Different Expression Patterns of Amyloid-β Protein Precursor Secretases in Human and Mouse Hippocampal Neurons: A Potential Contribution to Species Differences in Neuronal Susceptibility to Amyloid-β Pathogenesis

Abstract

Extensive loss of hippocampal neurons serves a pathological basis for irreversible cognitive impairment in patients with Alzheimer’s disease (AD). However, this characteristic cannot be replicated by transgenic mouse models, and its underlying mechanisms are unclear. Here, we present evidence that different expression patterns of amyloid-β protein precursor (AβPP) secretases in human and mouse hippocampal neurons are a decisive cause of species difference in the susceptibility to Aβ pathogenesis. Cell bodies of both pyramidal and granular neurons did not appear to undergo Aβ deposits in the 10-month-old transgenic mutant human AβPP/presenilin-1 (PS1) mice. They expressed high levels of non-amyloidogenic α-secretase, and its neuroprotective products soluble AβPPα, but low levels of amyloidogenic β-secretase and γ-secretase, and a neurotoxic product, Aβ₄₂ peptide. Unlike those found in the mouse, human hippocampal neuronal cell bodies expressed β-secretase and γ-secretase, but not α-secretase, which could increase Aβ generation, thus undergoing death in response to various pathological conditions. Increased hippocampal neuronal apoptosis at 48 h following local microinjection of α-secretase antibody ADAM10 into the hippocampus of AβPP/PS1 mice further suggests that high α-secretase expression in mouse neuronal cell bodies is a factor in the paucity of neuronal loss in AD-like pathology. Therefore, selective down-regulation of brain α-secretase in transgenic AD models will better replicate the disease spectrum, including decreased brain soluble AβPPα levels and massive neuronal loss in AD patients, and be beneficial for preclinical therapeutic evaluation of AD.

Keywords

Alzheimer’s disease amyloid-β metabolism AβPP/PS1 mice hippocampus

INTRODUCTION

Alzheimer’s disease (AD) is a devastating and debilitating disease of aging, imposing tremendous financial and social costs. Currently, due to the complex AD pathogenesis and pathology and lack of effective drug targets and reliable platforms for preclinical therapeutic evaluation, there are no effective therapies for the disease [1, 2].

Progressive loss of hippocampus can serve as a starting point for the pathogenesis of AD [3]. Even prior to amyloid-β (Aβ) plaque deposition, the hippocampal region, especially the entorhinal cortex, undergoes neuronal degeneration and loss [4]. Massive neuronal loss associated with hippocampal atrophy is a pathological basis for irreversible cognitive impairment in patients with AD [5]. Hippocampal atrophy is a strong diagnostic marker for the conversion of mild cognitive impairment to AD [6, 7].

Various transgenic rodent models, such as those overexpressing mutant human Aβ protein precursor (AβPP), presenilin-1 (PS1), or with both PS1 and tau, are valuable tools used to understand the pathophysiology of AD, and to test new therapeutic agents. However, none of these transgenic models of AD mimic massive neuronal loss in the hippocampus, and subsequently may not provide reliable results when testing therapeutic strategies [8 –10]. Undoubtedly, these shortcomings will reduce the practical value of AD basic research and translational medicine. In addition, there are several studies that have shown age-dependent hippocampal volume and neuron loss in the PDAPP-J20 (hAPP-J20) mouse model of AD, but with contradictory results in neurogenesis status and extent and region of hippocampal neuronal degeneration [11 –14]. Thus, optimization of the existing AD transgenic models, via induction of massive neuronal loss in the hippocampus, will help to fully recapitulate the pathological process in AD patients. To achieve this, one needs to understand the causes and mechanisms underlying the lack of hippocampal neuronal loss in the AD transgenicmice.

In order to determine why no massive neuronal death occurs in AD transgenic rodent models, we investigated lamina-specific Aβ deposition and distribution of Aβ metabolism-related molecules in the hippocampus of 10-month-old AβPP/PS1 mice. We also compared Aβ metabolism-related molecule expression in the hippocampus between mice and humans. Finally, we determined whether blocking α-secretase activity could induce hippocampal neuronal loss in AβPP/PS1 mice.

MATERIALS AND METHODS

Human specimens

Four cadavers aged 45–63 years, without known neurological or psychiatric illnesses (Table 1), were obtained within 4–6 h of death, via informed donation for the Medical Education and Research of Nanjing Medical University, with corresponding written consents prepared by the donors and their families. The utilization of human tissues was approved by theEthics Committee of Nanjing Medical University.

Animals

Ten-month old male AβPP/PS1 (AβPPSwe-PS1ΔE9 transgenic) mice and wild-type (WT) mice were housed under controlled illumination (12:12 h light/dark cycle), humidity (30–50% ), and temperature (18–22°C). This transgenic mouse strain coexpressed the human sequence within the Aβ domain with Swedish mutations K594 N/M595 L and the exon-9-deleted human presenilin-1 under the control of the mouse prion promoter with plaque deposition beginning at 5-6 months [15]. All experiments were conducted in accordance with international standards on animal welfare and the guidelines of the Institute for Laboratory Animal Research of Nanjing Medical University.

Tissue preparation

Cadavers were perfused through the internal carotid artery with 10,000 ml of a 10% formalin solution. The middle pats of the hippocampus on both sides were dissected from the cerebral temporal lobe cortex and post-fixed in 4% paraformaldehyde/PB at 4°C for 3 days. Mice were deeply anesthetized using 2% sodium pentobarbital (40 mg/kg, i.p.), then transcardially perfused with 0.9% saline followed by 4% freshly-prepared paraformaldehyde in phosphate buffer (PB, 0.1 M, pH 7.4). Forebrains were dissected and post-fixed overnight.

Microinjection of anti-ADAM10 antibody

AβPP/PS1 mice were intraperitoneally anesthetized using 3.5% chloral hydrate (350 mg/kg). Two microliters of rabbit anti-ADAM10 antibody (1:50; Millipore) in CSF was injected into the left hippocampal CA1 using a glass micropipette with 30 μm diameter tip attached to a 5-μl Hamilton microsyringe (coordinates relative to Bregma: -1.70 mm anterior/posterior, +1.25 mm medial/lateral, and +1.5 mm dorsal/ventral) [16]. The micropipette was withdrawn 5 min after injection. The right CA1 received 2 μl normal rabbit IgG (1:50; Santa Cruz Biotech) in CSF only. After recovery from anesthesia, mice were returned to their home cages and sacrificed 3 days later.

