Sustained Systemic Glucocerebrosidase Inhibition Induces Brain α-Synuclein Aggregation, Microglia and Complement C1q Activation in Mice

Abstract

Aims: Loss-of-function mutations in GBA1, which cause the autosomal recessive lysosomal storage disease, Gaucher disease (GD), are also a key genetic risk factor for the α-synucleinopathies, including Parkinson's disease (PD) and dementia with Lewy bodies. GBA1 encodes for the lysosomal hydrolase glucocerebrosidase and reductions in this enzyme result in the accumulation of the glycolipid substrates glucosylceramide and glucosylsphingosine. Deficits in autophagy and lysosomal degradation pathways likely contribute to the pathological accumulation of α-synuclein in PD. In this report we used conduritol-β-epoxide (CBE), a potent selective irreversible competitive inhibitor of glucocerebrosidase, to model reduced glucocerebrosidase activity in vivo, and tested whether sustained glucocerebrosidase inhibition in mice could induce neuropathological abnormalities including α-synucleinopathy, and neurodegeneration.

Results: Our data demonstrate that daily systemic CBE treatment over 28 days caused accumulation of insoluble α-synuclein aggregates in the substantia nigra, and altered levels of proteins involved in the autophagy lysosomal system. These neuropathological changes were paralleled by widespread neuroinflammation, upregulation of complement C1q, abnormalities in synaptic, axonal transport and cytoskeletal proteins, and neurodegeneration.

Innovation: A reduction in brain GCase activity has been linked to sporadic PD and normal aging, and may contribute to the susceptibility of vulnerable neurons to degeneration. This report demonstrates that systemic reduction of GCase activity using chemical inhibition, leads to neuropathological changes in the brain reminiscent of α-synucleinopathy.

Conclusions: These data reveal a link between reduced glucocerebrosidase and the development of α-synucleinopathy and pathophysiological abnormalities in mice, and support the development of GCase therapeutics to reduce α-synucleinopathy in PD and related disorders.

FIG. 1.

Chronic conduritol-β-epoxide (CBE) treatment inhibits glucocerebrosidase (GCase) activity and promotes the accumulation of glucosylceramide (GluCer) and glucosylsphingosine (GluSph). Chronic (28 days) systemic treatment of CBE at 100 mg/kg in mice inhibits GCase activity in the brain. Data show that CBE blocked GCase activity in forebrain tissue homogenates from CBE-treated mice at 24 h postfinal CBE injection (A). The same forebrain tissue homogenates were used to determine GluCer and GluSph accumulation by mass spectroscopy chromatography. GluCer and GluSph accumulated in the forebrain following chronic inhibition of GCase by CBE (B). To determine the kinetics of GCase activity recovery and reductions in GluSph levels over time, GCase and GluSph levels were measured at 3, 5, and 7 days postchronic (28 days) CBE treatment in the forebrain, midbrain, and cerebellum (C). GCase activity gradually increased and returned to basal levels by 7 days in the forebrain and by 5 days in the midbrain and cerebellum (C). GluSph levels gradually decreased and returned to basal levels by 3 days in the forebrain and 5 days in the midbrain and cerebellum (D). *p<0.05, unpaired t-test. n=3–5/group. Graphs are expressed as mean±SEM.

FIG. 2.

GCase inhibition induced widespread neuroinflammatory response in mice. Chronic 28-day treatment of CBE (100 mg/kg) increased the number of Iba-1-positive microglia throughout the brain and these microglia remained highly reactive even 7 days postcessation of CBE (A). CBE caused a distinct microglial response in cortical layer V (A), which can clearly be seen at magnification (B). CBE also caused a microglial response in the striatum and substantia nigra, as evidenced by Iba-1-positive microglial (C).

FIG. 3.

GCase inhibition resulted in dysregulation of proteins involved in synaptic transmission in mice. Western blots were performed to determine whether chronic 28-day treatment of CBE (100 mg/kg) altered expression levels of proteins necessary for synaptic transmission. Whole-cell lysates (20–30 μg) from the striatum, substantia nigra, and motor cortex were homogenized and processed. The data indicate that CBE induced the expression of C1q, the initiating protein of the classical complement in the striatum, substantia nigra, and motor cortex (A). Levels of synaptosomal-associated protein 25 (SNAP-25) remained unchanged in the striatum, substantia nigra, and motor cortex following CBE treatment (B). Protein levels of synaptophysin were increased in the striatum and motor cortex in mice treated with CBE. In contrast, synaptophysin was decreased in the substantia nigra of CBE-treated mice (C). Levels of synaptotagmin were decreased in the motor cortex and substantia nigra. Levels of synaptotagmin remained unchanged in the striatum (D). Levels of the presynaptic protein, synapsin, remained unchanged in the striatum, substantia nigra, and motor cortex (E). Levels of the postsynaptic protein, postsynaptic density protein 95 (PSD-95), remain unchanged in the striatum and motor cortex (F). Levels of PSD-95 were decreased in the substantia nigra in mice treated with CBE (F). Optical densities (ODs) of the individual bands were quantified using NIH ImageJ and normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) on the same gel. *p<0.05, unpaired t-test. n=3–5/group. Graphs are expressed as mean±SEM.

