American Journal of Medical Genetics 137A:241–248 (2005)

Schmid Type of Metaphyseal Chondrodysplasia andCOL10A1 Mutations—Findings in 10 PatientsOuti Makitie,1,2* Miki Susic,1 Leanne Ward,3,4 Catherine Barclay,1 Francis H. Glorieux,4 and William G. Cole1

1Division of Genetics and Genomic Biology, Research Institute, Hospital for Sick Children, Toronto, Ontario, Canada2Hospital for Children and Adolescents, Helsinki University Hospital, Helsinki, Finland3Department of Pediatrics, University of Ottawa, Ottawa, Ontario, Canada4Genetics Unit, Shriners Hospital for Children, McGill University, Montreal, Quebec, Canada

The Schmid type of metaphyseal chondrody-plasia (MCDS) is characterized by short stature,widened growth plates, and bowing of the longbones. It results from autosomal dominant muta-tions of COL10A1, the gene which encodes a1(X)chains of type X collagen. We report the clinicaland radiographic findings in 10 patients withMCDS and COL10A1 mutations. Six patients hadlower limb deformities, which necessitated ortho-pedic surgeries in all of them. One patient demon-strated no deformities and normal stature atage 11 years (height �1.2 SDS) while the othersmanifested severe short stature (<�3.5 SDS).Radiographs showed metaphyseal changes whichwere most pronounced at the hips and knees. Fiveof the identified 10 mutations in COL10A1 werenovel. Six mutations resulted in truncation of theNC1 domain while four mutations were singleamino-acid substitutions. Our findings suggestthat COL10A1 mutations result in a uniformpattern of growth plate abnormalities. However,the clinical variability in severity among affectedindividuals is greater than previously thought.� 2005 Wiley-Liss, Inc.

KEY WORDS: chondrodysplasia; growth plate;hypertrophic chondrocytes; col-lagen X; dominant mutation; coxavara

INTRODUCTION

The Schmid type of metaphyseal chondrodyplasia (MCDS;MIM #156500) is characterized by short stature and bowing ofthe long bones; radiographic features include widening andirregularity of the growth plates, especially in the distal and

proximal femora [Schmid, 1949; Lachman et al., 1988]. MCDSresults from autosomal dominant mutations in COL10A1(MIM #120110), the gene encoding a1(X) chains of type Xcollagen molecules [Kirsch and von der Mark, 1991; Warmanet al., 1993;McIntosh et al., 1994;Wallis et al., 1994;McIntoshet al., 1995]. Type X collagen is a short-chain, non-fibrillarcollagen which consists of three identical a1(X) chains [Kirschand von der Mark, 1991]. It is synthesized specifically byhypertrophic chondrocytes at sites of endochondral ossifica-tion [Kirsch and von der Mark, 1991]. The homotrimercontains a central short helical domain flanked by a smallnon-collagenous amino-terminal domain (NC2) and a large,highly conserved, carboxyl-terminal non-collagenous domain(NC1) [LuValle et al., 1988; Thomas et al., 1991]. The signalsequence is removed in the rough endoplasmic reticulum butthe NC1 and NC2 domains are retained in mature type Xcollagen molecules extracellularly and participate in interac-tions between molecules [Kirsch and von der Mark, 1991;Kwan et al., 1991]. The formation of the type X collagenmolecule is initiated at the NC1 domain, which is essential forcorrect association and alignment of the a chains before triplehelix formation [Engel and Prockop, 1991; Brass et al., 1992].

Apart from two exceptions, all mutations in MCDS havebeen identified in the NC1 domain [Warman et al., 1993;Dharmavaram et al., 1994; McIntosh et al., 1994; Wallis et al.,1994, 1996; Bonaventure et al., 1995; Chan et al., 1995, 1998;McIntosh et al., 1995; Pokharel et al., 1995;Matsui et al., 1996;Stratakis et al., 1996; Ikegawa et al., 1997; Savarirayan et al.,2000; Bateman et al., 2004]. About half of the mutationswould be expected to result in truncation of the NC1 domainand the other half in amino acid substitutions. In vitro and celltransfection studies have shown that a1(X) chains bearingsome of the reported NC1 domain mutations, in contrast tothe normal chains, do not form stable trimeric moleculesand are degraded intracellularly, resulting in a reduction inthe amount of functional collagen X within the growth platecartilage extracellular matrix [Chan et al., 1995, 1998; Wilsonet al., 2002; Bateman et al., 2004].

