TNC^−/− mice (Evers et al., 2002) were inbred on the C57BL/6 background. Ten- to twelve-week-old male TNC^−/− and TNC^+/+ littermates were obtained from heterozygous breeding and were kept under an inverted 12:12 h light:dark cycle (lights off at 07:00) and standard housing conditions (23 ± 1°C; 50% humidity; food and water ad libitum). Behavioral tests were performed in an experimental room adjacent to the animal facility and illuminated with dim red light. Experiments were performed in the middle of the dark phase of the cycle. All materials were cleaned with soapy water, water and ethanol (75%) between mice. Experiments were carried out in accordance with the European Community Council Directive (86/609/EEC), and the procedures used were approved by the State of Hamburg. Care was taken to minimize pain or discomfort for the animals.

The polyclonal antibodies against the Ca_v1.2 and Ca_v1.3 subunits of L-VGCCs were kindly provided by R. Westenbroek and W. Catterall and are described elsewhere (Hell et al., 1996). Monoclonal antibody against rabbit glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was obtained from Chemicon International (Temecula, CA, USA).

The immunohistochemical analysis of Ca_v1.2 and Ca_v1.3 expression was performed as described by Kochlamazashvili et al. (2010). For Western blotting, hippocampi were homogenized in 200 μl TE buffer (50 mM Tris/HCl, 5 mM EDTA, pH 8). After determination of the protein concentration using the BCA^TM Protein assay (Thermo Scientific, Rockford, IL, USA), 50 μg proteins per lane were subjected to SDS-PAGE on 10% gels followed by Western blot analysis. Proteins were transferred to a nitrocellulose membrane (Protran, Schleicher and Schuell, Dassel, Germany), and the membrane was blocked with 5% non-fat dry milk powder in PBS, pH 7.5. The membrane was incubated with primary antibody (1:1000) overnight at 4°C with shaking, washed in PBS with 0.05% Tween (PBS-T), and probed with HRP-conjugated secondary antibody (1:10000 in PBS containing 5% milk powder) for 1 h. After washing, immunodetection was performed using the chemiluminescent substrate with extended duration (Pierce, Bonn, Germany) on X-ray films (Kodak Biomax-ML, Sigma-Aldrich). Band intensities were densitometrically quantified using the image software TINA 2.09 (DesignSoft Inc., Budapest, Hungary).

After brief CO₂ sedation, decapitation and removal of the brain, transverse hippocampal sections were cut with a Leica VT 1000 M vibratome (Leica, Nussloch, Germany) in ice-cold artificial cerebrospinal fluid (ACSF) containing (in mM): 250 sucrose, 25 NaHCO₃, 25 glucose, 2.5 KCl, 1.25 NaH₂PO₄, 2 CaCl₂, and 1 MgCl₂ (pH 7.3, adjusted with NaOH), as described (Evers et al., 2002). The slices were then kept at room temperature for at least 2 h before the start of recordings in carbogen-bubbled ACSF, containing 125 mM NaCl, instead of 250 mM sucrose. Recordings of field excitatory postsynaptic potential (fEPSP) were performed in the stratum radiatum of the CA1b subfield with glass pipettes filled with ACSF and having a resistance of 1–2 MΩ. Schaffer collaterals/commissural fibers were stimulated with a bipolar electrode placed approximately 300 μm closer to the CA3 subfield than the recording electrode. Basal synaptic transmission was monitored at 0.05 Hz. Four TBSs were applied to induce LTP with the inter-TBS interval of 20 s. TBS consisted of 10 bursts delivered at 5 Hz. Each burst consisted of four pulses delivered at 100 Hz. The duration of pulses was 0.2 ms, and the stimulation strength was set to provide baseline fEPSPs with amplitudes of approximately 50% from the subthreshold maximum. To restore LTP in TNC^−/− mice, an activator of L-VGCC was added to ACSF for 20 min, starting 15 min before induction of LTP ((+)Bay K-8644, Tocris, Bristol, UK; stock of 100 mM in ethanol, applied at a final concentration of 10 μM).

