Abstract
Bone remodelling is a complex mechanism regulated by osteoclasts and osteoblasts and perturbation of this process leads to the onset of diseases, which may be characterised by altered bone erosion or formation. In this review, we will describe some bone formation-related disorders as sclerosteosis, van Buchem disease, hypophosphatasia and Camurati–Engelmann disease. In the past decades, the research focused on these rare disorders offered the opportunity to understand important pathways regulating bone formation. Thus, the identification of the molecular defects behind the etiopathology of these diseases will open the way for new therapeutic approaches applicable also to the management of more common bone diseases including osteoporosis.
Introduction
In healthy skeleton, the bone structure is maintained by the balanced relationship between osteoblasts and osteoclasts. These cells are responsible for the bone remodelling that begins with the recruitment and activation of bone-resorbing osteoclasts followed by the formation of new calcified tissue by osteoblasts (Del Fattore et al. 2012). This activity is well organised and the quality and quantity of bone structures are dependent on fine regulated mechanisms. Defects of bone resorption that cause excessive or impaired bone loss impact skeletal integrity as well as the immoderate, disorganised or reduced bone formation (Del Fattore et al. 2012).
Osteoblasts, the bone-making cells, derive from neural ectoderm to form craniofacial bones and from paraxial mesoderm and lateral plate mesoderm to form axial and appendicular skeleton (Tuan 1998). The bone formation process occurs in two distinct processes: intramembranous or endochondral ossification (Rutkovskiy et al. 2016). Bone formation starts with the recruitment of MSC or pericytes of the bone marrow (James & Peault 2019) and it is characterised by four phases: lineage commitment, proliferative expansion, synthesis of extracellular matrix and mineralisation (Rutkovskiy et al. 2016) (Fig. 1).
Bone formation phases. Schematic diagram of the osteoblastogenesis and bone matrix production. Osteoblasts arise from a pluripotent mesenchymal stem cell (MSC) which, under proper stimuli, is committed toward an osteoprogenitor, a preosteoblast and an active, mature osteoblast. In this process, the bone matrix is produced and mineralised. At the end of bone formation, some osteoblasts remain embedded in the calcified matrix as osteocytes. Specific markers are indicated for each differentiation stage. Runx2, Runt-related transcription factor 2; ALP, alkaline phosphatase; COL1A, type I collagen α; OCN, osteocalcin; OPN, osteopontin; BSP, bone sialoprotein; DMP-1, dentin matrix acidic phosphoprotein 1; MEPE, matrix extracellular phosphoglycoprotein; SOST, sclerostin. A full colour version of this figure is available at https://doi.org/10.1530/JOE-20-0285.
Citation: Journal of Endocrinology 248, 2; 10.1530/JOE-20-0285
Regarding the lineage commitment phase, osteoblast differentiation from MSC follows a specific programme of gene expression that takes place under the control of both wingless protein (WNT) and bone morphogenetic proteins (BMP). The activation of these pathways and their signal transduction lead to the stabilisation of β-catenin and activation of small mother against decapentaplegic (SMAD) signalling and to the expression of master transcriptional factors involved in osteoblastogenesis such as runt-related transcription factor 2 (Runx2) and transcription factor 7 (Sp7, Osterix) (Wu et al. 2016). Following initial lineage commitment, a phase of intensive proliferative activity of pre-osteoblasts occurs; these cells express Runx2, Osterix, ALP (alkaline phosphatase) and type I collagen ensues, are committed to the osteoblast lineage and represent an intermediate stage. At the end of the proliferation phase, pre-osteoblasts change their morphology in large cuboidal osteoblasts. Osteoblasts express high levels of ALP and generate the not mineralised and flexible bone osteoid (synthesis of extracellular matrix phase), by the secretion of matrix proteins including type I collagen (90%) and small percentage of proteoglycans, fibronectin and specific bone proteins, such as osteopontin and osteocalcin (Rutkovskiy et al. 2016). Subsequently, the mineralisation phase occurs and osteoblasts release matrix vesicles (MV) also known as ‘calcifying globules’ by budding, leading to the formation of hydroxyapatite crystals made of calcium and phosphate (Ca10(PO4)6(OH)2) (Del Fattore et al. 2012). MV contain several enzymes and transporters located on their membrane generating inorganic pyrophosphate (PPi) that is hydrolysed into inorganic phosphate (Pi) by tissue-nonspecific alkaline phosphatase (TNSALP) (Zhou et al. 2012). The presence of annexin and phosphatidyl serine into MV leads to the accumulation of calcium ions, which together with Pi promote the hydroxyapatite crystals nucleation and growth. The newly formed hydroxyapatite crystals are released by MV into extracellular matrix where the presence of Ca2+ and Pi supports crystals proliferation filling empty spaces between collagen fibrils (Del Fattore et al. 2012).
Bone quality is determined by the proportion of organic and mineral matrix that is characterised in healthy adult subjects by 60% mineral, 20% organic and 20% water but may vary in case of diseases and drug therapies. At the end of bone formation, some osteoblasts die by apoptosis or become lining cells, others remain embedded in the calcified matrix as osteocytes (Rutkovskiy et al. 2016).
The number of osteoblasts and their activity are controlled by epigenetic mechanisms and are regulated by local or systemic factors (Sims & Martin 2014).
Molecular dissections of genetic osteoblast disorders allowed to identify the role of proteins involved in osteoblast activity. In this review, we described rare genetic osteoblast-related diseases and the underlying defects of molecular pathways. A deeper analysis of the molecular alterations is relevant for the identification of new therapeutic approaches for these rare bone disorders.
Sclerosteosis and van buchem disease
Sclerosteosis and van Buchem disease (VBD) are two rare autosomal recessive sclerosing disorders, classified as craniotubular hyperostoses and caused by genetic defects of sclerostin gene SOST. The first description was reported among the Afrikaners community in South Africa (van Lierop et al. 2017). In 1958 Truswell et al. defined sclerosteosis for the first time as ‘osteopetrosis with syndactyly’ (Truswell 1958). Patients are affected by facial distortion due to bossing of the forehead and mandibular overgrowth, hypertelorism and proptosis (van Lierop et al. 2017). In the 66% of individuals, syndactyly is noted at birth and it mainly involves the index and third finger. In early childhood, affected patients become tall and heavy due to skeletal overgrowth. Hallmark complication affecting 93% of the patients is facial palsy, due to overgrowth of the skull. This feature is recurrent and appears early in the first months of life. Because of the bone thickening, blindness, deafness and increased intracranial pressure may be present (van Lierop et al. 2017). In the majority of patients, symptoms develop during early childhood and progress until adulthood when they stabilise after the third decade of life (van Lierop et al. 2013). Histopathological findings revealed accelerated bone formation with increased trabecular thickness and osteoidosis; the newly formed bone is lamellar and correctly mineralised (Hassler et al. 2014). Moreover, bone thickening and cortical hyperostosis are detected by imaging techniques. If not treated, patients have a short life expectancy (van Lierop et al. 2013).
