Identification and prediction of biomarkers, immune infiltration features, molecular regulatory mechanisms, and potential therapeutic agents associated with cuproptosis in osteoarthritic cartilage

doi:10.21203/rs.3.rs-4842592/v1

Download PDF

Article

Identification and prediction of biomarkers, immune infiltration features, molecular regulatory mechanisms, and potential therapeutic agents associated with cuproptosis in osteoarthritic cartilage

https://doi.org/10.21203/rs.3.rs-4842592/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: Osteoarthritis (OA) is a prevalent chronic degenerative joint disease among older adults. Given the incomplete comprehension of OA mechanisms and treatment approaches, there is an urgent need for novel insights into potential mechanisms and targeted treatment methods for this disease. Cuproptosis, a newly proposed copper-dependent mechanism, plays a crucial role in the development and progression of various diseases. However, its role in the development of OA remains inconclusive. This study aimed to ascertain the functional pathways, hub genes, and immune infiltration characteristics associated with cuproptosis in OA cartilage. In addition, this study investigated the regulatory mechanisms underlying hub genes and identified prospective therapeutic agents for OA.

Methods: Gene expression profiles were retrieved from the Gene Expression Omnibus database. Employing R software, we screened for cuproptosis-related genes (CRGs) exhibiting differential expression between OA and normal cartilage, followed by sample classification based on their expression levels. Differentially expressed genes between cuproptosis subtypes were subjected to weighted gene co-expression network analysis to identify candidate hub genes associated with cuproptosis and OA. The identified genes were subsequently subjected to functional enrichment analysis. Employing the CytoHubba plugin in Cytoscape and a combination of LASSO with support vector machine methods, we further eliminated the hub genes. Afterward, hub gene validation, correlation analysis between hub genes and immune cells, hub gene co-expression analysis, and correlation analysis between hub genes and CRGs were conducted. Finally, networks of miRNA regulation, transcription factor regulation, and drug-gene interactions were constructed for the hub genes.

Results: A total of 104 candidate hub genes were identified, with functional enrichment analysis revealing their association with metabolism and inflammation. Subsequently, seven hub genes were further screened and validated, showing consistent expression in both the training and validation datasets. Immune cell infiltration analysis revealed associations between immune cells, OA, and cuproptosis. Potential regulatory mechanisms and factors underlying OA and cuproptosis were unveiled through hub gene co-expression and regulatory networks. Finally, we identified potential therapeutic drugs for OA through the DGIdb.

Conclusions: Through comprehensive bioinformatics analysis, we identified hub genes, immune infiltration characteristics, and functional pathways associated with cuproptosis in OA cartilage. Our in-depth exploration of regulatory mechanisms, interrelationships, and potential therapeutic drugs associated with the hub genes provides numerous potential research directions for investigating OA and cuproptosis.

Biological sciences/Computational biology and bioinformatics

Health sciences/Biomarkers

Osteoarthritis

cuproptosis

machine learning

biomarker

pharmacotherapy

regulatory network

immune infiltration

Osteoarthritis (OA) is a chronic joint disease often observed in older adults, marked by synovitis, cartilage degeneration, subchondral bone remodeling, and the development of osteophytes. These pathological changes result in joint pain, swelling, stiffness, and physical impairment and can contribute to a decline in the quality of life and disability [1]. The pathogenesis underlying OA is intricate, and its precise mechanisms remain incompletely understood. Currently, the predominant clinical treatment approach focuses on pain reduction, joint function protection, and disease progression retardation. In the absence of surgical intervention, there are no non-surgical modalities available for controlling or reversing OA [2]. Hence, further investigation into the mechanisms of OA and the development of effective treatment strategies are urgently needed. An increasing amount of research suggests that the progression of OA is linked to diverse cell death mechanisms, including ferroptosis, necrosis, pyroptosis, and apoptosis [3]. Targeted interventions aimed at modulating chondrocyte death represent a potential strategy for treating OA.

Copper is an indispensable trace element that serves as a crucial catalytic cofactor in numerous biochemical reactions of significance. Physiologically, copper homeostasis contributes significantly to organismal development and survival [4]. Numerous pathological conditions, including cancer and cardiovascular and neurodegenerative diseases, are associated with cuprous ion imbalance [5]. Cuproptosis is a recently proposed copper-dependent cell death pathway. Tsvetkov discovered that an excess of intracellular copper ions can directly bind to lipid components of the tricarboxylic acid (TCA) cycle, resulting in acylated protein aggregation, degradation of iron-sulfur cluster proteins, and ultimately apoptosis. Additionally, several cuproptosis-related genes (CRGs) were identified in the study [6]. While limited research indicates dysregulation of certain CRGs in OA synovium [7], there is currently no investigation elucidating the specific role of cuproptosis in OA cartilage.

Bioinformatics and microarray technology are garnering increasing attention and are widely applied in predicting the pathogenesis, diagnosis, and treatment targets of various diseases. Figure 1 illustrates the flowchart for this study, which incorporated a range of bioinformatics and machine learning techniques. Initially, dysregulated CRGs in OA cartilage were identified. Subsequently, the immune microenvironmental characteristics, biological processes (BPs), and hub genes potentially associated with cuproptosis in OA progression were elucidated via a series of analyses. Further analysis revealed potential connections between hub genes and CRGs, along with their transcription factors and miRNA regulatory networks. Additionally, potential drugs for the treatment of OA were predicted based on the hub genes. The analytical findings suggest numerous prospective avenues for future OA research.

2.1 Raw data acquisition and processing

The Gene Expression Omnibus is a global public repository created to archive and provide open access to high-throughput datasets [8]. In this study, we retrieved and utilized two datasets, namely GSE114007 and GSE169077 (Table 1), designating GSE114007 as the training dataset and GSE169077 as the validation dataset. Given that this study relied on publicly available data and did not involve animals or humans, Institutional Review Board approval was not required.

Table 1

Basic information of each dataset.
Dataset	Platform	OA sample	Healthy sample	Platform	Author	Year
GSE114007	GPL11154	20	18	GPL96	Fisch KM	2018
GSE169077	GPL96	6	5	GPL570	Attur M	2021

2.2 Processing of datasets and identification of differentially expressed CRGs

The GSE114007 dataset was normalized, and the expression levels of CRGs were obtained using the limma package in R software (version 4.21). Differentially expressed CRGs with significant P-values (P < 0.05) were evaluated and filtered utilizing the linear model and Bayesian test provided in the limma package. Subsequently, the results were visualized using the ggpubr package.

