ECTS2013 Poster Presentations Other diseases of bone and mineral metabolism (48 abstracts)
1University of Aberdeen, Aberdeen, UK; 2Institute of Genetic and Biomedical Research (IRGB), Milan, Italy; 3Instituto Clinico Humanitas IRCCS, Rozzano, Italy; 4University Childrens Hospital, Ulm, Germany.
Osteoclast-poor autosomal recessive osteopetrosis is characterised by susceptibility to fracture despite high bone mineral density as a consequence of an absence of osteoclasts. One of the 12 receptor activator of NF-κB (RANK) mutations associated with this condition is a frameshift mutation encoding a protein that is truncated within the extracellular, N-terminal domain (R110Pfs). We investigated the effect of this mutation on osteoclast formation, receptor localisation and signalling downstream of RANK and the possibility that translation of the C-terminal region of the protein from alternative translation initiation sites may explain any phenotypes observed.
In addition to published data, we observed that the in vitro osteoclast formation data from this patient was intriguing: there appeared some osteoclast formation in the absence of RANK ligand but then no further increase in osteoclast formation when RANK ligand was added. We generated seven myc-tagged expression constructs representing the N-terminal and potential downstream C-terminal products. When each of these proteins were overexpressed in Hela cells, immunostaining and confocal microscopy revealed that, whilst wildtype-RANK, was localised to the plasma membrane, golgi and discrete intracellular vesicles, the mutant proteins showed distinct subcellular locations, and were diffusely expressed throughout the cytosol. Western blot analysis confirmed the expected mass of each protein and since the R110Pfs product lacks a transmembrane domain we analysed the culture supernatant for, but were unable to detect, secreted protein. p65 translocation experiments demonstrated that the R110Pfs product does not support ligand-dependent or ligand-independent activation of NF-κB, whereas the putative C-terminal products of the alternative translation start sites (at positions 299 and 326 in RANK) induced only ligand-independent activation of NF-κB.
Taken together, these results strongly suggest that the in vitro RANKL-independent osteoclast phenotype observed in osteoclast cultures derived from this osteopetrosis patient can be explained by expression of C-terminal RANK causing ligand-independent activation of NF-κB.