Interleukin-8 (IL-8) over-production and autocrine cell activation are key factors in monomethylarsonous acid [MMA(III)]-induced malignant transformation of urothelial cells

Abstract

The association between chronic human exposure to arsenicals and bladder cancer development is well recognized; however, the underlying molecular mechanisms have not been fully determined. We propose that inflammatory responses can play a pathogenic role in arsenic-related bladder carcinogenesis. In previous studies, it was demonstrated that chronic exposure to 50 nM monomethylarsenous acid [MMA(III)] leads to malignant transformation of an immortalized model of urothelial cells (UROtsa), with only 3 mo of exposure necessary to trigger the transformation-related changes. In the three-month window of exposure, the cells over-expressed pro-inflammatory cytokines (IL-1β, IL-6 and IL-8), consistent with the sustained activation of NFKβ and AP1/c-jun, ERK2, and STAT3. IL-8 was over-expressed within hours after exposure to MMA(III), and sustained over-expression was observed during chronic exposure. In this study, we profiled IL-8 expression in UROtsa cells exposed to 50 nM MMA(III) for 1 to 5 mo. IL-8 expression was increased mainly in cells after 3 mo MMA(III) exposure, and its production was also found increased in tumors derived from these cells after heterotransplantation in SCID mice. UROtsa cells do express both receptors, CXCR1 and CXCR2, suggesting that autocrine cell activation could be important in cell transformation. Supporting this observation and consistent with IL-8 over-expression, CXCR1 internalization was significantly increased after three months of exposure to MMA(III). The expression of MMP-9, cyclin D1, bcl-2, and VGEF was significantly increased in cells exposed to MMA(III) for 3 mo, but these mitogen-activated kinases were significantly decreased after IL-8 gene silencing, together with a decrease in cell proliferation rate and in anchorage-independent colony formation. These results suggest a relevant role of IL-8 in MMA(III)-induced UROtsa cell transformation.

Figure 1

Chronic exposure of UROtsa cells to 50 nM of MMA(III) leads to sustained IL-8 gene over-expression. The bars represent the mean of four independent biological samples processed under the same conditions. Asterisks indicate a significant difference [(***) P< 0.001] compared to non-exposed cells.

Figure 2

IL-8 production profile after chronic exposure of UROtsa cells to 50 nM MMA(III). The bars represent the mean ± SD of three independent experiments. Asterisk indicates a significant difference [(***) P< 0.001] compared to nonexposed cells as shown by two-way ANOVA followed by Bonferroni’s test.

Figure 3

IL-8 is expressed in tumors derived from heterotransplanted UROtsa cells. Tissue sections were assayed for IL-8 and Ki67 (proliferation marker) by specific IHC. Tumors derived from cells that were exposed to 50 nM MMA (III) for 3 mo showed higher IL-8 production with an intra-cytoplasmic localization in the basal layer. IL-8 production was also found in the superficial layer, compared to that produced from tumors derived from cells that were not exposed to MMA(III), which was found only in the basal layer with a peri-nuclear localization. These findings were consistent with the expression of the cell proliferation marker Ki67 and tumor size.

Figure 4

UROtsa cells constitutively express CXCR1 and CXCR2. Untreated UROtsa cells were processed to determine the expression of surface and intracellular CXCR1 and CXCR2. The receptors were found to be mainly intracellular, suggesting IL-8 autocrine activation. A) Histogram of surface and intracellular CXCR1 determination. B) and C) Percentage of intracellular and surface CXCR1 and CXCR2 on untreated UROtsa cells and the lymphoblastoid cell line LCL18564, respectively.

Figure 5

Increased internalization of CXCR1 in MMA(III)-exposed UROtsa cells. Flow cytometry analysis showed that the cells exposed to 50 nM MMA(III) for 3 or more months had an increased internalization of the IL-8 receptor CXCR1. The bars represent the mean ± SD of at least three independent experiments. Asterisks indicate a significant difference [(***) P< 0.001], compared to nonexposed cells as shown by two-way ANOVA followed by Bonferroni’s test.

Figure 6

UROtsa cells over-expressing IL-8 showed an increase in the expression of genes associated with tumor growth and invasion. RNA from cells exposed to MMA (III) for 3 mo was isolated and transformed to cDNA, and the expression of MMP-9, cyclin D1, VGEF and bcl-2 was assayed by real- time PCR. For comparative purposes, RNA recovered from UROtsa cells treated with 10 ng/ml rIL-8 for 24 h was processed under the same conditions to determine gene expression. Bars represent mean ± SD of four independent biological samples. Asterisks indicate a significant difference [(***) P< 0.001, (*) P< 0.05], compared to nonexposed UROtsa cells as shown by ANOVA followed by Bonferroni’s test.

Figure 7

The expression of IL-8 gene was silenced in MMA(III)-exposed UROtsa cells by shRNA lentivirus transduction particles. A) IL-8 gene expression was reduced 99.7 % in 3 mo (IL8-) cells. B) Protein production was reduced 60% compared to non-IL-8 sh RNA transfected cells that were exposed to MMA(III) for 3 mo and cells transfected with non-target control transduction particles (NTC).

Figure 8

IL-8 gene silencing in MMA(III)-exposed UROtsa cells leads to a decrease in growth rate and in colony formation ability. After IL-8 silencing, UROtsa cells exposed to MMA(III) for 3 mo (3 mo IL8 (−) cells) were assayed for cell proliferation (A) as well as growth in soft agar (B). Bars represent the mean ± SD of three independent experiments. Asterisks indicate a significant difference [(***) P< 0.001], compared to cells exposed to MMA(III) for 3 mo, which express normal IL-8 quantities, and to NTC-treated cells, as shown by ANOVA followed by Bonferroni’s test.

Figure 9

The expression of MMP-9, cyclin D1, bcl-2, and VGEF coding genes was significantly decreased in UROtsa cells exposed to MMA(III) for 3 mo, after IL-8 gene silencing compared with non-transduced cells or NTC-transduced cells expressing normal quantities of IL-8. The bars represent the mean ± SD of four different biological samples, and significant differences are shown with asterisks [(***) P< 0.001].

Figure 10

IL-8 gene silencing decreases tumor growth in MMA(III)-transformed UROtsa cells. UROtsa cells exposed chronically to 50 nM MMA (III) for 3 mo or stably transduced 3 mo MMA (III)-treated UROtsa cells (10 × 10⁶ cells in 0.1 ml PBS) were subcutaneously injected into the flank of SCID mice. (A) Tumor incidence represents the number of animals that developed tumors; tumor volume on day 60 after tumor injection (n=4). (B) Tumor size of 3 mo IL-8(−)decreases significantly compared to 3 mo Ctl cells that express “normal” IL-8 levels or with non target control (NTC) stably transfected cells. Results are shown as mean ± SEM. *Significantly different from controls (P≤0.05) as indicated by ANOVA followed by Bonferroni’s test.