Systemic inhibition of mammalian target of rapamycin inhibits fear memory reconsolidation

Abstract

Background: Established traumatic memories have a selective vulnerability to pharmacologic interventions following their reactivation that can decrease subsequent memory recall. This vulnerable period following memory reactivation is termed reconsolidation. The pharmacology of traumatic memory reconsolidation has not been fully characterized despite its potential as a therapeutic target for established, acquired anxiety disorders including posttraumatic stress disorder (PTSD). The mammalian target of rapamycin (mTOR) kinase is a critical regulator of mRNA translation and is known to be involved in various forms of synaptic plasticity and memory consolidation. We have examined the role of mTOR in traumatic memory reconsolidation.

Methods: Male C57BL/6 mice were injected systemically with the mTOR inhibitor rapamycin (1-40mg/kg), at various time points relative to contextual fear conditioning training or fear memory retrieval, and compared to vehicle or anisomycin-treated groups (N=10-12 in each group).

Results: Inhibition of mTOR via systemic administration of rapamycin blocks reconsolidation of an established fear memory in a lasting manner. This effect is specific to reconsolidation as a series of additional experiments make an effect on memory extinction unlikely.

Conclusions: Systemic rapamycin, in conjunction with therapeutic traumatic memory reactivation, can decrease the emotional strength of an established traumatic memory. This finding not only establishes mTOR regulation of protein translation in the reconsolidation phase of traumatic memory, but also implicates a novel, FDA-approved drug treatment for patients suffering from acquired anxiety disorders such as PTSD and specific phobia.

Figures

Figure 1. Systemic rapamycin inhibits fear memory consolidation

a, Injection of rapamycin 30 min prior to training impairs recall. Pre-train baseline (veh vs. rap, p = 0.54), 1 min post-train (veh vs. rap, p = 0.30), 24 h post-train (veh vs. rap, p < 0.0001). N = 12 in both groups. Legend applies to panels a, b. b, Injection of rapamycin 2 min following training impairs recall 24 h later (veh vs. rap, p < 0.0001). N= 12 in both groups. c, A dose-response curve for rapamycin administered immediately after training is shown. Bars represent percentage of freezing 24 h after training. A one-way ANOVA revealed a main effect of drug (F4,45 = 14.29, p < 0.0001; post hoc Tukey’s test, vehicle vs. 1.0 mg/kg, p = 0.044, vs. 10 mg/kg, p = 0.023, vs. 20 mg/kg, p = 0.00014, vs. 40 mg/kg, p = 0.00013). Rap 1.0 mg/kg does not differ from 10.0 mg/kg, (p = 0.99) but differs from 20 and 40 mg/kg, (p = 0.018, 0.0063 respectively). Rap 10 mg/kg differs from 20 and 40 mg/kg (p = 0.035, 0.013, respectively). Rap 20 mg/kg does not differ from 40 mg/kg (p = 0.99). No significant differences were observed during training (Pre-Train Baseline) across groups (p>0.05). N=10 in all groups. Error bars represent SEM in all figures.

Figure 2. Systemic rapamycin inhibits reconsolidation or facilitates extinction

a, Multiple trial post-training rapamycin inhibits subsequent fear memory recall. Repeated injection of rapamycin 2 min after extinction trials for 4 days impairs subsequent retrieval following a single pairing. This effect is long lasting as subsequent memory recall tested in the absence of rapamycin 7 days after the last extinction trial remains impaired. Repeated measures ANOVA revealed main effects of drug (F1,38 = 51.67, p < 0.0001), day (F4,152 = 56.12, p < 0.0001), and interaction between drug and day (F4,152 = 5.79, p = 0.00023). Student’s t-test of days 1–4 and 11 indicate p = 0.91, p < 0.0001, p = 0.00064, p = 0.0026, p < 0.0001, respectively). N = 20 in both groups. b, Reactivation of a fear memory is necessary for the rapamycin effect on subsequent recall. Multiple daily injections of rapamycin in the absence of memory reactivation do not affect subsequent memory recall (veh vs. rap, p = 0.43). No significant differences were observed during pre-training across groups (p > 0.05). N=10 in both groups.

Figure 3. Systemic rapamycin decreases fear memory reconsolidation

a Post-reactivation injection of rapamycin or anisomycin following a single memory reactivation significantly reduced subsequent recall of a fear memory 7 days later (probe) in the absence of drug. This effect was equivalent to that of the protein synthesis inhibitor anisomycin. One-way ANOVA revealed a main effect of condition (F2,33 = 6.19, p = 0.0052), post hoc Tukey’s test indicated that vehicle differed from rapamycin and anisomycin, which didn’t differ (veh vs. rap, p = 0.016, vs. aniso, p = 0.0099, aniso vs. rap, p = 0.98) b, Reminder shock had no effect on post-reactivation rapamycin effect on subsequent memory, similar to anisomycin. One-way ANOVA revealed a main effect of condition during probe test (F2,33 = 17.73, p = 0.000006). Post hoc Tukey’s test indicated that vehicle differed from rapamycin and anisomycin treated mice, while rapamycin and anisomycin effects were equivalent (veh vs. rap, p = 0.00032, vs. aniso, p = 0.00013, aniso vs. rap, p = 0.54). RS = reminder shock. c, Comparison of reminder shock (+ RS) versus no-reminder shock (no RS) groups on the final day of testing for each drug. One-way ANOVA: main effect of condition, F5,66 = 23.84, p < 0.00001, post hoc Tukey comparisons, veh no RS vs. veh + RS, p = 0.99, rap no RS vs. rap + RS, p = 0.93, aniso no RS vs. aniso + RS, p = 0.99. No significant differences were observed during training or reactivation (48 h) across groups (p > 0.5). N = 12 in all above groups. d, Reminder shock used in b & c (0.2 mAx1) does not cause significant learning/memory in naïve mice tested 24 h after training (solid circles). The same 0.2 mAx1 reminder shock reinstates contextual fear memory recall following 4 days of extinction after training with one (closed squares) or two (closed triangles) pairings of 0.5 mA footshocks with context.

Figure 4. The effect of systemic rapamycin after reactivation is long-lasting and stronger than anisomycin

Post-reactivation injection of rapamycin or anisomycin following a single memory reactivation significantly reduced subsequent recall of a fear memory 21 days later in the absence of drug (probe) (ANOVA: main effect of drug, F2,33 = 19.16, p = 0.000003; post hoc Tukey tests, veh vs. rap, p = 0.000001, veh vs. aniso, p = 0.032, aniso vs. rap, p = 0.00048). There was no difference between groups during training or reactivation (not shown, p > 0.05 for both).

Figure 5. Systemic rapamycin does not alter activity or anxiety-like behavior

a, Rapamycin-treated mice exhibited normal locomotor activity as measured 2 h in the locomotor apparatus (ANOVA: F1,22 = 1.36, p = 0.26). b, Anxiety-like behavior in the rapamycin-treated mice did not differ from vehicle controls as measured by time spent in the light side of the dark/light box (p = 0.48) or number of entries into the light side (not shown, p = 0.91). Legend in b applies to Panels c–d. c, There was no difference in time spent in the open arms in the elevated plus maze in rapamycin-treated mice (p = 0.71) or number of open arm entries (not shown, p = 0.93). d, Shock threshold to elicit flinching, vocalizing, and jumping was normal in rapamycin-treated mice (p = 0.72, p = 0.46, p = 0.48, respectively).