- Email: [email protected]
Osteoclasts derived from patients with neurofibromatosis 1 (NF1) display insensitivity to bisphosphonates in vitro
Bone 50 (2012) 798–803
Contents lists available at SciVerse ScienceDirect
Bone journal homepage: www.elsevier.com/locate/bone
Original Full Length Article
Osteoclasts derived from patients with neurofibromatosis 1 (NF1) display insensitivity to bisphosphonates in vitro Eetu Heervä a, Sirkku Peltonen b, Erkki Svedström c, Hannu T. Aro d, Kalervo Väänänen a, Juha Peltonen a,⁎ a
University of Turku, Department of Cell biology and Anatomy, Turku 20520, Finland Department of Dermatology, Turku University Hospital and University of Turku, Turku 20521, Finland Department of Diagnostic Imaging, Medical Imaging Centre of Southwest Finland, University of Turku, Turku 20521, Finland d Orthopaedic Research Unit, Turku University Hospital, Turku 20520, Finland b c
a r t i c l e
i n f o
Article history: Received 29 November 2011 Accepted 16 December 2011 Available online 27 December 2011 Edited by: Bjorn Olsen Keywords: BTM Fracture Farnesyl thiosalicylic acid Neurofibromatosis Osteoporosis
a b s t r a c t A total of 20 patients with neurofibromatosis 1 (NF1) were screened for NF1-related osteoporosis, and blood samples were collected for isolation of peripheral blood osteoclast progenitors. Patients with NF1 had higher levels of serum bone turnover markers (CTX and PINP) compared to controls. In addition, persons with high bone resorption in vitro on average had high levels of serum CTX. Of the 20 patients with NF1, 15 had low bone mineral density (osteopenia/osteoporosis), but these 15 patients did not have marked risk factors for low bone mineral density. Thus, we recommend screening for osteoporosis to all adult patients with NF1. Our aim was also to characterize the effects of bisphosphonates on NF1 osteoclasts in vitro. NF1 osteoclasts and osteoclasts from healthy controls in vitro were treated with zoledronic acid, alendronate and clodronate. These bisphosphonates caused a marked reduction in the number of normal control osteoclasts in vitro, while only a slight change was observed in the number of NF1 osteoclasts. Ras-inhibitor FTS counteracted this NF1-related insensitivity to zoledronic acid, suggesting that Ras may play a role in this phenomenon. © 2011 Elsevier Inc. All rights reserved.
Introduction Neurofibromatosis type 1 (NF1) is an autosomal dominant neuro-cutaneous-skeletal syndrome with an incidence of ~1/3000 [1–3]. The protein product of NF1 gene, neurofibromin, functions as a Ras-GTPase activating protein (Ras-GAP) negatively regulating Ras signaling pathway [4]. In bone, NF1 mRNA and protein have been detected in human chondrocytes, osteoblasts and osteoclasts [5,6]. Low bone mineral density (BMD) is a common feature in NF1, and is found in both sexes and also in children and adolescents with NF1 (Table 1) [6–12]. Low levels of serum vitamin D, high levels of serum parathyroid hormone, and increased collagen degradation products in urine have been reported in patients with NF1 [10–13]. In the subpopulation of patients with NF1 who have low BMD, high levels of serum parathyroid hormone, calcium, and tartrate resistant acid phosphatase 5b have been reported [12]. Analysis of whole-body subtotal age-
Abbreviations: NF1, neurofibromatosis type 1; BMD, bone mineral density; CTX, collagen type I C-terminal telopeptide; PINP, procollagen type I N-terminal propeptide; TRACP5b, tartrate resistant acid phosphatase 5b; MCSF, macrophage colony stimulating factor; RANKL, receptor activator of nuclear factor kappa-beta ligand; FTS, farnesyl thiosalicylic acid. ⁎ Corresponding author at: University of Turku, Department of Cell Biology and Anatomy, Kiinamyllynkatu 10, 20520 Turku, Finland. Fax: + 358 23337352. E-mail address: juha.peltonen@utu.fi (J. Peltonen). 8756-3282/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2011.12.011
matched BMD (Z-score), urine collagen degradation products and in vitro osteoclast formation capacity in 75 patients with NF1 aged 1–25 years showed no correlation between these parameters [13]. Bone biopsy histomorphometry of the patients with NF1 has revealed increased volume of osteoid, and increased number of both osteoclasts and osteoblasts compared to controls [10,11]. This suggests increased bone turnover in NF1, which can be assessed by bone turnover markers (BTMs), such as serum C-terminal cross-linking beta-telopeptide of type I collagen (CTX), tartrate resistant acid phosphatase 5b (TRACP5b), and the N-terminal propeptide of procollagen type I (PINP). CTX is representative of the catabolic rate of bone, TRACP5b has been used as an indicator of osteoclast activity, and PINP is secreted by osteoblasts during collagen synthesis, representing bone formation rate [14,15]. Amino and non-amino-bisphosphonates rapidly localize to bone and are used to treat osteoporosis. Non-amino-bisphosphonates, such as clodronate, generate toxic ATP analogues which disrupt the function of mitochondria, leading to apoptosis of osteoclasts [16–18]. Amino-bisphosphonates, such as alendronate and zoledronic acid, inhibit farnesyldiphosphate syntase, interfering with farnesylation of small GTPases including Ras, Rac and Rho. This leads to reduced ERK pathway signaling (downstream mediator of Ras), which in turn leads to mitochondrial membrane depolarization, activation of caspase cascade, and ultimately to cell death [18–21]. Osteoclasts are derived from haematopoietic progenitor cells, which can be isolated from peripheral blood samples [22,23]. In
E. Heervä et al. / Bone 50 (2012) 798–803 Table 1 Osteoporosis and osteopenia in neurofibromatosis 1 defined using dual energy X-ray absorptiometry. Study
Patients with NF1
Osteoporosis
Osteopenia
Kuorilehto (2005)
12 females, 14 males, aged 21–73 39 females, 45 males, aged 5–18 16 females, 16 males, aged 3–17 9 females, 5 males, aged 19–66 47 females, 26 males, aged 3–59 43 females, 29 males, aged 18–72 10 females, 10 males, aged 23–76
7 (27%)
13 (50%)
Stevenson (2007) Duman (2008) Seitz (2008) Brunetti-Pierri (2008) Tucker (2009) Current study
BMD reported lower compared to controls BMD reported lower compared to controls 8 (57%) Not reported 24 (32%)⁎
41 (57%)
14 (19%)
36 (50%)
4 (20%)
11 (55%)
⁎ Osteoporosis was defined as Z-score − 2.5 SD or less.
osteoclasts, Ras pathway functions as an anti-apoptotic and proosteoclastogenic pathway [24]. Osteoclasts derived from peripheral blood of patients with NF1 have been shown to be more numerous, resorb larger amounts of bone, display increased ERK and AKT activity, and tolerate serum-free conditions better compared to control osteoclasts [13,25–27]. Zoledronic acid has been shown to improve fracture healing in Nf1 +/− mice [28]. Since NF1 protein functions as a Ras-GAP and since the functions of amino-bisphosphonates are mediated via Ras pathway, the aim of this study was to characterize the effects of zoledronic acid, alendronate and clodronate in osteoclasts derived from peripheral blood of 20 NF1 patients and their controls. An additional aim was to characterize BMD and BTMs in these patients with NF1.
Subjects and methods NF1 patients A group of 21 Finnish Caucasian patients with NF1 from 19 families was recruited from the NF clinic of the Department of Dermatology, Turku University Hospital, Turku, Finland. In the statistical analyses we included 10 males aged 23–76 years (average 45) and 10 females aged 23–62 years (average 38). One 56-yearold female patient was excluded because she was medicated with alendronate. All patients fulfilled the NIH diagnostic criteria for NF1 [29]. This study was approved by Ethics Committee of Southwest Finland Hospital District, and the participants gave their written consents to BMD measurement, laboratory tests and osteoclast cultures. The study was carried out in Turku University Hospital and University of Turku during the years 2010–2011. Medical records of the patients were screened for the past 10 years. The body mass index of the patients was within normal range, three patients were smokers, and of 10 the female patients two were postmenopausal. Four patients were having vitamin D supplementation. Physical activity was evaluated by using a questionnaire [7]. No malignancies or other conditions known to affect bone were found, apart for alendronate in the one excluded patient, and lactose-intolerance in one patient. One patient with mild scoliosis and two patients with lytic bone lesions were found. Four patients had fractures during the past 10 years, which were located in ankle, distal radius, rib and patella. These fractures were noted in two patients with NF1-related osteoporosis and in two patients with osteopenia. In addition, the 56-year old female patient who was excluded from statistical analyses (see above) due to alendronate medication had a stress fracture of metatarsal bone during alendronate treatment.
