NeuropharmacologyVol. 31, No. 2, PP. 163-168,1992 Printed in GreatBritain. All rightsreserved
0028-3908/92$5.00+ 0.00 Copyright 0 1992Pergamon Press plc
THE EFFECTS OF DEXMEDETOMIDINE, AN ALPHA, AGONIST, ON LEARNING AND MEMORY, ASSESSED USING PASSIVE AVOIDANCE AND WATER MAZE TASKS IN RATS J. SIRVI~, P. RIEKKINENJR, T. EKONSALO,R. LAMMINTAUSTAand P. J. RIEKKINEN Department of Neurology, University of Kuopio, P. 0. Box 6, SF-7021 1 Kuopio, Finland and Orion Corporation,
Farrnos Research Group, Turku, Finland
(Accepted 19 September
1991)
Summary-The effects of dexmedetomidine, a specific and potent alpha, agonist, on the performance of rats in passive avoidance and water maze tasks were studied. Pre-training administration of subanaesthetic dose (9.0 pg/kg) of dexmedetomidine impaired the retention of the passive avoidance task (assessed 24 hr after training) but it did not affect the training of this task. Smaller doses (0.3,O.g and 3.0 pg/kg) did not affect the training or retention of this aversively motivated task. On the other hand, pre-training administration of 0.3 and 0.9 pg/kg dexmedetomidine impaired the acquisition of the water maze task, whereas larger doses (3.0 and g.Opgg/kg) had no significant effect on spatial learning. Pre-training administration of dexmedetomidine (0.3-9.0 pg/kg) increased swimming speed in rats. Only a large dose (300 lg/kg) of dexmedetomidine, administered immediately after training, impaired the retention of the passive avoidance task and the acquisition of the water maze task. These data agree with previous findings that pharmacological manipulation of the noradrenergic system affects the retention of aversively-motivated (passive avoidance) tasks. The present results suggest that the dose-response curve of dexmedetomidine for impairment of learning/memory differs between the passive avoidance and water maze tasks. Key words-alpha-2
adrenoceptor,
dexmedetomidine,
Aversively motivated (e.g. passive avoidance) and spatially cued (e.g. water maze) tasks have been widely
used in behavioural pharmacology (Olton and Wenk, 1987). These tasks partly differ in their underlying neural substrates; lesions of the amygdala impair the acquisition and retention of the inhibitory avoidance task but do not impair the acquisition of the water maze task (Cahill and McGaugh, 1990; Dunn and Everitt, 1988; Sutherland and McDonald, 1990). The learning and memory of a spatial task is critically dependent on the hippocampus and its connections with cortical and subcortical structures (Morris, Garrud, Rawlins and O’Keefe, 1982; Sutherland, Kolb and Whishaw, 1982; Sutherland and Rodriguez, 1989; Sutherland and McDonald, 1990). Anatomical and physiological studies suggest that noradrenergic neurones, projecting from the locus coeruleus to the forebrain, play a role in cognitive functions (Foote and Morrison, 1987; McCormick, 1989). Previously, lesions of the central noradrenergic system or the inhibition of the synthesis of noradrenaline have been found to impair learning/memory, assessed by the inhibitory avoidance task or different mazes (Anlezark, Crow and Greenway, 1973; Archer, 1983; Crow and Wendlandt, 1976; Randt, Quartermain, Goldstein and Anagnoste, 1971; Stein, Belluzi and Wise, 1975). Furthermore, psychopharmacological studies indicate that the noradrenergic system in the amygdala is involved in mechanisms of acquisition
learning, memory, noradrenergic systems.
and retrieval of memory, assessed using avoidance tasks (Gallagher, Kapp, Pascoe, Rapp, 1981; Ellis and Kesner, 1983; McGaugh, 1989; McGaugh, IntroiniCollison, Nagahara and Cahill, 1990). On the other hand, spatially cued learning is not impaired or may even be facilitated, by partial or extensive depletion of noradrenaline in the forebrain in adult rats (Hagan, Albert, Morris and Iversen, 1983; Riekkinen Jr, Sirviii, Valjakka, Pitkgnen, Partanen, Riekkinen, 1990; Selden, Cole, Everitt and Robbins, 1990a; Valjakka, Riekkinen Jr, SirviG, Nieminen, Airaksinen, Miettinen and Riekkinen, 1990). Neurochemical and electrophysiological studies suggest that the firing rate of the noradrenergic neurones in the locus coeruleus and the release of noradrenaline is regulated by alpha, autoreceptors (Cederbaum and Aghajanian, 1976; Birch and Fellenz, 1985). These receptors mediate the autoinhibition of noradrenergic neurones and the activation of these receptors decreases the turnover of noradrenaline (Aghajanian and VanderMaelen, 1982). The aim of the present study was to investigate the effects of pharmacologically decreased activity of noradrenergic neurones on learning and memory. Thus, the effects of dexmedetomidine, a selective and potent alpha, agonist (Savola and Virtanen, 1991; Scheinin, Virtanen, MacDonald, Scheinin and Lammintausta, 1990; Virtanen, Savola, Saano and 163
