CSF and blood samples from participants

The National Health Service Research Ethics Committee had approved this study (IRAS ID: 204,872 and FIS PI15/00513), and all subjects have provided written informed consent to participate in the study.

CSF supernatants were collected and centrifuged at 335 g for 10 min at 40 C to separate the cell pellet from the supernatant according to the standard procedures [16].

In this study, we analysed three distinct cohorts of CSF samples, including samples from individuals with MS and control subjects.

The first cohort included matched serum and CSF samples collected from 65 patients with RR-MS and 35 with OND in Hospital Universitario Ramon y Cajal of Madrid. The demographic and clinical data of MS and OND were reported in Supplementary Table 1.

Serum and CSF supernatants were analysed for the protein levels of sCD27. For surface antigen identification, cellular pellets obtained after CSF centrifugation were resuspended in their residual volumes, stained with appropriate amounts of monoclonal antibodies (according to manufacturer’s instructions) for 30 min at 4 °C in the dark, washed twice with phosphate-buffered saline (PBS), and analysed in a FACSCanto II flow cytometer (BD Biosciences).

For intracellular cytokine detection, cellular pellets were incubated for 4 h at 37 °C in 5% CO2 with 50 ng/ml Phorbol 12-myristate 13-acetate (PMA, Sigma-Aldrich, St. Louis, MO) and 750 ng/ml Ionomycin (Sigma-Aldrich) in presence of 2 µg/ml Brefeldin A (GolgiPlug, BD Biosciences) and 2.1 µM Monensin (Golgi Stop, BD Biosciences). After incubation, cells were washed once and stained with the appropriate amounts of monoclonal antibodies (according to the manufacturer’s instructions) for 30 min at 4 °C in the dark. Then, cells were washed once, fixed and permeabilized with Cytofix/Cytoperm Kit (BD Biosciences). After that, cells were washed twice and subjected to intracellular staining with monoclonal antibodies recognizing different cytokines, for 30 min at 4 °C in the dark. Finally, cells were washed and analysed in a FACSCanto II flow cytometer. The following monoclonal antibodies were used in the study: CD3-PerCP, CD3-BV421, CD5-APC, CD8-APC-H7, CD14-FITC, CD19-PE-Cy7, CD25-PE, CD27-FITC, CD38-PE-Cy5, CD45-V500, CD56-APC, CD127-BV421, IFNγ-FITC, TNF-PerCP-Cy5.5, and GM-CSF-PE (BD Biosciences, San Diego, CA), IL17-APC (R&D Systems, Minneapolis, MN). The different cell subsets were identified according to the combination of markers shown in supplementary Table 2.

The second and third cohorts include CSF from 71 patients with a diagnosis of RR-MS and SP-MS and 12 control donors with a diagnosis of headache, migraine, and idiopathic intracranial hypertension (IIH) (OND). The samples were collected from Charing Cross Hospital and only the CSF supernatants were available and investigated for the presence of inflammatory markers (cytokines and chemokines) and sCD27. The demographic and clinical data of those cohorts were reported in Supplementary Table 1.

Enzyme-linked immunosorbent assay (ELISA)

ELISA was performed to investigate protein levels of sCD27, BAFF, CXCL13, levels of cytokines IFN-g, TNF (DuoSet Elisa development System, R&D, UK) and NFL-light (Tecan, Switzerland) according to the manufacturing protocols. Briefly, the flat bottom 96 plates were coated with a working solution of capture antibody and filled with a blocking solution of PBS with 1% BSA (bovine serum albumin, Sigma). Then, the plates were filled with 100 ul of diluted samples and 100 ul of a serial dilution of standard analyte. After 2 h of incubation, the plates were washed with PBS with 1% Tween 20 and incubated with 100 ul of a working solution of the detection antibody for 1 h. Then the plates were washed and 100 ul of working solution of streptavidin-HRP was added for 20 min. After washes, the substrate solution was added for 20 min, and a working solution of sulphuric acid (2 N H2SO4) was added to block the reaction. The plates were run in a microplate reader (Promega) set at 450 and 560 nm. The readings (absorbance) at 450 nm were corrected by subtraction of the readings at 560 nm. All the samples were analysed in duplicate.

