Rat and mouse serum

Synovial mononuclear cellular fractions at a concentration of 10⁶/mL were stimulated with 0.1 μg/mL lipopolysaccharide (LPS) or 10 ng/mL phorbol 12-myristate 13-acetate (PMA) and 1 μg/mL ionomycin (all from Sigma). Synovial fibroblasts were cultured to confluency (5×10⁴/mL), treated for 3 h with IL-1β (10 ng/mL, Peprotech), washed and cultured for another 24 h. GM-CSF levels in supernatants were determined by BD Biosciences OptEIA ELISA according to the manufacturer's instructions. For intracellular cytokine staining, CD4+ T cells were stimulated with PMA and ionomycin for 5 h with 10 μg/mL Brefeldin A (Sigma) added after 1 h. Cells were then harvested, fixed and permeabilised and stained using the FoxP3/transcription factor staining buffer set (eBioscience) according to the manufacturer's instructions. Cells were preincubated with 2% mouse and rat serum (both Sigma) prior to antibody labelling.

For intracellular staining, cells were washed twice with staining buffer and fixed/permeabilized using fixation/permeabilization buffer (eBioscience) at 4 °C for 45 min. After two washes with permeabilization wash buffer (eBioscience), cells were blocked with mouse and rat serum (Sigma-Aldrich) for 10 min at 4 °C in the dark, then stained with anti-Helios-fluorescein isothiocyanate (FITC; Clone 22F6, Biolegend), anti-FoxP3-phycoerythrin cyanin 7 (PE/Cy7; Clone PCH101, eBioscience) and anti-CTLA-4-Peridinin Chlorophyll Protein Complex/e-Fluor™ 710 (PerCp-Fluor™ 710; Clone 14D3, eBioscience) antibodies for 30 min at 4 °C in the dark. Cells were washed twice with permeabilization buffer, and re-suspended in 300 µL of FACS staining buffer (eBioscience). Data were acquired by BD LSRFortessa X-20 flow cytometer (BD Biosciences) and analyzed by FlowJo v.10.0 software (Tree Star, Ashland, Covington, KY, USA).
The percentage of CD4⁺ T cells expressing a certain IC in activated PBMCs compared to control co-culture (breast cancer cell line + activated PBMCs) was used as a measure to determine upregulation or downregulation of IC expression. Similarly, we compared the percentage of CD4⁺ T cells expressing a certain IC in different co-culture conditions to that in control co-culture.

Mice were sacrificed and perfused by intracardial injection of PBS at 6 dpi. Subsequently, the brain and spleen were extracted and processed to obtain leukocytes as previously described (Claser et al, 2011; Teo et al, 2013). Isolated leukocytes were stained with LIVE/DEAD Aqua (Life Technologies), then blocked in 100 μl of blocking buffer consisting of a mix of 1% of rat and mouse serum (Sigma‐Aldrich) in FACS buffer [1% BSA, 2 μm EDTA in PBS]. Next, cells were incubated with PE‐labeled SQLLNAKYL‐H‐2D^b (Pb‐1) tetramer (Howland et al, 2013) on ices before addition of conjugated antibodies for another 20 min of incubation. Cells were fixed in IC fixation buffer (ebioscience) for 5 min before acquisition using a LSR II flow cytometer (BD Biosciences). Conjugated antibodies used were as follows: α‐CD45 (clone 30‐F11, BD Biosciences, 1:400 dilution), α‐CD3 (clone 17A2, BD Biosciences, 1:200 dilution), α‐CD4 (clone GK1.5, Biolegend, 1:400 dilution), α‐CD8 (clone 53–6.7, BD Biosciences, 1:400 dilution), α‐LFA‐1 (H155‐78; Biolegend, 1:200 dilution), α‐NK1.1 (clone PK136, ebioscience, 1:200 dilution), α‐CD11b (clone M1/70, Biolegend, 1:400 dilution), and α‐Ly6G (clone 1A8, Biolegend, 1:400 dilution). Gating strategy of T‐cell infiltrate in the brain is shown in Appendix Fig S7. Majority of the T‐cell infiltrate in the brain of the infected mice express LFA‐1 marker (Appendix Fig S7).