Study cohorts

MS patients, NND patients and healthy individuals were recruited through the MS day clinic and the Department of Neurology, University Medical Center Hamburg-Eppendorf (UKE). The study was approved by the local ethics committee (Hamburger Patienteninformationssystem Multiple Sklerose-HAPIMS, Ethikkommission der Ärztekammer Hamburg, registration number PV4405) and informed consent was obtained from all patients and healthy individuals. These patients were not on any immunomodulatory medication at the time of sampling. These cryopreserved samples were collected in the biobank of the UKE Institute of Neuroimmunology and Multiple Sclerosis (INIMS). Trans men were recruited through the medical practice amedes Medizinisches Versorgungszentrum Hamburg GmbH. The study was approved by the local ethics committee (Ethikkommission der Ärztekammer Hamburg, registration number PV5245) and informed consent was obtained from all individuals. These cryopreserved samples were collected at the Research Department Virus Immunology, Leibniz Institute of Virology, Hamburg. Study participants were screened for chronic infections, autoimmune and metabolic diseases and cancers by medical history and blood tests. Participants received 1000 mg testosterone undecanoate i.m. (Nebido, Jenapharm, Jena, Germany) injections. After the first testosterone injection, the second injection was scheduled 6 weeks later, followed by a steady rhythm of 3 months. Further characteristics of the cohorts are detailed in Tables S1–3.

GTEx analysis

Gene expression data of spleen samples from 87 women and 154 men was downloaded from the Genotype-Tissue Expression (GTEx) project at https://gtexportal.org (GTEx_Analysis_v8). Raw RNA-seq read counts were analyzed for differential expression between sexes using DESeq2_1.34.0 (Table S6). Genes with exclusive localization on one of the sex chromosomes were excluded. According to our cut off criteria (false discovery rate-adjusted P value < 0.05; absolute log2 fold change > 0.2), 408 differentially expressed genes were identified (153 higher in women, 255 higher in men; Table S7). To test for the influence of sex hormone levels on gene expression, premenopause samples (38 women, 63 men, age ≤ 49 years) were compared to postmenopause samples (49 women, 91 men, age ≥ 50 years) separately in both sexes using DESeq2_1.34.0 (Table S8, S9). Plotting of the gene expression data was performed using ggplot2_3.4.2 and tidyheatmaps_0.1.0 (https://github.com/jbengler/tidyheatmaps).

PBMC isolation and cryopreservation

Blood was collected in EDTA coated tubes (Sarstedt) and processing of blood samples was started immediately upon receipt of blood samples in the laboratory. For MS patients, NND patients and healthy individuals PBMCs were isolated by a 30 min gradient centrifugation at 860 ×g using Biocoll Separating Solution (Biochrom), followed by two washing steps. Cells were then cryopreserved in RPMI medium (PAN-Biotech) with 25% heat-inactivated fetal calf serum (FCS, Biochrom) and 10% (v/v) dimethyl sulfoxide (DMSO, AppliChem) and stored in liquid nitrogen until further analysis. For trans men PBMCs were isolated by a 30 min gradient centrifugation at 500 ×g using Lymphocyte Separation Media (Capricorn), followed by two washing steps. After lysis of erythrocytes with ACK Lysing Buffer (Lonza), cells were then cryopreserved in heat-inactivated fetal bovine serum (FBS) supplemented with 10% (v/v) dimethyl sulfoxide (SigmaAldrich) and stored in liquid nitrogen until further analysis.

Flow cytometrySurface staining

Staining of surface marker antigens was always performed at 4 °C using monoclonal fluorochrome labeled antibodies. In the case of whole blood samples subsequently erythrocytes were lysed and lymphocytes fixed using BD FACS™ Lysing Solution. PBMCs and lymphocytes derived from CSF were fixed with BD Cytofix™ Fixation Buffer after staining prior to analysis. Surface marker antibodies used in this study are listed in Table S4.

Absolute cell quantification

Absolute cell counts were quantified by using Precision Count Beads™ (BioLegend).

