Genetics and molecular biology of melorheostosis

Eunice Kennedy Shriver National Institute Of Child Health & Human Development

Investigators

Linked publications & trials

Abstract

Melorheostosis is a sporadic dysostosis characterized by asymmetric bone overgrowth and functional impairment. The classic form of melorheostosis has radiographs characterized by a dripping candle wax appearance. Prior to our project the genetic cause of melorheostosis was unknown; somatic mosaicism was postulated because there was no parent to child transmission and mutations were not detected in leukocytes. A collaboration of investigators in NICHD, NIAMS and the Institute of Osteology in Vienna came together to investigate the genetics of melorheostosis. To increase the ability of WES sequencing to detect causative mutations, 15 melorheostosis patients underwent biopsies of both affected and contralateral unaffected bone, thus allowing low percent mosaicism in the affected tissue to be detected. We identified somatic mosaicism for MAP2K1 and SMAD3 mutations in classical and endosteal melorheostosis, respectively. Eight of 15 biopsied patients, all of whom had dripping candle wax pattern on radiographs, had somatic mosaicism for mutations at MAP2K1 p.K57 or p.Q56 in the NRD (Kang, Jha, Deng et al., 2018). The MEK1 substitutions activated ERK1/2 MAPK, enhanced osteoblast proliferation and delayed osteoblast differentiation and ECM mineralization. These findings underlie the increased osteoid accumulation, increased cellularity and remodeling identified on histologic examination in deeper regions of affected bone. The outer zone of the candle wax lesion is composed of multi-layered primary lamellae formed by periosteal apposition, which confers hardness to the overgrowth bone. Increased vascularity was present in affected bone and overlying skin: cortical bone vascularity was increased; overlying skin was erythematous, with increased thickness of vessel walls and expression of transcripts related to angiogenesis. MAP2K1 mosaicism in overlying skin is diagnostic in many cases. Four of 15 biopsied patients, all of whom had endosteal overgrowth on radiographs, had somatic mutations in SMAD3 (p.S264) in affected, but not unaffected, bone (Kang, Jha, Ivovic et al., 2020). qBEI revealed higher matrix mineralization in affected than unaffected bone. The SMAD3 mutations increased canonical TGF-/SMAD signaling and inhibited melorheostotic osteoblast proliferation, indicating gain-of-function. The SMAD3 mutations stimulated osteoblast differentiation and ECM mineralization in affected cells; these pathways were inhibited when osteogenesis was driven in the presence of BMP2. RNA-Seq-based transcriptome profiling confirmed the SMAD3 mutation significantly influenced TGF- pathway and ossification-related processes, and uncovered insights into pathophysiology, including ECM organization, cell growth, and immune response-related pathways. Overall, we demonstrated that melorheostosis is genetically heterogeneous, with somatic mutations in two genes responsible for distinct radiographic and clinical patterns through different molecular and cellular mechanisms. Our current effort focuses on different aspects of melorheostosis in MAP2K1 and SMAD3 cases. For MAP2K1, we are studying the secretome of melorheostosis lesions. Since some lesions have a very low percent of mutant cells, it is likely that they secrete proteins that influence adjacent non-mutant cells to act in concert with mutant cells. The secretome of conditioned media from both unaffected and affected cells is being compared to identify candidate factors. The pandemic disrupted these studies and we are anticipating their early resumption. We have studied the functional characteristics of a SMAD3 mutation identified in a patient who does not have melorheostosis, but has a germ-line skeletal disorder. The resulting substitution (p.I67S) has not been previously reported. We conducted overexpression studies to compare this mutation to the melorheostosis mutation (p.S247Y) and found reduced SMAD3 stability and TGFb activation vs the gain-of-function melorheostosis mutation. For SMAD3, we have generated a knock-in mouse that carries the patient mutation. When activated in the germ line the mutation is lethal. Currently we are activating the mutation at different stages of osteoblast development to generate a viable melorheostosis model.