- Top of page
- MATERIALS AND METHOD
Individuals with nerofibromatosis Type 1 (NF1) frequently suffer a spectrum of bone pathologies, such as abnormal skeletal development (scoliosis, congenital bowing, and congenital pseudoarthroses, etc), lower bone mineral density with increased fracture risk. These skeletal problems may result, in part, from abnormal osteoclastogenesis. Enhanced RAS/PI3K activity has been reported to contribute to abnormal osteoclastogenesis in Nf1 heterozygous (Nf1+/−) mice. However, the specific downstream pathways linked to NF1 abnormal osteoclastogenesis have not been defined. Our aim was to determine whether mammalian target of rapamycin (mTOR) was a key effector responsible for abnormal osteoclastogenesis in NF1. Primary osteoclast-like cells (OCLs) were cultured from Nf1 wild-type (Nf1+/+) and Nf1+/− mice. Compared to Nf1+/+ controls, there were 20% more OCLs induced from Nf1+/− mice. Nf1+/− OCLs were larger and contained more nuclei. Hyperactive mTOR signaling was detected in Nf1+/− OCLs. Inhibition of mTOR signaling by rapamycin in Nf1+/− OCLs abrogated abnormalities in cellular size and number. Moreover, we found that hyperactive mTOR signaling induced abnormal osteoclastogenesis major through hyper-proliferation. Our research suggests that neurofibromin directly regulates osteoclastogenesis through mTOR signaling pathway. Inhibiting mTOR may represent a viable strategy to treat NF1 bone diseases. © 2011 Orthopaedic Research Society Published by Wiley Periodicals, Inc. J Orthop Res 30:144–152, 2012
Neurofibromatosis Type 1 (NF1) is one of the most common autosomal dominant genetic disorders, affecting one in 3,500 individuals worldwide.1 NF1 is a tumor predisposition syndrome. The most common tumor pathologies in afflicted individuals are benign neurofibromas and malignant peripheral nerve sheath tumors (MPNSTs). Humans with NF1 may also suffer a spectrum of bone pathologies. Abnormal skeletal development occurs in 10–20% of NF1 individuals, resulting in short stature, scoliosis, congenital bowing, and congenital pseudoarthroses of long bones.2 Fractures of dysplastic bones occur in about 3% young NF1 children, especially boys.3 Orthopedic surgery to correct NF1 bone defects often fails due to non-union and pseudoarthroses of the healing bones.4 Several studies report lower bone mineral density (BMD), a known risk factor for fractures, in NF1 humans.5–7 Specifically, NF1 humans have a unique generalized skeletal dysplasia: when subjects with NF1 were separated in groups with and without a skeletal abnormality, those who did not have a skeletal abnormality still had statistically significant decreases in BMD compared with control subjects, although they were less pronounced than in those with osseous abnormalities.7 Unlike post-menopausal or male osteoporosis, NF1 osteopenia/osteoporosis has been diagnosed in children and adolescents, especially in those with skeletal abnormality; children with NF1 also have a general tendency toward osteopenia.7–9 These clinical data suggest an abnormal bone phenotype in NF1. Currently, there is limited treatment for NF1 bone pathologies.
Neurofibromin is expressed in osteoclasts.10 NF1 children have an increase in the urinary excretion of pyridinium crosslinks, reflecting increased bone resorption.11 High serum bone tartrate resistant acid phosphatase (TRAP) concentration and high serum calcium concentration have been reported to associate with lower BMD among the NF1 patients.12 Histomorphometric analysis reveals increased numbers of osteoclasts in biopsies from NF1 patients.13 Osteoclasts from NF1 patients are larger in size; their nuclei are more numerous; actin rings are more frequent; and the resorption pits are more numerous and larger.14 These data indicate that abnormal osteoclastogenesis in NF1 individuals, which may be in part responsible for the skeletal pathologies in NF1. However, the molecular mechanisms of abnormal osteoclastogenesis in NF1 are not quite clear.
