Osteogenesis Imperfecta, Type V

A number sign (#) is used with this entry because of evidence that osteogenesis imperfecta type V (OI5) is caused by heterozygous mutation in the IFITM5 gene (614757), encoding interferon-induced transmembrane protein-5, on chromosome 11p15.

Description

Osteogenesis imperfecta is a connective tissue disorder characterized by bone fragility and low bone mass. Due to considerable phenotypic variability, Sillence et al. (1979) developed a classification of OI subtypes based on clinical features and disease severity: OI type I, with blue sclerae (166200); perinatal lethal OI type II, also known as congenital OI (166210); OI type III, a progressively deforming form with normal sclerae (259420); and OI type IV, with normal sclerae (166220). Most forms of OI are autosomal dominant with mutations in one of the 2 genes that code for type I collagen alpha chains, COL1A1 (120150) and COL1A2 (120160). Glorieux et al. (2000) described a novel autosomal dominant form of OI, which they designated OI type V, in 7 patients. The disorder was similar to OI type IV but had distinctive clinical, histologic, and molecular characteristics. OI type V is characterized by calcification of the forearm interosseous membrane, radial head dislocation, a subphyseal metaphyseal radiodense line, and hyperplastic callus formation (summary by Cho et al., 2012).

Clinical Features

Type V OI is moderately deforming and patients exhibit moderate to severe bone fragility of long bones and vertebral bodies. All type V patients (4 males and 3 females) described by Glorieux et al. (2000) had experienced fractures in the first year of life and had a history of frequent fractures (3.2 +/- 2.3 fractures/year). None of the patients had blue sclerae or dentinogenesis imperfecta. Radiographically, the patients were characterized by 3 distinctive features: hyperplastic callus formation at fracture sites, calcification of the interosseous membrane between the radius and ulna, and the presence of a radioopaque metaphyseal band adjacent to the growth plates. Hyperplastic callus presents as a hard painful swelling over affected bone and was present in 4 patients. All 7 patients had limitation of pronation/supination in one or both forearms, which was associated with a radiologically apparent calcification of the interosseous membrane. Three patients had anterior dislocation of the radial head. One patient had an additional calcified interosseous membrane in the left lower leg. A radiodense metaphyseal band immediately adjacent to the growth plates was a constant feature in growing patients. The band was most clearly visible in the metaphyses of the distal femora, proximal tibias, and the distal radii. Other radiologic findings included flattened, wedge-shaped, or biconcave vertebrae and wormian bones of the skull. Lumbar spine bone mineral density was low and similar to age-matched patients with OI type IV. Histology of iliac biopsy specimens revealed that lamellae were arranged in an irregular mesh-like pattern distinct from normal lamellar organization. Levels of biochemical bone markers were generally within the reference range, but serum alkaline phosphatase (171760) and urinary collagen type I N-telopeptide excretion (NTx) increased during periods of active hyperplastic callus formation.

Shapiro et al. (2013) studied 17 patients from 12 families with OI type V, all of whom were heterozygous for a c.-14C-T mutation in the IFITM5 gene (see MOLECULAR GENETICS). Typical features of OI type V in these patients included limited range of supination or pronation of the forearm in 15, radial head dislocation in 14, calcification of interosseous membranes in 13, and hyperplastic callus formation in 10; however, none of the patients exhibited a metaphyseal banding pattern. Other features included hyperextensible joints in 11 patients, triangular facies in 10, long bone bowing in 9, pes planus in 3, blue sclerae in 2, and mild unilateral mixed hearing loss in 1. Eleven affected individuals could ambulate without assistance. Ultrasound scans of the first child of an affected man revealed bowing of the femur at 22 weeks' gestation and a thin calvarium and angulated ribs suggesting intrauterine fracture at 33 weeks' gestation; at birth, the infant had 2 healing rib fractures, bowing of the femurs, and a thin calvarium but no other anomalies. Shapiro et al. (2013) noted that bone mineral density varied greatly among these patients, even within families. Three patients had chronically elevated levels of alkaline phosphatase, including 2 brothers in whom levels remained elevated even during bisphosphonate therapy.

