STAT1 transcriptionally activates CXCL13 to promote cell inflammation and proliferation and inhibit cell apoptosis in rheumatoid arthritis fibroblast-like synoviocytes

Abstract

Background

Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by synovial inflammation and joint destruction. C-X-C motif chemokine ligand 13 (CXCL13) is a chemokine involved in B-cell recruitment and lymphoid neogenesis, but its regulatory mechanisms in RA remain unclear.

Methods

Fibroblast-like synoviocytes (FLS) were isolated from synovial tissues of RA patients and traumatic knee disease volunteers. Differentially expressed genes between healthy and RA synovial tissues were analyzed using GEO datasets (GSE55235, GSE12021, and GSE89408). Quantitative real-time PCR was used to measure mRNA levels of CXCL13, signal transducer and activator of transcription 1 (STAT1), interleukin (IL-6), interleukin-1β (IL-1β), and interleukin-18 (IL-18), while Western blotting was used to detect CXCL13, STAT1, interleukin-8 (IL-8), and cyclooxygenase-2 (COX-2) protein expression. Enzyme-linked immunosorbent assays were used to quantify IL-6, IL-1β, and IL-18 levels. Cell viability, proliferation, and apoptosis were assessed via cell counting kit-8, 5-Ethynyl-2’-deoxyuridine, flow cytometry, and TUNEL assays. Chromatin immunoprecipitation (ChIP) and dual-luciferase reporter assays were used to verify the STAT1-CXCL13 interaction.

Results

CXCL13 and STAT1 were significantly upregulated in RA-FLS. Silencing CXCL13 suppressed pro-inflammatory cytokines (IL-6, IL-1β, IL-18) and proteins (IL-8, COX-2), inhibited cell viability and proliferation, and induced apoptosis. STAT1 knockdown reduced CXCL13 expression with inhibiting phosphorylated STAT1 expression. STAT1 acted as a transcriptional activator of CXCL13 in RA-FLS. Overexpressing CXCL13 reversed the anti-inflammatory, anti-proliferative, and pro-apoptotic effects of STAT1 knockdown in RA-FLS.

Conclusion

STAT1 transcriptionally activated CXCL13 to enhance inflammation, promote proliferation, and suppress apoptosis in RA-FLS. These findings highlight the STAT1/CXCL13 axis as a potential therapeutic target for modulating synovial hyperplasia and inflammation in RA.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13018-026-06699-2.

Introduction

Rheumatoid arthritis (RA) is a chronic systemic autoimmune disease. Its core pathological features involve persistent inflammatory responses and aberrant hyperplasia within the joint synovium, leading to the progressive destruction of articular bone and cartilage [1]. If left uncontrolled, this cascade of pathological changes ultimately results in joint deformities and severe functional impairment, significantly diminishing patients’ quality of life. Notably, significant advancements in therapeutic strategies in recent years have markedly improved the overall clinical outcomes for RA patients [2, 3]. However, a significant clinical challenge persists: achieving and maintaining complete disease remission remains difficult for a considerable proportion of patients [4]. This reality underscores the inherent complexity and marked individual heterogeneity of RA. Consequently, in-depth analysis of RA pathogenesis has emerged as a critical objective in current rheumatology research.

Chemokines are a family of structurally similar small-molecule cytokines that play a central regulatory role in mediating the directional migration and tissue homing of immune cells [5]. Among them, CXCL13, a key member of this family, was initially named B-cell-attracting chemokine 1 (BCA-1) or B-lymphocyte chemoattractant (BLC) due to its potent ability to recruit B lymphocytes [6]. Emerging evidence indicates that the function of CXCL13 extends far beyond its initial characterization. It specifically bound to its receptor CXCR5 to regulate various pathological processes such as inflammatory responses and tumor progression [7, 8]. Within the field of rheumatology, CXCL13 demonstrates significant clinical value. This factor serves as an independent predictive biomarker for long-term radiographic joint damage in patients with early RA [9]. In addition, CXCL13 levels were significantly elevated in RA patients compared to healthy controls and may be associated with disease progression in RA [10]. Nevertheless, its specific role and underlying mechanisms in RA development remain to be elucidated.

