Extinction requires new RNA and protein synthesis and the soma of the cell right pedal dorsal 1 in Lymnaea stagnalis

Abstract

Lymnaea stagnalis were operantly conditioned to not perform aerial respiratory behavior. This learned response was subsequently extinguished. Here, we show that spaced extinction training is more effective than massed extinction training, in addition to the occurrence of spontaneous recovery. We also find evidence of a critical period within the first hour after extinction training in which new RNA and protein synthesis must occur for a memory of extinction training to be established. The memory for extinction training can also be extended using cooling and by preventing aerial respiration from occurring after extinction training. In addition, we demonstrate that memory formation of extinction training requires the soma of the cell right pedal dorsal 1, a cell that we have previously shown to be necessary for long-term memory consolidation and reconsolidation. This finding implies that the events that lead to the formation of extinction memory occur in the same cell that is responsible for long-term memory of operant conditioning. All of these data are consistent with the hypothesis that, during extinction, a new associative memory is being formed and that this new memory covers up, but does not abolish, the "old" memory.

Figure 8.

Extinction requires the soma of RPeD1. A, Thirty-three animals received two 45 min operant conditioning sessions separated by 1 hr. One hour later, animals underwent surgery in which the soma of either LPeD1 (n = 22) or RPeD1 (n = 11) was ablated. Two days later (to allow for surgical recovery), extinction training was administered: two 45 min extinction sessions separated by 1 hr. Four hours later, all animals were tested for savings. LPeD1 soma-ablated animals demonstrated memory for the extinction training; TS (L) was significantly different (p < 0.01) than the last training session but not significantly different (p > 0.05) from the first training session. RPeD1 soma-ablated animals did not demonstrate memory for extinction training but of operant conditioning; TS (R) was not significantly different (p > 0.05) from the last training session and was significantly lower (p < 0.01) than the first training session (ANOVA; F_{(32, 2)} = 67.975; p < 0.0001). B, Fifteen animals received two 45 min operant conditioning sessions separated by 1 hr. One hour later, the soma of RPeD1 was ablated. A test for savings was administered 2 d later: TS was significantly lower (p < 0.01) than TR1 and not significantly different (p > 0.05) from TR2 (ANOVA; F_{(14, 2)} = 67.1629; p < 0.0001). A change of context (CC) test was administered to demonstrate that these animals are still capable of aerial respiratory behavior. CC was not significantly different (p > 0.05) from TR1 and was significantly higher (p < 0.01) than TR2.

Figure 7.

Protein and RNA synthesis are required to form a memory for extinction training. A, Seventy-five animals received three 45 min operant conditioning sessions (TR1 and TR2 were administered on the same day 1 hr apart, followed by TR3 the next day). One hour later, animals received three 45 min extinction sessions (E1 and E2 were administered on the same day 1 hr apart, followed by E3 the next day). Immediately after E3, animals were injected with either saline (n = 26), actinomycin D (n = 28), or anisomycin (n = 21). Two hours later, a test for savings was administered to all animals. Only the animals injected with saline showed memory for extinction training (ANOVA; F_{(74, 4)} = 80.9878; p < 0.0001); TS was significantly different (p < 0.01) from both TR1 and TR3, thus not meeting the criteria for operant memory designation. Animals injected with either actinomycin D or anisomycin still showed memory for operant conditioning; TS for both groups was significantly lower (p < 0.01) than TR1 and not significantly different (p > 0.05) than TR3. B, Thirty animals received three 45 min extinction sessions (E1 and E2 were administered on the same day 1 hr apart, followed by E3 the next day) without any prior operant conditioning. Immediately after E3, animals were injected with either saline (n = 10), actinomycin D (n = 10), or anisomycin (n = 10). Two hours later, a test for savings was administered to all animals. There were no differences (p > 0.05) between the three injected groups.

Figure 1.

