Glucocerebrosidase inhibition causes mitochondrial dysfunction and free radical damage

Abstract

Mutations of the gene for glucocerebrosidase 1 (GBA) cause Gaucher disease (GD), an autosomal recessive lysosomal storage disorder. Individuals with homozygous or heterozygous (carrier) mutations of GBA have a significantly increased risk for the development of Parkinson's disease (PD), with clinical and pathological features that mirror the sporadic disease. The mechanisms whereby GBA mutations induce dopaminergic cell death and Lewy body formation are unknown. There is evidence of mitochondrial dysfunction and oxidative stress in PD and so we have investigated the impact of glucocerebrosidase (GCase) inhibition on these parameters to determine if there may be a relationship of GBA loss-of-function mutations to the known pathogenetic pathways in PD. We have used exposure to a specific inhibitor (conduritol-β-epoxide, CβE) of GCase activity in a human dopaminergic cell line to identify the biochemical abnormalities that follow GCase inhibition. We show that GCase inhibition leads to decreased ADP phosphorylation, reduced mitochondrial membrane potential and increased free radical formation and damage, together with accumulation of alpha-synuclein. Taken together, inhibition of GCase by CβE induces abnormalities in mitochondrial function and oxidative stress in our cell culture model. We suggest that GBA mutations and reduced GCase activity may increase the risk for PD by inducing these same abnormalities in PD brain.

Fig. 1

CβE treatment led to a reduction in ADP phosphorylation. Glutamate (complexes I, III and IV) and succinate-dependent (complexes II, III and IV) ADP phosphorylation was significantly impaired following incubation with CβE for 20 days onwards. Ascorbate/TMPD-dependent (complex IV) ADP phosphorylation was not significantly altered. Solid bars: control; diagonal line bars: 1 days incubation; Dotted bars: 10 days incubation; horizontal striped bars: 20 days incubation; open bars: 30 days incubation. Activities (mean ± SEM) were corrected by cell number, measured from 3 experimental repeats. One-way ANOVA analysis (followed by Dunnett post test) was performed on the data and showed that both glutamate and succinate-linked activities were decreased significantly from 20 days onwards (^∗p < 0.05, ^∗∗p < 0.01).

Fig. 2

CβE treatment led to a progressive fall in mitochondria membrane potential (Ψ_m) and no change in mitochondrial content. (A) Deflections in the system are demonstrated by the use of rotenone (cross-hatched bar; ‘+rot’) and oligomycin (diagonal shaded bar; ‘+oligo’); ^∗∗p < 0.001. Data (mean ± SEM) for each time point where 195 cells from 5 coverslips representing 3 independent preparations were analysed, and is expressed as percentage changes (open bars) to the untreated control (solid bar). 50 μM CβE produced a significant and progressive fall in Ψ_m from 10 days (^∗p < 0.01 after ANOVA followed by Dunnett post test analysis). (B) Representative TMRM fluorescence images of untreated and SHSY-5Y cells treated with CβE for 30 days showing disruption of the mitochondrial network. (C) Image analysis reveals that both the circularity and (D) aspect ratio of the mitochondria upon CβE treatment are significantly increased (^∗p < 0.0001 by t-test), suggesting the mitochondria are more fragmented (solid bars: control; open bars: treated cells. (E) Levels of porin and ANT were measured in SHSY-5Y cells incubated with 50 μM CβE for 30 days (lanes 4–6) by Western blot analysis, against untreated cells (lanes 1–3) Blot (i) upper panel: β-actin; lower panel: ANT. Blot (ii) upper panel: β-actin; lower panel: porin. Graph (F) shows that there was no significant effect on porin or ANT levels (mean ± SEM, n = 3) by this treatment, (black bars, control; open bars: treated cells).

Fig. 3

CβE treatment led to increased free radical production and reduced aconitase activity. (A) As a positive control, paraquat pre-treatment at 300 μM for 1 day (cross-hatched bar; ‘+PQ’) significantly increased the rate of DHE oxidation, as analysed by paired t-tests (p = 0.002). Continuous treatment of SHSY-5Y cells with 50 μM CβE (open bars) showed a progressive increase in the rate of DHE oxidation after 20 days (^∗p < 0.01, mean percentage changes with respect to the untreated control (solid bar) ±SEM) as determined by one-way ANOVA followed by Dunnett post test. (B) Aconitase activity was significantly reduced (to 37% of control) after 30 days of CβE treatment (mean ± SEM, ^∗∗p < 0.001).

Fig. 4

CβE treatment led to increased alpha-synuclein. Levels of alpha-synuclein were measured in SHSY-5Y cells incubated with 50 μM CβE for 30 days by Western blot analysis, against untreated cells. (A) Shows the blots of untreated (lanes 1–3) or CβE-treated (lanes 4–6) SHSY-5Y cells, stained with anti-alpha-synuclein (lower panels) or anti-β-actin (upper panels). (B) Shows the graphical representation of alpha-synuclein levels from the scanned blots, normalised to β-actin levels (solid bar: control, open bar: treated cells). The level of alpha-synuclein protein was significantly (^∗p < 0.002; n = 3) higher (by 59%) than that of control levels.

Fig. 5

CβE treatment did not affect Lamp 1 levels. Levels of Lamp 1 were estimated by Western blotting of control cells (lanes 1–3) and treated ones (lanes 4–6) with anti-Lamp 1 and normalising with anti-β-actin. (A) Shows the blot and (B) shows a graphical representation of the result (solid bar: control, open bar: treated). No significant difference was seen between the two groups (n = 3).

Fig. 6

CβE treatment did not significantly alter basal LC3-II levels. LC3-II levels were estimated for treated and control cells. CβE did not affect the basal levels of LC3-II. (A): Shows Western blots of control and CβE cells (blot (i): without bafilomycin; blot (ii): with bafilomycin); control (lanes 1–3) or CβE (lanes 4–6). (B): Shows a graphical representation of the result (solid bars: control, open bars: CβE treated). The levels did not change significantly after CβE treatment.