- Email: [email protected]
Nitric oxide, but not serotonin, is involved in acquisition of food-attraction conditioning in the snail Helix pomatia
ELSEVIER
Neuroscience Letters 206 (1996) 29-32
llEURgSCI[HC[ LETTERS
Nitric oxide, but not serotonin, is involved in acquisition of food-attraction conditioning in the snail Helix pomatia Thomas Teyke Institut fiir Zoologie (II1)Biophysik, Johannes Gutenberg-Universitiit, 55099 Mainz, Germany
Received 18 November1995; revisedversion received 30 January 1996; accepted 30 January 1996
Abstract
The effects of inhibition of nitric oxide (NO) or serotonin (injection of nitro-L-arginine methyl ester (L-NAME) or 5,6dihydroxytryptamine (5,6-DHT), respectively) on food-attraction conditioning was investigated in Helix. Blocking NO synthase (NOS) prior to conditioning significantly impaired the food-finding ability of the snails. Food-conditioned snails, after inhibition of NOS, remained able to locate the conditioned food. These results indicate that the acquisition of memory depends on NO, whereas memory recall and olfactory orientation are not dependent. Ablating the serotonergic system did not influence food-attraction conditioning, suggesting that food-attraction conditioning may be at variance with conventional associative conditioning procedures. Keywords: Learning; Mollusk; Serotonin (5-HT); Nitric oxide; Food-finding; Olfactory orientation
Mollusks are capable of different forms of learning (see e.g. [1,6,12]) and some of the underlying neuronal and molecular mechanisms have been identified [1,8,11]. Most notably, serotonin has been implicated as the modulatory transmitter involved in sensitization as well as associative conditioning [8]. Other transmitters, however, might play a role in certain forms of learning. In the octopus, nitric oxide (NO) is involved in touch learning [15], without affecting retrieval of memory [16]. In the slug Limax, circumstantial evidence suggests that NO may be involved in aversive olfactory conditioning [5]. Investigation of the foraging strategy of Helix revealed a functionally different olfactory learning phenomenon. Snails, by feeding on a certain food, acquire the ability to subsequently locate this particular food [18]. Both transmitters serotonin and NO, implicated in learning in mollusks, have been identified in the nervous system of Helix. About 30 serotonin-containing neurons have been localized in the cerebral ganglion [9,10]. NO is also present in central neurons, especially in neurons and in the neuropil in the PC lobe of the cerebral ganglion [3,17]. Furthermore, peripheral olfactory neurons stain positively for NADPH-diaphorase [3]. Functionally, both serotonin and NO modulate field potential oscillations of the PC * Tel.: +49 6131 394483; fax: +49 6131 395443
lobe, which are believed to be involved in odor processing and olfactory learning in Limax [5,7]. In this study, the involvement of the two transmitters in food-attraction conditioning was investigated by selectively blocking one transmitter. Animals were collected locally and these 'naive' animals were first tested for their ability to locate carrot juice in the open field orientation test, described elsewhere [18]. To condition the snails, they were fed a 2 cm 2 piece of filter paper soaked with 1 ml carrot juice (Granini). The success of conditioning was assessed the next day by re-testing the snails' ability to locate carrot juice. The paths of the animals were recorded, and the number of snails locating the carrot site was scored. Serotonin was blocked by injecting animals with neurotoxin 5,6dihydroxytryptamine (5,6-DHT; 20 mg/kg in saline containing 0.5 mg/ml ascorbic acid), 3, 7, or 14 days before conditioning. NO synthesis was inhibited by injections of nitro-L-arginine methyl ester (L-NAME; 75 mg/kg) 1 h prior to conditioning. Both groups were accompanied by saline injected controls and tested in a blind manner. Observations indicated that neither drug caused major side effects at the time when the experiments were performed. Injection of L-NAME caused no apparent irregularities in the snails' spontaneous behavior. During the conditioning procedure, 1 h after the injection, the snails
0304-3940/96/$12.00 © 1996 ElsevierScience Ireland Ltd. All rights reserved PII: S0304-3940(96) 12434-2
30
A
T. Teyke / Neuroscience Letters 206 (1996) 29-32
~
]
Nai)ve
B
Saline
"O
C
D
5,6-DHT
20 cm
