Upregulated ankyrin repeat-rich membrane spanning protein contributes to tumour progression in cutaneous melanoma

Abstract

Background: We have previously demonstrated that overexpression of ankyrin repeat-rich membrane spanning (ARMS) protein facilitates melanoma formation via conferring apoptotic resistance. This study aims to investigate whether ARMS contributes to melanoma progression.

Method: Using immunohistochemistry, we graded the expression level of ARMS in 54 cases of primary melanoma and 46 cases of metastatic melanoma. The immunointensity of ARMS was statistically correlated with individual clinicopathological characteristics. By RNA interference, stable melanoma cell clones with ARMS-knockdown were constructed, and were used for in vitro scratch wound, transwell invasion assays, and in vivo lung metastasis experiment.

Results: Stronger immunointensity of ARMS was observed mostly in melanomas with Breslow tumour thickness >1.0 mm (Fisher's exact test, P=0.002) or with nodal metastasis (Fisher's exact test, P=0.026), and was correlated with a worse overall survival in melanoma patients (log-rank test, P=0.04). Depletion of ARMS inhibited migration, invasion, and metastatic potential of melanoma cells in vitro and in vivo. Moreover, ARMS mediated melanoma cell migration and invasion through activation of the extracellular signal-regulated kinase (ERK) kinase (MEK)/ERK signalling pathway.

Conclusion: Ankyrin repeat-rich membrane spanning expression, conjunctly with tumour thickness or ulceration, may serve as a prognostic factor in patients with cutaneous melanoma.

Figure 1

Immunohistochemical analysis of ARMS in cutaneous melanoma. Representative immunohistochemistry of ARMS in melanoma in situ and melanoma at vertical growth phase. Ankyrin repeat-rich membrane spanning staining was negative in melanoma in situ ( × 200, A). Stronger immunointensity of ARMS was shown at the invasive front ( × 200, B and C, arrow heads) and in tumour cells invading the nerve in the dermis at vertical growth phase ( × 100, D). Either cytoplasmic granular ( × 400, E) or membranous ( × 400, F) distribution pattern of ARMS was observed in cutaneous melanoma.

Figure 2

Kaplan–Meier survival analysis between ARMS expression and the overall survival in melanoma patients. Patients with negative-, weak-, or moderate-ARMS expression have a significantly better overall survival than those with strong-ARMS expression (P<0.05, log-rank test).

Figure 3

Depletion of ARMS deteriorates migratory and invasive abilities in melanoma cells via MEK/ERK signalling. (A) Western blot and quantification of ARMS protein levels in mouse melanoma B16-F0, -F1, and -F10 cell lines as normalised to α-tubulin using densiometry showed increased expression of ARMS in B16-F10 cells, the most aggressive melanoma cell line among them. Melan-A cells were given a value of 1 for comparison. Mouse brain lysate was used as a positive control. (B) Western blotting showed effective ARMS knockdown by siRNA in human A2058 and A375 melanoma cells. (C) Effective RNAi of endogenous ARMS in two independent mouse B16-F10 ARMS–RNAi stable cell clones as compared with their controls (scrambled). Levels of α-tubulin were shown as a loading control. (D and E) ARMS knockdown in human A2058 and A375 melanoma cells resulted in a reduction of migration velocity (bar graph, means±s.d.) in a wound-scratch assay (D) and decreased the invasive ability of the cells in a Boyden chamber invasion assay (E). Representative photomicrographs with the cresyl violet stain and bar graph (means±s.d.) demonstrated the status of cellular invasion between the control and ARMS–RNAi cells in a Boyden chamber invasion assay (magnification × 200). The experiments were performed in triplicate. ^**P<0.01, Student's t-test. (F and G) ARMS silencing leaded to a significant reduction of migratory and invasive ability of B16-F10 cells via MEK/ERK signalling. Forced expression of RNAi-resistant ARMS or constitutively active MEK-1 (CA-MEK1) into B16-F10 melanoma cells with ARMS-knockdown resulted in a significant increase in cellular migratory ability (F, representative images and bar graph) and invasive potential in a Boyden chamber invasion assay (G, representative images and bar graph). Inhibition of MEK/ERK by PD98059 treatment in control and ARMS–RNAi cells compromised cell migration (F, bar graph) and invasion (G, images and bar graph). ^*P<0.05; ^**P<0.01; Student's t-test. (H) Overexpression of RNAi-resistant ARMS or CA-MEK1 in ARMS–RNAi cells resulted in ERK1/2 phosphorylation in melanoma cells compared with ARMS–RNAi cells. (I) The number of lung metastatic melanoma nodules was significantly decreased in mice receiving intravenous injection of ARMS–RNAi melanoma cells compared with that of control mice. Tumour nodules were shown in representative grossly dissected lungs (upper panels) and mouse lung tissue sections stained with haematoxylin and eosin (lower panels, × 200) in each group. ^**P<0.01, Student's t-test.