Parkin-mediated ubiquitination of mutant glucocerebrosidase leads to competition with its substrates PARIS and ARTS

The following antibodies were used in this study: monoclonal anti-GCase 2E2 (Abnova), generated against a peptide corresponding to amino acids 146–236 of human GCase, rabbit polyclonal anti-GCase, generated against a peptide corresponding to amino acids 517–536 of human GCase (G4171, Sigma), rabbit anti-ERK (C16 Santa Cruz Biotechnology, Santa Cruz, CA, USA), mouse monoclonal anti-actin (MP Biomedicals, OH, USA), mouse monoclonal anti-myc and anti-GFP (Cell Signaling Technology, Beverly, MA, USA); Secondary antibodies: horseradish peroxidase-conjugated goat anti-mouse or goat anti-rabbit (Jackson Immuno Research Laboratories, West Grove, PA, USA), rabbit anti-caspase 3 and rabbit anti-cleaved caspase 9 (Cell Signaling Technology, Beverly, MA, USA).

Carbobenzoxy-L-leucyl-L-leucyl-L-leucinal (MG132), cyclohexamide (CHX), Leupeptin and phenylmethylsulfonyl fluoride (PMSF) were purchased from Sigma-Aldrich (Rehovot, Israel). Four-methyl-umbelliferyl-glucopyranoside (4-MUG) was purchased from Genzyme Corp (Cambridge, MA, USA). Nonidet P-40 (NP-40) was purchased from Roche Diagnostics (Mannheim, Germany). Absolute Blue qPCR SYBR Green ROX Mix was from TAMAR Laboratory Supplies (Mevaseret Zion, Israel).

Human primary skin fibroblasts cell lines are described in Table 1. Fibroblasts and SHSY5Y human dopaminergic cells were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 20% fetal calf serum (FCS), 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. SHSY5Y stably expressing different human GCase variants were described elsewhere [36].

Construction of myc-His-GCase plasmids was described elsewhere [36]. pSC2-6myc ARTS was described in [34]. MISSION shRNA plasmid (TRCN0000000285) encoding parkin-targeted shRNA was purchased from Sigma-Aldrich (St Louis, Mo, USA).

Total RNA was isolated using the EZ-RNA kit (Biological Industries, Beit Haemek, Israel), according to the manufacturer’s instructions.

Two micrograms of RNA were reverse transcribed with M-MLV reverse transcriptase (Promega corporation, CA, USA), in the presence of 1 μg oligo-dT primer in a total volume of 20 μl, at 42°C for 60 minutes. Reactions were stopped by incubation at 70°C for 15 minutes. One-two microliters of the resulting cDNA were amplified by quantitative real-time PCR.

One microliter of cDNA was used for real-time PCR. PCR was performed using the KAPA SYBR Fast Universal qPCR kit (Kapa Biosystems, Wilmington, MA, USA) in a Rotor-Gene 6000 (Corbett life sciences, Valencia, CA, USA). The reaction mixture contained 50% qPCR mix, 300 nM of forward primer (5′-ATCTGAAGGAGCAACATCTGG-3′) and 300 nM of reverse primer (5′-CACGGGCGAGTTTACTATGTAG-3′), in a final volume of 10 μl. Thermal cycling conditions were: 95°C (10 minutes), 40 cycles of 95°C (10 seconds), 60°C (20 seconds) and 72°C (20 seconds). Relative gene expression was determined by Ct value.

Cell monolayers were washed three times with ice-cold phosphate-buffered saline (PBS) and lysed at 4°C in 500 μl of lysis buffer (10 mM HEPES pH 8.0, 100 mM NaCl, 1 mM MgCl₂ and 1% Triton X-100) containing 10 μg/ml aprotinin, 0.1 mM PMSF and 10 μg/ml leupeptin. Lysates were incubated on ice for 30 minutes and centrifuged at 10,000 g for 15 minutes at 4°C. Samples, containing the same amount of protein, were electrophoresed through 10% SDS-PAGE and electroblotted onto a nitrocellulose membrane (Schleicher and Schuell BioScience, Keene, NH, USA). Membranes were blocked with 5% skim milk and 0.1% Tween-20 in Tris-buffered saline (TBS) for 1 hour at room temperature (RT) and incubated overnight with the primary antibody. The membranes were then washed three times in 0.1% Tween-20 in TBS and incubated with the appropriate secondary antibody for 1 hour at RT. After washing, membranes were reacted with ECL detection reagents (Santa Cruz Biotechnology Inc., CA, USA) and analyzed by luminescent image analyzer (X-OMAT 2000 Processor, Kodak, Rochester, NY, USA).

SHSY5Y cells were transfected using either a MP-100 Microporator (Digital Bio Tech, Seoul, South Korea) according to the manufacturer’s instructions, or Lipofectamine 2000™ (Invitrogen, CA, USA).