Loss-of-function mutations in NF1 tumor suppressor gene underlie the disease. The gene product, neurofibromin acts as a Ras-GTPase activating protein (GAP). Neurofibromin converts RAS from its active GTP to its inactive GDP isoform. Loss-of-function of NF1 deregulates RAS signaling pathway, thus promoting abnormal cellular proliferation and tumorigenesis. Enhanced RAS activity has been reported in osteoclasts from Nf1+/− mice and NF1 patients. Initial therapies for NF1-associated tumors focused on the use of RAS inhibitors. RAS activation requires isoprenylation for membrane localization, which can be blocked by farnesyltransferase inhibitors. In preclinical studies, farnesyltransferase inhibitors can reduce cell proliferation dramatically both in NF1-deficient human and Nf1−/− mouse cells15–17; however, these agents have little effect on tumor growth in NF1 patients.18 The limited success of RAS inhibitors in NF1 clinical trials suggests that downstream pathways or other signaling pathways may play important roles in NF1 pathologies. Mammalian target of rapamycin (mTOR) is the downstream pathway for RAS. Recent research has shown that mTOR is constitutively activated in NF1-deficient primary cells (Nf1−/− mouse embryonic fibroblasts, Nf1−/− astrocytes, and NF1−/− Schwann cells), as well as in human NF1-associated tumors. NF1 tumor cells derived from NF1 patients are highly sensitive to the mTOR inhibitor, rapamycin.19, 20 These results suggest that mTOR pathway may be one of the critical downstream of NF1-related RAS signaling. mTOR also appears as an essential signaling pathway engaged in the stimulation of osteoclast survival.21 Tumor necrosis factor-a (TNF-a), receptor activator of nuclear factor kappa B ligand (RANKL) and macrophage-colony-stimulating factor (M-CSF) promote osteoclast survival by signaling through mTOR/S6K.22, 23 Furthermore, rapamycin, the inhibitor for mTOR, disturbs osteoclast differentiation and survival.21 Because mTOR is constitutively activated in Nf1 primary cells as well as in NF1 tumors,19, 20 we hypothesize that the mTOR pathway may be deregulated in Nf1-deficient osteoclasts and linked to abnormal osteoclastogenesis in NF1.
- Top of page
- MATERIALS AND METHOD
Osteoclasts are produced by the fusion of mononuclear precursors to form polykaryons, a critical step to perform resorption.29 Osteoclasts normally acquire up to eight nuclei before dying by apoptosis, probably as a result of exposure to the high extracellular concentration of calcium that occurs during bone resorption.30 The number of osteoclasts is a faithful index of bone resorption.31 Abnormal osteoclast formation is usually related to abnormal bone resorption, such as that in Paget's disease, where abundant hypernucleated osteoclasts correlate with extensive bone resorption.32 Here, we find that the number and size of osteoclasts, as well as the number of their nuclear profiles, increase in OCL culture from Nf1+/− mice at 10 weeks old. These results suggest that neurofibromin is required for ex vivo osteoclast formation and function in adult mice.
mTOR signaling is constitutively activated in neurofibromin deficient primary cells such as Nf1−/− mouse embryonic fibroblasts, Nf1−/− astrocytes, and NF1−/− Schwann cells, as well as in human NF1-associated tumors.19, 20 Our findings demonstrate that elevated mTOR signaling also occurs in primary Nf1+/− OCLs (Fig. 2). Elevated mTOR signaling in Nf1+/− OCLs may be due to abnormal activation of upstream signaling pathways. Indeed, our early research found that p21-GTP and Akt phosphorylation were elevated in Nf1+/− osteoclasts.33 Genetically restoring PI3K level corrected abnormal osteoclast formation and function in Nf1+/− mice.33 Because mTOR is the downstream target of PI3K/Akt, increased PI3K/Akt activity likely underlies hyperactive mTOR signaling in Nf1+/− osteoclasts. mTOR plays a critical role in cellular growth control (where cell growth refers to an increase in both cell size and cell number) by regulating cell translational machinery via effects on ribosomes and protein synthesis.34, 35 Since mTOR is activated in Nf1+/− osteoclasts, it is not surprise to detect giant osteoclasts and more osteoclast numbers in Nf1+/− mice (Fig. 1). Correcting osteoclast formation by rapamycin further confirms the critical role of mTOR in abnormal osteoclast formation in Nf1+/− mice (Fig. 4).