2.3 Classification of dataset samples

Using the k-means algorithm and the ConsensusClusterPlus package, we categorized each sample into unique clusters according to the differentially expressed CRGs derived from section 2.2. Subtypes (k) were configured up to nine, with parameters set as reps = 50, pItem = 0.8, and pfeature = 1. On the basis of the cumulative distribution function (CDF) curve, cluster consensus score, and consistency matrix, we determined the optimal number of subclusters.

2.4 Differentially expressed genes (DEGs) between cuproptosis subtypes

DEGs associated with cuproptosis subtypes were identified using the limma package in R. The results were visualized through a volcano plot generated using the ggplot2 package.

2.5 ssGSEA and immunocyte infiltration analysis

Using the GSVA package in the R software, we derived immunocyte scores for each sample, and determined the enrichment levels of 28 immune features per sample [9]. Subsequently, employing Wilcoxon analysis, we compared and analyzed the levels of immunocyte infiltration between the OA and normal cartilage, as well as across various cuproptosis subtypes. To visualize the results, we used the Vioplot package to generate a violin plot.

2.6 Weighted gene co-expression network analysis (WGCNA) and candidate hub gene screening

WGCNA is a bioinformatic method that constructs gene coexpression network by clustering and dynamic pruning, which facilitates the exploration of critical pathogenic mechanisms and potential therapeutic targets [10]. A WGCNA network was constructed using an unsigned topological overlap matrix (TOM) with specific parameters: 60 RNAs in the module set and a cutting height of 0.25. Modules with a threshold of P < 0.05 were identified as key modules, and candidate hub genes were screened from these key modules based on criteria such as gene significance (GS) above 0.5 and module membership (MM) exceeding 0.8.

2.7 Functional enrichment analysis of the candidate hub genes

Using the clusterProfiler package, we conducted Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses to annotate biological terms and pathways associated with the candidate hub genes [11, 12]. For each term, P-values of 0.05 were used to determine statistical significance, and enrichment analysis results were visualized utilizing GOplot package.

2.8 Hub gene screening via support vector machine (SVM) and LASSO regression analysis.

LASSO logistic regression is a widely employed method for constructing clinical predictive models. To determine optimal parameters, it employs the minimum λ value and enhances the selection of hub genes through 10-fold cross-validation [13]. SVM-RFE, a widely employed machine learning algorithm in classification and regression analyses [14], determines the hub gene range via 10-fold cross-validation. Genes identified by both algorithms were screened using the VennDiagram package and included in the hub gene list.

2.9 Hub gene screening using Cytoscape

We constructed a protein-protein interaction (PPI) network using the STRING database (https://cn.string-db.org) and imposed a minimum connection rate threshold of 0.4 for network inclusion. Subsequently, genes in the PPI network were subjected to computational analysis using the CytoHubba tool in Cytoscape. Each of the 15 genes with the highest scores was extracted from the results generated by each of the five algorithms (MNC, Degree, Closeness, Radiality, and Stress). Subsequently, these genes were overlapped. Utilizing the VennDiagram package, we visualized the results, and the identified genes were integrated into the hub gene list.

2.10 Validation of hub genes in the training and validation datasets

Utilize the limma package in R for comparing gene expression levels in osteoarthritic and healthy cartilage. Statistically significant differences were considered in the following situations: *P < 0.05, **P < 0.01, and ***P < 0.001. The results were visualized using the ggpubr package. To assess the diagnostic value of hub genes for OA, we utilized the pROC package to construct ROC curves and determine the AUC values. A hub gene was considered to exhibit robust specificity and sensitivity for distinguishing healthy individuals from patients with OA when the AUC value exceeded 0.8.

2.11 Statistical analysis of correlations between immune cells and hub genes

The correlation between hub genes (determined in Sections 2.8 and 2.9) and immune cells associated with OA and cuproptosis (determined in Section 2.5) was assessed using the limma package. Visualization of the results was conducted by utilizing the ggplot2 package.

2.12 Construction of co-expression networks for hub genes

Utilizing the R package limma, we constructed a network of co-expression for hub genes in OA cartilage, which was then visualized with the help of tidyverse package.

2.13 Correlation analysis between hub genes and CRGs

R software packages dplyr and limma were used to analyze correlations between CRGs and hub genes in OA cartilage. and the results were presented using the ggplot2, linket, and tidyverse packages.

2.14 Prediction of transcription factors associated with the hub genes

To better understand the mechanism by which hub genes are regulated, we employed the transcription factor target interaction database (https://www.grnpedia.org/trrust/) to identify transcription factor binding sites associated with these hub genes. Subsequently, a transcription factor regulatory network was constructed by means of Cytoscape.

2.15 Construction of an miRNA-mRNA regulatory network for the hub genes

TargetScan, miRDB, and miRanda databases were utilized for the prediction of miRNAs regulating hub genes. The miRNAs associated with two or more hub genes were subsequently screened as highly predictive miRNAs. Subsequently, a miRNA-mRNA regulatory network was constructed using Cytoscape.

2.16 Construction of drug-gene interaction network

DGIdb is a specialized database designed for predicting potential drugs targeting genes. It amalgamates data from various sources, including drug-gene interaction databases, pharmacology databases, and gene annotation databases, enabling users to identify potential targeted drugs for specific genes of interest [15]. Leveraging the resources offered by the DGIdb, we identified potential targeted drugs for the hub genes and constructed a drug-gene interaction network using Cytoscape.

3.1 Differential expression analysis of CRGs

Following the annotation and standardization of genetic data, expression levels of CRGs were extracted from the samples, and differential analysis was conducted on the extracted data. As shown in Fig. 2A, variations in the expression of MTF1, PDHB, DLD, LIAS, DLAT, and NFE2L2 were observed between the OA and healthy samples. Figure 2B illustrates the chromosomal positions of these DEGs.

3.2 Clustering of samples and identification of DEGs between subtypes C1 and C2

Based on the CDF curve (Fig. 3A) and the delta area (Fig. 3B), setting the k value to 2 yielded the highest consistency scores for each subtype (Fig. 3C), indicating the most stable clustering. Therefore, on the basis of the differentially expressed CRGs, we classified the samples into two categories: C1 (n = 17) and C2 (n = 21) (Fig. 3D). Principal component analysis results, as displayed in Fig. 3E, revealed significant differences between the two clusters, suggesting a close relationship between sample classifications and the expression of CRGs. Differential analysis was performed on samples from different subtypes, preliminarily identifying 9632 DEGs potentially associated with cuproptosis typing (Fig. 3F).

