Molecular Genetics and Metabolism 111 (2014) 41–45

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism journal homepage: www.elsevier.com/locate/ymgme

Decreased bone mineral density in Costello syndrome Chiara Leoni a,b, David A. Stevenson b, Lucilla Martini a, Roberto De Sanctis a, Giovanna Mascolo a, Francesca Pantaleoni c, Sara De Santis a, Ilaria La Torraca a, Silvia Persichilli d, Paolo Caradonna e, Marco Tartaglia c, Giuseppe Zampino a,⁎ a

Center for Rare Diseases, Department of Pediatrics, Università Cattolica del Sacro Cuore, Rome, Italy Division of Medical Genetics, Department of Pediatrics, University of Utah, UT, USA Department of Hematology, Oncology and Molecular Medicine, Istituto Superiore di Sanità, Rome, Italy d Institute of Biochemistry and Clinical Biochemistry, Università Cattolica del Sacro Cuore, Rome, Italy e Department of Internal Medicine, Università Cattolica del Sacro Cuore, Rome, Italy b c

a r t i c l e

i n f o

Article history: Received 26 June 2013 Received in revised form 11 August 2013 Accepted 11 August 2013 Available online 16 August 2013 Keywords: Costello syndrome HRAS Bone mineral density RASopathies NF1 Noonan syndrome

a b s t r a c t Introduction: Costello syndrome (CS) is a multisystemic disorder characterized by postnatal reduced growth, facial dysmorphism, cardiac defects, cognitive impairment, skin and musculo-skeletal anomalies, and predisposition to certain cancers. CS is caused by activating germline mutations in the HRAS proto-oncogene. Similar to what is observed in other RASopathies, CS causative HRAS mutations promote enhanced signal flow through the RAF–MEK–ERK and PI3K–AKT signaling cascades. While decreased bone mineralization has been documented in other RASopathies, such as neurofibromatosis type 1 and Noonan syndrome, systematic studies investigating bone mineral density (BMD) are lacking in CS. Materials and methods: Dual-energy X-ray absorptiometry (DXA) was utilized to assess BMD and body composition (fat and fat-free mass) in a cohort of subjects with molecularly confirmed diagnosis of CS (n = 9) and age-matched control individuals (n = 29). Using general linear regression, subtotal body (total body less head), lumbar, femoral neck and femur BMD parameters were compared considering age, sex, body mass index (BMI) and Tanner stage. Blood and urine biomarkers of bone metabolism were also assessed. Results: All individuals with CS showed significantly lower mean values of subtotal, lumbar and femoral neck BMD compared to the control group (p ≤ 0.01). Similarly, mean total body mass and fat-free mass parameters were lower among the CS patients than in controls (p b 0.01). Low 25-OH vitamin D concentration was documented in all individuals with CS, with values below the reference range in two patients. No significant correlation between vitamin D levels and BMD parameters was observed. Discussion: CS belongs to a family of developmental disorders, the RASopathies, that share skeletal defects as a common feature. The present data provide evidence that, similar to what is recently seen in NF1 and NS, bone homeostasis is impaired in CS. The significant decrease in BMD and low levels of vitamin D documented in the present cohort, along with the risk for pathologic fractures reported in adult individuals with CS, testifies the requirement for a preventive treatment to alleviate evolutive complications resulting from dysregulated bone metabolism. © 2013 Elsevier Inc. All rights reserved.

1. Introduction Costello syndrome (CS) (OMIM #218040) is a developmental disorder characterized by distinctive facial features, postnatal reduced growth, cardiac defects and hypertrophic cardiomyopathy, intellectual disability, skin–muscle–skeletal anomalies and predisposition for Abbreviations: CS, Costello syndrome; NS, Noonan syndrome; CFCS, cardiofaciocutaneous syndrome; NF1, neurofibromatosis type 1; DXA, dual-energy X-ray absorptiometry; BMD, bone mineral density; S-BMD, subtotal bone mineral density; WBLH, whole body less head; L-BMD, lumbar bone mineral density; FN-BMD, femoral neck bone mineral density; F-BMD, femur bone mineral density. ⁎ Corresponding author at: Center for Rare Diseases, Department of Pediatrics, Università Cattolica del Sacro Cuore, Largo Agostino Gemelli, 8, 00168 Rome, Italy. E-mail address: [email protected] (G. Zampino). 1096-7192/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ymgme.2013.08.007

