Dopaminergic drugs reverse the impairment of radial-arm maze performance caused by lesions involving the cholinergic medial pathway

0306-4522/92 $5.00+ 0.00 PergamonPressLtd fc‘ 1992IBRO

Neuroscience Vol. 50,No. I, pp. 129-135,1992 Printedin Great Britain

DOPAMINERGIC DRUGS REVERSE THE IMPAIRMENT OF RADIAL-ARM MAZE PERFORMANCE CAUSED BY LESIONS INVOLVING THE CHOLINERGIC MEDIAL PATHWAY S. R. MCGURK,* tBox

E. D. LEVINPS and L. L. BUTCHER*

*Department of Psychology, University of California, Los Angeles, CA 90024, U.S.A. 3557, Department of Psychiatry, Duke University Medical Center, Durham, NC 27710, U.S.A

Abstract-Pharmacological studies have shown that both cholinergic and dopaminergic transmitter systems are crucial for optimal choice accuracy in the radial-arm maze and that these systems interact in a complex fashion. Lesion studies have provided evidence that the basal nuclear complex of the forebrain, the origin of cholinergic projections to the cerebral mantle, may be critical for the cholinergic modulation of learning and memory. We have shown that knife-cut lesions of the medial cholinergic pathway significantly impair radial-arm maze choice accuracy performance. The current study examined the effectiveness of D, and D, ligands in counteracting this lesion-induced deficit. The adverse effects of medial cholinergic pathway lesions were diminished or reversed by daily treatment with a D, agonist (SKF 38393), a D, agonist (LY 171555) or a D, antagonist (SCH 23390), but were not affected by treatment with a D, antagonist (raclopride). The three beneficial treatments have previously been found to attenuate the adverse effects of nictonic or muscarinic blockade on choice accuracy performance in the radial-arm maze. The finding that these dopaminergic drugs ameliorate the memory deficit caused by lesions involving the cholinergic medial pathway suggests the importance of interactions between cholinergic and dopaminergic systems in radial-arm maze performance. These results may provide leads for the development of novel therapeutic approaches for treating human disorders thought to result from cholinergic hypofunction.

The basal nuclear complex (comprised of the medial septal nucleus, diagonal band nuclei, magnocellular preoptic field, and nucleus basalis), has received considerable attention in the examination of central cholinergic involvement in learning and memory. Much of this interest has been spurred by the association of pathological changes in these cholinergic neurons with the disruption of memory in Alzheimer’s disease.“,‘6,20,47,5’ Lesions of this region significantly impair learning and memory in a variety of tasks, including the radial-arm

effects of relatively large lesions of the non-septal basal nuclear complex, but they have not attempted to discriminate between the roles of the medial and lateral projections. Our initial investigations demonstrated that the medial pathway is important for the working memory function as measured by choice accuracy in the radial-arm maze.33 Knife-cut lesions of the medial cortex that include the medial cholinergic pathway produce pronounced, transient deficits in radial-arm maze choice accuracy, which can be reversed by the cholinergic agonist arecoline. Others have reported that the memory-disrupting effects of lesions of the nucleus basalis itself can, under some circumstances, be reversed by cholinergic agonists.‘3,‘4.‘7,36 Dopaminergic and cholinergic systems interact in a complex manner with regard to cognitive function.2” For example, we found that radial-arm maze choice accuracy deficits caused by administration of cholinergic antagonist drugs can be reduced by simultaneous treatment with dopaminergic drugs.24.25.27.3”.3’ The present study was conducted to examine the interaction of dopaminergic systems with the effect of lesions of the medial cholinergic pathway on cognitive performance. Radial-arm maze behavior was chosen for this study because it is sensitive to a variety of pharmacological manipulations, including those affecting cholinergic and dopaminergic systems.23,26

maze,2.3.12,18,19.21.28,34,36.42.46

There are three major projections of cholinergic neurons in the basal nuclear complex: the septohippocampal pathway, a medial pathway innervating the cingulate, retrosplenial and medial occipital and frontal cortices, and a lateral pathway innervating the remaining lateral neocortex.4,9*22.43.4a50The importance of the septohippocampal pathway in cognitive function has been extensively studied. Several behavioral studies by others have examined the $To whom correspondence should be addressed. Ahbreuiarions: AChE, acetylcholinesterase; CY-208-243, (-)-4,6,6a,7,8,12b-hexahydro-7-methyl-indolo[4,3-ab]phenanthidine; LY 171555, 4,4a,5,6,7,8,8a,9_octahydro5-n-propyl-2H-pyrazolo-3,4,-g-glinoline; SCH 23390, (K)-( +)-8-chloro-2,3,4,5-tetrahydro-3-methyl-5-phenylIH-3-benzazepin-7-01; SKF 38393, 2,3,4,5_tetrahydro7.8-dihydroxy-I-phenyl-IH-3-benzazepines. 129

