The effect of mutant GBA1 on accumulation and aggregation of α-synuclein

Abstract

Gaucher disease (GD) patients and carriers of GD mutations have a higher propensity to develop Parkinson’s disease (PD) in comparison to the non-GD population. This implies that mutant GBA1 allele is a predisposing factor for the development of PD. One of the major characteristics of PD is the presence of oligomeric α-synuclein-positive inclusions known as Lewy bodies in the dopaminergic neurons localized to the substantia nigra pars compacta. In the present study we tested whether presence of human mutant GCase leads to accumulation and aggregation of α-synuclein in two models: in SHSY5Y neuroblastoma cells endogenously expressing α-synuclein and stably transfected with human GCase variants, and in Drosophila melanogaster co-expressing normal human α-synuclein and mutant human GCase. Our results showed that heterologous expression of mutant, but not WT, human GCase in SHSY5Y cells, led to a significant stabilization of α-synuclein and to its aggregation. In parallel, there was also a significant stabilization of mutant, but not WT, GCase. Co-expression of human α-synuclein and human mutant GCase in the dopaminergic cells of flies initiated α-synuclein aggregation, earlier death of these cells and significantly shorter life span, compared with flies expressing α-synuclein or mutant GCase alone. Taken together, our results strongly indicate that human mutant GCase contributes to accumulation and aggregation of α-synuclein. In the fly, this aggregation leads to development of more severe parkinsonian signs in comparison to flies expressing either mutant GCase or α-synuclein alone.

Introduction

Gaucher disease (GD) is a lysosomal storage disorder resulting from mutations in the GBA1 gene, which encodes the lysosomal enzyme acid-β-glucocerebrosidase (GCase), leading to its impaired activity. Due to the defective activity of mutant GCase, glucosylceramide (GlcCer) and glucosylsphingosine (GlcSph) accumulate, ultimately leading to lysosomal dysfunction (1–5).

Due to its heterogeneity, GD was divided into three types: type 1, primarily a non-neurological disease; type 2, the acute neuronopathic disease; and type 3, the sub-acute neuronopathic form (6).

Mutant GCase variants are recognized in the ER as misfolded and are retained there by the ER quality control machinery, in an attempt to correctly fold them. This ER retention leads to ER stress and eventually to activation of the ER stress response, known as the Unfolded Protein Response (UPR) and to ER-associated degradation (ERAD) (7). ERAD involves the retrotranslocation of mutant enzyme molecules through the ER translocons back to the cytosol, where they undergo ubiquitination and proteasomal degradation (8–10).

In recent years, a large number of studies established a link between GD and Parkinson’s disease (PD). Thus, GD patients and carriers of GD mutations have a higher risk of developing PD than the general population (11–13). GBA1-associated parkinsonism is characterized by an earlier onset and an increased likelihood of cognitive decline and dementia, compared with typical PD (14–16). Accumulation and aggregation of α-synuclein, the main component of Lewy bodies and Lewy neurites (17), was noted in GD-associated PD, both in animal models and in post-mortem analysis of brains from GD patients and carriers of GBA1 mutations who developed PD (18–21). Analysis of post-mortem frontal cortex tissue from GD carriers with Lewy body disease demonstrated a significantly reduced level of GCase protein and enzyme activity, as well as ER retention of mutant GCase (22). SDS-soluble α-synuclein oligomers were noted as well (23). Other studies noted decrease in GCase activity and protein levels in post-mortem of PD patients even without GBA1 mutations (20,24–28).

Several studies demonstrated that mutant GCase promotes α-synuclein accumulation in a dose-and time-dependent manner. Thus, Cullen et al. (18) observed that overexpression of human mutant GCase variants in neural MES23.5 and PC12 cells, stably expressing human α-synuclein, significantly raised human α-synuclein levels. The authors concluded that `mutant GCase promotes α-synuclein accumulation in a dose-and time-dependent manner, thereby demonstrating a biochemical link between GBA1 mutation carrier status and increased synucleinopathy risk’ [cited from Cullen (18)].

Using iPS-derived dopaminergic cells that originated from carriers of GD mutations who developed PD, Fernandes et al. recorded decreased GCase levels, elevated GlcCer amount, misprocessing of GCase, ER stress upregulation and autophagic/lysosomal dysfunction. The authors suggested that the combination of ER and autophagic/lysosomal disturbances impairs protein homeostasis in dopaminergic neurons, which leads to development of PD (29). In L444P heterozygous mice, which were knocked out for the endogenous mouse α-synuclein encoding gene and exogenously expressed human α-synuclein, the latter accumulated in hippocampal-derived cells in tissue culture. In parallel, the mice exhibited motor deficits (30). Kim et al. reported that heterozygous expression of the D409H mutation in human A53T α-synuclein transgenic mice, resulted in reduced GCase amount and activity, exacerbated death of dopaminergic cells, formation of insoluble α-synuclein aggregates, mitochondrial defects and glial activation which led to motor abnormalities and shortened the life span (31). A very recent publication has shown that in mice heterozygous for the L444P mutation, there was a reduction in general autophagy and mitochondrial priming, both in vivo in mouse brain tissue and in vitro in cultured hippocampal neurons. The findings were recapitulated in postmortem anterior cingulate cortical tissue from PD patients with heterozygous GBA1 mutations, suggesting that GD-carriers have mitophagy deficits that may contribute to PD-associated mitochondrial deficits (32).

