Introduction: Osteoporosis (OP) represents a metabolic bone disorder characterized by reduced bone density and increased fracture susceptibility, primarily caused by an imbalance between bone resorption and formation. Osteogenic differentiation plays a critical role in OP, as it promotes bone formation processes. RAD51, a crucial gene encoding a protein essential for homologous recombination repair of DNA double-strand breaks, is central to maintaining genomic stability and ensuring accurate DNA repair. Previous investigations conducted by our research team have suggested the involvement of RAD51 in the pathogenesis of OP. The objective of this study was to explore the signaling pathways associated with RAD51. Methods: MC3T3-E1 cells were induced to undergo osteogenic differentiation and exposed to a microgravity environment to simulate OP-like conditions. Furthermore, an ovariectomized (OVX) mouse model was established to mimic OP. Osteogenic differentiation was evaluated through Alizarin Red S staining for calcium deposition and alkaline phosphatase (ALP) staining to assess ALP activity. DNA damage was quantified using the comet assay, while protein expression profiles were analyzed via Western blot. Results: Experimental results showed a significant downregulation of RAD51 expression in OP models. Notably, RAD51 overexpression promoted osteoblast differentiation and mitigated DNA damage in these models. Mechanistic studies further revealed that RAD51 suppresses activation of the cGAS-STING signaling pathway, which has been shown to negatively regulate osteoblast differentiation. In OVX mice, RAD51 overexpression mitigated bone loss, promoted osteoblast differentiation, and reduced DNA damage. Conclusion: RAD51 facilitated osteogenic differentiation and attenuated DNA damage in OP by modulating the cGAS-STING signaling pathway, offering a potential novel therapeutic target for OP treatment.

ObjectivesPrimary Sjögren’s disease (SjD) is a systemic autoimmune disease whose major organ complications severely affect patient prognosis. Currently, there is a lack of efficient and accurate tools for early identification of high-risk patients.MethodsThis study retrospectively collected clinical data from 232 SjD patients, including demographic characteristics, clinical symptoms, and laboratory indicators. Three machine learning algorithms—logistic regression (LR), support vector machine (SVM), and random forest (RF)—were used to construct prediction models, and a soft voting ensemble strategy was applied to combine model outputs. Feature importance was analyzed using the SHAP framework.ResultsThe ensemble model performed best on the test set, with an AUC of 0.725, accuracy of 71%, and a negative predictive value (NPV) of 79%. SHAP analysis revealed that complement C3 and immunoglobulin G (IgG) were the most important predictors, with low C3 and high IgG levels significantly associated with complication risk.ConclusionThe ensemble model showed moderate discriminatory performance for identifying major organ complications in SjD using routinely available clinical variables. Complement C3 and IgG were identified as important predictors. The model may serve as a preliminary auxiliary risk-stratification tool, but further external validation and model optimization are required before clinical implementation.

Rheumatoid arthritis (RA) is a systemic inflammatory disease where osteoporosis and fragility fractures represent severe comorbidities. Conventional inflammatory markers like C-reactive protein (CRP) often fail to adequately stratify skeletal risk. This study aimed to investigate whether novel, composite inflammatory indices, derived from routine blood tests, are more strongly associated with osteoporosis and fracture history in patients with RA. In this cross-sectional study of 439 RA patients, we analyzed clinical data and a panel of inflammatory markers. Patients were stratified by bone mineral density (BMD). Correlation, multivariate logistic regression, and receiver operating characteristic (ROC) curve analyses were used to identify factors associated with osteoporosis and fracture risk. Osteoporosis was diagnosed in 55.4

Effective control of inflammatory cytokines is crucial for controlling ankylosing spondylitis (AS). However, due to the complexity of cytokine networks, current therapies targeting individual cytokines often fall short of achieving satisfactory outcomes. Here, we developed mesenchymal stem cell (MSC)-like nanodecoys (denoted "MSC-NDs") and evaluated their potential as a versatile anti-inflammatory therapeutic for AS. To improve membrane yield, microvesicles derived from MSCs via cytochalasin B (CB) stimulation were employed as substitutes for traditional membrane extractions. Proteomic analysis confirmed that CB-induced microvesicles retained a membrane protein profile comparable to that of conventionally isolated MSC membranes, while offering over twice the production efficiency. The resulting MSC-NDs effectively neutralized multiple proinflammatory cytokines and suppressed cytokine-induced osteogenic differentiation of MSCs in vitro. In a mouse model of AS, MSC-NDs significantly reduced systemic cytokine levels and effectively delayed pathological new bone formation. RNA sequencing of lumbar spine tissue further revealed widespread downregulation of genes involved in bone metabolism and inflammation. These findings underscore the therapeutic potential of MSC-like nanodecoys as a versatile anti-inflammatory platform for the treatment of AS and potentially other inflammatory disorders.