J. SIRVld et ai.
164
Nyman, 1988) on the performance of rats in passive avoidance and water maze tasks were studied. In order to differentiate between the effects of treatment with drugs on the acquisition and consolidation processes of memory, both pre- and post-training administrations of drug were used in these studies. METHODS
Animals
Male Wistar rats (27%35Og) were used in these experiments. The animals were housed in a temperature- (2O~C), humidity- (50-60%) and light- (lights on 0700-2100 hr) controlled environment, with free access to food and water. Drug
Dexmedetomidine (Orion Corporation, Farmos RCD Pharmaceuticals, Turku, Finland) was dissolved in saline and injected subcutaneously (s.c. 0.5 ml/kg). Controls were treated with saline (0.5 ml/kg). According to previous studies, dexmedetomidine has been shown to be a selective agonist of alpha, adrenoceptors and decrease the turnover of noradrenaline dose-dependently (l-300 pg/kg). Obvious sedation is induced at the doses of > 10 pg/kg. At doses of > 10 pg/kg, the turnover of serotonergic and dopaminergic neurones in the brain is also affected (Savola and Virtanen, 1991; MacDonald, Scheinin and Scheinin, 1988; Scheinin et al., 1990). For the estimation of sedation, visual inspection of the mobility of rats in their home cages was used, as well as measures of motor activity in behavioural tasks (the training latency of passive avoidance task and swimming speed in the water maze task). Passive avoidance experiments
The passive avoidance apparatus consisted of a rectangular plexiglass box, divided into dark and lighted compartments by a sliding guillotine door. The dark compartment had a metal grid floor, to which a shock generator (Campden Instruments Ltd, U.K.) was connected. In the training of the task, the rat was placed to the lit side. After entry into the dark compartment, a 3.0 mA electric shock was delivered to the feet of the rat. The shock remained on until the rat returned to the lit side. Training continued until a rat remained on the lit side for 60sec. Latency to the first entry into the dark chamber (training latency) and the number of re-entries (training trials) were recorded by the experimenter. In the retention test, the rat was placed on the lit side and, after 60 set the door was opened. The session continued until the rat entered the dark compartment (maximum time 360 set). The latency to enter into the dark chamber (testing latency) was recorded. Experiment 1. The rats were treated with saline or dexmedetomidine (0.3, 0.9, 3.0 or 9.0 pg/kg), 30 min before the training of the passive avoidance task. The rats were not treated immediately before the retention
test, which was performed 24 hr after training, as described above. Nine rats were tested in each group. Experiment 2. The rats were treated with saline or dexmedetomidine (0.3, 3.0, 30.0 or 300pg/kg), immediately (within 2min) after the training of the passive avoidance task. The rats were not treated immediately before the retention test, which was performed 24 hr after training. Ten rats were tested in each group. Water maze experiments
The water maze apparatus (purchased from San Diego Instruments, San Diego, California) has been described in detail previously (Riekkinen Jr, Sirviii and Riekkinen, 1990). During the training period of the water maze task, the rat was trained once a day for 15 days (maximum trial duration 70 set, 10 set reinforcement on the platform) to find a submerged platform, located in the middle annulus of the southwest quadrant in a black pool, filled with clear water. The rats which failed to find the submerged platform were placed on it for 10 sec. The computer calculated the escape latency (set) and distance [arbitrary units (pixels)], onto the hidden platform. Experiment 3. The rats were treated with saline or dexmedetomidine (0.3, 0.9, 3.0 or 9.0 pg/kg), 30 min before a daily training trial. Ten rats were tested in each group. Experiment 4. The rats were treated with saline or dexmedetomidine (0.3, 3.0, 30.0 or 300pg/kg), immediately (within 2 min) after a daily training trial. Ten rats were tested in each group. Statistics
Passive avoidance data (training latency, number of training trials, retention latency) was analyzed using one-way analysis of variance (ANOVA) (treatment as a variable), followed by Duncan’s test comparisons of different groups. Water maze data [escape latency and distance, as well as swimming speed (distance/latency)], was analyzed using ANOVA (treatment as a variable and training day as a covariant). RESULTS
Passive avoidance Experiment 1.
Pre-training administration of dexmedetomidine (0.3-9.0 pg/kg) did not affect the latency of first entry or the number of training trials needed to acquire the training criteria (Table 1). Analysis of variance revealed an effect of treatment in the testing latency between rats treated with saline or 0.3-300 pg/kg dexmedetomidine (F(4,40) = 9.0, The rats treated with 9.0pg/kg P < 0.05). dexmedetomidine had a shorter testing latency (assessed 24 hr after training), as compared to salinetreated rats (Table 1). The testing latency did not differ between saline- and 0.3, 0.9 or 3.0pg/kg dexmedetomidine-treated rats (Table I).
Effects of dexmedetomidine
on learning and memory
Table 1. The training latency (Latency I), the number of training
Table 2. The testing latency (Latency 2) of saline- or dexmedetomidine- (d-med) (0.3, 3.0, 30.0 or 300 pg/kg) post-training treated rats
trials (Trials) and the testing latency (Latency 2) of saline- or dexmedetomidine- (d-med) (0.3-9.0 pg/kg) pm-training treated rats Latency (set) Saline d-med 0.3 0.9 3.0 9.0 ANOVA F(4,40)
1
31+ 25 51*34 45 + 31 40 & 37 43 i 31 0.1 70.1
Trials
Latency 2 (set)
1.0 1.0 1.0 2.3 k 0.2 1.0 1.1 >O.l
360+0 350 f 10 309 f 59 360 k 0 180 * 701 9.9 10.05
Latency 2 W) 360*0 360*0 328 + 99 360*0 178 f lSO* 9.5 40.01
Saline d-med 0.3 3.0 30.0 ANOV?F(4,45)
Tbe results are expressed as mean f SD. *P < 0.05 Duncan’s test.
The results are expressed as mean f SD. *I’ c 0.05 Duncan’s test.
Experiment 2. Analysis of variance revealed the effect of treatment in the testing latency (assessed 24 hr after training and subsequent treatment with saline or 0.3-9.0 pug/kg dexmedetomidine) (F(4,45) = 9.5, P < 0.01). The rats treated with 300 pgg/kg dexmedetomidine had a shorter testing latency, as compared to saline-treated rats (P < 0.05, Duncan’s test) (Table 2). The testing latency did not differ between rats treated with saline and 0.3, 3.0 or 30 pg/kg dexmedetomidine (P > 0.1, Duncan’s test) (Table 2). Water maze task Experiment 3. Pre-training administration of dexmedetomidine increased the swimming speeds of the rats (a significant treatment effect; ANOVA F(4,669) = 9.0, P < 0.001); as compared to salinetreated rats, the effect was significant at all doses tested (0.3 pg/kg: F(1,267) = 5.3, P < 0.05; 0.9 pg/kg: F(l,267) = 10.2, P < 0.01); (3.0 pg/kg: F(l,267)=4.3, P 0.1). DISCUSSION
The difficulties in the interpretation of the effects of peripherally administered alpha, agonists include whether the effects are mediated centrally and/or peripherally, which areas of the brain are involved in
...................... ...................... ...................... ...................... ............ .......... ...................... ...................... ...................... ...................... ...................... ...................... ...................... ...... ................ ::::::::::::::::::::::
30 25
P g 20
...................... ...................... ...................... ...................... ...................... ...................... ...................... ...................... ...................... ...................... ...................... :::::::::::::::::::::: ...................... ...................... :::::::::::::::::::::: :::::::::::::::::::::: :::::::::::::::::::::: ...................... :::::::::::::::::::::: :::::::::::::::::::::: ::::::::::::::::::::::
v1 k
165
15
5 10 s 0
m
Saline d-mad
3.0
gj
d-mea
0.3
I:::::I
d-,,,ed
9.0
0
d-mad
0.9
Fig. 1. The swimming speeds (speed expressed in arbitrary units/s=) of rats, treated with saline or dexmedetomidine (0.3, 0.9, 3.0 or 9.0 pg/kg), 30 min before a daily training in the water maze task. The results are expressed as a group mean (+SEM) of 15 training trials of the water maze task.