We have previously suggested that the presence of misfolded mutant GCase contributes to the development of PD in carriers of GBA1 mutations. Development of parkinsonian signs could be recapitulated in transgenic flies expressing either the N370S or the L444P human GCase variants (10,33). These phenotypes could be partially rescued by growing the model flies in the presence of the pharmacological chaperone ambroxol (10,33). Two other groups documented the development of neurodegeneration (34) and of parkinsonian signs in flies (34,35) as a consequence of expression of human mutant GCase.

In the present study we aimed to further elucidate the possible contribution of mutant GCase to α-synuclein accumulation and aggregation. We examined the possible consequences of heterologous expression of mutant GCase in α-synuclein expressing cells. We used SHSY5Y cells stably expressing the N370S and the L444P human mutant GCases, as well as flies expressing both human α-synuclein and human mutant GCase variants in their dopaminergic cells. Half-life of α-synuclein in cells expressing mutant GCase was significantly elevated, which led to its accumulation and aggregation. Flies expressing human α-synuclein and human mutant GCase variants in their dopaminergic cells presented aggregated α-synuclein in their brain, premature death of dopaminergic cells and shorter life span compared with flies expressing human mutant GCase or α-synuclein alone. Our results highlight a possible role for mutant GCase in the turn-over and aggregation of α-synuclein and shed light on the involvement of mutant GCase in the development of PD.

Results

Stabilization of α-synuclein and GCase in SHSY5Y cells

We first aimed to test the effect of human mutant GCase on endogenous α-synuclein in human SHSY5Y neuroblastoma-derived cells (36), often used as a tissue culture model for dopaminergic cells (37). Using cycloheximide (CHX) chase analysis to inhibit de novo protein synthesis, the half-life of α-synuclein and of human mutant GCase was tested in cells stably expressing WT, N370S or L444P GCase (Fig. 1). The results demonstrated a significant stabilization of both α-synuclein and mutant GCase. The half-life of α-synuclein increased from 22 h in the SHSY5Y control cells and in the cells that expressed WT GCase (Fig. 1A–C) to 28 h in the presence of the N370S mild GCase mutant (Fig. 1A and D) and to 52 h in the presence of the L444P severe GCase mutant (Fig. 1A and E). Interestingly, the half-life of the different human mutant GCases increased as well (Fig. 1F–H). While WT GCase had a half-life of 40 hours, the N370S and the L444P GCases were significantly stabilized, making the half-life unmeasurable. The results strongly indicated that presence of mutant GCase leads to α-synuclein stabilization.

Figure 1

Mutual stabilization of α-synuclein and mutant GCase. Control SHSY5Y cells, or SHSY5Y cells stably expressing WT, N370S or L444P human GCase were incubated for different durations with CHX (100 μg/ml). Cell lysates were subjected to western blotting and the blots were interacted with anti-myc or anti-α-synuclein antibodies and with anti-erk antibodies as a loading control (A). (erk appears as two isoforms, erk 1 and erk 2) (69). Protein intensity was detected by chemiluminesnce and was quantified in control SHSY5Y cells (B) and in cells expressing WT (C), N370S (D) or L444P (E) GCase. The intensity of the different GCases was measured as well (F–H). The value obtained for α-synuclein or for GCase was normalized according to the amount of erk in the same lane. The value obtained for non-treated cells was considered as 1. The results presented in all the panels are the mean ± SEM of four different experiments.

Increased amounts of α-synuclein in cells expressing human mutant GCase

To record the levels of α-synuclein in cells expressing mutant GCase, western blot analysis was performed. The results showed that α-synuclein levels in SHSY5Y cells expressing WT human GCase did not significantly differ from those in control SHSY5Y cells. On the other hand, α-synuclein levels doubled and tripled in the presence of the N370S and L444P mutant GCases, respectively (Fig. 2A and B), in comparison to control SHSY5Y cells. To confirm the stabilization of α-synuclein in the presence of mutant GCase, immunofluorescence staining of the cells was performed. In SHSY5Y cells stably expressing the mutant N370S or L444P GCases, the amount of α-synuclein was significantly elevated in comparison to control SHSY5Y cells or cells expressing WT GCase (Fig. 2C and D). The results strongly indicated that α-synuclein accumulates in the presence of mutant GCase.

