Failures to reconsolidate memory in a mouse model of Alzheimer's disease

Abstract

Previous studies have demonstrated that the formation of spatial, contextual and trace conditioning memories are impaired in animal models of Alzheimer's disease (AD), consistent with the observations that the first sign of cognitive decline in AD includes difficulties in the acquisition of new information or memory formation. Evidence is accumulating that memory retrieval is a dynamic process in which stored information becomes labile again and needs to be restabilized. However, it is poorly understood how this process referred to as memory reconsolidation is affected in animal models of AD. The present study was designed to use contextual fear conditioning to compare the changes in memory formation and subsequent reconsolidation processes in transgenic mice that overexpress human APP and PS1 harboring five familial AD mutations (5XFAD model). The results clearly demonstrate that cognitive dysfunction starts to occur primarily as reduced levels of contextual learning or memory formation in 5XFAD mice, but it is exacerbated by additional retrieval-dependent retrograde amnesia due to deficient reconsolidation as disease further develops.

Fig. 1

Age-dependent impairments of contextual fear conditioning in 5XFAD mice. (A–C) 5XFAD mice at 3–4 months (A), 6–7 months (B) and 10–15 months (C) of ages, and their respective wild-type littermate mice were trained with 2 CS/US pairings for contextual fear conditioning. 5XFAD mice at 6–7 and 10–15 months of age but not at 3–4 months of age show significantly lower levels of contextual freezing than wild-type controls when tested 24 h after training (n = 15–21 mice per group). * p < 0.05 versus wild-type controls. All data are presented as mean ± SEM.

Fig. 2

Deficient reconsolidation of contextual fear memory in older 5XFAD mice. Experimental design used is presented at the top of each panel. (A) 5XFAD mice at 10–12 months of age and their wild-type littermate mice were trained with 5 CS/US pairings for contextual fear conditioning. Mice received a single 3-min re-exposure to the conditioning context 24 h after training, and were then tested for contextual memory 24 h later. Levels of freezing during re-exposure were not different between 5XFAD and wild-type mice. In contrast, levels of freezing during subsequent memory testing in 5XFAD mice were significantly lower than those of wild-type controls (* p < 0.05) and as compared to those during re-exposure (^#p < 0.05) (n = 12–14 mice per group). (B) When mice received no re-exposure intervening between training with 5 CS/US pairings and memory testing 48 h later, no difference in freezing levels was found between 5XFAD mice and wild-type controls (n = 11–14 mice per group). (C) Weaker training with 1 CS/US pairing dramatically reduced freezing during re-exposure in wild-type mice, but similar levels of freezing were retained during memory testing 24 h later (n = 12 mice per group). All data are presented as mean ± SEM.

Fig. 3

Normal reconsolidation of contextual fear memory in younger 5XFAD mice. Experimental design used is presented at the top of each panel. (A) 5XFAD mice at 3–4 months of age and their wild-type littermate mice were trained with 2 CS/US pairings for contextual fear conditioning. Mice received a single 3-min re-exposure to the conditioning context 24 h after training, and were then tested for contextual memory 24 h later (n = 9–12 mice per group). (B) 5XFAD mice at 6–7 months of age and their wild-type littermate mice were similarly tested for reconsolidation except that they were trained with 5 CS/US pairings (n = 9–10 mice per group). In either case, levels of freezing were not significantly different between 5XFAD (closed columns) and wild-type mice (open columns) or between the re-exposure and subsequent memory test sessions. However, a trend toward a decrease in reconsolidation was observed in 5XFAD mice at 6–7 months of age as compared to wild-type littermate controls during memory testing 24 h after re-exposure (p = 0.11). All data are presented as mean ± SEM.