Article Text
PDF
Abstract
Objectives T helper 9 (Th9) cells are recognised for their characteristic expression of the transcription factor PU.1 and production of interleukin-9 (IL-9), which has been implicated in various autoimmune diseases. However, its precise relationship with rheumatoid arthritis (RA) pathogenesis needs to be further clarified.
Methods The expression levels of PU.1 and IL-9 in patients with RA were determined by ELISA, western blotting (WB) and immunohistochemical staining. PU.1-T cell-conditional knockout (KO) mice, IL-9 KO and IL-9R KO mice were used to establish collagen antibody-induced arthritis (CAIA), respectively. The inhibitor of PU.1 and IL-9 blocking antibody was used in collagen-induced arthritis (CIA). In an in vitro study, the effects of IL-9 were investigated using siRNAs and IL-9 recombinant proteins. Finally, the underlying mechanisms were further investigated by luciferase reporter analysis, WB and Chip-qPCR.
Results The upregulation of IL-9 expression in patients with RA exhibited a positive correlation with clinical markers. Using CAIA and CIA model, we demonstrated that interventions targeting PU.1 and IL-9 substantially mitigated the inflammatory phenotype. Furthermore, in vitro assays provided the proinflammatory role of IL-9, particularly in the hyperactivation of macrophages and fibroblast-like synoviocytes. Mechanistically, we uncovered that PU.1 and IL-9 form a positive feedback loop in RA: (1) PU.1 directly binds to the IL-9 promoter, activating its transcription and (2) Th9-derived IL-9 induces PU.1 via the IL-9R-JAK1/STAT3 pathway.
Conclusions These results support that the PU.1-IL-9 axis forms a positive loop in Th9 dysregulation of RA. Targeting this signalling axis presents a potential target approach for treating RA.
- Arthritis, Rheumatoid
- Cytokines
- T-Lymphocytes
Data availability statement
Data are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
The pathogenesis of rheumatoid arthritis (RA) is complex and existing drugs have many shortcomings. A specific subtype of CD4+ T cells, T helper 9 (Th9) cells, known for secreting IL-9, has been implicated in various autoimmune diseases. Th9-derived IL-9 is highly expressed in patients with RA. Our previous study demonstrated that PU.1, an essential transcription factor, plays a promoting role in RA pathogenesis.
WHAT THIS STUDY ADDS
We show that PU.1 can cause T cell abnormality in RA by promoting Th9 differentiation and IL-9 production. Our results demonstrate that PU.1 and IL-9 form a positive feedback loop in RA: (1) PU.1 directly binds to the IL-9 promoter, activating its transcription and (2) Th9-derived IL-9 induces PU.1 via the IL-9R-JAK1/STAT3 pathway.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
We demonstrate that the PU.1-IL-9 axis forms a positive loop in Th9 dysregulation of RA. Targeting this signalling axis presents a potential target approach for treating RA.
Introduction
Rheumatoid arthritis (RA) is a common autoimmune disease (AID) with a global incidence of approximately 1%.1 Key clinical features of RA are immune system disorders, synovitis and bone/cartilage erosion and degradation.2 Despite significant research, the complete understanding of the pathogenesis of RA remains elusive,3 posing challenges in developing effective drugs. Currently, RA treatments include the application of disease-modifying antirheumatic drugs. However, these medications have some shortcomings, including adverse reactions, low response rates and drug resistance.4 Therefore, elucidating the pathogenesis of RA is essential in determining effective treatment for RA.5–7
PU.1, a transcription factor of the erythroblast transformation specific (EST) family, plays an important role in developing and differentiating various immune cells (such as T cells and dendritic cells) and in AIDs.8–10 However, the effect of PU.1, a key cellular component in RA, remains unclear. RA is characterised by disturbances in the balance of various T cell subtypes,6 11 12 including Th1, Th2, Th17 and Tregs, which are well-investigated contributors to the disease.13–15 In addition to these subtypes, Th9 is a relatively new subtype in the overall proportion of T cells. The role of Th9, along with its specific cytokine IL-9, has been investigated in several AIDs, including inflammatory bowel diseases, multiple sclerosis and systemic lupus erythematosus.16–18 Additionally, interactions between Th9 and Th17/Treg have been reported in previous studies.19–22 Few studies have suggested that Th9 cells may play a pathogenic role in RA progression23 24; however, definitive studies remain lacking. Therefore, this study aims to investigate the potential pathogenic role of Th9 cells in RA progression. PU.1 is a key transcription factor that promotes Th9 differentiation.25 Based on previous studies, including our investigation of PU.1,26 we hypothesise that PU.1 may promote the course of RA by activating Th9 differentiation and IL-9 production.
Materials and methods
Patients and normal volunteers
This study was approved by the Biomedical Ethics Committee of Anhui Medical University (83230273). The diagnosis of RA followed the 2010 clinical and radiological criteria outlined by the American College of Rheumatology and European League Against Rheumatism.27 Osteoarthritis (OA) diagnoses were based on the 2018 bone and joint diagnosis and treatment guidelines updated by the Orthopaedic Branch of the Chinese Medical Association.28 Blood samples were obtained from patients with RA and healthy controls from the First Affiliated Hospital of Anhui Medical University. The human knee synovial tissues used in this study were obtained from patients with RA and OA who had undergone knee arthroplasty. Normal synovial tissue was obtained during knee surgery for non-inflammatory knee diseases. Online supplemental table 1 presents detailed clinical characteristics. Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.
Supplemental material
Animal
UREΔ mice (Spi1tm1.3Dgt/J, JAX Strain #: 006083) were purchased from Jackson lab (USA). Male DBA/1JGpt mice, aged 7–8 weeks, were purchased from GemPharmatech. IL9-Luc-2A-EGFP mice, SPI1 flox/flox, Lck-Cre mice, IL9 knockout (KO) mice, IL9R KO mice, aged 7–8 weeks, were purchased from Cyagen. All mice were housed in a specific pathogen-free mouse facility (room temperature: 20°C–22°C; room humidity: 40%–60%) with free access to food and water under a 12 hours light/dark cycle. During the experimental period, experimental and control mice were housed in individual cages to maintain consistent environmental conditions. This study was approved by the Ethics Committee of the Institute of Clinical Pharmacology of Anhui Medical University (approval no. PZ-2022-024).
