Pharmacological inhibition of asparaginyl endopeptidase by δ-secretase inhibitor 11 mitigates Alzheimer’s disease-related pathologies in a senescence-accelerated mouse model

Animal experimental procedures were approved by the Animal Experimentation Ethics Committee and Institutional Animal Care and Use Committee (IACUC) at Shanghai Jiao Tong University School of Medicine, and carried out strictly in accordance with the guidelines of Association for Assessment and Accreditation of Laboratory Animal Care [21]. Male SAMP8 and SAMR1 mice were provided by the Beijing Vital River Laboratory Animal Science and Technology Co., Ltd., China. Animals were housed in the SPF animal facility at 24 ± 2 °C under a 12 h/12 h light/dark cycle with free access to water and a standard rodent diet. Four-month-old SAMP8 mice were chosen for experiments and were randomly divided into two groups (n = 15/group): δ-secretase inhibitor 11 treatment and vehicle treatment groups. The SAMP8 mice were treated with δ-secretase inhibitor 11 (10 mg/kg) or vehicle once daily via oral gavage over a period of 3 months.

The AEP inhibitor δ-secretase inhibitor 11 (7-morpholin-4-yl-benzo[1,2,5]oxadiazol-4-ylamine, PubChem CID: 1095027) was purchased from J&K Scientific Ltd. (Beijing, China). The δ-secretase inhibitor 11 was first dissolved in dimethyl sulfoxide as the stock solution and then diluted in 0.9% NaCl solution containing gum arabic for systemic treatment. The primary antibody for mature AEP was from R&D Systems Inc. (Minneapolis, MN). Antibodies for β-actin, Iba1, and microtubule-associated protein 2 (MAP-2) were purchased from Abcam (Cambridge, MA). Antibody for AEP-cleaved tau (N368) was from Sigma-Aldrich (St. Louis, MO). Antibody for phospho-tau (Thr231) was from Thermo Fisher Scientific (Carlsbad, CA). Antibodies for postsynaptic density protein 95 (PSD-95) and synaptophysin (SYP) were from Cell Signaling Technology (Danvers, MA).

Recombinant mouse AEP (R&D Systems Inc., Minneapolis, MN) was diluted to 50 μg/ml in activation buffer (0.1 M NaOAc, 0.1 M NaCl, pH 4.5), incubated for 6 h at 37 °C with or without AEP inhibitor, and then diluted to 2 ng/μl in assay buffer (50 mM MES, 250 mM NaCl, pH 5.5). Fifty microliter of 2 ng/μlL AEP was loaded in a plate, and added with 50 μl of 200 μM substrate Z-Ala-Ala-Asn-AMC (Bachem AG, Switzerland) diluted with assay buffer. Absorbance was read at excitation and emission wavelengths of 380 nm and 460 nm (top read), respectively, in kinetic mode for 45 min. Tissue homogenate (10 μg) was incubated in 200 μl of assay buffer containing 20 μM AEP substrate and assayed as above. The activity of AEP was expressed as the reading at 45 min minus the first reading. The levels of soluble Aβ_1–40 and Aβ_1–42 in the whole brain homogenate were determined using the ELISA kit (BlueGene Biotech, Shanghai, China) according to the manufacturer’s recommendations.

Mice were deeply anesthetized with 4% isoflurane (RWD Life Science, Shenzhen, China) and then decapitated. Brain tissues were collected and lysed in RIPA buffer (Beyotime, Nanjing, China) supplemented with protease inhibitor (Beyotime). The total protein concentration was determined using the BCA Protein Assay Kit (Sangon Biotech, Shanghai, China). Equal amounts of protein extracts (30 μg) were separated by SDS-PAGE and electrophoretically transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA). Then, the membranes were incubated with anti-AEP antibodies (R&D Systems Inc., Minneapolis, MN) at 4 °C overnight, followed by incubation with IRDye 680LT fluorescent secondary antibody (LI-COR Biosciences, Lincoln, NE). Proteins were visualized using the Odyssey Fc Imaging System (LI-COR Biosciences). Mouse β-actin was used as the loading control.

