Impairment of PARK14-dependent Ca2+ signalling is a novel determinant of Parkinson's disease

The etiology of idiopathic Parkinson's disease (idPD) remains enigmatic despite recent successes in identification of genes (PARKs) that underlie familial PD. To find new keys to this incurable neurodegenerative disorder we focused on the poorly understood PARK14 disease locus (Pla2g6 gene) and the store-operated Ca2+ signalling pathway. Analysis of the cells from idPD patients reveals a significant deficiency in store-operated PLA2g6-dependent Ca2+ signalling, which we can mimic in a novel B6.Cg-Pla2g6ΔEx2-VB (PLA2g6 ex2KO) mouse model. Here we demonstrate that genetic or molecular impairment of PLA2g6-dependent Ca2+ signalling is a trigger for autophagic dysfunction, progressive loss of dopaminergic (DA) neurons in substantia nigra pars compacta and age-dependent L-DOPA-sensitive motor dysfunction. Discovery of this previously unknown sequence of pathological events, its association with idPD and our ability to mimic this pathology in a novel genetic mouse model opens new opportunities for finding a cure for this devastating neurodegenerative disease.

(b) no effect of acute TG treatment (5μM, 20 min, like in Supplementary Fig. 1a) on TxNip and BiP expression in hPSF from control and idPD patients; (c) no effect of acute TG treatment (5μM, 5 min, like in Main Fig. 2d) on TxNip and BiP expression in MEFs from WT and PLA2g6 ex2 KO mice.
Each data point is average ±SE (n=2) expression of each target normalized to the level of GAPDH expression in corresponding sample.

mouse model (B6.129S-Pla2g6 ΔEx2-VB /J).
WT Pla2g6 locus and the targeting vector are schematically represented at the top of the panel. Exon 2 of Pla2g6, containing the translation initiation codon, is flanked by two loxP sites (open triangles), whereas the neomycin cassette (Neo) is immediately flanked by two FRT sites (double filled triangles). As depicted, the expected homologous recombination event creates the recombined (floxed) locus and removes Diphtheria Toxin A (DTA) negative selection marker. Crossing a recombined Pla2g6 locus mouse with a ubiquitous Flp recombinase C57BL/6 animal allowed for excision of the FRT-flanked region, creating an animal carrying conditional Pla2g6 Ex2 allele without neomycin selection cassette. Breeding the heterozygous recombined F1 mouse with a ubiquitous Cre recombinase C57BL/6 animal resulted in the Cre-mediated excision of the floxed exon 2 region, creating a total exon 2 knockout (ex2 KO   The genomic DNA of the 2 tested F1 mice (lanes 1 and 2) were compared with wild-type DNA (lane 3). The HpaI/NheI digested DNAs were blotted on nylon membrane and hybridized with the probe expected to anneal to the 3' end of homology arm of the targeting vector to validate the zygocity of the Pla2g6(ex2) constitutive knock-out gene mutation in these animals. The expected fragments are: 8.2, 9, and 6.1 kb for WT allele, recombined/floxed allele, and constitutive knock-out allele (floxed region deleted), respectively.

PCR-based genotyping and confirmation of the constitutive Exon 2 knockout at the genome and transcript level.
(A) Schematic representation of WT Pla2g6 locus with positions of four sets of primers used for PCR-based genotyping (sets 1 and 2) of the colony, or for confirmation of the lack of Pla2g6 exon 2 in transcripts from mouse brains (set 3 and 4). (B) and (C), Representative results of tail DNA genotyping for 9 animals from the colony using PCR primer sets 1 and 2. Expected length of PCR products for primer set 1 are 4028 (WT) and 2900 bp (ex2 KO allele), and for set 2 only WT allele (857 bp product) can be detected. Taken together, PCR with both sets of primers allowed for unambiguous determination of the Pla2g6 locus genotype for each animal within the colony. (D) and (E), Total RNA isolated from brains of two representative pairs of WT and exon 2 KO animals was reverse-transcribed and used as a template for PCR with primer sets 3 and 4. Expected length of PCR products for primer set 3 are 736 (WT) and 486 bp (ex2 KO allele), and for set 4 only WT allele (644 bp product) can be detected. As expected, for both animals previously genotyped as Exon 2 KO (using primer sets 1 and 2), transcripts coding for PLA2g6 are present in the brain, but are missing exon 2. Additionally, the product amplified with the primer set 3 from brains of ex2 KO mice was cloned and sequenced, and both the expected cDNA sequence and the lack of Exon 2 were confirmed (data not shown).