3.3 Immune microenvironment analysis

Figure 4 presents the results of the ssGSEA and immune cell infiltration analysis. Figure 4A displays the infiltration of immune cells in normal cartilage versus OA cartilage, while Fig. 4B compares subtypes C1 and C2. Notably, NKT, Th2, central memory CD8 T cells, and effector memory CD8 T cells exhibited differential infiltration in both comparisons, suggesting potential associations with both OA and cuproptosis.

3.4 WGCNA

Based on the expression levels of the CRGs delineated in Section 3.2, WGCNA was subsequently performed. Employing a soft threshold of 6 ensured that the gene distribution conformed to a scale-free network. The results of model fitting and average connectivity are depicted in Fig. 5A (R2 = 0.89). The clustering tree and heatmap of gene-trait correlations revealed five modules (Fig. 5B, C). The yellow and brown gene modules exhibited significant correlations with OA (Fig. 5D). A total of 104 candidate hub genes were selected from the yellow and brown modules (Figs. 5E, F) based on the screening criteria of GS greater than 0.5 and MM exceeding 0.8.

3.5 Functional enrichment analysis of the candidate hub genes

GO and KEGG enrichment analyses of the candidate hub genes revealed various BPs, including ossification, fat cell differentiation, glycogenesis, cellular response to steroid hormone stimulation, and osteoblast differentiation. Cellular components (CCs) were primarily associated with endoplasmic reticulum lumens and the collagen-containing extracellular matrix (ECM). Furthermore, molecular functions (MFs) were primarily associated with ECM structural constraints (Fig. 6A). KEGG analyses indicated that candidate hub genes were primarily involved in signaling pathways including PI3K/AKT, mitogen-activated protein kinase (MAPK), protein digestion and absorption, hypoxia-inducible factor-1 (HIF-1), and FoxO (Fig. 6B).

3.6 Hub gene screening via LASSO regression and SVM analyses

Based on the 104 hub genes identified in Section 3.2, LASSO logistic regression analysis (Fig. 7A, B) and SVM analysis (Fig. 7C, D) were conducted. The integration of results from both methods revealed two hub genes, namely DDIT3 and CISH, as the most predictive genes (Fig. 7E).

3.7 Hub gene screening using Cytoscape

The STRING database was utilized to construct a PPI network based on the candidate hub genes (Fig. 8A). Subsequently, the PPI network was analyzed and computed using the CytoHubba tool in Cytoscape. Five algorithms (MCC, MNC, Degree, Stress, and Closeness) were employed to analyze and compute the genes, yielding the top 15 genes from each algorithm's results. After intersecting these genes, six hub genes were identified: NFIL3, ARNTL, GADD45A, BCL6, RARA, and DDIT3 (Fig. 8B).

3.8 Validation of hub genes

Compared to healthy samples, the OA cartilage samples exhibited a significant decrease in hub gene expression (Fig. 9A). Hub genes were evaluated for their diagnostic value based on ROC curves. All hub genes demonstrated robust capability in distinguishing OA samples from normal ones, with each exhibiting AUC values exceeding 0.8 (Fig. 9B). We selected the GSE169077 dataset for validation, and the results were similar to those obtained from the training dataset (Fig. 9C). NFIL3, ARNTL, GADD45A, BCL6, RARA, DDIT3, and CISH emerged as crucial diagnostic biomarkers for OA (Fig. 9D).

3.9 Correlation analysis of hub genes and other molecular factors

Figure 10A illustrates the chromosomal distribution of the hub genes, while Fig. 10B demonstrates their significant correlation with immune cells associated with OA and cuproptosis in cartilage, as delineated in Section 3.3. Additionally, Fig. 10C displays the hub gene co-expression network, revealing that the expression of NFIL3 and ARNTL in OA cartilage is negatively correlated with other hub genes. Subsequently, we conducted an in-depth analysis of the association between CRGs and hub genes in OA cartilage, revealing significant correlations between them (Fig. 10D).

3.10 Prediction of hub gene regulatory networks

A network of miRNA regulatory genes for hub genes was constructed using Cytoscape software. MiRNAs associated with two or more hub genes were identified as highly predictive miRNAs (Fig. 11A). Utilizing the TRUST database, we predicted the transcription factors associated with hub genes (Fig. 11B), including STAT3, NFYA, JUND, BRCA1, and JUN. Finally, we searched the DGIdb to identify drugs interacting with the hub genes (Table 2) and constructed a drug-gene interaction network, as illustrated in Fig. 11C.

Table 2 Potential drugs for OA.

Copper ions possess the capacity to bind with diverse proteins or enzymes, serving as catalytic cofactors in numerous BPs [16]. Maintaining a balance of copper content within living organisms is crucial. Disruption in the equilibrium of copper ions may result in oxidative stress and aberrant autophagy, contributing to various diseases associated with copper or its ions [17]. Cuproptosis, identified as a novel form of cell death, relies on the accumulation of copper ions. It is characterized by the disruption of the TCA cycle, triggered by an excess of copper ions in the cellular environment, ultimately resulting in cellular apoptosis [6]. Despite the intricate association of cuproptosis with various inflammatory diseases, there remains a paucity of research investigating the mechanism of cuproptosis in OA cartilage [18]. Employing bioinformatics and machine learning techniques, our study delved into the biological mechanisms underlying cuproptosis in OA cartilage. We explored alterations in the immune microenvironment associated with cuproptosis and OA, identifying seven associated hub genes. Subsequently, we predicted transcription factors and miRNAs regulating the hub genes and identified potential drugs for the treatment of OA.

Six CRGs (NFE2L2, LIAS, DLD, DLAT, PDHB, and MTF1) exhibited differential expression between the OA and healthy samples, suggesting their potential involvement in the development of OA. NFE2L2 encodes transcription factors crucial in intricate regulatory networks governing metabolism, inflammation, autophagy, and mitochondrial physiology to prevent cell apoptosis [19]. Notably, NFE2L2 activation inhibits the TCA cycle by directly activating pyruvate dehydrogenase kinase 1, potentially constituting a mechanism to avert apoptosis [20]. MTF1 stimulates the expression of genes associated with metal homeostasis, such as metallothionein, thereby amplifying catabolic signaling in OA and ultimately contributing to disease progression [21, 22]. LIAS, DLD, DLAT, and PDHB localize to mitochondria and contribute to the formation of pyruvate dehydrogenase (PDH), which participates in the TCA cycle [23, 24]. In OA cartilage, the dysregulation of PDH triggers ECM degradation, fostering chondrocyte apoptosis [25]. Thus, we posit that mitochondrial PDH dysregulation is intricately linked to cuproptosis in OA cartilage.

