Background
Alzheimer´s disease (AD) is a proteinopathy characterized by the accumulation of aggregated extracellular amyloid-beta (Abeta, Aβ) peptides and intracellular hyperphosphorylated tau (revised in [
1]). Concomitant with appearance of extracellular Abeta deposits, another central pathological feature of the disease is the early formation of amyloid plaque-associated neuritic changes in the form of dystrophic neurites, together with a selective loss of connections and neuronal groups. Dystrophic neurites, defined as thickened or irregular neuronal processes, are considered to be expression of a widespread alteration of the neuronal cytoskeleton. In AD, dystrophic axons are particularly abundant in the hippocampal fiber systems originating from the subiculum, CA1, and the entorhinal cortex [
2]. However the exact molecular mechanisms underlying the pathogenesis of AD remain to be elucidated.
Dystrophic neurites were characterized by the presence of numerous vesicles of multiple origins [
3,
4]. Several lines of investigation support the notion that defective autophagy process, a cellular catabolic mechanism essential for degradation of aggregated proteins and organelles, significantly contributes to AD pathogenesis [
5‐
8]. Interestingly, autophagic compartments have been reported to participate in APP processing and Aβ peptides production [
9].
Abeta peptides, cytotoxic in their oligomeric state [
10‐
13] derive from the sequential cleavage of APP by beta- and gamma-secretases [
14,
15]. Although the exact intracellular localization of APP processing is unknown, the autophagic and endocytic pathways could be both involved in precursor protein (APP) processing and Abeta generation. In this sense, BACE-1 and gamma-secretase complex have been detected in many cellular locations, including early and late endosomes [
16], autophagic vacuoles [
17‐
19] and lysosomes [
20]. On the other hand, the Abeta degradation, in vivo, could be mediated by several proteases, as neprilysin, IDE, and several cathepsins as B, D and E [
21]. Abnormal processing of APP or Abeta accumulation in AD could be related to several mechanisms, including excessive production, abnormalities in transport, alteration of autophagic and endosomal pathways, and deficits in its degradation through the lysosome or the ubiquitin-proteasome system (UPS) [
22‐
24]. In fact, accumulation of autophagic vacuoles (AVs) has been observed in brains from AD patients [
3,
19] and in PS1/APP mice after Abeta deposition [
17,
19,
25]. Moreover, the AVs were principally accumulated within dystrophies and could reflect impairment in AVs clearance in AD brains [
5]. In this sense it has been reported that enhancing lysosomal cathepsin activity ameliorates Abeta toxicity [
26] and restoring the autophagy-lysosomal pathway (by deletion of cystatin B) reduced amyloid load and rescued memory performance [
9].
In the present work we investigated how the possible age-related relationship between aberrant Abeta generation and dysfunctions in axonal cytoskeleton as well as in lysosomal and proteasomal systems, manifested in our PS1/APP AD model. We proposed that the decrease in lysosomal proteolytic activities was implicated in increased Abeta production. Abnormal accumulation of Abeta could aggravate the axonal and cytoskeleton abnormalities linked to the pathology of AD.
