Breaking the rules for bone adaptation to mechanical loading.

The ability of bone to respond to mechanical stimuli has been known for over a century; however, it has only been in the past several decades that great gains have been made in terms of understanding the factors that influence this response. For instance, we now know that 1) bone preferentially responds to dynamic rather than static stimuli, 2) only short durations of loading are necessary to initiate an adaptive response, and 3) bone cells accommodate to customary mechanical loading environments (5). Applying these so-called “rules” for bone adaptation to mechanical loading, scientists and clinicians have been able to predict with some certainty the bone response that may be generated with a particular loading or exercise regimen (7). In particular, they have been able to utilize the principle of site specificity. The adaptive response of bone to mechanical loading is highly site specific. This is clearly evident on the whole bone (organ) level, with only the bones that are actually loaded undergoing adaptation. A clear example of this is in racquetsport players, wherein the playing or racquet arm has significantly greater bone mass and size than the contralateral, nonplaying arm (1). However, the site-specific nature of bone adaptation to mechanical loading can be localized even further than to the individual bone level. Bone experiences internal strain when mechanically loaded. From an engineering standpoint, strain refers to the change in length of a bone when load is applied. Although it is a unitless value, because it is small for bone it is often expressed in terms of microstrain ( ). As long bones are curved, they bend when axially loaded. This results in exposure of different tissue-level regions within the bone cross section to different levels of microstrain. Only those regions within the individual loaded bone that experience sufficient microstrain adapt. This has been demonstrated most evidently using the rodent ulna axial compression model, wherein tissue-level bone adaptation closely matches the tissue-level microstrain distribution (8). The site-specific depositing of new bone is functionally important. It adds new bone and increases bone strength where it is needed most, in the direction of loading, while not overtly increasing the overall weight of the bone. However, in this issue of the Journal of Applied Physiology, Zhang et al. (11) provide evidence that the site-specific response of bone to mechanical loading may not be so simplistic. Applying lowlevel, compressive loading to the proximal tibial epiphysis of mice, they induced bone adaptation at a distant, nonloaded site (4-mm distal on the periosteal surface of the tibial diaphysis). That is, they found mechanical loading to stimulate bone formation at a site distant from the site of loading and distant from a site of significant microstrain. Thus their data provide evidence for potential non-site-specific and non-strain-dependent bone adaptation in response to mechanical loading. As with most provocative, novel studies, the work of Zhang et al. (11) generates many more questions than answers. The first and most obvious question is how can load applied to one site induce bone adaptation at a site that is apparently not experiencing any appreciable load or microstrain? An answer to this question was theorized by Zhang et al., and it may lie in how mechanical signals applied to the skeleton are converted into a bone cell response: a process referred to as mechanotransduction. Bone is a porous tissue consisting of a fluid phase, a solid matrix, and cells. It is a common belief that mechanotransduction in the skeleton involves the movement of the fluid phase in relation to the solid matrix, which subsequently stimulates “detector” cells and triggers a cascade of adaptive molecular events (2). It is plausible that the epiphyseal loading introduced by Zhang et al. (11) resulted in intracortical fluid flow via load-induced alterations in intramedullary pressure (Fig. 1). This is an explanatory mechanism that the authors are actively exploring (10). The second question to ask is what is the effect of epiphyseal loading on both the loaded site and even more distant sites than the one assessed? Zhang et al. (11) report that epiphyseal loading stimulated periosteal (cortical) bone adaptation at a site 4-mm distant to the loading site. Based on their previous study loading the proximal ulna (10), one would expect that loading of the proximal tibia also generated localized cortical and trabecular bone adaptation at the loading site (epiphysis) and that this response was not related to some form of regional acceleratory phenomenon (trauma). Similarly, if load-induced alteration in intramedullary pressure is the mechanism for the adaptive response observed by Zhang et al. (11), one may expect adaptation to also occur at even more distant sites as the induced pressure change is propagated throughout the entire intramedullary cavity. Indeed, preliminary evidence indicates that this may be the case (12). The third question to ask is why does the apparent non-sitespecific bone adaptation to epiphyseal loading exhibit an inverse relationship to increasing loading frequency? Zhang et al. (11) found bone adaptation to be greatest at a loading frequency of 5 Hz, less significant at 10 Hz, and nonexistent at 15 Hz. This finding contravenes those of conventional loading studies that consistently show a positive relationship between