OBJECTIVE:Relapsing polychondritis (RP) is a rare inflammatory disease characterized by recurrent cartilaginous inflammation with unclear pathogenesis. The precise alterations in the peripheral immune landscape driving RP pathogenesis remain incompletely defined. METHODS:Single-cell RNA sequencing was performed on peripheral blood mononuclear cells from six patients with RP and five matched healthy controls. Bioinformatic analyses characterized cellular composition, transcriptomic profiles, differentially expressed genes, pathway enrichment, metabolic states, and intercellular communication. Plasma levels of alarmins and resistin were measured by enzyme-linked immunosorbent assay. RESULTS:Patients with RP exhibited a significant reduction in circulating T cells and a trend toward increased CD14+ monocytes, neutrophils, and plasma cells. Within T cells, we observed expansion of CD8+ subsets and depletion of FCER1G+ T, γδT, and mucosal-associated invariant T cells. Transcriptional programs were indicative of enhanced chemotaxis, activation, differentiation, and interferon responses, alongside cytolytic ability in cytotoxic T subsets. B cells exhibited an activated phenotype, increased plasma cell differentiation potential, and metabolic reprogramming. Myeloid cells display robust up-regulation of alarmins (S100A8/9/12) and proinflammatory pathways. Intercellular communication analysis identified CD14+ monocytes as dominant signaling hubs, exerting extensive crosstalk via annexin, major histocompatibility complex, resistin, and chemokine ligand-receptor pairs, with enrichment of the ANXA1-FPR1 axis. CONCLUSION:This study delineates a comprehensive, cell type-revolved atlas of peripheral immune dysregulation in RP, revealing coordinated alterations in T cell activation and redistribution, B cell differentiation, myeloid-driven inflammation, and prominent alarmin-related signatures. Monocyte-centric signaling networks, potentially orchestrated through the ANXA1-FPR1 pathway, may serve as key amplifiers of systemic inflammation and represent promising targets for therapeutic immune modulation in RP.

Introduction:This study investigated genetic polymorphisms and aberrant expression of anoctamin-6 (ANO6) in ankylosing spondylitis (AS) patients to clarify the role of ANO6 in AS pathogenesis. Material and methods:Genetic sequencing was performed to identify disease-associated single nucleotide polymorphisms (SNPs) within the exonic and transcriptional regulatory regions of ANO6 in 417 AS patients and 541 healthy controls (HCs). Linkage disequilibrium (LD) and haplotype analyses were conducted to assess the association of the candidate SNPs with disease susceptibility. A cohort of 2,144 AS patients was stratified based on the presence or absence of risk-associated haplotypes, and clinical phenotypes were compared between the 2 groups. Reverse transcription-polymerase chain reaction was employed to quantify messenger RNA (mRNA) expression levels in peripheral blood mononuclear cells (PBMCs) from 40 AS patients and 32 HCs. Immunohistochemistry was used to evaluate ANO6 protein expression in tendon tissues from 9 AS patients and 5 control subjects. Results:Five SNP loci (rs79662606, rs76186361, rs17095830, rs80224086, and rs75712006) exhibited significant associations with AS susceptibility, with strong LD observed between pairwise loci. Haplotype analysis revealed a higher prevalence of GTGCG, ATGTA, and ACGTA haplotypes in AS patients (p < 0.001). No significant differences in clinical phenotypes were observed between AS patients with or without risk-associated haplotypes. The mRNA expression in PBMCs was significantly lower in AS patients (0.49 ±0.21) compared to HCs (0.86 ±0.38, p < 0.001). Conversely, ANO6 protein expression was significantly elevated in tendon tissues of AS patients, with an average histoscore of 182.22 ±30.732 compared to 62.00 ±35.637 in controls (p = 0.001). Conclusions:ANO6 may represent a novel AS susceptibility gene, with GTGCG, ATGTA, and ACGTA as risk-associated haplotypes. Dysregulated ANO6 expression in PBMCs and tendon tissues suggests its potential role in AS pathogenesis.

