American Journal of Medical Genetics 116A:61–65 (2003)

Mesomelic and Rhizomelic Short Stature:The Phenotype of Combined Leri-Weill Dyschondrosteosisand Achondroplasia or Hypochondroplasia

Judith L. Ross,1,2* Gary Bellus,3 Charles I. Scott, Jr.,1,2 Jack Abboudi,1 Giedre Grigelioniene,4

and Andrew R. Zinn5

1Department of Pediatrics, Thomas Jefferson University, Philadelphia, Pennsylvania2Department of Dermatology, University of Colorado, Denver, Colorado3Department of Pediatrics, A.I. DuPont Hospital for Children, Wilmington, Delaware4Pediatric Endocrinology Unit, Karolinska Hospital, Stockholm, Sweden5McDermott Center for Human Growth and Development and Department of Internal Medicine,University of Texas Southwestern Medical School, Dallas, Texas

We studied two children with combinedgenetic skeletal disorders. Both had Leri-Weill dyschondrosteosis (LWD); one also hadachondroplasia and the other had hypochon-droplasia. Both had severe short stature andevidence of rhizomelia and mesomelia as wellas other phenotypic features of their indivi-dual genetic disorders. Achondroplasia wasdue to the G380R FGF3R mutation and hypo-chondroplasia to a N540K mutation in thesamegene.Thepatientwithhypochondropla-sia had a heterozygous SHOX deletion; noSHOX mutation was identified in the childwith achondroplasia. The phenotypes of com-bined LWD and achondroplasia or hypochon-droplasia appeared to be less than additive,suggesting that SHOX and FGFR3 act onoverlapping pathways of bone growth anddevelopment. � 2002 Wiley-Liss, Inc.

KEY WORDS: FGFR3; SHOX; double hete-rozygosity

INTRODUCTION

Leri-Weill dyschondrosteosis (LWD; MIM 127300),a dominantly inherited skeletal dysplasia, is character-ized phenotypically by Madelung wrist deformity,

mesomelia, and short stature [Leri and Weill, 1929].LWD can now be defined genetically by haploinsuffi-ciency of the SHOX gene located in the Xp-Yp pseu-doautosomal region [Belin et al., 1998; Shears et al.,1998; Ross et al., 2001]. A more severe skeletal dys-plasia, Langer-type mesomelic dysplasia (LMD, MIM249700), is thought to be the homozygous form of LWDresulting from complete loss of SHOX gene or proteinfunction. The LMD phenotype includes severe shortstature with predominant shortening of shanks andforearms [Langer, 1965; Espiritu et al., 1975a, 1975b;Kunze and Klemm, 1980].

Achondroplasia is an autosomal dominant disordercharacterized by rhizomelic short stature, large headwith frontal bossing, midface hypoplasia, lumbar lordo-sis, and trident hand. The genetic abnormality involvesmutations in thefibroblast growth factor receptor 3 gene(FGFR3) located at 4p16.3 [Bellus et al., 1995]. Hypo-chondroplasia may be caused by a different mutation atthe FGFR3 locus and is clinically similar but less severethan achondroplasia [Bellus et al., 2000; Grigelionieneet al., 2000].

We present four unusual offspring resulting from theunions of two siblingswith Langermesomelic dysplasia,each married to a spouse with either achondroplasia orhypochondroplasia. The four children all had Leri-Weilldyschondrosteosis; in addition, two also had achondro-plasia or hypochondroplasia. These patients present aremarkable combination of disorders causing rhizomelicand mesomelic short stature. The phenotypes of com-bined LWD plus hypochondroplasia or achondroplasiaappear to be less than additive, suggesting that SHOXand FGFR3 act partly in overlapping pathways of bonegrowth and development.

MATERIALS AND METHODS

Molecular Analysis of SHOX and FGFR3

SHOX deletions were detected by fluorescence in situhybridization (FISH) using cosmid probe LLNOYCO30-

Grant sponsor: National Institutes of Health; Grant number:NS35554.