certain rare cancers during childhood [1]. CS is a rare condition, with approximately 300 individuals reported worldwide [2]. It is caused by heterozygous activating mutations in HRAS, which encodes one member of a small family of monomeric GTPases functioning as a major node in intracellular signaling [3]. RAS signaling is implicated in several cellular functions, including cell fate determination, proliferation, survival, and differentiation, and within this network, signal flow through the RAF–MEK–ERK pathway mediates early and late developmental processes controlling morphology determination, organogenesis, synaptic plasticity and growth. Besides the role played by RAF–MEK–ERK pathway in oncogenesis [4], this intracellular pathway has recently been recognized as the cause of an emerging family of clinically related disorders known as RASopathies. This group of disorders includes an increasing number of conditions

42

C. Leoni et al. / Molecular Genetics and Metabolism 111 (2014) 41–45

among which Noonan syndrome (NS) and neurofibromatosis type 1 (NF1) are the most common [5]. These are caused by heterozygous mutations in genes encoding RAS proteins, regulators of RAS function and modulators of RAS interaction with downstream signal transducers effectors [6]. Several lines of evidence support the relevant role of RAS signaling in bone homeostasis and modeling [7–9]. Orthopedic problems represent an important diagnostic element of CS and related RASopathies. Bone mineral density (BMD) and skeletal abnormalities have been well studied in patients with NF1 [10–19], and recently investigated in Noonan syndrome [20]. Skeletal involvement in CS has been clinically reported in a small cohort of patients [7,21–23]. White et al. described the adult phenotype of CS, and reported six patients with osteoporosis and two with osteopenia. Stevenson et al. reported clinical BMD parameters from different body regions, using unadjusted z-score measurements in 7 affected cases. Dual energy X-ray absorptiometry (DXA) measurements in CS, however, have not yet been systematically examined with appropriate controls. In the present study, we investigated BMD in a cohort of individuals with CS compared to a control group controlling for potential confounders such as sex, age, body mass index (BMI) and Tanner stage. Biochemical markers of bone metabolism in individuals with CS are also provided. 2. Materials and methods Ten individuals with a clinical and molecular diagnosis of CS (7 F; 3 M) admitted to the Center for Rare Diseases of Catholic University (Rome, Italy) for regular clinical follow-up were recruited for the study. Informed written consent was obtained. DXA imaging was not performed on individuals b 5 years of age. A cohort of 29 healthy controls (18 F; 11 M) was selected from the DXA studies database of healthy subjects of the Department of Internal Medicine (Catholic University, Rome) using age and sex as selection criteria. Age range in the control group was 5.3 to 29.6 years (mean ± SD = 10.3 ± 5.3 years). All patients were examined by experienced pediatricians and a physical therapist. Growth measurements were evaluated by standard clinical balance (weight) and wall-stadiometer (height). Tanner stage [24,25] was assessed in both individuals with CS and controls. BMI was calculated using the standard formula: weight/height2 (kg/m2). No subject participating in the study had history of bone pain or evidence of bone fractures in the year prior to enrollment for BMD and biomarker evaluation. Family history of osteoporosis and non-traumatic fractures was collected in all individuals. Physical activity was evaluated qualitatively, counting the amount of hours spent playing sports or doing physical therapy per week, over a time frame of three months prior to the study. Body composition consisting of fat free mass (FFM) and fat mass (FM) as well as bone mineral density parameters were obtained by dual-energy X-ray absorptiometry scans (MOC-DXA, Lunar Prodigy Advance, GE Healthcare). Whole body scans were obtained with the patient supine, arms placed at one's side with palms down. Subtotal body measurements were evaluated by subtracting the head region from the rest of the body. The bone parameters evaluated by DXA in the studied were: subtotal bone mineral density (whole body less head) (S-BMD g/cm2), lumbar bone mineral density (L-BMD g/cm2), femoral neck bone mineral density (FN-BMD g/cm2), and femur bone mineral density (F-BMD g/cm2). Markers of bone metabolism were obtained from blood (i.e., calcium, phosphorus, magnesium, 25-hydroxy (25-OH) vitamin D, parathyroid hormone, serum c-terminal telopeptide of the β-1 chain of type 1 collagen and IGF1), and 24 hour urine collection (i.e., calcium, phosphorus, creatinine, and phosphate). 25-OH vitamin D concentrations were analyzed by automated chemiluminescence immunoassay, using a Liaison analyzer (DiaSorin-RIA). The lower detection value