Signal transducer and activator of transcription 1 (STAT1) is a key member of the STAT transcription factor family, playing a central role in cellular signal transduction in response to various cytokines and growth factors. The canonical activation pathway of STAT1 begins with the binding of a cytokine or growth factor to its specific cell surface receptor. This binding triggers the activation of downstream kinases. Specifically, Janus tyrosine kinases (JAKs) phosphorylate conserved tyrosine residues on STAT proteins, while mitogen-activated protein kinases (MAPKs) phosphorylate conserved serine residues. Following this phosphorylation, STAT proteins dimerize. The activated STAT dimers then translocate into the nucleus, where they recognize and bind to specific promoter regions of target genes, thereby initiating and enhancing their transcription [11, 12]. STAT1 mediates cellular responses to interferons (IFNs), serving as the primary signaling molecule for their biological effects [13]. Importantly, research has established a close association between STAT1 activity and RA progression [14]. For instance, aberrant activation of the STAT1 signaling pathway in murine arthritis models significantly increased joint rubor and swelling, enhanced local inflammation, abnormal synovial hyperplasia, and accelerated articular cartilage damage [15].

Given the well-established roles of both STAT1 and CXCL13 in regulating inflammation within the RA synovium, a potential mechanistic link between them warrants investigation. STAT1, as a master transcriptional regulator of inflammatory genes, is a prime candidate for controlling the expression of key chemokines like CXCL13. However, a direct transcriptional relationship between STAT1 and CXCL13 in the context of RA has not yet been established. Through the prediction of the JASPAR database (https://jaspar.elixir.no/analysis), the study revealed that STAT1 may be a potential transcriptional activator of CXCL13. Therefore, this study hypothesizes that STAT1 transcriptionally activates CXCL13, thereby regulating the progression of RA. Fibroblast-like synoviocytes (FLS) play a pivotal role in RA pathogenesis, serving as the primary cellular component driving synovial inflammation in RA patients [16]. In the inflamed joint, FLS exhibit an aggressively activated phenotype, such as increased proliferation and the production of various inflammatory mediators, including chemokines [17, 18]. To validate this hypothesis, this study utilized FLSs derived from RA patients. The goal is to elucidate the mechanism by which CXCL13 regulates inflammation in FLSs during RA progression, potentially offering novel therapeutic targets for RA treatment.

Materials and methods

Fibroblast-like synoviocyte isolation and culture

Synovial tissue specimens from RA patients (N = 3) were obtained during total knee arthroplasty procedures performed at Shandong Provincial Hospital Affiliated to Shandong First Medical University. Control samples lacking RA pathology originated from individuals (N = 3) undergoing traumatic limb amputation under emergency conditions. All healthy control donors were confirmed free of both rheumatoid arthritis and osteoarthritis diagnoses. Clinical characteristics of RA patients are shown in Table S1. The diagnosis of RA was made in accordance with the 2010 ACR/EULAR classification criteria for RA [19]. Inclusion Criteria: (1) Diagnosed with RA; (2) Aged between 18 and 60 years; (3) All patients provided written informed consent. Exclusion Criteria: (1) Patients who did not meet the aforementioned diagnostic criteria; (2) Patients with severe concomitant diseases of the heart, liver, kidney, or other major organs; (3) Patients aged under 18 or over 70 years; (4) Women who are pregnant or currently breastfeeding; (5) Patients who used biological agents during the study period; (6) Patients with comorbid psychiatric disorders who are unable to cooperate with the treatment. All participants were required to read and sign written informed consent forms. The institutional ethics review board at Shandong Provincial Hospital Affiliated to Shandong First Medical University formally authorized all experimental protocols (NO.2021-251). Isolation and cultivation of FLS employed an explant adhesion technique. Synovial specimens underwent five sequential washes employing phosphate-buffered saline solution containing 2% penicillin-streptomycin (HyClone, Logan, UT, USA). Following meticulous removal of extraneous connective tissues, the purified synovium was minced into approximately one cubic centimeter fragments. These tissue fragments were distributed into 25 cm² culture flasks for surface attachment and maintained without medium at 37 °C within a 5% CO₂ humidified incubator for four hours. Subsequently, 5 mL of complete growth medium (JSY-CC5545, Procell, Wuhan, China) enriched with 20% fetal bovine serum (Procell) and 1% penicillin-streptomycin—was introduced into each flask. FLS were typically harvested following an approximate three-week incubation period. Upon achieving 90% monolayer confluence, cellular subculturing was conducted at a 1:2 splitting ratio, with experimental investigations utilizing cells between passages three to six.

FLS characterization

Cultured FLS populations underwent microscopic evaluation to document characteristic morphological features. Post-fixation, cells were immersed in PBS buffer and subjected to overnight incubation at 4 °C using vimentin primary antibody (1:200, Polyclonal, PA5-27231, Thermo Fisher, Waltham, MA, USA). Fluorescent labeling employed goat anti-rabbit IgG secondary antibody (1:100 dilution, Polyclonal, Thermo Fisher), incubated at 37 °C for sixty minutes, followed by nuclear counterstaining via dropwise application of DAPI solution (Beyotime, Shanghai, China). The cells were incubated with FITC anti-CD90 (Monoclonal, Abcam, United Kingdom) for 30 min at 4 °C in the dark. After being washed three times, the cells were resuspended in buffer and analyzed by flow cytometry. A minimum of 10,000 total events was acquired for each sample to ensure statistical robustness.