Grover et al. (2013) reported a 5.5-year-old Hispanic girl with OI, originally classified as type III or severe type IV, who was found to carry the IFITM5 c.-14C-T mutation. She was small at birth, with weight and length at and below the 5th percentile, respectively. Forearm deformities and fussiness on handling prompted a skeletal survey, which showed multiple bilateral posterior rib fractures, bowing deformities of both forearms, and nondisplaced fractures of the left ulna and proximal right humeral shaft. Examination at 5.5 years revealed the characteristic triangular facies of OI with gray sclerae, midface hypoplasia, and hypotonia of the limbs. The patient had a history of multiple fractures of long bones after minimal or no trauma. Despite multiple fractures, there was no evidence of hyperplastic callus formation, interosseous membrane calcification, or radial head dislocation; in addition, she did not exhibit rhizomelia, limitation of forearm supination/pronation, vertebral compression fractures, or dentinogenesis imperfecta.

Balasubramanian et al. (2013) studied 4 patients from 3 families with OI type V and the c.-14C-T mutation. All had a similar facial gestalt with a broad forehead, a short up-turned nose, a small mouth with a thin upper lip, a prominent chin, and grayish blue sclerae.

Lazarus et al. (2014) described 9 patients from 8 families with OI type V and the c.-14C-T mutation. All displayed the characteristic radial head dislocation and calcification of the forearm interosseous membrane, and 5 developed hyperplastic callus. In contrast, bone fragility was quite variable, as shown by an 8-year-old patient who had experienced approximately 35 fractures and a 32-year-old patient who had never had a clinical fracture. Heterogeneity was also observed regarding height and physical activity level. None of the patients had blue sclerae or opalescent dentine.

Guillen-Navarro et al. (2014) reported a 5.75-year-old Spanish girl with OI diagnosed prenatally due to limb shortening, who was heterozygous for an S40L missense mutation in the IFITM5 gene (see MOLECULAR GENETICS). The authors stated that this was, to their knowledge, the first patient with an IFITM5 mutation who was prenatally diagnosed with bone shortening, a clinical feature characteristic of OI types II and III. The patient did not exhibit the typical signs of OI type V, although limited pronation/supination of the forearms was observed, in addition to characteristic radiodense lines on long bone x-rays. Despite the lack of hyperplastic callus formation, she had had elevated urinary excretion of the bone marker collagen type I N-telopeptide from the first months of life. Guillen-Navarro et al. (2014) also described a 30-year-old woman of Lithuanian origin with OI due to the c.-14C-T mutation who had a history of multiple fractures in childhood and short stature due to progressively severe skeletal deformities and loss of mobility. At age 13 years, her height was 160 cm (50th to 75th percentile), but by age 30 it had decreased to 125 cm (below the 3rd percentile). Radiologic examination revealed some of the typical clinical features of OI type V, including fracture-related hyperplastic callus formation and calcification of the interosseous membrane of the forearm; however, dislocation of the radial head and radiodense metaphysical banding were not detected.

Farber et al. (2014) studied a 25-year-old Caucasian woman with severe progressive OI without the typical features of OI type V who was heterozygous for the S40L missense mutation in the IFITM5 gene. Examination at age 25 revealed extreme short stature, with length of 87 cm (50th percentile for a 2-year-old girl) and weight of 33.3 kg (50th percentile for a 10-year-old girl), as well as relative macrocephaly, with a head circumference of 53.5 cm (25th percentile for an adult). She had a round face with a high bossed forehead and bluish sclerae. She also exhibited dentinogenesis imperfecta, as well as prominent and irregular ridges on the cutting surface of the teeth. She had a barrel chest with pectus excavatum, extreme bowing of all extremities, and S-curve scoliosis as well as prominent lordosis. She did not display radial head dislocation, hypertrophic callus formation, ossification of the interosseous membrane, or dense metaphyseal bands; rather, transiliac biopsy at 7 years of age revealed broad bands of unmineralized osteoid and a fish-scale pattern of lamellation, as seen in OI type VI (OI6; 613982).