Electro chemiluminescent assay

Meso Scale Discovery detection system (MSD; Meso Scale diagnostics, UK) was performed to quantify the concentration of selected biomarkers in the CSF supernatants. The MSD U-Plex plates were customised with the cytokines (IL-6, IL-10, TNF) and chemokines (CCL19, CXCL10, CXCL12) and developed as suggested by the supply (Meso Scale diagnostics, UK). All the samples were analysed in duplicate.

Preparing multiplex coating solution

200 ul of biotinylated antibody for each analyte was mixed with 300 ul of assigned linker and incubated for 30 min a room temperature. Then, 200 ul of stop solution was added to each vial and left for another 30 min. 600 ul of each U-Plex coupled antibody solution was mixed in one tube and 2.4 ml of stop solution was added to bring the final volume to 6 ml.

Coating the U-Plex plate

A binding plate was coated by adding 50 ul of the multiplex coating solution to each well and incubated for one hour on the shaker a 200 rpm at room temperature. Then, the coating solution was aspirated, and the plate was washed 3 times with PBS-T (PBS with 10% of Tween 20).

Assay procedure

50 ul of each diluted sample and serial dilution of standard curve were added to the well and the plate was incubated for one hour at room temperature on the shaker at 200 rpm. Then, the plate was washed three times with PBS-T and 50ul of working solution of detection antibody was added to each well and incubated for one hour at room temperature on the shaker a 200 rpm. Then, the washing process was repeated and PBS-T and 150 ul of 2fold-concentrated read buffer were added for MSD analysis.

Cell culture and protein extraction

Peripheral blood mononuclear cells (PBMCs) were isolated from fresh blood samples of 25 healthy donors by centrifugation on a gradient of Ficoll-plaque as described [23].

Briefly, 40 ml of whole blood was diluted with PBS and 35 ml of diluted blood was layered slowly on a gradient of 15 ml of Ficoll-plaque in 50 ml of falcon tube. The samples were centrifuged at 1800 rpm for 30 min with the break turned off. The layer of mononuclear cells was collected and washed twice with PBS through centrifugation at 1600 and 1400 rpm.

CD4, CD8 T and B cells were isolated from PBMCs using magnetic beads (CD4 + T cell isolation kit, CD8 + T cell isolation kit and CD19 Micro beads, Miltenyi Biotech, Germany) as suggested by the technical data sheet.

In a two-step procedure, CD4 + and CD8 + T cell isolation kits were used to isolate untouched CD4 + or CD8 + T cells. The PBMCs were stained with a cocktail of biotin monoclonal antibodies against CD4 + T cells (in CD8T cell isolation) or CD8 + T cells (in CD4 T cell isolation), monocytes, eosinophils, neutrophils, B cells, NK cells, γδ T cells, granulocytes, and erythroid cells in buffer containing PBS with 0.5% of FBS and 2 mM of Ethylenediaminetetraacetic (EDTA) for 5 min in ice. Then, CD4 + T or CD8 + T microbeads that are conjugated to anti-biotin antibody and monoclonal anti-CD61 antibody (in CD4 T cell isolation) were added to the cells and incubated for 10 min in ice. The cells were layered onto an LS column placed in a magnetic field previously activated with a rinsing buffer (PBS with 2 mM of EDTA). The column retains labelled cells and elutes unlabelled CD4 + or CD8 + T cells based on the isolation kit used for the separation.

The CD19 isolation kit is a one-step positive separation for isolating CD19 + B cells. First, the PBMCs were stained with microbeads conjugated with antibodies against CD19 B cells for 15 min in ice and then washed in a buffer at 300 x g for 10 min at 4 degrees. Then, the cell pellet was resuspended in 500 ul of buffer and layered onto an MS column previously placed in a magnetic field and activated with rinsing buffer. The column was washed with rinsing buffer to elute the negative fraction. The column was removed from the magnetic field and CD19 + B cells were collected by washing out the column with 1 ml of rinsing buffer. The cells were stained with fluorescent antibodies and their purity was evaluated by analysing them with Flow cytometry LSRII.

