Modeling of bone remodeling by three-dimensional co-culture of mouse embryonic stem cell-derived osteoblasts and osteoclast precursors (#70)
Bone is maintained by several specialized cells through a process called bone remodeling. In this process, osteoblasts and hematopoietic lineage-derived osteoclasts communicate with each other through bi-directional signaling that regulates their functions. Osteoclastogenesis and the activation of osteoclasts largely depend on their interactions with osteoblastic cells through direct contact and paracrine factors; functional osteoblasts have the ability not only to secrete bone matrix but also to support osteoclastogenesis in vivo. We have developed culture strategies for differentiating pluripotent stem cells (PSCs) into osteoblastic cells. A bone synthesis-related function of PSC-derived osteoblasts has been extensively demonstrated both in vitro and in vivo, whereas multicellular interactions remain unexplored. Here we developed a three-dimensional (3D) co-culture system allowing osteoblast differentiation and subsequent interaction with osteoclast precursors. First, 2.3-kb Col1a1-GFP mouse embryonic stem cells (ESCs) were differentiated into osteoblastic cells within atelocollagen porous scaffolds by using small molecule inducers under defined conditions. Gene expression, protein expression and histological analyses revealed the formation of calcified structures containing osteoblast/osteocyte populations. Importantly, they highly expressed Rankl, a key stimulator of osteoclastogenesis. Osteoclast precursors from bone marrows of Rank-Cre;Rosa26-tdTomato mice were then seeded in the 3D culture system. After 1 week of the co-culture, TRAP-positive multinucleated cells were detected. Observation by two-photon microscopy revealed direct contact between green fluorescent protein-positive osteoblastic cells and tdTomato-positive osteoclastic cells as well as mature bone-resorbing osteoclasts. The present strategy may thus provide a system that reconstitutes the bone microenvironment in a 3D manner. Our data further suggest that a PSC-derived osteogenic population has not just osteogenic capacity to form bone, but also supports osteoclastogenesis. This strategy will allow us to reproduce and visualize the bone remodeling process in vitro in real-time and thus potentially elucidate unknown intercellular events during the bone remodeling process.