799
Control persons A group of 20 healthy volunteers was recruited from Turku University personnel. Controls were matched for sex, age and menopausal status. Osteoclast progenitors were isolated from peripheral blood samples for cell culture experiments. CTX and PINP serum measurements were available from 18 controls. Bone densitometry A dual energy X-ray absorptiometry scan of lumbar spine (L1–L4) and left proximal femur was obtained, using a Hologic QDR 4500 densitometer (Hologic Inc., Waltham, MA, USA) for 11 patients, or an Osteocore 3 densitometer (Medilink Inc., Mauguio, France) for 8 patients, or a Lunar Prodigy densitometer (GE Healthcare, Madison, WI) for one patient, with adult standard measurement software. Corresponding T-scores and Z-scores were calculated using the equipment's standard software and Finnish Caucasian demographic databases provided by the manufacturers. The densitometers were calibrated each morning with a calibration phantom. The image quality and measurements were analyzed by an experienced radiologist (ES). The diagnosis of osteoporosis was made according to the NIH criteria with T-score −2.5 SD or below either in lumbar or femoral region, and the diagnosis of osteopenia with T-score between −1.0 SD and −2.5 SD [30,31]. Serum samples Laboratory analyses included total calcium, ionized calcium, inorganic phosphate, parathyroid hormone, 25-D3-vitamin, and alkaline phosphatase including its isoenzymes, and CTX. These measurements were purchased from Turku University Hospital, Turku, Finland, and carried out using Roche automatic analyzer (Roche diagnostics, Mannheim, Germany). TRACP5b was measured using BONETRAP immunofixation method, and analyses were purchased from Medix Laboratories, Helsinki, Finland. PINP measurements were purchased from Oulu University Hospital, Oulu, Finland, and were carried out using UniQ RIA analyzer (Orion, Espoo, Finland). All samples were collected in a similar way, and analyzed using a standardized protocol. In vitro analysis of osteoclast cultures In each case, osteoclast progenitors from 20 patients and 20 controls were isolated immediately after the patient's visit. Osteoclast progenitors were cultured as described initially by Yang et al. [25], with slight modifications [26]. Briefly, peripheral blood mononuclear cells were isolated using Ficoll-centrifugation (GE Healthcare, Uppsala, Sweden), and 0.5 million cells were seeded on bovine bone slices. Cells were cultured for 7 days in alpha-MEM (Gibco, NY) supplemented with 10% heat-inactivated fetal calf serum (Gibco, NY), receptor activator of nuclear factor kappa-beta ligand (RANKL, 20 ng/ml, Peprotech, Rocky Hill, NJ), macrophage colony stimulating factor (MCSF, 10 ng/ml, R&D systems, Minneapolis, MN) and antibiotics. This was followed by a 2-day culture with the same medium described above with RANKL and MCSF, and different bisphosphonates. We used amino-bisphosphonates zoledronic acid (Zometa ®, Novartis Finland, Finland) and alendronate (Sigma-Aldrich, Steinheim, Germany), and non-amino-bisphosphonate clodronate (Leiras, Finland). A Ras inhibitor Farnesyl Thiosalicylic Acid (FTS) 10E-6 M was purchased from Cayman Chemicals (Ann Arbor, MI) and dissolved in DMSO. For each drug 4 separate bone slices were prepared. Cells were fixed in formaldehyde and stained with tartrate resistant acid phosphatase staining kit (Sigma-Aldrich leukocyte acid phosphatase-kit, Steinheim, Germany), which is a direct osteoclastspecific staining method. Nuclei were visualized with Hoechst
800
E. Heervä et al. / Bone 50 (2012) 798–803
(Invitrogen, Eugene, OR) and actin rings with rhodamine-conjugated phalloidin (Cambrex, Walkersville, MD). TRACP-positive cells with 3 or more nuclei were considered as viable osteoclasts and those with fragmented nuclei and retracted cytoplasm as osteoclasts undergoing apoptosis [32]. Bone slices were imaged with Leica DMRB fluorescence microscope (Leica, Wetzlar, Germany). Osteoclasts were identified and counted from 6 approximately 0.5 × 0.5 mm rectangular areas selected in a routine manner [26].
Caspase-3 enzymatic activity Cells were cultured on bone as described above, with and without 10E-7 M zoledronic acid. A caspase-3 colorimetric assay kit (PromoCell, Heidelberg, Germany) was used to measure enzymatic activity of caspase-3, a marker of apoptosis. Cells were lysed and treated according to manufacturer's instructions. Absorbance was measured using Hidex Chameleon 96-plate reader (Hidex, Turku, Finland) at 405 nm.
Quantification of bone resorption in vitro We used quantification of cell culture media CTX instead of counting the resorption pits due to the fact, that CTX can be measured both before and after bisphosphonate treatment from the very same cultures. Levels of cell culture media CTX correlate with the number of resorption pits in the cultures [26]. CTX measurement was purchased from Turku University Hospital, Turku, Finland, and carried out using Roche automatic analyzer (Roche diagnostics, Mannheim, Germany).
Statistics The statistical data had normal distribution as estimated by Shapiro–Wilk analysis. Two-sample t-test was used to analyze the differences between patients and controls, males and females, or osteoporotic and non-osteoporotic groups. Correlation was calculated using linear regression analysis. In cell culture experiments the 20 patients and controls were matched for age, sex and menopausal status, and two-sample t-test was used. Paired t-test was used to compare treated and non-treated samples of the same person. The difference in the effect of different bisphosphonates between NF1 and control cell cultures was analyzed with two-sample t-test, followed by repeated measures analysis of variance. P-values less than 0.05 were considered statistically significant. All statistical analyses were made with SPSS for Windows version 19 (SPSS Inc, Chicago, IL).
Results Characterization of patients with NF1 BMD of lumbar spine (L1–L4) and proximal femur were obtained from the 20 patients with NF1. The lowest T-score of the NF1 patients was −1.7 SD ±1.2 SD on average, and the respective lowest Z-score −1.2 SD ±1.0 SD. Four out of 20 patients had osteoporosis and 11 patients had osteopenia (Table 1), consistent with the previous literature [30,31]. In addition, three patients had osteoporosis in one lumbar vertebra, but the average T-score of lumbar spine was between −2.1 SD and −2.2 SD, thus not fulfilling the diagnostic criteria for osteoporosis. Out of the 15 patients with low BMD (osteoporosis or osteopenia) three were smokers, one had lactose intolerance, two were postmenopausal women, and one male was aged 76 years. However, the remaining 8 patients with low BMD did not have any apparent risk factors for low BMD apart from NF1, and none of the 15 patients with low BMD had two or more risk factors. All 20 patients were also screened for selected serum values (Table 2). Elevated levels of parathyroid hormone were found in 4 patients, elevated levels of serum CTX in 8 patients, elevated levels of PINP in 3 patients and low levels of vitamin D were found in 2 patients. Other values were within reference range. Vitamin D levels were similar in winter and summer samples, possibly due to vitamin D supplementation. No gender or age-related differences were observed. The levels of serum total and ionized calcium correlated with the patients' T-scores. Thus, patients with NF1 and low BMD had higher levels of total and ionized serum calcium than the patients with normal BMD (p = 0.02, linear regression). These serum values were essentially similar as reported previously for larger cohorts [10,12]. Serum CTX and PINP have been proposed to be the first-line bone turnover markers [15], but have not been characterized in NF1 before. Patients with NF1 had higher levels of serum CTX and PINP compared to healthy controls (Table 2). Also, the CTX/PINP ratio was higher in patients with NF1 compared to controls (Table 2), possibly reflecting catabolic bone turnover rate in these patients [33]. No correlations between laboratory parameters and T- or Z-scores were found. Osteoclasts derived from patients with NF1 display insensitivity to zoledronic acid, alendronate and clodronate Osteoclast progenitors isolated from peripheral blood of 20 NF1 and 20 control persons were differentiated into osteoclasts and treated with different bisphosphonates for 2 days. Untreated NF1 samples yielded a higher number of osteoclasts compared to controls
Table 2 Serum values associated with bone health in the current study, presented as average ± SD. a
Serum measurements
NF1, n = 20
Reference values
Total calcium (mmol/l) Ionized calcium (mmol/l) Phosphate (mmol/l)
2.27 ± 0.09 1.24 ± 0.04 0.97 ± 0.17
Parathyroid hormone (ng/ml) Total alkaline phosphates (U/l) Bone-specific alkaline phosphatase (U/l) D-25-vitamin (nmol/l), no supplementation, n = 16 D-25-vitamin (nmol/l), supplementation, n = 4
53 ± 20 57 ± 18 27 ± 13 54 ± 22 60 ± 17
2.15–2.51 1.16–1.3 0.71–1.23 males 0.76–1.41 females 15–65 35–105 Below 69 Over 40 Over 40
Bone turnover markers CTX (μg/l) PINP (μg/l) CTX/PINP ratio TRACP5b (U/l) a b
Control, n = 18 0.29 ± 0.19 38 ± 16 b 6.2 ± 3.2 b n/a
b
NF1, n = 20 0.45 ± 0.17 52 ± 17 b 9.0 ± 4.1 b 2.9 ± 0.8
Reference values b
Given reference values are from the corresponding laboratory measurement manufacturer for adults. Value higher in patients with NF1 compared to controls, p b 0.05.
a
Below 0.57 premenopausal females 20–76 Below 5.3 for males, below 4.8 for females
E. Heervä et al. / Bone 50 (2012) 798–803
801
(p = 0.005). Thus, the drug reduced the number of osteoclasts in both NF1 and control samples, but had much weaker effect on the NF1 osteoclasts. It should be noted that, apoptotic osteoclasts with fragmented nuclei [32] were more frequent in control samples compared to NF1 samples treated with 10E-6 M zoledronic acid, 2.3% versus 0.3%, respectively (p = 0.04, n = 10 males). To assess the role of apoptosis, enzymatic activity of caspase-3 in five cell lysates was evaluated using an absorptiometry kit. Following treatment with 10E-7 M zoledronic acid, the caspase-3 levels were increased by +75% of baseline in control samples. The respective increase was only +11% in NF1 samples (Fig. 1E, p = 0.02). Due to limited number of available cells the effects of alendronate and clodronate were tested on osteoclasts derived from a subgroup of 4 male and 2 female NF1 patients and their corresponding controls. In analogy to findings on zoledronic acid, NF1 osteoclasts displayed insensitivity to 10E-6 M alendronate (p = 0.04, n = 6, Fig. 1F), and 10E-7 M clodronate (p = 0.05, n = 6, Fig. 1F). It should also be noted, that large multinuclear osteoclasts were present in NF1 samples treated with zoledronic acid (Fig. 2), and also with alendronate and clodronate. Addition of Ras inhibitor FTS counteracts the insensitivity to zoledronic acid observed in NF1 osteoclasts
Fig. 1. Effects of zoledronic acid (ZOL), alendronate (ALN) and clodronate (CLD) in the cell culture. The data is presented as average ± SD. (A) The number of viable osteoclasts in the cell culture with and without 10E-7 M zoledronic acid, 10 male patient/control pairs. (B) The number of viable osteoclasts in the cell culture with and without 10E7 M zoledronic acid, 10 female patient/control pairs. (C) The effect of 10E-6–10E-8 M zoledronic acid on the number of viable osteoclasts in the cell culture, 10 male patient/control pairs. (D) The same data as in (C), represented as percentages compared to untreated samples, 10 male patient/control pairs. (E) The proportional levels of caspase-3, a marker of apoptotic activity, after treatment with ZOL. Five male patient/control pairs pooled together. (F) The respective effects of 10E-6 M alendronate, 10E-7 M clodronate, 10E-7 M zoledronic acid, the combination of 10E-7 M zoledronic and 10E-6 M Ras inhibitor FTS (ZOL + FTS), and 10E-6 M Ras inhibitor FTS alone (FTS). 4 male and 2 female patient/control pairs pooled together. *p b 0.05. **p b 0.01.
