TMEM175 deficiency impairs lysosomal and mitochondrial function and increases α-synuclein aggregation

Significance

Genetic studies have identified potential risk factors that have helped elucidate biological pathways involved in the development of Parkinson disease (PD). However, the majority of genome-wide association studies-implicated loci contain multiple genes requiring further investigation to establish the causality of any nominated gene to the pathology of associated disease. Here, we provide functional evidence that deficiency of the lysosomal K⁺ channel transmembrane protein 175, discovered under the third most-significant PD genome-wide association study peak, is critical for pathogenesis of PD by impairing lysosomal and mitochondrial function.

Abstract

Parkinson disease (PD) is a neurodegenerative disorder pathologically characterized by nigrostriatal dopamine neuron loss and the postmortem presence of Lewy bodies, depositions of insoluble α-synuclein, and other proteins that likely contribute to cellular toxicity and death during the disease. Genetic and biochemical studies have implicated impaired lysosomal and mitochondrial function in the pathogenesis of PD. Transmembrane protein 175 (TMEM175), the lysosomal K⁺ channel, is centered under a major genome-wide association studies peak for PD, making it a potential candidate risk factor for the disease. To address the possibility that variation in TMEM175 could play a role in PD pathogenesis, TMEM175 function was investigated in a neuronal model system. Studies confirmed that TMEM175 deficiency results in unstable lysosomal pH, which led to decreased lysosomal catalytic activity, decreased glucocerebrosidase activity, impaired autophagosome clearance by the lysosome, and decreased mitochondrial respiration. Moreover, TMEM175 deficiency in rat primary neurons resulted in increased susceptibility to exogenous α-synuclein fibrils. Following α-synuclein fibril treatment, neurons deficient in TMEM175 were found to have increased phosphorylated and detergent-insoluble α-synuclein deposits. Taken together, data from these studies suggest that TMEM175 plays a direct and critical role in lysosomal and mitochondrial function and PD pathogenesis and highlight this ion channel as a potential therapeutic target for treating PD.

Genome-wide association studies (GWAS) have identified a large number of loci that alter the risk of developing Parkinson disease (PD) (1). Given the nature of idiopathic PD as a complex, multifaceted disease, these non-Mendelian variants typically result in small increases or decreases in disease risk. However, the causal variants under the majority of GWAS peaks have not been definitively identified (2, 3). Characterizing the biological function of genes located in GWAS-implicated regions in the context of molecular and cellular development of PD may help to identify pathways that are critical in the etiology of the disease.

Several genes associated with PD (both Mendelian and GWAS) function in the lysosomal degradation pathway. Lysosomal degradation serves as a key final step to resolve protein aggregation in the process of macroautophagy and chaperone-mediated autophagy, and mutations that impair this process have been suggested to cause α-synuclein aggregation and downstream cellular toxicity. Mutations in the lysosomal hydrolase glucocerebrosidase (GBA) substantially increase the risk of PD (4, 5) and have been reported to compromise lysosomal degradation and increase α-synuclein aggregation (6). Mutations in lysosomal type 5 P-type ATPase (ATP13A2), a lysosomal ATPase, cause early-onset Parkinsonian syndrome (7) and have also been shown to cause lysosomal impairment. ATP13A2 mutations are linked to impaired acidification, decreased proteolytic processing, reduced degradation of lysosomal substrates, diminished lysosomal-mediated clearance of autophagosome, and neurotoxicity via the accumulation of α-synuclein (8, 9). Vacuolar protein sorting-associated protein 35 (VPS35) deficiency, also linked to familial PD (10), has been shown to perturb the maturation step of cathepsin D (CTSD) by increasing mannose-6-phosplate receptor turnover, resulting in accumulation of α-synuclein in the lysosomes (11). VPS35 gene deletion in dopamine neurons has also been reported to result in accumulation of α-synuclein and dopamine neuron loss (12, 13).

Mitophagy, the selective autophagy of mitochondria, also requires lysosomal degradation (14) and has also been implicated in the pathogenesis of PD. Phosphatase and tensin homolog deleted on chromosome 10-induced putative kinase 1 (PINK1), a mitochondrial ubiquitin kinase, and parkin, an E3 ubiqutin ligase, together initiate ubiquitin-mediated autophagosome formation around mitochondria (15–18). Mutations in genes encoding these proteins cause early-onset parkinsonian syndrome (19, 20). Mutations in F-box protein 7 that also cause early-onset autosomal recessive PD (21, 22) were recently shown to act on the parkin-mediated mitophagy by directly interacting with PINK1 and parkin (23). Although there are varying degrees of α-synuclein pathology in PINK1 or parkin-associated PD cases, reduced respiratory capacity and increased susceptibility to oxidative stress by the impairment of mitochondria quality control was demonstrated to lead to dopaminergic neuronal damage (24, 25). Additionally, overexpression of PINK1 or parkin can protect cells against toxicity associated with reduced proteasome function induced by synuclein aggregation (26, 27). Although it remains to be determined whether dysregulation of mitophagy also contributes to sporadic PD, maintenance of mitochondrial health is clearly linked to the molecular pathology of PD.

Human transmembrane protein 175 (TMEM175), recently identified from a lysosomal proteome (28), is a K⁺ channel located in late endosomes and lysosomes (29). A PD GWAS meta-analysis (1) identified a highly significant risk loci on chromosome 4, which covers several genes, including TMEM175. Identification of the candidate gene residing in this peak could be critical, not only to understanding the pathogenesis of PD but also in identifying a novel genetically validated therapeutic target for treating the disease. If in fact TMEM175 is the candidate gene, then as a K⁺ channel there is a high potential for druggability and there is a tractable therapeutic strategy. TMEM175 appears a likely candidate, as it has been shown to regulate lysosomal membrane potential, pH stability, and organelle fusion via potassium conductance on lysosomal and endosomal membranes (29). To help resolve a potential link between TMEM175 and PD, we have characterized the effects of TMEM175 in the context of lysosomal and mitochondrial function, as well as the susceptibility to α-synuclein aggregation in both neuroblastoma and primary neuron cellular systems. We show that TMEM175 deficiency impairs lysosomal degradation, lysosome-mediated autophagosome clearance, and mitochondrial respiratory capacity. These cellular defects result in increased neuronal susceptibility to exogenously applied α-synuclein fibrils, a well-studied model of α-synuclein uptake, processing, and degradation. Our findings suggest that TMEM175 loss-of-function mutations likely impair autophagy-mediated degradation of α-synuclein oligomeric species and, moreover, that variation in TMEM175 may be in part responsible for the linkage of the chromosome 4:951,947 GWAS peak to PD risk.

Results

TMEM175 Deficiency Decreased Lysosomal Catalytic Activity in Neuronal Cells via Destabilized Lysosomal pH.

