Dual Role of DUOX1-Derived Reactive Oxygen Species in Melanoma

Abstract

Melanoma is the most serious type of skin cancer. Inflammation and oxidative stress play an essential role in the development of several types of cancer, including melanoma. Although oxidative stress promotes tumor growth, once cells escape from the primary tumor, they are subjected to a more hostile environment, with higher levels of oxidative stress typically killing most cancer cells. As Dual Oxidase 1 (DUOX1) is a major producer of reactive oxygen species (ROS) in epithelia, we used allotransplantation and autochthonous melanoma models in zebrafish together with in silico analysis of the occurrence and relevance of DUOX1 expression of the skin cutaneous melanoma (SKCM) cohort of The Cancer Genome Atlas (TCGA) to address the role of this enzyme in the aggressiveness of melanoma cells in vivo. It was found that high transcript levels of the gene encoding DUOX1 were associated with the poor prognosis of patients in the early-stage melanoma of TCGA cohort. However, DUOX1 transcript levels were not found to be associated to the prognosis of late-stage SKCM patients. In addition, the transcript level of DUOX1 in metastatic SKCM was lower than in primary SKCM. Using zebrafish primary melanoma and allotransplantation models, we interrogated the role of DUOX1 in vivo. Our results confirmed a dual role of DUOX1, which restrains melanoma proliferation but promotes metastasis. As this effect is only observed in immunocompromised individuals, the immune system appears to be able to counteract this elevated metastatic potential of DUOX1-deficient melanomas.

Keywords: DUOX1; melanoma; metastasis; oxidative stress; reactive oxygen species; zebrafish.

Figure 1

DUOX1 expression correlates with survival in SKCM patients and decreases in metastatic SKCM. (A) Kaplan–Meier survival analysis of TCGA cohort of early- (I + II + III Clark levels at diagnosis) and late- (IV and V Clark levels at diagnosis) stage SKCM patients according to their DUOX1 transcript levels: first quartile (Low), second and third (Medium) and fourth (High). Early-stage melanoma: Low, n = 22; Medium, n = 59; High, n = 16. Late-stage melanoma: Low, n = 57; Medium, n = 100; High, n = 60. Statistical significance was calculated with Mantel–Cox test. (B) Kaplan–Meier survival analysis of TCGA cohort of SKCM patients according to Clark-level classification at diagnosis: early-stage (I + II + III, n = 99) and late-stage (IV + V, n = 217). Statistical significance was calculated with Mantel–Cox test. (C) DUOX1 transcript levels in primary (n = 102) and metastatic (n = 360) SKCMs from TCGA cohort. Each dot represents a patient, and the mean is also shown. **** p < 0.0001 according to unpaired Student t test. a.u., arbitrary units.

Figure 2

Melanocyte DUOX1 inhibition does not affect melanocyte transformation and early melanoma progression. (A) Schematic representation of the procedure to co-express oncogenic NRAS-Q61R and DN-DUOX1 in melanocytes. Zebrafish one-cell Casper zebrafish embryos were injected with MinicoopR mitfa:NRAS-Q61R and either MinicoopR mitfa:DN-DUOX1 or MinicoopR mitfa:EGFP (control). Larvae were examined at 5 dpf for the presence of melanocytes and images were acquired monthly for 3 months to track melanoma development. (B) Representative images of the five different categories established to classify tumor progression. (C) Percentages of fish in the different categories at 30, 60, 75 and 90 dpf. (D) Tumor free curve. Representation of the percentage of fish without nodular tumors.

Figure 3

DUOX1 inhibition autonomously reduces aggressiveness and growth of transplanted SKCMs. (A) Schematic diagram showing adult allotransplantation procedure. One-year-old Casper zebrafish were irradiated 2 days before transplantation to prevent tumor rejection. Three hundred thousand cells from the nodular tumors of either MinicoopR mitfa:NRAS-QW61R/mitfa:DN-DUOX1 or MinicoopR mitfa:NRAS-Q61R/mitfa:EGFP (control) fish were subcutaneously injected in pre-irradiated recipients, and images were taken weekly during the following 4 weeks after transplantation and analyzed as indicated in the Methods and Materials section. Arrow, timeline; dpi, days post-injection; wpt, weeks post-transplant. (B) Representative images of transplanted melanoma growth rate of MinicoopR mitfa:NRAS-QW61R/mitfa:DN-DUOX1 and MinicoopR mitfa:NRAS-Q61R/mitfa:EGFP in pre-irradiated adult Casper zebrafish. (C) Average tumor size for each week post-transplant. Each dot represents a recipient-transplanted fish, and the mean is also shown. ** p < 0.01, **** p < 0.0001 according to unpaired Student t test. (D) Growth rate of transplanted MinicoopR mitfa:NRAS-QW61R/mitfa:DN-DUOX1 and MinicoopR mitfa:NRAS-Q61R/mitfa:EGFP SKCMs. DN-DUOX1: n = 5 tumors and 111 recipient fish; EGFP: n = 3 tumors and 72 recipient fish.

Figure 4

DUOX1 deficiency in SKCM promotes metastasis. (A) Schematic diagram of adult allotransplantation assays. (B) Representative images of the progression of metastasis (arrows). (C) Metastasis-free curve of adult zebrafish transplanted with MinicoopR mitfa:NRAS-Q61R/mitfa:DN-DUOX1 and MinicoopR mitfa:NRAS-Q61R/mitfa:EGFP (control). ** p < 0.01 according to a Log rank Mantel–Cox test. DN-DUOX1: n = 5 tumors and 111 recipient fish; EGFP: n = 3 tumors and 72 recipient fish.