Skeletal abnormalities including osteoporosis and osteopenia occur frequently in both pediatric and adult neurofibromatosis type 1 (NF1) patients. NF1 (Nf1) haploinsufficient osteoclasts and osteoclast progenitors derived from both NF1 patients and Nf1+/− mice exhibit increased differentiation, migration, and bone resorptive capacity in vitro, mediated by hyperactivation of p21Ras in response to limiting concentrations of macrophage-colony stimulating factor (M-CSF). Here, we show that M-CSF binding to its receptor, c-Fms, results in increased c-Fms activation in Nf1+/− osteoclast progenitors, mediating multiple gain-in-functions through the downstream effectors Erk1/2 and p90RSK. PLX3397, a potent and selective c-Fms inhibitor, attenuated M-CSF mediated Nf1+/− osteoclast migration by 50%, adhesion by 70%, and pit formation by 60%. In vivo, we administered PLX3397 to Nf1+/− osteoporotic mice induced by ovariectomy (OVX) and evaluated changes in bone mass and skeletal architecture. We found that PLX3397 prevented bone loss in Nf1+/−-OVX mice by reducing osteoclast differentiation and bone resorptive activity in vivo. Collectively, these results implicate the M-CSF/c-Fms signaling axis as a critical pathway underlying the aberrant functioning of Nf1 haploinsufficient osteoclasts and may provide a potential therapeutic target for treating NF1 associated osteoporosis and osteopenia.
Citation: He Y, Rhodes SD, Chen S, Wu X, Yuan J, Yang X, et al. (2012) c-Fms Signaling Mediates Neurofibromatosis Type-1 Osteoclast Gain-In-Functions. PLoS ONE 7(11): e46900. https://doi.org/10.1371/journal.pone.0046900
Editor: Michael Glogauer, University of Toronto, Canada
Received: June 11, 2012; Accepted: September 6, 2012; Published: November 7, 2012
Copyright: © 2012 He et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported in part by the Department of Defense NF043032 (FCY), NF073112 (FCY) and National Institutes of Health/National Cancer Institute/5R01CA074177 (FCY and DWC), Indiana Clinical and Translational Sciences Institute Career Development Award PHS NCCR Grant No: 5TL1RR025759-03 (SDR), and Children’s Tumor Foundation Young Investigator Award 2011-01-010 (SDR). Plexxikon Inc. provided the PLX3397 compound. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Neurofibromatosis type 1 (NF1) is one of the most common genetic diseases with an incidence of 1 out of 3000 individuals . NF1 is caused by mutations of the NF1 tumor suppressor gene, which encodes neurofibromin, a GTPase activating protein for p21ras . Individuals with NF1 have a high incidence of both malignant and non-malignant complications , . Clinical studies have reported that NF1 patients are at a significant risk for both generalized osteoporotic abnormalities , ,  and focal skeletal abnormalities including dystrophic kyphoscoliosis and pseudarthrosis , . NF1 patients have an increased prevalence of osteoporosis beginning from childhood and adolescence , , , leading to greater risk of fracture later in life . Given that osteoporosis occurs in a younger patient population and the predisposition to pseudarthrosis is 2–5% in individuals with NF1 , , , the ultimate health costs and sequelae of this condition in NF1 patients may be significantly greater. Although NF1 related osteopenia commonly presents in the childhood years, there is no efficient treatment so far. Despite evidence of low serum Vitamin D levels in some NF1 patients, clinical studies involving Vitamin D supplementation have yielded conflicting results on whether improvements in bone mineral density (BMD) can be achieved , , .
Osteoclasts are specialized bone resorbing cells which differentiate from the myeloid monocyte/macrophage lineage. Many skeletal diseases, in particular diseases with decreased bone mineral density (BMD), occur as a consequence of a skeletal imbalance that favors bone resorption. Although a significant number of skeletal diseases, including skeletal manifestations in NF1 patients, have been linked to abnormal osteoclast function(s) , , , , the intracellular mechanisms by which osteoclasts normally function or contribute to disease states are poorly understood.
Ras signaling pathways are highly relevant to bone formation and the maintenance of skeletal homeostasis. Several Ras-activating growth factors, including M-CSF, are known to affect skeletal development and remodeling. Mitogen-activated protein kinase, a major downstream effector of Ras, is critical in the mitogenic response to extracellular stimuli including growth, podosome formation, and bone resorption of the osteoclast . Previously, we reported that haploinsufficient Nf1 (+/−) myeloid progenitors are hypersensitive to M-CSF, leading to increased osteoclast formation and bone erosive activity in vitro . Furthermore, we have shown that the phenotype of Nf1+/− osteoclasts appears to be associated with hyperactivation of the MAPK pathway .
Upstream of the Ras/MAPK pathway, M-CSF binding to its membrane receptor, c-Fms, stimulates phosphorylation of Y807 in the activation loop, resulting in a conformational shift that enhances intrinsic kinase activity and docking of adaptor proteins such as Grb2 and Sos with subsequent activation of the Ras/Raf/MEK/ERK cascade. Here we show that Nf1 haploinsufficient osteoclast progenitors exhibit increased c-Fms activation in response to M-CSF, resulting in multiple osteoclast gain-in-functions including migration, adhesion, and bone resorptive capacity, which are correlated with hyperphosphorylation of the downstream effectors Erk1/2 and p90rsk. Administration of PLX3397, a potent and selective small molecule inhibitor of c-Fms receptor tyrosine kinase activity, was sufficient to mitigate hyperfunctioning Nf1+/− osteoclast phenotypes in vitro. In vivo, ovariectomized (OVX) Nf1+/− mice exhibit accelerated bone loss as compared to WT controls in response to bone resorptive stress. In the present study, we found that treatment with PLX3397 was sufficient to correct this phenotype, normalizing bone mineral density and trabecular bone mass in Nf1+/− -OVX mice vs. vehicle treated controls. Collectively, these data implicate the M-CSF/c-Fms signaling axis as a critical pathway underlying the aberrant functioning of Nf1 haploinsufficient osteoclasts and warrant further investigation of c-Fms as a potential therapeutic target for treating NF1 associated osteoporosis and osteopenia.
