Antibodies and probes
Rabbit anti-QSOX2 antibody (ab121376, RRID:AB_11128050), Monoclonal ANTI-FLAG® M2 antibody (Sigma Aldrich F3165, RRID:AB_259529), Rabbit anti-Phospho-Stat5 antibody (Tyr694) (Cell Signalling Technology D47E7, RRID:AB_10544692), Rabbit anti-phospho-Stat3 antibody (Tyr705) (Cell Signalling Technology D3A7, RRID:AB_2491009), Rabbit anti-phospho-Stat1 antibody (Tyr701) (Cell Signalling Technology Clone 58D6, RRID:AB_561284), Rabbit anti-Tom20 antibody (Cell Signalling Technology D8T4N, RRID:AB_2687663), Rabbit anti-phospho-DRP1 (Ser616) (Cell Signalling Technology D9A1, RRID:AB_11178659), Rabbit anti-GAPDH antibody (ab9485, RRID:AB_307275), Mouse anti-Actin beta monoclonal antibody (ab6276, RRID:AB_2223210), Mouse anti-Histone Deacetylase 1 antibody (Santa-Cruz biotechnology sc-81598, RRID:AB_2118083), Rabbit anti-GFP antibody (ab290, RRID:AB_303395), Fluorescent probe - MitoTracker™ Red (M22425, Thermo Fisher Scientific), Rat anti-Human phospho-STAT5a/b Y694/Y699 (R&D Systems Clone MAB4190), Mouse anti-alpha Tubulin antibody DM1A (ab7291, RRID:AB_2241126), Total OXPHOS Rodent WB Antibody Cocktail (ab110413, RRID:AB_2629281), Rabbit anti-Phospho-Akt (Ser473) antibody (Cell Signalling Technology D9E, RRID:AB_2315049), Rabbit anti-Akt (pan) antibody (Cell Signalling Technology C67E7, RRID:AB_915783), Rabbit anti-MAP Kinase (ERK-1, ERK-2) antibody (Sigma Aldrich M5670, RRID:AB_477216), Monoclonal anti-MAP Kinase, Activated (Diphosphorylated ERK-1&2) antibody (Sigma Aldrich M9692, RRID:AB_260729), Goat anti-Rat IgG (H+L) Highly Cross-Adsorbed Secondary Antibody Alexa Fluor Plus 488 (A48262 RRID:AB_2896330), Goat anti-mouse IgG (H+L) Highly Cross-Adsorbed Secondary Antibody Alexa Fluor Plus 488 (A32723, RRID:AB_2633275), Goat anti-Rabbit IgG (H+L) Highly Cross-Adsorbed Secondary Antibody, Alexa Fluor Plus 647 (A32733, RRID:AB_2633282), IRDye® 800CW Goat anti-Mouse IgG (RRID:AB_10793856), IRDye® 800CW Goat anti-Rabbit IgG (RRID:AB_10796098), IRDye® 680RD Goat anti-Mouse IgG (RRID:AB_2651128), IRDye® 680RD Goat anti-Rabbit IgG (RRID:AB_2721181), Tetramethylrhodamine, ethyl ester (TMRE, ab113852).
QSOX2 Variant Detection and Confirmation
Variants in QSOX2 were found on whole exome/genome sequencing and confirmed by Sanger sequencing using primers amplifying exon 8 (forward: 5′-CCAGGACAGGGAGACTTG-3′ and reverse: 5′-GGTGGAGAGCACCTCAG-3′), exon 10 (forward: 5′-CCCAGTCAAGAAGGCAG-3′ and reverse: 5′-AGTACATGCCTTTGCACAC-3′) and exon 12 (forward: 5′-GAGTGGGAGTCCGGTTG-3′ and reverse: 5′-CATCCGATGTGAAACCAG-3′) of QSOX2. Pathogenicity of both variants was evaluated using a combination of predictive tools: Sorting Intolerant from Tolerant, Polymorphism Phenotyping v2, Combined Annotation Dependent Depletion and Mutation taster.
Protein Structure Modelling and Thermostability Analysis
Protein 3D modelling of the Alpha Fold Protein Structure Database35 QSOX2 crystal structure Q6ZRP7 was performed using the tool PyMOL (Schrodinger, LLC. 2010. The PyMOL Molecular Graphics System, Version X.X) with thermostability of the missense mutant protein assessed using computational platforms: DynaMut36, I-Mutant37, SDM38, DUET39, MUpro_SVM40 and mCSM41.
