Consecutive Analysis of BACE1 Function on Developing and Developed Neuronal Cells

Hippocampal slice culture

The experiments in this study were approved by the committees on animal experiments of Juntendo University. Animals were purchased from Nihon SLC (Hamamatsu, Japan) and treated according to our institutional guidelines. Rat hippocampal slice cultures were prepared following conventional methodology [26, 28]. Briefly, whole brains were excised from anaesthetized Sprague-Dawley rats at postnatal day 7–8 and the hippocampi were isolated. Slices of 400-μm thickness were obtained from the central region of the hippocampi, using a tissue chopper (McIllwain). The slices were placed on a polytetrafluoroethylene membrane filter (Millicell-CM; Millipore, Carrigtwohill, Ireland), and culture medium (see below for composition) was added up to the bottom surface of the filter. These prepared cultures were maintained at 37°C with 5% CO₂-enriched humidified atmosphere (Fig. 1B, C). The slice culture medium (SCM) contained 50% minimal essential medium based on Hanks’ salts (Nacalai, Kyoto, Japan), 25% Hank’s balanced salt solution (Nacalai), and 25% heat-inactivated horse serum (Life Technologies, Gaithersburg, MD). The culture medium was replaced twice a week with fresh medium during the entire culture period.

Drug application

Test drugs were dissolved into water to concentrations 1000 times higher than the final levels, kept at 4 or –20°C until use, and diluted into medium unless otherwise stated. The α-secretase inhibitor TAPI-1(tumor necrosis factor-α processing inhibitor-1) and BACE inhibitor (β-secretase inhibitor IV) (Millipore) were dissolved into DMSO (Sigma-Aldrich, MI, USA) to concentrations of 500 and 1000 times higher than the final levels, respectively, and kept at –20°C until use. TAPI-1 and BACE1 were added to fresh SCM before analysis.

Immunohistochemical staining

For the immunohistochemical staining of presynaptic structures, the cultured slices were fixed with 4% paraformaldehyde in phosphate-buffer for 1 h at 4°C. The fixed preparations were rinsed with phosphate-buffered saline (PBS) and then treated with PBS containing 0.1% Triton X-100 and 5% fetal bovine or horse serum at 24°C for 30 min. The treated slices were incubated with a primary antibody against NeuN (a marker protein for neuronal nuclei; mouse monoclonal, 1:200 dilution, 24 h, 4°C; Chemicon, Temecula, CA, USA), synaptophysin (a marker protein for presynapses; mouse monoclonal, 1:200 dilution, Millipore), PSD95 (a marker protein for postsynapses; monoclonal, 1:200 dilution, Sigma-Aldrich), MAP2 (Microtubule-associated protein 2; monoclonal 1:200 dilution, Abcam), Neurofilament 200 (a marker protein for axon; monoclonal 1:200 dilution, Sigma-Aldrich), and BACE1 (rabbit monoclonal, D10E5, 1:100 dilution, 24 h, 4°C; Cell signaling Technology, Danvers, MA, USA). The primary antibody-treated slices were incubated with a secondary antibody conjugated with Alexa Fluor 488 or 594 at 24°C temperature for 2–2.5 h. The slices were examined using a Leica SP5/TCS confocal laser scanning microscope (Leica Microsystems, Wetzlar, Germany) or an Olympus IX71 (Olympus, Tokyo, Japan) with Orca-ER cooled CCD camera (Hamamatsu Photonics, Hamamatsu, Japan).