Albumin (ALB) and globulin (GLB) are key liver function parameters. While hyperhomocysteinemia (HHcy) is a known risk factor for various diseases, population-based evidence linking these markers to HHcy remains limited. This study investigated the associations between ALB, GLB, and HHcy. A cross-sectional analysis of triennial health examination data (2020, 2021, 2023) enrolled 1,050 eligible participants. Restricted cubic splines (RCS) examined nonlinear associations, while multivariate logistic regression and subgroup analyses assessed relationships. Random forest and Boruta algorithms validated predictive power. Among 1050 participants, 764 had HHcy. The HHcy group showed significantly higher ALB but lower GLB levels. In adjusted models, each 1-g/L ALB increase was associated with a 13

OBJECTIVES:Aberrant new bone formation is a hallmark of ankylosing spondylitis (AS), yet the cellular mechanisms remain unclear. Bone marrow-derived mesenchymal stem cells (BMSCs) are crucial for skeletal homeostasis and may be pathologically reprogrammed under chronic inflammation. We aimed to characterize the cellular heterogeneity and differentiation states of BMSCs in AS, identify BMSC subpopulations associated with structural damage, and explore their molecular features and potential therapeutic targets. METHODS:Single-cell RNA sequencing was performed on BMSCs from 12 AS patients stratified into severe structural damage (SSD, n = 9) and no structural damage (NSD, n = 3) groups based on mSASSS and SPARCC-SSS assessments. After quality control, 47 628 cells were analysed by unsupervised clustering, trajectory inference, metabolic profiling, pathway enrichment and in silico drug screening to characterize pathogenic subpopulations. RESULTS:The analysis revealed eight BMSC subpopulations forming a continuum from progenitors to lineage-committed cells. A distinct senescence-related cluster (SRC) was identified as the predominant population in patients with advanced syndesmophytes. The SRC displayed a dual phenotype of cellular senescence and hyperactive metabolism, characterized by elevated glycolysis and oxidative phosphorylation. This state supported the secretion of matrix-remodeling factors (MMP2) and inflammatory cytokines (IL-6). A specific gene signature (CDR1, PLPPR2, CMBL, SRXN1 and PPP1R3C) linked oxidative stress to this aberrant differentiation trajectory. Furthermore, computational prediction highlighted JAK inhibitors (baricitinib and filgotinib) as potential therapeutics to attenuate the SRC phenotype. CONCLUSION:This study identifies a senescence-associated, hypermetabolic BMSC subpopulation that is closely associated with structural progression in AS, providing a cellular basis for pathological bone formation and suggests that JAK inhibitors may serve as effective agents to reverse this pathogenic BMSC signature.

Purpose:Previous studies have revealed alterations of the functional connectivity of the brain networks in ankylosing spondylitis (AS). Fractional amplitude of low-frequency fluctuations (fALFF) and regional homogeneity (ReHo) are both voxel-based functional metrics capable of estimating local spontaneous neural activities. This study aimed to investigate the local spontaneous neural activities in AS patients by utilizing the analytical approaches of fALFF and ReHo. Patients and Methods:A total of 78 AS patients and 59 healthy controls (HCs) underwent resting-state functional magnetic resonance imaging. fALFF and ReHo maps were generated to identify brain regions with aberrations of spontaneous neural activities in AS patients. Different frequency bands, including the standard frequency band (0.01-0.1 Hz), slow-5 (0.01-0.027 Hz), and slow-4 (0.027-0.073 Hz), were adopted in the fALFF analysis. Results:Compared with HCs, AS patients exhibited extensive alterations of fALFF and ReHo values in brain regions belonging to the default mode network (DMN), salience network (SN), frontoparietal network (FPN), sensorimotor network, visual network and the cerebellum. Clusters found in the slow-5 band only showed significantly decreased fALFF values, whereas the slow-4 was the major contributor to the elevated fALFF values in the standard band. Conclusion:The fALFF and ReHo analyses consistently revealed significantly altered local spontaneous neural activities in AS patients, especially in the DMN, SN and FPN, comprising the triple network model. The slow-4 band might be more sensitive to the elevated fALFF values in AS patients than the slow-5 band. Our findings provide further evidence that the aberrations of the triple network model serve as an important feature of AS from the perspective of local neural activities.