130

S.K.MCGURK r~ al EXPERIMENTAL

PROCEDURES

Animals

Female Sprague-Dawley rats (Bantin and Kingman Co., Fremont, CA, U.S.A.) housed individually, on a reverse 12 h light: dark cycle, were used. All behavioral testing was conducted during the dark phase (6.00-18.00). Rats had access to water continuously and were kept on a restricted feeding schedule to maintain their body weights at 80--85% of free-feeding levels. adjusted for growth. Behavioral apparatus and procedures

The eight-arm maze used in this study was similar to the one originally developed by Olton and coworkers.37 It was constructed of wood and elevated 30 cm from the floor. The arms (80 cm long x 10 cm wide) extended from an octagonal central platform (35 cm across). Food cups (1 cm deep) were located midway across the arm 2cm from the end. The testing room contained many extra-maze cues and was dimly lit during testing. Rats were tested once per day, five days per week. At the beginning of each session, all maze arms were baited with food reinforcement (Kellogg’s Froot Loop). The rat was placed in the center octagon and was allowed to roam freely. An arm choice was recorded when a rat placed all four paws into an arm. The maze arms were not rebaited during a session so that only the first entry into an arm was rewarded. Re-entries into an arm during a testing session were counted as errors. The session continued until the rat entered all eight arms or 5 min had elapsed. The subjects underwent at least 20 sessions of training before surgery. At that time, they were performing at asymptotic levels of choice accuracy. Behavioral testing for all groups resumed two days post-surgery and continued for ten days. Data recorded included a measure of choice accuracy (entries to repeat; i.e. number of correct entries before the first error), and a measure of response duration (seconds per entry; i.e. total number of arm entries divided by the time to complete the session). Surgical procedures

Stereotaxic surgery was performed on rats anesthetized with sodium pentobarbital (55 mgjkg, ip.). After the skull was exposed, a trough approximately 2.0 x 6.0mm was drilled slightly anterior to bregma, parallel to the interaural line. Lesions of the medial cholinergic pathway were made by lowering a thin surgical knife fixed to the stereotaxic apparatus through this opening 3.5 mm into the brain at approximately 1.Omm anterior to bregma and moving the knife across the brain, parallel to the interaural line, approximately 2.0mm lateral from the midline in both directions. As described and illustrated in Figs 2 and 5 in McGurk ef al.,33 this knife-cut completely severed the medial cholinergic pathway, as well as other fiber bundles transversing the same region. The incision was then closed and animals were returned to their home cages. Animals subjected to control sham surgery were treated identically to the lesioned groups but did not receive any damage to neural tissue (see McGurk et al.“). Drug administration

All drugs were dissolved in 0.9% saline and administered (i.p.) 15-20 min before the beginning of the maze testing session. All injections were made in a volume of 1 ml/kg body weight. Drugs used were: SKF 38393 (3.0mg/kg, Smith, Kline, and French, Philadelphia, PA), a D, dopaminergic agonist; LY 171555 (0.05 mg/kg, Eli Lilly, Indianap&s, IN), a D, dopaminer‘gic agonist; SCH 23590 maleate (0.05 me/kn. Scherinn. Bloomtield. NJ), a D, dopaminergic antago&trand rat&pride (0.5 x&/kg, A&a, -SiidertaGe, Sweden), a D, dopaminergic antagonist. All drug doses are expressed in terms of the salt. The doses of the dopaminergic drugs were selected on the basis of previous experiments and

determined to be the highest doses which did not cause excessive balking (non-responsiveness) in the radial-arm maze.24.25.32 Rats (n = 52) were trained on the maze and subjected to either lesions of the medial cholinergic pathway (n = 43) or sham surgery (n = 9). After surgery, lesioned animals were randomly assigned to one of five drug treatment conditions: SKF 38393 (n = 8). LY 171555 (n = 9). SCH 23390 (n = 9), raclopride (n = 8), or saline (n = 9). The control rats (n = 9) received only saline treatment, Histological procedures

