Neuropeptides (1992) 23,87-92 0 Longman GroupUK Ltd 1992

The NMDA-Receptor Antagonist Dizocilpine (MK-801) Suppresses the Memory Facilitatory Action of Thyrotropin-releasing Hormone S. A. KASPAROV and B. A. CHIZH Department of Pharmacology, Moscow Sechenov Medical Academy, Moscow, B. Pirogovskaya Z/6, 199881, Moscow, Russia (Reprint requests to SK)

Abstract -Thyrotropin-releasing hormone (TRH) has been shown to improve memory both in animal amnesia models and in humans. In an earlier study we have found that respiratory stimulant action of TRH is mediated through the N-methyl-D-aspattate (NMDA) receptors. Since brain NMDA receptors are implicated in neuronal plasticity, we have investigated whether the non-competitive NMDA antagonist dizocilpine antagonizes the memoryfacilitatory action of TRH. TRH (5 mg/kg i.p.) increased passive avoidance scores in the rats in both memory retention and retrieval tests. Dizocilpine (0.2 mg/kg i.p.) caused no significant performance changes. However, it completely blocked the improvement of retention caused by TRH and reversed its effect on retrieval. We conclude that the facilitatory effect of TRH on avoidance retention and retrieval may be mediated by NMDA receptors.

found in rats, not subjected to any amnestic treatment (3, 4). Memory improvement with TRH has recently been demonstrated also in clinical conditions in Alzheimer’s disease patients (5). The mechanisms of this effect are not known. In a previous study we have found that another effect of TIW, namely its respiratory-stimulant action, is effectively antagonized by competitive (D-Ap5) and non-competitive (dizocilpine, M&801) NMDA-receptors blockers (6). There is a considerable evidence for the involvement of NMDA receptors in the processes of memory and learning (7-l 0). Therefore, we have tested whether dizocilpine is able to antagonize the memory improvement caused by TRH and thus if the NMDA receptors are involved in its effect.

Introduction Thyrotropin-releasing hormone (TRH) has a variety of neurophysiological effects probably independent from its original hormonal action. One of them, improvement of memory, is of considerable interest and potential clinical importance. Using various models of amnesia this effect has been shown by Oka et al (1, 2) in rats and mice. A tendency to improve the performance in the passive avoidance test in the 12-armed radial maze test has also been Date received 27 February 1992 Date accepted 18 May 1992 Correspondence to: S. A. Kasparov, Department of Pharmacology, Moscow Sechenov Medical Academy, Moscow, B. Pirogovskaya 216,199881, Moscow, Russia

87

88 Previously statistically significant improvement by TRH of active and passive avoidance was found in animals subjected to different amnestic procedures (1, 2). However, as dizocilpine and other NMDA antagonists are able to impair learning and memory themselves (7-10) we decided not to use any model of amnesia in these experiments. Instead, we attempted to demonstrate the effect of TRH in normal animals in which a low dose of dizocilpine did not impair avoidance performance. Materials and Methods Male ICWWistar hybrid rats (250-280 g) were used. Rats were kept in 12 h light-dark cycle with free access to food and water, 6-8 animals per cage. All experiments were performed between 6-9 h of the light period. Each experimental group included in the present study consisted of the animals from at least 3 different offsprings to minimize the influence of spontaneous genetic variations. Passive avoidance was tested using a light-dark step-through apparatus (symmetrical, 2 compartments of 15 x 35 cm, 15 cm height) with a grid floor and a sliding door between the compartments. The periods of time spent in the compartments were recorded by a computer. Before the trials each animal for two consecutive days was injected once a day with saline and placed into the apparatus for 3 min to adapt it to the procedure. Passive avoidance traininghestingprocedure During the training session the animal was placed into the light compartment, heading its blind end. From that moment on, its relocations from one compartment to the other were recorded for 3 min. Two parameters were measured: the latency of the first entering into the dark part (step-through latency, STL) and darkness preference ratio (DPR), the ratio between the time periods spent in the dark and light compartments. After 180 s the sliding door was closed and in the dark part the animal was delivered one electric shock (50 V, 0.2 s). This stimulus produced flinching but usually no vocalization or strong running. If the animal did not enter the dark compartment at the end of 180 s observation period it was gently placed there by hand.