Fig. 1. Representativepaths of eight snails before ((A), naive) and after (B-D) conditioning with food. Snails were released between a carrot juice feeding site (filled circle) and a control site containing water (dotted). Three groups of experimental animals are shown: (B) salineinjected controls; (C) snails injected with L-NAME; and (D) snails injected with 5,6-DHT. Note that injection of L-NAMEdid not promote successful food-finding,indicating that food-attractionconditioning did not manifest. behaved and ate in a manner indistinguishable from that of saline-injected controls (but see [4] for effects on feeding in L y m n a e a ) . Injection of 5,6-DHT (or 5,7-DHT) caused short-term abnormalities, which completely disappeared after a few hours [2,13]. In this study, the snails were conditioned 3, 7, or 14 days after injection, at a time when DHT-injected animals exhibit only subtle behavioral modifications, such as altered arousal build-up [13], and impaired associative conditioning [2]. It thus seems justified to conclude that changes in the snails' foodfinding ability reflect drug-related effects on learning, rather than indirect effects stemming from the drugs' influence on other behavioral systems. All 'naive' snails were tested for their ability to spontaneously locate the carrot juice feeding site. As shown in Fig. 1A, naive snails moved in different directions, and only 1/8 animals located the feeding site. Evaluating all runs yielded a total of 8% of naive snails locating the feeding site (Fig. 2, naive), which matched the previously reported rate [18]. Following these tests, all snails in the experimental groups (saline-injected controls, L-NAME-, and DHT-injected snails) were conditioned and re-tested the following day. As shown in Fig. 1B, the majority of saline-injected snails moved directly towards the food. A total of 73% of the animals located and consumed the food (Fig. 2), which indicated that the saline-injected control animals had become conditioned to the food. In
contrast, L-NAME injection affected the ability of the snails to locate the food. Only 3/8 snails shown in Fig. 1C moved in the direction of the food. As shown in Fig. 2, a total of 44% of the L-NAME-injected animals located the food. This is a highly significant reduction from the rate of saline-injected controls (P < 0.001). Although the food-finding rate is significantly lower than that of controis, it is significantly higher than that of naive animals (P < 0.001), which indicates that some injected snails had become conditioned. Increasing the dosage to 150 mg/kg L-NAME (n = 8), or using a lower dosage of 10 mg/kg (n = 12) reduced the food-finding rate to 50% and 45%, respectively, which in both cases does not differ from that reported above. Using a different blocker of NO synthase (NOS), NMMA (50 mg/kg; n =7), led to comparable results of 57% of the snails locating the food. In general, there was no evidence of a physiological defect in the snails that failed to locate the food. Locomotion (behavior, speed, etc.) of the snails and the paths taken closely resembled that of naive animals (compare Fig. 1A,B). The effects of L-NAME are short-lived, and it was possible to condition snails 24 h after they had been injected with L-NAME. When tested the day following the conditioning, 77% of injected snails (n = 12) were able to locate the food. This rate is indistinguishable from that of normal animals (P = 0.5). The finding that inhibition of NO impairs the snails' food-finding ability does not permit one to determine if the acquisition of memory, retention or recall of memory, or the olfactory orientation has been disturbed. NO has also been located in peripheral olfactory sensory neurons [3], indicating that NO might be involved in the processing of olfactory information. To test for these alternatives, food-conditioned animals were injected with L-NAME, and 1 h after injection, tested for their food-finding ability. As shown in Fig. 3, prior to injection 83% of the conditioned snails (cond.) located the food. One hour after L-NAME injection, 75% of the animals were able to n.$,
~3
o o
100 -
°,.q
o
50
~rj
I naive [92]
I
I
uline [31]
(X~X L-NAME [34]
5,6-DHT [27]
Fig. 2. Food-findingrate of snails before and after injection with saline, L-NAME or DHT, respectively, as an indicator of successful foodattraction conditioning.Z2 tests: ***P < 0.001; n.s., not significant.