Primary mouse macrophage isolation
After euthanasia, the mice were sterilised with 75% ethanol. Subsequently, 10 mL of precooled phosphate-buffered saline (PBS) was intraperitoneally administered, followed by gently massaging the abdomen for 5 min. The peritoneal cavity contents were aspirated into a 15 mL centrifuge tube, placed on ice and centrifuged (300 g, 10 min). The supernatant was discarded, and 10% Dulbecco’s Modified Eagle Medium (DMEM) medium was added to resuspend the cells into a six-well plate. After 4 hours, the cells were washed thrice with PBS to remove other suspended cells. Flow cytometry analysis was performed to confirm the primary mouse macrophage purity using anti-CD11b (BioLegend, B342456) and anti-F4/80 (BioLegend, B353283) antibodies.
Primary human RA-fibroblast-like synoviocytes isolation
The RA synovium was washed with PBS, minced and evenly arranged at the bottom of a culture flask. The flask was inverted to face upwards. Subsequently, 2 mL of DMEM with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin was added, and the cells were cultured at 37°C and 5% CO2. The medium was replaced every 3 days. After 10–14 days of culture, the adherent cells were trypsinised with trypsin-EDTA and transferred to new flasks. Fibroblast-like synoviocytes (FLS) morphology was confirmed using optical microscopy. To ensure primary cell purity, flow cytometry analysis with an anti-human CD90 antibody (BioLegend, 328107) was performed. The RA-FLS used in these experiments were between first and third passages.
Collagen antibody-induced arthritis model
On day 0, mice underwent intraperitoneal injection with a mixture containing 4 mg of five-clonal collagen antibody (Chondrex, 53010). On day 3, an additional intraperitoneal injection of 40 µg lipopolysaccharide was administered. On day 14, the mice were euthanised under anaesthesia. For the assessment of joint swelling, the scoring criteria involved counting ankle and toe joints. Each instance of joint swelling was assigned a score of one point, with a maximum of 22 swollen joints per mouse. The arthritis index was graded on a scale from 0 to 4 according to the following criteria: 0=absence of local redness or swelling; 1=swelling of the finger joints; 2=slight swelling of the ankle or wrist joints; 3=severe swelling of the entire foot and paw and 4=stiffness or deformity of the paw.
Collagen-induced arthritis model
In this experimental protocol, 10 mg chicken type II collagen (Chondrex, 20011) was dissolved in 5 mL of acetic acid (0.01 mol/L) and emulsified by combining an equal volume with 5 mg/mL Complete Fuchs adjuvant (Chondrex, 7023). On day 0, mice underwent intradermal injection of 0.15 mL of the collagen type II (CII) emulsion after depilation on the back and tail root. On day 21, a subsequent injection of 0.1 mL of CII emulsion was administered to enhance stimulation. The experimenters, blinded to the treatments, recorded joint swelling, arthritis index and body weight every 3 days. On day 28, failed models were excluded from further analysis. Following the appearance of joint swelling on day 29, mice received intraperitoneal injections of 0.3 mg DB2313 (MCE, HY-124629) every 3 days while the model group received an equivalent volume of PBS for a duration of 3 weeks. In a separate set of experiments, 200 µg IL-9mAb (Biolegend, MM9C1) was administered through intraperitoneal injections every 3 days, with the model group receiving an equivalent amount of IgG for 2 weeks. Subsequently, collagen-induced arthritis (CIA) mice were euthanised, and blood, spleen, knee and ankle joints were collected for subsequent analyses.
Bioluminescence imaging
Following the establishment of the collagen antibody-induced arthritis (CAIA) model in il9-Luc-2A-EGFP mice, longitudinal in vivo imaging sessions were conducted on days 4, 7, 10, 13 and 16. To optimise imaging quality, mice underwent depilation and targeted treatment at the abdomen and joints to minimise potential interference with the imaging signal. The experimental subjects were anaesthetised using isoflurane and subsequently received an intraperitoneal injection of 150 µg/g D-Luciferin (potassium salt). After a 15 min incubation period, the mice were carefully positioned within a specialised imaging instrument (Spectral Instruments, Amix). The optical wavelength parameters were meticulously configured to 430–510 nm to capture the emitted bioluminescent signals. Subsequently, images were acquired and subjected to analysis using Aura Imaging software for comprehensive evaluation.
Immunohistochemical staining
Tissues were meticulously embedded, sectioned onto slides and subjected to a baking process at 60°C for a duration of 2 hours. Subsequently, deparaffinisation was carried out using xylene and a gradient of alcohol. Immunohistochemical (IHC) staining was performed using the Multiplex Fluorescence Immunohistochemistry Kit (Panovue, 10002100050) and the Double Staining Detection Kit (ZSGB-Bio, DS-0003), following the manufacturer’s stipulated protocols. Primary antibodies targeting specific proteins or modifications were employed, including anti-human IL-9 (Abcam, ab181397, 1:100), anti-human PU.1 (Abcam, ab88082, 1:100) and anti-human CD4 (Proteintech, 67786-1-Ig, 1:200). Following the application of neutral resin to seal the sections, the stained tissues were meticulously examined, and the outcomes were interpreted under a light microscope (3DHISTECH, Pannoramic MIDI II).
ELISA analysis
To measure the inflammatory cytokines IL-9 in the plasma, ELISA was performed with the commercially available kits (Signalway Antibody, EK1226).
Immunofluorescence
The human and mouse synovium were embedded in paraffin and sectioned into 5 µm slices. These sections were subjected to a 2-hour incubation at 60℃, followed by deparaffinisation and hydration using a gradient of ethanol and then washed with PBS three times. Permeabilisation with 0.5% Triton X-100 was performed for 15 min, before washing thrice with PBS. After heat-induced antigen retrieval in citrate buffer (pH 6.8), the sections underwent a 10 min blockage of endogenous peroxidase, rinsed thrice with PBS and blocked with serum for 15 min. Subsequently, primary antibodies, including anti-human IL-9 (Abcam, ab181397, 1:100), anti-human PU.1 (Abcam, ab88082, 1:50) and anti-human CD4 (Proteintech, 67786-1-Ig, 1:100), were added and incubated at 4°C overnight. The next day, after incubation with fluorescent secondary antibodies for 1 hour at room temperature, the cells were rinsed thrice with PBS. Nuclei were stained with DAPI for 8 min under light-free conditions, washed three times with PBS and sealed with a blocking solution containing an anti-fluorescence quencher. Finally, the stained cells were observed under a fluorescence microscope (Leica, SP8).