Lower back pain comprises the majority of the disease burden of patients with ankylosing spondylitis (AS), while the alterations of the large-scale brain networks could be implicated in the neuropathophysiology of pain. The frontoparietal network (FPN) is known as a pain modulation hub, with key nodes dorsolateral prefrontal cortex (dlPFC) and ventrolateral prefrontal cortex (vlPFC) participating in the pain modulation and reappraisal process. In this study, we adopted the analytical approaches of independent component analysis (ICA) and seed-based correlation analysis (SCA) to examine the resting-state functional connectivity (rsFC) of the large-scale brain networks, notably FPN, between 82 AS patients and 61 healthy controls (HCs). We also investigated the correlation between the rsFC and the clinical measures of AS patients. Both ICA and SCA consistently showed that the rsFC between FPN and mPFC, a key node of the default mode network (DMN), was significantly increased in AS. In addition, SCA also identified a cluster at the right posterior lobe of cerebellum which exhibited increased rsFC with the posterior cingulate cortex, and the right lateral prefrontal cortex also showed increased rsFC with the right dlPFC. Correlation analysis showed that the rsFC between mPFC and the left anterior prefrontal cortex was significantly correlated with C-reactive protein in AS. The increased FPN-DMN connectivity could contribute to the neuropathophysiology of lower back pain in AS, with potential association with faulty pain modulation and reappraisal mechanisms facilitated by the FPN.

OBJECTIVE:This study aims to analyze the immune metabolism mechanism of Ankylosing Spondylitis (AS) and explain the causal relationship between immune cells and metabolites in the progression of AS disease through Mendelian randomization (MR) analysis, providing a theoretical basis for targeted therapy. METHODS:The causal relationship between 486 metabolites and 731 immunophenotypes and AS was evaluated by the two-step MR method, and key mediating variables were screened. RESULTS:Univariate MR identified 42 immunophenotypes and 8 metabolites that were significantly associated with the risk of AS. Two-step MR showed that the proportion of CD33dim HLA DR+ CD11b- %CD33dim HLA DR+ cells decreased, the secretion level of betaine increased, and the risk of AS increased significantly. CONCLUSION:It was found that immune cells CD33dim HLA DR+ CD11b- %CD33dim HLA DR+ and its metabolite betaine have a causal relationship with the occurrence and development of AS, which is worth verifying in the clinical diagnosis and treatment of AS.

This study systematically evaluated the immunomodulatory function of PD-L1-positive mesenchymal stem cells (PD-L1(+) MSCs) using single-cell RNA sequencing (scRNA-seq) and investigated their roles in suppressing inflammation and regulating pathological bone formation in curdlan-induced SKG ankylosing spondylitis (AS) mouse models. scRNA-seq identified MSC subpopulations with high immunomodulatory capacity and key biomarker PD-L1 for subpopulation classification. In vitro co-culture experiments were conducted to evaluate the effects of MSC subpopulations on T-cell proliferation and TNF-α levels. In vivo experiments were performed in forty-eight SKG mouse models to analyze the effects of MSC subpopulations on joint inflammation scores, T-cell subset proportions, inflammatory cytokines, histopathology, and pathological bone formation. scRNA-seq revealed significant heterogeneity in MSCs under inflammatory stimulation, with the immunomodulatory subpopulation exhibiting high expression of PD-L1 and IDO. In vitro experiments demonstrated that PD-L1(+) MSCs significantly suppressed T-cell proliferation and reduced TNF-α levels. Joint redness and swelling scores showed that the PD-L1(+) MSC group exhibited the most significant improvement in arthritis, while the IL-17Ai, PD-L1(-) MSC, and MSC groups also effectively reduced inflammation, with significantly lower scores than the model control(MC) group. Histological analysis revealed severe inflammatory cell infiltration in the MC group, while the IL-17Ai, PD-L1(+) MSC, and MSC groups exhibited reduced infiltration. Immunohistochemical analysis further confirmed these findings, with PD-L1(+) MSCs exhibiting a significant reduction in TNF-α and IL-17A-positive cells (P < 0.0001 and P < 0.01, respectively).PD-L1(+) MSCs regulated immune responses by reducing Th17 cell proportions, increasing Th2 and Treg cell proportions, and significantly lowering pro-inflammatory cytokines IFN-γ, IL-17A, and TNF-α. MicroCT analysis indicated that the PD-L1(+) MSC, MSC, and IL-17Ai group effectively suppressed pathological bone formation through immunomodulation, whereas the PD-L1(-) MSC group showed weaker effects, underscoring the importance of PD-L1 in regulating bone formation. hUC-MSCs demonstrated significant therapeutic effects in the AS mouse model, particularly the PD-L1(+) MSCs, which inhibited joint inflammation and pathological new bone formation through immunomodulatory mechanisms. These findings provide valuable insights into the therapeutic mechanisms of AS treatment.