Knock out of Exon 2 of Pla2g6 gene did not affect the level of transcripts of the (L) and (S) splice variants of PLA2g6 in the brain (a), MEF cells (b) and testis (c).
Quantitative Real Time PCR analysis of expression levels of (L) and (S) splice variants of PLA2g6 in the brains, testis and MEF cells from WT and ex2 KO mice. Please, notice significantly higher expression of PLA2g6(L) in testis. Data are normalized to GAPDH in each sample, and shown as average ±SE from 2-3 experiments.

Full length PLA2g6(L) is present in WT, while only truncated protein is present in ex2 KO mice.
Representative Western blot probed with custom-made PIN antibody that specifically targets PIN domain encoded by exon 8b that is present only in (L) splice variant of PLA2g6, and β-actin staining of the same samples. Specificity of PIN ab is shown in Supplementary Fig. 10. Images have been cropped for presentation. Full size images are presented in Supplementary Fig. 22.
WT: endogenous protein from testis of WT mouse; ex2 KO : endogenous protein from testis of ex2 KO mouse; PLA2g6 179-806 : recombinant N terminally truncated myc/his-tagged PLA2g6(L) 179-806 protein expressed in FreeStyle™ 293-F cells. Please, notice that recombinant protein contains myc and his tags on its N and C termini, respectively, which slightly increase its MW in comparison with equivalent endogenous protein in ex2 KO mice.

Validation of custom mPIN ab that specifically recognizes (L), but not (S) variant of PLA2g6.
(a) Representative Western blot shows that recombinant myc PLA2g6(S) his and myc PLA2g6(L) his protein can be detected with Myc antibody, while only myc PLA2g6(L) his protein can be recognized by custom-made PIN antibody that specifically targets PIN domain (encoded by exon 8b), which is present in (L), but spliced out in (S) variant of PLA2g6. Images have been cropped for presentation. Full size images are presented in Supplementary Fig. 22. (b) Western blot shows that not only recombinant myc (L) his , but also endogenous PLA2g6(L) protein from WT mouse can be specifically recognized by PIN antibody. Blot on the bottom shows the same membrane stained for β-actin. Please, notice that recombinant protein contains myc and his tags on its N and C termini, respectively, which slightly increase its MW in comparison with equivalent endogenous protein in WT mice. Images have been cropped for presentation. Full size images are presented in Supplementary Fig. 22.   Representative images show the results of immunostaining of iPSC-derived DA neurons (31 DIV) for tyrosine hydroxylase (TH from Abcam, red), DAT (Abcam, cyan), and VMAT2 (Abcam, green); the nuclei were stained with DAPI (blue). Enlarged images (on the right) focus on individual DA neurons and represent corresponding areas on the less detailed images that show larger area. The composite images show co-localization of TH with other dopaminergic neuronal proteins (DAT and VMAT2) typical for mature DA neurons.

Representative images of VTA (a) and SNc (b) areas of the brain of WT and ex2 KO littermates (24 months old).
(a) Immunostaining for tyrosine hydroxylase (TH, brown) of VTA area.

Results of blinded stereological analysis of TH+ neurons in SN of 8, 16 and 24-month old WT and ex2 KO littermates.
(a) Representative immunostaining for tyrosine hydroxylase (TH from Calbiochem, brown) in brain slices and corresponding (enlarged) nigrostriatal area of the brain of 16-month old WT and ex2 KO littermates.
(b) Results of blinded stereological analysis of the total numbers of TH+ neurons in SN (both sides) of WT and ex2 KO mice (data for littermate pairs are connected with lines).

(b) Enlarged images of the corresponding areas identified in (a).
(c) Negative control showing the results of staining with secondary Alexa594 and Alexa488 antibodies (both from Molecular Probes) in the absence of primary antibodies.

Accumulation of endogenous LC3 in TH+ neurons in SNpc of ex2 KO , but not WT 16-month old mice.
Representative images show the results of immunostaining for tyrosine hydroxylase (αTH from Abcam, red), LC3 (αLC3B from Cell Signaling, green) and DAPI (blue) of neurons in SNpc area of the brain from WT and ex2 KO littermates (16-month old).