KEGG and GO analysis

The differently expressed genes between synovial tissues from healthy joints and RA joints were analyzed through the GEO datasets (GSE55235, GSE12021 and GSE89408). FDR correction in the gene selection is listed in supplementary excel sheet. The overlapping genes were identified using an online tool (https://www.majorbio.com/tools) and then subjected to KEGG and GO analysis through the Sanger box bioinformatics tool (http://sangerbox.com/tool.html).

Cell transfection

CXCL13-specific small interfering RNA (si-CXCL13), STAT1-targeted siRNA (si-STAT1), CXCL13 overexpression plasmid (oe-CXCL13), alongside their corresponding negative control constructs (si-NC and oe-NC), were commercially sourced from GenePharma (Shanghai, China). Transfection procedures were performed using Lipofectamine 3000 reagent (Invitrogen, Carlsbad, CA, USA) according to established protocols.

Quantitative real-time PCR

Total RNA extraction was performed using TRIzol reagent (Ambion Inc., Austin, TX, USA). Complementary DNA synthesis was performed using HiScript II Q RT SuperMix (Vazyme Inc., Nanjing, China). Quantitative assessment of mRNA expression levels was performed using SYBR Green Fast qPCR Mix (ABclonal Inc., Wuhan, China) with primer sequences (supplied by Tsingke Inc., Beijing, China) detailed in Table 1. Relative quantification of target gene expression, normalized against the GAPDH housekeeping gene, was computed by applying the comparative 2^−ΔΔCt method. The assay was performed with three independent biological replicates.

Table 1.

Primer sequences used for PCR

Name	Primers for PCR (5′-3′)
GAPDH	Forward	GAAATGAATGGGCAGCCGTT
	Reverse	ATCACCCGGAGGAGAAATCG
CXCL13	Forward	GTCCAAGGTGTTCTGGAGGT
	Reverse	TGAGGGTCCACACACACAAT
IL-6	Forward	TGAACTCCTTCTCCACAAGCG
	Reverse	GGGCGGCTACATCTTTGGAA
IL-18	Forward	TGGCTGCTGAACCAGTAGAAG
	Reverse	GAGGCCGATTTCCTTGGTCA
IL-1β	Forward	CCAGCTACGAATCTCCGACC
	Reverse	TGGACCAGACATCACCAAGC
STAT1	Forward	TGTGAAGTTGAGAGATGTGAATGA
	Reverse	TTGGAGATCACCACAACGGG

Western blotting

Total cellular protein extracts from FLS lysates were prepared using RIPA lysis buffer (Solarbio, Beijing, China), with protein concentrations subsequently determined via BCA assay kit (Solarbio). Equivalent aliquots containing 40 µg of total protein underwent electrophoretic separation on 10% SDS-polyacrylamide gels. Proteins were then transferred onto nitrocellulose membranes (Membrane Solutions, Shanghai, China). Following a ninety-minute blocking step employing 5% non-fat milk dissolved in TBST, membranes were incubated overnight at 4 °C with distinct primary antibodies targeting CXCL13 (PA5-28827, 1:5000, Thermo Fisher), STAT1 (1:500, AHO0832, Thermo Fisher), phosphorylated STAT1 (700349, 1:1000, Thermo Fisher), interleukin-8 (IL-8, 1:1000, PA5-79113, Thermo Fisher), cyclooxygenase-2 (COX-2, 1:1000, MA5-14568, Thermo Fisher), and GAPDH (1:50000, MA5-35235, Thermo Fisher). Subsequent incubation for one hour utilized secondary antibodies (31430 and 31460, Thermo Fisher). Following five additional TBST washing cycles, target protein bands were visualized employing an enhanced chemiluminescence detection substrate (Beyotime). Protein expression was quantified using imageJ software. The assay was performed with three independent biological replicates.

Enzyme-linked immunosorbent assays (ELISAs)

Culture supernatants were collected and subjected to analysis. These measurements were conducted in strict compliance with the manufacturer-provided protocols accompanying the specific ELISA kits (PI325 for IL-6, PI305 for IL-1β, PI558 for IL-18, PI640 for IL-8, KOA0432 for CXCL13, Beyotime). Optical density determinations for all sample wells were quantified utilizing an ELISA plate reader set to 450 nm, with readings finalized within ten minutes following reaction cessation. The assay was performed with three independent biological replicates.