In the present study we describe genetic findings, includingfive novel mutations, and the resulting phenotypes in 10patients with MCDS and COL10A1 mutations. Six of thesemutations resulted in truncation of theNC1 domainwhile fourmutations were single amino-acid substitutions; the resultingskeletal phenotypeswere radiographically very similar but thedegree of severity varied among the reported patients.

PATIENTS AND METHODS

Patients

As part of an ongoing research program of molecularmechanisms in skeletal dysplasias, which was approved bythe Research Ethics Board of The Hospital for Sick Children,Toronto, Canada, DNA samples of 31 patients with a diagnosisof MCDS were studied for COL10A1 mutations. The patients

This article contains supplementary material, which may beviewed at the American Journal of Medical Genetics websiteat http://www.interscience.wiley.com/jpages/1552-4825/suppmat/index.html.

Grant sponsor: Foundation for Paediatric Research; Grantsponsor: Paivikki and Sakari Sohlberg Foundation, Helsinki,Finland; Grant sponsor: Novo Nordisk A/S (OM); Grant sponsor:Shriners of North America; Grant sponsor: Canadian Institutes ofHealth Research.

*Correspondence to: Outi Makitie, M.D., Ph.D., Hospital forChildren and Adolescents, Metabolic Bone Clinic, University ofHelsinki, Helsinki, Finland. E-mail: outi.makitie@helsinki.fi

Received 31 January 2005; Accepted 11 April 2005

DOI 10.1002/ajmg.a.30855

� 2005 Wiley-Liss, Inc.

were ascertained thorough clinical genetic, orthopaedic, andpediatric services. The diagnosis ofMCDSwas based clinicallyon short-limbed short stature and absence of evident extra-skeletal manifestations and radiographically, on compatiblemetaphyseal changes on skeletal radiographs [Lachman et al.,1988; Warman et al., 1993]. Causative mutations in COL10A1were identified in 11 patients. One of these patients haspreviously been described [Baric et al., 2000]; the remaining10 patients were included in the present report.

Clinical and Radiographic Characteristics

Hospital records and radiographs were available for 7 of the10 patients and they were reviewed for clinical and radio-graphic characteristics of the skeletal dysplasia; no clinical orradiographic data other than the known diagnosis of MCDSwere available for the remaining 3 patients. The hospitalrecords were reviewed for data on growth, symptoms, clinicalfindings, and surgeries. The growth data was compared withpublished reference data for healthy children [Kuczmarskiet al., 2000]. A total of 109 skeletal radiographs for 7 patients(median 13 per patient) were available for review. Long-itudinal data, with two or more skeletal examinations over aminimum of 5 years, were available for four patients; crosssectional data only was available for three patients. Theradiographic surveys were most complete for the lowerlimbs since the clinical problems mainly affected the hipsand knees. All radiographs were first evaluated cross-sectionally to characterize typical features in different agegroups. Longitudinal radiographs were then used to seewhether similar sequence of changes with age occurred inindividual patients.

Characterization of Mutations

Blood samples were collected for DNA extraction andmutational analysis. Mutational screening of the COL10A1gene was performed by direct sequencing of five overlappingfragments covering exons 1–3 and their exon/intron bound-aries. In PCR reactions, genomic DNA was amplified in avolume of 20 ml using 50 ng of each primer and 1.0 U of Taqpolymerase (Qiagen) in 1� PCR buffer (67 mM Tris-HCl,pH 8.8, 6.7 mM MgCl2, 98 mM b-mercaptoethanol, 16.6 mM(NH4)2SO4, 6.7 mM EDTA, 328 mg/ml BSA, 1.5 mM of eachdNTP, and10%DMSO.ThePCRconditionswere denaturationat 948C for 20 sec, annealing at an appropriate temperaturefor 30 sec, extension at 728C for 40 sec for 40 cycles, andan additional 8 min at 728C. The reaction products wereresolved and extracted from the 1% agarose gel (EM Science).PCR fragments were directly sequenced using ThermoSequenase Cy 5 and Cy 5.5 Dye Terminator Kit (Amersham

Biosciences) on OpenGene System (Visible Genetics) auto-mated sequencer.

RESULTS

Clinical Features

The study included 10 patients (5males) ranging in age from4 to 40 years (Table I). Patients presented with progressivegrowth failure, which had its onset postnatally, usually duringthe second year of life. It usually resulted in significant shortstature (height z-score <�3.5), but one patient (MCDS# 51)demonstrated normal height at the age of 11 years (z-score�1.2). Patients usually came to medical attention after age2 years with short stature, waddling gait and varus or valgusdeformity of the knees. Lower-limb deformities progressedwith age in all the patients andnecessitated surgical correctionof coxa vara in six patients, genua vara in four patients, andof genua valga in two patients. None of the patients hadsignificant clinical complaints apart from the skeletal mani-festations. Three representative patients (MCDS #2, #9, #10)with a typical clinical course of MCDS and one patient (MCDS#51) with unusually mild presentation are described here indetail.