J. SIRVI~er al.
166
Training
Saline d-med
3.0
day
@
d-med
0.3
a
d-med
90
a
d-med
0.9
Fig. 2. The acquisition of the water maze task (escape distance, expressed as arbitrary units), in rats treated with saline or dexmedetomidine (0.3,0.9,3.0 or 9.0 pg/kg), 30 min before a daily training trial. The results are expressed as a group mean of a daily training trial.
possible central effects and whether the findings are due to the effects on postsynaptic and/or presynaptic receptors of noradrenergic neurones, if central mechanisms are involved. In favour of central effects, there is clear evidence that dexmedetomidine decreases the turnover of noradrenaline in the brain, induces mydriasis and sedation tScheinin et al., 1990). However, it has become evident that interactions between peripheral mechanisms and central noradrenergic system are important in some aspects of behaviour (McCaugh, 1989; Svensson, 1987). The present rest&s suggest that pre-training administration of a subanaesthetic dose (9.0 gg/kg) of dexmedetomidine impaired the retention of passive avoidance task but the smaller doses (0.3-3.0 @g/kg)
did not affect the retention of this task. However, possible improvement of retention of passive avoidance by small doses of noradrenergic agonists can not be excluded, due to the ceiling effect (i.e. control rats showed maximum retention). Although peripheral effects of dexmedetomidine on the performance of rats in learning and memory tasks cannot be excluded, it would be tempting to speculate that decreased activity of the central noradrenergic system is involved in the deficit in retention induced by pre-training administration of 9.0 pg/kg dexmedetomidine. The turnover of noradrenaline may be affected more markedly at the dose of 9.0 pgjkg, than at the dose of 3.0 @g/kg but the turnover of dopaminergic and serotonergic systems are not affected at the
Training
m m
Saline
B
d-mad
0.3
d-med
c:::::L
d-med
300
30
day
m
d-med
3.0
Fig. 3. The acquisition of the water maze task (escape latency expressed as set) in rats treated with saline or dexmedetomidine (0.3, 3.0, 30.0 or 300 pg/kg), immediately after a daily training trial. The results are expressed as a group mean of a daily training trial.
Effects of dexmedetomidine
dose of 9.0pgg/kg. It is important to note that the acquisition of the passive avoidance task (the latency of the first entry, the number of training trials) was not affected by pre-training administration of dexmedetomidine (0.3-9.0 pg/kg), possibly excluding some non-mnemonic effects (motivation, locomotion, analgesia) of dexmedetomidine on the retention of aversively-motivated tasks. On the other hand, the acquisition of the spatial memory task was impaired, using pre-training administration of small doses (0.3 and 0.9pg/kg) of dexmedetomidine, whereas larger doses (3.0 and 9.0 pg/kg) did not markedly affect spatial learning. These results do not suggest that depressed noradrenergic activity is related to impairment of the learning of spatial reference memory tasks, because the smallest dose may not decrease the release of noradrenaline in brain and it is reasonable to believe that largest dose (9.0 pg/kg) depressed the activity of central noradrenergic neurones (MacDonald et al., 1988). The present results, showing a marked difference in a dose-response curve of pre-training administration of dexmedetomidine in the retention of passive avoidance and acquisition of the water maze, may be of some interest in the context of recent findings about the differences between the effects of noradrenergic lesions on conditioning to explicit or contextual cues of aversively-motivated tasks and on the acquisition of the spatially-cued water maze task (Selden et al., 1990a; Selden, Everitt, Jarrard and Robbins, 1990b; Selden, Robbins and Everitt, 1990~). However, it is important to note that the duration of treatment (and training) differed between the passive avoidance (a single dose) and water maze (a subchronic administration) tasks. Only the largest dose (300 pg/kg) administered immediately after training, impaired the retention of the passive avoidance task and impaired the acquisition of the water maze task. Since this dose is highly sedative and also decreases the turnover of dopamine and serotonin (MacDonald et al., 1988; Scheinin et al., 1990), it is questionable whether the effects of treatment with the drug on consolidation of memory were due to decreased activity of central noradrenergic neurones. The effects of pre-training administration of dexmedetomidine on swimming speeds in rats are worth mentioning. The present study showed a dosedependent (in a dose range 0.3-9.0 pg/kg) increase in the swimming speeds of dexmedetomidine-treated rats. Previously, it was found that a small dose (1.0 pg/kg) of guanfacine, a partial alpha, agonist, slightly increased the swimming speed, as compared to saline- or 100 pg/kg guanfacine-treated rats, whereas 10 and 100 pg/kg guanfacine did not have any significant effect on swimming speeds, as compared to saline-treated rats (Sirvio, Riekkinen Jr, Vajanto, Koivisto and Riekkinen, 1991). Furthermore, administration of clonidine or noradrenaline,
on learning and memory
167
into the lateral ventricles of the locus coeruleus, dose-dependently affected swim-motivated locomotion; smaller doses increased and larger doses decreased swimming. On the other hand, the injections of clonidine or noradrenaline, posterior to the locus coeruleus, induced opposite effects to those of injections into the locus coeruleus (Weiss, Simson, Hoffman, Ambrose, Cooper and Webster, 1986). In conclusion, systemic pre-training administration of subsedative doses of the alpha, agonist, dexmedetomidine, impaired retention of the passive avoidance task, the acquisition of the water maze task and increased swimming speed in rats. However, the dose-response curves between these effects.
of dexmedetomidine
differed
Acknowledgement-Ewen
MacDonald, Ph.D., is gratefully acknowledged for the revision of the language of the manuscript. REFERENCES
Aghajanian G. K. and VanderMaelen C. P. (1982) Alpha,adrenoceptor-mediated hyperpolarization of locus coeruleus neurons: Intracellular studies in oivo. Science 215: 13941396. Anlezark G. M., Crow T. J. and Greenway A. P. (1973) Impaired learning and decreased cortical norepinephrine after bilateral locus coeruleus lesions. Science 181: 682684. Archer T. (1983) Signalled and unsignalled avoidance impairments following noradrenaline depletion with DSP-4: An hypothesis incorporating an associative and a non-associative factor. Stand. J. Psychol. 24: 75-87. Birch P. J. and Fillenz M. (1985) The regulation of noradrenaline synthesis in central nerve terminals by autoreceptors and heteroreceptors in rat. Neurosci. Lerr. 59: 197-202.
Cahill L. and McGaugh J. L. (1990) Amygdaloid complex lesions differentially affect retention of tasks using appetitive and aversive reinforcement. Behav. Neurosci. 104: 532-543.
