You are here

Article Text

Article menu

Download PDFPDF

Letters to JMG
Spondylo-epi-metaphyseal dysplasia (SEMD) matrilin 3 type: homozygote matrilin 3 mutation in a novel form of SEMD
Free
  1. Z U Borochowitz1,*,
  2. D Scheffer2,*,
  3. V Adir1,
  4. N Dagoneau2,
  5. A Munnich2,
  6. V Cormier-Daire2
  1. 1The Simon Winter Institute for Human Genetics, Bnai-Zion Medical Center, Rappaport - Faculty of Medicine, Technion - Israel Institute of Technology, Haifa, Israel
  2. 2Department of Medical Genetics and INSERM U393, Hôpital Necker Enfants Malades, Paris, France
  1. Correspondence to:
 Zvi U Borochowitz, MD
 The Simon Winter Institute for Human Genetics, Bnai-Zion Medical Center, PO Box 4940, Haifa, 31048, Israel; mdzvitechnion.ac.il

http://dx.doi.org/10.1136/jmg.2003.013342

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Osteochondrodysplasia is a heterogeneous group of conditions caused by impaired development of osseous skeleton. Within this group, spondylo-epi-metaphyseal dysplasia (SEMD) represents a subgroup which includes a number of conditions associated with vertebral, epiphyseal, and metaphyseal anomalies. The International Classification recognises at least 15 distinct entities within this group as defined by a combination of clinical, radiological, and molecular data.1 Mutations in the genes encoding structural proteins of the cartilage extracellular matrix (that is, cartilage oligomeric matrix protein, type II collagen, perlecan) or involved in posttranslational processing and transport (lysosomal storage disorders, sulfation protein (PAPSS2), regulator of chromatin (SMARCAL1), transcription initiation factor kinase (EIFKA3)) have been previously shown to be responsible for different forms of SEMD.1–6 However, a large number of SEMD conditions are still defined on purely descriptive features. Here, we describe five patients from a large consanguineous family of Arabic-Muslim origin with short-limb dwarfism with distinct skeletal dysplasia which, to our knowledge, has not been described previously. The manifestations and a constellation of radiographic bony abnormalities are unique in this entity. As there are abnormalities in the vertebral bodies, long bones, and specifically the lower limbs involving epiphyses and metaphyses, we suggest that this entity be called spondylo-epi-metaphyseal dysplasia, bowed-legs type. This hitherto unreported form of autosomal recessive SEMD is caused by homozygous mutation in matrilin 3 (MATN3), a component of the extracellular matrix of cartilage. This finding not only provides a molecular explanation for a novel form of SEMD but also expands the phenotype range of MATN3 defects which have been previously identified in autosomal dominant epiphyseal dysplasia and idiopathic hand osteoarthritis.7,8

    METHODS

    In this family, five affected individuals were found to have disproportionate dwarfism with similar clinical and radiological features. Patient 1 (figs 1 and 2) was born to healthy parents of Arabic-Muslim origin who are first cousins and of normal stature (father 175 cm (75th centile) and mother 168 cm (50th centile)), after a term pregnancy and normal delivery (birth weight 3100 g, birth length 45 cm (-2 SD), head circumference 33 cm). At 1 year, she had short-limbed short stature (height 68 cm (-3 SD), weight 8 kg (3rd centile), head circumference 43 cm (-3 SD), arm span 71 cm, palm length 5.2 cm (50th centile), middle finger length 3.2 cm (3rd centile)). Neurological examination, psychomotor development, hearing, and visual acuity were otherwise normal. Routine chemistry, thyroid function tests, and renal ultrasounds were normal as well. Re-evaluation at 6 years showed further disproportion of limbs, with bowed legs, waddling gait, wide joints (especially the knees), with limited elbow extension, pectus excavatum, and lumbar lordosis. At age 10 years, she underwent bilateral osteotomy of both proximal tibiae due to severe bowing. Examination at age 12 years showed the same changes (height at −4 SD), severe waddling gait, and very pronounced bowing of lower limbs.