MCDS #2, a female, appeared normal at birth. Her growthvelocity was normal during the first year of life but subse-quently decelerated and remained below the 3rd centile afterthe age of 2 years. On physical examination at age 2.7 years,she demonstrated disproportionate short stature with rhizo-melic shortening of the limbs and long trunk, bow-leggedappearance, increased lumbar lordosis, and limited hip move-ments. The major joints were prominent. Lower limb defor-mities progressed with age and necessitated bilateralintertrochanteric valgus osteotomies of the hips with bilateralpercutaneous adductor longus release at the age of 6.7 years.Her height at 10 years was 115 cm (�3.6 SDS). At the age of 14,she complained of severe joint and leg pain after walkingshort distances; she had mild bilateral varus deformity of theknees, significant bilateral coxa varawithwell formed andwellcovered femoral heads.

MCDS #9, a male, was born of clinically healthy parents.He appeared normal at birth. Hewalked at 1 year of age with awaddling gait and with increasing bowing of his legs. Hisgrowth within the first year was normal but slowed fromabout age 18 months. Linear growth remained below the thirdcentile and his adult height was 137 cm (�5.6 SDS). The varusdeformities of the femoral necks and tibiae were surgicallycorrected at age 5, 12, and 16 years of age.

MCDS #10, a female, was born of clinically normal parents.She developed the typical clinical and radiographic features

TABLE I. Clinical Features and COL10A1 Mutations in the 10 Patients With Schmid Type ofMetaphyseal Dysplasia

# Sex AgeHeight(SDS) Deformities Surgeries Mutation

COL10A1Aa length

2 F 18 �3.6 Genua vara OT 1952G>A 6505 M 20 NA NA NA 1845T>A 6149 M 17 �5.6 Genua vara OT�2 1844delA 62010 F 20 �3.8 Genua vara OT�2 2001T>G 66618 M 13 �5.7 Genua valga OT�2 1989C>G 66225 F 13 �3.6 Genua valga OT, EPD 1783G>A 68035 M 4 NA NA NA 1771T>C 68037 F 6 NA NA NA 1851T>A 68039 M 14 NA Genua vara OT�2 1860delT 62051 F 11 �1.2 No No 1942G>A 680

Aa, aminoacid; NA, data not available; OT, osteotomy; EPD, epiphyseodesis.

242 Makitie et al.

of MCDS, which were of similar severity to those observed inpatients #2 and #9. Her height was normal up to about18 months of age but beyond 2 years age it remained below thethird centile. Her adult height of 139 cm (�3.8 SDS) wasachieved at 17 years of age. Bilateral proximal femoral andhigh tibial osteotomies were undertaken at 12 years of ageand again at 17 years of age.

MCDS #51, a female, was born to healthy parents. She wasnoticed to have a limp at age 4 years. At age 8 years she wasreferred to the orthopaedic clinic for evaluation of bilateral hippain and abnormal gait. She was found to have restricted hipmovements and bilateral coxa vara on radiographs. She hashad no surgeries but complained of persistent hip symptoms.Her height at age 11 years was 135 cm (�1.2 SDS) andpredicted adult height 153 cm (�1.6 SDS).

Radiographic Features

The characteristics anddistribution of radiographic findingswere very similar in all the seven patients for whom radio-graphs were available for review. However, the degree ofseverity varied between patients but in individual patients

the degree of severity tended to be uniform at all affected sites.The changes were more marked in the weight-bearing joints,especially in the hips, and less severe in other sites.

Lower limbs. In every patient, the proximal femur wasthemost severely affected site. All the patients had significant,progressive coxa vara deformity (Figs. 1 and 2), often with analmost horizontal femoral neck, sagging of the femoral headin the femoral neck and evidence of mechanical failure. Themetaphyseal changes which resulted in a short, irregular, andwide femoral neck were apparent in early infancy (as early asat 4 months of age in one patient) and tended to progress withage. The proximal femoral epiphysis was always normal inshape but it was enlarged relative to the child’s age and to thewidth of the adjacent metaphysis in early childhood.

