Multiple Mechanisms of the Matricellular Protein Periostin in the Pathogenesis of Hepatic Fibrosis

Emory University, Atlanta GA

Investigators

Linked publications, trials & patents

Abstract

PROJECT ABSTRACT Cirrhosis is a major source of morbidity and mortality in the United States, however patient treatment for hepatic fibrosis is essentially non-existent. Fibrillogenesis, defined by chemical cross-linking of alpha I and other fibrillar collagens, is critical to the progression of liver fibrosis. The matricellular protein, periostin, resides in ECM but serves as a regulatory molecule for fibrillogenesis by working in concert with lysyl oxidase-like proteins (LOXLs) to promote collagen cross-linking. We have found periostin mRNA and protein are increased in C57BL/6 mice subjected to carbon tetrachloride (CCl4) gavage. Conversely, in vivo studies performed in periostin global knockout mice revealed protection against CCl4-induced liver stiffness. Periostin strongly co-localized with alpha smooth muscle actin expression in myofibroblasts in either CCl4-gavaged WT mice, or WT mice subject to BDL. The preliminary data acquired along with published literature provide a firm scientific premise to test the central hypothesis of this proposal: Periostin produced by myofibroblasts (MyoFs) promotes hepatic fibrosis via a novel signaling pathway by direct binding to DDR1, activating Pyk2, and increasing liver stiffness by activation of lysyl oxidase (LOX) activity which together amplify the fibrotic response. This hypothesis will be tested by the following three aims. SPECIFIC AIM 1: Demonstrate that hepatic stellate cells (HSCs) are the primary cellular source of periostin during induction of liver fibrosis in vivo, and that subsequently hepatic MyoFs preferentially synthesize and secrete the pro-fibrogenic periostin splice variant periostin 2 containing exon-17. Using HSC-specific Lrat-Cre reporter mice, we will lineage map HSCs to establish the spatiotemporal hepatic source of periostin during liver fibrosis. We have found the splice variant exon-17 containing periostin is uniquely present in MyoFs, but absent in quiescent HSCs. We will isolate both quiescent HSCs and MyoFs by fluorescence-activating cell sorting (FACS) to identify downstream effectors of the profibrogenic effects of periostin by RNA-Seq. SPECIFIC AIM 2: Establish that periostin is a DDR1 ligand which promotes fibrogenesis by triggering a Pyk2-mediated signal transduction pathway. Immunoprecipitation (IP) studies have demonstrated that periostin binds DDR1 in MyoFs and activates DDR1-RTK activity. We will identify the periostin binding site for DDR1 via solid phase binding assay, and by use of His-tagged periostin protein isoforms. SPECIFIC AIM 3: Establish an in vivo mechanism whereby periostin functions independently of TGF?1 as a pro-fibrogenic stimulus through increased LOX expression and activity. Periostin and LOX have been previously thought to be dependent on TGF?1?activity, but preliminary data indicate periostin can directly drive activation of Smad binding element-4 and Smad phosphorylation. Therefore we will determine the direct contribution of periostin following either CCl4gavage or BDL in TGF?1 null mice crossed with Postn+/lacZreporter mice to demonstrate that periostin actions are TGF?1/Rag1 independent.