Single cell RNA-Seq analysis

A single-cell transcriptomics dataset for peripheral blood mononuclear cells (PBMCs) derived from four healthy human donors was obtained from the Gene Expression Omnibus (GSM3087619, GSM3087622, GSM3087624, GSM3087626). Data preprocessing, including cell quality control, has been described in a previous study (Chen et al. 2018). Principal component analysis was performed on quality cells, followed by unsupervised hierarchical clustering by t-distributed stochastic neighbor embedding algorithm (t-SNE) analysis. The marker genes used for cluster annotation are listed in Additional file 8: Table S1 (Zheng et al. 2017; Li et al. 2019; Sinha et al. 2018; Karamitros et al. 2017). The top genes displayed in the heatmap have the most significant P value and are differentially expressed marker genes for each cell type. Violin plots showing the expression levels of marker genes or SIGIRR by different populations of PBMCs.

Bulk transcript profiling (RNA-Seq) for immune lineages in blood and PBMCs in human

The bulk RNA-seq datasets for major immune lineages in whole blood and PBMCs are available from v21.0.proteinatlas.org (Uhlen et al. 2019). The read counts were normalized to transcripts per million protein coding genes (pTPM) for each cell type. The gating strategy for flow sorting is indicated in Additional file 9: Table S2.

Healthy donor and patient samples

Cryopreserved peripheral blood for flow cytometry analysis and sorting was collected at the Department of Rheumatology, West China Hospital, Sichuan University. This study was approved by the West China Hospital Ethical Vetting Board, Sichuan University (number 2018–199). Diagnosis for RA was based on ACR (American College of Rheumatology)/EULAR (European League Against Rheumatism) 2010 rheumatoid arthritis classification criteria (Kay and Upchurch 2012). The demographic and clinical laboratory data of these patients were obtained from hospital records and are summarized in Additional file 10: Table S3. RA cases and controls did not reveal any significant differences with regard to age and sex. DAS28 (CRP) = 0.56*√(TJC28) + 0.28*√(SJC28) + 0.014*GH + 0.36*ln(CRP + 1) + 0.96, where TJC = tender joint count and SJC = swollen joint count (Wells et al. 2009).

Cell preparation

Total peripheral blood mononuclear cells (PBMCs) were isolated from fresh blood of healthy volunteers and RA patients by density gradient centrifugation using Percoll (Solarbio). Human PBMCs were kept in liquid nitrogen before thawing in a 37 °C water bath. Cells were resuspended in T-cell culture medium (RPMI-1640, 10% fetal bovine serum (HyClone), 2 mM L-glutamine (Solarbio), 100 U/ml penicillin, 100 μg/ml streptomycin (both Solarbio), and 5 mM 2-mercaptoethanol (Sigma–Aldrich)). Samples were then rested overnight at 37 °C before restimulation with 50 ng ml −1 phorbol 12-myristate 13-acetate (PMA, Sigma–Aldrich) and 500 ng ml−1 ionomycin (ION, Sigma–Aldrich) in the presence of brefeldin A (BD Biosciences) for 4 h at 37 °C. Synovial tissue samples were disaggregated into single-cell suspensions as previously described (Carter et al. 2011). Murine synovial tissues were isolated from the knee joints of AIA mice and incubated with 1 mg/mL collagenase IV (Thermo Fisher Scientific) and 100 μg/ml DNase I (Roche) in RPMI in a 37 °C water bath for 45 min. Single-cell suspensions were obtained by passing cells through 70 μm cell strainers for flow cytometry staining.

Flow cytometry

For surface staining, the single-cell suspensions were washed and stained with fluorescence-conjugated antibodies for 15 min at room temperature. For intracellular cytokine staining, the combined surface- and live/dead-stained sample was fixed with fixation buffer (Thermo Fisher Scientific) for 20 min at room temperature. Fixed cells were washed twice with permeabilization buffer (Thermo Fisher Scientific). Cells were resuspended in 100 μl of intracellular antibody mixture in permeabilization buffer for 30 min at room temperature. The sample was washed, the supernatant was removed, and the cells were resuspended in cell staining medium. All fluorescence-conjugated antibodies used for flow cytometry are listed in Additional file 11: Table S4. Cells were acquired and analyzed with FACSAria SORP flow cytometers (BD Biosciences). All flow cytometry data were analyzed with NovoExpress software (Agilent). Memory CD4 T cells (CD4+ CD45RO+ CD45RA−) from both healthy and RA cohorts were separated by flow-assisted cell sorting (FACSAria SORP, BD Biosciences) for subsequent stimulation in cell culture.