Cell counting kit-8 (CCK-8)

Cells were plated within 96-well microplates at a density of 6 × 10³ cells per well and subsequently cultured under standard conditions (37 °C, 5% CO₂) for a 24-hour duration. Following 48-hour transfection periods, CCK-8 reagent (Beyotime) was introduced into every well, followed by an additional two-hour incubation at 37 °C. Absorbance values for each sample were then recorded at a wavelength of 450 nm employing a microplate spectrophotometer. The assay was performed with three independent biological replicates.

5-Ethynyl-2′-deoxyuridine assay

Cellular proliferation rates were evaluated employing a Cell-Light EdU DNA Cell Proliferation Kit (Uelandy, Suzhou, China). Subsequent to incubation with 50 µM EdU solution for two hours, FLS cultures underwent fixation using 4% paraformaldehyde and were stained utilizing Apollo Dye solution. Nuclei were visualized through counterstaining with DAPI. Proliferation-positive cells were subsequently imaged and quantified by fluorescence microscopy. The assay was performed with three independent biological replicates.

Flow cytometry

Post-transfection, approximately 1 × 10⁵ FLS cells derived from each experimental cohort were rinsed with PBS. Harvested cells were then resuspended in binding buffer and incubated under light-protected conditions with Annexin V-FITC (10 µL) and propidium iodide (PI; 10 µL) (BD Biosciences, San Jose, CA, USA). Apoptotic cell populations were ultimately assessed via flow cytometry using Flow J software. At least 10,000 total events were collected per sample. The assay was performed with three independent biological replicates.

TUNEL assay

Cells were seeded within 24-well plates (1 × 10⁵ cells per well) and fixed using 4% paraformaldehyde for fifteen minutes. TUNEL reaction mixture (Beyotime) was applied to each well and incubated at 37 °C for sixty minutes. Following PBS washes, cellular nuclei were stained with DAPI for five minutes at 37 °C. Specimens were coverslipped with Fluoromount-G™ mounting medium (Thermo Fisher). Fluorescence microscopy was employed to capture images from three randomly selected visual fields per sample, enabling calculation of the proportion of TUNEL-positive cells. The assay was performed with three independent biological replicates.

Chromatin immunoprecipitation (ChIP)

ChIP analyses were conducted employing the Simple ChIP Enzymatic Chromatin IP Kit (Cell Signaling Technology, Shanghai, China), strictly adhering to the supplier’s guidelines. Cultured cells underwent fixation via cross-linking using 1% formaldehyde diluted within their growth medium, subsequently harvested and suspended in specialized lysis buffer. The lysate fractions underwent overnight incubation with either non-specific rabbit IgG or a specific anti-STAT1 antibody (AHO0832, Thermo Fisher). Immunoprecipitated DNA fragments were ultimately amplified via quantitative real-time PCR. Cell lysates without any treatment were used as Input. The assay was performed with three independent biological replicates.

Dual-luciferase reporter assay

Bioinformatic database JASPAR (https://jaspar.elixir.no/analysis) screening was conducted to identify putative STAT1 binding motifs located within the CXCL13 gene promoter. DNA sequences corresponding to the wild-type CXCL13 promoter or versions containing mutated binding sites were cloned directionally into the pGL3-basic promoter vector (GenePharma), generating reporter constructs designated WT-CXCL13 and MUT-CXCL13. FLS were subsequently co-transfected for a 48-hour duration with these luciferase reporter plasmids alongside either si-STAT1 or si-NC (non-targeting control siRNA) utilizing Lipofectamine™ 3000 reagent (Invitrogen). Luminescence measurements were ultimately performed employing a dual-luciferase reporter assay system (Promega, Shanghai, China). The assay was performed with three independent biological replicates.

Statistical analysis

Statistical analyses were performed employing GraphPad Prism software. Continuous variable results are presented as arithmetic means ± standard deviations (SD). Inter-group comparisons involving two independent samples utilized the unpaired Student’s t-test. Comparisons encompassing three experimental groups employed one-way analysis of variance (ANOVA) with Tukey’s test. P-value less than 0.05 defined the threshold for statistical significance.

