Distinct responses of neurons and astrocytes to TDP-43 proteinopathy in amyotrophic lateral sclerosis

Smethurst et al. develop a human, clinically relevant modelling platform that recapitulates key aspects of sporadic ALS. They demonstrate cell type-specific vulnerability to both seeded aggregation (induced by post-mortem tissue extracts) and upon treatment with highly purified recombinant TDP-43 oligomers.


Human iPSC derived MN and AC culture
For all experiments a minimum of 3 different control hiPSC lines were utilized (each in technical triplicate). The hiPSC derived MNs and ACs were cultured and differentiated as previously described (Hall et al., 2017), Briefly, for differentiation, control iPSC-derived ventral spinal neural progenitor cells (NPCs) were propagated in N2B27 medium (DMEM/F12 Glutamax, Neurobasal, L-Glutamine, N2 supplement, non-essential amino acids, B27 supplement, β-mercaptoethanol (all from Life Technologies) and insulin (Sigma) with FGF-2 (Peprotech) for up to 30 days. For terminal differentiation of the NPCs to MNs the cells were plated out onto 0.07% polyethyleneimine (PEI) (Sigma) and Geltrex (Life Technologies) coated 24 well plates into N2B27 media containing compound E (Enzo Life Sciences) to promote cell cycle exit. To produce ventral spinal ACs the NPCs were propagated in FGF-2 for a minimum of 60 days, during which time they had become glial precursor cells (GPCs). FGF-2 withdrawal and co-treatment with 10ng/ml Bone Morphogenetic protein 4 (BMP4) (R&D systems) and 10ng/ml Leukaemia inhibitory factor (LIF) (Sigma) were used to differentiate for a minimum of 3 weeks.

TDP-43 seeding to hiPSC derived MNs and ACs
The hiPSC-MNs were differentiated for a minimum of 3 days and then 5μg of either control spinal cord extract or serially passaged ALS spinal cord extract containing seeds of pathological TDP-43 were either mixed with 4μl of PULSin (Polyplus transfection) reagent in HEPES buffer for 15 mins or with 1.5ul of Lipofectamine 3000 for 5 mins, and then complexes were added to the cells and incubated for 6 hours. After this the complexes were removed, and the media was changed to fresh N2B27 media with compound E. The cells were then fixed at either 3, 7-or 14-days post treatment. For proteasome inhibition, the cells were co-treated with 2µM of MG132 (Sigma) for 6 hours at the same time as transfection of the serially passaged ALS SC.

Spreading and propagation of TDP-43 aggregates in co-culture
To investigate the spreading and propagation of TDP-43 aggregates between MNs and ACs we performed co-culture experiments. For MN to AC propagation, MNs were plated on to 24 well plate coverslips and seeded with serially passaged sALS (spALS) SC extract and MG132 treatment and then ACs were added in a 1:1 ratio the day after the MNs had been washed 3 times with N2B27 media changes. Cells were then fixed in 4% PFA at 3, 7 and 14 days post co-culture. For AC to MN propagation, the ACs were seeded with spALS SC extract and MG132 treatment and washed extensively with media changes. The following day the ACs were dissociated with trypsin and added to MNs in a 1:1 ratio. These co-cultures were then left for 3, 7 and 14 days with regular media changes and fixation with 4% PFA at each of these time points.