Western blotting
PBS-washed cells or ground tissues were treated with Radiation Immunoprecipitation Assay Buffer along with protease inhibitors and phenylmethylsulfonyl fluoride in a ratio of 100:1:1. The mixture was incubated on ice for 30 min. Subsequently, the sample was centrifuged at 4°C, 12 000 rpm for 15 min, and the supernatant was collected. Protein concentration was determined using the BCA method. Protein samples were loaded onto an SDS-PAGE gel for electrophoretic separation. Proteins were then transferred onto a polyvinylidene-fluoride membrane (PVDF). The membrane was briefly immersed in a rapid containment solution (NCS Biotech, P30500) for 10 min, followed by three washes with Tris-buffered saline containing Tween 20 (TBST) and one wash with PBS, each lasting 8 min. The PVDF membrane was incubated with specific primary antibodies overnight at 4°C. The primary antibodies used included beta-Actin Antibody (Affinity, T0022, 1:5000), anti-human IL-9 (Abcom, ab181397, 1:1000), anti-mouse IL-9 (LS Bio, LS-C487693, 1:1000), anti-PU.1/Spi1 antibody (Abcam, ab230336, 1:1000), anti-JAK1 antibody (Cell Signaling, 3344S, 1:1000), anti-phospho-JAK1 Antibody (Cell Signaling, 3344S, 1:1000), anti-STAT3 antibody (Abcam, ab119352, 1:1000), anti-phospho-STAT3 Antibody (Cell Signaling, 9145S, 1:1000). After three washes with TBST, the membrane was incubated with the corresponding secondary antibodies for 1 hour at room temperature. Subsequently, the membrane was washed three times with TBST and once with PBS. Signal detection was performed using a developer (Biosharp, BL520B) according to the manufacturer’s instructions. The grey scale values of each band were analysed using image software. The grey scale values of the target proteins were normalised to the grey scale values of the control proteins for statistical analysis.
Ultrasound assessment
Before ultrasonography, each mouse was consistently anaesthetised with 2.5% isoflurane, and hair was removed from the knee joint using a depilatory cream. A coupling agent was applied to the surface of the knee joint using the Doppler flow pattern of a high-resolution small-animal ultrasound real-time imaging system (VisualSonics, 1800568S). Recording parameters and probe positions were adjusted to optimise results.
MicroCT
After euthanising the mice, knee and knuckle joints were carefully extracted, and excess surrounding tissues were removed. The samples were then subjected to fixation by immersing them in paraformaldehyde. Subsequently, the prepared samples were securely positioned on a holder for imaging. Micro-CT scans of the hind limbs were conducted using a specialised micro-CT system (NEMO, NMC-100). The scanning procedure involved capturing images from different directions and angles to ensure comprehensive visualisation of the joint structures. The specific scanning conditions were set as follows: scanning resolution of 7.5 µm, supply voltage of 90 kV and a current of 0.06 mA. The acquired projection data underwent image reconstruction using the acquisition software Cruiser, tailored for the micro-CT system. Following reconstruction, the obtained data were subjected to analysis using Avatar software (V.1.6.5.3).
H&E staining
The tissue sections were deparaffinised, hydrated with an alcohol gradient, stained with H&E, dehydrated and sealed with neutral glue. The histopathological changes were observed, before being recorded using a light microscope.
Masson staining
The paraffin sections were deparaffinised and then stained with Weigel Iron Haematoxylin for 5 min. Subsequently, they underwent alcohol fractionation with hydrochloric acid. Following that, they were stained with Masson-Lichon Red acidic reddish solution for 8 min. Following this, a series of washing steps were performed: 1 min with glacial acetic acid in water, 1 min with 1% phosphomolybdic acid solution, and another 1 min with glacial acetic acid solution. The samples were stained with an aniline blue dye solution for 1 min, washed with glacial acetic acid solution for 1 min, permeabilised with xylene and sealed with neutral glue. The staining results were observed under a light microscope.
Safranin O staining
The sections were dewaxed, soaked in Safranin O staining solution for 3 min, washed with distilled water for 1 min, soaked in Fast Green staining solution for 2 min, washed with distilled water for 1 min, fractionated with 1% glacial acetic acid for 1 min, dehydrated in 95% ethanol and permeabilised with xylene. Finally, changes in the staining were observed after the slices were sealed with neutral glue.
TRAP staining
Paraffin sections were deparaffinised, and the substrate solution was prepared following the instructions of the TRAP staining kit (BestBio, BB-44 212-3) and placed on a 37°C water bath for 10 min. Following this, the sections were mixed with the substrate solution, incubated for 1 hour in the dark, washed with water for 1 min and counterstained with haematoxylin for 2 min. Images were captured after sealing the slices with a neutral gel, and a positive reaction was identified by the presence of a bright or dark red granule localised in the cytoplasm.
Flow cytometry
After centrifugation and washing with PBS, surface flow cytometry antibodies were added according to the instruction of the manufacturer: anti-mouse CD4 (Biolegend, 100438), anti-mouse CD25 (Biolegend, 102011), anti-mouse CD11b (Biolegend, 101236), anti-mouse F4/80 (Biolegend, 123108), anti-mouse CD86 (Miltenyi Biotec, 130-102-604), anti-human CD4 (Biolegend, 302604) and anti-human CD86 (Biolegend, 374208). The mixture was incubated for 30 min at 4°C in the dark, followed by centrifugation of cells at 1500 rpm for 5 min. The supernatant was discarded, and 100 µL of fixative was added for 15 min at room temperature. After centrifuging at 1500 rpm for 5 min, the supernatant was discarded, and 100 µL of cell lysis buffer was added and incubated at 4°C for 15 min. After washing with PBS, intracellular staining antibodies were added: anti-mouse IL-17A (eBioscience, 2317033), anti-mouse Foxp3 (eBioscience, 2403335), anti-mouse IL-9 (Biolegend, 514106), anti-mouse CD206 (Biolegend, 141708), anti-human IL-9 (Biolegend, 507605), anti-human CD68 (eBioscience, 1984137) and anti-human CD206 (Biolegend, 321125). The cells were incubated at 4°C in the dark for 30 min. Subsequently, cells were centrifuged at 1500 rpm for 5 min, unbound antibodies were discarded and the cells were resuspended in 200 µL PBS for testing using a flow cytometer (Beckman Coulter, A00-1-1102).