The growth plates of the distal femora and proximal tibiaeand fibulae were irregular, thickened, and flared (Fig. 3),though the changes were less marked than those seen at theproximal femora. Widening of the proximal tibia was usuallymore pronounced medially than laterally. There was varusor valgus deformity of the long bones, resulting fromasymmetric growth of both the tibia and the femur, in allpatients exceptMCDS #51, who also presented withmuch less

Fig. 1. Radiographs of MCDS #51 show progressive coxa vara with metaphyseal irregularity ((a) 8.5 years, (b) 11 years). Metaphyseal flaring andthickening is less remarkable at the distal femora and proximal tibiae ((c) 8.5 years) and the distal radius ((d) 11 years). The spine is normal ((e) 11 years).

Schmid Metaphyseal Chondrodysplasia 243

marked changes of the growth plates (Fig. 1). The secondaryossification centers were often very large (both in width and inthickness) and asymmetric but normal in contour (Fig. 3).Metaphyseal changes at the distal tibiae and fibulae weresimilar to those seen at the knee with flaring, irregularity, andthickening of the growth plates (Fig. 4). Radiographs of the feetdid not show any growth plate anomalies or other deformities.

Upper limbs. The growth plates of the proximal humeri,distal radii, and distal ulnae were irregular and abnormallythick; usually onlymildwideningwas seen.Onlyminimal ornochanges in the growth plates of the distal humerus, proximalradius, metacarpals, and phalanges were seen; the elbow wasnormal in all patients. The metaphyseal changes were mostmarked at the distal radius and ulna (Fig. 4) but tended tobecome less severe with age. In every patient the growth plateabnormalities were less marked at the upper limb than atthe lower limb (Figs. 1 and 4). The age of appearance and thesubsequent development as well as the shape and contour ofthe secondary ossification centers were normal. Cone epi-physes were seen in none of the patients.

Spine. Radiographs of the spine were available for fivepatients.All showednormal shapeand size of the vertebrae butone had mild scoliosis and two, increased lumbar lordosis.

COL10A1 Mutations

All 10 patients were found to have a mutation in COL10A1.Eight of the mutations consisted of a single nucleotide sub-stitution and two patients had a deletion of a nucleotide(Table I). These sequence changes were predicted to result inpremature termination of the reading frame in six patients,with resulting truncated COL10A1 polypetides of 614–666amino acids. In four patients the mutation resulted in single

amino-acid substitution (Table I). Five of thesemutations havebeen previously reported in other cohorts of MCDS and fiveof the mutations were novel (see the online Suppl. 1 at http://www.interscience.wiley.com/jpages/1552-4825/suppmat/index.html).

In each patient, the remaining regions of exon 3 as well asexons 1 and 2 and their exon/intron boundaries did not showany anomalies on direct sequencing.

DISCUSSION

In this study, we report the clinical and radiographicfindings in a large cohort of patientswithMCDSand confirmedmutations in COL10A1. The clinical features of the patientswere similar to those reported previously for MCDS, char-acterized by disproportionate short stature with short limbsand bowed legs [Schmid, 1949; Lachman et al., 1988]. Most ofthe patients presented during the second year of life withbowed legs and with a postnatal, progressive growth failure.Lower-limb deformity, usually varus deformity of the hips andknees, was a significant feature of MCDS and often requiredsurgical intervention.

The main radiographic features in patients with MCDS andCOL10A1 mutation included coxa vara, an enlarged capitalfemoral epiphysis in early childhood, generalizedmetaphysealflaringand irregularity, andnormal spine.Thesefindingswerein accordancewith those previously described for a cohort of 20patients with a clinical diagnosis of MCDS [Lachman et al.,1988]. However, radiographic analysis in the present studysuggests that themost characteristic radiographic findings arefound in the proximal femoral metaphysis which showedmetaphyseal irregularity, coxa vara, and vertical growth platein all the patients. This contrasts with the findings in the

Fig. 2. The radiographic changes at the proximal femur are characterized by significant coxa vara with almost horizontal femoral neck, sagging of thefemoral head in the femoral neck and evidence of mechanical failure (arrows). The metaphyseal changes result in irregular, short, and wide femoral neck.Theproximal femoral epiphysis isnormal in shapebut enlarged relative to the child’s ageand to thewidth of the adjacentmetaphysis.a:MCDS#25at5years,b: MCDS #9 at 9 years, c: MCDS #18 at 6 years, d: MCDS #39 at 7 years.

244 Makitie et al.

previous study [Lachman et al., 1988], which reported onepatient with normal and three patients with very mildlyabnormal proximal femoral metaphyses. Further, in thepresent cohort the proximal femur was invariably the mostseverely affected site, while Lachman et al. [1988] reported thedistal femoral metaphyses to be more severely affected thanthe proximal femoral metaphyses. The diagnosis of MCDS intheir study was based on clinical and radiographic data onlyand it is possible that the cohort included some patients withmetaphyseal chondodysplasia caused by genetic defects otherthan COL10A1 mutation.

