A mechanistic understanding of the role of Col1a2 (the third chain of the type I collagen trimer) in brittle bone disease

Context Brittle bone disease (osteogenesis imperfecta, OI) occurs in 1 in 10,000 individuals. Symptoms range from mild to lethal before birth, but commonly include easily-broken bones, loose joints, and weak muscles. Most cases are caused by mutations in one of the two genes for type I collagen, the major structural component of tissues and organs. Collagen precursors form trimers and assemble into string-like fibrils outside the cell. Type I collagen trimers comprise two copies of the collagen a1 chain and one a2 chain, encoded by the Col1a1 and Col1a2 genes respectively. A mouse model of OI, termed ‘oim’ (osteogenesis imperfecta murine) which is widely used for preclinical testing, carries a Col1a2 mutation such that only a1 homotrimers form in mice with two mutant genes. Carrying one copy of the defective gene produces mild symptoms, whereas two copies cause spontaneous bone fractures. We have shown that bone fragility occurs due to lack of normal (a1)2 a2 heterotrimers AND the presence of the defective oim collagen gene; whereas inactivation of both copies of Col1a2 does not substantially alter bone structure or biomechanics. Challenge Although the oim model is widely used for preclinical testing of treatments for OI, including anabolic, anti-resorptive and stem cell therapies, the pathogenetic mechanism is unknown. We showed that inactivation of both copies of the second collagen gene (Col1a2) in mice does not substantially alter bone structure or mechanical function. However, one copy of the oim mutation in Col1a2 is more detrimental when the other Col1a2 gene is inactivated, raising the question as to whether normal Col1a2 could counteract other OI-causing mutations. Intriguingly, the abnormal a2 chain precursor protein was not detected in the oim model with two mutant copies of Col1a2, despite the presence of the messenger RNA required to make the mutant protein. Nor was there activation of cellular stress pathways, as seen in other cases with abnormal collagen precursors. This suggests that mechanisms active before or during protein production could play a role in the development of OI. Aims and objectives We aim to determine whether OI is caused by an unusual pre-translational defect in the oim model, and whether other OI mutations also have defective protein translation, as well as the role of normal Col1a2 in counteracting bone fragility. This will potentially reveal important new mechanisms of OI pathogenesis and shed light on the suitability of the oim mouse model in drug testing. The objectives are: 1) To determine whether the mutant Col1a2 gene is actually made into protein in the cell. 2) The understand the rate of collagen synthesis, fibril assembly, protein composition, and cellular function in the oim mutant model. 3) To establish whether lack of a normal a2 collagen chain and/or the presence of homotrimeric type I collagen can exacerbate OI caused by other type I collagen mutations. Applications and benefits There is no curative therapy for OI. We therefore need a reliable model for early-stage testing of the wide range of potential OI treatments. This work will establish the extent to which the osteogenesis imperfecta murine (oim) model is a suitable system for such testing. Furthermore, if as we hypothesise, the mutation influences pathways preceding protein production, or protein production itself, this opens up the opportunity for new approaches targeting these pathways as future frontline treatments for OI.