Psychopharmacology (1988) 94:491-495
© Springer-Verlag 1988
Attenuation of scopolamine-induced impairment
of spontaneous alternation behaviour by antagonist
but not inverse agonist and agonist fl-carbolines
Martin Sarter, Geert Bodewitz, and David N. Stephens
Research Laboratories of Schering AG, D-t000 Berlin 65, and D-4619 Bergkamen, Federal Republic of Germany
Abstract. Mice were tested in a simple automated Y-maze.
Total number of arm entries and alternation behaviour
were measured. The latter is thought to reflect working
memory capacity at a rudimentary level. During an 8-rain
session, vehicle-treated mice performed 32.4+7.4 arm en-
tries, 51.0+ 12.4% of which were organized in alternations
(triplets). The two variables showed a negative correlation.
Scopolamine (1.0 mg/kg) significantly enhanced activity, re-
duced alternation behaviour and diminished the correlation
between the two variables. The effects of benzodiazepine
receptor inverse agonist, antagonist and agonist/?-carbo-
lines on this spontaneous behaviour and on the effects of
scopolamine were examined. The effects of inverse agonists
and agonists on locomotor activity were complex in interac-
tion with both vehicle and scopolamine. The scopolamine-
induced reduction of alternation behaviour was significant-
ly reversed by the antagonist ZK 93426 but not by inverse
agonists; furthermore, partial agonists and agonists showed
no effects. It is hypothesized that the interaction of antago-
nist /%carbolines with scopolamine is based on a direct
GABA-ergic control of cholinergic neurotransmission, and
suggests an ability of antagonist p-carbolines to antagonize
amnestic properties of scopolamine.
Key words: Spontaneous a l t e r n a t i o n - S c o p o l a m i n e - / / -
Carbolines - Memory - Acetylcholine - Mice
Since the cholinergic system has become the centre of dis-
cussion in Alzheimer's disease research, two animal models
of disturbed memory processes have received growing atten-
tion, the direct "pathoanatomical model", i.e. basal fore-
brain lesions, and the "pharmacological model", namely,
blockade of cholinergic receptors by scopolamine or atro-
pine (Spencer and Lal 1983). Scopolamine has been shown
to impair acquisition in a variety of paradigms in different
species, a finding which is frequently attributed to atten-
tional deficits. However, the effects of scopolamine do not
seem to be limited to working memory but affect also refer-
ence memory abilities (e.g. Spencer et al. 1985).
Cholinomimetic drugs have been demonstrated to an-
tagonize the effects of scopolamine in various paradigms
(for review see Warburton and Wesnes 1984); physostig-
mine, for instance, has been shown to antagonize the scopo-
Offprint requests to: Martin Sarter, Department of Neuropsycho-
pharmacology, Schering AG, P.O. Box 650311, D-1000 Berlin 65,
Federal Republic of Germany
Psychopharmacology
lamine-induced deficit of spontaneous alternation behav-
iour (Tobe et al. 1983). Since there are some reports of
an interaction between diazepam, muscimol and acetylcho-
line turnover (Zsilla et al. 1976; Nagy and Desci 1978; Tan-
ganelli et al. 1985; Casamenti et al. 1986), suggesting that
the cortical acetylcholine release is under the control of
GABA-ergic neurons, it could be speculated that benzodi-
azepine (BZ) receptor inverse agonist ligands might enhance
cortical acetylcholine turnover. If this is true, the enhanced
availability of the transmitter would displace scopolamine
from the receptor and, therefore, antagonize the behaviour-
al effects of scopolamine.
We would therefore predict that inverse agonist/~-carbo-
lines would display scopolamine antagonist properties in
behavioural models. However, bearing in mind the power-
ful anxiogenic and convulsant properties of full inverse ago-
nist benzodiazepine receptor ligands (Braestrup et al. 1982;
Stephens and Kehr 1985), it is to be expected that improve-
ments in scopolamine-disrupted behaviour by inverse ago-
nists would take place over a very limited dose range; for
this reason weak partial inverse agonists may be considered
as being more likely to show anti-scopolamine activities.