UK Biobank (UKBB) data analysis
We included 420,162 samples of European ancestry in the UKBB for exome-wide association tests. For the 450K release of exome-sequencing data in the UKBB, we performed individual and variant level quality control procedures previously described by Gardner et al.42 Variants were annotated using ENSEMBL Variant Effect Predictor (VEP) v10443. Protein truncating variants were defined as stop gain, frameshift, splice acceptor and splice donor variants. The burden test assumed the presence or absence of variants of interest in a gene as an indicator variable, which was regressed against the phenotype in a linear mixed model using BOLT-LMM v2.3.644 on the UKBB Research Analysis Platform (RAP). Covariates adjusted in the burden test included age at assessment (UKBB Data-field 21003), age squared, the whole-exome sequencing batches (as a categorical variable, either 50K, 200K, or 450K) and the first 10 genetic principal components (UKBB Data-field 22009.1-10).
Quality check for rs61744120 imputation and data analysis
To study the quality of the imputed SNP rs61744120, we compared the genotypes between WGS and FinnGen imputed data in FINRISK participants where data was available for both formats. The FINRISK cohorts comprise the respondents of representative, cross-sectional population surveys that are carried out every 5 years since 1972 (to assess the risk factors of chronic diseases and health behaviour in the working age population) in 3-5 large study areas of Finland. THL Biobank host samples were collected in the following survey years: 1992, 1997, 2002, 2007, and 2012. Genome-wide imputation was done as part of the FinnGen project using Sequencing Initiative Suomi (SISu) project data as reference.
Individuals with the minor/minor genotype were identical between WGS and both releases of the imputed data. However, there were variations in minor/major and major/major genotypes in 10 individuals producing an error rate of 0.25%. The additive genetic association model was utilised to estimate the proportional risk of disease i.e. reduction in height associated with this single nucleotide polymorphism. Calculation of height standard deviation scores based on raw height data of minor/minor homozygotes was performed using Finnish population based references for healthy subjects as outlined by Saari et. al (2011)45.
In-vitro splicing assay
An in-vitro splicing assay was designed, as previously described by Maharaj et al.46, using the Exontrap vector pET01 (MoBiTec). A designated DNA sequence, including exons 7 and 8 of QSOX2 as well as intervening introns, was selectively cloned into the multiple cloning site of the exontrap splicing machinery. Clones were selected and verified by sanger sequencing using vector-specific primers ET 06 (forward: 5′-GCGAAGTGGAGGATCCACAAG-3′) and ET 07 (reverse: 5′-ACCCGGATCCAGTTGTGCCA-3′). Site directed mutagenesis to generate the c.1055C>T (p.T352M) variant was performed using the QuikChange II XL site-directed mutagenesis kit (Agilent, 200521) according to the manufacturer’s instructions. Empty pET01 vector, QSOX2-WT and variant clones were transfected into mammalian HEK293 cells for 24 hours followed by RNA extraction. cDNA synthesis was performed using the vector-specific hexamer GATCCACGATGC and RT-PCR conducted using pET01 primer 02 (forward: 5′-GAGGGATCCGCTTCCTGGCCC-3′) and primer 03 (reverse: 5′-CTCCCGGGCCACCTCCAGTGCC-3′). PCR products were analysed on a 2% agarose gel and bands gel extracted, column purified and confirmed by Sanger sequencing.
Site-directed Mutagenesis
Site-directed mutagenesis of a QSOX2 (NM_181701.4) Human Tagged ORF Clone (GenScript, ID: OHu07590C) was performed using the QuikChange II XL site-directed mutagenesis kit (Agilent, 200521) according to the manufacturer’s instructions. Primers for generation of QSOX2 variants were designed using the online tool https://www.agilent.com/store/primerDesignProgram.jsp.
