JNK inhibitor CC-930 reduces fibrosis in a murine model of Nf1-deficient fracture repair

Abstract

Tibial pseudarthrosis often features deficient bone formation, excessive bone resorption, and extensive pathological fibrosis, particularly in individuals with Neurofibromatosis type I (NF1). It was hypothesized that overactive NF1-Ras-JNK signalling may underlie the pathological fibrosis, and that this could be treated via a JNK antagonist. CC-930, a small molecule JNK inhibitor, was trialed in closed fractures in wild type mice CC-930 (25 mg/kg/twice daily) was dosed throughout fracture healing (D2–21) and during the latter stages of repair (D11–21). All fractures healed by D21, regardless of treatment, with some of the CC-930 (D11–21) treatment group showing early bridging. CC-930 (D11–21) was tested in an Nf1-null fracture model where Nf1 was inactivated by Ad-Cre virus injection in Nf1flox/flox mice; these mice also possessed a Cre-responsive tdTomato transgene. CC-930 resulted in a significant decrease in non-unions (93% vehicle vs. 64% CC-930, p<0.01). Local treatment with the bone anabolic rhBMP-2 (10μg) increased union and callus bone volume, but also increased the fibrotic tissue at the fracture site. Fractures treated with a combination of rhBMP-2 (10μg) and CC-930 were all partially or fully bridged by D21 (p<0.01 vs. vehicle) and there was a significant decrease in fibrosis vs. rhBMP-2 alone (p<0.01). In untreated Nf1-null fractures, the tdTomato transgene was expressed in fibrous tissue at the fracture site, but not in the newly forming bone. These data suggest that JNK inhibition may be an effective therapeutic approach for reducing pathological fibrosis in NF1 tibial pseudarthrosis, however other adjunctive strategies may be required to augment bone formation.