Fig 1.
Antisense-SK2 treatment reduced expression of hippocampal SK2 channels.
(A) RT-PCR analysis was performed on RNA extracted from a single hippocampus of naïve mice or from a single hippocampus of mice that were pre-treated with vehicle, control oligonucleotides (ODNs) or SK2 antisense ODNs. Each reaction mixture contained a set of primers specific for the cDNA of hypoxantine-phosphoribosyl-transferase (HPRT), used as internal control. (B) Bar graphs show the relative band intensities on the basis of densitometric analysis as ratios of SK2 and HPRT mRNA after 27 cycles of co-amplification from 7–8 mRNA samples. Representative western blots showing the analysis of SK2 (C), SK3 (D) and SK1 (G) protein homogenates from single hippocampi of mice pre-treated with vehicle, control oligonucleotides (ODNs) or SK2 antisense ODNs. Bars represent mean western blot band intensities ± SEM for SK2 (D), SK3 (F) and SK1 (H) proteins from hippocampal homogenates (n = 9) (Bonferroni multiple comparisons test: * p < 0.05).
Fig 2.
Enhanced basal synaptic transmission and LTP in hippocampal slices from mice pre-treated with SK2 antisense ODNs.
The distance between the stimulating and recording electrodes was kept constant between slices. (A) Input-output curve of fEPSP slope (mV/ms) versus stimulus (V) at the SC-CA1 pyramidal cell synapse in naive mice and mice pre-treated with vehicle, antisense ODNs against SK2 channels and control ODNs. The maximal fEPSP slopes were significantly larger in the SK2 antisense-treated mice than those in the naive, vehicle and control ODNs-treated mice. Data are presented as mean ± s.e.m. (B) Relationship between the slope of the evoked fEPSPs from panel A and the corresponding fiber volley amplitude. SK2 antisense-treated mice exhibit a greater postsynaptic response than control groups to similar presynaptic depolarization. Data are presented as mean ± s.e.m. (C) Comparison of PPF in naive mice and mice pre-treated with vehicle, antisense ODNs against SK2 channels and control ODNs. No differences were found between these four groups of mice. Data presented are the mean ± SEM of the facilitation of the second response relative to the first response. Insets: Field EPSPs recorded in response to paired-pulse stimulation at an interstimulus interval of 50 ms in slices from all four treatment groups as indicated. (D) TBS-LTP elicited in slices from mice that were pre-treated with SK2 antisense ODNs was significantly enhanced when compared to LTP induced in slices from naive mice. There was no statistical difference between control ODNs-injected mice and mice that were pre-injected with vehicle. Insets: Responses shown are fEPSPs recorded during baseline (upper row) and 55–60 min (bottom row) after the induction of LTP (post-induction). Traces are averages of five consecutive responses. Statistical significance was determined by two-way ANOVA followed by Bonferroni multiple comparisons test (*p < 0.05).
Fig 3.
Inhibition of SK2 channel function impaired contextual fear memory.
(A, B) Mice injected intrahippocampally (i.h.) or intracerebroventricularly (i.c.v.) with antisense-SK2 ODNs showed reduced freezing when compared to naive animals. No significant difference in freezing scores was seen in mice injected with vehicle or control ODNs (n = 9–11/ group). (C) Intrahippocampal injection of the selective SK2 antagonist Lei-Dab7 impaired freezing when compared to non-injected animals, whereas vehicle-injection had no effect (n = 8–9/group). (D) When Lei-Dab7 was injected intracortically (i.c.), freezing was not different from non-injected animals. Similarly, vehicle-injected mice showed no significant difference in freezing when compared to non-injected mice (n = 6/ group). Percentage of freezing during pre-shock and post-shock was not significantly different in any treatment group from percentage of freezing of non-injected animals. Freezing was measured in the memory test 24 h after training. Data presented are the mean ± SEM. Statistics was performed by repeated measures two-way ANOVA with Bonferroni multiple comparisons test (*p < 0.05).
Fig 4.
Pre-exposure of mice overcomes impaired contextual fear after inhibition of SK2 channel function.