Hippocampal slices (350 μm) of 4- to 6-week-old mice were prepared as described previously (Balaban et al., 2004) and investigated under submerged conditions at 32–34°C. Perfusion medium contained (in mM) 125 NaCl, 2.5 KCl, 2 CaCl₂, 1.25 NaH₂PO₄, 25 NaHCO₃, 1.5 MgCl₂, 25 D-glucose and 0.5 L-glutamine and was bubbled with 95% O₂ and 5% CO₂. Recordings from pyramidal cells in the CA1 region of the hippocampus were conducted with patch electrodes, containing (in mM) 127 K-gluconate, 20 KCl, 2 MgCl₂, 2 Na₂ATP, 10 HEPES and 0.025 of the fluorescent calcium sensitive dye Oregon Green 488 BAPTA 1 (Molecular Probes, Eugene, OR, USA). To induce Ca²⁺ influx, we applied either depolarizing pulses through the recording pipette, or synaptic stimuli through extracellular electrodes. Synaptic stimuli were applied in four theta bursts at 5 Hz. The strength of the synaptic stimuli was adjusted to evoke 1–4 action potentials in each burst. Depolarizing pulses were applied via the recording pipette in a theta-burst-like manner as five pulses at 50 Hz; the duration of each pulse was 10 ms. The strength of the injected current was adjusted so that 3–5 action potentials were evoked by a burst. The recording of Ca²⁺ fluorescence started 20–30 min after rupturing the membrane to let the dye penetrate the cell. For imaging, a CCD camera SenSys1400 (Photometrics, Muenchen, Germany) was used. Acquisition of the imaging data and its synchronization to intracellular stimulation and recording of electrophysiological data was carried out using MetaMorph software (Universal Imaging Corporation, Downingtown, PA, USA). Fluorescence changes of Oregon Green were measured with single wavelength excitation (470 ± 20 nm) and emission >510 nm. Ca²⁺ concentration changes were expressed as ΔF/F₀, where F₀ is the fluorescence intensity when the cell is at rest, and ΔF is the change in fluorescence during stimulation. The signals were normalized per number of action potentials. Ca²⁺ imaging experiments and analyses were performed from two groups of mice without knowledge of their genotypes.

This paradigm was used to test long-term memory and extinction of the conditioned response. Mice were conditioned in the context, which consisted of a chamber (23.5 × 23.5 cm and 19.5 cm high) with Plexiglas walls and ceiling, and a stainless-steel grid-floor from which an electric shock could be elicited. The chamber was surrounded by a black curtain and illuminated by white light (10 lux). The test was performed over three consecutive days. On day 1, mice were placed in the center of the cage and received three electric footshocks (0.25 mA, 1 s) at 120, 160, and 200 s. At 240 s the recording was terminated, and mice were immediately returned into their home cage. On day 2, 24 h after conditioning, mice underwent an extinction trial and were placed again in the context for 28 min without receiving any footshock. On day 3, mice underwent a recall trial that consisted of being placed in the conditioning chamber for 4 min without receiving any footshock. The conditioned response was analyzed by quantifying the percentage of time spent freezing (defined as absence of body movements for at least 1 s). Freezing behavior was automatically analyzed using a modified version of the infrared sensor Mouse-E-Motion (Infra-e-motion, Hamburg, Germany). Long-term memory for the context was evaluated by quantification of the amount of time spent freezing during the first 4 min of the extinction trial on day 2. Short- and long-term extinction of the conditioned responses was evaluated by quantification of time spent freezing during the last 4 min of the extinction trial on day 2 and during the recall trial on day 3.

Nifedipine (25 mg/kg) was suspended in 10% Cremophor EL/PBS vehicle and diltiazem (15 mg/kg) was suspended in saline (0.9% NaCl in water). Intraperitoneal injections were performed 50 min (nifedipine) or 20 min (diltiazem) prior to the recall trial and extinction protocol performed on day 2 of the contextual fear conditioning test.