After treatment, the SAMP8 and SAMR1 mice were trained in a round water pool (diameter 150 cm) with extra-maze cues. Each animal received four training trials per day for 5 consecutive days, to learn to find the hidden platform located 1.5 cm below the water surface. In each trial, the mice were given 60 s to find the platform in one of four different positions. The animals were allowed to stay at the platform for 10 s if they found the platform within the given time. However, if the animals failed to find the platform within the given time, they were manually guided to the platform and left there for 10 s. The escape latency was recorded for up to 60 s. After each trial, the mice were dried and kept in a warm cage. The inter-trial interval for each mouse was 10 min. For the probe test, the platform was removed and the mice were allowed to swim for 60 s. The probe test was conducted 24 h after the last training trial. Data were analyzed by an investigator who was blinded to the treatment condition.

Mice were deeply anesthetized and cardinally perfused with 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS). The entire brain was removed and cut into coronal sections at 10-μm thickness using a cryostat microtome (Leica CM1950, Germany) and stored at − 80 °C until staining. Immunofluorescence staining of phospho-tau, Iba1 and MAP-2 was performed as previously described [22]. Permeabilization of brain sections was performed in PBS with 0.1% Triton X-100 for 15 min at room temperature. After being blocked, the sections were incubated with primary antibody overnight at 4 °C. Next day, after three washes with PBS, the sections were incubated with the Alexa Fluor 488-conjugated donkey anti-rabbit or Alexa Fluor 594 anti-mouse IgG secondary antibody (Invitrogen Life Technologies, Carlsbad, CA). Photographs were taken and analyzed by using a Leica SP8 confocal microscope (Leica Microsystems, Germany). Quantification was carried out in six slices of each brain spaced 120 μm apart to estimate the average number of immune-labeled cells per unit area and the average intensity of the immunostaining. Quantification and analysis were conducted by an experimenter who was blinded to the treatment condition.

Data are expressed as mean ± SEM and were analyzed with the Prism 7 software (La Jolla, CA). The concentration of the inhibitor yielding half-maximal inhibition (IC₅₀) of enzymatic activity was calculated using the following equation: Fractional Enzymatic Activity = Bottom + (Top-Bottom)/(1 + 10^((LogIC₅₀-C)*n)), where C is the logarithm of inhibitor concentration and n is the Hill coefficient. The statistical difference between two independent groups was analyzed with the unpaired Student’s t-test, and that among more than two groups was assessed with the parametric one-way ANOVA followed by a Tukey’s post-hoc test. Post-hoc tests were conducted when the F value achieved the necessary level (P < 0.05) and there was no significant variance inhomogeneity. For the Morris water maze test, a two-way ANOVA repeated measures was used to compare the acquisition data. Differences were considered to be significant when P < 0.05.

We first assessed whether the mature/cleaved form of AEP differs between pathological and healthy conditions. It has been reported that both the expression and enzymatic activities of mature AEP in the brains of AD patients are higher than those in healthy individuals [8, 10, 12, 23]. Here we found that the active AEP fragments in the cortex and hippocampus of 4- and 6-month-old SAMP8 mice were elevated compared to the age-matched SAMR1 mice (Fig. 1a and c). The enzymatic activity of mature AEP in 4- and 6-month SAMP8 mice was significantly higher than that in age-matched SAMR1 mice (Fig. 1b and d).

The chemical structure of δ-secretase inhibitor 11 is illustrated in Figure S1. This δ-secretase inhibitor 11 has been reported to selectively inhibit AEP activity with an IC₅₀ value of 0.70 ± 0.18 μM [14]; this effect is 46- to > 282-fold more potent than its inhibition over other cysteine proteases, including caspase-3, caspase-8, cathepsin-S and cathepsin-L [14]. Here, we tested again the inhibitory effect of δ-secretase inhibitor 11 on AEP activity in vitro and the dose-response relationship revealed that the δ-secretase inhibitor 11 had an IC₅₀ value of 0.15 ± 0.09 μM (Fig. 2a and b).