Van Buchem disease, first described in 1955 as ‘hyperostosis corticalis generalizata familiaris’, is a milder type of sclerosteosis (van Buchem et al. 1955). It is characterised by the same symptoms of sclerosteosis, except for syndactyly and tendency to tall stature (van Buchem 1976, van Lierop et al. 2013). Endosteal hyperostosis produces dense cortical bone with a narrow medullary cavity (Ihde et al. 2011). The diagnosis is mainly based on the analysis of skull and mandible abnormalities and on radiological images.
Biochemical analysis of patients affected by both diseases displays increased levels of procollagen type I aminoterminal propeptide (P1NP) and ALP and normal level of calcium, phosphorus and parathyroid hormone (PTH) (van Lierop et al. 2013).
The similar features that characterise the diseases are attributable to genetic defects of SOST gene, encoding sclerostin. Sclerosteosis is caused by inactivating mutations of SOST gene (van Lierop et al. 2017) whereas van Buchem’s patients harbour a 52 kb homozygous deletion on chromosome 17q12, compromising a downstream enhancer of SOST important for gene transcription in adult bone (Balemans et al. 2002). Consequently, sclerostin serum levels are undetectable in sclerosteosis, but low levels can be revealed in van Buchem disease (van Lierop et al. 2013).
In these two diseases, osteoclast abnormalities have not been reported suggesting that osteoclasts are not involved in the pathological mechanisms responsible for the skeletal changes observed in patients (Stein et al. 1983). However, it was demonstrated that sclerostin stimulates osteocyte support of osteoclast activity by a RANKL (receptor activator of nuclear factor kappa-Β ligand)-dependent pathway. The catabolic action of sclerostin was further demonstrated; in a rat model of postmenopausal osteoporosis, sclerostin neutralisation protected mice from bone loss induced by ovariectomy due to increased osteoblast surface and reduced osteoclast number (Li et al. 2009).
SOST pathway
The genetic analysis of human sclerosteosis and VBD allowed the identification of SOST gene (Balemans et al. 2002). SOST encodes for sclerostin that is a potential new member of the Dan/Cerberus family of BMP (Brunkow et al. 2001). Sclerostin mRNA is widely expressed in the organism, especially during embryogenesis; postnatally the protein has been only revealed in osteocytes, chondrocytes and cementocytes (Holdsworth et al. 2019). Sclerostin is a monomeric glycoprotein with a cystine knot-like domain containing two N-linked glycosylation sites and many positively charged lysine and arginine residues. The mature structure of sclerostin includes C- and N-terminal arms and three loops around the cystine knot motif (Brunkow et al. 2001, Balemans et al. 2002). Loop 2 has been identified as the critical region that binds low-density lipoprotein (LRP) 5/6. LRP 5/6 is a co-receptors of the WNT canonical signalling pathway that plays an important role in the regulation of bone formation. The association of sclerostin to the WNT canonical pathway was suggested by the observation of similar clinical traits between patients with high bone mass diseases due to sclerostin absence and those with LRP5 defects (Li et al. 2005). Although sclerostin can bind also to LRP4 with structural similarities to LRP5/6, this link is not able to create a co-receptor complex with frizzled that is required to bind WNT proteins (Fig. 2).
Sclerostin in bone physiology and in Sclerosteosis/VBD. In osteoblasts, binding of Wnt ligands to the LRP5/6 and frizzled co-receptor stimulates bone formation. Sclerostin also binds to LRP5/6 and inhibits the Wnt signalling pathway, an action facilitated by LRP4. In physiological condition (upper panel), the bone-forming role of Wnt is well balanced with the inhibitory effect of sclerostin on bone formation. In Sclerosteosis or van Buchem disease (lower panel), the mutation/deletion in SOST gene, that causes a defect of sclerostin production, reduces the levels of sclerostin and determines an increase of bone formation. VBD, van Buchem disease; Wnt, wingless protein; LRP, low-density lipoprotein receptor-related protein. The figure was produced using Servier Medical Art. A full colour version of this figure is available at https://doi.org/10.1530/JOE-20-0285.
Citation: Journal of Endocrinology 248, 2; 10.1530/JOE-20-0285
In spite of that, LRP4 overexpression enhances the efficacy of sclerostin in the WNT pathway while its silencing reduces sclerostin function (Leupin et al. 2011). Interestingly, mutations in LRP4 cause Sclerosteosis type 2, which shows features comparable to sclerosteosis (Leupin et al. 2011), suggesting LRP4 as a regulator of sclerostin activity.
Sclerostin plays an essential role in the inhibition of osteoblast differentiation and function, leading to reduced bone formation (Li et al. 2008). Indeed, sclerostin limits mesenchymal lineage cell differentiation towards the osteoblast lineage, decreases bone matrix formation by osteoblasts, promotes osteoblast apoptosis, maintains bone lining cells in a quiescent state and inhibits differentiation of late osteoblasts into osteocytes (Holdsworth et al. 2019). The production of sclerostin is regulated by mechanical loading, estrogens and PTH and increases with age (van Lierop et al. 2017). Sost gene deletion in mice causes increased bone mineral density (BMD) in the entire skeleton, while the overexpression leads to osteopenic phenotype (Li et al. 2008). In mice, the van Buchem phenotype is created via deletion of the non-coding enhancer ECR5, that reproduces a milder phenotype than sost−/− mice (Collette et al. 2012).
Treatment
There are no specific treatments for sclerosteosis and VBD. The management of the diseases aims to surgical correction of syndactyly, decompression of cranial nerves and reduction of mandibular overgrowth. In some patients, skull bones are replaced by prosthetic cap or are replaced after bone reduction (Hamersma et al. 2003). Life expectancy for sclerosteosis patients is 33 years and the major causes of death are herniation of the brain or perioperative complications to correct intracranial pressure (Hamersma et al. 2003); instead for van Buchem affected people life expectancy can reach up to 82 years (van Lierop et al. 2017).