Materials and Methods
This study was carried out in strict accordance Indiana University's Institutional Animal Care and Use Committee (IACUC). Animals and records of their ovariectomy surgery were maintained in compliance with Indiana University's Institutional Animal Care and Use Committee with approval protocol ID #3401- A4. All surgery was performed under isoflurane anesthesia, and all efforts were made to minimize suffering.
Animal and material preparation
Nf1+/− mice were obtained from Tyler Jacks at the Massachusetts Institute of Technology (Cambridge, Massachusetts, USA) in a C57BL/6J.129 background and backcrossed for 13 generations into a C57BL/6J strain . Studies were conducted with a protocol approved by the Indiana University Laboratory Animal Research Center. Chemicals were purchased from Sigma (St. Louis, MO) unless otherwise stated. All cytokines were purchased from PeproTech (Rocky Hills, NC). PLX3397 was provided by Plexxikon Inc. (Berkeley, CA).
Clonogenic progenitor assays
Colony-forming unit-macrophage/monocyte (CFU-M) of bone marrow mononuclear cells (BMMNCs) was assayed, as described previously . BMMNCs were isolated from 6- to 8-week-old WT and Nf1+/− mice by flushing the bone marrow and using Ficoll density gradient centrifugation. 2.5×104 BMMNCs were seeded onto a 35-mm gridded dish containing methylcellulose supplemented with 30 ng/mL M-CSF and 20 ng/mL murine recombinant receptor activator of nuclear factor kappa-B ligand (RANKL) at 37°C in a 5% CO2 incubator for 7 days. Colonies were counted using an inverted light microscope.
Generation of murine osteoclasts
Murine osteoclasts were obtained in vitro from BMMNCs, as described previously . BMMNCs were cultured in alpha minimum essential medium (α-MEM) supplemented with 10% fetal bovine serum (FBS, Sigma) and M-CSF 30 ng/mL and RANKL 20 ng/mL for 3 days. Cell culture media was switched on day 4 to α-MEM supplemented with 10% FBS, M-CSF (30 ng/mL), and RANKL (60 ng/mL) for an additional 3 days of culture. Adherent cells were fixed and stained for tartrate resistant acid phosphatase (TRACP) to identify osteoclasts, as described previously . Osteoclasts were visualized using a Nikon TE2000-S microscope (Nikon Inc., Melville, NY) and images were taken by a QImaging camera and QCapture-Pro software (Fryer Company Inc., Cincinnati, OH). Multinucleated TRACP+ cells containing more than three nuclei were scored as mature osteoclasts. The numbers and areas of multinucleated, TRACP+ osteoclasts were calculated using Image Pro Express (Media Cybernetics, In., MD).
Osteoclast migration was evaluated using a transwell assay, as described previously with minor modifications . Briefly, equivalent numbers of pre-osteoclasts were loaded onto the upper chamber of an 8 µm polycarbonate transwell (Corning Incorporated, Corning, NY) coated with vitronectin (Takara, Japan). The lower chamber contained α-MEM supplemented with M-CSF (30 ng/mL) and 0.1% bovine serum albumin with phosphate buffered saline (PBS) vehicle or PLX3397 (200 nM). Cells that migrated to the bottom chamber were stained with crystal violet and counted. Six fields per experimental condition were counted.
A single-cell suspension of osteoclast precursors were pipetted into 96-well plates pre-coated with vitronectin at a density of 1×104 cells/well as previously described . After 30 minutes of incubation in the presence of M-CSF (30 ng/mL) with PBS vehicle or PLX3397 (200 nM), nonattached cells were gently removed with PBS and adherent cells were stained with crystal violet and counted.
Bone resorption assay
Osteoclasts can form a specialized cell-extracellular matrix to initiate degradation of bone matrix by secreting proteinases. To directly assess the effect of PLX3397 on osteoclast bone resorption, independent of adhesion and differentiation, pit formation assay was performed using cells that were cultured for 3 days in media containing M-CSF and RANKL as described previously . Briefly, preosteoclasts were detached from plates after 3 days culture in α-MEM supplemented with M-CSF (30 ng/mL) and RANKL (20 ng/mL) and replated at 1×105/well on dentine discs (ALPCO Diagnostic, Windham, NH) in a 96-well plate. The cells were incubated at 37°C, 5% CO2 in the presence of 30 ng/mL M-CSF and 60 ng/mL RANKL and allowed to adhere overnight. On the following day, vehicle (PBS) or PLX3397 (200 nM) was added to the culture with media changed every three days. Wells containing media alone, without M-CSF, were used as negative controls (data not shown). Following 7 days of culture, the dentine discs were rinsed with PBS, immersed in 1 M ammonium hydroxide overnight, and stained with a 1% toluidine blue in 0.5% sodium tetraborate solution. The resorptive areas or “pits” were imaged using reflective light microscopy. Pit area (µm2) per low power field was quantitated using Image Pro Express software. Six fields per experimental condition were evaluated.
