Antibodies and reagents

BCKDK (sc-374425; immunoblotting: 1:100, immunostaining: 1:25) and BCKDE1A (sc-67200; immunoblotting: 1:1000) antibodies was purchased from Santa Cruz Biotechnology. NDUFS1 (68253-1-Ig, 12444-1-1AP; immunoblotting: 1:1000, immunostaining: 1:400), NDUFS6 (14417-1-AP; immunoblotting: 1:1000), and LONP1 (15440-1-AP; immunoblotting: 1:3000), BCAT1 (13640-1-AP, immunoblotting: 1:1000), and ATPB (17247-1-AP; immunoblotting: 1:3000) antibodies were from Proteintech. ß-actin (A1978, immunoblotting: 1:10000) and IgG (12–370; coimmunoprecipitation: 2ug) were purchased from Sigma-Aldrich. αSyn phospho-S129 (ab168381; immunostaining: 1:1000), αSyn aggregate (ab209538; immunostaining: 1:1000), tyrosine hydroxylase (ab76442; immunostaining: 1:500), and the OXPHOS complex antibody cocktail (ab110413; immunoblotting: 1:1000) were from Abcam. Tyrosine hydroxylase (MAB318) antibody was purchased from Millipore. MPP+ (D048), protease inhibitor, and protein phosphatase inhibitor were purchased from Sigma-Aldrich. DAPI (D1308) and secondary antibodies (Alexa Fluor #A-11008, #A-11001, #A-11011, #A-11004, #A-21449) were purchased from Thermo Fisher Scientific.

Cell culture

SH-SY5Y cells stably expressing either a pLKO3.1 control shRNA (Addgene; #8453) or BCKDK shRNA plasmid (Sigma-Aldrich; TRCN0000199200) were cultured in 1:1 DMEM/F12 media with 10% FBS, 100ug/ml penicillin, 100ug/ml streptomycin, and 1ug/ml puromycin. Myc, myc-αSyn WT, or myc-αSyn A53T [34] expressing stable SH-SY5Y cell lines were cultured in FBS (10%) and penicillin/streptomycin (1%) supplemented DMEM/F12 media treated with 400ug/ml G418. BCKDK-myc plasmid was purchased from Origene (RC203601). Leucine-free media was from ThermoFisher (#30030); cells were cultured in 1:1 leucine-free DMEM/F12 media supplemented with FBS (10%) and penicillin/streptomycin (1%). HEK293 cells were maintained in DMEM media supplemented with FBS (10%) and penicillin/streptomycin (1%). Cells were kept in a 37 °C, 5% CO2 incubator.

Induced pluripotent stem cells and neuronal differentiation

The PD iPS cell line (RUCDR Infinite Biologics; αSyn A53T; NN0004337) and its isogenic control (NN0004344) were differentiated into dopaminergic neuron-enriched cultures as described in [37, 50]. Briefly, iPS cell colonies were plated on Matrigel (2.5%) coated plates in mTeSR media, treated with SB431542 (10 μm) and Noggin (100ng/ml) in Neural Media (NM; Neurobasal and DMEM/F12 (1:1), B-27 Supplement Minus Vitamin A, N2 Supplement, GlutaMAX, 100units/ml penicillin and 100 µg/ml streptomycin) with FGF2 (20ng/µl) and EGF (20ng/µl) for 7 days, then treated with human recombinant Sonic hedgehog (SHH, 200ng/ml) in neuronal differentiation medium (Neurobasal and DMEM/F12 (1:3), B27, N2, GlutaMax and P/S) for 4 days. Cells were then incubated in neuronal differentiation medium containing BDNF (20ng/ml), ascorbic acid (200 μm), SHH (200ng/ml), and FGF8b (100ng/ml) for 3 days, then treated with BDNF, ascorbic acid, GDNF (10ng/ml), TGF-b (1ng/ml), and cAMP (500 μm). All growth factors were from Pepro Tech. Cells were plated on poly-D-lysine/laminin coated coverslips. Three independent differentiations were performed, each lasting 27 days. Differentiating iPSCs were infected at day 25 with either GFP or BCKDK-GFP lentivirus using a multiplicity of infection (MOI) of one, then fixed using 4% paraformaldehyde after 48 h for staining. Lentiviral vectors (GFP: pLV[Exp]-EF1A>/EGFP; BCKDK-GFP: pLV[Exp]-EF1A > hBCKDK[NM_005881.4]/EGFP) were constructed and packaged by VectorBuilder.