Hypophosphatasia
It is important to note that not only osteoblast-specific alterations cause defects to bone mineral density. Indeed, hypophosphatasia (HPP) is characterised by loss-of-function mutation of ALPL gene, localised on chromosome 1p36-34, that encodes for tissue-nonspecific alkaline phosphatase (TNSALP) protein (Whyte 2017). The disease is classified in seven clinical forms: perinatal, infantile, childhood, adult, odonto-HPP, pseudo-HPP and benign prenatal HPP. The major clinical manifestations are hypomineralisation of bone and teeth, which results in defective ossification, osteomalacia and loss of dentition (Whyte 2017). HPP could already manifest in utero, causing profound skeletal hypomineralisation that can lead to death due to asphyxia after few days of life. Defects of bony fusion of cranial sutures can lead to increased intracranial pressure and cerebral damage. The typical manifestation of childhood form is premature loss of deciduous teeth (Whyte 2016). The deposition of calcium pyrophosphate dihydrate cristals in articular cartilage is the main cause of musculoskeletal pain in HPP adult patients (Shapiro & Lewiecki 2017). In severe forms patients are affected by respiratory failure, short stature, hypotonia, fractures (Fig. 3) and muskuloskeletal pain (Shapiro & Lewiecki 2017).
X-ray picture from a 46-year-old patient affected by hypophosphatasia. Asterisks indicate multiple vertebral fractures. A full colour version of this figure is available at https://doi.org/10.1530/JOE-20-0285.
Citation: Journal of Endocrinology 248, 2; 10.1530/JOE-20-0285
The hypomineralisation is caused by accumulation of inorganic pyrophospate (PPi) in the extracellular matrix, which inhibits hydroxyapatite crystals’ formation and affects the cementum in the outside faces of teeth root (Foster et al. 2012). Indeed, osteoblasts from tnsalp null mice differentiate normally but are unable to initiate mineralisation in vitro while osteoclastogenesis and bone resorption are not altered (Wennberg et al. 2000).
Laboratory tests may reveal low levels of ALP activity, increased levels of urinary phoshoethanolamine (PEA), serum PPi and serum pyridoxal 5’-phospate (PLP), and normal serum values of calcium, ionised calcium, inorganic phosphate, PTH, 25-hydroxy and 1,25-hydroxy vitamin D (Whyte 2017). These parameters may be useful for HPP diagnosis, which is based on genetic analysis, even in prenatal context. Severe HPP could be explained by recessive inheritance whereas milder form could have an autosomal dominant inheritance.
The heterogeneity in clinical manifestations reflects the mutations identified in ALPL gene; so far, 409 variants associated with HPP disease have been described (http://www.sesep.uvsq.fr/03_hypo_mutations.php) (Mornet 2018).
ALPL gene
ALPL gene encodes for tissue nonspecific alkaline phosphatase enzyme (TNSALP) that is, with the placental, the intestinal and the placental-like forms, one of the four metalloenzymes ALPs in humans. The ALPL gene is located on chromosome 1, at the tip of the short arm, whereas other ALPs genes are found on the long arm of chromosome 2 (2q34–q37) (Mornet 2018). No inherited disorders have been described for tissue-specific ALPs (Millan & Whyte 2016).
TNSALP is widely expressed in the body, particularly in the liver, bone and kidney. While its role in bone calcification is well known, the role in extra-calcified tissue is still under investigation (Millan & Whyte 2016).
This enzyme is a membrane glycoprotein linked to the membrane by glycosylphosphatidylinositol (GPI). It has five main domains: catalytic site, calcium-binding site, crown domain, homodimer interface and N-terminal alpha helix (Mornet 2018). During the synthesis of the peptide, the N-terminal domain guides the protein to the endoplasmic reticulum (ER) lumen where it folds in the tertiary structure and creates a quaternary structure as homodimer. Moreover, in the ER, the neo-synthetised peptide is modified by N-glycans, disulfide bonds and GPI. In the Golgi the immature 66 kDa form of TNSALP becomes the 80 kDa mature form modifying the high-mannose type N-glycans to complex type N-glycans (Mornet 2018). Subsequently, mature protein is transported to the plasma membrane to localise in the surface via GPI where it dephosphorylates PPi releasing inorganic phosphate (Satou et al. 2012). The essential role of TNSALP for skeletal mineralisation came from the study of HPP in the 80s. The accumulation of PPi, PEA and PLP observed in HPP was essential to identify these compounds as natural substrates of TNSALP (Salles 2020). Mutation of TNSALP reduces enzyme activity leading to high level of PPi that is an inhibitor of hard tissue mineralisation; this results in the blocking of hydroxyapatite crystal growth (Whyte 2017) (Fig. 4).
TNSALP function in physiological condition and in HPP. TNSALP enzyme is responsible for the formation of Pi cleaving PPi (left panel). PPi is a potent inhibitor of hydroxyapatite crystals, while Pi is the main actor of crystals formation with calcium. In HPP, mutations in TNSALP cause inactivation of the enzyme and an accumulation of PPi that inhibit bone mineralisation in patients (right panel). TNSALP, tissue nonspecific alkaline phosphatase; Pi, inorganic phosphate; PPi, inorganic pirophosphate. The figure was produced using Servier Medical Art. A full colour version of this figure is available at https://doi.org/10.1530/JOE-20-0285.
Citation: Journal of Endocrinology 248, 2; 10.1530/JOE-20-0285
Treatment
The hypomineralisation that characterises HPP often caused a misdiagnosis of the disease as osteoporosis and, as a consequence, led to treatment with antiresorptive drugs, such as bisphosphonates. The wrong treatment worsened the prognosis of HPP because bisphophonates are analogous of PPi and accumulates in HPP (Whyte 2017). The implementation of diagnosis and treatment is still necessary. The therapeutic options were based on treatment of hypercalcemia with fluid hydration, low calcium diet and surgery for fractures. To ameliorate the reduced ALP level, infusion with ALP-rich plasma obtained from Paget’s patients was performed (Choida & Bubbear 2019). Whyte et al. reported life-treating intravenous infusion in four patients with severe HPP and after 5 weeks partial or complete correction of the deficiency of circulating ALP activity was observed, but PEA and PPi remain elevated and skeletal deformities persisted (Whyte et al. 1986).
Bone marrow transplantation (BMT) was also used as therapeutic approach for HPP. In the short-term, BMT was efficient to improve bone mineralisation; however the return of host hematopoiesis was observed few months post-BMT; restoration of ALP activity was limited, and normal bony architecture could not be achieved (Taketani et al. 2015).
A positive effect was produced by bone anabolic drug Teriparatide, a recombinant peptide mimicking PTH, which leads to increased ALP levels and fracture healing (Whyte et al. 2007, Righetti et al. 2018); however patient osteoblasts still produced a defective enzyme (Subbiah et al. 2010).
Due to the poor effects of aforementioned treatments, the new frontier is the development of a recombinant TNASALP enzyme. Asfotase alfa is the human recombinant enzyme created to replace mutated TNSALP; it is a glycoprotein that contains the catalytic domain of TNSALP thus being able to degrade PPi. This enzyme replacement therapy (ERT), approved by FDA in 2015, is administered from perinatal to adulthood via subcutaneous injection (Kitaoka et al. 2017).
