Prognostic significance of Sox4 expression in human cutaneous melanoma and its role in cell migration and invasion

Abstract

The Sox4 transcription factor is involved in various cellular processes, such as embryonic development and differentiation. Deregulated expression of Sox4 in several human cancers has been reported to date, but its role in melanoma is unknown. We explored the role of Sox4 in melanoma pathogenesis in vivo and in vitro. Using tissue microarray, we evaluated Sox4 expression in 180 melanocytic lesions and investigated its role in melanoma cell migration and invasion. Sox4 expression was remarkably reduced in metastatic melanoma compared with dysplastic nevi (P < 0.05) and primary melanoma (P < 0.01). This reduction was correlated with a poorer disease-specific survival of melanoma patients (P = 0.039). Multivariate Cox regression analysis revealed that reduced Sox4 expression is an independent prognostic factor (P = 0.049). Knockdown of Sox4 enhanced melanoma cell invasion, migration, and stress fiber formation. The increased migration and invasion on Sox4 knockdown depends on the presence of nuclear factor (NF)-κB p50 and is abrogated when p50 is knocked down. We further observed inhibition of NF-κB p50 transcription by Sox4, in addition to a reverse pattern of expression of Sox4 and NF-κB p50 in different stages of melanocytic lesions. Our results suggest that Sox4 regulates melanoma cell migration and invasion in an NF-κB p50-dependent manner and may serve as a prognostic marker and potential therapeutic target for human melanoma.

Figure 1

Expression of Sox4 protein in cutaneous melanoma. A: Representative images of dysplastic nevi (DN) with strong nuclear staining, primary melanoma (PM) with moderate staining, and metastatic melanoma (MM) with negative staining. Scale bar = 100 μm. B: Kruskal-Wallis test for differences in Sox4 staining among DN, PM, and MM. The median is depicted as a horizontal line inside each box. *P < 0.05; **P < 0.01.

Figure 2

Kaplan-Meier analysis of correlation between Sox4 expression and 5-year survival. A and B: Overall and disease-specific 5-year survival of all primary and metastatic melanoma patients. C and D: Overall and disease-specific 5-year survival of metastatic melanoma and high-risk primary melanoma patients (thickness >1.5 mm).

Figure 3

Enhancement of melanoma cell migration and invasion on Sox4 knockdown in a NF-κB p50-dependent manner. MMRU melanoma cells were transfected with either siCTR, siSox4, sip50, or co-transfected with siSox4/sip50. A: Wound-healing assay was performed on monolayers of MMRU melanoma cells 48 hours after transfection. Original magnification, ×100. B: Quantitation of A, *P < 0.05; **P < 0.01. C: For Boyden chamber assay, cells were suspended in serum-free medium and seeded on Matrigel, incubated at 37°C for 24 hours, stained by crystal violet, and quantified. *P < 0.05; **P < 0.01, ***P < 0.001. D: Protein extracts were prepared at 72 hours after transfection and analyzed for the expression of Sox4 and NF-κB p50 Western blot analysis. β-actin was used as a loading control.

Figure 4

Sox4 knockdown induces stress fiber formation. MMRU cells were transfected with siSox4 or siCTR, followed by serum starvation overnight and serum stimulation for 30 minutes. For Rock inhibitor treatment, cells were treated with serum-free medium containing 10 ìM Y27632 overnight, and then incubated with complete medium containing 10% fetal bovine serum and 10 ìM Y27632 for 30 minutes. A: Increased formation of stress fiber on Sox4 knockdown compared to the control. Treatment with ROCK inhibitor Y27632 (RI) abrogates the Sox4-KD effect. Original magnification ×630. B: Quantitation of A, ***P < 0.001.

Figure 5

Inhibition of NF-κB p50 transcription by Sox4. A: For quantitative reverse transcription-PCR analysis, cells were transfected with siCTR or siSox4, and lysed for total RNA extraction and reverse transcription 72 hours after transfection. Expression of NF-κB p50 and Sox4 mRNAs was measured by real-time quantitative PCR and normalized with β-actin as loading control. **P < 0.01, ***P < 0.001. B: Protein extracts were prepared at 24 hours after transfection and analyzed for the expression of 3×Flag-Sox4. β-actin was used as a loading control. C: Sox4 protein binds to the NF-κB p50 gene promoter sequence in MMRU cells in vivo, demonstrated by chromatin immunoprecipitation assay. Binding of Sox4 protein to Dicer1 promoter sequence was used as a positive control.

Figure 6

Inverse correlation between expression of Sox4 and NF-κB p50 in melanoma. A: Inverse correlation between Sox4 and NF-κB p50 expression in 169 melanocytic lesions at different stages. A significant reduced expression of Sox4 (P = 0.0260, χ² test) and elevated expression of NF-κB p50 (P = 0.0147, χ² test) in metastatic melanoma compared with the primary melanoma and dysplastic nevi is observed. B: Combined analysis of Sox4 and NF-κB p50 staining. Each sample was categorized based on the expression of Sox4 and NF-κB p50 as follows: 1, low (negative-weak) Sox4/high (moderate-strong) NF-κB p50; 2, high Sox4/low NF-κB p50; 3, high Sox4/high NF-κB p50. C: Representative images of serial sections of a high-risk nodular primary melanoma stained for NF-κB p50 and Sox4. Scale bar = 100 μm for left and right panels and 400 μm for the middle panels.