The elementary fusion modalities of osteoclasts.

The last step of the osteoclast differentiation process is cell fusion. Most efforts to understand the fusion mechanism have focused on the identification of molecules involved in the fusion process. Surprisingly, the basic fusion modalities, which are well known for fusion of other cell types, are not known for the osteoclast. Here we show that osteoclast fusion partners are characterized by differences in mobility, nuclearity, and differentiation level. Our demonstration was based on time-laps videos of human osteoclast preparations from three donors where 656 fusion events were analyzed. Fusions between a mobile and an immobile partner were most frequent (62%), while fusion between two mobile (26%) or two immobile partners (12%) was less frequent (p<0.001). In general, the immobile fusion partner contained more nuclei than the mobile one (p<0.01). Furthermore, enrichment in nuclei of an osteoclast with three or more nuclei resulted from fusion with a mono-nucleated cell in 67% of the cases (p<0.001), while mono-nucleated cells fused with a multinucleated cell in 61% of the cases (p<0.05). This observation suggested that a more mature osteoclast prefers to fuse with a less mature pre-osteoclast. This hypothesis was supported by a nucleus-tracing approach in a co-culture of more and less differentiated pre-osteoclasts/osteoclasts. Furthermore, we found that osteoclast fusion proceeds through primarily two different types of cell contacts: phagocytic-cup and broad-contact-surfaces (>80% of all fusions). We conclude that osteoclasts most often gain nuclei by addition of one nucleus at a time, and that this nucleus is most often delivered by a moving cell to an immobile cell. These characteristics fit the in vivo observations where mono-nucleated precursors migrating from the bone marrow fuse with more mature osteoclasts sitting on the bone surface. They also fit the fusion modalities of other cell types.