*Correspondence to: Dr. Judith L. Ross, Department ofPediatrics, Thomas Jefferson University, 1025 Walnut Street,Philadelphia, PA 19107. E-mail: judith.ross@mail.tju.edu

Received 30 April 2002; Accepted 7 June 2002

DOI 10.1002/ajmg.a.10807

� 2002 Wiley-Liss, Inc.

M034F5 as previously described [Wei et al., 2001]. Atleast five metaphases were scored for each sample.SHOX mutations were detected by denaturing high-performance liquid chromatography (dHPLC) anddirect sequencing of genomic PCR products (SHOX-DNA-DxTM, Esoterix Endocrinology, Calabasas Hills,CA).

PCR of FGFR3 exons 10 and 13 was carried out aspreviously described [Bellus et al., 1995, 1996]. The exon10 G380R (G1138A) mutation was detected by digestionof the PCR products with the restriction enzyme Sfc-I(New England Biolabs), the mutation creating a novelrestriction site not present in wild type. The exon 13N540K mutations were detected by digestion of thePCR products with the restriction enzyme BspM-I(New England Biolabs). Both N540K mutations(C1620A and C1620G) destroy the single BspM-I sitein exon 13. The C1620G, but not the C1620A mutation,creates a unique Alu-I restriction site and its presencewas ruled out by failure of exon 13 PCR products todigest with Alu-I.

Clinical Assessment

The clinical assessment included measurement ofheight, lower segment, arm span, and forearm length.High arched palate, increased carrying angle of thearms, and scoliosis were evaluated as previously de-scribed [Zinn et al., 1998]. Upper arm length was mea-sured from the acromion to the olecranon. Lower armlength was measured from olecranon to radial pro-minence at wrist. Arm span was measured as the dis-tance from right to left third fingertip with patientsfacing the wall with outstretched arms held parallelto the ground. Results were converted to z-scoreswhere possible using age- and gender-specific norms[McKusick, 1972; Hamill et al., 1979; Hall et al., 1995].Comparisons with achondroplasia norms for head cir-cumference and upper and lower segment measure-ments were made from previously published results[Horton et al., 1978].

Metacarpal-phalangeal profiles measured from handX-rays were used to determine the presence of shortfourth or fifthmetacarpals [Poznanski et al., 1972]. Age-and gender-specific height z-scoreswere calculated fromNational Center for Health Statistics [Hamill et al.,1979]. Z-scores for right radial length and ratio of upperto lower segment were calculated from published norms[McKusick, 1972]. A positive z-score means the ratio(upper to lower segment) was greater than normal,indicating relatively short legs.

RESULTS

Molecular Analysis of SHOX and FGFR3

The demographics and clinical and molecular diag-noses of themembers of the two families are shown in thepedigree (Fig. 1), photographs (Fig. 2), and Table I. Twounrelated spouses, onewith achondroplasia (I-4) andonewith hypochondroplasia (I-1), of two siblings with LMD(I-2 and I-3) and their children (II-1–II-4) were studied.The two groups of children were first cousins. All were

Caucasian. The phenotypes of the parents with LMDwere previously described [Zinn et al., 2002]. Deletion ofone SHOX allele was demonstrated in both parents, butno molecular defect was identified in their other SHOXallele; however, both had classic LMD phenotypes [Zinnet al., 2002]. Patient I-1 and his daughter II-2 bothcarried an FGFR3 mutation, N540K, associated withhypochondroplasia and both had clinical features ofhypochondroplasia. Inaddition, thechildII-2wasdeletedfor SHOX and had clinical features of LWD. Her malesibling II-1 was deleted for SHOX and manifested LWDfeatures only. Patient I-4 and her daughter II-3 both hadan FGFR3 mutation associated with achondroplasia,G380R, and both had clinical features of achondroplasia.The child II-3 also had clinical features of LWD but noabnormalities of her SHOX gene were detected. Inaddition, her brother II-4 was deleted for SHOX andhad LWD clinical features only.

