Hippocampal Estrogen Signaling Mediates Sex Differences in Retroactive Interference

Abstract

Despite being a crucial physiological function of the brain, the mechanisms underlying forgetting are still poorly understood. Estrogens play a critical role in different brain functions, including memory. However, the effects of sex hormones on forgetting vulnerabilitymediated by retroactive interference (RI), a phenomenon in which newly acquired information interferes with the retrieval of already stored information, are still poorly understood. The aim of our study was to characterize the sex differences in interference-mediated forgetting and identify the underlying molecular mechanisms. We found that adult male C57bl/6 mice showed a higher susceptibility to RI-dependent memory loss than females. The preference index (PI) in the NOR paradigm was 52.7 ± 5.9% in males and 62.3 ± 13.0% in females. The resistance to RI in female mice was mediated by estrogen signaling involving estrogen receptor α activation in the dorsal hippocampus. Accordingly, following RI, females showed higher phosphorylation levels (+30%) of extracellular signal-regulated kinase1/2 (ERK1/2) in the hippocampus. Pharmacological inhibition of ERK1/2 made female mice prone to RI. The PI was 70.6 ± 11.0% in vehicle-injected mice and 47.4 ± 10.8% following PD98059 administration. Collectively, our data suggest that hippocampal estrogen α receptor-ERK1/2 signaling is critically involved in a pattern separation mechanism that inhibits object-related RI in female mice.

Keywords: ERK1/2; estrogens; forgetting; hippocampus; object recognition memory; retroactive interference; sex.

Figure 1

Female mice are resistant to retroactive interference. (A–C): Schematic representation of different object recognition memory paradigms used. Standard NOR procedure (A). Retroactive interference procedure (RI; (B)). Proactive interference procedure (PrI; (C)); white squares represent object 1, black triangle represents object 2, yellow circles represent object 3, red triangle represents object 4, and blue triangle represents object 5. (D): Histograms (mean ± SD) showing preference indexes for male mice undergoing Std-NOR, RI, and PrI. Asterisks indicate statistically significant differences between groups assessed by one-way ANOVA (F_(2;25) = 6.731; p = 0.005; Std-NOR vs. RI p = 0.007, Holm–Sidak post hoc test; Std-NOR vs. PrI p = 0.604, Holm–Sidak post hoc test; Std-NOR n = 9; RI n = 11; PrI n = 8). (E): Histograms (mean ± SD) showing preference index for female mice undergoing Std-NOR, RI, and PrI. Asterisks indicate statistically significant differences between groups assessed by one-way ANOVA (F_(2;25) = 0.590; p = 0.562. Std-NOR n = 10; RI n = 10; PrI n = 8). * p < 0.05; ** p < 0.01; n.s., not significant. All graphs and images were realized using CorelDraw21 (Corel Corporation, Ottawa, Ontario, Canada).

Figure 2

Estrogen receptors play a key role in resistance to retroactive interference. (A): Schematic representation of 4-hydroxytamoxifen (4-OHT) or vehicle (Veh) intrahippocampal (I.h.) injection during RI procedure; white squares represent object 1, yellow circles represent object 2, and red triangle represents object 3. (B): Histograms (means ± SD) showing preference indexes of animals undergoing RI after 4-OHT treatment. Note that 4-OHT injection induces memory loss. Asterisks indicate statistically significant differences between groups assessed by Kruskal–Wallis one-way analysis of variance on ranks (p < 0.001; Veh n = 8; 4-OHT n = 9). (C): Schematic representation of RI procedure with I.h. injection of Veh, MPP, or PHTPP, antagonists of the estrogen receptors α and β, respectively. (D): Histograms (means ± SD) showing percentages of PI in animals undergoing RI in the presence of Veh, MPP, or PHTPP. Asterisk indicates statistically significant difference between groups assessed by one-way ANOVA (F_(2;21) = 4.339; p = 0.026; Veh vs. PHTPP p = 0.906, Holm–Sidak post hoc test; Veh vs. MPP p = 0.042, Holm–Sidak post hoc test; n = 8 animals for each group). * p < 0.05; *** p < 0.001; n.s., not significant. All graphs and images were realized using CorelDraw21 (Corel Corporation, Ottawa, ON, Canada).

Figure 3

ERK1/2 inactivation leads to RI-induced memory loss in female mice. (A): Representative Western blots for Akt and ERK1/2 phosphorylation; white squares represent object 1, yellow circles represent object 2, and red triangle represents object 3. (B): Histograms (means ± SD) showing fold induction variation of Akt phosphorylation at S473. One-way ANOVA, F_(1;7) = 0.116; p = 0.745. n = 4 for both groups. (C): Histograms (means ± SD) showing fold induction variation of ERK1/2 phosphorylation at Thr202/Tyr204. One-way ANOVA, F_(1;7) = 16.885; p = 0.006; n = 4 for both groups. (D): Schematic representation of experimental protocols including vehicle (Veh) or PD98059 intrahippocampal (I.h.) injections. (E): Histograms (means ± SD) showing preference indexes of female mice undergoing the RI paradigm in the presence of vehicle or PD98059. Asterisks indicate statistically significant differences between groups assessed by one-way ANOVA (F_(1;14) = 18.368, p < 0.001; n = 8 for both groups). ** p < 0.01; *** p < 0.001; n.s., not significant. All graphs and images were realized using CorelDraw21 (Corel Corporation, Ottawa, Ontario, Canada).