Impairment of natural killer (NK) cell activity is an important mechanism of tumor immunoevasion. T cell immunoglobulin domain and mucin domain-3 (Tim-3) is an activation-induced inhibitory molecule, inducing effector lymphocyte exhaustion in chronic viral infection and cancers. However, its function in NK cells in human esophageal cancer remains unclear. We prospectively collected peripheral blood and tumor samples from 53 patients with esophageal cancer. Peripheral and tumor-infiltrating NK cells were analyzed for Tim-3, Annexin V, CD69, CD107a and IFN-γ expression by flow cytometry. Quantitative real-time PCR was used to test relative mRNA expression of IFN-γ, granzyme B, perforin and NKG2D in sorted Tim-3+ NK cells and Tim-3− NK cells, respectively. NK cells isolated from healthy donors were treated with recombinant TNF-α to induce Tim-3 expression. Tim-3 and TNF-α mRNA levels in tumor tissues were measured in both humans and mice. Finally, associations between NK cell frequencies with pathological parameters were investigated. We observed up-regulation of Tim-3 expression on NK cells from esophageal cancer patients, especially at the tumor site. Furthermore, tumor-infiltrating NK cells with high Tim-3 expression exhibited a phenotype with enhanced dysfunction. In vitro, Tim-3 expression on NK cells isolated from blood of healthy donors can be induced by recombinant TNF-α via NF-κB pathway. In both animal models and patients, the Tim-3 level was positively correlated with TNF-α expression in esophageal cancer tissues. Finally, higher Tim-3 level on tumor-infiltrating NK cells is correlated with tumor invasion, nodal status and poor stage in patients with esophageal cancer. Taken together, Tim-3 may play a crucial role to induce NK cell dysfunction in tumor microenvironment and could serve as a potential biomarker for prognosis of esophageal cancer.

Lung adenocarcinoma (LUAD), largely remains a primary cause of cancer-related death worldwide. The molecular mechanisms in LUAD metastasis have not been completely uncovered. In this study, we identified differentially expressed genes (DEGs), miRNAs (DEMs) and lncRNAs (DELs) underlying metastasis of LUAD from The Cancer Genome Atlas database. Intersection mRNAs were used to perform gene ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and co-expression network analysis. In addition, survival analyses of intersection mRNAs were conducted. Finally, intersection mRNAs, miRNAs and lncRNAs were subjected to construct miRNA-mRNA-lncRNA network. A total of 1015 DEGs, 54 DEMs and 22 DELs were identified in LUAD metastasis and non-metastasis samples. GO and KEGG pathway analysis had proven that the functions of intersection mRNAs were closely related with many important processes in cancer pathogenesis. Among the co-expression interactions network, 22 genes in the co-expression network were over the degree 20. These genes imply that they have connections with many other gene nodes. In addition, 14 target genes (ARHGAP11A, ASPM, HELLS, PRC1, TMPO, ARHGAP30, CD52, IL16, IRF8, P2RY13, PRKCB, PTPRC, SASH3 and TRAF3IP3) were found to be associated with survival in patients with LUAD significantly (log-rank P < 0.05). Two lncRNAs (LOC96610 and ADAM6) acting as ceRNAs were identified based on the miRNA-mRNA-lncRNA network. Taken together, the results may provide a novel perspective to develop a multiple gene diagnostic tool for LUAD prognosis, which might also provide potential biomarkers or therapeutic targets for LUAD.