Animal model and PD patient tissue

αSyn A53T [B6.Cg-Tg(Prnp-SNCA*A53T)23Mkle/J, JAX Stock No: 006823] breeders (C57Bl/6J genetic background) were purchased from Jackson Laboratories. All animal experiments were conducted in accordance with protocols approved by the Institutional Animal Care and Use Committee of Case Western Reserve University and were performed based on the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Male mice were used and were maintained at 12 h light/dark cycles.

Postmortem patient samples were obtained from the National Institutes of Health (NIH) NeuroBioBank under the approval of the Institutional Review Board (IRB). All donated specimens were assessed by board-certified neuropathologists to establish a disease condition diagnosis. Age-matched tissue sections from 3 healthy individuals (1 female, 2 male) and 3 PD patients (1 female, 2 male) were used in immunostaining. For immunoblotting, midbrain lysates from a total of 7 healthy male subjects and 6 male PD patients were assessed. Patient information is provided in the Supplemental Information (Suppl. Figure 1e, Suppl. Figure 4b).

Immunoblotting

Total cell lysates were extracted by first washing cells with 1X PBS, then incubating them in total lysis buffer (10mM HEPES-NaOH [pH 7.9], 150mM NaCl, 1mM EGTA, 1% Triton X-100, protease inhibitor cocktail, and phosphatase inhibitor cocktail) at 4 °C for 30 min. Lysates were collected and centrifuged for 10 min at 12,000 g; the supernatant was collected for immunoblotting analysis. Protein concentrations were measured via Bradford assay, and samples were prepared in Laemmli buffer with equal amounts of total protein. Samples were run on SDS-PAGE, transferred to nitrocellulose membranes, and probed with the appropriate antibodies. Blots were visualized with ECL. The resulting bands were analyzed using ImageJ; intensity of each band was measured and normalized to the indicated loading control. Mouse samples run concurrently were normalized to the average of the control (WT) samples.

Blue native PAGE

Mitochondrial fractions were isolated by washing cells with PBS, then incubating them in mitochondrial lysis buffer (250 mM sucrose, 20 mM HEPES–NaOH, pH 7.9, 10 mM KCl, 1.5 mM MgCl2, 1 mM EDTA, 1 mM EGTA, protease inhibitor cocktail, and phosphatase inhibitor cocktail) for 30 min at 4 °C. Lysates were collected and homogenized using a 25-gauge needle. Homogenates were centrifuged at 800 g for 10 min at 4 °C; supernatants were then centrifuged at 10,000 g for 20 min at 4 °C. The remaining pellets were washed with mitochondrial lysis buffer and spun again at 10,000 g for 20 min. Mitochondrial pellets were incubated on ice in 1x NativePAGE Sample Buffer (NativePAGE™ Sample Prep Kit; Thermo Fisher; BN2008) containing 2% Digitonin (sigma, D141) for 30 min, and then centrifuged at 20,000 g for 30 min at 4 °C. Protein concentrations of the resulting supernatant were determined, and samples were prepared with NativePAGE 5% G-250 Sample Additive. Samples were run on pre-cast native 4–16% Bis-Tris gels using blue native polyacrylamide gel electrophoresis, then either incubated in Brilliant Blue R staining solution or transferred onto nitrocellulose membranes and probed using the indicated antibodies. Gels incubated in Brilliant Blue were transferred to Coomassie destaining solution overnight. Band densities for immunoblots were normalized to LonP1, which was used as a loading control, and stained gels were normalized to total protein levels.

Immunostaining

Mouse sections were obtained from 12-month control and A53T-αSyn expressing mice that were deeply anesthetized and transcardially perfused with 4% paraformaldehyde in PBS. Frozen brain sections were used for all tissue staining; sections were permeabilized using 0.5% Triton X-100 in PBS buffer for 5 min, blocked in 6% normal goat serum with 0.3% Triton X-100 for one hour, then incubated in primary antibody diluted in 3% normal goat serum with 0.3% Triton X-100 overnight at 4 °C, then with the appropriate 488/568/633/647nm secondary antibodies for 2 h at room temperature. All incubation was performed in a humidified chamber, and slides were washed three times for 5 min each using PBS with 0.1% Triton X-100 after each step. Slides were incubated with DAPI for 10 min at room temperature to stain nuclei, then mounted (Faramount Aqueous Mounting Medium; Agilent; S302580-2).