NEUROPEPTIDES

Experimental protocols Protocal A. In the preliminary experiments we attempted to find an optimal training procedure to demonstrate the effect of TRH on memory retention. In these experiments TRH (2.5-l 0 mg/kg i.p. before the training session) failed to significantly improve passive avoidance scores measured 24 h after a single training session with different intensity of negative reinforcement (l-5 footshocks). Therefore, a continuous 6&y training and testing protocol was employed. According to this protocol rats were daily injected with saline, TRH, dizocilpine or TRH plus dizocilpine and then subjected to one testing/training session as described above. On day 7 the rats were left in their cages and on day 8 all groups were injected with saline and the avoidance was measured to estimate the retention of the memory trace. Protocol B. This protocol was applied to more specifically test the effects of the substances on the memory retrieval. Four groups of rats were injected with saline and subjected to a single 180~second recording session upon the termination of which they were delivered 1 footshock. 24 hours later they were injected with saline, TRH, dizocilpine or TRH + dizocilpine and then the passive avoidance was registered as described above. In both protocols the rats of all groups were tested in parallel. The number of animals used in the protocol A were: control, n = 48; TRH, n = 43; dizocilpine, n = 22; TRH + dizocilpine, n = 47, in protocol B: control, n= 30; TRH, n= 30; dizocilpine, n = 20; TRH + dizocilpine, n = 20. Nociception test. The animal was placed in a cage with a grid floor and current pulses (0.5 s, 20 s interstimulus interval) of different amplitude were delivered through the floor. Threshold voltage necessary to evoke an escape response (flinching, running or vocalization) was recorded 15 min after the i.p. administration of the substances. Each group consisted of 6 animals. Administration of drugs TRH (Sigma, USA) and dizocilpine (Merck, Sharp and Dohme, UK) were dissolved in 0.2 ml saline and administered i.p. 15 min before placement into the apparatus. In series where TRH and dizocilpine were

DJZOCILPINE (MK-801) SUPPRESSES

MEMORY FAClLITATORY

administered simultaneously they were injected in separate syringes into different sides of the abdominal cavity. Care was taken to minimize stress caused by the injection. In the preliminary experiments a dose of 5 mg/kg of TRH was found to be more effective as compared to 2.5 of 10 m&g. The dose of dizocilpine (0.2 mgikg) was chosen so as to produce just minimal signs of the ‘phencyclidine-like’ behavior and not to cause significant impairment ofthe avoidance (see Results). Statistical procedure

The changes in both recorded parameters (STL and DPR) were estimated using an alternative approach because after a single training session the distribution of the values significantly differed from a normal one. Hence, on the day prior to the first training episode, the mean values and standard deviations of STL and DPR were calculated. These were similar in all groups, and therefore were pooled. In groups of the protocol A they were: STL: M = 8.4 s, SD = 13.6 s; DPR: M = 83.1%, SD = 14.1%, n = 160, in the protocol B - STL: M = 12.3 s, SD = 16.3 s; DPR: M = 74.3%, SD = 17.1%, n = 100. Consecutively, in protocol A avoidance was scored as positive if STL exceeded 49.2 s (M + 3 s.d.) or DPR was less than 40.8% (M - 3 s.d.). The equivalent threshold values for the protocol B were: STL: M > 61.3 set or DPR < 23%. In most cases STL criterion was used for its better reproducibility. The significance of the difference between the groups was evaluated using the Fisher exact probability test. The significance of the changes in the nociception test was determined by Student’s t-test.

Results Protocol A Days O-5, testing and training. The values of STL on day 0 before the tirst training episode were similar in all groups (M + m): control, 6.93 f 1.42; TRH, 6.11 f 0.78; dizocilpine, 10.24 f 3.71; TRH + dizocilpine, 8.18 f 2.31 s (P > 0.1 in all cases). Similarly, DRP did not differ between the groups: control, 91.48 + 1.52; TRH, 89.29 f 2.35; dizocilpine, 89.38 + 2.79; TRH + dizocilpine, 87.89 f 2.56%, (P > 0.1 in all cases).

89

ACTION OF TRH

80

60

I Y

2

5

DAYS

Fig. 1 Effect of TRH (5 mgkg), dizocilpine (0.2 rngkm their combination on darkness avoidance in rats (STL criterion). Protocol A, testing and training. Significance (exact Fisher test) as compared to the saline-treated controls: J - 0.05 < p < 0.1, JJ - 0.01 < p < 0.05, JJJ - p < 0.01.

