SEARCH

SEARCH BY CITATION

Keywords:

  • OSTEOPETROSIS;
  • SNX10;
  • VÄSTERBOTTEN;
  • HSCT;
  • DIAGNOSIS

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Note added in proof
  8. Disclosures
  9. Acknowledgements
  10. References
  11. Supporting Information

Human Autosomal Recessive Osteopetrosis (ARO) is a genetically heterogeneous disorder caused by reduced bone resorption by osteoclasts. In 2000, we found that mutations in the TCIRG1 gene encoding for a subunit of the proton pump (V-ATPase) are responsible for more than one-half of ARO cases. Since then, five additional genes have been demonstrated to be involved in the pathogenesis of the disease, leaving approximately 25% of cases that could not be associated with a genotype. Very recently, a mutation in the sorting nexin 10 (SNX10) gene, whose product is suggested to interact with the proton pump, has been found in 3 consanguineous families of Palestinian origin, thus adding a new candidate gene in patients not previously classified. Here we report the identification of 9 novel mutations in this gene in 14 ARO patients from 12 unrelated families of different geographic origin. Interestingly, we define the molecular defect in three cases of “Västerbottenian osteopetrosis,” named for the Swedish Province where a higher incidence of the disease has been reported. In our cohort of more than 310 patients from all over the world, SNX10-dependent ARO constitutes 4% of the cases, with a frequency comparable to the receptor activator of NF-κB ligand (RANKL), receptor activator of NF-κB (RANK) and osteopetrosis-associated transmembrane protein 1 (OSTM1)-dependent subsets. Although the clinical presentation is relatively variable in severity, bone seems to be the only affected tissue and the defect can be almost completely rescued by hematopoietic stem cell transplantation (HSCT). These results confirm the involvement of the SNX10 gene in human ARO and identify a new subset with a relatively favorable prognosis as compared to TCIRG1-dependent cases. Further analyses will help to better understand the role of SNX10 in osteoclast physiology and verify whether this protein might be considered a new target for selective antiresorptive therapies. © 2013 American Society for Bone and Mineral Research.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Note added in proof
  8. Disclosures
  9. Acknowledgements
  10. References
  11. Supporting Information

Autosomal recessive osteopetrosis (ARO) is a rare inherited disease resulting from a deficit in bone resorption caused by osteoclast malfunction or absence. It presents early in life with extreme sclerosis of the skeleton and is associated with a number of severe clinical manifestations, including cranial nerve dysfunction, such as deafness and blindness, pancytopenia, and hepatosplenomegaly caused by severe bone marrow failure beginning in early infancy or in fetal life. The disease is usually early lethal unless treated with hematopoietic stem cell transplantation (HSCT). The estimated incidence is 1 in 250,000 newborns, although it is much higher in specific geographic areas with a consanguineous population, such as Costa Rica,1 the Chuvash Republic of Russia,2 and the Province of Västerbotten in Northern Sweden.3 Current knowledge on the molecular bases of human ARO has identified the gene responsible for the disease in the Costa Rican and the Chuvashian subgroups, whereas the Västerbottenian cluster represents a somewhat less severe form3 of yet unknown etiology.

Recently, a mutation in the Sorting Nexin 10 (SNX10) gene has been found in 3 consanguineous families of Palestinian origin in which the patients displayed a milder presentation than in classic ARO.4 On this basis, we reasoned that SNX10 could be considered a good candidate gene associated with “Västerbottenian osteopetrosis” as well as in other cases of ARO with an unknown molecular defect.

Here we describe the identification of 9 novel mutations in the SNX10 gene in 12 reportedly unrelated families; importantly, we demonstrate that this gene is responsible for Västerbottenian osteopetrosis. Our findings highlight the molecular heterogeneity of SNX10-dependent ARO, which corresponds to 4% of patients in our cohort. Clinical data show that the severity of the disease can vary; HSCT may not always be required, but it can cure the disease even when performed later in life.

Altogether these data confirm the molecular and clinical complexity of human ARO and the importance of a precise molecular diagnosis to facilitate treatment-related decisions and to provide information regarding the prognosis of individual patients.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Note added in proof
  8. Disclosures
  9. Acknowledgements
  10. References
  11. Supporting Information

Samples

A cohort of more than 310 ARO patients was gathered in our Institutes through a network of worldwide collaborations. Criteria for inclusion in the study were diagnosis in infancy, high bone mineral density, bone marrow failure with consequent pancytopenia, hepatosplenomegaly, and in some cases cranial nerve palsies.

Specimens, including blood and DNA samples, were collected from patients after their parents provided informed consent. This research complies with the standards established by the Independent Ethical Committee of the Humanitas Clinical and Research Centre, the “Ethikkommission der Universität Ulm,” and the Ethical Committee of the Charité-Universitaetsmedizin Berlin.

Molecular studies

The molecular analysis of genes known to be responsible for the different types of the disease (TCIRG1, CLCN7, OSTM1, PLEKHM1, receptor activator of NF-κB ligand [RANKL], and receptor activator of NF-κB [RANK]) was performed by amplification and direct sequencing of exons and intron-exon boundaries as described.1, 5–10

The SNX10 gene was amplified using primers and conditions kindly provided by Aker and colleagues (Hebrew University Medical Center, Jerusalem). The mutation nomenclature conforms to www.hgvs.org/mutnomen.

In silico studies

The human SNX10 protein sequence was retrieved from Uniprot (http://www.uniprot.org/uniprot/Q9Y5x0). The PX sequence (aa10-127) was searched in PDB using the program Psi-Blast (http://www.rcsb.org/pdb/home/home.do).