B cell activation

Isolated CD19 cells were suspended in a completed medium (RPMI-1640 medium completed with 10% of FBS) (Gibco, ThermoFisher Scientific, UK) and seeded in a round bottom 96-well plate at the concentration of 2 × 105 cells/well and then stimulated with F(ab)2-goat anti-human IgMIgG, (H + L) functional grade (5ug/ml, eBioscience), CD40L (1ug/ml, functional grade, Invitrogen), CpG ODN 2006 (TLR9 ligand, 2.5ug/ml, InvivoGen) and resiquimod (R848, TLR7 and TLR8 ligand, 1ug/ml, Sigma-Aldrich). The cell culture supernatants were collected at 12, 24, and 48 h and analysed for the presence of sCD27 and TNF.

B cell activation for western blotting

Isolated CD19 B cells were suspended in a completed medium and seeded in a 48-well plate at the concentration of 5 × 105 cells/well and then stimulated as reported above for 48 h.

The cell pellets were harvested and processed for protein extraction as reported below.

CD4 and CD8 T cell activation

Isolated CD4 + or CD8 + T cells were suspended in complete medium and seeded at the concentration of 2 × 105 cells per well in a sterile round bottom 96-well plate pre-coated with 1, 5 and 10 ug/ml of anti-CD3 (clone UCHT1, Biosciences) and incubated a 37 °C for 72 and 96 h. Supernatants were collected at 72, and 96 h and analysed for protein levels of IFNg and sCD27. 1 × 106 cells of purified CD4 and CD8 T cells were seeded on 48 well plates, pre-coated with 5ug/ml of aCD3 (clone UCHT1,eBiosciences) for 72 h. The cells were harvested and centrifuged at 1400 rpm. The cell pellets were processed for protein extraction and levels of sCD27 and CD27 have been analysed as reported below.

Protein extraction and quantification

Cell pellets were washed 2 times in PBS at 2,500 rpm for 10 min at 4 degrees and then resuspended in cell extraction buffer (RIPA buffer with added protease and phosphate inhibitor cocktail solution, Invitrogen, UK) for 30 min on the ice. Then, the extracts were centrifuged for 10 min at 16,000 rpm at 4 °C. The supernatants were collected and stored at -80 °C. Protein quantification was performed with BCA protein assay (Pierce) as suggested by the supply and 30 ug of protein was loaded for each condition either T or B cells.

CRISPR/Cas9 RNPs

The CD27 gene ablation on CD4 T cells was performed as described [24].

T-cell activation

A sterile, non-treated 48-well tissue culture plate was coated with anti-CD3 antibody at the concentration of 10ug/ml. The pellet of isolated CD4 T cells was suspended in complete RPMI medium (high-glucose RPMI-1640, 10% FBS, penicillin-streptomycin 50ug/ml, sodium pyruvate 5mM, HEPES 5mM) with anti-CD28 (5ug/ml, clone CD28.2) and IL-2 (20U/ml) at the final concentration of 2,5 × 106 cells/ml. 500 ul of cell suspension was added at each well of the anti-CD3 plate bound plate and incubated at 37 °C with 5%CO2 for cell stimulation and activation. After 48–72 h, the cells were collected, washed two times in PBS and suspended in PBS at the concentration of 2,5 × 106 cells/ml. The volume of 2 × 105 cells was centrifuged and the cell pellet was suspended in Neon R buffer before nucleoporation (Neon™ transfection system, Thermo Fisher scientific, UK).

crRNP generation

crRNPs were synthesized in vitro by incubation of 1ul of tracrRNA (160 μm) and 1ul of gRNA (160 μm) in a sterile PCR tube to form 2ul of complex gRNA: tracrRNA at the concentration of 80 μm. A mix of positive and negative controls was prepared at the same time. Then, the mixtures were incubated at 37 °C for 30 min in the thermocycler to promote the duplex formation. 2ul of Cas9-NLS protein was added to the RNA mixtures and then incubated at 37 °C for 15 min to form crRNPs.

crRNP nucleotrasfection

500 ul of fresh medium (high-glucose RPMI-1640, 10% FBS, sodium pyruvate 5mM, HEPES 5mM) with 20U/ml of IL-2 were added at 24-well plate.

