Osteopontin is a novel downstream target of SOX9 with diagnostic implications for progression of liver fibrosis in humans - PubMed (original) (raw)

. 2012 Sep;56(3):1108-16.

doi: 10.1002/hep.25758.

Emma Harvey, Varinder Athwal, Andrew Berry, Cliff Rowe, Fiona Oakley, Anna Moles, Derek A Mann, Nicoletta Bobola, Andrew D Sharrocks, Brian J Thomson, Abed M Zaitoun, William L Irving, Indra N Guha, Neil A Hanley, Karen Piper Hanley

Affiliations

Free PMC article

Osteopontin is a novel downstream target of SOX9 with diagnostic implications for progression of liver fibrosis in humans

James Pritchett et al. Hepatology. 2012 Sep.

Free PMC article

Abstract

Osteopontin (OPN) is an important component of the extracellular matrix (ECM), which promotes liver fibrosis and has been described as a biomarker for its severity. Previously, we have demonstrated that Sex-determining region Y-box 9 (SOX9) is ectopically expressed during activation of hepatic stellate cells (HSC) when it is responsible for the production of type 1 collagen, which causes scar formation in liver fibrosis. Here, we demonstrate that SOX9 regulates OPN. During normal development and in the mature liver, SOX9 and OPN are coexpressed in the biliary duct. In rodent and human models of fibrosis, both proteins were increased and colocalized to fibrotic regions in vivo and in culture-activated HSCs. SOX9 bound a conserved upstream region of the OPN gene, and abrogation of Sox9 in HSCs significantly decreased OPN production. Hedgehog (Hh) signaling has previously been shown to regulate OPN expression directly by glioblastoma (GLI) 1. Our data indicate that in models of liver fibrosis, Hh signaling more likely acts through SOX9 to modulate OPN. In contrast to Gli2 and Gli3, Gli1 is sparse in HSCs and is not increased upon activation. Furthermore, reduction of GLI2, but not GLI3, decreased the expression of both SOX9 and OPN, whereas overexpressing SOX9 or constitutively active GLI2 could rescue the antagonistic effects of cyclopamine on OPN expression.

Conclusion: These data reinforce SOX9, downstream of Hh signaling, as a core factor mediating the expression of ECM components involved in liver fibrosis. Understanding the role and regulation of SOX9 during liver fibrosis will provide insight into its potential modulation as an antifibrotic therapy or as a means of identifying potential ECM targets, similar to OPN, as biomarkers of fibrosis.

Copyright © 2012 American Association for the Study of Liver Diseases.

PubMed Disclaimer

Figures

Fig. 1

Fig. 1

IHC of SOX9 and OPN in healthy liver. Consecutive 5-μm sections of healthy liver in rat and human (at 18 weeks post-conception [wpc] and adulthood) stained for SOX9 and OPN (brown) and counterstained with toludine blue. Note detection only in the round nuclei (SOX9) and cytoplasm (OPN) of biliary epithelial cells. Size bar represents 50 μm.

Fig. 2

Fig. 2

IHC of SOX9, OPN, and desmin in fibrotic liver. (A) Dual IF in fibrotic tissue from rat and mouse showing nuclear Sox9 (red) in biliary cells (asterisks) and in cells with cytoplasmic staining for desmin (green) (white arrowheads). (B) Consecutive 5-μm tissue sections shown from fibrotic rat and mouse liver stained with Sox9 and Opn (brown) counterstained with toluidine blue. Note similarly located staining for Sox9 and Opn in cells with more spindle-shaped nuclei (arrows) as well as in biliary cells (asterisk). Size bars represent 100 μm.

Fig. 3

Fig. 3

SOX9 and OPN expression in activated HSCs. (A-E) Quantification of SOX9 and OPN in quiescent and activated rHSCs and LX2 cells by qPCR (A and C) and immunoblotting (B, D, and E) (in [A], Sox9 was up-regulated 8.0-fold). In (B), induction of Sox9, Opn, and Col1 is shown during activation of rHSCs in culture (relative to quiescent; day 0). Representative immunoblotting images for (B) and (D) are shown as inset (B) or as an individual image (E), respectively. All immunoblotting quantification was normalized to β-actin. *P < 0.05; †P < 0.005, compared to quiescent day 0 cells (A and B) or 0% serum (C and D).