TMEM175 was identified as a K⁺ channel that mediates potassium conductance on lysosomal and endosomal membranes, and thus regulates lysosomal membrane potential and pH in RAW 246.7 murine macrophages (29). To determine if TMEM175 serves a similar function in cells of neuronal origin, knockout of TMEM175 (KO) was engineered in the SH-SY5Y neuroblastoma cell line using a CRISPR/Cas9 approach (Fig. 1A). Under fed conditions, in media containing 10% (vol/vol) serum, lysosomal pH of TMEM175 KO SH-SY5Y was slightly more acidic than WT. Upon starvation in Earle’s Balanced Salt Solution (EBSS) for 3 h, a significant increase in lysosomal pH was observed, whereas WT SH-SY5Y cells maintained lysosomal pH in both fed and starved conditions (Fig. 1B). This result agrees with previous findings in TMEM 175 KO macrophage, and is expected given universal expression of TMEM175 in lysosomes throughout multiple tissues. Protein levels of lysosomal aspartyl protease, CTSD, and cysteine protease, cathepsin B (CTSB), were significantly decreased in TMEM175 KO cells relative to WT both in fed and starved conditions (Fig. 1D), suggesting lysosomal dysfunction. Faster migration of lysosome-associated membrane glycoprotein 1 (LAMP1) was also observed in TMEM175 KO cells relative to WT potentially by deglycosylation or proteolytic cleavage that could lead to their instability (Fig. 1D, second row) (30, 31). TMEM175 KO did not change the number of LAMP1⁺ lysosomes (Fig. 1C). Further evaluation of the activities of these lysosomal enzymes revealed that CTSD, CTSB, and GBA, another lysosomal hydrolase linked to PD, were significantly decreased by 20–30% in TMEM175 KO compared with WT, as indicated by the smaller rate constant (k) (Fig. 1 E–G). Starvation did not further amplify the degree of enzyme activity reduction between WT and KO. This finding would suggest that additional mechanisms (beyond impaired pH regulation) may play a role in the reduced enzyme activity resulting from TMEM175 KO. Consistent with decreased activities of proteases, the rate of l-Homopropargylglycine (HPG)-labeled protein degradation was decreased in KO relative to WT (Fig. 1H).

Fig. 1.

Unstable lysosomal pH and reduced lysosomal enzyme activity in TMEM175 KO cells (A) Western blotting and quantitative real-time PCR (qRT-PCR) showing levels of TMEM175 in SH-SY5Y WT and TMEM175 KO cells. (B) Lysosomal pH of WT and KO was measured in fed (white) and starved (stv, black, EBSS 3 h) conditions (n = 6). (C) Number of LAMP1⁺ lysosomes determined by immunohistochemistry and shown relative to WT (n = 5–7). (D) Protein levels of lysosomal CTSD and CTSB were quantified by Western blotting (Left) and shown normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Right, faster migrating LAMP1 in the second row is indicated with red asterisk) (n = 4). (E–G) Lysosomal enzyme activity of CTSD, CTSB, and GBA was determined by the relative fluorescence from enzyme activities of crude lysosomal fractions of WT and KO cells in fed (black) and starved (red) conditions (n = 3–4). Relative fluorescence as a function of time was fit to first order reaction with rate constants, k, for each group indicated on the right. (H) Relative intensity of HPG-labeled proteins after incorporation of HPG was plotted as a function of time and fit to one phase exponential decay (n = 3–5). The rate constant k is indicated on the right. Data are given in relative units (RU). Data presented are mean + SEM. Two-way ANOVA in B, D, E–H; Bonferroni's test in C, *P < 0.05, **P < 0.01, and ***P < 0.001 WT vs. KO.

We confirmed these results in primary rat hippocampal neuronal cultures. Transient knockdown (KD) of TMEM175 in primary neurons by siRNA (Fig. S1A) led to unstable pH compared with control siRNA, consistent with the above result in SH-SY5Y (Fig. S1B). TMEM175 knockdown did not affect the number of LAMP1⁺ lysosomal counts (Fig. S1C). Depletion of TMEM175 also significantly decreased enzyme activity of CTSD, CTSB, and GBA by 20–35% relative to controls (Fig. S1 E–G), even though only CTSB protein level was decreased in TMEM175-depleted rat hippocampal neurons (Fig. S1D). This result could be because of a transient knockdown in this model instead of complete KO of TMEM175 in SH-SY5Y. The degradation rate of HPG-labeled proteins was also lower in TMEM175-depleted cells compared with controls (Fig. S1H). Taken together these results indicate that TMEM175 plays critical role in maintaining lysosomal pH and proper lysosomal enzyme function in neuronal cells.

Fig. S1.

Unstable lysosomal pH and impaired lysosomal enzyme activity in TMEM175-depleted rat hippocampal neurons. (A) Levels of TMEM175 mRNA from rat hippocampal cells treated with control siRNA (siCTL) or TMEM175 targeting siRNA (siTM175) were measured by qRT-PCR (n = 20). (B) Lysosomal pH was measured from rat hippocampal neurons treated with control siRNA (siCTL) or TMEM175 mRNA targeting siRNAs (siTM175) in fed (white) and starved (stv, black, EBSS 4 h) conditions (n = 6). (C) Number of LAMP1⁺ lysosomes determined by immunohistochemistry and shown relative to WT (n = 6). (D) Protein levels of lysosomal CTSD and CTSB were determined by Western blotting and presented normalized to GAPDH (n = 4; Right). (E–G) Lysosomal CTSD (E), CTSB (F), and GBA (G) activity were measured in fed (black) and starved (red) conditions (n = 3). (H) Relative intensity of HPG-labeled proteins after incorporation of HPG was plotted as a function time and fit to one-phase exponential decay. Rate constant k is indicated on the right (n = 4–5). Data are given in relative units (RU). Data presented are mean + SEM. Student t test in A; two-way ANOVA in B and E–H, Mann–Whitney test in C. *P < 0.05, **P < 0.01, and ***P < 0.001. siCTL vs. siTM175.

TMEM175 Deficiency Impairs Clearance of Autophagosome After Accelerated Autophagosome–Lysosome Fusion.

