Sex-dependent up-regulation of two splicing factors, Psf and Srp20, during hippocampal memory formation

Abstract

Gene transcription is required for long-term memory (LTM) formation. LTM formation is impaired in a male-specific manner in mice lacking either of the two Ca(2+)/calmodulin-dependent kinase kinase (Camkk) genes. Since altered transcription was suggested to cause these impairments in LTM formation, we used microarrays to screen for CaMKKbeta-dependent gene expression changes. Here we show that the hippocampal mRNA expression of two splicing factors, splicing factor arginine/serine-rich 3 (Sfrs3/Srp20) and polypyrimidine tract-binding protein-associated splicing factor (Psf), is altered in CaMKKbeta-deficient males. In wild-type (WT) mice, the basal expression level in the hippocampus is higher in males than in females, and the sex difference in Srp20 expression is detectable before puberty. Training in two hippocampus-dependent learning tasks, the spatial version of the Morris water maze (MWM) and background contextual fear conditioning, increases the hippocampal mRNA expression of both splicing factors in WT males. However, the increase in Srp20 mRNA expression occurs only in males and not in females, whereas the up-regulation of Psf expression occurs in both sexes. Importantly, control experiments demonstrate that the up-regulation of both splicing factors is specific for the learned associations after contextual fear conditioning. In summary, we provide the first evidence for a regulation of splicing factors during LTM formation and we suggest that alternative splicing contributes to sex differences in LTM formation.

Figure 1.

Lack of CaMKKβ results in altered Srp20 and Psf mRNA expression in the naïve hippocampus in males. (A) Hippocampal Srp20 mRNA expression was significantly reduced in naïve CaMKKβ-deficient males (n = 5) compared with naïve WT males (n = 5). (B) Hippocampal Srp20 mRNA expression levels did not differ between naïve mutant females (n = 6) and naïve WT females (n = 8). (C) Hippocampal Psf mRNA expression was significantly higher in naïve mutants males (n = 7) compared with naïve WT males (n = 6). (D) Hippocampal PSF mRNA levels did not differ significantly between naïve WT (n = 9) and mutant females (n = 6), although increasing the number of animals per group might have revealed a small difference between the female genotypes. Data are means ± SEM; (*) P < 0.05.

Figure 2.

Sex differences in hippocampal Srp20 and Psf mRNA expression levels in WT mice. (A) Hippocampal Srp20 mRNA expression was significantly lower in females than in males independently of the age group considered (P21: five females, four males; adults: seven females, eight males). Sex differences in adult naïve animals were further confirmed (see Figs. 5A, 6B). (B) Srp20 is expressed in the three major hippocampal subfields in both male and female adult mice as determined by in situ hybridization. (C) Hippocampal Psf mRNA expression differed with age (significance stars not shown) and post hoc analysis suggested that the expression was significantly lower in adult females (n = 5) than in adult males (n = 8) and not in P21 mice (six females, four males). Higher hippocampal Psf mRNA expression in adult males than in adult females was also found in Figures 5C and 6D. (D) Psf is expressed in the three major hippocampal subfields in both male and female adult mice as determined by in situ hybridization. Data are means ± SEM; (*) P < 0.05; (***) P < 0.001.

Figure 3.

Hippocampal lesions in female 129B6F1 mice impair spatial memory formation in the Morris water maze. (A) Time to reach the hidden platform differed between the sham-lesioned and lesioned mice. Two-way ANOVA with repeated measures showed that there was a significant effect of treatment (F_(1,13) = 5.9, P < 0.05) and training (F_(8,104) = 14.9, P < 0.001), but no significant interaction (F_(8,104) = 1.5, P = 0.18). (B) During the probe trial on day 3, the sham-lesioned mice (n = 8) searched selectively in the target quadrant (TQ) where the platform had been located (F_(3,28) = 10.8, P < 0.0001), whereas the lesioned mice (n = 7) searched randomly (F_(3,27) = 2.7, P = 0.07). Analysis of time spent in the target quadrant showed a significant effect of lesion (F_(1,13) = 9.0, P < 0.01). (C) Minimum (black) and maximum (gray-shaded) of hippocampal lesions. Note that the maximal lesions are an overestimate, because neocortex overlying the hippocampus became detached during slice preparation. Data are means ± SEM; (**) P < 0.01. Adapted with permission from Elsevier © 2001, Paxinos and Franklin 2001.