Figure 2

Expression of mutant GCase in SHSY5Y cells leads to α-synuclein accumulation. (A) Lysates of SHSY5Y cells or cells expressing different human GCases were subjected to western blot analysis using anti-α-synuclein antibodies following by anti erk antibodies as a loading control. (B) Quantification of α-synuclein in the blots presented in (A). The quantity of α-synuclein was normalized according to the amount of erk in each lane and the value obtained for SHSY5Y cells was considered as 1. The results represent the mean ± SEM of three independent experiments. (C) Representative confocal images of cells expressing WT, N370S or L444P human GCases and stained with anti-α-synuclein antibodies. (D) Quantification of α-synuclein signal intensity obtained from cells, as explained in Materials and Methods. Results represent the mean ± SEM of 80 different cells. ^*P < 0.05, ^***P < 0.005. Bar 10 μm.

Evidence for α-synuclein aggregation in cells expressing mutant GCase

Since α-synuclein accumulated in SHSY5Y cells expressing human mutant GCase, we further tested whether it also oligomerizes and aggregates in these cells. For this purpose, we used anti-oligomeric antibodies (A11) which recognize a structure-specific epitope of amyloidogenic proteins, including α-synuclein (38), and react with all the soluble aggregates, such as oligomers and protofibrils (39). Control cells or cells expressing WT human GCase or the mild N370S mutant did not present any visible oligomeric proteins. However, cells expressing the severe L444P GCase mutant presented a considerable amount of oligomeric proteins, most probably α-synuclein (Fig. 3A and B). The fact that only the L444P but not the N370S presented positive A11 staining may indicate low sensitivity of the antibody toward oligomeric α-synuclein forms.

Figure 3

Expression of mutant GCase leads to intracellular α-synuclein aggregation. (A) Representative confocal images of SHSY5Y cells expressing different human GCases stained with anti-oligomeric antibodies (A11). (B) Quantification of aggregated α-synuclein signal intensity obtained from cells as shown in (A). (C) Western blot analysis (using anti-α-synuclein antibodies) of lysates prepared from control SHSY5Y cells or cells expressing different GCases and treated with the indicated concentrations of PK. The same blot was interacted with anti-β tubulin antibodies to show the initial loading amounts. Loading controls in the PK-treated lanes are not shown for the treated samples since the PK treatment led to protein degradation, as expected. (D) Representative confocal images of control SHSY5Y cells or cells expressing different human GCases stained with anti Ser129 phosphorylated α-synuclein. (E) Quantification of phosphorylated α-synuclein signal intensity obtained from cells as presented in (D). Results represent the mean ± SEM of 80 different cells. ^***P < 0.005. P-α-synuclein: S129 phosphorylated α-synuclein, PK-proteinase K. Bar 10 μm.

Oligomeric forms of α-synuclein can be biochemically distinguished by their susceptibility to proteinase K (PK) digestion. PK-resistant α-synuclein accumulates in brains of PD patients and in animal models of PD, including mice and Drosophila (40,41). We treated lysates of SHSY5Y cells, expressing different human GCases, with increasing concentrations of PK and performed western blot analysis to assess the resistance of α-synuclein in these cells. In control cells or cells that expressed WT GCase, α-synuclein was completely degraded following treatment with 0.5 μg/μl PK. In contrast, in cells that expressed mutant N370S or L444P GCase, α-synuclein was present even after treatment with 1 μg/μl PK, strongly suggesting that in SHSY5Y cells expressing mutant GCases, α-synuclein is PK-resistant (Fig. 3C).

The most prevalent α-synuclein modification observed in Lewy body deposits in PD, as well as in dementia with Lewy bodies and multiple system atrophy, is phosphorylation on Ser129. This modification was indicated to affect α-synuclein solubility, membrane-binding properties and subcellular distribution (42). Based on the mentioned observations, we tested whether α-synuclein is Ser129 phosphorylated in SHSY5Y cells expressing human mutant GCase. The results (Fig. 3D and E) clearly indicated that the level of Ser129-phosphorylated α-synuclein was significantly higher in cells expressing mutant GCase, in comparison to control cells or cells expressing WT GCase. These results confirm that in the presence of human mutant GCase, α-synuclein is not only stabilized and forms oligomers but is also phosphorylated on Ser129, a hallmark of aggregated α-synuclein in Lewy bodies.

Cellular defects in SHSY5Y cells expressing mutant GCase

Actin was previously shown to colocalize with α-synuclein in rat neuronal CSM 14.1 cells and in SHSY5Y cells (43), and altered actin levels were found in Caenorhabditis elegans (44) and Drosophila PD models (45,46). In mouse neuroblastoma Neuro2A cells, α-synuclein interacted with actin, as tested by FRET analysis, slowed down actin polymerization and resulted in cytoskeletal defects. However, changes in actin amounts were not noted (47). We assessed cytoskeleton status in SHSY5Y cells expressing human mutant GCase by phalloidin staining (Fig. 4A). In cells expressing the severe L444P GCase mutant, phalloidin staining was significantly altered. Instead of the barbed-type appearance of actin, it appeared fragmented and dispersed with no characteristic cytoskeletal appearance.