for this method was 7.0 ng/ml, with a total CV (coefficient of variation) lower than 10% [26,27]. Serum 25-OH vitamin D concentrations were categorized into 3 classes: optimal ≥ 30 ng/ml, insufficient 20– 29 ng/ml, and deficient ≤ 20 ng/ml. Dietary calcium intake was assessed by 7-day diet record evaluation [28]. A general linear model was used to compare the bone parameters between cases and controls, while controlling for potential confounders (i.e., age, sex, BMI, and Tanner stage). Significant values were set at p = 0.05. Pearson correlation was used to assess the relationship between 25-OH vitamin D concentrations, and S-BMD and L-BMD. 3. Results Thirty-nine individuals were enrolled in the study. Age range in the CS cohort (n = 10) was 2.2 to 28.9 years, whereas in the control group (n = 29) it was 5.3 to 29.6 years (Table 1). Within the CS cohort, all subjects fulfilled the criteria for a clinical diagnosis of CS and harbored disease causative mutations in HRAS. Specifically, the common c.34G NA missense change (p.Gly12Ser) was documented in 9 cases, while one individual was heterozygous for the less common c.37G NT substitution (p.Gly13Cys) (Table 2). One individual was younger than 5 years of age; for this reason she did not undergo DXA imaging, but did have biochemical markers of bone metabolism obtained. As expected, height and weight were significantly lower in the CS cohort compared to the control group (Table 1). All individuals were ambulant, even though multiple skeletal abnormalities, such as scoliosis, pes planus, and tight heel cords were present. Level of physical activity was annotated, together with any other significant sign or feature during physical examination (Table 2). In particular skin findings such as skin hyperpigmentation, acanthosis nigricans and hyperkeratosis were described in Table 2 and clinically compared with vit D concentrations. The review of medical records documented that CS individual #8 was on treatment with valproate; similarly, CS individual #10 was on statin and thyroid therapy due to hypercholesterolemia and hypothyroidism, respectively. Both individuals exhibited delayed puberty (ID#8 at 14 y of age and ID#10 showed delayed menarche at 19 y old but she started to have signs of puberty at 10 y of age). There was no family history of osteoporosis or non-traumatic fractures among participants. DXA analysis documented significantly lower S-BMD, L-BMD, and FN-BMD values in the CS cohort compared to the control group (Table 3), while no statistically significant differences between cohorts in femur BMD and fat mass measures were observed. There was no statistically significant correlation between 25-OH vitamin D concentrations and S-BMD measures (p = 0.11) or L-BMD parameters (p = 0.12). Evaluation of the biochemical markers of bone metabolism in both blood and urine samples documented that serum phosphorus levels were below the assay reference interval in 4/10 individuals (range 4–7 mg/dl), while serum 25-OH vitamin D was insufficient in 3/10, deficient in 5/10, and lower than the cutoff of 7 ng/ml in 2/10 (Table 4). No information on sunlight exposure by standardized questionnaire was collected, ruling out the possibility of exploring correlations between vitamin D concentrations and hours of sunlight

Table 1 Clinical characteristics. Subjects

Costello syndrome (n = 10)

Controls (n = 29)

