Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP, M0896), ammonium chloride (NH₄Cl, A9434), Chloroquine diphosphate salt (CQ, C6628), 3-methyladenine (3-MA, M9281) and cycloheximide (CHX, C7698), proteinase K (P2308) were purchased from Sigma. Lactacystin (BML-P1104-0200) was purchased from Enzo. Allstars Negative Control siRNA (scramble, 1027280), siRNA for LAMP2 (1027416/GS3920) and siRNA for DJ-1 (1027416/GS11315) were purchased from Qiagen. SiRNA for LAMP1 (A10001) and scramble were purchased from GenePharma. Cell Premix Ex Taq (Til RNaseH Plus) was purchased from Takara (RR420B). A Lysosome Enrichment Kit for Tissue and Cultured Cells (89839) and Protein Extraction Reagent Kit (23225) were purchased from Thermo Scientific. Phenylmethylsulfonyl fluoride (PMSF, ST 506) and 3-(N-Morpholino) propanesulfonic acid (MOPS, ST302) was purchased from Beyotime Biotechnology. Protease and phosphatase inhibitor cocktails were purchased from Roche (4693124001/4693159001). Dulbecco’s Modified Eagle’s Medium (DMEM), fetal bovine serum (FBS) and penicillin and streptomycin (PS) were purchased from Gibco. Polyvinylidene fluoride (PVDF) membranes were purchased from Millipore. The antibodies for immunoblot analysis are listed as follows: DJ-1 (Abcam, ab18257, 1:2500), LAMP2A (Abcam, ab18528, 1:1000), HSC70 (Abcam, ab1427/ab2788, 1:1000), SQSTM1/p62 (abcam, ab56416, 1:1000), α-synuclein (BD, 610787, 1:1000), α-synuclein (Invitrogen, AHB0261, 1:1000), LAMP-1 (CST, 3243S, 1:1000), GAPDH (CST, 2118S, 1:2500), LC3B (CST, 3868S, 1:1000), COX IV (CST, 4805, 1:1000), c-Raf (CST, 12552S, 1:1000), actin (Sigma, A5441, 1:2500) and horseradish peroxidase-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories, 115-035-144/146, 1:10000). All other chemicals were from Sigma-Aldrich unless otherwise specified.

Human neuroblastoma SH-SY5Y cells were obtained from the American Type Culture Collection and cultured in DMEM plus 10% FBS and 100 U/mL PS at 37°C in a humidified incubator with 5% CO₂. For pharmacological studies, NH₄Cl, CQ, 3-MA, lactacystin and CHX were added at the indicated times and concentrations.

Human WT α-synuclein and WT DJ-1 cDNAs were cloned into a pcDNA3.1-flag vector (Gui et al., 2012) and pcDNA3-HA vector (Wang et al., 2011) respectively. Lipofectamine 2000 was used to transfect SH-SY5Y cells with plasmid DNA and/or siRNA in accordance with the manufacturer’s instructions. Briefly, the cells were cultured in a 6-well plate to 80%–90% confluence, and transfected with the indicated amounts of DNA and/or siRNA. Transfection efficiency was verified by Western blotting 48 h post-transfection.

Protein extraction and Western blotting were carried out as previously described (Jiang et al., 2015). The cells were lysed in RIPA buffer (50 mM Tris-HCl pH 8.0, 150 mM NaCl, 0.1% SDS, 0.5% sodium deoxycholate and 1% NP-40) supplemented with protease and phosphatase inhibitor cocktails and 1 mM PMSF for 30 min on ice. After centrifugation at 14,000 g for 15 min at 4°C, the protein concentration in the supernatants was determined with a BCA protein assay kit (Thermo Scientific, 23225). Equal amounts of protein (20 μg/lane) were separated by SDS-PAGE and then transferred to PVDF membranes. The membranes were incubated in blocking buffer for 1 h at room temperature and subsequently incubated with primary antibodies overnight at 4°C. The membranes were then incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature. Finally, the signals were detected by chemiluminescence (BIO-RAD ChemiDoc MP Imaging System) according to the manufacturer’s instructions.

