PMC

Gαs* mice exhibit PKA-dependent deficits in short-term and long-term fear memory. (A) Wild-type (WT, n = 14), Gαs* (n = 8), R(AB) (n = 12), and Gαs* × R(AB) mice (n = 7) were trained to associate a conditioned auditory stimulus (CS) with a 2-sec 1.5-mA footshock (unconditioned stimulus, US). Small vertical arrow indicates onset of the US. (B) Gαs* mice show normal levels of freezing during the training session but significantly lower levels of freezing relative to all other genotypes during both short-term memory (+30 min) and long-term memory (+24 h) tests. (*) Post-hoc vs. WT, R(AB), and Gαs* × R(AB), P < 0.001; (@) post-hoc vs. WT, P < 0.05; (#) post-hoc vs. WT and R(AB), P < 0.02.

Following extended training, Gαs* mice show normal levels of freezing to the CS during the short-term but not long-term memory test. (A) Wild-type (WT, n = 5) and Gαs* transgenic mice (n = 7) were trained to associate a conditioned auditory stimulus (CS) with an unconditioned 1.5-mA footshock stimulus (US). Small vertical arrow indicates four presentations of a 2-sec 1.5-mA footshock. Both short-term (+30 min) and long-term memory (+24 h) for the training context and cue were assessed. (B) Across retrieval tests, Gαs* mice (n = 7) exhibit significantly lower levels of freezing to the training context relative to wild-type littermates (n = 5). In contrast, Gαs* mice show normal levels of freezing to the cue 30 min, but not 24 h, following training; *post-hoc vs. WT, P < 0.01–0.001.

Gαs* mice exhibit PKA-dependent deficits in neuronal activity within cortex as indicated by reduced basal Arc mRNA expression. Arc mRNA levels (% WT) were quantified in wild-type (WT, n = 16–17), Gαs* (n = 12–13), R(AB) (n = 16–17), and Gαs* × R(AB) mice (n = 6–7) across (A) cortical regions surrounding the hippocampus (Vis, visual; Par, parietal association; Som, somatosensory; Ent, entorhinal), (B) the orbital cortex (Orb), (C) the hippocampal subregions, and (D) the medial caudate putamen (CPu; med, medial; lat, lateral) (see for Arc expression in lateral CPu). Data are collapsed across two cohorts of animals, the first naive to fear conditioning and the second sacrificed 8 d following training (see Materials and Methods for further details). (E) Autoradiographs show Arc mRNA expression in animals from the second cohort. *Post-hoc vs. WT, R(AB), and Gαs* × R(AB), P < 0.02; ^@post-hoc vs. WT and R(AB), P < 0.05.

Gαs*-induced decreases in cortical Arc mRNA persist following extended training with four CS US pairings. Arc mRNA levels (% WT home cage) were measured in wild-type (WT) and Gαs* mice sacrificed 30 min following training with four CS–US pairings (n = 6 for each genotype). Arc levels in trained mice were compared to those found in animals sacrificed in parallel from the home cage (WT, n = 5; Gαs*, n = 7). Animals sacrificed from the home cage had undergone testing 8 d previously (data in ; see Materials and Methods for further details). (A) In cortices surrounding the hippocampus and (B) orbital cortex, training with four CS–US pairings significantly increases Arc mRNA levels across genotypes; however, Gαs* mice still show significantly lower levels of Arc mRNA relative to wild-type littermates across behavioral conditions. (C) In the hippocampus, training with four CS–US pairings also increases Arc mRNA levels; however, there is no significant effect of genotype. (D) There is no significant effect of training nor genotype in medial caudate putamen (CPu); however, Gαs* mice show significantly higher levels of Arc across training conditions in lateral CPu. (E) In the amygdala, training with four CS–US pairings significantly increases Arc expression in both wild-type and Gαs* mice, with no significant difference between genotypes. (Abbreviations as in ) (*)Effect of genotype, P < 0.02–0.002; (#) effect of training, P < 0.02.