Fig. 6: Bioinformatic analysis based on mRNA profile indicated TPP2 as the potential target of XBP-1s.

A Scatter plot of relative mRNA expression in CAL-27 treated with shXBP-1s compared with negative control. Genes in green and red are differentially expressed at significant levels (n = 3 per group). B GSEA (KEGG, REACTOME, and BIOCARTA) pathway distribution for shXBP-1s versus negative control of CAL-27. Horizontal line denotes FDR significance cutoff of 0.05. Immune-, metabolism-, post-translation- and cancer-related gene sets were demarcated by dots in indicated colors, respectively. C Volcano plot of differentially expressed genes on the basis of fold change and p value. Genes to be focused on were labeled. D Gene sets upregulated in shXBP-1s CAL-27 compared to the negative control (FDR < 0.05 and NES > 1.5). Color gradation is based on GSEA NES. Gene sets demarcated in panel (C) were specified. E Gene sets downregulated in shXBP-1s compared to negative control of CAL-27 (FDR < 0.05 and NES < −1.5). Color gradation is based on GSEA NES. Gene sets demarcated in panel (C) were specified. Gene sets containing TPP2 were highlighted. NES indicates normalized enrichment score. F Diagrammatic drawing of associated genes divided into four groups based on respective functions that are recognized in cancer biology. G Downregulated genes from the two gene sets. Color gradation is representative of log2 fold change over negative control. Relevant genes were labeled and TPP2 was labeled in red. shCtrl indicates negative control. H Distribution of XBP-1 occupancy frequencies in TPP2 promoter in three different cancer cell lines based on ChIP-seq database, respectively. The most enriched peaks are highlighted. I Motif analysis (motif-counter) showed the enriched XBP-1 motif in TPP2 promoter, the arrow indicates that the highest score binding sites consistently located in the forward strand, which was highlighted in panel (H). J ChIP-qPCR analysis of the XBP-1 genomic occupancy in the TPP2 promoter in SCC-9 and CAL-27 as indicated. Immunoprecipitated DNA was measured by qRT-PCR with primers to amplify the TPP2 promoter region, including the distal site (mean ± s.e.m; n = 4; p < 0.0001, p < 0.0001, p = 0.3961 for SCC-9 and p < 0.0001, p < 0.0001, p = 0.408 for CAL-27; **p < 0.001 by one-way ANOVA followed by Dunnett’s tests for multiple comparisons). K Luciferase assay demonstrated that knockdown of XBP-1 inhibited TPP2 promoter’s activity in SCC-9 and CAL-27. SCC-9 and CAL-27 cells with stable expression of wild-type (wt) or mutant (mut) TPP2 promoter delivered pGL4.20-Basic vectors were co-transfected with shXBP-1 or shCtrl (mean ± s.e.m; n = 5; p < 0.0001 for SCC-9 and CAL-27; **p < 0.001 by one-way ANOVA followed by Dunnett’s tests for multiple comparisons). L Luciferase reporter assay demonstrated that XBP-1 activated TPP2 promoter (mean ± s.e.m; n = 3; p < 0.0001 for SCC-9 and CAL-27; **p < 0.001 by one-way ANOVA followed by Dunnett’s tests for multiple comparisons). M qRT-PCR verified the downregulation of TPP2 from microarrays in TSCCs after XBP-1s knockdown (mean ± s.e.m; n = 4; p < 0.0001 for SCC-9 and CAL-27; **p < 0.001 by two-tailed t-test). N Immunoblotting further verified the downregulation of TPP2 in TSCCs after XBP-1s knockdown. MW molecular weight. β-actin, loading control. O Overall survival of cancer patients with different levels of TPP2 using The Cancer Genome Atlas (TCGA) database (p = 0.0196, 0.00482 and 0.0227) HNSC head and neck squamous cell carcinoma, LUSC lung squamous cell carcinoma, SARC sarcoma.