Subconfluent skin fibroblasts were treated overnight with 25 μM MG-132, after which cells were washed 3 times with ice-cold PBS and lysed at 4°C in 1 ml of lysis buffer (10 mM Hepes pH = 8, 100 mM NaCl, 1 mM MgCl₂, and 0.5% NP-40) containing 10 μg/ml aprotinin, 0.1 mM PMSF and 10 μg/ml leupeptin (Sigma-Aldrich, Rehovot, Israel). Following centrifugation at 10,000 g for 15 minutes at 4°C, the supernatants were pre-cleared for 2 hours at 4°C with protein A-agarose (Roche Diagnostic, Mannheim, Germany). Following washes with 1 ml of lysis buffer containing protease inhibitors, proteins were eluted for 10 minutes at 100°C with 5X loading buffer, electrophoresed through 10% SDS-PAGE and blotted. The corresponding blot was interacted with the appropriate antibodies.

Sub-confluent skin fibroblasts were treated overnight with 25 μM MG-132, after which stringent immunoprecipitation conditions were applied as previously described [37], with some modifications. The medium was aspirated and the cells were harvested and lysed in 100 μl of denaturing buffer (1% SDS, 50 mM Tris pH7.4, 140 mM NaCl). Following vigorous vortexing, the lysates were immediately boiled for 10 minutes, cleared by centrifugation at 10,000 g for 10 minutes and diluted 10 fold into buffer containing 2% Triton X-100, 50 mM Tris pH7.4, 140 mM NaCl. After additional centrifugation to remove any insoluble material, immunoprecipitation was carried out with monoclonal anti-GCase antibody. Following three washes (5% sucrose, Tris pH7.5, 50 mM NaCl, 0.5% NP40, EDTA 5 mM) the ubiquitin-GCase conjugates were resolved through SDS-PAGE and the corresponding blots were interacted with polyclonal anti-GCase antibodies.

Cells were treated for 24 hours with 25 μM MG132, after which they were processed according to the experiment performed (western blotting with or without immunoprecipitation).

The blots were scanned using Image Scan and the intensity of each band was measured by the Image Master 1DPrime densitometer (both from Amersham Pharmacia Biotech, Buckinghamshire, England). To quantify the results, the intensity of the tested protein at each lane was divided by the intensity of the loading control (actin or ERK) in the same lane. The value obtained for non-transfected cells or WT cells was considered as 1.

We have already shown that the L444P mutant GCase variant undergoes ubiquitination [38]. In order to confirm ubiquitination of other mutant variants, GCase-containing complexes were immunoprecipitated from lysates that originated from fibroblasts of GD patients, separated through SDS-PAGE and the corresponding blot was interacted with anti-GCase antibodies. As presented in Figure 1A and B, there was a visible ubiquitination of severe mutant GCase variants, while ubiquitination of the N370S mutation was non visible, as expected for a mild mutation [2, 39]. It is worth mentioning that the N370S homozygous cells were derived from a mild patient (patient no. 8, Table 1).

We have shown in the past that parkin and mutant GCase, expressed in heterologous systems, interact with each other. Moreover, parkin mediates polyubiquitination and proteasomal degradation of mutant, but not normal transfected GCase [36]. In the present study, we aimed at confirming the interaction between endogenous parkin and mutant GCase. In order to do so, we first set out to demonstrate the expression of parkin in skin fibroblasts (Figure 1C). At the next stage, parkin-containing complexes were immunoprecipitated from lysates prepared from GD skin fibroblasts and from normal skin fibroblasts, after which the corresponding blot was interacted with anti-GCase antibody. The results presented in Figure 1D and E show that mutant GCase, but not its normal counterpart, interacted with parkin.

To confirm that parkin not only interacts with endogenous GCase but also mediates its degradation, the effect of overexpression of normal or mutant parkin (T240R ligase dead parkin) was tested in skin fibroblasts that originated from GD patient, homozygous for the L444P mutation, or in normal skin fibroblasts. The results, presented in Figure 2A and B, indicated that overexpression of normal, but not mutant, parkin in cells that originated from GD patient led to a significant decrease in GCase level. However, the level of GCase in normal skin fibroblasts was not affected by parkin overexpression. These results strongly suggest that parkin mediates degradation of mutant but not normal GCase.To further confirm that parkin mediates degradation of mutant GCase, normal skin fibroblasts or GD-derived skin fibroblasts were transfected with shRNA directed against endogenous parkin. As shown in Figure 2C, down-regulation of parkin led to stabilization of mutant GCase in GD-derived fibroblasts originated from patient but not in control fibroblasts, indicating that parkin mediates degradation of mutant GCase.