Metaphyseal changes were observed also in the upper limbsand in the ribs but these were less severe than in the lowerlimbs. Differences in the rates of growth may influence thesusceptibility of each growth plate to the disrupting effects oftype X collagenmutations. For example, the proximal humeraland distal radial and ulnar growth plates, which normallyaccount for 80% of the longitudinal growth of the arm, werethickened and had irregular margins. In contrast, the othergrowth plates of the arm bones, which normally grow at amuch slower rate, were of normal or near normal thicknesses.Radiographs of the legs also showed marked changes in thegrowth plates of the fast growing long bones but apparentlynormal growth plates in the slower growing bones of the feet.

Overall, the clinical features and the radiographic metaphy-seal changes, their distribution, and severity showed very littlevariability among individual patients except MCDS #51 whohad a significantly milder presentation, characterized by

normal height, late onset of symtoms, and lack of lower limbdeformity. She had short femoral necks and coxa vara and thegrowth plates showed evidence of mechanical failure andstress fracture. Otherwise her radiographs showed very littlemetaphyseal involvement. We have previously studied pedia-tric patients with isolated coxa vara for COL10A1 mutationsbut identified no mutations (WG Cole, unpublished observa-tion). The findings in MCDS #51 however suggest that theclinical variability in severity of COL10A1 associated skeletaldysplasia is greater than previously thought. More impor-tantly, our findings confirm that type X collagen plays a keyrole in the femoral neck development andmay be an importantdeterminant of its length, width, and neck-shaft angle. It isrecognized that individual variability in these characteristicsexists and that they may have important implications forsusceptibility to hip fracture [Carpintero et al., 2003; Gnudiet al., 2004; Pulkkinen et al., 2004]. Whether the previouslyidentifiedCOL10A1 polymorphisms (see the online Suppl. 2 athttp://www.interscience.wiley.com/jpages/1552-4825/suppmat/index.html) playa role in the femoralneckgeometry remains tobe elucidated in future studies.

All reported MCDS mutations in COL10A1, except two,are localized to the C-terminal NC1 domain of the protein(summarized in Suppl. 1). The triple helical domain contains anumber of imperfections in the Gly-X-Y triple helical sequencesuggesting that point mutations may be well tolerated in thetriple helical domain. The exon structure is also quite differentfrom the long chain helical type I, II, and III collagens, so

Fig. 3. The growth plates at the distal femora and proximal tibiae and fibulae demonstrate flaring, thickening, and irregularity. The changes wereusually assymmetrical and resulted in genua vara ((a) MCDS #39 at 1 year 10 months, (b) MCDS #2 at 2 years 9 months) or genua valga ((c) MCDS #25 at5 years, (d) MCDS #18 at 6 years) deformity.

Schmid Metaphyseal Chondrodysplasia 245

that the dominant-negative type of splicing mutationswould not be expected. The two mutations outside the NC1domain are in the signal peptide and have only been identifiedin Japanese cases [Ikegawa et al., 1997]; these are likely toresult in inhibition of the signal peptide cleavage and retentionof the mutant chains attached to the microsome membranes[Chan et al., 2001]. All of the other 36 identified mutationsexist in the NC1 domain and produce either terminationsignals (nonsense or framesift mutations) (53%) or missensemutations (47%) (Suppl. 1). Controversy exists concerning themolecular pathology of MCDS in these cases. While severalstudies suggest haploinsufficiency to be the main pathology atthe tissue level [Chan et al., 1998;Wilson et al., 2002; Batemanet al., 2003, 2004], a dominant negative effect of the mutanta1(X) chainonnormal trimerassemblyhasalsobeen suggestedas an alternative mechanism [Chan et al., 1999; Bogin et al.,2002]. The very mild presentation in one of the patients in thepresent study (MCDS #51) with a single amino acid change inthe a1(X) chain (1942G>A; D648N) is more likely to be theresult of a dominant negative effect than of haploinsufficiency.Interestingly, anothermutation in the same codon (1943A>G;D648G, Suppl. 1) has been described but no details otherthan the diagnosis of MCDS were given for the patient[Bonaventure et al., 1995].

Previous studies have reported patients with clinicaldiagnosis of MCDS with no identifiable mutations in theCOL10A1 gene [Bonaventure et al., 1995; Wallis et al., 1996;

Bateman et al., 2004]. This was seen in the present cohort aswell: of the 31 patients with a clinical diagnosis of MCDS only11 were found to have COL10A1 mutations. The clinical andradiographic pattern in these patients was uniform, whereasthe patients without an identified mutation comprised amore heterogenous group both clinically and radiographically.Three of these patients however, had clinical and radiographicfindings identical to those seen in patients with COL10A1mutations, providing further evidence for locus heterogeneityin MCDS [Bateman et al., 2004].