Frozen tissue sections from postmortem healthy individuals and PD patients were used for staining; sections were permeabilized using 0.5% Triton X-100 in PBS for 5 min, then blocked for one hour in 6% normal goat serum with 0.3% Triton X-100. Slides were incubated in the indicated primary antibodies, diluted using 3% normal goat serum with 0.3% Triton X-100, and placed in a humidity chamber at 4 °C overnight. Slides were then incubated with 488/568nm secondary antibodies for 1 h at room temperature in a humidified chamber. Slides were washed using 0.1% PBS-Triton X-100 three times after every step. Stained tissue was mounted using Faramount Aqueous Mounting Medium.

All immunostained sections were imaged using an Olympus confocal microscope. Staining was quantified as fluorescence intensity per cell expressing the indicated marker, as measured using ImageJ, and averages were taken for each biological repeat.

Proximity ligation assay

Mouse tissue sections were permeabilized using 0.1% PBS-Triton X-100 for 5 min, then blocked with PLA blocking buffer for 1 h at 37 °C. Sections were then incubated overnight at 4 °C in BCKDK and NDUFS1 antibodies; this was followed by incubation in the provided PLA probes (Rabbit PLUS, Mouse MINUS In Situ PLA Probes; Sigma-Aldrich; DUO92002) for 1 h at 37 °C. Samples were incubated in ligase buffer for 30 min at 37 °C, then in amplification buffer for 100 min at 37 °C. Slides were mounted using Duolink In Situ Mounting Medium with DAPI (Sigma-Aldrich; DUO82040), and mounted slides were imaged using confocal microscopy. Fluorescence intensity was measured for each field of view using ImageJ software, then normalized to the number of cells, determined using DAPI-stained nuclei.

AlphaFold2

BCKDK (O14874) and NDUFS1 (P28331) amino acid sequences were obtained from the UniProtKB database. Structure prediction was conducted as described in [36] using the provided code.

αSyn BIFC

SH-SY5Y or HEK293T cells were cultured on coverslips coated with a gelatin-fibronectin solution (1% gelatin; 0.5% fibronectin) and a poly-D-lysine-laminin solution (0.1 mg/ml PDL overnight at 4 °C; 5ug/ml laminin for 1 h at 37 °C), respectively. Cells were transfected with two tagged αSyn plasmids, VN-αSyn (Addgene #89470) and αSyn-VC (Addgene #89471) for 48 h, then treated with DAPI for 10 min, washed 3 times with PBS, and mounted using Faramount Aqueous Mounting Medium. Where indicated, cells were co-transfected with BCKDK shRNA (Sigma-Aldrich; TRCN0000199200), NDUFS1 shRNA (Sigma-Aldrich; TRCN0000064631), or NDUFS expression plasmid (DNASU; HsCD00860978). Mounted coverslips were imaged using an Olympus confocal microscope at 405 nm and 488 nm. Fluorescence intensity at 488 nm was quantified per cell, indicated by DAPI signal, using ImageJ software. Averages were taken for each biological repeat.

Acetyl CoA ELISA

Cells were pelleted, washed, and lysed in PBS using a water bath sonicator (3 × 1 min, 80 kHz). Lysates were centrifuged for 10 min at 1500 g, and kit instructions (A-CoA ELISA kit; Elabscience; E-EL-0125) were used to measure Acetyl CoA levels. In brief, samples were added to the Micro ELISA plate for 90 min, followed by the provided Biotinylated Detection Ab, HRP Conjugate, and Substrate reagent. Absorbance was measured at 450 nm. Mouse samples run concurrently were normalized to the average of the control (WT) samples.

Measurement of Complex I activity

Control and BCKDK KD SH-SY5Y cell pellets were collected, washed, and lysed in PBS using a water bath sonicator (3 × 1 min, 80 kHz). Detergent (Complex I Enzyme Activity Assay Kit; Abcam; ab109721) was added and samples were left on ice for 30 min, then centrifuged at 16,000 g for 20 min. Complex I activity was then determined over the course of 30 min by following manufacturer instructions. Briefly, plates were incubated with lysates for 3 h, then washed and incubated with the provided NADH/dye solution. Absorbance was measured at 450 nm every 20 s for 30 min.

