Introduction

Osteosclerotic metaphyseal dysplasia (OSMD; MIM 615198) is a rare sclerosing bone disorder characterized clinically by developmental delay, hypotonia and late-onset spastic paraplegia.1, 2 The radiographic manifestations consist of osteosclerosis localized predominantly to the metaphyses of the long tubular bones, which also have osteopenic diaphyses.2, 3, 4, 5 The osteosclerotic changes also occur to a lesser degree in metaphyseal equivalents of other bones, including ends of ribs and clavicles, iliac crests, ischio-pubic synchondroses and vertebral end plates. The skull is reported to be spared. In some cases, aberrant results of clinical biochemistry test have been reported, including elevated levels of serum alkaline phosphatase, aspartate aminotransferase and creatine kinase.2, 3, 4, 5

OSMD is considered to be an autosomal recessive disorder.2 Our previous study identified a homozygous frame-shift mutation in leucine-rich repeat kinase 1 gene (LRRK1, MIM 610986) in one OSMD patient from consanguineous parents.6 The study also showed that close phenotypic similarity of Lrrk1 knockout mice with OSMD, suggesting that loss of LRRK1 function would result in OSMD.6, 7 As no LRRK1 mutations were found in other two OSMD patients examined in the study, OSMD was presumed to be a genetically heterogeneous condition.6 The idea is supported by the difference of phenotypes between the OSMD patients with and without LRRK1 mutations.

In this study, by conducting whole-exome sequencing in sib cases of OSMD, we discovered a novel homozygous LRRK1 mutation.

Materials and methods

Patients

We identified an Indian family with OSMD (Supplementary Figure S1). The non-consanguineous healthy couple had two affected children, a healthy daughter and three abortions.

Patient 1

The proband (II-5 in Supplementary Figure S1) was a 14-year-old boy who suffered from recurrent fractures by trivial trauma. He was born at term by normal vaginal delivery. He was noticed to have abnormal bones when he was investigated for vomiting at the age of 1 year (Figures 1a–d). He had eight fractures since 2 years of age. A radiograph at age 3 years showed marked resolution of the sclerotic lesions of hands and forearms (Figure 1e). He had valgus deformity of the knee, which was surgically repaired bilaterally by medial hemi-epiphysiodesis of the femur and tibia at age 14 years.

Figure 1
figure 1

Radiographs of the proband (II-5) in infancy. (a–d) The proband (II-5) at age 1 year. (a) Mild sclerosis of the skull base. (b) The right hand. Sclerosis of metaphyses in the distal radius and ulna, carpal bones and metacarpals with mild under-modeling. (c) Whole body. Sclerosis of spine, ribs, epiphyses and metaphyses of proximal humeri and pelvic margins. (d) Lateral spine. Rugger-jersey vertebral bodies and broad ribs. No platyspondyly. (e) The proband at age 3 years. Both forearm. The sclerotic lesions of hands and forearms and under-modeling of the long tubular bones markedly resolved.

Full size image

On examination at age 12 years, his height and weight were 147 cm (−1 SD) and 52 kg (0.7 SD), respectively. Occipito-frontal circumference was 53 cm (2 SD). He had mild facial dysmorphism, including large ears with upturned ear lobules, hypoplastic alae nasi, enamel hypoplasia and crowding of teeth (Figures 2a–c). He had mild joint laxity and bilateral pes planus. Fingers were swollen bilaterally with shortened finger tips of index fingers (Figure 2d). Hepatomegaly (3 cm below right costal margin) was noted. Complete blood count, peripheral smear and serum calcium, phosphorous and alkaline phosphatase were within normal limits.

Figure 2
figure 2

Appearances of the proband (II-5) at age 12 years. (a, b) Mild facial dysmorphism including large ear with upturned ear lobules and hypoplastic alae nasi. (c) Abnormal dentition. Persistent primary dentition, irregular teeth and enamel hypoplasia. (d) Swollen fingers and shortened finger tip of the index finger (arrow). A full color version of this figure is available at the Journal of Human Genetics journal online.

Full size image

Skeletal survey at age 12 years (Figure 3) showed osteosclerosis of spine (rugger-jersey spine), marginal sclerosis of pelvis, metaphyseal sclerosis of the long and short tubular bones with under-modeling. Acro-osteolysis was noted in both index fingers and the right middle finger. Variable severity of osteosclerosis of metacarpals, metatarsals and phalanges were observed. The lesions showed significant asymmetry, particularly at lower ends of femur and metatarsal bones.