Hence, the administration of TRH, dizocilpine or their combination caused no significant change in the spontaneous behavior of the untrained animals in the light-dark box. From day 1 the control rats demonstrated progressively increasing avoidance scores (Fig. 1). The degree of avoidance as estimated from STL increased steadily from day 1 to day 5. TRH, injected daily, tended to increase the avoidance as compared to controls starting from day 1 to day 5 (Fig. 1). However, the difference in the percentage of positive avoidance scores according to STL was statistically significant (p < 0.05) only on day 3, when it reached plateau, while the animals of the control group approached the same degree of avoidance on day 5. According to the DPR the TRHtreated rats on all days also showed an obvious tendency to avoid the dark compartment more efficiently, than the control group, but these changes did not reach statistical significance (0.1 > p > 0.05; data not shown). Dizocilpine (0.2 mgkg) caused no significant

90

NEUROPEPTIDES

change in the avoidance behavior, the percentage of positive scores being close to the control group on days l-5 (Fig. 1). However, in the preliminary experiments in which we attempted to adjust the dose of dizocilpine devoid of amnestic effect under the conditions of the present study, a higher dose (1 mg/kg) strongly impaired the avoidance on days 1-5, (p < 0.05 on days 1-4, p < 0.01 on day 5, n = 10, data not shown). The combined administration of TRH and dizocilpine caused dramatic changes in the behavior of the rats. In many of them clear signs of ataxia developed, the animals made chaotic and irregular movements over the cage, looking completely ‘disorientated’. Nevertheless, this did not prevent them from purposeful behavior in the light-dark shuttle, where they headed to the dark compartment faster than the controls (Fig. 1) and spent significantly more time there (p < 0.01 on days 2 and 3, p < 0.05 on days 1 and 4). Thus, the avoidance behavior was strongly inhibited. OCCNTROL R3

TM-I

K OF MIIDANCE

m -

DIZOCILPHE TRHtDlZOCILPlNE

SCORES PER GROUP

7, retention test. On day 7,48 h after the last training session, the degree of avoidance in the control group decreased as compared to day 5 (STL 45.8%, DPR - 54.2%; Fig. 2). Dizocilpine-treated rats had somewhat better scores, but this difference was not significant (p > 0.1). The rats treated during the training/testing sessions with TRH demonstrated significantly better retention of the memory trace on day 7 according to both STL and DPR. On the other hand, the rats, injected with TRH + dizocilpine had the same degree of avoidance on day 7 as controls.

Day

Protocol B TRH significantly improved the retrieval of the memory trace when injected before the test 24 h after a single training session (Fig. 3). Dizocilpine (0.2 mgkg) did not significantly change the avoidance scores in this paradigm. Combined administration of TRH and dizocilpine completely blocked memory retrieval. Nociception test. Dizocilpine (0.2 mgkg) or its combination with TRH caused no change in the nociceptive threshold, while TRH (5 mg/kg) tended to increase it, p = 0.056 (control, 31.7 f 6.54, dizocilpine, 28.3 f 4.01, TRH, 53.3 f 3.33, TRH + dizocilpine, 28.3 f 4.01 V).

‘::1r

Discussion

Fig. 2 Effect of TRH, dizocilpine and their combination on the retention of darkness avoidance in rata (STL and DPR criteria). Protocol A, retention test on day 7. The substances were administered during training-testing session on days O-5. Before the retention test all groups were injected with saline. Significance marksasinFigure1.

The memory-facilitatory effect of TRH has been demonstrated earlier using several experimental modelsofamnesiainratsandmice. Thus, animprovement in the active avoidance by TRH (10 mgkg i.p.) was reported in rats subjected to the lesions of the internal capsule by Oka et al (2). Oka et al (1) have also shown that TRH (2-20 mg/kg i.p. or s.c.) increases the active avoidance in ‘poorly-performing’ mice or mice subjected to anoxia. It also antagonized cycloheximide-induced amnesia in a passive avoidance test. On the other hand, Yamamum et al ( 10) did not fmd any improvement using several models of amnesia in rats and mice after administration of large doses of TRH (30-300 mgikg). Hence, memory facilitatory effect of TRH is likely to have a complex doseresponse relationship. The aim of the present study - to investigate the interaction of TRH with dizocilpine, which itself