(p = 0.001 males, p = 0.05 females, Figs. 1A and B), which is in agreement with previous findings [13,25,26]. No gender, age, or menopause-related differences were observed. First, the effect of 10E-7 M zoledronic acid was analyzed. On average, the number of viable control osteoclasts was reduced to half, representing the expected result of the drug. Surprisingly, a reduction of 25% only was noted in NF1 samples (p = 0.03 for males, p = 0.05 for females, two-sample t-test, Figs. 1A and B, and 2). Thus, a higher proportion and higher number of osteoclasts remained viable in NF1 samples treated with zoledronic acid compared to controls. Again, no age or gender-related differences were observed. Further analyses showed that the effect of zoledronic acid was dose-dependent, as tested in the 10 male samples treated with 10E8–10E-6 M zoledronic acid (Figs. 1C and D). Subsequently, the effect of zoledronic acid to number of osteoclasts was evaluated using a different statistical method (repeated measures ANOVA), and the effect of the drug was different between NF1 and control groups
Ras inhibitor FTS was added into the culture medium at the same time with zoledronic acid. 10E-6 M FTS alone did not have an apparent effect on the number of osteoclasts (Fig. 1F), as tested in samples from subgroup defined above in Osteoclasts derived from patients with NF1 display insensitivity to zoledronic acid, alendronate and clodronate. However, the combination of 10E-6 M FTS and 10E-7 M zoledronic acid had more profound effect than zoledronic acid alone. The number of osteoclasts was reduced 70% in both NF1 and control samples treated with the combination of zoledronic acid and FTS (Fig. 1F), and thus no difference was observed between NF1 and control samples treated with FTS and zoledronic acid. Zoledronic acid reduces frequency of actin rings in control but not in NF1 samples Actin rings, markers of active osteoclasts, were visualized by rhodamine-conjugated phalloidin staining. In osteoclasts derived from NF1 and control persons, 44–48% of osteoclasts had an actin ring. Zoledronic acid reduced the percentage of active control osteoclasts to 25% (p = 0.03), but no change was observed in the frequency of active NF1 osteoclasts as estimated by the presence of actin rings. Zoledronic acid prevents bone resorption in control but not in NF1 samples Total bone resorption in the cell culture was estimated by measuring levels of CTX from cell culture media of 20 patient/control pairs before (day 7) and after (day 9) the addition of 10E-7 M zoledronic acid. Culture medium CTX levels have been shown to correlate with the number of resorption pits in bone slice [26], and can be measured both before and after zoledronic acid in the very same cultures. CTX levels were readily detectable in both NF1 and control samples on day 7 indicating that the cultures harbored resorbing osteoclasts. Before addition of zoledronic acid (day 7), NF1 samples had higher levels of cell culture media CTX, 340 μg/l versus 580 μg/l, respectively (p = 0.003). NF1 samples also had higher CTX/osteoclast ratio compared to control samples (p = 0.05). No gender, age or menopause-related differences were observed. Treatment with 10E-7 M zoledronic acid for 2 days prevented marked increase in levels of cell culture media CTX in control samples (+130 μg/l, p = non-significant). However, a marked increase of CTX
802
E. Heervä et al. / Bone 50 (2012) 798–803
Fig. 2. Representative micrographs of osteoclasts in the cell culture. The panel shows typical NF1 and control samples, with and without 10E-7 M zoledronic acid. Viable osteoclasts are marked with dotted white lines. Arrows depict osteoclasts considered undergoing apoptosis, showing the fragmented nuclei or absence of nuclei. Scale bars: 50 μm.
(+500 μg/l, p = 0.0001) was noted in NF1 cell culture media. These results also indicate insensitivity of NF1 osteoclasts to zoledronic acid. Serum CTX correlates with osteoclast culture medium CTX Persons with high serum CTX often displayed high bone resorption in vitro. Specifically, the amount of CTX in untreated cell culture samples was considered as an indicator of bone resorption in vitro, and serum CTX as an estimate of catabolic rate of bone [15,26]. Serum and cell culture medium CTX levels correlated in the NF1 group (p = 0.03, linear regression, n = 20, coefficient of determination 30%). Similar results were obtained with both NF1 and control samples (n= 40). However, no correlations between serum CTX (and other laboratory parameters), osteoclast number in vitro, and T- or Z-scores were found. No gender or age-related differences were found. Discussion The in vitro exposure of normal control osteoclasts on bone slices to zoledronic acid, alendronate and clodronate induced a marked reduction in the number of osteoclasts. Also the addition of zoledronic acid prevented increase in cell culture media CTX levels in the control samples. These are expected findings and attest to the validity of the experimental setup. An unexpected finding of the current study was that NF1 osteoclasts displayed insensitivity to zoledronic acid, alendronate and clodronate, as estimated by the fact that the number of osteoclasts was only moderately reduced, and levels of cell culture CTX continued to rise in NF1 samples. NF1 osteoclasts have been shown to display an enhanced formation capacity [13,25,26], which could explain the high number of NF1 osteoclasts after bisphosphonate treatment. However, in NF1 samples treated with bisphosphonates, large multinuclear osteoclasts were noted. Thus, it seems unlikely that these large multinuclear
osteoclasts could have been formed during bisphosphonate exposure in vitro. In addition, the lower increase in caspase-3 activity in NF1 samples treated with zoledronic acid supports the view of insensitivity of NF1 osteoclasts to apoptotic stimuli. NF1 osteoclasts have been shown to display hyperactive Ras [13] and Ras has been shown as an anti-apoptotic pathway in osteoclasts derived from healthy controls [24]. Thus we hypothesize that hyperactive Ras in NF1 osteoclasts is the reason for high number of osteoclasts observed in NF1 samples treated with bisphosphonates. To test this hypothesis, Ras inhibitor FTS was used. FTS counteracted the insensitivity to zoledronic acid observed in NF1 osteoclasts. This suggests that hyperactive Ras provides anti-apoptotic effect against amino- and non-amino-bisphosphonates in NF1 osteoclasts. The combination of Ras-inhibitor FTS and zoledronic acid had more profound effect than zoledronic acid alone. Thus, it is feasible to speculate that osteoclasts may upregulate Ras to counteract apoptotic signals. The question thus emerges whether the effect of bisphosphonates is compromised in selected patients with NF1-related osteoporosis. The 56-year-old female patient who was excluded from the statistical analyses had suffered a stress fracture during alendronate treatment. In addition, her osteoclasts were insensitive to zoledronic acid in vitro. However, in general population fractures are reported in 20–30% of patients medicated with alendronate, and in 1–3% of those medicated with zoledronic acid, during follow-up of three years [34,35]. The limitation of this study is that the in vitro findings do not necessarily predict the effects of bisphosphonates in patients with NF1. It is possible that as NF1 patients are medicated with bisphosphonates for years, the cumulative exposure to the drugs overcomes the insensitivity to these drugs observed in vitro. An investigation on how often patients with NF1-related osteoporosis do not respond to bisphosphonate treatment is called for. In a previous study of a cohort of 75 patients with NF1 aged 1–25, no correlation between age-matched total body BMD excluding head
E. Heervä et al. / Bone 50 (2012) 798–803
(Z-score), number of osteoclasts in the cell culture and urine bone turnover markers was found [13]. Our results on bone resorption marker CTX from serum and osteoclast culture media show that these parameters correlate statistically significantly with each other, both in NF1 and control samples. Thus persons with high bone resorption in vitro on average have high levels of serum CTX. The differences in these two studies include the facts, that the patients in the current study are older and different bone turnover markers were measured. We suggest serum CTX as an interesting candidate for future studies identifying useful markers for bone health in NF1. Conclusions Osteoclasts isolated from patients with NF1 display in vitro insensitivity to zoledronic acid, alendronate and clodronate, possibly due to hyperactive Ras. The current and previous studies show that most of patients with NF1 and low BMD have no apparent risk factors for low BMD apart from NF1. Thus, we recommend that all adult patients with NF1 should be evaluated for possible osteoporosis. Conflicts of interest Hannu T. Aro has received institutional grants from Novartis and Eli Lilly. Acknowledgments The current study was financially supported by the Academy of Finland, Turku University Foundation, Foundation of Gerda & Ella Saarinen, Turku University Hospital EVO funding, Finnish Culture Foundation, and Emil Aaltonen Foundation. Jonas Nyman and Kaisa Ivaska are greatly acknowledged for helpful discussions. Tero Wahlberg is acknowledged for statistical assistance. Eetu Heervä is a member of Turku Graduate School of Clinical Sciences. References [1] Riccardi VM, Eichner JE. Neurofibromatosis: phenotype. Johns Hopkins University Press, Baltimore, MD, USA: Natural History and Pathogenesis; 1986. [2] Lammert M, Friedman J, Kluwe L, Mautner V. Prevalence of neurofibromatosis 1 in German children at elementary school enrollment. Arch Dermatol 2005;141:71–4. [3] Jouhilahti EM, Peltonen S, Heape AM, Peltonen J. The pathoetiology of neurofibromatosis 1. Am J Pathol 2011;178:1932–9. [4] Xu G, O'Connell P, Viskochil D, Cawthon R, Robertson M, Culver M, Dunn D, Stevens J, Gesteland R, White R. The neurofibromatosis type 1 gene encodes a protein related to GAP. Cell 1990;62:599–608. [5] Leskelä H, Kuorilehto T, Risteli J, Koivunen J, Nissinen M, Peltonen S, Kinnunen P, Messiaen L, Lehenkari P, Peltonen J. Congenital pseudarthrosis of neurofibromatosis type 1: impaired osteoblast differentiation and function and altered NF1 gene expression. Bone 2009;44:243–50. [6] Elefteriou F, Kolanczyk M, Schindeler A, Viskochil D, Hock J, Schorry E, Crawford A, Friedman J, Little D, Peltonen J, Carey J, Feldman D, Yu X, Armstrong L, Birch P, Kendler D, Mundlos S, Yang F, Agiostratidou G, Hunter-Schaedle K, Stevenson D. Skeletal abnormalities in neurofibromatosis type 1: approaches to therapeutic options. Am J Med Genet A 2009. [7] Kuorilehto T, Pöyhönen M, Bloigu R, Heikkinen J, Väänänen K, Peltonen J. 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 2005;16:928–36. [8] Stevenson DA, Moyer-Mileur LJ, Murray M, Slater H, Sheng X, Carey JC, Dube B, Viskochil DH. Bone mineral density in children and adolescents with neurofibromatosis type 1. J Pediatr 2007;150:83–8. [9] Duman O, Ozdem S, Turkkahraman D, Olgac ND, Gungor F, Haspolat S. Bone metabolism markers and bone mineral density in children with neurofibromatosis type-1. Brain Dev 2008;30:584–8. [10] Brunetti-Pierri N, Doty SB, Hicks J, Phan K, Mendoza-Londono R, Blazo M, Tran A, Carter S, Lewis RA, Plon SE, Phillips WA, O'Brian Smith E, Ellis KJ, Lee B. Generalized metabolic bone disease in neurofibromatosis type I. Mol Genet Metab 2008;94:105–11. [11] Seitz S, Schnabel C, Busse B, Schmidt HU, Beil FT, Friedrich RE, Schinke T, Mautner VF, Amling M. High bone turnover and accumulation of osteoid in patients with neurofibromatosis 1. Osteoporos Int 2010;21:119–27.