In order to develop a rapid and reliable screening meth-
od to test the interaction of drugs with scopolamine-induced
behavioural impairments, we have looked for a simple,
spontaneous behaviour which could be easily measured,
which was sensitive to scopolamine and which reflected -
even if at a very rudimentary level - the cognitive status
of the animal. Although tests on spontaneous alternation
behaviour were introduced more than 20 years ago (e.g.
Douglas and Isaacson 1965), Anisman's reports (Anisman
1975) have made them popular. Such a test is based on
a simple maze and on the tendency of rodents to enter
that arm of the maze which was least recently explored,
i.e. the arm which was not entered in the last two choices
(in the case of a Y-maze).
Anisman (1975) found that the activity of the animals
is a poor predictor of and negatively related to the alterna-
tion behaviour and discussed his results only in terms of
habituation processes. Subsequently, alternation ability has
been interpreted in terms of working memory (Warburton
and Heise 1972; Beninger et al. 1986).
Materials and methods
Subjects. Male N M R I mice weighing 25 + 2 g (Department
Tierzucht und -haltung, Schering AG) were used.
492
l°{~m..... A- . . . . /~ photocells
walls:lO cm high
plug box A A' ~
Y-MAZE
Fig. 1. Schematic illustration of the Y-maze. The photo beams are
indicated by broken lines. For further details see text
Apparatus. The Y-maze used for all tests is illustrated in
Fig. 1. It was made of black Plexiglas with transparent ceil-
ings. It has been automated by D. Niemeyer (Bioelectronic
Lab. Schering AG) using photo beams to detect locomo-
tion. The beams in the centre area are used to differentiate
real entries into an arm from subsequent movements within
an arm of the maze. The data of each trial are collected
and computed (see below) by a logic device also created
by D. Niemeyer.
Testing procedure. Following random assignment to experi-
mental groups, subjects were treated with either vehicle
[Cremophor EL/BASF Ludwigshafen (CEL); 10% in sa-
line; 10 ml/kg; IP] or scopolamine (scopolamine H C L ;
Sigma; 1 mg/kgjl0 ml vehicle IP) and the test compound
(see below). Both injections were carried out 30 rain before
the test; the mouse was put into the maze, the ceiling closed,
and, after 8 min, the trial terminated automatically. Follow-
ing each trial, the maze was placed on a new sheet of paper.
Testing was performed between I p.m. and 5 p.m. For each
dose, eight animals were tested for the interaction with vehi-
cle and with scopolamine. The three doses selected for test-
ing were based on calculations from in vivo binding values;
for two compounds (B-CCM; ZK 95962) an additional
fourth dose was tested because of a positive trend seen
in the variable "scopolamine-induced reduction of alterna-
tion behaviour."
Data analyses. Locomotor activity, i.e. total number of arm
entries, was collected cumulatively over the 8 min. The al-
ternation behaviour defined as consecutive entries into all
three arms without repeated entries was expressed in per-
cent of the total arm entries. If, for instance, the arms are
called A, B, C, and the animal performs:
ABCBABCBA CBAACB,
the total alternation opportunities would be 13 (total entries
minus 2) and the percent alternation behaviour would be
Table l. Classification offl-carbolines and doses used
Compound Action at the Doses used (mg/kg)
benzodiazepine receptor
ZK 93423 Full agonist 0.001, 0.025, 0.1
ZK 95962 Partial agonist 0.39, 1.56, 6.25, 12.5
ZK 91296 Partial agonist 0.1, 0.39, 1.56
ZK 9 3 4 2 6 Antagonist 1.56, 6.25, 25
ZK 90886 Partial inverse agonist 0.39, 1.56, 3.13
FG 7142 Partial inverse agonist 1.56, 6.25, 25
fl-CCM Inverse agonist 0.1, 0.39, 1.56, 3.13
DMCM Inverse agonist 0.1, 0.39, 1.56
Note: Classifications are based on biochemical evidence of the
effect of GABA agonists on the affinity of the fl-carbolines to
the benzodiazepine receptor, on the influence of the fl-carbolines
on the binding of [asS]TBPS, a ligand for the GABA-modulated
chloride ion channel, and on pharmacological effects; see, e.g. Ste-
phens and Kehr (1985)
61.5% (8 out of 13). (Note that " B A C " but not " B A A C "
is counted as an alternation according to the definition giv-
en above and that, therefore, this computation differs from
that used by Anisman and results in an apparently lower
degree of spontaneous alternation.)