Results

CXCL13 expression was upregulated in RA-FLS

The differently expressed genes between synovial tissues from healthy joints and RA joints were analyzed through the GEO datasets (GSE55235, GSE12021 and GSE89408). The overlapping genes were identified using an online tool (https://www.majorbio.com/tools), and the results are shown in Fig. 1A. The 58 genes were subjected to KEGG pathway analysis. The results showed the pathway most relevant to RA is the cytokine-cytokine receptor interaction pathway (Fig. 1B). Moreover, subsequent GO pathway analysis showed that these overlapping genes were associated with the cytokine-cytokine receptor interaction pathway-related regulation of immune response, cell-cell signaling, and signaling receptor activity (Fig. 1C). The genes in significantly enriched KEGG pathways (P < 0.05) are displayed in Fig. 1D. The researcher found that the cytokine-cytokine receptor interaction pathway-associated genes, such as CXCL10, CXCL9, CCL5, TNFRSF17, CXCL13 and CCL18 were included. CXCL13 was selected for the following study due to its limited study in RA. The study then identified RA-FLS. Under a microscope, the researchers observed that RA-FLS exhibited a spindle-shaped morphology (Fig. 1E–G). The purity of the isolation cells was 95.03% marked with FITC anti-CD90 (Figure S1). Next, the study analyzed CXCL13 expression in FLS and RA-FLS. As shown in Fig. 1H and I, its expression at the mRNA and protein levels was upregulated in RA-FLS in comparison with FLS. Thus, CXCL13 might be an RA-related gene and its expression was upregulated in RA-FLS.

Fig. 1.

CXCL13 expression was upregulated in RA-FLS.A The illustration showed that the differently expressed genes between synovial tissues from healthy joints and RA joints were analyzed through the GEO datasets (GSE55235, GSE12021 and GSE89408). B The 58 overlapping genes were subjected to KEGG pathway enrichment analysis. C The 58 overlapping genes were subjected to GO pathway enrichment analysis, including BP, CC, and MF. D The genes involved in the KEGG pathway (P < 0.05) were displayed in a circle graph. E The growth status of RA-FLS was observed under a microscope. F and G Vimentin protein expression in RA-FLS was analyzed by fluorescence microscopy. (H and I) The mRNA and protein expression of CXCL13 were analyzed by qRT-PCR and Western blotting assays in FLS and RA-FLS. **P < 0.01 and ***P < 0.001

CXCL13 Silencing inhibited cell inflammation and proliferation and induced cell apoptosis in RA-FLS

The study then transfected si-CXCL13 and si-NC into RA-FLS to determine the consequential effects on cell inflammation, proliferation and apoptosis. The efficiency of CXCL13 knockdown is shown in Fig. 2A and Figure S2A. Subsequently, the results showed that CXCL13 depletion decreased levels of IL-6 by 69.62%, IL-1β by 60.87% and IL-18 by 70.80 in cell supernatant (Fig. 2B-D), decreased mRNA levels of IL-6 by 67.00%, IL-1β by 59.00% and IL-18 by 72.00% (Fig. 2E–G), and reduced protein levels of IL-8 by 74.67% and COX-2 by 68.67% (Fig. 2H). ELISAs also showed that CXCL13 knockdown inhibited IL-8 production in cell supernatant (Figure S2B). In addition, the results showed that CXCL13 silencing inhibited cell viability by 44.91% and proliferation by 74.86% (Fig. 3A–C). As shown in Figure S3, CXCL13 overexpression significantly increased the production of IL-6 and IL-8 in the cell supernatant by 5.52- and 8.52-fold, respectively. The study also revealed that CXCL13 silencing induced cell apoptosis by 4.06-fold (Fig. 3D) and 2.29-fold (Fig. 3E). Thus, these data demonstrate that CXCL13 knockdown represses cell inflammation and proliferation and induces cell apoptosis in RA-FLS.

Fig. 2.

CXCL13 silencing inhibited cell inflammation in RA-FLS. RA-FLS was transfected with si-CXCL13 and si-NC. A CXCL13 protein expression was analyzed by Western blotting assay. B–D The levels of IL-6, IL-1β and IL-18 were analyzed by ELISAs. E–G The mRNA levels of IL-6, IL-1β and IL-18 were quantified by qRT-PCR. H IL-8 and COX-2 protein expression was detected by Western blotting assay. **P < 0.01 and ***P < 0.001

Fig. 3.

CXCL13 silencing inhibited RA-FLS proliferation and induced cell apoptosis. RA-FLS was transfected with si-CXCL13 and si-NC. A Cell viability was analyzed by CCK-8 assay. B and C Cell proliferation was analyzed by EdU assay. D and E Cell apoptosis was analyzed by flow cytometry and TUNEL assay. *P < 0.05, **P < 0.01 and ***P < 0.001

STAT1 transcriptionally activated CXCL13 in RA-FLS

STAT1 is a transcriptional factor that was investigated in the study. The study showed the mRNA levels of STAT1 and phosphorylated STAT1 were upregulated in RA-FLS when compared with FLS (Fig. 4A and Figure S4). The study then transfected si-STAT1 and si-NC into RA-FLS to determine the effect on CXCL13 expression. The results showed that the protein expression of STAT1 and CXCL13 was downregulated after transfection with STAT1 siRNA (Fig. 4B). The study also showed that the mRNA levels of CXCL13 and STAT1 were reduced after transfection with STAT1 siRNA (Fig. 4C and D). Subsequently, the ChIP assay showed that the promoter region of CXCL13 was significantly enriched by the STAT1 antibody in RA-FLS (Fig. 4E). Moreover, STAT1 knockdown inhibited the luciferase activity of the wild-type CXCL13 reporter plasmid, but it did not affect the luciferase activity of the mutant CXCL13 reporter plasmid (Fig. 4F). The binding sites of STAT1 for the promoter region of CXCL13 are shown in Fig. 4G. Thus, STAT1 transcriptionally activated and upregulated CXCL13 in RA-FLS.