Electrophoresis and immunoblotting

Acute or cultured hippocampal slices were solubilized in ice-cold RIPA buffer [1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 25 mM Tris-HCl, pH 7.5, 137 mM NaCl, and 3 mM KCl with a protease inhibitor cocktail (Complete, Roche Diagnostics, Mannheim, Germany)]. Samples were resolved by SDS-PAGE and transferred onto polyvinylidene fluoride membranes (Immobilon-P, Millipore). Membranes were incubated in 5% skim milk in TBST (1× Tris-buffered saline with 0.1% Tween 20) at 25°C for 30–60 min to block nonspecific binding. Membranes were incubated at 4°C overnight or 25°C for 2 h with primary antibodies: anti-synaptophysin (guinea pig, 1:2000 dilution; Frontier Institute, Hokkaido, Japan, or mouse monoclonal, 1:200 dilution, Millipore), anti-PSD95 (mouse monoclonal, 1:3000 dilution, Sigma-Aldrich), anti-BACE1 (rabbit monoclonal, D10E5, 1:2000 dilution; Cell signaling Technology), anti-AβPP (mouse monoclonal, 22C11, 1:1000 dilution; Millipore, or rabbit, AβPP-C15 1:1000 dilution [7]), anti-phospho AβPP (Thr668) (rabbit, 1:1000 dilution, Cell Signaling), anti-βIII tubulin (mouse monoclonal, TuJ-1, 1:5000 dilution; R&D systems Minneapolis, MN, USA), or anti-β-actin (a loading control, mouse monoclonal, 1:5000 dilution; Wako Chemicals, Osaka, Japan). AβPP-C15 antibody was raised against C-terminal 12-aa peptide of AβPP, and adsorbed against corresponding C-terminal peptides of APLP1 and APLP2. After four changes of 1× TBST and three 5-min washes at 25°C, membranes were incubated in horseradish peroxidase-conjugated secondary antibodies. After four washes at 25°C, membranes were incubated with ECL solution (Thermo Fisher Scientific, MA, USA). For quantification, chemiluminescence light signals in Super Signal Dura substrate (Thermo Fisher Scientific) were captured by a cooled charge-coupled device camera system (LAS-3000plus; Fuji Film, Japan) that ensured wide ranges of linearity. Densitometric quantification of synaptic protein expression in culture slices were normalized to β-actin or βIII tubulin. The protein content was estimated using bicinchoninic acid (BCA) reagent (Thermo Fisher Scientific).

Amyloid-β assay and sAβPPβ assay

The levels of Aβ₄₀ and Aβ₄₂ were measured using a solid phase by sandwich ELISA kit following the supplier’s information (IBL, Gunma, Japan). The test samples and standards were added to the 96-well plate (precoated with anti-Aβ (35–40) (1A10) mouse IgG for Aβ₄₀, precoated with anti-Aβ (38–42) rabbit IgG for Aβ₄₂) and incubated overnight at 4°C. After several washes with EIA wash buffer, 100 μL of HRP-labeled anti-mouse/rat Aβ (1–16) rabbit IgG solution was added to each well (except for the wells corresponding to reagent blanks) and incubated at 4°C for 1 h. After a thorough wash, 100 μL of tetramethylbenzidine solution was added to each well and incubated in the dark at 24°C for 30 min. The reaction was stopped by adding 100 μL of 1 N sulfuric acid, and the readings were taken at 450 nm. The levels of Aβ in the test samples were calculated by incorporating the unknown values into the standard curve obtained in the assay. The levels of soluble AβPP β (sAβPPβ) were measured using the IBL assay kit (IBL, 96 well plate precoated with anti-sAβPPβ rabbit IgG, HRP-labeled anti-AβPP18 mouse IgG) as described above.

Propidium iodide staining

Propidium iodide (PI; Nacalai) is a red-fluorescent nuclear and chromosome counterstain. Since livecells are not permeable to PI, it is commonly used to detect dead cells in a population. We used PI fluorescence intensity as an index of total neuronal degeneration. Cultured slices were exposed to SCM containing 10 μg/mL PI for 12 h. The whole slice was photographed with identical incident light intensity and exposure time on the fluorescence microscope (IX70 with a 4x objective lens, Olympus). PI fluorescence intensity was measured on all micrographs using MetaMorph software (Molecular Devices, CA, USA).

Electrophysiology

Extracellular recording of field excitatory postsynaptic potential (fEPSP) was performed in a conventional manner [26]. Briefly, a glass microcapillary electrode for recording (3 M NaCl or artificial cerebrospinal fluid (aCSF); see below, 3–16 MΩ; input resistance) was placed in the CA1 stratum pyramidale. For evoking synaptic responses, a bipolar tungsten stimulating electrode was placed in the CA3 stratum pyramidale, and a pulse of 100 μs was delivered once in 30 s. An Axopatch 1D amplifier (Axon Instruments) was used, and the signal was filtered at 1 kHz, digitized at 10 kHz, and stored on a personal computer (pClamp 10, Molecular Devices). Input-output curves were collected by varying the stimulation strength applied in the range 100–400 μA. D-2-amino-5-phosphonovaleric acid (D-APV; 25 μM) was applied to prevent NMDA receptor-mediated synaptic responses. The composition of the aCSF was (mM) 126 NaCl, 5 KCl, 1.25 NaH₂PO₄, 2.5 CaCl₂, 2 MgSO₄, 22 NaHCO₃, and 10 glucose. This aCSF was equilibrated with 95% O₂ and 5% CO₂ by constant bubbling. Experiments were performed at 32±1°C.