Cerderbaum J. M. and Aghajanian G. K. (1976) Noradrenergic neurons of the locus coeruleus: inhibition of epinephrine and activation by the alpha-antagonist piperoxane. Brain Res. 112: 413419. Crow T. J. and Wendlandt S. (1976) Impaired acquisition of a passive avoidance response after lesions induced in the locus coeruleus by 6-OH-dopamine. Nafure 259: 42-44. Dunn L. T. and Everitt B. J. (1988) Double dissociations of the effects of amygdala and insular cortex lesions on a conditioned taste aversion, passive avoidance, and neophobia in the rats using ibotenic acid. Behav. Neurosci. 102: 3-23.
Ellis M. E. and Kesner R. P. (1983) The noradrenergic system in the amygdala and aversive information processing. Behav. Neurosci. 97: 399415. Foote S. L. and Morrison J. H. (1987) Extrathalamic modulation of cortical function. A. Rev. Neurosci. 10: 67-95. Gallagher M., Kapp B. S., Pascoe J. P. and Rapp P. R. (1981) A neuropharmacology of amygdaloid systems which contribute to learning and memory. In: The Amygdaloid Complex (Ben-Ari Y., Ed.), pp. 343-354. Elsdvier/North-Holland; Amsterdam. _Hanan J. J.. Alpert J. E.. Morris R. G. M. and Iversen S. D. (i983) The effects of central catecholamine depletions on spatial learning in rats. Behav. Brain Res. 9: 83-104.
168
J. Sr~vnS et al.
MacDonald E., Scheinin H. and Scheinin M. (1988) Behavioural and neurochemical effects of medetomidine, a novel veterinary sedative. Eur. J. Pharmac. 158: 119-127. McCormick D. A. (1989) Cholinergic and noradrenergic modulation of thalamocortical processing. Trends Neuroxi. 12: 215-221. McGaugh J. L. (I 989) Involvement of hormonal and neuromodulatory systems in the regulation of memory storage. A. Rev. Neurosci. 12: 255-287.
McGaugh J. L., Introini-Collinson I. B., Nagahara A. H. and Cahill L. (1990) Involvement of the amygdaloid complex in neuromodulatory influences on memory storage..Neurosci. Biobehav. Rei. 14: 42543 1. Morris R. G. M.. Garrud P.. Rawlins J. N. P. and O’Keefe J. (1982) Place-navigation impaired in rats with hippocampal lesions. Nature 297: 365-395. Olton D. and Wenk G. L. (1987) Dementia: Animal models of the cognitive impairments produced by degeneration of the basalforebrain-cholinergic system. In: Psjchopharmacoloav. The Third Generation of the Proaress (Meltzer H. Y., Ed.). Raven Press, New “York. Randt C. T., Quartermain D., Goldstein M. and Anagnoste B. (1971) Norepinephrine biosynthesis inhibition: effects on memory in mice. Science 172: 498499. Riekkinen P. Jr, Sirvio J. and Riekkinen P. (1990) Similar memory impairments found in medial septal-vertical diagonal band of Broca and nucleus basalis lesioned rats: are memory defects induced by nucleus basalis lesions related to the degree of non-specific subcortical cell loss? Behav. Brain Res. 37: 81-88.
Riekkinen P. Jr, Sirvio J., Valjakka A., Pitklnen A., Partanen J. and Riekkinen P. J. (1990) The effects of concurrent manipulations of cholinergic and noradrenergic systems on neocortical EEG and spatial learning. Behav. Neural. Biol. 54: 204-210.
Savola J.-M. and Virtanen R. (1991) Central ar-adrenoceptors are highly stereoselective for dexmedetomidine, the dextro enantiomer of medetomidine. Eur. J. Pharmac. 195: 193-199.
Scheinin H., Virtanen R., MacDonald E., Scheinin M. and Lammintausta R. (1990) Medetomidine-a novel a2adrenoceptor agonist: a review of its pharmacodynamic effects. Prog. Neuropsychopharmac. 13: 63565 1. Selden N. W. R., Cole B., Everitt B. J. and Robbins T. W. (1990a) Damage to coeruleo-cortical noradrenergic projections impairs locally cued, but enhances spatially cued water maze acquisition. Behav. Brain Res. 39: 29-51.
Selden N. W. R., Everitt B. J., Jarrard L. E. and Robbins T. W. (1990b) Complementary roles for the amygdala and hippocampus in aversive conditioning to explicit and contextual cues. Neuroscience 42: 335-350. Selden N. W. R., Robbins R. W. and Everitt B. J. (199Oc) Dorsal noradrenergic bundle lesions impair conditioning to explicit but enhance conditioning to contextual aversive cues: support for an attentional theory of central noradrenergic function. J. Neurosci. 10: 53 l-539. Sirviij J., Riekkinen P. Jr, Vajanto I., Koivisto E. and Riekkinen P. J. (1991) The effects of guanfacine, alpha., agonist, on the performance of young and aged rats in spatial navigation task. Behav. Neural. Biol. 56: 101-107. Stein L., Belluzzi J. D. and Wise C. D. (1975) Memory enhancement by central administration of norepinephrine. Brain Res. 84: 329-335. Sutherland R. J., Kolb B. and Whishaw I. Q. (1982) Spatial mapping: definitive disruption by hippocampal or medial frontal cortex damage in the rat. Neurosci. Left. 31: 271-276.
Sutherland R. J. and McDonald R. J. (1990) Hippocampus, amygdala and memory deficits in rats. Behav. Brain Res. 37: 57-80.
Sutherland R. J. and Rodrigues A. J. (1989) The role of fornixifimbria and some related subcortical structures in place learning and memory. Behav. Brain Res. 32: 265-277.
Svensson T. (1987) Peripheral, autonomic regulation of locus coeruleus noradrenergic neurons in the brain: Putative implications for psychiatry and psychopharmacology. Psychopharmact?o~y 92: -1-7. _ _ _ Valiakka A.. Riekkinen P. Jr. Sirviij J.. Nieminen S.. Airaksinen’M., Miettinen R. and Riekkinkn P. (1990) The effects of dorsal noradrenergic bundle lesions on spatial learning, locomotor activity, and reaction to novelty. Behav. Neural. Biol. 54: 323-329.
Virtanen R., Savola J.-M., Saano V. and Nyman L. (1988) Characterization of the selectivity, specificity and potency of medetomidine as an a,-adrenoceptor agonist. Eur. J. Pharmac. 150: 9-14. Weiss J. M., Simson P. G., Hoffman L. J., Ambrose M. J., Cooper S. C. and Webster A. (1986) Infusion of adrenergic receptor agonist and antagonist into the locus coelureus and ventricular system of the brain. Neuropharmacology 25: 367-384.