Discussion
In this work, we studied the progression of hippocampal neuronal pathology, in the PS1M146L/APP751SL model, that connects cytoskeleton and protein degradation dysfunction with dystrophy formation and synaptic Abeta production. As we reported previously, this model displayed the formation of axonal dystrophies, associated to extracellular Abeta deposition since early ages [
17]. These dystrophies accumulated numerous vesicles of, among different origins, autophagic/lysosomal nature. Also, the axonal pathology was associated with the presence of aberrant presynaptic terminals in close proximity to Abeta plaques. In agreement with others [
36‐
38] we observed that extracellular Abeta, probably by increasing the local calcium concentration, might produce cytoskeletal abnormalities that could induce transport defects (see also [
39‐
43]). In fact, data presented in this work demonstrated the existence of a progressive increase in neurofilaments and tau hyperphosphorylation (SMI31, AT8 and AT100 epitopes). This increase was prevalently observed in dystrophies, surrounding the Abeta plaques. In conjunction with this apparent cytoskeletal dysfunction, we also observed a decrease in the levels of kinesin and dynein motor proteins. Taken together, these data strongly suggest the existence of a progressive, age-dependent, disorganization of the axonal cytoskeleton, which could impair the normal axonal transport, at local points in contact with extracellular Abeta deposits, (see also [
25,
43]). Furthermore, our data also demonstrated the accumulation of LC3-II and ubiquitinated proteins, principally at dystrophies. Whereas these observations could be interpreted as a simple accumulation of autophagosomes/lysosomes due to transport defects, our data also demonstrated the existence of a marked inhibition of the intracellular proteolytic activities. In fact, we observed a profound decrease in the cathepsin B and D activities and, in a minor extent, the proteasomal chymotrypsin activity. In parallel, we also observed a reduction in the mature forms of both cathepsins B and D in PS1/APP samples. The reasons that determined this marked and early proteolytic inhibition are unknown. It has been proposed that FAD mutations in PS1 impaired the maturation of V0a1 subunit of the vATP-ase [
6]. However, as mentioned (see results), in our model, the V0a1 subunit displayed no apparent defects on its maturation. Furthermore, 6-month-old PS1M146L transgenic mice displayed no variations on cathepsin B activity. Thus, the observed decrease in cathepsin B and D activities seemed to be independent of the proton pump maturation (see also [
44]). Nevertheless, our data were indeed compatible with defects on lysosomal acidification and/or maturation. In this sense, it has been also proposed that axonal transport was crucial for lysosomal maturation and function [
33]. Thus, in our AD model, the cytoskeletal disorganization (probably mediated by extracellular Abeta and calcium misbalance) could also impair the lysosomal maturation (reflected by a decrease in the mature forms of both cathepsins tested) and, in consequence, the proteolytic function (decrease in the cathepsin activities). Further experiments should be done to ascertain this point.
We also analyzed the consequences of this neuritic pathology on APP processing and Abeta generation. In this sense, the overexpression of cathepsin B [
45], the enhancement of cathepsin activities [
9] or the positive lysosomal modulation [
26] reduced the Abeta accumulation, the synaptic deficiencies and restored the cognitive function in transgenic mice. Thus, the progressive decrease of cathepsin activities and, in a minor extent, proteasomal activity, observed in our model, could be paralleled by an increase in the amyloidogenic APP processing. In fact, the observed age-dependent accumulation of hAPPfl and C99, together with the increase in BACE-1 and PS1-ctf proteins and their enzymatic activities, confirmed this proposition. Furthermore, our in vitro experiments also corroborated the rapid and progressive accumulation of APP derived C-terminal fragments, intracellular Abeta and both BACE-1 and PS1-ctf proteins after lysosomal and/or proteasomal inhibition (see also [
46,
47]). Therefore, based on these data, it is tempting to speculate that the formation of axonal dystrophies, probably due to the presence of extracellular Abeta [
48], might produce a decrease in the intracellular proteolysis. This was paralleled by the accumulation of APP derived fragments inducing, in consequence, the production of higher amount of Abeta peptides (see also [
49]). This self-progressing pathological scenario could be implicated in the progressive increase of Abeta and BACE-1 observed in AD patients (data not shown; see also [
50]) and in the expansion of the pathology (see below).