Fig. 4.

STAT1 transcriptionally activated CXCL13 in RA-FLS.A STAT1 mRNA expression was analyzed by qRT-PCR in RA-FLS and FLS. B The protein expression of STAT1 and CXCL13 was detected by Western blotting assay in RA-FLS transfected with si-STAT1 or si-NC. C and D The mRNA expression of CXCL13 and STAT1 was detected by qRT-PCR in RA-FLS transfected with si-STAT1 or si-NC. E The ChIP assay was performed to identify the association of STAT1 and CXCL13. F The dual-luciferase reporter assay was performed to identify the association of STAT1 and CXCL13. G The binding sites of STAT1 for the promoter region of the CXCL13 gene. TSS, transcription start site, **P < 0.01 and ***P < 0.001

CXCL13 overexpression attenuated STAT1 knockdown-induced effects on cell inflammation, proliferation and apoptosis in RA-FLS

The study further transfected STAT1 siRNA, CXCL13 overexpression plasmid, and the matched controls (si-NC and oe-NC) into RA-FLS to determine the consequential effects on cell inflammation, proliferation and apoptosis in RA-FLS. The efficiency of CXCL13 overexpression is shown in Figure S5. As shown in Fig. 5A, the transfection with si-STAT1 inhibited CXCL13 and STAT1 protein expression, whereas the co-transfection with si-STAT1 and oe-CXCL13 attenuated STAT1 knockdown-induced inhibitory effect on CXCL13 protein expression. Subsequently, the results showed that STAT1 knockdown reduced the levels of IL-6, IL-1β and IL-18 in cell supernatant and inhibited their mRNA expression in cells, accompanied by the decreased protein expression of COX-2 and IL-8; however, these effects were relieved after CXCL13 overexpression (Fig. 5B–H). In addition, the results showed that STAT1 silencing inhibited cell viability and proliferation and induced cell apoptosis, whereas these effects were rescued after CXCL13 overexpression (Fig. 6A–D). Thus, these data demonstrate that STAT1 knockdown represses cell inflammation and proliferation and induces cell apoptosis by regulating CXCL13 in RA-FLS.

Fig. 5.

STAT1 silencing inhibited cell inflammation by regulating CXCL13 in RA-FLS. RA-FLS was divided into the si-NC + oe-NC group, the si-STAT1 + oe-NC group and the si-STAT1 + oe-CXCL13 group. A CXCL13 and STAT1 protein expression were analyzed by Western blotting assay. B–D The levels of IL-6, IL-1β and IL-18 were analyzed by ELISAs. E–G The mRNA levels of IL-6, IL-1β and IL-18 were quantified by qRT-PCR. H COX-2 and IL-8 protein expression was detected by Western blotting assay. *P < 0.05, **P < 0.01 and ***P < 0.001

Fig. 6.

STAT1 silencing inhibited RA-FLS proliferation and induced cell apoptosis by regulating CXCL13. RA-FLS was divided into the si-NC + oe-NC group, the si-STAT1 + oe-NC group and the si-STAT1 + oe-CXCL13 group. A Cell viability was analyzed by CCK-8 assay. B Cell proliferation was analyzed by EdU assay. C and D Cell apoptosis was analyzed by flow cytometry and TUNEL assay. *P < 0.05, **P < 0.01 and ***P < 0.001

Discussion

The precise etiology of RA and its complex pathogenic network remain incompletely elucidated, which to some extent limits the development of more effective therapeutic strategies. This study revealed that within the RA milieu, the transcription factor STAT1 was aberrantly upregulated. Specifically, the researchers observed that the expression of phosphorylated STAT1 was significantly higher in RA-FLS compared to normal FLS. Furthermore, knockdown of STAT1 not only reduced total STAT1 levels but also inhibited the expression of phosphorylated STAT1, leading to a reduction in CXCL13 expression. This increased STAT1 expression directly transcriptionally regulated and significantly upregulated the expression of the pro-inflammatory chemokine CXCL13. The overexpression of CXCL13 not only exacerbated the inflammatory state of FLS themselves but also abnormally promoted their proliferative capacity while suppressing apoptosis.