Foot-shock intensity was reduced from 0.7 mA to 0.5 mA. (A) Intrahippocampal injection of the SK2 antagonist Lei-Dab7 before training impaired freezing when compared to non-treated mice. The same treatment showed no impairment when animals were pre-exposed for 5 min to the conditioning context 24 h before the training. Vehicle-treated animals were not different from non-treated animals (n = 7/group). (B) The effects of SK2 blockade before pre-exposure to the conditioning context on subsequent contextual fear conditioning in which 20 sec placement to shock interval is employed (CTX-S 20s). Injection of Lei-Dab7 before 5 min pre-exposure to context eliminated facilitation of the acquisition of context conditioning at a 20 sec placement to shock interval. Injection of vehicle before pre-exposure or 20 sec placement to shock training phase did not influence contextual pre-exposure effect (n = 8–9/group). (C) Lei-Dab7 was injected at the indicated time points after the training and before the memory test. Statistical comparison was made versus non-injected animals (n = 7–9/group). Arrows in the schematic experimental diagram indicate time points of Lei-Dab7 injection. Freezing was measured in the memory test 24 h after training. Data presented are the mean ± SEM. Statistics was performed by two-way ANOVA with Bonferroni multiple comparisons test (*p < 0.05).
Fig 5.
Cleavage of SK2 channel protein after contextual fear conditioning training.
Western blot analysis of SK2 protein levels in (A) hippocampal tissue obtained from naïve mice or 1 h and 3 h after training (context+shock). In the context group, mice were exposed to the training context without receiving a foot-shock. In the shock group, mice received a foot-shock immediately after they were exposed to the training context and were removed immediately after the foot-shock. Experiment was performed twice. (B) Representative immunoblot of SK2 protein levels in hippocampal tissue from naïve, trained and trained vehicle- or Lei-Dab7-injected mice. Mice were injected either 0.5 h before or 0 h after training as indicated. Hippocampal tissue was removed 1 h after training. The SK2(538–555) antibody recognized the 64-kDa SK2 protein and a 10-kDa SK2 C-terminal fragment (top). (C) SK2 protein levels in hippocampal tissue from naïve non-injected, Lei-Dab7-injected naïve, or vehicle-injected trained mice. Hippocampal tissue was removed 90 or 210 min after injection (1h and/or 3h after training). The number of individual samples per treatment was five. Data presented are the mean ± SEM. Statistically significant differences: *p < 0.05 versus naive, #p < 0.005 versus 3 hours context and conditioning groups, ap<0.05 versus training + vehicle.
Fig 6.
SK2 leucine zipper (SK2-LZ) domain peptide enhances fear conditioning and TBS-LTP by inhibiting SK2 current.
(A) Mice intrahippocampally (i.h.) injected at 30 before or immediately after training with either the SK2-LZ peptide (488–526), a random sequenced control peptide or vehicle showed no difference in freezing when compared to non-injected animals. Injection of the selective SK2 antagonist Lei-Dab7 0.5 h before training impaired freezing when compared to non-injected animals. This impairment was rescued if mice were injected with the SK2-LZ peptide (488–526) immediately after training. I.h. injection of SK2-LZ peptide (488–526) alone, either 30 min before training or immediately following training did not affect contextual fear. Freezing was measured in the memory test 24 h after training. n = 7–10 mice/group. Statistics were performed by one-way ANOVA with Bonferroni multiple comparisons test (*p < 0.05). (B) TBS-LTP elicited in slices from mice that were pre-treated with SK2-LZ peptide (488–526) was significantly enhanced when compared to LTP induced in slices from naive mice. There was no statistical difference between control peptide-injected mice and naïve mice. Statistics were performed by two-way ANOVA with Bonferroni multiple comparisons test (*p < 0.05). Insets: Responses shown are fEPSPs recorded during baseline (upper row) and 55–60 min (bottom row) after the induction of LTP (post-induction). Traces are averages of five consecutive responses. (C) Average whole-cell currents were recorded in Jurkat cells endogenously expressing SK2 channels. Voltage ramps were from -100 mV to 100 mV for 100 ms delivered at 2 s intervals. SK2 current was measured at 80 mV and normalized to cell size (in pF) to show current density over time. After 200 seconds external solutions containing 100 μM SK2-LZ peptide (filled circles, n = 6) was applied to the cells for 150 seconds or no external solution application (control; open circles, n = 4). Arrows at 196 s and 348 s indicate the time points at which the current-voltage (I-V) curves in D were obtained (D) Representative I-V curves for endogenous SK2 in Jurkat cells taken after 196 s, right before application (dotted line) and at the end of the experiment at 348 s (solid line) in unexposed control cells (left panel) or when exposed to 100 μM SK2-LZ peptide. Data presented are the mean ± SEM.