High content image assay
As described above, murine preosteoclasts were obtained in vitro from BMMNCs isolated from 6- to 8-week-old WT and Nf1+/− mice. After being cultured in α-MEM supplemented with M-CSF (30 ng/mL) and RANKL (20 ng/mL), preosteoclasts were starved in 0.5%BSA in α-MEM for 6 hours, pretreated with or without PLX3397 in serial dilution for 2 hours and then stimulated with M-CSF (30 ng/mL) for 5 minutes. The cells were then fixed and incubated with anti-phospho-c-Fms antibody (Cell signaling) at 1∶100 for immunofluorescence staining. Thereafter, the cells were incubated with Alexa Fluor® 488 goat anti-rabbit IgG (H+L) (Invitrogen) at 1∶500 and Hoechst (Invitrogen) at 1∶1000 in PBS at room temperature for 90 minutes and imaged on a BD AttoVisionTM instrument. The mean fluorescence intensity of phospho-c-Fms was acquired and normalized to each cell based on the nuclear counts. Data readout was analyzed as phospho-c-Fms mean fluorescence intensity. The half maximal inhibition concentration (IC50) of PLX3397 was calculated by Prism software.
To determine if application of PLX3397 prevents bone loss in Nf1+/−-OVX mice in vivo, two groups of 8 week-old WT or Nf1+/− female mice were subject to ovariectomy surgery (OVX) to induce low bone mass as previously reported . Briefly, the mice were anesthetized via inhalation of 1.5% isoflurane (IsoFlo; Abbott Laboratories, North Chicago, IL) mixed with O2 (1.5 liter/minute). Ovariectomy was preceded by a 2 cm midline dorsal skin incision. The peritoneal cavity was incised to access the ovaries, which were then excised with scissors. The success of OVX was verified by examining the uterine weight of the mice following sacrifice (Figure S1A–B).
After two weeks of recovery from the OVX surgery, the mice were treated with either the PBS vehicle or PLX3397 (80 mg/kg/day) by daily gavage as recommended by the Plexxikon Inc. for a period of twelve weeks. At the end of the treatment, the mice were euthanized for analysis. To ensure that these mice received sufficient amounts of PLX3397, blood was collected in heparin coated tubes by cardiac puncture and plasma was separated to determine the concentration of PLX3397 in the blood (Figure S1C).
Utilizing our previously established methods , , , the femurs of syngeneic experimental mice were fixed with 4% buffered paraformaldehyde (PFA) and decalcified for paraffin embedding. Histological sections from the distal metaphysis were stained for the osteoclast enzyme TRACP. The area of TRACP positive (TRACP+) osteoclasts per unit trabecular surface was counted under the microscope with a QImaging camera and QCapture-Pro software (Version 5.1, Fryer Company Inc., Cincinnati, OH).
Detection of C-terminal telopeptide of type I collagen (CTX) in plasma
To evaluate the impact of PLX3397 on osteoclast activity in vivo, the CTX levels in the plasma were measured. Briefly, blood was collected by heart puncture and transferred into ethylenediaminetetraacetic acid (EDTA) coated tubes. The tubes were immediately placed on ice, and centrifuged within 30 minutes for plasma separation. All samples were analyzed in the same assay. The concentration of CTX was determined by using a commercial enzyme immunoassay kit (RatlapsTM EIA, Immunodiagnostic systems Inc., Fountain Hills, AZ).
Liquid chromatography/mass spectroscopy (LC/MS) analysis of mice plasma
Plasma was collected following centrifugation at 3,000 rpm at 4°C for 20 minutes and processed for liquid chromatography/mass spectroscopy (LC/MS) analysis. Briefly, murine plasma samples were prepared for LC/MS analysis by precipitating plasma with 3x volumes of ice cold acetonitrile. The precipitated samples were centrifuged and the clarified samples were diluted with 5×0.2% formic acid in water. Samples were analyzed by LC/MS in a formic acid, water, methanol chromatography system using an MDS SCIEX API 3000 spectrometer (Applied Biosystems, Carlsbad, California).
Measurement of BMD in vivo
Bone mineral density (BMD) of the femur was evaluated in vivo using peripheral dual energy x-ray absorptiometry (pDEXA) designed for analyzing mouse skeletons (GE-Lunar Corp., Madison, USA). The mice were anesthetized via inhalation of 1.5% isoflurane (IsoFlo; Abbott Laboratories, North Chicago, IL) mixed with O2 (1.5 liter/minute) and were placed in the prone position on a specimen tray for scanning. For this experiment, 8 week-old female mice were scanned for BMD once before surgery. Then, after 12 weeks of drug treatment, the mice were scanned for BMD again before euthanasia. The BMD of the left femoral metaphysis was measured by defining a region of interest of 11 pixels x 10 pixels proximal to the distal growth plate, which contains a high content of trabecular bone.
Micro (µ) computed tomography (CT)
After euthanasia, bone volume and microarchitecture of the femoral distal metaphysis were evaluated using a desktop µCT (µCT 20; Scanco Medical AG, Bassersdorf, Switzerland). Each specimen was scanned with a slice increment of 9 µm starting at 15% of the total femur length measured from the tip of femoral condyle and extending proximally for 200 slices. Microarchitecture parameters including bone volume (BV, mm3) and bone volume fraction (BV/TV, %) were measured.
