Caspase-9 mediates synaptic plasticity and memory deficits of Danish dementia knock-in mice: caspase-9 inhibition provides therapeutic protection

In 1928 Ramon y Cajal predicted that weakening of synapses leads to dementia. Long-term potentiation (LTP) is a synaptic plasticity phenomenon that underlies the strengthening of synaptic functions during memory acquisition. Consistent with Ramon y Cajal’s prediction, LTP is defective in the hippocampal Schaffer collateral pathway of FDD_KI mice. However, basal synaptic transmission and paired-pulse facilitation are normal in FDD_KI mice, suggesting that no changes in Ca²⁺ mobilization or alterations in the probability of neurotransmitter release are driven by the Danish mutation [3]. To examine the role of caspases in synaptic plasticity, we analyzed the effect of the cell-permeable, irreversible pan-caspase inhibitor Z-VAD-fmk on LTP. Hippocampal slices were perfused either with Z-VAD-fmk (at 10 μM concentration) or vehicle for 60 min before inducing LTP. Z-VAD-fmk reversed the LTP deficit of Danish samples and did not alter LTP in wild-type mice (Figure 1).

Most caspases are expressed in the hippocampus. To start dissecting which caspase(s) play(s) a role in LTP deficits in FDD_KI mice, we analyzed the effect of Z-LEHD-fmk and Z-DEVD-fmk, which have partially overlapping inhibitory patterns of caspases inhibition. As shown in Figure 1, Z-LEHD-fmk behaved similarly to Z-VAD-fmk (i.e. it fully rescued the LTP deficit of FDD_KI mice, without imposing on normal synaptic plasticity). In contrast, LTP Z-DEVD-fmk delayed, but did not rescue, the insurgence of LTP deficits in FDD_KI mice (Figure 1). The evidence indicates that some, but perhaps not any, caspases are involved in the pathogenesis of LTP deficits of FDD_KI mice.

We reasoned that if caspases have a causative role in dementia, then inhibiting caspase activity should in addition rescue memory deficits of FDD_KI mice. To test for this, we analyzed the effect of Z-LEHD-fmk and Z-DEVD-fmk on the memory deficits of FDD_KI mice in a longitudinal study. Memory was analyzed using novel object recognition (NOR), a non-aversive memory test that relies on the mouse’s natural exploratory behavior. The first NOR study showed that during training, FDD_KI and WT mice spent the same amount of time exploring two identical objects (Figure 2A, left panel), showing no discrimination between these two identical objects (Figure 2A, right panel). The following day, one of the two old objects was replaced with a new one to test the mouse’s memory. WT mice preferentially explored the novel object; conversely FDD_KI mice spent the same amount of time exploring the two objects as if they were both novel to them, showing that they had no memory of the objects from the previous day (Figure 2B, the left panel shows the time spent exploring each object, while the right panel show the discriminatory ratio). After this first test, the mice were rested for one day before re-testing. In this second experiment, the mice were injected in the lateral ventricle with Z-LEHD-fmk 1 h before the training/testing trials. Treated FDD_KI mice spent significantly more time exploring the novel object, just as treated controls did (Figure 2B). Following 2 days of rest, a new NOR test performed without treatments showed that FDD_KI mice had relapsed into amnesia (Figure 2B), demonstrating that the therapeutic effect of Z-LEHD-fmk is short-lived. One day later, mice were injected 1 h before the training/testing with Z-DEVD-fmk. Z-DEVD-fmk neither improved memory of FDD_KI mice nor altered performance of WT animals (Figure 2B). Thus, Z-LEHD-fmk (which fully corrected the synaptic deficit of FDD_KI mice) rescued, albeit temporarily, the memory deficit of FDD_KI mice. In contrast, Z-DEVD-fmk (which, as show in Figure 1, was inefficient in normalizing LTP in FDD_KI mice) did not. Although Z-LEHD-fmk and Z-DEVD-fmk are commonly referred to as specific caspase-9 and caspase3/7 inhibitors, respectively, these compounds show overlapping selectivity [22]. Therefore from these experiments it is difficult to pinpoint the caspase(s) involved in these pathogenic processes. However, altogether these data suggest that one or more caspases mediate synaptic/memory deficits of FDD_KI mice.