FIG. 4.

GCase inhibition caused dysregulation of proteins involved in retrograde transport in mice. Western blots were performed to determine whether chronic 28-day treatment of CBE (100 mg/kg) altered expression levels of proteins necessary for axonal transport. Whole-cell lysates (20–30 μg) from the striatum, substantia nigra, and motor cortex were homogenized and processed. Data indicate that the levels of the retrograde transport proteins, dynactin and dynein, were increased in the substantia nigra of mice treated with CBE (A, B). Levels of the anterograde transport proteins, KIF 2A and KIF 17, remained unchanged in the striatum, substantia nigra, and motor cortex (C, D). ODs of the individual bands were quantified using NIH ImageJ and normalized to GAPDH on the same gel. *p<0.05, two-way ANOVA. n=3–5/group. Graphs are expressed as mean±SEM.

FIG. 5.

GCase inhibition induced altered protein levels of cytoskeletal proteins in mice. Western blots were performed to determine whether chronic 28-day treatment of CBE (100 mg/kg) altered expression levels of cytoskeletal proteins. Whole tissue lysates (20–30 μg) from the striatum, substantia nigra, and motor cortex were homogenized and processed. Data indicate that levels of the cytoskeletal protein, α-tubulin, were decreased in the striatum and increased in the substantia nigra following CBE treatment (A). Levels of α-tubulin remained unchanged in the motor cortex (A). Levels of β-tubulin were decreased in the substantia nigra; however, they remained unchanged in the striatum and motor cortex (B). Levels of γ-tubulin remained unchanged following chronic CBE treatment (C). ODs of the individual bands were quantified using NIH ImageJ and normalized to GAPDH on the same gel. *p<0.05, unpaired t-test. n=3–5/group. Graphs are expressed as mean±SEM.

FIG. 6.

GCase inhibition caused accumulation of proteinase-K-resistant insoluble α-synuclein aggregates and induced autophagy–lysosomal proteins in the nigrostriatal pathway. Mice were treated with either vehicle or CBE and immunostained for insoluble α-synuclein using a proteinase-K digestion protocol. α-Synuclein insoluble aggregates were analyzed by size (μm²) using NIH ImageJ. CBE caused a fivefold increase in the number of aggregates between 1 and 10 μm² (A). CBE also promoted the accumulation of the larger insoluble aggregates (between 11–20, 21–50, 51–100, and >151 μm²) (A). Western blots were performed using total cell lysate (20–30 μg) from the substantia nigra and striatum. The chaperone-mediated autophagy receptor, lysosomal-associated membrane protein (LAMP)-2A, was increased in the substantia nigra, striatum, and motor cortex of CBE-treated mice (B). The lipidated form of LC3 (LC3-II) was also increased in the substantia nigra, striatum, and motor cortex of CBE-treated mice (C). The lysosomal protease, cathepsin D (CSTD), was elevated in the substantia nigra, striatum, and motor cortex of CBE-treated mice (D). The ubiquitin-like protein, p62, remained unchanged in all three brain regions of CBE-treated mice (E). ODs of the individual bands were quantified using NIH ImageJ and normalized to GAPDH on the same gel. *p<0.05, unpaired t-test. n=4/group. Graphs are expressed as mean±SEM.

FIG. 7.

GCase inhibition resulted in histopathological changes in mice. Stereological counts of neuronal cell bodies in cortical layers V and VI indicated that chronic (28 days) systemic treatment of CBE caused a reduction of NeuN-positive neurons (A). Stereological cell counts in the substantia nigra indicated that CBE did not alter the number of dopaminergic (TH+NeuN-labeled) neurons in the substantia nigra pars compacta or the number of NeuN-labeled neurons in the pars reticulata (B). There was also no effect of CBE treatment on dopaminergic striatal fiber density (C) *p<0.05, unpaired t-test. n=3–4/group. Graphs are expressed as mean±SEM.