Role of the somatostatin system in contextual fear memory and hippocampal synaptic plasticity

Abstract

Somatostatin has been implicated in various cognitive and emotional functions, but its precise role is still poorly understood. Here, we have made use of mice with somatostatin deficiency, based upon genetic invalidation or pharmacologically induced depletion, and Pavlovian fear conditioning in order to address the contribution of the somatostatin system to associative fear memory. The results demonstrate an impairment of foreground and background contextual but not tone fear conditioning in mice with targeted ablation of the somatostatin gene. These deficits were associated with a decrease in long-term potentiation in the CA1 area of the hippocampus. Both the behavioral and the electrophysiological phenotypes were mimicked in wild-type mice through application of the somatostatin-depleting substance cysteamine prior to fear training, whereas no further deficits were observed upon application in the somatostatin null mutants. These results suggest that the somatostatin system plays a critical role in the acquisition of contextual fear memory, but not tone fear learning, and further highlights the role of hippocampal synaptic plasticity for information processing concerning contextual information.

Figure 1.

SST mutation disrupts both foreground and background contextual fear memory in a gene dosage-dependent manner. (A) In foreground contextual fear conditioning, SST^−/− mice (n = 8) showed significantly decreased freezing compared with SST^+/− (n = 10) and SST^+/+ mice (n = 8). Furthermore, a significant deficit was also evident for SST^+/− animals during retrieval sessions 2 and 3. (B) After auditory cued fear conditioning, SST^−/− mice (n = 13) and SST^+/− mice (n = 16) displayed significantly less freezing to the background context when compared with SST^+/+ mice (n = 12), whereas groups did not differ in their freezing behavior during CS− and CS+ presentation. All data depicted as mean ± SEM; (◊) P < 0.05, (◊◊) P < 0.01, (◊◊◊) P < 0.001 in all plots.

Figure 2.

Acute somatostatin depletion with cysteamine dose-dependently reduces contextual fear memory, while post-training cysteamine induces an unspecific enhancement of fear responses. (A) Pre-training cysteamine treatment (50 mg/kg, n = 10; and 150 mg/kg, n = 8) dose-dependently led to a reduction of conditioned freezing to the background context in C57Bl/6 mice, when compared with saline-treated mice (n = 10). In contrast, all groups showed a comparable amount of freezing upon CS− and CS+ re-exposure. (B) The freezing duration of SST^−/− mice was unaffected by cysteamine application prior to training regardless of the retrieval type studied (context, CS−, and CS+; n = 9 for all groups). (C) C57Bl/6 mice treated post-training with 150 mg/kg cysteamine (n = 11) displayed significantly increased freezing during all parts of the retrieval session (context, CS−, and CS+), when compared with saline-controls (n = 10). This difference was also evident for context and CS− re-exposure when animals treated with 150 mg/kg were compared with the group receiving 50 mg/kg cysteamine (n = 11). (D) Similar to wild-type animals, SST^−/− mice also exhibited generally increased freezing behavior during context, CS−, or CS+ exposure, when treated with 150 mg/kg cysteamine immediately after training (cf. C; n = 9 for all groups). All data depicted as mean ± SEM; (◊) P < 0.05, (◊◊) P < 0.01, (◊◊◊) P < 0.001 in all plots.

Figure 3.

Basal hippocampal synaptic transmission in SST^+/+ and SST^−/− mice. Field EPSPs were determined in CA1 in response to Schaffer-collateral stimulation. (A) Input–output relationships, with fEPSP slope plotted against stimulation intensity. Note the lack of differences between genotypes. (B) Responses to paired pulse stimulation, with interstimulus intervals (ISIs) varied between 50 and 500 msec. Plotted is the ratio of the second to the first response against ISI. Note paired pulse facilitation at short ISIs and lack of difference between genotypes. P > 0.05 for all values in SST^+/+ vs. SST^−/− mice.

Figure 4.

CA1–LTP and paired-pulse facilitation after LTP-induction in SST^+/+ and SST^−/− mice. (A) Field EPSPs in CA1 in response to Schaffer-collateral stimulation at 1/30 Hz, with tetanic stimulation (three trains of 50 pulses at 100 Hz; 30-sec interval between trains) applied at times zero. Tetanic stimulation induced stable LTP, which was significantly smaller in amplitude in SST^−/− as compared with SST^+/+ animals. All but the first value post tetani were significantly different between genotypes for 1 h after induction of LTP (P < 0.05 in Student’s t-test). Representative voltage traces illustrate responses before (gray traces) and after (black traces) tetanization in slices obtained from SST^+/+ and SST^−/− mice, as indicated. Traces are scaled to baseline fEPSP amplitude to allow for better comparison of fEPSP slopes. Scale bars: 10 msec in horizontal, and unit baseline amplitude in vertical direction. (B) Responses to paired-pulse stimulation, with interpulse intervals varied between 50 and 500 msec, determined 1 h after induction of LTP. Plotted is the ratio of the second to the first response (fEPSP slope) against interstimulus interval (ISI). Note paired-pulse facilitation similar to that in Fig. 3B, and the lack of difference between genotypes. P > 0.05 for all ISIs in SST^+/+ vs. SST^−/− mice.

Figure 5.

Effects of pharmacological depletion of somatostatin by application of cysteamine on CA1–LTP. (A) Input–output relationships, with fEPSP slope plotted against stimulation intensity. Note the lack of differences between slices obtained from animals treated with either saline or cysteamine, as indicated (P > 0.45 for all stimulation intensities in Student’s t-test). (B) LTP in CA1 of C57Bl/6 mice ex vivo, with slices prepared 4 h after application of cysteamine (50 mg/kg i.p.) and in a saline-injected control group. Note the reduction of LTP after cysteamine application. All post-tetanic values were significantly different across groups (P < 0.05 for all time points post tetani, Student’s t-test). Representative voltage traces illustrate responses before (gray traces) and 30 min after (black traces) tetanization obtained in slices from animals after saline and cysteamine treatment, as indicated. Traces are scaled to baseline fEPSP amplitude to allow for better comparison of fEPSP slopes. Scale bars: 10 msec in horizontal, and unit baseline amplitude in vertical direction. (C) LTP in CA1 ex vivo 4 h after application of cysteamine (50 mg/kg i.p.) in SST^−/− mice and in a pharmacologically naïve SST^−/− control group. Application of cysteamine to SST^−/− animals did not result in a further reduction of LTP as compared with untreated SST^−/− mice. Induction protocol for LTP in B and C is as in Fig. 4A. Representative voltage traces illustrate responses before (gray traces) and after (black traces) tetanization in slices obtained from SST^−/− mice treated with cysteamine and in untreated SST^−/− controls. Traces are scaled to baseline fEPSP amplitude. Scale bars are as in B.