No abstract

Dysregulation of cysteine cathepsin protease activity is pivotal in tumorigenic transformation. However, the role of cathepsin protease in lung cancer remains unknown. Here, we analyzed GEO database and found that lung cancer presented high expression of cathepsin V (CTSV). We then performed immunohistochemistry assay in 73 paired lung cancer tissues and normal lung tissues and confirmed that CTSV is overexpressed in lung cancer and correlates with poor prognosis. The mass spectrometry experiment showed that the N-glycosylation locus of CTSV are N221 and N292, glycosylated CTSV (band 43 kDa) was particularly expressed in lung cancer samples and correlated with lymph node metastasis. Mechanistic studies showed that only glycosylated CTSV (43-kDa band) are secreted to extracellular matrix (ECM) and promoted the metastasis of lung cancer. Importantly, the Elisa detection in serum of 12 lung cancer patients and 12 healthy donors showed that the level of CTSV in serum distinguished lung cancer patients from healthy donors. Together, our findings reveal the clinical relevance of CTSV glycosylation and CTSV drives the metastasis of lung cancer, suggesting that the glycosylated CTSV in serum is a promising biomarker for lung cancer.

Currently, the federated graph neural network (GNN) has attracted a lot of attention due to its wide applications in reality without violating the privacy regulations. Among all the privacy-preserving technologies, the differential privacy (DP) is the most promising one due to its effectiveness and light computational overhead. However, the DPbased federated GNN has not been well investigated, especially in the sub-graph-level setting, such as the scenario of recommendation system. The biggest challenge is how to guarantee the privacy and solve the non independent and identically distributed (non-IID) data in federated GNN simultaneously. In this paper, we propose DP-FedRec, a DP-based federated GNN to fill the gap. Private Set Intersection (PSI) is leveraged to extend the local graph for each client, and thus solve the non-IID problem. Most importantly, DP is applied not only on the weights but also on the edges of the intersection graph from PSI to fully protect the privacy of clients. The evaluation demonstrates DP-FedRec achieves better performance with the graph extension and DP only introduces little computations overhead.

Non-small-cell lung cancer (NSCLC) is a malignancy with high overall morbidity and mortality due to a lack of reliable methods for early diagnosis and successful treatment of the condition. We identified genes that would be valuable for the diagnosis and prognosis of lung cancer. Common DEGs (DEGs) in three GEO datasets were selected for KEGG and GO enrichment analysis. A protein-protein interaction (PPI) network was constructed using the STRING database, and molecular complex detection (MCODE) identified hub genes. Gene expression profiling interactive analysis (GEPIA) and the Kaplan-Meier method analyzed hub genes expression and prognostic value. Quantitative PCR and western blotting were used to test for differences in hub gene expression in multiple cell lines. The CCK-8 assay was used to determine the IC50 of the AURKA inhibitor CCT137690 in H1993 cells. Transwell and clonogenic assays validated the function of AURKA in lung cancer, and cell cycle experiments explored its possible mechanism of action. Overall, 239 DEGs were identified from three datasets. AURKA, BIRC5, CCNB1, DLGAP5, KIF11, and KIF15 had shown great potential for lung cancer diagnosis and prognosis. <i>In vitro</i> experiments suggested that AURKA significantly influenced the proliferation and migration of lung cancer cells and activities related to the dysregulation of the cell cycle. AURKA, BIRC5, CCNB1, DLGAP5, KIF11, and KIF15 may be critical genes that influence the occurrence, development, and prognosis of NSCLC. AURKA significantly affects the proliferation and migration of lung cancer cells by disrupting the cell cycle.