Section preparation

The postfixed human or mouse hippocampal tissues were dehydrated in a series of graded ethanol solutions, then embedded in paraffin. Coronal brain sections were serially cut at 5 μm using a paraffin slicing machine (Leica RM2135, Nussloch, Germany). Every tenth section was collected as 1 set, and 10 sets per brain sample were obtained.

Immunohistochemistry

Paraffin-embedded hippocampal sections were dewaxed in xylene, rehydrated successively in ethanol, then rinsed in sterile water. The sections were incubated with one of the following primary antibodies: rabbit anti-Aβ₄₂ (1:300; Abcam), mouse anti-sAβPPα (1:100; Immuno-Biological Laboratories), rabbit anti-ADAM10 (1:400; Millipore), mouse anti-BACE1 (1:400; Millipore), rabbit anti-PS1 (1:300; Sigma-Aldrich), rabbit anti-NEP (1:200; Millipore), rabbit anti-LRP1 (1:400; Santa Cruz Biotechnology), rabbit anti-AβPP (1:400; Millipore), mouse anti-synaptophysin (1:1000; Millipore), or mouse anti-β-tubulin III (1:2000; Sigma-Aldrich) overnight at 4°C. Following PBS rinsing, the sections were incubated with biotinylated goat anti-mouse or rabbit (1:400) for 1 h at room temperature and visualized using the diaminobenzidine (DAB) visualization method (Elite ABC Kit, Vector, Burlingame, CA, USA). To obtain stable results of immunohistochemistry, all experimental conditions were kept in consistence with each other. Especially, the antibody-bound peroxidase is visualized by incubating sections for 3–4 min in 0.05% DAB with H₂O₂ to avoid staining too strong. Sections labeled by AβPP were counterstained by Congo red.

TUNEL assay

The paraffin-embedded hippocampal sections from AβPP/PS1 mice received CA1 microinjection of anti-ADAM10 antibody were dewaxed in xylene,rehydrated successively in ethanol, then rinsed in sterile water. After permeabilized with 0.1% Triton X-100 (dissolved in 0.1% sodium citrate) for 5 min, the coverslips or sections were incubated with 20 μg/ml proteinase K for 15 min at 37°C, then incubated with 100 μl TUNEL reaction mixture for 1 h at 37°C. The reaction was stopped by washing with PBS 3 times. The sections were then counterstained with Hoechst 33342. The number of TUNEL-positive cells on the CA1 pyramidal layer was counted on three sections per mouse. Only those animals found to have microinjection needle tip positioned correctly were used for the statistical evaluation of the results.

Hippocampal neuronal counting

Stereological quantification of hippocampal neurons of AβPP/PS1 mice and WT mice was performed as recently described [14]. Briefly, every tenth coronal section containing the hippocampus per animal was stained with NeuN as described above. Total number of NeuN cells in the pyramidal layer of the CA1 and CA2/3 subregions and granule cell layer of the dentate gyrus was quantified using the formula: T = (N × V) ÷ t. Where N, V, and t are the cell density, volume of the structure, and thickness of the section, respectively. The area of CA1 pyramidal layer, CA2/3 pyramidal layer and granular cell layer on each section (S1 … Sn) was respectively measured with Image-Pro Plus 6.0 Analysis System (Media Cybernetics Inc., San Francisco, CA, USA). Then, the volume of the above structures was calculated using the following formula: V = accumulative total area (0.5S1 + S2 + S3+ … +Sn-1 + 0.5Sn μm²) × inter-section distance (50 μm) [17].

Quantitative analyses of images

The light or fluorescence micrographs were captured at × 100 magnification using a Leica DM4000B digital microscope with constant exposure time, offset, and gain for each staining marker, and analyzed by the Image-Pro Plus 6.0 Analysis System. The laminar cytoarchitecture of human and mouse hippocampus was identified according to β-tubulin III and synaptophysin immunostainings using The Human Central Nervous System: A Synopsis and Atlas [18], and The Mouse Brain in Stereotaxic Coordinates [16], respectively. The hippocampal laminae on each micrograph were manually delineated. For analysis of Aβ load in the hippocampus of AβPP/PS1 mice, the percentage of area covered by Thioflavin S-positive Aβ plaques in each hippocampal lamina was measured using interest grayscale threshold analysis with constant settings for minimum and maximum intensities as described previously [19]. The percentage area of positive signal was calculated by dividing the area of positive signal to the total area in the region of interest. The average of 5 sections was used to represent a plaque load for each mouse. For analysis of lamina-specific expression of Aβ₄₂, sAβPPα, ADAM10, BACE1, PS1, AβPP, IDE, NEP, or LRP1, the baseline threshold values for each antibody were obtained by measuring the mean immunoreactive optical density (MIOD = IOD/total area) on the corresponding sections of the immunohistochemical controls that the primary antibodies were omitted and replaced with an equivalent concentration of normal mouse or rabbit serum. And then, the relative MIOD levels of immunostaining with each antibody in each hippocampal lamina were measured. The average of 5 human or mouse sections was used to estimate the MIOD levels of immunostaining with each antibody.

Statistical analysis

Using SPSS software, version 16.0 (SPSS Inc., USA), data were analyzed by Student’s t-test orOne-Way-ANOVA, followed by Newman-Keuls post-hoc multiple comparison test, with the significance level set to p < 0.05.

RESULTS

Cytoarchitecture and lamina of the mouse hippocampus

For ease in understanding the result descriptions, we briefly introduced the cytoarchitecture and lamina of the mouse hippocampus. Figure 1A is a cartoon-like image showing the two main types of hippocampal neurons, pyramidal and granular, whose somata are located at the CA1-CA3 pyramidal (Py) layer and dentate gyrus granular (Gr) layer, respectively [20, 21]. As shown by immunostaining with β-tubulin III (Fig. 1B), the pyramidal neurons send their axons, basal dendrites, and apical dendrite trunks along with branches, to occupy the alveus (alv), oriens (Or) layer, radiatum layer (Rad), and lacunosum moleculare layer (LMol), respectively [20]. The dendrites of granular neurons extend towards the moleculare layer (Mol), finally reaching the hippocampal fissure, while their axons extend towards the polymorph layer (Po), terminating in the stratum lucidum (Slu) of CA3 [20]. The hippocampus also receives numerous synaptic inputs, including those originating from the entorhinal cortex via the perforant path and contralateral hippocampus via the associational commissural pathway [22]. These synaptic inputs also exhibit clear-cut lamination, as shown by the immunostaining of a presynaptic marker synaptophysin (Fig. 1 C).