Of note, our data also demonstrated the existence of a preferential accumulation of Abeta peptides in isolated synaptosomal fractions. Whereas this preferential accumulation could reflect the high gamma-secretase activity at the presynaptic terminals (data not shown but see [
51]), these data also indicated that Abeta peptides were produced in compartment(s) different to, or in addition to, the autophagic vesicles [
52]. This proposition was based on the observation that LC3-II, an autophagic marker, was preferentially concentrated in the microsomal fractions and it was scarce in synaptosomes. Although autophagic vesicles were indeed concentrated in pathological presynaptic terminals [
17], the clear disproportion between LC3-II and Abeta in microsomes and synaptosomes, at all ages tested, strongly indicated a different compartmentalization. In this sense, it has been recently proposed that Abeta was synaptically produced and secreted from endosomal compartment [
53]. Independently of the intracellular compartment, the synaptic production of Abeta peptides, observed in this work, could be implicated in the synaptic dysfunction in AD patients and in the pathological progression between synaptically connected areas. In this sense, it is widely accepted that the severity of the disease correlated better with the synaptic dysfunction rather than the plaque load. Thus, the age-dependent accumulation and, probably, release of Abeta peptides by presynapses could be directly implicated in the dendritic spine alterations observed in AD patients and in AD models. Furthermore, recent publications have been highlight the progression of Abeta [
54,
55] and tau [
56,
57] pathology between synaptically connected regions. In this sense, we have also reported the existence of a preferential Abeta deposition, microglial activation and neurodegeneration of principal neurons in layers V-VI of the entorhinal cortex [
58]. However, these particular pyramidal cells did not express the transgenic hAPP and, in consequence, did not produce Abeta peptides. Thus, the presynaptic accumulation of Abeta peptides, reported here, might be implicated in the Abeta deposition and pathological spreading from Abeta producing region/layer into other brain regions.
In sum, our data demonstrated the existence of a progressive, age-dependent, cytoskeletal pathology (probably due to the extracellular Abeta deposition) that could be implicated in a reduction of the intracellular proteolytical processes. This impairment was associated to a progressive accumulation of APP derived fragments (and Abeta peptides) according with the increase of BACE-1 and gamma-secretase activities. This retard in the APP metabolism seemed to be directly implicated in the synaptic Abeta accumulation and, in consequence, in the pathology progression between synaptically connected regions.
Methods
Transgenic mice
Generation and initial characterization of PS1M146L/APP751sl (PS1/APP) tg mice has been reported previously [
10,
58,
59]. Heterozygous PS1/APP double tg mice (C57BL:6 background) were generated by crossing homozygous PS1 tg mice with heterozygous Thy1-APP751SL mice. Only male mice were used in this work. Age-matched non-transgenic male mice (WT) of the same genetic background (C57BL:6) were used as controls.
Mice were first anesthetized with sodium pentobarbital (60 mg/kg), the hippocampi were dissected and immediately frozen and stored at −80°C until use. For immunohistochemistry, anesthetized mice were perfused transcardially with a paraformaldehyde-based solution (see details below). All animal experiments were performed in accordance with the guidelines of the Committee of Animal Research of the University of Seville (Spain) and the European Union Regulations.
RNA and total protein extraction
Total RNA from mice hippocampi or cultured cells was extracted using Tripure Isolation Reagent (Roche) as described previously [
10,
11,
27,
58,
59]. After isolation, RNA integrity was assessed by agarose gel electrophoresis. The yield of total RNA was determined by measuring the absorbance (260:280 nm) of isopropanol-precipitated aliquots of the samples. The recovery of RNA was comparable in all studied groups (1.2-1.5 μg/mg of tissue).
The protein pellets, obtained using the Tripure Isolation Reagent and isopropanol-mediated precipitation, were resuspended in 4% SDS and 8M urea in 40 mM Tris–HCl, pH 7.4 and rotated overnight at room temperature to get complete protein solubilization.
Reverse transcription and real-time RT-PCR
Retrotranscription (RT) was performed using random hexamers, 4 μg of total RNA as template and High-Capacity cDNA Archive Kit (Applied Biosystems) following the manufacturer recommendations [
10,
27]. For real time RT-PCR, commercial Taqman™ probes (Applied Biosystems) were used for amplification. Alternatively, SYBRgreen dye and designed specific primers were used for amplification of human APP751 (forward: 5´-GGATATGAAGTTCATCATCA-3´; reverse: 5´-TCACTGTCGCTATGACAACA-3´), and human PS1 (forward: 5´-TGGCTCATCTTGGCTGTG-3´; reverse: 5´-ACCAGCATACGAAGTGG-3´). PCR reactions were carried out using either ABI Prism 7000 or 7900HT sequence detector systems (Applied Biosystems). A standard curve was first constructed for every assay, using increasing amounts of cDNA. In all cases, the slope of the curves indicated optimal PCR conditions (slope 3.2-3.4). The cDNA levels of the different mice were determined using GAPDH as housekeeper. Therefore, GAPDH amplification was done in parallel with the gene to be analyzed, and this dada used to normalize target gene results.