Analysis of LC3 mCherry/eGFP autophagic flow in WT and ex2 KO MEFs, and WT MEFs treated with thapsigargin.
Representative images (the whole cells are shown on the left, and enlarged part of the cells are shown on the right) and correlation maps of tandem mCherry (red) / eGFP(green) tagged LC3 in live MEF cells. Experiments were done 48 hours after transfection. Thapsigargin treatment was 10nM for 24 hours before the experiment. Below are the summary results of comparative analysis of (a) correlation coefficient of mCherry and eGFP, and (b) the size of mCherry particles; the data show average ± SE; summary data from a total of 15 cells per condition (5 cells from each of 3 independent experiments), *(p<0.05), ***(p<0.001).  (d) Progressive increase in IM-induced Ca 2+ release following prolonged culture of MEFs from Orai1 KO mice demonstrate the ability of MEFs to compensate for Orai1 deficiency, and to restore their ER stores.

Primary MEFs from Orai1 knockout (Orai1 KO ) mice have impaired SOCE, depleted ER Ca 2+ stores and significant autophagic dysfunction, which mimic deficiencies found in MEFs from
(e) Representative images (left) and summary data for correlation coefficient (right) of tandem mCherry (red) / eGFP(green) tagged LC3 in live MEF cells from Orai1 KO mice. Experiments similar to those described in main Figure 4d,f. The data show average ± SE; summary data from a total of 15 cells per condition (5 cells from each of 3 independent experiments). ** p<0.01***(p<0.001).

Rescue of LC3 mCherry/eGFP autophagic flow in ex2 KO MEFs by WT PLA2g6(L), but not F72L mutant.
Representative images (the whole cells are shown on the left, and enlarged part of the cells are shown on the right) and correlation maps of tandem mCherry (red) / eGFP(green) tagged LC3 in live MEF cells from ex2 KO mice. The cells where transfected with LC3 mCherry/eGFP together with either WT PLA2g6(L) (images on the top), or its human PD-associated F72L mutant (images on the bottom). Experiments were done 48 hours after transfection. Below are the summary results of comparative analysis of (a) correlation coefficient of mCherry and eGFP, and (b) the size of mCherry particles; the data show average ± SE; summary data from a total of 15 cells per condition (5 cells from each of 3 independent experiments), ***(p<0.001).   Figure 5c and Supplementary Figures 9, 10a, and 10b. Orange rectangles show the parts of the blots that have been cropped for presentation. control1  ND34770  P2  --72  --control2  ND29179  P3  --68  --control3  ND35044  P2  --77  --control4  ND29178  P3  --66  --control5  ND38530  P5  --

Animal models.
A novel PARK14 (Pla2g6) ex2 KO transgenic mouse model (B6.Cg-Pla2g6 ΔEx2-VB /J), in which ex2 of Pla2g6 gene was constitutively deleted, was custom created by GenOway (France). The strategy is outlined in Supplementary Fig. 5. Briefly, the DNA fragment containing exon 2 and adjacent intron regions of the Pla2g6 gene was isolated by PCR from the 129Sv/Pas genetic background mouse, and subcloned into the pCR4-TOPO vector (Invitrogen). To construct a targeting vector, a fragment including exon 2 (containing ATG 1 codon) and a fragment located in the third intron of the Pla2g6 gene were used to flank a neomycin selection cassette (FRT site-MC1-Neo-FRT site-loxPsite), and a distal loxP site in intron 1. The 129Sv ES cells (GenOway, France) were electroporated with the linearized targeting construct and homologous recombination was assessed in 1408 selected ES cell clones via PCR and Southern blot (Supplementary Fig. 6). One of the Pla2g6 recombined ES cell clones was microinjected into C57BL/6 blastocysts, and gave rise to male chimeras with significant ES cell contribution as determined by an agouti coat color 50%. After mating the chimeras with C57BL/6 female, the agouti colored F1 offspring were genotyped for germ line transmission of the Pla2g6 recombined allele. Floxed heterozygous Pla2g6 conditional knockout animals were generated by Flp-mediated excision of the neomycin resistance gene. The heterozygous constitutive ex2 knockout mice were generated by breeding of floxed conditional heterozygous mice with ubiquitous Cre recombinase C57BL/6 mice, and Cre-mediated excision of targeted exon 2 was verified by genotyping of tail DNA via PCR ( Supplementary Fig. 7). Oligonucleotides used as PCR genotyping primers were as follows: set 1: gtgaacacacaggctaaggctccaatcta and tcaacaagcaaaggacagacatcccac; set 2: agcagaggggcaggctgggtctctc and aggaacacagttgttgggctggggttgtc; set 3: tatcttctcgagttctctagcctccaatcctggg and cacatagaattcgtccccttgcacagcgtaatgg; and set 4: agcagaggggcaggctgggtctctc and cacatagaattcgtccccttgcacagcgtaatgg.
Heterozygote ex2 KO males were backcrossed to C57BL/6 females for 9 generations, and congenic B6.Cg-Pla2g6 ΔEx2-VB /J line was established. Due to infertility of homozygous ex2 knockout males, and the inability of homozygous females to produce/sustain live pups, cross-breeding of heterozygous mice was used to produce homozygous constitutive ex2 knockout (ex2 KO ) animals that were used in this study. Ageing male mice were used for all live animal studies, while female mice were used for MEF cell preparation. Experimental sets of homozygous ex2 KO (ex2 KO ) and wild type (WT) littermate males were housed and aged together.
Constitutive Orai1 knockout (Orai1 KO ) mice (Supplementary Reference 1 ) were kindly provided by Dr. Monica Vig. Because homozygous Orai1 KO mice have a very limited life span (up to 4-6 weeks), crossbreeding of heterozygous mice was used to produce embryos for Orai1 KO MEF cell isolation.
Animal number for each study group was determined for the experimental results to reach statistical significance with a power of 90% at p<0.05, or to demonstrate that there is no difference between the groups. Animals were maintained in an advanced pathogen-free facility with veterinary service and unlimited access to food and water. All experimental procedures were compliant with ethical regulations and approved by the Institutional Animal Care and Use Committee of Boston University.