Primary fibroblast cell culture
Fibroblast isolation was performed from skin punch biopsies of proband 2, parents and a healthy control. Immediately after excision, the specimen was incubated in DMEM high glucose supplemented with 10% Fetal Bovine Serum (FBS) and 1% Penicillin/Streptomycin. The skin specimen, chopped into 1mm cubes, was subsequently digested using a mixture of nutrient media (DMEM high glucose supplemented with 10% FBS, 1% penicillin/streptomycin and 1:100 non-essential amino acids), 0.25% collagenase and 0.05% DnaseI. The mixture, incubated at 37 °C in 5% CO2 overnight, was centrifuged at 1000rpm for 5min and the pellet resuspended in fibroblast primary culture media (DMEM high glucose with 10 % FBS, 1% penicillin/streptomycin and 1:100 non-essential amino acids). The resuspended mixture was plated in a 0.1% gelatin coated T25 flask and left in an incubator at 37°C in 5% CO2 until fibroblast cultures were established.
Cell culture, GH/IGF-1 stimulation and nuclear fractionation
Dermal fibroblasts and C28/I2 chondrocytes were cultured in DMEM high glucose supplemented with 10% FBS and 1% penicillin/streptomycin. HEK 293-hGHR cells47 were similarly cultured in DMEM high glucose base media with selection antibiotic, G-418 (Sigma Aldrich) at a concentration of 400μg/ml. Prior to GH treatment, cells were serum deprived for at least 24hours in serum-free media supplemented with 0.1% Bovine serum albumin (BSA). Optimal standardised human GH (Cell Guidance Systems) concentration (500ng/ml) was used for all experiments with a stimulation time of 20minutes at 37 °C in 5% CO2. For IGF-1 stimulation, cells were similarly serum deprived for 24hours prior to treatment with recombinant human IGF-1 (Peprotech, 100ng/ml) for 30minutes at 37 °C in 5% CO2. Nuclear and cytoplasmic cell fractions were prepared using the NE-PER™ Nuclear and Cytoplasmic Extraction Reagents (Thermo Fisher) according to the manufacturer’s instructions. Cross contamination of cellular fractions was negligible.
CRISPR-Cas9 Engineered Knockout of QSOX2 in C28/I2 Human Chondrocyte Cell Line
CRISPR gene editing was achieved utilizing the protocol outlined by Ran et. al48. Guide sequences were designed using the online CRISPR Design Tool (http://tools.genome-engineering.org). The single guide RNA oligos (Forward 5’-GGGACCTGCGCTGAGAG-3’ and Reverse 5’-GCGGTAAGGAAAGAAATACGG-3’) were then cloned into pSpCas9(BB)-2A-GFP (PX458), a gift from Feng Zhang (Addgene plasmid #48138; http://n2t.net/addgene:48138; RRID:Addgene_48138, https://www.addgene.org/48138)48 and introduced into immortalized C28/I2 (Sigma Aldrich™, Catalog no. SCC043) human chondrocyte cells via transfection using Lipofectamine™ 3000 according to manufacturer’s instructions. After 72 hours, GFP-positive cells were cell sorted by fluorescence-activated cell sorting into prepared 96-well plates, to ensure single cell clonal expansion. Colonies were expanded and genotyped after 4 to 6 weeks.
Co-immunoprecipitation
In order to probe the interaction between QSOX2 and endogenous STAT5B, 7µg of QSOX2 cDNA was transfected into 2x106 HEK 293-hGHR cells (10cm dish) using Lipofectamine™ 3000 according to manufacturer’s instructions. After 48hours cells were lysed with 0.5% NP-40 buffer (0.5% NP-40, 20 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, 10% glycerol, 1 mM PMSF). The lysate was added to a micro-centrifuge tube, placed on a rotary mixer for 1 hour at 4°C, then centrifuged for 20 minutes at 14,000g. Protein concentration was quantified using a Bradford protein assay (Bio-Rad). Lysate was equally divided into three separate micro-centrifuge tubes and Immunoprecipitation carried out at 4°C overnight following addition of primary antibodies (5µg anti-STAT5B, 5µg anti-QSOX2 and 5µg Goat anti-mouse IgG - H&L - Fab Fragment Polyclonal Antibodies) and Protein G Sepharose beads (Sigma-Aldrich). Bound proteins were extracted from coated beads and analysed by immunoblotting.
