The role of extrinsic and intrinsic outer-vascular mechanics on sprouting angiogenesis during early bone regeneration: an in silico study
Sprouting angiogenesis is a key process in early bone regeneration. During sprouting, endothelial cells (ECs) are known to be sensitive and generate specific responses to local mechanical stimuli. Externally applied loads (extrinsic loads) impact tissue deformations and affect sprout patterning (Rosenfeld et al, Proc Natl Acad Sci USA, 2016). On a cell-cell communication level, ECs interact mechanically with stromal cells (SCs) through cell traction forces (intrinsic loads). How these two distinct mechanical cues impact angiogenesis during early bone regeneration remains largely unknown. Here we aim to investigate in silico the relative role of extrinsic and intrinsic mechanical signals on early sprout patterning. Multi-scale computer models of mechanics-driven sprouting angiogenesis and SCs organization were developed to represent the healing region of a mouse osteotomy subjected to physiological loading and stabilized with a rigid or a semirigid fixator (Borgiani et al, J Bone Miner Res, 2019). Finite element models at the tissue scale, to compute mechanical strains, were iteratively coupled to agent-based models at the cell scale, describing ECs and SCs response to local mechanical cues (e.g. durotaxis). Vessel sprouts were predicted to align along preferential directions in agreement with dedicated in vivo experiments. Furthermore, under reduced mechanical stability (semirigid) a lack of sprouts was experimentally observed at the core of the osteotomy that was also predicted by the in silico analyses. To our knowledge, this is the first in silico model of early sprout patterning comprising both cell intrinsic forces and extrinsic loads. Collectively, our results suggest that high extrinsic loads might overrule the mechanical communication on a cell-cell level. Future studies are needed to investigate the influence of further mechano-biological interactions. Acknowledgements: This work was supported by grant CRC 1444 from the DFG.