Western blotting analysis
The phosphorylation levels of c-Fms, Erk1/2 and p90RSK in preosteoclasts were determined by western blot using phospho-specific antibodies (Cell Signaling, Danvers, MA). Phosphorylation levels were quantitated with Image J and standardized by total ß-actin expression levels (Cell Signaling). Briefly, preosteoclasts were deprived of growth factors for 12 hours in α-MEM supplemented with 0.5% bovine serum albumin (BSA) before being pretreated with PLX3397 at different concentrations as indicated, and were stimulated with M-CSF (30 ng/mL) for 5 minutes in the presence of PLX3397. The cells were then lysed in ProteoJET lysis buffer (Fermentas, Glen Burnie, MD) supplemented with Complete Mini protease inhibitor cocktail (Roche, Indianapolis, IN) for protein extraction. Protein separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) were transferred to nitrocellulose filters and probed with specific antibodies.
The data are presented as mean ± standard error of the mean (SEM). Differences among experimental groups were analyzed by repeated measures of paired t-tests or analysis of variance (ANOVA) with appropriate post hoc t-tests. P values less than 0.05 were considered significant. Statistical analyses were performed with Prism 5.0 software (GraphPad, La Jolla, CA).
Aberrant c-Fms activation mediates M-CSF hyperresponsiveness in Nf1 haploinsufficient osteoclast progenitors
Although Nf1+/− osteoclasts and osteoclast progenitors are hyperresponsive to limiting doses of M-CSF, the molecular mechanisms mediating this phenotype remain unclear. We therefore sought to determine whether Nf1 haploinsufficiency may result in aberrant M-CSF signaling via hyperactivation of its receptor, c-Fms. We first examined the phosphorylation level of c-Fms by immunofluorescence staining of WT and Nf1+/− preosteoclasts. Nf1+/− preosteoclasts exhibited increased phospho-c-Fms staining both at base line and after being stimulated with M-CSF (30 ng/mL) for 5 minutes as compared with WT preosteoclasts. The expression levels of phospho-c-Fms decreased markedly following treatment with a small molecule inhibitor of the c-Fms receptor tyrosine kinase, PLX3397, at 62.5 nM and reached the lowest level at 1 µM in WT cultures and 0.1 µM in Nf1+/− cultures, respectively (Figure 1A and 1B).
(A) Expression level of phospho-c-Fms in WT and Nf1+/− preosteoclasts was evaluated by immunofluorescence staining. Preosteoclasts were obtained by 3 days culture in the presence of M-CSF (30 ng/mL) and RANKL (20 ng/mL). After serum deprivation in 0.5% BSA in α-MEM for 6 hours, cells were pretreated with PLX3397 or PBS vehicle before stimulated with M-CSF (30 ng/mL). The cells were then fixed and stained for phospho-c-Fms (Arrows indicate the phospho-c-Fms positive cells). (B) Nf1+/− osteoclast progenitors show enhanced phosphorylation of c-Fms both at baseline and in the presence of M-CSF. Treatment with PLX3397 significantly reduced the activation of c-Fms both in the WT and Nf1+/− preosteoclasts. Maximal inhibitory concentration (IC50) indicates the half maximal (50%) inhibitory concentration (IC) of a substance (50% IC, or IC50). (C) After being stimulated with M-CSF (30 ng/mL) for 5 minutes, the phosphorylation levels of c-Fms, Erk, and P90RSK in both WT and Nf1+/− preosteoclasts were detected by western blot. Pretreatment of PLX3397 completely blocked the activation of Erk and p90RSK in those cells. These experiments were repeated on three independent occasions with similar results.
To further validate these findings and determine the impact of increased c-Fms activation with respect to the downstream MAPK pathway, the phosphorylation levels of c-Fms, p90RSK, and Erk were detected by western blotting. Nf1+/− preosteoclasts demonstrated increased phosphorylation levels of c-Fms, p90RSK and Erk as compared with WT preosteoclasts when stimulated by M-CSF. Pretreatment of PLX3397 blocked the phosphorylation of these proteins both in WT and Nf1+/− preosteoclasts (Figure 1C and Figure S2).
Pharmacologic c-Fms inhibition reduces osteoclast progenitor frequency and osteoclast differentiation in vitro.
Given that Nf1 haploinsufficiency leads to M-CSF induced hyperactivation of c-Fms and the downstream MAPK effector pathway, we sought to investigate the effect of pharmacologic c-Fms inhibition on Nf1+/− osteoclast differentiation in vitro. Since osteoclasts are derived from monocyte/macrophage progeny within the bone marrow, we quantified the number of colony forming unit-macrophage (CFU-M) per femur following culture in semisolid methylcellulose media. Consistent with our previous data , a significantly increased number of CFU-M was observed in Nf1+/− bone marrow mononuclear cells (BMMNCs) as compared to WT BMMNCs. Addition of PLX3397 significantly reduced the frequencies of CFU-M at concentrations of 200 nM both in WT and Nf1+/− cultures (Figure 2A).
(A) Clonogenic assays were performed to establish the number of osteoclast progenitors, CFU-M, per femur from WT and Nf1+/− mice. Addition of PLX3397 reduced CFU-M at a dose dependent manner. *p<0.05 and **p<0.01. (B) The ability of WT and Nf1+/− BMMNCs to form osteoclasts in response to M-CSF with PLX3397 or PBS vehicle was evaluated by TRACP staining. The y-axis shows the osteoclast area per cell. (C) The numbers of TRACP+ osteoclasts in WT and Nf1+/− culture was evaluated by TRACP staining. Data are represented as mean ± SEM, with n = 3. **p<0.01 and ***p<0.001 compared with the control group. Three independent experiments were conducted with similar results.
