Loss of adhesion in the circulation converts amelanotic metastatic melanoma cells to melanotic by inhibition of AKT

Abstract

Direct injection of murine K-1735 melanoma cells into the subcutis, lung, or brain of syngeneic mice produces amelanotic tumors, whereas intravenous injection into the lateral tail vein or internal carotid artery produces both amelanotic and melanotic foci in the lung and the brain respectively. We hypothesized that loss of adhesion in the circulation may contribute to the melanogenic phenotypes of cells. To test this, we used enforced suspension culture of K-1735 cells by consistent rotating culture of K-1735 cells. We found that the expression of the microphthalmia transcription factor (MITF) and melanin-stimulating hormone receptor (MSHR) were upregulated in cells growing in suspension and were accompanied by inhibitions of AKT and ERK, which were reversed in cells upon regrowth as an adherent monolayer. Inhibition of the AKT pathway was responsible for MITF induction by suspension culture. Stable expression of constitutively active AKT significantly repressed the melanogenesis of K-1735 cells injected via circulation. An amelanotic clone of K-1735 cells was resistant to suspension culture-induced MITF, although the inhibition of AKT pathway was intact. Collectively, these data suggest that the inhibition of AKT pathway due to loss of adhesion within the circulation renders a subpopulation of K-1735 cells to produce melanin.

Figure 1

Induction of melanogenesis in circulating K-1735 cells. Tumors produced by K-1735 cells injected directly into the subcutis subcutaneously (s.c.), lung intrapleural (i.p.), or brain intracerebroventricularly (i.c.) are amelanotic. In contrast, K-1735 cells introduced into the lung by injection into the lateral tail vein intravenously (i.v.), or into the brain by injection into the internal carotid artery (i.c.a.) produce melanotic lesions (bar = 10 mm). The heterogeneity of melanogenesis in brain lesions is shown by the H&E-stained ICA section. Bar = 20 µm.

Figure 2

Inhibition of AKT in circulating tumor cells. Sequential frozen sections of the brain (K-1735 cells injected through the internal carotid artery), lung (K-1735 cells injected intravenously through the tail vein), and the subcutis (subcutaneously) containing GFP-labeled K-1735 cells (arrowheads, green) were stained for pAKT (arrowhead, red), CD31 (arrow, red), and cell nuclei (Hoechst, blue). Note that the K-1735 melanoma cells within the vasculature are negative for pAKT and that the tumor cells in the subcutis are positive. Bar = 50 µm.

Figure 3

Suspension culture upregulation of melanogenesis-related genes inhibits the AKT and ERKpathways. (A) Western blot analysis of pAKT, tAKT, phosphorylated ERK (pERK1/2), total ERK (tERK1/2), and MITF. K-1735 cells growing in suspension culture (Sp) for 3 hours exhibit upregulated MITF expression and inhibited pAKT and pERK1/2 expression. These changes are not found in adherent K-1735 cells (Ad). The target gene of MITF, Bcl-2, was also upregulated accordingly. Actin was used as loading control. Cell lysate from B16-F10 melanoma cells served as positive control. (B) Northern blot analysis of MSHR mRNA expression in K-1735 cells grown in suspension or in adherent monolayers. K-1735 cells growing in suspension culture exhibit upregulated MSHR expression. Tubulin was used as loading control. Total mRNA from B16-F10 melanoma cells served as positive control.

Figure 4

Western blot analysis of MITF expression in adherent K-1735 cells treated with AKT inhibitor LY294002 or ERK inhibitor and PD98095. (A) The AKT inhibitor LY294002 inhibited AKT phosphoryla tion and induced MITF expression without affecting levels of either tAKT or phosphorylated ERK1/2 (pERK1/2). (B) The ERK inhibitor PD98095 did not induce MITF expression, but significantly inhibited pERK1/2. Actin was used as loading control.

Figure 5

Suspension culture leads to reversible upregulation of melanogenesis-related genes. Western blot analysis shows that the inhibition of pAKT and the upregulation of MITF in suspended cells can be reversed by allowing the suspended cells to readhere for 6 hours. Actin was used as loading control.

Figure 6

Constitutive expression of activated AKT can inhibit melanogenesis in circulating K-1735 cells. (A) Detection of membrane-bound pAKT (m-pAKT) expression by Western blot analysis in vector control cells and constitutively active AKT (myr-AKT1)-expressing cells grown in an adherent monolayer (Ad) or in suspension culture (Sp). Culturing cells in suspension for 6 hours significantly inhibits the phosphorylation of AKT in control cells, but not in myr-AKT1-expressing cells. (B) Gross anatomy of lungs with foci produced by control cells shows foci heterogeneous for melanogenesis. The foci in lungs from mice injected with myr-AKT1-expressing cells are amelanotic. (C) Gross anatomy of brain with control cells or myr-AKT1-expessing cells injected into the internal carotid artery. The brain with myr-AKT1-expressing tumor cell foci have melanin less than that of the foci produced by control cells. (D) Histologic analysis of brain tissues with foci produced by control and myr-AKT1-expressing cells. Heterogeneous expression of melanin is visible in sections containing control tumor cells. Tumor cells expressing myr-AKT1 are amelanotic. Bar = 50 µm.

Figure 7

Amelanotic C2 clone of K-1735 is resistant to suspension culture-induced melanogenic signaling, as revealed by Western blot analysis. Both the melanotic clone (C4) and the amelanotic clone (C2) respond to suspension culture (Sp) compared with adherent culture (Ad) by inhibiting the expression ofpAKT without altering the level of tAKT. However, unlike the C4 clone, in the C2 clone, MITF expression is not upregulated by growth in suspension. Actin was used as loading control.