Based on the CRGs identified previously, the samples were classified into two subtypes relevant to cuproptosis. Through immunocyte infiltration analysis, we discerned four types of immune cells associated with both OA and cuproptosis, namely NKT cells, Th2 cells, factor memory CD8 T cells, and central memory CD8 T cells. NKT cells commonly participate in viral infection, cellular transformation, and the elimination of abnormal cells. They exert influence on the immune response mediated by T and B cells through the rapid release of cytokines and growth factors and are often observed in inflamed synovial tissue [26]. Th2 cells modulate the activities of dendritic cells, B cells, and eosinophils by secreting cytokines, including IL-4, IL-5, IL-10, and IL-13. They are key contributors to the pathogenesis of allergies and are essential for the host's defense against multicellular parasites [27]. Despite the presence of the anti-inflammatory factors IL-4 and IL-10, secreted by Th2 cells, in the peripheral blood and synovial fluid of patients with OA, further comprehensive explorations are required to elucidate their precise roles [28]. Central memory CD8 + T cells and effector memory CD8 + T cells both constitute subsets of memory CD8 + T cells. Upon activation, this heterogeneous cell group differentiates into distinct memory subgroups, which are crucial for immune memory, facilitating rapid responses upon subsequent encounters with a pathogen [29]. Memory-like CD8 + T cell populations are evident in certain OA samples, yet their functions remain incompletely understood [30].

Through subtyping differential analysis and WGCNA, a total of 104 candidate hub genes associated with OA and cuproptosis in cartilage were identified. KEGG enrichment analysis highlighted the predominant involvement of these candidate hub genes in pathways such as protein digestion and absorption, PI3K/AKT, MAPK, HIF-1, and FoxO. The PI3K/AKT signaling pathway is crucial for maintaining cartilage homeostasis, with its activation intricately involved in the synthesis and metabolism of the ECM. The PI3K/AKT pathway has been observed to be downregulated in cartilage tissues affected by OA [31]. Research has demonstrated that exposure to copper inhibits the PI3K/AKT signaling pathway within cells [32], suggesting a potential mechanism through which cuproptosis participates in OA. The MAPK pathway plays a critical role in OA progression. Certain pro-inflammatory mediators trigger the release of MMPs and ADAMTS from chondrocytes via the MAPK pathway. These cascade reactions promote the degradation of ECM proteins, ultimately resulting in chondrocyte apoptosis [33]. MAPK inhibition is frequently employed as a strategy for treating OA [34]. Although there is evidence linking the MAPK pathway with certain CRGs, the specific processes involved have not been fully elucidated [35]. HIF-1 is a survival factor for hypoxic chondrocytes, enabling their survival in a vascular hypoxic environment. It is essential for maintaining aerobic glycolysis in chondrocytes and for ECM synthesis. Elevated expression of HIF-1 can enhance the expression of ECM genes, promote cell viability, and protect chondrocytes from apoptosis [36]. Furthermore, the HIF-1 signaling pathway is accountable for inducing glutathione expression in cells, establishing various endogenous defense mechanisms against copper-induced cytotoxicity [37]. FoxO plays a key role in preventing cell and organ senescence. In OA chondrocytes, FoxO inhibits the production of inflammatory mediators and degradative enzymes while concurrently promoting the expression of protective genes. Notably, its expression diminishes as OA progresses [38]. Maintaining or restoring FoxO expression holds promise for averting the onset of OA and impeding disease progression. Studies have shown frequent phosphorylation of FoxO1a in cells exposed to cupric ions, resulting in its inactivation and expulsion from the nucleus [39]. This phenomenon may contribute to the decline of FoxO expression in OA cartilage. GO enrichment analysis revealed that the candidate hub genes were predominantly enriched in the ECM and associated with BPs such as gliosis and ossification. This observation corroborates prior research [40], suggesting a close association between the cuproptosis process and alterations in the ECM of chondrocytes during OA.

Further analysis revealed seven hub genes associated with both OA and cuproptosis: NFIL3, ARNTL, GADD45A, BCL6, RARA, DDIT3, and CISH, all of which exhibited low expression levels in the OA samples. NFIL3, a critical immune regulatory factor predominantly expressed in various immune cells [41], is associated with severe joint inflammation and arthritis [42]. Although NFIL3 exhibits a correlation with OA, the underlying mechanism remains unclear. ARNTL serves as a central component in the biological clock loop, directing the intricate circadian expression of genes regulated by the biological clock [43]. Patients with OA often exhibit lower levels of ARNTL expression in chondrocytes, possibly due to altered microenvironments and HIF pathways [44]. GADD45 is a member of a protein family characterized by a molecular weight of 18 kDa. Members of this protein family are instrumental in maintaining chondrocyte homeostasis by modulating apoptosis and chondrocyte differentiation through the regulation of stress-responsive mitogenic protein kinase cascades [45]. Nevertheless, the molecular mechanism governing its expression in chondrocytes remains unknown. BCL6 is a transcription factor that recruits co-inhibitory factors in the germinal centers of B and T cells, serving as a key control point in adaptive immunity [46]. While considered a biomarker associated with immune infiltration in OA progression, the exact mechanism remains unclear [47]. Through its interaction with retinoic acid (RA), the retinoic acid receptor (RARA) initiates downstream signaling pathways governing cell differentiation, proliferation, and tissue and organ formation [48]. While some evidence suggests RARA's involvement in the OA process, its specific role at different stages remains unclear [49]. DDIT3 is a crucial transcription factor and pro-apoptotic molecule induced during endoplasmic reticulum stress. It mediates apoptosis by inhibiting the expression of the pro-survival protein Bcl-2 [50]. Although overexpression of DDIT3 promotes chondrocyte apoptosis [51], its pathogenic mechanism in OA has not been fully elucidated, potentially providing novel avenues for future research on OA. CISH is a family of endogenous inhibitory factors crucial in JAK/STAT signaling regulation. Human OA chondrocytes demonstrate lower expression levels of CISH compared to chondrocytes from non-OA cartilage samples, which may significantly contribute to OA progression and changes in chondrocyte ECM structure [52].

This study revealed decreased expression levels of all hub genes in OA compared to healthy cartilage samples. Despite this, the hub gene co-expression network and gene-immunocyte correlation analyses showed that the expression of NFIL3 and ARNTL was negatively correlated with the expression of other hub genes in OA cartilage. This observation suggests potential divergent roles for certain hub genes across different stages of OA. Furthermore, we explored the association between hub genes and CRGs in OA cartilage. Each hub gene was found to be linked with multiple CRGs, providing insights into the process of cuproptosis in OA for future research.