Sample analyses

All samples were acquired on a BD FACS LSR II analyzer or FACSymphony A3 (BD Biosciences). Flow cytometry-based cell sorting was performed on a FACSAria III cell sorter (BD Biosciences). Data was analyzed with FlowJo software (Version 10.8.1, BD Biosciences).

Human T cell proliferation

Cryopreserved PBMCs from healthy women and men were thawed and CD3+ T cells were isolated from single-cell suspension using the Human Pan T Cell Isolation Kit (Miltenyi) according to the manufacturer’s protocol and labelled with CFSE (CellTrace™ CFSE Cell Proliferation Kit, ThermoFisher) according to the manufacturer’s protocol. CD3+ T cells were seeded at a density of 25,000 cells per well in an anti-CD3 (0.5 µg/ml, clone OKT3, BioLegend) coated 96-well plate. Cells were supplemented with soluble anti-CD28 (5 µg/ml, clone CD28.2, BioLegend) and anti-CD99 (5 µg/ml, clone HCD99, BioLegend) or respective isotype control (5 µg/ml, clone MOPC-173, BioLegend). Proliferation was tracked by cluster formation in the IncuCyte® for 7 days.

Human T cell activation

Cryopreserved PBMCs from healthy women were thawed and seeded at a density of 2 × 105 cells per well in an anti-CD3 (10 µg/ml, clone OKT3, BioLegend) coated 96-well plate. Cells were supplemented with soluble anti-CD28 (2.5 µg/ml, clone CD28.2, BioLegend) and anti-CD99 (10 µg/ml, clone hec2, BioLegend) or respective isotype control (10 µg/ml, clone MOPC-21, BioLegend). Samples were incubated for 72 h at 37 °C and 5% CO2, stained with antibodies listed in Table S4 and analyzed by flow cytometry after 2, 24, 48 and 72 h.

In vitro testosterone treatment

For testosterone treatment, we used RPMI 1640 Medium, without L-glutamine, without phenol red (Capricorn) supplemented with 1% GlutaMAX™ Supplement (Gibco), 1% Pen Strep (Gibco) and 5% charcoal stripped FBS Standard (PAN Biotech). For charcoal stripping of FBS, 2 g dextran-coated charcoal (DCC, Merck) were added to 100 ml FBS and incubated on a shaker overnight at 4 °C followed by centrifugation and filtration. Cryopreserved PBMCs from male healthy individuals were thawed and seeded at a density of 2 × 105 cells per well in an anti-CD3 (10 µg/ml, clone OKT3, BioLegend) coated 96-well plate and supplemented with soluble anti-CD28 (2.5 µg/ml, clone CD28.2, BioLegend) or left unstimulated. Subsequently, testosterone (3, 30 and 300 ng/ml, Sigma Aldrich) or 5α-dihydrotestosterone (0.3, 3 and 30 ng/ml, SigmaAldrich) was added. Samples were incubated for 48 h at 37 °C and 5% CO2, stained with antibodies listed in Table S4 and analyzed by flow cytometry after 2, 24 and 48 h.

Mice

C57BL/6J WT and C57BL/6J-CD99em2−8/Hhtg (Cd99-deficient) were kept under specific pathogen-free conditions in the animal facility of the ZMNH, University Medical Centre Hamburg-Eppendorf. 8–12-week-old mice were used for experiments. All animal experimental procedures were in accordance to international and national animal welfare guidelines. Ethical approvals were obtained from the State Authority of Hamburg, Germany (Approval no. 083/19 and 107/21).