Magnetic cell sorting of primary human/mouse CD4+ T cells
Peripheral blood mononuclear cell (PBMC) was obtained from human peripheral blood by differential centrifugation and washed with PBS. Anti-human CD4 (Miltenyi Biotec, 130-045-101) was added following the instruction of the manufacturer, incubated at 4°C for 15 min, centrifuged at 1500 rpm for 5 min, and the supernatant was discarded. The cells were resuspended in 1 mL PBS. After sorting column treatment, the negative fraction flowed out, column was removed from the magnetic sorter and positive fraction was eluted to obtain a purified fraction. Flow cytometry was used to confirm the purity of primary cells. A portion of the mouse spleen was excised, and a 5 mL of mouse lymphatic isolate was added. The spleen was thoroughly ground under a gauze screen, and the resulting grist was transferred to a 15 mL centrifuge tube. Subsequently, 10% RPMI-1640 was slowly added along the wall to obtain lymphocytes through differential centrifugation. Anti-mouse CD4 antibody (Miltenyi Biotec, 130-117-043) was added according to the instruction of the manufacturer, followed by the same procedure.
Human Th9 cell differentiation
In preparation for the experiment, 24-well plates were coated 1 day in advance with PBS containing 2.5 µg/mL anti-human CD3 (eBioscience, 16-0037-81) and 2.5 µg/mL anti-human CD28 (eBioscience, 16-0289-81) for overnight incubation at 4°C. PBMCs were isolated from human peripheral blood using differential centrifugation and washed with PBS. Magnetic cell sorting of primary human CD4+ T cells as mentioned before. Subsequently, the PBS in the precoated well plate was removed, and 1×106 Naive CD4 T cells were resuspended and added to the well plate along with 3 ng/mL Recombinant human TGF-β1 (R&D Systems, 240-B-002), 10 ng/mL Recombinant human IL-4 (R&D Systems, 204-IL-010) and 10 ng/mL anti-human IFN-γ (eBioscience, 16-7318-81). The cells were cultured for 4 days to generate Th9 cells for subsequent experiments.
Cell culture and transfection
EL4, Jurkat, THP-1 and primary CD4+ T cells were cultured in RPMI-1640 medium while RA-FLS and HEK293T cells were cultured in DMEM. In addition to the basic medium, a mixture of 10% FBS and 1% penicillin/streptomycin was added and cultured routinely at 37°C in a 5% CO2 incubator. For siRNA transfection, cells were cultured in six-well plates in a basal medium without penicillin/streptomycin for 12 hours. Subsequently, a 8 µL siRNA was dissolved in 200 µL OPTI- MEM medium (Gibco) and 6 µL Lipofectamine RNAiMAX (Invitrogen, 13778-150) was dissolved in 200 µL OPTI-MEM. The two solutions were incubated separately for 5 min. Once prepared, the siRNA and Lipofectamine IMAX solutions were combined in equal amounts and left to stabilise at room temperature for 15 min. After removing the original cell culture medium, each well received 1600 µL of the fresh medium and 400 µL of the prepared siRNA transfection mixture. The cells were then incubated in a cell incubator for 6 hours. Subsequently, the transfected medium was replaced with a normal medium, and the cells were cultured for 48 hours for subsequent experiments.
Dual-luciferase reporter assay
In this experimental procedure, 200 ng of either overexpression plasmid or mutant plasmid was first dissolved in a mixture comprising 1.25 µL P3000 Reagent (Invitrogen, 100022058) and 30 µL OPTI-MEM medium. Simultaneously, 1 µL Lipofectamine 3000 (Invitrogen, 100022052) was mixed with 30 µL OPTI-MEM medium. These two solutions were incubated separately for 5 min and then combined in equal proportions, stabilising at room temperature for an additional 15 min. Following the removal of the original cell culture medium, each well of a 96-well plate containing HEK 293 T cells received 50 µL of fresh medium and 50 µL of the prepared plasmid transfection mixture. The cells were then incubated in a cell culture incubator for 6 hours, after which the medium was replaced with a normal medium. After 24 hours of cell culture, the Firefly luciferase and Renilla luciferase intensities were measured according to the instructions of the Dual-Luciferase Reporter Assay Kit (Promega, E2920), and the ratio of the two fluorescence intensities in each well was calculated.
Transwell assay
The RA-FLS was resuspended in serum-free medium and placed in the upper chamber (8 µm) at a density of 5×103 cells/well. Simultaneously, the lower chamber was filled with 500 µL of normal medium and incubated at 37°C with 5% CO2 for 24 hours. In another experiment, the upper chamber contained untreated RA-FLS while the lower chamber was treated with RA-Th9 cell culture medium supernatant for 24 hours. After this incubation period, the cells remaining in the upper chamber were carefully removed. Subsequently, the migrated cells under the membrane were fixed with methanol, stained with 0.1% crystal violet for 15 min, and washed three times with PBS to remove unbound crystal violet. The upper side of the chambers was gently wiped with a cotton swab, and five random areas of each well were counted under a microscope.
RT-qPCR
Total RNA was extracted from cells on ice using TRIzol, and its concentration and purity were determined. Samples of acceptable purity were used for reverse transcription and PCR amplification assays according to the instructions of the manufacturer (Vazyme, R222-01). The PCR volumes were 20 µL, and the amplification conditions were determined in accordance with the instructions of the manufacturer (Vazyme, Q111-02). The relative quantification of gene expression was performed using the 2-ΔΔCT method, using the GAPDH gene as an internal reference.
