups with AF64A infusions finding significant differences in retention levels among them. Other studies in our laboratory (4, 5) have shown that implanted rats with vehicle infusions do not significantly differ in their retention scores in inhibitory avoidance tasks from those of intact animals. Thus. in the present study the vehicle-control group was obviated and two intact groups, a positive and a negative control were added. The relevance of striatal involvement in long-term memory was made evident by the fact that only striatal transplants were able to prevent the amnestic effects of the AF64A lesion. Animals with cerebellar and sham transplants showed lower scores than those of the shocked naive (NAWS) group. Similar results have been obtained after intrastriatal electrolytic lesions followed by striatal or mesencephalic transplants. Only striatal transplants reversed the initial memory impairment induced by the electrolytic lesion and such effect was correlated with the acetylcholine synthetic enzyme, choline acefyltransferase (I 3). In contrast, the presence of AChE, which hydrolyzes acetylcholine, appears not to be correlated with behavioral recovery (28). The present results support this notion since both cerebellar and striatal transplants showed AChE positive staining but only striatal transplants resulted in a significant recovery of retention deficits. It has been suggested that functional efticacy of striatal transplants is related to the presence of striatal specific P-zone regions (29) and that this zones are more abundant when the donor tissue is obtained from the lateral ganglionic eminence (30). In the present study, striatal transplant derived from the entire ganglionic Further studies are needed to assess the eminence, prevented AF64A-induced retention deficits. separate contribution of medial versus lateral ganglionic eminence transplants in this behavioral model. Intracerebral fetal transplants are believed to exert their effect through the establishment of afferent and efferent connections with the host brain and through indirect mechanisms such as unregulated release of trophic factors or neurotransmitters (3 I). In the present study, the testing and training sessions were scheduled 6 weeks after the transplant since at this time striatal transplants can be expected to have established afferent and efferent connections with the host brain (19). At the embryonic a.ges selected, cholinergic cells are already present in the developing striatum (32), and since both the striatum and the cerebellum continue to develop postnatally (33, 34), the tissue had potential for continued growth. The fact that only striatal and not cerebellar transplants prevented long-term memory loss, indicates that non-specific factors are insufficient to induce complete The finding that behavioral recovery in a learning task, and that homotopic transplants are required.
1960
AF64A and Striatal Transplants
Vol. 63, No. 22, 1998
volume of transplant did not significantly correlate with retention latencies, suggests that other factors besides volume, such as type of tissue transplanted, are more relevant for functional recovery in this behavioral paradigm. With regard to locomotor behavior, AF64A striatal lesions turned out to be innocuous inasmuch as all lesioned groups (SH-TR, STR, CEREB) showed only a transient increment in this behavior. Other studies (35) have shown significant increases in spontaneous locomotor behavior five weeks after similar lesions, in animals kept on a light/dark cycle; and striatal lesioned animals have been shown to increase their spontaneous activity levels during the dark phase (15). Since the animals used in the present study were kept under continuous light, it is possible that their waking-rest cycle was different and that this fact may be related to the short-lived effect of the lesion on spontaneous activity levels. In conclusion, striatal AF64A lesions do not interfere with short-term memory but disrupt long-term memory of an inhibitory avoidance task, and such deficit is prevented by transplantation of fetal striatal transplants. Acknowledgments
This work was supported by DGAPA-UNAM. We wish to thank Carlos Medina, Leopold0 GonzalezSantos and Rafael Favila Humara for technical support, and the staff at the Animal Laboratory Facility of the Faculty of Medicine, UNAM, for providing the timed-pregnant female and male rats. References
1. 2. 3.
4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20.