Studies of affected infants, children, adolescents and adults assessed the efficacy and positive effects on skeletal mineralisation, growth, mobility and survival up to 5 years of therapy (Whyte et al. 2016). Despite good results in treating HPP, asfotase alfa requires lifetime treatment administration with frequent injections (3–6 months) and has several adverse effects as injection-site reactions and possible long-term tachyphylaxis. Alternative therapies are still under development including gene therapies that use lentiviral vector (LV) delivery of TNSALP or transduction of bone marrow cells ex vivo with LV-TNSALP and, interestingly, anabolic treatment with anti-sclerostin antibody (Matsumoto et al. 2011, Whyte et al. 2016, Seefried et al. 2017).
Camurati–Engelmann disease
Camurati–Engelmann disease (CED), also known as progressive diaphyseal dysplasia (PDD), is a rare genetic skeletal disorder, first described in 1920 by Cockayne (Cockayne 1920), later by Camurati in 1922 (Camurati 1922) and by Engelmann in 1929 (Engelmann 1929). This is a very rare disease, indeed about 300 cases have been reported to date; it has an incidence of 1:1 million births and displays an autosomal dominant inheritance. It is characterised by limb pain, muscular weakness and atrophy, cortical thickening of long bones diaphyses, cranial hyperostosis that usually appears during childhood. Clinical manifestations and severity of disease are extremely variable within patients for the age of onset, rate of progression and bone involvement, making the diagnosis challenging (van Hul et al. 2019). The main radiographic feature of CED is a thickening of the diaphysis due to abnormal periosteal and endosteal apposition of new bone, which sometimes affects metaphyses. The most involved regions are tibiae, femora, skull and pelvis (Spranger 2018) (Fig. 5). Bone scintigraphy identifies accumulation in the affected sites of the skeleton and it is a valuable tool for early stages diagnosis (Clybouw et al. 1994). Thickening of skull bone can lead to blindness and/or deafness; minor other orthopaedic defects are reported as kyphosis, scoliosis and flat feet (Yuldashev et al. 2017). Muscle weakness, easy fatigability and leg pain may occur especially during childhood, wrongly leading to a diagnosis for muscular disorders. Radiographic manifestations are usually detected before age of 30 (van Hul et al. 2019).
X-ray images from a 41-year-old patient affected by Camurati–Engelmann disease. Arrows indicate the thickening and irregularity of endosteal and periosteal sides of long bones including tibia, femur and humerus. A full colour version of this figure is available at https://doi.org/10.1530/JOE-20-0285.
Citation: Journal of Endocrinology 248, 2; 10.1530/JOE-20-0285
In 2000, Janssens et al. identified the transforming growth factor β1 (TGFβ1) gene (chromosome 19q13.2) as the primary causative gene in CED (Janssens et al. 2000).
At present, 13 different mutations have been reported in patients, that are located in the latency-associated peptide (LAP) or in the signal peptide of the protein (van Hul et al. 2019). Three of them are the most recurrent: R218C, R218H and C225R (Janssens et al. 2006). The result of the mutations found in LAP is a reduction of the LAP ability to bind TGFβ thus leading to increased activity. Regarding the mutations found in the TGFβ signal peptide, they cause intracellular accumulation of TGFβ and increased signalling (Janssens et al. 2003). CED mutations lead to a net increase of bone mass in some regions of the skeleton. Despite some conflicting findings, most experiments suggest that TGFβ promotes proliferation, early differentiation of osteoblast progenitor cells, and matrix production, while inhibiting later differentiation and matrix mineralisation. At the same time, bone erosion is also enhanced in CED as also shown by high levels of bone resorption markers in patients (Hernandez et al. 1997). Moreover, in vitro experiments confirmed an increase of osteoclastogenesis and bone resorption in cell cultures from CED patients.
So far it is not completely understood why symptoms are mostly in the skeleton, even though TGFβ1 and its receptors are ubiquitously expressed (Weiss & Attisano 2013).
TGF beta pathway
TGFβ1 is a member of the TGFβ family and is a potent stimulator of cell migration, proliferation, differentiation and apoptosis (Del Fattore et al. 2012). The protein is produced as a precursor with an N-terminal signal peptide, a latency-associated peptide necessary for folding and dimerisation, and the C-terminal growth factor. In the ER, the precursor forms cystine bridges and the signal peptide is cleaved. After glycosylation in the Golgi, LAP is also cleaved by furin (Nuchel et al. 2018). The homodimer with the mature peptide, non-covalently linked to LAP subunits, is secreted and embedded in the bone matrix (Miyazono et al. 1988). It could be found in either a latent form composed of a mature peptide homodimer, a LAP homodimer and a latent TGFβ binding protein, or in an active form characterised by the mature peptide homodimer. The mature peptide may also form heterodimers with other TGFβ family members.
TGFβ is the most abundant cytokine in the bone matrix, deposited by osteoblasts, and can have both a positive and negative role in the development and maintenance of the skeleton when released by osteoclast activity during bone resorption (Wu et al. 2016); indeed the growth factor is released from the extracellular matrix during bone resorption and the acid pH of the Howship’s lacuna breaks the interaction between mature peptide and LAP, causing the activation of TGFβ1. TGFβ1 regulates both osteoblast and osteoclast differentiation and activity. It regulates bone remodelling by the interaction with its constitutively active type II receptor, recruiting also type I receptor to form heteromeric complex (Shi et al. 2011). TGFβ1 stimulates osteoblast differentiation in the early stages, and inhibits the later phase of proliferation and mineralisation that are more susceptible to BMP signalling. Moreover, TGFβ1 inhibits further osteoclast differentiation and activation, coupling bone formation with bone resorption. TGFβ isoforms and their receptors, type I receptor (TGFβRI) and type II receptor (TGFβRII), play important roles in endochondral and intramembranous ossification (Chen et al. 2012).
The binding of TGFβ1 with its receptors modulates the activation of SMAD transcription factors. SMAD2 and SMAD3 (R-SMAD) respond to TGFβ receptors, and, competing with SMAD7, they form complexes with SMAD4, one of the common mediator of TGFβ pathway (Shi & Massague 2003) (Fig. 6). However other growth factors including interferon gamma (IFNγ) and tumor necrosis factor alpha (TNFα) can modulate SMAD proteins expression with attendant inhibition of TGF-β signalling (Fig. 6). In recent papers R-SMAD were reported to be involved in a molecular complex that impacts on TNF receptor-associated factor 6 (TRAF6), confirming the inhibitory effect of the growth factor on the RANKL-induced osteoclastogenesis (Yasui et al. 2011).