803
[12] Tucker T, Schnabel C, Hartmann M, Friedrich RE, Frieling I, Kruse HP, Mautner VF, Friedman JM. Bone health and fracture rate in individuals with neurofibromatosis 1 (NF1). J Med Genet 2009;46:259–65. [13] Stevenson DA, Yan J, He Y, Li H, Liu Y, Zhang Q, Jing Y, Guo Z, Zhang W, Yang D, Wu X, Hanson H, Li X, Staser K, Viskochil DH, Carey JC, Chen S, Miller L, Roberson K, Moyer-Mileur L, Yu M, Schwarz EL, Pasquali M, Yang FC. Multiple increased osteoclast functions in individuals with neurofibromatosis type 1. Am J Med Genet A 2011;155A:1050–9. [14] Glover SJ, Gall M, Schoenborn-Kellenberger O, Wagener M, Garnero P, Boonen S, Cauley JA, Black DM, Delmas PD, Eastell R. Establishing a reference interval for bone turnover markers in 637 healthy, young, premenopausal women from the United Kingdom, France, Belgium, and the United States. J Bone Miner Res 2009;24:389–97. [15] Vasikaran S, Eastell R, Bruyère O, Foldes AJ, Garnero P, Griesmacher A, McClung M, Morris HA, Silverman S, Trenti T, Wahl DA, Cooper C, Kanis JA, Group I-IBMSW. Markers of bone turnover for the prediction of fracture risk and monitoring of osteoporosis treatment: a need for international reference standards. Osteoporos Int 2011;22:391–420. [16] Reszka AA, Rodan GA. Mechanism of action of bisphosphonates. Curr Osteoporos Rep 2003;1:45–52. [17] Rogers MJ. New insights into the molecular mechanisms of action of bisphosphonates. Curr Pharm Des 2003;9:2643–58. [18] Sato M, Grasser W, Endo N, Akins R, Simmons H, Thompson DD, Golub E, Rodan GA. Bisphosphonate action. Alendronate localization in rat bone and effects on osteoclast ultrastructure. J Clin Invest 1991;88:2095–105. [19] Fujita H, Utsumi T, Muranaka S, Ogino T, Yano H, Akiyama J, Yasuda T, Utsumi K. Involvement of Ras/extracellular signal-regulated kinase, but not Akt pathway in risedronate-induced apoptosis of U937 cells and its suppression by cytochalasin B. Biochem Pharmacol 2005;69:1773–84. [20] Bergstrom JD, Bostedor RG, Masarachia PJ, Reszka AA, Rodan G. Alendronate is a specific, nanomolar inhibitor of farnesyl diphosphate synthase. Arch Biochem Biophys 2000;373:231–41. [21] Dunford JE, Rogers MJ, Ebetino FH, Phipps RJ, Coxon FP. Inhibition of protein prenylation by bisphosphonates causes sustained activation of Rac, Cdc42, and Rho GTPases. J Bone Miner Res 2006;21:684–94. [22] Scheven B, Visser J, Nijweide P. In vitro osteoclast generation from different bone marrow fractions, including a highly enriched haematopoietic stem cell population. Nature 1986;321:79–81. [23] Massey H, Flanagan A. Human osteoclasts derive from CD14-positive monocytes. Br J Haematol 1999;106:167–70. [24] Tiedemann K, Hussein O, Sadvakassova G, Guo Y, Siegel P, Komarova S. Breast cancer-derived factors stimulate osteoclastogenesis through the Ca2+/protein kinase C and transforming growth factor-beta/MAPK signaling pathways. J Biol Chem 2009;284:33662–70. [25] Yang F, Chen S, Robling A, Yu X, Nebesio T, Yan J, Morgan T, Li X, Yuan J, Hock J, Ingram D, Clapp D. Hyperactivation of p21ras and PI3K cooperate to alter murine and human neurofibromatosis type 1-haploinsufficient osteoclast functions. J Clin Invest 2006;116:2880–91. [26] Heervä E, Alanne MH, Peltonen S, Kuorilehto T, Hentunen T, Väänänen K, Peltonen J. Osteoclasts in neurofibromatosis type 1 display enhanced resorption capacity, aberrant morphology, and resistance to serum deprivation. Bone 2010;47:583–90. [27] Ma J, Li M, Hock J, Yu X. Hyperactivation of mTOR critically regulates abnormal osteoclastogenesis in neurofibromatosis type 1. J Orthop Res 2011. [28] Schindeler A, Birke O, Yu NY, Morse A, Ruys A, Baldock PA, Little DG. Distal tibial fracture repair in a neurofibromatosis type 1-deficient mouse treated with recombinant bone morphogenetic protein and a bisphosphonate. J Bone Joint Surg Br 2011;93:1134–9. [29] Stumpf D, Annergers J, Brown S, Conneally P, Housman D, Leppert M, Miller J, Moss M, Pileggi A, Rapin I, Strohman R, Swanson L, Zimmersman A. Neurofibromatosis. Conference statement. National Institutes of Health Consensus Development Conference. Arch Neurol 1988;45:575–8. [30] Consensus NIH. Development Panel on Osteoporosis Prevention, Diagnosis, and Therapy. Osteoporosis prevention, diagnosis, and therapy. JAMA 2001;285: 785–95. [31] Lewiecki EM, Gordon CM, Baim S, Binkley N, Bilezikian JP, Kendler DL, Hans DB, Silverman S, Bishop NJ, Leonard MB, Bianchi ML, Kalkwarf HJ, Langman CB, Plotkin H, Rauch F, Zemel BS. Special report on the 2007 adult and pediatric Position Development Conferences of the International Society for Clinical Densitometry. Osteoporos Int 2008;19:1369–78. [32] Hentunen TA, Reddy SV, Boyce BF, Devlin R, Park HR, Chung H, Selander KS, Dallas M, Kurihara N, Galson DL, Goldring SR, Koop BA, Windle JJ, Roodman GD. Immortalization of osteoclast precursors by targeting Bcl-XL and Simian virus 40 large T antigen to the osteoclast lineage in transgenic mice. J Clin Invest 1998;102:88–97. [33] Chopin F, Biver E, Funck-Brentano T, Bouvard B, Coiffier G, Garnero P, Thomas T. Prognostic interest of bone turnover markers in the management of postmenopausal osteoporosis. Joint Bone Spine 2011;30. [34] Adami S, Isaia G, Luisetto G, Minisola S, Sinigaglia L, Gentilella R, Agnusdei D, Iori N, Nuti R, Group IS. Fracture incidence and characterization in patients on osteoporosis treatment: the ICARO study. J Bone Miner Res 2006;21:1565–70. [35] Black DM, Delmas PD, Eastell R, Reid IR, Boonen S, Cauley JA, Cosman F, Lakatos P, Leung PC, Man Z, Mautalen C, Mesenbrink P, Hu H, Caminis J, Tong K, Rosario-Jansen T, Krasnow J, Hue TF, Sellmeyer D, Eriksen EF, Cummings SR, Trial HPF. Onceyearly zoledronic acid for treatment of postmenopausal osteoporosis. N Engl J Med 2007;356:1809–22.