T. Teyke / Neuroscience Letters 206 (1996) 29-32
100
-
!1o$o
o O
rd o
50 ¸
cond. [2O]
saline [8]
L-NAME [121
Fig. 3. Food-findingrate of conditioned snails before and after injection of L-NAMEor vehicle. The high food-finding rates indicate that memory recall and olfactory orientation are not impaired by injection of LNAME.Z2 tests: n.s., not significant. find the food. This rate is similar to that of saline-injected controls (83%; P = 0.4), and to the food-finding rate before injection (P = 0.4). These results suggest that recall and retention of memory is independent of NO, and that blocking NOS does not impair olfactory orientation. Blocking the serotonergic system by injection of 5,6DHT did not affect food-attraction conditioning. As shown in Fig. 1D, 7/8 of the injected snails were able to locate the food. The performance of snails that were injected 3, 7, or 14 days prior to conditioning was indistinguishable; thus the results were combined. After conditioning the snails for carrot, a total of 85% of the DHTinjected snails were able to locate the carrot (Fig. 2). This is a rate similar to that of saline-injected controls (77%; P = 0 . 2 ) , indicating that depletion of serotonin does not impair food-attraction conditioning. However, DHTinjected snails showed a behavioral particularity beyond those reported earlier [2,13]: they moved more slowly than control animals, suggesting a role of serotonin in locomotion (see also [14]). The average crawling speed of DHT-injected animals during their orientation to food was 1.2 _+0.4 cm/min (n = 12), which is considerably slower than that of control snails (5.4 _+ 1.5 cm/min; n = 12). Using selective blockers for NO and for serotonin, it is demonstrated that NO is involved in the acquisition of food-attraction conditioning in snails, whereas serotonin is not involved in the learning. The acquisition of foodattraction was not completely abolished by injection of LN A M E and about 40% of the injected snails still became conditioned. The reason for this is unclear and might reflect problems with the experimental procedure, or general problems regarding inhibition of learning. On the experimental side, the time window of the efficacy of LN A M E on NOS has not been determined in snails. In this study, L-NAME was injected 1 h prior to conditioning, corresponding to the protocol used in octopus ([15]; but see [4]). General problems might reflect the fact that in a
31
variety of learning phenomena, for unknown reasons, treatments attempting to block the acquisition of learning are often only partially successful and a certain percentage of animals remains able to learn. Based upon the effects of nitric oxide on neurons presumably involved in olfactory learning, Gelperin [5] suggested that NO might be involved in aversive olfactory conditioning in Limax. As documented, NO is involved in olfactory conditioning in Helix, although this particular type of conditioning differs considerably from the conditioning paradigm employed in Limax. Contrary to conventional associative conditioning phenomena, foodattraction conditioning is not influenced by depletion of serotonin. As reported earlier, depletion of serotonin abolishes aversive associative conditioning in Helix [2], which corroborates findings in Aplysia that serotonin is involved in associative conditioning [8]. The finding that food-attraction conditioning is not impaired in serotonindepleted snails suggests that food-attraction conditioning is functionally different from conventional associative conditioning paradigms and might therefore utilize a different transmitter. In conclusion, the evidence presented in this paper suggests that the acquisition of food-attraction in Helix involves NO and is independent of serotonin, thus setting apart food-attraction