Inheritance

In 3 patients with OI type V studied by Glorieux et al. (2000), the family history was positive for OI with documented father-to-son transmission in 2 families, consistent with an autosomal dominant pattern of inheritance.

Clinical Management

Zeitlin et al. (2006) described the results of 2 years of pamidronate treatment in 11 children and adolescents (5 boys, 6 girls) with OI type V (age at start of therapy, 1.8 to 15.0 years). Pamidronate was given in intravenous cycles at a cumulative yearly dose of 9 mg/kg. After 2 years, pamidronate treatment led to a decrease in the urinary excretion of N-terminal telopeptide of type I collagen to 50% of baseline levels. Both the size and volumetric bone mineral density of lumbar vertebrae increased compared to age- and sex-matched reference data (P less than 0.05 in both cases). Histomorphometry of transiliac bone samples in 7 patients showed an average increase of 86% in cortical thickness (p = 0.005). Fracture incidence decreased from 1.5 fractures per year before treatment to 0.5 fractures per year during the first 2 years of treatment. Ambulation status improved in 4 patients and remained unchanged in the others. Zeitlin et al. (2006) concluded that intravenous pamidronate therapy has a similar effect in OI type V as it has in other OI types.

Mapping

Cho et al. (2012) performed genomewide linkage analysis on a 4-generation family (family 1) with OI type V. Using 407 microsatellite markers with an average interval of 10 cM, they genotyped 14 family members (9 affected and 5 unaffected). A maximum lod score of 2.52 was obtained at marker D11S4046. Additional markers defined the locus at the 11pter-p15.4 region spanning 9.1 Mb from D11S4149.

Molecular Genetics

Cho et al. (2012) studied 19 Korean individuals with OI type V, including 13 affected individuals from 3 families and 6 simplex individuals. Cho et al. (2012) performed whole-exome sequencing in an affected simplex individual and 3 unaffected members of her family, and manually selected sequence variations (including those of the 5-prime and 3-prime UTRs and intron regions) unique to the proband. Among the variations located in the linked region of family 1, Cho et al. (2012) focused on a heterozygous change in the IFITM5 gene (c.-14C-T; 614757.0001). Sanger sequencing confirmed that this variation completely cosegregated with the disease in family 1. Furthermore, it was not found in 200 unrelated normal chromosomes from individuals with the same ethnic background. Cosegregation in the other 2 families (families 2 and 3) and de novo occurrence in the 5 other simplex individuals confirmed that this variation is a disease-causing mutation of OI type V.

Semler et al. (2012) independently performed whole-exome sequencing in a female with OI type V and her unaffected parents and identified a heterozygous de novo mutation in the 5-prime UTR of IFITM5 (c.-14C-T). They subsequently identified the identical heterozygous de novo mutation in a second individual with OI type V by Sanger sequencing.

Rauch et al. (2013) sequenced exon 1 of the IFITM5 gene in 42 patients with OI type V (ages, 2-67 years; 18 females) from 23 different families and identified the c.-14C-T mutation in all. Despite the presence of the same mutation, there was marked interindividual phenotypic variability. Indicators of disease severity varied widely: height z-scores in 38 patients ranged from -8.7 to -0.1, median -3.5; median final height was 147 cm in 15 men and 145 cm in 10 women; lumbar spine areal bone mineral density z-score in the absence of bisphosphonate treatment was between -7.7 and -0.7 in 29 men, median -5.3; scoliosis was present in 57% and vertebral compression fractures in 90% of all patients. Rauch et al. (2013) suggested that the IFITM5 mutation leads to a dysregulation of periosteal bone formation in addition to the bone formation deficit in trabecular bone.

In 17 affected individuals from 12 families with OI type V, 13 of whom were known to be negative for mutation in the COL1A1 (120150) and COL1A2 (120160) genes, Shapiro et al. (2013) identified heterozygosity for the c.-14C-T mutation in the IFITM5 gene. The authors noted strikingly variable phenotypic expressivity, both within and between affected OI type V families.

