Halofuginone inhibition of COL1A2 promoter activity via a c-Jun-dependent mechanism

Objective. The naturally occurring compound halofuginone has been shown to antagonize collagen synthesis by fibroblasts both in vitro and in vivo. We previously demonstrated that this inhibitory property was related to the ability of halofuginone to disrupt transforming growth factor beta signal transduction. The present study further analyzed the ability, of halofuginone to affect transcription factors that can regulate type I collagen gene expression by examining its effect on c-Jun, the negative regulator of collagen gene transcription. Methods. The phosphorylation state of c-Jun in the presence of halofuginone was examined via direct Western blotting, and the transcriptional activity of the activator protein 1 (AP-1) binding element via electrophoretic mobility, shift assay and luciferase reporter assay. We determined whether the effect of halofuginone on collagen synthesis was dependent on the presence of c-Jun by, ectopic expression of a wild-type or dominant-negative c-Jun construct in the presence of halofuginone and assaying alpha2(I) collagen promoter strength via luciferase reporter assay. The effect of halofuginone on alpha2(I) collagen message levels in fibroblasts when wildtype or dominant-negative c-Jun was overexpressed was determined. We also determined whether halofuginone had an effect on the phosphorylation state of c-Jun in the skin of TSK/+ mice via immunohistochemistry. Results. Treatment of fibroblasts with 10(-8)M halofuginone enhanced basal and mitogen-mediated phosphorylation of c-Jun in culture. This elevated phosphorylation of c-Jun correlated with enhanced DNA binding and transcriptional activation of an AP-1 complex consisting of c-Jun and Fos but lacking the c-Jun antagonist JunB. Overexpression of c-Jun enhanced in a dose-dependent manner the ability of halofuginone to inhibit the activity of a luciferase reporter construct under control of the -3200-bp to +54-bp COL1A2 promoter, whereas the expression of a dominant-negative c-Jun construct abolished this effect. Northern blotting showed that overexpression of c-Jun enhanced the ability of halofuginone to reduce collagen alpha2(I) messenger RNA levels in fibroblasts, whereas expression of the dominant-negative c-Jun abolished this effect. Topical administration of a halofuginone-containing cream for 20 days to TSK mice, which spontaneously develop dermal fibrosis, greatly increased the phosphorylated form of c-Jun in the skin; this was followed by, a decrease in skin thickness and type I collagen messenger RNA expression. Conclusion. Our findings illustrate the powerful down-regulatory property of c-Jun toward type I collagen and establish that halofuginone exerts its effect on collagen synthesis in a c-jun-dependent manner.