Contents lists available at SciVerse ScienceDirect
Bone journal homepage: www.elsevier.com/locate/bone
Original Full Length Article
Osteoclasts derived from patients with neurofibromatosis 1 (NF1) display insensitivity to bisphosphonates in vitro Eetu Heervä a, Sirkku Peltonen b, Erkki Svedström c, Hannu T. Aro d, Kalervo Väänänen a, Juha Peltonen a,⁎ a
University of Turku, Department of Cell biology and Anatomy, Turku 20520, Finland Department of Dermatology, Turku University Hospital and University of Turku, Turku 20521, Finland Department of Diagnostic Imaging, Medical Imaging Centre of Southwest Finland, University of Turku, Turku 20521, Finland d Orthopaedic Research Unit, Turku University Hospital, Turku 20520, Finland b c
a r t i c l e
i n f o
Article history: Received 29 November 2011 Accepted 16 December 2011 Available online 27 December 2011 Edited by: Bjorn Olsen Keywords: BTM Fracture Farnesyl thiosalicylic acid Neurofibromatosis Osteoporosis
a b s t r a c t A total of 20 patients with neurofibromatosis 1 (NF1) were screened for NF1-related osteoporosis, and blood samples were collected for isolation of peripheral blood osteoclast progenitors. Patients with NF1 had higher levels of serum bone turnover markers (CTX and PINP) compared to controls. In addition, persons with high bone resorption in vitro on average had high levels of serum CTX. Of the 20 patients with NF1, 15 had low bone mineral density (osteopenia/osteoporosis), but these 15 patients did not have marked risk factors for low bone mineral density. Thus, we recommend screening for osteoporosis to all adult patients with NF1. Our aim was also to characterize the effects of bisphosphonates on NF1 osteoclasts in vitro. NF1 osteoclasts and osteoclasts from healthy controls in vitro were treated with zoledronic acid, alendronate and clodronate. These bisphosphonates caused a marked reduction in the number of normal control osteoclasts in vitro, while only a slight change was observed in the number of NF1 osteoclasts. Ras-inhibitor FTS counteracted this NF1-related insensitivity to zoledronic acid, suggesting that Ras may play a role in this phenomenon. © 2011 Elsevier Inc. All rights reserved.
Introduction Neurofibromatosis type 1 (NF1) is an autosomal dominant neuro-cutaneous-skeletal syndrome with an incidence of ~1/3000 [1–3]. The protein product of NF1 gene, neurofibromin, functions as a Ras-GTPase activating protein (Ras-GAP) negatively regulating Ras signaling pathway [4]. In bone, NF1 mRNA and protein have been detected in human chondrocytes, osteoblasts and osteoclasts [5,6]. Low bone mineral density (BMD) is a common feature in NF1, and is found in both sexes and also in children and adolescents with NF1 (Table 1) [6–12]. Low levels of serum vitamin D, high levels of serum parathyroid hormone, and increased collagen degradation products in urine have been reported in patients with NF1 [10–13]. In the subpopulation of patients with NF1 who have low BMD, high levels of serum parathyroid hormone, calcium, and tartrate resistant acid phosphatase 5b have been reported [12]. Analysis of whole-body subtotal age-
Abbreviations: NF1, neurofibromatosis type 1; BMD, bone mineral density; CTX, collagen type I C-terminal telopeptide; PINP, procollagen type I N-terminal propeptide; TRACP5b, tartrate resistant acid phosphatase 5b; MCSF, macrophage colony stimulating factor; RANKL, receptor activator of nuclear factor kappa-beta ligand; FTS, farnesyl thiosalicylic acid. ⁎ Corresponding author at: University of Turku, Department of Cell Biology and Anatomy, Kiinamyllynkatu 10, 20520 Turku, Finland. Fax: + 358 23337352. E-mail address: juha.peltonen@utu.fi (J. Peltonen). 8756-3282/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2011.12.011
matched BMD (Z-score), urine collagen degradation products and in vitro osteoclast formation capacity in 75 patients with NF1 aged 1–25 years showed no correlation between these parameters [13]. Bone biopsy histomorphometry of the patients with NF1 has revealed increased volume of osteoid, and increased number of both osteoclasts and osteoblasts compared to controls [10,11]. This suggests increased bone turnover in NF1, which can be assessed by bone turnover markers (BTMs), such as serum C-terminal cross-linking beta-telopeptide of type I collagen (CTX), tartrate resistant acid phosphatase 5b (TRACP5b), and the N-terminal propeptide of procollagen type I (PINP). CTX is representative of the catabolic rate of bone, TRACP5b has been used as an indicator of osteoclast activity, and PINP is secreted by osteoblasts during collagen synthesis, representing bone formation rate [14,15]. Amino and non-amino-bisphosphonates rapidly localize to bone and are used to treat osteoporosis. Non-amino-bisphosphonates, such as clodronate, generate toxic ATP analogues which disrupt the function of mitochondria, leading to apoptosis of osteoclasts [16–18]. Amino-bisphosphonates, such as alendronate and zoledronic acid, inhibit farnesyldiphosphate syntase, interfering with farnesylation of small GTPases including Ras, Rac and Rho. This leads to reduced ERK pathway signaling (downstream mediator of Ras), which in turn leads to mitochondrial membrane depolarization, activation of caspase cascade, and ultimately to cell death [18–21]. Osteoclasts are derived from haematopoietic progenitor cells, which can be isolated from peripheral blood samples [22,23]. In
E. Heervä et al. / Bone 50 (2012) 798–803 Table 1 Osteoporosis and osteopenia in neurofibromatosis 1 defined using dual energy X-ray absorptiometry. Study
Patients with NF1
Osteoporosis
Osteopenia
Kuorilehto (2005)
12 females, 14 males, aged 21–73 39 females, 45 males, aged 5–18 16 females, 16 males, aged 3–17 9 females, 5 males, aged 19–66 47 females, 26 males, aged 3–59 43 females, 29 males, aged 18–72 10 females, 10 males, aged 23–76
7 (27%)
13 (50%)
Stevenson (2007) Duman (2008) Seitz (2008) Brunetti-Pierri (2008) Tucker (2009) Current study
BMD reported lower compared to controls BMD reported lower compared to controls 8 (57%) Not reported 24 (32%)⁎
41 (57%)
14 (19%)
36 (50%)
4 (20%)
11 (55%)
⁎ Osteoporosis was defined as Z-score − 2.5 SD or less.
osteoclasts, Ras pathway functions as an anti-apoptotic and proosteoclastogenic pathway [24]. Osteoclasts derived from peripheral blood of patients with NF1 have been shown to be more numerous, resorb larger amounts of bone, display increased ERK and AKT activity, and tolerate serum-free conditions better compared to control osteoclasts [13,25–27]. Zoledronic acid has been shown to improve fracture healing in Nf1 +/− mice [28]. Since NF1 protein functions as a Ras-GAP and since the functions of amino-bisphosphonates are mediated via Ras pathway, the aim of this study was to characterize the effects of zoledronic acid, alendronate and clodronate in osteoclasts derived from peripheral blood of 20 NF1 patients and their controls. An additional aim was to characterize BMD and BTMs in these patients with NF1.
Subjects and methods NF1 patients A group of 21 Finnish Caucasian patients with NF1 from 19 families was recruited from the NF clinic of the Department of Dermatology, Turku University Hospital, Turku, Finland. In the statistical analyses we included 10 males aged 23–76 years (average 45) and 10 females aged 23–62 years (average 38). One 56-yearold female patient was excluded because she was medicated with alendronate. All patients fulfilled the NIH diagnostic criteria for NF1 [29]. This study was approved by Ethics Committee of Southwest Finland Hospital District, and the participants gave their written consents to BMD measurement, laboratory tests and osteoclast cultures. The study was carried out in Turku University Hospital and University of Turku during the years 2010–2011. Medical records of the patients were screened for the past 10 years. The body mass index of the patients was within normal range, three patients were smokers, and of 10 the female patients two were postmenopausal. Four patients were having vitamin D supplementation. Physical activity was evaluated by using a questionnaire [7]. No malignancies or other conditions known to affect bone were found, apart for alendronate in the one excluded patient, and lactose-intolerance in one patient. One patient with mild scoliosis and two patients with lytic bone lesions were found. Four patients had fractures during the past 10 years, which were located in ankle, distal radius, rib and patella. These fractures were noted in two patients with NF1-related osteoporosis and in two patients with osteopenia. In addition, the 56-year old female patient who was excluded from statistical analyses (see above) due to alendronate medication had a stress fracture of metatarsal bone during alendronate treatment.
799
Control persons A group of 20 healthy volunteers was recruited from Turku University personnel. Controls were matched for sex, age and menopausal status. Osteoclast progenitors were isolated from peripheral blood samples for cell culture experiments. CTX and PINP serum measurements were available from 18 controls. Bone densitometry A dual energy X-ray absorptiometry scan of lumbar spine (L1–L4) and left proximal femur was obtained, using a Hologic QDR 4500 densitometer (Hologic Inc., Waltham, MA, USA) for 11 patients, or an Osteocore 3 densitometer (Medilink Inc., Mauguio, France) for 8 patients, or a Lunar Prodigy densitometer (GE Healthcare, Madison, WI) for one patient, with adult standard measurement software. Corresponding T-scores and Z-scores were calculated using the equipment's standard software and Finnish Caucasian demographic databases provided by the manufacturers. The densitometers were calibrated each morning with a calibration phantom. The image quality and measurements were analyzed by an experienced radiologist (ES). The diagnosis of osteoporosis was made according to the NIH criteria with T-score −2.5 SD or below either in lumbar or femoral region, and the diagnosis of osteopenia with T-score between −1.0 SD and −2.5 SD [30,31]. Serum samples Laboratory analyses included total calcium, ionized calcium, inorganic phosphate, parathyroid hormone, 25-D3-vitamin, and alkaline phosphatase including its isoenzymes, and CTX. These measurements were purchased from Turku University Hospital, Turku, Finland, and carried out using Roche automatic analyzer (Roche diagnostics, Mannheim, Germany). TRACP5b was measured using BONETRAP immunofixation method, and analyses were purchased from Medix Laboratories, Helsinki, Finland. PINP measurements were purchased from Oulu University Hospital, Oulu, Finland, and were carried out using UniQ RIA analyzer (Orion, Espoo, Finland). All samples were collected in a similar way, and analyzed using a standardized protocol. In vitro analysis of osteoclast cultures In each case, osteoclast progenitors from 20 patients and 20 controls were isolated immediately after the patient's visit. Osteoclast progenitors were cultured as described initially by Yang et al. [25], with slight modifications [26]. Briefly, peripheral blood mononuclear cells were isolated using Ficoll-centrifugation (GE Healthcare, Uppsala, Sweden), and 0.5 million cells were seeded on bovine bone slices. Cells were cultured for 7 days in alpha-MEM (Gibco, NY) supplemented with 10% heat-inactivated fetal calf serum (Gibco, NY), receptor activator of nuclear factor kappa-beta ligand (RANKL, 20 ng/ml, Peprotech, Rocky Hill, NJ), macrophage colony stimulating factor (MCSF, 10 ng/ml, R&D systems, Minneapolis, MN) and antibiotics. This was followed by a 2-day culture with the same medium described above with RANKL and MCSF, and different bisphosphonates. We used amino-bisphosphonates zoledronic acid (Zometa ®, Novartis Finland, Finland) and alendronate (Sigma-Aldrich, Steinheim, Germany), and non-amino-bisphosphonate clodronate (Leiras, Finland). A Ras inhibitor Farnesyl Thiosalicylic Acid (FTS) 10E-6 M was purchased from Cayman Chemicals (Ann Arbor, MI) and dissolved in DMSO. For each drug 4 separate bone slices were prepared. Cells were fixed in formaldehyde and stained with tartrate resistant acid phosphatase staining kit (Sigma-Aldrich leukocyte acid phosphatase-kit, Steinheim, Germany), which is a direct osteoclastspecific staining method. Nuclei were visualized with Hoechst
800
E. Heervä et al. / Bone 50 (2012) 798–803
(Invitrogen, Eugene, OR) and actin rings with rhodamine-conjugated phalloidin (Cambrex, Walkersville, MD). TRACP-positive cells with 3 or more nuclei were considered as viable osteoclasts and those with fragmented nuclei and retracted cytoplasm as osteoclasts undergoing apoptosis [32]. Bone slices were imaged with Leica DMRB fluorescence microscope (Leica, Wetzlar, Germany). Osteoclasts were identified and counted from 6 approximately 0.5 × 0.5 mm rectangular areas selected in a routine manner [26].
Caspase-3 enzymatic activity Cells were cultured on bone as described above, with and without 10E-7 M zoledronic acid. A caspase-3 colorimetric assay kit (PromoCell, Heidelberg, Germany) was used to measure enzymatic activity of caspase-3, a marker of apoptosis. Cells were lysed and treated according to manufacturer's instructions. Absorbance was measured using Hidex Chameleon 96-plate reader (Hidex, Turku, Finland) at 405 nm.
Quantification of bone resorption in vitro We used quantification of cell culture media CTX instead of counting the resorption pits due to the fact, that CTX can be measured both before and after bisphosphonate treatment from the very same cultures. Levels of cell culture media CTX correlate with the number of resorption pits in the cultures [26]. CTX measurement was purchased from Turku University Hospital, Turku, Finland, and carried out using Roche automatic analyzer (Roche diagnostics, Mannheim, Germany).
Statistics The statistical data had normal distribution as estimated by Shapiro–Wilk analysis. Two-sample t-test was used to analyze the differences between patients and controls, males and females, or osteoporotic and non-osteoporotic groups. Correlation was calculated using linear regression analysis. In cell culture experiments the 20 patients and controls were matched for age, sex and menopausal status, and two-sample t-test was used. Paired t-test was used to compare treated and non-treated samples of the same person. The difference in the effect of different bisphosphonates between NF1 and control cell cultures was analyzed with two-sample t-test, followed by repeated measures analysis of variance. P-values less than 0.05 were considered statistically significant. All statistical analyses were made with SPSS for Windows version 19 (SPSS Inc, Chicago, IL).
Results Characterization of patients with NF1 BMD of lumbar spine (L1–L4) and proximal femur were obtained from the 20 patients with NF1. The lowest T-score of the NF1 patients was −1.7 SD ±1.2 SD on average, and the respective lowest Z-score −1.2 SD ±1.0 SD. Four out of 20 patients had osteoporosis and 11 patients had osteopenia (Table 1), consistent with the previous literature [30,31]. In addition, three patients had osteoporosis in one lumbar vertebra, but the average T-score of lumbar spine was between −2.1 SD and −2.2 SD, thus not fulfilling the diagnostic criteria for osteoporosis. Out of the 15 patients with low BMD (osteoporosis or osteopenia) three were smokers, one had lactose intolerance, two were postmenopausal women, and one male was aged 76 years. However, the remaining 8 patients with low BMD did not have any apparent risk factors for low BMD apart from NF1, and none of the 15 patients with low BMD had two or more risk factors. All 20 patients were also screened for selected serum values (Table 2). Elevated levels of parathyroid hormone were found in 4 patients, elevated levels of serum CTX in 8 patients, elevated levels of PINP in 3 patients and low levels of vitamin D were found in 2 patients. Other values were within reference range. Vitamin D levels were similar in winter and summer samples, possibly due to vitamin D supplementation. No gender or age-related differences were observed. The levels of serum total and ionized calcium correlated with the patients' T-scores. Thus, patients with NF1 and low BMD had higher levels of total and ionized serum calcium than the patients with normal BMD (p = 0.02, linear regression). These serum values were essentially similar as reported previously for larger cohorts [10,12]. Serum CTX and PINP have been proposed to be the first-line bone turnover markers [15], but have not been characterized in NF1 before. Patients with NF1 had higher levels of serum CTX and PINP compared to healthy controls (Table 2). Also, the CTX/PINP ratio was higher in patients with NF1 compared to controls (Table 2), possibly reflecting catabolic bone turnover rate in these patients [33]. No correlations between laboratory parameters and T- or Z-scores were found. Osteoclasts derived from patients with NF1 display insensitivity to zoledronic acid, alendronate and clodronate Osteoclast progenitors isolated from peripheral blood of 20 NF1 and 20 control persons were differentiated into osteoclasts and treated with different bisphosphonates for 2 days. Untreated NF1 samples yielded a higher number of osteoclasts compared to controls
Table 2 Serum values associated with bone health in the current study, presented as average ± SD. a
Serum measurements
NF1, n = 20
Reference values
Total calcium (mmol/l) Ionized calcium (mmol/l) Phosphate (mmol/l)
2.27 ± 0.09 1.24 ± 0.04 0.97 ± 0.17
Parathyroid hormone (ng/ml) Total alkaline phosphates (U/l) Bone-specific alkaline phosphatase (U/l) D-25-vitamin (nmol/l), no supplementation, n = 16 D-25-vitamin (nmol/l), supplementation, n = 4
53 ± 20 57 ± 18 27 ± 13 54 ± 22 60 ± 17
2.15–2.51 1.16–1.3 0.71–1.23 males 0.76–1.41 females 15–65 35–105 Below 69 Over 40 Over 40
Bone turnover markers CTX (μg/l) PINP (μg/l) CTX/PINP ratio TRACP5b (U/l) a b
Control, n = 18 0.29 ± 0.19 38 ± 16 b 6.2 ± 3.2 b n/a
b
NF1, n = 20 0.45 ± 0.17 52 ± 17 b 9.0 ± 4.1 b 2.9 ± 0.8
Reference values b
Given reference values are from the corresponding laboratory measurement manufacturer for adults. Value higher in patients with NF1 compared to controls, p b 0.05.
a
Below 0.57 premenopausal females 20–76 Below 5.3 for males, below 4.8 for females
E. Heervä et al. / Bone 50 (2012) 798–803
801
(p = 0.005). Thus, the drug reduced the number of osteoclasts in both NF1 and control samples, but had much weaker effect on the NF1 osteoclasts. It should be noted that, apoptotic osteoclasts with fragmented nuclei [32] were more frequent in control samples compared to NF1 samples treated with 10E-6 M zoledronic acid, 2.3% versus 0.3%, respectively (p = 0.04, n = 10 males). To assess the role of apoptosis, enzymatic activity of caspase-3 in five cell lysates was evaluated using an absorptiometry kit. Following treatment with 10E-7 M zoledronic acid, the caspase-3 levels were increased by +75% of baseline in control samples. The respective increase was only +11% in NF1 samples (Fig. 1E, p = 0.02). Due to limited number of available cells the effects of alendronate and clodronate were tested on osteoclasts derived from a subgroup of 4 male and 2 female NF1 patients and their corresponding controls. In analogy to findings on zoledronic acid, NF1 osteoclasts displayed insensitivity to 10E-6 M alendronate (p = 0.04, n = 6, Fig. 1F), and 10E-7 M clodronate (p = 0.05, n = 6, Fig. 1F). It should also be noted, that large multinuclear osteoclasts were present in NF1 samples treated with zoledronic acid (Fig. 2), and also with alendronate and clodronate. Addition of Ras inhibitor FTS counteracts the insensitivity to zoledronic acid observed in NF1 osteoclasts
Fig. 1. Effects of zoledronic acid (ZOL), alendronate (ALN) and clodronate (CLD) in the cell culture. The data is presented as average ± SD. (A) The number of viable osteoclasts in the cell culture with and without 10E-7 M zoledronic acid, 10 male patient/control pairs. (B) The number of viable osteoclasts in the cell culture with and without 10E7 M zoledronic acid, 10 female patient/control pairs. (C) The effect of 10E-6–10E-8 M zoledronic acid on the number of viable osteoclasts in the cell culture, 10 male patient/control pairs. (D) The same data as in (C), represented as percentages compared to untreated samples, 10 male patient/control pairs. (E) The proportional levels of caspase-3, a marker of apoptotic activity, after treatment with ZOL. Five male patient/control pairs pooled together. (F) The respective effects of 10E-6 M alendronate, 10E-7 M clodronate, 10E-7 M zoledronic acid, the combination of 10E-7 M zoledronic and 10E-6 M Ras inhibitor FTS (ZOL + FTS), and 10E-6 M Ras inhibitor FTS alone (FTS). 4 male and 2 female patient/control pairs pooled together. *p b 0.05. **p b 0.01.
(p = 0.001 males, p = 0.05 females, Figs. 1A and B), which is in agreement with previous findings [13,25,26]. No gender, age, or menopause-related differences were observed. First, the effect of 10E-7 M zoledronic acid was analyzed. On average, the number of viable control osteoclasts was reduced to half, representing the expected result of the drug. Surprisingly, a reduction of 25% only was noted in NF1 samples (p = 0.03 for males, p = 0.05 for females, two-sample t-test, Figs. 1A and B, and 2). Thus, a higher proportion and higher number of osteoclasts remained viable in NF1 samples treated with zoledronic acid compared to controls. Again, no age or gender-related differences were observed. Further analyses showed that the effect of zoledronic acid was dose-dependent, as tested in the 10 male samples treated with 10E8–10E-6 M zoledronic acid (Figs. 1C and D). Subsequently, the effect of zoledronic acid to number of osteoclasts was evaluated using a different statistical method (repeated measures ANOVA), and the effect of the drug was different between NF1 and control groups
Ras inhibitor FTS was added into the culture medium at the same time with zoledronic acid. 10E-6 M FTS alone did not have an apparent effect on the number of osteoclasts (Fig. 1F), as tested in samples from subgroup defined above in Osteoclasts derived from patients with NF1 display insensitivity to zoledronic acid, alendronate and clodronate. However, the combination of 10E-6 M FTS and 10E-7 M zoledronic acid had more profound effect than zoledronic acid alone. The number of osteoclasts was reduced 70% in both NF1 and control samples treated with the combination of zoledronic acid and FTS (Fig. 1F), and thus no difference was observed between NF1 and control samples treated with FTS and zoledronic acid. Zoledronic acid reduces frequency of actin rings in control but not in NF1 samples Actin rings, markers of active osteoclasts, were visualized by rhodamine-conjugated phalloidin staining. In osteoclasts derived from NF1 and control persons, 44–48% of osteoclasts had an actin ring. Zoledronic acid reduced the percentage of active control osteoclasts to 25% (p = 0.03), but no change was observed in the frequency of active NF1 osteoclasts as estimated by the presence of actin rings. Zoledronic acid prevents bone resorption in control but not in NF1 samples Total bone resorption in the cell culture was estimated by measuring levels of CTX from cell culture media of 20 patient/control pairs before (day 7) and after (day 9) the addition of 10E-7 M zoledronic acid. Culture medium CTX levels have been shown to correlate with the number of resorption pits in bone slice [26], and can be measured both before and after zoledronic acid in the very same cultures. CTX levels were readily detectable in both NF1 and control samples on day 7 indicating that the cultures harbored resorbing osteoclasts. Before addition of zoledronic acid (day 7), NF1 samples had higher levels of cell culture media CTX, 340 μg/l versus 580 μg/l, respectively (p = 0.003). NF1 samples also had higher CTX/osteoclast ratio compared to control samples (p = 0.05). No gender, age or menopause-related differences were observed. Treatment with 10E-7 M zoledronic acid for 2 days prevented marked increase in levels of cell culture media CTX in control samples (+130 μg/l, p = non-significant). However, a marked increase of CTX
802
E. Heervä et al. / Bone 50 (2012) 798–803
Fig. 2. Representative micrographs of osteoclasts in the cell culture. The panel shows typical NF1 and control samples, with and without 10E-7 M zoledronic acid. Viable osteoclasts are marked with dotted white lines. Arrows depict osteoclasts considered undergoing apoptosis, showing the fragmented nuclei or absence of nuclei. Scale bars: 50 μm.
(+500 μg/l, p = 0.0001) was noted in NF1 cell culture media. These results also indicate insensitivity of NF1 osteoclasts to zoledronic acid. Serum CTX correlates with osteoclast culture medium CTX Persons with high serum CTX often displayed high bone resorption in vitro. Specifically, the amount of CTX in untreated cell culture samples was considered as an indicator of bone resorption in vitro, and serum CTX as an estimate of catabolic rate of bone [15,26]. Serum and cell culture medium CTX levels correlated in the NF1 group (p = 0.03, linear regression, n = 20, coefficient of determination 30%). Similar results were obtained with both NF1 and control samples (n= 40). However, no correlations between serum CTX (and other laboratory parameters), osteoclast number in vitro, and T- or Z-scores were found. No gender or age-related differences were found. Discussion The in vitro exposure of normal control osteoclasts on bone slices to zoledronic acid, alendronate and clodronate induced a marked reduction in the number of osteoclasts. Also the addition of zoledronic acid prevented increase in cell culture media CTX levels in the control samples. These are expected findings and attest to the validity of the experimental setup. An unexpected finding of the current study was that NF1 osteoclasts displayed insensitivity to zoledronic acid, alendronate and clodronate, as estimated by the fact that the number of osteoclasts was only moderately reduced, and levels of cell culture CTX continued to rise in NF1 samples. NF1 osteoclasts have been shown to display an enhanced formation capacity [13,25,26], which could explain the high number of NF1 osteoclasts after bisphosphonate treatment. However, in NF1 samples treated with bisphosphonates, large multinuclear osteoclasts were noted. Thus, it seems unlikely that these large multinuclear
osteoclasts could have been formed during bisphosphonate exposure in vitro. In addition, the lower increase in caspase-3 activity in NF1 samples treated with zoledronic acid supports the view of insensitivity of NF1 osteoclasts to apoptotic stimuli. NF1 osteoclasts have been shown to display hyperactive Ras [13] and Ras has been shown as an anti-apoptotic pathway in osteoclasts derived from healthy controls [24]. Thus we hypothesize that hyperactive Ras in NF1 osteoclasts is the reason for high number of osteoclasts observed in NF1 samples treated with bisphosphonates. To test this hypothesis, Ras inhibitor FTS was used. FTS counteracted the insensitivity to zoledronic acid observed in NF1 osteoclasts. This suggests that hyperactive Ras provides anti-apoptotic effect against amino- and non-amino-bisphosphonates in NF1 osteoclasts. The combination of Ras-inhibitor FTS and zoledronic acid had more profound effect than zoledronic acid alone. Thus, it is feasible to speculate that osteoclasts may upregulate Ras to counteract apoptotic signals. The question thus emerges whether the effect of bisphosphonates is compromised in selected patients with NF1-related osteoporosis. The 56-year-old female patient who was excluded from the statistical analyses had suffered a stress fracture during alendronate treatment. In addition, her osteoclasts were insensitive to zoledronic acid in vitro. However, in general population fractures are reported in 20–30% of patients medicated with alendronate, and in 1–3% of those medicated with zoledronic acid, during follow-up of three years [34,35]. The limitation of this study is that the in vitro findings do not necessarily predict the effects of bisphosphonates in patients with NF1. It is possible that as NF1 patients are medicated with bisphosphonates for years, the cumulative exposure to the drugs overcomes the insensitivity to these drugs observed in vitro. An investigation on how often patients with NF1-related osteoporosis do not respond to bisphosphonate treatment is called for. In a previous study of a cohort of 75 patients with NF1 aged 1–25, no correlation between age-matched total body BMD excluding head
E. Heervä et al. / Bone 50 (2012) 798–803
(Z-score), number of osteoclasts in the cell culture and urine bone turnover markers was found [13]. Our results on bone resorption marker CTX from serum and osteoclast culture media show that these parameters correlate statistically significantly with each other, both in NF1 and control samples. Thus persons with high bone resorption in vitro on average have high levels of serum CTX. The differences in these two studies include the facts, that the patients in the current study are older and different bone turnover markers were measured. We suggest serum CTX as an interesting candidate for future studies identifying useful markers for bone health in NF1. Conclusions Osteoclasts isolated from patients with NF1 display in vitro insensitivity to zoledronic acid, alendronate and clodronate, possibly due to hyperactive Ras. The current and previous studies show that most of patients with NF1 and low BMD have no apparent risk factors for low BMD apart from NF1. Thus, we recommend that all adult patients with NF1 should be evaluated for possible osteoporosis. Conflicts of interest Hannu T. Aro has received institutional grants from Novartis and Eli Lilly. Acknowledgments The current study was financially supported by the Academy of Finland, Turku University Foundation, Foundation of Gerda & Ella Saarinen, Turku University Hospital EVO funding, Finnish Culture Foundation, and Emil Aaltonen Foundation. Jonas Nyman and Kaisa Ivaska are greatly acknowledged for helpful discussions. Tero Wahlberg is acknowledged for statistical assistance. Eetu Heervä is a member of Turku Graduate School of Clinical Sciences. References [1] Riccardi VM, Eichner JE. Neurofibromatosis: phenotype. Johns Hopkins University Press, Baltimore, MD, USA: Natural History and Pathogenesis; 1986. [2] Lammert M, Friedman J, Kluwe L, Mautner V. Prevalence of neurofibromatosis 1 in German children at elementary school enrollment. Arch Dermatol 2005;141:71–4. [3] Jouhilahti EM, Peltonen S, Heape AM, Peltonen J. The pathoetiology of neurofibromatosis 1. Am J Pathol 2011;178:1932–9. [4] Xu G, O'Connell P, Viskochil D, Cawthon R, Robertson M, Culver M, Dunn D, Stevens J, Gesteland R, White R. The neurofibromatosis type 1 gene encodes a protein related to GAP. Cell 1990;62:599–608. [5] Leskelä H, Kuorilehto T, Risteli J, Koivunen J, Nissinen M, Peltonen S, Kinnunen P, Messiaen L, Lehenkari P, Peltonen J. Congenital pseudarthrosis of neurofibromatosis type 1: impaired osteoblast differentiation and function and altered NF1 gene expression. Bone 2009;44:243–50. [6] Elefteriou F, Kolanczyk M, Schindeler A, Viskochil D, Hock J, Schorry E, Crawford A, Friedman J, Little D, Peltonen J, Carey J, Feldman D, Yu X, Armstrong L, Birch P, Kendler D, Mundlos S, Yang F, Agiostratidou G, Hunter-Schaedle K, Stevenson D. Skeletal abnormalities in neurofibromatosis type 1: approaches to therapeutic options. Am J Med Genet A 2009. [7] Kuorilehto T, Pöyhönen M, Bloigu R, Heikkinen J, Väänänen K, Peltonen J. 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 2005;16:928–36. [8] Stevenson DA, Moyer-Mileur LJ, Murray M, Slater H, Sheng X, Carey JC, Dube B, Viskochil DH. Bone mineral density in children and adolescents with neurofibromatosis type 1. J Pediatr 2007;150:83–8. [9] Duman O, Ozdem S, Turkkahraman D, Olgac ND, Gungor F, Haspolat S. Bone metabolism markers and bone mineral density in children with neurofibromatosis type-1. Brain Dev 2008;30:584–8. [10] Brunetti-Pierri N, Doty SB, Hicks J, Phan K, Mendoza-Londono R, Blazo M, Tran A, Carter S, Lewis RA, Plon SE, Phillips WA, O'Brian Smith E, Ellis KJ, Lee B. Generalized metabolic bone disease in neurofibromatosis type I. Mol Genet Metab 2008;94:105–11. [11] Seitz S, Schnabel C, Busse B, Schmidt HU, Beil FT, Friedrich RE, Schinke T, Mautner VF, Amling M. High bone turnover and accumulation of osteoid in patients with neurofibromatosis 1. Osteoporos Int 2010;21:119–27.