Figure 4

Cellular defects in SHSY5Y cells expressing mutant GCase. Representative confocal images of SHSY5Y cells expressing human myc-WT, N370S or L444P GCases, stained with phalloidin which specifically binds to F-actin filaments. (B) Western blot analysis of lysates of SHSY5Y cells expressing different GCases using anti-ubiquitin antibodies. (C) Quantification of ubiquitin intensity in the blots as shown in (B). The quantity of ubiquitin was normalized according to the amount of erk in each lane and the value obtained for control SHSY5Y cells was considered as 1. The results represent the mean ± SEM of three independent experiments. (D) SHSY5Y cells expressing different GCases were stained with trypan blue. The total number of cells as well as the number of trypan blue positive cells were recorded, and percentage of dead cells was calculated. (E) Representative confocal images of cells stained with TUNEL (white; apoptotic cells) and DAPI (all cells). E. Quantification of TUNEL positive cells. Results represent the mean ± SEM of 80 cells. ^*P < 0.05, ^***P < 0.005. Bar 10 μm (A), 20 μm (E).

Figure 5

Aggregation of α-synuclein in Drosophila dopaminergic cells. (A, B) An illustration of the dopaminergic cells in the posterior region of the fly brain (A) [modified from (70)] and their staining with anti-TH antibodies which stain dopaminergic cells (B). (C, D) Representative confocal images of brains dissected from Ddc-GAL4 control flies (;;Ddc-GAL4/+), from Syn flies expressing human α-synuclein (;;Ddc-GAL4/UAS-α-synuclein), from Syn + WT flies epressing human α-synuclein and WT human GCase (;WT GCase/+;Ddc-GAL4/UAS-α-synuclein), from Syn + N370S GCase flies expressing human α-synuclein and the N370S mutant GCase (;N370S GCase/+;Ddc-GAL4/UAS-α-synuclein) or from Syn + L444P flies, expressing human α-synuclein and the human L444P mutant GCase (;L444P GCase/+;Ddc-GAL4/UAS-α-synuclein). Fixed brains were stained with anti-oligomeric A11 antibodies, that also recognize α-synuclein, at day 12 (C) or at day 22 (D) post-eclosion. Shown are Z-projections of confocal sections. (E) Quantification of aggregated α-synuclein signal intensity obtained from the images is shown in panels C (brains of 12 days old flies, blue columns) and D (brains of 22 days old flies, orange columns). (F, G) Representative confocal images of brains non treated (F) and treated (G) with PK at day 12 post-eclosion, and stained with anti-α-synuclein antibody. (H) Quantification of α-synuclein signal intensity obtained from the images as shown in (F) and (G). Results represent the mean ± SEM of 12 different brains. ^**P < 0.01, ^***P < 0.005, Syn: α-synuclein, Ddc-GAL4:;;Ddc-GAL4/+, Syn + GCase:;GCase/+;Ddc-GAL4/UAS-α-synuclein, PK-proteinase K. Bar 50 μm.

Figure 6

Death of dopaminergic cells in flies coexpressing α-synuclein and human mutant GCase. Representative confocal images of brains isolated from 12 days old flies not expressing (None) or expressing human WT, N370S or L444P human GCases without (A) or with (B) expression of α-synuclein, following fixation and staining with anti-TH antibodies. Shown are Z-projections of confocal sections. (C) Quantitative representation of the average number of cells in dopaminergic clusters. Results represent the mean ± SEM of 12 brains. Blue columns represent quantification of panels in (A), orange columns represent quantification of panels in (B). ^**P < 0.01, ^***P < 0.005. DA: dopaminergic cells. Bar 50 μm. Nomenclature of flies: Control flies: Ddc-GAL4 (genotype:;;Ddc-GAL4/+), Flies expressing α-synuclein: α-synuclein in (A) Syn in (B) (genotype:;GCase/+;Ddc-GAL4/UAS-α-synuclein), Flies expressing α-synuclein with either WT or N370S or L444P GCase: Syn + (WT or N370S or L444P) GCase (genotypes:;WT GCase/+;Ddc-GAL4/UAS-α-synuclein,;N370S GCase/+;Ddc-GAL4/UAS-α-synuclein,;L444P GCase/+;Ddc-GAL4/UAS-α-synuclein, respectively).

Similar to other neurodegenerative disorders, PD is characterized by impaired ubiquitin-proteasome system. Accordingly, increased ubiquitination of intraneuronal proteins was documented in PD-derived cells (48). We therefore tested whether there is a change in ubiquitination levels in SHSY5Y cells expressing human mutant GCases. The results (Fig. 4B and C) showed that in cells expressing mutant GCases, the amount of total ubiquitinated proteins was significantly higher than in control cells or in cells expressing WT GCase.

Since PD is associated with death of dopaminergic cells, we also tested whether stabilization and aggregation of α-synuclein, as well as the overall increase in the amount of ubiquitinated proteins, is accompanied by cell death. Trypan blue staining of SHSY5Y cells expressing different human GCases indicated that the relative amount of death was significantly higher in cells expressing the L444P mutant GCase compared with control cells or with cells expressing other GCases (Fig. 4D). To determine whether the cells presented an apoptotic death, Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay was performed (Fig. 4E). We found that cells expressing mutant GCases presented a significant increase in apoptotic cell death in comparison to control cells and cells expressing WT GCase (Fig. 4F). The fact that we observed a small percentage of apoptotic cells following the expression of the mild N370S mutant GCase, whereas no cell death was observed using trypan blue, reflects, most probably, the fact that the TUNEL assay detects preliminary stages of apoptotic cell death, namely DNA fragmentation (49).