As shown thus far, parkin interacts with mutant GCase and mediates its degradation. We showed in a previous study that parkin, expressed in HEK293 cells, mediates K48 polyubiquitination and proteasomal degradation of mutant, but not normal, transfected GCase [36]. Since PD is characterized by death of dopaminergic neurons in the substantia nigra pars compacta region of the brain, and accumulation of pro-apoptotic parkin substrates have been documented, we wondered whether a competition exists between known pathogenic parkin substrates and mutant GCase.

One of the substrates of parkin is PARIS (PArkin Interactiong Substrate, ZNF746). PARIS undergoes parkin-dependent polyubiquitination and proteasomal degradation. Non-ubiquitinated PARIS shuttles to the nucleus where it serves as a major transcriptional repressor of PGC1α (peroxisome proliferator-activated receptor gamma [PPARγ] coactivator-1α). The latter is a master regulator of mitochondrial biogenesis through induction of NRF1 (nuclear respiratory factor-1) gene expression [33]. One of the targets of NRF1 is the ATPase5β gene that encodes the beta subunit of complex V of the oxidative phosphorylation machinery in the mitochondria (see Figure 3A). PARIS was shown to accumulate in the brains of PD patients, with no concomitant PARIS mRNA increase, indicating regulation of PARIS at the protein level [33].

Based on the above, and on our results showing that mutant GCase is a parkin substrate, we chose to test whether transcription of PGC1α, NRF1 and ATPase5β is repressed in GD-derived fibroblasts. The results, presented in Figure 3B, show a significant decrease in mRNA level of the three tested genes. Our results indicated the same decrease in mRNA level of PGC1α, NRF1 and ATPase5β in human dopaminergic neuroblastoma-derived cells in tissue culture (SHSY5Y), stably expressing mutant GCase [36] in comparison to their level in the same cells expressing normal GCase. These results indicated that the presence of mutant GCase leads to decreased transcription of genes regulated by PARIS, suggesting that a competition exists between PARIS and mutant GCase. To confirm these results, we tested whether increasing amounts of GFP-tagged PARIS in SHSY5Y cells, stably expressing normal or mutant GCase, leads to stabilization of mutant but not normal GCase. The results, presented in Figure 4, showed that with elevated amounts of PARIS in the cells, the level of the N370S mutant GCase increased 2-fold and the amount of the L444P mutant GCase variant increased 4-fold in comparison to the normal human GCase, expressed under the same competitive conditions.

To summarize, our results imply that mutant GCase and PARIS compete on the availability of parkin. Thus, presence of mutant GCase leads to accumulation of cytoplasmic PARIS and down-regulation of genes associated with mitochondrial biogenesis, which may lead to cell death.

Another parkin substrate is the pro-apoptotic protein ARTS (Apoptosis Related protein in TGFβ Signaling pathway, Sept4_i2) [34]. ARTS is a resident of the mitochondrial outer membrane, but upon apoptotic stimuli it rapidly translocates to the cytoplasm where it binds to and inhibits XIAP (E3 ubiquitin ligase X-linked inhibitor of apoptosis), leading to caspase activation and cell death [40, 41]. Accumulation of ARTS in cells renders them more susceptible to apoptosis [40, 42‐44]. We tested a possible competition between ARTS and mutant GCase in dopaminergic cells. SHSY5Y cells, stably expressing normal or the N370S mutant GCase, were transfected with increasing amounts of ARTS-expressing plasmid. The results (Figure 5A and B) strongly indicate that overexpression of ARTS leads to accumulation of the N370S mutant but not of normal GCase.

Translocation of ARTS from the mitochondria to the cytosol following apoptotic stimuli leads to cleavage of caspase 3 and caspase 9 [42]. We have shown that mutant GCase competes with ARTS on the availability of parkin. Therefore, upon apoptotic stimuli and in the presence of mutant GCase, ARTS should accumulate, inducing cleavage of caspase 3 and 9. To test this assumption, cleavage of caspase 3 and 9 was assayed in staurosporine-stimulated SHSY5Y cells stably expressing normal or N370S mutant GCase. The results, presented in Figure 6A and B, indicate that the apoptotic stimuli was followed by an increase in the amount of cleaved caspase 3 and caspase 9 in SHSY5Y cells expressing N370S mutant GCase in comparison to SHSY5Y cells expressing normal GCase. We also tested whether this staurosporine-induced cleavage of caspase 3 and caspase 9 as well as another stimulus (hydrogen peroxide) lead to apoptosis. As presented in Figure 6C, SHSY5Y cells stably expressing the N370S or the L444P mutant GCase were more susceptible to apoptotic stimulus than SHSY5Y cells stably expressing the normal GCase.

Our results strongly indicate that mutant GCase leads to sensitization of cells to apoptotic stimuli and death, involving caspase cleavage.