The PsiBlast Hit 1OCU corresponding to the crystal structure of the yeast Grd19p protein with a bound ligand was chosen as the closest homologue to be used as a template for homology modeling.11 The sequence of 1OCU and SNX10 PX were realigned with other sequences that belong to the PX-only subfamily using ClustalW. Models of PX wild-type and mutants were predicted using the program Modeller (ver 9.10). Molecular graphics and analyses were performed with the UCSF Chimera package. PovRay was used to generate high-quality images.

The effect of the mutation c.111 + 5G > C was tested using the software www.fruitfly.org, www.cbs.dtu.dk, and www.linux1.softberry.com.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Note added in proof
  8. Disclosures
  9. Acknowledgements
  10. References
  11. Supporting Information

Genetic findings

In the last decade, more than 310 osteopetrotic patients from all over the world have been screened by our groups for genes responsible for human ARO. This allowed us to select 60 patients who did not have an identified genotype and could be investigated for mutations in the SNX10 gene.

The molecular analysis led to the identification of biallelic mutations in the SNX10 gene in 14 patients from 12 unrelated families, for a total of 9 novel mutations: 3 missense, 3 nonsense, and 3 splicing defects (Table 1).

Table 1. Molecular Findings in 14 New SNX10-Dependent Patients
Patient#Genomic changeacDNA changebPredicted effect
  • a

    Accession number genomic sequence of the SNX10 gene: NC_000007.

  • b

    Accession number of the SNX10 cDNA: NM_001199835.1; the numbering used starts with nucleotide +1 for the A of the ATG-translation initiation codon.

  • c

    The change is expected to affect splicing since the splice donor is changed.

  • d

    A different amino acid substitution at the same position is already reported.

  • e

    The change might affect splicing since it is close to the splice donor.

1g.72742G > Tc.212 + 1G > Tr.spl?c
 g.72742G > Tc.212 + 1G > Tr.spl?c
2g.72742G > Tc.212 + 1G > Tr.spl?c
 g.72742G > Tc.212 + 1G > Tr.spl?c
3g.72742G > Tc.212 + 1G > Tr.spl?c
 g.72742G > Tc.212 + 1G > Tr.spl?c
4g.69151A > Cc.95A > Cp.Tyr32Ser
 g.69151A > Cc.95A > Cp.Tyr32Ser
5A-5Bg.69151A > Cc.95A > Cp.Tyr32Ser
 g.69151A > Cc.95A > Cp.Tyr32Ser
6g.72681G > Cc.152G > Cp.Arg51Prod
 g.72681G > Cc.152G > Cp.Arg51Prod
7A-7Bg.72681G > Cc.152G > Cp.Arg51Prod
 g.72681G > Cc.152G > Cp.Arg51Prod
8g.69102C > Tc.46C > Tp.Arg16X
 g.69102C > Tc.46C > Tp.Arg16X
9g.69103G > Tc.47G > Tp.Arg16Leu
 g.72713C > Tc.184C > Tp.Gln62X
10g.73252G > Tc.311 + 1G > Tr.spl?c
 g.73252G > Tc.311 + 1G > Tr.spl?c
11g.69143C > Ac.87C > Ap.Tyr29X
 g.69143C > Ac.87C > Ap.Tyr29X
12g.69746G > Cc.111 + 5G > Cr.(spl?)e
 g.69746G > Cc.111 + 5G > Cr.(spl?)e

Three patients (patients 1, 2, and 3) were all found to be homozygous for the same nucleotide change at the donor splice site of exon 4 (c.212 + 1G > T).

Patient 4 was homozygous for a c.95A > C mutation predicted to cause a p.Tyr32Ser amino acid change, and her parents were shown to be heterozygous for the same nucleotide change. We found the same mutation in the homozygous state in 2 affected siblings (patients 5A and 5B) from an unrelated family; both their parents and their healthy sister bore the mutation in the heterozygous state.

Patient 6 was homozygous for a transversion (c.152G > C) predicted to lead to an amino acid substitution (p.Arg51Pro); both his consanguineous parents were heterozygous for this nucleotide change. The same mutation was found in the homozygous state in 2 affected siblings (patients 7A and 7B) from an unrelated family. Their consanguineous parents were heterozygous carriers while their unaffected sibling showed the wild-type codon.

Patient 8 was homozygous for a transition (c.46C > T) predicted to cause a nonsense mutation at codon 16 (p.Arg16X). Her parents' DNA was not available for analysis.

Patient 9 was compound heterozygous for 2 nucleotide changes, c.47G > T and c.184C > T, predicted to cause an amino acid substitution at codon 16 (p.Arg16Leu) and a stop at codon 62 (p.Gln62X), respectively. His mother was heterozygous for the first mutation and his father for the second mutation.

Patient 10 was homozygous for a nucleotide change at the donor splice-site of exon 5 (c.311 + 1G > T). The same mutation was found in the heterozygous state in his consanguineous parents and in 4 out of his 5 healthy siblings.

Patient 11 was homozygous for a transversion (c.87C > A) predicted to cause a stop at codon 29 (p.Tyr29X).

Patient 12 was homozygous for a nucleotide change close to the donor splice-site of exon 3 (c.111 + 5G > C). The same mutation was found in the heterozygous state in his consanguineous parents.

The missense substitutions (p.Arg16Leu, p.Tyr32Ser, and p.Arg51Pro) were not found in more than 100 chromosomes from healthy unrelated individuals from the same geographical areas, and were not present in dbSNP135; therefore, they are unlikely to be neutral polymorphisms. Alignment of SNX10 protein sequences from several species further supports this idea, because the mutated residues are strongly conserved in evolution (Supplementary Fig. 1).