Xeligekimab is a novel immunoglobulin G4 (IgG4) monoclonal antibody targeting interleukin-17A (IL-17A). In a phase III trial in patients with plaque psoriasis, xeligekimab showed efficacy and safety consistent with other IL-17A inhibitors, supporting its potential application in the treatment of spondyloarthritis. This phase III trial aimed to investigate the efficacy and safety of xeligekimab in patients with radiographic axial spondyloarthritis (r-axSpA). This was a phase III study conducted at multiple centers in China. Eligible patients were randomly assigned (1:1:1) to receive xeligekimab 100 mg, xeligekimab 200 mg, or placebo. Randomization was stratified by medication history (biologic-experienced vs. biologic-naïve) and weight (≥ 70 kg vs. < 70 kg). The primary endpoint was the proportion of patients achieving an Assessment of SpondyloArthritis International Society 20 (ASAS20) response at week 16. A key secondary endpoint was the ASAS40 response rate at the same time point. A total of 465 patients were recruited. A significantly higher proportion of patients receiving xeligekimab 200 mg (n = 114 (74.0

AIM:Tumor necrosis factor inhibitors (TNFi) are widely used for the treatment of autoimmune diseases, with latent tuberculosis infection (LTBI) reactivation being a significant unresolved issue. The pathogenic mechanisms are not fully understood. Integrated transcriptomic analysis could provide insights into monitoring tuberculosis progression after TNFi therapy and help reduce LTBI reactivation. METHODS:We selected six transcriptomic datasets from studies related to TNFi treatment and tuberculosis. Pathway enrichment, pseudotime, and transcription factor analyses were performed to explore the underlying mechanisms. RESULTS:Our analysis revealed distinct transcriptional changes in memory B cells during tuberculosis progression and TNFi therapy. In active tuberculosis (ATB), ROR1+ memory B cells were identified in a noncanonical differentiation trajectory, characterized by downregulation of B cell-related genes (e.g., CD22, EBF1, MS4A1), reduced translational capacity, and suppression of immune response pathways, accompanied by upregulation of oxidative phosphorylation, which highlighted metabolic alterations during tuberculosis progression. A similar subtype also emerged in TNFi-treated patients, suggesting that metabolic reprogramming of memory B cells may disrupt immune balance, thereby contributing to LTBI reactivation and ATB development following TNFi therapy. CONCLUSIONS:The study integrates bioinformatics and single-cell RNA sequencing to reveal the role of memory B cells in ATB progression and TNFi treatment, offering insights into TNFi-associated TB susceptibility and potential therapeutic targets.

BACKGROUND:Non-radiographic axial spondyloarthritis (nr-axSpA) is an early stage of axial spondyloarthritis characterized by the absence of definitive radiographic changes. Interleukin-17 inhibitors (IL-17i) have shown efficacy in treating nr-axSpA, but a significant proportion of patients fail to respond. Identifying predictive biomarkers for IL-17i response is crucial for optimizing treatment strategies. METHODS:This retrospective study analyzed pre-treatment serum samples from nr-axSpA patients treated with secukinumab. Responders (R) and non-responders (NR) were defined based on changes in ASDAS-CRP scores after 12 weeks of therapy. Serum protein profiles were analyzed using liquid chromatography-tandem mass spectrometry (LC-MS/MS) with label-free quantification, followed by validation of candidate proteins using enzyme-linked immunosorbent assay (ELISA). Statistical analyses included Student's t-test, Mann-Whitney U-test, and Spearman's correlation. RESULTS:LC-MS/MS identified 47 differentially expressed proteins (DEPs) between R and NR groups, with 31 upregulated and 16 downregulated in NR. Seven proteins including heat shock protein HSP 90-alpha (HSP90α) and serotransferrin were selected for ELISA validation. ELISA confirmed significantly higher HSP90α (p = 0.027) and lower serotransferrin (p = 0.034) levels in NR compared to R. HSP90α showed a mild negative correlation with the reduction of BASDAI scores (r = -0.30, p = 0.035), while serotransferrin exhibited a mild positive correlation (r = 0.287, p = 0.048). Both proteins correlate mildly with the disease activity index. CONCLUSION:Serum levels of HSP90α and serotransferrin may serve as potential biomarkers for predicting IL-17i treatment response in nr-axSpA patients. Elevated HSP90α and reduced serotransferrin levels may reflect chronic inflammation and disease activity, particularly from a subjective symptom perspective. Further validation in prospective studies is needed to confirm their utility in guiding personalized treatment strategies.

