Cathepsin B Improves ß-Amyloidosis and Learning and Memory in Models of Alzheimer's Disease

Abstract

Amyloid-ß (Aß) precursor protein (APP) metabolism engages neuronal endolysosomal pathways for Aß processing and secretion. In Alzheimer's disease (AD), dysregulation of APP leads to excess Aß and neuronal dysfunction; suggesting that neuronal APP/Aß trafficking can be targeted for therapeutic gain. Cathepsin B (CatB) is a lysosomal cysteine protease that can lower Aß levels. However, whether CatB-modulation of Aß improves learning and memory function deficits in AD is not known. To this end, progenitor neurons were infected with recombinant adenovirus expressing CatB and recovered cell lysates subjected to proteomic analyses. The results demonstrated Lamp1 deregulation and linkages between CatB and the neuronal phagosome network. Hippocampal injections of adeno-associated virus expressing CatB reduced Aß levels, increased Lamp1 and improved learning and memory. The findings were associated with the emergence of c-fos + cells. The results support the idea that CatB can speed Aß metabolism through lysosomal pathways and as such reduce AD-associated memory deficits.

Keywords: Adeno-associated virus; Gene therapy; Lysosomal degrading enzyme; Proteomics; Radial arm water maze.

Fig. 1

CatB suppresses Aß production from mouse NPC-derived neurons. a, b NPC-derived neurons were infected with AdAPPsw or co-infected with AdAPPsw and AdGFP, AdCysB or AdCatB. Aß40 (a) or Aß42 (b) production was quantified, and Aß42/Aß40 ration c was calculated. d Aß retention in neurons was examined using 6E10 antibody (top). Expression of CatB (HA-tagged CatB), CysB (Flag-tagged CysB) and GFP were to validate experimental condition (bottom). Bars represent mean ± S.E.M. *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA, Newman-Keuls post hoc test

Fig. 2

Proteomics changes induced by AdCatB. a Heatmap of 320 proteins that were significantly altered with AdCatB administration, common to both the AdGFP and untreated controls, indicating a near even amount of up- (164 proteins) and down- (156 proteins) regulation. b PANTHER analysis revealed enrichment of proteins involved in metabolic process. Expansion of this family of proteins identified primary metabolic processes as the major influencing subgroup, and a further expansion yielded protein metabolic processes as a primary enriched metabolic function with AdCatB treatment. c KEGG pathway analysis revealed that LAMP1 and vATPase, involved in the phagosomal compartments were significantly altered with AdCatB treatment as compared to AdGFP treated and untreated controls

Fig. 3

Optimization for AAV-GFP and CatB in vitro. a Immunoblots show expression of GFP and CatB in a dose-dependent manner in NPC-derived neurons transduced with AAVs at 1 × 10^7–9 vg/10,000 cells/well. b Overexpression of CatB in NPC-derived neurons has no effect on cell viability as compared to control (Con) or GFP group. c Hippocampal frozen sections were immunostained for HA to identify exogenous CatB expression. HA-immunostaining and GFP fluorescent images in the AAV-GFP or CatB-injected hippocampus were shown. Scale bar =100 μm (50 μm in high magnified images). d Hippocampal frozen sections were immunostained for GFAP (astrocyte) or Iba1 (microglia). Scale bar =200 μm. e Quantification of GFAP-positive cells in the hippocampus. f Quantification of Iba-positive cells in the hippocampus

Fig. 4

CatB attenuates Aß loads in the hippocampus of APP/PS1 mice. a Representative images of Aß staining in the hippocampus of AAV-GFP and AAV-CatB-injected APP/PS1 mice at 7 months of age. Scale bar =200 μm. b Quantification of total Aß loads in the hippocampal region (n = 7 per group, 12 sections per brain). c The levels of Aß40 and Aß42 in extracellular and intracellular-enriched fractions were measured by human Aß40 or Aß42-specific ELISA (n = 5). Bars represent mean ± S.E.M. *p < 0.05, one-way ANOVA, Newman-Keuls post hoc test

Fig. 5

AAV-mediated CatB transduction increases Lamp1 expression in the hippocampus of APP/PS1 mice and cultured neurons. a Immunoblot of Lamp1 in intracellular fraction of the mouse hippocampus. b Quantification of Lamp1 expression (n = 6). c Confocal microscopy shows cellular localization of Lamp1 lysosomal compartment (red) in neurons. d Lamp1 expression levels were quantified using ImageJ (n = 3). Bars represent mean ± S.E.M. *p < 0.05, **p < 0.01, one-way ANOVA, Newman-Keuls post hoc test

Fig. 6

AAV-CatB-mediated transduction improves learning and memory in APP/PS1 mice. a Non-Tg (n = 10), AD (n = 9), AAV-GFP (n = 8) or CatB (n = 10)-injected APP/PS1 mice were tested by the RAWM task at 6–7 months of age. Non-Tg serves as a positive control for the spatial learning task. The compiled average errors for day 1–3, 4–6 and 7–9 are shown. Bars represent mean ± S.E.M. *p < 0.05, two-way ANOVA, Bonferroni post hoc test. b Immunohistochemical detection of c-fos-labeled cells in the dentate GCL. Scale bar, 200 μm. c Quantification of the number of c-fos-labeled cells (n = 7 mice per group, 12 sections per mouse). Bars represent mean ± SEM. ^a,b,c p < 0.05, ^aaa p < 0.001, ^a,aaa vs non-Tg, ^b vs uninjected APP/PS1, ^c vs AAV-GFP-injected APP/PS1, one-way ANOVA, Newman-Keuls post hoc test