At the conclusion of the behavioral portion of the experiment the animals were killed, perfused intracardially with a 10% formalin solution, and their brains removed. The brains were sectioned in a horizontal plane, mounted on glass slides, and processed for acetylcholinesterase (AChE) as detailed in Butcher et ~1.~ The extent of the lesions was assessed for each animal. Histological analyses revealed that lesions of the medial cholinergic pathway, but not control sham lesions, caused a marked depletion of cortical AChE caudal to the lesion (see McGurk ef al.“). Data analysis

Choice accuracy (entries to repeat) and choice latency (seconds per entry) data were evaluated by analysis of variance for repeated measures. Significant group x session interactions were followed-up by tests of the simple main effects of each group at each session. Dunnett’s test (twotailed) was used for specific comparisons between the untreated lesioned group and the other groups. RESULTS

The effects of dopamine agonists on choice accuracy are depicted in Figs 1 and 2. Overall analysis of variance for entries to repeat revealed significant effects for experimental group treatment [F(5,46) = 4.47, P < O.OOS],test session [F(9,414) = 5.85, P < O.OOOOl],and the interaction of group x session [F(45,414) = 2.04, P < O.OOl]. Significant group effects were found for sessions 1M (P < 0.025). Post hoc Dunnett’s tests were performed at each session comparing the performance of the lesioned group with that of the control and drug treated groups. Performance of the untreated lesion group was significantly worse than control for sessions l-5 (P < 0.05). Thereafter, no discernible lesion-related effect on choice accuracy was detected. The performance of the lesioned group treated with the D, agonist SKF 38393 was significantly better than the untreated lesioned group for sessions l-5 (P < 0.05) as shown in Fig. 1. The laioned group treated with the D, antagonist SCH 23390, however, was significantly better than the untreated lesioned group only on the second session (P < 0.05; Fig. 2). By contrast, performance of the lesioned group treated with the D, agonist LY 171555 was significantly better than the untreated lesioned group on the first and fourth sessions (P -c0.05). There was no discernible difference between the lesioned group treated with the D2 antagonist raclopride and the untreated lesioned group. The results of analysis of the effects of the dopaminergic drug treatments when performance for sessions l-5 are combined are shown in Table 1.

Dopaminergic

drugs

and cholinergic

medial

131

pathway

8 7

-

Control

-

Lesion

Medial Cortex

1

/

I

I

I

I

I

11

11

Lesion

-

SKF 38393

-

LY 171555

+

1

9

12345678

10

Trial Fig. 1. Effects of dopaminergic agonists on choice accuracy performance (entries to repeat). The lesioned group treated with SKF 38393, a D, agonist, was significantly more accurate than the untreated lesioned group on sessions 1-5 (P < 0.05). The lesioned group treated with LY 171555 (quinpirole). a D, agonist, was significantly

more accurate

on the first and fourth

sessions

(P < 0.05).

of simple main effects showed significant group effects for each session (Figs 3, 4). The lesion alone did not significantly affect latency scores and treatment of lesioned animals with SKF 38393 caused a significant (P < 0.05) increase in latency only on session 9. By contrast, treatment with LY 171555 significantly increased latency on sessions 2-10 and SCH 23390 treatment increased latency on sessions 1 and 7 (Figs 3, 4). Raclopride treatment increased latency on sessions 2, 3, and 5510 (Figs 3, 4). Response duration data collapsed over all IO postlesion behavioral sessions are displayed in Table 2. Analyses of these data, using Dunnett’s tests, showed significant increases in latency with SCH 23390, LY 171555 and raclopride.

These data were collapsed because this is the time period during which there was a discernable effect on the lesion alone on behavior (Table 1). For this analysis, there was a significant overall group effect [F(5,46) = 9.37, P < O.OOl], and Dunnett’s test showed a significant lesion-induced deficit. The lesion-induced deficit was significantly attenuated by SCH 23390, LY 171555 and SKF 38393. Raclopride treatment, however, was not effective in reversing the lesion-induced deficit (Table 1). Analysis of variance of the response duration data revealed a significant effect for experimental group [F(5,46) = 12.78, P < O.OOl], test session [F(9,414) = 15.34, P < O.OOl], and group x session interaction [F(45,414) = 2.22, P < O.OOl]. Follow-up

tests

Control Lesion Medial -

Cortex

Lesion + SCH-23390 Raclopride

11 1

2

3

4

5

6

7

8

9

10

Trial Fig. 2. Effects of dopaminergic antagonists on choice accuracy performance (entries to repeat). The lesioned group treated with SCH 23390, a D, antagonist, was significantly more accurate than the untreated lesioned group on session 2 (P < 0.05). No significant differences were seen between the untreated lesioned group and the lesioned group treated with raclopride, a D, antagonist.