In silico studies

All the mutations identified fall in the Phox-homology (PX) domain of SNX10, which is the only functional domain.12 Modeling of the PX domain of SNX10 protein showed that this region is folded into a globular shape, composed of 3 exposed beta-strands followed by 3 alpha-helices that well overlie the alpha-helices and beta-strands of the chosen template Grd19p. The model also showed the electropositive basic pocket that is responsible for binding to the negatively charged phosphate groups (Fig. 1A).

thumbnail image

Figure 1. Mutations in relation to the SNX10 protein structure. (A) Three-dimensional model of PX domain of human SNX10 protein (in blue), determined by homology modeling using the protein structure of yeast Grd19p (1OCU, in gray) as a template. The three residues affected by missense mutations in our patients are depicted in magenta; putative functional residues are depicted in green. (B) Alignment of PX domains of human SNX10 (Q9Y5X0-1), Grd19p (PDB ID: 1OCU), and human SNX3 (O60493-1). Secondary structure alignment was obtained from the structure of Grd19p.11 Numbers on the left hand side of each sequence denote the full-length sequence position. Amino acid colors as in A.

Download figure to PowerPoint

SNX10 mutations occurred at residues not involved in phosphatidylinositol binding and located in the three beta-strands at the opposite side of the binding pocket, so they could not directly affect the binding activity.

Mutation of the exposed positively charged residue Arg16 into Leucine might disrupt contacts to other proteins and therefore affect SNX10 function.

Tyr32 occupies a buried position. Mutation p.Tyr32Ser might locally destabilize the internal packing and affect the secondary structure of the neighboring beta-strand; this could indirectly impair binding to other proteins. Alternatively, it could affect phosphatidylinosytol-3-phosphate binding by altering the side chain conformation of the conserved Arg94 that is known to directly bind the phosphate group.13

Arg51 makes contact to the neighboring beta-strand and might be important for local main chain conformation. Mutation p.Arg51Pro could either distort the main chain conformation and/or affect binding to another protein.

Grd19p is the yeast homologue of human SNX3, which is another member of the PX-only family. Alignment of the PX domain of human SNX10 protein, Grd19p and human SNX3 showed that Tyr32 and Arg51 are conserved, whereas Arg16 was not (Fig. 1B). However, multispecies alignment of SNX10 proteins highlighted that this residue is maintained, suggesting a possible role specific to SNX10 and not shared by other sorting nexins (Supplementary Fig. 1).

Clinical evaluation of patients

We carefully reviewed the clinical history of these patients in order to identify the specific features of SNX10-dependent ARO (Table 2).

Table 2. Clinical Features at Diagnosis and HSCT Outcome in SNX10-Dependent ARO
Patient#Age at onsetBone fracturesNeurological defectsHematological defectsOther featuresAge at HSCTHSC origin; donorOutcome and follow-up
  • HSCT = hematopoietic stem cell transplantation; ARO = autosomal recessive osteopetrosis; HSC = hematopoietic stem cell; Hb = hemoglobin; MUD = matched unrelated donor; VOD = veno-occlusive disease; MRD = matched related donor; RD = related donor; MMRD = mismatched related donor.

  • a

    Defects documented after diagnosis.

13 monthsNoHydrocephalus, nystagmus, impaired visionAnemia (Hb 10.2 g/dL)33 monthsBone marrow; MUDBone rescue; alive and well at 11 years; vision partially maintained
22.5 monthsNoStrabismusAnemia (Hb 7.9 g/dL)Hypertelorism, frontal bossing, upper airway obstruction4 monthsBone marrow; MUDDead shortly after HSCT due to VOD and pulmonary hypertension
3Early infancyNoBlindness, hydrocephalus Upper airway obstructionNot doneDead at 10 years due to bacterial infection
43 monthsNoBlindnessAnemia (Hb 8.7 g/dL)Stridor, airway difficultiesNot doneDead at 20 months
5A4 monthsNoImpaired visionAnemia (Hb 11.3 g/dL after transfusion at 6 months)aFever, decreased urine output, failure to thrive7 monthsBone marrow; MRDEngrafted; dead at 1 year due to multiple infections and pulmonary hemorrhage
5BAt birthNoImpaired visionHepatosplenomegalyHypocalcaemia, frontal bossing6 monthsPeripheral blood; partially matched RDEngrafted; bone improvement; dead at 18 months due to multisystem organ failure
64 monthsNoImpaired vision, deafnessHepatosplenomegaly anemia (Hb 9.5 g/dL)Not done Alive at 4 years
7A3 monthsNoMild strabismusHepatosplenomegaly, anemia (Hb 6.7 g/dL), thrombocytopenia (86.000/µL)Frontal bossing, upper airway obstruction7 monthsBone marrow; MRDEngrafted; alive and well at 22 years, normal bone, impaired vision
7B3 monthsNoImpaired vision, developmental delayHepatosplenomegaly, anemia (Hb 9.4 g/dL)Dolichocephalous, coxa vara, scoliosis4 monthsBone marrow; MRD25% chimerism; alive at 18 years, intermediate osteopetrosis, blind
86 monthsNoImpaired visionAnemia (Hb 8.9 g/dL)Frontal bossingNot reportedLost to follow-up
94 monthsNoHydrocephalusAnemia (Hb 7.9 g/dL at 6 months)a20 monthsBone marrow; MRDAlive and reasonably well at 30 months
101 yearSeveralImpaired visionAnemia (Hb 10.4 g/dL)Not done Alive at 22 years, wheelchair-bound due to recurrent non-healing fractures
117 monthsNoImpaired visionAnemia (Hb 8.1 g/dL)1st at 15 months; 2nd at 81 months1st peripheral blood, 2nd bone marrow; MUDAlive and well at 20 years
123.5 monthsNoBlindnessSplenomegaly, anemia (Hb 9 g/dL)Upper airway obstruction9 monthsPeripheral blood; MMRDAlive and well at 3.5 years