To predict and identify additional therapeutic targets for OA, we constructed regulatory networks involving transcription factors and miRNAs associated with the identified hub genes. Notably enriched in the transcription factor regulatory network were STAT3, NFYA, JUND, BRCA1, and JUN. The involvement of these transcription factors in OA has been established in several studies [53–57]. However, the role of pathways formed by interactions between transcription factors and hub genes in OA cartilage remains poorly understood. Research into these pathways may provide novel avenues for understanding OA progression in the future. Regarding the miRNA regulatory network, several miRNAs have demonstrated predictive and therapeutic potential for OA [58]. Additionally, it is plausible that other unreported miRNAs in the regulatory network could serve as potential targets for OA therapy. Furthermore, the drug-gene interaction network revealed interactions between specific drugs and certain hub genes. Colchicine, dexamethasone, curcumin, and celecoxib, which have been employed in clinical practice, exhibit certain alleviative or therapeutic effects on OA [59–62]. In vitro studies have indicated that genistein, tretinoin, suramin, bortezomib, and sirolimus all demonstrate anti-apoptotic and protective effects on cartilage, indicating their potential therapeutic capabilities for OA [63–67]. Notably, tretinoin, acting as an RA agonist, can inhibit inflammation of articular cartilage both in vitro and in vivo [64]. Similarly, tamibarotene and alitretinoin, recognized as RA agonists [68], appear to hold substantial potential in the treatment of OA.

This study exhibits certain limitations. The relatively small sample size employed may introduce some bias into the results. In addition, while we identified correlations in this study, causal relationships remain unelucidated. Additional research is needed to ascertain the precise mechanisms at play.

We elucidated the correlation between cuproptosis and OA progression. Employing bioinformatics analysis, we identified seven hub genes linked to both cuproptosis and OA in cartilage. Our findings underscore the significant research potential of these hub genes as potential biomarkers for OA. Additionally, we predicted transcription factors, miRNAs, and drugs associated with the hub genes, thereby providing numerous avenues for future research.

Ethics approval and consent to participate

Since the data in this study are all derived from public databases and do not involve human or animal experiments, no ethical approval and consent to participate were required.

Consent for publication

all authors agreed to publish the paper at JOURNAL OF INFLAMMATION RESEARCH.

Availability of data and materials

All data and materials for this article are derived from the geo database and are genuinely available.The detailed website is as follows:

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE114007,

https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE169077,

Competing interests

The authors confirm that there are no conflicts of interest.

Funding

There is no financial support for this article.

Authors' contributions

XL organized the research results and wrote the manuscript. ZD devised the concept and performed the statistical analysis. RG collected the original data. AW supervised the study. HG and FZ revised this manuscript. All authors contributed to the article and approved the submitted version.

Acknowledgements

We thank each author for his contributions.