132

S. R. MCGURK et al.

Table I. Entries to repeat: sessions l--5 Group

Entries to repeat

Significance vs lesion

Control Lesion only Lesion + SKF 38393 Lesion + SCH 23390 Lesion + LY 171555 Lesion + raclopride

6.56 _+0.27 4.33 * 0.31 6.62 & 0.24 5.64 + 0.36 6.07 + 0.53 4.88 + 0.22

P < 0.01

anticholinergic-induced working memory detici ts in the radial-arm maze. We have previously shown that the D, agonist LY 171555, but not the D1 blocker raclopride, was effective in attenuating the choice accuracy deficit caused by nicotinic blockade with mecamylamine.25~32 We have also found that both the D, agonist SKF 38393 and the D, antagonist SCH 23390 were effective in reversing the choice accuracy impairment caused by muscarinic blockade with scopolamine.24~2’ Although the knife-cut lesions made in the current experiment interrupts cholinergic innervation of the cortex and results in persistent lowering of cortical acetylcholinesterase activity,33 it is also the case that other pathways are interrupted by this procedure as well. It is not known, however, if the D, and D2 interactions with the lesions produced in the present study directly involve the pathway actually interrupted or perhaps more likely operate at other loci to indirectly modify the effect of medial cortex knifecuts. Nonetheless, they are remarkably similar to the interactions of the two dopamine systems with muscarinic or nicotinic blockade.24,27 Possible actions at non-lesioned sites include the following. First, SCH 23390 is known to have antagonistic effects at 5-HT, as well as D, receptors.’ This may contribute to its efficacy. Serotonin inhibits acetylcholine release in the hippocampus, an effect which is reversed by 5-hydroxytryptamine blockade.?’ SKF 38393 may have acted in a fashion similar to the D, antagonist SCH23390 because it inhibits dopamine synthesis by a non-D, receptor mediated process.6 However, this effect occurs only at relatively high concentrations and, thus, was not likely to have accounted for the substantial effect of SKF 38393 seen in the current study. Finally, it is possible that SCH 23390 and SKF 38393 may not act on precisely the same population of D, receptors. A variety of evidence points to subtypes of D, receptors.’

P < 0.01 P < 0.05 P < 0.01

NS

Effects of dopaminergic agonists and antagonists on choice accuracy performance (entries to repeat) combined across postlesion sessions 1-5 (mean f S.E.M.). Compared with the untreated lesioned group, significant improvements in choice accuracy were seen in the

lesioned rats treated with the D, agonist, SKF38393 (P -c 0.01); the D, antagonist, SCH 23390 (P i 0.05), and the D, agonist, LY 171555(P < 0.01). Lesioned rats treated with the D, antagonist, raclopride, were not significantly different (NS) from untreated lesioned rats.

DISCUSSION Previous results we have obtained have demonstrated the importance of the medial cholinergic projection from the basal nuclear complex to the cerebral cortex for accurate radial-arm maze performance.33 Knife-cut lesions of this pathway in the cortex significantly impaired radial-arm maze performance. This deficit was attenuated by the cholinergic agonist arecoline. The results of the present study demonstrated that this lesion-induced deficit can also be reversed with D, and D, treatments. The D, agonist LY 171555 effectively attenuated the lesion-induced choice accuracy deficit, whereas the D2 antagonist raclopride was without effect. Both the D, agonist SKF 38393 and the D, antagonist SCH 23390 were effective in attenuating the lesioninduced choice accuracy deficit. The effects of these dopaminergic ligands in reversing the lesion-induced memory deficits are consistent with our previous observations of their reversal of

60 ,

-

Control

__O_

Lesion

Medial Cartex

O’, 1

, 2

I

,

I

I

I

1

3

4

5

6

7

8

I

I

9

10

-

SKF 38393

-

LY 171555

Trial

Fig. 3. Effectsof dopaminergic agonists on response duration. The lesioned group treated with SRF 38393, a D, agonist, had signikantiy gmater response duration than the untreated f&smed group on~tiion

9

only (P < 0.05). The lesioned group treated with LY 171555(quinpirole), a D2 agonist, had significantly greater response duration on sessions 2-10 (P < 0.05).