    Figure 1
    Figure 1

    Pedigree of the family and genotype analysis of eight markers within the linked region on chromosome 2.

    Figure 2
    Figure 2

    Patients 1 and 2: total AP and lateral view of the body. Note the short trunk and severe bowing of legs.

    Key points

    • We investigated a large consanguineous Arabic-Muslim family in which five affected individuals were diagnosed with a novel form of autosomal recessive spondylo-epi-metaphyseal dysplasia (SEMD). All five affected individuals presented with disproportionate early-onset dwarfism, bowing of the lower limbs, lumbar lordosis, and normal hands. Skeletal findings included short, wide, and stocky long bones with severe epiphyseal and metaphyseal changes, hypoplastic iliac bones, and flat-ovoid vertebral bodies.

    • Genome wide homozygosity mapping mapped the disease locus gene to chromosome 2p25–24 with a maximum LOD score of 3.46 at locus D2S305. Two recombination events defined the minimal critical region interval between loci D2S320 and D2S215 (2.24 Mb). Using a candidate gene approach, we identified, in all affected individuals, a homozygous substitution (973T>A) in the gene encoding for Matrilin 3 (MATN3), a cartilage specific protein. The mutations changed a cysteine into a serine in the EGF-domain of MATN3, in a region highly conserved across species.

    • We conclude that homozygous MATN3 mutations are responsible for this novel form of SEMD.

    Patient 2, the brother of the propositus (figs 1 and 2), had short limbs at birth (length 45 cm (-2 SD), head circumference 34 cm). At 4 months, he had bowed lower legs, and mild lumbar lordosis (length 55 cm, −3 SD). On follow-up, he was noted to have short limbs, severe short stature, small chest, lumbar lordosis, and waddling gait. No hyperlaxity of joints was noted.

    Patient 3 is the second brother of the propositus (fig 1). At 33 weeks of gestation, ultrasound had detected short limbs. He was delivered normally at 39 weeks and examination at birth showed short limbs (length 44 cm (-2 SD), head circumference 33 cm). Recent examination at 4 months showed short limbs and bowed thighs. No chest narrowing or joint hyperlaxity was noted.

    Patient 4, the brother of the father of the propositus (fig 1), was born to parents who were first cousins. He was first examined at 23 years of age (head circumference 52.5 cm (-2.5 SD), height 136 cm (-5 SD), arm span 147 cm, palm length 11 cm (90th centile), middle finger length 7.6 cm (50th centile)). He had a waddling gait since he started walking (at 1 year), short limbs with severe short stature, bowing of lower limbs, and severe lumbar lordosis. At 13 years of age, he underwent bilateral osteotomy of both proximal tibiae for severe bowing. On follow-up at age 32, he is married with four children, all with regular height.

    Patient 5 is the half-cousin of the propositus (fig 1). He was born to parents who were first cousins after a 41 week pregnancy (birth weight 3000 g). Waddling gait was noted when he began walking at approximately 1 year of age. At age 11.5 years, he had short-limbed short stature (height of 117 cm (-4 SD), head circumference 51 cm (10th centile), arm span 122 cm, palm length 5.2 cm (50th centile), middle finger length 3.2 cm (3rd centile)). He had a flat face, normal shaped chest, short limbs, hyperlordosis of the lumbar area, wide joints, with limited extension of elbows, and bowed lower limbs. On follow-up at age 23 years (height 130 cm), the same clinical features were noted. He conducted a normal life and drove a car.

    x Ray radiographs of patient 3 at birth were similar to those at 3 months (fig 3A) and showed ribs of normal length, with posteriorly marked cupping. In the lateral projection, most of the vertebral bodies had an oval shape. In the AP projection, vertebral bodies were flat. The long tubular limb bones (except those of the hands and feet) were abnormally short with a stocky appearance and relatively wide metaphyses with lateral spurs. The epiphyses of the knees were irregularly shaped and small, and those of the proximal femurs were markedly underossified for age. The tubular bones of the hands were normal. The iliac wings were dysplastic and square with narrowing of the greater sciatic notch. The acetabular roof was flat, projecting medially from the roof with a bony spike, giving rise to a trident configuration of the iliac bone. The pubic bones were hypoplastic with small and underossified ischia. Epiphyseal and metaphyseal changes as well as spinal changes increased in severity with time (fig 3B,D).