Patients 1, 2, and 3 were born in unrelated families from the same region in Northern Sweden (Fig. 2A) with no evidence for parental consanguinity. Patient 1 came to attention because of a large skull at 3 months of age raising the suspicion of hydrocephalus. Nystagmus, diminished visual acuity, and late dentition were also present. Radiological investigation performed at 2 years of age showed a generalized increase in bone density (Fig. 2B, C; left panel), thus leading to the diagnosis of osteopetrosis, Västerbottenian type. A moderate anemia (hemoglobin 10.2 g/dL) was also present; therefore, at 33 months of age the patient received an HSCT from a human leukocyte antigen (HLA)-matched unrelated donor, after conditioning according to a protocol recently recommended by the European Group for Bone Marrow Transplantation-European Society for Immunodeficiencies (EBMT-ESID) (www.esid.org/downloads/OPGuidelines-2011). In the post-HSCT period, veno-occlusive disease (VOD) with cytomegalovirus (CMV) reactivation and hypercalcemia occurred and were treated with no further complications. He achieved full engraftment and an almost complete rescue of his sclerotic bone phenotype (Fig. 2D). Two months after transplantation he was back home in reasonably good conditions. Vision was partially maintained; at 11 years he is alive and well (Fig. 2C; right panel), attending school with no specific assistance.

thumbnail image

Figure 2. (A) Map of Sweden with the Province of Västerbotten in dark brown. The patients with the “Västerbottenian form” come from a small village north of the Province capital Umeå. (B) X-rays of patient 1 showing a diffuse increase in bone density, signe du masque, and bone-in-bone appearance. (C) Pictures of patient 1. Before HSCT (left panel) frontal bossing is particularly pronounced, whereas after HSCT (right panel) the patient displays a normal facies. (D) X-rays of a lower limb of patient 1 before (left panel) and after HSCT (right panel), showing a clear reduction in bone density 1 year after transplantation.

Download figure to PowerPoint

Patient 2 was diagnosed at 2.5 months due to hypertelorism, frontal bossing, upper airway obstruction, strabismus, severe anemia (hemoglobin 7.9 g/dL), and a dense sclerotic skeleton. She received a transplant at 4 months of age with an HLA-matched unrelated donor, after conditioning according to EBMT-ESID guidelines. Full donor chimerism was achieved; however, the transplant was complicated by VOD and pulmonary hypertension, leading to her death.

Patient 3 displayed manifestations of the disease early in infancy: she presented with upper airway obstruction, blindness, and hydrocephalus requiring a ventriculoperitoneal (VP) shunt; in addition, delayed tooth eruption was noted at 30 months. An X-ray of the skull clearly demonstrated an osteopetrotic phenotype. The family refused HSCT and she died at 10 years of age due to a bacterial infection.

Patient 4 was born from healthy nonconsanguineous parents of Mexican descent. At 3 months of age she presented with stridor and airway difficulties; therefore, radiological examination was performed, leading to the diagnosis. The patient also showed anemia (hemoglobin 8.7 g/dL) and narrowing of the optic canals with significant visual impairment. She underwent a surgery for the unroofing of the optic canal at 4 months, but despite this she progressively lost all her visual capacity and by 6 months she was essentially blind with documented optic atrophy. At 15 months of age she developed hydrocephalus requiring a VP shunt and hypocalcemia, treated with calcitriol without benefit. At 20 months of age, during preparation for transplantation, she displayed mild to moderate developmental delay, both in regard to language and to gross motor milestones, frontal bossing, and roving eye movements. Ear, nose, and throat (ENT) assessment revealed severe apnea due to choanal obstruction, whereas hearing was normal; in addition, she developed a right VII nerve palsy. Severe anemia was noted (hemoglobin 6.2 g/dL) as well as thrombocytopenia (platelet count 96,000/µL). At this point, just prior to the admission for transplantation, she acutely arrested and died, probably due to her airway compromise.

Patient 5A was the second child of healthy, unrelated, Spanish-speaking American parents. She was diagnosed at 4 months of age, when she was admitted to the local hospital for fever, decreased urine output, and failure to thrive. Radiological examination revealed increased bone density with skeletal changes suggestive of osteopetrosis, whereas neurological investigation showed impaired vision. She presented to the transplant center at 6 months of age with failure to thrive (4.8 kg and 59.9 cm), anemia, thrombocytopenia, and obstructive sleep apnea. Prior to being admitted for transplantation, she had a tracheostomy tube placed due to the severity of sleep apnea. At 7 months of age she received a T-replete bone marrow graft from an HLA-matched family donor, after conditioning with busulfan and cyclophosphamide, and cyclosporine for graft-versus-host disease (GVHD) prophylaxis. She achieved full engraftment and donor chimerism. Post-HSCT complications included colonization with multiple viruses, such as adenovirus (tracheal, day +18 through day +29; and stool, day +71 through day +115) and CMV (tracheal, day +93 through day +98). Haemophilus influenzae was cultured from the tracheostomy tube on day –6 through day +4. On day +82, she began having hematemesis that may have been related to diffuse alveolar hemorrhage, diagnosed on day +90. She did not recover and died on day +135.

