- Email: [email protected]
Dextrorphan blocks long- but not short-term memory in a passive avoidance task in rats
UP 20949
czmarek Nencki Instinue of Erpprimental
Biobgy, Pasteura 3, 02-093 Wmaw.
Received IO May 1991, revised MS received
4
Pahnd
September 19%. accepted IO September I991
In the present study a mixed u and PCP (phencyclidine) site ligand, dextrorphan (22 mg/kg), blocked ton short-term memory in a passive avoidance task. This effect was not accompanied by any behavioral alterations th.at could interfere with passive avoidance performance. The action nf dextrorphan was shared by a selective NMDA (N-methyl-D-aspariate) receptor antagonist, MK-801 (5methyl-10.1 l-dihydro-5H-dibenzocyclohepten-5,E0-imine mafeate, 0.1 mg/kgJ The results suggest that dextrorphan affects long-term memory, probably via blockade of NMDA receptors. Dextrorphan;
MK-801; NMDA receptors; Passive avoidance behavior; Memory (tong term)
1. Introduction
It has been postulated that the N-methyl-D-aspartate (NMDA) receptor subtype for excitatory amino acids plays a central role in many physiological and pathological situations in the central nervous system (Wroblewski and Danysz, 1989). Among physiological processes dependent on activation of NMDA receptors developmental plasticity as well as learning and memory have been particularly well studied. Blocking NMDA receptors with specific antagonists disrupts learning (Morris et al., 1986; Danysz and Wroblewski, 1989). Similarly, it has been clearly shown that NMDA receptor activation is critical for the establishment of most forms of long term potentiation (LTP), an electrophysiological model of memory formation (Colingridge and Bliss, 1987). However, LTP is composed of several phases of different duration and different susceptibility to disruption by NMDA antagonists i.e. with an initial phase of LTP being less dependent on NMDA receptor activation (Matthies, 1989). This assumption was the basis for comparing the effects of NMDA receptor antagonists on long- versus short-term memory formation. A passive avoidance test with a specially designed testing situation was used. Dextrorphan, the precursor of which, dextromethorphan, is a commonly
Correspondence to: L. Kaczmarek, Nencki Institute of Experimental Biology, Pasteura 3, 02-093 Warsaw, Poland. * Present address: Merz and Co., Department of Pharmacology, Eckenheimer Landstrasse IOl-104,600O Frankfurt/Main i, F.R.G.
used cough suppressant, has been shown to be a ent non-competitive NMDA antagonist (Church et al., 1985), acting through the PCP (phencyclidine) site. it has been shown to prevent NMDA receptor-related excitoxicity, seizures and &hernia (Choi et al., 1987; Leander, 1989; Prince and Feeser, 19883. A potential clinical value of dextrorphan has been postulated (Choi et al., 19871, hence the evaluation of possible side-effects such as memory disruption seems necessary.
2. Materials and methods Male hooded rat weighing 200-250 g were used for all the experiments except the MK-ml-treated groups for which Wistar rats were used. The animals were kept on a natural dark/light cycle with food and water provided ad libitum. The passive avoidance apparatus consisted of two compartmenrs. The large one (60 X 60 X 23 cm) was painted white. The smaller compartment (19 x 17 X 14 cm) was painted black and was equipped with a stainless steel rod floor, through which the footshocks were delivered. During the basic training/ testing procedure the animals were placed in the white compartment and as soon as the rat entered the black compartment the latency L, was recorded, the guillotine door was closed and five brief footshocks (1 s, 30 s interval, 1.3 mA, DC current) were delivered within 2 min. Next, the rat was transferred to the white compartment. Observation continued until the rat entered the black compartment (latency L,) or 5 min elapsed, whichever came first.
