Reagents and antibodies
Acid citric dextrose (ACD) tubes were from BD Vacutainer. Ficoll-Paque was purchased from GE Healthcare. Azidohomoalanine (AHA), bicinchonic acid assay kit, Click-it protein reaction kit, biotin alkyne probe, dithiothreitol (DTT), Triton-X100m, C18 tips, goat polyclonal anti-biotin and goat anti-rabbit Alexa FluorVR 680 IgG antibodies were from ThermoFisher. RPMI 1640 Met-, P8340 protease inhibitor cocktail, magnetics streptavidin beads, ammonium bicarbonate (ABC), iodoacetamide (IAA), formic acid (FA), mouse monoclonal anti-actin (clone AC-15), Donkey anti-goat IgG, and goat anti-rabbit IgG HRP-conjugated antibodies were bought from MilliporeSigma. Trypsin/lys-C and the enhanced chemiluminescence kit (ECL) were from Promega and PerkinElmer respectively. Rabbit monoclonal anti-ILK (EP1593Y), rabbit monoclonal anti-ANXA2 (ERP13052B) and mouse monoclonal anti-FERMT3 (3D6) were from Abcam. Mouse monoclonal anti-ATP2A3 (PL/IM430) was from Santa Cruz Biotechnology. Recombinant chicken anti-VCL antibody was from Immune Biosolutions. Goat anti-mouse IgG HRP-conjugated was from Jackson IR. Goat anti-mouse IRDyeVR 800CW IgG was from LI-COR Biosciences.
Study population and ethics
The study population included 7 FX patients and 7 healthy controls. Participants were all males and were matched for age. The recruitment was performed through the Fragile X Clinic, at the CIUSSS de l’Estrie-CHUS (Sherbrooke, Québec, Canada). Informed written consent was obtained from healthy controls and from a caregiver for FXS participants in accordance with requirements of the Ethics Review Board of the CIUSSS de l’Estrie-CHUS. All participants had blood draw in the morning to decrease potential diurnal variation.
PBMCs isolation
PBMCs isolation was carried out using Ficoll-Paque following the manufacturer instructions with some minor modifications. Briefly, blood sample were collected by venipuncture into 8 mL ACD tubes and centrifuged at 300g for 10 minutes to allow plasma collection. A volume of PBS, equal to the volume of plasma collected, was then added to each tube and resulting blood samples were place onto a layer of Ficoll-Paque (blood/Ficoll-Paque ratio of 4:3). Afterwards, samples were centrifuged at 500g for 30 minutes. PBMCs were subsequently collected, washed two times with PBS and counted on a flow cytometer (DXH-9000 hematology analyzer, Beckman Coulter®).
AHA labeling
We used the metabolic labeling of the nascent proteins by BONCAT to isolate the nascent proteins from the whole proteome. In this approach, newly synthesis proteins are labelled with azidohomoalanine, an azide bearing methionine analogue, before being conjugated to a biotin alkyne probe. Nascent proteins are subsequently purified by a streptavidin pull-down and analyzed by mass spectrometry [22].
Freshly extracted PBMCs were first resuspended into warm (37°C) RPMI 1640 Met- (supplemented with 2 mM L-Glutamine) and incubated for 30 minutes at 37°C under gentle agitation to deplete the intracellular reserve of methionine. PBMCs were then diluted to a concentration of 3 million cells/mL and labeled with 100 μM AHA for two hours. Cells were pelleted and stored at -80°C after labelling. The optimised labeling conditions as well as the specificity of the AHA labeling and subsequent Click reaction toward nascent proteins were determined by Western blot using a specific anti-biotin antibody (Figure 1).
PBMCs protein extract
PBMCs were lysed in ice-cold 50 mM Tris pH 8.0, 1% NaDoc, 1% P8340 protease inhibitor cocktail and 250 U/mL Benzonase nuclease. After 15 minutes on ice, cells were homogenised with a 28G needle and proteins were pelleted for 25 minutes at 20 000g (4°C). Protein quantification in the supernatant was determined by bicinchonic acid assay.