    Figure 3
    Figure 3

    Skeletal findings in affected individuals. At 3 months (patient 3). The long tubular limb bones are abnormally short with a stocky appearance and relatively wide metaphyses with lateral spurs. The epiphyses of the knees are irregularly shaped and small, and those of the proximal femurs are markedly underossified for age. The tubular bones of the hands are normal. Ribs are of normal length, with posteriorly marked cupping, and vertebral bodies are oval on lateral projection and flattened on AP projection. The iliac wings are dysplastic and square with narrowing of the greater sciatic notch. The acetabular roof is flat, projecting medially from the roof with a bony spike, giving rise to a trident configuration of the iliac bone. The pubic bones are hypoplastic with small and underossified ischia.

    It should be emphasised that in the family reported here, none of the obligate heterozygote parents or the carrier healthy sib sister of the propositus (fig 1) ever complained of knee or hip pain or hand osteoarthritis or underwent surgery for early-onset osteoarthritis. A complete skeletal radiographic evaluation of the propositus’ parents as adults, and of the healthy sib sister as a teenager, did not reveal any skeletal changes or any epiphyseal dysplasia.

    Approval was obtained from the National Helsinki Committee for Genetic Research in Humans, and informed consent was given by family members or parents of the affected minors. Homozygosity mapping of the disease gene was performed using a panel of 400 markers at an average distance of 10 cM. Genomic DNA was extracted from leukocytes according to standard procedures. The MLINK program of the linkage software package was used to calculate two-point LOD scores between the disease phenotype and each of the markers, assuming a recessive disease with an allele frequency of 0.001.

    A series of 18 primers was used to amplify the eight coding exons of the MATN3 gene (including intron–exon boundaries) as previously described.7

    Amplified products were purified and sequenced using the fluorescent dideoxyterminator method on an ABI 3100 automatic sequencer.

    RESULTS

    Using a homozygosity mapping strategy, we found linkage of the disease gene on chromosome 2p25–24 (Zmax = 3.46 at 𝛉 = 0 at locus D2S305, table 1).

    View this table:
    Table 1

    Two-point LOD scores for linkage of an SEMD locus to chromosome 2p24

    A recombination event between loci D2S320 and D2S387 in patients 3 and 5 defined the proximal boundary of the genetic interval and another recombination event between loci D2S305 and D2S2150 in patients 2 and 4 defined the distal boundary of the candidate region (2.4 Mb, fig 1).

    A search of the human genome resources of the National Center for Biotechnology Information (NCBI) disclosed several putative disease genes within the candidate region including lysosomal-associated protein transmembrane 4 (LAPTM4A), matrilin 3 (MATN3), and Syndecan 1 (SDC1). Of these genes, MATN3 was considered the strongest candidate gene owing to its restrictive expression in cartilage.9,10 Moreover, dominant mutations in the region encoding the von Willebrand domain (vWFA) of MATN3 have been recently associated with multiple epiphyseal dysplasia.7

    Sequence analysis of MATN3 (GenBank accession number NM_002381) in the affected individuals revealed a homozygous substitution (973 T>A), which changed a cysteine into a serine in the first EGF domain of MATN3 (C304S).

    The mutation cosegregated with the disease (fig 4) and was not found by direct sequencing in 100 chromosomes from the same ethnic origin.