TGFβ1 in physiological state and in CED. TGFβ1 is responsible of osteoblast differentiation, proliferation and activity. In physiological conditions (left panel), its binding to TGFβ receptor II activates SMAD proteins via phosphorylation. SMAD2 and SMAD3 trimerise with SMAD4. The trimer enters into the nucleus and activate gene transcription. Smad7 attenuates phophorylation and nuclear translocation of Smad2/Smad3. IFNγ inhibits the TGFβ/Smad signalling pathway by activation of Smad7. TNFα inhibits the pathway by AP-1 that directly interfere with the interaction of the Smad2/3-Smad4 complex with DNA. In CED (right panel), the hyperactivation of TGFβ1 leads to the increase of signal transduction and downstream pathway causing an iper-activation of the genic transcription. TGFβ1, transforming growth factor beta 1; IFNγ. interferon gamma; TNFα, tumor necrosis factor alpha; CED, Camurati-Engelmann disease; SMAD, small mother against decapentaplegic. The figure was produced using Servier Medical Art. A full colour version of this figure is available at https://doi.org/10.1530/JOE-20-0285.
Citation: Journal of Endocrinology 248, 2; 10.1530/JOE-20-0285
TGFβ1 has also a non-canonical pathway, via mitogen-activated protein kinase (MAPK), that promotes osteoprogenitor proliferation and differentiation (Matsunobu et al. 2009). Both canonical and non-canonical pathways converge at RUNX2 gene to control mesenchymal cell osteogenic differentiation (Lee et al. 2002).
Treatment
The treatment of CED is not well established and different therapeutic options are based on short case reports. In some cases, corticosteroid-based treatment could solve pain and improve muscle function (Wallace et al. 2004). A dose of 1 mg/kg/day of prednisone is often administered to patients with severe symptoms. Histological analyses after steroid therapy have shown an increase of bone resorption and new bone formation but in several studies, bone scintigraphy reported no regression of sclerosis (Bas et al. 1999). An angiotensin II type receptor that downregulates TGFβ1, was used to treat CED. However, conflicting effects are reported (Yuldashev et al. 2017). Preclinical studies that target TGFβ1 type I receptor have been reported to treat the high bone turnover disease (Akhurst 2017), but no clinical trials data are still available. Few clinical trials of TGFβ inhibitor could open the way to a new therapy for CED.
Discussion
In the physiological process of bone remodelling, bone formation is a tightly regulated process. Signals that influence the differentiation and function of osteoblasts play an important role in the mechanism of action of anabolic agents in the skeleton. So far, the use of Teriparatide, analogue of PTH, has been the only safe, effective and approved anabolic drug treatment for osteoporosis, with limitation in the use and in the time of administration.
The study of rare diseases paved the way to the comprehension of molecular mechanisms involved in the physiology of bone, revealing also possible targets to improve bone formation. In this review, we described four rare diseases characterised by bone formation defects and the molecular pathways involved in the onset of these disorders that suggested or can suggest new targets for anabolic treatments.
Sclerosteosis and van Buchem diseases highlighted the importance of sclerostin in bone physiology and made patients a unique model to study the role of this protein in bone metabolism. The generation of sost−/− mice confirmed that the lack of sclerostin only impaired bone tissue, without affecting other organs.
Sclerostin could represent also a target by which nature counteracts the postmenopausal bone loss. Indeed, increased methylation of SOST promoter has been revealed in bone biopsies of postmenopausal women. The hyper-methylation of the promoter is associated to the reduced expression of SOST gene thus leading to reduced inhibition of WNT signalling. This could be a compensatory mechanism to stimulate bone formation that is reduced in postmenopausal condition (Reppe et al. 2015).
All these observations indicated sclerostin as a possible novel target for the development of new anabolic therapies for osteoporosis without side effects on other organs.
Indeed, several sclerostin inhibitors have been tested for osteoporosis treatment. In an elegant review, McClung reported the positive effect on the bone formation of Romosozumab, a human MAB against sclerostin (McClung 2018), that was recently approved by Food and Drug Administration for its use in osteoporotic patients with the high risk of fractures. Beside teriparatide, romozumab is another safe and efficacious bone anabolic treatment.
Although patients with sclerosteosis or van Buchem disease contributed greatly to understand the function of sclerostin in bone and to identify new therapy for osteoporosis, they still not have efficacious treatment for their diseases.
The development of gene therapy or enzyme replace therapy could be a safe way to treat bone formation defects as for hypophosphatasia. The comprehension of TNSALP role in bone led to the production of Asfotase Alfa enzyme replacement therapy to treat HPP. This therapy has a good safe profile (Whyte et al. 2012), with evident effects in infants, children and adults disease. Since potential adverse effects such as site injection reaction (in about 75% of patients), anaphylactoid reaction, exacerbation of calcification in vascular and kidney and immune reaction to treatment (Whyte et al. 2012) have been observed, further studies to ameliorate these features are needed. Thus, also for this disease innovative and life-saving therapy with long-term efficacy and reduced side effects is still under investigation.
The bone-specificity, that let sclerostin be a good target for bone diseases without side-effects, is not the main property of TGFβ that, beyond the important role in all phases of chondrogenesis, mesenchymal condensation, matrix deposition and skeletal homeostasis maintenance, has a broad spectrum of function in other pathologies, such as cancers and vascular diseases (Chen et al. 2012).
The pleiotropic effect of TGFβ1 on the skeletal cells, with roles both in osteoclasts and in osteoblasts, makes the growth factor appealing as a target for the treatment of bone diseases (Chen et al. 2012). Tgfβ1−/− mice display reduced bone growth and mineralisation (Geiser et al. 2005) while the osteoporotic mice treated with TGFβ displayed increased BMD (Gazit et al. 1999). Despite its ubiquitous distribution, TGFβ is one of the growth factor to be considered as anabolic agent to promote fracture healing or to reverse bone resorption in osteoporosis. The limit of its use is related to the wide distribution of TGFβ that could cause adverse effects highlighting the requirement to develop a good carrier that can deliver drugs specifically to bone or, for genetic diseases’ treatment, to develop a tissue-specific gene therapy.
The implications of TGFβ in several mechanisms reveal that additional studies are required to better identify its role in body physiology and to evaluate a possible use as bone-forming drug.
Advances in the understanding of cellular and molecular signalling contributing to osteoblast differentiation and function, based also on the study of genetic diseases, have indicated new targets to treat either rare or common skeletal pathologies. Further development and study of therapeutic approaches targeting bone formation fulfil a clinical need for treatment option for bone loss to promote bone regeneration despite targeting only bone resorption.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.
Funding
This work was supported by funding from Italian Ministry of Health (Ricerca Corrente #RC2019 to A D F).