Total RNA was extracted from cultured cells using TRIzol reagent (Invitrogen). Reverse transcription was carried out using PrimerScript Reverse Transcriptase. Real-time quantitative PCR was performed using Roche LightCycler^® 480 system with SYBR Green reagents (Takara). The results were determined by the 2^−ΔΔCt method to compare gene expression with GAPDH expression as a loading control. Primer sequences were as follows: LAMP2A forward: 5′-GCACAGTGAGCACAAATGAGT-3′; reverse: 5′-CAGTGGTGTGTATGGTGGGT-3′. GAPDH forward: 5′-GCACAGTGAGCACAAATGAGT-3′, reverse: 5′-CAGTGGTGTGTATGGTGGGT-3′.

Lysosomes from human neuroblastoma SH-SY5Y cells were isolated with Lysosome Enrichment Kit for Tissue and Cultured Cells (Thermo Scientific, 89839) following manufacturer’s instructions. The kit uses OptiPrep Cell Separation Media for the density-based separation of lysosomes from contaminating cell structures. After three washes with PBS, the cultured cells were homogenized with extraction buffer from a Lysosome Enrichment Kit. Then, the lysosomes were isolated by density gradient centrifugation (145,000× g for 2 h). For the assessment of cyt-HSC70, 30 μl volume of the incubation buffer (10 mM MOPS, pH 7.3, 0.3 M sucrose) supplemented with protease and phosphatase inhibitor cocktails and 1 mM PMSF was added into the freshly isolated intact lysosomes, and then proteinase K (5 μl of a 1 mg/ml stock in 1 mM CaCl₂, 50 mM Tris-HCl, pH 8) was added into the incubation buffer for 10 min at 0°C (Kaushik and Cuervo, 2009). After that, 1 ml volume of incubation buffer supplemented with protease and phosphatase inhibitor cocktails and 1 mM PMSF were added and centrifuged all samples at 25,000 g for 5 min at 4°C (Kaushik and Cuervo, 2009). Aspirate the supernatant fractions and wash the pellet fractions twice with incubation buffer to eliminate any protein bound non-specifically to the lysosome surface (Kaushik and Cuervo, 2009). Finally, the lysosomes were lysed in RIPA buffer, assayed by Western blotting for the levels of LAMP1, c-Raf and COX IV and used in the following experiments. A total of 5 μg of protein in lysosome-enriched fractions was loaded into each lane according to the protein concentration measured.

DJ-1 knockout mice (kindly provided by Dr. Jie Shen, Harvard Medical School) were back-crossed with C57BL/6 mice at least 6 times. The mice were housed 5 per cage in a controlled environment under 22–25°C with 40%–60% relative humidity, and a 12 h light/dark cycle with free access to food and water. The study was approved by the Ethics Committee of Ruijin Hospital affiliated to Shanghai Jiao Tong University School of Medicine, and all procedures were carried out in accordance with the guidelines of the laboratory animal ethical standards of Shanghai Jiao Tong University School of Medicine.

DJ-1 knockout and WT littermate male mice (8–10 weeks old) used in the experiments received MPTP as previously described (Jackson-Lewis and Przedborski, 2007). In brief, the mice were treated with MPTP (30 mg/kg/day, intraperitoneally) once a day for five consecutive days, while control mice were injected with saline only (Jackson-Lewis and Przedborski, 2007). At the indicated times (0, 4, 7, 14 and 21 days) after the last MPTP injection, the mice were sacrificed. The ventral midbrain tissue was isolated, weighed and then stored at −80°C until used as described (Jackson-Lewis and Przedborski, 2007).