For each nucleotransfection reaction, 4ul of crRNPs were added to the cell suspension prepared in R buffer. Then, 10 ul of the cells/crRNP mixture were transferred into 10ul Neon tip and electroporated at the following conditions: 1600 V, 10 ms, 3 pulses. Then, the cells were transferred to a 24-well plate and dynabeads were added at the ratio beads: cells 1 to 1. The electroporation was repeated for positive and negative controls from each donor. The transfected cells were incubated in a humidified 37 °C, 5%CO2 incubator for 48–72 h.

Cell activation and expansion

48–72 h after nucleofection, fresh medium with 20U/ml of IL-2 was added to the cells and 48–72 h later, the cells were stained with anti-CD27 and the efficiency of nucleofection was measured as the percentage of CD27 on CD4 T cells detected by flow cytometer. The cells were fed with fresh medium and IL-2 (20U/ml) for two weeks and then, the cells were washed, suspended in fresh medium, and seeded at the concentration of 2 × 105 cells/well in a U-bottom, 96-well plate pre-coated with 10 ug/ml of anti-CD3. At 24- and 48-hours supernatant was analysed for levels of IFNg and protein extraction from cell pellet was performed.

Purification of extracellular vesicles

Extracellular vesicles were isolated from supernatants of isolated CD4 T cells cultured with or without anti-CD3 (5 μg/ml) for 72 h using the methodology described in the paper [25].

In brief, the culture supernatants were centrifuged at 2,000 g for 20 min to remove debris and dead cells. Then the supernatants were centrifuged at 16,500 g for 45 min to isolate endosomal vesicles of 10k. The supernatants were centrifuged at 100,000 g for 2 h at 4 degrees and the pellets were resuspended in PBS and centrifuged at 100,000 g for 2 h a room temperature to isolate endosomal microvesicles of size 100k (exosomes) [26]. Cell and endosomal vesicle pellets were resuspended in a cell extraction buffer for total protein extraction as described above.

Western blot analysis

Proteins extracted from whole cells and exosomes were separated by 4–12% SDS-PAGE (Mini protean TGX stain-free gels, Bio-Rad, UK) and transferred onto PVDF membranes previously activated in methanol for 30 s. The blots were blocked with 5% non-fat dry milk a room temperature for 1 h and incubated with the working concentration of primary antibodies overnight at 4 °C and then incubated with HRP-conjugated secondary antibodies (Invitrogen) a room temperature for 1 h. The membranes were developed with an enhanced chemiluminescence (ECL) detection kit (Pierce). Information about the primary and secondary antibodies is reported in Supplementary Table 3.

Post-mortem CSF and brain tissues of patients with MS and controls

Snap-frozen brain and/or paraffin-embedded 4% paraformaldehyde-fixed (FFPE) tissues and matched CSF from a total of 55 cases with a diagnosis of SPMS and 17 control cases were provided by the UK Multiple Sclerosis Society Tissue Bank (Imperial College London, UK). Demographic and clinical features of MS cases were described in Supplementary Tables 4 and control cases in Supplementary Table 5. FFPE tissues from 35 out of the 55 examined post-mortem MS cases were used for neuropathological investigation. The combined CSF samples were analysed for protein levels of sCD27 and sCD21 by using the R&D system Elisa kit and analysed by Simple Plex (ProteinSimple, San Jose, CA) following the manufacturer’s instructions. In addition, the protein levels of CCL19, CXCL10, CXCL12, CXCL13, IL-6, IL-10, BAFF, and IFN-g were determined using customised immune-assay multiplex Luminex technology (Bio-Plex X200 System equipped with a magnetic workstation, Bio-Rad, Hercules, CA, USA), following the procedure previously optimized [27]. NFL proteins were measured using the ELISA assay (MyoBioSource San Diego, CA, USA) according to procedures previously optimized [27].