Figure 4.

Performance of male and female WT mice in the hidden-platform version of the Morris water maze (MWM). (A) Time to find the hidden platform decreased with training in all groups; planned comparisons suggest that selective males (n = 5) acquired the task faster than their nonselective counterparts (n = 6), while both female groups acquired the task at the same rate (seven selective, five nonselective). (B) In a probe trial given at the end of training, selective animals spent significantly more time swimming in the target quadrant than in any other quadrant (significance stars are omitted); while this was not observed in nonselective animals; no sex effects were observed. (TQ) Target quadrant; (AR) adjacent right quadrant; (AL) adjacent left quadrant; (OP) opposite quadrant. Data are means ± SEM; (***) P < 0.001.

Figure 5.

Hippocampal Srp20 and Psf mRNA expressions are up-regulated after training in the MWM in WT mice, and the up-regulation of Srp20 expression occurs only in males. (A) Hippocampal Srp20 mRNA expression was significantly up-regulated in selective males, but not females. Post-hoc analysis also showed that naïve males expressed higher levels of Srp20 mRNA than naïve females (P < 0.05), which is not indicated for the ease of presentation. Males: seven naïve, five selective; females: seven naïve, six selective. (B) In males, Srp20 mRNA up-regulation occurred in the swim control and nonselective groups, but this was significantly smaller than in the selective group: seven naïves, seven swim controls, five nonselective, five selective. (C) Hippocampal Psf mRNA expression was significantly up-regulated by spatial training in both sexes as shown by two-way ANOVA. Additionally, males expressed significantly higher levels of hippocampal Psf mRNA than females. Males: eight naïve, five selective; females: five naïve, seven selective. (D) Psf mRNA up-regulation was not specific to MWM learning, as nonselective groups and swim controls express higher levels of Psf mRNA than the naïve group. Eight naïves, eight swim controls, six nonselective, five selective. Data are means ± SEM; (*) P < 0.05; (**) P < 0.01; (***) P < 0.001.

Figure 6.

Hippocampal Srp20 and Psf mRNA expressions are regulated by learned associations after contextual fear conditioning in WT mice; the up-regulation of Srp20 expression is male-specific. (A) Male (n = 7) and female mice (n = 3) showed contextual freezing 24 h after conditioning in comparison with a Box control group (eight males, four females). (B) Hippocampal Srp20 mRNA expression was significantly up-regulated by contextual fear conditioning in males, but not in females, as shown by post-hoc analysis. Females: six naïve, five contextually fear conditioned (FC); males: seven naïve, eight FC. (C) In males, hippocampal Srp20 mRNA expression levels did not differ between naïve mice (n = 7) and Box control mice (n = 6), while there was significantly increased expression in FC mice (n = 8). The LI shock control (n = 5) did not result in an up-regulation in comparison with naïve mice (n = 4). (D) Hippocampal Psf mRNA expression was significantly up-regulated by contextual fear conditioning in males (eight naïve, seven FC) and females (five naïve, seven FC) as shown by two-way ANOVA. Additionally, males expressed significantly higher levels of hippocampal Psf mRNA than females. (E) In males, hippocampal Psf mRNA expression levels did not differ between naïve mice (n = 8) and Box control mice (n = 6), while there was significantly increased expression in FC mice (n = 7). There was no difference in expression between naïve (n = 4) and LI control mice (n = 5). Data are means ± SEM; (*) P < 0.05; (**) P < 0.01; (***) P < 0.001.