In order to demonstrate the principle of the task, two
groups of mice were injected either with vehicle/vehicle or
vehicle/scopolamine and the results compared with the t-
test (n 1 = n2 = 12; c~= 0.05). Statistical comparisons of drug-
treated groups to controls were carried out with four inde-
pendent Dunnett tests (each including six to seven drug-
treated groups, i.e. doses of two compounds, and a control
group (each group: n = 8; ~ = 0.05 for each test). All statisti-
cal analyses were carried out by Dr. A. van der Linde,
Dept of Biostatistics, Schering AG.
Drugs. All the drugs were prepared as ultrasonic microsus-
pensions in 10% Cremophor in saline (see above). ZK
93 423 (6-benzyloxy-4-methoxymethyl-/~-carboline-3-car-
boxylic acid ethyl ester), ZK 93 426 (5-isopropoxy-4-methyl-
/~-carboline-3-carboxylic acid ethyl ester), and ZK 95962
(5-isopropoxy-4-methoxymethyl-fl-carboline-3-carboxylic
acid ethyl ester) were synthesized by Dr. D. Seidelmann,
Schering AG; ZK 91296 (5-benzyloxy-4-methoxymethyl-fl-
carboline-3-carboxylic acid ethyl ester) and D M C M (6,7-di-
methoxy-4-ethyl-fl-carboline-3-carboxylic acid methyl ester)
by Dr. G. Neef, Schering AG, ZK 90886 (5-methoxy-4-
ethyl-fl-carboline-3-carboxylic acid ethyl ester) by Dr. V.
Eder, Schering AG, and F G 7142 (fl-carboline-3-carboxylie
acid methyl amide) a n d / / - C C M (fl-carboline-3-carboxylic
acid methyl ester) were synthesized by AS Ferrosan, Copen-
hagen, Denmark. These compounds are classified in Table 1
according to their main actions at the GABA/benzodiaze-
pine receptor complex.
Results
Although the effects of fl-carbolines on the scopolamine-
induced reduction of alternation behaviour are the main
subject of this study, their direct effects (vehicle interaction)
are also briefly presented.
Effects of scopolamine
During the 8-rain session, the vehicle-treated mice per-
formed 32.4_+7.4 (M; SD) arm entries and 51.0_+12.4%
 493

~6 Spontaneousalternation:Baselines Spontaneousalternation:ZK93426
Total arm entries M, SD alternation(%) % alternation all trials (M, SD)
-70

60. -60 60

7 -50 50-
l 2_
40- -40 40-

30- -30 30-

20- 20 20-

10- -10 10-

GEL
+
CEL
+
CEL GEL
+
SC SC SC SC
CEL/CEL ;EL/SC CEWCEL CEL/SC
n=12 n=12 CEL ZK ZK ZK GEL zK
+ z+K z+K
1.56 6.25 25.0 1.56 6.25 25.0

Fig. 2. The effect of scopolamine (1 mg/kg; IP) on spontaneous
alternation performance and total arm entries in the Y-maze in
comparison to vehicle-treated mice (vehicle - vehicle and vehicle
- scopolamine). CEL Cremophor EL (vehicle); SC scopolamine;
M mean; SD standard deviation; * P < 0.05 (t-test)
Fig. 3. The effects of the antagonist fl-carboline ZK 93426 in inter-
action with vehicle (Cremophor EL; 10 ml/kg; IP) or scopolamine
(1 mg/kg; IP), respectively, on alternation performance. Whereas
there is no effect of ZK 93426 (ZK; doses in mg/kg) when coin-
jected with vehicle (CEL), it antagonized the scopolamine (SC)-
induced impairment of alternation performance at 6.25 mg/kg. M
mean; SD standard deviation; * P<0.05 (Dunnett test)

alternations. The variables were negatively correlated (r = Table 2. Correlation coefficients: total arm entries and alternation
- 0 . 4 9 ) , i.e. the higher the n u m b e r o f total arm entries, all trials
the lower the alternation performance. Scopolamine (1 mg/
kg) significantly enhanced l o c o m o t o r activity ( t = - 4 . 1 1 ; Compound Interaction with
P = 0 . 0 0 0 7 ) but reduced alternation behaviour ( t = 2 . 1 8 ;
CEL Scopolamine
P = 0.043). Furthermore, scopolamine abolished the interre-
lationship between the two variables seen in the vehicle- CEL -- 0.46 0.19
controls (r = 0.19; see Fig. 2).