Metaphyseal chondrodysplasias are a heterogeneous groupof disorders characterized by flaring and irregularity ofvarious metaphyses. The most common forms after MCDSare Cartilage-hair hypoplasia (CHH, McKusick type),Shwachman–Diamond syndrome (SDS) and Jansenmetaphy-seal chondrodysplasia. Responsible genes have recently beenidentified for each of these disorders [Schipani et al., 1996;Ridanpaa et al., 2001; Boocock et al., 2003]. The differentialdiagnosis between MCDS and other metaphyseal chondrodys-plasias may be challenging. The hallmarks of Shwachman–Diamond syndrome are, in addition to the specific skeletalfindings [Makitie et al., 2004], the associated extra-skeletalmanifestations (pancreatic insufficiency and bone marrowdysfunction) [Aggett et al., 1980; Ginzberg et al., 1999]. Jansentype of metaphyseal chondrodysplasia is characterized byextremedisorganization of themetaphyses and by severe shortstature and is associated with biochemical abnormalities

Fig. 4. The characteristic metaphyseal changes were present at the distal radius and ulnae ((a) MCDS #2 at 2 years 9 months, (b) MCDS #18 at 6 years,(c)MCDS#10at 9 years) andat the distal tibiae andfibulae (d)MCDS#39at 1 year 10months, (e)MCDS#2at 2 years 9months, and (f)MCDS#25at 5 years)but they were less remarkable than at the hips and knees.

246 Makitie et al.

[Schipani et al., 1996]. Differentiating MCDS from CHH maybeparticularly challenging.While themajority of patientswithCHHhave extra-skeletal manifestations (sparse hair, compro-mised cell-mediated and/or humoral immunity, hematologicalabnormalities, Hirschsprung disease) [Makitie and Kaitila,1993], a subgroup of patients with CHH only present with askeletal phenotype [Bonafe et al., 2002]. In fact 2 of our 20patients with a clinical diagnosis of MCDS and no COL10A1mutation were found to be homozygous for a base substitutionG forAat nucleotide 70 of theRMRP gene that encodes anRNAsubunit of the MRP RNAse complex; neither of these patientshad extra-skeletal manifestations suggestive of CHH [Ridan-paa et al., 2003].

ACKNOWLEDGMENTS

This study was supported by the Foundation for PaediatricResearch and the Paivikki and Sakari Sohlberg Foundation,Helsinki, Finland, and by ESPE Research Fellowship, spon-sored by Novo Nordisk A/S (OM); Shriners of North America(FHG and LW); and by a grant from the Canadian Institutes ofHealth Research (WGC).

REFERENCES

Aggett PJ, CavanaghNPC,MatthewDJ, Pincott JR, Sutcliffe J, Harries JT.1980. Shwachman’s syndrome. A review of 21 cases. Arch Did Child55:331–347.

Baric I, Skrabic V, Begovic D, Sarnavka V, Superti-Furga A. 2000. A17-month-old boy with bowed legs. Eur J Pediatr 159:863–865.

Bateman JF, Freddi S, Nattrass G, Savarirayan R. 2003. Tissue-specificRNA surveillance? Nonsense-mediated mRNA decay causes collagen Xhaploinsufficiency in Schmidmetaphyseal chondrodysplasia.HumMoleGenet 12:217–225.

Bateman JF,Freddi S,McNeil R, ThompsonE,HermannsP, SavarirayanR,Lamande SR. 2004. Identification of four novel COL10A1 missensemutations in schmid metaphyseal chondrodysplasia: Further evidencethat collagen X NC1 mutations impair trimer assembly. Hum Mutat23:396.

Bogin O, Kvansakul M, Rom E, Singer J, Yayon A, Hohenester E. 2002.Insight into Schmid metaphyseal chondrodysplasia from the crystalstructure of the collagen X NC1 domain trimer. Structure 10:165–173.

Bonafe L, Schmitt K, EichG, Giedion A, Superti-Furga A. 2002.RMRP genesequence analysis confirmsa cartilage-hair hypoplasia variantwith onlyskeletal manifestations and reveals a high density of single-nucleotidepolymorphisms. Clin Genet 61:146–151.

Bonaventure J, Chaminade F, Maroteaux P. 1995. Mutations in threesubdomains of the carboxy-terminal region of collagen typeXaccount formost of the Schmid metaphyseal dysplasias. Hum Genet 96:58–64.

