Osteopontin is induced by TGF-β2 and regulates metabolic cell activity in cultured human optic nerve head astrocytes

Abstract

The aqueous humor (AH) component transforming growth factor (TGF)-β2 is strongly correlated to primary open-angle glaucoma (POAG), and was shown to up-regulate glaucoma-associated extracellular matrix (ECM) components, members of the ECM degradation system and heat shock proteins (HSP) in primary ocular cells. Here we present osteopontin (OPN) as a new TGF-β2 responsive factor in cultured human optic nerve head (ONH) astrocytes. Activation was initially demonstrated by Oligo GEArray microarray and confirmed by semiquantitative (sq) RT-PCR, realtime RT-PCR and western blot. Expressions of most prevalent OPN receptors CD44 and integrin receptor subunits αV, α4, α 5, α6, α9, β1, β3 and β5 by ONH astrocytes were shown by sqRT-PCR and immunofluorescence labeling. TGF-β2 treatment did not affect their expression levels. OPN did not regulate gene expression of described TGF-β2 targets shown by sqRT-PCR. In MTS-assays, OPN had a time- and dose-dependent stimulating effect on the metabolic activity of ONH astrocytes, whereas TGF-β2 significantly reduced metabolism. OPN signaling via CD44 mediated a repressive outcome on metabolic activity, whereas signaling via integrin receptors resulted in a pro-metabolic effect. In summary, our findings characterize OPN as a TGF-β2 responsive factor that is not involved in TGF-β2 mediated ECM and HSP modulation, but affects the metabolic activity of astrocytes. A potential involvement in a protective response to TGF-β2 triggered damage is indicated, but requires further investigation.

Figure 1. Results of the Oligo GEArray microarray.

Densitometric measurements demonstrate induction of seven putative TGF-β2 responsive genes in human optic nerve head (ONH) astrocytes. (A) Oligo GEArray Human Extracellular Matrix and Adhesion Molecules microarray layout. (B) Chemiluminescence signals of microarrays incubated with total mRNA from ONH astrocytes treated with 1 ng/ml TGF-β2 and control cells (co). Regulated genes are marked. (C) Densitometric analysis of the gene induction of TGF-β2 responsive genes in ONH astrocytes compared to untreated cells. TGF-β2-dependent up-regulation of collagen 6α2 (COL6A2), collagen 8α1 (COL8A1), catenin delta 1 (CTNND1), integrin beta 8 (ITGB8), epsilon sarcoglycan (SGGE), osteopontin (OPN) and heat shock protein 90 (HSP90).

Figure 2. Quantification of OPN induction in TGF-β2 treated ONH astrocytes.

Analysis of OPN expression in ONH astrocytes after treatment with 1/ml TGF-β2 for 72 hours. (A) Semiquantitative (sq) RT-PCR analysis indicates an up-regulation of OPN transcript in TGF-β2 treated ONH astrocytes compared to control cells (co). (B) Western blot analysis demonstrates an increase in secreted OPN protein (MMP-cleaved) in TGF-β2 treated ONH astrocytes compared to controls. (C) Densitometric quantification of sq RT-PCR reveals significant induction of OPN mRNA (2.3-fold, p = 0.008). OPN signals are normalized to GAPDH. (D) Quantification of western blot results reveals significantly increased OPN secretion into supernatant (2.5-fold, p = 0.0054). OPN western blots are normalized within the β-actin signal on the same nitrocellulose membrane. (E) Statistical analysis of real time RT-PCR shows significant up-regulation of OPN transcript (6-fold, p = 0.0073). Values represent mean ± SD of 11 independent experiments (n = 11).

Figure 3. Semiquantitative RT-PCR analyses of most prevalent OPN receptors in cultivated ONH astrocytes and effect of TGF-β2.

(A) Representative RT-PCR results of OPN receptor gene expression in untreated (co) and TGF-β2 (1 ng/ml, 72 h) treated ONH astrocytes. (B) Densitometric analysis of sq RT-PCR results does not demonstrate any regulation of OPN receptors in TGF-β2 treated cells compared to controls. OPN signal is normalized to GAPDH. Values represent mean ± SD of 11 independent experiments (n = 11).

Figure 4. Localization of most prevalent OPN receptors on cultured human ONH astrocytes.

Immunofluorescence signals (green) indicate expression of CD44 (A) and integrin receptor subunits IntαV (B), Intβ3 (C) and Intβ5 (D). The nuclei (blue) are counterstained with DAPI. Pictures are representative for three independent experiments. Scale bar 50 μm.

Figure 5. Semiquantitative RT-PCR analysis of glaucoma-associated ECM components in ONH astrocytes after treatment with OPN.

Treatment with 250/ml OPN for 72 hours has no effect on gene expression of glaucoma-associated ECM components (A, B), genes of the ECM degradation system (C–F), as well as stress response genes (G, H) in cultivated ONH astrocytes compared to controls (0 OPN). (A, C, E and G) Representative RT-PCR results upon treatment with 250 ng/ml OPN () and controls (0 OPN). (B, D, F and H) Densitometric analysis of sq RT-PCR results normalized to GAPDH. Values represent mean ± SD of 11 independent experiments (n = 11).

Figure 6. Semiquantitative RT-PCR analysis of key regulator of ECM synthesis in ONH astrocytes after treatment with different OPN concentrations.

Increasing OPN concentration has no effects to the gene expression of key regulator of ECM synthesis (CTGF) or degeneration (PAI-1) in cultivated ONH astrocytes. (A) Representative RT-PCR results of connective tissue growth factor (CTGF) and plasminogen activator inhibitor-1 (PAI-1) in untreated (co) and OPN treated ONH astrocytes. (B) Densitometric analysis of sq RT-PCR results reveal no gene induction of CTGF and PAI-1 after treatment with 250, 1000 or 2000 ng/ml OPN compared to untreated cells (co). OPN signal is normalized to GAPDH. Values represent mean ± SD of 11 independent experiments (n = 11).

Figure 7. TGF-β2 and OPN effects on metabolic activity in cultured ONH astrocytes.

Cell viability was assessed by MTS assays. Quantification of the metabolic cell activity of ONH astrocytes treated with TGF-β2 (1 ng/ml) and different OPN concentrations (0, 250, 1000, 2000 ng/ml) for 48, 96 and 144 hours. Values represent mean ± SD of 11 independent experiments (n = 11). Statistical significance was calculated by student’s t-test (*p<0.05; **p<0.01; ***p<0.001).

Figure 8. Effects of selective OPN receptor blocking on metabolic activity in cultured ONH astrocytes.

Metabolic cell activity of ONH astrocytes incubated with an anti-CD44 antibody or a synthetic RGD peptide for 36 and 72 hours compared to untreated cells (co). Values represent mean ± SD of 11 independent experiments (n = 11). Statistical significance is calculated by student’s t-test (*p<0.05; **p<0.01; ***p<0.001).