In this study we examined the relationship between circulating sclerostin levels and bone in subjects with chronic SCI (more than 2 years post-injury). We found that sclerostin levels increase significantly with age. These results are in agreement with a recent report that circulating sclerostin levels increase with age, and are positively associated with bone density in elderly men and women (16
). When considering the SCI subjects only, age-adjusted sclerostin levels were significantly reduced in people using a wheelchair compared to those who walk. Similarly, age-adjusted sclerostin levels were significantly reduced in people with SCI who use a wheelchair compared to people without SCI. Subjects with SCI who use a wheelchair had lower bone density at sublesional skeletal sites (distal femur and proximal tibia) compared to subjects with SCI who walk. We found no difference in age-adjusted sclerostin or bone density when comparing subjects with SCI who walk to subjects without SCI. Because there are few subjects in each of these 2 groups, these findings should be confirmed in a larger study. However, this suggests that, from a bone perspective, people with SCI who walk are similar to people without SCI.
For people with SCI, sclerostin levels were significantly associated with bone mineral density at the knee (distal femur, proximal tibia) but not the radius. For the majority of subjects with SCI in this study, the knee but not the radius, was subjected to partial or total mechanical unloading due to paralysis. This suggests that in the case of complete SCI with associated severe osteoporosis, local reductions in sclerostin production in osteoporotic sublesional bones can be meaningfully detected in circulating sclerostin assays. It is possible that in less severe SCI (such as incomplete SCI with preservation of ambulation) sclerostin levels are changed regionally to a lesser degree than seen in complete SCI, yet when assessed systemically these changes are not detected. This should be explored further in animal models where local levels of sclerostin within the bone microenvironment can be compared across injury severities.
We report lower, not higher, sclerostin levels associated with SCI and wheelchair use. We speculate that our findings are due to the fact that we studied chronic SCI. Prior studies have attempted to characterize bone loss following SCI (17
). Based on these reports, bone loss is thought to occur in 2 phases: 1) rapid, acute bone loss that plateaus somewhere between 18–24 months post-injury and 2) chronic, ongoing bone loss that is more gradual in nature. In the chronic phase of SCI, the majority of bone already has been resorbed and it is likely that ongoing bone loss is more gradual. Based on a rodent model of SCI-induced osteoporosis (22
), nearly all of the trabeculae in the sublesional long bones have been resorbed after injury with extensive cortical thinning. The osteocytes that produce sclerostin are found in trabecular and cortical bone. Therefore, dramatic sublesional bone loss would result in fewer sclerostin-producing cells and a lower basal level of sclerostin. Our findings of reduced sclerostin levels may be reflective of the severity of bone loss that occurs in chronic SCI. This suggests that unloading leads to elevated sclerostin levels during the acute phase, which inhibits bone formation by suppressing osteoblastic differentiation and/or function (23
). If inhibition of bone formation proceeds long term, without reintroduction of mechanical loading, extreme bone loss occurs. In severe osteoporosis, fewer bone cells exist to produce sclerostin, and sclerostin levels fall to levels lower than those seen under normal conditions. These findings suggest that, in chronic SCI, circulating sclerostin is a biomarker
of osteoporosis severity, not a mediator of ongoing bone loss.
Sclerostin, encoded by the sost
gene, is a potent inhibitor of bone growth (5
). Several elegant animal studies have shown sclerostin levels are inversely proportional to bone mass (5
) and that production of sclerostin by osteocytes is dramatically reduced by mechanical loading in rodents (6
). The currently accepted paradigm for the Wnt signaling pathway states that Wnt binds to a co-receptor complex involving Frizzled receptor and low-density lipoprotein receptor-related protein (LRP)-5 or LRP-6, both present on osteoblasts. This binding stabilizes cytoplasmic β-catenin and causes it to translocate to the nucleus. Translocation of β-catenin, in turn, activates the transcription of genes that promote osteoblast proliferation, differentiation, and function, ultimately resulting in new bone formation. Several antagonists have been described that can inhibit this signaling pathway. For instance, molecules like secreted frizzled-related proteins, Wif (Wnt inhibitor factor), and Cerberus can bind Wnt and functionally block the pathway. Dickkopfs (Dkk1) and sclerostin, on the other hand, inhibit the Wnt pathway by preventing the formation of the Wnt- Frizzled-LRP5 complex either by promoting the internalization of the LRP5/6 co-receptor (Dkk1) or by competitive binding to LRP5 (sclerostin) (27
). These studies have established the central role of Wnt signaling antagonists in the pathogenesis of disuse osteoporosis and provide a basis for the regulation of bone responses to unloading via enhanced or reduced Wnt signaling due to mechanical stimulation or unloading, respectively.
Animal models clearly demonstrate elevated sclerostin levels in response to mechanical unloading that is reversed with reloading. These are short term studies that are thought to mimic the acute phase of SCI-induced bone loss. There are no animal studies reported in the literature that are long-term or mimic chronic SCI. Based on these reports and our results in chronic SCI, we propose a conceptual model of sclerostin mediated bone loss following SCI (). Mechanical unloading (paralysis) in acute SCI subjects causes greater sclerostin levels. This increase contributes to decreased bone formation and subsequent bone loss during the acute phase of SCI. The ability to walk (mechanical loading) modulates the response of bone to paralysis by causing a smaller increase in sclerostin levels, thereby partially protecting from bone loss. In the chronic phase, bone wasting results in lower sclerostin levels than those observed in the able bodied. This effect is due to the reduction of sclerostin-producing osteocytes in the osteoporotic bone. In the chronic phase, similar to the acute phase, the ability to walk partially protects from bone loss.
Conceptual Model of Sclerostin-Mediated Bone Loss Following SCI
Our novel finding of decreased sclerostin levels in extreme osteoporosis strongly supports a relatively narrow therapeutic window for targeting this pathway in disuse osteoporosis. Based on the literature supporting emerging therapeutic trials (29
), treatment with an anti-sclerostin antibody would be ineffective in chronic SCI when sclerostin levels already are suppressed. This suggests that the sclerostin pathway should be targeted immediately after SCI, before excessive bone loss occurs. Anti-sclerostin antibodies are being tested in post-menopausal osteoporosis, and may be effective in acute-SCI. We found that wheelchair use in chronic SCI is associated with low sclerostin levels and low BMD at SCI- specific sites. This suggests that physical therapy programs that reintroduce mechanical loading soon after SCI may effectively reduce or block sclerostin-mediated bone loss. These findings have important implications for rehabilitation following acute SCI.
The effect of bisphosphonate therapy on the relationship between sclerostin and bone is also unknown. In this study only 3 subjects were actively taking a bisphosphonate. Despite bisphosphonate treatment, each subject had very low bone density at the distal femur (0.42–0.73 grams/cm2) compared to the mean BMD (0.97 grams/cm2) at the distal femur in the group with no SCI. This suggests that this therapy was started to treat severe osteoporosis, not prevent bone loss. This also suggests that in the context of severe osteoporosis, as seen in our subjects with chronic SCI, bisphosphonate therapy has little impact on the relationship between sclerostin and bone. In fact, when excluding the 3 subjects taking a bisphosphonate, the relationships between bone density and sclerostin did not change. These observations need to be confirmed in a larger study and also assessed in the acute phase of SCI.
The optimal time frame for targeting sclerostin in disuse osteoporosis, either with medication or with physical therapy, is currently unknown. Future studies to measure sclerostin in acute SCI are needed to clarify pathogenesis and prior to the consideration of anti-sclerostin antibodies to treat bone loss in the acute phase of SCI.