One major function of lysosomes as cellular catalytic machinery is to degrade autophagy cargos through fusion with autophagosomes at the final step of autophagy. Digestion of engulfed materials is important for recycling amino acids and preventing the accumulation of aggregated proteins and aged organelles (14, 32). To determine if TMEM175 deficiency affects the process of autophagy, we evaluated the fusion of autophagosome to lysosome. To monitor the autophagosome, a RFP–GFP–LC3 construct was transiently transfected into TMEM175 KO SH-SY5Y cells. The signal of autophagosome-associated LC3 under fed conditions was evaluated. In this assay, pH-sensitive GFP signal (Fig. 2, green) is quenched by the acidic environment of lysosomes, although RFP signal (Fig. 2, red) is pH-insensitive. Thus, autophagosomes (AP, not fused to lysosomes) are marked by both GFP and RFP, whereas autophagolysosome (APL, fused to lysosomes) are marked only by RFP (Fig. 2A). The number of GFP puncta was significantly decreased in TMEM175 KO cells relative to WT cells, whereas the number of the total puncta mildly decreased, resulting in an overall decreased GFP (AP) and increased RFP-only (APL) puncta relative to total puncta (Fig. 2 B and C). This result suggests that a larger portion of autophagosomes are fused to lysosomes in TMEM175 KO cells, consistent with the previous finding that deficiency of TMEM175 leads to accelerated fusion of autophagosomes to lysosomes in RAW246.7 macrophage (29). To further determine whether TMEM175 deficiency affects the completion of autophagy, clearance of autophagy substrates was assessed. The endogenous level of microtubule-associated protein 1A/1B-light chain 3-II (LC3-II), which increases upon initiation of autophagy and is degraded as a part of autophagic cargo, was evaluated at multiple time points by Western blotting or immunocytochemistry in the absence and the presence of bafilomycin A. Without bafilomycin, autophagosome LC3 formation upon starvation (LC3-II/I ratio or LC3 puncta) peaked at 1 h and decreased afterward in WT, whereas there was a delayed degradation of LC3 in TMEM175 KO cells (Fig. 2 D and E and Fig. S2A), resulting in a significant increase in LC3 levels in TMEM175 KO cells relative to control at 3 and 5 h after starvation (Fig. 2 F and G, Upper). Adding bafilomycin resulted in accumulation of LC3 over time after the initiation of starvation to the same degree in both WT and TMEM175 KO cells, eliminating the difference in LC3 levels between WT and TMEM175 KO cells (Fig. 2 D and E and Fig. S2A) (+BafA1). This result indicates that reduction of LC3 in WT was indeed because of lysosomal degradation, and that inefficient lysosomal clearance of autophagosome but not autophagosome biogenesis was the cause of increased LC3 levels in TMEM175-deficient cells. Similarly, in rat primary hippocampal neurons, the autophagy substrate sequestosome1 (p62), as well as LC3 levels, were increased in TMEM175-depleted cells relative to control hippocampal neurons in both fed and starved conditions by immunocytochemistry (Fig. 2 F and G, Lower). Protein levels determined by Western blotting confirmed these findings (Fig. S2B), suggesting accumulation of autophagosomes. Given that the number of total autophagosomes is mildly decreased in TMEM175-depleted cells, increase levels of autophagosome substrates suggests that completion of autophagy is stalled because of impaired lysosomal degradation.

Fig. 2.

TMEM175 deficiency leads to accelerated autophagosome–lysosome fusion and impairs clearance of autophagosome. (A) Schematic describing autophagosome–lysosome fusion assay with the RFP- and GFP-tagged LC3 construct transfected into WT and TMEM175 KO cells. (B) The number of green and total LC3 puncta from WT and KO cells was quantified (n = 5). (C) Ratio of GFP puncta or RFP only puncta to total is quantified (n = 5). (D and E) Level of autophagy substrate LC3 was measured at indicated time points by Western blotting in WT and KO cells with or without bafilomycin A1 (100 nM). Quantified ratio of LC3-II to LC3-I relative to time = 0 is shown (n = 4). Extra space between LC3-I and LC3-II was cropped to improve the clarity of images. (F and G) Autophagosome-associated LC3 puncta [red, Top, 5-h starvation, SH-SY5Y; Middle, 4-h starvation, rat hippocampal neurons treated with control (siCTL) or TMEM175 targeting (siTM175) shRNA], and p62 puncta (green, Bottom, rat hippocampal neurons) in fed and starved conditions were assessed by immunohistochemistry. Black and white contrast images of LC3 were highlighted with puncta quantified (yellow) (F). Quantified intensity of each puncta is plotted (n = 5) (G). (Scale bars, 10 μm.) Data presented are mean + SEM. Two-way ANOVA in B, E, G; Mann–Whitney test in C, *P < 0.05, **P < 0.01, and ***P < 0.001. WT vs. KO or siCTL vs. siTM175.

Fig. S2.

TMEM175 deficiency results in accumulation of autophagy substrate by the inefficient degradation in the lysosome. (A) Autophagosome-associated LC3 puncta from SH-SY5Y WT and KO cells were assessed by immunohistochemistry at the indicated time points after starvation with or without bafilomycin A1 (100 nM, ±BafA1). Puncta intensity was quantified (Right) (n = 4). (Scale bar, 10 μm.) (B) Protein levels of p62, LC3-I, and LC3-II at fed and starved conditions were quantified from rat hippocampal neurons treated with control siRNA (siCTL) or TMEM175 mRNA targeting siRNAs (siTM175) by Western blotting and values were normalized to GAPDH (n = 4) (Right). LC3-II* indicates a longer exposure. Data presented are mean + SEM. Two way ANOVA, *P < 0.05, **P < 0.01 and ***P < 0.001. WT vs. KO or siCTL vs. siTM175.

TMEM175 Deficiency Decreases Mitochondrial Respiration.

Autophagy regulates mitochondrial turnover through multiple mechanisms of mitophagy, which establish energetic homeostasis and prevent mitochondrial aging in response to environmental cues (14, 33). Inhibition of autophagy can lead to decreased mitochondrial respiration through accumulation of dysfunctional mitochondria and deficits in metabolic substrates (34, 35). To determine whether decreased clearance of autophagosomes in TMEM175-deficient cells also affects mitochondrial energetic capacity, we measured mitochondrial oxygen consumption rate (OCR). There were significant decreases in basal as well as maximal mitochondrial OCR in TMEM175 KO SH-SY5Y cells relative to WT cells (Fig. 3 A and B). Maximal OCR was determined from the difference between carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP)-induced OCR and Oligomycin A-induced OCR. Oligomycin A inhibited complex V of the electron transport chain to decrease oxygen consumption to a similar degree between TMEM175 KO and WT cells. Therefore, maximal OCR was reduced primarily because of a decrease in uncoupled respiration induced by a range of FCCP concentrations (Fig. S3 A and B), which cause maximal electron transfer through the electron transport chain independent of ATP synthesis. This finding indicates considerably compromised respiratory capacity in TMEM175-deficient cells compared with WT cells. Transient depletion of TMEM175 (siTM175) in rat hippocampal neurons also led to decrease in maximal OCR (Fig. S3 C and D), although to a lesser degree than complete KO. Consistent with this result, total ATP levels were decreased in TMEM175-depleted cells relative to controls because of a decrease in mitochondrial oxidative phosphorylation (Fig. 3C and Fig. S3E). In primary neurons, this decrease also coincides with reductions of MitoTracker staining, which accumulates in active mitochondria, particularly in neurites, suggesting diminished migratory ability of active mitochondria (Fig. 3 D and E). Because TMEM175 is localized to lysosomes and endosomes, and does not appear to be expressed in mitochondria as shown by MitoFates (36) and MitoCarta (37) prediction of mitochondrial localization, decreased respiration is likely mediated by lysosomal impairment. Together, these findings suggest that reduction of TMEM175 expression leads to deficits in energy homeostasis through impaired lysosome-mediated mitophagy.