conditioning from other associative conditioning phenomena. Although learning is commonly classified in narrowly defined categories, there appears to be a variety of distinctly different conditioning paradigms, which can be distinguished according to their behavioral function, as well as their underlying biochemical mechanisms. Many thanks to M. Peschel and S. Fuss for testing the animals. Supported in part by a grant from 'Stiftung Rheinland-Pfalz flJr Innovation' and DFG Te 138/3-1. [1] Alkon, D.L., Conditioning-induced changes of Hermissenda channels: relevance to mammalian brain function. In N.M. Weinberger, J.L. McGaugh and G. Lynch (Eds.), Memory Systems of the Brain, Guilford, New York, 1985, pp. 9-26. [2] Balaban, P.M., Vehovszky, A., Maksimova, O.A. and Zakharov, I.S., Effect of 5,7-dihydroxytryptamine on the food-aversive conditioning in the snail Helix lucorum L., Brain Res., 404 (1987) 201-210. [3] Cooke, I.R.C., Edwards, S.L. and Anderson, C.R., The distribution of NADPH diaphorase activity and immunoreactivityto nitric oxide synthase in the nervous system of the pulmonate mollusc Helix aspersa, Cell Tissue Res., 277 (1994) 565-572. [4] Elphick, M.R., Kemenes, G., Staras, K. and O'Shea, M., Behavioral role for nitric oxide in chemosensoryactivation of feeding in a mollusc, J. Neurosci., 15 (1995) 7653-7664. [5l Gelperin, A., Nitric oxide mediates network oscillations of olfactory interneurons in a terrestrial mollusc, Nature, 369 (1994) 6163. [6] Gelperin, A., Rapid food-aversion learning by a terrestrial mollusk, Science, 189 (1975) 567-570. [7] Gelperin, A., Rhines, L.D., Flores, J. and Tank, D.W., Coherent network oscillations by olfactory intemeurons: modulation by endogenous amines, J. Neurophysiol., 69 (1993) 1930-1939.
32
T. Teyke / Neuroscience Letters 206 (1996) 29-32
[8] Hawkins, R.D., Kandel, E.R. and Siegelbaum, S.A., Learning to modulate transmitter release: themes and variations in synaptic plasticity, Ann. Rev. Neurosci., 16 (1993) 625~65. [9] Hermidi, L., Elekes, K. and S.-R6zsa, K., Distribution of serotonin-containing neurons in the central nervous system of the snail Helix pomatia, Cell Tissue Res., 257 (1989) 313-323. [10] Hermidi, L., Kemenes, G. and R6zsa, K., Selective in vivo labeling of serotonergic neurones by 5,6-dihydroxytryptamine (5,6DHT) in the snail Helix pomatia. In H.H. Boer, W.P.M. Geraerts and J. Joosse (Eds.), Neurobiology: Molluscan Models, NorthHolland, Amsterdam, 1987, pp. 22-25. [11] Kandel, E.R. and Schwartz, J.H., Molecular biology of learning: modulation of transmitter release, Science, 218 (1982) 433-443. [12] Kandel, E.R., Cellular Basis of Behavior, Freeman, New York, 1976, 727 pp. [13] Kemenes, G. and S.-R6zsa, K., The role of serotonergic mechanisms in food-induced arousal of the snail Helix pomatia L. In H.H. Boer, W.P.M. Geraerts and J. Joosse (Eds.), Neurobiology:
[14]
[15]
[16]
[17]
[18]
Molluscan Models, North-Holland, Amsterdam, 1987, pp. 277286. Mackey, S. and Carew, T.J., Locomotion in Aplysia: triggering by serotonin and modulation by bag cell extract, J. Neurosci., 3 (1983) 1469-1477. Robertson, J.D., Bonaventura, J. and Kohm, A.P., Nitric oxide is required for tactile learning in Octopus vulgaris, Proc. R. Soe. London B, 256 (t994) 269-273. Robertson, J.D., Bonaventura, J. and Kohm, A., Nitric oxide synthase inhibition blocks octopus touch learning without producing sensory or motor dysfunction, Proc. R. Soc. London B, 261 (1995) 167-172. S~inchez-Alvarez, M., Le6n-Olea, M., Talavera, E., Pellicer, F., SLnchez-lslas, E. and Martfnez-Lorenzana, G., Distribution of NADPH-diaphorase in the perioesophageal ganglia of the snail, Helix aspersa, Neurosci. Lett., 169 (1994) 51-55. Teyke, T., Food-attraction conditioning in the snail, Helix pomatia, J. Comp. Physiol., 177 (1995) 409-414.