CCK-8 assay
Following the designated treatment, 5×105 /mL of cell suspension (100 µL per well) was added to a 96-well plate with three replicate wells per set. Subsequently, 10 µL CCK-8 solution was added and incubated for 3 hour. After the 3 hours incubation period, absorbance was measured at 450 nm using microplate readers (Tecan, Infinite M1000 PRO).
CHIP-qPCR
The reaction was stopped by adding glycine to a final concentration of 125 mM and then prepared in Protease Inhibitor Cocktail. Next, MNase and SCIENTZ ultrasonic cell shredders were used to break down the DNA fragments into 150–900 bp fragments, which were identified by 2% agarose electrophoresis. Each IP reaction used approximately 5–10 µg of digested cross-linked chromatin, 5 µg of PU.1 antibody (Cell Signaling, 2266S) or STAT3 antibody (Cell Signaling, 12 640S), and non-specific IgG antibodies were incubated overnight at 4°C for the co-immunoprecipitation reaction. Next, the chromatin was eluted from the antibody-ProteinA/G Magnetic Beads and unlinked, then input DNA and ChIP DNA (Cell Signaling Technology; 14 209S) were purified, and resultant DNA was used for subsequent real-time PCR (Cell Signaling Technology; 88989) analysis.
Apoptosis
Following the successful induction of RA-Th9 cells, the cells were treated with 100 ng/mL of recombinant IL-9 (Abcam, AB50119) and either 10 µM of Tofacitinib (MCE, HY-40354) or 2.5 µM of Stattic (MCE, HY-13818) for a 48-hour incubation period. Apoptosis induction was achieved by treating the cells with 20 ng/mL of TNF-α (Peprotech, 300–01A) for 2 hours, serving as a positive control for apoptosis detection. After the 2 hours incubation, cells were collected, washed and subjected to flow cytometry studies and analysis to assess the impact of IL-9 on apoptosis. The experimental procedures and data analysis were conducted following the instructions provided in the kit (BestBio, BB-41033).
RNA-seq
PBMC samples from CIA mice were analysed through high-throughput transcriptome sequencing using BGI (Shenzhen, China) to identify aberrantly and specifically expressed genes. Kyoto Encyclopaedia of Genes and Genomes pathway analysis was performed to estimate the function of dysregulated genes in CAIA mice. Sequencing and analysis were performed according to the standard protocol for RNA-Seq from the BGI.
Statistical analysis
Statistical analysis was performed by using GraphPad PRISM V.8.2.1 (GraphPad, USA). Data presented in the figures are expressed as means±SD. To assess differences between experimental groups, one-way or two-way analysis of variance, Student’s t-tests, and rank sum tests were used. p<0.05 was considered significant.
Results
The relationship of PU.1 and IL-9 in patients with RA and arthritic mice
The blood expression of IL-9 was significantly higher in patients with RA than in normal controls (figure 1A). Pearson’s correlation test showed that Disease Activity Score 28, erythrocyte sedimentation rate, C reactive protein and anti-citrullinated protein antibodies values were positively correlated with IL-9 levels (r=0.4504, p<0.0001; r=0.3707, p<0.0001; r=0.2963, p=0.0055; r=0.1914, p<0.0001, respectively) (figure 1B–E). In our previous studies, PU.1, a transcriptional factor, was found to be upregulated in the joint synovium of patients with RA. IHC double staining revealed significant co-localisation of PU.1 and IL-9 in the synovial tissue of patients with RA than that of normal individuals and patients with OA (figure 1F), suggesting a positive correlation between PU.1 and IL-9 in the joint synovium of patients with RA. Western blotting confirmed elevated IL-9 expression in the joint synovium of patients with RA (figure 1G). Immunofluorescence multiple staining results revealed co-localised expression of PU.1, IL-9 and CD4+ T cells in the joint synovium of patients with RA (figure 1H). Furthermore, an il9-Luc-2A-EGFP mice was used to establish a CAIA model (online supplemental figure 1A–D). The IL-9 signal in the joint was positively correlated with the progress of CAIA and IL-9 increased in the joint synovium of CIA (figure 1I,online supplemental figure 1E). These results indicate a potential role for PU.1 and IL-9 in RA. To explore PU.1 as a potential arthritis therapeutic target, the PU.1 inhibitor DB2313 was tested in CIA mouse model (online supplemental figure 1F,G). Beyond typical arthritis relief, DB2313 downregulated IL-9 expression in the spleen and PBMCs of CIA mice (figure 1J), suggesting the role of PU.1 in arthritis development via Th9 cell modulation.
The relationship of PU.1 and IL-9 in patients with RA and arthritic mice. (A) IL-9 expression was analysed in peripheral blood from the normal control group and patients with RA (n=52 and 207, respectively). (B–E) Pearson correlation test examined IL-9 correlation with DAS28, ESR, CRP and ACPA (n=204, 209, 209 and 202, respectively). (F) Co-localisation of IL-9 and PU.1 in the synovium was compared among normal controls, patients with OA and RA (n=3). (G) IL-9 protein expression in RA and OA synovium was compared (n=4). (H) Co-localisation of PU.1, IL-9 and CD4 in the joint synovium of patients with RA (n=3). (I) Bioluminescence imaging demonstrates IL-9 signal in joint was positively correlated with progress of CAIA (n=4). (J) DB2313 effect on the proportions of Th9 in the spleen and PBMCs of collagen-induced arthritis (CIA) mice assessed by flow cytometry (n=4–5). ACAP, anticitrullinated protein antibodies; CAIA, collagen antibody-induced arthritis; CRP, C reactive protein; DAS28, Disease Activity Score 28; ESR, erythrocyte sedimentation rate; OA, osteoarthritis; PBMCs, peripheral blood mononuclear cells; RA, rheumatoid arthritis.