Fig. 3.

Mitochondrial respiration is decreased by TMEM175 depletion. (A) OCR of SH-SY5Y WT and TMEM175 KO cells were measured in real time. (B) Basal OCR and the difference between FCCP-induced OCR and Oligomycin A-induced OCR were plotted (n = 9–11). (C) Total intracellular ATP normalized to protein was plotted (n = 9). (D and E) Representative images of MitoTracker Red staining of siCTL and siTM175-treated rat hippocampal neurons were shown (D). (Scale bar, 50 μm.) Quantification of intensity, area and number of MitoTracker Red puncta in neurites was plotted (E) (n = 6). Data presented are mean + SEM. Mann–Whitney test in B and C; two-way ANOVA in E, *P < 0.05, **P < 0.01, and ***P < 0.001. WT vs. KO or siCTL vs. siTM175.

Fig. S3.

Mitochondrial respiration is decreased by TMEM175 depletion. (A and B) OCR of WT or KO cells was measured in real time with FCCP concentrations in the range of 0.1 μM to 1 μM (A). The difference between FCCP-induced OCR and Oligomycin A-induced OCR at FCCP concentrations from 0.03 μM to 10 μM was plotted (n = 9–11) (B). (C and D) OCR of control siRNA (siCTL) or TMEM175 siRNAs-treated (siTM175) rat hippocampal neurons was measured in real time (C). Difference between FCCP-induced OCR and Oligomycin A-induced OCR from a representative experiment was plotted (n = 11) (D). (E) Total intracellular ATP normalized to protein was plotted (n = 13). Data presented are mean + SEM. Two-way ANOVA in B; Mann–Whitney test in D and E. *P < 0.05, **P < 0.01 and ***P < 0.001. WT vs. KO or siCTL vs. siTM175.

TMEM175 Deficiency Increases Phosphorylated α-Synuclein Aggregates in Rat Primary Hippocampal Neurons.

Impaired completion of autophagy in TMEM175-deficient cells strongly suggests that TMEM175 might influence α-synuclein aggregation, the clearance of which is often mediated by autophagy. To determine if TMEM175 plays a role in α-synuclein aggregation, we used a previously published method of seeding preformed α-synuclein fibrils (PFF) (38, 39) in rat hippocampal neuronal cultures. In this model, exogenously provided human α-synuclein PFF recruits endogenously expressed rat α-synuclein, which is essential to form aggregates that are characterized by hyperphosphorylation and detergent insolubility. Rat hippocampal neurons were seeded with human α-synuclein PFF after preinfection with lentiviral control shRNA (shCTL) or TMEM175 shRNA (shTM175) and incubated for an additional 14 and 21 d, after which they were evaluated for the phosphorylated α-synuclein by immunocytochemistry and Western blotting (Fig. 4A). shRNA against TMEM175, which resulted in ∼75–80% KD of TMEM175 relative to control shRNA in 14- and 21-d cultures (Fig. 4B). In PFF-treated TMEM175-KD neurons, a significant ∼2- to 3-fold (14 d) (Fig. 4 C and D) and ∼1.5 fold (21 d) (Fig. 4 G and H) increase in phosphorylated α-synuclein staining was observed compared with control PFF-treated neurons. This increase was independent of types of shRNAs used and correlated with the degree of TMEM175 depletion (Fig. S4A). Depletion of TMEM175 alone did not change the endogenous levels of α-synuclein in PBS-treated groups (Fig. S4B). There was no change in cell viability assessed by the number of healthy nuclei or metabolic activity in TMEM175-KD neurons at 14 d after PFF (Fig. 4E). Viable cells decreased at 21 d after PFF (Fig. 4I), potentially contributing to attenuated fold-increase in phosphorylated α-synuclein compared with 14 d after PFF. Importantly, the pathologically relevant, phosphorylated oligomers of α-synuclein in the detergent-insoluble fraction were also increased by ∼1.7-fold (14 d) (Fig. 4F, Upper) and ∼1.4-fold (21 d) (Fig. 4J, Upper) in TMEM175-deficient neurons relative to control neurons. Phosphorylated α-synuclein monomers in the Triton X-100 soluble fraction were also increased in TMEM175-deficient neurons relative to control neurons (Fig. S4C). Total α-synuclein levels detected by antibody that recognizes both human and rat α-synuclein remained unchanged in both detergent-insoluble and insoluble fractions (Fig. 4 F and J, Lower, and Fig. S4C). Taken together, these results suggest that TMEM175 appears to play a critical role in α-synuclein dynamics and depletion of TMEM175 leads to increased propensity for accumulation of α-synuclein aggregates, which is the key factor in α-synucleinopathies and the development of PD.

Fig. 4.

TMEM175 deficiency increases phosphorylated α-synuclein aggregates in rat primary hippocampal neurons. (A) Diagram depicting the experimental design of PFF-seeding model. (B) Levels of TMEM175 mRNA from rat hippocampal cells treated with control shRNA (shCTL) or TMEM175 targeting shRNA (shTM175) at 14 and 21 d after PFF. (C and G) Immunocytochemistry staining of phosphorylated α-synuclein (p-a-syn) aggregate (green) and βIII-tubulin outlining neuronal cell body and neurites (red) at 14 d (C,14 d) and 21 d (G, 21 d) after PBS/PFF treatment in control (shCTL) and TMEM175 (shTM175) targeting shRNA infected neurons. Blue indicates nuclear staining by Hoechst. (Scale bars, 100 μm.) (D and H) Quantification of intensity, area and number of aggregates was plotted (n = 3). (E and I) Cell viability was assessed by the number of healthy nuclei (Left) and metabolic activity (Right) (n = 3). (F and J) Protein amounts of phosphorylated α-synuclein (Upper) and total α-synuclein (Lower) aggregate in insoluble fraction were quantified by Western blotting, and intensity of monomer and oligomers normalized to tubulin was plotted on the right (n = 3–5). Data presented are mean + SEM. Student t test in B; two-way ANOVA in D and H; Bonferroni's test in E, F, I, and J *P < 0.05, **P < 0.01, and ***P < 0.001. shCTL vs. shTM175.

Fig. S4.