The procedures were carried out as previously described (Roberts et al., 2015) with minor modifications. For the sequential extraction, the following buffers were used: (1) TBS buffer: 50 mM Tris-HCl pH 7.4, 175 mM NaCl and 5 mM EDTA; (2) TBST buffer: TBS with 1% Triton X-100; (3) RIPA buffer (SDS buffer): 50 mM Tris-HCl pH 7.4, 175 mM NaCl, 5 mM EDTA, 1% NP-40, 0.5% sodium deoxycholate and 0.1% SDS; and (4) Urea buffer: 8 M urea/5% SDS. Ventral midbrain tissue was homogenized in sequence with the four sequential extraction buffers listed above and supplemented with cocktails and PMSF to generate TBS, TBST, SDS and urea-soluble fractions. First, the sample was homogenized in twenty volumes of TBS buffer on ice. The homogenate was centrifuged for 5 min at 1000 g at 4°C, and then the supernatant was ultracentrifuged for 30 min at 120,000 g at 4°C. The resulting supernatant was the TBS-soluble fraction. Second, the pellet was washed twice in TBS buffer, resuspended by sonication in TBST buffer and centrifuged for 20 min at 120,000 g at 4°C. The resulting supernatant was the TBST-soluble fraction. Third, the pellet was washed twice, resuspended by sonication in RIPA buffer and centrifuged for 20 min at 120,000 g at 4°C. The resulting supernatant was the SDS soluble fraction. Fourth, the pellet was washed twice with TBS buffer at room temperature and then resuspended by sonication in urea buffer for the urea-soluble fraction.

All values are expressed as the mean ± SEM. Data were from three or more independent experiments. The relative intensities were normalized to internal loading control and further normalized to the control levels. Statistical analysis was performed using Prism 5 (GraphPad Software). Statistical significance was determined by one-way analysis of variance (ANOVA) and two-way ANOVA with Bonferroni’s post hoc test. Values of p < 0.05 were considered statistically significant.

To explore the underlying molecular mechanisms, we established a cellular system transiently co-transfected with siRNA against DJ-1 and WT α-synuclein plasmid into SH-SY5Y cells, and the transfection efficiency was verified by Western blotting (Supplementary Figure S1). In agreement with previous research (Shendelman et al., 2004), Western blotting analysis confirmed that DJ-1 deficiency markedly accelerated α-synuclein accumulation (Figures 1A–C), while DJ-1 overexpression inversely inhibited α-synuclein accumulation in SH-SY5Y cells (Figures 1D–F).

Furthermore, we studied the relationship between DJ-1 and α-synuclein in PD animal models (Figures 2A–H). In this experiment, we established a MPTP mouse model of PD in DJ-1 knockout and WT littermate mice to determine the effect of DJ-1 deficiency on α-synuclein aggregation in vivo. First, WT mice received free base MPTP 30 mg/kg daily for five consecutive days and were sacrificed at different times to analyze the changes in α-synuclein levels in ventral midbrain regions. Consistent with previous research (Vila et al., 2000), α-synuclein total tissue protein peaked at day 4 after MPTP administration and then returned to control level until 21 days (Supplementary Figure S2). Second, to further determine whether the effect of DJ-1 deficiency on α-synuclein aggregation with or without MPTP administration, we analyzed α-synuclein levels in DJ-1 knockout and WT mice at day 4 after administration (Figures 2A–H). In agreement with the previous results (Yamaguchi and Shen, 2007), there was no significant difference between the two groups that received salinein all the fractions (Figures 2A–H), whereas soluble and insoluble α-synuclein oligomers in DJ-1 knockout mice that received MPTP increased more prominently in the SDS and urea-soluble fractions compared with those in WT control mice (Figures 2E–H). Taken together, our results showed that DJ-1 deficiency aggravated accumulation and aggregation of α-synuclein in both SH-SY5Y cells and PD animal models.

The proteolytic pathways of a given protein are determined by cell type, cellular conditions, or activity of the pathway (Park and Cuervo, 2013). To determine whether WT α-synuclein can be degraded by the lysosomal pathway in our cellular system, we treated SH-SY5Y cells with NH₄Cl or CQ (inhibitors of lysosomes) at different concentrations. Endogenous α-synuclein levels increased after administration with NH₄Cl or CQ (Figures 3A–D), whereas administration with lactacystin (a proteasome inhibitor) in SH-SY5Y cells failed to upregulate α-synuclein levels (Figures 3G,H), suggesting the lysosomal pathway is the primary proteolytic pathway for endogenous α-synuclein but not proteasome. In accordance with a previous study (Cuervo et al., 2004), α-synuclein was mainly degraded in lysosomes by CMA but not macroautophagy. Because endogenous α-synuclein levels were augmented when LAMP2 was knocked down in SH-SY5Y cells but not LAMP1, a known marker for lysosome (Figures 3I–N), whereas 3-MA at low concentrations (an inhibitor of macroautophagy) had little effect on the levels of α-synuclein (Figures 3E,F). The transfection efficiency of siRNA against LAMP2 and LAMP1, and the role of 3-MA in macroautophagy were determined by Western blotting (Supplementary Figure S3). These results suggested that CMA is the primary degradation pathway for endogenous WT α-synuclein in our cellular system and the accumulation of α-synuclein in DJ-1-deficient cells may result from the inhibition of CMA-dependent degradation of α-synuclein.