Gender Age (y) Weight (kg) Height (cm) BMI (kg/m2) Tanner stage

7 F/3 M 13.13 ± 8.31 25.53 ± 13.09 116.13 ± 25.18 17.47 ± 2.50 4 F/stage I 2 M/stage I 3 F/stage IV 1 M/stage IV

11 M/18 F 10.38 ± 5.35 35.26 ± 14.67 134.28 ± 18.56 18.76 ± 3.02 8 F/stage I 10 M/stage I 10 F/stages II–V 1 M/stage V

C. Leoni et al. / Molecular Genetics and Metabolism 111 (2014) 41–45

43

Table 2 Clinical findings and molecular diagnosis in individuals with Costello syndrome. ID

Gender

1 2

F M

3 4

Age

HRAS amino acid change

Skeletal anomalies

Skin hyperpigmentation

Acanthosis nigricans

Hyperkeratosis

25-OH vit D ng/ml

SBMD-WBLH z-score

L-BMD z-score

Weekly physical activitya

2.2 5.2

Gly12Ser Gly13Cys

PC, T, PP PC, T, PP

− +

− +

+ +

26 12.5

NA −1.2

NA −1.4

M F

7.8 8.3

Gly12Ser Gly12Ser

PC, H, T, PP PC, T, PP

+ +

− −

+ −

b7 24.3

−4.3 −2.6

−4.4 −1.6

5

F

9.3

Gly12Ser

PC, T, PP

+

+

+

18.4

−1.2

−1.8

6

F

11.6

Gly12Ser

PC, S, T, PP

+

+

+

18.7

−1.6

−1.9

7 8 9 10

F F M F

19.1 18.5 20.6 28.9

Gly12Ser Gly12Ser Gly12Ser Gly12Ser

PC, S, T, PP S, T, PP PC, S, T, PP PC, S, T, PP

+ + + +

+ + + −

+ + + +

28.1 b7 19.3 18.7

−3.1 −3.7 −1.2 −2.2

−3.4 −3.1 −0.9 −3.0

2 h PT 2 h PT/1 h SPA 3 h PT 1 h SPA/2 h swimming 1 h SPA/1 h swimming 1 h PT/1 h SPA/1 h swimming 1 h SPA 3 h PT 3 h karate 3 h modern dance

PC, pectus carinatum; S, scoliosis; H, hip dysplasia; T, tight heel cord; PP, pes planus. NA: not assessed. SBMD-WBLH: subtotal body mineral density of whole body less head; L-BMD, lumbar bone mineral density. a Physical activity: total number of hours spent per week in physical therapy (PT), sport at school (SPA, school physical activity) or playing individual sports.

exposure. However, 6/10 patients stated frequent exposure to the sunlight (both during summer and winter seasons). The concentration of serum c-terminal telopeptide of the β-1 chain of type 1 collagen (β-CTX) was above the normal assay reference interval for age in one affected subject and at the upper normal reference in an additional two individuals with CS. IGF1 was lower than normal values in all affected individuals, with two of them having values below the lower limit detected by the assay (25 ng/ml value). The 24 hour urine collection showed increased calcium excretion in one individual, increased phosphorus excretion in seven individuals, and tubular phosphate reabsorption above the normal range in five individuals. Mean daily calcium intake was 732 ± 454 mg/day, with 50% of individuals showing low calcium intake compared to Italian Level of Recommend Nutrients [29]. 4. Discussion Recent studies have explored the impact of RAS/MAPK pathway signaling dysregulation on skeletal development and the role of this signaling cascade on bone homeostasis and modeling [7–9]. NF1 has particularly provided relevant insights [14–16,30]. NF1 mouse models consistently have documented occurrence of skeletal defects as a result of abnormalities in both osteoblastic [9,31] and osteoclastic [32] cell lineages, leading to decrease bone stiffness, reduction in the mineral content, increased degree of porosity and decreased stability of cortical bone. Clinically, individuals affected by RASopathies have an overlapping musculoskeletal phenotype characterized by spine curvature (scoliosis, kyphosis), anterior wall chest deformity (pectus carenatus/excavatum), tight tendons and contractures (elbow, wrist, hip, knee, ankle) and feet abnormalities (pes planus/cavus, toe crowding) [19,21,22]. In