To examine the chronic effect of δ-secretase inhibitor 11 on AEP activity in the brain, we treated 4-month-old SAMP8 and SAMR1 mice with δ-secretase inhibitor 11 (10 mg/kg) or vehicle once daily via oral gavage for 3 months. The body weight was recorded over the treatment period and did not show any difference among the four groups (Fig. S1). As the mammalian AEP is present in the brain, liver, kidney and many other tissues, AEP activity was measured in these tissues after the last treatment. We again found that AEP activity in the SAMP8 mouse brain was higher than that in the SAMR1 mouse brain. Oral administration of δ-secretase inhibitor 11 significantly suppressed brain AEP activity compared to the vehicle treatment (Fig. 2c and d). The concentrations of Aβ_1–40 and Aβ_1–42 in brain lysates of SAMP8 mice were significantly reduced due to AEP inhibition by δ-secretase inhibitor 11 (Fig. 2e and f). Interestingly, AEP activity (RFU ≈ 50,000) in kidney and liver tissues of these mice was nearly 20 times higher than that (RFU ≈ 2,500) in the brain tissue and was not altered by the 3-month treatment (Fig. S1).

Given that the systemic administration of δ-secretase inhibitor 11 efficiently repressed brain AEP activity, we examined whether this reagent had therapeutic efficacy in this SAD model. We treated 4-month-old SAMP8 and SAMR1 mice with δ-secretase inhibitor 11 (10 mg/kg, p.o.) once daily via oral gavage for 3 months, and then conducted Morris water maze test to assess cognitive function related with spatial learning and memory [24]. Results showed that the swimming speed did not differ among animals treated with either vehicle or δ-secretase inhibitor 11 (Fig. 3a). The escape latency of each group decreased progressively over the 5-day training period (Fig. 3b), suggesting the development of spatial learning and memory. The SAMP8 mice learned slower than the SAMR1 mice (P < 0.05), while the SAMP8 mice treated with δ-secretase inhibitor 11 learned significantly faster than the vehicle-treated SAMP8 mice (P < 0.05). In the probe test on day 6, the SAMP8 mice treated with δ-secretase inhibitor 11 spent significantly longer time and travelled a longer distance in the target quadrant than the vehicle-treated SAMP8 mice (Fig. 3c, d). These data indicated that the 3-month treatment with δ-secretase inhibitor 11 ameliorated the cognitive impairment of SAMP8 mice.

We next investigated the cellular mechanisms involved in the therapeutic efficacy of δ-secretase inhibitor 11 in SAMP8 mice. Previous studies have revealed that the activated AEP induces hyperphosphorylation of tau by cleaving tau protein at N255 and N368, and the tau N368 fragments have been detected in human AD brains [12, 13]. Western blot analysis showed that the protein levels of AEP and tau N368 were significantly suppressed after 3-month treatment with δ-secretase inhibitor 11 (Fig. 4a, b). Immunofluorescence staining showed that the phospho-tau was significantly attenuated in the cortex and hippocampus of SAMP8 mice with δ-secretase inhibitor 11 treatment (Fig. 4c-e).

Tau hyperphosphorylation has been reported to drive microglial activation in the cortex of AD mice [25]. Therefore, we assessed microglial activation in SAMP8 mouse brain using Iba1 staining. We found that the Iba1 signal was markedly decreased in the cortex and hippocampus of SAMP8 mice treated with δ-secretase inhibitor 11 (Fig. 5a-c). Whole-brain lysate analysis showed that the level of Iba1 was significantly reduced by δ-secretase inhibitor 11 (Fig. 5d). To detect pathological changes of neuronal structure in SAMP8 mouse brain, we performed immunofluorescence staining of neuronal marker MAP-2, which mainly localizes in the soma and dendrites of most neurons. MAP-2 immunoreactivity was expressed in a fragment pattern in the cortex and hippocampus of vehicle-treated mice, and the MAP-2 density was significantly lower than that in SAMP8 mice treated with δ-secretase inhibitor 11 (Fig. 6a-c). Whole-brain lysate analysis showed that both synapse-associated protein SYP and PSD-95 were significantly up-regulated by drug treatment (Fig. 6d, e). These data indicated that the dendritic disruption and synaptic loss in SAMP8 mouse brain were substantially alleviated by treatment with δ-secretase inhibitor 11.