No obvious hippocampal neuronal loss in AβPP/PS1mice

The AβPP/PS1 mice is one of the most widely used AD models, replicating several key features of AD, including Aβ deposition with dystrophic neurites, synaptic dysfunction and cognitive deficits, but fails to develop extensive neuronal death [8 –10]. To prove this, we first compared the number of NeuN-positive neurons in the Py and Gr in 10-month-old AβPP/PS1 mice and age-matched WT mice. As expected, AβPP/PS1 mice did not show vast neuronal loss in the Py of the CA1 and CA2/3 subregions and Gr of the dentate gyrus (Fig. 2A-D).

Lamina-specific deposition of Aβ plaques in the hippocampus of AβPP/PS1 mice

Aβ accumulation is a primary cause for neuronal death in the AD process [23]. To better understand why there is no marked neuronal loss in the hippocampus of 10-month-old AβPP/PS1 mice, we first characterized Aβ deposition in each of the hippocampal laminae of the animals using thioflavin-S florescence staining. The deposition of Aβ plaques was predominant in the LMol and Mol, especially around blood vessels aligned with the hippocampal fissure (Fig. 3A). Quantitative image analysis of amyloid plaque burden revealed that the percentage area occupied by thioflavin-S positive plaques was highest in the LMol and Mo, moderate in the alv, Or, Rad and Po, and lowest in the Py and DG [F(6, 54) = 30.06, p < 0.001] (Fig. 3B). More than 75% of Aβ plaques in the whole hippocampus were present at LMol and Mo, while less than 5% in the Py and DG (Fig. 3 C).

Lamina-specific distribution of Aβ₄₂ and sAβPPα in the hippocampus of AβPP/PS1 mice

An imbalance between the amyloidogenic pathway and non-amyloidogenic pathway is one cause for Aβ deposition [23]. Thus, we determined whetherlamina-specific plaque deposition was associated with lamina-specificity of Aβ₄₂ and soluble AβPPα (sAβPPα), the products of these two mutually exclusive pathways in the hippocampus of AβPP/PS1 mice. Aβ₄₂ immunohistochemistry revealed that Aβimmunoreactivity was highest in the LMol, Mol, and Po, and lowest in the Py and DG [F(6, 54) =37.194, p < 0.001] (Fig. 4A, B, E). In contrast, Py and DG had higher expression levels of sAβPPα than other hippocampal laminae [F(6, 54) = 8.314, p < 0.001] (Fig. 4 C, D, F). These results demonstrated that hippocampal neuronal somata in the Py and DG produce high levels of neurotrophic and neuroprotective sAβPPα and low levels of neurotoxic Aβ₄₂ in middle-aged AβPP/PS1 mice.

Lamina-specific distribution of AβPP in the hippocampus of AβPP/PS1 mice

AβPP is a precursor molecule that generates Aβ during proteolysis [23]. Therefore, we determined whether laminar vulnerability to Aβ deposition was due to lamina-specific distribution of AβPP in the hippocampus of AβPP/PS1 mice using AβPP immunohistochemistry and Congo red staining. Dense AβPP immunoreactive products were localized within the Slu and Po, where only a few Congo red-positive plaques were observed [F(6, 54) = 11.547, p < 0.001] (Fig. 5A, C). This suggested that lamina-specific distribution of AβPP was present in the hippocampus of AβPP/PS1 mice, but not related to lamina-specific plaque deposition and Aβ₄₂ accumulation. In addition, dystrophic neurites around amyloid plaques, primarily scattered within the LMol and Mol, were immunopositive for AβPP (Fig. 5B). This was consistent with previous studies reporting that amyloid plaques-related AβPP in the hippocampus are synthesized at distal sites and axonally transported to synaptic terminals via the perforant pathway [24, 25].

Lamina-specific distribution of AβPP secretases in the hippocampus of AβPP/PS1 mice

AβPP is cleaved by β-secretase and γ-secretase to generate Aβ, but, in contrast, α-secretase releases sAβPPα, precluding Aβ generation [23]. Therefore, we determined whether laminar differences in Aβ₄₂ and sAβPPα expression, and Aβ deposition, were due to lamina-specific distribution of α-, β- and γ-secretases in the hippocampus of AβPP/PS1 mice. α-secretase, labeled by anti-disintegrin and metalloproteinase 10 (ADAM10), showed high expression in the alv, Or, Py, Rad, Gr, Po, and SLu, and low expression in the LMoL and Mol [F(6, 54) = 11.174, p < 0.001] (Fig. 6A, G). β-secretase, labeled by anti-β-site amyloid precursor protein-cleaving enzyme 1 (BACE1), was expressed in the Or, Rad, LMol, Mol, Po, and SLu, but very little expression in the alv, Py and Gr [F(6, 54) = 10.405, p < 0.001] (Fig. 6B, H). γ-secretase, labeled by anti-PS1, was only highly presented at the Rad, Mol and SLu [F(6, 54) = 10.066, p < 0.001] (Fig. 6 C, I).

Examinations of hippocampal sections at higher magnification revealed that α-secretase was primarily expressed by cell bodies of pyramidal and granular neurons, and axons in the alv (Fig. 6D). It was also localized to the axonal terminates in the Or, Rad, Mol, Po, and SLu, sharing a similar expression pattern with synaptophysin, a presynaptic vesicle protein (data not shown) [26]. β-secretase was localized to all elements of the dendrites (basal dendrites and both trunks and branches of apical dendrites) of pyramidal and granular neurons (Fig. 6E), while γ-secretase was limitedly located to the proximal dendritic portions of pyramidal and granular neurons(Fig. 6F).

In summary, the above immunolocalization provides clear evidence that α-secretase, β-secretase and γ-secretase exhibited distinctive patterns of the hippocampal laminar distribution. The cell bodies of hippo-campal pyramidal and granular neurons expressed high levels of α-secretase, but low levels of β-secretase and γ-secretase, which would prevent Aβ generation and neurotoxicity. In contrast, low expression of α-secretase, but high expression of β-secretase within the LMol and MoL resulted in axonally transported AβPP to be substantially hydrolyzed into Aβ peptides that are subsequently released and deposited as plaques in the extracellular spaces of these hippocampal laminae.