Immunohistochemistry

Snap-frozen 10 μm sections were fixed with cold methanol for 5 min. The sections were then washed with PBS, treated with a solution 0.3% hydrogen peroxidase blocked with a solution of 2.5% horse serum and incubated with a working solution of primary antibody (anti-human CD3, anti-human CD27 and anti-human CD20) overnight at 4oC in a damp container. Antibody binding was detected by horseradish peroxidase-conjugated secondary antibody using an automated chromogenic detection system. The slides were washed in running water for 5 min, stained in haematoxylin for 1 min, washed for 5 min in water and dipped in solutions with rising percentages of ethanol 70%, 80%, 100% and 100% of Xylene. A drop of aqueous mounting solution was added on each slide that was closed with a coverslip and left to dry overnight. The images were acquired by using slide scanner Leica biosystem Aperio AT2 and the cell count was performed by ImageJ Fiji software.

Immunofluorescence

An Immunofluorescence assay was performed on snap-frozen sections from the brain tissues of 2 patients. A blocking solution of 2.5% of horse serum in PBS was added to each section and then a working solution of a combination of primary antibodies (one combination of antibodies included mouse anti-human CD4, goat anti-human CD27, rabbit anti-human CD8, and a second combination included mouse anti-human CD3, goat anti-human CD27 and rabbit anti-human CD20) was added overnight in a humid container at 4 °C. After three washes with PBS, a solution of anti-rabbit biotin antibody in 2.5% horse serum was added for 1 h. The sections were washed three times and a working solution of secondary fluorescence antibodies (donkey anti-mouse 555, donkey anti-goat 488 and donkey streptavidin-647) was added for 2 h. Subsequently, the sections were washed three times with PBS and a DAPI solution (Roche Diagnostics GmbH, Germany) was added for one minute. After washes, the slides were mounted with antifade medium (Vectashield, Vector laboratories), coverslipped and dried for almost 2 h in a dark room. The images were obtained by using a Leica TCS SP8 confocal microscope. Antibodies used in immunohistochemistry and immunofluorescence are given in Supplementary Table 6.

Neuropathological characterization of brain tissues

A semiquantitative scoring system was developed in a subgroup of 35 MS cases to have a standardised measure of the degree of demyelination and inflammation in the different compartments examined (meninges, perivascular spaces, choroid plexus). The presence and extent of demyelination and the activity of MS cases were assessed by combining immunohistochemical detection of myelin oligodendrocyte glycoprotein (MOG) and major histocompatibility complex (MHC) class II combined with Luxol fast blue (LFB) staining on serial sections, as previously described [28, 29]. The percentage area of demyelinated white matter (WM) and grey matter (GM) was measured using Qpath [30]. The entire GM and WM fraction for each MOG immunostaining was traced and measured, then the area of individual GM and WM lesions was measured. Finally, the mean GM and WM lesion area was reported per section and MS case as the percentage of total demyelinated GM or WM, respectively. In addition to the detection of lymphoid-like structures in the meninges, a score of the degree of inflammation in the different compartments (choroid plexus and perivascular infiltrates) was performed on a semi-quantitative basis, rating the degree of inflammation in three different levels of intensity by examining the haematoxylin-eosin staining of all the examined blocks. According to previously published procedures [31], score 0 corresponds to absent/scarce (0–5 cells), 1 corresponds to moderate (6–25 cells), 2 corresponds to high (25–50 cells), and 3 (> 50 cells) corresponds to elevated inflammatory infiltrate. All semi-quantitative scoring was performed on images acquired at 20x objective (field area 556.5 μm x 312,9 μm equal to 174128,85 μm²).

R analysis

PCA and HCA were made with R Studio version 4.0.2, Factoextra (10.7) and factoMiner packages, missMDA package for missing values., Corrplot and ggplot2 packages for hierarchical clustering, To establish whether sCD27 is a predicting factor for a specific neuropathological variable, we adopted the Random Forest Recursive Feature Elimination method (RF-RFE) [32]. In detail, a 10-iteration 5-fold cross-validation protocol was adopted, and performances were measured using the Mean Squared Error. The analyses were performed with MATLAB version R2022b.

Statistical analysis

Statistical analyses were performed using Prism version 9.2 (GraphPad Software Inc.) and MATLAB. Differences between groups were compared using the Mann-Whitney test, 2way Anova multiple comparisons, Ratio paired t-test, one sample t-test, Kruskal-Wallis test, and Dunn test. Correlations were assessed using the Spearman rank correlation. Flow cytometry analyses were performed using Flow-jo software version 10.6.1 (FlowJo LLC).