Boocock GRB, Morrison JA, Popovic M, Richards N, Ellis L, Durie PR,Rommens JM. 2003. Mutations in SBDS are associated withShwachman–Diamond syndrome. Nature Genet 33:97–101.

Brass A, Kadler KE, Thomas JT, Grant ME, Boot-Handford R. 1992.The fibrillar collagens, collagen VIII, collagen X and C1q complementproteins share the similar domain in their C-terminal non-collagenousregions. FEBS Lett 303:126–128.

Carpintero P, Leon F, ZafraM, Serranto-Trenas JA, RomanM. 2003. Stressfractures of the femoral neck and coxa vara. Arch Orthop Trauma Surg123:273–277.

Chan D, ColeWG, Rogers JG, Bateman JF. 1995. Type X collagen multimerassembly in vitro is prevented by a Gly618 to Val mutation in the alpha1(X) NC1 domain resulting in Schmid metaphyseal chondrodysplasia.J Biol Chem 270:4558–4562.

Chan D, Weng YM, Graham HK, Sillence DO, Bateman JF. 1998. Anonsense mutation in the carboxyl-terminal domain of type X collagencauses haploinsufficiency in schmid metaphyseal chondrodysplasia.J Clin Invest 101:1490–1499.

Chan D, Freddi S, Weng YM, Bateman JF. 1999. Interaction of collagenalpha1(X) containing engineered Nci mutations with normal alpha1(X)in vitro. Implications for the molecular basis of schmid metaphysealchondrodysplasia. J Biol Chem 274:13091–13097.

Chan D, Ho MSP, Cheah SE. 2001. Aberrant signal peptide cleavage ofcollagen X in Schmid metaphyseal chondrodysplasia. Implications forthe molecular basis of the disease. J Biol Chem 11:7992–7997.

Dharmavaram RM, Elberson MA, Peng M, Kirson LA, Kelley TE, JimenezSA. 1994. Identification of amutation in type X collagen in a family withSchmid metaphyseal chondrodysplasia. Hum Mol Genet 3:507–509.

Engel J, Prockop D. 1991. The zipper-like folding of collagen helices and theeffects of mutations that disrupt that zipper. Annu Rev Biophys Chem20:137–152.

Ginzberg H, Shin J, Ellis L, Morrison J, Ip W, Dror Y, Freedman M,Heitlinger LA, Belt MA, Corey M, Rommens JM, Durie PR. 1999.Shwachman syndrome: Phenotypic manifestations of sibling sets andisolated cases in a large patient cohort are similar. J Pediatr 135:81–88.

Gnudi S, Malavolta N, Testi D, Viceconti M. 2004. Differences in proximalfemur geometry distinguish vertebral from femoral neck fractures inosteoporotic women. Br J Radiol 77:219–223.

IkegawaS,NakamuraK,NaganoA,HagaN,NakamuraY. 1997.Mutationsin the N-terminal globular domain of the type X collagen gene(COL10A1) in patients with schmid metaphyseal chondrodysplasia.Hum Mutat 9:131–135.

Kirsch T, von der Mark K. 1991. Isolation of human type X collagen andimmunolocalization in fetal human cartilage. Eur J Biochem 196:575–580.

Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, Flegal KM, Guo SS,Wei R, Mei Z, Curtin LR, Roche AF, Johnson CL. 2000. CDC growthcharts: United States. Adv Data 314:1–27.

KwanAPL, Cummings CE, Chapman JA, GrantME. 1991.Macromolecularorganisation of chicken type X collagen in vitro. J Cell Biol 114:597–604.

LachmanRS,RimoinDL, Spranger J. 1988.Metaphyseal chondrodysplasia,Schmid type. Clinical and radiographic deliniation with a review ofliterature. Pediatr Radiol 18:93–102.

LuValleP,NinomiyaY,RosenblumND,OlsenBR. 1988. The typeX collagengene. Intron sequences split the 50-untranslated region and separate thecoding regions for thenon-collagenous amino-terminal and triple-helicaldomains. J Biol Chem 263:18378–18385.

Matsui Y, Kimura T, Tsumaki N, Yasui N, Ochi T. 1996. A recurrent1992delCT mutation of the type X collagen gene in a Japanese patientwith Schmidmetaphyseal chondrodysplasia. Jpn JHumGenet 41:339–342.

McIntosh I, Abbott MH, Warman ML, Olsen BR, Francomano CA. 1994.Additional mutations of type X collagen confirm COL10A1 as theSchmid metaphyseal chondrodysplasia locus. Hum Mol Genet 3:303–307.