Clinical/Auxological Findings

All subjects were short; the LWD cases were tallerthan combined LWDþachondroplasia or hypochondro-plasia cases, who in turn were taller than the achon-droplasia or hypochondroplasia cases (Fig. 2; Tables II–IV). Males and females were similarly affected, accord-ing to diagnosis. All subjects had a high arched palateand normal nails and none had scoliosis. The carryingangle of II-3 and I-4 was increased. Tibial bowing wasmore pronounced in the children with LWD plusachondroplasia or hypochondroplasia compared to theirLWD only siblings. The radius, upper arm, entire arm,hand, and the lower limb was more shortened in thechildren with LWD plus achondroplasia or hypochon-droplasia compared to the children with only LWD ortheir parents with achondroplasia or hypochondropla-sia. The tibial length was less severely affected in thechildren with LWD plus achondroplasia or hypochon-droplasia compared to their parents with hypochondro-plasia or achondroplasia.

The hand X-rays of all affected subjects wereassessed for features ofMadelung deformity of the wrist(Table V). All four children with LWD had someradiographic features ofMadelung deformity, includingdistal radial lucency, carpal wedging, or radial bowing.No Madelung deformity features were noted in theadults with achondroplasia or hypochondroplasia.Patient II-3 had carpal wedging consistent with early

Fig. 1. Pedigree showing molecular diagnoses. Numbers beneath sym-bols indicate age (years) at the time of study.

62 Ross et al.

Madelung deformity. She did not have any radiallucencies.

The most prominent features associated with achon-droplasia include large head, frontal bossing, midfacehypoplasia, narrowed interpedicular distance in thelumbar column, lordosis/kyphosis, shortened iliac wing,rhizomelia, distal tibia vara, metaphysical flaring, andlong fibulae. These features were most notable in thesingle adult patient with achondroplasia and somewhatless notable in her affected (achondroplasiaþLWD)

daughter (Table V). Less severe and fewer featureswereobserved in the adult or child with hypochondroplasia.

Head circumference SD score was increased in theadults with only hypochondroplasia or achondroplasia,whoalsohad frontal bossing, butnot in the childrenwithhypochondroplasia or achondroplasia andLWD.Noneofthe childrenhad tridenthands.Brachydactylywasmorenotable in the children with both LWD and achondro-plasia or hypochondroplasia compared to their LWDsiblings (Table V).

TABLE I. Molecular Diagnostic Results*

I-1 I-4 II-1 II-2 II-3 II-4

Clinical diagnosis H A LWD LWDþH LWDþA LWDFGFR3 mutation

cDNA C1620A G1138A None C1620A G1138A NoneProtein N540K G380R None N540K G380R None

SHOX deletion No Not tested Yes (X) Yes No Yes (Y)

*H, hypochondroplasia; A, achondroplasia; LWD, Leri-Weill dyschondrosteosis.

Fig. 2. Overall appearance of patients.

SHOX and FGFR3 Double Heterozygosity 63

DISCUSSION

We described two children (cousins), one with bothLWD and achondroplasia and one with both LWD andhypochondroplasia. We compared their phenotypeswith those of their unrelated parents who had achon-droplasia or hypochondroplasia and those of theirsiblings who had LWD only. Confirmation of these diag-noses was obtained through molecular genetic techni-ques in three of the four children. One child had theFGFR3 mutation commonly associated with achondro-plasia, G380R, and one had an FGFR3 mutation com-monly associated with hypochondroplasia, N540K, asdid their respective parents. The phenotype associatedwith the N540K mutation is quite variable, rangingclinically from very mild to severe, as seen with theachondroplasia-associated G380R mutations.