Neuroscience Letters 206 (1996) 29-32
llEURgSCI[HC[ LETTERS
Nitric oxide, but not serotonin, is involved in acquisition of food-attraction conditioning in the snail Helix pomatia Thomas Teyke Institut fiir Zoologie (II1)Biophysik, Johannes Gutenberg-Universitiit, 55099 Mainz, Germany
Received 18 November1995; revisedversion received 30 January 1996; accepted 30 January 1996
Abstract
The effects of inhibition of nitric oxide (NO) or serotonin (injection of nitro-L-arginine methyl ester (L-NAME) or 5,6dihydroxytryptamine (5,6-DHT), respectively) on food-attraction conditioning was investigated in Helix. Blocking NO synthase (NOS) prior to conditioning significantly impaired the food-finding ability of the snails. Food-conditioned snails, after inhibition of NOS, remained able to locate the conditioned food. These results indicate that the acquisition of memory depends on NO, whereas memory recall and olfactory orientation are not dependent. Ablating the serotonergic system did not influence food-attraction conditioning, suggesting that food-attraction conditioning may be at variance with conventional associative conditioning procedures. Keywords: Learning; Mollusk; Serotonin (5-HT); Nitric oxide; Food-finding; Olfactory orientation
Mollusks are capable of different forms of learning (see e.g. [1,6,12]) and some of the underlying neuronal and molecular mechanisms have been identified [1,8,11]. Most notably, serotonin has been implicated as the modulatory transmitter involved in sensitization as well as associative conditioning [8]. Other transmitters, however, might play a role in certain forms of learning. In the octopus, nitric oxide (NO) is involved in touch learning [15], without affecting retrieval of memory [16]. In the slug Limax, circumstantial evidence suggests that NO may be involved in aversive olfactory conditioning [5]. Investigation of the foraging strategy of Helix revealed a functionally different olfactory learning phenomenon. Snails, by feeding on a certain food, acquire the ability to subsequently locate this particular food [18]. Both transmitters serotonin and NO, implicated in learning in mollusks, have been identified in the nervous system of Helix. About 30 serotonin-containing neurons have been localized in the cerebral ganglion [9,10]. NO is also present in central neurons, especially in neurons and in the neuropil in the PC lobe of the cerebral ganglion [3,17]. Furthermore, peripheral olfactory neurons stain positively for NADPH-diaphorase [3]. Functionally, both serotonin and NO modulate field potential oscillations of the PC * Tel.: +49 6131 394483; fax: +49 6131 395443
lobe, which are believed to be involved in odor processing and olfactory learning in Limax [5,7]. In this study, the involvement of the two transmitters in food-attraction conditioning was investigated by selectively blocking one transmitter. Animals were collected locally and these 'naive' animals were first tested for their ability to locate carrot juice in the open field orientation test, described elsewhere [18]. To condition the snails, they were fed a 2 cm 2 piece of filter paper soaked with 1 ml carrot juice (Granini). The success of conditioning was assessed the next day by re-testing the snails' ability to locate carrot juice. The paths of the animals were recorded, and the number of snails locating the carrot site was scored. Serotonin was blocked by injecting animals with neurotoxin 5,6dihydroxytryptamine (5,6-DHT; 20 mg/kg in saline containing 0.5 mg/ml ascorbic acid), 3, 7, or 14 days before conditioning. NO synthesis was inhibited by injections of nitro-L-arginine methyl ester (L-NAME; 75 mg/kg) 1 h prior to conditioning. Both groups were accompanied by saline injected controls and tested in a blind manner. Observations indicated that neither drug caused major side effects at the time when the experiments were performed. Injection of L-NAME caused no apparent irregularities in the snails' spontaneous behavior. During the conditioning procedure, 1 h after the injection, the snails
0304-3940/96/$12.00 © 1996 ElsevierScience Ireland Ltd. All rights reserved PII: S0304-3940(96) 12434-2
30
A
T. Teyke / Neuroscience Letters 206 (1996) 29-32
~
]
Nai)ve
B
Saline
"O
C
D
5,6-DHT
20 cm
