The autophagy and endocytic pathways converging on the lysosome are major routes for the processing of amyloid precursor protein (APP) and the generation and degradation of Aβ and β-cleaved C-terminal fragments. A continuum of endosomal-autophagic-lysosomal abnormalities in neurons of the AD brain is well established, including very early appearing swelling of neuronal endosomes reflecting pathologically accelerated endocytosis, increased lysosome biogenesis and striking autophagic-lysosomal pathway pathology characterized by robust accumulation of autophagic vacuoles (AVs) and lysosomes in dystrophic neurites throughout the AD brain. These AVs, most of which contain cathepsins, are filled with incompletely digested “waste” proteins (including Aβ), implying that lysosomal proteolysis is defective. Growing genetic and biochemical evidence has identified lysosomal proteolysis failure as the principal basis for autophagy dysfunction in AD.
Although boosting autophagy induction has been used, with some promise, to delay disease onset in AD mouse models, we have considered that remediation at the lysosomal level may be necessary in AD once proteolysis becomes impaired. To provide proof of principle that selectively targeting lysosomal dysfunction may be a therapeutic approach, we investigated TgCRND8 mice overexpressing mutant human APP695. Our evaluations of this mouse model documented marked deficits of autophagic-lysosomal function. Neurons in affected brain regions of TgCRND8 mice, but not wild-type mice (WT), exhibit grossly enlarged cathepsin (Cat) D-positive lysosomal compartments and relatively fewer normal-sized lysosomes. These giant neuronal lysosomal compartments appear ultrastructurally as electron-dense, single-membrane-limited vesicles of 1.5 to 5.0 µm diameter containing amorphous granular and membranous material and a minor lipopigment component and are distinguishable from lipofuscin. By double-immunofluorescence labeling, these compartments contain markers of both autophagosomes/autolysosomes and late endosomes. Collectively, these studies identified enlarged compartments as autolysosomes.
Further evidence indicated that enlarged autolysosomes are filled with incompletely digested autophagic substrates, reflecting inefficient degradation by lysosomal hydrolases. AV and lysosome fractions isolated from the brains of TgCRND8 mice contain abnormally high levels of LC3-II, ubiquitinated proteins and Aβ detected by immunoblotting. These antigens are also abundant in giant autolysosomes of CA1 neurons of TgCRND8 brain labeled immunocytochemically. Cathepsin activities measured in extracts of TgCRND8 brain are also significantly lowered relative to those in WT brains.
Given this evidence for impaired lysosomal proteolysis, we sought to restore more normal cathepsin activities in lysosomes of TgCRND8 mice by deleting the gene for CstB, an endogenous inhibitor of cysteine proteases. Using a highly specific affinity-purified polyclonal antibody, we showed that CstB mainly localizes to lysosomal compartments in mouse neurons. CstB deletion, as expected, partially relieves the suppression of multiple cathepsins, including, surprisingly, CatD: enzymatic activities of CatB, CatL and CatD are higher in CstB knockout mouse (CBKO) brains than in WT brains. Moreover, the rate of degradation of long-lived proteins is raised over WT levels in mouse primary fibroblasts and neurons from CBKO mice.