The chordoma cell lines U-CH17P, U-CH17M, U-CH17S, U-CH11R, UM-Chor1 and MUG-chor1 were generously provided by the Chordoma Foundation (www.chordomafoundation.org). The cells were grown IMDM/RPMI 1640 (4:1; Sarstedt Inc) supplemented with 10% FBS and penicillin-streptomycin-glutamine (PSG; 100 U/mL penicillin, 100 µg/mL streptomycin, 292 µg/mL L-glutamine) (Gibco, Ontario, Canada). The cells were maintained in a humidified 37°C incubator with 5% CO2 until confluency.
Silica-bead based cell surface protein capture
Cell surface protein capture was performed as previous described (11, 12) with minor modifications. The chordoma cell lines (U-CH17P, U-CH17M, U-CH17S and U-CH11R) were washed three times with ice-cold MBS buffer (25 mM MES, 150 mM NaCl pH 6.5) and a monolayer of cells were overlaid with 20 mL of 1% colloidal silica bead (LUDOX® CL colloidal silica suspension in water) with gentle shaking for 15 minutes at 4°C. Excessive beads were removed, and the plates were washed three times with MBS buffer. 20 mL of 0.1% Polyacrylic acid (PAA)in MBS buffer was added to the cells with gentle shaking for 15 minutes at 4°C. PAA was removed from the plates and fresh sucrose/HEPES buffer (250 mM sucrose, 25 mM HEPES pH 7.4 and 20 mM KCL) was added. Cells were scraped from the plate and centrifuged at 300 xg for 5 minutes. The supernatant was discarded, and the pellet was added to fresh sucrose/HEPES buffer. Cells were lysed by 3 cycles of sonication (35% amplitude, 10 seconds). A discontinuous Histodenz density gradient in sucrose/HEPES was prepared at different concentrations (45%,50%,55% and 60%) and layered in a 13.2 mL (Beckman) ultracentrifugation tube. The samples were diluted with Histodenz to a final concentration of 20% and added on top of the Histodenz gradient. The samples were centrifuged at 100,000 xg for 2 hours in a SW 41Ti rotor (Beckman). After the centrifugation, the density gradient layers were removed, and the resulting pellet was washed with sodium carbonate (pH 12) solution via rotation for 15 minutes at 4°C. The beads were centrifuged at 5,000 xg for 20 minutes in a benchtop centrifuge and the supernatant was removed. Elution buffer (400mM NaCl, 25mM HEPES, 1% Triton X- 100, pH 7.4) was added to the beads and rotated overnight at 4°C to elute membrane proteins from the silica beads. The eluted proteins were precipitated with ice cold acetone overnight at -20°C. The samples were pelleted at 10,000 xg for 10 mins and the resulting pellet was solubilized with 100 µL of lysis buffer (50% (v/v) 2,2,2,-Trifluoroethanol (TFE) and 50% PBS).