DPR

I

DiZOCILPINE

(MK-801) SUPPRESSES

MEMORY FACILITATORY

ACTION OF TRH

91

DlZOCtLPtNE

TRH+DIZOCILPINE

Fig. 3

Effect of TRH, dizocilpine and their combination on the memory retrieval (STL criterion). Protocol B. Hatched area represents percentage of the positive avoidance scores per group. Note strong reduction of the avoidance by the combined administration of TRH and dizocilpine. Doses and significance marks as in Figure 1.

under certain conditions is able to impair memory (8-10, 12), made the employment of amnestic procedures highly undesirable. In preliminary experiments we were unable to find a significant improvement of the avoidance retention after a single training session using TRH in normal rats, However, 6-days training/testing protocol with a relatively weak negative reinforcement (1 stimulus) has shown that TRH is able to significantly facilitate the passive avoidance in animals, not subjected to any amnestic procedure. The effects of the substances on the avoidance scores during the training/testing, could result from the changes of retention or retrieval of the memory trace or the combination of both. The significantly better avoidance in the TRH-treated group 48 h after the completion of the 6-days training (day 7 in protocol A) suggests a more effective retention of the avoidance after the administration of TRH. On the other hand, TFW was able to increase avoidance when injected before the trial performed 24 h after a single training session with saline (Protocol B). Thus, retrieval of the memory trace was also facilitated by TRH.

Dizocilpine (0.2 mg/kg) had no significant effects on the retention or retrieval of the passive avoidance under the conditions of the present experiment. Obviously, the ability of dizocilpine to impair memory depends considerably on the applied procedure with tests based on more complex spatial tasks being generally more sensitive (10, 12-14). Also, in the light-dark passive avoidance paradigm the effect of dizocilpine may even reverse dependent on the dose and regime of its administration (9). The combined administration of TRH and dizocilpine caused a dramatic impairment in the avoidance on days l-5 (protocol A). Unexpectedly, the retention on day 7, when all groups were injected with saline, was not different in the control and TRH + dizocilpine treated group. Hence, it is likely, that suppression of avoidance by the combination of TRH and dizocilpine on days l-5 resulted from impairment of the retrieval of the memory trace. Indeed, this combination completely blocked the retrieval in protocol B. Several data demonstrate the involvement of NMDA receptors in the memory acquisition and retention, rather than in the retrieval (13, 15-17).

92

Impairment of the retrieval observed in our experiments after the combined administration of TRH and dizocilpine may be explained by the use-dependence of the dizocilpine effect (18, 19). It is possible, that under normal conditions the role of NMDA receptors in memory retrieval is not essential. However, if NMDA-mediated processes were stimulated by TRH, the NMDA-operated channels may become highly sensitive to dizocilpine. As a result, the animal’s behavior would be strongly affected and retrieval of the previously learned information impaired. The observed effects of the substances on the avoidance scores did not result from a nonspecific change in locomotor activity: neither TRH nor dizocilpine or their combination changed the STL of the untrained rats on day 0. The time spent in the dark compartment was also unaltered, as demonstrated by DPR. The effects of the substances could not either be related to the changes in the nociception. Moreover, TRH tended to increase nociceptive threshold, and thus might even be expected to decrease negative reinforcement efficacy. In conclusion, the results of the present study suggest that in rats, not subjected to any amnestic treatment TRH is able to increase retention and retrieval of passive avoidance through the activation of NMDA-receptor mediated processes. Acknowledgements The authors are grateful to Dr. A. Foster (Merck, Sharp and Dohme, UK) for the gift of dizocilpine and Dr. M. L. Zeise (MPI for Psychiatry, Munich) for valuable scientific discussions.

References 1. Oka, M., Ochi, Y., Furukawa, K., Ito, T., Karasawa, T. and Kadowkawa. T. (1989). L-6-ketooineridine-2-Carbonvl-Lleucyl-L-pro&e amide’as a novel thyrotropin releasing hormone analogue with improving effects on impaired central nervous system functions. Arznebn-Forsch. 39: 297-303. 2. Oka,M., Ito, T., Furukawa,K.,Karasawa, T. andKadowawa, T. (1990). Memory deficits following internal capsule lesions in rats and their improvement by L-6-ketopiperidine-2-carbonyl-L-leucyl-L-prollne amide (RGH-2202), athyrotropinreleasing hormone analogue. Arch. Int. Pharmacodyn. 306: 18-33. 3. Zakusov, V., Ostrovskaya,R., Kosoi, M., Gugasheva, T. and Skoldinov, A. (1986). On the mechanism of antidepressant and learning facilitatory effect ofthyroliberin and its analogs. Bull. Expl. Biol. and Med. 7: 39-41 (in Russ.).

