Identification of WDR74 and TNFRSF12A as biomarkers for early osteoarthritis using machine learning and immunohistochemistry

Abstract

Background: Osteoarthritis (OA) is a chronic joint condition that causes pain, limited mobility, and reduced quality of life, posing a threat to healthy aging. Early detection is crucial for improving prognosis. Recent research has focused on the role of ubiquitination-related genes (URGs) in early OA prediction. This study aims to integrate URG expression data with machine learning (ML) to identify biomarkers that improve diagnosis and prognosis in the early stages of OA.

Methods: OA single-cell RNA sequencing datasets were collected from the GEO database. Single-cell analysis was performed to investigate the composition and relationships of chondrocytes in OA. The potential intercellular communication mechanisms were explored using the CellChat R package. URGs were retrieved from GeneCards, and ubiquitination scores were calculated using the AUCell package. Gene module analysis based on co-expression network analysis was conducted to identify core genes. Additionally, ML analysis was performed to identify core URGs and construct a diagnostic model. We employed XGBoost, a gradient-boosting ML algorithm, to identify core URGs and construct a diagnostic model. The model's performance was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve. In addition, we explored the relationship between core URGs and immune processes. The ChEA3 database was utilized to predict the transcription factors regulated by core ubiquitination-related genes. The expression of select URGs was validated using qRT-PCR and immunohistochemistry (IHC).

Results: We identified WDR74 and TNFRSF12A as pivotal ubiquitination-related genes associated with OA, exhibiting considerable differential expression. The diagnostic model constructed with URGs exhibited remarkable accuracy, with area under the curve (AUC) values consistently exceeding 0.9. The expression levels of WDR74 and TNFRSF12A were significantly higher in the IL-1β-induced group in an in vitro qPCR experiment. The IHC validation on human knee joint specimens confirmed the upregulation of WDR74 and TNFRSF12A in OA tissues, corroborating their potential as diagnostic biomarkers.

Conclusions: WDR74 and TNFRSF12A as principal biomarkers highlighted their attractiveness as therapeutic targets. The identification of core biomarkers might facilitate early intervention options, potentially modifying the illness trajectory and enhancing patient outcomes.

Keywords: diagnosis; machine learning; osteoarthritis; single-cell RNA sequencing; ubiquitination.

Figure 1

Identification of osteoarthritis (OA) cartilage cell composition and relationships through single-cell sequencing. (A) UMAP plot depicting the clustering of cells into 10 groups. (B) Bubble chart illustrating marker genes associated with each cell type. (C) Stacked bar graph showing the proportions of cell types. (D, E) CellChat communication analysis elucidating the quantity and intensity of cell-to-cell communication. (F) Density plot displaying the core cell type composition in control and OA groups. (G) Bubble chart indicating receptor-ligand pairs involved in cell communication interactions.

Figure 2

High-dimensional weighted gene co-expression network analysis (hdWGCNA) of single-cell data. (A) Ubiquitination scoring to identify ubiquitination-associated cell types. (B) Soft threshold selection plot to determine the optimal threshold. (C) Gene module clustering diagram. (D–F) Identification of specific gene modules and their correlation analysis with cell types, highlighting gene clusters most relevant to cell types.

Figure 3

Screening of differentially expressed ubiquitination-related genes. (A) GO enrichment analysis of genes in the blue module. (B) Volcano plot from differential expression analysis of OA dataset, showcasing differentially expressed genes. (C) Venn diagram identifying OA ubiquitination-related genes from bulk sequencing data. (D) Correlation analysis of differentially expressed ubiquitination-related genes. (E) GENEMANIA analysis of protein-protein interaction networks of differentially expressed ubiquitination-related genes.

Figure 4

Machine learning-based selection of core ubiquitination-related genes. (A) Box plots of differential expression for ubiquitination-related genes. (B, C) XGBoost machine learning algorithm for the selection of core ubiquitination-related genes. (D–F) ROC curves of individual core ubiquitination-related genes across three validation cohorts.

Figure 5

Diagnostic model of core ubiquitination-related genes. (A) Nomogram of the diagnostic model for core ubiquitination-related genes. (B–D) ROC curves of the diagnostic model in different OA datasets. (E–G) Decision curve analysis (DCA) of the diagnostic model across various OA datasets.

Figure 6

Immunological relevance of core ubiquitination-related genes. (A) Heatmap of the correlation between core ubiquitination-related genes and immune cells and processes. (B) Heatmap illustrating the correlation between core ubiquitination-related genes and immune checkpoint molecules. (C) Correlation matrix showing the association between core ubiquitination-related genes and chemokine and TNF families.

Figure 7

Construction of ubiquitination-related molecular subtypes. (A–C) Consensus clustering analysis to identify molecular subtypes in different OA datasets. (D–F) PCA analysis based on molecular subtypes. (G–I) Expression heatmaps of core ubiquitination-related genes across identified molecular subtypes.

Figure 8

Drug and transcription factor analysis of core ubiquitination-related genes. (A, B) Drug screening based on the DSigDB database. (C, D) Prediction of transcription factors regulating core ubiquitination-related genes using the ChEA3 database.

Figure 9

The expression level WDR74 and TNFRSF12A in ATDC5 cells with different concentrated IL-1β and human OA cartilage. (A) qRT-PCR measurement of Col2a1, Sox9, Mmp3, Mmp13, WDR74 and TNFRSF12A in ATDC5 cells treated with different concentrated IL-1β. ** p<0.01; *** p<0.001. (B, C) Immunohistochemistry assay with anti-WDR74 in undamaged cartilage tissues and OA cartilage tissues. (D, E) Immunohistochemistry assay with anti-TNFRSF12A in undamaged cartilage tissues and OA cartilage tissues. Scale bar, Left, 100μm; Right, 20μm. (F, G) Relative expression level in OA (n=55) and corresponding undamaged (n=55) cartilage tissues based on an immunohistochemistry assay and significance was evaluated by paired Student’s t test.