Lamina-specific distribution of Aβ-clearance related molecules in the hippocampus of AβPP/PS1 mice

Neprilysin (NEP) and insulin degrading enzyme (IDE) mediate proteolytic degradation, while low density lipoprotein receptor-related protein-1 (LRP1) mediate cellular uptake and lysosomal degradation, constituting the two major pathways for eliminating Aβ from brain parenchyma [27]. Thus, we determined whether no marked loss of pyramidal and granular neurons in the hippocampus of AβPP/PS1 mice was also associated with high expression of IDE, NEP and LRP1 in these neurons. As expected, there was high staining intensity for IDE and NEP in the cytoplasm and LRP1 on the cytomembrane of both pyramidal neurons and granular neurons (Fig. 7A-C). Quantitative analysis further demonstrated that the pixel intensity values of IDE, NEP and LRP1 in the Py and Go were higher than in other hippocampal laminae [IDE: F(6, 54) = 4.481, p < 0.001; NEP: F(6, 54) = 11.866, p < 0.001; LRP1: F(6, 54) = 9.173, p < 0.001] (Fig. 7D-F).

Differences in expression pattern of AβPP secretases between the human and mouse hippocampal neurons

The above results demonstrate that mouse hippocampal neurons highly express non-amyloidogenic α-secretase, and Aβ clearance-related proteins including IDE, NEP, and LRP1, and low levels of β-secretase and γ-secretase. However, it was not clear which one plays the key role in preventing neurotoxicity in hippocampal neurons. To address this, we examined Aβ metabolism-related molecule expression in the normal hippocampus between mice and humans.

Human α-secretase was confined to the alv consisting of axons originating from hippocampal pyramidal neurons, and was not expressed by cell bodies, dendrites of pyramidal neurons (Fig. 8A, G). In contrast, mouse α-secretase was localized to cell bodies and axons of pyramidal neurons, but not to dendrites (Fig. 8D, J). In addition, α-secretase was expressed by hippocampal axon terminals of mice, but not humans. Human β-secretase was expressed by all neuronal elements and extensively distributed within the hippocampus (Fig. 8B, H). β-secretase on mouse pyramidal neurons was primarily localized to basal and apical dendrites and their branches, with very low levels on the cell bodies and no expression at the axons and terminals (Fig. 8E, K). Human γ-secretase was expressed by pyramidal neuronal cell bodies and axons (Fig. 8 C, I), while, mouse pyramidal neurons γ-secretase was mainly localized to basal and apical dendrites and axons, with low levels on the cell bodies and dendritic ends (Fig. 8F, L).

The semi-quantitative analysis further confirmed differences in expression patterns of AβPP secretases between normal human and mouse hippocampal laminae. Both human and mouse alv and Or had high expression of α-secretase, but levels of α-secretase were much low in the human Py, RAd, LMOL, MOL, Gr and Po, compared to those in mice (t = 6.493, 11.616, 9.667, 7.332, 7.802, 7.310, respectively; all p < 0.001, with 6 degrees of freedom) (Fig. 8M). Human β-secretase expression was high in the alv, Py, and Gr, but low in the Rad, LMol and Mol, relative to that in the corresponding laminae of mice (t = 3.467, 3.079, 3.445, 3.576; 3.137; 3.767, respectively; all p < 0.05, with 6 degrees of freedom) (Fig. 8 N). Similarly, human γ-secretase levels were high in the alv, Py, and Gr, but low in the Rad and Mol, compared with those of mice (t = 3.378, 3.119, 2.717, 3.391, 2.485, respectively; all p < 0.05, with 6 degrees of freedom) (Fig. 8O).

In addition, both human and mouse hippocampal neurons expressed NEP, IDE, and LRP1 (Fig. 9A-F). The lmamina-specific expression of these Aβ-clearance related molecules was generally similar between humans and mice (Fig. 9G-I).

Together, the above results reveal species differences in the expression pattern of AβPP secretases between human and mouse hippocampal neurons. No expression of α-secretase in human neuronal cell bodies, contrasted to high expression in the mouse, may be a determining factor for the widespread neuronal loss in the human hippocampus, but not in the mouse, during the AD process.

Blocking α-secretase activity induced hippocampal neuronal apoptosis in AβPP/PS1 mice

To further confirm that high expression of α-secretase plays a decisive role for protecting hippocampal neurons from death in AβPP/PS1 mice, we observed hippocampal neuronal apoptosis after blocking α-secretase activity by microinjection of anti-ADAM10 antibody. As expected, anti-ADAM10 antibody treatment significantly increased the number of terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) positive pyramidal neurons as compared with IgG control (Fig. 10A, B).

DISCUSSION

AD is a common form of age-related neurodegeneration, and currently lacks effective therapy. Transgenic mouse models are a unique tool used to explore disease pathogenesis and investigate therapeutic strategies for AD, but cannot recapitulate the complete human neuropathological spectrum, including extensive hippocampal neuronal death and hippocampal atrophy. This limitation produces a large number of positive results obtained when employing therapeutic strategies in these AD models, which failed to translate into clinically beneficial outcomes. Thus, finding the mechanisms underlying species differences between humans and mice in neuronal susceptibility to Aβ pathogenesis is an important premise for establishing better models that are able to avoid false positive findings during the preclinical therapeutic evaluation of AD. In the present study, we have presented evidences that different expression patterns of AβPP secretases in human and mouse hippocampal neurons are a decisive cause of species difference in the susceptibility to Aβ pathogenesis. The main findings are as follows: i) characterizing the hippocampal laminal distribution and interrelation of Aβ metabolism-related molecules and products in 10-month-old AβPP/PS1 mice; ii) revealing the species differences in expression patterns of AβPP secretases in human and mouse hippocampal neurons; iii) demonstrating high expression of α-secretase is a key mechanism for protecting hippocampal neurons from massive loss in AD-like pathology.