0028-3908/92$5.00+ 0.00 Copyright 0 1992Pergamon Press plc
THE EFFECTS OF DEXMEDETOMIDINE, AN ALPHA, AGONIST, ON LEARNING AND MEMORY, ASSESSED USING PASSIVE AVOIDANCE AND WATER MAZE TASKS IN RATS J. SIRVI~, P. RIEKKINENJR, T. EKONSALO,R. LAMMINTAUSTAand P. J. RIEKKINEN Department of Neurology, University of Kuopio, P. 0. Box 6, SF-7021 1 Kuopio, Finland and Orion Corporation,
Farrnos Research Group, Turku, Finland
(Accepted 19 September
1991)
Summary-The effects of dexmedetomidine, a specific and potent alpha, agonist, on the performance of rats in passive avoidance and water maze tasks were studied. Pre-training administration of subanaesthetic dose (9.0 pg/kg) of dexmedetomidine impaired the retention of the passive avoidance task (assessed 24 hr after training) but it did not affect the training of this task. Smaller doses (0.3,O.g and 3.0 pg/kg) did not affect the training or retention of this aversively motivated task. On the other hand, pre-training administration of 0.3 and 0.9 pg/kg dexmedetomidine impaired the acquisition of the water maze task, whereas larger doses (3.0 and g.Opgg/kg) had no significant effect on spatial learning. Pre-training administration of dexmedetomidine (0.3-9.0 pg/kg) increased swimming speed in rats. Only a large dose (300 lg/kg) of dexmedetomidine, administered immediately after training, impaired the retention of the passive avoidance task and the acquisition of the water maze task. These data agree with previous findings that pharmacological manipulation of the noradrenergic system affects the retention of aversively-motivated (passive avoidance) tasks. The present results suggest that the dose-response curve of dexmedetomidine for impairment of learning/memory differs between the passive avoidance and water maze tasks. Key words-alpha-2
adrenoceptor,
dexmedetomidine,
Aversively motivated (e.g. passive avoidance) and spatially cued (e.g. water maze) tasks have been widely
used in behavioural pharmacology (Olton and Wenk, 1987). These tasks partly differ in their underlying neural substrates; lesions of the amygdala impair the acquisition and retention of the inhibitory avoidance task but do not impair the acquisition of the water maze task (Cahill and McGaugh, 1990; Dunn and Everitt, 1988; Sutherland and McDonald, 1990). The learning and memory of a spatial task is critically dependent on the hippocampus and its connections with cortical and subcortical structures (Morris, Garrud, Rawlins and O’Keefe, 1982; Sutherland, Kolb and Whishaw, 1982; Sutherland and Rodriguez, 1989; Sutherland and McDonald, 1990). Anatomical and physiological studies suggest that noradrenergic neurones, projecting from the locus coeruleus to the forebrain, play a role in cognitive functions (Foote and Morrison, 1987; McCormick, 1989). Previously, lesions of the central noradrenergic system or the inhibition of the synthesis of noradrenaline have been found to impair learning/memory, assessed by the inhibitory avoidance task or different mazes (Anlezark, Crow and Greenway, 1973; Archer, 1983; Crow and Wendlandt, 1976; Randt, Quartermain, Goldstein and Anagnoste, 1971; Stein, Belluzi and Wise, 1975). Furthermore, psychopharmacological studies indicate that the noradrenergic system in the amygdala is involved in mechanisms of acquisition
learning, memory, noradrenergic systems.
and retrieval of memory, assessed using avoidance tasks (Gallagher, Kapp, Pascoe, Rapp, 1981; Ellis and Kesner, 1983; McGaugh, 1989; McGaugh, IntroiniCollison, Nagahara and Cahill, 1990). On the other hand, spatially cued learning is not impaired or may even be facilitated, by partial or extensive depletion of noradrenaline in the forebrain in adult rats (Hagan, Albert, Morris and Iversen, 1983; Riekkinen Jr, Sirviii, Valjakka, Pitkgnen, Partanen, Riekkinen, 1990; Selden, Cole, Everitt and Robbins, 1990a; Valjakka, Riekkinen Jr, SirviG, Nieminen, Airaksinen, Miettinen and Riekkinen, 1990). Neurochemical and electrophysiological studies suggest that the firing rate of the noradrenergic neurones in the locus coeruleus and the release of noradrenaline is regulated by alpha, autoreceptors (Cederbaum and Aghajanian, 1976; Birch and Fellenz, 1985). These receptors mediate the autoinhibition of noradrenergic neurones and the activation of these receptors decreases the turnover of noradrenaline (Aghajanian and VanderMaelen, 1982). The aim of the present study was to investigate the effects of pharmacologically decreased activity of noradrenergic neurones on learning and memory. Thus, the effects of dexmedetomidine, a selective and potent alpha, agonist (Savola and Virtanen, 1991; Scheinin, Virtanen, MacDonald, Scheinin and Lammintausta, 1990; Virtanen, Savola, Saano and 163