A SHOX gene deletion was detected in one child andno apparent SHOX deletion or coding region mutationwas detected in the second child. That she had dyschon-drosteosis relies on the diagnosis of her father’s Langermesomelic dysplasia, which is the homozygous form ofLWD and is therefore transmitted to all his progeny.Also, she did have certainLWDfeatures, including earlyMadelung deformity wrist changes. SHOX defects canbe identified in most but not all LWD subjects. As therehas been no convincing report of locus heterogeneityin LWD, it is probable that some mutations affectingSHOX expression lie outside of the coding region andare not detected by present methodology. We favor thehypothesis that both LMD siblings I-1 and I-2 andconsequently proband II-3 carry suchaSHOXmutation.Althoughwe cannot formally exclude the involvement ofgenes other than SHOX in their phenotype, there is noevidence for digenic inheritance of LMD. This child maydevelop the classic LWD phenotype as she goes throughadolescence. Regardless, it is indisputable that her

cousin had the genetic defects associated with hypo-chondroplasia and LWD.We will assume for the sake ofdiscussion that an undetected SHOX defect was pre-sent in II-3. To our knowledge, this represents the firstcase study describing the combined effects of thesegenetic skeletal dysplasias in children. The combinationof hypochondroplasia and achondroplasia has been pre-viously described [McKusick et al., 1973].

The children with both achondroplasia or hypochon-droplasia and LWDwere short but no shorter than theiraffected parent with achondroplasia or hypochondro-plasia. This was unexpected in view of the predicted oc-currence of both rhizomelia and mesomelia. Perhapsthe children will become relatively shorter as they growinto adulthood. We have previously not noted an effectof age on height SD score in LWD [Ross et al., 2001],but there is an age effect in achondroplasia/hypochon-droplasia [Horton et al., 1978], with children becom-ing progressively shorter than the normal population[Brook and de Vries, 1998]. Not surprisingly, the targetbones of SHOX appear to be quite specific, mainly thebones of the forearm and lower leg with relative sparingof the head, hands, feet, etc. The children with achon-droplasia or hypochondroplasia plus LWD had somemild facial features of achondroplasia or hypochondro-plasia (large head, frontal bossing), and their hands andmetacarpals were relatively smaller than those of theirLWD-affected siblings. The effects of the FGFR3 muta-tion on the fibula are distinct and appear to override

TABLE III. Auxologic Measurement SD Scores Derived FromAchondroplasia Norms

I-1 I-4 II-2 II-3

Height SD 0.8 0.9 0.4 �0.8Upper segment SD 3.4 0 0.5 �2.1Lower segment SD �1.5 1.4 �0.4 0.2Head circumference SD 2.3 �0.4 �2.1 �1.2

TABLE IV. Clinical Features Related to Achondroplasia*

I-1 I-4 II-1 II-2 II-3 II-4

Frontal bossing þ þþ � þ � �Midface hypoplasia � þ � � � �Mandibular hypoplasia � � þ � þ �Cuboid vertebral body þ þþ � � þþ �Short interpeduncular distance � þþ � � þ �Lordosis/kyphosis þ þþ � � þ �Short iliac wing þ þþ � þ þ �Tibia vara � þ � � � �Metaphyseal flaring � þþ � þ þþ �Long fibulae þ þþ � � þþ �Tibial bowing � � � þþ þ �Rhizomelia � þþ � � þ �

*þ, mild; þþ, marked; �, absent.

TABLE II. Auxologic Measurements

I-1 I-4 II-1 II-2 II-3 II-4

Bone age (year) 18 18 2.7 5 6.8 8Height SD �6.6 �5.7 �2.7 �5.2 �4.9 �1.9Head circumference SD 2.7 6.2 0.4 1.7 0.2 1.8Upper to lower segment ratio 2 1.3 1.4 1.5 1.7 1.2Upper to lower segment ratio SD 14.2 35.4 NA 14.5 14.4 4.9Arm span SDa �6.6 �3.4 �7.3 �6.5 �6.1 �3.3Radial length SDa �3.4 �3 NA �5.5 �4.6 �3.5Arm length SDa �4.8 NA �3.1 �3.9 �3.9 �1.1Upper arm length SDa �3.3 �3.9 NA �4.7 �5.2 �2.2Hand length SDa �2.1 �2.3 �1 �1.9 �3.3 �0.2Tibia length SDa �4.9 �5.5 NA �1.7 �2.5 1

aMeasured on right extremity.