The &hed
-
9
injected i.p. either with drug disin physiotogical saline or vehicle alone, l.O- 1.5
400
G
the ttaining,! testing session. S-Methyf-10.1 I-dihydro-SH-dibenzocycloheptenj,~g)_imine maleate (MK-SO11 was obtained from RBI and dextrorphan was a generous gift from Dr. Denis Cboi. h b&xc
El
300
5 5
200
z :
100
CL
3. Results
i 0
In the preliminary experiments 36 rats (4-5 per group1 were injected with increasing doses of dextrorpban (up to 30 mg/kgj. Doses up to 11 me/kg did not produce any significant effects on rat performance in the test (fig. 1). The 22 mg/kg dose resulted in a clear-cut characteristic memory deficit not accompanied by gross behavioral disturbances. In most cases, neither control nor dextrorphan-treated animals entered the black compartment immediately after exposure to footshock (L,). This behavior apparently cannot be explained by electric shock-induced suppression of motility_ In fact, rats exposed to an inescapable electric shock in a dark compartment (doors closed). during a second exposure 30 s later escaped to the lighted compartment immediately (< 30 s; results not shown). Control animals also showed passive avoidance behavior 2 days later CL,). in contrast, dextrorphantreated rats entered the black compartment significantly faster 38 h later (L,) (fig. 1). Similar results were obtained with a more selective ligand for the PCP site i.e. MK-SGl (0.1 mg/kg) (fig. 1). Dextrorphan failed to affect the latency (motivation) to re-enter the black compartment. provided no footshock was given. To analyze the possible effect of dextrorphan on retention and/or retrieval of already acquired memory, the animals were injected before retrieval (L,) instead of training (TJ. No effect of dextrorphan was observed under these conditions (table 1). To exclude a state-dependent learning phenomenon, the rats were
TABLE
-
L 3
Control
1,
Dextrorphan
22
0.1
mg/Lrg
MK-801
Fig. 1. Effect of dextrorphan and MK-801 on passive avoidance latencies during tests (L, and L,). The results are expressed as medians (Med + semi-interquartile range (Q) of groups consisting of 6- 13 animals). Drugs were injected 1 h before training. Dextrorphan and MK-801 were tested in separate experiments in which hooded and Wistar rats were used respectively. Since their control groups did not differ, one group consisting of hooded rats was used for graphic representation. * P < 0.05 as compared to control group (Mann-Whitney two-tailed U-test).
given injections twice, both before the first-day session CL,, + L I) and before the third-day session (L,). The dextrorphan injected animals were still clearly deficient (table 1). Additional experiments addressed the possible involvement of behavioral alterations as a factor interfering with the passive avoidance performance. Dextrorphan failed to affect either pain threshold (measured by a shock titration method) or an open field behavior (results not shown).
4. Discussion Dextrorphan has been suggested to be a potentially useful therapeutic tool antagonizing pathological effects of undesirable NMDA receptor activation (Choi et al., 1987). Moreover, dextromethorphan, which is
1
Effects of dextrorphan on passive avoidance latency fin s) during iraining CL,) and tests (L, and L,). The results are expressed as medians (Medti semi-interquartile range (Q) of groups consisting of 6-9 animals. See Results for further explanations. E?iperirnent
Latency for control (0 L,,
f I) Short-/long-term
memory
(2) Retention/retrieval
f-31State-dependent learning
Med Q Med Q Med Q
and dextrorphan (D) treated animals Ll
L2
D
C
D
C
D
C
42 20 49 4 34 18
30 14 35 23 22 13
300 0 300 0 300 0
300 0
143 a 125 300 0 68 = 19
300 3 300 0 300 n
d P < 0.05 as compared to control (C) group (Mann-Whitney,
two-tailed test).
300 0 300 0
111
easily metabolized within the body to dextrorphan, is already widely used as an antitussive and has recently been applied in an opioid withdrawal syndrome (Koyoncuoglu and Saydam, 1990). The behavioral effects of dextrorphan were, however, not thoroughly tested. The present communication is the first report showing that dextrorphan may produce memory deficits. Results of the experiments presented herein suggest that dextrorphan has a direct effect on longterm memory formation. In order to test this notion we have developed a training/ testing version of a passive avoidance design which enabled us to address this problem directly. Both the short- and long-term memory tests were essentially the same in terms of performance required. The main difference was the time component. The effects of dextrorphan treatment seem to be related to memory acquisition as there was no interference with memory retention and retrieval. No effect was observed on short-term memory. The effect on passive avoidance performance seems to be specific, since at the dose tested, no clear alterations in general behavior of the rats were observed. This was documented in the open field experiment as well as by the L, latency values. This last index, while variable from one experiment to another could not distinguish between control and drug-treated ariimals. This suggests that both groups or rats displayed similar motivation and motility in the test. However, it should be noted that higher doses of dextrorphan (> 30 mg/kg) did result in gross behavioral abnormalities. The resuits also could not be explained by the difference in pain threshold for both groups of animals, since the reaction to footshock was not changed. In addition to its action on the NMDA receptorcoupled PCP site, dextrorphan may also be active on the so called u site, probably not linked directly to the NMDA receptor complex. Therefore, a more selective NMDA receptor antagonist, MK-801 was tested In addition to dextrorphan, with similar results, which is
in line with other reports (Danysz and Wroblewski, 1989). Together with the previous report that 1,34-2tolyguanidine (DTG), a selective u ligand, failed to produce a passive avoidance deficit iDanysz and Wroblewski, 19891, the present data suggest that it is the NMDA receptor rather than the cr site that is tramducing the learning deficit induced by dextrorpban. Therefore, it can be concluded that activation of NMDA receptors is necessary for long- bu: not shortmemory formation.