Dopaminergic

drugs

and cholinergic

medial

133

pathway

_

Control Lesion

Medial Cortex

O’, 1

I

I

I

I

I

I

I

1

2

3

4

5

6

7

8

9

Lesion

+

-

SCH-23390

_d_

Raclopride

1’ 10

Trial Fig. 4. Effects of dopaminergic antagonists on response duration. The lesioned group treated with SCH 23390, a D, antagonist, had significantly greater response duration than the untreated lesioned group on sessions 1 and 7 (P < 0.05). The lesioned group treated with raclopride, a D, antagonist, had significantly greater response duration on sessions 2. 3 and 5-10 (P < 0.05).

Differential sites of action of SCH 23390 and SKF 38393 on different D, receptor subtypes may underlie their similar effects. Dopaminergic input has inhibitory influence over septohippocampal cholinergic cells.‘0.4’ Dopamine antagonist infusion into the septum increases the firing rate of septohippocampal acetylcholine neurons and improves cognitive performance.” This may be a possible site for the beneficial effects of the D, antagonist SCH 23390. In contrast, dopamine stimulation in the nucleus accumbens4’ and frontal cortex7,44 is important for working memory function. D, receptors in particular are important for dopamine effects in the frontal cortex.44 These may be likely sites for the beneficial effects of the D, agonist SKF 38393. The choice accuracy deficits caused by medial cholinergic pathway lesions did not appear to be related to alterations in motor or motivational systems. Others have reported that lesions of the basal nuclear complex produce a spectrum of effects, such as motor dysfunction, loss of appetite, in addition to

Table 2. Response duration: sessions l-10 Seconds/ entry

Group Control Lesion Lesion Lesion Lesion Lesion

only + SKF 38393 + SCH 23390 + LY 171555 + raclopride

15.6 14.8 19.6 25.5 32.8 32.6

k + + & k &

1.5 1.3 1.6 2.6 3.2 2.5

Significance vs lesion NS NS

P < 0.01 P < 0.01 P < 0.01

Effects of dopaminergic agonists and antagonists on response duration combined across postlesion sessions l- 10 (mean k S.E.M.). Compared with the untreated lesioned group, significant increases in response duration were seen in the lesioned rats treated with the D, antagonist, SCH 23390 (P < O.Ol), the D, agonist, LY 171555 (P -c0.01) and the D, antagonist, raclopride (P < 0.01). Lesioned rats treated with the D, agonist SKF 38393 were not significantly different (NS) from untreated lesioned rats in terms of response duration.

profound alterations in learning and memory.‘9~34~38 However, the results of the present study cannot be accounted for by lesion-induced motor impairment or alterations in motivational state. Whereas rats with lesions of the medial cholinergic pathway were severely impaired in terms of choice accuracy, they did not differ from controls in response duration to respond or number of reinforcers consumed. These rats also did not show the deleterious health effects or weight loss seen with more global nucleus basalis the appetite and motor effects lesionsJ4 Perhaps reported by others after lesions of cell bodies in the nucleus basalis were attributable to disruption of the lateral pathway of that nucleus. In pilot experiments, we found that interdiction of the lateral cholinergic projections from the basal nuclear complex by extensive bilateral knife-cuts in a sagittal plane produced animals with profound motor disturbances and loss of appetite. The effects of the dopaminergic drugs on choice accuracy also did not appear to be due to alterations in motor function. Even though several of the dopaminergic drugs increased response latency, there was no consistent relationship between speed of responding and choice accuracy. For example, animals receiving lesion + LY 171555 were significantly slower than the untreated lesion group in their latency to respond but their choice accuracy was better. By comparison, animals receiving lesion + raclopride were significantly slower than lesion alone rats but their choice accuracy was very similar.