The hippocampus is composed of a series of clearly defined laminae. Each hippocampal lamina shows differences in their anatomical, chemical and functional properties [28 –30]. It is also a region vulnerable to Aβ pathology in human and rodents [5, 6]. However, to the best of our knowledge, lamina-specific Aβ accumulation and aggregation in the hippocampus of AD patients or animal models has yet to be addressed. In the present study, we determined the lamina-specific distribution of sAβPPα, Aβ₄₂, and Aβ plaques in the hippocampus of 10-month-old AβPP/PS1 mice. Our results show that LMol and Mo layers had the highest amyloid plaque burden, while Py and Gr layers had the lowest. The similar lamina distribution of Aβ plaques also occurs in 6-12 month AβPP/PS1 mice [30 –35] and PDAPP mice [11–14 , 37], according to the hippocampal micrographs of Aβ burden in these published papers. There was a lack of obvious Aβ load in the Py and Gr, even in 16-18 old AβPP/PS1 mice [38 –40], suggesting that mouse hippocampal neuronal cell bodies have extremely high ability to locally suppress Aβ generation and improve its clearance.

Our results further revealed that the laminar heterogeneity of sAβPPα and Aβ, both in soluble and plaque forms, is correlated with laminar variability in expression of AβPP secretases in AβPP/PS1 mice. AβPP cleavage is controlled by both α-secretase mediated non-amyloidogenic pathway and β-, γ-secretase-mediated amyloidogenic pathway [23]. Consistent with high expression of sAβPPα, low expression of Aβ₄₂ and the near absence of Aβ plaques in the Py and Gr, cell bodies of both pyramidal neurons and granular neurons express high levels of α-secretase but low levels of β-secretase and γ-secretase. In contrast, high levels of Aβ accumulation, and low levels of sAβPPα in the LMol and Mo, are related to high expression of β-secretase and low expression of α-secretase. In addition, it is interesting that Aβ plaques within the LMol and Mo are mainly located in the brain parenchyma adjacent to microvessels aligned with the hippocampal fissure. Recent studies have suggested that the paravascular pathway facilitates the clearance of interstitial solutes (ISF), including a large proportion of soluble Aβ from the brain parenchyma [41–42]. ISF drainage is impaired following chronic exposure to Aβ in AβPP/PS1 mice, in turn, causing amyloid deposits along the perivascular spaces [43]. The results also show that Rad and Slu are rich in expression of α-secretase, β-secretase and γ-secretase, with relatively low Aβ accumulation. This phenomenon supports the view that the majority of AβPP is cleaved within the Aβ sequence by α-secretase, subsequently precluding the formation of Aβ peptides [23].

Brain accumulation of Aβ is a consequence of the imbalance between Aβ generation and Aβ clearance [24]. Apart from AβPP secretases, lamina-specific expression of Aβ clearance-related proteins may also contribute to the laminar heterogeneity of Aβ load in AβPP/PS1 mouse hippocampus. High expression of LRP1 on the cytomembrane of pyramidal and granular neurons will facilitate uptake of Aβ peptides in the adjacent extracellular space. This will result in subsequent proteolytic degradation within the cytoplasm by high levels of IDE and NEP, avoiding Aβ accumulation in the Py and Gr [24 , 45]. By contrast, low expression levels of the above proteins in the LMol and Mo may hamper clearance of ISF Aβ, gradually resulting in plaque deposits. Together, the above evidences indicate that laminar heterogeneity of Aβ accumulation is correlated with laminar variability in Aβ generation and/or clearance-related molecule expression and microvascular distribution in the hippocampus of AβPP/PS1 mice.

The second key finding of the present study reveals distinctive expression patterns of AβPP secretases between human and mouse hippocampal neurons, although the underlying mechanism for this species difference is not clear. Human neuronal cell bodies express β-secretase and γ-secretase, but not α-secretase, which may facilitate Aβ generation, subsequently causing neuronal death under pathological conditions. In contrast, mouse neuronal cell bodies express a relatively high level of α-secretase and a relatively low level of β-secretase and γ-secretase. This leads to most AβPP being hydrolyzed into neuroprotective sAβPPα, while a minority becomes neurotoxic Aβ, thus allowing them to survive long periods, even in environments with high concentrations of AβPP. Increased hippocampal neuronal apoptosis in AβPP/PS1 mice, following blockage of α-secretase activity, provides strong evidence to support this conclusion.

Furthermore, the present results suggest high levels of α-secretase activity and sAβPPα production are a determining factor for protecting hippocampal neurons from massive death in transgenic AβPP/PS1 mice. Apart from its neurotrophic and neuroprotective properties [46 –48], sAβPPα has been implicated in improving neurite outgrowth [49], synaptogenesis [50], long-term potentiation [51], and synaptic plasticity [52], all of which are necessary for ameliorating Aβ-induced neurodegeneration. Additionally, sAβPPα suppresses Aβ generation in transgenic AD mice [51 –54], and its mechanism may interact directly with BACE1 and interferes with the β-secretase/AβPP interaction [55]. Overexpressing human sAβPPα ameliorates Aβ pathogenesis in AβPP/PS1 mice. Correspondingly, immunoneutralization of sAβPPα enhances amyloidogenic AβPP processing in this AD model [56]. Several risk factors for sporadic AD, such as oxidative stress [56, 57], hypoxia [59, 60], and abnormal cholesterol metabolism [61, 62], have been shown to decrease sAβPPα. Consistent with animal experimental data, mutations in the ADAM10 gene with reduced levels of α-secretase activity or sAβPPα, are associated with familial late-onset AD [63 –65]. These evidences together highlight that restoration of α-secretase activity, and/or sAβPPα levels, may have significant therapeutic potential for AD.

Finally, it should be mentioned that the present study investigated expression of Aβ metabolism-related molecules and products in the hippocampal laminae AβPP/PS1 mice only at ages of 10 months. Whether there are age-dependent expression patterns of AβPP metabolic proteins needs to be determined in the future. In addition, PDAPP-J20 mice show age-dependent progressive loss of neurons in the CA1, but not in the CA3, which correlates strongly with the expression of total Aβ [11]. Therefore, it will be interesting to compare hippocampal lamina-specific expression of AβPP metabolic proteins between AβPP/PS1 mice and PDAPP-J20 mice at different ages. Another limitation of the present study is that we have not compared Aβ metabolism-related molecule expression in the hippocampus of AβPP/PS1 mice and cadavers who have suffered from AD. All these potential findings would help to further reveal the underlying mechanisms for species difference in the susceptibility of hippocampal neurons to Aβ pathogenesis.

In conclusion, the present study reveals that expression patterns of mouse AβPP secretases are different from that of human, which prevents massive loss of mouse hippocampal neurons during Alzheimer-like pathology. Currently, we are establishing an α-secretase condition knockout gene in hippocampal neurons of 3xTg mice. Hopefully, this new transgenic AD mouse model will more fully mimic the typical pathophysiological changes in AD patients, thus be more beneficial for testing therapeutic strategies for AD.