References D.W., S. Peters and V. Viseskul, 1987, Dextrorphan and Levorphanol selectively block N-methyl-D-aspartame receptormediated neurotoxicity on cortical neurons, J. Pharmacol. Exp. Ther. 242, 713. Church, J., D. Lodge and SC. Berry, 1985, Differential effects of dextrorphan and levorphanol on the excitation of rat spinal neurons by amino acids, Eur. J. Pharmacol. 111, 185. Collingridge, G.L. and T.V.P. Bliss, 1987, NMDA receptors - their role in long-term potentiation, Trends Neurosci. 10, 288. Danysz, W. and J.T. Wroblewski, 1989, Amnesic properties of glutamate receptor antagonists, Neurosci. Res. Commun. 479, 9. Koyuncuoglu, H. and B. Saydam. 1990, The treatment of heroin addicts with dextromethorphan: a double-blind comparison of dextromethorphan with chlorpromazine, Int. J. Clin. Pharmacol. Ther. Toxicol. 28, 147. Leander. J.D., 1989. Evaluation of dextromethorphan and carberspentane as anticonvulsants and N-methyl-D-aspartic acid antagonists in mice, Epilepsy Res. 4, 28. Matthies, H., 1989, Neurobiological aspects of learning and memory, Ann. Rev. Psychol. 40, 381. Morris, R.G.M., E. Anderson, G.S. Lynch and M. Baudry, 19%. Selective impairment of learning and blockade of long-term potentiation by an N-methyl-D-aspartate receptor antagonist. AP5. Nature 319, 774. Prince, D.A. and H.R. Feeser. 1988, Dextromethorphan protects against cerebral infarction in a rat model of hypoxia-ischemia Neurosci. Len. 85, 291. Wroblewski, J.T. and W. Danysz, 1989, Modulation of glutamate receptors: molecular mechanisms and functional implications. Ann. Rev. Pharmacol. Toxicol. 29, 441.
Choi,
czmarek Nencki Instinue of Erpprimental
Biobgy, Pasteura 3, 02-093 Wmaw.
Received IO May 1991, revised MS received
4
Pahnd
September 19%. accepted IO September I991
In the present study a mixed u and PCP (phencyclidine) site ligand, dextrorphan (22 mg/kg), blocked ton short-term memory in a passive avoidance task. This effect was not accompanied by any behavioral alterations th.at could interfere with passive avoidance performance. The action nf dextrorphan was shared by a selective NMDA (N-methyl-D-aspariate) receptor antagonist, MK-801 (5methyl-10.1 l-dihydro-5H-dibenzocyclohepten-5,E0-imine mafeate, 0.1 mg/kgJ The results suggest that dextrorphan affects long-term memory, probably via blockade of NMDA receptors. Dextrorphan;
MK-801; NMDA receptors; Passive avoidance behavior; Memory (tong term)
1. Introduction
It has been postulated that the N-methyl-D-aspartate (NMDA) receptor subtype for excitatory amino acids plays a central role in many physiological and pathological situations in the central nervous system (Wroblewski and Danysz, 1989). Among physiological processes dependent on activation of NMDA receptors developmental plasticity as well as learning and memory have been particularly well studied. Blocking NMDA receptors with specific antagonists disrupts learning (Morris et al., 1986; Danysz and Wroblewski, 1989). Similarly, it has been clearly shown that NMDA receptor activation is critical for the establishment of most forms of long term potentiation (LTP), an electrophysiological model of memory formation (Colingridge and Bliss, 1987). However, LTP is composed of several phases of different duration and different susceptibility to disruption by NMDA antagonists i.e. with an initial phase of LTP being less dependent on NMDA receptor activation (Matthies, 1989). This assumption was the basis for comparing the effects of NMDA receptor antagonists on long- versus short-term memory formation. A passive avoidance test with a specially designed testing situation was used. Dextrorphan, the precursor of which, dextromethorphan, is a commonly
Correspondence to: L. Kaczmarek, Nencki Institute of Experimental Biology, Pasteura 3, 02-093 Warsaw, Poland. * Present address: Merz and Co., Department of Pharmacology, Eckenheimer Landstrasse IOl-104,600O Frankfurt/Main i, F.R.G.
used cough suppressant, has been shown to be a ent non-competitive NMDA antagonist (Church et al., 1985), acting through the PCP (phencyclidine) site. it has been shown to prevent NMDA receptor-related excitoxicity, seizures and &hernia (Choi et al., 1987; Leander, 1989; Prince and Feeser, 19883. A potential clinical value of dextrorphan has been postulated (Choi et al., 19871, hence the evaluation of possible side-effects such as memory disruption seems necessary.
