In vivo loading impairs bone microstructure and endocortical bone formation in chickens
Isabela Vitienes1, 2, Mayumi Umebayashi1, 2, Nabhaan Farooqi1, 2, Catherine Julien1, 2, Ana Rentsch3, Tina Widowski3, Russell P. Main4, Bettina M. Willie1, 2
1Department of Biological and Biomedical Engineering; Faculty of Dental Medicine and Oral Health Sciences, McGill University, Montreal, Canada, 2Research Center, Shriners Hospital for Children-Canada, Montreal, Canada, 3Department of Animal Biosciences, University of Guelph, Guelph, Canada, 4Weldon School of Biomedical Engineering, Purdue University, West Lafayette, USA
In contrast to mammals, where calcium homeostasis is weighted towards bone accrual and maintenance, in female birds there is also a drive to resorb bone to release calcium for eggshell production. This model is therefore valuable to study bone adaptation dynamics in circumstances of high calcium demand. We aimed to investigate the effect of controlled in vivo loading on bone in young female chickens of different genotypes and exercise history. We hypothesized that, as seen in mammals, in vivo loading would lead to bone formation and improved bone density and microstructure, with similar effects across genotypes and a stronger effect in chickens with a more sedentary exercise history. Female chickens of two genetic strains (brown (B) and white (W) feathered) were raised in 3 housing systems allowing for sedentary, moderate, or high activity levels. At the onset of sexual maturity, the right tibiotarsus was subjected to controlled in vivo axial compressive loading while the left tibiotarsus served as a non-loaded internal control (once-daily, 2 weeks). Calcein was administered during the loading period to label newly formed bone. The effect of loading was assessed by static morphometry, histomorphometry and by measuring serum remodeling markers (TRAP5b, P1NP). Baseline serum remodeling markers were also measured from a separate group of nonloaded chickens.
Compared to the non-loaded limb, in the loaded limb we observed lower cortical area, total area, cortical area fraction, cortical thickness, trabecular thickness, and medullary bone volume fraction. P1NP was unaffected by loading whereas we saw a genetic strain dependent effect of loading on TRAP5b: only W loaded chickens had lower TRAP5b compared to W nonloaded controls. Periosteal bone formation (Ps.sl/Ps.Pm) was unaffected by loading whereas, at the endocortical surface of W chickens, there was a decrease in bone formation (Ec.sl/Ec.Pm) in the loaded compared to nonloaded limb.
Our results surprisingly show a negative effect of in vivo loading on mid-diaphyseal tibiotarsal microstructure and endocortical bone formation, despite there being decreased circulating TRAP5b in the loaded compared to non-loaded W chickens. Further investigation is warranted of bone microstructure at other regions, such as the proximal and digtal metaphyses, which may exhibit a different response to loading due to the local strain environment; microCT analyses at other anatomical sites and finite element modeling is ongoing.