To assess whether inhibition of M-CSF signaling affected osteoclast formation, we established osteoclast cultures in the presence of M-CSF/RANKL with PBS vehicle or PLX3397. Osteoclast differentiation was determined by calculating the TRACP+ area per osteoclast (Figure 2B) and by counting the number of TRACP+ osteoclasts per low power field (Figure 2C). Nf1+/− cultures contained significantly increased TRACP+ area and osteoclast numbers as compared to WT cultures Addition of PLX3397 reduced osteoclast formation in both WT and Nf1+/− cultures.
M-CSF/c-Fms signaling mediates Nf1+/− preosteoclast migration and adhesion
The ability of osteoclasts to migrate across the bone surface is a key cellular function required for bone resorption. We have previously reported that haploinsufficiency of Nf1 resulted in enhanced preosteoclast migration in response to M-CSF . To assess the impact of c-Fms inhibition on preosteoclast motility mediated by M-CSF, preosteoclast migration was assessed by transwell assays. Bottom chambers containing media alone without M-CSF were used as negative controls (data not shown). After four hours of stimulation, M-CSF induced a significantly higher level of migration in Nf1+/− cultures than that of WT cultures (Figure 3A). Addition of PLX3397 reduced Nf1+/− preosteoclast migration to WT levels.
(A) A transwell assay was used to investigate the effect of the c-Fms inhibitor PLX3397 on osteoclast migration. Experiments were conducted in triplicates per condition. Five fields were counted per well. Three independent experiments were performed with similar results. **p<0.01;. (B) Adhesion assays were used to investigate the effect of the PLX3397 on osteoclast attachment to vitronectin. Experiments were conducted in triplicates per condition. Five fields were counted per well. Three independent experiments were performed with similar results. ***p<0.001. (C) Quantitative evaluation of bone resorption. Data represent the mean ± SEM of 6 fields per genotype. Each condition was carried out in triplicates and four independent experiments were performed. *p<0.05; **p<0.01; ***p<0.001.
To determine the impact of M-CSF/c-Fms signaling on preosteoclast adhesion, a key process for initiation of bone resorption, adhesion assays were performed. Consistent with our previous findings , significantly increased Nf1+/− preosteoclasts were observed to adhere to vitronectin following M-CSF stimulation. Pretreatment with PLX3397 (200 nM) dramatically decreased the number of adherent cells in both WT and Nf1+/− preosteoclasts mediated by M-CSF (Figure 3B).
PLX3397 inhibits osteoclast bone resorption
To assess the impact of PLX3397 on osteoclast bone resorptive activity, pit formation assays were performed , . Consistent with our previous study , Nf1+/− osteoclasts induced a 2–3 fold increased pit forming area as compared to WT osteoclasts (Figure 3C). Importantly, PLX3397 reduced pit forming areas in both WT and Nf1+/− cultures. These data indicate that c-Fms inhibition not only inhibits osteoclast differentiation but also diminishes osteoclast bone resorptive activity.
Pharmacologic c-Fms inhibition prevents bone loss in ovariectomized Nf1+/− mice
Based on our findings that Nf1+/− osteoclasts exhibit increased bone lytic activity and that PLX3397 inhibits osteoclast formation and bone resorptive activity in vitro, we postulated that pharmacologic inhibition of c-Fms with PLX3397 may exert anti-catabolic effects on bone in vivo. To further test our hypothesis that PLX3397 may prevent bone loss in vivo, we gavage-fed PLX3397 to Nf1+/− mice undergoing ovariectomy (OVX), an established osteoporotic stress model . 80 mg/kg/day of PLX3397 were administered to both WT-OVX and Nf1+/−-OVX mice for a period of twelve weeks, and PBS was used as vehicle in control groups.
Similar to our previous study, the OVX surgery resulted in reduced BMD in both genotypes of mice . Moreover, the Nf1+/− -OVX mice lost significantly more bone mass than the WT- OVX mice (Figure 4A), confirming that a pro-resorptive challenge induces a greater osteoclastic response in Nf1 haploinsufficient mice. Nf1+/−-OVX mice fed with PLX3397 showed a significantly increased BMD as compared to the vehicle control group.
(A) The BMD of femoral metaphysis was measured by pDEXA and analyzed with Lunar Piximus. *p<0.05; ***p<0.001 (B) Micro-CT (μCT) scans were performed on the distal femur of all groups of mice. Representative reconstructed 3D pictures are shown. The cortical portion of each femur has been removed to allow visualization of the metaphyseal architecture. (C) Bone volume fraction (BV/TV) was assessed by μCT. *p<0.05. (D) The concentration of CTX in plasma in WT and Nf1+/− mice was measured after OVX. Data are represented as mean ± SEM, with n = 8. *p<0.05, **p<0.01 and ***p<0.001 compared with the control group.
To further investigate the role of PLX3397 on modulating trabecular bone, microcomputed tomography (μCT) was performed to examine the architecture of trabecular bone. Representative μCT 3D reconstructions are shown (Figure 4B). Nf1+/−-OVX mice displayed significantly less trabecular bone, as determined by bone volume/tissue volume (BV/TV), and as compared to WT-OVX mice that received vehicle treatment. PLX3397 significantly increased the trabecular bone in Nf1+/−-OVX mice (Figure 4C).