CONCLUSION

The interactions of D, and D1 systems with cholinergic mechanisms revealed by this and previous studies may be important for the development of treatments of Alzheimer’s disease.39 Alzheimer’s disease has been characterized, in significant part. as

S. R. MCGURK et (11

134

a disorder of cholinergic function.“,39,4”.4’ Given that D, and D2 ligands were found in the current study to reverse the choice accuracy impairment resulting from lesions of the medial cholinergic pathway and have also been found to be effective in reversing the memory deficits which result from pharmacological blockade of nicotinic and muscarinic acetylcholine receptors, 24,‘5.27 it is not unreasonable to consider the possibility that similar treatments may be useful in attenuating the cognitive deficits associated with Alzheimer’s disease. Regardless of their clinical implications, the results of the present experiments demonstrate the importance of medial cholinergic

pathways in cognitive function and, perhaps more importantly, the significance of interactions between cholinergic and dopaminergic systems in modifying such activities. research was supported by NIH grant NS-10928 to L.L.B.; S.R.M. was supported by the Ursula Mandel Scholarship from UCLA and an American Psychological Association Dissertation Research Award E.D.L. was supported by NIDA grant DA02665 and an Alzheimer’s Association/Neil Bluhm Pilot Research Grant. The authors wish to thank Smith, Kline, and French for supplying SKF 38393, Eli Lilly for supplying LY 171555, Schering for supplying SCH 23390, and Astra for supplying raclopride. Acknowledgements-This

REFERENCES 1. Andersen Dopamine 2. Bartus R. following

P. H., Gingrich J. A., Bates M. D., Dearry A., Falardeau P., Senogles S. E. and Caron M. G. (1990) receptors: beyond the D,/D, classification. Trends Pharmac. Sci. 11, 231-236. T., Flicker C., Dean R. L., Pontecorvo M., Figueiredo J. C. and Fisher S. K. (1985) Selective memory loss nucleus basalis lesions: long term behavioral recovery despite persistent cholinergic deficiencies. Pharmac.

Biochem. Behav. 23, 125-135. 3. Beninger R. J., Wirsching B. A., Jhamanda K., Boegman R. J. and El-Defrawy S. R. (1986) Effects of altered cholinergic function on working and reference memory in the rat. Can. J. Physiol. Pharmac. 64, 376-382. 4. Big1 V., Woolf N. J. and Butcher L. L. (1982) Cholinergic projections from the basal forebrain or frontal, parietal, temporal, occipital, and cingulate cortices: a combined fluorescent tracer and acetylcholinesterase analysis. Bruin Res. Bull. 8, 727-749. 5. Bischoff S., Heinrich M., Sontag J. and Kraus J. (1986) The D-l dopamine receptor antagonist SCH 23390 also interacts potently with brain serotonin (5-HT2) receptors. Eur. J. Pharmac. 129, 367-370. 6. Brooderson R. J., White F. J. and Galloway M. P. (1990) Inhibition of dopamine synthesis in striatal slices by the D, agonist SKF 38393 is not mediated by D, dopamine receptors. Synapse 6, 395-397. 7. Brozoski T. J., Brown R. M., Rosvold H. E. and Goldman P. S. (1979) Cognitive deficit caused by regional depletion of dopamine in prefrontal cortex of rhesus monkey. Science 205, 929-932.

8. Butcher L. L., Talbot K. and Bilezikjian L. (1975) Acetylcholinesterase neurons in dopamine-containing regions of the brain. J. Neural Transm. 37, 127-153. 9. Butcher L. L. and Woolf N. J. (1986) Central cholinergic systems: synopsis of anatomy and overview of physiology and pathology. In The Biological Substrates ofAi(zheimer’s Disease (eds Scheibel A. B. and Weschler A. F.), pp. 73-86. Academic Press, New York. 10. Costa E., Panula P., Thompson H. K. and Cheney D. L. (1983) The transsynaptic regulation of the septo-hipFocampa1 cholinergic neurons. Life Sci. 32, 165-179. 11. Doucette R., Fisman M., Hachinski V. C. and Mersky H. (1986) Cell loss from the nucleus basalis of Meynert in Alzheimer’s disease. Can. J. neural. Sci. 13, 435440. 12. Dubois B., Mayo W., Agid Y., LeMoal M. and Simon H. (1985) Profound disturbances of spontaneous and learned behaviors following lesions of the nucleus basalis magnocellularis in the rat. Bruin Res. 338, 249-258. 13. Dunnett S. B. (1985) Comparative effects of cholinergic drugs and lesions of nucleus basalis or fimbria-fornix on delayed matching in rats. Pyschopharmacology 87, 357-363. 14. Dunnett S. B., Low W., Bunch S. T., Thomas S. R., lversen S. D., Lewis S. D., BjBrklund P. R. and Stenevi U. (1981) Septal transplant reinnervation of the hippocampus: cholinergic enhancement of radial maze performance. Behav. Brain. Res. 2, 258-259.