J. SIRVld et ai.
164
Nyman, 1988) on the performance of rats in passive avoidance and water maze tasks were studied. In order to differentiate between the effects of treatment with drugs on the acquisition and consolidation processes of memory, both pre- and post-training administrations of drug were used in these studies. METHODS
Animals
Male Wistar rats (27%35Og) were used in these experiments. The animals were housed in a temperature- (2O~C), humidity- (50-60%) and light- (lights on 0700-2100 hr) controlled environment, with free access to food and water. Drug
Dexmedetomidine (Orion Corporation, Farmos RCD Pharmaceuticals, Turku, Finland) was dissolved in saline and injected subcutaneously (s.c. 0.5 ml/kg). Controls were treated with saline (0.5 ml/kg). According to previous studies, dexmedetomidine has been shown to be a selective agonist of alpha, adrenoceptors and decrease the turnover of noradrenaline dose-dependently (l-300 pg/kg). Obvious sedation is induced at the doses of > 10 pg/kg. At doses of > 10 pg/kg, the turnover of serotonergic and dopaminergic neurones in the brain is also affected (Savola and Virtanen, 1991; MacDonald, Scheinin and Scheinin, 1988; Scheinin et al., 1990). For the estimation of sedation, visual inspection of the mobility of rats in their home cages was used, as well as measures of motor activity in behavioural tasks (the training latency of passive avoidance task and swimming speed in the water maze task). Passive avoidance experiments
The passive avoidance apparatus consisted of a rectangular plexiglass box, divided into dark and lighted compartments by a sliding guillotine door. The dark compartment had a metal grid floor, to which a shock generator (Campden Instruments Ltd, U.K.) was connected. In the training of the task, the rat was placed to the lit side. After entry into the dark compartment, a 3.0 mA electric shock was delivered to the feet of the rat. The shock remained on until the rat returned to the lit side. Training continued until a rat remained on the lit side for 60sec. Latency to the first entry into the dark chamber (training latency) and the number of re-entries (training trials) were recorded by the experimenter. In the retention test, the rat was placed on the lit side and, after 60 set the door was opened. The session continued until the rat entered the dark compartment (maximum time 360 set). The latency to enter into the dark chamber (testing latency) was recorded. Experiment 1. The rats were treated with saline or dexmedetomidine (0.3, 0.9, 3.0 or 9.0 pg/kg), 30 min before the training of the passive avoidance task. The rats were not treated immediately before the retention
test, which was performed 24 hr after training, as described above. Nine rats were tested in each group. Experiment 2. The rats were treated with saline or dexmedetomidine (0.3, 3.0, 30.0 or 300pg/kg), immediately (within 2min) after the training of the passive avoidance task. The rats were not treated immediately before the retention test, which was performed 24 hr after training. Ten rats were tested in each group. Water maze experiments
The water maze apparatus (purchased from San Diego Instruments, San Diego, California) has been described in detail previously (Riekkinen Jr, Sirviii and Riekkinen, 1990). During the training period of the water maze task, the rat was trained once a day for 15 days (maximum trial duration 70 set, 10 set reinforcement on the platform) to find a submerged platform, located in the middle annulus of the southwest quadrant in a black pool, filled with clear water. The rats which failed to find the submerged platform were placed on it for 10 sec. The computer calculated the escape latency (set) and distance [arbitrary units (pixels)], onto the hidden platform. Experiment 3. The rats were treated with saline or dexmedetomidine (0.3, 0.9, 3.0 or 9.0 pg/kg), 30 min before a daily training trial. Ten rats were tested in each group. Experiment 4. The rats were treated with saline or dexmedetomidine (0.3, 3.0, 30.0 or 300pg/kg), immediately (within 2 min) after a daily training trial. Ten rats were tested in each group. Statistics
Passive avoidance data (training latency, number of training trials, retention latency) was analyzed using one-way analysis of variance (ANOVA) (treatment as a variable), followed by Duncan’s test comparisons of different groups. Water maze data [escape latency and distance, as well as swimming speed (distance/latency)], was analyzed using ANOVA (treatment as a variable and training day as a covariant). RESULTS
Passive avoidance Experiment 1.
Pre-training administration of dexmedetomidine (0.3-9.0 pg/kg) did not affect the latency of first entry or the number of training trials needed to acquire the training criteria (Table 1). Analysis of variance revealed an effect of treatment in the testing latency between rats treated with saline or 0.3-300 pg/kg dexmedetomidine (F(4,40) = 9.0, The rats treated with 9.0pg/kg P < 0.05). dexmedetomidine had a shorter testing latency (assessed 24 hr after training), as compared to salinetreated rats (Table 1). The testing latency did not differ between saline- and 0.3, 0.9 or 3.0pg/kg dexmedetomidine-treated rats (Table I).
Effects of dexmedetomidine
on learning and memory
Table 1. The training latency (Latency I), the number of training
Table 2. The testing latency (Latency 2) of saline- or dexmedetomidine- (d-med) (0.3, 3.0, 30.0 or 300 pg/kg) post-training treated rats
trials (Trials) and the testing latency (Latency 2) of saline- or dexmedetomidine- (d-med) (0.3-9.0 pg/kg) pm-training treated rats Latency (set) Saline d-med 0.3 0.9 3.0 9.0 ANOVA F(4,40)
1
31+ 25 51*34 45 + 31 40 & 37 43 i 31 0.1 70.1
Trials
Latency 2 (set)
1.0 1.0 1.0 2.3 k 0.2 1.0 1.1 >O.l
360+0 350 f 10 309 f 59 360 k 0 180 * 701 9.9 10.05
Latency 2 W) 360*0 360*0 328 + 99 360*0 178 f lSO* 9.5 40.01
Saline d-med 0.3 3.0 30.0 ANOV?F(4,45)
Tbe results are expressed as mean f SD. *P < 0.05 Duncan’s test.
The results are expressed as mean f SD. *I’ c 0.05 Duncan’s test.
Experiment 2. Analysis of variance revealed the effect of treatment in the testing latency (assessed 24 hr after training and subsequent treatment with saline or 0.3-9.0 pug/kg dexmedetomidine) (F(4,45) = 9.5, P < 0.01). The rats treated with 300 pgg/kg dexmedetomidine had a shorter testing latency, as compared to saline-treated rats (P < 0.05, Duncan’s test) (Table 2). The testing latency did not differ between rats treated with saline and 0.3, 3.0 or 30 pg/kg dexmedetomidine (P > 0.1, Duncan’s test) (Table 2). Water maze task Experiment 3. Pre-training administration of dexmedetomidine increased the swimming speeds of the rats (a significant treatment effect; ANOVA F(4,669) = 9.0, P < 0.001); as compared to salinetreated rats, the effect was significant at all doses tested (0.3 pg/kg: F(1,267) = 5.3, P < 0.05; 0.9 pg/kg: F(l,267) = 10.2, P < 0.01); (3.0 pg/kg: F(l,267)=4.3, P 0.1). DISCUSSION
The difficulties in the interpretation of the effects of peripherally administered alpha, agonists include whether the effects are mediated centrally and/or peripherally, which areas of the brain are involved in
...................... ...................... ...................... ...................... ............ .......... ...................... ...................... ...................... ...................... ...................... ...................... ...................... ...... ................ ::::::::::::::::::::::
30 25
P g 20
...................... ...................... ...................... ...................... ...................... ...................... ...................... ...................... ...................... ...................... ...................... :::::::::::::::::::::: ...................... ...................... :::::::::::::::::::::: :::::::::::::::::::::: :::::::::::::::::::::: ...................... :::::::::::::::::::::: :::::::::::::::::::::: ::::::::::::::::::::::
v1 k
165
15
5 10 s 0
m
Saline d-mad
3.0
gj
d-mea
0.3
I:::::I
d-,,,ed
9.0
0
d-mad
0.9
Fig. 1. The swimming speeds (speed expressed in arbitrary units/s=) of rats, treated with saline or dexmedetomidine (0.3, 0.9, 3.0 or 9.0 pg/kg), 30 min before a daily training in the water maze task. The results are expressed as a group mean (+SEM) of 15 training trials of the water maze task.