In addition, CTX, a sensitive marker of bone loss, was significantly increased in the plasma of both WT-OVX and Nf1+/−-OVX mice as compared to controls (*p<0.05 for Nf1+/−-OVX vs. Nf1+/− sham) (Figure 4D). CTX was even higher in the plasma of Nf1+/−-OVX mice than that of WT-OVX mice (*p<0.05 for Nf1+/−-OVX vs. WT-OVX). Importantly, administration of PLX3397 attenuated the plasma CTX levels in both WT-OVX mice and Nf1+/−-OVX mice. Collectively, these data suggest that pharmacologic c-Fms inhibition is sufficient to prevent bone loss by abrogating osteoclast bone resorptive activity in vivo.
c-Fms hyperactivity mediates expansion of the osteoclast progenitor pool in Nf1+/− OVX mice in vivo
Since osteoclasts are considered to be the key lineage mediating bone loss in ovariectomy induced resorptive stress, we next investigated the effect of ovariectomy on the number of osteoclast progenitors in the bone marrow by performing CFU-M assays. Ovariectomy surgery significantly increased the frequency of CFU-M in both WT-OVX and Nf1+/−-OVX groups as compared to the sham group. PLX3397, but not by vehicle gavage-feeding, suppressed the CFU-M formation to the level of sham groups in the Nf1+/− mice (Figure 5A).
(A) The CFU-M number of the BMMNC from WT and Nf1+/− mice after OVX with vehicle or PLX3397 treatment was counted. *p<0.05. (B) Quantitative analysis of the percentage of TRACP+ osteoclast area/bone surface area (Oc./BS). Data are represented as mean ± SEM, with n = 5. *p<0.05; **p<0.01; ***p<0.001. (C) Femur samples were processed for TRACP staining. Representative photographs (40x upper panels, 100x lower panels) of the trabecular bone following TRACP staining are shown. The red stained area indicates TRACP+ osteoclasts.
To examine the impact of PLX3397 on osteoclastogenesis in vivo, the TRACP+ cell area/bone surface area of the femur was scored following TRACP staining. A significant increase in the TRACP+ area/bone surface area was observed in Nf1+/− Sham as compared to the WT Sham group. Furthermore, OVX surgery induced an even more dramatic increase in osteoclastogenesis in Nf1+/− mice as evidence by the increased TRACP+ cells compared to WT-OVX mice. Importantly, PLX3397 treatment reduced the number of TRACP+ osteoclasts in Nf1+/−-OVX mice (Figure 5B and 5C). Collectively, these data indicate that elevated c-Fms signaling, following ovariectomy induced resorptive stress, results in increased numbers of osteoclast progenitors and mature osteoclasts in Nf1+/− mice, and consequently accelerated bone loss.
Osteoporosis and osteopenia affect both children and adults with NF1 , , , , , . As Nf1+/− mice have normal bone mass, ovariectomy is an established experimental model to induce low bone mass in Nf1+/− mice . Previously, we reported an increased frequency of osteoclast progenitors in Nf1+/− bone marrow induced by M-CSF as assessed by the colony forming unit-macrophage (CFU-M) assay . We also showed that haploinsufficiency of Nf1 results in hypersensitivity of osteoclast progenitors to M-CSF correlated with hyperactivation of the Ras/MAPK pathway . In the present study, we show that the M-CSF receptor, c-Fms, is hyperactivated in Nf1 haploinsufficient osteoclasts and progenitors following M-CSF stimulation. Utilizing a small molecule c-Fms kinase inhibitor, PLX3397, we demonstrate that c-Fms overactivity mediates multiple Nf1+/− osteoclast gain-in-function phenotypes both in vitro and in vivo including expansion of osteoclast progenitors in the bone marrow, increased osteoclast migration, adhesion, and differentiation, and accelerated bone loss following ovariectomy induced resorptive stress.
M-CSF is a critical cytokine for the development of macrophages and their subsequent differentiation to the osteoclast cell lineage . It has been reported recently that macrophage colony stimulating factor (M-CSF), together with signals from a family of biologically related tumor necrosis factor receptor like proteins (particularly RANKL), control the induction of osteoclast differentiation and are key to the physiologic and pathologic responses of osteoclasts , . Post-menopausal patients, as well as those with common chronic diseases such as arthritis, have a high incidence of osteoporosis, due to increased production of M-CSF and RANKL, which leads to increased osteoclast activation and consequent bone loss . It has been shown that PI3K is a critical downstream effector of the M-CSF/c-Fms pathway . Once activated by M-CSF, PI3K modulates osteoclast survival as well as actin remodeling , . Pharmacologic inhibitors of PI3K, including LY294002 and wortmannin, show a dramatic reduction of osteoclast formation in cultures treated with M-CSF and RANKL. Furthermore, treating osteoclasts with pharmacologic inhibitors of PI3K also impairs the resorptive activity of osteoclasts, which is associated with impaired membrane ruffling, deficient actin ring formation, and reduced pit formation . Unfortunately, the clinical utility of commercially available PI3K inhibitors such as wortmannin and LY294002 is limited as these compounds target the ATP-binding site of all PI3K family members and interfere with other PI3K related kinases at therapeutic concentrations. As such, these inhibitors are broad spectrum, non-specific and are associated with significant toxicity in vivo. PLX3397 is a novel small molecule that potently and selectively inhibits c-Fms, Kit, and Flt3-ITD kinases , . Since osteoclasts cultured from Nf1+/− mice and human NF1 patients exhibit hyperactivation of the c-Fms receptor kinase in response to M-CSF, we reasoned that pharmacologic inhibition of c-Fms with PLX3397 may exert anti-catabolic effects on bone in the NF1 murine model.