Statistical analysis

All quantitative data were presented as mean±standard deviation (SD) or standard error of the mean (SEM). Statistical analysis was performed using either Student’s t-test (two-group comparison) or ANOVA (more than two groups) followed by post hoc comparison. The level of significance was indicated by asterisks: ^*p < 0.05, ^**p < 0.01, ^***p < 0.001

Preparation of rat hippocampal slice culture

First, to evaluate the BACE1 activity and its role in neural tissue consecutively, we assessed organotypic hippocampal slice cultures by immunohistochemistry. We have previously reported that cultured rat hippocampal slices retained neuronal activity for several weeks [25, 29]. Immunofluorescent staining with antibodies against NeuN, a neuronal nuclear protein used as a neuron-specific marker [30], showed that gross patterns of neuron arrangement were preserved for 40 days in vitro (DIV) (Fig. 1A). Next, we stained cultured slices with antibodies against MAP2, neurofilament 200, synaptophysin, a synaptic vesicle protein and thus a marker of presynapses [31], and postsynaptic density protein 95 (PSD95), a postsynaptic scaffold protein and thus a marker of postsynapses [32]. The punctate structures immunopositive for these synapse markers were observed after 40 DIV (Fig. 1B, C). Furthermore, we examined the synaptic activity by using extracellular recording of fEPSP. Synaptic transmission was recorded at CA3-CA1 synapses in hippocampal slice cultures after 39 DIV (the results are shown in Fig. 8). These data indicate that synapses remained in cultured hippocampal slices for over 6weeks.

Next, we examined the spatial expression pattern of BACE1 in acute (0 DIV) and cultured (3–40 DIV) hippocampal slices by immunofluorescent staining with antibodies against BACE1. BACE1 expression in acute and cultured hippocampal slices accumulated at the stratum radiatum (Fig. 1D), similar to in vivo distribution [33]. Antibody specificity was confirmed by previous reports [34, 35]. These results suggested that cultured hippocampal slices conserved BACE1 expression, neural function, and organization forseveral weeks.

Chronological changes of BACE1 and synaptic protein expression

In organotypic cultures, conditions are changed every hour, just after plating on filter membrane cups or culture dishes. Previous reports showed that reorganization and maturation of synapses occurred in cultured hippocampal slices [23, 24]. To elucidate the time-dependent changes in synapse and AβPP processing, we evaluated the expression level of synaptic proteins, BACE1, and AβPP by immunoblotting assays for several weeks (Supplementary Figure 1). Immunoblotting signals prepared from cultured slices indicated corresponding molecular weights (approximately BACE1: 70 kDa; AβPP: 100 kDa; synaptophysin: 38 kDa; PSD95: 95 kDa; β-actin: 43 kDa, Fig. 2A-E).

Similarly to immunohistological analysis (Fig. 1B, C), immunoblotting analyses revealed that synapse marker proteins synaptophysin, PSD95, and GluA2/3, the alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor, decreased from 1 to 4 DIV and increased after 7 DIV. In addition, our data indicated that expression levels of synaptophysin and PSD95 were stable after 20 DIV (Fig. 2H, I, and Supplementary Figure 1). In contrast, quantitative immunoblotting analyses suggested that expression levels of BACE1 and AβPP were not changed significantly after 7 DIV (Fig. 2F, G).Immature, mature and phosphorylated AβPP were detected in acute and cultured hippocampal slices (Supplementary Figure 2). Consistent with the previous studies [23–26, 28], these results suggested that 3–20 DIV were a critical period of synapse formation and maturation in hippocampal slice cultures. Furthermore, our data indicated that synapse conditions and expression of BACE1 and AβPP were stable after 20 DIV. In this study, we define the 3–18 DIV as the synaptic developing stage, and the 26 day after in vitro as the synaptic developedstage.