    Figure 4
    Figure 4

    Skeletal findings in affected individuals. At 1 year (patient 2). Epiphyses are irregularly shaped and small in all long bones, especially at the proximal femurs. Metaphyses are wide with same changes as at 3 months. Vertebral bodies continue to have the flattened-oval shape seen at the earlier age as well as posteriorly marked cupping.

    Figure 5
    Figure 5

    and DSkeletal findings in affected individuals. (C) At 12 years (patient 1). Note the severe epi-metaphyseal changes of long bones and the severe small proximal epiphyses of femur. Iliac bones are hypoplastic with steep acetabular roofs, and with no trident configuration as seen at an earlier age. (D) At 25 years (patient 5). Note the severe epi-metaphyseal changes seen in all long bones.

    Figure 6
    Figure 6

    Sequence analysis of MATN3 in the family. The homozygous mutant (973T>A), heterozygous and normal sequences are shown.

    DISCUSSION

    We report here homozygous mutations (C304S) in the region encoding for the first EGF like-domain of matrilin-3 in an apparently new autosomal recessive form of SEMD.

    Matrilin-3 belongs to a family of four oligomeric extracellular proteins and has a restricted expression pattern to cartilage.9–12 Although the tissue specific expression patterns of matrilin-1 and -3 are similar, their levels of expression are different during embryonic development and matrilin-3 appears to be more highly expressed than matrilin-1 during early development of the epiphyseal growth plate.11 The functions of matrilins have not been completely defined but they may play a role in stabilising the extracellular matrix structure. They can self-associate into supramolecular structures resulting in the formation of filamentous networks which can be associated with collagen fibrils. Matrilin-1 has been shown to bind to collagen type 2 in cartilage and also to interact with agregan.13 All matrilins share a similar structure including a Von Willebrand factor A domain, a varying number of epidermal growth factor domains, and a C-terminal coiled-coil domain.9,10,14 While matrilin-1 only contains one EGF domain, matrilin-3 lacks one A domain and contains four EGF-like domains. The matrilins form homo- and hetero-multimers mediated though the C-terminal coiled-coil domain and matrilin-3 has been shown to form a heteromultimer only with matrilin-1.10,14 The vWFA domain is a collagen binding domain and engineered mutations within the metal ion-dependent adhesion site motif of the vWFA domain of matrilin-1 reduces its ability to associate with the ECM network.13 The mutations that we have identified in the SEMD family reported here changed a cysteine residue located in the first EGF domain. Each EGF-like domain contains six cysteines and all of them (including the C304) are highly conserved across species. Although the role of these cysteine residues in matrilin-3 has not yet been demonstrated, they are presumably involved in intermolecular disulfide bonds as previously shown for cysteine residues in EGF-like domains in other human diseases such as Marfan syndrome or CADASIL.15,16

    This finding also expands the phenotypic spectrum of matrilin-3 mutations. Indeed, dominant mutations in the region encoding the von Willebrand domain (vWFA) of matrilin-3 have been identified in families with multiple epiphyseal dysplasia (MED).7,17. These mutations were missense mutations (V194D, R121W, A128P) located within the vWFA domain of the protein. More recently, a dominant mutation in the region encoding the first EGF domain has been found to cosegregate with hand osteoarthritis (OA) in several families.8 The mutation was a missense mutation changing a threonine to a methionine. In the family reported here, none of the heterozygote parents or the healthy sibs of the affected individuals ever complained of knee or hip pain or hand osteoarthritis or underwent surgery for early-onset osteoarthritis nor did radiographic evaluation reveal any degree of epiphyseal dysplasia. It seems therefore that a heterozygote state does not have any clinical or radiographic effects in this disorder. Although mutations in MATN3 causing MED or this novel form of SEMD are all missense, their consequences are certainly distinct as they are clearly not located in the same domains (that is, the von Willebrand/EGF domain). By contrast, the mutation identified in hand OA is also located in the first EGF domain, but presumably it does not have the same effect on the function of protein based on the amino-acid change (that is, threonine change in hand OA/cysteine change in SEMD).