M-CSF is critical for the proliferation of osteoclast precursors and addition of an M-CSF neutralizing antibody considerably reduced macrophage differentiation at the early precursor stage , . Consistent with these findings, our in vivo study indicates that administration of PLX3397 significantly reduced the frequency of osteoclast progenitors as shown by reduced numbers of CFU-M. In addition, M-CSF is also essential for osteoclast formation and maturation by contributing to proliferation and survival [1√8] . In the present study, we show that pharmacologic c-Fms inhibition reduces osteoclast differentiation and bone resorption. In addition, our data demonstrates that PLX3397 inhibited M-CSF mediated MAPK signaling pathway by diminishing the activation of c-Fms, p90RSK and Erk, which are key to the hyperfunctioning of Nf1+/− osteoclasts in response to M-CSF.
Underscoring the critical role of c-Fms in mediating excessive osteoclast bone resorptive activity in NF1, administration of PLX3397 was sufficient to prevent bone loss in an established murine model of NF1 low bone mass. This normalization of bone mineral density and trabecular bone mass in response to pharmacologic c-Fms inhibition in vivo was associated with the attenuation of osteoclast development and cellular functions, including osteoclast differentiation, migration and bone resorption. Collectively, these data suggest that the M-CSF signaling through its receptor c-Fms may play a critical role in osteoclast-mediated osteoporotic manifestations in NF1. Biochemically, our data show that administration of PLX3397 can prevent the activation of the MAPK signaling cascade in response to M-CSF, thereby correcting aberrant osteoclast functions in the Nf1+/− murine model. As such, further studies are needed to assess whether targeting M-CSF/c-Fms signaling might provide a potential treatment for osteoporosis in NF1.
(A) Uteri of WT or Nf1+/− mice undergoing OVX surgery were compared to sham operated controls. (B) Uterine weight of WT and Nf1+/− mice undergoing OVX surgery vs. sham controls. Data are represented as mean ± SEM, with n = 5. *p<0.05; **p<0.01; ***p<0.001. (C) The plasma concentration of PLX3397 in mice receiving treatment is shown.
We thank Lisa Harris for administrative support. Plexxikon Inc. provided the PLX3397 compound.
Conceived and designed the experiments: YH FCY TAG. Performed the experiments: YH SDR SC XW JY XY LJ XL NT. Analyzed the data: YH KSM TAG FCY MX. Contributed reagents/materials/analysis tools: NT XL. Wrote the paper: YH FCY MX.
- 1. Lammert M, Friedman JM, Kluwe L, Mautner VF (2005) Prevalence of neurofibromatosis 1 in German children at elementary school enrollment. Arch Dermatol 141: 71–74.
- 2. Hiatt KK, Ingram DA, Zhang Y, Bollag G, Clapp DW (2001) Neurofibromin GTPase-activating protein-related domains restore normal growth in Nf1−/− cells. J Biol Chem 276: 7240–7245.
- 3. Riccardi VM (1981) Von Recklinghausen neurofibromatosis. N Engl J Med 305: 1617–1627.
- 4. Friedman JM (2002) Neurofibromatosis 1: clinical manifestations and diagnostic criteria. J Child Neurol 17: 548–554; discussion 571–542, 646–551.
- 5. Kuorilehto T, Poyhonen M, Bloigu R, Heikkinen J, Vaananen K, et al. (2005) Decreased bone mineral density and content in neurofibromatosis type 1: lowest local values are located in the load-carrying parts of the body. Osteoporos Int 16: 928–936.
- 6. Lammert M, Kappler M, Mautner VF, Lammert K, Storkel S, et al. (2005) Decreased bone mineral density in patients with neurofibromatosis 1. Osteoporos Int 16: 1161–1166.
- 7. Illes T, Halmai V, de Jonge T, Dubousset J (2001) Decreased bone mineral density in neurofibromatosis-1 patients with spinal deformities. Osteoporos Int 12: 823–827.
- 8. Jacques C, Dietemann JL (2005) [Imaging features of neurofibromatosis type 1]. J Neuroradiol 32: 180–197.
- 9. Stevenson DA, Zhou H, Ashrafi S, Messiaen LM, Carey JC, et al. (2006) Double inactivation of NF1 in tibial pseudarthrosis. Am J Hum Genet 79: 143–148.
- 10. Brunetti-Pierri N, Doty SB, Hicks J, Phan K, Mendoza-Londono R, et al. (2008) Generalized metabolic bone disease in Neurofibromatosis type I. Mol Genet Metab. 94: 105–111.
- 11. Seitz S, Schnabel C, Busse B, Schmidt HU, Beil FT, et al. (2010) High bone turnover and accumulation of osteoid in patients with neurofibromatosis 1. Osteoporos Int 21: 119–127.
- 12. Stevenson DA, Moyer-Mileur LJ, Murray M, Slater H, Sheng X, et al. (2007) Bone mineral density in children and adolescents with neurofibromatosis type 1. J Pediatr 150: 83–88.
- 13. Heerva E, Koffert A, Jokinen E, Kuorilehto T, Peltonen S, et al.. (2012) A controlled register based study of 460 neurofibromatosis 1 (NF1) patients: Increased fracture risk in children and adults over 41 years. J Bone Miner Res.
- 14. Crawford AH Jr, Bagamery N (1986) Osseous manifestations of neurofibromatosis in childhood. J Pediatr Orthop 6: 72–88.
- 15. Friedman JM, Birch PH (1997) Type 1 neurofibromatosis: a descriptive analysis of the disorder in 1,728 patients. Am J Med Genet 70: 138–143.
- 16. Elefteriou F, Kolanczyk M, Schindeler A, Viskochil DH, Hock JM, et al. (2009) Skeletal abnormalities in neurofibromatosis type 1: approaches to therapeutic options. Am J Med Genet A 149A: 2327–2338.