Fig. 1. Representativepaths of eight snails before ((A), naive) and after (B-D) conditioning with food. Snails were released between a carrot juice feeding site (filled circle) and a control site containing water (dotted). Three groups of experimental animals are shown: (B) salineinjected controls; (C) snails injected with L-NAME; and (D) snails injected with 5,6-DHT. Note that injection of L-NAMEdid not promote successful food-finding,indicating that food-attractionconditioning did not manifest. behaved and ate in a manner indistinguishable from that of saline-injected controls (but see [4] for effects on feeding in L y m n a e a ) . Injection of 5,6-DHT (or 5,7-DHT) caused short-term abnormalities, which completely disappeared after a few hours [2,13]. In this study, the snails were conditioned 3, 7, or 14 days after injection, at a time when DHT-injected animals exhibit only subtle behavioral modifications, such as altered arousal build-up [13], and impaired associative conditioning [2]. It thus seems justified to conclude that changes in the snails' foodfinding ability reflect drug-related effects on learning, rather than indirect effects stemming from the drugs' influence on other behavioral systems. All 'naive' snails were tested for their ability to spontaneously locate the carrot juice feeding site. As shown in Fig. 1A, naive snails moved in different directions, and only 1/8 animals located the feeding site. Evaluating all runs yielded a total of 8% of naive snails locating the feeding site (Fig. 2, naive), which matched the previously reported rate [18]. Following these tests, all snails in the experimental groups (saline-injected controls, L-NAME-, and DHT-injected snails) were conditioned and re-tested the following day. As shown in Fig. 1B, the majority of saline-injected snails moved directly towards the food. A total of 73% of the animals located and consumed the food (Fig. 2), which indicated that the saline-injected control animals had become conditioned to the food. In
contrast, L-NAME injection affected the ability of the snails to locate the food. Only 3/8 snails shown in Fig. 1C moved in the direction of the food. As shown in Fig. 2, a total of 44% of the L-NAME-injected animals located the food. This is a highly significant reduction from the rate of saline-injected controls (P < 0.001). Although the food-finding rate is significantly lower than that of controis, it is significantly higher than that of naive animals (P < 0.001), which indicates that some injected snails had become conditioned. Increasing the dosage to 150 mg/kg L-NAME (n = 8), or using a lower dosage of 10 mg/kg (n = 12) reduced the food-finding rate to 50% and 45%, respectively, which in both cases does not differ from that reported above. Using a different blocker of NO synthase (NOS), NMMA (50 mg/kg; n =7), led to comparable results of 57% of the snails locating the food. In general, there was no evidence of a physiological defect in the snails that failed to locate the food. Locomotion (behavior, speed, etc.) of the snails and the paths taken closely resembled that of naive animals (compare Fig. 1A,B). The effects of L-NAME are short-lived, and it was possible to condition snails 24 h after they had been injected with L-NAME. When tested the day following the conditioning, 77% of injected snails (n = 12) were able to locate the food. This rate is indistinguishable from that of normal animals (P = 0.5). The finding that inhibition of NO impairs the snails' food-finding ability does not permit one to determine if the acquisition of memory, retention or recall of memory, or the olfactory orientation has been disturbed. NO has also been located in peripheral olfactory sensory neurons [3], indicating that NO might be involved in the processing of olfactory information. To test for these alternatives, food-conditioned animals were injected with L-NAME, and 1 h after injection, tested for their food-finding ability. As shown in Fig. 3, prior to injection 83% of the conditioned snails (cond.) located the food. One hour after L-NAME injection, 75% of the animals were able to n.$,
~3
o o
100 -
°,.q
o
50
~rj
I naive [92]
I
I
uline [31]
(X~X L-NAME [34]
5,6-DHT [27]
Fig. 2. Food-findingrate of snails before and after injection with saline, L-NAME or DHT, respectively, as an indicator of successful foodattraction conditioning.Z2 tests: ***P < 0.001; n.s., not significant.