DMCM 0.07 - 0.29
Direct effects of fl-carbolines (interaction with vehicle) fl-CCM - 0.07 - 0.11
FG 7142 -0.63 0.00
D M C M (0.39 mg/kg) reduced significantly the activity o f ZK 90886 -0.22 -0.19
the mice at a dose o f 0.39 mg/kg ( t = 3 . 1 9 ; P < 0 . 0 5 ) . The ZK 93426 --0.39 0.15
full agonist Z K 93 423 at 0.1 m g / k g also drastically reduced ZK 91296 -0.14 -0.18
activity; since at this dose of this c o m p o u n d some animals ZK 95962 -0.40 --0.03
ZK 93423 --0.48 -0.33
were completely immobile, it has not been included in the
statistical analysis. The only c o m p o u n d which enhanced
activity was the partial agonist Z K 95962 (12.5 mg/kg;
Interrelationships between the interdependent variables. The
t = 2.79; P < 0.05). N o n e o f the other c o m p o u n d s affected
negative correlation coefficient between total arm entries
the n u m b e r o f total arm entries, or alternation behaviour
and alternation behaviour in the case of the vehicle interac-
when coinjected with vehicle.
tion was attenuated by all c o m p o u n d s except F G 7142 and
Z K 93423 (over all doses for each c o m p o u n d ; see Table 2).
Effects of fi-carbolines in interaction with scopolamine
In interacting with scopolamine, for none of the c o m p o u n d s
Effects on scopolamine-induced enhancement of locomotor (including the vehicle) was any interrelationship found (see
activity. The effects of different fl-carbolines on this variable Table 2).
were complex. D M C M (1.56 mg/kg; t = 3 . 0 8 ; P < 0 . 0 5 ) as
well, as Z K 93 423 (0.025 mg/kg; t = 2.81 ; P < 0.05) reduced
Discussion
the l o c o m o t o r activity enhanced by scopolamine, but the
other c o m p o u n d s and other doses o f Z K 93423 showed The main result o f this study suggests that the antagonist
no significant effects. fl-carboline Z K 93426 was able to antagonize the effects
of scopolamine on spontaneous alternation behaviour. The
Effects on scopolamine-induced reduction of alternation inverse agonists D M C M and fl-CCM and the partial inverse
behaviour. The inverse agonists D M C M (0.39 mg/kg) and agonist Z K 90886, but not F G 7142, all showed a non-
fl-CCM (0.1 mg/kg) and the partial inverse agonist Z K significant tendency to attenuate effects on the scopol-
90 886 (0.39 mg/kg) tended to attenuate the alternation im- amine-induced impairment, whereas all the agonist fl-carbo-
pairment induced by scopolamine (all P > 0.05) but the par- lines were ineffective.
tial inverse agonist F G 7142 showed no effect. The antago-
nist Z K 93426 significantly attenuated the effects of scopo- Effects on locomotor activity. The mixed effects of inverse
lamine at a dose o f 6.25 mg/kg ( t = 2 . 8 6 ; P < 0 . 0 5 ; see agonist and agonist fl-carbolines on l o c o m o t o r activity, irre-
Fig. 3). spective o f the interaction with scopolamine or vehicle,
494
might be related to the sedative effects in the case of full
agonists and, in the case of inverse agonists, to their poten-
tial anxiogenic properties (Stephens and Kehr 1985; e.g.
Beck and Cooper 1986). During this experiment, we careful-
ly observed the animals for signs of convulsions but they
never occurred at the doses used. Therefore, it seems unlike-
ly that the locomotor inhibition is due to convulsive proper-
ties of the inverse agonists. The one activity-enhancing com-
pound, ZK 95962, is a partial agonist without sedative
properties (e.g. Stephens et al. 1987).