Figure 3
figure 3

Radiographs of the proband (II-5) at age 12 years. (a) Skull. Mild sclerosis of the posterior part of the skull base (arrow). (b) Right hand. Acro-osteolysis of the index and middle fingers (arrows). Mild under-modeling of the distal radius. The metaphyses of the short tubular bones show only minimal sclerosis. (c) Left foot. Various sclerosis in phalanges and metatarsals. (d) Lateral spine. Rugger-jersey vertebral bodies and absent platyspondyly. (e) Both knees. Asymmetric metaphyseal sclerosis and under-modeling of the distal femora and proximal tibiae. Diaphyseal bone density is normal. (f) Marginal sclerosis of the sacrum, ilia, ischia and pubes and sclerosis of the proximal femora.

Full size image

Patient 2

The elder sister of the proband (II-2 in Supplementary Figure S1) also had recurrent fractures (the left olecranon at age 13 years, the proximal phalanx of the left little finger at 15 years, lateral condyle of the right tibia and head of fibula at 20 years and the left femur shaft at 25 years). She had been told by the surgeons that her bones could not be drilled with ease and the narrow medullary canal did not accommodate a regular intramedullary rod.

She was evaluated at 25 years of age. She was 158 cm tall (−0.6 SD), weighed 65 kg (0.5 SD) and had a head circumference of 54 cm (2 SD). Her cognition was normal. She did not have facial dysmorphism, joint laxity and swollen finger tips.

Radiographs showed variable degree of sclerosis of end plates of the vertebral bodies, mild marginal sclerosis of iliac wing, well remodeled proximal femora with dense cortex, and under-modeling of radius, ulna, metacarpals and phalanges (Figure 4). Her skeletal lesions were far milder than those in the proband.

Figure 4
figure 4

Radiographs of the elder sister (II-2) at age 25 years. (a) Lumbar spine. Only mild sclerosis of the vertebral end plates. (b) Pelvis. Mild sclerosis of iliac wings and proximal femora. (c) Left hand. Under-modeling of the distal radius and ulna, metacarpals and phalanges. Acro-osteolysis is absent.

Full size image

Whole-exome sequencing and variant calling

The study protocol was approved by the ethical committee of RIKEN and participating institutions. Peripheral bloods were obtained from the family members after the informed consent. Genomic DNAs were extracted from peripheral bloods using QIAamp DNA Blood Midi Kit (Qiagen, Hilden, Germany) according to the standard protocol. DNAs concentrations were measured by using a Qubit V.2.0 Fluorometer (Life Technologies, Carlsbad, CA, USA). Whole-exome sequencing was performed as previously described.8, 9 Briefly, DNA (3 μg) was sheared with a S2 ultrasonicator (Covaris, Wobum, MA, USA) and processed by SureSelectXT Human All Exon V5 (Agilent Technologies, Santa Clara, CA, USA). Captured DNA was sequenced using HiSeq 2000 (Illumina, San Diego, CA, USA) with 101 bp pair-end reads with seven indices. Image analysis and base calling were completed using HCS, RTA and CASAVA software (Illumina). Reads were mapped to the reference human genome (hg19) by Novoalign-3.02.04. Aligned reads were processed by Picard to remove PCR duplicates. Variants were called by GATK v2.7-4 following GATK Best Practice Workflow v310 and annotated by ANNOVAR.11

PCR and sanger sequencing

Sanger sequencing was performed to confirm the mutation identified by the whole-exome sequencing. A fragment including the mutation was amplified by PCR using primers, 5′-AGTGGTRCATCTCAGCACAC-3′ and 5′-CTTCCCTGGGGCTGTAAAAT-3′, and was sequenced in both strands. Sanger sequencing was performed on a 3730 DNA analyzer (Life Technologies). Sequencher V.4.7 (Gene Codes) and Genetyx (Genetyx Inc.) were used for aligning sequencing chromatographs to reference sequences.

Evaluation of the mutation identified in LRRK1

The mutation was evaluated by five databases, dbSNP (http://www.ncbi.nlm.nih.gov/projects/SNP/), 1000 genomes (http://www.1000genomes.org/), ExAC (http://exac.broadinstitute.org/), ESP6500 (http://evs.gs.washington.edu/EVS/) and Human Gene Mutation Database (https://portal.biobase-international.com/hgmd/pro/start.php). Homozygosity mapping was performed by using Homozygosity Mapper as previously described12 on whole-exome sequencing data.