Death of dopaminergic cells and aggregation of α-synuclein in flies expressing human α-synuclein and mutant GCase

We have previously shown that transgenic flies ectopically expressing mutant GCases develop parkinsonsian signs (10,33). Based on the aggregation of α-synuclein in the presence of mutant GCase, shown in the present study, we tested the effect of mutant GCase expression on the aggregation of exogenous α-synuclein in transgenic flies (the fly genome does bot contain an α-synuclein encoding gene) (50). Mutant human GCase and α-synuclein, coupled to the Upstream Activating Sequences (UAS) promoter, were activated in the brain of flies which expressed the Dopadecarboxylase (Ddc)-GAL4. The GAL4-UAS system is utilized to study targeted gene expression during the life cycle of the fly. GAL4 is a transcription factor identified in the yeast Saccharomyces cerevisiae. It regulates the transcription of genes by directly binding to UAS element, analogous to an enhancer element defined in multicellular eukaryotes. Utilizing the GAL4-UAS system, gene expression occurs strictly by GAL4 that stimulates transcription of genes under UAS element. The GAL4 gene can be under the control of different native gene promoters (driver genes). The absence of GAL4 in lines containing only UAS coupled to the gene of interest maintains the transgene(s) in a transcriptionally silent state. To activate their transcription, the UAS-regulated gene of interest strains are mated to flies expressing GAL4 and the resulting progeny then express the gene of interest in a transcriptional pattern that reflects the GAL4 pattern of the respective driver (51).

Brains of flies expressing human α-synuclein and human GCase variants in their dopaminergic cells, using the Ddc-GAL4 driver, were stained with the anti-oligomeric A11 antibodies at days 12 and 22 post-eclosion (Fig. 5A and B, respectively). The results indicated toxic oligomers in the brains of 12 days old flies expressing both the severe human L444P mutant GCase and α-synuclein in their dopaminergic cells (Fig. 4C and E). As expected, at day 22 post-eclosion, all flies expressing human α-synuclein in their dopaminergic cells presented toxic aggregates (Fig. 5D and E).

We further evaluated PK-resistance of α-synuclein in brains of 12 days old flies expressing human α-synuclein together with mutant GCases (Fig. 5F–H). Flies expressing α-synuclein alone, or α-synuclein together with WT or the mild N370S mutant, did not present any PK-resistant α-synuclein (Fig. 5G and H). However, flies expressing α-synuclein together with the severe L444P mutant showed a significant amount of PK-resistant α-synuclein in their brains (Fig. 5G and H).

To follow death of dopaminergic cells, TH expressing cells were counted at day 12 post-eclosion in the posterior region of the flies’ brains, where approximately 70 such cells are located (52). In flies expressing both α-synuclein and mutant GCase, the amount of dopaminergic cells was significantly lower in comparison to their amount in flies expressing mutant GCase or flies expressing α-synuclein alone (Fig. 6A and B). The results strongly indicated that expression of both α-synuclein and the severe L444P GCase mutant in the dopaminergic cells of the flies leads to devastating results.

Locomotor dysfunction and decreased survival in flies expressing human α-synuclein and mutant GCase

We have previously shown that expression of mutant GCases in dopaminergic cells leads to decreased locomotor activity, as tested by climbing assays (10,33). In the present study, we evaluated the climbing behavior of flies expressing both α-synuclein and mutant GCase in their dopaminergic cells at days 2, 12 and 22 post-eclosion, using the counter-current apparatus. Flies expressing both α-synuclein and the severe L444P or mild N370S GCase mutants presented significantly lower locomotor skills at day 12 or at day 22 post-eclosion, respectively, in comparison to flies expressing either protein alone or α-synuclein together with WT GCase (Fig. 7A). Flies expressing both α-synuclein and the severe L444P mutant survived only 38 days in comparison to flies that expressed either the severe L444P mutant or α-synuclein alone, which survived 45 or 53 days, respectively (Table, Fig. 7). Flies expressing the mild N370S together with α-synuclein survived 45 days, in comparison to flies expressing only the N370S mild GCase or flies expressing α-synuclein alone, which survived 48 and 53 days, respectively (Table, Fig. 7B). These results strongly indicated a synergistic deleterious effect of α-synuclein and mutant GCase in dopaminergic cells.

Figure 7

Parkinsonian signs in flies coexpressing α-synuclein and human mutant GCase. (A) Thirty control (Ddc-GAL4) or transgenic flies expressing different GCase variants, with and without α-synuclein, were analyzed for climbing behavior at 2, 12 and 22 days post-eclosion. Flies were grown at 10 flies per vial and were transferred to fresh food every other day. (B) Kaplan Meier curve showing the overall survival rates of 100 flies per line. Eight different lines were used with the genotypes denoted in the legend to Figure 6. (C) Table summarizing survival (in days) of the different fly strains. ^*P < 0.05, ^**P < 0.01, ^***P < 0.005.