Based on the evidence that caspases are pathogenic in FDD_KI mice, we sought biochemical evidence of caspase activation and/or activity. Because FDD_KI mice have deficits in hippocampal-dependent memory and synaptic activity, which are associated with learning and memory, we tested whether we could detect signs of caspase activation in hippocampal synaptic preparations of 12 month-old mice. As discussed above, caspases are synthesized as zymogens (FL-caspase). Effector caspases are cleaved by initiator caspases (cl.-caspase) and this cleavage leads to activation of effector caspases. To allow unequivocal identification of active caspase we used an unbiased in vivo active caspase-trapping assay [23]. The caspase activity probe bVAD is the best way to determine whether caspases are active since bVAD binds irreversibly to all caspases that are active. In other words, if a caspase is active and its active site is available, bVAD will bind to it. Because bVAD is biotinylated, it can be isolated on streptavidin agarose along with any active caspase that is bound to it. This strategy has also the advantage of enriching for the apical active caspase rather than the downstream caspases in a pathway that involves a cascade of caspase activation. To determine which caspases are active, FDD_KI and WT mice were injected in one hippocampus with 100 nmol of bVAD. In these experiments, we utilized 6 (Figure 3A) or 5 (Figure 3B) month-old mice since the memory deficits of FDD_KI mice start at around 4–5 months of age [3]. Two hrs post treatment, the injected region and the contra-lateral non-injected area were dissected, and bVAD-caspase complexes were isolated on streptavidin-agarose beads and analyzed by Western blotting. bVAD captured greatly more FL-caspase-9, but not FL-caspase-8, from the hippocampus of the FDD_KI sample as compared to the WT littermate sample (Figure 3A). The binding was specific because streptavidin-agarose beads did not pull-down active FL-caspase-9 from homogenates prepared from the contra-lateral, non-injected sample. Cl.-caspase-3 and −6 were not trapped by bVAD (Figure 3A). The inability to isolate cl.-caspase-3 and cl.-caspase-6 may depend on the fact that bVAD inhibits caspase-9 activity, thereby inhibiting processing of effector caspases-3 and −6 by active caspase-9. This possibility is not very likely because in FDD_KI mice there is probably ongoing caspase activation and bVAD will bind to any active caspase present at the moment of bVAD administration. Alternatively, cl.-caspase-3 may not be available for bVAD-binding because it is complexed in vivo with endogenous inhibitor of apoptosis proteins (IAPs). Lastly, cl.-caspase-3 and cl.-caspase-6 may be captured by bVAD at very low levels that are below the detection power of our experimental system. This is indeed a possibility given the low level of material that can be harvested in this experimental setting and the evidence that cl-caspase-3 and cl-caspase-6 are not detectable in the input material either.

To determine whether active caspase-9 was present in synaptic fractions, we repeated the experiment and performed bVAD pull-downs from synaptosomal fractions. As shown in Figure 3B, active caspase-9 was also isolated from synaptosomal fractions of FDD_KI but not WT mice. Blotting for caspase-3, -6 and −8 showed once more absence of detectable active caspase-3, -6 or −8 in these synaptosomal preparations (data not shown). To formally exclude that the differences between WT and FDD_KI mice illustrated above did not depend on disparity of bVAD delivery in vivo, we prepared organotypic hippocampal cultures from 5 month-old WT and FDD_KI mice. BVAD trapped significantly more active caspase-9 from organotypic hippocampal culture of FDD_KI mice than WT littermates (Figure 3C). Once again, we could not detect active FL-caspase-8, cl.-caspase-3 and cl.-caspase-6 neither in WT nor in FDD_KI sample. Altogether these data indicate that caspase-9 is excessively activated in Danish dementia mice. Moreover, the data suggest that, if the Danish mutation triggers a cascade of caspase activation, caspase-9 is the apical caspase in such a cascade.