Fig. 4

Fig. 4

SOX9 and OPN expression in activated hHSCs and rHSCs. IF showing nuclear SOX9 (red) and cytoplasmic α-SMA (green; left panel) or OPN (green; right panel) in rat and human activated HSCs (aHSCs). Size bar represents 20 μm.

Fig. 5

Fig. 5

SOX9 regulation of OPN in HSCs. (A-C) siRNA abrogation of Sox9 in activated rHSCs (A and B) and LX2 cells (C) standardized against scrambled siRNA control for mRNA (A) and protein (B and C). Example immunoblotting is shown as inset in (B) and (C). *p < 0.05; **P < 0.01; †P < 0.005; ‡P < 0.001, compared to control. (D) Alignment of the upstream OPN enhancer region with conserved SOX9-binding motif highlighted in black (human sequence shown is −3,886 to −3,842 base pairs relative to transcriptional start site). Conserved nucleotides indicated by asterisk (*). The core SOX-binding motif is CAAT with increased binding affinity for SOX9 conferred by additional flanking nucleotides. (E) ChIP assay for SOX9-binding element in conserved upstream OPN enhancer element in LX2 cells cultured in high serum and activated rHSCs. Negative control is immunoglobulin G (IgG), and positive control is input (diluted 10-fold).

Fig. 6

Fig. 6

Hh regulates SOX9 and OPN expression in HSCs. (A-D) SOX9 and OPN protein levels quantified from immunoblotting of activated rHSCs and LX2 cells after 24-hour treatment with the Hh antagonist, CYC, or the Hh agonist, SAG. (E) Protein levels for SOX9 and OPN after treatment with SAG for 24 hours and knockdown of SOX9 (by 87%) using siRNA or scrambled control in LX2 cells. (F) Quantification of OPN protein after overexpression of SOX9 in LX2 cells in the presence or absence of CYC. Example immunoblotting image shown in inset. Change in expression is compared to vehicle treated cells (DMSO) for all experiments and, in the case of (F), EV control. Experiments standardized against β-actin. *P < 0.05; **P < 0.01; †P < 0.005; ‡P < 0.001, compared to control.

Fig. 7

Fig. 7

Gli2 mediates the expression of Sox9 and Opn in HSCs. (A) Expression of Gli factors in quiescent and activated rHSCs by qPCR. (B and C) siRNA for GLI2 (B, 67% knockdown) and GLI3 (C, 86% knockdown) or scrambled control in LX2 cells, followed by qPCR for GLI2 (B) or GLI3 (C), SOX9, and OPN. *P < 0.05; †P < 0.005; ‡P < 0.001, compared to scrambled siRNA treatment.

Fig. 8

Fig. 8

Gli2 overexpression rescues antagonistic effects of CYC on the expression of Sox9 and Opn in HSCs. (A-D) Quantification of SOX9 and OPN protein after overexpression of constitutively active GLI2 (GLI2ΔN; A and B) or active GLI3 (GLI3A; C and D) in LX2 cells in the presence or absence of CYC. (E) IF showing nuclear Gli2 (red) and cytoplasmic α-Sma (green) in activated rHSCs (aHSCs) and bright-field IHC showing nuclear Gli2 (brown staining) in CCl4-treated fibrotic rat liver. Arrows indicate Gli2 expression in a bile duct. *P < 0.05; **P < 0.01; †P < 0.005, ‡P < 0.001, compared to EV transfection. Size bar represents 20 μm.

References

    1. Friedman SL. Mechanisms of hepatic fibrogenesis. Gastroenterology. 2008;134:1655–1669. - PMC - PubMed
    1. Iredale JP. Models of liver fibrosis: exploring the dynamic nature of inflammation and repair in a solid organ. J Clin Invest. 2007;117:539–548. - PMC - PubMed
    1. Rangaswami H, Bulbule A, Kundu GC. Osteopontin: role in cell signaling and cancer progression. Trends Cell Biol. 2006;16:79–87. - PubMed
    1. Chiodoni C, Colombo MP, Sangaletti S. Matricellular proteins: from homeostasis to inflammation, cancer, and metastasis. Cancer Metastasis Rev. 2010;29:295–307. - PubMed
    1. El-Tanani MK. Role of osteopontin in cellular signaling and metastatic phenotype. Front Biosci. 2008;13:4276–4284. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources