RESUMO
Circular RNA (circRNA), which is a newly discovered non-coding RNA, has been documented to play important roles in miRNA sponges, and the dysregulation of which is involved in cancer development. However, circRNA expression profiles and their role in initiation and progression of Wilms tumor (WT) remain largely unclear at present. Here, we used paired WT samples and high-throughput RNA sequencing to identify differentially expressed circRNAs (DE-circRs) and mRNAs (DE-mRs). A total of 314 DE-circRs and 1612 DE-mRs were identified. The expression of a subset of differentially expressed genes was validated by qRT-PCR. A complete circRNA-miRNA-mRNA network was then constructed based on the common miRNA targets of DE-circRs and DE-mRs identified by miRanda prediction tool. The Gene set enrichment analysis (GSEA) indicated that several signaling pathways involving targeted DE-mRs within the ceRNA network were associated with cell cycle and immune response, which implies their participation in WT development to some extent. Subsequently, these targeted DE-mRs were subjected to implement PPI analysis and to identify 10 hub genes. Four hub genes were closely related to the survival of WT patients. We then filtered prognosis-related hub genes by Cox regression and least absolute shrinkage and selection operator (LASSO) regression analysis to construct a prognosis-related risk score system based on a three-gene signature, which showed good discrimination and predictive ability for WT patient survival. Additionally, we analyzed the mutational landscape of these genes and the associations between their expression levels and those of immune checkpoint molecules and further demonstrated their potential impact on the efficacy of immunotherapy. qRT-PCR and western blotting (WB) analysis were used to validate key differentially expressed molecules at the RNA and protein levels, respectively. Besides these, we selected a key circRNA, circEYA1, for function validation. Overall, the current study presents the full-scale expression profiles of circRNAs and the circRNA-related ceRNA network in WT for the first time, deepening our understanding of the roles and downstream regulatory mechanisms of circRNAs in WT development and progression. We further constructed a useful immune-related prognostic signature, which could improve clinical outcome prediction and guide individualized treatment.
RESUMO
Background: Neuroblastoma (NB) is the most frequent solid tumor in pediatrics, which accounts for roughly 15% of cancer-related mortality in children. NB exhibited genetic, morphologic, and clinical heterogeneity, which limited the efficacy of available therapeutic approaches. Recently, a new term 'cuproptosis' has been used to denote a unique biological process triggered by the action of copper. In this instance, selectively inducing copper death is likely to successfully overcome the limitations of conventional anticancer drugs. However, there is still a gap regarding the role of cuproptosis in cancer, especially in pediatric neuroblastoma. Methods: We characterized the specific expression of cuproptosis-related genes (CRGs) in NB samples based on publicly available mRNA expression profile data. Consensus clustering and Lasso-Cox regression analysis were applied for CRGs in three independent cohorts. ESTIMATE and Xcell algorithm was utilized to visualize TME score and immune cell subpopulations' relative abundances. Tumor Immune Dysfunction and Exclusion (TIDE) score was used to predict tumor response to immune checkpoint inhibitors. To decipher the underlying mechanism, GSVA was applied to explore enriched pathways associated with cuproptosis signature and Connectivity map (CMap) analysis for drug exploration. Finally, qPCR verified the expression levels of risk-genes in NB cell lines. In addition, PDHA1 was screened and further validated by immunofluorescence in human clinical samples and loss-of-function assays. Results: We initially classified NB patients according to CRGs and identified two cuproptosis-related subtypes that were associated with prognosis and immunophenotype. After this, a cuproptosis-related prognostic model was constructed and validated by LASSO regression in three independent cohorts. This model can accurately predict prognosis, immune infiltration, and immunotherapy responses. These genes also showed differential expression in various characteristic groups of all three datasets and NB cell lines. Loss-of-function experiments indicated that PDHA1 silencing significantly suppressed the proliferation, migration, and invasion, in turn, promoted cell cycle arrest at the S phase and apoptosis of NB cells. Conclusions: Taken together, this study may shed light on new research areas for NB patients from the cuproptosis perspective.