Results and discussion

We carried out whole-exome sequencing in the proband, and harvested about 2.6 Gb sequences. The sequences were successfully mapped to all human RefSeq. At least 96% of all coding regions were covered in a depth of 10 reads (Supplementary Table S1). We identified a homozygous 1 bp insertion (c.5971_5972insG) in the last exon of LRRK1. The insertion was not deposited in any available databases, including dbSNP, 1000 genomes, Human Gene Mutation Database and ESP6500. Homozygosity mapping using the whole-exome sequencing data showed that LRRK1 is residing in a 1.8 Mb homozygous stretch of the patient genome (Supplementary Table S2). Mutations in known candidate genes of osteopetrosis were not identified in the dataset. By Sanger sequencing, we confirmed the homogenous mutation in both patients (Supplementary Figure S2).

LRRK1 was identified as an oncogene that promotes tumor formation in nude mice.13 The LRRK1 gene (NM_024652) consists of 34 exons, which encompass about 150 kb on chromosome 15q26.3. LRRK1 encodes 2 015 amino acids and has a multi-domain that contains ankyrin repeats, leucine-rich repeats, Ras of complex proteins, C-terminal of Roc and serine threonine kinase domain, as well as seven tryptophan-aspartic acid dipeptide (WD) 40 domains at the C-terminus (Supplementary Figure S3).14, 15, 16 The 1 bp insertion (c.5971_5972insG) is located after the WD40 domain in the last exon of LRRK1. The mutation is predicted to result in a frame-shift and produce an elongated protein (p.A1991Gfs*31) without nonsense-mediated mRNA decay. The position of the mutation is very similar to another LRRK1 elongation mutation previously identified by us in one patient with OSMD (Supplementary Figure S3).6

OSMD is a very rare disorder, and only seven patients in six families have been reported.2, 3, 4, 5, 6 We previously identified a homozygous mutation in LRRK1 in one OSMD patient from consanguineous parents.6 The mutation is a 7 bp deletion (c.5938_5944del7) within the last exon and is predicted to produce an elongated mutant protein (p.E1980Afs*66) (Supplementary Figure S3). Overexpression of the elongated mutant protein cannot rescue the bone resorption defect of Lrrk1-deficient osteoclasts, suggesting its hypomorphic effect. This is consistent with the phenotypic similarity of Lrrk1 knockout mice with the OSMD patient carrying the elongation mutation.6 Interestingly, the novel mutation (c.5971_5972insG) identified in two siblings in this study is also located in the last exon and is predicted to produce a similar elongated mutant protein (p.A1991Gfs*31).

The phenotype of OSMD with LRRK1 mutations has not been clarified because only one infant case complicated by Duchenne muscular dystrophy has been identified6. The current study could add two cases with different ranges of the age until after skeletal maturity. The three cases with LRRK1 mutations did not have intellectual disability and epilepsy that were previously reported in OSMD2 (Table 1). The first case with LRRK1 mutation reported by Iida et al.6 presented with failure to thrive, hypotonia and a delay in psychomotor development, but these findings were absent in our sibs. The features were probably due to the complication of Duchenne muscular dystrophy in the first case. In contrast, recurrent fractures characteristic to our sib cases were not found in the first case. This is probably because of age. The first case was under 2-year old and the fractures in the sibs started after infancy. Only the proband of the sib cases had facial dysmorphysm, which have not been described in other OSMD patients. Also, dental abnormalities were found only in the proband, whereas this abnormality is common in osteopetrosis. Radiographic features of the three cases shared similarities: lack of platyspondyly, osteosclerosis of spine, pelvic margins and metaphyses of the long and short tubular bones with under-modeling (Table 1); however, the severity and distribution of osteosclerosis are variable. The comparison of the three cases of different ages indicated that the sclerotic change and under-modeling resolve with advancing age, which may make clinical diagnosis difficult. Further studies are necessary to clarify the phenotype of OSMD with LRRK1 mutation.

Table 1 Clinical and radiographic features of patients with osteosclerotic metaphyseal dysplasia and LRRK1 mutation
Full size table