Two Strains of Lymnaea stagnalis and the Progeny from Their Mating Display Differential Memory-Forming Ability on Associative Learning Tasks

Abstract

The pond snail Lymnaea stagnalis learns and forms long-term memory (LTM) following both operant conditioning of aerial respiratory behavior and classical conditioning of taste aversive behavior. In the present study, we examined whether there are interstrain differences in the ability to form LTM following these two types of conditioning. A strain of Lymnaea (TC1) collected in Alberta, Canada exhibits superior memory-forming ability following aerial respiratory operant conditioning compared to a laboratory-reared strain of Lymnaea from Netherlands known as the Dutch strain. We asked whether the offspring of the Canadian TC1 and Dutch snails (i.e., filial 1 (F₁) cross snails) would have the superior memory ability and found, rather, that their memory ability was average like the Dutch snails. That is, the Canadian TC1 snails have superior ability for LTM formation following aerial respiratory operant conditioning, but the Dutch and the generated F₁ cross have average ability for memory forming. We next examined the Canadian TC1, Dutch and F₁ cross snails for their ability to learn and form memory following conditioned taste aversion (CTA). All three populations showed similar associative CTA responses. However, both LTM formation and the ratio of good-to-poor performers in the memory retention test were much better in the Dutch snails than the Canadian TC1 and F₁ cross snails. The memory abilities of the Canadian TC1 and F₁ cross snails were average. Our present findings, therefore, suggest that snails of different strains have different memory abilities, and the F₁ cross snails do not inherit the memory ability from the smart strain. To our knowledge, there have been a limited number of studies examining differences in memory ability among invertebrate strains, with the exception of studies using mutant flies.

Keywords: F1 cross; Lymnaea; aerial respiratory operant conditioning; conditioned taste aversion; interstrain differences.

Figure 1

Training procedures for aerial respiratory operant conditining and conditioned taste aversion (CTA) in Lymnaea. (A) Aerial respiratory operant conditining. The pond water was made hypoxic (≤5% O₂) by bubbling with N₂ before training, and bubbling was then continued throughout training. Snails were acclimatized for 10 min before the training session. During the 0.5 h training session, each time a snail attempted to open its pneumostome at the water’s surface, the pneumostome was gently poked. For memory test, the protocol used for 0.5 h training session was repeated at 24 h after the training session. (B,C) CTA. (B) For forward conditioning (i.e., CTA training), snails were subjected to 10 pairings of sucrose administration (100 mM, conditioned stimulus, CS) for 5 s and electric shock (9 V, 0.4 μA, unconditioned stimulus, US) for 0.2 s. The inter-trial interval was 65.2 s. For backward conditioning, the stimuli were reversed. That is, 10 pairings of application of electric shock (US) for 0.2 s and application of sucrose (CS) for 5 s were used. The inter-trial interval was also 65.2 s in the backward confitioning. For naive control experiments, the CS and the US were replaced with the application of DW. The timng was the same as that of the forward training. (C) The time schedule is shown for training. Snails were deprived of food for 1 day before training (except Figure 5). A pretest and two post-tests were performed to count the number of feeding responses to sucrose (bites/min) for 1 min after the application of CS for 5 s. One post-test was performed 10 min after training, and the other post-test was done 24 h after training.

Figure 2

Comparison of the pneumostome opening attempts between the training session (training) and the memory test (test) in the Dutch snails (black, N = 60), the Canadian TC1 snails (white, N = 60) and the F₁ cross snails (hatched, N = 28) in the operant conditioning of aerial respiratory behavior. The number of pneumostome opening attempts was counted over 0.5 h during the training session and the memory test 24 h later. The data are shown in percentages, indicating that the results of F₁ cross snails resembled those of the Dutch snails. We calculated the statistical significant using the raw data. **P < 0.01; N.S., not-significant. The statistical analysis was performed usign a paired t-test.

Figure 3

Comparison of the feeding responses to the sucrose CS among the Dutch snails (black), the Canadian TC1 snails (white) and the F₁ cross snails (hatched) in the classical conditioning trial of taste aversion behavior. The feeding response was examined at pretest, 10 min post-test and 24 h post-test in response to the CS. (A) Forward conditioning (i.e., CTA training) for the three cohorts (i.e., the Dutch, the Canadian TC1 and the F₁ cross snails) was performed. The results showed that (1) there was a significant difference in the feeding responses shown in (A) (two-way repeated measures ANOVA: for strains F_(2,91) = 3.84, P < 0.05; for time course F_(2,182) = 118.18, P < 0.01; for interaction F_(4,182) = 2.66, P < 0.05). The simple effect test showed a significangt difference for strains at 24 h post-test (F_(2,91) = 5.04, P < 0.01). Further, a Holm’s multiple comparison showed *P < 0.05 between the Dutch and F₁ cross snails; *P < 0.05 between the Dutch and Canadian TC1; and P > 0.05 between the Canadian TC1 and F₁ cross snails. (2) There were no significant differences in the feeding response to the sucrose CS at the pretest (F_(2,91) = 6.07, P > 0.05). (3) The feeding responses to the CS were significantly suppressed in all three cohorts (for example, P < 0.01 for the Canadian TC1 snails), but there were no significant differences among the three cohorts at the 10 min post-test (F_(2,91) = 1.27, P > 0.05). (B) Backward conditioning (pairings of the US first and then the CS) for the Dutch, the Canadian TC1 and the F₁ cross snails showed that (1) there were no significant differences among the three cohorts at the pretest, the 10 min post-test and the 24 h post-test (two-way repeated measures ANOVA: for strains F_(2,52) = 0.42, P > 0.05); but (2) there were significant differeces in the feeding responses shown in B (two-way repeated measures ANOVA: for time course F_(2,79.32) = 9.22, P < 0.01. Further, a Holm’s multiple comparison showed *P < 0.05 between the pretest and 10 min post-test; *P < 0.05 between the pretest and 10 min post-test; *P < 0.05 between the 10 min post-test and 24 h post-test). There were no significant interaction (two-way repeated measures ANOVA: for interaction F_(3.05,79.32) = 0.81, P > 0.05). (C) The naive snails also showed that (1) there were no significant differences among the three cohorts at the pretest, the 10 min post-test and the 24 h post-test (two-way repeated measures ANOVA: for strain F_(2,52) = 0.81, P > 0.05). (2) There were no significant differences in the feeding responses (two-way repeated measures ANOVA: for time course F_(1.57,81.69) = 2.95, P > 0.05). There were no significant interaction (two-way repeated measures ANOVA: for interaction F_(3.05,79.32) = 0.81, P > 0.05).

Figure 4

Comparison of the ratio between the good and poor performers at the memory retention tests among the Dutch, the Canadian TC1 and the F₁ cross snails. Pie charts express the ratio (percentage) of the good and poor performers. (A) At 10 min post-test, no significant differences were found (P > 0.05) in the ratios of good performers (black) or poor performers (white) among the three cohorts. (B) At 24 h post-test, there was a significant difference among the three cohorts (P < 0.01). That is, the Canadian TC1 and the F₁ cross snails exhibited poor formation of CTA-long-term memory (LTM).

Figure 5

CTA in the Canadian TC1 snails without 24-h food deprivation. If the food deprivation before CTA training had been stressful to the Canadian TC1 snails, we expected that the memory scores would have declined. We therefore trained the food-satiated snails by the CTA training paradigm. However, there was no significant difference among the pretest (black), the 10 min post-test (white) and the 24 h post-test (hatched). That is, the learning and memory were worse in snails in the satiated state than in snails with modest food deprivation (i.e., food deprivation of 1 day), as previously reported.