KRT8 upregulation promotes tumor metastasis and is predictive of a poor prognosis in clear cell renal cell carcinoma

Abstract

Keratin 8 (KRT8) plays an essential role in the development and metastasis of multiple human cancers. However, its role in clear cell renal cell carcinoma (ccRCC) remains unexplored. Here, we investigated the expression pattern, clinical significance, and function of KRT8 in ccRCC. KRT8 mRNA and protein levels were determined in two large cohorts using quantitative real-time polymerase chain reaction (qRT-PCR) and tissue microarray (TMA) immunohistochemistry (IHC), respectively. We found that KRT8 expression was upregulated in ccRCC and vein tumor thrombi (VTTs). KRT8 overexpression in ccRCC was significantly correlated with aggressive characteristics and was predictive of a poor prognosis in ccRCC patients. Moreover, KRT8 overexpression in renal cancer cell lines promoted cell migration and invasion. In contrast, KRT8 knockdown suppressed ccRCC metastasis both in vitro and in vivo. In addition, our findings showed that KRT8 promoted ccRCC metastasis by increasing IL-11 expression, causing IL-11 autocrine induction, and triggering STAT3 signaling. Overall, this study established the significance of KRT8-IL-11 axis activation in aggressive ccRCC and defined a novel critical signaling mechanism that drives human ccRCC invasion and metastasis.

Keywords: IL-11; KRT8; biomarker; clear cell renal cell carcinoma; metastasis.

Figure 1. KRT8 upregulation in primary metastatic ccRCC tissues

(A) KRT8 mRNA expression levels in 109 paired ccRCC and adjacent NT tissues were evaluated by qRT-PCR. (B) Western blots showing KRT8 protein expression in tumor tissues (T) and paired adjacent NT tissues (N) from 32 ccRCC patients. (C) Relative IHC staining for KRT8 expression in paired ccRCC tissue samples (n=189). KRT8 expression levels were significantly upregulated in tumor tissues compared with corresponding adjacent non-tumor renal tissues. (D) IHC characteristics of KRT8 in ccRCC and adjacent NT tissues. Representative staining results for KRT8 are shown. (E) Relative KRT8 expression levels in 15 pairs of ccRCC and VTT tissues, as assessed by qRT-PCR. (F) Representative IHC images of matched VTT/tumor/peritumor tissues.

Figure 2. KRT8 upregulation serves as a prognostic factor for patients with ccRCC

(A and B) The high KRT8 subgroup had significantly shorter PFS and OS than the low KRT8 subgroup in the qRT-PCR cohort. (C and D) Similar results were observed in the TMA cohort comprising 147 primary non-metastatic patients with ccRCC. (E and F) The prognostic value of KRT8 was also observed in patients with early stage localized ccRCC (pT stage I and II). Statistical significance was assessed by two-sided log-rank tests.

Figure 3. KRT8 promotes ccRCC metastasis in vitro

Scratch wound healing assays and transwell assays showed that KRT8 knockdown inhibited the migratory and invasive properties of the renal cancer cell lines Caki-1 (A, C and E) and ACHN (B, D and F). The representative results and statistical analysis are shown.

Figure 4. KRT8 promotes ccRCC metastasis in vivo

(A) Images of lung metastases that developed in the Caki-1 cell lines in the lateral tail vein injection models. The images were acquired using an IVIS Imaging System. Representative luciferase signals captured in each group at the time of the initial injection: 0 days, 7 days and 42 days after cell injection are shown. The statistical analysis is shown in (B). (C) Representative H&E-stained images of lung metastatic loci from each group in (A). The statistical analysis is shown in (D).

Figure 5. KRT8 activates IL-11/STAT3 signaling in ccRCC

(A) Genes that are differentially expressed in Caki-1-SH cells compared with Caki-1-NC cells. (B) Relative IL-11 mRNA and protein expression levels in KRT8-knockdown cells, KRT8-overexpression cells and paired controls. IL-11 expression was analyzed by western blotting and normalized to β-actin. (C) IL-11 concentrations in the culture media from stable KRT8-knockdown or KRT-overexpression cells and paired controls were determined by ELISA. (D) Relative p-STAT3 and STAT3 expression levels in stable KRT8-knockdown cells, KRT8-overexpression cells and paired controls were analyzed using western blotting and were normalized to β-actin. (E) IL-11 mRNA expression levels in paired ccRCC tissue samples (n=109). (F) The correlation between KRT8 mRNA levels and IL-11 mRNA levels was measured in the same set of ccRCC tissues that was assessed in (E).

Figure 6. KRT8 is required for the IL-11-mediated metastatic phenotype in ccRCC

(A) IL-11 mRNA levels were determined by qRT-PCR, and IL-11, p-STAT3, and STAT3 protein levels were determined by western blotting. (B) IL-11 overexpression restored Caki-1-SH-mediated migration and invasion ability. In contrast, IL-11 knockdown significantly reduced 786-O-KRT8 cell migration and invasion. (C) Representative images of the mice over time after the tail vein injections with each type of renal cancer cell. The statistical analysis is shown in (D). (E) Representative H&E-stained images of the lung metastatic loci in each group in (C). The statistical analysis is shown in (F).