Katz JN, Arant KR, Loeser RF. Diagnosis and Treatment of Hip and Knee Osteoarthritis: A Review. JAMA. 2021 Feb 9;325(6):568-578. doi: 10.1001/jama.2020.22171.
Abramoff B, Caldera FE. Osteoarthritis: Pathology, Diagnosis, and Treatment Options. Med Clin North Am. 2020 Mar;104(2):293-311. doi: 10.1016/j.mcna.2019.10.007.
Yang J, Hu S, Bian Y, Yao J, Wang D, Liu X, Guo Z, Zhang S, Peng L. Targeting Cell Death: Pyroptosis, Ferroptosis, Apoptosis and Necroptosis in Osteoarthritis. Front Cell Dev Biol. 2022 Jan 18;9:789948. doi: 10.3389/fcell.2021.789948.
Baker ZN, Cobine PA, Leary SC. The mitochondrion: a central architect of copper homeostasis. Metallomics. 2017 Nov 15;9(11):1501-1512. doi: 10.1039/c7mt00221a.
Chen L, Min J, Wang F. Copper homeostasis and cuproptosis in health and disease. Signal Transduct Target Ther. 2022 Nov 23;7(1):378. doi: 10.1038/s41392-022-01229-y.
Tsvetkov P, Coy S, Petrova B, Dreishpoon M, Verma A, Abdusamad M, Rossen J, Joesch-Cohen L, Humeidi R, Spangler RD, Eaton JK, Frenkel E, Kocak M, Corsello SM, Lutsenko S, Kanarek N, Santagata S, Golub TR. Copper induces cell death by targeting lipoylated TCA cycle proteins. Science. 2022 Mar 18;375(6586):1254-1261. doi: 10.1126/science.abf0529. Epub 2022 Mar 17. Erratum in: Science. 2022 Apr 22;376(6591):eabq4855.
Wang W, Chen Z, Hua Y. Bioinformatics Prediction and Experimental Validation Identify a Novel Cuproptosis-Related Gene Signature in Human Synovial Inflammation during Osteoarthritis Progression. Biomolecules. 2023 Jan 7;13(1):127. doi: 10.3390/biom13010127.
Barrett T, Suzek TO, Troup DB, Wilhite SE, Ngau WC, Ledoux P, Rudnev D, Lash AE, Fujibuchi W, Edgar R. NCBI GEO: mining millions of expression profiles--database and tools. Nucleic Acids Res. 2005 Jan 1;33(Database issue):D562-6. doi: 10.1093/nar/gki022.
Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics. 2013 Jan 16;14:7. doi: 10.1186/1471-2105-14-7.
Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008 Dec 29;9:559. doi: 10.1186/1471-2105-9-559.
The Gene Ontology Consortium. The Gene Ontology Resource: 20 years and still GOing strong. Nucleic Acids Res. 2019 Jan 8;47(D1):D330-D338. doi: 10.1093/nar/gky1055.
Du J, Yuan Z, Ma Z, Song J, Xie X, Chen Y. KEGG-PATH: Kyoto encyclopedia of genes and genomes-based pathway analysis using a path analysis model. Mol Biosyst. 2014 Jul 29;10(9):2441-7. doi: 10.1039/c4mb00287c.
Zhou ZR, Wang WW, Li Y, Jin KR, Wang XY, Wang ZW, Chen YS, Wang SJ, Hu J, Zhang HN, Huang P, Zhao GZ, Chen XX, Li B, Zhang TS. In-depth mining of clinical data: the construction of clinical prediction model with R. Ann Transl Med. 2019 Dec;7(23):796. doi: 10.21037/atm.2019.08.63.
Huang S, Cai N, Pacheco PP, Narrandes S, Wang Y, Xu W. Applications of Support Vector Machine (SVM) Learning in Cancer Genomics. Cancer Genomics Proteomics. 2018 Jan-Feb;15(1):41-51. doi: 10.21873/cgp.20063.
Cotto KC, Wagner AH, Feng YY, Kiwala S, Coffman AC, Spies G, Wollam A, Spies NC, Griffith OL, Griffith M. DGIdb 3.0: a redesign and expansion of the drug-gene interaction database. Nucleic Acids Res. 2018 Jan 4;46(D1):D1068-D1073. doi: 10.1093/nar/gkx1143.
Guo H, Wang Y, Cui H, Ouyang Y, Yang T, Liu C, Liu X, Zhu Y, Deng H. Copper Induces Spleen Damage Through Modulation of Oxidative Stress, Apoptosis, DNA Damage, and Inflammation. Biol Trace Elem Res. 2022 Feb;200(2):669-677. doi: 10.1007/s12011-021-02672-8.
Jian Z, Guo H, Liu H, Cui H, Fang J, Zuo Z, Deng J, Li Y, Wang X, Zhao L. Oxidative stress, apoptosis and inflammatory responses involved in copper-induced pulmonary toxicity in mice. Aging (Albany NY). 2020 Jul 4;12(17):16867-16886. doi: 10.18632/aging.103585.
Wang M, Zheng L, Ma S, Lin R, Li J, Yang S. Cuproptosis: emerging biomarkers and potential therapeutics in cancers. Front Oncol. 2023 Nov 7;13:1288504. doi: 10.3389/fonc.2023.1288504.
Che J, Yang X, Jin Z, Xu C. Nrf2: A promising therapeutic target in bone-related diseases. Biomed Pharmacother. 2023 Dec;168:115748. doi: 10.1016/j.biopha.2023.115748.
Chang CW, Chen YS, Tsay YG, Han CL, Chen YJ, Yang CC, Hung KF, Lin CH, Huang TY, Kao SY, Lee TC, Lo JF. ROS-independent ER stress-mediated NRF2 activation promotes warburg effect to maintain stemness-associated properties of cancer-initiating cells. Cell Death Dis. 2018 Feb 7;9(2):194. doi: 10.1038/s41419-017-0250-x.
Lee M, Won Y, Shin Y, Kim JH, Chun JS. Reciprocal activation of hypoxia-inducible factor (HIF)-2α and the zinc-ZIP8-MTF1 axis amplifies catabolic signaling in osteoarthritis. Osteoarthritis Cartilage. 2016 Jan;24(1):134-45. doi: 10.1016/j.joca.2015.07.016.
Won Y, Shin Y, Chun CH, Cho Y, Ha CW, Kim JH, Chun JS. Pleiotropic roles of metallothioneins as regulators of chondrocyte apoptosis and catabolic and anabolic pathways during osteoarthritis pathogenesis. Ann Rheum Dis. 2016 Nov;75(11):2045-2052. doi: 10.1136/annrheumdis-2015-208406.
Mayr JA, Feichtinger RG, Tort F, Ribes A, Sperl W. Lipoic acid biosynthesis defects. J Inherit Metab Dis. 2014 Jul;37(4):553-63. doi: 10.1007/s10545-014-9705-8.
Škerlová J, Berndtsson J, Nolte H, Ott M, Stenmark P. Structure of the native pyruvate dehydrogenase complex reveals the mechanism of substrate insertion. Nat Commun. 2021 Sep 6;12(1):5277. doi: 10.1038/s41467-021-25570-y.
Yang X, Jiang Q, Luan T, Yu C, Liu Z, Wang T, Wan J, Huang J, Li K. Pyruvate Dehydrogenase Kinase 1 inhibition mediated oxidative phosphorylation enhancement in cartilage promotes osteoarthritis progression. BMC Musculoskelet Disord. 2023 Jul 20;24(1):597. doi: 10.1186/s12891-023-06585-6.
Dar HY, Azam Z, Anupam R, Mondal RK, Srivastava RK. Osteoimmunology: The Nexus between bone and immune system. Front Biosci (Landmark Ed). 2018 Jan 1;23(3):464-492. doi: 10.2741/4600.
Vyas SP, Hansda AK, Goswami R. Rheumatoid arthritis: 'melting pot' of T helper subsets. Int Rev Immunol. 2019;38(5):212-231. doi: 10.1080/08830185.2019.1621865.
Li YS, Luo W, Zhu SA, Lei GH. T Cells in Osteoarthritis: Alterations and Beyond. Front Immunol. 2017 Mar 30;8:356. doi: 10.3389/fimmu.2017.00356.
Harris KM, Clements MA, Kwilasz AJ, Watkins LR. T cell transgressions: Tales of T cell form and function in diverse disease states. Int Rev Immunol. 2022;41(5):475-516. doi: 10.1080/08830185.2021.1921764. Epub 2021 Jun 21.
Ponchel F, Burska AN, Hensor EM, Raja R, Campbell M, Emery P, Conaghan PG. Changes in peripheral blood immune cell composition in osteoarthritis. Osteoarthritis Cartilage. 2015 Nov;23(11):1870-8. doi: 10.1016/j.joca.2015.06.018.
Sun K, Luo J, Guo J, Yao X, Jing X, Guo F. The PI3K/AKT/mTOR signaling pathway in osteoarthritis: a narrative review. Osteoarthritis Cartilage. 2020 Apr;28(4):400-409. doi: 10.1016/j.joca.2020.02.027.
Xu J, Li Y, Kang M, Chang C, Wei H, Zhang C, Chen Y. Multiple forms of cell death: A focus on the PI3K/AKT pathway. J Cell Physiol. 2023 Sep;238(9):2026-2038. doi: 10.1002/jcp.31087.
Yao M, Zhang C, Ni L, Ji X, Hong J, Chen Y, Wang J, Li C, Lin J, Lu T, Sheng Y, Sun M, Shi M, Zhou C, Cai X. Cepharanthine Ameliorates Chondrocytic Inflammation and Osteoarthritis via Regulating the MAPK/NF-κB-Autophagy Pathway. Front Pharmacol. 2022 Jun 21;13:854239. doi: 10.3389/fphar.2022.854239.
Chen WP, Jin GJ, Xiong Y, Hu PF, Bao JP, Wu LD. Rosmarinic acid down-regulates NO and PGE2 expression via MAPK pathway in rat chondrocytes. J Cell Mol Med. 2018 Jan;22(1):346-353. doi: 10.1111/jcmm.13322.
Li J, Zhang W, Ma X, Wei Y, Zhou F, Li J, Zhang C, Yang Z. Cuproptosis/ferroptosis-related gene signature is correlated with immune infiltration and predict the prognosis for patients with breast cancer. Front Pharmacol. 2023 Jul 13;14:1192434. doi: 10.3389/fphar.2023.1192434.
Zhang FJ, Luo W, Lei GH. Role of HIF-1α and HIF-2α in osteoarthritis. Joint Bone Spine. 2015 May;82(3):144-7. doi: 10.1016/j.jbspin.2014.10.003.
Xue Q, Kang R, Klionsky DJ, Tang D, Liu J, Chen X. Copper metabolism in cell death and autophagy. Autophagy. 2023 Aug;19(8):2175-2195. doi: 10.1080/15548627.2023.2200554.