Footnotes

ACKNOWLEDGMENTS

This research was supported by grants from the National Natural Science Foundation of China (No 81271210) and the Natural Science Foundation of Jiangsu Educational Department (14KJA320001).

Authors’ disclosures available online ().

References

Holtzman

, Morris

, Goate

(2011) Alzheimer’s disease: The challenge of the second century. Sci Transl Med 3, 77sr1.

Huang

, Mucke

(2012) Alzheimer mechanisms and therapeutic strategies. Cell 148, 1204–1222.

Cassel

, Mathis

, Majchrzak

, Moreau

, Dalrymple-Alford

(2008) Coexisting cholinergic and parahippocampal degeneration: A key to memory loss in dementia and a challenge for transgenic models? Neurodegener Dis 5, 304–317.

Caselli

, Reiman

(2013) Characterizing the preclinical stages of Alzheimer’s disease and the prospect of presymptomatic intervention. J Alzheimers Dis 33, S405–S416.

Whitwell

(2010) Progression of atrophy in Alzheimer’s disease and related disorders. Neurotox Res 18, 339–346.

Dhikav

, Anand

(2011) Potential predictors of hippocampal trophy in Alzheimer’s disease. Drugs Aging 28, 1–11.

Barnes

, Bartlett

, van de Pol

, Loy

, Scahill

, Frost

, Thompson

, Fox

(2009) A meta-analysis of hippocampal atrophy rates in Alzheimer’s disease. Neurobiol Aging 30, 1711–1723.

Bales

(2012) The value and limitations of transgenic mouse models used in drug discovery for Alzheimer’s disease: An update. Expert Opin Drug Discov 7, 281–297.

Bilkei-Gorzo

(2014) Genetic mouse models of brain ageing and Alzheimer’s disease. Pharmacol Ther 142, 244–257.

10.

Braidy

, Muñoz

, Palacios

, Castellano-Gonzalez

, Inestrosa

, Chung

, Sachdev

, Guillemin

(2012) Recent rodent models for Alzheimer’s disease: Clinical implications and basic research. J Neural Transm 119, 173–195.

11.

Wright

AL1

, Zinn

, Hohensinn

, Konen

, Beynon

, Tan

, Clark

, Abdipranoto

, Vissel

(2013) Neuroinflammation and neuronal loss precede Aβ plaque deposition in the hAPP-J20 mouse model of Alzheimer’s disease. PLoS One 8, e59586.

12.

Jin

, Galvan

, Xie

, Mao

, Gorostiza

, Bredesen

, Greenberg

(2004) Enhanced neurogenesis in Alzheimer’s disease transgenic (PDGF-APPSw, Ind) mice. Proc Natl Acad Sci U S A 101, 13363–13367.

13.

Lilja

, Rojdner

, Mustafiz

, Thome

, Storelli

, Gonzalez

, Unger-Lithner

, Greig

, Nordberg

, Marutle

(2013) Age dependent neuroplasticity mechanisms in Alzheimer Tg2576 mice following modulation of brain amyloid-beta levels. PLoS One 8, e58752.

14.

Beauquis

, Vinuesa

, Pomilio

, Pavía

, Galván

, Saravia

(2014) Neuronal and glial alterations, increased anxiety, and cognitive impairment before hippocampal amyloid deposition in PDAPP mice, model of Alzheimer’s disease. Hippocampus 24, 257–269.

15.

Jankowsky

, Slunt

, Ratovitski

, Jenkins

, Copeland

, Borchelt

(2001) Co-expression of multipletransgenes in mouse CNS: A comparison of strategies. Biomol Eng 17, 157–165.

16.

Franklin

KBJ

, Paxinos

(2008), The mouse brain in stereotaxic coordinates, 2nd Edition, Elsevier, Amsterdam.

17.

Pereda-Pérez

, Popović

, Otalora

, Popović

, Madrid

, Rol

, Venero

(2013) Long-term social isolation in the adulthood results in CA1 shrinkage and cognitive impairment. Neurobiol Learn Mem 106, 31–39.

18.

Nieuwenhuys

, Voogd

, Van Huijzen

(1988) The human central nervous system: A synopsis and atlas, 3rd edn. Springer, Berlin Heidelberg New York 97.

19.

Kim

, Castellano

, Jiang

, Basak

, Parsadanian

, Pham

, Mason

, Paul

, Holtzman

(2009) Overexpression of low-density lipoprotein receptor in the brain markedly inhibits amyloid deposition and increases extracellular A beta clearance. Neuron 64, 632–644.

20.

Spruston

(2008) Pyramidal neurons: Dendritic structure and synaptic integration. Nat Rev Neurosci 9, 206–221.

21.

Claiborne

, Amaral

, Cowan

(1990) Quantitative, three-dimensional analysis of granule cell dendrites in the rat dentate gyrus. J Comp Neurol 302, 206–219.

22.

Rolls

(2013) A quantitative theory of the functions of the hippocampal CA3 network in memory. Front Cell Neurosci 7, 98.

23.

Tanzi

, Bertram

(2005) Twenty years of the Alzheimer’s disease amyloid hypothesis: A genetic perspective. Cell 120, 545–555.

24.

Sheng

, Price

, Koliatsos

(2003) The beta-amyloid-related proteins presenilin 1 and BACE1 are axonally transported to nerve terminals in the brain. Exp Neurol 184, 1053–1057.

25.

Buxbaum

, Thinakaran

, Koliatsos

, O’Callahan

, Slunt

, Price

, Sisodia

(1998) Alzheimer amyloid protein precursor in the rat hippocampus: Transport and processing through the perforant path. J Neurosci 18, 9629–9637.

26.

Sze

CI1

, Troncoso

, Kawas

, Mouton

, Price

, Martin

(1997) Loss of the presynaptic vesicle protein synaptophysin in hippocampus correlates with cognitive decline in Alzheimer disease. J Neuropathol Exp Neurol 56, 933–944.

27.

Tanzi

, Moir

, Wagner

(2004) Clearance of Alzheimer’s Abeta peptide: The many roads to perdition. Neuron 43, 605–608.

28.

Brückner

, Grosche

, Hartlage-Rübsamen

, Schmidt

, Schachner

(2003) Region and lamina-specific distribution of extracellular matrix proteoglycans, hyaluronan and tenascin-R in the mouse hippocampal formation. J Chem Neuroanat 26, 37–50.