Independent of the analyzed gene, results were always expressed using the comparative Ct method, following the Bulletin number 2 from Applied Biosystems. As a control condition, we selected 6-month-old WT mice. In consequence, the expression of all tested genes, for all ages and mouse strains, was referenced to the expression levels observed in 6-month-old WT mice.
Western blot
Western blots were performed as described [
60]. Briefly, 5–20 μg of proteins from the different samples were loaded on 16%-SDS-tris-tricine-PAGE or 12%-SDS-tris-glycine-PAGE and transferred to nitrocellulose (Hybond-C Extra; Amersham).
After blocking, using 5% non-fat milk, membranes were incubated overnight, at 4°C, with the appropriate antibody: phospho-neurofilament (clone SMI-31; 1:1,000; Abcam), total-neurofilament (clone SMI-32; 1:1000; Abcam), phospho-Ser199:202-Thr205-PHF-tau (clone AT8; 1:1000; Pierce), phospho-Thr212-Ser214-PHF-tau (clone AT100; 1:1,000; Innogenetics), kinesin-1 heavy chain (1:1,000; Abcam), dynein-1 intermediate chain (1:1000; Millipore), ATP-synthetase-Beta (1:1000; BD Transduction Laboratories), Abeta peptide (clone 6E10; 1:5000; Signet), APP C-terminal (1:6000; Calbiochem), PS1 C-terminal (1:2000; Millipore), BACE-1 (1:1000; Abcam), V0a1-proton-pump subunit (1:1000; Synaptic Systems), LC3B (1:1000; Cell Signaling), cathepsin B (1:1000, Santa Cruz); cathepsin-D (1:5000; ABFrontier Co. Ltd), Lamp-1 (1:1000; Developmental Studies Hybridoma Bank; University of Iowa), ubiquitin (1:1000; Sigma-Aldrich) and Beta-actin (1:5000; Sigma-Aldrich). Membranes were then incubated with the corresponding horseradish-peroxidase-conjugated secondary antibody (Dako, Denmark) at a dilution of 1:8000. Each blot was developed using the ECL-plus detection method (Amersham).
For quantification, the scanned (Epson 3200) images were analyzed using PCBAS program. For normalization purposes, proteins were first estimated by Lowry and protein loading corrected by beta-actin. The intensity of bands from 6-month-old WT or PS1/APP (Figures
4A, E and F; Figures
5C and D) mice were averaged and considered as 1 relative unit. All other data were then normalized by the specific signal observed in 6-month-old WT or PS1/APP group or negative control for “in vitro” experiments.
Enzymatic activity determination
BACE-1, Cathepsin B and D activities were determined using commercial kits (R&D Systems, Germany; Calbiochem, Germany and Sigma-Aldrich, respectively) following the manufacturer instructions. Briefly, fresh hippocampal samples were homogenized in the buffer supplied by the manufacturer or in PBS (for proteasome activity), centrifuged at 10,000xg (15 min at 4°C) and the supernatant (100–200 μg of protein per assay) was used. BACE-1 and cathepsin B and D activities were determined using substrates provided by the manufacturer
Proteasome chymotrypsin-like activity was determined as described [
61]. Briefly, soluble fractions (30–50 μg of protein per assay) were diluted in 50 μM reaction substrate Suc-Leu-Leu-Val-Tyr-aminomethylcoumarin (AMC) (Sigma-aldrich), 0.1mM EDTA, 5mM DTT, 0.01% (w:v) CHAPS; 100mM NaCl; 1% (v:v) Glycerol, 50mM HEPES-KOH pH 7.5. Duplicated reactions were placed in a 96-well black polystyrene microplate (BD Transduction Labs.) and incubated at 37°C. Fluorescence was determined at excitation 360-380nm and emission 460-480 nm.
For each enzymatic assay, the fluorescence intensity was determined every 15 minutes (starting by the addition of the substrate) for a 1–2 hours final incubation time, using a Synergy HT Multi-mode microplate reader (Biotek). The activities were calculated from the maximal slope of the fluorescence intensity vs time curves and corrected by the amount of protein added. The results were then normalized by the activity observed in 6 months WT mice or 6 months PS1/APP mice (for BACE-1).