TMEM175 deficiency increases phosphorylated α-synuclein aggregates without affecting endogenous level of α-synuclein or PFF uptake. (A) Amounts of phosphorylated α-synuclein (p-a-syn) aggregates in neurites at 14 d after PBS/PFF treatment in control (CTL) and TMEM175-KD (TM175#1, TM175#2, TM175#3) neurons were assessed by immunocytochemistry and intensity of aggregates was plotted (n = 3) (Right). Levels of TMEM175 mRNA from CTL or TM175s are shown (Left). (B) Endogenous levels of rat α-synuclein from cells treated with shCTL or shTM175 at 14 or 21 d after PBS were assessed by immunocytochemistry and shown quantified in graphs. (C) Protein amounts of phosphorylated α-synuclein and total α-synuclein in Triton X-100 soluble fraction were determined by Western blotting (Upper), and quantified intensity of monomers normalized to tubulin was plotted (Lower) (n = 3–5). (D) Amounts of Alexa 555-PFF internalized after 3-h incubation from shCTL or shTM175-treated cells were assessed by live-cell imaging and shown quantified (n = 4). Data presented are mean + SEM. One-way ANOVA in A; Mann–Whitney test in B and C; two-way ANOVA in D; *P < 0.05 and ***P < 0.001. CTL vs. TM175. ns, not significant.

Discussion

Genetic studies have identified potential risk factors that have helped elucidate biological pathways involved in the development PD. However, the majority of GWAS-implicated loci contain multiple genes requiring further work, such as sequencing, integrative approaches such as Mendelian randomization, and wet-laboratory biological validation to establish the causality of any nominated gene to the pathology of associated disease. In this study, we have experimentally shown multiple cellular/physiological results that raise the possibility that TMEM175 is at least in part responsible for the increased risk of PD from the third-most significant PD GWAS peak. Cyclin-G–associated kinase (GAK) is a candidate for this GWAS peak because of its interaction with LRRK2 (1, 40). Lysosomal pH was altered and catalytic capacity of the lysosome was decreased by the depletion of TMEM175, as illustrated by the decreased activity of major lysosomal proteases. In TMEM175-defecient cells, autophagosome clearance was thus delayed with accumulation of autophagic substrates. Mitochondrial energetic capacity was also reduced potentially through impairment in maintaining a functional mitochondrial pool and providing sufficient metabolite to fuel oxidative phosphorylation. Taking these data together, we find that deficiency of TMEM175 leads to increased formation of phosphorylated α-synuclein aggregates, potentially contributed by impaired lysosomal and mitochondrial function. We suggest that TMEM175, like GAK, should be considered a candidate based on functional data. These data demonstrate links to two established PD genetic risk factors: α-synuclein, through increased α-synuclein aggregation; and GBA, through reduced GBA activity in TMEM175-deficient cells. It was noted in the most recent meta-analysis that two independent signals are present in the Chr 4:951,947 GWAS peak (1), raising the possibility that variants in both GAK and TMEM175 could be contributing to disease risk. Similarly, genes in this genomic region may be involved in the same functional pathways.

Dysregulation of ion currents across lysosomal membrane and lysosomal pH alters lysosomal function, and disruption of channel function has been implicated in neurological diseases, including PD. In familial PD, disruption of ATP13A2, which regulates cation transport across lysosomes, leads to dysfunctional lysosomal degradation of proteins and autophagosomes via impaired lysosomal acidification (8, 9). Overexpression of two-pore channels, which mediate endolysosomal Ca²⁺ signals, alkalinizes lysosomal pH to specifically inhibit autophagosomal-lysosomal fusion (41) and are also shown to act downstream of pathogenic leucine-rich repeat kinase 2 (LRRK2) to regulate trafficking within the endolysosomal system (42). This evidence stresses the importance of ion current regulation in lysosomal function in PD, and the association of TMEM175 with sporadic PD would further support this pathway as a potential area of intervention for disease-modification therapies. Depletion of TMEM175 leads to unstable pH and the impairment of lysosomal function via decreased protease and GBA activity, the enzymatic function of which is also influenced by the pH of the surrounding environment. Whether the integrity of the lysosome is also compromised by the dysregulation of lysosomal pH requires further investigation; however, complete abrogation of TMEM175 resulted in decreased protein levels of mature CTSD, CTSB, and LAMP1, suggesting that the absence of TMEM175 contributes to early disintegration of dysfunctional lysosomes.

It is important to notice that this impairment of lysosomal catalytic activity also resulted in decreased capacity of macroautophagy, where enzymatic degradation after fusion of the autophagosome to lysosome is critical for complete turnover of unwanted proteins and organelles. Inhibition of basal autophagy is particularly detrimental in neurons that are postmitotic and cannot dilute unfolded proteins or damaged organelles through cell division (43, 44). TMEM175 depletion led to accumulation of autophagy substrates and accelerated fusion of autophagosomes to lysosomes. Given that the formation of autophagosomes is mildly affected by TMEM175 deficiency, these results suggest that decreased lysosomal catalytic activity by TMEM175 deficiency also impairs lysosomal capacity to clear autophagosomes, and thus stalls the completion of autophagy. Cellular fusion of autophagosomes to lysosomes might be increased, in part, as an adaptive response to compensate for stalled degradation of autophagosomes.

The impact of mitigated lysosomal function culminates when cells are challenged with a proteostatic burden, such as α-synuclein aggregates, which are handled by various mechanisms that converge on lysosomal degradation (45). Depletion of TMEM175 resulted in increased aggregate formation in the α-synuclein PFF seeding model, where endocytosed PFF seeds induce endogenous α-synuclein recruitment and aggregation (39). Importantly, there was no difference in the uptake of α-synuclein PFF seeds between control and TMEM175-depleted cells (Fig. S4D), suggesting the effect of impaired lysosomal degradation by TMEM175 depletion manifested during the maturation of phosphorylated, insoluble α-synuclein aggregates. Similarly, impaired lysosomes in TMEM175-deficient cells may simply be unable to handle α-synuclein aggregates, which are also resistant to autophagic degradation, reenforcing the vicious cycle of aggregate maturation and inhibition of macroautophagy. However, exactly which parts of any cellular attempt to resolve α-synuclein aggregation are compromised by lysosomal dysfunction resulting from the depletion TMEM175 remains to be studied. The PFF seeding model we adopted was designed to induce minimal cellular toxicity. Hence, we could not determine if TMEM175 KD leads to synergistically increased susceptibility to PFF-induced cell death. Rather, we focused on initial processing/aggregation events of α-synuclein aggregates that we believe to be far upstream from actual cell death, and more proximal to the early cellular insults/stress in the disease. As cellular models of α-synuclein aggregate-mediated cell death become better understood, a logical next step in this research would be to investigate whether TMEM175 deficiency renders cells more vulnerable to aggregate-induced cell death.

