Gadolinium-induced fibrosis is counter-regulated by CCN3 in human dermal fibroblasts: a model for potential treatment of nephrogenic systemic fibrosis

Abstract

We recently show that CCN3 is a counter-regulatory molecule for the pro-fibrotic protein CCN2, and a potentially novel fibrosis therapy. The goal of this study was to assess the role of CCN3 in fibroproliferative/fibrotic responses in human dermal fibroblasts exposed to Omniscan, one of the gadolinium-based contrast agents associated with development of nephrogenic systemic fibrosis (NSF) a rare but life-threatening disease thought to be complication of NMR diagnostics in renal impaired patients. Human dermal fibroblasts were exposed to Omniscan; or to platelet-derived growth factor (PDGF) and transforming growth factor-β (TGF-β) as controls. Proliferation was assessed along with matrix metalloproteinase-1, tissue inhibitor of metalloproteinases-1 and type 1 procollagen in the absence and presence of CCN3. In parallel, CCN3 production was assessed in control and Omniscan-treated cells. The results showed that PDGF stimulated fibroblast proliferation, production of Timp-1 and MMP-1 whereas exogenous CCN3 inhibited, in a dose response manner, cell proliferation (approx. 50 % max.) and production of MMP-1 (approx 35 % max.) but had little effect on TIMP-1. TGF-β stimulated type 1 procollagen production but not proliferation, Timp-1 or MMP-1 compared to non-TGF-ß treated control cells, and CCN3 treatment blocked (approx. 80 % max.) this up-regulation. Interestingly, untreated, control fibroblasts produced high constitutive levels of CCN3 and concentrations of Omniscan that induced fibroproliferative/fibrogenic changes in dermal fibroblasts correspondingly suppressed CCN3 production. The use of PDGF and TGF-β as positive controls, and the study of differential responses, including that to Omniscan itself, provide the first evidence for a role of fibroblast-derived CCN3 as an endogenous regulator of pro-fibrotic changes, elucidating possible mechanism(s). In conclusion, these data support our hypothesis of a role for fibroblast-derived CCN3 as an endogenous regulator of pro-fibrotic changes in these cells, and suggest that CCN3 may be an important regulatory molecule in NSF and a target for treatment in this and other fibrotic diseases.

Fig. 1

Effects of CCN3 on human dermal fibroblast proliferation and production of MMP-1, TIMP-1 and type I procollagen in response to PDGF or TGF-β. Proliferation was assessed by cell counting; MMP-1, TIMP-1 and Type I procollagen by ELISA. Values shown are means and standard errors based on fibroblast isolates from 5 different normal donors for proliferation and MMP-1 assays, and 3 donors for TIMP-1 and type I procollagen assays. Statistical significance of the data was assessed by ANOVA, followed by paired-group comparisons. *indicates statistically significant increase compared to controls at p < 0.05 level. **indicates statistically significant decrease compared to positive control at p < 0.05 level

Fig. 2

Effects of CCN3 on human dermal fibroblast proliferation and production of MMP-1, in response to Omniscan. Upper panel: Proliferation: Values shown are means and standard errors based on separate experiments with fibroblast isolates from five different individuals. Middle panel. MMP-1 assessed by Western blotting. Insert: Example Western blot: lane 1 = control; lane 2 = Omniscan; lane 3 = Omniscan and CCN3. Lower panel. MMP-1 assessed by ELISA. Values shown are means and standard errors based separate experiments with fibroblast isolates from three different individuals. For all analysis, statistical significance of the data was assessed by ANOVA, followed by paired-group comparisons. *indicates statistically significant increase compared to control at p < 0.05 level. **indicates statistically significant decrease compared to positive control at p < 0.05 level

Fig. 3

Effects of Omniscan on CCN3 production by dermal fibroblasts. Upper panel. Proliferation: Values shown are means and standard errors based on separate experiments with fibroblast isolates from five different individuals. Lower panel. Values for CCN3 were determined by ELISA and are mean and standard errors based on the same 5 separate donors. For both panels, statistical significance of the data was assessed by ANOVA, followed by paired-group comparisons. *indicates statistically significant change compared to untreated control at p < 0.05 level

Fig. 4

Effects of Omniscan on CCN2 production by dermal fibroblasts. Values for CCN2 were determined by ELISA and are mean and standard deviations based on 5 separate donors