    Similar situations, where heterozygotes for a recessive mutations do not exhibit the dominant phenotype, have been observed.18–21 Indeed, heterozygous ROR2 mutations cause brachydactyly type B, whereas homozygous mutations cause Robinow syndrome and heterozygous parents of those with Robinow syndrome do not have brachydactyly.18,19 Similarly, heterozygous IHH mutations cause brachydactyly type A1 (BDA1), whereas homozygous mutations cause acrocapital femoral dysplasia (ACFD) with no manifestations in the heterozygote parents.20,21 The mutations in IHH causing either BDA1 or ACFD are both missense mutations, but their accurate locations within the amino terminal domain differ. One possible explanation is that dominant mutations may exert a dominant negative effect whereas homozygous mutations would cause a loss of function.

    We conclude that MATN3 is another component of the intracellular matrix involved in the SEMD group. A better understanding of the interactions of MATN3 with the other extracellular matrix proteins, particularly COMP, type II and IX collagens, will certainly shed light on the pathogenesis of these chondrodysplasias.

    REFERENCES

    1. ↵
      Hall C. International Nosology and Classification of Constitutional Disorders of Bone (2001). Am J Med Genet2002;113:65–77.
      OpenUrlCrossRefPubMedWeb of Science
    2. Tiller GE, Polumbo PA, Weis MA, Bogaert R, Lachman RS, Cohn DH, Rimoin DL, Eyre DR. Dominant mutations in the type II collagen gene, COL2A1, produce spondyloepimetaphyseal dysplasia, Strudwick type. Nat Genet1995;11:87–9.
      OpenUrlCrossRefPubMedWeb of Science
    3. Nicole S, Davoine CS, Topaloglu H, Cattolico L, Barral D, Beighton P, Hamida CB, Hammouda H, Cruaud C, White PS, Samson D, Urtizberea JA, Lehman-Horn F, Weissenbach J, Hentati F, Fontaine B. Perlecan, the major proteoglycan of basement membranes, is altered in patients with Schwartz-Jampel syndrome (chondrodystrophic myotonia). Nat Genet2000;26:480–3.
      OpenUrlCrossRefPubMedWeb of Science
    4. Delepine M, Nicolino M, Barrett T, Golamaully M, Lathrop GM, Julier C. EIF2AK3, encoding translation initiation factor 2-alpha kinase 3 is mutated in patients with Wolcott-Rallison syndrome. Nat Genet2000;25:406–9.
      OpenUrlCrossRefPubMedWeb of Science
    5. Xu ZH, Otterness DM, Freimuth RR, Carlini EJ, Wood TC, Mitchell S, Moon E, Kim UJ, Xu JP, Siciliano MJ, Weinshilboum RM. Human 3′ phosphoadenosine 5′-phosphosulfate synthetase 1 (PAPSS1) and PAPSS2: gene cloning, characterization and chromosomal localization. Biochem Biophys Res Commun2000;268:437–44.
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      Boerkoel CF, Takashima H, John J, Yan J, Stankiewicz P, Rosenbarker L, Andre JL, Bogdanovic R, Burguet A, Cockfield S, Cordeiro I, Frund S, Illies F, Joseph M, Kaitila I, Lama G, Loirat C, McLeod DR, Milford DV, Petty EM, Rodrigo F, Saraiva JM, Schmidt B, Smith GC, Spranger J, Stein A, Thiele H, Tizard J, Weksberg R, Lupski JR, Stockton DW. Mutant chromatin remodeling protein SMARCAL1 causes Schimke immuno-osseous dysplasia. Nat Genet2002;30:215–20.
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      Chapman KL, Mortier GR, Chapman K, Loughlin J, Grant ME, Briggs MD. Mutations in the region encoding the von Willebrand factor A domains of matrilin-3 are associated with multiple epiphyseal dysplasia. Nat Genet2001;28:393–6.