Motor coordination tests.
Age-dependent progression of the overall motor deficit was determined during monthly evaluation of ex2 KO and WT mice for the signs and severity of clinical symptoms. Motor deficit was assessed in arbitrary units (AU) using the following scale: 0 = No abnormalities noted; 1 = first subtle signs of motor dysfunction; 2 = clear signs of movement impairment, but sustained postural stability; 3 = obvious signs of impairment in movement and occasional postural instability; 4 = strong ataxia and instability, but no difficulty with eating, drinking and grooming; 5 = very strong ataxia resulting in difficulty with keeping sternal/upright position, and frequently falls when walking, but still able to eat, drink and groom, although with some difficulty. Figure 3a present the time course of motor deficit development in age matched groups of WT and ex2 KO mice, and each point indicates median severity ±SE of the symptoms in multiple animals. The number of animals tested at different time points is identified on the graph.
The balance beam test (Supplementary Reference 2, 3 ) was used to assess the ability of ageing ex2 KO and WT mice to maintain balance while walking along a narrow (2cm) beam placed 20 cm above a soft mattress. Each mouse was placed on a beam for 2 minutes, and its movement was recorded by video camera. The total travel distance, the number of missteps (paw faults, or slips) during travel, and the number of falls off the beam were analyzed for each mouse, and data summarized for each group as mean ±SE. The numbers of animals used for each group are identified on the graphs. L-DOPA challenge test (Supplementary Reference 3 ) was performed on three age groups (12,16 and 20 months old) ex2 KO male mice with motor deficits. Control balance beam test was performed in the morning a day before L-3,4-dihydroxyphenylalanine (L-DOPA) challenge. In the morning of the following day, all mice received a single dose of L-DOPA (5, 10 or 25 mg/kg, Sigma) via peritoneal injection. Twenty minutes before L-DOPA administration, mice were given 6 mg/kg of benserazide (Sigma) to inhibit peripheral DOPA decarboxylase. The balance beam test was done 1 h after L-DOPA injection, as described above. Videorecorded data for each mouse were analyzed later, and summarized for each group as mean ±SE. The numbers of animals used for each age group are identified next to the graph.
The pole test was performed using the standard approach (Supplementary Reference 3 ). Briefly, animals were placed head-up on top of a vertical wooden pole (50 cm in length, 1 cm in diameter). The base of the pole was placed in a cage filled with bedding material. When placed on the pole, the animals need to balance on the tip of the pole to orient downward, before they can descend back into the cage. Balance time that is needed to orient downward was analyzed. After 5 training runs, and 1 day resting, each animal received 5 test trials and average of 5 measurements was determined. Summary data show results (mean ±SE). The numbers of animals used for these studies is identified on the graphs.
The rotarod test was performed using the standard approach (Supplementary Reference 2 ). Agematched ex2 KO and WT males (16-18 months of age) were tested for the length of time each mouse can maintain its balance and stay on a rotating rod (3 cm diameter). After 4 training sessions and one day resting, the mouse was placed on the rod, and then the rod started to rotate at 5 rpm with acceleration to 40 rpm within 5 minutes until the animal fell from the rod. Average of 4 measurements of the latency to fall (in seconds) was determined for each mouse and data summarized for each group as mean ±SE. The numbers of animals used for these studies is identified on the graphs.
The grip test was performed using the standard approach (Supplementary Reference 2 ). Grip Strength Meter (GSM) (Columbus Instruments, Columbus, Ohio) was used to objectively quantify the muscular strength of the forelimbs and hind limbs of age-matched ex2 KO and WT animals (16-18 months of age). All tests (4 repetitions) were performed at the same time in the morning. Strength force was normalized to body weight, which was measured each time after the test. The data were summarized for each group as mean ±SE. The numbers of animals used for these studies is identified on the graphs.