Patient 5B was the younger brother of patient 5A. He started treatment with calcitriol soon after birth due to hypocalcemia and was diagnosed with osteopetrosis at approximately 2 months of age. He displayed frontal bossing, impaired vision, and hepatosplenomegaly. Prior to HSCT, he underwent a tonsillectomy for apnea. At 6 months of age he received an HLA-partially matched, granulocyte colony-stimulating factor (G-CSF)-mobilized, and CD3-depleted family donor transplant after a reduced intensity preparative regimen consisting of fludarabine, thiotepa, melphalan, OKT3, rituximab, and cyclosporine. He achieved myeloid engraftment on day +17 and platelet engraftment on day +15. Posttransplantation complications included acute skin GVHD (grade 2), beginning at day +41 and lasting for a week. He later developed limited chronic skin GVHD. He achieved full donor chimerism and a reduction in bone mineral density by almost 50% was seen by 6 months post-HSCT by quantitative computed tomography. In addition, while pre-HSCT auditory brainstem response (ABR) showed moderate conductive hearing loss bilaterally, the 3- and 6-month post-HSCT ABRs were normal with bilateral middle ear dysfunction. Unfortunately, he developed methicillin resistant Staphylococcus aureus sepsis and died on day +353 of multisystem organ failure.

Patient 6 was born from healthy consanguineous parents (first degree consanguinity) of Turkish origin. At 4 months he displayed generalized osteosclerosis, visual impairment, deafness, anemia (hemoglobin 9.5 g/dL), and hepatosplenomegaly. HSCT has not yet been performed due to lack of a suitable donor. At present, he is alive at 4 years of age with anemia and thrombocytopenia.

Patients 7A and 7B are the second and third child of healthy consanguineous Turkish parents (second degree consanguinity). Patient 7A was admitted to the clinic at 3 months of age due to upper airway obstruction, massive hepatosplenomegaly, anemia (hemoglobin 6.7 g/dL), and thrombocytopenia (platelet count 86,000/µL). The diagnosis of osteopetrosis was confirmed by X-rays of extremities. At the age of 7 months he received a bone marrow transplant from an HLA-identical family donor after receiving myeloablative conditioning with busulfan and cyclophosphamide combined with fractionated radiation of the enlarged spleen (cumulative dose 7 Gy). After prolonged aplasia, complete donor engraftment was achieved. He experienced acute skin GVHD (grade 1-2) and limited chronic GVHD. The patient is alive and well at 21 years after HSCT with full donor chimerism; however, his vision is impaired with bilateral optical nerve atrophy.

Patient 7B is the younger sister of patient 7A. Osteopetrosis was diagnosed at the age of 3 months due to frontal bossing, anemia (hemoglobin 9.4 g/dL), liver enlargement, and abnormal eye movements. At the age of 4 months she received a bone marrow transplant from a matched related donor and initially showed the expected radiological improvement. However, after 5 years she again developed an osteopetrotic phenotype (Supplementary Fig. 2) with hydrocephalus and multiple fractures, and indeed the overall donor chimerism was determined to be 25% with predominance of donor T-cells and complete lack of donor macrophages. The patient is alive at 18.5 years of age with mental and growth retardation (height at 3rd centile), coxa vara, scoliosis, blindness, epilepsy, and behavioral problems.

Patient 8 was the sixth child of healthy consanguineous parents of Pathan descent. He was diagnosed at 6 months of age due to frontal bossing, impaired vision, and anemia (hemoglobin 8.9 g/dL); an elder sister had expired at 7 years of age with the same diagnosis. Soon after, the family moved and the patient was lost to follow-up.

Patient 9 was born from healthy unrelated parents from Belgium. At the age of 4 months he presented with a gradual increase in head circumference, resulting from hydrocephalus, for which he underwent a third ventriculostomy procedure. From the age of 6 months, a decreased rate of growth was noted. Because of failure to thrive and persistent fever, he was hospitalized; anemia was documented (hemoglobin 7.9 g/dL) as well as increased bone density on radiography of the thorax. Further radiographic examination confirmed generalized osteopetrosis, “bone-in-bone,” and sandwich vertebrae. Bone biopsy at 15 months of age showed the presence of osteoclasts (Supplementary Fig. 3). Physical examination at the time of diagnosis revealed scaphocephaly with frontal bossing. A small fontanel could be palpated. Horizontal nystagmus was present, although vision was thought to be preserved. Abdominal palpation and ultrasound did not reveal hepatosplenomegaly. Magnetic resonance imaging (MRI) of the brain revealed atrophy of the optic nerves, stenosis of the petrosal canal, and narrowing of the internal acoustic canal. He received an HSCT from an HLA-identical parental donor at 20 months of age. The most important posttransplantation complications were a CMV reactivation and acute skin GVHD (grade 3), which was successfully treated with prednisone and tacrolimus. Presently, at 30 months of age, he is very active with normal neuromotor development. Hearing and visual capacity also remains normal.

Patient 10 was born from healthy Muslim consanguineous parents (first-degree consanguinity) from Pakistan. At 12 months of age he presented with increased bone density, several fractures, impaired vision, and mild anemia (hemoglobin 10.4 g/dL). He did not receive HSCT because there was no donor available. Currently he has impaired but functional vision and normal blood counts. He has had recurrent non-healing fractures of the femurs and is therefore wheelchair-bound. He is alive at 22 years of age without HSCT.

