Fig 2: The effect of DDX39A knockdown on the wound healing of melanoma cells. The scratch width and the relative migration rates of A375 (A,C) and SK-MEL28 (B,C) cells in the shDDX39A group and the shCtrl group were imaged and counted using the Celigo Cell Counting application system at 0, 6, and 24 h, separately. **P < 0.01. The mean and standard deviation (SD) values of at least three repeated experiments are reported.

Fig 3: The impact of DDX39 downregulation on the proliferation, invasion, and migration of melanoma cells was investigated by RNA interference mediated by lentivirus. The effect of DDX39A knockdown on the cell proliferation of A375 (A) and SKMEL28 (B) cells was measured by the Celigo Cell Counting application system. Beginning on day 2, the cells were imaged and counted daily for 5 days. The effect of DDX39A knockdown on the cell growth of A375 (C) and SK-MEL28 (D) cells was assessed by the MTT experiment in triplicate daily for 5 days. The mean and standard deviation (SD) values of at least three repeated experiments are reported. ***P < 0.001, compared with the shCtrl group.

Fig 4: The prognostic value of DDX39 overexpression in melanoma patients. The overall survival time was counted from the day of operation to the end of follow-up. The disease-free survival time was counted from the date of surgery to the time of disease recurrence or metastasis. (A) Patients with a higher expression of DDX39 had a shorter disease-free survival time. (B) Patients with a higher expression of DDX39 also had a shorter overall survival time.

Fig 5: (A) By analyzing the Affymetrix expression profile in the GEO database GSE 46517, DDX39, a member of the DEAD box RNA helicase family, was identified as an overexpressed gene in melanoma but not in pigmented nevi tissues. Immunohistochemical staining of DDX39 was weak in pigmented nevus tissues (B), while it was stronger in human melanoma tissues (C). Magnification, 400×. ***P < 0.001.