It is also important that impaired autophagy might link TMEM175 deficiency to decrease in mitochondrial energetic capacity. Direct inhibition of autophagy through genetic and pharmacological methods was shown to decrease mitochondrial oxygen consumption and energy production, through accumulation of dysfunctional mitochondria and reduced availability of metabolites to fuel oxidative phosphorylation (34, 35, 46, 47). Decreased mitochondrial respiration and ATP generation in TMEM175-deficient cells could also be caused by these mechanisms. Although the exact mechanism remains to be elucidated, mitochondrial dysfunction could independently increase the risk of PD by promoting the loss of dopaminergic neurons through production of oxidative stress, energy depletion, and ultimately cell death (48, 49).

Together, these findings establish TMEM175 as a potential risk factor and candidate point of intervention for PD with both genetic and biological evidence tying TMEM175, lysosomal function, macroautophagy, and mitochondrial function to α-synucleinopathy.

Methods

Isolation of Lysosomal Fraction.

Lysosomal fraction was isolated by differential sedimentation in in sucrose homogenization buffer (0.25 M sucrose, 20 mM Hepes). Details are provided in SI Methods.

Lysosomal Enzyme Activity Assay.

CTSD and CTSB activity was measured using a cathepsin D and B activity assay kit (Abcam, Ab65302, Ab65300), according to the manufacturer’s instructions. Details are provided in SI Methods.

Lysosome pH Imaging and Lysosomal Staining.

Lysosome pH measurement was adapted from a previously described method (29). Details are provided in SI Methods.

Mitochondrial Analysis.

Mitochondrial oxygen consumption was measured according to the instruction of XF Cell Mito Stress Test Kit (Seahorse Bioscience). Details are provided in SI Methods.

α-Synuclein PFF Seeding.

PFF was prepared according to previously described method (50) using human α-synuclein (rPeptide, S-1001-2). Details are provided in SI Methods.

Immunocytochemistry.

Cells were fixed with 4% (vol/vol) paraformaldehyde/4% (wt/vol) sucrose. After primary and secondary antibody staining and rinsing, antibody-specific fluorescence was visualized with imaging using ArrayScan XTI Live High Content Platform (Thermo Fisher). Details are provided in SI Methods. Table S1 for a list of antibodies used.

Table S1.

List of antibodies used

Application	Antibody	Manufacturer	Catalog no.	Dilution
Western blot	TMEM175	Proteintech Group	19925-1-AP>	1:500
	pSer129-a-synuclein	Abcam	ab51253	1:1,000
	Syn-1	BD	610787	1:1,000
	betaIIITubulin	Abcam	ab107216	1:1,000
	CTSD, human	Abcam	ab75852	1:500
	CTSD, rat	Proteintech Group	21327-1-AP	1:500
	CTSB, human	Abcam	ab58802	1:400
	CTSB, rat	Cell Signaling	31718S	1:500
	GAPDH	Abcam	ab9484	1:2,000
	LAMP1, human & rat	Abcam	ab24170	1:500
	LAMP2, human	DSHB	H4B4	1:500
	LC3	Cell Signaling	4108S	1:500
	p62	Abcam	ab56416	1:1,000
Immunocytochemistry	pSer129-a-synuclein	Abcam	ab59264	1:250
	βIIITubulin	Abcam	ab107216	1:1,000
	LC3	Cell Signaling	4108S	1:200
	p62	Abcam	ab56416	1:100
	LAMP1, human	Abcam	ab25630	1:20
	LAMP1, rat	Calbiochem	426017	1:100

Statistical Analyses.

Statistical analysis was done by the GraphPad Prism v7.0 (GraphPad Software). Details are provided in SI Methods.

SI Methods

Cell Culture.

SH-SY5Y (Merck) (51) cells were maintained in DMEM+F12 (Gibco) supplemented with 10% (vol/vol) HI-FBS (Invitrogen), 0.5 mg/mL zeocin (Invitrogen), 7.5 μg/mL Blasticidin (Invitrogen), and 1× antibiotics (100 units/mL penicillin and 100 μg/mL streptomycin; Invitrogen). For culturing primary rat hippocampal neurons, dissected hippocampus from embryonic day 18 rat (Brain Bits, hp) was transferred to 50-mL conical tubes and left to settle. Next, 20 mL of prewarmed Dulbecco’s phosphate-buffered saline (DPBS, Gibco) was added to the tissue to wash for 1 min with gentle mixing. After DPBS removal, the tissue was incubated in 3 mL of TrypLE Select (Gibco) at 37 °C in a water bath for 20 min with occasional flicking to stir up the settled material. After incubation, 1 mL of media was added and the tissue was manually triturated with pipetting. Dissociated cells were spun down at 650 × g and the pellet was resuspended in 10 mL of Neurobasal A (Gibco) medium supplemented with 1× B27 (Gibco), 1× GlutaMax (Gibco), and antibitotics (100 units/mL penicillin and 100 μg/mL streptomycin; Gibco), passed through a 100-µm cell strainer, and plated onto a poly-l-lysine-coated 96-well plate (Biocoat). Cells were fed weekly by 50% media change. For starvation, cells were incubated in EBSS (Gibco) for the indicated times. For blocking of autophagy, cells were treated with 100 nM Bafilomycin A1 (Alfa Aesar, J61835) for the indicated amount of times in EBSS.

CRISPR KO.

To knock out the human TMEM175 in SH-SY5Y cells using the CRISPR/Cas9 technique, TMEM175 sequence AGTTCGACAGAAGTGTACAGAGG (underlined, PAM sequence) was targeted using GeneArt Cas9 Nuclease Protein and CRISPR IVT gRNA via a Neon Transfection protocol. Individual clones were diluted from the transfected population, isolated, expanded, and screened for TMEM175 disruption by PCR amplification of genomic DNA. Clones positive for TMEM175 disruption are confirmed by next-generation sequencing.

Isolation of Lysosomal Fraction.

Cells were washed in PBS, trypsinized, quenched with media, and centrifuged 500 × g for 5 min to produced cell pellets. After media removal, pellets were washed with PBS and centrifuged again. Rat primary neurons were scraped directly in sucrose homogenization buffer after washing with PBS and centrifuged 500 × g to collect cell pellets. Pellets were resuspended in sucrose homogenization buffer (0.25 M sucrose, 20 mM Hepes) containing protease inhibitor and phosphatase inhibitor (Cell Signaling), and homogenized by sonication (10 pulses of 0.5 s). Homogenates were centrifuged at 500 × g for 5 min to discard unbroken cell debris. The resulting supernatant (S1) was centrifuged at 6,800 × g for 10 min to yield P2 (mitochondrial fraction). The remaining supernatant (S2) was centrifuged again at 20,000 × g for 30 min to yield P3 (lysosomal fraction), which was washed with fresh sucrose homogenization buffer and centrifuged at 20,000 × g for 15 min. P2/P3 fractions were resuspended in RIPA buffer containing protease inhibitor and phosphatase inhibitor. For fractions used for lysosomal cathepsin activity assays, protease inhibitor was not added.