Patient 11 has been already described (patient 4 in Mazzolari and colleagues, 2009).14 Briefly, she was born from healthy nonconsanguineous parents of Italian origin. She was diagnosed at 7 months due to a severe osteopetrotic phenotype, received 2 HSCTs (the second when she was almost 7 years old), and is now alive and well, 13 years after transplantation. Another affected sibling was born to this family 10 years later; she's described in the same article (Patient 10 in Mazzolari and colleagues, 2009)14; the molecular analysis could be performed only on the former sibling, due to lack of sample from the latter.

Patient 12 was the first child of healthy, related, German-speaking Sinti parents (second-degree consanguinity). At the age of 3.5 months he was admitted for hematuria, upper airway infection, and gastroenteritis. Hepatosplenomegaly, thrombocytopenia, anemia (hemoglobin 9 g/dL), as well as binocular nystagmus were noted and a CMV infection and Norovirus enteritis were diagnosed. At the age of 6.5 months a bone marrow evaluation was performed because of persisting anemia and thrombocytopenia, which excluded leukemia. Because of the clinical presentation, osteopetrosis was within the differential diagnosis and confirmed by X-rays. He presented to the transplant center at 7 months of age with anemia (hemoglobin 9.2 g/dL), splenomegaly, frontal bossing, upper airway obstruction, and blindness. Bone biopsy was performed and showed the presence of osteoclasts (data not shown). At 9 months of age he received a CD34+-selected peripheral stem cell graft from his HLA-haploidentical father, after conditioning according to the EBMT-ESID guidelines. The transplant course was uneventful; he achieved engraftment and full donor chimerism and is alive and well 3 years after HSCT.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Note added in proof
  8. Disclosures
  9. Acknowledgements
  10. References
  11. Supporting Information

ARO is a genetic bone disease characterized by increased bone density due to a failure in bone resorption. In the last decade, it has been demonstrated to be genetically heterogeneous and it has been shown that the specific molecular defect defines subgroups with different prognostic and therapeutic implications.15 Of note, to date, in approximately 25% of patients the specific genetic defect is not known, confirming that our understanding of bone biology related to this disorder is still incomplete.

Recently, a new gene, Sorting Nexin 10 (SNX10), has been shown to be involved in the process of osteoclast differentiation and function upon RANKL-stimulation.16 SNX10 belongs to a family of about 30 proteins sharing a common phospholipid binding motif, the PX domain, which is involved in protein sorting and membrane trafficking.17 The important role of SNX10 in bone physiology has been further demonstrated by the identification of a single mutation in this gene in 8 ARO patients from the same geographical area.4

Here we report the identification of 9 novel mutations in 14 patients, consisting of 3 nonsense, 3 missense, and 3 splicing defects, all located in the PX domain.

The nonsense mutations (p.Arg16X, p.Tyr29X, and p.Gln62X) are predicted to lead to the production of highly truncated forms of SNX10 protein, in which a large part of the functional domain is lacking. The missense substitutions (p.Arg16Leu, p.Tyr32Ser, and p.Arg51Pro) affect residues that are strongly conserved in evolution. Of note, a different amino acid change has been already reported affecting Arginine 51 by Aker and colleagues.4 On the basis of sequence and structural similarity of the PX domains, the missense substitutions herein reported are predicted not to directly affect the phosphoinositide binding property of SNX10. According to our theoretical 3D model of the SNX10 PX domain based on homology modeling, Arg16Leu might disrupt interactions with other proteins, Tyr32Ser might destabilize the internal packing, and Arg51Pro might either alter the main chain conformation or affect binding to another protein. Regarding the mutations predicted to cause splicing defects, the consequence at the cDNA level could not be verified due to lack of samples. However, the c.212 + 1G > T and c.311 + 1G > T mutations are expected to impair the splicing process because the nucleotide changes affect exactly the donor splice sites of exons 4 and 5, respectively, whereas the c.111 + 5G > C mutation is importantly predicted to reduce the strength of the donor splice site of exon 3.

The clinical data of our patients show that SNX10-dependent ARO presents a range of severity, without any obvious relation to the specific molecular defect in this gene. In all of them the clinical onset of the disease was apparent in early infancy, leading to the establishment of the diagnosis. All the patients presented with anemia at diagnosis or during their clinical history, whereas none of them displayed overt immunological deficits. The large majority had secondary neurological defects; hydrocephalus was reported in 5 out of 14 (patients 1, 3, and 9 at diagnosis; patient 4 at 15 months; patient 7B 3 months after HSCT) and in 1 patient a right VII nerve palsy was present (patient 4). Bone biopsy was performed in 2 patients (patients 9 and 12) and showed an osteoclast-rich osteopetrosis. Nine patients received HSCT (patients 1, 2, 5A, 5B, 7A, 7B, 9, 11, and 12), and the age at transplantation was between 4 and 81 months; 3 of them died due to post-HSCT complications (patients 2, 5A, and 5B). Interestingly, 3 of the patients that are still alive underwent transplantation later in life (patients 1, 9, and 11). Of the remaining 5 patients, 1 died prior to starting her scheduled transplant procedure (patient 4); in 4 patients HSCT was not performed (patients 3, 6, 8, and 10) and 1 of them remains alive at 22 years old (patient 10). Overall, the outcome seems to be more benign as compared to the TCIRG1-dependent form.