Effects on alternation behaviour. The only compound which
significantly antagonized the effects of scopolamine on al-
ternation behaviour was the antagonist ZK 93426. The
compound was not effective in antagonizing the scopol-
amine-induced enhancement of locomotor activity, which
might support the hypothesis of two different neuronal
mechanisms underlying the effects of scopolamine on loco-
motion and alternation, respectively. Furthermore, the
compound did not show any effect when coadministered
with the vehicle. The antagonist ZK 93 426, therefore, was
selectively active in interaction with the effects of scopol-
amine on alternation behaviour. Since alternation behav-
iour may reflect a primitive working memory capacity, this
result suggests that ZK 93 426 may have a specific interac-
tion with a cholinergic system involved in memory pro-
cesses.
Although D M C M , fl-CCM and ZK 90 886, but not F G
7142, also showed a tendency to antagonize the effects of
scopolamine on spontaneous alternation, the initial hypoth-
esis that the potency of/%carbolines to antagonize the ef-
fects of scopolamine on alternation behaviour may be di-
rectly related to their degree of exerting inverse agonist
properties was not confirmed by our results. However, the
inverse agonists have been shown to antagonize the effects
of scopolamine in other tests as, for instance, passive avoid-
ance performance (Jensen et al. 1987). Moreover, the fact
that agonists did not markedly enhance the effects of sco-
polamine suggests that in general the interactions of fl-car-
bolines with scopolamine are not predicted from their status
in terms of a benzodiazepine receptor agonist or inverse
agonist pharmacological profile.
Mode of action of the antagonist ~-carboline Z K 93426.
Surprisingly, the benzodiazepine receptor antagonist ZK
93426 which has only weak intrinsic properties (Jensen
et al. 1984; File et al. 1986) most potently antagonized the
effects of scopolamine. However, in some experiments ZK
93 426 exerts definite pharmacological effects, including ef-
fects consistent with an anxiogenic profile (Jensen et al.
1984; File et al. 1986), state dependency in a step-down
passive avoidance task, increased exploratory activity in a
hole-board in senescent but not young rats, and, similarly,
it improved spatial delayed alternation learning in senescent
rats but impaired it in young ones (Sarter and Stephens
1988). Therefore, the question remains of the mechanism
by which this compound is able to antagonize the effects
of scopolamine and exert beneficial effects only in otherwise
behaviourally impaired animals.
A direct interaction at cholinergic receptors can be ex-
cluded, since ZK 93426 shows very low or no affinity for
3H-QNB binding sites (quinuclidinyl benzilate; T. Honore,
personal communication). Since, however, lesion of the
cholinergic neurons of the basal nucleus of Meynert selec-
tively affects alternation performance in rats (Beninger et al.
1986), it might be speculated that the cortical cholinergic
afferents are involved in the effects of scopolamine on alter-
nation behaviour and other systems (perhaps striatal inter-
neurons) in the mediation of locomotor activity (Karczmar
1976). There is, however, only poor evidence for a differen-
tiation of the cholinergic systems in terms of receptor types
or binding parameters (but see Richardson et al. 1987).
GABA-ergic agonists as well as benzodiazepines (mostly
muscimol and diazepam) have been shown to reduce corti-
cal acetylcholine turnover (Petkov et al. 1983; Casamenti
et al. 1986). The effects on acetylcholine turnover are re-
stricted to the cortex and midbrain and are not found in
the striatum and hippocampus (Zsilla et al. 1976). Injections
of muscimol into the basal forebrain similarly reduced ace-
tylcholine turnover (Wood 1985) and high affinity choline
uptake, the rate limiting step in the acetylcholine synthesis
(Wenk 1984). Recently, the anatomical correlate of such
an interaction has been demonstrated by Zaborsky et al.
(1986), who showed that GAD-immunopositive terminals
make synaptic contacts with cholinergic neurons of the bas-
al forebrain, and we have shown that there is a high density
of benzodiazepine receptor binding sites in the substantia
innominata (Sarter et al. 1988).