Lysosomal Enzyme Activity Assay.

CTSD activity was measured using a CTSD activity assay kit (Abcam, Ab65302). Ten micrograms of lysosomal fraction was adjusted to 50 µL per well with Cell Lysis Buffer for experimental samples. Fifty microliters of blank cell lysis buffer was used for measuring background and 10 μg lysosomal fraction with 1× protease inhibitor (CST #5871) were used as negative control. Next, 52 µL of reaction mix containing 2 µL CTSD substrate and 50 µL reaction buffer was added to each well, incubated at 37 °C for the indicated amounts of time, and fluorescence from converted substrate was measured at excitation/emission (Ex/Em) = 328/460 by a Tecan Infinite M1000 plate reader. CTSB activity was measured using a CTSB activity assay kit (Abcam, Ab65300). Briefly, 20 μg of lysosomal fraction was adjusted to 50 µL per well with cell lysis buffer for experimental samples. Fifty microliters of blank cell lysis buffer was used for measuring background and 10 μg lysosomal fraction with 1× CB inhibitor (Abcam, Ab65300) was used as negative control. Next, 50 µL CB Reaction Buffer followed by 2 µL of 10 mM CB Substrate Ac-RR-AFC (200 µM final concentration) were added to each well, incubated at 37 °C for the indicated amounts of time, and fluorescence from converted substrate was measure at Ex/Em = 400/505 nm by a Tecan Infinite M1000 plate reader. GBA activity was measured using 4-Methylumbelliferyl β-d-glucopyranoside (4-MU, Sigma). Five micrograms of lysosomal fraction was adjusted to 50 µL per well with activity assay buffer for experimental samples. Fifty microliters of blank activity assay buffer was used for measuring background and 5 μg lysosomal fraction with 1 mM conduritol-β-epoxide (CBE, Sigma) was used as negative control. Fifty microliters of 4-MU reaction mix with 2 mM 4-MU and 2% (vol/vol) BSA in activity assay buffer [Sigma; 0.25% Triton X-100, 0.25% Taurocholic acid, 1 mM EDTA in citrate phosphate buffer (Sigma; 400 mM citric acid, 800 mM Dibasic Na phosphate, pH 5.4)] was added to each well, incubated at 37 °C for indicated amounts of time, and fluorescence from converted substrate was measured at Ex/Em = 355/460 nm by a Tecan Infinite M1000 plate reader. The first-order rate constant (k) of enzymatic reaction was obtained by GraphPad Prism v7.0 (GraphPad Software) one-phase association fit with shared plateau.

Lysosome pH Imaging and Lysosomal Staining.

Cells plated on 96-well plates were loaded overnight with 250 µg/mL Oregon-Green 488-conjugated Dextran (Invitrogen) and chased with media (or EBSS for starvation) for 3 h before imaging. Cells were washed with live-cell imaging solution (Invitrogen). The solution used for imaging starved cells contained 140 mM NaCl, 2.5 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, 20 mM Hepes (pH 7.4) and, for the nonstarvation groups, 1× amino acids (made from 50× amino acids mixture; Life Technologies) was added. Fluorescence was imaged using ArrayScan XTI Live High Content Platform (Thermofisher) with a 20× objective. In situ pH calibrations were performed at the end of imaging for each well using Intracellular pH Calibration Buffer Kit (Invitrogen) in which Isotonic K⁺ solutions supplemented with 10 µM of nigericin and valinomycin with pH ranging from 4.5 to 6.5 were used as pH standard solutions. The resulting fluorescence intensity was used to obtain the pH standard curve for each group. Fluorescence intensity was quantified from Oregon-Green 488⁺ puncta from images using HCS Studio 2.0 Cell Analysis Software’s compartmental analysis Bioapplication. The pH value for each group was calculated by fitting the measured average fluorescence intensity of each group to pH standard curve generated from in situ pH calibrations from each group.

Protein Degradation Assay.

Degradation of protein was monitored using Click-iT HPG Alexa Fluor 488 Protein Synthesis Assay Kit (Invitrogen) according to the manufacturer’s instructions. Briefly, cells were incubated with 50 μM HPG in methionine-free DMEM (Gibco) for 30 min (SH-SY5Y) or 1 h (rat primary culture) at 37 °C, washed with EBSS, and chased in complete medium. Cells were fixed in 4% (vol/vol) PFA at indicated time points after chasing and processed for Click-iT reaction that ligates Alexa Fluor 488 to HPG that were incorporated to proteins. The amount of remaining HPG-labeled proteins was determined by the total signal intensity of Alexa Fluor 488 in the ring around the nucleus stained with NuclearMask Blue Stain.

Mitochondrial Analysis.

Cells (SH-SY5Y, 40,000 per well; rat primary hippocampal neurons, 20,000 per well) were seeded in a 96-well XF Cell Culture Microplate. The following day cells were washed 2× with and incubated in Mito Stress Test Assay Medium (1 mM pyruvate, 2 mM glutamine, and 10 mM glucose in XF base medium) in a 37 °C non-CO₂ incubator for 1 h. OCRs were measured with an XF96 extracellular flux analyzer (Seahorse Bioscience) at baseline, after addition of 2 μM oligomycin to evaluate respiration associated with cellular ATP production, after addition of 0.3 μM (0.03–10 μm for titration) FCCP to evaluate uncoupled respiration, and after addition of 1 μM antimycin/rotenone to measure nonmitochondrial respiration. To visualize intact mitochondria, cells were stained with MitoTracker Deep Red FM (Invitrogen) according to the manufacturer’s instructions. Briefly, cells were stained with media containing 100 nM MitoTracker Deep Red for 30 min at 37 °C and washed 2× with Live Cell Imaging buffer (Invitrogen). Stained cells were imaged in Live Cell Imaging buffer using an ArrayScan XTI Live High Content Platform (Thermofisher) with a 20× objective and analyzed as indicated as below.

Measurement of ATP.

The total level of intracellular ATP was assessed using CellTiter-Glo Luminescent Assay (Promega) following the manufacturer’s instructions. Obtained luminescent value was normalized to total protein amount determined by Pierce bicinchoninic acid Protein Assay Kit.

Transient KD.

For transient KD with lentiviral shRNA, rat hippocampal neurons were transduced with lentiviral particle (MISSION Lentiviral Transduction Particle, Sigma) targeting 5′ CCAGCATCTTCCAGTTTGC 3′ (TM175#1), 5′ ATGACATTTCTAATCGTAA 3′ (TM175#2), or 5′GTGATGAACTTGAGCGTGT 3′ (TM175#3) of TMEM175 mRNA at multiplicity of infection of 2 on day in vitro 7. Nonmammalian shRNA was used as control (MISSION Nonmammalian shRNA Control, Sigma). For transient KD with siRNA, rat hippocampal neurons were treated with 20 nM cholesterol conjugated-siRNA diluted in media targeting 5′ ATGACATTTCTAATCGTAA 3′ of TMEM175 mRNA at day in vitro 4. Nontargeting siRNA was used as control.