Generation of Cd99-deficient mice

To generate Cd99-deficient mice, the second coding exon was targeted by the CRISPR/Cas9 genome editing system. A single sgRNA was designed using the CRISPOR design tool (http://crispor.tefor.net) [40]. The template for transcription with the targeting sequence (GCGAGTGACGACTTCAACCT) was generated by fill-in reaction with Klenow DNA Polymerase (Thermo Fisher Scientific). Transcription was performed using the HiScribe™ T7 High Yield RNA Synthesis Kit (#E2040S, New England Biolabs), with subsequent purification of the transcript with the MEGAClear™ Transcription Clean-Up Kit (#AM1908, Thermo Fisher Scientific), both according to the manufacturer’s instructions. The sgRNA (600 ng/µL) and Cas9 protein (Alt-R® S.p. Cas9 Nuclease V3, #1,081,058, Integrated DNA Technologies (IDT)) (500 ng/µL) in Gibco™ Opti-MEM™ (Thermo Fisher Scientific) were electroporated into one-cell-stage embryos derived from superovulated C57BL/6J mice using the NEPA 21 electroporator (Nepa Gene) [41]. Embryos were implanted into F1 foster mothers (C57BL/6 × CBA) and the resulting offspring was analyzed by PCR amplification on genomic tail DNA using primers F (5′-CGG GGC CCG GAT TGG ATG TAA ATG CTG-3′) and R (5′- AGA GCC CCG GGT ATG TAA ATG ACT C-3′) and subsequent Sanger sequencing. For our experiments, we used animals with an insertion of 2 bp in exon 2, which resulted in CD99 protein deficiency. These offspring were analyzed by separate PCRs with either WT allele-specific forward primer (CCT CAG CGA GTG ACG ACT TCA ACC) or mutant allele-specific forward primer (CCT CAG CGA GTG ACG ACT TCA AAA CC) and common reverse primer (CCC AGA GCC CCG GGT ATG TAA ATG ACT C). WT specific PCR resulted in a PCR product of 180 bp and mutant specific PCR resulted in a product of 182 bp.

Mouse T cell proliferation

Lymph nodes (axillary, brachial, inguinal) and spleen from homo- and heterozygous Cd99-deficient and C57BL/6 WT mice were collected in ice-cold PBS. CD3+ T cells were isolated from single-cell suspension using the MojoSort™ Mouse CD3 T Cell Isolation Kit (BioLegend) according to the manufacturer’s protocol and labelled with CFSE (CellTrace™ CFSE Cell Proliferation Kit, ThermoFisher) according to the manufacturer’s protocol. CD3+ T cells were seeded at a density of 3 × 105 cells per well in an anti-CD3 (1 µg/ml, clone 145-2CL11, BioLegend) coated 96-well plate. Cells were supplemented with soluble anti-CD28 (0.5 µg/ml, clone 37.51, BioLegend) and IL-2 (20 IU/ml, Peprotech). Samples were incubated for 72 h at 37 °C and 5% CO2, stained with antibodies listed in Table S4 and analyzed by flow cytometry.

EAE inductionActive immunization

Mice were immunized subcutaneously with 200 μg of MOG35–55 peptide (Peptide and elephants) in complete Freund’s adjuvant (CFA) containing 2 mg/ml Mycobacterium tuberculosis (BD Bioscience). Additionally, 200 ng pertussis toxin (Merck Millipore) in PBS were injected intraperitoneally (i.p.) on the day of immunization and 48 h later.

EAE scoring

Weight and clinical signs of disease were scored daily starting at day 7 by the following system: 0, no clinical deficits; 1, tail weakness; 2, hind limb paresis; 3, partial hind limb paralysis; 3.5, full hind limb paralysis; 4, full hind limb paralysis and forelimb paresis; 5, premorbid or dead. Animals reaching a clinical score of ≥ 4 or having more than 25% body weight loss (from starting weight) were sacrificed according to regulations of the local Animal Welfare Act. Scoring of EAE experiments was performed blinded for the genotype. Mice without any disease symptoms 20 days post immunization were excluded from the analyses. To minimize cage-specific effects, experimental groups were mixed within cages and littermates were used as control animals.

Immune cell isolationSpleen and LN

Inguinal, brachial and axillary lymph nodes as well as spleen samples were homogenized through a 70 µm cell strainer and washed with PBS (300 ×g, 10 min, 4 °C). Splenic cells were resuspended in erythrocyte lysis buffer (10 mM potassium bicarbonate, Merck Millipore; 0.15 M ammoniochloride, Merck Millipore; 0.1 mM Na2EDTA, Thermo Fisher Scientific; in ddH2O; pH 7.4) and incubated for 5 min on ice to lyse red blood cells before it was stopped with PBS.