Discussion

In recent years, increasing evidence has shown a link between GD and PD. In a previous work, we used Drosophila melanogaster to show that expression of human mutant GCase in the dopaminergic cells of flies leads to Parkinson-like disease (33). In the present study, we further analyzed the contribution of human mutant GCase to the accumulation and aggregation of α-synuclein, a key factor in PD pathology. Aiming to recapitulate the situation in carriers of GD mutations, we used two model systems, both comprising normal endogenous GCase activity, with no substrate accumulation, in parallel to the presence of a mutant GCase.

We found that in SHSY5Y cells endogenously expressing α-synuclein and stably expressing the N370S or the L444P human mutant GCases, the half-life of endogenous α-synuclein increased, leading to its accumulation in the cells (Fig. 1, A–E), as previously documented in several models (18,30,53). In parallel, the half-life of human mutant GCases was elevated as well (Fig. 1, F–H). The accumulated α-synuclein was phosphorylated at Ser129 (Fig. 3D) and PK-resistant (Fig. 3C), strongly indicating its aggregation.

We also tested the effect of human mutant GCase on α-synuclein in D. melanogaster flies expressing both human mutant GCase and human α-synuclein in their dopaminergic cells. Our results strongly indicated that there was an enhanced α-synuclein aggregation in brains of flies expressing human α-synuclein and the L444P mutant GCase, in comparison to flies expressing the WT or the N370S GCases together with α-synuclein, or α-synuclein alone (Fig. 5). Flies expressing α-synuclein and the severe L444P GCase mutant presented a significantly shorter life span in comparison to flies expressing only one of the proteins. The same was true for flies expressing the mild N370S GCase in concert with α-synuclein, in comparison to flies expressing each one of the proteins alone (Fig. 7). Taken together, our results strongly indicated that presence of mutant GCase has a pivotal effect on aggregation of α-synuclein, death of dopaminergic cells and survival of the flies. They also showed that a more severe GCase mutation, such as the L444P mutation, has a more pronounced effect on α-synuclein aggregation than the milder N370S mutation. These results are in accordance with previously published data showing that PD patients who are carriers of more severe GBA1 mutations develop a more severe PD, as tested by the age of PD onset, as well as motor, psychiatric, cognitive and olfactory symptoms (54,55).

As clearly stated in the introduction, the effect of mutant GCase on accumulation and aggregation of α-synuclein has been previously shown. However, this is the first study to show increased stability of both α-synuclein and mutant GCase in SHSY5Y cells, stably expressing mutant GCases. This is also the first documentation of the effect of mutant GCases on WT α-synuclein in flies.

Two different mechanisms have been proposed to explain GD-associated PD, a gain-of-function or a loss-of-function models. While we favor the gain-of-function theory, presented in this study, showing that presence of mutant GCase leads to accumulation of α-synuclein, another model assumes loss of GBA1 function (in heterozygotes it is more appropriate to be named haploinsufficiency), which causes some lysosomal substrate accumulation and leads to failure of autophagic pathways and to reduced disposal of α-synuclein (56). Several studies documented the contribution of substrate buildup to the aggregation of α-synuclein. Sardi et al. (21) showed that mice homozygous for the D409V GBA1 mutation presented hippocampal accumulation of α-synuclein-ubiquitin aggregates, PK-resistant α-synuclein and impaired memory. These mice did not accumulate GlcCer in their brain but showed increased levels of GlcSph. Hippocampal administration of recombinant AAV1 vector expressing WT human GCase reduced the amounts of GlcSph and α-synuclein-ubiquitin in the brain, and impaired memory deficits (21,57). In another attempt to reduce substrate accumulation in the homozygous D409V mice, animals were treated with the GlcCer synthase inhibitor GZ667161 (58). Treated mice showed less GlcSph, decreased α-synuclein-ubiquitin levels, and improved memory-associated deficits (58). In SHSY5Y cells treated for 10 days with the non-competitive GCase inhibitor conduritol-beta-epoxide (CBE) (59), we also documented a significant increase in the amount of α-synuclein (Supplementary Material, Fig. S1A–F), which also aggregated (Supplementary Material, Fig. S1G and H), in parallel to GlcCer accumulation (Supplementary Material, Fig. S1A and B). These results implicate that GlcCer accumulation leads to α-synuclein accumulation and aggregation. However, it is important to emphasize that no substrate accumulation was detected in brains of PD patients who were carriers of GD mutations (60) and there is no documentation of substrate accumulation in GD-carrier animals, which expressed human α -synuclein (30,31).

Degradation of α-synuclein occurs via both lysosomal and proteasomal pathways. In the proteasomal pathway, degradation of α-synuclein depends on its SIAH-mediated ubiquitination (61), while its lysosomal degradation depends on its autophagy and fusion of autophagosomes with lysosomes, where α-synuclein is degraded by cathepsin D (62). It is possible that the presence of mutant GCases, which undergo ERAD (9), attenuates the proteasomal degradation of α-synuclein. Defective proteasomal degradation (63) can lead to defective autophagy, as previously reported in iPSC-derived neurons from carriers of GD mutations (64) and defects in mitophagy (32).