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      Stefansson SE, Johnson H, Ingvarsson T, Manolesxu I, Johnson HH, Olafsdottir G, Palsdottir E, Stefansdotir G, Sveinbjornsdottir G, Frigge ML, Kong A, Gulcher JR, Stefabsson K. Genomewide scan for osteoarthritis: a novel mutation in Matrilin-3. Am J Hum Genet2003;72:1448–59.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      Zhang Y, Chen Q. Changes of Matrilin forms during endochondral ossification. J Biol Chem2000;42:32628–34.
    10. ↵
      Belluoccio D, Schenker T, Baici A, Trueb B. Characterization of human Matrilin-3 (MATN3). Genomics1998;53:391–4.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      Deak F, Wagener R, Kiss I, Paulsson M. The matrilins: a novel family of four oligomeric extracellular matrix proteins. Matrix Biol1999;18:55–64.
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      Segat D, Frie C, Nitsche PC, Klatt AR, Piecha D, Korpos E, Deak F, Wagener R, Paulsson M, Smyth N. Expression of matrilin-1, -2 and -3 in developing mouse limbs and heart. Matrix Biol2000;19:649–55.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      Chen Q, Zhang Y, Johnson DM, Goetinck PF. Assembly of a novel cartilage matrix protein filamentous network: molecular basis of differential requirement of von Willebrand factor A domains. Mol Cell Biol1999;10:2149–62.
      OpenUrl
    14. ↵
      Franck S, Schulthess T, Landwehr R, Lustig A, Mini T, Jeno P, Engel J, Kammerer RA. Characterization of the Matrilin coiled-coil domains reveals seven novel isoforms. J Biol Chem2002;21:19071–9.
    15. ↵
      Schrijver I, Liu W, Brenn T, Furthmayr H, Francke U. Cysteine substitutions in epidermal growth factor-like domain of fibrillin-1: distinct effects on biochemical and clinical phenotypes. Am J Hum Genet1999;65:1007–20.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      Dichgans M, Ludwig H, Miuller-Hocker J, Messerschmidt A, Gasser T. Small in-frame deletions and missense mutations in CADASIL: 3D models predict misfolding of Notch3v EGF-like repeat domains. Eur J Hum Genet2000;4:280–5.
    17. ↵
      Mostert AK, Dijkstra PF, Jansen BRH, Van Horn JR, de Graaf B, Heutink P, Lindhpout D. Familial multiple epiphyseal dysplasia due to a matrilin-3 mutation: further delineation of the phenotype including 40 years follow up. Am J Med Genet2003;120A:490–7.
    18. ↵
      Schwabe GC, Tinschert S, Buschow C, Meinecke P, Woolff G, Gillessen-Kaesbach G, Oldridge M, Wilkie AOM, Komec R, Mundlos S. Distinct mutations in the receptor tyrosine kinase gene ROR2 cause brachydactyly type B. Am J Hum Genet2000;67:822–31.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      Afzal R, Rajab A, Fenske CD, Oldridge M, Elanko N, Ternes-Pereira E, Tuyuz B, Murday VA, Patton MA, Wilkie AO, Jeffery S. Recessive Robinow syndrome, allelic to dominant brachydactyly type B, is caused by mutations of ROR 2. Nat Genet2000;25:419–22.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      Gao B, Guo J, She C, Shu A, Yang M, Tan Z, Yang X, Guo S, Feng G, He L. Mutations in the IHH, encoding Indian hedgehog, cause brachydactyly type A1. Nat Genet2001;28:386–8.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      Hellemans J, Coucke PJ, Giedion A, De Paepe A, Kramer P, Beemer F, Mortier GR. Homozygous mutations in IHH cause acrocapitofemoral dysplasia an autosomal recessive disorder with cone-shaped epiphyses in hands and hips. Am J Hum Genet2003;72:1040–6.
      OpenUrlCrossRefPubMed

    Footnotes

    • ↵* These authors equally contributed to this work.

    • Conflicts of interest: none declared.