References
Akhurst RJ 2017 Targeting TGF-beta signaling for therapeutic gain. Cold Spring Harbor Perspectives in Biology 9 a022301. (https://doi.org/10.1101/cshperspect.a022301)
Balemans W, Patel N, Ebeling M, Van Hul E, Wuyts W, Lacza C, Dioszegi M, Dikkers FG, Hildering P & Willems PJ et al. 2002 Identification of a 52 kb deletion downstream of the SOST gene in patients with van Buchem disease. Journal of Medical Genetics 39 91–9 7. (https://doi.org/10.1136/jmg.39.2.91)
Bas F, Darendeliler F, Petorak I, Sadikoglu B, Bilir A, Bundak R, Saka N & Gunoz H 1999 Deflazacort treatment in progressive diaphyseal dysplasia (Camurati-Engelmann disease). Journal of Paediatrics and Child Health 35 401–40 5. (https://doi.org/10.1046/j.1440-1754.1999.00381.x)
Brunkow ME, Gardner JC, Van Ness J, Paeper BW, Kovacevich BR, Proll S, Skonier JE, Zhao L, Sabo PJ & Fu Y et al. 2001 Bone dysplasia sclerosteosis results from loss of the SOST gene product, a novel cystine knot-containing protein. American Journal of Human Genetics 68 577–5 89. (https://doi.org/10.1086/318811)
Camurati M 1922 Di un raro caso di osteite simmetrica ereditaria degli arti inferiori. Chirurgia degli Organi di Movimento 6 662–665.
Chen G, Deng C & Li YP 2012 TGF-beta and BMP signaling in osteoblast differentiation and bone formation. International Journal of Biological Sciences 8 272–2 88. (https://doi.org/10.7150/ijbs.2929)
Choida V & Bubbear JS 2019 Update on the management of hypophosphatasia. Therapeutic Advances in Musculoskeletal Disease 11 1759720X19863997. (https://doi.org/10.1177/1759720X19863997)
Clybouw C, Desmyttere S, Bonduelle M & Piepsz A 1994 Camurati-Engelmann disease: contribution of bone scintigraphy to genetic counseling. Genetic Counseling 5 195–19 8.
Cockayne EA 1920 Case for diagnosis. Proceedings of the Royal Society of Medicine 13 132–13 6. (https://doi.org/10.1177/003591572001301935)
Collette NM, Genetos DC, Economides AN, Xie L, Shahnazari M, Yao W, Lane NE, Harland RM & Loots GG 2012 Targeted deletion of Sost distal enhancer increases bone formation and bone mass. PNAS 109 14092–1409 7. (https://doi.org/10.1073/pnas.1207188109)
Del Fattore A, Teti A & Rucci N 2012 Bone cells and the mechanisms of bone remodelling. Frontiers in Bioscience 4 2302–2 32 1. (https://doi.org/10.2741/543)
Engelmann G 1929 Ein fall von osteopathia hyperostotica (sclerotisans) multiplex infantilis. Fortschr Geb Roentgen 39 1101–1106.
Foster BL, Nagatomo KJ, Nociti FH, Fong H, Dunn D, Tran AB, Wang W, Narisawa S, Millán JL & Somerman MJ 2012 Central role of pyrophosphate in acellular cementum formation. PLoS ONE 7 e38393. (https://doi.org/10.1371/journal.pone.0038393)
Gazit D, Zilberman Y, Turgeman G, Zhou S & Kahn A 1999 Recombinant TGF-beta1 stimulates bone marrow osteoprogenitor cell activity and bone matrix synthesis in osteopenic, old male mice. Journal of Cellular Biochemistry 73 379–3 89. (https://doi.org/10.1002/(SICI)1097-4644(19990601)73:3<379::AID-JCB9>3.0.CO;2-U)
Geiser AG, Hummel CW, Draper MW, Henck JW, Cohen IR, Rudmann DG, Donnelly KB, Adrian MD, Shepherd TA & Wallace OB et al. 2005 A new selective estrogen receptor modulator with potent uterine antagonist activity, agonist activity in bone, and minimal ovarian stimulation. Endocrinology 146 4524–45 35. (https://doi.org/10.1210/en.2005-0024)
Hamersma H, Gardner J & Beighton P 2003 The natural history of sclerosteosis. Clinical Genetics 63 192–19 7. (https://doi.org/10.1034/j.1399-0004.2003.00036.x)
Hassler N, Roschger A, Gamsjaeger S, Kramer I, Lueger S, Van Lierop A, Roschger P, Klaushofer K, Paschalis EP & Kneissel M et al. 2014 Sclerostin deficiency is linked to altered bone composition. Journal of Bone and Mineral Research 29 2144–21 51. (https://doi.org/10.1002/jbmr.2259)
Hernandez MV, Peris P, Guanabens N, Alvarez L, Monegal A, Pons F, Ponce A & Munoz-Gomez J 1997 Biochemical markers of bone turnover in Camurati-Engelmann disease: a report on four cases in one family. Calcified Tissue International 61 48–51. (https://doi.org/10.1007/s002239900293)
Holdsworth G, Roberts SJ & Ke HZ 2019 Novel actions of sclerostin on bone. Journal of Molecular Endocrinology 62 R167–R185. (https://doi.org/10.1530/JME-18-0176)
Ihde LL, Forrester DM, Gottsegen CJ, Masih S, Patel DB, Vachon LA, White EA & Matcuk Jr GR 2011 Sclerosing bone dysplasias: review and differentiation from other causes of osteosclerosis. RadioGraphics 31 1865–18 82. (https://doi.org/10.1148/rg.317115093)
James AW & Peault B 2019 Perivascular mesenchymal progenitors for bone regeneration. Journal of Orthopaedic Research 37 1221–1228. (https://doi.org/10.1002/jor.24284)
Janssens K, Gershoni-Baruch R, Guanabens N, Migone N, Ralston S, Bonduelle M, Lissens W, Van Maldergem L, Vanhoenacker F & Verbruggen L et al. 2000 Mutations in the gene encoding the latency-associated peptide of TGF-beta 1 cause Camurati-Engelmann disease. Nature Genetics 26 273–27 5. (https://doi.org/10.1038/81563)
Janssens K, Ten Dijke P, Ralston SH, Bergmann C & Van Hul W 2003 Transforming growth factor-beta 1 mutations in Camurati-Engelmann disease lead to increased signaling by altering either activation or secretion of the mutant protein. Journal of Biological Chemistry 278 7718–77 24. (https://doi.org/10.1074/jbc.M208857200)
Janssens K, Vanhoenacker F, Bonduelle M, Verbruggen L, Van Maldergem L, Ralston S, Guanabens N, Migone N, Wientroub S & Divizia MT et al. 2006 Camurati-Engelmann disease: review of the clinical, radiological, and molecular data of 24 families and implications for diagnosis and treatment. Journal of Medical Genetics 43 1–11. (https://doi.org/10.1136/jmg.2005.033522)
Kitaoka T, Tajima T, Nagasaki K, Kikuchi T, Yamamoto K, Michigami T, Okada S, Fujiwara I, Kokaji M & Mochizuki H et al. 2017 Safety and efficacy of treatment with asfotase alfa in patients with hypophosphatasia: results from a Japanese clinical trial. Clinical Endocrinology 87 10–19. (https://doi.org/10.1111/cen.13343)
Lee KS, Hong SH & Bae SC 2002 Both the Smad and p38 MAPK pathways play a crucial role in Runx2 expression following induction by transforming growth factor-beta and bone morphogenetic protein. Oncogene 21 7156–71 63. (https://doi.org/10.1038/sj.onc.1205937)
