Recall and reconsolidation of contextual fear memory: differential control by ERK and Zif268 expression dosage

Abstract

Compelling evidence points to the existence of independent cellular processes involved in the consolidation and reconsolidation of memory. For instance, a double dissociation has been reported between hippocampal Extracellular-Regulated Kinase-1/2 (ERK1/2) activity being necessary for contextual fear conditioning (CFC) consolidation but not reconsolidation. Conversely, hippocampal expression of the immediate early gene Zif268 is necessary for CFC reconsolidation but not consolidation. Since we previously reported that ERK1/2 controls the transcription of Zif268 in the hippocampus, we examined the precise role of ERK1/2 activity and Zif268 gene expression dosage in CFC memory processing. For this, we first assessed performance of Zif268 homozygous and heterozygous mutant mice in a CFC paradigm. Whereas Zif268-/- mice displayed a deficit of both consolidation and reconsolidation, Zif268+/- mice displayed a selective deficit of reconsolidation only, therefore pointing to the relationship between Zif268 gene expression dosage and CFC memory processing. Zif268 gene expression dosage interfered with the reconsolidation process if and only if CFC memory was relatively recently encoded and directly reactivated. Furthermore, CFC memory strengthening previously reported to involve Zif268 expression in the hippocampus was spared in Zif268+/- mice. Finally, blocking ERK1/2 activity prior to CFC retrieval prevented the deficit of reconsolidation observed in Zif268+/- mice. Collectively, these results highlight a tight relationship between Zif268 gene expression dosage and CFC memory processing. They also suggest that ERK1/2 activity upon CFC memory recall is necessary for its retrieval, a prerequisite for its reactivation and subsequent reconsolidation.

Figure 1. Effect of Zif268 gene dosage on CFC consolidation and reconsolidation.

A) Experimental design. Zif268+/+, Zif268+/− and Zif268−/− mice were trained (Tr.) and retrieval sessions were conducted 1 day (Ret.1), 2 days (Ret.2) and 9 days (Ret.3) after training. B) Freezing behavior was measured in Zif268+/+ (white circles), Zif268+/− (grey circles) and Zif268−/− (black circles) mice during each retrieval session. Zif268−/− mice show a deficit of consolidation whereas Zif268+/− mice display a selective impairment of reconsolidation. Data are means ± SEM; n = 10 mice per group. ^###p<0.001, Zif268+/− or −/− versus Zif268+/+; *p<0.05; ***p<0.001 present versus past retrieval.

Figure 2. Effect of Zif268 gene dosage on reactivated and non-reactivated CFC memories.

A) Experimental design. Zif268+/− mice were trained (Tr.) and retrieval sessions were conducted 1 day (Ret.1), 2 days (Ret.2) and 9 days (Ret.3) after training. B) Freezing behavior was measured in Zif268+/− mice during each test session. One group was repeatedly submitted to the 3 retrieval sessions (Multiple retrieval – grey circles, solid line) and three groups were submitted only once to one of the 3 retrieval sessions (Single retrieval – grey circles, dotted line). The deficit of reconsolidation observed in Zif268+/− mice is dependent on re-exposure to the context. Data are means ± SEM; n = 8–10 mice per group.

Figure 3. Effect of Zif268 gene dosage on the reconsolidation of 1 and 8 days-old CFC memories.

A) Experimental design. Zif268+/− mice were trained (Tr.) and retrieval sessions (Ret.1 and Ret.2) were conducted in independent groups 1 and 9 days, 2 and 9 days or 8 and 9 days after training. B) Freezing behavior was measured in Zif268+/− mice during each test session 1 and 9 days (grey circles), 2 and 9 days (grey triangles) or 8 and 9 days (grey squares) following training. The reconsolidation deficit observed in Zif268+/− mice is not observed when the memory has been encoded 8 days earlier. Data are means ± SEM; n = 9–10 mice per group. **p<0.01, present versus past retrieval.

Figure 4. Effect of Zif268 gene dosage on CFC memory strengthening.

A) Experimental design. Zif268+/+ and Zif268+/− mice were trained for two successive days (Tr.1 and Tr.2) and a retrieval (Ret.) session was conducted on subsequent day. B) Freezing behavior was measured in Zif268+/+ (white circles) and Zif268+/− (grey circles) mice before each shock-pairings as well as during the retrieval session. Zif268 knockdown did not affect the strengthening of CFC memory. Data are means ± SEM; n = 11 mice per group.

Figure 5. Effect of MEK inhibition on Zif268 gene dosage-dependent impairment of CFC reconsolidation.

A–B) Experimental designs. Zif268+/+ mice and Zif268+/− mice were trained (Tr.) and on subsequent day, vehicle (white diamonds) or SL327 (black diamonds) were administrated 1h before retrieval (Ret.). An additional retrieval session was conducted on the next day. C–D) Freezing behavior was measured in vehicle (VEH, circles) and MEK inhibitor (SL327, triangles) treated Zif268+/+ (white) and Zif268+/− (grey) mice during each retrieval session. SL327 transiently impaired freezing behavior of Zif268+/+ when injected before the first retrieval session. SL327 had a protecting effect against the deficit of reconsolidation observed in Zif268+/− mice. Data are means ± SEM; n = 9–10 mice per group in C; n = 8–9 mice per group in D. ^#p<0.05, SL327 versus Vehicle; *p<0.05; ***p<0.001, present versus past retrieval.