Matsuzaki T, Alvarez-Garcia O, Mokuda S, Nagira K, Olmer M, Gamini R, Miyata K, Akasaki Y, Su AI, Asahara H, Lotz MK. FoxO transcription factors modulate autophagy and proteoglycan 4 in cartilage homeostasis and osteoarthritis. Sci Transl Med. 2018 Feb 14;10(428):eaan0746. doi: 10.1126/scitranslmed.aan0746.
Barthel A, Ostrakhovitch EA, Walter PL, Kampkötter A, Klotz LO. Stimulation of phosphoinositide 3-kinase/Akt signaling by copper and zinc ions: mechanisms and consequences. Arch Biochem Biophys. 2007 Jul 15;463(2):175-82. doi: 10.1016/j.abb.2007.04.015.
Xu C, Liu Y, Zhang Y, Gao L. The role of a cuproptosis-related prognostic signature in colon cancer tumor microenvironment and immune responses. Front Genet. 2022 Oct 12;13:928105. doi: 10.3389/fgene.2022.928105.
Yin J, Zhang J, Lu Q. The role of basic leucine zipper transcription factor E4BP4 in the immune system and immune-mediated diseases. Clin Immunol. 2017 Jul;180:5-10. doi: 10.1016/j.clim.2017.03.013.
Schlenner S, Pasciuto E, Lagou V, Burton O, Prezzemolo T, Junius S, Roca CP, Seillet C, Louis C, Dooley J, Luong K, Van Nieuwenhove E, Wicks IP, Belz G, Humblet-Baron S, Wouters C, Liston A. NFIL3 mutations alter immune homeostasis and sensitise for arthritis pathology. Ann Rheum Dis. 2019 Mar;78(3):342-349. doi: 10.1136/annrheumdis-2018-213764.
Poulsen RC, Hearn JI, Dalbeth N. The circadian clock: a central mediator of cartilage maintenance and osteoarthritis development? Rheumatology (Oxford). 2021 Jul 1;60(7):3048-3057. doi: 10.1093/rheumatology/keab197.
Saito T, Kawaguchi H. HIF-2α as a possible therapeutic target of osteoarthritis. Osteoarthritis Cartilage. 2010 Dec;18(12):1552-6. doi: 10.1016/j.joca.2010.10.006.
Shimada H, Otero M, Tsuchimochi K, Yamasaki S, Sakakima H, Matsuda F, Sakasegawa M, Setoguchi T, Xu L, Goldring MB, Tanimoto A, Komiya S, Ijiri K. CCAAT/enhancer binding protein β (C/EBPβ) regulates the transcription of growth arrest and DNA damage-inducible protein 45 β (GADD45β) in articular chondrocytes. Pathol Res Pract. 2016 Apr;212(4):302-9. doi: 10.1016/j.prp.2016.01.009.
Lai CY, Marcel N, Yaldiko AW, Delpoux A, Hedrick SM. A Bcl6 Intronic Element Regulates T Follicular Helper Cell Differentiation. J Immunol. 2022 Sep 15;209(6):1118-1127. doi: 10.4049/jimmunol.2100777.
Chen Z, Wang W, Hua Y. Identification and validation of BCL6 and VEGFA as biomarkers and ageing patterns correlating with immune infiltrates in OA progression. Sci Rep. 2023 Feb 13;13(1):2558. doi: 10.1038/s41598-023-28000-9.
Zhang W, Peng Q, Zhang X, Guo J, Tong H, Li S. Vitamin A Promotes the Repair of Mice Skeletal Muscle Injury through RARα. Nutrients. 2023 Aug 22;15(17):3674. doi: 10.3390/nu15173674.
Davies MR, Ribeiro LR, Downey-Jones M, Needham MR, Oakley C, Wardale J. Ligands for retinoic acid receptors are elevated in osteoarthritis and may contribute to pathologic processes in the osteoarthritic joint. Arthritis Rheum. 2009 Jun;60(6):1722-32. doi: 10.1002/art.24550.
McCullough KD, Martindale JL, Klotz LO, Aw TY, Holbrook NJ. Gadd153 sensitizes cells to endoplasmic reticulum stress by down-regulating Bcl2 and perturbing the cellular redox state. Mol Cell Biol. 2001 Feb;21(4):1249-59. doi: 10.1128/MCB.21.4.1249-1259.2001.
Uehara Y, Hirose J, Yamabe S, Okamoto N, Okada T, Oyadomari S, Mizuta H. Endoplasmic reticulum stress-induced apoptosis contributes to articular cartilage degeneration via C/EBP homologous protein. Osteoarthritis Cartilage. 2014 Jul;22(7):1007-17. doi: 10.1016/j.joca.2014.04.025.
Malemud CJ. Negative Regulators of JAK/STAT Signaling in Rheumatoid Arthritis and Osteoarthritis. Int J Mol Sci. 2017 Feb 24;18(3):484. doi: 10.3390/ijms18030484.
Huang W, Tang H, Liu Y, Li W, Shimu AS, Li B, Zhu C. ROR1/STAT3 positive feedback loop facilitates cartilage degeneration in Osteoarthritis through activation of NF-κB signaling pathway. Int Immunopharmacol. 2023 Aug;121:110433. doi: 10.1016/j.intimp.2023.110433.
Li G, Han N, Li Z, Lu Q. Identification of transcription regulatory relationships in rheumatoid arthritis and osteoarthritis. Clin Rheumatol. 2013 May;32(5):609-15. doi: 10.1007/s10067-012-2143-9.
Chen K, Fang H, Xu N. LncRNA LOXL1-AS1 is transcriptionally activated by JUND and contributes to osteoarthritis progression via targeting the miR-423-5p/KDM5C axis. Life Sci. 2020 Oct 1;258:118095. doi: 10.1016/j.lfs.2020.118095.
Pellicelli M, Picard C, Wang D, Lavigne P, Moreau A. E2F1 and TFDP1 Regulate PITX1 Expression in Normal and Osteoarthritic Articular Chondrocytes. PLoS One. 2016 Nov 1;11(11):e0165951. doi: 10.1371/journal.pone.0165951. Erratum in: PLoS One. 2016 Nov 23;11(11):e0167530.
Xie T, Ren X, Zhuang H, Jiang F, Zhang Y, Zhou P. Down-regulation of Jun induces senescence through destabilizing chromatin in osteoarthritis chondrocytes. Am J Transl Res. 2023 Jul 15;15(7):4873-4886.
Zhang QY, Zhou H, Zhou XX, Yu FB, Liu YY, Chen ZY, Ma YQ, Li XL, Tian B. Small non-coding RNAome changes during human chondrocyte senescence as potential epigenetic targets in age-related osteoarthritis. Genomics. 2023 Mar;115(2):110574. doi: 10.1016/j.ygeno.2023.110574.
Singh A, Molina-Garcia P, Hussain S, Paul A, Das SK, Leung YY, Hill CL, Danda D, Samuels J, Antony B. Efficacy and safety of colchicine for the treatment of osteoarthritis: a systematic review and meta-analysis of intervention trials. Clin Rheumatol. 2023 Mar;42(3):889-902. doi: 10.1007/s10067-022-06402-w.
Ummarino A, Pensado-López A, Migliore R, Alcaide-Ruggiero L, Calà N, Caputo M, Gambaro FM, Anfray C, Ronzoni FL, Kon E, Allavena P, Torres Andón F. An in vitro model for osteoarthritis using long-cultured inflammatory human macrophages repeatedly stimulated with TLR agonists. Eur J Immunol. 2023 Dec;53(12):e2350507. doi: 10.1002/eji.202350507.
Zhao J, Liang G, Zhou G, Hong K, Yang W, Liu J, Zeng L. Efficacy and safety of curcumin therapy for knee osteoarthritis: A Bayesian network meta-analysis. J Ethnopharmacol. 2024 Mar 1;321:117493. doi: 10.1016/j.jep.2023.117493.
Zeng M, Wu Z, Liang J, Gong A. Efficacy and safety of Sodium hyaluronate combined with Celecoxib for Knee osteoarthritis: A systematic review and meta-analysis. Asian J Surg. 2023 Nov 25:S1015-9584(23)01855-9. doi: 10.1016/j.asjsur.2023.11.077.
Zou Y, Liu Q, Guo P, Huang Y, Ye Z, Hu J. Anti‑chondrocyte apoptosis effect of genistein in treating inflammation‑induced osteoarthritis. Mol Med Rep. 2020 Sep;22(3):2032-2042. doi: 10.3892/mmr.2020.11254.
Zhu L, Kamalathevan P, Koneva LA, Zarebska JM, Chanalaris A, Ismail H, Wiberg A, Ng M, Muhammad H, Walsby-Tickle J, McCullagh JSO, Watt FE; Oxford Hand Surgical Team; Sansom SN, Furniss D, Gardiner MD, Vincent TL, Riley N, Spiteri M, McNab I, Little C, Cogswell L, Critchley P, Giele H, Shirley R. Variants in ALDH1A2 reveal an anti-inflammatory role for retinoic acid and a new class of disease-modifying drugs in osteoarthritis. Sci Transl Med. 2022 Dec 21;14(676):eabm4054. doi: 10.1126/scitranslmed.abm4054.
Shen PC, Huang SH, Liu ZM, Lu CC, Chou SH, Tien YC. Suramin ameliorates osteoarthritis by acting on the Nrf2/HO-1 and NF-κB signaling pathways in chondrocytes and promoting M2 polarization in macrophages. Int Immunopharmacol. 2023 Jul;120:110295. doi: 10.1016/j.intimp.2023.110295.
Hu W, Zhang W, Li F, Guo F, Chen A. Bortezomib prevents the expression of MMP-13 and the degradation of collagen type 2 in human chondrocytes. Biochem Biophys Res Commun. 2014 Sep 26;452(3):526-30. doi: 10.1016/j.bbrc.2014.08.102.
Dhanabalan KM , Gupta VK , Agarwal R . Rapamycin-PLGA microparticles prevent senescence, sustain cartilage matrix production under stress and exhibit prolonged retention in mouse joints. Biomater Sci. 2020 Aug 7;8(15):4308-4321. doi: 10.1039/d0bm00596g.
Wu Z, Graf FE, Hirsch HH. Acitretin and Retinoic Acid Derivatives Inhibit BK Polyomavirus Replication in Primary Human Proximal Renal Tubular Epithelial and Urothelial Cells. J Virol. 2021 Jul 12;95(15):e0012721. doi: 10.1128/JVI.00127-21.