In several publications, reduced GCase activity was documented in older individuals who developed sporadic PD (26,65). We suggest that under normal conditions, misfolded WT GCase molecules are refolded by the ER quality control machinery. However, at older ages, these quality control mechanisms are debilitated, leading to retention of more misfolded WT GCase molecules in the ER. This ER retention results in ER stress which activates the UPR, retards autophagy, mitophagy and causes dysregulated lysosomal activity, thus leading to α-synuclein accumulation, death of dopaminergic cells and development of PD. Such molecules can be removed from the ER using chemical chaperones (66).

To summarize, our results confirm previous studies and argue that the presence of mutant GCase both in SHSY5Y cells and in Drosophila dopaminergic cells leads to accumulation and aggregation of α-synuclein and to faster death of these cells, implying an important role for mutant GCases in the development of PD in carriers of GD mutations.

Materials and Methods

Cell lines

SHSY5Y human neuroblastoma cells (ATCC/CRL-2266) were grown in Dulbecco’s modified Eagle’s medium supplemented with 20% fetal calf serum, 100 U/ml penicillin–streptomycin, 1 mm sodium pyruvate and 2 mm L-glutamine (Biological Industries, Beit-Haemek, Israel), at 37°C in the presence of 5% CO2.

Generation of SHSY5Y cell lines stably expressing myc-His-GCase

SHSY5Y cells stably expressing myc-His-GCase variants were generated by co-transfecting SHSY5Y cells with pBABE-puro plasmid (Aldgene, Inc., Cambridge, MA, USA) and WT or mutant (N370S or L444P) myc-His-GCase, at 1:10 (w:w) ratio, respectively, using Lipofectamine 2000TM (Invitrogen), according to the manufacturer’s protocols. Forty-eight h later, cells were split and grown in a medium containing 0.8 μg/ml of puromycin (Sigma-Aldrich, Rehovot, Israel). Puromycin-resistant cultures were tested for myc-His-GCase expression by western blot analysis, using anti-myc antibody. These cells were previously described by us (67).

Fly strains and maintenance

All strains were grown on standard cornmeal-molasses medium at 25°C. Transgenic lines, harboring pUASTmycHisGCase, pUASTmycHisN370SGCase or pUASTmycHisL444PGCase on the second chromosome were established by Bestgene (Chino Hills, CA, USA). Transgenic flies harboring UAShWTαSyn (#8146) and Dopadecarboxylase (Ddc)-GAL4 (#7010) were obtained from Bloomington stock center (Bloomington, IN, USA).

Antibodies

Primary antibodies: mouse anti-myc and mouse anti-GFP (Cell Signaling Technology, Beverly, MA, USA), rabbit anti-α-synuclein (C20, Santa Cruz Biotechnology, Santa Cruz, CA, USA), mouse anti-actin (Sigma Aldrich, St. Louis, MO, USA), rabbit anti phosphorylated Ser129 α-synuclein (EP156Y, Abcam, Cambridge, UK), mouse anti-α-synuclein (BD Bioscience, Franklin lakes, NJ, USA), rabbit anti-amyloid oligomers (A11, Millipore, Billerica, MA, USA), rabbit anti-erk (Santa Cruz, Dallas, TX, USA), rabbit anti-TH (AB152, Millipore, Billeria, MA, USA), and rabbit anti-GlcCer antibodies (Glycobiotech, Kukels, Germany). Secondary antibodies used were: Horseradish peroxidase-conjugated goat anti-mouse, Horseradish peroxidase-conjugated goat anti-rabbit antibodies, Cy™2-conjugated goat anti-rabbit and Cy™3-conjugated goat anti-rabbit (all from Jackson Immuno Research Laboratories, West Grove, PA, USA).

CBE treatment

Cells were treated with 200 nm CBE (Sigma-Aldrich) for 10 days as described elsewhere (33).

SDS-PAGE and western blot

Cell lysates were prepared essentially as described elsewhere (33) and samples containing equal amounts of protein were electrophoresed through 10% SDS/PAGE and electroblotted onto a nitrocellulose membrane (Schleicher and Schuell BioScience, Keene, NH, USA). Membranes, treated as previously described (33) were incubated with the appropriate antibodies, and following washing were incubated with enhanced chemiluminescence detection reagent (Santa Cruz Biotechnology, Dallas, TX, USA) and analyzed using a luminescent image analyzer (ChemiDoc XRS+ System, Bio-Rad, Hercules, CA, USA).

CHX chase experiments

Cells were treated with CHX (100 μg/ml) to inhibit de novo protein synthesis. At the indicated times, cell lysates were prepared as outlined above, and equal amounts of proteins were subjected to western blot analysis using the appropriate antibodies.