Pull down assay
To assess whether the presence or absence of QSOX2 impacts dimerization of STAT5B, QSOX2 wild type and knockout C28/I2 cells were transfected in parallel with pCMV6-AC-GFP-STAT5B and pCMV6-AC-STAT5B-FLAG plasmids using Lipofectamine™ 3000 according to manufacturer’s instructions. After 12hours, complete media was removed and cells cultured in serum free media supplemented with 0.1% BSA for a further 24hours. Cells were treated with GH 500ng/ml for 20 minutes prior to addition of lysis Buffer (50mM Tris HCl, pH 7.4, with 150mM NaCl, 1mM EDTA, and 1% TritonX-100). Lysates were placed on a rotary mixer for 1hour at 4°C prior to clarification by centrifugation at 14,000xg for 15minutes. ANTI-FLAG M2-Agarose Affinity Gel beads (Sigma Aldrich) were equilibrated with TBS prior to addition of protein samples and incubated at 4°C overnight on a rotary mixer. Coated beads were collected and washed with TBS (twice). Samples were eluted using SDS sample buffer, separated by SDS-PAGE gel electrophoresis and probed by immunoblotting using monoclonal anti-FLAG and monoclonal anti-GFP antibodies.
Immunoblotting
Whole cell lysates were prepared by addition of RIPA buffer (Sigma Aldrich) supplemented with protease and phosphatase inhibitor tablets (Roche). Protein concentrations were quantified using a Bradford protein assay (Bio-Rad) and lysates denatured by addition of SDS sample buffer 6× (Sigma Aldrich) and boiled for 5 minutes at 98°C. A 20-µg bolus of protein was loaded into the wells of a 4% to 20% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel (Novex) prior to electrophoretic separation using MOPS buffer. Protein transfer to nitrocellulose membrane was achieved by electroblotting at 15 V for 45 minutes. The membrane was blocked with either 5% fat-free milk or BSA in tris-buffered saline/0.1% Tween-20 (TBST) and left to gently agitate for 1 hour. Primary antibody was added at a concentration of 1:1000 with housekeeping control at a concentration of 1:10,000. Primary antibody incubation was left overnight at 4°C with gentle agitation. The membrane was then washed for 5 minutes (3 times) with TBST. Secondary antibodies were added at a concentration of 1:5000 to blocking buffer and the membrane incubated at 37°C for 60 to 90 minutes. The membrane was subsequently washed 3 times (5 minutes each) with TBST and visualized with the LI-COR Image Studio software for immune-fluorescent detection.
Mitochondrial Membrane Potential Assay
Fibroblasts were seeded in clear bottomed 96 well plates (1x105 cells/well) and cultured at 37°C in 5% CO2 overnight. Culture medium was aspirated, replaced with serum free base media supplemented with 0.1% BSA and cells incubated at 37°C for a further 8hours. GH (500ng/ml) and depolarisation control carbonilcyanide p-triflouromethoxyphenylhydrazone, FCCP (20μM) were added to relevant wells and plate incubated at 37°C in 5% CO2 for 10minutes. Tetramethylrhodamine ethyl ester (TMRE) was then added at a concentration of 500nM and cells incubated for a further 20minutes at 37°C in 5% CO2. Media was aspirated from wells and replaced by 100μl of PBS/0.2% BSA. This process was repeated prior to fluorescence measurement (Ex/Em = 549/575nm) using the CLARIOstar Multimode Plate Reader (BMG Labtech).
GHRE Luciferase reporter assay
HEK 293-hGHR cells were seeded in six-well plates and transiently transfected with 2.5μg DNA per well: 1.0μg pGL2 8xGHRE (growth hormone response element) luciferase reporter plasmid, 0.5µg STAT5B WT, 0.5µg QSOX2 WT/mutant cDNA /empty vector and 0.5µg pRL-SV40 (Renilla luciferase). After overnight incubation, culture medium was replaced with serum free DMEM supplemented with 0.1% BSA and incubated for a further 8hours. Cells were stimulated with GH (500 ng/ml) for 24 hours and lysates collected and assayed using the Dual-Luciferase® Reporter Assay System (Promega, E1910) on the CLARIOstar Multimode Plate Reader (BMG Labtech).