J. SIRVI~er al.
166
Training
Saline d-med
3.0
day
@
d-med
0.3
a
d-med
90
a
d-med
0.9
Fig. 2. The acquisition of the water maze task (escape distance, expressed as arbitrary units), in rats treated with saline or dexmedetomidine (0.3,0.9,3.0 or 9.0 pg/kg), 30 min before a daily training trial. The results are expressed as a group mean of a daily training trial.
possible central effects and whether the findings are due to the effects on postsynaptic and/or presynaptic receptors of noradrenergic neurones, if central mechanisms are involved. In favour of central effects, there is clear evidence that dexmedetomidine decreases the turnover of noradrenaline in the brain, induces mydriasis and sedation tScheinin et al., 1990). However, it has become evident that interactions between peripheral mechanisms and central noradrenergic system are important in some aspects of behaviour (McCaugh, 1989; Svensson, 1987). The present rest&s suggest that pre-training administration of a subanaesthetic dose (9.0 gg/kg) of dexmedetomidine impaired the retention of passive avoidance task but the smaller doses (0.3-3.0 @g/kg)
did not affect the retention of this task. However, possible improvement of retention of passive avoidance by small doses of noradrenergic agonists can not be excluded, due to the ceiling effect (i.e. control rats showed maximum retention). Although peripheral effects of dexmedetomidine on the performance of rats in learning and memory tasks cannot be excluded, it would be tempting to speculate that decreased activity of the central noradrenergic system is involved in the deficit in retention induced by pre-training administration of 9.0 pg/kg dexmedetomidine. The turnover of noradrenaline may be affected more markedly at the dose of 9.0 pgjkg, than at the dose of 3.0 @g/kg but the turnover of dopaminergic and serotonergic systems are not affected at the
Training
m m
Saline
B
d-mad
0.3
d-med
c:::::L
d-med
300
30
day
m
d-med
3.0
Fig. 3. The acquisition of the water maze task (escape latency expressed as set) in rats treated with saline or dexmedetomidine (0.3, 3.0, 30.0 or 300 pg/kg), immediately after a daily training trial. The results are expressed as a group mean of a daily training trial.
Effects of dexmedetomidine
dose of 9.0pgg/kg. It is important to note that the acquisition of the passive avoidance task (the latency of the first entry, the number of training trials) was not affected by pre-training administration of dexmedetomidine (0.3-9.0 pg/kg), possibly excluding some non-mnemonic effects (motivation, locomotion, analgesia) of dexmedetomidine on the retention of aversively-motivated tasks. On the other hand, the acquisition of the spatial memory task was impaired, using pre-training administration of small doses (0.3 and 0.9pg/kg) of dexmedetomidine, whereas larger doses (3.0 and 9.0 pg/kg) did not markedly affect spatial learning. These results do not suggest that depressed noradrenergic activity is related to impairment of the learning of spatial reference memory tasks, because the smallest dose may not decrease the release of noradrenaline in brain and it is reasonable to believe that largest dose (9.0 pg/kg) depressed the activity of central noradrenergic neurones (MacDonald et al., 1988). The present results, showing a marked difference in a dose-response curve of pre-training administration of dexmedetomidine in the retention of passive avoidance and acquisition of the water maze, may be of some interest in the context of recent findings about the differences between the effects of noradrenergic lesions on conditioning to explicit or contextual cues of aversively-motivated tasks and on the acquisition of the spatially-cued water maze task (Selden et al., 1990a; Selden, Everitt, Jarrard and Robbins, 1990b; Selden, Robbins and Everitt, 1990~). However, it is important to note that the duration of treatment (and training) differed between the passive avoidance (a single dose) and water maze (a subchronic administration) tasks. Only the largest dose (300 pg/kg) administered immediately after training, impaired the retention of the passive avoidance task and impaired the acquisition of the water maze task. Since this dose is highly sedative and also decreases the turnover of dopamine and serotonin (MacDonald et al., 1988; Scheinin et al., 1990), it is questionable whether the effects of treatment with the drug on consolidation of memory were due to decreased activity of central noradrenergic neurones. The effects of pre-training administration of dexmedetomidine on swimming speeds in rats are worth mentioning. The present study showed a dosedependent (in a dose range 0.3-9.0 pg/kg) increase in the swimming speeds of dexmedetomidine-treated rats. Previously, it was found that a small dose (1.0 pg/kg) of guanfacine, a partial alpha, agonist, slightly increased the swimming speed, as compared to saline- or 100 pg/kg guanfacine-treated rats, whereas 10 and 100 pg/kg guanfacine did not have any significant effect on swimming speeds, as compared to saline-treated rats (Sirvio, Riekkinen Jr, Vajanto, Koivisto and Riekkinen, 1991). Furthermore, administration of clonidine or noradrenaline,
on learning and memory
167
into the lateral ventricles of the locus coeruleus, dose-dependently affected swim-motivated locomotion; smaller doses increased and larger doses decreased swimming. On the other hand, the injections of clonidine or noradrenaline, posterior to the locus coeruleus, induced opposite effects to those of injections into the locus coeruleus (Weiss, Simson, Hoffman, Ambrose, Cooper and Webster, 1986). In conclusion, systemic pre-training administration of subsedative doses of the alpha, agonist, dexmedetomidine, impaired retention of the passive avoidance task, the acquisition of the water maze task and increased swimming speed in rats. However, the dose-response curves between these effects.
of dexmedetomidine
differed
Acknowledgement-Ewen
MacDonald, Ph.D., is gratefully acknowledged for the revision of the language of the manuscript. REFERENCES
Aghajanian G. K. and VanderMaelen C. P. (1982) Alpha,adrenoceptor-mediated hyperpolarization of locus coeruleus neurons: Intracellular studies in oivo. Science 215: 13941396. Anlezark G. M., Crow T. J. and Greenway A. P. (1973) Impaired learning and decreased cortical norepinephrine after bilateral locus coeruleus lesions. Science 181: 682684. Archer T. (1983) Signalled and unsignalled avoidance impairments following noradrenaline depletion with DSP-4: An hypothesis incorporating an associative and a non-associative factor. Stand. J. Psychol. 24: 75-87. Birch P. J. and Fillenz M. (1985) The regulation of noradrenaline synthesis in central nerve terminals by autoreceptors and heteroreceptors in rat. Neurosci. Lerr. 59: 197-202.
Cahill L. and McGaugh J. L. (1990) Amygdaloid complex lesions differentially affect retention of tasks using appetitive and aversive reinforcement. Behav. Neurosci. 104: 532-543.