T cell culture ex vivo

Sorted human memory CD4 T cells were cultured in 48-well plates (5 × 105 cells per well) containing plate-bound anti-CD3 (5 μg/ml, BD Bioscience) and soluble anti-CD28 (1 μg/ml, BD Bioscience) supplemented with 20 ng/ml IL-1β (Peprotech). For quantitative real-time PCR and ELISA, cells or supernatants were collected on days 1 and 2. Murine splenic CD4 T cells were enriched by a mojoSort™ mouse CD4 T cell isolation kit (Biolegend), cultured with plate-bound anti-CD3 (2 μg/ml, BD Bioscience) and soluble anti-CD28 (1 μg/ml, BD Bioscience), and stimulated with 20 ng/ml IL-1β (Peprotech) for 30 min to study the phosphorylation of signaling molecules or 48 h for cytokine production by ELISA or flow cytometry.

Mouse primary CD4 T cell transfection

Mouse primary CD4 T cells were transfected with Dharmacon Accell nontargeting control (Horizon Discovery, D-00190–10) or mouse C/EBPβ-specific siRNA pool (composed of the following four antisense sequences: I, UGGUUUACAUGUCGACUAA; II, UGGUUUACAUGUUUUCUGA; III, UGGUUUACAUGUUUUCCUA; IV, UGGUUUACAUGUUGUGUGA. Horizon Discovery, E-043110–00) using Accell delivery medium, following the manufacturer’s instructions.

Activation of memory CD4 T cells ex vivo

Memory CD4+ T cells were flow-sorted from splenocytes of CFA/mBSA-immunized mice as responder cells. To obtain stimulatory cells (providing additional costimulatory signals for T cell activation), single splenocyte suspensions were obtained from synergetic mice and treated with red blood cell lysis solution at room temperature for 5 min to remove erythrocytes, followed by mitomycin C treatment at 37 ℃ for 20 min, washed and used as stimulatory cells. Then, responder cells (1 × 105) were cultured with mitomycin C-treated stimulatory cells (5 × 105) in 96-well flat-bottom plates in the presence of 25 μg/ml soluble mBSA at 37 ℃ for 3 days. The procedures to establish the coculture system of stimulatory cells/responder cells were previously described (Kruisbeek et al. 2004; Wong et al. 2009) with minor modifications. Cytokine production in memory CD4 T cells in this coculture system was analyzed with intracellular staining by flow cytometry.

ELISA

Cytokine production of human or mouse CD4 T cell culture supernatants was measured using uncoated ELISA kits (Thermo Fisher Scientific) or specific mouse ELISA kits (Neobioscience). All procedures were performed as recommended by the manufacturer.

Plasmids, retrovirus production and transduction

The SIGIRR cDNA was PCR amplified and cloned into pMIG (SIGIRR-IRES-GFP). Phoenix helper-free retrovirus producer lines (Ivanov et al. 2006; Schraml et al. 2009) were transfected with 10 μg of the indicated plasmids using polyethylenimine (Polysciences). The viral supernatant was collected and supplemented with 5 mg/ml polybrene (Yeasen). For viral transduction, sorted mouse CD4 T cells were plated, fresh retrovirus supernatant was added, and the cells were spun at 2500 rpm for 1.5 h at 30 °C. After spin infection, the cells were cultured in T cell culture medium and harvested on day 3 for intracellular cytokine staining and quantitative real-time PCR analysis.

Generation of Sigirr −/− mice

Sigirr +/− founder mice on a C57BL/6 background were generated and purchased from Cyagen Biosciences. Male founder mice were bred with female mice to generate Sigirr +/+ and Sigirr −/− mice. SIGIRR KO mice were identified by genotyping PCRs with F1, F2 and R1 primers (listed in Additional file 12: Table S5) performed on tails at 3–4 weeks of age. Mice were housed in an animal facility (SPF condition) at Sichuan University in compliance with the guidelines set by West China Hospital Institutional Animal Care and Use Committee, Sichuan University.

Induction of antigen-induced arthritis (AIA)

AIA was induced as previously described (Engdahl et al. 2013). C57/BL6 mice were immunized with 0.2 mg of methylated bovine serum albumin (mBSA; Sigma–Aldrich) dissolved in phosphate buffered saline (PBS) and emulsified with an equal volume of Freund’s complete adjuvant (Sigma–Aldrich). One week later, the mice received a second injection of 0.3 mg of mBSA dissolved in 30 μl of vehicle into the right knee joint (day 0). The left knee was injected with vehicle and used as an internal control. Animals were inspected daily for arthritis development by measuring the knee joint diameter using a micrometer. Arthritis severity was assessed through a comparison between the experimental right limb and the control left limb.