- 17. Mondal R, Nandi M, Chandra PK Neurofibromatosis, pathological fracture and hypervitaminosis-D. Indian Pediatr 47: 881–882.
- 18. Yang FC, Chen S, Robling AG, Yu X, Nebesio TD, et al. (2006) Hyperactivation of p21ras and PI3K cooperate to alter murine and human neurofibromatosis type 1-haploinsufficient osteoclast functions. J Clin Invest 116: 2880–2891.
- 19. Yan J, Chen S, Zhang Y, Li X, Li Y, et al. (2008) Rac1 mediates the osteoclast gains-in-function induced by haploinsufficiency of Nf1. Hum Mol Genet 17: 936–948.
- 20. Heerva E, Alanne MH, Peltonen S, Kuorilehto T, Hentunen T, et al. (2010) Osteoclasts in neurofibromatosis type 1 display enhanced resorption capacity, aberrant morphology, and resistance to serum deprivation. Bone 47: 583–590.
- 21. Alanne MH, Siljamaki E, Peltonen S, Vaananen K, Windle JJ, et al.. (2012) Phenotypic characterization of transgenic mice harboring Nf1(+/−) or Nf1(−/−) osteoclasts in otherwise Nf1(+/+) background. J Cell Biochem.
- 22. He Y, Staser K, Rhodes SD, Liu Y, Wu X, et al. (2011) Erk1 positively regulates osteoclast differentiation and bone resorptive activity. PLoS One 6: e24780.
- 23. Jacks T, Shih TS, Schmitt EM, Bronson RT, Bernards A, et al. (1994) Tumour predisposition in mice heterozygous for a targeted mutation in Nf1. Nat Genet 7: 353–361.
- 24. Wu X, Chen S, He Y, Rhodes SD, Mohammad KS, et al. (2011) The haploinsufficient hematopoietic microenvironment is critical to the pathological fracture repair in murine models of neurofibromatosis type 1. PLoS One 6: e24917.
- 25. Zhang W, Rhodes SD, Zhao L, He Y, Zhang Y, et al. (2011) Primary osteopathy of vertebrae in a neurofibromatosis type 1 murine model. Bone 48: 1378–1387.
- 26. Yilmaz K, Ozmen M, Bora Goksan S, Eskiyurt N (2007) Bone mineral density in children with neurofibromatosis 1. Acta Paediatr 96: 1220–1222.
- 27. Dulai S, Briody J, Schindeler A, North KN, Cowell CT, et al. (2007) Decreased bone mineral density in neurofibromatosis type 1: results from a pediatric cohort. J Pediatr Orthop 27: 472–475.
- 28. Yoshida H, Hayashi S, Kunisada T, Ogawa M, Nishikawa S, et al. (1990) The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature 345: 442–444.
- 29. Lacey DL, Timms E, Tan HL, Kelley MJ, Dunstan CR, et al. (1998) Osteoprotegerin ligand is a cytokine that regulates osteoclast differentiation and activation. Cell 93: 165–176.
- 30. Yasuda H, Shima N, Nakagawa N, Yamaguchi K, Kinosaki M, et al. (1998) Osteoclast differentiation factor is a ligand for osteoprotegerin/osteoclastogenesis-inhibitory factor and is identical to TRANCE/RANKL. Proc Natl Acad Sci U S A 95: 3597–3602.
- 31. Raisz LG (2005) Pathogenesis of osteoporosis: concepts, conflicts, and prospects. J Clin Invest 115: 3318–3325.
- 32. Golden LH, Insogna KL (2004) The expanding role of PI3-kinase in bone. Bone 34: 3–12.
- 33. Pilkington MF, Sims SM, Dixon SJ (1998) Wortmannin inhibits spreading and chemotaxis of rat osteoclasts in vitro. J Bone Miner Res 13: 688–694.
- 34. Palacio S, Felix R (2001) The role of phosphoinositide 3-kinase in spreading osteoclasts induced by colony-stimulating factor-1. Eur J Endocrinol 144: 431–440.
- 35. Nakamura I, Takahashi N, Sasaki T, Tanaka S, Udagawa N, et al. (1995) Wortmannin, a specific inhibitor of phosphatidylinositol-3 kinase, blocks osteoclastic bone resorption. FEBS Lett 361: 79–84.
- 36. Anthony SP, Puzanov I, Lin PS, Nolop KB, West B, et al.. (2011) Pharmacodynamic activity demonstrated in phase I for PLX3397, a selective inhibitor of FMS and Kit. J Clin Oncol (suppl; abstr 3093) 29.
- 37. DeNardo DG, Brennan DJ, Rexhepaj E, Ruffekk B, Shiao SL, et al. (2011) Leukocyte complexity predicts breast cancer survival and functionally regulates response to chemotherapy. Drug Discovery 1: 54–67.
- 38. Liggett W Jr, Shevde N, Anklesaria P, Sohoni S, Greenberger J, et al. (1993) Effects of macrophage colony stimulating factor and granulocyte-macrophage colony stimulating factor on osteoclastic differentiation of hematopoietic progenitor cells. Stem Cells 11: 398–411.
- 39. Biskobing DM, Fan X, Rubin J (1995) Characterization of MCSF-induced proliferation and subsequent osteoclast formation in murine marrow culture. J Bone Miner Res 10: 1025–1032.
- 40. Otero K, Turnbull IR, Poliani PL, Vermi W, Cerutti E, et al. (2009) Macrophage colony-stimulating factor induces the proliferation and survival of macrophages via a pathway involving DAP12 and beta-catenin. Nat Immunol 10: 734–743.