GBA mutation promotes early mitochondrial dysfunction in 3D neurosphere models

Abstract

Glucocerebrosidase (GBA) mutations are the most important genetic risk factor for the development of Parkinson disease (PD). GBA encodes the lysosomal enzyme glucocerebrosidase (GCase). Loss-of-GCase activity in cellular models has implicated lysosomal and mitochondrial dysfunction in PD disease pathogenesis, although the exact mechanisms remain unclear. We hypothesize that GBA mutations impair mitochondria quality control in a neurosphere model.We have characterized mitochondrial content, mitochondrial function and macroautophagy flux in 3D-neurosphere-model derived from neural crest stem cells containing heterozygous and homozygous _N370SGBA mutations, under carbonyl cyanide-m-chlorophenyl-hydrazine (CCCP)- induced mitophagy.Our findings on mitochondrial markers and ATP levels indicate that mitochondrial accumulation occurs in mutant _N370SGBA neurospheres under basal conditions, and clearance of depolarised mitochondria is impaired following CCCP-treatment. A significant increase in TFEB-mRNA levels, the master regulator of lysosomal and autophagy genes, may explain an unchanged macroautophagy flux in _N370SGBA neurospheres. PGC1α-mRNA levels were also significantly increased following CCCP-treatment in heterozygote, but not homozygote neurospheres, and might contribute to the increased mitochondrial content seen in cells with this genotype, probably as a compensatory mechanism that is absent in homozygous lines.Mitochondrial impairment occurs early in the development of GCase-deficient neurons. Furthermore, impaired turnover of depolarised mitochondria is associated with early mitochondrial dysfunction.In summary, the presence of GBA mutation may be associated with higher levels of mitochondrial content in homozygous lines and lower clearance of damaged mitochondria in our neurosphere model.

Keywords: Gaucher's disease; Parkinson's disease; autophagy; mitochondria; neurospheres.

Figure 1

Characterization of neurosphere development by analysis of β-III tubulin, MAP2 and nestin markers. Neuronal markers β-III tubulin and MAP2 were determined by Western Blot (A). Both neural markers showed stabilized expression on day 4 of development. β-III tubulin and MAP2 levels confirmed neural properties of neurospheres (B). Increasing nestin mRNA levels confirmed neural stem cell properties of neurospheres during development (C). No significant differences were found at 4 day of development AU, arbitrary units. Results are expressed by mean± SEM.

Figure 2

GCase, HEX and β-gal enzymatic activities stabilized from 0 to 6 days of differentiation. GCase activity decreased as expected for the respective genotypes (33% decrease in wt/_N370SGBA and 98% decrease in _N370SGBA/_N370SGBA) when measured at pH 5.4 in the presence of sodium taurocholate. Measurement of GCase at pH 4.5 in the absence of the GCase activator sodium taurocholate yielded similar results, while HEX and β-gal were unaffected. Results are expressed by mean± SEM.

Figure 3

Confirmation of mitochondrial uncoupling with CCCP in the neurosphere model. Representative blots of OPA1 isoforms. Expected changes from long to short OPA1 isoforms were observed upon uncoupler treatment in all the genotypes, as mitochondria undergo depolarisation/fission following mitochondrial uncoupling (24h, 10 μM CCCP) (A and B). TOM20 levels progressively decrease from 0 to 24h in a control line (C). L-OPA1, long isoform of optic atrophy 1 protein OPA1; S-OPA1, short isoform of OPA1; UT, untreated.

Figure 4

Mitochondrial content under basal conditions in control, wt/ _N370SGBA and _N370SGBA/_N370SGBA neurospheres was assessed by western blotting for the mitochondrial proteins VDAC1 (outer mitochondrial membrane), SDHA (inner mitochondrial membrane), and TFAM (matrix), normalized by β-actin content. All markers tended to increase in _N370SGBA/_N370SGBA neurospheres in basal conditions, when compared to control lines (A) (SDHA was significant). As expected in control lines, protein levels of VDAC1, SDHA and TFAM and ATP levels decreased following CCCP treatment for 24 hours (B and C). Despite an apparent increase in mitochondrial mass in untreated conditions, ATP levels remained unchanged in wt/_N370SGBA and _N370SGBA/_N370SGBA neurospheres (C). Mitochondrial markers of wt/_N370SGBA did not show a marked response to CCCP uncoupling, in fact, TFAM significantly increased in the heterozygous lines and the decrease in ATP levels were not so pronounced (B and C) and ATP levels of homozygous _N370SGBA/_N370SGBA neurospheres showed a null response to CCCP uncoupling. All of the above data suggest that mitochondrial function may be impaired in GBA mutant neurospheres. AU: arbitrary units. Thin line indicates p value <0.05. Results are expressed by mean± SEM.

Figure 5

Steady-state mRNA levels of TFEB and PGC1α during uncoupling treatment. TFEB mRNA levels tended to decrease in wt/_N370SGBA, and _N370SGBA/_N370SGBA but PGC1α significantly increased upon CCCP induction in wt/_N370SGBA, probably in an attempt to compensate for the underlying mitochondrial defects. UT, untreated. AU: arbitrary units. Thin line indicates p value <0.05. Results are expressed by mean± SEM.

Figure 6

Autophagy flux measurement. Macroautophagy flux showed trends to further increase in the control lines. p62 and LC3-2 levels are similar in the 3 untreated cell lines, however, when autophagy was blocked under bafilomycin treatment, p62 and LC3-2 levels tended to increase the most in controls. The lack of significant differences suggests that preexistent mitochondrial impairment may account for the pathogenesis of GBA mutant neurospheres. AU, arbitrary units; UT, untreated; BAF, bafilomycin. Results are expressed by mean± SEM.

Figure 7

Global summary of the observed molecular parameters in GBA mutant neurospheres. Compared to the controls, dysfunctional mitochondrial accumulation occurs in mutant GBA neurospheres under basal conditions, as shown by the accumulation of healthy and altered mitochondria, represented as elongated and rounded structures, respectively. Mitochondrial elimination is correct under mitophagy induction (CCCP) in the control line, as observed by the decrease of ATP, represented by the battery, and the macroautophagy induction, represented by the garbage bin. There is an impaired turnover of depolarized mitochondria under induced mitophagy in GBA neurospheres. This is not a result of impaired autophagosome formation and degradation, but at an earlier stage, where defective mitochondrial dysfunction was observed.