Comparisons between two groups were performed with the two-tailed t-test. Behavioral data were evaluated using a multifactorial analysis of variance (ANOVA) followed by Newman-Keuls post hoc tests when appropriate: two-way ANOVA (with “genotype” and “treatment” as between groups factors), mixed two-way (with “genotype” as between groups factor and “time” as within group factor) and mixed three-way ANOVA (with “genotype” and “treatment” as between groups factors and “time” as within group factor). All tests were two tailed, and the level of significance was set at p < 0.05. Data are presented as the mean ± standard error of the mean (SEM).

Our previous work (Evers et al., 2002) demonstrated impairment of LTP in the CA1 area of the hippocampus induced by TBS of Schaffer collaterals/commissural fibers in TNC^−/− mice. Here, we first aimed at reproducing these findings. LTP was induced by four episodes (one per 20 s) of TBS consisting of 10 bursts (one per 200 ms) of four pulses at 100 Hz. Figure 1A shows impairment of LTP in TNC^−/− mice. The magnitude of LTP in TNC^−/− mice (111.3 ± 2.4%, n = 11 slices) was significantly reduced compared to LTP in wild-type (TNC^+/+) mice (132.6 ± 4.8%, n = 9, p = 0.002). Since our prior work showed that this reduction in LTP magnitude in TNC^−/− mice was mimicked by treating slices from TNC^+/+ mice with the L-VGCC blocker nifedipine and that the effects of TNC deficiency and nifedipine showed full occlusion, we hypothesized that the function of L-VGCCs was impaired in TNC^−/− mice (Evers et al., 2002). This hypothesis predicts that impaired LTP in TNC^−/− mice might be restored by the upregulation of L-type channel activity. To test this assumption, we used BAY K-8644, an activator of L-VGCCs. Induction of LTP in TNC^−/− mice in the presence of BAY K-8644 (bath-applied for 20 min starting 15 min before induction of LTP) restored LTP magnitude (133.2 ± 3.6%, n = 9) to the levels seen in TNC^+/+ mice (132.6 ± 4.8%, n = 9, p = 0.922; Figure 1B). In TNC^+/+ mice, the level of LTP induced in the presence of BAY K-8644 (138.4 ± 2.9%, n = 5) was not different from the untreated control (p = 0.323; Figure 1B). These data support the idea that TNC^−/− mice have a partial deficit in the activity of L-VGCCs, which can be compensated by pharmacological activation of these channels. Next, we tested whether reduced activity of L-VGCCs is due to reduced expression or impaired functionality of L-VGCCs.

We first performed immunohistochemical analysis to estimate whether TNC deficiency affects the overall expression or localization of Ca_v1.2 and Ca_v1.3 α1 subunits of neural L-VGCCs. No reduction in expression of either channel subunit was detected in the CA1 area of the hippocampus (Figure 2). Consistent with a previous study (Hell et al., 1996), we observed the strongest expression of L-VGCCs in the somata and proximal dendrites of CA1 pyramidal neurons, but these channels were also apparent in the distal part of dendritic trees.

Next, we performed semiquantitative Western blot analysis with equal protein amounts of total homogenate prepared from the hippocampi of TNC^−/− and TNC^+/+ mice. We used affinity purified polyclonal antibodies against the Ca_v1.2 and Ca_v1.3 subunits of neural L-VGCCs, which recognized two bands of 210–220 and 180–190 kDa (Hell et al., 1996). The total levels of these two forms were quantified by densitometry and normalized against the GAPDH levels. The levels of GAPDH were nearly identical in the compared probes, showing that it is a proper loading and expression control that can be used for normalization of L-VGCC levels. The results indicated elevated, rather than reduced, expression of these two forms of Ca_v1.2 and Ca_v1.3 subunits in the hippocampus of TNC^−/− mice compared to TCN^+/+ mice (Figure 3). It needs to be pointed out that in addition to the bands with the cognate molecular weights for the channel subunits additional bands of unknown identity are seen in Western blots with these antibodies (Figure 2). We therefore decided that these additional bands will not be measured.