addition, osteoporosis has been observed in NF1, NS, anecdotally in cardiofaciocutaneous syndrome (CFCS), and now in CS. Increased bone resorption as measured by urine pyridinium crosslinks (pyridinoline and deoxypyridinoline) has been demonstrated in NF1 [8] as well as in other RASopathies (CFCS, CS, and NS), suggesting abnormalities in osteoclast function [7]. However, studies examining in vivo and in vitro bone cellular function in CS and CFCS are needed. Our data provide further support that decreased BMD is a frequent finding in CS, with a significant reduction in the subtotal body, lumbar and femoral neck. Interestingly, in our cohort it appeared that lumbar BMD represented one of the most significantly decreased value from the DXA imaging which may predispose to the development of non-dystrophic scoliosis frequently seen in Costello syndrome. We think that the orthopedic abnormalities detected in our cohort of the Costello individuals, did not significantly limit them in participation of physical activity such as swimming, karate either dance. Given the significant short stature in CS, standard z-scores from DXA are likely not optimal in evaluating BMD. Size should be taken into account when evaluating BMD in these subjects on a clinical basis as decreases in BMD may be overestimated due to differences in bone size [33]. Future studies with additional participants will allow control for other factors such as height, weight, activity level, calcium intake, vitamin D concentrations and co-morbidities that would be helpful to clarify the etiology of the decreased BMD. Although we detected low serum 25-OH vitamin D concentrations in all individuals with CS, a direct correlation with BMD could not be established. Bone homeostasis is part of a complex balance of physiologic and environmental aspects. The former include bone cellular function (e.g., osteoblast and osteoclast activity), extracellular matrix integrity (collagen and mesenchymal stromal cells), and vitamin D homeostasis (regulating excretion and reabsorption of electrolytes such as calcium, phosphorus, magnesium). Environmental factors, such as sunlight exposure, geographical latitude, diet and physical activity, also play an

Table 3 Dual energy X-ray absorptiometry (DXA) scan parameters. Densitometry

Costello syndrome Controls (n = 29) p valuea

Body composition

SBMD-WBLH g/cm2

L-BMD g/cm2

FN-BMD g/cm2

F-BMD g/cm2

TBM kg

FM kg

FFM kg

0.69 ± 0.17 0.75 ± 0.20 0.01

0.64 ± 0.19 0.71 ± 0.20 b0.01

0.67 ± 0.19 0.70 ± 0.17 b0.01

0.72 ± 0.12 0.70 ± 0.16 0.44

26.9 ± 12.82 31.72 ± 15.76 b0.01

6.17 ± 3.77 8.75 ± 5.67 0.15

18.01 ± 8.90 22.10 ± 10.50 b0.01

SBMD-WBLH: subtotal body mineral density of whole body less head; L-BMD, lumbar bone mineral density; FN-BMD, femoral neck bone mineral density; F-BMD, femur bone mineral density; TBM, total body mass, FM, fat mass; FFM, fat free mass. a Comparison of values among patients and control group using general linear model controlling for age, gender, BMI and Tanner stage.

44

C. Leoni et al. / Molecular Genetics and Metabolism 111 (2014) 41–45

Table 4 Biochemical markers of bone metabolism in individuals with Costello syndrome (n = 10). Marker (units)

Mean ± SD

Normal range 8.6-10.2 mg/dl 4.0-7.0 mg/dl 1.8-2.4 mg/dl N30 ng/ml

Parathyroid hormone β-CTX IGF1

9.73 ± 0.41 4.08 ± 0.64 2.23 ± 0.24 20.75 ± 5.04a 2/10 patients: b7 ng/ml 30.62 ± 8.96 0.96 ± 0.61 98.75 ± 65.20

Urine Calcium Phosphorus Phosphate tubular reabsorption

142.00 ± 114.00 826.90 ± 530.14 88.50 ± 6.45

50–300 mg/l 270–870 mg/l 80–90%

Blood Calcium Phosphorus Magnesium 25-OH vit D

agents targeting other down-stream targets of the RAS pathway may also help improve the bone phenotype in CS. Conflict of interest