Effects of PU.1 on Th9 of SPI1: LCK-cre mice-established CAIA model
To investigate the effect of PU.1 on T cells further, T cell-conditional KO mice (SPI1 flox/flox,Lck-Cre ) were used to create a CAIA model (figure 2A). As expected, PU.1-specific KO in T cells alleviated arthritis development in SPI1 flox/flox, Lck-Cre mice, evidenced by reduced arthritis index, joint swelling, angiogenesis and joint synovium pathology (figure 2B–E). IL-9 secretion in the spleen and PBMC of the CAIA model constructed using SPI1 flox/flox,Lck-Cre mice was significantly reduced (figure 2F,I). SPI1 flox/flox, Lck-Cre CAIA mice exhibited decreased Th9 and Th17 and elevated Treg cell proportions (figure 2G,H,J,K). Moreover, the proinflammatory phenotype of peritoneal macrophages (figure 2L,M) and hyperplasia of FLS (figure 2N) were also repressed in SPI1 flox/flox, Lck-Cre CAIA mice. These results confirmed that PU.1 promotes the development of arthritis by specifically regulating T cells.
Effects of PU.1 on Th9 of SPI1: LCK-cre mice-established CAIA model. (A) A CAIA model was constructed using PU.1 LCK-cre conditional knockout mice (SPI1flox/flox,Lck-Cre). (B) PU.1 KO effect in T cells on joint swelling of CAIA model (n=8). (C, D) PU.1 KO effect in T cells on joint swelling number and arthritis index of CAIA model. (E) H&E and Safranin O staining were used to evaluate PU.1 KO effect in T cells on joint pathology of the CAIA model (n=3). (F–K) Flow cytometry analysis shows PU.1 KO effect in T cells on Th9, Th17 and Treg in spleen and PBMC of CAIA model (n=3–8). (L, M) Flow cytometry analysis demonstrates PU.1 KO effect in T cells on peritoneal macrophage polarisation of CAIA model (n=5). (N) IHC shows PU.1 KO effect in T cells on FLS (FAP-α+) of CAIA model (n=3). CAIA, collagen antibody-induced arthritis; FLS, fibroblast-like synoviocytes; IHC, immunohistochemical; KO, knockout; LPS, lipopolysaccharide; PBMC, peripheral blood mononuclear cell.
Effects of PU.1-IL9 axis on Th9, macrophage and RA-FLS in vitro
Bioinformatics and literature screening suggest that PU.1 may directly activate IL-9 transcription in T cells.29 30 In vitro differentiation of sorted human CD4+ T cells into Th9 cells (figure 3A) revealed that PU.1 promotes IL-9 production (figure 3B). Luciferase reporter and CHIP-qPCR results confirmed that PU.1 could directly bind to the IL-9 promoter region, activating its transcription (figure 3C,D). At in vitro level, the focus shifted to understanding how IL-9 influences other cells in the joint synovium of patients with RA, specifically macrophages and FLS.7 In RA, synovial macrophages exhibit proarthritis phenotypes while FLS hyperplasia contributes to bone and cartilage erosion. Recombinant IL-9 induced RA-FLS proliferation and migration and promoted proinflammatory M1 polarisation of macrophages in vitro. Conditioned medium from PU.1 knockdown Th9 cells repressed proliferation, migration and MMPs expression of RA-FLS; these effects were countered by exogenous IL-9 recombinant protein (figure 3E–H). Furthermore, IL-9 facilitated proinflammatory M1 polarisation of macrophages (figure 3I–K, online supplemental figue 2), indicating the PU.1-IL9 axis promotes proinflammatory functions in macrophages and RA-FLS in vitro.
Effects of PU.1-IL9 axis on Th9, macrophage and RA-FLS in vitro. (A) Illustration depicting the in vitro differentiation process of CD4+ naive T cells from patients with RA into Th9 cells. (B) Effect of siRNA-mediated PU.1 knockdown on Th9 differentiation (n=5). (C) Luciferase reporter assay demonstrates direct binding of PU.1 to IL-9 promoter region (n=4). (D) CHIP-qPCR verifies the binding of PU.1 protein to the IL-9 promoter region (n=3). (E) Experimental design diagram of the effect of PU.1-IL9 axis on RA-FLS in vitro. (F) Transwell assay to verify the effect of PU.1-IL9 axis on the migration of RA-FLS in vitro (n=5). (G) CCK-8 method to verify PU.1-IL9 axis effect on proliferation of RA-FLS in vitro (n=3). (H) RT-qPCR was employed to assess the expression levels of MMPs in RA-FLS (n=3). (I) Experimental design diagram of PU.1-IL9 axis effect on macrophages in vitro. (J) Flow cytometry analysis shows PU.1-IL9 axis effect on macrophage polarisation in vitro (n=4–5). (K) RT-qPCR was employed to assess the expression levels of proinflammatory factors in macrophages (n=3). FLS, fibroblast-like synoviocytes; PBMC, peripheral blood mononuclear cell; RA, rheumatoid arthritis.
Effects of PU.1 on Th9 of UREΔ mice-established CAIA model
PU.1 knockdown mice (UREΔ mice) in the CAIA model had significant arthritis relief (online supplemental figure 3). Interestingly, replenishing UREΔ CAIA mice with recombinant IL-9 protein exacerbated these alleviated arthritic phenotypes (figure 4A–E). Flow cytometry results revealed increased Th9 proportion in the spleen and PBMCs (figure 4F,I), indicating PU.1 knockdown inhibited IL-9 production during arthritis development. Increased Th17 and induced Treg in the spleen and PBMCs of recombinant IL-9 protein-treated UREΔ CAIA mice were also confirmed (figure 4G,H,J,K).
Effects of PU.1 on Th9 of UREΔ mice-established CAIA model. (A) CAIA model was established using PU.1 knockdown mice (UREΔ mice). (B, C) Arthritis index and the count of joint swellings in the CAIA model established in UREΔ mice with exogenous IL -9 supplementation (n=4). (D) Images illustrating joint swelling in the UREΔ mice-established CAIA model with exogenous IL-9 supplementation. (E) H&E and Safranin O staining performed to evaluate joint pathology in the CAIA model established in UREΔ mice with exogenous IL-9 supplementation (n=3). (F–K) Flow cytometry was used to determine the proportions of Th9, Th17 and Treg cells in the spleen and PBMC of UREΔ mice in the CAIA model with exogenous IL-9 supplement (n=4). CAIA, collagen antibody-induced arthritis; PBMC, peripheral blood mononuclear cell.