15. Durkin T., Galey D., Micheau J., Beslon H. and Jaffard R. (1986) The effects of intraseptal injection of haloperidol in vivo on hippocampal cholinergic function in the mouse. Bruin Res 376, 420-124. 16. Erzin-Waters C. and Resch L. (1986) The nucleus basalis of Meynert. Gun. J. neural. Sci. 13, 8-14. 17. Fibiger H. C. (1986) Learning and memory deficits after ibotenate lesions of the nucleus basalis: reversal by cholinomimetics. Can. J. neural. Sci. 13, 498. 18. Flicker C., Dean R. L., Watkins S. L., Fisher S. K. and Bartus R. T. (1983) Behavioral and neurochemical effects following neurotoxic lesions of a major cholinergic input to the cerebral cortex in the rat. Pharmac. Biochem. Eehav. 18, 973-98 1. 19. Helper D. J., Wenk G. L., Cribbs B. L., Olton D. S. and Coyle J. T. (1985) Memory impairments following basal forebrain lesions. Brain Res. 346, S-14. 20. Jacobs R. W. and Butcher L. L. (1986) Pathology of the basal forebrain in Alzheimer’s disease and other dementias. In The Biological Substrates qfAlzheimer’s Disease (eds Scheibel A. B. and Wechsler A. F.), pp. 87-100. Academic Press, New York. 21. Kesner R. P., DiMattia B. V. and Crutcher K. A. (1987) Evidence for neocortical involvement in reference memory, Behav. neural. Biol. 47, 4tL-53. 22. Kristt, D. A., McGowan R. A., Martin-MacKinnon

N. and Soloman J. (1985) Basal forebrain innervation of rodent neocortex: studies using acetylcholinesterase histochemistry, Golgi and lesion’strategies. Bruin Res. 337, 19-39. 23. Levin E. D. (1988) Psychopharmacological effects in the radial-arm maze. Neurosci. Biobehav. Rev. 12. 169-175. 24. Levin E. D. i1988j Scbpoimine interactions with D, and D, antagonists on radial-arm maze performance in rats. Behav. neural Biol. SO, 240-245. 25. Levin E. D., McGurk S. R., Rose J. E. and Butcher L. L. (1989) Reversal of a mecamylamine-induced with the D, agonist, LY 171555. Pharmac. Biochem. Behav. 33, 919-922.