10.0-65.0 pg/ml 0.5–2.5 mg/lb 30–330 ng/mlb

We excluded those patients from the mean ± SD reported in this table. a 25-OH vitamin D concentrations were lower than the cutoff detected by RIA method in 2/10 patients. b Age and gender related range of reference.

important role in regulating bone remodeling and facilitating vitamin D absorption/synthesis. In the absence of in vitro studies bone cellular function in CS is difficult to assess, but we speculate that individuals with CS have a primary defect in bone cellular function and this is supported by previous data suggesting increased bone resorption [7]. Still, the osteopenia in CS is likely multifactorial. Secondary contributors such as low 25-OH vitamin D and IGF1 concentrations, low calcium intake or reduced physical activity, may worsen the primary affected bone structure. Vitamin D can be ingested from diet and or supplements but the endogenous synthesis mediated by sunlight exposure usually provides the adequate amount in humans. Photochemical synthesis of vitamin D occurs cutaneously, where pro-vitamin D3 (7 dehydrocholesterol) is converted to pre-vitamin D (pre-D) in response to ultraviolet B exposure (sunlight). In CS, skin findings such as increased pigmentation, hyperkeratosis and acanthosis nigricans might affect proper pro-vitamin conversion to pre-vitamin. Even though 6 individuals with CS had frequent sunlight exposure, all affected individuals showed low concentrations of vitamin D. We speculate that the increased skin thickness and pigmentation might interfere with vitamin D synthesis. DXA imaging in adults with CS and those who sustain fractures should be considered clinically, in addition to supplementation with vitamin D and periodic monitoring of 25-OH vitamin D concentrations in all individuals with CS. A limitation of the study is the relatively small cohort size and potential confounders such as the ability to control for physical activity and sunlight exposure due to lack of standardized data in both groups. Future longitudinal studies on larger cohort of patients are needed.

5. Conclusions Our data demonstrate that BMD is decreased in CS, consistent with other anecdotal reports [7,22]. We also report low 25-OH vitamin D concentrations in individuals with CS. Although none of the subjects included in the study had developed pathologic fractures or suffered bone pain, based on the average young age of the cohort, we speculate these complications might represent a clinically relevant issue at older ages. This is also supported by White et al., who documented bone pain and vertebral fractures in a few adults with CS [23]. In the future, therapies may be developed to specifically target the RAS/ MAPK pathway, which could improve bone homeostasis in CS. Using a mouse model, He et al. demonstrated that osteoclastogenesis and osteoclast functions depend upon Erk1, suggesting a potential role for a selective Erk1-targeted chemical kinase inhibitor as therapeutic agent in osteoporotic conditions involving RAS/MAPK pathway [34]. Additional