Leupin O, Piters E, Halleux C, Hu S, Kramer I, Morvan F, Bouwmeester T, Schirle M, Bueno-Lozano M & Fuentes FJ et al. 2011 Bone overgrowth-associated mutations in the LRP4 gene impair sclerostin facilitator function. Journal of Biological Chemistry 286 19489–19 500. (https://doi.org/10.1074/jbc.M110.190330)
Li X, Liu P, Liu W, Maye P, Zhang J, Zhang Y, Hurley M, Guo C, Boskey A & Sun L et al. 2005 Dkk2 has a role in terminal osteoblast differentiation and mineralized matrix formation. Nature Genetics 37 945–9 52. (https://doi.org/10.1038/ng1614)
Li X, Ominsky MS, Niu QT, Sun N, Daugherty B, D’Agostin D, Kurahara C, Gao Y, Cao J & Gong J et al. 2008 Targeted deletion of the sclerostin gene in mice results in increased bone formation and bone strength. Journal of Bone and Mineral Research 23 860–86 9. (https://doi.org/10.1359/jbmr.080216)
Li X, Ominsky MS, Warmington KS, Morony S, Gong J, Cao J, Gao Y, Shalhoub V, Tipton B & Haldankar R et al. 2009 Sclerostin antibody treatment increases bone formation, bone mass, and bone strength in a rat model of postmenopausal osteoporosis. Journal of Bone and Mineral Research 24 578–5 88. (https://doi.org/10.1359/jbmr.081206)
Matsumoto T, Miyake K, Yamamoto S, Orimo H, Miyake N, Odagaki Y, Adachi K, Iijima O, Narisawa S & Millan JL et al. 2011 Rescue of severe infantile hypophosphatasia mice by AAV-mediated sustained expression of soluble alkaline phosphatase. Human Gene Therapy 22 1355–13 64. (https://doi.org/10.1089/hum.2010.210)
Matsunobu T, Torigoe K, Ishikawa M, De Vega S, Kulkarni AB, Iwamoto Y & Yamada Y 2009 Critical roles of the TGF-beta type I receptor ALK5 in perichondrial formation and function, cartilage integrity, and osteoblast differentiation during growth plate development. Developmental Biology 332 325–3 3 8. (https://doi.org/10.1016/j.ydbio.2009.06.002)
McClung MR 2018 Romosozumab for the treatment of osteoporosis. Osteoporosis and Sarcopenia 4 11–15. (https://doi.org/10.1016/j.afos.2018.03.002)
Millan JL & Whyte MP 2016 Alkaline phosphatase and hypophosphatasia. Calcified Tissue International 98 398–416. (https://doi.org/10.1007/s00223-015-0079-1)
Miyazono K, Hellman U, Wernstedt C & Heldin CH 1988 Latent high molecular weight complex of transforming growth factor beta 1. Purification from human platelets and structural characterization. Journal of Biological Chemistry 263 6407–64 15.
Mornet E 2018 Hypophosphatasia. Metabolism: Clinical and Experimental 82 142–155. (https://doi.org/10.1016/j.metabol.2017.08.013)
Nuchel J, Ghatak S, Zuk AV, Illerhaus A, Morgelin M, Schonborn K, Blumbach K, Wickstrom SA, Krieg T & Sengle G et al. 2018 TGFB1 is secreted through an unconventional pathway dependent on the autophagic machinery and cytoskeletal regulators. Autophagy 14 465–486. (https://doi.org/10.1080/15548627.2017.1422850)
Reppe S, Noer A, Grimholt RM, Halldorsson BV, Medina-Gomez C, Gautvik VT, Olstad OK, Berg JP, Datta H & Estrada K et al. 2015 Methylation of bone SOST, its mRNA, and serum sclerostin levels correlate strongly with fracture risk in postmenopausal women. Journal of Bone and Mineral Research 30 249–2 56. (https://doi.org/10.1002/jbmr.2342)
Righetti M, Wach J, Desmarchelier R & Coury F 2018 Teriparatide treatment in an adult patient with hypophosphatasia exposed to bisphosphonate and revealed by bilateral atypical fractures. Joint Bone Spine 85 365–367. (https://doi.org/10.1016/j.jbspin.2017.12.001)
Rutkovskiy A, Stenslokken KO & Vaage IJ 2016 Osteoblast differentiation at a glance. Medical Science Monitor Basic Research 22 95–106. (https://doi.org/10.12659/msmbr.901142)
Salles JP 2020 Hypophosphatasia: biological and clinical aspects, avenues for therapy. Clinical Biochemist. Reviews 41 13–27. (https://doi.org/10.33176/AACB-19-00031)
Satou Y, Al-Shawafi HA, Sultana S, Makita S, Sohda M & Oda K 2012 Disulfide bonds are critical for tissue-nonspecific alkaline phosphatase function revealed by analysis of mutant proteins bearing a C(201)-Y or C(489)-S substitution associated with severe hypophosphatasia. Biochimica et Biophysica Acta 1822 581–58 8. (https://doi.org/10.1016/j.bbadis.2012.01.007)
Seefried L, Baumann J, Hemsley S, Hofmann C, Kunstmann E, Kiese B, Huang Y, Chivers S, Valentin MA & Borah B et al. 2017 Efficacy of anti-sclerostin monoclonal antibody BPS804 in adult patients with hypophosphatasia. Journal of Clinical Investigation 127 2148–2158. (https://doi.org/10.1172/JCI83731)
Shapiro JR & Lewiecki EM 2017 Hypophosphatasia in adults: clinical assessment and treatment considerations. Journal of Bone and Mineral Research 32 1977–1980. (https://doi.org/10.1002/jbmr.3226)
Shi Y & Massague J 2003 Mechanisms of TGF-beta signaling from cell membrane to the nucleus. Cell 113 685–700. (https://doi.org/10.1016/s0092-8674(0300432-x)
Shi M, Zhu J, Wang R, Chen X, Mi L, Walz T & Springer TA 2011 Latent TGF-beta structure and activation. Nature 474 343–34 9. (https://doi.org/10.1038/nature10152)
Sims NA & Martin TJ 2014 Coupling the activities of bone formation and resorption: a multitude of signals within the basic multicellular unit. BoneKEy Reports 3 481. (https://doi.org/10.1038/bonekey.2013.215)
Spranger JBP, Hall C, Nishimura G, Superti-Furga A & Unger S 2018 Bone Dysplasias: an Atlas of Genetic Disorders of Skeletal Development. (https://doi.org/10.1093/med/9780190626655.001.0001)
Stein SA, Witkop C, Hill S, Fallon MD, Viernstein L, Gucer G, Mckeever P, Long D, Altman J & Miller NR et al. 1983 Sclerosteosis: neurogenetic and pathophysiologic analysis of an American kinship. Neurology 33 267–2 77. (https://doi.org/10.1212/wnl.33.3.267)
Subbiah V, Madsen VS, Raymond AK, Benjamin RS & Ludwig JA 2010 Of mice and men: divergent risks of teriparatide-induced osteosarcoma. Osteoporosis International 21 1041–104 5. (https://doi.org/10.1007/s00198-009-1004-0)
Taketani T, Oyama C, Mihara A, Tanabe Y, Abe M, Hirade T, Yamamoto S, Bo R, Kanai R & Tadenuma T et al. 2015 Ex vivo expanded allogeneic mesenchymal stem cells with bone marrow transplantation improved osteogenesis in infants With severe hypophosphatasia. Cell Transplantation 24 1931–19 43. (https://doi.org/10.3727/096368914X685410)
Truswell AS 1958 Osteopetrosis with syndactyly; a morphological variant of Albers-Schonberg’s disease. Journal of Bone and Joint Surgery. British Volume 40–B 209–218.