803
[12] Tucker T, Schnabel C, Hartmann M, Friedrich RE, Frieling I, Kruse HP, Mautner VF, Friedman JM. Bone health and fracture rate in individuals with neurofibromatosis 1 (NF1). J Med Genet 2009;46:259–65. [13] Stevenson DA, Yan J, He Y, Li H, Liu Y, Zhang Q, Jing Y, Guo Z, Zhang W, Yang D, Wu X, Hanson H, Li X, Staser K, Viskochil DH, Carey JC, Chen S, Miller L, Roberson K, Moyer-Mileur L, Yu M, Schwarz EL, Pasquali M, Yang FC. Multiple increased osteoclast functions in individuals with neurofibromatosis type 1. Am J Med Genet A 2011;155A:1050–9. [14] Glover SJ, Gall M, Schoenborn-Kellenberger O, Wagener M, Garnero P, Boonen S, Cauley JA, Black DM, Delmas PD, Eastell R. Establishing a reference interval for bone turnover markers in 637 healthy, young, premenopausal women from the United Kingdom, France, Belgium, and the United States. J Bone Miner Res 2009;24:389–97. [15] Vasikaran S, Eastell R, Bruyère O, Foldes AJ, Garnero P, Griesmacher A, McClung M, Morris HA, Silverman S, Trenti T, Wahl DA, Cooper C, Kanis JA, Group I-IBMSW. Markers of bone turnover for the prediction of fracture risk and monitoring of osteoporosis treatment: a need for international reference standards. Osteoporos Int 2011;22:391–420. [16] Reszka AA, Rodan GA. Mechanism of action of bisphosphonates. Curr Osteoporos Rep 2003;1:45–52. [17] Rogers MJ. New insights into the molecular mechanisms of action of bisphosphonates. Curr Pharm Des 2003;9:2643–58. [18] Sato M, Grasser W, Endo N, Akins R, Simmons H, Thompson DD, Golub E, Rodan GA. Bisphosphonate action. Alendronate localization in rat bone and effects on osteoclast ultrastructure. J Clin Invest 1991;88:2095–105. [19] Fujita H, Utsumi T, Muranaka S, Ogino T, Yano H, Akiyama J, Yasuda T, Utsumi K. Involvement of Ras/extracellular signal-regulated kinase, but not Akt pathway in risedronate-induced apoptosis of U937 cells and its suppression by cytochalasin B. Biochem Pharmacol 2005;69:1773–84. [20] Bergstrom JD, Bostedor RG, Masarachia PJ, Reszka AA, Rodan G. Alendronate is a specific, nanomolar inhibitor of farnesyl diphosphate synthase. Arch Biochem Biophys 2000;373:231–41. [21] Dunford JE, Rogers MJ, Ebetino FH, Phipps RJ, Coxon FP. Inhibition of protein prenylation by bisphosphonates causes sustained activation of Rac, Cdc42, and Rho GTPases. J Bone Miner Res 2006;21:684–94. [22] Scheven B, Visser J, Nijweide P. In vitro osteoclast generation from different bone marrow fractions, including a highly enriched haematopoietic stem cell population. Nature 1986;321:79–81. [23] Massey H, Flanagan A. Human osteoclasts derive from CD14-positive monocytes. Br J Haematol 1999;106:167–70. [24] Tiedemann K, Hussein O, Sadvakassova G, Guo Y, Siegel P, Komarova S. Breast cancer-derived factors stimulate osteoclastogenesis through the Ca2+/protein kinase C and transforming growth factor-beta/MAPK signaling pathways. J Biol Chem 2009;284:33662–70. [25] Yang F, Chen S, Robling A, Yu X, Nebesio T, Yan J, Morgan T, Li X, Yuan J, Hock J, Ingram D, Clapp D. Hyperactivation of p21ras and PI3K cooperate to alter murine and human neurofibromatosis type 1-haploinsufficient osteoclast functions. J Clin Invest 2006;116:2880–91. [26] Heervä E, Alanne MH, Peltonen S, Kuorilehto T, Hentunen T, Väänänen K, Peltonen J. Osteoclasts in neurofibromatosis type 1 display enhanced resorption capacity, aberrant morphology, and resistance to serum deprivation. Bone 2010;47:583–90. [27] Ma J, Li M, Hock J, Yu X. Hyperactivation of mTOR critically regulates abnormal osteoclastogenesis in neurofibromatosis type 1. J Orthop Res 2011. [28] Schindeler A, Birke O, Yu NY, Morse A, Ruys A, Baldock PA, Little DG. Distal tibial fracture repair in a neurofibromatosis type 1-deficient mouse treated with recombinant bone morphogenetic protein and a bisphosphonate. J Bone Joint Surg Br 2011;93:1134–9. [29] Stumpf D, Annergers J, Brown S, Conneally P, Housman D, Leppert M, Miller J, Moss M, Pileggi A, Rapin I, Strohman R, Swanson L, Zimmersman A. Neurofibromatosis. Conference statement. National Institutes of Health Consensus Development Conference. Arch Neurol 1988;45:575–8. [30] Consensus NIH. Development Panel on Osteoporosis Prevention, Diagnosis, and Therapy. Osteoporosis prevention, diagnosis, and therapy. JAMA 2001;285: 785–95. [31] Lewiecki EM, Gordon CM, Baim S, Binkley N, Bilezikian JP, Kendler DL, Hans DB, Silverman S, Bishop NJ, Leonard MB, Bianchi ML, Kalkwarf HJ, Langman CB, Plotkin H, Rauch F, Zemel BS. Special report on the 2007 adult and pediatric Position Development Conferences of the International Society for Clinical Densitometry. Osteoporos Int 2008;19:1369–78. [32] Hentunen TA, Reddy SV, Boyce BF, Devlin R, Park HR, Chung H, Selander KS, Dallas M, Kurihara N, Galson DL, Goldring SR, Koop BA, Windle JJ, Roodman GD. Immortalization of osteoclast precursors by targeting Bcl-XL and Simian virus 40 large T antigen to the osteoclast lineage in transgenic mice. J Clin Invest 1998;102:88–97. [33] Chopin F, Biver E, Funck-Brentano T, Bouvard B, Coiffier G, Garnero P, Thomas T. Prognostic interest of bone turnover markers in the management of postmenopausal osteoporosis. Joint Bone Spine 2011;30. [34] Adami S, Isaia G, Luisetto G, Minisola S, Sinigaglia L, Gentilella R, Agnusdei D, Iori N, Nuti R, Group IS. Fracture incidence and characterization in patients on osteoporosis treatment: the ICARO study. J Bone Miner Res 2006;21:1565–70. [35] Black DM, Delmas PD, Eastell R, Reid IR, Boonen S, Cauley JA, Cosman F, Lakatos P, Leung PC, Man Z, Mautalen C, Mesenbrink P, Hu H, Caminis J, Tong K, Rosario-Jansen T, Krasnow J, Hue TF, Sellmeyer D, Eriksen EF, Cummings SR, Trial HPF. Onceyearly zoledronic acid for treatment of postmenopausal osteoporosis. N Engl J Med 2007;356:1809–22.