Three extinction sessions are needed for successful extinction. Forty-two animals received three 45 min training sessions (TR1 and TR2 were administered on the same day 1 hr apart, followed by TR3, which was given the next day). Animals were then randomly divided into three groups (n = 14 each). Only animals that received three extinction sessions demonstrated memory for extinction at the 2 hr test for savings (ANOVA; F_{(41, 4)} = 74.5208; p < 0.0001). A, The first group received one 45 min extinction session (E1) 1 hr after TR3 and then received a test for savings (TS) 2 hr later. These animals did not show extinction but instead still demonstrated memory for operant conditioning: TS was significantly lower than TR1 (p < 0.01) and not significantly different than TR3 (p > 0.05). B, The second group received two 45 min extinction sessions (E1, E2) 1 hr after TR3, each separated by 1 hr. A test for savings (TS) 2 hr after E2 revealed memory for operant conditioning and not for the extinction training. TS was significantly lower than TR1 (p < 0.01) and not significantly different than TR3 (p > 0.05). C, The third group received three 45 min extinction sessions (E1, E2, E3) beginning 1 hr after TR3, each separated by 1 hr. The test for savings (TS) 2 hr after E3 revealed memory for the extinction training and not for operant conditioning. TS was significantly different (p < 0.01) from both TR1 and TR3, thus not meeting both criteria for operant memory designation.

Figure 3.

Spaced extinction training is more effective than massed extinction training. All animals were first operantly conditioned and subsequently received either spaced or massed extinction training. Only those animals that received the spaced extinction training demonstrated memory for extinction during the test for savings (ANOVA; F_{(82, 5)} = 127.4121; p < 0.0001). A, All animals (n = 83) received three 45 min operant conditioning training sessions (TR1 and TR2 were administered on the same day 1 hr apart, followed by TR3 the next day). Animals were then divided into four groups (B–E). B, Animals (n = 28) received three 45 min extinction sessions (E1–E3), each separated by 1 hr, beginning 1 hr after TR3 and tested 2 hr after E3. The test for savings was significantly different (p < 0.01) from both TR1 and TR3, thus not meeting the criteria for operant memory designation. C, Animals (n = 28) received one 135 min extinction session beginning 1 hr after TR3 and tested 2 hr after E1. The test for savings was significantly lower (p < 0.01) than TR1 and not significantly different (p > 0.05) than TR3, thus meeting both criteria for operant memory designation. D, Animals (n = 14) received one 135 min extinction session beginning 1 hr after TR3 and tested 4 hr after E1. The test for savings was significantly lower (p < 0.01) than TR1 and not significantly different (p > 0.05) than TR3, thus meeting both criteria for operant memory designation. E, Animals (n = 13) received one 135 min extinction session beginning 3 hr after TR3 and tested 2 hr after E1. The test for savings was significantly lower (p < 0.01) than TR1 and not significantly different (p > 0.05) than TR3, thus meeting both criteria for operant memory designation.

Figure 5.

Cooling can extend the memory for extinction training. Twenty-six animals received three 45 min operant conditioning sessions (TR1 and TR2 were administered on the same day 1 hr apart, followed by TR3 on the next day). One hour later, animals received three 45 min extinction sessions (E1 and E2 were administered on the same day 1 hr apart, followed by E3 the next day). One hour later, all animals were immediately placed in prechilled 4°C water for 22 hr, after which they were transferred to room temperature water. Two hours later, animals were tested for savings. Animals still showed memory for extinction training when they normally demonstrate spontaneous recovery for operant conditioning (Fig. 2). TS was not significantly different (p > 0.05) from TR1 and was significantly higher (p < 0.01) than TR3, thus not meeting the criteria for operant memory designation. (ANOVA; F_{(25, 2)} = 46.8073; p < 0.001).

Figure 4.