Cytotoxicity assay
Toxicity was assayed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2Htetrazoliumbromide (MTT) assay which was performed using MTT salt (Sigma) resuspended at 1mg/ml in sterile water. Cells were plated out in 96 well plates at 20,000 cells per well treated with either control or ALS treated cell extracts and then 20μl of the 1 mg/ml MTT solution was added to each well and left to incubate at 37°C for 2-4 h. The media from each well was decanted and replaced with 100μl of dimethyl sulfoxide (DMSO) and agitated until the entire purple formazan product was dissolved and the plate was read at 570 nm.
Absorbance readings were taken as a percentage of the mean value of non-seeded untreated control cells.
Dot blots 2ul of sample was added to a nitrocellulose membrane and allowed to air dry. The membrane was then blocked in 5% non-fat milk for 30 mins then incubated in primary antibody (polyclonal rabbit anti-TDP-O (1:500, Gift from Dr Yu Run Chen)) for 1 hour in 5% non-fat milk in PBS + 0.01% Tween (PBST) and then washed 3 times for 5 mins each in PBST. The membrane was then incubated with secondary fluorescent IR-Dye 800 at (1:10,000, Abcam) in 5% non-fat milk in PBST for 1 hour and then membrane was washed again 3 time in PBST and developed using the Odyssey Scanner (Li-Cor) Construction of vector DNA encoding human TAR DNA-binding protein 43 (TDP 43) was purchased from Eurofins Scientific UK. The codons were optimised for expression in HEK293 cells. The expression vector was prepared using the pTriEX Ek/LIC cloning kit (Novagen, USA). Briefly, the gene was amplified by PCR to incorporate the LIC extensions, gel purified, treated with T4 DNA polymerase, and the resulting fragment was annealed with the pTriEX-4 Ek/LIC vector according to the manufacturers' instructions. The resulting TDP-43-pTriEX Ek/LIC plasmid was transformed into NovaBlue Competent Cells (Novagen). Positive clones were selected on LB agar plates containing ampicillin (100μg/ml) and were analysed for presence of insert by PCR and then sent for DNA sequencing to check the sequence.

Characterisation of TDP-43 by SEC-MALS
The protein was loaded onto a Size Exclusion Chromatography Superdex S200 10/30 (GE Healthcare) equilibrated in 20mM Tris, 150mM NaCl, pH10 and eluted using a HPLC instrument (Agilent) chained with a Dawn HeleosII MALS detector and Optilab dRX refractometer (Wyatt Technology). Molar masses were calculated from the intensity of scattered light at 18 different angles, and SEC elution volume using ASTRA software (Wyatt Technology).

Image analysis
All images were analysed using the Volocity software. Nuclear cytoplasmic TDP-43 was calculated using imageJ by separating the TDP-43 and DAPI channel and thresholding and making the DAPI channel binary to use as a template to make images of the nuclear and cytoplasmic TDP-43 only. These images where then measured for pixel intensity and the integrated intensity was used as the final measurement value. These values were added together to make a total value and then percentage of each channel was calculated as the final values. Cell counting was performed using 5 random fields from each coverslip and counted manually using the cell counter facility on the Volocity software.

Statistical analysis
An unpaired two-tailed student's t-test was used when comparing between two individual groups cases to generate a statistical p value. A one-way ANOVA was used when comparing 2 or more groups with a post hoc Tukey test to compare all groups. Any p value below 0.05 was considered to be statistically significant (*p<0.05, **p<0.01, ***p< 0.001).

Compliance with ethical standards
For human iPSC work, informed consent was obtained from all patients and healthy controls in this study. Experimental protocols were all carried out according to approved regulations and guidelines by UCLH's National Hospital for Neurology and Neurosurgery and UCL Queen Square Institute of Neurology Joint Research Ethics Committee (09/ 0272). The human postmortem spinal cord samples were obtained from the tissue bank NeuroResource, UCL Queen Square Institute of Neurology, London, UK. Samples were donated to the tissue bank with written tissue donor informed consent following ethical review by the NHS NRES Committee London-Central and stored under a Research Sector Licence from the UK Human Tissue Authority (HTA).

Figure S1. TDP-43 seeded aggregation in non-overexpressing HEK293 cells.
Immunostaining demonstrating the seeded aggregation of endogenous TDP-43 3 days after the transfection with either PULSin or Lipofectamine of spALS SC extract that is not attributable to the extraction procedure itself (Control SC). White arrows indicate cells with nuclei cleared of endogenous TDP-43 in the presence of cytoplasmic TDP-43 aggregates. SC = spinal cord, spALS = serially passaged sporadic ALS post-mortem tissue, pTDP-43 = phospho TDP-43 (pS409/410). All scale bars represent 10 μm.