2. Materials and methods Male hooded rat weighing 200-250 g were used for all the experiments except the MK-ml-treated groups for which Wistar rats were used. The animals were kept on a natural dark/light cycle with food and water provided ad libitum. The passive avoidance apparatus consisted of two compartmenrs. The large one (60 X 60 X 23 cm) was painted white. The smaller compartment (19 x 17 X 14 cm) was painted black and was equipped with a stainless steel rod floor, through which the footshocks were delivered. During the basic training/ testing procedure the animals were placed in the white compartment and as soon as the rat entered the black compartment the latency L, was recorded, the guillotine door was closed and five brief footshocks (1 s, 30 s interval, 1.3 mA, DC current) were delivered within 2 min. Next, the rat was transferred to the white compartment. Observation continued until the rat entered the black compartment (latency L,) or 5 min elapsed, whichever came first.
The &hed
-
9
injected i.p. either with drug disin physiotogical saline or vehicle alone, l.O- 1.5
400
G
the ttaining,! testing session. S-Methyf-10.1 I-dihydro-SH-dibenzocycloheptenj,~g)_imine maleate (MK-SO11 was obtained from RBI and dextrorphan was a generous gift from Dr. Denis Cboi. h b&xc
El
300
5 5
200
z :
100
CL
3. Results
i 0
In the preliminary experiments 36 rats (4-5 per group1 were injected with increasing doses of dextrorpban (up to 30 mg/kgj. Doses up to 11 me/kg did not produce any significant effects on rat performance in the test (fig. 1). The 22 mg/kg dose resulted in a clear-cut characteristic memory deficit not accompanied by gross behavioral disturbances. In most cases, neither control nor dextrorphan-treated animals entered the black compartment immediately after exposure to footshock (L,). This behavior apparently cannot be explained by electric shock-induced suppression of motility_ In fact, rats exposed to an inescapable electric shock in a dark compartment (doors closed). during a second exposure 30 s later escaped to the lighted compartment immediately (< 30 s; results not shown). Control animals also showed passive avoidance behavior 2 days later CL,). in contrast, dextrorphantreated rats entered the black compartment significantly faster 38 h later (L,) (fig. 1). Similar results were obtained with a more selective ligand for the PCP site i.e. MK-SGl (0.1 mg/kg) (fig. 1). Dextrorphan failed to affect the latency (motivation) to re-enter the black compartment. provided no footshock was given. To analyze the possible effect of dextrorphan on retention and/or retrieval of already acquired memory, the animals were injected before retrieval (L,) instead of training (TJ. No effect of dextrorphan was observed under these conditions (table 1). To exclude a state-dependent learning phenomenon, the rats were
TABLE
-
L 3
Control
1,
Dextrorphan
22
0.1
mg/Lrg
MK-801
Fig. 1. Effect of dextrorphan and MK-801 on passive avoidance latencies during tests (L, and L,). The results are expressed as medians (Med + semi-interquartile range (Q) of groups consisting of 6- 13 animals). Drugs were injected 1 h before training. Dextrorphan and MK-801 were tested in separate experiments in which hooded and Wistar rats were used respectively. Since their control groups did not differ, one group consisting of hooded rats was used for graphic representation. * P < 0.05 as compared to control group (Mann-Whitney two-tailed U-test).
given injections twice, both before the first-day session CL,, + L I) and before the third-day session (L,). The dextrorphan injected animals were still clearly deficient (table 1). Additional experiments addressed the possible involvement of behavioral alterations as a factor interfering with the passive avoidance performance. Dextrorphan failed to affect either pain threshold (measured by a shock titration method) or an open field behavior (results not shown).