Whereas in the original report from Aker and colleagues4 all the patients were of Palestinian origin, those described herein come from very different geographic areas. In particular, 3 of them come from the Province of Västerbotten, a small region in Northern Sweden originally settled by Sami people. Despite no consanguinity being reported in these 3 families, it is likely that a founder effect occurred in this case, as is reported for the Costa Rican and the Chuvashian populations.1, 2 Accordingly, a recent study aimed at characterizing the genetic structure of the Swedish population highlighted the presence of an increased autozygosity in the northernmost counties also comprising Västerbotten, likely arisen due to a small population and vast geographical distances between settlements in the North.18 Regarding the mutations identified in patients 4, 5A, and 5B, on one hand, and patients 6, 7A, and 7B, on the other, the possibility of a founder effect could not be excluded based on the similar ethnic origin of the patients and on the pattern of single-nucleotide polymorphisms (SNPs) genotyped in SNX10 gene, even though further genetic analyses would be required for confirmation.

In conclusion, our data confirm the molecular and clinical complexity of human ARO. Further studies are required to better understand the pathogenesis of the SNX10-dependent form that represents 4% of cases in our cohort, a frequency comparable to the RANKL-, RANK-, and OSTM1-dependent subsets. In addition, our findings have important therapeutic implications: the response to HSCT, together with the absence of primary neurological involvement that would otherwise preclude this option, underlines that screening for SNX10 mutations must be considered in the workup of newly diagnosed patients with ARO.

The primary cellular localization of SNX10 in mammalian cells seems to be in endosomes and its overexpression has been shown to induce the formation of giant vacuoles, which is dependent of V-ATPase activity.13, 19 Accordingly, an interaction with the V-ATPase subunit d1 was shown.19 These findings, together with SNX10 upregulation upon RANKL-induced osteoclastogenesis16 and the strong expression of Snx10 in osteoclasts during murine embryonic development suggest that it could play a critical role in targeting the V-ATPase to the ruffled border.20 Besides the regulation of endosomal trafficking, a role of SNX10 in the formation of primary cilia was suggested in zebrafish.19 However, none of our patients showed any typical sign of a ciliopathy. The availability of samples from non-transplanted SNX10-dependent patients will allow further clarifying the function of this protein in osteoclast physiology. At the same time, our molecular data, together with the absence of symptoms not related to the bone defect, suggest that SNX10 could be evaluated as a new target for the development of selective antiresorptive therapies.

Note added in proof

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Note added in proof
  8. Disclosures
  9. Acknowledgements
  10. References
  11. Supporting Information

We have identified an additional unrelated patient displaying Vasterbottenian osteopetrosis and bearing the same mutation as the others with the same origin herein described.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Note added in proof
  8. Disclosures
  9. Acknowledgements
  10. References
  11. Supporting Information

This work was partially supported by the Telethon Foundation (grant GGP12178 to CS); by the PRIN Project (200999KRFW-002 to PV); by Giovani Ricercatori from Ministero della Salute (grant GR-2008-1134625 to CS); and by PNR-CNR Aging Program 2012-2014. This work was partially performed on behalf of the ESID and the Inborn Error Working Party of the EBMT. We are grateful to the affected individuals and their families for their cooperation. In particular, we thank the parents of patient 1 for giving the consent to publish the pictures of their son. Richard Lovins (clinical research nurse) is acknowledged for his assistance with data collection. Dr Victoria Bordon (HSCT coordinator, Ghent University Hospital) and Dr David Creytens (Department of Pathology, Ghent University Hospital) are acknowledged for contributing clinical data and histological images on patient 9.