Immunofluorescence
Cells seeded on glass coverslips (24 well plate) were fixed with 4% paraformaldehyde for 15minutes. Cells were then washed three times in PBS and permeabilized in ice cold 100% methanol for 10minutes at -20°C. After three further PBS washes, coverslips were incubated in Blocking buffer (1X PBS / 5% goat serum / 0.3% Triton™ X-100) at room temperature for 60minutes. Primary antibody (rat anti-STAT5B, rabbit anti-QSOX2, rabbit anti-Tom20, rabbit anti-phospho-DRP1, mouse anti-alpha tubulin) reconstituted in dilution Buffer (1X PBS / 1% BSA / 0.3% Triton™ X-100 buffer) was added to cells and left at 4°C overnight with gentle agitation. Cells were then washed three times with PBS prior to addition of fluorescent secondary antibody and left at room temperature for 90minutes (protected from light). Coverslips were counterstained with DAPI and washed with PBS to mounting on microscope slides.
MitoTracker immunostaining
For MitoTracker staining of mitochondria, fibroblast and C28/I2 cells were seeded at a density of 2.5 × 103 per well (24 well plate) on glass coverslips. The MitoTracker lyophilized probe was reconstituted in anhydrous DMSO to a stock concentration of 1mM. A working concentration of 100nM was established by dilution in nutrient media prior to addition to cells and incubated at 37°C in 5% CO2 for 30minutes. After incubation, cells were washed twice with phosphate buffered saline (PBS) and coverslips fixed with 4% paraformaldehyde for 15minutes. Permeabilization was achieved by addition of 0.2% TritonX-100 for 5minutes. Coverslips were counterstained with DAPI and washed with PBS to mounting on microscope slides. Images were obtained using the 63x oil objective of the confocal Laser scanning microscope 710.
Generation of Nanoluc SmBiT and LgBiT (STAT5B-N-small BiT and QSOX2-N-Large BiT fusion vectors) by Gibson Assembly
Wild type STAT5B and QSOX2 constructs were generated by cloning Nanoluc small BiT (SmBiT) and large BiT (LgBiT) sequences to the N terminus of each receptor using a flexible Glycine-(gly)-Serine-(ser) linker by Gibson assembly. Primers were designed using the Benchling assembly wizard (Benchling Biology Software 2020, https://benchling.com). Constructs were generated following the Gibson assembly methodology according to the manufacturer’s instructions (Gibson Assembly Master Mix, NEB®). A Phusion High-Fidelity PCR Kit (NEB®) was used to amplify target sequences. Thermocycling conditions were as follows: Denaturation at 98°C for 3minutes, amplification 35 x (98°C for 30 seconds and 72°C for 20-30seconds/Kb) and elongation at 72°C for 10minutes. Gel electrophoresis was used to visualise products prior to DpnI digestion. Fragments were ligated using NEBuilder® HiFi DNA Assembly Master Mix (NEB®) and transformed using NEB® competent E. coli cells. Single colonies were selected for mini-preparation, and accurate assembly of constructs verified by Sanger sequencing. QSOX2 (p.T352M, p.V325Wfs*26, p.F474del) and STAT5B (p.Q177P) variant constructs were generated by site directed mutagenesis as outlined above.
NanoBiT complementation assays
Protein–protein interactions were assessed with NanoBiT complementation assays using the STAT5B WT/mutant and QSOX2 WT/mutant plasmids N terminally fused with NanoBiT fragments (LgBiT and SmBiT). HEK 293-hGHR cells (1x105 cells/well) were seeded in poly-D-lysine coated white bottom 96-well plates and plasmids were reverse-transfected using Lipofectamine™ 3000 according to the manufacturer’s instructions. The optimal DNA concentration required for maximum bioluminescence signal was determined to be 200ng per well; 100ng SmBiT-STAT5B and 100ng LgBiT-QSOX2. 24hours post-transfection, cell culture medium was removed and replaced with 100µL NanoBiT assay buffer (pH 7.4, HBSS 1X, HEPES 24mM, NaHCO3 3.96mM, CaCl2 1.3mM, MgSO4 1mM, BSA 0.1%) per well and equilibrated for 1 hour at 37°C in 5% CO2. Following equilibration, six (6) baseline luminescence readings were recorded using the CLARIOstar Multimode Plate Reader (BMG Labtech). Furimazine (Nanolight Technology) was prepared in a 1:50 dilution with assay buffer and 25µl added to each well following baseline measurements and readings continued for 1hour.
Statistics
Statistical analysis was performed using either a 2-tailed Student’s t test or one-way ANOVA (where three or more data groups were compared) to generate P values. P ≤0.05 was considered statistically significant. Data are presented as mean ± SD in all figures in which error bars are shown.