The molecular mechanisms under which mTOR regulates osteoclast formation and differentiation are still not quite clear. Abnormal osteoclast formation in Nf1+/− mice may be a consequence from hyper-proliferation and/or defective apoptosis. Our research demonstrates that inhibition of mTOR by rapamycin during the first 4-day culture, or during the entire 8-day culture period shows similar effect in reducing abnormal osteoclast formation (Fig. 5). These results suggest hyperactive mTOR signaling induces abnormal osteoclast formation mostly through hyper-proliferation in Nf1+/− mice. mTOR is a key regulator of cell proliferation, in part through regulation of cyclin D1.36 Cyclin D1 activates cyclin-dependent kinases 4 and 6 to regulate the passage of cells through the critical G1-S restriction point of the cell cycle.37, 38 Overexpression of cyclin D1 has been reported in NF1 Schwann cells as well as MPNSTs.39, 40 Inhibition of cyclin D1 expression inhibits growth of NF1 deficient tumor xenografts in mice.41 Based on these published data, we speculate that elevated mTOR may act via cyclin D1 to deregulate proliferation in Nf1+/− osteoclasts.
Published literatures show that rapamycin significantly inhibits proliferation in MC3T3-E1, bone marrow stromal cells (BMSCs), and ROS 17/2.8 (ROS) cells through reducing levels of cyclin A and D1 protein.42, 43 Some studies have demonstrated that rapamycin promotes osteogenic differentiation from mesenchymal stromal cells (MSCs),44 human embryonic stem cell (hESC),45 and ROS cells.43 On the other hand, one study reports rapamycin inhibits osteogenic differentiation in the early differentiating stage of MC3T3-E1 and BMSCs.42 These contradicting results indicate that rapamycin may play different roles during osteoblastic differentiation. We have found hyperactive RAS signaling in Nf1+/− osteoblasts.46 mTOR signaling might also function through osteoblasts to regulate osteoclastogensis in Nf1+/− mice. Further experiments are required to determine this possibility.
Rapamycin is an antifungal agent originally purified from Streptomyces hygroscopicus with potent immunosuppressive actions.47 Rapamycin is used in combinations with other drugs to prevent organ rejection after renal transplantation. Renal transplant patients who receive rapamycin daily for several years do not seem to have obligatory problems from sustained blockage of TOR. Among the side effects of rapamycin, there are no pronounced immunosuppressive problems (viral, bacterial, and fungal infections, Kaposi's sarcoma or osteoporosis).48 Ironically, it was assumed that rapamycin would predispose patients to osteoporosis and cancer, but instead rapamycin exhibited anti-cancer and bone-sparing activities.49, 50 Bone-resorbing osteoclasts depend on the TOR pathway for their formation, function, and survival. Rapamycin and other inhibitors of TOR inhibit osteoclast formation and activity, thus prevents the OVX-induced loss of cancellous bone.21 In the present study, we also find similar inhibition of osteoclast formation by rapamycin. Because Rapamycin and its derivatives inhibit mTOR-dependent mRNA translation both in osteoclasts and tumor cells, they are considered as bi-functional molecules affecting simultaneously bone and tumor metabolisms.51 Because mTOR signaling is constitutively activated in NF1-deficient Schwann cells and tumor cells,19, 20 rapamycin and its derivatives are in clinical trial for NF1 tumors in multicenter clinical trials. In the present study, we found that rapamycin can inhibit mTOR signaling in primary Nf1+/− OCLs and correct abnormal osteoclast formation in Nf1+/− mice. Our findings suggest that rapamycin may service as bi-functional molecules to treat NF1 individuals with osteoporosis/osteopenia.
In summary, our research suggests that neurofibromin directly regulates osteoclastogenesis through mTOR signal transduction. Inhibiting mTOR may represent a viable strategy to target NF1 bone diseases.