Brain slices: preparation, immunostaining and analysis.
The brains were extracted following paraformaldehyde (PFA, 4% in standard PBS) perfusion, and stored in 4% PFA at 4 o C for 24 hours. The brains were then cryopreserved in PBS containing 15% and 30% (w/v) sucrose and stored at 4 o C. Brain sections were prepared and stained using standard protocols (Supplementary Reference 4 ). Briefly, each brain was embedded and frozen in OCT Tissue-Tek, and a small cut was placed on the right side of the frozen OCT block near the right cortex for side identification. WT and ex2 KO brains were sectioned coronal (30μm thickness) with a cryostat microtome and collected as free-floating sections in a 24-well plate and stored at 4 o C. The sections containing the substantia nigra pars compacta (SNc) were collected using a staggering method, and six sets of six brain sections were collated: each set contained similar sections from the rostral, middle, and caudal parts of the SNc region. Investigators were blinded during sectioning, TH staining, unbiased stereology and analysis of PAS staining; mouse phenotypes were revealed/confirmed only after all data was generated. No samples were excluded from analysis in targeted age groups.
Blinded unbiased stereological analysis: For stereological analysis (described below) one set of slices from each brain was immunostained with tyrosine hydroxylase (TH) rabbit polyclonal antibody (Calbiochem) and 3,3'-Diaminobenzidine (DAB) HRP substrate (Vector Laboratories) using the standard techniques. Briefly, the endogenous peroxidase activity was blocked by 3% hydrogen peroxide in PBS. The sections were washed with PBS, permeabilized in 0.1% Triton X-100/PBS and blocked with 10% normal goat serum in 0.1% Triton/PBS. The TH antibody was diluted 1:1000 in blocking buffer and incubated overnight at 4 o C. The next day, the sections were washed with 0.1% Triton/PBS and incubated in Envision™+ Rabbit (Dako) solution at room temperature for 1 hour. After washing with 0.1% Triton/PBS, DAB (Vector Laboratories) staining was developed. The stained sections were mounted on slides, counterstained with Harris-modified hematoxylin and sealed with Permount. To estimate the number of TH positive (TH+) neurons in SNc of age-matched pairs of ex2 KO and WT littermates, matching sets of 6 sections from SNc area of the brain of each experimental animal were analyzed using a Nikon Eclipse E600 microscope and the Stereo-Investigator v11.01.2 software. The total number of TH+ cells in the substantia nigra was estimated using the optical fractionator technique (Supplementary Reference 5 ). Briefly, both sides (left/right) of the substantia nigra regions were outlined separately using a 4X/0.1 air objective (Nikon). Then the TH+ cells were manually counted with a 10X/0.25 air objective (Nikon) using a 60x60 µm counting frame within a 180x180µm grid with a 18µm optical dissector height. 100-200 total cells were counted per section to ensure that the coefficient of error (CE) was less than 0.1. The data were summarized for each group as average ±SE. Paired t test was used for statistical analysis of the differences within the matching pairs of WT and ex2 KO littermates. The numbers of littermate pairs used for these studies is identified on the graphs.
Periodic acid-Schiff (PAS) staining was performed using the standard procedure and kit available from Sigma (#395B). Briefly, the tissue sections were rehydrated in deionized water and placed in periodic acid solution for 5 minutes. After washing with deionized water, the sections were immersed in Schiff's reagent for 15 minutes. The slides were washed under running tap water for 5 minutes and then counterstained in Gill's Hematoxylin No. 3 for 90 seconds. After a final wash under running tap water, the tissue was dehydrated in alcohol and xylene and mounted. PAS positive cells (stained rose to magenta with blue to black nuclei) were counted in SNc, hippocampus (Hp) and M1/M2 motor cortex regions. Summary data show average number of PAS+ neurons per mm 2 (±SE) from 3 pairs of age-matched WT and ex2 KO 16 months old animals.
Immunostaining for TH and LC3 was done using the chicken polyclonal anti-TH antibody (Abcam ab76442), and rabbit polyclonal anti-LC3 antibody (Cell Signaling #2775). Briefly, the free-floating sections were washed with PBS and incubated in 0.1M glycine/PBS for 30 minutes. After additional washing, the sections were transferred to the antigen-retrieval buffer solution (10mM citric acid, 0.05% Tween-20, pH 6.0) and incubated in the steam phase of a 85°C water bath. The sections were then blocked with 10% goat serum/0.1% Triton X-100 for 60 minutes at room temperature. The TH and LC3 antibodies were diluted 1:250 and 1:20 respectively and applied to the sections overnight at 4°C. The following day, the sections were washed and the secondary antibodies goat anti-rabbit Alexa488 (Invitrogen A11034), and goat anti-chicken Alexa594 (Abcam AB150172) were diluted 1:1000 and applied for 60 minutes at room temperature. The nuclei were stained blue with 1ug/ml DAPI (Sigma D9542). Sections were mounted with Vectashield (H-1000) and sealed with nail polish.
Nissl staining (cresyl violet staining) was performed using the standard procedure. Briefly, the tissue was de-fatted for 10 minutes in xylene, followed by 10 minutes in 100% ethanol. The sections were stained with 0.1% cresyl violet acetate (Sigma C5042) solution for 10 minutes and rinsed in tap water to remove the excess stain. After a final 5 minute wash with 80% ethanol, the tissue was cleared in xylene for 5 minutes and mounted with Permount. The Nissl substance in the cytoplasm of neurons stains dark blue and confirms the reduction of TH+ neurons in the substantia nigra.

