Fig 2: TNFSF10 (TRAIL) expression levels associated with prognosis for patients with metastatic RCCTo determine TNFSF10 (TRAIL) expression levels, clinical samples from patients with metastatic RCCs were analyzed by qRT-PCR with GAPDH normalization (left panel). The black line indicates the average value of TNFSF10 expression. The RCC patients were categorized in high or low TNFSF10 expression groups based on the average value. Kaplan-Meier survival curves for TNFSF10 in patients with metastatic RCC (right panel). The high TNFSF10 group was associated with significantly shorter survival periods compared with the low TNFSF10 group after chemotherapy (log-rank test, p = 0.015).

Fig 3: Comparison of TNFSF10 expression between EA and AA in breast and other tumors. TNFSF10 expression in the pan-cancer RNA-seq dataset was analyzed by (A) molecular subtypes of breast tumors (n = 1060); (B) by ethnicity and molecular subtypes of breast tumors; and (C) by ethnicity in five different types of tumors. The values are presented as log2 FPKM. Basal, basal-like; Her2, Her2-enriched, LumA, luminal A; LumB; NL, normal-like subtype of breast cancer. Normal indicates normal breast tissues. COAD, colon adenocarcinoma (n = 272); HNSC, head and neck squamous cell carcinoma (n = 492); KIRC, kidney renal clear cell carcinoma (n = 518); KIRP, kidney renal papillary cell carcinoma (n = 267); LUAD, lung adenocarcinoma (n = 440).

Fig 4: Abundant expression of TRAIL in starvation-ressistant renal cell carcinoma(A) RPKM value of TNFSF10 (TRAIL) in the starvation-resistant SW839 RCC cell line and the starvation-sensitive NC65 RCC cell line. (B) Quantitative RT-PCR of TRAIL in the starvation-resistant (SW839, VMRC-RCW, and KMRC-1) and starvation-sensitive (NC65, ACHN, Caki1, and Caki2) RCC cell lines. (C) Data shown in panel B for TRAIL were confirmed by western blotting. (D) Quantitative graphic representation of data shown in panel C. Transcriptional and protein expressions were normalized against GAPDH and α-tubulin, respectively. Error bars represent standard errors from three independent experiments. ANOVA: F (6, 14) = 336.47, p = 2.614e–14; p < 0.05, pairwise comparisons using t tests with pooled SD vs. SW839 (*). Note that TRAIL mRNA and protein expression was higher in starvation-resistant cells than in starvation-sensitive cells.

Fig 5: Decreased infiltration of CD4+ T cells into TNFSF10-KO breast tumors, in vivo. (A and B) Representative images from IHC analyses of CD4+ or CD8+ T cells in 4 T1 breast tumors with TNFSF10-KO or WT genotypes engrafted into syngeneic mice. Tumor-infiltrating CD4+ or CD8+ T cells were stained with anti-mouse CD4 or anti-mouse CD8 antibodies (in brown) and compared between TNFSF10-KO and WT tumors. Original magnification of 200× was used for scoring. The scale bar of 200 μm was included in all images. (B) The density (number) of CD4 or CD8 positive cells was estimated by counting all immuno positive cells in five high-powered fields. The result was presented as a number of positively stained cells per 1 mm2 of each tumor (the mean and SD) from five tumors (**P < 0.01).