Isolation of nascent proteins and preparation for MS analysis
Two experimental groups were formed (Control and FX) by pooling an equal amount of protein from each participant. The conjugation of the biotin probe was performed on 150 μg of pooled proteins using the Click-it protein reaction kit, following the manufacturer instructions and using a biotin alkyne probe. Nascent biotinylated proteins were purified by an overnight incubation with magnetics streptavidin beads at 4°C. The beads were washed 4 times with PBS containing 0.1% Triton X-100 and 5 times with a 20 mM pH 8.0 ABC (ammonium bicarbonate) buffer. Cysteines reduction was realized by incubating the beads with 10 mM DTT at 60°C for 30 minutes and subsequent alkylation with 15 mM iodoacetamide for 30 minutes at room temperature in the dark. Proteins digestion into peptide was performed overnight at 37°C with 1 μg of trypsin/lys-C. The digestion was stopped by adding formic acid to a final concentration of 1%. The supernatant was transferred to a clean tube and the beads were washed two times with 60% acetonitrile in 0.1% formic acid. All supernatants were combined and evaporated. The peptide samples were reconstituted in 0.1% trifluoroacetic acid, desalted with a C18 tip and dried. Nascent proteins were suspended in 1% formic acid and kept at -20°C. Four technical replicates were made for each experimental group, and each replicate was injected twice into the mass spectrometer (intra-assay replicate).
Preparation of PBMCs total proteome for MS analysis
The same experimental groups used in the previous section were also used for the preparation of the total proteome. Briefly, 5 μg of pooled protein extracts were solubilized with 8M urea in 10 mM HEPES pH 8.0 (protein/urea ratio (p/v) of 1:1). Reduction of proteins was performed with 10 mM DTT at 60°C for 30 minutes and alkylation with 15 mM IAA at room temperature in the dark for 30 minutes. Urea concentration was lowered below 1M by adding 50 mM pH 8.0 ABC buffer. Proteins digestion into peptide was performed overnight at 37°C with 0.25 μg (ratio trypsin/protein of 1:20) of trypsin/lys-C and stopped by adding 1% formic acid. Peptide samples were evaporated, desalted with a C18 tip and evaporated. The total proteome was suspended in 1% formic acid and kept at -20°C. Four technical replicates were made for each experimental group, and each replicate was injected twice into the mass spectrometer (intra-assay replicate).
LC-MS/MS analysis
Peptide were injected into an HPLC (nanoElute, Bruker Daltonics) and loaded onto a trap column with a constant flow of 4 µl/min (Acclaim PepMap100 C18 column, 0.3 mm id x 5 mm, Dionex Corporation) and eluted onto an analytical C18 Column (1.9 µm beads size, 75 µm x 25 cm, PepSep). Peptides were eluted over a 2-hour gradient of acetonitrile (5-37%) in 0.1% formic acid at 400 nL/min while being injected into a TimsTOF Pro Mass Spectrometer equipped with a Captive Spray nano electrospray source (Bruker Daltonics). Data were acquired using data-dependent auto-MS/MS with a 100-1700 m/z mass range, with PASEF enabled with a number of PASEF scans set at 10 (1.27 seconds duty cycle) and a dynamic exclusion of 0.4 minute, m/z dependent isolation window and collision energy of 42.0 eV. The target intensity was set to 20,000, with an intensity threshold of 2,500.
Mass spectrometry data analysis
Proteins identification from the raw data was accomplish with the MaxQuant software[23] (version 1.6.10.0). Peaks list was searched against the Uniprot human database (09/2019). Trypsin was set as digestion enzyme with specificity for arginine and lysine (but not before proline). A maximum of two miss cleavages was tolerated. Oxidation of methionine and acetylation of proteins N-terminus were set as variable modifications, while carbamidomethylation of cysteine was set as a fixed modification. Carbamylation of lysine was set as variable modification only for the analysis of the total proteome. False discovery rate of peptide (minimum of 7 amino acids) and proteins was set to 0.05 using a reverse database. The mass tolerance was set to 7 ppm for precursor ions and 20 ppm for fragment ions. The MaxQuant label free quantification (LFQ) was used, with a minimum ratio count of two, for accurate intensity based quantification of proteins between samples[24].
