F4 80 clone bm8

FACS analysis was performed using 200-μL tumor or splenocyte single-cell suspensions per antibody panel. Cells were first incubated at 4°C for 30 minutes in PBS with anti-CD16/CD32 Fc block (BD Biosciences) at 1: 100 and live/death staining (Zombie Aqua, BioLegend) at 1: 1,000, or Live/Dead fixable blue dead cell stain (Invitrogen). Next, cells were washed and stained at 4°C for 30 minutes in PBS with antibodies against CD45 (clone 30-F11), B220 (clone RA3–6B2), CD3 (clone 17A2), CD8 (clone 53–6.7), MHC-I (clone AF6–88.5), PD-1 (clone RPM1–30), PD-L1 (clone 10F.9G2), NK1.1 (clone PK136), CD11b (clone M1/70), Ly6G (clone 1A8) and F4/80 (clone BM8), all from BioLegend, at 1 μg/mL. PE-labeled H-2Kb - SIINFEKL pentamer (ProImmune) was used at 2.5 μL per staining. In some sets of experiments, absolute cell numbers were calculated by adding 10 μL counting beads (1 × 10⁶ beads/mL, Spherotech) to each sample before acquisition.

CD4⁺ T cells-depleted splenic cells were cultured with IL-23 (20 ng/ml, Biolegend) with or without butyrate (0.5 mM) for 16 h, and then stimulated with ionomycin (750 ng/ml) and phorbol-12-myristate 13-acetate (50 ng/ml) for 2 h, followed by addition of brefeldin (BD Biosciences) for another 3 h. After Fc blocking, cells were stained with surface markers (PE/Cy7-anti-Thy1, 1:200, Clone#30-H12, Cat#105326; FITC-lineage (CD3, Clone#145-2C11, Cat#100306; CD11b, Clone#M1/70, Cat#101206; CD11c, Clone#N418, Cat#117306; B220, Clone#RA3-6B2, Cat#103206; F4/80, Clone#BM8, Cat#123108; NK1.1, Clone#PK136, Cat#108706; and Gr1, Clone#RB6-8C5, Cat#108419), 1:100), which were purchased from Biolegend. Cells were permeabilized using Foxp3/Transcription Factor Fixation/Permeabilization set (Cat#00-5523-00, ThermoFisher), followed by intracellular staining (anti-IL-22, 1:200, Clone#1H8PWSR, Cat#12-7221-82, Invitrogen; anti-IL-17, 1:200, Clone#TC11-18H10.1, Cat#506916, Biolegend). For ILC3, cells were also stained with RORγt (1:200, Clone#B2D, Cat#17-6981-82, Invitrogen).
All the events were collected by BD FACS Diva software. IL-22 and IL-17 production in ILCs (Thy1⁺ Lineage⁻ cells), or in ILC3 (Thy1⁺ Lineage⁻ RORγt⁺ cells) was analyzed using FlowJo. Gating strategies are included in Supplementary Fig. 15.

Tissues were obtained from some of the animals previously used for intravital microscopy. Cryosections of 5 μm of the brain encephalic region were obtained and fixed with ice-cold acetone (Merck, Rahway, NJ, USA) for 10 min. The blockage of non-specific binding was made using a goat serum (Vector Laboratories, Newark, NJ, USA) for 20 min. F4/80 clone BM8 and CD3 clone 17A2 (Biolegend, San Diego, CA, USA) were used in the analysis and were incubated overnight after dilution on PBST. The ImmPRESS^® HRP Goat Anti-Rat IgG Polymer Detection Kit with DAB (Dako, Santa Clara, CA, USA) was used to detect the antibodies. The counterstain used was Harris’s Hematoxylin Solution (EasyPath-São Paulo, Brazil) for 1 min. The slides were rinsed with tap water and then dehydrated chemically and mounted with Erv-mount mounting media (EasyPath). The counting of CD3+ and F4/80+ cells was performed manually. Representative images were acquired in Motic Panthera L Microscope Built-in Smart CAM & ImagingOnDevice System (Motic-Xiamen, Xiamen, China).

A single‐cell suspension was derived from the BALF of LysMCre:Shp2^fl/fl and Shp2^fl/fl mice 24 hour after secondary S aureus infection. After blocking the Fc receptor, cells were stained with CD11b (clone M1/70, BioLegend, San Diego), F4/80 (clone BM8, BioLegend) and Dectin‐1 (clone RH1, BioLegend), or their corresponding IgG isotype controls. After extracellular markers were stained, the cells were fixed, permeabilized and stained with CD206 (clone C068C2, BioLegend) or its isotype control. Cell samples were detected by CytoFLEX flow cytometer (Beckman Coulter, Inc.), and FlowJo software (TreeStar, Inc.) was used for analysis.