Also for each enzymatic activity, a reaction without substrate, a reaction without sample, and reaction with an inhibitor were used as negative controls. The inhibitors used were: Cathepsin B, Inhibitor Ref. 219385 (Callbiochem); Cathepsin D, pepstatin A Solution Ref. P3749 (Sigma-Aldrich); Proteasome 10 μM MG-132 (Sigma-Aldrich). Independently of the enzymatic activity assay, each experiment was repeated, at least, three times for each age and genotype.
Gamma-secretase activity was determined following previously described protocols with some modifications [
62]. Briefly, membrane pellets from PS1/APP animals (n=6 per age) were thawed and resuspended (at 3 mg of protein per ml) in 150 mM Citrate Buffer, pH 6.4, containing protease inhibitors (Roche). Aliquots (150 μg of proteins) were used for each assay. As negative control, 100 μM L-685-458 gamma-secretase inhibitor (Calbiochem) was added prior assay. Samples were then incubated, at 37°C with orbital shaking at 400 rpm, for 2 hours. After incubation, membranes were sonicated (at 80 W for 30 seconds) and centrifuged at 30,000xg (30 minutes, 4°C). Supernatants were used to determine AICD production by western blot using anti-APP C-terminal as primary antibody.
Tissue preparation for immunohistochemistry
After anesthesia with sodium pentobarbital (60 mg/kg), 6, 12, and 18-month-old control (WT) and PS1/APP tg male mice (n=4/age/genotype) were perfused transcardially with 0.1 M phosphate-buffered saline (PBS), pH 7.4 followed by 4% paraformaldehyde, 75 mM lysine, 10 mM sodium metaperiodate in 0.1 M phosphate buffer (PB), pH 7.4. Brains were then removed, post-fixed overnight in the same fixative at 4°C, cryoprotected in 30% sucrose, sectioned at 40 μm thickness in the coronal plane on a freezing microtome and serially collected in wells containing cold PBS and 0.02% sodium azide. All animal experiments were approved by the Committee of Animal Use for Research of the Malaga University (Spain) and the European Union Regulations.
Immunohistochemistry
Coronal free-floating brain sections (40 μm thick) from 6 and 12–18 month-old control (WT) and PS1/APP mice were processed simultaneously in the same solutions and conditions to prevent processing variables. Sections were first treated with 3% H2O2/3% methanol in PBS and with avidin-biotin Blocking Kit (Vector Labs, Burlingame, CA, USA), and then incubated overnight at room temperature with one of the following antibodies: anti-APP-C-terminal rabbit polyclonal (1:20,000; Sigma-Aldrich), anti-phospho-Ser199:202:Thr205-PHF-tau mouse monoclonal (clone AT8; 1:500; Pierce) anti-ubiquitin rabbit polyclonal (1:2,000, Dako), anti-LC3 goat polyclonal (1:1,000, Santa Cruz Biotechnology), anti-BACE-1 rabbit polyclonal (1:1,000, Abcam) or anti-Lamp2 rabbit polyclonal (1:500, Abcam).
The tissue-bound primary antibody was detected by incubating with the corresponding biotinylated secondary antibody (1:500 dilution, Vector Laboratories), and then followed by streptavidin-conjugated peroxidase (Sigma Aldrich) diluted 1:2000. The reaction was visualized with 0.05% 3-3’-diaminobenzidine tetrahydrochloride (DAB, Sigma Aldrich), 0.03% nickel ammonium sulphate and 0.01% hydrogen peroxide in PBS. When required, immunolabeled sections were then incubated for 3 minutes in a solution of 20% Congo red. Sections were then mounted on gelatin-coated slides, air dried, dehydrated in graded ethanols, cleared in xylene and coverslipped with DPX (BDH) mounting medium. Specificity of the immune reactions was controlled by omitting the primary antiserum.