Tuan RS 1998 Cellular and molecular regulation of embryonic skeletal development and morphogenesis. Cells and Materials 8 3–18.
Van Buchem FS, Hadders HN & Ubbens R 1955 An uncommon familial systemic disease of the skeleton: hyperostosis corticalis generalisata familiaris. Acta radiologica 44 109–1 20. (https://doi.org/10.3109/00016925509170789)
Van Buchem FSPP, Ptick JJG & Jaspar HHJ 1976 Hyperostosis Corticalis Generalisata Familiaris (Van Buchem’s Disease).
Van Hul W, Boudin E, Vanhoenacker FM & Mortier G 2019 Camurati-Engelmann disease. Calcified Tissue International 104 554–560. (https://doi.org/10.1007/s00223-019-00532-1)
van Lierop AH, Appelman-Dijkstra NM & Papapoulos SE 2017 Sclerostin deficiency in humans. Bone 96 51–62. (https://doi.org/10.1016/j.bone.2016.10.010)
van Lierop AH, Hamdy NA, Van Egmond ME, Bakker E, Dikkers FG & Papapoulos SE 2013 Van Buchem disease: clinical, biochemical, and densitometric features of patients and disease carriers. Journal of Bone and Mineral Research 28 848–8 54. (https://doi.org/10.1002/jbmr.1794)
Wallace SE, Lachman RS, Mekikian PB, Bui KK & Wilcox WR 2004 Marked phenotypic variability in progressive diaphyseal dysplasia (Camurati-Engelmann disease): report of a four-generation pedigree, identification of a mutation in TGFB1, and review. American Journal of Medical Genetics: Part A 129A 235–2 47. (https://doi.org/10.1002/ajmg.a.30148)
Weiss A & Attisano L 2013 The TGFbeta superfamily signaling pathway. Wiley Interdisciplinary Reviews: Developmental Biology 2 47–63. (https://doi.org/10.1002/wdev.86)
Wennberg C, Hessle L, Lundberg P, Mauro S, Narisawa S, Lerner UH & Millan JL 2000 Functional characterization of osteoblasts and osteoclasts from alkaline phosphatase knockout mice. Journal of Bone and Mineral Research 15 1879–18 88. (https://doi.org/10.1359/jbmr.2000.15.10.1879)
Whyte MP 2016 Hypophosphatasia – aetiology, nosology, pathogenesis, diagnosis and treatment. Nature Reviews: Endocrinology 12 233–2 46. (https://doi.org/10.1038/nrendo.2016.14)
Whyte MP 2017 Hypophosphatasia: an overview for 2017. Bone 102 15–25. (https://doi.org/10.1016/j.bone.2017.02.011)
Whyte MP, Magill HL, Fallon MD & Herrod HG 1986 Infantile hypophosphatasia: normalization of circulating bone alkaline phosphatase activity followed by skeletal remineralization. Evidence for an intact structural gene for tissue nonspecific alkaline phosphatase. Journal of Pediatrics 108 82–8 8. (https://doi.org/10.1016/s0022-3476(8680773-9)
Whyte MP, Mumm S & Deal C 2007 Adult hypophosphatasia treated with teriparatide. Journal of Clinical Endocrinology and Metabolism 92 1203–120 8. (https://doi.org/10.1210/jc.2006-1902)
Whyte MP, Greenberg CR, Salman NJ, Bober MB, Mcalister WH, Wenkert D, Van Sickle BJ, Simmons JH, Edgar TS & Bauer ML et al. 2012 Enzyme-replacement therapy in life-threatening hypophosphatasia. New England Journal of Medicine 366 904–9 13. (https://doi.org/10.1056/NEJMoa1106173)
Whyte MP, Rockman-Greenberg C, Ozono K, Riese R, Moseley S, Melian A, Thompson DD, Bishop N & Hofmann C 2016 Asfotase alfa treatment improves survival for perinatal and infantile hypophosphatasia. Journal of Clinical Endocrinology and Metabolism 101 334–3 42. (https://doi.org/10.1210/jc.2015-3462)
Wu M, Chen G & Li YP 2016 TGF-beta and BMP signaling in osteoblast, skeletal development, and bone formation, homeostasis and disease. Bone Research 4 16009. (https://doi.org/10.1038/boneres.2016.9)
Yasui T, Kadono Y, Nakamura M, Oshima Y, Matsumoto T, Masuda H, Hirose J, Omata Y, Yasuda H & Imamura T et al. 2011 Regulation of RANKL-induced osteoclastogenesis by TGF-beta through molecular interaction between Smad3 and Traf6. Journal of Bone and Mineral Research 26 1447–14 56. (https://doi.org/10.1002/jbmr.357)
Yuldashev AJ, Shin CH, Kim YS, Jang WY, Park MS, Chae JH, Yoo WJ, Choi IH, Kim OH & Cho TJ 2017 Orthopedic manifestations of Type I Camurati-Engelmann disease. Clinics in Orthopedic Surgery 9 109–115. (https://doi.org/10.4055/cios.2017.9.1.109)
Zhou X, Cui Y, Zhou X & Han J 2012 Phosphate/pyrophosphate and MV-related proteins in mineralisation: discoveries from mouse models. International Journal of Biological Sciences 8 778–7 90. (https://doi.org/10.7150/ijbs.4538)