29.

Schönberger

, Draguhn

, Both

(2014) Lamina-specific contribution of glutamatergic and GABAergic potentials to hippocampal sharp wave-ripple complexes. Front NeuralCircuit 8, 103.

30.

Bas Orth

, Vlachos

, Del Turco

, Burbach

, Haas

, Mundel

, Feng

, Frotscher

, Deller

(2005) Lamina-specific distribution of Synaptopodin, an actin-associated molecule essential for the spine apparatus, in identified principal cell dendrites of the mouse hippocampus. J Comp Neurol 487, 227–239.

31.

Scott

, Feng

, Kiss

, Needle

, Atchison

, Kawabe

, Milici

, Hajós-Korcsok

, Riddell

, Hajós

(2012) Age-dependent disruption in hippocampal θ oscillation in amyloid-β overproducing transgenic mice. Neurobiol Aging 33, 1481. e13–e23.

32.

Donkin

, Stukas

, Hirsch-Reinshagen

, Namjoshi

, Wilkinson

, May

, Chan

, Fan

, Collins

, Wellington

(2010) ATP-binding cassette transporter A1 mediates the beneficial effects of the liver X receptor agonist GW3965 on object recognition memory and amyloid burden in amyloid precursor protein/presenilin 1 mice. J Biol Chem 285, 34144–34154.

33.

Ermini

, Grathwohl

, Radde

, Yamaguchi

, Staufenbiel

, Palmer

, Jucker

(2008) Neurogenesis and alterations of neural stem cells in mouse models of cerebral amyloidosis. Am J Pathol 172, 1520–1528.

34.

, Green

, Fromholt

, Borchelt

(2012) Reduction of low-density lipoprotein receptor-related protein (LRP1) in hippocampal neurons does not proportionately reduce, or otherwise alter, amyloid deposition in APPswe/PS1dE9 transgenic mice. Alzheimers Res Ther 4, 12.

35.

Oksman

, Iivonen

, Hogyes

, Amtul

, Penke

, Leenders

, Broersen

, Lütjohann

, Hartmann

, Tanila

(2006) Impact of different saturated fatty acid, polyunsaturated fatty acid and cholesterol containing diets on beta-amyloid accumulation in APP/PS1 transgenic mice. Neurobiol Dis 23, 563–572.

36.

, Ruberu

, Muñoz

, Jenner

, Spiro

, Zhao

, Rassart

, Sanchez

, Ganfornina

, Karl

, Garner

(2015) Apolipoprotein D modulates amyloid pathology in APP/PS1 Alzheimer’s disease mice. Neurobiol Aging 36, 1820–1833.

37.

Schenk

, Barbour

, Dunn

, Gordon

, Grajeda

, Guido

, Hu

, Huang

, Johnson-Wood

, Khan

, Kholodenko

, Lee

, Liao

, Lieberburg

, Motter

, Mutter

, Soriano

, Shopp

, Vasquez

, Vandevert

, Walker

, Wogulis

, Yednock

, Games

, Seubert

(1999) Immunization with amyloid-beta attenuates Alzheimer-disease-like pathology in the PDAPP mouse. Nature 400, 173–177.

38.

Hooijmans

, Graven

, Dederen

, Tanila

, van Groen

, Kiliaan

(2007) Amyloid beta deposition is related to decreased glucose transporter-1 levels and hippocampal atrophy in brains of aged APP/PS1 mice. Brain Res 1181, 93–103.

39.

van Groen

, Kiliaan

, Kadish

(2006) Deposition of mouse amyloid beta in human APP/PS1 double and single AD model transgenic mice. Neurobiol Dis 23, 653–662.

40.

Huang

, Wang

, Cao

, Marshall

, Gao

, Xiao

, Hu

, Xiao

(2015) Isolation housing exacerbates Alzheimer’s disease-like pathophysiology in aged APP/PS1 mice.pyu. Int J Neuropsychopharmacol 18, 116.

41.

Iliff

, Wang

, Liao

, Plogg

, Peng

, Gundersen

, Benveniste

, Vates

, Deane

, Goldman

, Nagelhus

, Nedergaard

(2012) A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid β. Sci Transl Med 4, 147ra111.

42.

Xie

, Kang

, Xu

, Chen

, Liao

, Thiyagarajan

, O’Donnell

, Christensen

, Nicholson

, Iliff

, Takano

, Deane

, Nedergaard

(2013) Sleep drives metabolite clearance from the adult brain. Science 342, 373–377.

43.

Arbel-Ornath

, Hudry

, Eikermann-Haerter

, Hou

, Gregory

, Zhao

, Betensky

, Frosch

, Greenberg

, Bacskai

(2013) Interstitial fluid drainage is impaired in ischemic stroke and Alzheimer’s disease mouse models. Acta Neuropathol 126, 353–364.

44.

Baranello

, Bharani

, Padmaraju

, Chopra

, Lahiri

, Greig

, Pappolla

, Sambamurti

(2015) Amyloid-beta protein clearance and degradation (ABCD) pathways and their role in Alzheimer’s disease. Curr Alzheimer Res 12, 32–46.

45.

Kanekiyo

, Cirrito

, Liu

, Shinohara

, Li

, Schuler

, Shinohara

, Holtzman

, Bu

(2013) Neuronal clearance of amyloid-β by endocytic receptor LRP1. J Neurosci 33, 19276–19283.

46.

Furukawa

, Sopher

, Rydel

, Begley

, Pham

, Martin

, Fox

, Mattson

(1996) Increased activity-regulating and neuroprotective efficacy of alpha-secretase-derived secreted amyloid precursor protein conferred by a C-terminal heparinbinding domain. J Neurochem 67, 1882–1896.

47.

Copanaki

, Chang

, Vlachos

, Tschape

, Muller

, Kogel

, Deller

(2010) sAPPalpha antagonizes dendritic degeneration and neuron death triggered by proteasomal stress. Mol Cell Neurosci 44, 386–393.

48.

Kogel

, Deller

, Behl

(2012) Roles of amyloid precursor protein family members in neuroprotection, stress signaling and aging. Exp Brain Res 217, 471–479.

49.

Gakhar-Koppole

, Hundeshagen

, Mandl

, Weyer

, Allinquant

, Muller

, Ciccolini

(2008) Activity requires soluble amyloid precursor protein alpha to promote neurite outgrowth in neural stem cell-derived neurons via activation of the MAPK pathway. Eur J Neurosci 28, 871–882.