Thus, it could be speculated that antagonist and inverse
agonist fl-carbolines will enhance cortical acetylcholine
turnover via the GABA-ergic control of the activity of cho-
linergic efferents of the basal forebrain. In the intact organ-
ism, these cholinergic neurons might be strongly inhibited
by an (unknown) endogeneous ligand of the GABA-benzo-
diazepine-receptor complex. The antagonist might therefore
exert pharmacological actions by disinhibiting the choliner-
gic turnover. The increased release of acetylcholine would
then be able to displace scopolamine at the receptor.
Furthermore, it could be speculated that the antagonist
fl-carboline is more effective than inverse agonists in antag-
onizing behavioural effects of scopolamine, because the dis-
inhibiting effects of inverse agonists on acetylcholine turn-
over might result in a level of acetylcholine release too high
to mediate effectively a behavioural act. According to these
speculations, antagonist fl-carbolines should enhance the
relearning performance in animals which are behaviourally
impaired as a result of a reduction of the acetylcholine
turnover (Sarter 1986). This assumption contrasts to the
interpretation of recently reported beneficial effects of fl-
CCM on acquisition of a passive avoidance task which has
been attributed to an arousal-enhancing property of fl-
CCM (Venault et al. 1986). Since, however, the antagonist
fl-carboline did not exert effects on arousal as far as can
be seen from its ineffectiveness on locomotion, the potential
enhancement of cognitive abilities may be mediated via a
different psychological process as, for instance, enhanced
vigilance (Stephens and Sarter 1988).
Acknowledgements. We are grateful to D. Niemeyer and R. Koll-
berg for automating the Y-maze apparatus, to Dr. A. van der
Linde for the statistical work, and to Dr. T. Honore for allowing
us to cite his unpublished observations on 3H-QNB binding.
References
Anisman H (1975) Dissociation of disinhibitory effects of scopol-
amine: Strain and task factors. Pharmacol Biochem Behav
3:613-618
Beck CHM, Cooper SJ (1986) The effect of the fl-carboline FG

7142 on the behaviour of male rats in a living cage: An ethologi-
cal analysis of social and nonsocial behaviour. Psychopharma-
cology 89:203-207
Beninger RJ, Jhamandas K, Boegman RJ, EL-Defrawy SR (1986)
Effects of scopolamine and unilateral lesions of the basal fore-
brain on T-maze spatial discrimination and alternation in rats.
Pharmacol Biochem Behav 24: 1353-1360
Braestrup C, Schmiechen R, Neef G, Nielsen M, Petersen EN
(1982) Interaction of convulsive ligands with benzodiazepine
receptors. Science 216 : 1241-1243
Casamenti F, Deffenu G, Abbamondi AL, Pepeu G (1986) Chang-
es in cortical acetylcholine output induced by modulation of
the nucleus basalis. Brain Res Bull 16:689-695
Douglas RJ, Isaacson RL (1965) Homogeneity of single trial re-
sponse tendencies and spontaneous alternation in the T-maze.