None. Acknowledgments The authors would like to thank all patients and their families for participation in this study, and the “Associazione Italiana per le sindromi Costello e cardiofaciocutanea” (AISC). The authors acknowledge Xiaoming Sheng and Francesco Dotto for helping with the statistical analysis. References [1] J.M. Costello, A new syndrome, N. Z. Med. J. 74 (1971) 397. [2] K.W. Gripp, A.E. Lin, Costello syndrome: a Ras/mitogen activated protein kinase pathway syndrome (rasopathy) resulting from HRAS germline mutations, Genet. Med. 14 (2012) 285–292. [3] N. Mitin, K.L. Rossman, C.J. Der, Signaling interplay in Ras superfamily function, Curr. Biol. 15 (2005) R563–R574. [4] A.E. Karnoub, R.A. Weinberg, Ras oncogenes: split personalities, Nat. Rev. Mol. Cell Biol. 9 (2008) 517–531. [5] M. Tartaglia, G. Zampino, B.D. Gelb, Noonan syndrome: clinical aspects and molecular pathogenesis, Mol. Syndromol. 1 (2010) 2–26. [6] M. Tartaglia, B.D. Gelb, Disorders of dysregulated signal traffic through the RAS–MAPK pathway: phenotypic spectrum and molecular mechanisms, Ann. N. Y. Acad. Sci. 1214 (2010) 99–121. [7] D.A. Stevenson, E.L. Schwarz, J.C. Carey, D.H. Viskochil, H. Hanson, S. Bauer, H.Y. Weng, T. Greene, K. Reinker, J. Swensen, R.J. Chan, F.C. Yang, L. Senbanjo, Z. Yang, R. Mao, M. Pasquali, Bone resorption in syndromes of the Ras/MAPK pathway, Clin. Genet. 80 (2011) 566–573. [8] D.A. Stevenson, E.L. Schwarz, D.H. Viskochil, L.J. Moyer-Mileur, M. Murray, S.D. Firth, J.L. D'Astous, J.C. Carey, M. Pasquali, Evidence of increased bone resorption in neurofibromatosis type 1 using urinary pyridinium crosslink analysis, Pediatr. Res. 63 (2008) 697–701. [9] X. Yu, S. Chen, O.L. Potter, S.M. Murthy, J. Li, J.M. Pulcini, N. Ohashi, T. Winata, E.T. Everett, D. Ingram, W.D. Clapp, J.M. Hock, Neurofibromin and its inactivation of Ras are prerequisites for osteoblast functioning, Bone 36 (2005) 793–802. [10] T. Illés, V. Halmai, T. de Jonge, J. Dubousset, Decreased bone mineral density in neurofibromatosis-1 patients with spinal deformities, Osteoporos. Int. 12 (2001) 823–827. [11] T. Kuorilehto, M. Pöyhönen, R. Bloigu, J. Heikkinen, K. Väänänen, J. Peltonen, 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 (2005) 928–936. [12] S. Dulai, J. Briody, A. Schindeler, K.N. North, C.T. Cowell, D.G. Little, Decreased bone mineral density in neurofibromatosis type 1: results from a pediatric cohort, J. Pediatr. Orthop. 27 (2007) 472–475. [13] D.A. Stevenson, L.J. Moyer-Mileur, M. Murray, H. Slater, X. Sheng, J.C. Carey, B. Dube, D.H. Viskochil, Bone mineral density in children and adolescents with neurofibromatosis type 1, J. Pediatr. 150 (2007) 83–88. [14] N. Brunetti-Pierri, S.B. Doty, J. Hicks, K. Phan, R. Mendoza-Londono, M. Blazo, A. Tran, S. Carter, R.A. Lewis, S.E. Plon, W.A. Phillips, E. O'Brian Smith, K.J. Ellis, B. Lee, Generalized metabolic bone disease in neurofibromatosis type I, Mol. Genet. Metab. 94 (2008) 105–111. [15] T. Tucker, C. Schnabel, M. Hartmann, R.E. Friedrich, I. Frieling, H.P. Kruse, V.F. Mautner, J.M. Friedman, Bone health and fracture rate in individuals with neurofibromatosis 1 (NF1), J. Med. Genet. 46 (2009) 259–265. [16] N. Kossler, S. Stricker, C. Rödelsperger, P.N. Robinson, J. Kim, C. Dietrich, M. Osswald, J. Kühnisch, D.A. Stevenson, T. Braun, S. Mundlos, M. Kolanczyk, Neurofibromin (Nf1) is required for skeletal muscle development, Hum. Mol. Genet. 20 (2011) 2697–2709. [17] T.A. Delucia, K. Yohay, R.F. Widmann, Orthopaedic aspects of neurofibromatosis: update, Curr. Opin. Pediatr. 23 (2011) 46–52. [18] Z. Wang, Y. Liu, Research update and recent developments in the management of scoliosis in neurofibromatosis type 1, Orthopedics 33 (2010) 335–341. [19] D.A. Stevenson, F.C. Yang, The musculoskeletal phenotype of the RASopathies, Am. J. Med. Genet. C: Semin. Med. Genet. 157 (2011) 90–103. [20] K.S. Choudhry, M. Grover, A.A. Tran, E. O'Brian Smith, K.J. Ellis, B.H. Lee, Decreased bone mineralization in children with Noonan syndrome: another consequence of dysregulated RAS MAPKinase pathway? Mol. Genet. Metab. 106 (2012) 237–240. [21] W.K. Yassir, B.E. Grottkau, M.J. Goldberg, Costello syndrome: orthopaedic manifestations and functional health, J. Pediatr. Orthop. 23 (2003) 94–98. [22] K.A. Reinker, D.A. Stevenson, A. Tsung, Orthopaedic conditions in Ras/MAPK related disorders, J. Pediatr. Orthop. 31 (2011) 599–605. [23] S.M. White, J.M. Graham Jr., B. Kerr, K. Gripp, R. Weksberg, C. Cytrynbaum, J.L. Reeder, F.J. Stewart, M. Edwards, M. Wilson, A. Bankier, The adult phenotype in Costello syndrome, Am. J. Med. Genet. A 136 (2005) 128–135.