BARTUS X., R.L DEAN., and C. FLICKER Psychopharmacology: The Third Generation Of Progress, H.Y. Meltzer (Ed), 219-232, Raven Press, New York (1987). P.E. PHELPS, C.R. HOUSER and J E. VAUGH, J. Comp. Neurol. 238 286-307 (1985). R.A. PRADO-ALCALA, Plasticity in the central nervous system: Learning and memory, J.L.McGaugh, F. Bermtidez-Rattoni and R.A. Prado-Alcala (Eds.), 57-65, Erlbaum, New Jersey (1995). R.A. PRADO-ALCALA, J. FERNANDEZ-RUIZ, and G.L. QUIRARTE, Synaptic Transmission 2, T.W. Stone (Ed.), 57-69, Taylor and Francis, London (1993). R.A. PRADO-ALCALA, L. SIGNORET-EDWARD, M. FIGUEROA and M.A. BARRIENTOS, Behav. Neural Biol. 42 81-84 (1984). K. SANDBERG, I. HANIN, A. FISHER and J.T.COYLE, Brain Res 293 49-55 (1984). Z. Q. XIONG, Y.F. HAN, and X.C. TANG, NeuroReport 6 2221-2224 (1995). K. SANDBERG, P.R. SANBERG, I. HANIN, A. FISHER and J.T. COYLE, Behav. Neurosci. 98 162-165 (1984). DUNNETT, 0. ISACSON, D.J.S. SIRINATHSINGHJI, D.J. CLARKE and A. BJORKLUND, Neuroscience 24 813-820 (1988). R.A. FRICKER, L.E. ANNETT., E.M. TORRES and S.B. DUNNET, Brain Res. Bull. 41 409416 (1996). M.GIORDANO, S. HAGENMEYER-HOUSER and P.R. SANBERG, Brain Res. 446 183-188 (1988). 0. ISACSON, S.B. DUNNETT and A. BJGRKLUND, P. Natl. Acad. Sci. USA 83 2728-2732 (1986). A.L. PINA, C.E. ORMSBY, M.E. MIRANDA, N. JIMENEZ, R. TAPIA, and F. BERMUDEZRATTONI, J. Neural Transplantation and Plasticity 5 1l-16 (1994). L.L. PUNDT, T. KONDOH, J.A. CONRAD and W.C. LOW, Neurosci. Res. 24 415-420 (1996). P.R. SANBERG, M.A. HENAULT, and A.W. DECKEL, Pharmacol. Biochem and Behav. 25 297-300 (1986). CAMPBELL, P. KALEN, K. WICTORIN, C. LUNDBERG, R.J. MANDEL and A. BJGRKLUND, Neuroscience 53 403-415 ( 1993). D.J CLARKE and S.B DUNNETT,, J Chem Neuroanat. 6 147-58 (1993). A. RUTHERFORD, M. GARCIA-MUNOZ, S.B. DUNNETT and G.W. ARBUTHNOTT, Neurosci. Lett. 83 275-281 (1987). K. WICTORIN, Prog. Neurobiol. 38 611-639 (1992). Z.C. XU, C.J. WILSON and P.C. Emson, Neuroscience 29 539-550 (1989).
Vol. 63, No. 22, 1998
AF64A and Striatal Transplants
l%l
21. Z.C. XU, C.J. WILSON and P.C. Emson, J. Comp. Neural. 303 2-14 (1990). 22. G. PAXINOS and C. WATSON, The Rat Brain in Stereotaxic Coordinates, Academic Press, San Diego ( 1986). 23. S.A. VAN OOTEGHEM and M.T SHIPLEY, Brain Res. Bull. 12 :543-553 (1984). 24. R.P. MlCHEL and L.M. CRUZ-ORIVE, J. Microscopy, 150:117-136 (1988). 25. A.M. GRAYBIEL, F.-C. LB-J and S.B. DUNNETT, J. Neurosci. 9 3250-3271 (1989). Neuroscience 16 799-817 26. 0. ISACSON, P. BRUNDIN, F.H. GAGE and A. BJGRKLUND, (1985). 27. I. HANIN, A. FISHER, H. HGRTNAGL H., S.M. LEVENTER, P.E. POTTER and T.J. WALSH, Psychopharmacology: The Third Generation Of Progress, H.Y. Meltzer (Ed), 341-349 (1987). M. GIORDANO and K.H. 28. P.R. SANBERG, M.A. HENAULT, S.H. HAGENMEYER-HOUSER, RUSSELL, Cell and Tissue Transplantation into the Adult Brain, E.C. Azmitia and A. Bjiirkhmd (Eds.), 781-785, Annals of the New York Academy of Science 495 (1987). N. NAKAO, 0. LINDVALL and P. BRUNDIN, Neuroscience 29. E.M GRASBON-FRODL., 73 :171-183, (1996). 30. T.B FFEEMAN., P.R SANBERG and 0. ISACSON, Cell Transplantation 4 :539-545 (1995). 31. BJGRKLUND, 0. LINDVALL, 0. ISACSON, P. BRUNDIN, K. WICTORIN, R.E STRECKER,., D.J. CLARKE and S.B. DUNNETT. Trends Neurosci. 10 509-516 (1987). 32. U.B. SCHAMBRA, K.K SULIK., P. PETRUSZ and J.M LAUDER, J. Comp. Neurol. 288 : 1Ol122 (1989). 33. M. BEFRY and P. BRADLEY, Brain Res.112 :l-35, (1976). COWAN, Anat. Embryol. 163 : 275-298, 34. J.C. FENTRESS, B.B. STANFIELD and W.M. (1981). 35. K. SANDBERG, P.R. SANBERG and J.T. COYLE, Brain Res. 299 339-343 (1984).