Background Ankylosing spondylitis (AS) is a chronic autoimmune disease that primarily affects the axial joints. Immune cells play a key role in the pathogenesis of AS. This study integrated bioinformatics methods with experimental validation to explore the role of natural killer (NK) cells in AS.Methods Two microarray datasets, GSE25101 and GSE73754, were selected, and the scRNA-seq data were obtained from GSE194315 and Liu’s research. Differentially expressed genes (DEGs) and functional enrichment analysis were performed respectively. Weighted gene co-expression network analysis (WGCNA) was conducted to identify key modules of co-expressed genes and genes involved in NK cell function. The diagnostic value of the identified key genes was evaluated using ROC curves, logistic regression analysis, and a nomogram. Real-time PCR (RT-PCR) was used to quantified the expression of genes. Statistical analysis was conducted using the R software package, and a p-value of less than 0.05 was considered statistically significant.Results Pathways enrichment analysis revealed the involvement of NK cell-mediated immune pathways and regulation of the innate immune response, indicating the crucial role of innate immunity, especially NK cells, in AS pathogenesis. The construction of a co-expression network revealed that the MElightyellow module was most relevant to the NK cell-mediated immune pathway. IL2RB, CD247, PLEKHF1, EOMES, S1PR5, FGFBP2 from the MElightyellow module were identified as key genes involved in NK cell-mediated immune response and served as potential diagnostic biomarkers for AS, with moderate to high diagnostic values based on AUC values. Further analysis using scRNA-seq profiling revealed the higher expression level of IL2RB, CD247, PLEKHF1, S1PR5, FGFBP2 in NK cells compared to that in other cell types. CD247, PLEKHF1, EOMES, S1PR5, and FGFBP2 were reduced expressed in AS patients as compare to control group verified by scRNA-seq data, CD247, EOMES, FGFBP2, IL2RB and S1PR5 were reduced expressed verified by RT-PCR, and PLEKHF1, S1PR5, and FGFBP2 was upregulated after TNF-α blocker therapy.Conclusion The study revealed the potential role of NK cells and identified IL2RB, CD247, PLEKHF1, EOMES, S1PR5, and FGFBP2 as key genes associated with NK cells in the pathogenesis of AS.

INTRODUCTION:Ankylosing spondylitis (AS) is a chronic inflammatory disease affecting the axial skeleton, characterized by immune microenvironment dysregulation and elevated cytokines like TNF-α and IL-17. Mitochondrial oxidative phosphorylation (OXPHOS), crucial for immune cell function and survival, is implicated in AS pathogenesis. This study explores OXPHOS-related mechanisms in AS, identifies key genes using machine learning, and highlights potential therapeutic targets for precision medicine. MATERIALS AND METHODS:Peripheral blood mononuclear cells (PBMCs) bulk transcriptomic and single-cell RNA sequencing (scRNA-seq) data from AS patients were analyzed to investigate the role of the OXPHOS pathway in AS. Weighted gene co-expression network analysis (WGCNA) was performed to identify key gene modules associated with OXPHOS. Machine learning techniques, including support vector machine with recursive feature elimination (SVM-RFE), random forest, and least absolute shrinkage and selection operator (LASSO), were applied to identify significant AS-related genes. Real-time PCR (RT-PCR) was used to quantify gene expression, examine their patterns in specific cell subtypes, and explore their functional implications. RESULTS:Pathway enrichment analysis identified OXPHOS as a significantly enriched pathway distinguishing AS patients from healthy controls, with high normalized enrichment scores and significant group separation in principal component analysis. ScRNA-seq revealed significantly higher OXPHOS scores in AS patients, especially in dendritic cells (DCs) and monocytes, highlighting cell type-specific dysregulation. WGCNA identified two key gene modules (MEyellow and MEtan) that are closely associated with OXPHOS. Three hub genes-LAMTOR2, APBB1IP, and DGKQ-were screened using machine learning methods and validated by RT-PCR and scRNA-seq. Among them, LAMTOR2 was significantly more highly expressed in patients with AS, and functional analyses showed that it plays a role in promoting TH17 cell differentiation, which highlights its potential as a therapeutic target for ankylosing spondylitis. CONCLUSION:This multi-omics study provides valuable insights into the complex interplay between OXPHOS and AS. The identified genes, particularly LAMTOR2, serve as potential therapeutic targets, contributing to our understanding of AS mechanisms and paving the way for precision medicine in AS treatment.