C. Leoni et al. / Molecular Genetics and Metabolism 111 (2014) 41–45 [24] W.A. Marshall, J.M. Tanner, Variations in the pattern of pubertal changes in boys, Arch. Dis. Child. 45 (1970) 13–23. [25] W.A. Marshall, J.M. Tanner, Variations in the pattern of pubertal changes in girls, Arch. Dis. Child. 44 (1969) 291–303. [26] C.J. Farrell, S. Martin, B. McWhinney, I. Straub, P. Williams, M. Herrmann, State-of-the-art vitamin D assays: a comparison of automated immunoassays with liquid chromatography–tandem mass spectrometry methods, Clin. Chem. 58 (2012) 531–542. [27] C. Carrozza, S. Persichilli, G. Canu, J. Gervasoni, E. Torti, L. Tazza, C. Zuppi, Measurement of 25-hydroxyvitamin vitamin D by liquid chromatography tandem-mass spectrometry with comparison to automated immunoassays, Clin. Chem. Lab. Med. 50 (2012) 2033–2035. [28] J.R. Hebert, I.S. Ockene, T.G. Hurley, R. Luippold, A.D. Well, M.G. Harmatz, Development and testing of a seven-day dietary recall, J. Clin. Epidemiol. 50 (1997) 925–937. [29] Societa' Italiana di Nutrizione Umana, Livelli di Assunzione Raccomandati di Energia e Nutrienti per la Popolazione Italiana (LARN), S.I.N.U., Milan, Italy, 1996.

45

[30] E. Heervä, M.H. Alanne, S. Peltonen, T. Kuorilehto, T. Hentunen, K. Väänänen, J. Peltonen, Osteoclasts in neurofibromatosis type 1 display enhanced resorption capacity, aberrant morphology, and resistance to serum deprivation, Bone 47 (2010) 583–590. [31] X. Wu, S.A. Estwick, S. Chen, M. Yu, W. Ming, T.D. Nebesio, Y. Li, J. Yuan, R. Kapur, D. Ingram, M.C. Yoder, F.C. Yang, Neurofibromin plays a critical role in modulating osteoblast differentiation of mesenchymal stem/progenitor cells, Hum. Mol. Genet. 15 (2006) 2837–2845. [32] F.C. Yang, S. Chen, A.G. Robling, X. Yu, T.D. Nebesio, J. Yan, T. Morgan, X. Li, J. Yuan, J. Hock, D.A. Ingram, D.W. Clapp, Hyperactivation of p21ras and PI3K cooperate to alter murine and human neurofibromatosis type 1-haploinsufficient osteoclast functions, J. Clin. Invest. 116 (2006) 2880–2891. [33] R.I. Gafni, J. Baron, Overdiagnosis of osteoporosis in children due to misinterpretation of dual-energy X-ray absorptiometry (DEXA), J. Pediatr. 144 (2004) 253–257. [34] Y. He, K. Staser, S.D. Rhodes, Y. Liu, X. Wu, S.J. Park, J. Yuan, X. Yang, X. Li, L. Jiang, S. Chen, F.C. Yang, Erk1 positively regulates osteoclast differentiation and bone resorptive activity, PLoS One 6 (2011) e24780.