As an independent unfavorable prognostic factor, IL-8 promotes metastasis of nasopharyngeal carcinoma through induction of epithelial-mesenchymal transition and activation of AKT signaling

Abstract

Nasopharyngeal carcinoma (NPC) has the highest metastatic potential among head and neck cancers. Distant metastasis is the major cause of treatment failure. The role of interleukin-8 (IL-8) in NPC progression remains unknown. Our multivariate survival analyses of 255 patients with NPC revealed that higher IL-8 expression in primary NPC tissue was an independent prognostic factor for overall survival, disease-free survival, and distant metastasis-free survival of the patients. In vitro study revealed that IL-8 was highly expressed in the established high-metastasis NPC clone S18 relative to the low-metastasis cells. Suppression of IL-8 by short-hairpin RNA reduced the expression of IL-8 in S18 cells and subsequently inhibited migration, invasion, and hepatic metastasis of the cells without influencing cellular growth. Overexpression of IL-8 in S26 cells resulted in increased migration, invasion, and metastasis capabilities of the cells without affecting cellular growth. Exogenous IL-8 enhanced the migration and invasion of low-metastasis CNE-2 cells in a dose-dependent manner. An epithelial-mesenchymal transition (EMT) could be induced by IL-8 in various NPC cell lines. The high level of phosphorylated AKT in S18 cells could be suppressed by knocking down IL-8 expression. Further, IL-8-promoted migration and invasion could be abolished by either the application of the phosphoinositide-3-kinase inhibitor LY294002 or the knock down of AKT expression by using small-interfering RNA. In summary, IL-8 serves as an independent prognostic indicator of overall survival, disease-free survival, and metastasis-free survival for patients with NPC. IL-8 promotes NPC metastasis via autocrine and paracrine means, involving activation of AKT signaling and inducing EMT in NPC cells.

Fig. 1.

High IL-8 expression correlates with shorter overall survival, disease-free survival, and distant-metastasis-free survival in patients with NPC. (A) Tissue microarrays representative of those from 255 patients with NPC diagnosed at the M0 stage immunohistochemically stained using antibody against IL-8, at low (40×) and high (400×) magnification in a light microscope. Black scale bars0 500 µm; yellow scale bars, 100 µm. (B) The median follow-up time of these 255 patients was 69 months. (C) Overall survival (OS) curve. (D) Disease-free survival (DFS) curve. (E) Distant-metastasis-free survival (DMFS) curve. P values were calculated by the log-rank test.

Fig. 2.

IL-8 expression level correlates with the in vivo metastasis ability of NPC cells. (A) Relative mRNA levels of IL-8 in NPC cells using quantitative PCR. (B) Concentration of secreted IL-8 in conditioned medium as measured by ELISA. (C) mRNA levels of IL-8 in S18 cells expressing IL-8 shRNAs (IL-8 KD1 and KD2) or scrambled shRNA as measured using quantitative PCR. (D) Concentration of secreted IL-8 in conditioned medium of S18 cells expressing IL-8 shRNAs or scrambled shRNA as measured by ELISA. (E) Migratory and invasive abilities of S18 cells expressing IL-8 shRNAs or scrambled shRNA evaluated by Transwell assay. *P < 0.001 relative to scrambled controls (Student’s t-test). Photomicrographs are at 100× (right panel). Bars correspond to mean ± standard deviation (SD) of three independent experiments. (F) In vivo liver metastasis ability of S18 cells expressing IL-8 shRNA or scrambled shRNA evaluated in nude mice. Bars correspond to mean ± standard error (SE), with P value calculated using Student’s t-test (left panel). The six livers having the largest number of metastatic nodules from each group are shown in the right panel. Scale bars, 1cm.

Fig. 3.

Overexpression of IL-8 promotes migration, invasion, metastasis, and EMT in NPC cells. (A) The concentration of secreted IL-8 in conditioned medium of S26 and HONE-1 cells expressing IL-8 or empty vector. (B) Migratory and invasive abilities of S26 and HONE-1 cells expressing IL-8 or empty vector as evaluated by Transwell assay. *P < 0.001, **P = 0.0027, ^# P = 0.0016 relative to that of vector as controls (Student’s t-test). Photomicrographs are at 100× (right panel). Bars correspond to mean ± SD of three independent experiments. (C) In vivo liver metastasis ability of S26 cells expressing IL-8 or empty vector as evaluated in nude mice. Bars correspond to mean ± SE, with P value calculated using Student’s t-test (left panel). The six livers having the largest number of metastatic nodules from each group are shown (right panel). Scale bars, 1cm. (D) Immunoblots were performed with the indicated antibodies in whole-cell lysates.

Fig. 4.

IL-8 promotes NPC cell migration and invasion through AKT activation. (A) Immunoblots of whole-cell lysates using anti-phospho-AKT (Ser473) or anti-AKT (pan) antibody. (B) Immunoblots of whole-cell lysates from S26 cells expressing IL-8 or the empty vector after pretreatment with 20 μM LY294002 for 30min, using anti-phospho-AKT (Ser473) or anti-AKT (pan) antibody. (C) Migratory and invasive abilities of S26 cells expressing IL-8 or the empty vector evaluated by Transwell assay after pretreatment with LY294002. (D) Immunoblots of whole-cell lysates from S26 cells expressing IL-8 or the empty vector 36h after transfection with AKT siRNA or a scrambled siRNA, using anti-phospho-AKT (Ser473) or anti-AKT (pan) antibody. (E) Migratory and invasive abilities of S26 cells expressing IL-8 or the empty vector as evaluated by Transwell assay after transfection with AKT siRNA or the scrambled siRNA. Photomicrographs are at 100×. Bars correspond to mean ± SD of three independent experiments. β-actin served as the loading control.

Fig. 5.

Recombinant IL-8 promotes migration and invasion through activation of AKT in NPC cells. (A) Immunoblots of whole-cell lysates from CNE-2 cells stimulated with recombinant IL-8 (rIL-8) for 30min, using anti-phospho-AKT (Ser473) or anti-AKT (pan) antibody. (B) Migratory and invasive abilities of CNE-2 cells stimulated with various concentrations of rIL-8 as evaluated by Transwell assay. *P = 0.02132, **P = 0.0025, ^# P = 0.00267, ^## P < 0.001, relative to cells without rIL-8 stimulation (Student’s t-test). (C) Immunoblots of whole-cell lysates from CNE-2 cells transfected with AKT siRNA or scrambled siRNA and stimulated with rIL-8 (1.5ng/ml) for 30min at 36h after siRNA transfection, using anti-phospho-AKT (Ser473) or anti-AKT (pan) antibody. (D) Migratory and invasive abilities of CNE-2 cells transfected with AKT siRNA or scrambled siRNA and stimulated with rIL-8 as evaluated by Transwell assay. The photomicrographs are at 100×. Bars correspond to mean ± SD of three independent experiments. β-Actin served as the loading control.