PK digestion analysis

PK digestion in cells: cells were homogenized in PBS and centrifuged at 1000 g for 5 min at 4°C. The supernatants were collected and the samples were treated with PK at a final concentration of 0.1, 0.5 or 1 μg/ml for 30 min at 37°C. The reaction was stopped by addition of 5× sample buffer.

PK digestion in flies: protocol was modified from (41). Briefly, brains were dissected and digested with 1 μg/μl PK in PBS for 30 min at room temperature. After digestion, the samples were fixed with 4% paraformaldehyde for 60 min. Following rinsing with PBT (1× PBS supplemented with 0.3% Triton X-100), samples were reacted with anti α-synuclein antibody (1:50) diluted in BBT (1X PBS supplemented with 0.1% BSA, 0.1% Tween-20 and 250 mm NaCl) overnight, at 4°C with shaking. Following three washes with PBT, Cy™3-conjugated goat anti-mouse (1:250) secondary antibodies were added and incubated with shaking for 2 h at room temperature.

Immunofluorescence

Cells, grown on coverslips, and treated as previously described (67), were incubated for 1 h with the corresponding primary antibodies (1:500 for mouse anti-myc; 1:100 for rabbit anti-α-synuclein, 1:100 for phospho- α-synuclein and 1:100 for A11) in 1% BSA/PBS at RT. Cells were washed three times with PBS and then immunostained with Cy™²-conjugated goat anti-rabbit or Cy™³-conjugated goat anti-rabbit antibodies (1:250 dilution) in 1% BSA/PBS for 30 min at room temperature. Following three washes with PBS, the coverslips were mounted in VECTASHIELD hard set mounting medium with DAPI (Vector laboratories, Berligame, CA, USA).

For fly brains staining, adult brains were dissected in PBS and fixed with 4% paraformaldehyde for 60 min. Following rinsing with PBT (1X PBS supplemented with 0.3% Triton X-100), samples were reacted with anti-TH antibodies (1:80) or with anti amyloid oligomers (A11) antibodies (1:100), diluted in BBT (1X PBS supplemented with 0.1% BSA, 0.1% Tween-20 and 250 mm NaCl), overnight at 4°C with shaking. Following three washes with PBT, Cy™2-conjugated goat anti-rabbit or Cy™3-conjugated goat anti-rabbit (1:250) secondary antibodies were added and incubated with shaking for 2 h at room temperature. After several washes with PBT, the preparations were mounted in VECTASHIELD hard set mounting medium with DAPI (Vector laboratories, Berligame, CA, USA).

All slides were visualized using LSM510 Meta (ZEISS) confocal microscope. For quantitative studies, all images of a given experiment were exposed and processed identically. Captured images were analyzed using ImageJ software. Pixel intensity (in arbitrary units) was used to quantify fluorescence in the indicated experiments. Data was statistically evaluated using Student’s t-test.

Analysis of cell death

Cell death was evaluated by direct cell count using optic microscopy (Olympus) following staining with trypan blue (Biological industries, Beit Haemek, Israel). Total number of cells and number of stained (dead) cells were recorded.

Flies climbing assay

Climbing behavior of adult flies was measured using a countercurrent apparatus, essentially as described in (68). Briefly, groups of approximately 30 flies (both males and females) were given 5 min to adapt in the starting tube, which can slide along the apparatus, and then 20 s to move upwards against gravity to the upper frame’s tube. The top frame of tubes was then shifted to the right so that the start tube comes into register with a second bottom tube and flies, which successfully climbed up, were tapped down again, falling into tube 2. The upper frame was then returned to the left and the flies are once again allowed to climb into the upper tube. After five runs, the number of the flies in each tube was counted. For each time point, at least four cohorts from each genotype were scored. The Climbing Index (CI) was calculated using the following formula: CI (the weighted mean) = Σ (mnm)/N. CI is ranged from 1 (min) to 6 (max). Here: m – number of test vial, nm – number of flies in the mth vial, N – total number of flies.

TUNEL analysis

TUNEL assays were performed using the Apoptag Red In Situ Apoptosis Detection Kit (Millipore, Billerica, MA, USA), according to the manufacturer’s protocol. Briefly, cells were fixed in 4% paraformaldehyde for 15 min at 25°C and were then washed in PBS twice. Equilibrium buffer was added to the cells at the appropriate volume and then working strength TdT enzyme was added for 1 h at 37°C. Cells were washed in stop buffer and Anti-Digoxigenin-Fluorescein was added for 30 min at 25°C in the dark. The cells were washed three times with PBS and mounted in VECTASHIELD hard set mounting medium with DAPI (Vector laboratories, Berligame, CA, USA).

Survival assay

For each fly strain, 10 vials, each containing 10 flies, were maintained on food from day one post-eclosion. Fresh food was supplied every other day and deaths were recorded. Kaplan–Meier was used to plot survival using the XLSTAT 2017 software.

Funding

Buchmann fellowship for outstanding PhD students (to G.M.); the Israel Science Foundation (Grant 1300/13); Pfizer; SHIRE; NNE Research Program administered by TEVA (to M.H.).