Identification of cuproptosis-related subtypes, characterization of immune microenvironment infiltration, and
Jiao Nong1, Guanyu Lu2, Yue Huang2
1Teaching Department, First Affiliated Hospital of the Guangxi University of Chinese Medicine, Nanning, China.
Background:
Osteoarthritis (OA) is a prevalent chronic joint disease with an obscure underlying molecular signature. Cuproptosis plays a crucial role in various biological processes. However, the association between cuproptosis-mediated immune infifiltration and OA progression remains unexplored. Therefore, this study elucidates the pathological process and potential mechanisms underlying cuproptosis in OA by constructing a columnar line graph model and performing consensus clustering analysis.
Methods:
Gene expression profifile datasets GSE12021, GSE32317, GSE55235, and GSE55457 of OA were obtained from the comprehensive gene expression database. Cuproptosis signature genes were screened by random forest (RF) and support vector machine (SVM). A nomogram was developed based on cuproptosis signature genes. A consensus clustering was used to distinguish OA patients into different cuproptosis patterns. To quantify the cuproptosis pattern, a principal component analysis was developed to generate the cuproptosis score for each sample. Single-sample gene set enrichment analysis (ssGSEA) was used to provide the abundance of immune cells in each sample and the relationship between these significant cuproptosis signature genes and immune cells.To quantify the cuproptosis pattern, a principal component analysis technique was developed to generate the cuproptosis score for each sample. Cuproptosis-related genes were extracted and subjected to differential expression analysis to construct a disease prediction model and confifirmed by RT-qPCR.
Results:
Seven cuproptosis signature genes were screened (DBT, LIPT1, GLS, PDHB, FDX1, DLAT, and PDHA1) to predict the risk of OA disease. A column line graph model was developed based on these seven cuproptosis signature genes, which may assist patients based on decision curve analysis. A consensus clustering method was used to distinguish patients with disorder into two cuproptosis patterns (clusters A and B). To quantify the cuproptosis pattern, a principal component analysis technique was developed to generate the cuproptosis score for each sample. Furthermore, the OA characteristics of patients in cluster A were associated with the inflflammatory factors IL-1b, IL-17, IL-21, and IL-22, suggesting that the cuproptosis signature genes play a vital role in the development of OA.
Discussion:
In this study, a risk prediction model based on cuproptosis signature genes was established for the fifirst time, and accurately predicted OA risk. In addition, patients with OA were classifified into two cuproptosis molecule subtypes (clusters A and B); cluster A was highly associated with Th17 immune responses, with higher IL-1b, IL-17, and IL-21 IL-22 expression levels, while cluster B had a higher correlation with cuproptosis. Our analysis will help facilitate future research related cuproptosis-associated OA immunotherapy. However, the specifific mechanisms remain to be elucidated.