In summary, the combined results suggest that there is no deficit in expression of Ca_v1.2 and Ca_v1.3 subunits of L-VGCCs in the hippocampus of TNC^−/− mice, suggesting that impairment of L-VGCC-dependent LTP mechanisms in TNC^−/− mice is due to alteration of either functional properties or cell surface expression of these channels.

To directly test whether TNC deficiency leads to impaired activity of L-VGCCs, we measured Ca²⁺ influx in TNC^−/− and TNC^+/+ neurons using Ca²⁺ imaging. We filled CA1 pyramidal cells with the Ca²⁺ sensitive dye Oregon Green 488 BAPTA-1 and recorded changes in the fluorescence induced by TBS or trains of depolarization pulses applied via a patch pipette (Figures 4A,B). To assess the contribution of L-VGCCs, we measured Ca²⁺ influx in the control conditions and after the addition of nifedipine to the recording medium. The difference between Ca²⁺ signals recorded under these two conditions characterizes the contribution of L-VGCCs. We revealed a significant difference between nifedipine-sensitive components in TNC^+/+ and TNC^−/− mice. In TNC^+/+ mice, application of nifedipine led to a clear reduction in the Ca²⁺ signals in the somata of CA1 pyramidal neurons in response to TBS. In TNC^+/+ mice, the Ca²⁺ signals were reduced by ~10% (to 90.1% of control, reduction by 9.9 ± 3.1%, n = 13, p = 0.008). In contrast, in TNC^−/− mice, the application of nifedipine did not reduce Ca²⁺ signals (101.3% of the control, change of 1.3 ± 3.6%, n = 10; p = 0.726; Figures 4C,D). The difference between genotypes in the reduction of Ca²⁺ signals by nifedipine (and, thus, in the contribution of L-type channels in the two genotypes) was significant (p = 0.022).

To exclude a possible contribution of presynaptic influences, we performed the same measurements but induced Ca²⁺ influx by theta-like bursts of spikes evoked by brief depolarization pulses through the patch pipette at the somata of CA1 pyramidal neurons (five pulses, 10 ms, at 50 Hz, current strength adjusted to evoke 3–5 action potentials per burst). Similar to results with synaptic stimulation, nifedipine significantly reduced Ca²⁺ signals in TNC^+/+ mice (reduction by 6.2 ± 2.4%, n = 13, p = 0.024) but not in TNC^−/− mice (increase by 2.8 ± 2.7%, n = 10, p = 0.327). The difference between genotypes in the reduction of Ca²⁺ signals by nifedipine was significant (p = 0.017).

Thus, Ca²⁺ entry mediated by L-VGCCs during TBS or theta-like direct postsynaptic activation is impaired in the hippocampus of TNC^−/− mice. These results provide strong support for the hypothesis that TNC regulates the activity of L-VGCCs in the CA1 area of the hippocampus (Evers et al., 2002). Next, we asked whether this mechanism plays a role in the regulation of hippocampus-dependent behavior.

To test the possible behavioral consequences of impaired L-VGCC-dependent LTP in TNC^−/− mice, we used a paradigm of contextual fear conditioning and extinction, a cognitive function that requires synaptic plasticity in the hippocampus (for review see Izquierdo et al., 2016). On experimental day 1, mice underwent a conditioning protocol to associate a conditioned context (CC) to an unconditional stimulus (a footshock; US). Mice were placed in a test chamber for 4 min, and three series of electric shock pulses (0.25 mA for 1 s) were delivered at 120 s, 160 s and 200 s. Twenty-four hours after conditioning (day 2), we analyzed the behavior during an extinction trial. Mice were re-exposed for 28 min to the CC (same test chamber) in the absence of the US, and the time spent freezing during the extinction trial was measured. Figure 5A shows the time spent freezing during seven 4-min intervals of the extinction trial. The mixed two-way ANOVA (with “genotype” and “time” as between and within groups factors, respectively) showed a significant effect of the interaction between “genotype” and “time” on time spent freezing during the extinction trial (F_6,132 = 10.0; p < 0.001). Post hoc analyses indicated that there was no difference between TNC^+/+ and TNC^−/− mice during the first 4 min interval (48.3 ± 4.4%, n = 12 vs. 49.3 ± 4.9%, n = 12, p = 0.881). TNC^−/− mice spent more time freezing compared to TNC^+/+ mice during the later phases of the extinction trial (Figure 5A). Thus, whereas long-term memory was intact (as shown by the unaltered freezing time during the first 4 min), short-term extinction was impaired in TNC^−/− mice. To test the longer lasting effects of the extinction, we measured time spent freezing 24 h after the extinction trial. On day 3, mice were placed in the CC for 4 min, and TNC^−/− mice spent more time freezing compared to TNC^+/+ animals, indicating that the difference between genotypes acquired during the extinction session was preserved (Figure 5B; p = 0.008).

