Molecular biology of Paget’s Disease of Bone: role of p62 and novel genes

Paget’s disease of bone (PDB) is an age-related metabolic bone disease characterized by focal lesions of increased bone resorption and formation, eventually leading to bone deformities. The cause of PDB and the mechanisms that give rise to focal lesions are yet to be understood, but findings suggest that the disease is driven by aberrant, highly nucleated, osteoclasts (OCs). In recent years evidences of a genetic involvement were found: mutations in UBA domain of SQSTM1, which encodes for p62, have been reported in both familial and sporadic cases of PDB (P392L most commonly). Although, their actual pathogenicity has been controversial in experimental studies. Moreover, mutations only involve a part of PDB cases and, although some novel genes have been more recently associated to PDB (e.g.ZNF687), the genetic background of PDB remains in part unknown. In an attempt to establish an experimental model of PDB and better understand p62 role in the disease, we compared two genetically modified murine models, systemic p62 knock-out (p62KO) and mutated p62-P394L (P394L) mice. To further characterize the genetic background of PDB we investigated PDB-associated genes and novel genes in SQSTM1-negative patients. In vitro bone marrow-derived macrophages (BMMs) showed a reduced RANKL-induced osteoclastogenesis (OCgenesis) in p62KO-mice, also seen by TRAP staining on bone sections. BMMs of P394L mice showed a higher sensitivity to RANKL and an increase in OC size and number of nuclei, resembling PDB. Such alterations did not result in a bone phenotype at 6 months of age in either model. However, we found that, with ageing, 47% of P394L mice do develop focal osteolytic lesions. Surprisingly, 78% of p62KO mice developed severe lesions. Although further histological characterization is needed, both animals showed focal, PDB-like, osteolytic/sclerotic features. In vitro analysis of aged p62KO BMMs no longer showed a reduction in OCgenesis potential. Taken together, our findings suggest that p62 mutations in UBA cause a loss of function mechanism in PDB, further exacerbated by total loss of the protein. In support of our hypothesis, proteomics showed that aged p62KO and P394L BMMs are primed for OCgenesis and both present similar expression profiles. Investigating possible molecular mechanisms, we found that UBA-dependent p62 functions of autophagy and NF-κB signalling are not altered in either p62KO and P394L cells. Genetic analysis of 34 patients was performed on genes SQSTM1, TNFRSF11A, VCP, ZNF687 and two variants on TM7SF4 and RIN3. The majority of our cohort was negative for rare mutations on such genes, apart from three cases, carrying TNFRSF11A_M566L, SQSTM1_S275N and ZNF687_P937R. Finally, taking advantage of a large pedigree of a severely affected PDB family we performed Exome NGS to identify novel causal gene. Analysis of impact, familial segregation and allelic frequency identified a novel mutation: PFN1_D107Rfs*3, that causes loss of C-terminal domain of PFN1. This gene encodes for profilin1, a regulator of actin polymerization and cell motility. Given the essential role of cytoskeleton reorganization in OCs biology and the previous findings of bone focal deformities in PFN1 OC-conditional knock-out mice, we started investigating its potential pathogenicity. Silencing of PFN1 in murine BMMs resulted in larger OCs with a higher number of nuclei and increased resorption activity. Screening of PFN1 mutations on other PDB cases is ongoing. Overall, our data demonstrated that both p62 depletion and P394L mutation and are sufficient to cause PDB-like disease in mice. Based on the available molecular data the likely role of p62 in PDB is UBA-dependent but autophagy and NF-κB independent. The genetic background of PDB remains largely unknown, as demonstrated by our screening. Finally, the gene discovery part of the project allowed to identify a likely novel gene for PDB, associated with an early onset and aggressive phenotype.