Abnormal monocytes are involved in the pathogenesis of ankylosing spondylitis (AS). We investigated the association between an imbalance of monocyte subpopulations and bone destruction in AS. Compared to controls, AS patients exhibited increased CD14+CD11c+HLA-II- monocytes and decreased CD14+CD11c+HLA-II+ monocytes in peripheral blood. LPS stimulation promoted a shift from CD14+CD11c+HLA-II- monocytes toward CD14+CD11c+HLA-II+ monocytes, enhancing the former's capacity to promote CD4+ T cell proliferation. Micro RNA-seq analysis indicated that AS CD14+CD11c+HLA-II- monocytes were involved in osteoclast differentiation and overproduced soluble osteoclastogenic cytokine CSF-1. In 40 AS patients, evaluated CD14+CD11c+HLA-II- monocytes correlated positively with Spondyloarthritis Research Consortium of Canada (SPARCC) MRI inflammation score (r = 0.336, P = 0.034) and bone erosion score (r = 0.423, P = 0.007), but inversely with ankylosis score (r = -0.346, P = 0.029). Conversely, CD14+CD11c+HLA-II+ monocytes showed the opposite correlation. Our findings demonstrate that expansion of CD14+CD11c+HLA-II- monocytes contribute to bone erosion in AS, potentially mediated through CSF-1-driven osteoclast differentiation.

Primary Sjogren’s syndrome (pSS) and autoimmune thyroiditis (AIT) share overlapping genetic and immunological profiles. This retrospective study evaluates the efficacy of machine learning algorithms, with a focus on the Random Forest Classifier, to predict the presence of thyroid-specific autoantibodies (TPOAb and TgAb) in pSS patients. A total of 96 patients with pSS were included in the retrospective study. All participants underwent a complete clinical and laboratory evaluation. All participants underwent thyroid function tests, including TPOAb and TgAb, and were accordingly divided into positive and negative thyroid autoantibody groups. Four machine learning algorithms were then used to analyze the risk factors affecting patients with pSS with positive and negative for thyroid autoantibodies. The results indicated that the Random Forest Classifier algorithm (AUC = 0.755) outperformed the other three machine learning algorithms. The random forest classifier indicated Age, IgG, C4 and dry mouth were the main factors influencing the prediction of positive thyroid autoantibodies in pSS patients. It is feasible to predict AIT in pSS using machine learning algorithms. Analyzing clinical and laboratory data from 96 pSS patients, the Random Forest model demonstrated superior performance (AUC = 0.755), identifying age, IgG levels, complement component 4 (C4), and absence of dry mouth as primary predictors. This approach offers a promising tool for early identification and management of AIT in pSS patients. This retrospective study was approved and monitored by the Ethics Committee of The Third Affiliated Hospital of Sun Yat-sen University (No.II2023-254-02).

OBJECTIVES:To evaluate the potential of large language models (LLMs) in health education for patients with ankylosing spondylitis (AS)/spondyloarthritis (SpA), focusing on the accuracy of information transmission, patient acceptance and performance differences between different models. DESIGN:Cross-sectional, single-blind study. SETTING:Multiple centres in China. PARTICIPANTS:182 volunteers, including 4 rheumatologists and 178 patients with AS/SpA. PRIMARY AND SECONDARY OUTCOME MEASURES:Scientificity, precision and accessibility of the content of the answers provided by LLMs; patient acceptance of the answers. RESULTS:LLMs performed well in terms of scientificity, precision and accessibility, with ChatGPT-4o and Kimi models outperforming traditional guidelines. Most patients with AS/SpA showed a higher level of understanding and acceptance of the responses from LLMs. CONCLUSIONS:LLMs have significant potential in medical knowledge transmission and patient education, making them promising tools for future medical practice.