iPSC-derived A9 midbrain dopaminergic neurons.
Derivation of iPSC from MEFs was performed as previously described (Supplementary Reference 6, 7 ). Briefly, ~100,000 MEFs were plated in MEF media and transduced with the constitutive STEMCCA vector at an MOI of 2.5 for 24hr. Media was then changed to ESC media (DMEM supplemented with 15% FBS, 1X GlutaMAX, 350k units of ESGRO leukemia inhibitory factor, and 0.1mM 2-Mercaptoethanol), with media changes every second day until appearance of colonies. Putative iPSC colonies were picked and expanded for STEMCCA excision and characterization. Excision was done as described (Supplementary Reference 8 ) using Adeno-Cre infection and confirmed by PCR. Colonies positive for Alkaline Phosphatase, Oct3/4, Zfp96, Nanog and ERas were expanded and banked for neural differentiation. The cultures were routinely checked and confirmed to be negative for mycoplasma.
Differentiation of iPSC into DA neurons was done using standard protocol (Supplementary Reference 9 ) by first inducing formation of embryoid bodies in non-adherent conditions for 4 days in knockout serum replacement (KSR) media. Cells were then transferred to adherent plates and incubated in ITS media (DMEM/F12 (Gibco) + 1X ITS Supplement (Sigma I13146) + 1µg/ml Bovine fibronectin (Sigma F1141)) for 6 to 10 days to induce ectoderm formation. Cells were then expanded onto polyornithine/fibronectin coated coverslips in media containing N2 Max supplement, FGF2, FGF-8b, Shh-N and Ascorbic Acid for 5-7 days until cells reach confluency. Neural identity was confirmed by staining against III tubulin and nestin. Final differentiation into dopaminergic neurons was done by incubating the cells for 10 more days in minimal media (DMEM/F12 (Gibco) + 1X N2 Max + 200µM Ascorbic Acid (Sigma A4403)). Confirmation of DA neuron identity was done by staining the cells for TH (Abcam ab76442), Dopamine transporter (DAT, Abcam ab5990) and Vesicular monoamine transporter 2 (VMAT2, Abcam 70808).