The findings that reducing caspase activity with commercial peptide inhibitors rescues synaptic/memory deficits and that caspase-9 is active in FDD_KI mice, suggest that caspase-9 is involved in the pathogenesis of these deficits. To specifically determine the functional relevance of caspase-9 activity in memory loss pathogenesis, we specifically inhibited caspase-9. As a control, we also used a specific inhibitor of caspase-8 activity. Mammals express a family of cell death inhibiting proteins known as IAPs. IAPs contain BIR domains (Baculovirus Inhibitor of apoptosis protein Repeats), which perform specific functions. One member of this family, XIAP, is a potent specific inhibitor of active caspase-9, caspase-3, and caspase-7. The XIAP-BIR3 domain is a specific inhibitor of active caspase-9, and the XIAP-BIR2-linker domain inhibits active caspase-3 and caspase-7 [24]. Serpins are also caspases inhibitors and CrmA (a cowpox serpin) inhibits caspase-8 (as well as caspase-1, which is involved in inflammatory responses) but not other murine caspases [25]. To provide intracellular delivery, XIAP-BIR3 and CrmA were disulfide-linked to Penetratin1 (Pen1), a cell-penetrating peptide [23]. Upon entry into the cell the reducing environment of the cytoplasm reduces the disulfide linkage. This releases the peptide cargo and allows it to act at its target. We have previously shown that Pen1-XBIR3 inhibits caspase-9 dependent cell death using primary hippocampal neuron cultures, and that Pen1-XBIR3 delivery to the CNS blocks caspase-9 in an in vivo model of cerebral ischemia [23]. NOR experiments were used to assess the effect of Pen1-XBIR3 on memory. Six groups of mice (3 groups of FDD_KI mice and 3 groups of WT littermates) were injected in the lateral ventricle either with vehicle alone, Pen1-XBIR3 or Pen1-CrmA 1 hr before the training/testing trials. Pen1-XBIR3 treated FDD_KI mice spent significantly more time exploring the novel object showing reversal of the memory deficits (Figure 4A and B). On the contrary, Pen1-CrmA treated FDD_KI mice showed memory deficits comparable to those observed in vehicle-treated FDD_KI mice. Neither Pen1-XBIR3 nor Pen1-CrmA altered memory in WT animals. Following 5 days of rest, a new NOR test performed without treatments showed that the therapeutic effect of Pen1-XBIR3 persisted for at least 5 days post injection (Figure 4C and D). Our previous studies showed that one dose of Pen1-XBIR3 provided functional protection against ischemia for 3 weeks post-infarction [23]. Thus, Pen1-XBIR3 rescued the memory deficit of FDD_KI mice, while Pen1-CrmA did not. These data indicate that excessive activation of caspase-9 in FDD_KI mice is an essential step in the pathogenesis of memory loss.

Mice were handled according to the Ethical Guidelines for Treatment of Laboratory Animals of Albert Einstein College of Medicine. The procedures were described and approved in animal protocol number 20100404.

FDD_KI, mice are on a C57BL/6 J background and were generated and maintained at the Animal facility of the Albert Einstein College of Medicine. Mice were handled according to the Ethical Guidelines for Treatment of Laboratory Animals of Albert Einstein College of Medicine. The procedures were described and approved in animal protocol number 200404.

Transverse hippocampal slices (400 μm) from 13–14 month old WT and FDD_KI mice were transferred to a recording chamber where they were maintained at 29°C and perfused with artificial cerebrospinal fluid (ACSF) continuously bubbled with 95% O₂ and 5% CO₂. The ACSF composition in mM was: 124 NaCl, 4.4 KCl, 1 Na₂HPO₄, 25 NaHCO₃, 2 CaCl₂, 2 MgSO₄, and 10 glucose. CA1 field-excitatory-post-synaptic potentials (fEPSPs) were recorded by placing both the stimulating and the recording electrodes in CA1 stratum radiatum. After 90 minutes incubation, 10 μM Z-VAD-fmk was added. For LTP experiments, a 30 min baseline was recorded every minute at an intensity that evoked a response approximately 35% of the maximum evoked response. LTP was induced using a θ-burst stimulation (four pulses at 100 Hz, with bursts repeated at 5 Hz and each tetanus including one ten-burst train). Responses were recorded for 90 min after tetanization and plotted as percentage of baseline fEPSP slope. Z-VAD-fmk is from R&D Systems.

Dr. Xiaosong Li at the Animal Physiology core of the Albert Einstein College of Medicine surgically implanted the cannula. Using stereotaxic surgery performed under ketamine/xylazine anesthesia, mice were implanted with cannula (Plastics One Inc.) into the lateral ventricle (coordinates from bregma: A/P −0.4 mm, M/L − 1 mm, D/V −2.5 mm) or hippocampus (coordinates from bregma: A/P −2.45 mm, M/L − 1.5 mm, D/V −1.7 mm). Z-LEHD-fmk (800 nM), Z-DEVD-fmk (800 nM), or saline were delivered into the lateral ventricle at the rate of 1 μl per minute using a CMA 400 syringe pump, for a total volume of 1 μl. Pen1-XBIR3 (23 μM), Pen1-CrmA (16 μM), or saline were delivered into the lateral ventricle at a rate of 1 μl per minute using a CMA 400 syringe pump, for a total volume of 2 μl. Z-LEHD-fmk and Z-DEVD-fmk are from R&D Systems.