Since LAMP2A acts as both the receptor and essential components of the CMA translocation complex, LAMP2A levels at the lysosomal membrane are directly proportional to CMA activity (Schneider and Cuervo, 2013; Patel and Cuervo, 2015). Mild WT α-synuclein overexpression upregulated the total cellular levels of LAMP2A, but this upregulation was inhibited in DJ-1-deficient cells (Figures 4A,B). Meanwhile, the downregulation of LAMP2A was accompanied by an upregulation of α-synuclein in DJ-1-deficient cells (Figure 4A). Therefore, DJ-1 deficiency may accelerate α-synuclein accumulation by inhibiting the upregulation of LAMP2A under the condition of mild WT α-synuclein overexpression. To further confirm the role of DJ-1 deficiency in LAMP2A levels at the lysosomal membrane, we examined the levels of LAMP2A in the lysosomal fractions isolated from SH-SY5Y cells in the same cellular system (Figures 4E,F). The purity of lysosomal fractions was verified by Western blotting for LAMP1 (the lysosome marker), c-Raf (the cytoplasmic marker) and COX IV (the mitochondria marker) in both homogenate and lysosome-enriched fractions (Supplementary Figure S4). Consistent with the total whole cell lysate results, DJ-1 deficiency suppressed the increase in LAMP2A levels induced by mild WT α-synuclein overexpression in the lysosomal fraction (Figures 4E,F). In addition, there were no significant differences in LAMP1 levels among the groups (Figures 4A,D), suggesting a selective effect of DJ-1 on LAMP2A.

We then analyzed the effect of DJ-1-deficiency on HSC70 levels of in our established cellular system, because only HSC70-positive lysosomes are capable of performing CMA (Cuervo et al., 1997). WT α-synuclein overexpression slightly upregulated the levels of cyt-HSC70, but there was no obvious difference (Figures 4A,C), while DJ-1 deficiency strongly decreased the levels of lys-HSC70 compared with those in the control group (Figures 4G,H). Since HSC70 is a constitutively expressed chaperone protein and lys-HSC70 accounts for only a small proportion, the increase in cyt-HSC70 is not informative regarding CMA activity (Agarraberes et al., 1997; Patel and Cuervo, 2015). Therefore, DJ-1 may play an important role in the activation of CMA induced by mild WT α-synuclein accumulation and DJ-1 deficiency may decrease CMA-mediated degradation of α-synuclein by downregulating the key CMA components.

Since LAMP2A is the rate-limiting step in the CMA pathway and it is tightly regulated by de novo synthesis and degradation at the lysosomal membrane (Kaushik and Cuervo, 2012), we thus investigated LAMP2A reduction in DJ-1-deficient cells in the above mentioned contexts. DJ-1 deficiency upregulated LAMP2A mRNA levels, suggesting that LAMP2A reduction induced by DJ-1 deficiency was not due to de novo synthesis inhibition (Figure 5A). We then evaluated whether LAMP2A downregulation was due to enhanced degradation of LAMP2A by the protein synthesis inhibitor CHX to exclude the effect of translation. Compared with the control, DJ-1 deficiency notably increased the rate of degradation for LAMP2A (Figures 5B,C). In line with previous research (Cuervo and Dice, 2000a), our results showed that LAMP2A was degraded in lysosomes but not the ubiquitin-proteasome system (Figures 5D–I). In brief, DJ-1 deficiency may aggravate the accumulation of α-synuclein by accelerating the degradation of LAMP2A in lysosomes.