Single Cell Transcriptome and Surface Epitope Analysis of Ankylosing Spondylitis Facilitates Disease Classification...

Area of Science:

• Immunology
• Genomics
• Systems Biology

Background:

• Ankylosing spondylitis (AS) is an immune-mediated inflammatory disease affecting the axial skeleton, causing chronic pain and spinal ankylosis.
• The precise pathogenesis of AS requires further elucidation, hindering effective treatment development.

Purpose of the Study:

• To investigate molecular features of peripheral blood mononuclear cells (PBMCs) in AS patients using single-cell CITE-seq.
• To identify novel biomarkers and potential therapeutic targets for AS.
• To develop machine learning models for AS classification based on cellular profiles.

Main Methods:

• Single-cell CITE-seq analysis of PBMCs from AS patients and healthy controls.
• Quantification of RNA and surface protein expression across various immune cell subsets.
• Development and validation of machine learning models for AS classification.

Main Results:

• CD52 overexpression was observed in multiple cell types in AS patients.
• Specific monocyte and T cell subsets showed altered expression of cytotoxicity-related genes.
• An overrepresented NK cell subset with elevated CD16, CD161, and CD38 expression was identified in AS.
• Tregs exhibited underexpression of CD39, indicating reduced functionality.
• Machine learning models achieved high accuracy (>0.95 AUROC) in classifying AS.

Conclusions:

• CITE-seq revealed distinct molecular signatures in immune cells associated with AS pathogenesis.
• Identified genes and cell subsets offer potential therapeutic targets and diagnostic biomarkers for AS.
• Machine learning models demonstrate the utility of CITE-seq data for accurate AS classification.