T. Teyke / Neuroscience Letters 206 (1996) 29-32
100
-
!1o$o
o O
rd o
50 ¸
cond. [2O]
saline [8]
L-NAME [121
Fig. 3. Food-findingrate of conditioned snails before and after injection of L-NAMEor vehicle. The high food-finding rates indicate that memory recall and olfactory orientation are not impaired by injection of LNAME.Z2 tests: n.s., not significant. find the food. This rate is similar to that of saline-injected controls (83%; P = 0.4), and to the food-finding rate before injection (P = 0.4). These results suggest that recall and retention of memory is independent of NO, and that blocking NOS does not impair olfactory orientation. Blocking the serotonergic system by injection of 5,6DHT did not affect food-attraction conditioning. As shown in Fig. 1D, 7/8 of the injected snails were able to locate the food. The performance of snails that were injected 3, 7, or 14 days prior to conditioning was indistinguishable; thus the results were combined. After conditioning the snails for carrot, a total of 85% of the DHTinjected snails were able to locate the carrot (Fig. 2). This is a rate similar to that of saline-injected controls (77%; P = 0 . 2 ) , indicating that depletion of serotonin does not impair food-attraction conditioning. However, DHTinjected snails showed a behavioral particularity beyond those reported earlier [2,13]: they moved more slowly than control animals, suggesting a role of serotonin in locomotion (see also [14]). The average crawling speed of DHT-injected animals during their orientation to food was 1.2 _+0.4 cm/min (n = 12), which is considerably slower than that of control snails (5.4 _+ 1.5 cm/min; n = 12). Using selective blockers for NO and for serotonin, it is demonstrated that NO is involved in the acquisition of food-attraction conditioning in snails, whereas serotonin is not involved in the learning. The acquisition of foodattraction was not completely abolished by injection of LN A M E and about 40% of the injected snails still became conditioned. The reason for this is unclear and might reflect problems with the experimental procedure, or general problems regarding inhibition of learning. On the experimental side, the time window of the efficacy of LN A M E on NOS has not been determined in snails. In this study, L-NAME was injected 1 h prior to conditioning, corresponding to the protocol used in octopus ([15]; but see [4]). General problems might reflect the fact that in a
31
variety of learning phenomena, for unknown reasons, treatments attempting to block the acquisition of learning are often only partially successful and a certain percentage of animals remains able to learn. Based upon the effects of nitric oxide on neurons presumably involved in olfactory learning, Gelperin [5] suggested that NO might be involved in aversive olfactory conditioning in Limax. As documented, NO is involved in olfactory conditioning in Helix, although this particular type of conditioning differs considerably from the conditioning paradigm employed in Limax. Contrary to conventional associative conditioning phenomena, foodattraction conditioning is not influenced by depletion of serotonin. As reported earlier, depletion of serotonin abolishes aversive associative conditioning in Helix [2], which corroborates findings in Aplysia that serotonin is involved in associative conditioning [8]. The finding that food-attraction conditioning is not impaired in serotonindepleted snails suggests that food-attraction conditioning is functionally different from conventional associative conditioning paradigms and might therefore utilize a different transmitter. In conclusion, the evidence presented in this paper suggests that the acquisition of food-attraction in Helix involves NO and is independent of serotonin, thus setting apart food-attraction