50.

Bell

, Zheng

, Fahrenholz

, Cuello

(2008) ADAM-10 over-expression increases cortical synaptogenesis. Neurobiol Aging 29, 554–565.

51.

Taylor

, Ireland

, Ballagh

, Bourne

, Marechal

, Turner

, Bilkey

, Tate

, Abraham

(2008) Endogenous secreted amyloid precursor protein-alpha regulates hippocampal NMDA receptor function, long-term potentiation and spatial memory. Neurobiol Dis 31, 250–260.

52.

Turner

, O’Connor

, Tate

, Abraham

(2003) Roles of amyloid precursor protein and its fragments in regulating neural activity, plasticity and memory. Prog Neurobiol 70, 1–32.

53.

Postina

, Schroeder

, Dewachter

, Bohl

, Schmitt

, Kojro

, Prinzen

, Endres

, Hiemke

, Blessing

, Flamez

, Dequenne

, Godaux

, van Leuven

, Fahrenholz

(2004) A disintegrin-metalloproteinase prevents amyloid plaque formation and hippocampal defects in an Alzheimer disease mouse model. J Clin Invest 113, 1456–1464.

54.

Ring

, Weyer

, Kilian

, Waldron

, Pietrzik

, Filippov

, Herms

, Buchholz

, Eckman

, Korte

, Wolfer

, Müller

(2007) The secreted beta-amyloid precursor protein ectodomain APPs alpha is sufficient to rescue the anatomical, behavioral, and electrophysiological abnormalities of APP-deficient mice. J Neurosci 27, 7817–7826.

55.

Suh

, Choi

, Romano

, Gannon

, Lesinski

, Kim

, Tanzi

(2013) ADAM10 missense mutations potentiate β-amyloid accumulation by impairing prodomain chaperone function. Neuron 80, 385–401.

56.

Obregon

, Hou

, Deng

, Giunta

, Tian

, Darlington

, Shahaduzzaman

, Zhu

, Mori

, Mattson

, Tan

(2012) sAPP-α modulates β-secretase activity and amyloid-β generation. Nat Commun 3, 777.

57.

Tamagno

, Guglielmotto

, Aragno

, Borghi

, Autelli

, Giliberto

, Muraca

, Danni

, Zhu

, Smith

, Perry

, Jo

, Mattson

, Tabaton

(2008) Oxidative stress activates a positive feedback between the γ- and β-secretase cleavages of the beta-amyloid precursor protein. J Neurochem 104, 683–695.

58.

Quiroz-Baez

, Rojas

, Arias

(2009) Oxidative stress promotes JNK-dependent amyloidogenic processing of normally expressed human APP by differential modification of α-, β- and γ-secretase expression. Neurochem Int 55, 662–670.

59.

Auerbach

, Vinters

(2006) Effects of anoxia and hypoxia on amyloid precursor protein processing in cerebral microvascular smooth muscle cells. J Neuropathol Exp Neurol 65, 610–620.

60.

Webster

, Green

, Settle

, Peers

, Vaughan

(2004) Altered processing of the amyloid precursor protein and decreased expression of ADAM 10 by chronic hypoxia in SH-SY5Y: No role for the stress-activated JNK and p38 signalling pathways. Brain Res Mol Brain Res 130, 161–169.

61.

Racchi

, Baetta

, Salvietti

, Ianna

, Franceschini

, Paoletti

, Fumagalli

, Govoni

, Trabucchi

, Soma

(1997) Secretory processing of amyloid precursor protein is inhibited by increase in cellular cholesterol content. Biochem J 322, 893–898.

62.

Wahrle

, Das

, Nyborg

, McLendon

, Shoji

, Kawarabayashi

, Younkin

, Golde

(2002) Cholesterol-dependent γ-secretase activity in buoyant cholesterol-rich membrane microdomains. Neurobiol Dis 9, 11–23.

63.

Kim

, Suh

, Romano

, Truong

, Mullin

, Hooli

, Norton

, Tesco

, Elliott

, Wagner

, Moir

, Becker

, Tanzi

(2009) Potential late-onset Alzheimer’s disease-associated mutations in the ADAM10 gene attenuate α-secretase activity. Hum Mol Genet 18, 3987–3996.

64.

Lannfelt

, Basun

, Wahlund

, Rowe

, Wagner

(1995) Decreased alpha-secretase-cleaved amyloid precursor protein as a diagnostic marker for Alzheimer’s disease. Nat Med 1, 829–832.

65.

Manzine

, de França Bram

, Barham

, do Vale Fde

, Selistre-de-Araújo

, Cominetti

, Iost Pavarini

(2013) ADAM10 as a biomarker for Alzheimer’s disease: A study with Brazilian elderly. Dement Geriatr Cogn Disord 35, 58–66.

Different Expression Patterns of Amyloid-β Protein Precursor Secretases in Human and Mouse Hippocampal Neurons: A Potential Contribution to Species Differences in Neuronal Susceptibility to Amyloid-β Pathogenesis

Abstract

Keywords

INTRODUCTION

MATERIALS AND METHODS

Human specimens

Animals

Tissue preparation

Microinjection of anti-ADAM10 antibody

Section preparation

Immunohistochemistry

TUNEL assay

Hippocampal neuronal counting

Quantitative analyses of images

Statistical analysis

RESULTS

Cytoarchitecture and lamina of the mouse hippocampus

No obvious hippocampal neuronal loss in AβPP/PS1mice

Lamina-specific deposition of Aβ plaques in the hippocampus of AβPP/PS1 mice

Lamina-specific distribution of Aβ42 and sAβPPα in the hippocampus of AβPP/PS1 mice

Lamina-specific distribution of AβPP in the hippocampus of AβPP/PS1 mice

Lamina-specific distribution of AβPP secretases in the hippocampus of AβPP/PS1 mice

Lamina-specific distribution of Aβ-clearance related molecules in the hippocampus of AβPP/PS1 mice

Differences in expression pattern of AβPP secretases between the human and mouse hippocampal neurons

Blocking α-secretase activity induced hippocampal neuronal apoptosis in AβPP/PS1 mice

DISCUSSION

Footnotes

ACKNOWLEDGMENTS

References

Lamina-specific distribution of Aβ₄₂ and sAβPPα in the hippocampus of AβPP/PS1 mice