Chronological analysis of production of sAβPPβ and Aβ

For the continuous monitoring of the BACE1 activity, we measured BACE1 cleavage products sAβPPβ and Aβ₄₀ in SCM. BACE1 cleaves AβPP at extracellular sites and generates soluble extracellular fragments: sAβPPβ and a membrane-bound carboxyl terminal fragment, AβPP-CTFβ. The AβPP-CTFβ is subsequently cleaved by γ-secretase to produce Aβ, released to extracellular SCM. To estimate BACE1 activity, we measured sAβPPβ and Aβ in SCM from cultured hippocampal slices with a two-site ELISA assay (see Materials and Methods). To quantify sAβPPβ and Aβ₄₀ secretion levels, we collected SCM cultured hippocampal slices for 24 h after washing three times with 1 mL fresh SCM. Secretion of sAβPPβ was between 3 to 6 ng/mL (Fig. 3A), and secretion of Aβ₄₀ was between 200 to 300 pg/mL during the monitoring period (Fig. 3B). Furthermore, secretion of Aβ₄₂ was between 23 and 50 pg/mL for 24 h and showed a chronological change similar to Aβ₄₀ in cultured hippocampal slices (data not shown). Production of sAβPPβ and Aβ from cultured rat hippocampal slices and ratio of Aβ to sAβPPβ were stable after 7 DIV (Fig. 3C).

Pharmacological manipulation of BACE1

AβPP is alternatively cleaved by the members of the α-secretase family or BACE. The α-secretase is considered to be a part of the non-amyloidogenic pathway in AβPP processing, as it precludes Aβ formation [36, 37]. To check enzyme dependencies on AβPP metabolism in our system, we investigated the effect of each secretase inhibitor on hippocampal slice cultures. We cultured hippocampal slices for 24 h with fresh SCM containing vehicle (DMSO) or inhibitors after washing three times with 1 mL fresh SCM (Fig. 4). In accordance with previous reports [38], application of BACE inhibitor reduced secretion of Aβ₄₀ (Fig. 4B, D). AβPP is alternatively cleaved within the Aβ sequence by α-secretase, and application of the α-secretase inhibitor TAPI-1 increased secretion of Aβ₄₀ (Fig. 4C, D). Because BACE1 enzymatic activities were observed throughout the culture period, we investigated the effect of BACE inhibitor during the developing or developed stage of the synapse (Fig. 5). During both stages, application of BACE inhibitor IV reduced secretion of Aβ₄₀ (Fig. 5A, B) and sAβPP (Fig. 5C, D).

To evaluate the cytotoxicity of BACE inhibitor treatment, we used PI staining assays. Since live cells are not permeable to PI, PI fluorescence intensity can be used as an index for total neuronal degeneration. In contrast to kainic acid-induced cell death, long-term treatment with BACE inhibitor failed to affect survival of cultured hippocampal neurons (Fig. 6 and Supplementary Figure 3). These results indicated that alternative AβPP processing occurred and we could control AβPP cleavage in cultured hippocampal slices without toxic effects.

BACE1 activity is known to be important for synapse formation and maturation in the hippocampus [33]. Thus, we investigated the effect of continuous BACE1 inhibition on synapses in cultured hippocampal slices. When BACE inhibitor was applied during the period of synaptogenesis and maturation (3.5–17.5 DIV, cf. Fig. 2), PSD95 and GluA2/3 decreased (Fig. 7A-F). Conversely, application of BACE inhibitor after the period of synaptogenesis and maturation (26–42 DIV) did not affect pre- and post-synapse (Fig. 7G-L). In both periods, the addition of BACE inhibitor IV to SCM suppressed β cleavage of AβPP and accumulated mature AβPP (Fig. 7E, K). Furthermore, immunofluorescent staining analysis revealed that PSD95, but not synaptophysin, decreased in each punctate (Supplementary Figure 4, PSD95; 66.9±31.9%, synaptophysin; 114.4±15.9%). In concordance with change of protein level of GluA2/3, electrophysiological analysis indicated that application of BACE inhibitor during the developing stage decreased synaptic strength in the CA3-CA1 pathway (Fig. 8). These data suggested the importance of BACE1 activity in synaptogenesis and maturation; however, it has less function in matured synapses.