A novel nonosteocytic regulatory mechanism of bone modeling

Osteocytes, cells forming an elaborate network within the bones of most vertebrate taxa, are thought to be the master regulators of bone modeling, a process of coordinated, local bone-tissue deposition and removal that keeps bone strains at safe levels throughout life. Neoteleost fish, however, lack osteocytes and yet are known to be capable of bone modeling, although no osteocyte-independent modeling regulatory mechanism has so far been described. Here, we characterize a novel, to our knowledge, bone-modeling regulatory mechanism in a fish species (medaka), showing that although lacking osteocytes (i.e., internal mechanosensors), when loaded, medaka bones model in mechanically directed ways, successfully reducing high tissue strains. We establish that as in mammals, modeling in medaka is regulated by the SOST gene, demonstrating a mechanistic link between skeletal loading, SOST down-regulation, and intense bone deposition. However, whereas mammalian SOST is expressed almost exclusively by osteocytes, in both medaka and zebrafish (a species with osteocytic bones), SOST is expressed by a variety of nonosteocytic cells, none of which reside within the bone bulk. These findings argue that in fishes (and perhaps other vertebrates), nonosteocytic skeletal cells are both sensors and responders, shouldering duties believed exclusive to osteocytes. This previously unrecognized, SOST-dependent, osteocyte-independent mechanism challenges current paradigms of osteocyte exclusivity in bone-modeling regulation, suggesting the existence of multivariate feedback networks in bone modeling—perhaps also in mammalian bones—and thus arguing for the possibility of untapped potential for cell targets in bone therapeutics.Bone’s ability to change its morphology in response to load is widely attributed to osteocytes. A study of fish shows that bone can respond to load even in the absence of osteocytes, using a molecular mechanism that is conserved across vertebrates, albeit with different cellular effectors.Author summary: Bone is a “smart” tissue, able to sense loads within its bulk and change its morphology when needed by a process named bone modeling. This process is carried out by bone-depositing cells (osteoblasts) and bone-resorbing cells (osteoclasts) and is regulated by osteocytes—cells that reside in small cavities within the bone tissue. Osteocytes are considered to function as mechanosensors, detecting areas of high loads that require modeling, and master regulators of osteoblasts and osteoclasts. Curiously, evolutionarily advanced fish do not have osteocytes in their bones, although more basal fish and all other bony vertebrates have them. In this paper, we show how the bones of advanced fish can respond to load in a mechanically efficient way despite the absence of osteocytes. We describe the molecular mechanism, which we found to be the same as in all other vertebrates; however, we show that the cellular effectors are different. The protein sclerostin, which is produced by osteocytes in mammals and is a potent suppressor of bone building by osteoblasts, is produced by a variety of nonosteocytic cells in medaka and zebrafish, and nonosteocytic skeletal cells serve as sensors and responders in these species. These results challenge current paradigms of osteocyte exclusivity in the regulation of bone modeling.