cognitive deficit

Dopaminergic

drugs

and cholinergic

medial

pathway

135

26. Levin E. D., McGurk S. R., Rose J. E. and Butcher L. L. (1990) Cholinergic-dopaminergic interactions in cognitive performance. Behav. neural Biol. 54, 271-299. 27. Levin E. D. and Rose J. E. (1991) Interactive effects of D, and Dz agonists with scopolamine on radial-arm maze performance. Pharmac. Biochem. Behav. 38, 243-246. 28. Lo Conte G., Bartolini L., Casamenti F., Pepeu M. and Pepeu G. (1982) Lesions of cholinergic forebrain nuclei: changes in avoidance behavior and scopolamine actions. Pharmac. Biochem. Behav. 17, 933-937. 29. Mash D. C., Flynn D. D. and Potter L. T. (1985) Loss of M, muscarine receptors in the cerebral cortex in Alzheimer’s disease and experimental cholinergic denervation. Science 228, 11 t 5-1 I 17. 30. McGurk S., Levin E. D. and Butcher L. L. (1988) Cholinergic-dopaminergic interactions in radial-arm maze performance. Behav. neural Biol. 49, 234-239. 31. McGurk S. R., Levin E. D. and Butcher L. L. (1989) Nicotinic-dopaminergic relationships and radial-arm maze performance in rats. Behaa. neuraf Biol. 52, 78-86. 32. McGurk S. R., Levin E. D. and Butcher L. L. (1989) Radial-arm maze performance in rats is impaired by a combination of nicotinic-choiinergic and D1 dopaminergic drugs. Ps,vchopharmaco~ogy 99, 371-373. 33. McGurk S. R., Levin E. D. and Butcher L. L. (1991) Impairment of radial-arm maze performance in Kits following lesions of the cholinergic medial pathway: reversal by arecoline and differential effects of muscarinic and nicotinic antagonists. Neuroscience 44, 131.-147. 34. Miyamoto M., Kato J., Narumi S. and Nagaoka A. (1987) Characteristics of memory impairment following lesioning of the basal forebrain and medial septal nucleus in rats. Brain Res. 419, 19-31. 35. Muramatsu M., Tamaki-Ohashi J., Usuki C., Araki H. and Aihara H. (1988) Serotonin-2 receptor-mediated reguiation of release of acetylcholine by minaprine in cholinergic nerve terminal of hippocampus of rat. l~eur~p~arrn~~~~~~g~ 27, 603609. 36. Murray C. L. and Fibiger H. C. (I 985) Learning and memory deficits after lesions of the nucleus basalis magnocellularis: reversal by physostigmine. Neuroscience 14, 1025-1032. 37. Olton D. S. and Samuelson R. J. (1976) Remembrance of places passed: spatial memory in rats. J. ~.xP. Psychol. unim. Behav. Proc. 2, 97-l 15. 38. Olton D. S. and Wenk G. L. (1987) Dementia: animal models of the cognitive impairments produced by degeneration of the basal forebrain cholinergic system. In Ps,lchopr2arnlacolog3’: The Third Generafj~n ~~Pr~gre.T.~fed. Meltzer H. Y.), pp. 941-953. Raven Press, New York. 39. Perry E. K. (1987) Cortical neurotransmitter chemistry in Alzheimer’s disease. In Psychopharmaro!ogy: The Third Generation of‘ Progress (ed. Meltzer H. Y.), pp. 887.-895. Raven Press, New York. 40. Rinne J. O., Lonnberg P., Marjamiiki P. and Rinne U. K. (1989) Brain muscarinic receptor subtypes are differentially affected in Alzheimer’s disease and Parkinson’s disease. Brain Res. 483, 402406. 41. Robinson S. E., Malthe-Ssrenssen D., Wood P. L. and Commissiong J. (1979) Dopaminergic control of the septo-hippocampal cholinergic pathway. .r. Pharmac. exp. Ther. 208, 476479. 42. Salamone J. D., Beart P. M., Alpert J. E. and fversen S. D. (1984) Impairment in T-maze reinforced alternation performance following nucleus basahs magnocellularis lesions in rats. Behav. Brain Res. 13, 63-70. 43. Saper C. B. (1984) Organization of cerebral cortical afferent systems in the rat. Il. Magnocellular basal nucleus. J. coq. Neural. 222, 3 I3--342. 44. Sawaguchi T. and Goldman-Rakic P. S. (1991) DI dopamine receptors in prefrontal cortex: involvement in working memory. Science 251, 9477950. 45. Taghzouti K., Simon H., Louilot A. and Le Moal M (1986) Behavioral study after local injection of &hydroxydopamine into the nucleus accumbens in the rat. Brain Res. 344, 9-20. 46. Wenk G. L., Markowska A. L. and Olton D. S. (f989) Basal forebrain lesions and memory: alterations in neurotensin. not acetylcholine, may cause amnesia. Behm. Neurosci. 103, 765-~769. 47. Whitehouse P. J., Martin0 A. M., Antuono P. G.. Lowenstein P. R., Coyle J. T., Price D. L. and Keller K. J. (t986) Nicotinic acetylcholine binding sites in Alzheimer’s disease. Brain Res. 371, 146-151. 48. Woolf N. J., Eckenstein F. and Butcher L. L. (1983) Cholinergic projections from the basal forebrain to the frontal cortex: a combined fluorescent tracer and immunohistochemical analysis in the rat. Neurosci. Lerr.40, 93-98. 49. Woolf N. J., Eckenstein F. and Butcher L. L. (1984) Chohnergic systems in the rat brain: 1. Projections to the limbic telencephalon. Brain Res. BuX 13, 751-784. 50. Woolf N. J., Hernit M. C. and Butcher L. L. (1986) Cholinergic and non-cholinergic projections from the rat basal forebrain revealed by combined choline acetyltransferase and Phase&s vulgaris leucoagglutinin immunohistochemistry. Neurosri. Lell. 66, 281-286. 51. Woolf N. J., Jacobs R. W. and Butcher L. L. (1989) The pontomesencephalotegmental cholinergic system does not degenerate in Alzheimer’s disease. N~urosci. Leti. 96, 277-282. (Accepted 26 parch

1992)