64 Ross et al.

any SHOX deficiency effects, as the fibula was long inpatient II-3 who had both achondroplasia and LWD.

The prevalence of neither LWD nor hypochondropla-sia is knownbecause of overlap of themilder phenotypeswith the short-stature population. In addition, the twosyndromes can look similar. Distinguishing featuresinclude the presence of Madelung wrist deformity inLWD only. In addition, the bones of the forearm andlower leg are specifically affected in LWD, while inhypochondroplasia the face, spine, hands, and feet arealso affected. There is less phenotypic overlap betweenachondroplasia and LWD. Surprisingly, the child af-fected with both disorders was not relatively shorterthan her parent with achondroplasia alone. Perhaps theFGFR3 mutation has the maximal damaging effecton the growth plate without any incremental inhibi-tion from SHOX abnormalities. The LWD features inchildren affected with LWD plus achondroplasia orhypochondroplasia were similar and not more severethan those present in their LWD-affected sibs. If SHOXand FGFR3 act on independent pathways of skeletaldevelopment, we would expect the phenotype of com-bined LWD and achondroplasia or hypochondroplasiato be additive or even synergistic. By contrast, thephenotypes of these children who are doubly hetero-zygous for SHOX and FGFR3 mutations appear to beless than completely additive, implying that these genesact in partially overlapping pathways of bone growth.The mixed phenotypes of these rare children with com-binedMendeliandisordershighlight the complexnatureof gene action and interaction in humans.

ACKNOWLEDGMENTS

We acknowledge the contribution of Dr. Pia Marttilafor mutational analysis and Ling Zhang for FISHanalyses. We also thank the family for participating inthis study.

REFERENCES

Belin V, Cusin V, Viot G, Girlich D, Toutain A, Moncla A, Vekemans M,Lemerrer M, Munnich A, Cormierdaire V. 1998. Shox mutations indyschondrosteosis (Leri-Weill syndrome). Nat Genet 19:67–69.

Bellus GA, Hefferon TW, Ortiz de Luna RI, Hecht JT, Horton WA,Machado M, Kaitila I, McIntosh I, Francomano CA. 1995. Achon-droplasia is definedby recurrentG380Rmutations ofFGFR3.AmJHumGenet 56:368–373.

Bellus GA, Gaudenz K, Zackai EH, Clarke LA, Szabo J, Francomano CA,Muenke M. 1996. Identical mutations in three different fibroblastgrowth factor receptor genes in autosomal dominant craniosynostosissyndromes. Nat Genet 14:174–176.

Bellus GA, Spector EB, Speiser PW, Weaver CA, Garber AT, Bryke CR,Israel J, Rosengren SS, Webster MK, Donoghue DJ, Francomano CA.2000. Distinct missensemutations of the FGFR3 lys650 codonmodulatereceptor kinase activation and the severity of the skeletal dysplasiaphenotype. Am J Hum Genet 67:1411–1421.

Brook CG, de Vries BB. 1998. Skeletal dysplasias. Arch Dis Child 79:285–289.

Espiritu C, Chen H, Woolley PV Jr. 1975a. Mesomelic dwarfism as thehomozygous expression of dyschondrosteosis. Am J Dis Child 129:375–377.

Espiritu CE, Chen H, Woolley PV Jr. 1975b. Probable homozygosity forthe dyschondrosteosis genes. Birth Defects Orig Artic Ser 11(6):127–132.