Autophagosome–Lysosome Fusion Assay.

The autophagosome–lysosome fusion assay was carried out in SH-SY5Y cells transfected with the RFP- and GFP-tagged LC3. Cells were grown in regular media. The number of intensity of GFP or RFP puncta in each cell was imaged live using an ArrayScan XTI High Content Platform (Thermofisher) with a 20× objective and analyzed as indicated as below.

Immunocytochemistry.

After rinsing with DPBS, cells were fixed with 4% (vol/vol) paraformaldehyde/4% (wt/vol) sucrose at room temperature for 15 min. For analysis of LC3 puncta, cells were fixed with 100% (vol/vol) methanol for 15 min at −20 °C. Cells were then rinsed 3× with DPBS, permeabilized, and blocked with permeabilization/blocking buffer [0.2% BSA, 10% (vol/vol) goat serum, 0.2% Triton X-100 in DPBS] for 1 h at room temperature. Then cells were incubated with primary antibodies diluted in blocking buffer [5% (vol/vol) goat serum, 0.2% BSA in DPBS] at 4 °C overnight. After rising 2× with 0.05% tween20 in DPBS and 1× with DPBS, each well was incubated with matching goat anti-mouse/rabbit/chicken Alexa Fluor-conjugated secondary antibody (1:1,000; Life Technologies) and Hoechst 33342 (1:10,000; Invitrogen) in blocking buffer for 1 h at room temperature. After rinsing, antibody-specific fluorescence was visualized with imaging using an ArrayScan XTI Live High Content Platform (Thermo Fisher) with a 20× or 40× (p62, MitoTracker) objective.

α-Synuclein PFF Seeding and Uptake Assay.

For the α-synuclein PFF seeding and uptake assay, 100 μg/mL sonicated α-synuclein PFF were diluted in media and added to neurons [1:100 dilution (1 μg/mL), 0.1 mL per 96-well; 1:25 dilution (4 μg/mL), 2 mL per 6-well] 4 d after lentivirus transduction. Neurons were incubated for 14 and 21 d after PFF addition. Approximately 50% of media was replaced with fresh media once a week. To measure uptake, sonicated α-PFF were conjugated to Alexa Flour 555 (Invitrogen A30007) and indicated amounts of Alexa 555-PFFs in a complete medium were given to rat hippocampal neurons at 37 °C, 5% (vol/vol) CO₂ for 3 h, the last 15 min of which was also with Hoechst 33342 (Life Technologies; 1:20,000) to label nuclei. After incubation, cells were rinsed 2×, supplied with fresh media with red background suppressor (Life Technologies), and live-cell imaged using an ArrayScan XTI Live High Content Platform with 20× objective.

High-Content Image Analysis.

The ArrayScan XTI Live High Content Platform was used for the high-content imaging. At least nine fields per well were imaged, analyzed, and averaged. A total of three to six wells containing ∼20,000 cells per well were used per treatment condition (n = 3–6). Image analyses and calculations were performed using HCS Studio Cell Analysis Software (ThermoFisher). Hoechst staining was used to label cell nuclei. For analysis of phosphorylated α-synuclein and MitoTracker staining, neuronal profiling BioApplication where a single cell was identified based on total βIII-tubulin or Mitotracker staining that marks both the cell cytoplasm and neurites was used. Puncta were determined positive if their presented fluorescence intensity was higher than a fixed threshold within a given pixel. Total intensity, area, and number of p-a-syn and MitoTracker-positive puncta per cell were quantified within a defined neurite region stemming out of the cell body. For analysis of other puncta, including Oregon Green 488, LAMP1, LC3, p62, and Alexa555-PFF, compartmental analysis BioApplication was used. In this analysis, nucleus detected based on Hoechst staining was enlarged to cover the size of neuronal cell body and defined as a region to look for marker-specific puncta. Average intensity per cell within a defined region was quantified for Oregon Green 488-positive puncta. Total intensity per cell was quantified within the defined cell region for LC3⁺ and p62⁺ puncta.

Cell Viability.

Cell viability was assessed by CellTiter-Blue Cell Viability Assay (Promega) following the manufacturer’s instructions.

RNA Isolation and qRT-PCR.

Total RNAs from cells were isolated using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. RNA was reversed-transcribed to cDNA using a high-capacity cDNA reverse transcription kit (Applied Biosystems) following the manufacturer’s instructions. Gene expression was detected by Taqman gene-expression assay (Thermofisher, Human TMEM175, Hs01018100_g1; Human GAPDH, 4310884E; Rat TMEM175, Rn01531050_m1; Rat GAPDH, Rn01775763_g1).

Sequential Extraction of Detergent-Insoluble Proteins.

Detergent-insoluble fraction was prepared according to previously described method (50). Cells were rinsed 2× with PBS, lysed in 150 μL ice-cold 1% Triton X-100/TBS with protease and phosphatase inhibitor (Cell Signaling), and placed in a polyallomar high-speed microcentrifuge tube (Beckman Coulter, 357448). Lysate was sonicated 10 times, 0.5-s pulse (Qsonica Model Q55, Set 4), incubated on ice for 30 min, and centrifuged at 100,000 × g at 4 °C for 30 min. Supernatant was collected into a new tube, as soluble fraction, and pellet was resuspended in 100 μL ice-cold 1% Triton X-100/TBS, and centrifuged again at 100,000 × g at 4 °C for 30 min. The resulting supernatant was discarded and pellet was resuspended as an insoluble fraction in 2% (vol/vol) SDS/TBS.

Immunoblotting Analyses.

Proteins were separated by 4–12% NuPAGE Bis-Tris gel, and transferred to a PVDF membrane using iBlot (Invitrogen). Blots were blocked with Odyssey Block Buffer (Licor) for 1 h and probed overnight with primary antibodies (Table S1). Appropriate IRDye-conjugated secondary antibodies (1:10,000; Licor) were used to visualize proteins with Odyssay imaging system (Licor).

Statistical Analyses.

Statistical analysis was done by the GraphPad Prism version 7.0 (GraphPad Software). Two-way ANOVA followed by Fisher’s least-significant difference multiple comparison test was used in the studies that analyze more than two parameters assuming each comparison is independent from others. One-way ANOVA was used for analysis of data from three or more groups. Student’s t test or Mann–Whitney test was used to compare data from two groups. Statistical significance was determined by *P < 0.05, **P < 0.01, and ***P < 0.001.