IL-9 deficiency and IL-9 neutralising monoclonal antibody ameliorate arthritis progression
To elucidate the role of IL-9 in arthritis progression directly, a CAIA model was constructed using IL-9 KO mice (figure 5A, online supplemental figure 4). As expected, joint swelling in the IL-9 KO CAIA model was reduced compared with controls (figure 5B), as confirmed by quantitative analysis of the arthritis index and joint swelling number (figure 5C). H&E and Safranin O staining of the knee and ankle joints indicated a protective effect on joint injury with IL-9 loss (figure 5D). IL-9 deficiency also repressed Th17 cells while inducing Tregs in PBMC and the spleen in the CAIA model (figure 5E,F). The proarthritic effects of IL-9 indicate its potential as a therapeutic target for RA. We evaluated the efficacy of the IL-9 neutralising monoclonal antibody (mAb) in treating CIA mice (figure 5G). IL-9 blockade significantly inhibited endogenous IL-9 in CIA mice, leading to a significant reduction in arthritic phenotypes (figure 5H–J). Moreover, IL-9 blockade mitigated bone damage in the joints of CIA mice (figure 5K). Accordingly, IL-9 was repressed in IL-9 neutralising mAb-treated CIA model (figure 5L). As expected, Th17 was downregulated and Treg was upregulated in IL-9 neutralising mAb-treated CIA model (figure 5L). Overall, these study findings suggest IL-9 is a potential therapeutic target for RA.
IL-9 deficiency and IL-9 neutralising monoclonal antibody ameliorate arthritis progression. (A) CAIA model is contrasted by IL-9 knockout mice (IL-9 KO mice). (B) Joint swelling images of IL-9 KO mice-established CAIA model. (C) Arthritis index and the count of joint swellings were measured in the IL-9 KO mice-established CAIA model (n=4). (D) H&E and Safranin O staining were performed to evaluate the joint pathology in the IL-9 KO mice-established CAIA model (n=3). (E, F) Flow cytometry used to detect the proportions of Th9, Th17 and Treg cells in the spleen and PBMC of IL-9 KO mice-established CAIA model (n=4). (G) Evaluate the efficacy of IL-9 neutralising mAb treatment in CIA mice. (H) Examine the relieving effect of IL-9 blockade on joint swelling in CIA mice (n=3). (I) IL-9 blockade effect on the number of joint swellings and arthritis index in CIA mice. (J) Evaluate the effect of IL-9 blockade on joint pathology in CIA mice using H&E，Safranin O and TRAP staining (n=3). (K) Examine the effect of IL-9 blockade on joint bone injury in CIA mice using microCT (n=3). (L) Examine the effect of IL-9 blockade on Th9, Th17 and Treg subpopulations in CIA mice using flow cytometry (n=3). CAIA, collagen antibody-induced arthritis; CIA, collagen-induced arthritis; KO, knockout.
IL-9R KO alleviates CAIA arthritis progression
As expected, joint arthritis in the IL-9 receptor (IL-9R) KO CAIA model was significantly reduced compared with controls (figure 6A,D and online supplemental figure 5). Quantitative analysis of the arthritis index and joint swelling number was used to verify this result (figure 6B,C). H&E and Safranin O staining of the knee and ankle joints suggested that IL-9R deficiency had a protective effect on joint injury (figure 6E). In PBMC and spleens of IL-9R KO CAIA mice, Th17 was repressed and Tregs were induced (figure 6F,G), confirming that loss of IL-9R alleviates the IL-9-mediated proarthritic phenotype in vivo. Additionally, Western blotting of mouse spleen tissues showed reduced phosphorylated expression of JAK1 and STAT3 in the IL-9R KO group (figure 6H), suggesting that IL-9 may activate JAK1-STAT3 signalling pathway.
IL-9R KO alleviates CAIA arthritis progression. (A) CAIA model is constructed using IL-9 receptor knockout mice (IL-9R KO). (B, C) IL-9R KO effect on number of arthritis index and joint swelling in CAIA mice. (D) IL-9R KO relieving effect on joint swelling in CAIA mice (n=8). (E) Evaluate the effect of IL-9R KO on knee and ankle joint pathology of the CAIA model using H&E and Safranin O staining (n=3). (F, G) Using flow cytometry to analyse changes in Th9, Th17 and Treg in PBMC and spleen of the CAIA model constructed using IL-9R KO mice (n=8). (H) Western blot of JAK1/STAT1 in spleen from control and IL-9 KO mice-established CAIA models (n=4). CAIA, collagen antibody-induced arthritis; KO, knockout; PBMC, peripheral blood mononuclear cell.
IL-9 promotes PU.1 expression in Th9 cells via JAK1/STAT3 pathway
PU.1 expression was reduced in the spleen of the IL-9 monoclonal antibody-treated CIA, IL-9 and IL-9R KO CAIA models (figure 7A), suggesting IL-9 may promote PU.1 in RA. Literature screening31 and bioinformatics predictions led to the speculation that IL-9 promotes PU.1 expression through the JAK/STAT pathway (online supplemental figure 6A). Western blotting results confirmed that exogenous IL-9 protein stimulates PU.1 expression in T cells, with the JAK inhibitor tofacitinib inhibits PU.1 upregulation (figure 7B). Western blot screening revealed IL-9 primarily affects PU.1 expression in T cells through JAK1/STAT3 (figure 7C). Luciferase reporter assay and ChIP-qPCR further confirmed that STAT3 activates PU.1 by directly binding to its promoter region (figure 7D,E). Furthermore, the JAK inhibitor tofacitinib, STAT3 inhibitor stattic and PU.1 inhibitor DB2313 suppressed IL-9 release from Th9 cells (figure 7F). IL-9 treatment reduced TNF-α -induced apoptosis of Th9 cells, and this effect was reversed by tofacitinib and stattic administration (figure 7G, online supplemental figure 6B–D). Th9 cell supernatants enhanced RA-FLS migration, whereas tofacitinib-treated, stattic (STAT3 inhibitor)-treated and IL-9 mAb-treated Th9 cell supernatants inhibited this effect (figure 7H,I). Additionally, Th9 cell supernatant induced M1-type macrophage polarisation, reversed by tofacitinib, stattic and IL-9 mAb (figure 7J,K). These results suggest that IL-9 promotes PU.1 in Th9 cells via the JAK1/STAT3 pathway.