Subcellular fractionation and protein extraction
Subcellular fractionations were performed as previously described (13, 14) with minor modifications. Chordoma cell lines (U-CH17P, U-CH17M, U-CH17S and U-CH11R) were pelleted and washed three times with PBS. Cells were homogenised in lysis buffer (50 mM Tris-HCL (pH 7.4), 5mM MgCl2, 0.1% Triton X-100 and Protease inhibitors) and kept on ice for 10 mins, and further homogenised with a loose-fitting pestle. Sucrose was added to the lysates to a final concentration of 250 mM (isotonic solution). All the subsequent steps were done at 4°C. The lysates were centrifuged at 800 xg for 15 minutes in a benchtop centrifuge (Eppendorf 5430R) at 4°C to separate the nuclear fraction. The resulting supernatant served as the source of cytosol, mitochondria and microsomes (i.e., mixed membranes). The nuclear pellet was further resuspended in 2.5 mL of cushion buffer (2M sucrose, 50 mM Tris-HCl, 5 mM MgCl2, 1 mM dithiothreitol (DTT) and Protease inhibitors – Roche) and overlaid on 2 mL of cushion buffer in ultra-clear open top 5 mL ultracentrifugation tube (Beckman) and centrifuged at 80,000 xg for 45 minutes (Beckman SW 55Ti rotor). The mitochondrial fraction was isolated from the crude lysate by centrifugation at 8000 xg for 15 minutes. The resulting pellet was washed with a lysis buffer and spun again to retrieve the mitochondrial fraction. The resulting supernatant was centrifuged at 150,000 xg for 1 hour (Beckman SW 55Ti) to isolate the microsomal pellet (mixed membranes). The supernatant served as the cytosolic fraction. Nuclear proteins were extracted from the nuclear fraction with a lysis buffer (20 mM HEPES, 400 mM NaCl, 0.2mM EDTA) and rotated for 30 mins at 4 degrees. The pellet was passed through 18-guage needle several times and centrifuged at 9000 xg for 10 mins to isolate the soluble nuclear fraction and insoluble pellet. The resulting organelle pellets (mitochondria, nuclear and microsome) were lysed in 100 µL of lysis buffer (50% (v/v) 2,2,2,-Trifluoroethanol (TFE) and 50% PBS).
Sample preparation for shotgun proteomics
The pellets obtained from the subcellular fractions were lysed by repeated freeze-thaw cycles (5 cycles, switching between a dry ice/ethanol bath and 60°C water bath) in lysis buffer. Samples were sonicated on a ultrasonic block sonicator for five 10s cycles at 10 watts per tube (Hielscher VialTweeter) followed by extraction at 60°C for 1 hour. Disulphide bonds were reduced with 5mM DTT, followed by 30 min incubation at 60°C. Free sulfhydryl groups were alkylated by incubating the samples with 25 mM iodoacetamide in the dark for 30 minutes at room temperature. The samples were diluted (1:5) with 100 mM ammonium bicarbonate (pH 8.0) and 2 mM CaCl2 was added. Proteins were digested overnight with 2 µg of trypsin/Lys-C enzyme mix (Promega) at 37°C. Peptides were desalted by C18-based solid phase extraction, then dried in a SpeedVac vacuum concentrator. Peptides were solubilized in mass spectrometry-grade water with 0.1% formic acid. Liquid chromatography was directly coupled to an Orbitrap Fusion Tribrid (Thermo Scientific) and data was acquired as previously described (15, 16). Raw files were searched using the MaxQuant software (version 22.214.171.124) against a Uniprot human sequence database (number of sequences 42,041) with an FDR set to 1% for positive peptide spectral matches and protein identification using a target-decoy strategy (14). Searches were performed with maximum of two missed cleavages, oxidation of methionine residues as a variable modification, and carbamidomethylation of cysteine residues as a fixed modification. Intensity-based absolute quantification (iBAQ) and label-free quantitation (LFQ) were enabled, with match between runs function disabled due to the differences in organellar proteomes. Subsequent analyses were performed using the proteinGroups.txt file. Contaminant sequences and matching decoy were removed, and proteins identified with two or more unique peptides were carried forward. iBAQ intensities were used for protein quantitation. The data was median normalised and missing values were imputed with low values (between 1 and 1.2 log2 values).