STAT1 was significantly upregulated in RA-FLS, aligning with reported upregulation in RA joint tissues [20] and its role in FLS-mediated RA pathogenesis [15]. While prior studies implicated STAT1 in M1 macrophage polarization and chondrocyte injury [20], angiogenesis by interacting with interleukin-35 [21], and the development of FLS-mediated RA by upregulating SNX10 [15], the current findings revealed another mechanistic pathway: STAT1 acted as a transcriptional activator of CXCL13 in RA-FLS. CXCL13 upregulation in RA-FLS was functionally significant, as its silencing suppressed key pro-inflammatory cytokines (IL-6, IL-1β, IL-18) and proteins (IL-8, COX-2), inhibited cell viability and proliferation, and induced apoptosis. Critically, CXCL13 overexpression reversed the anti-inflammatory, anti-proliferative, and pro-apoptotic effects induced by STAT1 knockdown, establishing CXCL13 as the downstream effector of STAT1 in mediating these pathogenic processes in RA-FLS. This identification of the STAT1-CXCL13 regulatory axis and its comprehensive functional validation extends beyond existing knowledge of STAT1’s roles involving KLF4, SNX10, or interleukin-35 signaling in RA.

Integrating our findings with emerging mechanisms in RA pathogenesis, the STAT1/CXCL13 axis may functionally intersect with critical processes involving autophagy, extracellular vesicle-mediated communication, and post-translational modifications (PTMs). A hallmark of the RA microenvironment is the presence of multiple stressors, including hypoxia, ER stress, and pro-inflammatory cytokines like TNF-α, which are potent inducers of both autophagy and STAT1 activation. Autophagy serves as a stress response that generates autoantigens through PTMs, specifically the citrullination and carbamylation of peptides [22]. Notably, recent evidence suggests that autophagic cells exhibit elevated levels of these modified proteins, which are subsequently recognized by the immune system [23]. It is plausible that STAT1-driven inflammation within FLS could exacerbate cellular stress, thereby promoting autophagy and the subsequent generation of neoantigens that fuel the autoimmune response. Furthermore, autophagy serves as a key processing event for antigens that may be loaded onto extracellular vesicles (EVs), such as exosomes and microvesicles.

Based on literature studies, CXCL13 was consistently overexpressed in the synovium of RA patients [24]. CXCL13 interacted CXCR5 to promote endothelial progenitor cell homing and angiogenesis during RA [25]. Its pro-inflammatory mechanism involves activating the ERK/p38 MAPK pathway and suppressing miR-330-3p, thereby driving the synthesis of the key inflammatory mediator TNF-α [10]. This subsequently may trigger a cascade of downstream inflammatory factors/proteins such as IL-6, IL-1β, IL-8, and COX-2. Simultaneously, the CXCL13/CXCR5 axis promoted synovial angiogenesis by mediating the homing of endothelial progenitor cells [25], providing support for synovial hyperplasia and inflammation. TNF-α exhibits anti-apoptotic effects by activating pro-survival pathways such as NF-κB [26, 27]. Upon silencing CXCL13, pro-survival signaling is attenuated, leading to the derepression of apoptotic programs.

It is important to acknowledge that STAT1 functions within the broader context of the JAK-STAT signaling network, a complex system where multiple members often exhibit overlapping or cross-regulatory functions [28]. In this study, while we have demonstrated that STAT1 acts as a transcriptional activator of CXCL13 and mediates the inflammatory and proliferative phenotypes of RA-FLS, the observed effects may not be exclusively attributable to STAT1. Other STAT family members, such as STAT3, is also known to be activated in rheumatoid arthritis [29]. It remains unclear whether STAT3 may also contribute to the regulation of CXCL13 expression in RA-FLS, either independently or through cooperative interactions with STAT1. Future studies are warranted to dissect the specific and redundant roles of different STAT proteins in the STAT-CXCL13 axis to fully elucidate the regulatory landscape.

In addition, ectopic germinal center, which are sites of local autoantibody production, are a hallmark of severe disease and are critically dependent on CXCL13 for the recruitment of CXCR5⁺ B cells and T follicular helper cells [6]. Therefore, it is plausible to hypothesize that FLS, through STAT1-driven CXCL13 secretion, could act as primary stromal organizers within the synovium, initiating the aggregation of lymphocytes necessary for ectopic lymphoid neogenesis. This would significantly expand the role of FLS from mediators of inflammation and joint destruction to active architects of the local autoimmune response. However, the study acknowledges that our in vitro findings are insufficient to prove this complex process. The formation of a functional GC involves a multicellular interplay that is beyond the scope of our current model. This insightful question nonetheless provides a robust foundation for future research. Subsequent studies utilizing advanced co-culture systems, animal models with cell-specific gene deletions, and high-resolution spatial analysis of human synovial tissue will be essential to definitively determine whether FLS-derived CXCL13 is a critical driver of GC formation in RA.