IL-9 promotes PU.1 expression in Th9 cells via JAK1/STAT3 pathway. (A) Western blot analysis was performed to detect PU.1 expression in the spleen of CIA mice treated with IL-9 blocking antibody (n=6), CAIA model mice with IL-9 KO (n=4) and CAIA model mice with IL-9R KO (n=4), respectively. (B) Western blot analysis was performed to detect PU.1 expression in Jurkat and EL4 cells treated with recombinant IL-9 and the JAK inhibitor tofacitinib (n=3). (C) Western blot analysis was performed to detect the IL-9 effect on the JAK1/STAT3 pathway in RA-Th9 cells (n=3). (D, E) Luciferase reporter assay and CHIP-qPCR were used to validate how STAT3 activates PU.1 transcription by directly binding to the promoter region of PU.1 (n=3). (F) Flow cytometry was used to detect the JAK inhibitor tofacitinib effect, STAT3 inhibitor statics and PU.1 inhibitor DB2313 on the release of IL-9 from Th9 cells (n=5). (G) Flow cytometry was used to detect IL-9 and statin effect on TNF-α-induced apoptosis in Th9 cells (n=4). (H) FLS is cultured with supernatant from RA-Th9. (I) Assess the effects of tofacitinib, stattics and IL-9mAb on the cell migration of RA-FLS after treatment with the supernatant from RA-Th9 cells. (J) macrophages are co-cultured with RA-Th9. (K) Effects of tofacitinib, stattics and IL-9mAb on macrophage polarisation after co-culturing with RA-Th9 (n=5). CAIA, collagen antibody-induced arthritis; CIA, collagen-induced arthritis; FLS, fibroblast-like synoviocytes.
Discussion
In the initial exploration of PU.1, attention primarily focused on its role in differentiating various immune cells, including T, B, DC and neutrophils.32–34 Subsequent studies focused on the role of PU.1 in immune system-associated cancers, including leukaemia and lymphoma.35 36 A clinical trial (NCT0409234) is currently evaluating PU.1 expression in paediatric patients with acute lymphoblastic leukaemia. Recently, given the significant role of PU.1 in various immune cells, our interest shifted to its potential role in AIDs.33 The study elucidates the relationship between PU.1 and IL-9, highlighting that PU.1 plays a role in promoting IL-9 production during arthritis.37 This relationship is further confirmed through PU.1 knockdown mice, PU.1 T cell-conditional KO mice, IL-9 KO mice, IL-9R KO mice and Th9 in vitro differentiation, providing a comprehensive understanding of the PU.1-IL-9 feedback loop and its significance in arthritis development (online supplemental figure 7).
Assessing the therapeutic potential of a PU.1 inhibitor (DB2313) and an IL-9 blockade monoclonal antibody in the mouse CIA model, the study reveals significant arthritis relief. Despite the challenges associated with targeting PU.1 owing to its broad downstream influence, the modification of PU.1 inhibitors demonstrates promise, paving the way for further exploration of these therapeutic strategies in RA treatment. Additionally, the study suggests a correlation between IL-9 expression and key clinical biomarkers in patients with RA, opening avenues for investigating IL-9 as a potential diagnostic and therapeutic target in RA.
It has been reported that Th17 in RA can promote the production of IL-9,38and IL-9 in the joint synovium of RA can also promote the differentiation of Th17 cells,39 suggesting that there is likely to be a positive feedback pathway between Th9 and Th17 in RA. In addition, the relationship between Th9 and Treg is complex, with IL-9 can inhibit the function of Treg in RA,22 but showing a mutually reinforcing relationship between Th9 and Treg in other disease states, such as tumours and asthma.21 This suggests that the relationship between Th9 and Treg may be inconsistent in different states, the crosstalk and related mechanisms between Th9 and Treg in RA will also be one of our future research directions.
Most IL-9 originates from Th9 cells, with other immune cells, including ILC2, mast cells and Th17 cells, producing other amounts.40 41 Another study validated the ameliorating effects of CIA model via IL-9 blocking.42 However, IL-9 deficiency was inversely correlated with the regression phase of adjuvant-induced arthritis model.43 This inconsistent result may be related to the cellular source of IL-9. We have verified that IL-9 expression is significantly upregulated and positively correlated with several clinical RA biomarkers in many clinical RA samples. In the future, we need to further investigate the specific role of IL-9 in different stages of RA pathogenesis.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study was approved by the Ethics Committee of the First Affiliated Hospital of Anhui Medical University (PJ2023-06-37). The participants were identified by number, not by name. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
We thank the staff at BGI for performing RNA sequencing and data analyzing. We also thank the affected patients and healthy volunteers in this study.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Handling editor Josef S Smolen
JT, WC, WH, XW and YF contributed equally.
Contributors WW, JT and CZ conceived the project. WC, WH, XWang, YF, XWu, HZ, CL, XT, XZ and HW performed the experiments. CY collected the blood samples. DH, YC, AW, YZ, ZX, HX, SY, LZ, ZL, YD and SZ revised the manuscript. WW acts as guarantor for this work.
Funding This study was supported by National Natural Science Foundation of China (82373877, 82104185, 82272512), the Research Project of Distinguished Youth of University in Anhui Province (2022AH020052), Natural Science Foundation of Outstanding Youth of Anhui Province (2308085Y51), the Natural Science Foundation of Anhui Province, Distinguishing Youth Project (2108085J40), Support Program of Outstanding Young Talent in University (gxyqZD2022024), the Project of Improvement of Scientific Ability of Anhui Medical University (2020xkjT009), Natural Science Foundation of Anhui Province for young scholars (1908085QH379) and The Open Fund of Key Laboratory of Anti-inflammatory and Immune Medicine, Ministry of Education, P.R. China (Anhui Medical University, KFJJ-2020-01).
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.