McIntosh I, Abbott MH, Francomano CA. 1995. Concentration of mutationscausing Schmid metaphyseal chondrodysplasia in the C-terminalnoncollagenous domain of type X collagen. Hum Mutat 5:121–125.

MakitieO,Kaitila I. 1993.Cartilage-hair hypoplasia-clinicalmanifestationsin 108 Finnish patients. Eur J Pediatr 152:211–217.

Makitie O, Ellis L, Durie PR, Morrison JA, Sochett EB, Rommens JM, ColeWG. 2004. Skeletal phenotype in patients with Shwachman–Diamondsyndrome and mutations in SDBS. Clin Genet 65:101–112.

Pokharel RK, Alimsardjono H, Uno K, Fujii S, Shiba R, Matsuo M. 1995.A novelmutation substituting tryptophanwith arginine in the carboxyl-terminal, non-collagenous domain of collagen X in a case of Schmidmetaphyseal chondrodysplasia. Biochem Biophys Res Commun 217:1157–1162.

Pulkkinen P, Partanen J, Jalovaara P, Jamsa T. 2004. Combination of bonemineral density and upper femur geometry improves the prediction ofhip fracture. Osteoporosis Int 15:274–280.

Ridanpaa M, van Eenennaam H, Pelin K, Chadwick R, Johnson C, Yuan B,vanVenrooij W, Pruijn G, Salmela R, Rockas S, Makitie O, Kaitila I,de la Chapelle A. 2001.Mutations in the RNA component of RNaseMRPcause a pleiotropic human disease, cartilage-hair hypoplasia. Cell 104:195–203.

Ridanpaa M, Ward LM, Rockas S, Sarkioja M, Makela H, Susic M, GlorieuxFH, ColeWG,Makitie O. 2003. Genetic changes in the RNA componentsof RNase MRP and RNase P in Schmid metaphyseal chondrodysplasia.J Med Genet 40:741–746.

Savarirayan R, Cormier-Daire V, Lachman RS, Rimoin DL. 2000. Scmidtype metaphyseal chondrodysplasia: A spondylometaphyseal dysplasiaidentical to the ‘‘Japanese’’ type. Pediatr Radiol 30:460–463.

Schipani E, Langman CB, Parfitt AM, Jensen GS, Kikuchi S, Kooh SW,Cole WG, Juppner H. 1996. Constitutively activated receptors for

Schmid Metaphyseal Chondrodysplasia 247

parathyroid hormone and parathyroid hormone-related peptide inJansen’s metaphyseal chondrodysplasia. N Engl J Med 335:708–714.

Schmid F. 1949. Beitrag zur Dysostosis enchondralis metaphysaria.Monatsschr Kinderheilk 97:393.

Stratakis CA, Orban Z, Burns AL, Vottero A, Mitsiades CS, Marx SJ,Abbassi V, Chrousos GP. 1996. Dideoxyfingerprinting (ddF) analysis ofthe type X collagen gene (COL10A1) and identification of a novelmutation (S671P) in a kindred with schmid metaphyseal chondrodys-plasia. Biochem Mol Med 59:112–117.

ThomasJT,Cresswell CJ,RashB,NicolaiH, JonesT, SolomonE,GrantME,Boot-Handford RP. 1991. The human collagen X gene. Completeprimary translated sequence and chromosomal localization. Biochem J280:617–623.

Wallis GA, Rash B, Sweetman WA, Thomas JT, Super M, Evans G,Grant ME, Boot-Handford RP. 1994. Amino acid substitutions of

conserved residues in the carboxyl-terminal domain of the alpha 1(X)chain of type X collagen occur in two unrelated families with metaphy-seal chondrodysplasia type Schmid. Am J Hum Genet 54:169–178.

Wallis GA, Rash B, Sykes B, Bonaventure J, Maroteaux P, Zabel B, Wynne-Davies R, Grant ME, Boot-Handford RP. 1996. Mutations within thegene encoding the alpha 1(X) chain of type X collagen (COL10A1) causemetaphyseal chondrodysplasia type Schmid but not several other formsof metaphyseal chondrodysplasia. J Med Genet 33:450–457.

Warman ML, Abbott M, Apte SS, Hefferon T, McIntosh I, Cohn DH,Hecht JT, Olsen BR, Francomano CA. 1993. A type X collagen mutationcauses Schmid metaphyseal chondrodysplasia. Nat Genet 5:79–82.

Wilson R, Freddi S, Bateman JF. 2002. Collagen X chains harboring Schmidmetaphyseal chondrodysplasia NC1 domain mutations are selectivelyretained and degraded in stably transfected cells. J Biol Chem 277:12516–12524.

248 Makitie et al.