Histology and immunohistochemistry of inflammatory arthritis

Four days after the intra-articular injection of mBSA, the mice were anesthetized and sacrificed. Knees from the immunized mice were separately placed in 4% formaldehyde, decalcified, and embedded in paraffin. Sections were stained with hematoxylin and eosin. For immunohistochemical staining, 5-μm-thick sections of paraformaldehyde-fixed mouse joints were deparaffinized and hydrated with distilled water. Heat-induced epitope retrieval was performed by incubation in citrate buffer (pH = 6.0) at 98 °C for 5 min, and endogenous peroxidase was quenched by treatment with 3% H2O2 in PBS for 5 min. Next, goat serum (ZSGB-Bio) was applied for 30 min to block nonspecific binding on the tissue sections. CD4 T cells in the synovium were detected by incubation with a rabbit anti-CD4 primary antibody (Abcam) overnight at 4 °C in a humidified chamber. After washing with PBS, the histological slides were incubated with biotinylated anti-rabbit secondary Ab (ZSGB-Bio) and HRP-labeled streptavidin complex (ZSGB-Bio) for 30 min. The staining reaction for CD4 was visualized by a DAB Peroxidase Substrate Kit (MXB Biotechnologies), and positive signals of DAB chromogen were developed as brown precipitates. The slides were counterstained with hematoxylin (Beyotime Biotechnology) to detect nuclei. Images of the stained slides of joints were obtained with a ZEISS microscope. Five random fields of view (400 ×) of each mouse psoriatic plaque section stained with H&E were assessed using the Baker Score system. Similarly, mast cells were quantified by counting five randomly selected fields in brown-stained slides. Immunohistochemical staining was quantified by computer-assisted methods.

Isolation of total and Naïve CD4+ T Cells

CD4+ T cells were purified from spleens using anti-CD4 magnetic microbeads (Biolegend) (purity > 90%). For naïve CD4+ T cells, single-cell suspensions were first incubated with Fixable Viability Stain 780 (BD Bioscience, 1:1000) to discriminate live cells, followed by incubation with anti-CD25-APC, anti-CD4-BV421, anti-CD62L-PE, and anti-CD44-PerCP/Cy 5.5. Cell sorting was performed on an ARIA II SORP cytometer (BD Bioscience) to obtain a pure population of naïve CD4+ CD25− CD44−CD62L+ T cells (> 97% purity). A total of 5 × 106 naïve CD4 T cells derived from the spleen of Sigirr +/+ or Sigirr −/− mice were intravenously transferred into Rag1−/− mice. One day later, the recipient mice were immunized with mBSA as described above.

Quantitative real-time PCR (QRT-PCR)

Total RNA was extracted from tissues using TRIzol (Thermo Fisher Scientific), reverse-transcribed into cDNA using PrimeScript™ RT reagent Kit with gDNA Eraser (Takara) and followed by PCR amplification in triplicate using ChamQ SYBR qPCR Master Mix (Vazyme) in the CFX96 touch real-time PCR detection system (Bio-Rad). The primer sets for real-time PCR can be found in Additional file 12: Table S5.

Immunoblots

Murine CD4 T cells were collected after 30 min of treatment with IL-1, and total cell lysates were subjected to SDS–PAGE and then blotted using the indicated antibodies (Additional file 13: Table S6). The quantitative analysis of integral OD of the bands in the Western blot was performed using.

ImageJ software.

Transcriptional regulation of C/EBPβ on TNF gene

DNA binding motifs of transcription factor C/EBPβ were predicted by retrieving JASPAR, the online collection of transcription factor DNA-binding preferences (Fornes et al. 2020). Experimental binding of C/EBPβ to TNF promotor regions was revealed by ChIP-seq data deposited in the Cistrome Data Browser (Zheng et al. 2020).

Statistical analysis

All statistical analyses were performed with GraphPad Prism software. Analysis of the differences in SIGIRR expression and intracellular cytokines between healthy and RA populations was assessed using the Mann–Whitney test, and the relationship between SIGIRR expression and disease activity DAS28 was established by Spearman’s correlation test. Other comparison analyses in this study were assessed by unpaired t test (between two groups) or by multiple t tests using the Holm‒Sidak Method as indicated.