4. Discussion Dextrorphan has been suggested to be a potentially useful therapeutic tool antagonizing pathological effects of undesirable NMDA receptor activation (Choi et al., 1987). Moreover, dextromethorphan, which is
1
Effects of dextrorphan on passive avoidance latency fin s) during iraining CL,) and tests (L, and L,). The results are expressed as medians (Medti semi-interquartile range (Q) of groups consisting of 6-9 animals. See Results for further explanations. E?iperirnent
Latency for control (0 L,,
f I) Short-/long-term
memory
(2) Retention/retrieval
f-31State-dependent learning
Med Q Med Q Med Q
and dextrorphan (D) treated animals Ll
L2
D
C
D
C
D
C
42 20 49 4 34 18
30 14 35 23 22 13
300 0 300 0 300 0
300 0
143 a 125 300 0 68 = 19
300 3 300 0 300 n
d P < 0.05 as compared to control (C) group (Mann-Whitney,
two-tailed test).
300 0 300 0
111
easily metabolized within the body to dextrorphan, is already widely used as an antitussive and has recently been applied in an opioid withdrawal syndrome (Koyoncuoglu and Saydam, 1990). The behavioral effects of dextrorphan were, however, not thoroughly tested. The present communication is the first report showing that dextrorphan may produce memory deficits. Results of the experiments presented herein suggest that dextrorphan has a direct effect on longterm memory formation. In order to test this notion we have developed a training/ testing version of a passive avoidance design which enabled us to address this problem directly. Both the short- and long-term memory tests were essentially the same in terms of performance required. The main difference was the time component. The effects of dextrorphan treatment seem to be related to memory acquisition as there was no interference with memory retention and retrieval. No effect was observed on short-term memory. The effect on passive avoidance performance seems to be specific, since at the dose tested, no clear alterations in general behavior of the rats were observed. This was documented in the open field experiment as well as by the L, latency values. This last index, while variable from one experiment to another could not distinguish between control and drug-treated ariimals. This suggests that both groups or rats displayed similar motivation and motility in the test. However, it should be noted that higher doses of dextrorphan (> 30 mg/kg) did result in gross behavioral abnormalities. The resuits also could not be explained by the difference in pain threshold for both groups of animals, since the reaction to footshock was not changed. In addition to its action on the NMDA receptorcoupled PCP site, dextrorphan may also be active on the so called u site, probably not linked directly to the NMDA receptor complex. Therefore, a more selective NMDA receptor antagonist, MK-801 was tested In addition to dextrorphan, with similar results, which is
in line with other reports (Danysz and Wroblewski, 1989). Together with the previous report that 1,34-2tolyguanidine (DTG), a selective u ligand, failed to produce a passive avoidance deficit iDanysz and Wroblewski, 19891, the present data suggest that it is the NMDA receptor rather than the cr site that is tramducing the learning deficit induced by dextrorpban. Therefore, it can be concluded that activation of NMDA receptors is necessary for long- bu: not shortmemory formation.
References D.W., S. Peters and V. Viseskul, 1987, Dextrorphan and Levorphanol selectively block N-methyl-D-aspartame receptormediated neurotoxicity on cortical neurons, J. Pharmacol. Exp. Ther. 242, 713. Church, J., D. Lodge and SC. Berry, 1985, Differential effects of dextrorphan and levorphanol on the excitation of rat spinal neurons by amino acids, Eur. J. Pharmacol. 111, 185. Collingridge, G.L. and T.V.P. Bliss, 1987, NMDA receptors - their role in long-term potentiation, Trends Neurosci. 10, 288. Danysz, W. and J.T. Wroblewski, 1989, Amnesic properties of glutamate receptor antagonists, Neurosci. Res. Commun. 479, 9. Koyuncuoglu, H. and B. Saydam. 1990, The treatment of heroin addicts with dextromethorphan: a double-blind comparison of dextromethorphan with chlorpromazine, Int. J. Clin. Pharmacol. Ther. Toxicol. 28, 147. Leander. J.D., 1989. Evaluation of dextromethorphan and carberspentane as anticonvulsants and N-methyl-D-aspartic acid antagonists in mice, Epilepsy Res. 4, 28. Matthies, H., 1989, Neurobiological aspects of learning and memory, Ann. Rev. Psychol. 40, 381. Morris, R.G.M., E. Anderson, G.S. Lynch and M. Baudry, 19%. Selective impairment of learning and blockade of long-term potentiation by an N-methyl-D-aspartate receptor antagonist. AP5. Nature 319, 774. Prince, D.A. and H.R. Feeser. 1988, Dextromethorphan protects against cerebral infarction in a rat model of hypoxia-ischemia Neurosci. Len. 85, 291. Wroblewski, J.T. and W. Danysz, 1989, Modulation of glutamate receptors: molecular mechanisms and functional implications. Ann. Rev. Pharmacol. Toxicol. 29, 441.
Choi,