Sundar Srinivasan, Ph.D.
Dept. of Orthopaedics & Sports Medicine
Orthopaedic Science Laboratories
"Modeling Bone Mechanotransduction as an Emergent Phenomenon"


ABSTRACT

The central focus of our laboratory is to understand how bone cells and tissues perceive and respond to mechanical stimuli and lack thereof.  For instance, we seek to understand how exercise makes bone bigger and stronger and why disuse accompanying bed rest or space flight causes bone mass to be lost.  While the response of bone to altered mechanical states is rapid, it is also occurs in a highly localized fashion (e.g., tennis player have bigger bones in playing vs non-playing arms).  Additionally, bone response to mechanical stimuli is focal even within a given bone and appears to occur, paradoxically, at sites of minimal strain magnitude.  More recently, we observed that simply inserting a 10-s unloaded rest-interval between load cycles transforms impotent cyclic loading regimens into stimuli capable of dramatically enhancing bone formation in the adult and aged skeletons.  While these results have attractive potential for application, the paradoxical bone responses at sites of minimal strain and counterintuitive osteogenic potency of rest-inserted loading highlight the general lack of knowledge of how the process of mechanotransduction functions within bone.  We proposed that exploring bone mechanotransduction as an emergent adaptive phenomenon might offer unique explanatory insights into how bone cells and tissues perceive and respond to an epigenetic factor critical to bone's form and function.  We have begun to explore this proposal by developing agent-based models of bone mechanotransduction, an approach suited for the analysis of general classes of complex adaptive systems.  Agent based models are unique in that they permit examination of how local, agent (or cell) level functions and interactions between functions gives rise to emergent properties at the global or network levels.  Our current agent based models examine signaling induced in bone cell networks by mechanical stimuli.  Our data indicate that when networked cells are exposed to cyclic stimuli, their collective signaling operates at extremely poor efficiencies.  In contrast, exposing networked cells to rest-inserted stimuli synchronizes, enhances and sustains signaling within the network.  As well, our models permit correlation and prediction of adaptive events that occur weeks downstream with signaling that is induced in bone cells by mechanical stimuli on the order of seconds.  In sum, our agent based models provide unique insights into mechanisms underlying the counterintuitive osteogenic potency of regimens such as rest-inserted loading and hold promise in the broader exploration of how mechanotransduction functions within bone.