Chordoma cell line pellets and frozen chordoma tissue samples were lysed in RIPA buffer (50 mmol/L Tris pH 7.5, 150 mmol/L NaCl, 2 mmol/L EDTA pH 8.0, 0.5% (v/v) Triton X-100, and Complete protease inhibitor cocktail - Roche, Switzerland). The cells were kept on ice for complete lysis. Cell debris was removed by centrifugation at 16,000 xg for 10 mins at 4°C. The protein concentration was determined by BCA assay (Thermo scientific). Commercially available normal tissue lysates from various organs (Brain cortex, cerebellum, skin, stomach, esophagus and spleen) were purchased from Takarabio (USA). Gels were loaded with 10 µg of protein lysates per lane and proteins were separated on 7, 8 or 13% SDS-PAGE gels. The resolved proteins were wet transferred to polyvinylidene fluoride membrane (PVDF) and membranes were incubated in 5% (w/v) milk in Tris-buffered saline Tween-20 (TBST; 10 mmol/L Tris-Base, 150 mmol/L NaCl, 0.05% Tween-20; pH 7.4) for 1 hour. After blocking, membranes were incubated with primary antibodies (1:1000 mouse anti-human monoclonal PLA2R1, [Sigma], 1:1000 mouse anti-human polyclonal SLC6A12 [ThermoFisher Scientific], 1:1000 rabbit anti-human monoclonal Brachyury [Cell Signaling], 1:1000 mouse anti-human monoclonal Lamin B1 [abcam] and 1:1000 rabbit anti-human monoclonal ß-actin [Novus biologicals]) overnight at 4°C.
A cohort of 25 chordoma patients with formalin-fixed paraffin (FFPE) embedded slides and clinical follow up data were obtained from the University Health Network Brain Tumor Biobank (REB# − 18-5820). Slides with 5 µm FFPE tissue sections were rehydrated with serial dilutions of ethanol followed by water and pH 6 sodium citrate dihydrate buffer was used for heat-mediated antigen retrieval. A 3% hydrogen peroxide in methanol solution was utilized to block endogenous peroxidase activity. Blocking solution (5% bovine serum albumin in phosphate buffered saline plus 0.1% Triton X-100) was applied to slides for 1 hour at room temperature. Subsequently, primary antibodies including anti-SLC6A12 (Invitrogen, PA5-57099, rabbit polyclonal antibody) and anti-PLA2R (Millipore Sigma, MABC942, mouse monoclonal antibody) were applied overnight at 4˚C diluted (1:200, 1:200, 1:250, respectively) in blocking solution. A 1 hour incubation with secondary antibody was performed followed by processing with the DAKO polymer-HRP system and DAB peroxidase kit, counterstaining with hematoxylin, tissue dehydration, and slide cover slipping. Whole slide digital scanning was performed on all slides and images were analyzed using the HALO Image Analysis Platform (Indica Labs). Each slide was annotated with multiple regions of interest to delineate chordoma tumor tissue. Three independent reviewers assessed slides for all cases (JAZ, OS, and an experienced neuropathologist AG) and representative images were selected. Proportions of stain positive cells were quantified using the HALO software algorithm, defined to identify cells with either membranous or cytoplasmic staining as a fraction of all cells. This algorithm was applied to all annotated tissue sections in an unbiased systematic manner. Wilcoxon's rank sum test was used to compare values between skull base and spinal chordomas. Univariable and multivariable Cox models were utilized to assess the prognostic utility of stain proportions together with known major prognostic clinical factors (extent of surgical resection and radiotherapy use). For survival analyses results, the upper tertile of PLAR2 values (samples with higher marker positivity) were compared to the lower two tertiles of values.
siRNA knockdown and Clonogenic assay
The chordoma cell lines (U-CH17M, U-CH17S and UM-chor1) were seeded in 6-well plates at a density of 2,000 cells per well in DMEM/RPMI media and transfected with three siRNA for PLA2R1 (Cat# SR307882, Origene), SLC6A12 (Cat# SR304423, Origene) and scrambled siRNA (negative control) (Cat# SR30004, Origene) at a concentration of 5 nM using lipofectamine RNAiMax transfection reagent (Invitrogen). After 2 weeks colonies were stained with crystal violet staining (0.5% crystal violet, 25% methanol) and the quantification of the colonies was performed using ImageJ (version).
All the proteomics experiments were performed in triplicates. Applicable data were analyzed and represented using the R statistical environment (v3.6.3). Differential expression analysis was performed using unpaired Welch's t-test for statistical analysis, and Benjamini & Hochberg adjusted p-value < 0.05 deemed as statistically significant. Visualization in R was performed using the ggplot2 (3.2.1) and Complexheatmap (v2.2.2).