conditioning from other associative conditioning phenomena. Although learning is commonly classified in narrowly defined categories, there appears to be a variety of distinctly different conditioning paradigms, which can be distinguished according to their behavioral function, as well as their underlying biochemical mechanisms. Many thanks to M. Peschel and S. Fuss for testing the animals. Supported in part by a grant from 'Stiftung Rheinland-Pfalz flJr Innovation' and DFG Te 138/3-1. [1] Alkon, D.L., Conditioning-induced changes of Hermissenda channels: relevance to mammalian brain function. In N.M. Weinberger, J.L. McGaugh and G. Lynch (Eds.), Memory Systems of the Brain, Guilford, New York, 1985, pp. 9-26. [2] Balaban, P.M., Vehovszky, A., Maksimova, O.A. and Zakharov, I.S., Effect of 5,7-dihydroxytryptamine on the food-aversive conditioning in the snail Helix lucorum L., Brain Res., 404 (1987) 201-210. [3] Cooke, I.R.C., Edwards, S.L. and Anderson, C.R., The distribution of NADPH diaphorase activity and immunoreactivityto nitric oxide synthase in the nervous system of the pulmonate mollusc Helix aspersa, Cell Tissue Res., 277 (1994) 565-572. [4] Elphick, M.R., Kemenes, G., Staras, K. and O'Shea, M., Behavioral role for nitric oxide in chemosensoryactivation of feeding in a mollusc, J. Neurosci., 15 (1995) 7653-7664. [5l Gelperin, A., Nitric oxide mediates network oscillations of olfactory interneurons in a terrestrial mollusc, Nature, 369 (1994) 6163. [6] Gelperin, A., Rapid food-aversion learning by a terrestrial mollusk, Science, 189 (1975) 567-570. [7] Gelperin, A., Rhines, L.D., Flores, J. and Tank, D.W., Coherent network oscillations by olfactory intemeurons: modulation by endogenous amines, J. Neurophysiol., 69 (1993) 1930-1939.
32
T. Teyke / Neuroscience Letters 206 (1996) 29-32
[8] Hawkins, R.D., Kandel, E.R. and Siegelbaum, S.A., Learning to modulate transmitter release: themes and variations in synaptic plasticity, Ann. Rev. Neurosci., 16 (1993) 625~65. [9] Hermidi, L., Elekes, K. and S.-R6zsa, K., Distribution of serotonin-containing neurons in the central nervous system of the snail Helix pomatia, Cell Tissue Res., 257 (1989) 313-323. [10] Hermidi, L., Kemenes, G. and R6zsa, K., Selective in vivo labeling of serotonergic neurones by 5,6-dihydroxytryptamine (5,6DHT) in the snail Helix pomatia. In H.H. Boer, W.P.M. Geraerts and J. Joosse (Eds.), Neurobiology: Molluscan Models, NorthHolland, Amsterdam, 1987, pp. 22-25. [11] Kandel, E.R. and Schwartz, J.H., Molecular biology of learning: modulation of transmitter release, Science, 218 (1982) 433-443. [12] Kandel, E.R., Cellular Basis of Behavior, Freeman, New York, 1976, 727 pp. [13] Kemenes, G. and S.-R6zsa, K., The role of serotonergic mechanisms in food-induced arousal of the snail Helix pomatia L. In H.H. Boer, W.P.M. Geraerts and J. Joosse (Eds.), Neurobiology:
[14]
[15]
[16]
[17]
[18]
Molluscan Models, North-Holland, Amsterdam, 1987, pp. 277286. Mackey, S. and Carew, T.J., Locomotion in Aplysia: triggering by serotonin and modulation by bag cell extract, J. Neurosci., 3 (1983) 1469-1477. Robertson, J.D., Bonaventura, J. and Kohm, A.P., Nitric oxide is required for tactile learning in Octopus vulgaris, Proc. R. Soe. London B, 256 (t994) 269-273. Robertson, J.D., Bonaventura, J. and Kohm, A., Nitric oxide synthase inhibition blocks octopus touch learning without producing sensory or motor dysfunction, Proc. R. Soc. London B, 261 (1995) 167-172. S~inchez-Alvarez, M., Le6n-Olea, M., Talavera, E., Pellicer, F., SLnchez-lslas, E. and Martfnez-Lorenzana, G., Distribution of NADPH-diaphorase in the perioesophageal ganglia of the snail, Helix aspersa, Neurosci. Lett., 169 (1994) 51-55. Teyke, T., Food-attraction conditioning in the snail, Helix pomatia, J. Comp. Physiol., 177 (1995) 409-414.