Measurement of cell viability

Stable control and BCKDK knockdown cells were cultures in 96-well plates and were assayed using a Cytotoxicity Detection Kit (Sigma-Aldrich; #11644793001). Cells were incubated in fresh media for 6 h prior to assay; 30 min prior to assay, 1% Triton X-100 was added to the “high control” group. Media was collected from each well into a new 96-well plate and freshly prepared reaction solution, provided in the kit, was added to each well. Plates were incubated at room temperature in the dark for 30 min, then absorbance was measured at 492 nm and 692 nm (for background). Results were quantified as a percentage of LDH release in comparison to the “high control” group.

Measurement of cell proliferation

Stable control and BCKDK knockdown cells were plated in 96-well plates, treated with MPP+ (1mM, 48 h), then incubated at 37 °C with the two reagents (MTT labeling agent, 4 h; Solubilization solution, 16 h) provided in the MTT assay kit (Sigma-Aldrich; #11465007001). Absorbance was measured at 590 nm and 690 nm (as background). Sample absorbances were normalized to those of the control cells.

Measurement of mitochondrial membrane potential

Stable control and BCKDK knockdown SH-SY5Y cells were cultured on coverslips coated overnight at 4 °C with a gelatin-fibronectin solution (1% gelatin; 0.5% fibronectin), treated with MPP+ (1mM, 48 h), then incubated with 0.25 μm TMRM and DAPI for 25 min at 37 °C. Coverslips were washed 3 times with PBS, then mounted using a mounting medium (Faramount Aqueous Mounting Medium; Agilent; S302580-2). Coverslips were imaged using a Keyence BZ-X700 microscope, and the resulting images were analyzed using the ImageJ software. Total fluorescence intensity of the red channel and number of cells (using DAPI) were quantified for each field of view. The average fluorescence intensity for each group was taken for each biological repeat.

Measurement of mitochondrial reactive oxygen species (ROS)

Stable control and BCKDK knockdown SH-SY5Y cells were cultured on coverslips coated overnight at 4 °C with gelatin-fibronectin, treated with MPP+ (1mM, 48 h), then incubated in 5 μm MitoSOX Red fluorescent dye (Invitrogen, M36008) and DAPI for 10 min at 37 °C. Coverslips were washed 3 times with PBS, then mounted using a mounting medium (Faramount Aqueous Mounting Medium; Agilent; S302580-2). Coverslips were imaged using a Keyence BZ-X700 microscope, and the resulting images were analyzed using the ImageJ software. Total fluorescence intensity of the red channel and number of cells (using DAPI) were quantified for each field of view. The average fluorescence intensity for each group was taken for each biological repeat.

Co-immunoprecipitation

Total cell lysates were prepared from SH-SY5Y cells as described above. Total protein concentrations were measured using Bradford assay, and 1 mg of protein in 1 ml of total lysis buffer was incubated with 2ug of either BCKDK primary antibody or control IgG antibody overnight at 4 °C with slow rotation. Protein A/G beads (30ul; Santa Cruz Biotechnology; sc-2003) were added and slowly shaken for 2 h at 4 °C, then centrifuged and washed 4 times with total lysis buffer (4000 g x 3 min). Laemmi buffer (15ul) was added and samples were boiled at 100 °C for 10 min, cooled, and centrifuged again; immunoblotting was then performed on the resulting supernatant using the indicated antibodies. NDUFS1 protein levels were quantified using ImageJ and normalized to BCKDK protein levels for each pulldown group. Total lysates not subjected to immunoprecipitation were also assessed to confirm expression of the proteins of interest.

Proteomic analysis

Control and BCKDK knockdown cells were harvested, lysed using 2% SDS, and probe sonicated twice. Samples were prepared using filter-aided sample preparation (FASP), as described in [51]. Samples were washed in acetone, reduced in 10mM DTT in 3 M urea with water bath sonication, alkylated with 25mM iodoacetamide, then digested using trypsin/Lys-C. Samples were diluted in 0.1% formic acid for liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis.

Lysed proteins from each group were loaded onto a column with blanks in between. Data were acquired using an Orbitrap Velos Elite mass spectrometer equipped with a Waters nanoACQUITY LC system. Peptides were desalted on a trap column and resolved on a reversed-phase column, then were introduced into the nanospray mode. Peptides in the range of 380 to 1800amu were scanned for; the proteins were identified with Mascot and quantified.

Statistical analysis

All cell culture and animal experiments were conducted with at least three independent repeats. Data were analyzed using either the Student’s t-test or one-way ANOVA with post-hoc Tukey’s test for comparison across groups. All data are expressed as mean ± SEM. A p value < 0.05 was considered statistically significant.