In TNC^−/− mice, the L-VGCC blocker nifedipine neither affects Ca²⁺ influx into CA1 pyramidal neurons (Figure 4) nor does it reduce the size of LTP (Evers et al., 2002). We thus asked whether the behavioral effects of nifedipine would also be impaired in TNC^−/− mice. Pharmacological blockade of L-VGCCs is known to inhibit the extinction of fear (Cain et al., 2002; Izquierdo et al., 2016), although it remains a matter of debate whether these effects are due to impairment of specific cognitive processes mediating extinction or caused by a more generalized motor inhibition (Waltereit et al., 2008; Singewald et al., 2015). We performed contextual fear conditioning and extinction tests as described above using TNC^+/+ and TNC^−/− mice injected with either vehicle or the L-type VGCC blockers nifedipine or diltiazem 20 min before the extinction trial on day 2. Diltiazem was used to confirm that possible effects induced by the pharmacological blockade of the L-type VGCCs are not caused by side effects already described for nifedipine (Waltereit et al., 2008).

The results of these experiments (Figures 5C–E,G–I) revealed a significant effect of the interaction between “genotype”, “treatment” and “time” on time spent freezing (nifedipine experiment: F_(6,240) = 10.29; p < 0.0001; diltiazem experiment: F_(6,264) = 19.4; p < 0.0001; mixed three-way ANOVA with “genotype” and “treatment” as between groups factors and “time” as a within groups factor). The results of post hoc analyses indicated that L-VGCC blockers enhance the time spent freezing after 5 min of the extinction trial on day 2 in TNC^+/+ mice compared to vehicle treated TNC^+/+ mice (Figures 5C,G). The L-VGCC blockers had similar effects, although less prominently, in TNC^−/− mice: time spent freezing was enhanced after 13 min in nifedipine- (Figure 5E) and diltiazem-treated mice (Figure 5H) compared to TNC^−/− mice injected with the vehicle. Because the extinction of the conditioned response starts after approximately 5 min of the extinction trial, we performed a two-way ANOVA evaluation on time spent freezing from 5 min to the end of the extinction trial and detected a significant effect of the interaction between “genotype” and “treatment” (F_(1,44) = 14.05; p = 0.0005). Post hoc analyses indicated that nifedipine (Figure 5E) and diltiazem (Figure 5I) significantly enhanced the time spent freezing by TNC^+/+ mice, whereas in the TNC^−/− group values obtained with blockers did not reach a statistically significant difference (nifedipine: p = 0.061; diltiazem: p = 0.087). Moreover, in the vehicle group TNC^−/− mice spent more time freezing than TNC^+/+ mice, but less time than TNC^+/+ mice in the nifedipine (Figure 5E) and diltiazem (Figure 5I) groups. Similar effects of the interaction between “genotype” and “treatment” were observed on time spent freezing during the 4 min trial on day 3, 24 h after the extinction trial (Figure 5F for nifedipine: F_(1,40) = 10.1; p = 0.0029; Figure 5J for diltiazem: F_(1,44) = 7.4; p = 0.0093).

In conclusion, the inhibitory effects of L-VGCCs blockers on extinction were less pronounced in TNC^−/− than in TNC^+/+ mice, indicating an occlusion of the involved mechanisms.