NEUROPEPTIDES

4. Chepurnova, N., Guseva, A., Efimova, E., Mart’ianov, A. and Chepurnov, S. (1989). Comparison of the effects of thyroliberin and ACTI-U-7 PGP on the learning of rats during the solving of spatial orientation tasks. Fiziol. Zh. SSSR. 75: 677-683 (in Russ.). 5. Mellow, A., Sunderland, T., Cohen, R. et al. (1989). Acute effects of high-dose thyrotropin releasing hormone infusions in Alzheimer’s disease. Psychophrumacol. 98: 403-407. 6. Kharkevich, D., Chizh, B. and Kasparov, S. (1991). Stimulant effect of thyrotropin-releasing hormone and its analog, RGH 2202 on the diaphragm respiratory activity, and their antagonism with morphine: possible involvement of the N-methyl-D-aspartate receptors. Brain. Res. 551: 110-115. 7. Alesadri, B., Bgttig, K. and Wetzl, H. (1989). Effects of ketamine on tunnel maze and water maze performance in the rat. Behav. andNeuro1. Biol. 52: 194-212. 8. Venable, N. and Kelly, P. (1990). Effects of NMDA receptor antagonists on passive avoidance learning and retrieval in rats and mice. Psychopharmacol. 100: 2 15-22 1. 9. Mondadori, C., Weiskrsntz, L., Buerki, H., Petschke, F. and Fagg, G. (1989). NMDA receptor antagonists can enhance or impair learning performance in animals. Exp. Brain Res. 75: 449-456. 10. Ward, L., Mason, S. and Abraham, W. (1990). Effects of the NMDA antagonists CPP and MK-801 on radial maze performance inrats.Pharmacol. Biochem. Behav. 35: 785-790. 11. Yamamura, M., Konishita, K., Nakagawa, H and Ishida, R. (1991). Pharmacological study of TA-0910, a new thyrotropin-releasing hormone (TRH) analog (IV): Effects on experimental memory impairment ln mice and rats. Jpn. J. Pharmacol. 55: 241-253. 12. Whishaw, I. and Auer, R. (1989). Immediate and long-lasting effects of MK-801 on motor activity, spatial navigation in a swimming pool and EEG in the rat. Psychopharmacol. 98: 500-507. 13. Wozniak, D., Olney, J., Kettinger, III L., Price, M. and Miller, J. (1990). Behavioral effects of MK-801 in the rat. Psychopharmacol. 101: 47-56. 14. Malenfant, S.,O’Hearn, S. andFleming,A. (1991). MK801, an NMDA antagonist, blocks acquisition of a special task but does not block maternal experience effects. Physiol. Behav. 49: 1129-l 137. 15. Jones, K., Schaeffer, C. and DeNoble, V. (1989). Systemically administered N-methyl-D-aspartate interferes with acquisition of a passive avoidance in rats. Pharmacol. Biochem. Behav. 34: 181-185. 16. Turska, P. and Turski, W. (1990). Excitatory amino acids and memory: effects of drugs acting at N-methyl-D-aspartate receptors in learning and memory tasks. Neuropharmacol. 29: 1111-1116. 17. Heale, V. and Harley, C. (1990). MK-801 and AP5 impair acquisition, but not retention, of the Morris milk maze. Pharmacol. Biochem. Behav. 36: 145-149. 18. Halliwell, R., Peters, J. and Lambert, J. (1989). The mechanism of action and pharmacological specificity of the anticonvulsantNMDA antagonist MK-801: a voltage clamp study on neuronal cells in culture. Br. J. Pharmacol. 96: 48-94. 19. Schwsrz, S., Zhou, G., Katki, A. and Rothbart, D. (1990). Lhomocysteate stimulates [aH]MK801 binding to the phencyclidine recognition site and is thus an agonist for the M-methyl-D-aspartate-operated cation channel. Neuroscience. 37: 193-200.