No competing interests reported.

Download PDF

Editor invited by journal
10 Aug, 2024
Submission checks completed at journal
10 Aug, 2024
First submitted to journal
01 Aug, 2024

You are reading this latest preprint version

Identification and prediction of biomarkers, immune infiltration features, molecular regulatory mechanisms, and potential therapeutic agents associated with cuproptosis in osteoarthritic cartilage

Status:

Version 1

Abstract

Figures

1 Introduction

2 Materials and methods

2.1 Raw data acquisition and processing

2.2 Processing of datasets and identification of differentially expressed CRGs

2.3 Classification of dataset samples

2.4 Differentially expressed genes (DEGs) between cuproptosis subtypes

2.5 ssGSEA and immunocyte infiltration analysis

2.6 Weighted gene co-expression network analysis (WGCNA) and candidate hub gene screening

2.7 Functional enrichment analysis of the candidate hub genes

2.8 Hub gene screening via support vector machine (SVM) and LASSO regression analysis.

2.9 Hub gene screening using Cytoscape

2.10 Validation of hub genes in the training and validation datasets

2.11 Statistical analysis of correlations between immune cells and hub genes

2.12 Construction of co-expression networks for hub genes

2.13 Correlation analysis between hub genes and CRGs

2.14 Prediction of transcription factors associated with the hub genes

2.15 Construction of an miRNA-mRNA regulatory network for the hub genes

2.16 Construction of drug-gene interaction network

3 Results

3.1 Differential expression analysis of CRGs

3.2 Clustering of samples and identification of DEGs between subtypes C1 and C2

3.3 Immune microenvironment analysis

3.4 WGCNA

3.5 Functional enrichment analysis of the candidate hub genes

3.6 Hub gene screening via LASSO regression and SVM analyses

3.7 Hub gene screening using Cytoscape

3.8 Validation of hub genes

3.9 Correlation analysis of hub genes and other molecular factors

3.10 Prediction of hub gene regulatory networks

4 Discussion

5 CONCLUSION

Declarations

References

Additional Declarations

Status:

Version 1