CNS

Before the CNS was taken, mice were intracardially perfused with 10 ml PBS. The CNS was minced with a razor blade, digested for 45 min at 37 °C in a shaking water bath (1 mg/ml Collagenase A, Roche; 0.1 mg/ml DNaseI, Merck Millipore; in RPMI-1640 medium, PAN Biotech) and subsequently homogenized through a 70 µm cell strainer. After washing with PBS, immune cells were isolated by percoll gradient centrifugation (30%/78% 1.13 g/ml, GE Healthcare) at 2500 rpm, 30 min, 4 °C, w/o brake, harvested from the interphase and washed twice with PBS.

Jurkat T cell proliferation

WT Jurkat T cells, CD99-deficient Jurkat T cells or transduced Jurkat T cells were fluorescently labelled (CellTrace™ Violet Cell Proliferation Kit, ThermoFisher) according to the manufacturer’s protocol. Cells were seeded at a density of 3 × 105 cells per well in a 12-well plate. Samples were incubated for 96 h at 37 °C and 5% CO2 and analyzed by flow cytometry after 0, 24, 48, 72 and 96 h. Jurkat T cells are derived from a male donor and harbor X and Y chromosomes.

Vector construction and lentiviral production

To insert human CD99 (hCD99) into a lentiviral vector harboring a CMV promoter, we first performed PCR using Primer_f_1 and Primer_r_1 from a human CD99 gene ORF cDNA clone expression plasmid (Biozol). Restriction and ligation were performed using BamHI and XbaI restriction sites. To generate hCD99 CRISPR-KO cells, we used four guide RNAs (Oligo_f/r_1, Oligo_f/r_2, Oligo_f/r_3, Oligo_f/r_4) targeting different protein coding exonic regions of hCD99 that were annealed and inserted using Esp3I restriction sites into the Lenti-Cas9-gRNA-GFP. Lenti-Cas9-gRNA-GFP was a gift from Jason Sheltzer (Addgene #124,770; http://n2t.net/addgene:124770; RRID: Addgene_124770). Jurkat T cells were either transfected by Lipofectamine 2000 (ThermoFisher) according to the standard protocol or by lentiviral transduction (see below for detailed description). GFP-expressing transfected cells were identified and sorted by flow cytometry using a FACSAria III cell sorter (BD Biosciences). Genetic KO was confirmed using flow cytometry and Sanger sequencing. To produce lentiviruses, we first transfected HEK293T cells with 10 µg expression plasmid, 10 µg pMDLg/pRRE, 5 µg pRSV-Re, 2 µg pMD2.G. pMDLg/pRRE was a gift from Didier Trono (Addgene #12,251; http://n2t.net/addgene:12251; RRID: Addgene_12251). pRSV-Rev was a gift from Didier Trono (Addgene #12,253; http://n2t.net/addgene:12253; RRID:Addgene_12253). pMD2.G was a gift from Didier Trono (Addgene #12,259; http://n2t.net/addgene:12259; RRID: Addgene_12259). Briefly, HEK293T cells were seeded out with an 80% confluency in DMEM with glutamine and high glucose (ThermoFisher), the next day the plasmids were mixed in 1 × HEPES buffered saline (HBS) and 125 mM CaCl2 and were applied to the HEK293T cells for 6 h. Subsequently, medium was changed and after 48 h the supernatant was filtered through a 0.45 µm PES filter, immediately snap frozen and stored at − 80 °C. Primers, oligonucleotides, and the respective restriction sites are listed in Table S5.

Statistical analysis

Bar graphs represent mean ± standard error of the mean (SEM). Statistical analyses were performed using Prism 9 software (GraphPad Software). Normal distribution was tested using Kolmogorov–Smirnov or Shapiro–Wilk test. Significant results are indicated by *P < 0.05, **P < 0.01, ***P < 0.001.