Quantitative analysis of Abeta plaques-associated dystrophic neurites
The number of APP-immunopositive dystrophic neurites per plaque was quantified over Congo red stained Abeta deposits in sections from PS1/APP animals at young (4 and 6 month-old group) and old (12 and 18 months-old group) ages (n= 6/group; 5 sections per animal through the antero-posterior extent of the hippocampus). Quantification was done in CA1 subfield which was defined using a 10x objective and the number of dystrophic neurites was counted using a 100x objective in an Olympus BX61 microscope equiped with NewCAST software package (Olympus, Glostrup, Denmark). All plaques present in the CA1 region of each section were quantified. The number of dystrophic neurites per plaque was normalized to the mean plaque area to allow comparisons between groups.
Synaptosome and microsome fractions isolation
Synaptosomal fractions were obtained basically as described previously [
17,
63]) with some modifications. Briefly, one mouse hemicortex was gently homogenized with a glass Dounce homogenizer in cold Buffer A (0.32 M sucrose, 1 mM EDTA, 1mM EGTA in 20 mM Tris–HCl pH 7.5, plus 1mM sodium orthovanadate, 50 mM sodium floride and a complete protease inhibitor cocktail (Roche) at a ratio of 40-50 mg of tissue per ml of buffer. This homogenate was first centrifuged at 1500 g and the post-nuclear supernatant was again centrifuged at 12,600 g for 20 minutes at 4°C to get the crude synaptosomes fraction. This pellet was resuspended in 13% (w:v) (final concentration) Ficoll PM400 (in buffer A) and layered on the bottom of a discontinuous gradient, composed by buffer A and 7% Ficoll (in buffer A). The gradient was centrifuged at 100,000g (45 minutes, 4°C) in a TLS-55 swimming bucket rotor (Beckman-Coulter), and synaptosomes were isolated at the 7.5–13% interface. After washing (twice with buffer A), the pellet of synaptosomes was resuspended in Buffer A. The possible contamination with vacuolated postsynapses was evaluated by testing the presence of tau (presynaptic) and MAP2 (postsynaptic) proteins. Results (not shown) indicated the existence of a minimal contamination with vacuolated postsynapses.
On the other hand, the microsomal fraction was obtained after additional centrifugation of the 12,600 g supernatant at 100,000 g (1 hour, 4°C) in a TLA-110 rotor (Beckman-Coulter). The pellet of microsomes was also resuspended in Buffer A. The protein content of the both synaptosomal and microsomal fractions was determined by Lowry.
APPswe-expressing N2a cultures
APPswe-stably transfected Neuroblastoma cells were generously donated by Dr. Gopal Thinakaran (University of Chicago) [
35]. N2aAPPswe cells were cultured in high glucose DMEM-Optimem (50%-50%) supplemented with 2 mM glutamine and 5% fetal bovine serum (PPA Company), in presence of Penicillin and Streptomycin (100 units/ml and 0.01 mg/ml respectively) and G418 (PPA Company) as clonal selection antibiotic (0.2 μg/ml) [
35]. For cell drug treatments, a 0.2μm filtered stock solution of Chloroquine diphosphate salt (Sigma-Aldrich) or MG132 (Sigma-Aldrich) were diluted in the same media at a final concentration of 10 or 5 μM, respectively. This media was kept for 6 or 24 hours before collecting the cells and isolating RNA and protein as described above.
Statistical analysis
Data were expressed as mean ± SD. The comparison between two mice groups (WT and PS1/APP mice) was done by
t test. For comparison between several age groups, we used one-way ANOVA followed by Tukey post hoc multiple comparisons test (Statgraphics plus 3.1). As stated above, for most experiments, the different xgroups were compared with 6-month-old WT mice. In some cases (Figures
4A, E and F; Figure
5C and D) 6-month-old PS1/APP mice were used as reference. The significance was set at 95% of confidence.
Competing interest
The authors declare that they have no competing interests.
Authors’ contributions
MT, SJ, IC, VN and MV carried out the molecular experiments; R S-V, L T-E, and E M-S carried out the immunohistochemical experiments, JCD and MV participated in the design of experiment and revising the manuscript, AG and JV design the experiments, analyzed the data and wrote the manuscript. All authors read and approved the final manuscript.