Organotypic hippocampal slices were prepared and cultured as described previously [32]. Briefly, 400 μm slices were prepared using a tissue chopper. Slices were transferred onto a cell culture insert that was placed into a 6-well plate in an incubator with 5% CO₂ and 78% O₂. The plate contained 1 ml of culture media (MEM with Glutamax-1 supplemented with D-glucose, horse serum, nystatin, HEPES, EBSS, and Pen-Strep). The slices were cultured in the interface method. After 24 h of culture, the media was replaced with culture media containing 45 μM bVAD for 3 h after which the slices were collected and lysed separately in 10% glycerol, 150nM NaCl, 0.2% NP-40, 20 mM Tris–HCl (pH 7.3) with protease and phosphatase inhibitors. bVAD-caspase complex precipitation was performed by preclearing with Sepharose beads and isolation with streptavidin-agarose beads as described in the preceding chapter. After the final wash/pelleting, caspase-bVAD complexes were boiled off of streptavidin beads into 1× SDS sample buffer without reducing agent. Beads were pelleted at 14,000 rpm for 10 min, and the supernatant was transferred to a fresh tube and resolved by SDS-PAGE.

The mice were acclimated to the testing room for 30 min after being moved. Each mouse was placed into a 40 cm × 40 cm open field chamber with opaque walls, 2 ft high. Each mouse was allowed to habituate to the normal open field box for 10 min, and repeated again 24 h later, in which we manually recorded the number of entries into and time spent in the center of the locomotor arena. As previously reported [3], open field studies showed that FDD_KI mice have no defects in habituation, sedation, risk assessment and anxiety-like behavior in novel environments.

Novel object recognition began 24 h after the second open field session, and was performed as previously described [3, 33]. Briefly, NOR consisted of two sessions 24 h apart. In the first session, the mice were placed into the open field chamber with two identical, non-toxic objects, 12 cm from the back and sidewalls of the open field box, and 16 cm apart from each other. An 8 min session, in which the time exploring each object was recorded; an area 2 cm² surrounding the object is defined such that nose entries within 2 cm of the object were recorded as time exploring the object. We will refer to this as training trial. The animal was then returned to its home cage and 24 h late, placed into the open field box again. This time, there was one object identical to the previous one, and one novel object. We will refer to this as the test trial. The mice were given another 6 min to explore, and the amount of time exploring each object was recorded. Mice that spent < 7 seconds exploring the objects were omitted from the analysis [33]. Results were recorded as Time Spent Exploring each object an object discrimination ratio (ODR), which is calculated by dividing the time the mice spent exploring object 1 (for the training trial) or the novel object (for the test trial) by the total amount of time exploring the two objects.

For synaptic preparations, isolated hippocampi were homogenized (w/v = 10 mg tissue/100 ml buffer) in Hepes-sucrose buffer (20 mM Hepes/NaOH pH 7.4, 1 mM EDTA, 1 mM EGTA, 0.25 M sucrose) supplemented with protease and phosphatase inhibitors. Homogenates were centrifuged at 800 g for 10 min. The supernatant (S1) was separated into supernatant (S2) and pellet (P2) by spinning at 9,200 g for 15 min. Synaptosome fractions represent: S1, post-nuclear-supernatant; S2, cytosol, soluble proteins and light membrane; P2, crude synaptosomal fraction. The S1 and P2 fractions were analyzed by western blot using the following antibodies: α-caspase-3 (9662/Cell signaling); α-caspase-6 (9762/Cell Signaling); α-caspase-8 (4790/Cell Signaling); α-caspase-9 (ab28131/Abcam. Secondary antibodies conjugated with horse-radish peroxidase are from Southern Biotechnology.

Western blot images were scanned with Epson Perfection 3200 Photo scanner and were analyzed with NIH ImageJ software.

All data are shown as mean ± s.e.m. Statistical tests included two-way ANOVA for repeated measures and t-test when appropriate. All experiments were performed in a blinded fashion.