In a 5.5-year-old Hispanic girl with OI, originally classified as type III or severe type IV, who was negative for mutation in 10 known OI-associated genes, Grover et al. (2013) performed exome sequencing and identified heterozygosity for a de novo c.-14C-T mutation in IFITM5 that was not present in her unaffected parents. The patient did not exhibit any of the classic features of OI type V described in previous patients with this mutation, including hyperplastic callus formation, calcification of interosseous membranes, and radial head dislocation. Noting the marked phenotypic variability of OI type V, Grover et al. (2013) suggested that all unsolved OI patients should be screened for this recurrent IFITM5 mutation.

In 9 patients from 8 families with OI type V, Lazarus et al. (2014) sequenced the IFITM5 gene and identified heterozygosity for the c.-14C-T mutation in all. The mutation segregated with disease in the 3 families for which unaffected members were available.

In a 5.75-year-old Spanish girl with OI diagnosed prenatally on the basis of limb shortening, who was negative for mutation in 11 known OI-associated genes, Guillen-Navarro et al. (2014) identified heterozygosity for a de novo missense mutation in the IFITM5 gene (S40L; 614757.0002). Despite the lack of hyperplastic callus, the patient had had elevated urinary excretion of the bone marker collagen type I N-telopeptide from the first months of life; the authors recommended screening of IFITM5 in OI patients who lack the classic signs of OI type V but in whom urinary excretion of collagen type I N-telopeptide is above normal. Guillen-Navarro et al. (2014) also identified heterozygosity for the c.-14C-T mutation in IFITM5 in a 30-year-old woman of Lithuanian origin with OI. The patient had a history of multiple fractures in childhood, with progressively severe skeletal deformities; she exhibited fracture-related hyperplastic callus formation and calcification of the interosseous membrane of the forearm.

In a 25-year-old Caucasian woman who had extremely severe progressive OI without the typical features of OI type V and who was negative for mutation in 8 known OI-associated genes, including PEDF (SERPINF1; 172860), Farber et al. (2014) performed exome sequencing and identified a de novo missense mutation (S40L) in the IFITM5 gene. IFITM5 expression was normal in proband fibroblasts and osteoblasts, and BRIL protein level was similar to control. Farber et al. (2014) noted, however, that the patient's iliac biopsy had shown an OI6-associated fish-scale pattern of lamellation and that secretion of PEDF by patient fibroblasts was barely detectable; they thus hypothesized that the mutant gene was in a pathway with and/or interacted with PEDF. Analysis of patient osteoblasts confirmed minimal secretion of PEDF; in addition, COL1A1 (120150) expression and protein secretion was about one-fifth of control, and expression of alkaline phosphatase (171760) and osteocalcin (OC; 112260) were significantly reduced, whereas osteopontin (OPN; 166490) and bone sialoprotein (BSP; 147563) expression was 7- and 24-fold increased, respectively. Conversely, during a differentiation assay using osteoblasts from a patient with the c.-14C-T mutation and a typical OI type V phenotype, expression of PEDF increased 2-fold compared to control osteoblasts, with a consequent increase in secreted PEDF. Farber et al. (2014) concluded that BRIL and PEDF have a relationship that connects the genes for OI types V and VI and their roles in bone mineralization.

Exclusion Studies

Glorieux et al. (2000) screened the coding regions and exon/intron boundaries of both type I collagen genes (COL1A1 and COL1A2) in their patients and detected no mutations affecting glycine residues or splice sites.

Animal Model

Lietman et al. (2015) generated transgenic mice with the Ifitm5 variant corresponding to the recurrent human mutation, c.-14C-T, and observed perinatal lethality. Skeletal preparations and radiographs of embryonic day (E) 15.5 and E18.5 transgenic embryos revealed delayed/abnormal mineralization and skeletal defects, including abnormal rib cage formation, long bone deformities, and fractures. Primary osteoblast cultures derived from mutant calvaria at E18.5 showed decreased mineralization and reduced expression of osteoblast differentiation markers, such as osteocalcin, compared to wildtype. Overexpression of wildtype Ifitm5 did not, however, manifest a significant bone phenotype. Noting that previous studies had shown that knockout of Ifitm5 also did not result in significant bone abnormalities, Lietman et al. (2015) concluded that the c.-14C-T mutation acts in a neomorphic fashion.