Gpnmb is a melanosome-associated glycoprotein that contributes to melanocyte/keratinocyte adhesion in a RGD-dependent fashion

Abstract

Gpnmb is a glycosylated transmembrane protein implicated in the development of glaucoma in mice and melanoma in humans. It shares significant amino acid sequence homology with the melanosome protein Pmel-17. Its extracellular domain contains a RGD motif for binding to integrin and its intracellular domain has a putative endosomal and/or melanosomal-sorting motif. These features led us to posit that Gpnmb is associated with melanosomes and involved in cell adhesion. We showed that human Gpnmb is expressed constitutively by melanoma cell lines, primary-cultured melanocytes and epidermal melanocytes in situ, with most of it found intracellularly within melanosomes and to a lesser degree in lysosomes. Our newly developed monoclonal antibody revealed surface expression of Gpnmb on these pigment cells, albeit to a lesser degree than the intracellular fraction. Gpnmb expression was upregulated by UVA (but not UVB) irradiation and by alpha-melanocyte-stimulating hormone (MSH) (but not beta-MSH); its cell surface expression on melanocytes (but not on melanoma cells) was increased markedly by IFN-gamma and TNF-alpha. PAM212 keratinocytes adhered to immobilized Gpnmb in a RGD-dependent manner. These results indicate that Gpnmb is a melanosome-associated glycoprotein that contributes to the adhesion of melanocytes with keratinocytes.

Figure 1. Protein expression of Gpnmb by melanoma and melanocytes

(A) Human Gpnmb-Fc, mouse IgG (2 µg/lane), or whole cell extracts (40 µg/lane) from COS-1 transfected with vector or phGpnmb vector (encoding human Gpnmb), or 293T cells were examined by immunoblotting for immunoreactivity to 3D5 mouse anti-human Gpnmb mAb. (B) Protein expression of Gpnmb in SK-MEL-28 cells was compared with that in primary-cultured human normal melanocytes. Expression of β-actin was also examined. (C) Mouse Gpnmb-Fc, human IgG, or whole cell extracts (40 µg/lane) from COS-1 cells transfected with pmGpnmb (encoding mouse Gpnmb) or empty vector, EL-4 or B16F10 cells were also subjected to immunoblotting using UTX-103 rabbit anti-mouse Gpnmb mAb. Protein bands immunoreactive to 3D5 or UTX-103 mAb are shown by closed arrowheads. Open arrowhead indicates non-specific band. (D) Whole cell extract from SK-MEL-28 or B16F10 was treated with (+)/without (−) N-glycosidase (N-Gly), followed by immunoblotting using anti-human (3D5) or anti-mouse Gpnmb mAb (UTX-103 or 1E4). (E) P815 cells transfected with pmGpnmb or vector alone were examined by flow-cytometry for surface expression of Gpnmb using UTX-103 mAb. B16F10 melanoma cells were left untreated (for surface expression) or permealized (for intracellular expression) and stained with UTX-103 mAb (open histogram) or control IgG (filled in gray). Expression was examined by flow-cytometry.

Figure 2. In situ expression of Gpnmb in human skin

Skin biopsy of human healthy adult was snap-frozen, thin-sectioned and stained with 3D5 mAb (green fluorescence) and anti-MITF mAb (red) (A) or corresponding control IgG (biotinylated and unmodified mouse IgG for 3D5 and anti-MITF, respectively) (B). Stained specimen was examined under confocal microscope at a magnification of 63x. Fluorescent images are merged (Merge, a scale bar represents 20 nm). The dotted line indicates location of basement membrane. A phase contrast picture is inserted in the merge of control staining.

Figure 3. Intracellular localization of Gpnmb in melanoma cells and DC

(A) B16F10 melanoma cells were fixed, permealized, and doubly-stained with UTX-103 (in green fluorescence) and anti-Mel-5 or anti-LAMP-1 mAb (in red), followed by confocal image analysis (Scale bar represents 10 nm). (B) DC generated from bone marrow cells were immunofluorescently stained with UTX-103 mAb and LAMP-1 mAb. Staining with control IgG is shown under staining with specific Ab. Fluorescence images of cells are merged (Merge), in which phase contrast image of stained cells is inserted. Data shown is representative of three independent experiments.

Figure 4. Fractionation of lysosomes and melanosomes

Crude fractions containing melanosomes (A) and lysosomes (B) were prepared from B16F10 melanoma cells and then further fractionated by a discontinuous gradient of 1.0–2.0 M and 0.3–1.8 M sucrose, respectively. An aliquot of each fraction was examined for expression of Gpnmb (using 1E4 mAb) and a marker protein (Mel-5 for melanosomes or LAMP-1 for lysosomes) by immunoblotting. Coomasie staining of blotted membranes is shown under the immunoblotting. Blotting images shown in B were acquired by longer exposure than those in A. Second experiment also showed consistent results.

Figure 5. UVA irradiation upregulates expression of Gpnmb

B16F10 cells were irradiated with different doses of UVA (A) or UVB (B and C) and cultured for 24 h (A and B) or 48 h (C). B16F10 cells were also cultured with β-MSH (0.2 µM) plus IBMX (0.1mM) (D) for the indicated time or cultured with varying doses of α-MSH for indicated time (E). Whole cell extracts (20 µg/lane) were prepared from treated cells and assayed by immunoblotting for expression of Gpnmb or β-actin. Expression levels shown under the immunoblotting are expressed as fold increase over control that is calculated by dividing experimental relative expression level (intensity of Gpnmb band/that of β-actin) by control expression level (untreated cells). Data shown are representative of three experiments (A and B) and 2 experiments (C through E).

Figure 6. Cytokine-dependent regulation of Gpnmb surface expression

Human melanocytes were cultured in the presence of different agents. After culturing for 1 day, cells were stained with 3D5 mAb (open histograms) or control IgG (closed histograms). Staining was examined by flow-cytometry. Mean fluorescence intensity (MFI) of staining by 3D5 mAb is shown in the histograms. Second experiment showed similar results.

Figure 7. PAM212 keratinocytes adhere to immobilized Gpnmb in a RGD-dependent manner

(A) Dose-dependent adhesion: ³H-thymdine-labeled PAM212 cells were allowed to adhere immobilized Gpnmb-Fc at varying doses. Adherent cells were lysed and measured for ³H-cpm. Adhesive activity is expressed as percent ³H-cpm relative to total input cpm. (B) Incubation time: At different time points after incubation of PAM212 cells with immobilized Gpnmb-Fc (20 µg/ml), adhesion to Gpnmb was assayed by ³H-cpm. (C) RGD-dependency: Adhesion of PAM212 cells to Gpnmb was blocked by addition of RGDS or ALAL (tetramer peptide) at varying doses. Adhesive activity is shown as % of maximum adhesion (without inhibitors). (D) RGD-deficient Gpnmb: PAM212 cells were incubated in 96-ELISA wells precoated with wild-type Gpnmb-Fc or RAA mutant (20 µg/ml) for 1 h. Adhesion is expressed by % adhesion of input PAM212 cells. (E and F) Effect of inhibitors on adhesion: Adhesion assay was performed in the presence of EDTA (E) or polysaccharides (F) (F, fucoidan; C, chondroitin sulfate; M, mannan; and H, heparin) at different concentrations. (G) Comparison with adhesion to fibronectin: ³H–labeled PAM212 cells were incubated for 60 min in ELISA wells precoated with IgG, Gpnmb-Fc (each 40 µg/ml) or fironectin (80 µg/ml) and adhesion measured. All data are representative of at least three independent experiments.