Psychol Rep 16:87-92
File SE, Pellow S, Jensen LH (1986) Actions of the/~-carboline
ZK 93426 in an animal test of anxiety and the holeboard:
Interactions with Ro 15-1788. J Neural Transm 65 : 103-114
Jensen LH, Petersen EN, Braestrup C, Honore T, Kehr W, Ste-
phens DN, Schneider H, Seidelmann D, Schmiechen R (1984)
Evaluation of the/~-carboline ZK 93 426 as a benzodiazepine
receptor antagonist. Psychopharmacology 83:249-256
Jensen LH, Stephens DN, Sarter M, Petersen EN (1987) Bidirec-
tional effects of p-carbolines and benzodiazepines on memory
processes. Brain Res Bull 19:359-364
Karczmar AG (1976) Central actions of acetylcboline, cholinomi-
metics, and related drugs. In: Goldberg AM, Hanin I (eds)
Biology of cholinergic function. Raven Press, New York, pp
395 449
Nagy J, Decsi L (1978) Physostigmine, a highly potent antidote
for acute experimental diazepam intoxication. Neuropharma-
cology 17:469-475
Petkov V, Georgiev V, Getova D, Petkov VV (1983) On the effects
of diazepam, hyoscine and oxotremorine on acetylcholine re-
lease from the cerebral cortex. Acta Physiol Pharmacol Bulg
9:3-13
Richardson PJ, Brown SJ, Bailyes EM, Luzio JP (1987) Ectoen-
zymes control adenosine modulation of immunoisolated cholin-
ergic synapses. Nature 327:232-234
Sarter M (1986) Some considerations of different modes of action
of nootropic drugs. Neuropsychobiology 15:192-200
Sarter M, Stephens DN (1988) ]~-carbolines as tools in memory
research: Animal data and speculations. In: Hindmarch I, Ott
H, Roth T (eds) Benzodiazepine receptor ligands : Memory and
information processing. Psychopharmacology [Suppl] (in press)
Sarter M, Schneider HH, Stephens DN (1988) Treatment strategies
for senile dementia : Antagonist fl-carbolines. Trends Neurosci
(in press)
Spencer DG Jr, Lal H (1983) Effects of anticholinergic drugs on
learning and memory. Drug Dev Res 3 :489-502
Spencer DG Jr, Pontecorvo M J, Heise GA (1985) Central choliner-
495
gic involvement in working memory: Effects of scopolamine
on continuous nonmatching and discrimination performance
in the rat. Behav Neurosci 99:1049-1065
Stephens DN, Sarter M (1988) Bidirectional properties of/~-carbo-
line ligands at benzodiazepine receptors extend to effects on
vigilance. In: Hindmarch I, Ott H, Roth T (eds) Benzodiazepine
receptor ligands : Memory and information processing. Psycho-
pharmacology [Suppl] (in press)
Stephens DN, Kehr W (1985)/%carbolines can enhance or antago-
nize the effects of punishment in mice. Psychopharmacology
85 : 143-147
Stephens DN, Schneider HH, Kehr W, Jensen LH, Petersen E,
Honore T (1987) Modulation of anxiety by fl-carbolines and
other benzodiazepine receptor ligands: Relationship of phar-
macological to biochemical measures of efficacy. Brain Res Bull
19:309-318
Tanganelli S, Bianchi C, Beani L (1985) The modulation of cortical
acetylcholine release by GABA, GABA-like drugs and benzo-
diazepines in freely moving guinea-pigs. Neuropharmacology
24:291 299
Tobe A, Egawa M, Nagai R (1983) Effect of 4-(o-benzylphenoxyl)-
n-methylbutylaminehydrochloride (MCI-2016) on the scopol-
amine-induced deficit of spontaneous alternation behaviour in
rats. Jpn J Pharmacol 33 : 775-784
Venault P, Chapouthier G, de Carvalho LP, Simiand J, Morre
M, Dodd RH, Rossier J (1986) Benzodiazepine impairs and
p-carboline enhances performance in learning and memory
tasks. Nature 321 : 364-366
Warburton DM, Heise GA (1972) Effects of scopolamine on spa-
tial double alternation in rats. J Comp Physiol Psychol
81 : 523-532
Warburton DM, Wesnes K (1984) Drugs as research tools in psy-
chology: Cholinergic drugs and information processing. Neu-
ropsychobiology 11 : 121-132
Wenk GL (1984) Pharmacological manipulations of the substantia
innominata-cortical cholinergic pathway. Neurosci Lett
51 : 99-103
Wood PL (1986) Pharmacological evaluation of GABAergic and
ghitamatergic input to the nucleus basalis-cortical and the sep-
tal-hippocampal cholinergic projections. Can J Physiol Phar-
macol 64 : 325 328
Zaborsky L, Heimer L, Eckenstein F, Leranth C (1986) GA-
BAergic input to cholinergic forebrain neurons: An ultrastruc-
rural study using retrograde tracing of HRP and double immu-
nolabelling. J Comp Neurol 250:282-295
Zsilla G, Cheney DL, Costa E (1976) Regional changes in the
rate of turnover of acetylcholine in rat brain following diaze-
pare or muscimol. Arch Pharmacol 294:251-255
Received February 2, 1987 / Final version October 21, 1987