Cooling blocks memory formation for extinction training. Forty animals received three 45 min operant conditioning sessions (TR1 and TR2 were administered on the same day 1 hr apart, followed by TR3 on the next day). One hour later, animals received three 45 min extinction sessions (E1 and E2 were administered on the same day 1 hr apart, followed by E3 the next day). Half the animals (n = 20) were maintained at room temperature (RT) whereas the other half (n = 20) of the animals were immediately placed in prechilled 4°C water for 1 hr and then immediately transferred to room temperature water. All animals were tested for savings 2 hr after extinction training. The test for savings for animals kept at room temperature was significantly higher (p < 0.01) than the test for savings for animals that were first put in 4°C water for 1 hr (ANOVA; F_{(39, 3)} = 60.4611; p < 0.0001). Animals maintained at room temperature demonstrated memory for extinction training; TS was significantly different (p < 0.01) from both TR1 and TR3, thus not meeting the criteria for operant memory designation. Animals that were placed in 4°C water for 1 hr immediately after extinction training did not show memory for extinction training but instead demonstrated memory for operant conditioning. TS was significantly different (p < 0.01) than TR1 and not significantly different (p > 0.05) from TR3.

Figure 2.

Demonstration of spontaneous recovery at 24 hr. Twenty-six animals received three 45 min training sessions (TR1 and TR2 were administered on the same day 1 hr apart, followed by TR3, which was given the next day). One hour later, all animals then received three 45 min extinction sessions (E1 and E2 were administered on the same day 1 hr apart, followed by E3, which was given the next day). Animals were then divided randomly into two groups. All animals demonstrated spontaneous recovery at 24 hr whether or not they received a test for savings 2 hr after extinction training (ANOVA; F_{(25, 4)} = 30.0463; p < 0.0001). A, The first group (n = 14) received two tests for savings (TS1 and TS2): one at 2 hr after E3 and one at 24 hr after E3. Animals demonstrated memory for extinction training at 2 hr; TS1 was significantly different from both TR1 (p < 0.05) and TR3 (p < 0.01), thus not meeting the criteria for operant memory designation. At 24 hr, however, animals demonstrated spontaneous recovery for the memory for operant conditioning; TS2 was significantly lower than TR1 (p < 0.01) and not significantly different than TR3 (p>0.05). B, The second group (n = 12) received only one test for savings (TS1), 24 hr after E3. These animals demonstrated spontaneous recovery for the memory for operant conditioning; TS1 was significantly lower than TR1 (p < 0.01) and not significantly different than TR3 (p > 0.05).

Figure 6.

Submersion can extend the memory for extinction training. Forty-two animals received three 45 min operant conditioning sessions (TR1 and TR2 were administered on the same day 1 hr apart, followed by TR3 the next day). One hour later, animals received three 45 min extinction sessions (E1 and E2 were administered on the same day 1 hr apart, followed by E3 the next day). The animals were then divided into three groups (n = 14 each). A, The first group received two tests for savings (TS1 and TS2): one at 2 hr after E3 and one at 24 hr after E3. Memory for extinction training was demonstrated at TS1; TS1 was significantly different (p < 0.01) from both TR1 and TR3, thus not meeting the criteria for operant memory designation. Spontaneous recovery for the memory of operant conditioning was demonstrated at TS2; TS2 was significantly lower (p < 0.01) than TR1 and not significantly different (p > 0.05) from TR3, thus meeting the criteria for operant memory designation. B, The second group was immediately submerged after E3 and only taken out for the two tests for savings: one at 2 hr after E3 and one at 24 hr after E3. Memory for extinction training was demonstrated at both tests for savings; TS1 and TS2 were both significantly different (p < 0.01) from both TR1 and TR3, thus not meeting the criteria for operant memory designation. C, The third group was submerged immediately after E3 and only taken out for a test for savings at 24 hr after E3. These animals demonstrated memory for extinction training; TS was significantly different (p < 0.01) from both TR1 and TR3, thus not meeting the criteria for operant memory designation (ANOVA; F_{(41, 6)} = 36.8938; p < 0.0001).