Cerderbaum J. M. and Aghajanian G. K. (1976) Noradrenergic neurons of the locus coeruleus: inhibition of epinephrine and activation by the alpha-antagonist piperoxane. Brain Res. 112: 413419. Crow T. J. and Wendlandt S. (1976) Impaired acquisition of a passive avoidance response after lesions induced in the locus coeruleus by 6-OH-dopamine. Nafure 259: 42-44. Dunn L. T. and Everitt B. J. (1988) Double dissociations of the effects of amygdala and insular cortex lesions on a conditioned taste aversion, passive avoidance, and neophobia in the rats using ibotenic acid. Behav. Neurosci. 102: 3-23.
Ellis M. E. and Kesner R. P. (1983) The noradrenergic system in the amygdala and aversive information processing. Behav. Neurosci. 97: 399415. Foote S. L. and Morrison J. H. (1987) Extrathalamic modulation of cortical function. A. Rev. Neurosci. 10: 67-95. Gallagher M., Kapp B. S., Pascoe J. P. and Rapp P. R. (1981) A neuropharmacology of amygdaloid systems which contribute to learning and memory. In: The Amygdaloid Complex (Ben-Ari Y., Ed.), pp. 343-354. Elsdvier/North-Holland; Amsterdam. _Hanan J. J.. Alpert J. E.. Morris R. G. M. and Iversen S. D. (i983) The effects of central catecholamine depletions on spatial learning in rats. Behav. Brain Res. 9: 83-104.
168
J. Sr~vnS et al.
MacDonald E., Scheinin H. and Scheinin M. (1988) Behavioural and neurochemical effects of medetomidine, a novel veterinary sedative. Eur. J. Pharmac. 158: 119-127. McCormick D. A. (1989) Cholinergic and noradrenergic modulation of thalamocortical processing. Trends Neuroxi. 12: 215-221. McGaugh J. L. (I 989) Involvement of hormonal and neuromodulatory systems in the regulation of memory storage. A. Rev. Neurosci. 12: 255-287.
McGaugh J. L., Introini-Collinson I. B., Nagahara A. H. and Cahill L. (1990) Involvement of the amygdaloid complex in neuromodulatory influences on memory storage..Neurosci. Biobehav. Rei. 14: 42543 1. Morris R. G. M.. Garrud P.. Rawlins J. N. P. and O’Keefe J. (1982) Place-navigation impaired in rats with hippocampal lesions. Nature 297: 365-395. Olton D. and Wenk G. L. (1987) Dementia: Animal models of the cognitive impairments produced by degeneration of the basalforebrain-cholinergic system. In: Psjchopharmacoloav. The Third Generation of the Proaress (Meltzer H. Y., Ed.). Raven Press, New “York. Randt C. T., Quartermain D., Goldstein M. and Anagnoste B. (1971) Norepinephrine biosynthesis inhibition: effects on memory in mice. Science 172: 498499. Riekkinen P. Jr, Sirvio J. and Riekkinen P. (1990) Similar memory impairments found in medial septal-vertical diagonal band of Broca and nucleus basalis lesioned rats: are memory defects induced by nucleus basalis lesions related to the degree of non-specific subcortical cell loss? Behav. Brain Res. 37: 81-88.
Riekkinen P. Jr, Sirvio J., Valjakka A., Pitklnen A., Partanen J. and Riekkinen P. J. (1990) The effects of concurrent manipulations of cholinergic and noradrenergic systems on neocortical EEG and spatial learning. Behav. Neural. Biol. 54: 204-210.
Savola J.-M. and Virtanen R. (1991) Central ar-adrenoceptors are highly stereoselective for dexmedetomidine, the dextro enantiomer of medetomidine. Eur. J. Pharmac. 195: 193-199.
Scheinin H., Virtanen R., MacDonald E., Scheinin M. and Lammintausta R. (1990) Medetomidine-a novel a2adrenoceptor agonist: a review of its pharmacodynamic effects. Prog. Neuropsychopharmac. 13: 63565 1. Selden N. W. R., Cole B., Everitt B. J. and Robbins T. W. (1990a) Damage to coeruleo-cortical noradrenergic projections impairs locally cued, but enhances spatially cued water maze acquisition. Behav. Brain Res. 39: 29-51.
Selden N. W. R., Everitt B. J., Jarrard L. E. and Robbins T. W. (1990b) Complementary roles for the amygdala and hippocampus in aversive conditioning to explicit and contextual cues. Neuroscience 42: 335-350. Selden N. W. R., Robbins R. W. and Everitt B. J. (199Oc) Dorsal noradrenergic bundle lesions impair conditioning to explicit but enhance conditioning to contextual aversive cues: support for an attentional theory of central noradrenergic function. J. Neurosci. 10: 53 l-539. Sirviij J., Riekkinen P. Jr, Vajanto I., Koivisto E. and Riekkinen P. J. (1991) The effects of guanfacine, alpha., agonist, on the performance of young and aged rats in spatial navigation task. Behav. Neural. Biol. 56: 101-107. Stein L., Belluzzi J. D. and Wise C. D. (1975) Memory enhancement by central administration of norepinephrine. Brain Res. 84: 329-335. Sutherland R. J., Kolb B. and Whishaw I. Q. (1982) Spatial mapping: definitive disruption by hippocampal or medial frontal cortex damage in the rat. Neurosci. Left. 31: 271-276.
Sutherland R. J. and McDonald R. J. (1990) Hippocampus, amygdala and memory deficits in rats. Behav. Brain Res. 37: 57-80.
Sutherland R. J. and Rodrigues A. J. (1989) The role of fornixifimbria and some related subcortical structures in place learning and memory. Behav. Brain Res. 32: 265-277.
Svensson T. (1987) Peripheral, autonomic regulation of locus coeruleus noradrenergic neurons in the brain: Putative implications for psychiatry and psychopharmacology. Psychopharmact?o~y 92: -1-7. _ _ _ Valiakka A.. Riekkinen P. Jr. Sirviij J.. Nieminen S.. Airaksinen’M., Miettinen R. and Riekkinkn P. (1990) The effects of dorsal noradrenergic bundle lesions on spatial learning, locomotor activity, and reaction to novelty. Behav. Neural. Biol. 54: 323-329.
Virtanen R., Savola J.-M., Saano V. and Nyman L. (1988) Characterization of the selectivity, specificity and potency of medetomidine as an a,-adrenoceptor agonist. Eur. J. Pharmac. 150: 9-14. Weiss J. M., Simson P. G., Hoffman L. J., Ambrose M. J., Cooper S. C. and Webster A. (1986) Infusion of adrenergic receptor agonist and antagonist into the locus coelureus and ventricular system of the brain. Neuropharmacology 25: 367-384.