A significant and primary limitation of this study is the complete absence of in vivo validation, as all experimental data presented are derived exclusively from in vitro models using RA-FLS. While our findings robustly demonstrate the STAT1-mediated transcriptional activation of CXCL13 and its subsequent effects on inflammation, proliferation, and apoptosis within a controlled cell culture system, the lack of corroboration in a living organism substantially diminishes the physiological relevance, translational significance, and potential clinical applicability of our conclusions. The complex pathophysiology of rheumatoid arthritis involves a dynamic interplay of various cell types within the synovial microenvironment, including immune cell infiltration, cytokine gradients, and intricate stromal-vascular crosstalk. These critical elements, which are essential for understanding disease progression in a holistic context, cannot be fully recapitulated in a two-dimensional monolayer culture. Consequently, the true pathophysiological role and therapeutic tractability of the STAT1-CXCL13 axis within an intact, arthritic joint remain unverified. The reliance on cell culture data alone raises questions about whether the observed mechanisms would operate with the same efficacy in the face of systemic regulatory factors and heterogeneous cellular interactions present in vivo. Therefore, it must be explicitly stated that the lack of in vivo data is a major constraint on the interpretation of our results. Future research is unequivocally required to address this gap. This should be considered a top priority for future studies. Incorporating well-established animal models of RA, such as the collagen-induced arthritis (CIA) mouse model, will be essential to validate the in vivo operability of the STAT1-CXCL13 axis. Such in vivo confirmation is a critical next step before the therapeutic potential of targeting this pathway can be seriously considered for clinical translation. Another limitation of our study is the observed discrepancy regarding the role of STAT1 in regulating IL-6 production in RA-FLSs, which contrasts with the findings of a high-impact study [30] that reported no such influence. A factor could account for this difference. Variations in experimental models are a likely contributor. The biological behavior of primary FLSs is highly dependent on donor-specific characteristics, such as age, disease stage, and medication history, as well as on their passage number and activation state, leading to a different regulatory outcome. Another potential limitation of this study is the relatively small sample size, as the experiments were performed using FLS isolated from only three RA patients and three healthy controls. Given the known heterogeneity among primary fibroblast-like synoviocytes derived from different individuals, this limited sample size may affect the generalizability of our findings. Although the consistent results obtained from the in vitro assays suggest a robust biological trend, future studies with a larger cohort of patients are warranted to validate the role of the STAT1/CXCL13 axis across diverse RA subtypes and to minimize the potential impact of inter-individual variability. Additionally, while the researchers have established the functional impact of CXCL13 on FLS pathogenicity, our study did not specifically distinguish whether these effects are mediated through an autocrine mechanism involving CXCR5 or via paracrine interactions. The specific contribution of the CXCL13/CXCR5 autocrine loop to the STAT1-driven phenotype remains to be fully elucidated and will be a priority for our future research.

The identification of the STAT1/CXCL13 axis provides a mechanistic rationale for novel therapeutic strategies for RA. Targeting this axis offers several tangible benefits over existing therapeutic modalities. First, CXCL13 serves as a critical chemokine for the recruitment of CXCR5 + B cells and T follicular helper cells; thus, its inhibition could disrupt the formation of ectopic lymphoid structures (ELS) within the synovium, which are known drivers of local autoantibody production and severe joint damage. Second, as an upstream regulator, blocking CXCL13 has the potential to simultaneously dampen a broad spectrum of downstream inflammatory mediators, including TNF-α, IL-6, and IL-1β, rather than targeting a single cytokine. This multitarget approach may offer broader efficacy and could be particularly beneficial for patients with refractory RA who exhibit inadequate responses to single-cytokine blockade. Collectively, therapies aimed at STAT1 or CXCL13 inhibition could simultaneously suppress synovial inflammation, mitigate hyperplasia, and restore apoptotic sensitivity in pathogenic FLS, representing a promising strategy to achieve comprehensive disease control.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

M.W. and X.W. conducted the experiments and drafted the manuscript. Y.W. collected and contributed the methodology. N.Z. analyzed the data, operated the software and edited the manuscript. L.Y. prepared figures. N.H. designed and supervised the study. All authors read and approved the final manuscript.

Funding

This study was supported by the cross-sectional project (NO.HSRF.2021-0012).

Data availability

The datasets are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The institutional ethics review board at Shandong Provincial Hospital Affiliated to Shandong First Medical University formally authorized all experimental protocols, and complied with the guidelines and principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.