Authors' roles: Molecular studies: AP, CSchlack, EC, NLI, and LS. In silico studies: AS. Clinical evaluation of patients: AF, PJO, KAK, JR, CA, DA, OMV, BDM, AV, LDN, GS, J-SK, UK, and AS. Data collection and drafting manuscript: AP, PV, AV, and CSobacchi. All the authors critically reviewed the manuscript and approved the final version.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Note added in proof
  8. Disclosures
  9. Acknowledgements
  10. References
  11. Supporting Information
  • 1
    Sobacchi C, Frattini A, Orchard P, Porras O, Tezcan I, Andolina M, Babul-Hirji R, Baric I, Canham N, Chitayat D, Dupuis-Girod S, Ellis I, Etzioni A, Fasth A, Fisher A, Gerritsen B, Gulino V, Horwitz E, Klamroth V, Lanino E, Mirolo M, Musio A, Matthijs G, Nonomaya S, Notarangelo LD, Ochs HD, Superti Furga A, Valiaho J, van Hove JL, Vihinen M, Vujic D, Vezzoni P, Villa A. The mutational spectrum of human malignant autosomal recessive osteopetrosis. Hum Mol Genet. 2001;10:176773.
  • 2
    Bliznetz EA, Tverskaya SM, Zinchenko RA, Abrukova AV, Savaskina EN, Nikulin MV, Kirillov AG, Ginter EK, Polyakov AV. Genetic analysis of autosomal recessive osteopetrosis in Chuvashiya: the unique splice site mutation in TCIRG1 gene spread by the founder effect. Eur J Hum Genet. 2009;17:66472.
  • 3
    Fasth A. Osteopetrosis—more than only a disease of the bone. Am J Hematol. 2009;84:46970.
  • 4
    Aker M, Rouvinski A, Hashavia S, Ta-Shma A, Shaag A, Zenvirt S, Israel S, Weintraub M, Taraboulos A, Bar-Shavit Z, Elpeleg O. An SNX10 mutation causes malignant osteopetrosis of infancy. J Med Genet. 2012;49:2216.
  • 5
    Kornak U, Kasper D, Bösl MR, Kaiser E, Schweizer M, Schulz A, Friedrich W, Delling G, Jentsch TJ. Loss of the ClC-7 chloride channel leads to osteopetrosis in mice and man. Cell. 2001;104:20515.
  • 6
    Frattini A, Pangrazio A, Susani L, Sobacchi C, Mirolo M, Abinun M, Andolina M, Flanagan A, Horwitz EM, Mihci E, Notarangelo LD, Ramenghi U, Teti A, Van Hove J, Vujic D, Young T, Albertini A, Orchard PJ, Vezzoni P, Villa A. Chloride channel ClCN7 mutations are responsible for severe recessive, dominant, and intermediate osteopetrosis. J Bone Miner Res. 2003;18:17407.
  • 7
    Pangrazio A, Poliani PL, Megarbane A, Lefranc G, Lanino E, Di Rocco M, Rucci F, Lucchini F, Ravanini M, Facchetti F, Abinun M, Vezzoni P, Villa A, Frattini A. Mutations in OSTM1 (grey lethal) define a particularly severe form of autosomal recessive osteopetrosis with neural involvement. J Bone Miner Res. 2006;21:1098105.
  • 8
    Sobacchi C, Frattini A, Guerrini MM, Abinun M, Pangrazio A, Susani L, Bredius R, Mancini G, Cant A, Bishop N, Grabowski P, Del Fattore A, Messina C, Errigo G, Coxon FP, Scott DI, Teti A, Rogers MJ, Vezzoni P, Villa A, Helfrich MH. Osteoclast-poor human osteopetrosis due to mutations in the gene encoding RANKL. Nat Genet. 2007;9:9602.
  • 9
    Van Wesenbeeck L, Odgren PR, Coxon FP, Frattini A, Moens P, Perdu B, MacKay CA, Van Hul E, Timmermans JP, Vanhoenacker F, Jacobs R, Peruzzi B, Teti A, Helfrich MH, Rogers MJ, Villa A, Van Hul W. Involvement of PLEKHM1 in osteoclastic vesicular transport and osteopetrosis in incisors absent rats and humans. J Clin Invest. 2007;117:91930.
  • 10
    Guerrini MM, Sobacchi C, Cassani B, Abinun M, Kilic SS, Pangrazio A, Moratto D, Mazzolari E, Clayton-Smith J, Orchard P, Coxon FP, Helfrich MH, Crockett JC, Mellis D, Vellodi A, Tezcan I, Notarangelo LD, Rogers MJ, Vezzoni P, Villa A, Frattini A. Human osteoclast-poor osteopetrosis with hypogammaglobulinemia due to TNFRSF11A (RANK) mutations. Am J Hum Genet. 2008;83:6476.
  • 11
    Zhou CZ, Li de La Sierra-Gallay I, Quevillon-Cheruel S, Collinet B, Minard P, Blondeau K, Henckes G, Aufrère R, Leulliot N, Graille M, Sorel I, Savarin P, de la Torre F, Poupon A, Janin J, van Tilbeurgh H. Crystal structure of the yeast Phox homology (PX) domain protein Grd19p complexed to phosphatidylinositol-3-phosphate. J Biol Chem. 2003;278:503716.
  • 12
    Teasdale RD, Collins BM. Insights into the PX (phox-homology) domain and SNX (sorting nexin) protein families: structures, functions and roles in disease. Biochem J. 2012;441:3959.
  • 13
    Qin B, He M, Chen X, Pei D. Sorting nexin 10 induces giant vacuoles in mammalian cells. J Biol Chem. 2006;281:368916.
  • 14
    Mazzolari E, Forino C, Razza A, Porta F, Villa A, Notarangelo LD. A single-center experience in 20 patients with infantile malignant osteopetrosis. Am J Hematol. 2009;84:4739.
  • 15
    Villa A, Guerrini MM, Cassani B, Pangrazio A, Sobacchi C. Infantile malignant, autosomal recessive osteopetrosis: the rich and the poor. Calcif Tissue Int. 2009;84:112.
  • 16
    Zhu CH, Morse LR, Battaglino RA. SNX10 is required for osteoclast formation and resorption activity. J Cell Biochem. 2012;113:160815.
  • 17
    Worby CA, Dixon JE. Sorting out the cellular functions of sorting nexins. Nat Rev Mol Cell Biol. 2002;3:91931.
  • 18
    Humphreys K, Grankvist A, Leu M, Hall P, Liu J, Ripatti S, Rehnström K, Groop L, Klareskog L, Ding B, Grönberg H, Xu J, Pedersen NL, Lichtenstein P, Mattingsdal M, Andreassen OA, O'Dushlaine C, Purcell SM, Sklar P, Sullivan PF, Hultman CM, Palmgren J, Magnusson PK. The genetic structure of the Swedish population. PLoS One. 2011;6:e22547.
  • 19
    Chen Y, Wu B, Xu L, Li H, Xia J, Yin W, Li Z, Shi D, Li S, Lin S, Shu X, Pei D. A SNX10/V-ATPase pathway regulates ciliogenesis in vitro and in vivo. Cell Res. 2012;22:33345.
  • 20
    Hecht J, Seitz V, Urban M, Wagner F, Robinson PN, Stiege A, Dieterich C, Kornak U, Wilkening U, Brieske N, Zwingman C, Kidess A, Stricker S, Mundlos S. Detection of novel skeletogenesis target genes by comprehensive analysis of a Runx2(−/−) mouse model. Gene Expr Patterns. 2007;7:10212.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Note added in proof
  8. Disclosures
  9. Acknowledgements
  10. References
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
jbmr_1849_sm_SupplInfo.pdf327KSupplementary Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.