Proteomic data processing
Processing of proteomics data was carried out with the Perseus software[25] (version 1.6.7.0.). Proteins identified as "potential contaminant", "only identified by site" or "reverse" by MaxQuant were excluded. Data was normalized (average intensities of each sample equal within each group) before further analyzes. Proteins only identified in at least 50% of the samples from each group with a minimum of 1 unique peptide and 2 total peptides were kept. The statistical significance of proteins expression between the two groups was evaluated by a two-tailed student T-test, carried out in the Perseus software (p < 0.05 for the nascent proteome; p < 0.01 after correction with a permutation-based FDR, for the total proteome analysis). We choose to be more stringent for the differential expression analysis carried out for the total proteome to lower the numbers of statistical hits. The corresponding volcano plots were drawn using the ggplot2 package in R.
For the nascent proteome analysis, a "negative control" was produced to eliminate contaminating and endogenous biotinylated proteins. To achieve this, an unlabeled protein extract from a control PBMCs was subjected to all the steps of the workflow describe above (Click reaction, streptavidin pulldown, on beads proteins preparation for MS analysis etc.). Proteins identified with a fold change (experimental sample/negative control) superior to 1.1 where kept for further analysis. Proteins that do not fulfill this criterion where excluded.
Bioinformatic analysis
Functional annotation enrichment analyses were carried out with the Panther classification system ( www.pantherdb.org ) using the Gene Ontology, Panther protein class and REACTOME pathways annotations sets [26–28]. Protein-protein interaction network (PPI) analysis were obtained from the web-based LENS tool (Lens for Enrichment and Network Studies of Proteins) at the website: http://severus.dbmi.pitt.edu/LENS [29].
Western Blots
Optimal AHA labeling conditions and specificity towards newly synthetized proteins were confirmed by an anti-biotin Western blot (Figure 1). Briefly, 10 μg of proteins sample were resolved on a 10% SDS-PAGE, transferred onto a nitrocellulose membrane, block with 5% non-fat dry milk and incubated with the following antibodies: anti-biotin (1:1000) and anti-actin (1:5000). Blots were revelated using an enhanced chemiluminescence (ECL) kit and imaged with ChemiDoc (Bio-Rad).
Western blots were also used to validate 5 proteins found dysregulated in fragile X PBMCs by the proteomic screening (ILK, ATP2A3, ANXA2, FERMT3 and VCL). Briefly, 15 μg of proteins from each participant were resolved on a 9% (ILK, ATP2A3 and VCL) or a 12% (ANXA2 and FERMT3) SDS-PAGE, transferred onto a nitrocellulose membrane, block with 5% non-fat dry milk and incubated with the following antibodies: anti-ILK (1:2000), anti-ANXA2 (1:2000), anti-ATP2A3 (1:250), anti-FERMT3 (1:2000), anti-VCL (1:1000) and anti-actin (1:5000).Anti-goat IgG (1:10 000), anti-mouse IgG (1:10 000) and anti-rabbit IgG (1:10 000) HRP-conjugated secondary antibodies were used for ECL revelation, while anti-mouse IRDyeVR 800CW IgG (1:10 000) and anti-rabbit Alexa FluorVR 680 IgG (1:10 000) were used for fluorescence-based immunostaining. All immunoblots were analysed with the Image-J software (NIH). Immunoblots were revelated either by ECL (imaged with ChemiDoc, BioRad) or by fluorescence (imaged with the Odyssey Infrared Imaging System, LI-COR Biosciences).
Statistical analyses
Fisher exact tests were used with R studio to perform Gene Ontology enrichment analysis. A p-value inferior to 0.05 (corrected with a permutation-based FDR) was considered significative. For the Western blot analysis, statistically significant difference was determined by a two-tailed student t-test calculated in GraphPad Prism (version 8.3.0). A p-value inferior to 0.05 was considered significative.