Identification of sex-specific biomarkers related to programmed cell death and analysis of immune cells in ankylosing spondylitis

Abstract

AbstractAnkylosing spondylitis (AS) stands as a persistent inflammatory ailment predominantly impacting the axial skeleton, with the immune system and inflammation intricately entwined in its pathogenesis. This study endeavors to elucidate gender-specific patterns in immune cell infiltration and diverse forms of cell demise within the AS milieu. The aim is to refine the diagnosis and treatment of gender-specific AS patients, thereby advancing patient outcomes. In the pursuit of our investigation, two datasets (GSE25101 and GSE73754) pertinent to ankylosing spondylitis (AS) were meticulously collected and normalized from the GEO database. Employing the CIBERSORT algorithm, we conducted a comprehensive analysis of immune cell infiltration across distinct demographic groups and genders. Subsequently, we discerned differentially expressed genes (DEGs) associated with various cell death modalities in AS patients and their healthy counterparts. Our focus extended specifically to ferroptosis-related DEGs (FRDEGs), cuproptosis-related DEGs (CRDEGs), anoikis-related DEGs (ARDEGs), autophagy-related DEGs (AURDEGs), and pyroptosis-related DEGs (PRDEGs). Further scrutiny involved discerning disparities in these DEGs between AS patients and healthy controls, as well as disparities between male and female patients. Leveraging machine learning (ML) methodologies, we formulated disease prediction models employing cell death-related DEGs (CDRDEGs) and identified biomarkers intertwined with cell death in AS. Relative to healthy controls, a myriad of differentially expressed genes (DEGs) linked to cell death surfaced in AS patients. Among AS patients, 82 FRDEGs, 29 CRDEGs, 54 AURDEGs, 21 ARDEGs, and 74 PRDEGs were identified. In male AS patients, these numbers were 78, 33, 55, 24, and 94, respectively. Female AS patients exhibited 66, 41, 40, 17, and 82 DEGs in the corresponding categories. Additionally, 36 FRDEGs, 14 CRDEGs, 19 AURDEGs, 10 ARDEGs, and 36 PRDEGs exhibited differential expression between male and female AS patients. Employing machine learning techniques, LASSO, RF, and SVM-RFE were employed to discern key DEGs related to cell death (CDRDDEGs). The six pivotal CDRDDEGs in AS patients, healthy controls, were identified as CLIC4, BIRC2, MATK, PKN2, SLC25A5, and EDEM1. For male AS patients, the three crucial CDRDDEGs were EDEM1, MAP3K11, and TRIM21, whereas for female AS patients, COX7B, PEX2, and RHEB took precedence. Furthermore, the trio of DDX3X, CAPNS1, and TMSB4Y emerged as the key CDRDDEGs distinguishing between male and female AS patients. In the realm of immune correlation, the immune infiltration abundance in female patients mirrored that of healthy controls. Notably, key genes exhibited a positive correlation with T-cell CD4 memory activation when comparing male and female patient samples. This study engenders a more profound comprehension of the molecular underpinnings governing immune cell infiltration and cell death in ankylosing spondylitis (AS). Furthermore, the discernment of gender-specific disparities among AS patients underscores the clinical significance of these findings. By identifying DEGs associated with diverse cell death modalities, this study proffers invaluable insights into potential clinical targets for AS patients, taking cognizance of gender-specific nuances. The identification of gender-specific biological targets lays the groundwork for the development of tailored diagnostic and therapeutic strategies, heralding a pivotal step toward personalized care for AS patients.