Background

Bone formation is a fundamental process required for the development of the skeleton, for the progression of skeletal growth and to maintain bone structure throughout life as the skeleton is constantly remodeled. Bone formation is a process that occurs by three steps: (1) deposition of collagen-rich osteoid, (2) mineralization of that osteoid which occurs by deposition of bioapatite crystals around the collagen fibres, and (3) maturation of the mineralized bone substance (Blank and Sims, 2019). These three steps are mediated by two cell types derived from a common progenitor. Osteoblasts deposit osteoid and initiate mineralization, and as the osteoid is deposited, some osteoblasts become embedded within the bone matrix and differentiate into osteocytes (Dallas and Bonewald, 2010); these embedded cells also control the extent and nature of mineralization of bone (Vrahnas et al., 2019).

Identifying agents that can stimulate bone formation is critical for understanding the biology of the skeleton, and for developing new agents that can be used to promote bone formation in conditions of bone fragility such as osteoporosis, osteogenesis imperfecta, and osteomalacia, and to develop agents that can promote fracture healing or to stimulate bone formation in surgical interventions. Typically, the simplest system to test such agents is to use primary cultured osteoblasts (Orriss et al., 2014), or to differentiate stromal cell lines such as MC-3T3-E1 (Quarles et al., 1992) and Kusa 4b10 cells (Allan et al., 2003) in osteoblastogenic media. These systems are commonly used to study osteoblast differentiation by assessing effects of various agents on mRNA levels of osteoblast marker genes, alkaline phosphatase enzyme activity, and mineral deposition (Allan et al., 2003; McGregor et al., 2010; Walker et al., 2010). However, these systems do not always reflect the in vivo response. A clear example is the contrast between early in vivo works showing that leukemia inhibitory factor stimulates bone formation (Metcalf and Gearing, 1989; Cornish et al., 1993), and in vitro tests in which leukemia inhibitory factor could both stimulate and inhibit osteoblast differentiation (Malaval et al., 1995; Malaval and Aubin, 2001; Malaval et al., 2005; Falconi and Aubin, 2007).

In this protocol, we describe the use of a calvarial injection model which tests the ability of agents to stimulate bone formation in vivo. This method requires only small amounts of the stimulatory agent, and was first described by Cornish (Cornish et al., 1993) to resolve the controversy of whether leukemia inhibitor factor could promote bone formation. We have adapted the Cornish method to include calcein labelling to allow measurement of bone formation, and we have made use of it to test a range of cytokines including oncostatin M (Walker et al., 2010), cardiotrophin-1 (Walker et al., 2008), and most recently IL-6 acting through its soluble receptor (McGregor et al., 2019). We have also used this method to determine whether responses to cytokines are modified in mice with a cell-specific deletion of gp130, the common receptor used by these cytokines (Johnson et al., 2014). We provide here a full description of how to carry out the in vivo protocol, and how to embed and section tissues using undecalcified histology techniques; micro-computed tomography can also be used for assessment, but this does not allow measurement of bone formation rate using calcein labels. An abbreviated form of this method using single calvarial injections or two days of calvarial injections can also be used to assess gene responses elicited by cytokines in vivo by Western blot (Walker et al., unpublished) and effects on protein expression by immunohistochemistry (Walker et al., 2010).