GrigelionieneG,EklofO, LaurencikasE,OllarsB,HertelNT,Dumanski JP,Hagenas L. 2000. Asn540Lys mutation in fibroblast growth factorreceptor 3 and phenotype in hypochondroplasia. ActaPaediatr 89:1072–1076.

Hall JG, Froster-Iskenius UG, Allanson JE. 1995. Handbook of normalphysical 12 measurements. Oxford: Oxford University Press.

Hamill PV, Drizd TA, Johnson CL, Reed RB, Roche AF, Moore WM. 1979.Physical growth: National Center for Health Statistics percentiles.Am J Clin Nutr 32:607–629.

HortonWA, Rotter JI, RimoinDL, Scott CI, Hall JG. 1978. Standard growthcurves for achondroplasia. J Pediatr 93:435–438.

Kunze J, Klemm T. 1980. Mesomelic dysplasia, type Langer: a homozygousstate for dyschondrosteosis. Eur J Pediatr 134:269–272.

Langer LO Jr. 1965. Dyschondrosteosis, a hereditable bone dysplasia withcharacteristic roentgenographic features. Am J Roentgenol 95:178–188.

Leri A, Weill J. 1929. Une affection congenitale et symetrique du devel-oppement osseux: la dyschondrosteose. Bull Mem Soc Med Hop (Paris)53:1491.

McKusick VA. 1972. Heritable disorders of connective tissue. Saint Louis,MO: Mosby.

McKusick VA, Kelly TE, Dorst JP. 1973. Observations suggesting allelismof the achondroplasia and hypochondroplasia genes. J Med Genet10:11–16.

Poznanski AK, Garn SM, Nagy JM, Gall JC Jr. 1972. Metacarpophalangealpattern profiles in the evaluation of skeletal malformations. Radiology104:1–11.

Ross JL, ScottC Jr,Marttila P,KowalK,NassA,PapenhausenP,Abboudi J,Osterman L, Kushner H, Carter P, Ezaki M, Elder F, Wei F, Chen H,Zinn AR. 2001. Phenotypes Associated with SHOX Deficiency. J ClinEndocrinol Metab 86:5674–5680.

ShearsDJ,VassalHJ,GoodmanFR,PalmerRW,ReardonW,Superti-FurgaA, Scambler PJ, Winter RM. 1998. Mutation and deletion of thepseudoautosomal gene SHOX cause Leri-Weill dyschondrosteosis. NatGenet 19:70–73.

Wei F, Cheng S, BadieN, Elder F, Scott C Jr, Nicholson L, Ross JL, Zinn AR.2001. A man who inherited his SRY gene and Leri-Weill dyschondro-steosis from his mother and neurofibromatosis type 1 from his father.Am J Med Genet 102:353–358.

Zinn AR, Tonk VS, Chen Z, Flejter WL, Gardner HA, Guerra R, Kushner H,SchwartzS,SybertVP,VanDykeDL,RossJL.1998.Evidence foraTurnersyndrome locus or loci at Xp11.2-p22.1. AmJHumGenet 63:1757–1766.

Zinn AR, Wei F, Zhang L, Elder FF, Scott CI Jr, Marttial P, Ross JL. 2002.Complete SHOX deficiency causes Langer mesomelic dysplasia. Am JMed Genet 110:158–163.

TABLE V. Radiographic Features of Madelung Deformity of the Wrist*

I-1 I-4 II-1 II-2 II-3 II-4

Clinical Madelung deformity No No Yes Yes Yes NoRadial lucency, % radial head 0 0 50 100 0 50Radial bowing Yes No No Yes Yes YesCarpal wedging SD �0.8 �0.3 ND ND �2.4 �2.5Fourth metacarpal length SD �6.6 �5.3 �2.7 �5 �2.8 �1.2Fifth metacarpal length SD �7.4 �4.4 �3.2 �6 �3.8 �1.3

*ND, not done; could not be assessed on X-ray.

SHOX and FGFR3 Double Heterozygosity 65