The Gene for Cherubism Maps to Chromosome 4p16.3

Am. J. Hum. Genet. 65:151–157, 1999

151

Jonathan Mangion,

1,∗

Nazneen Rahman,

1∗

Sarah Edkins,

Rita Barfoot,

Trang Nguyen,

Asgeir Sigurdsson,

John V. Townend,

David R. Fitzpatrick,

Adrienne M. Flanagan,

and

Michael R. Stratton

Section of Cancer Genetics, Institute of Cancer Research, Sutton, Surrey, United Kingdom;

Department of Maxillofacial Surgery, St.

Richard’s Trust, Chichester, West Sussex, United Kingdom;

Human and Clinical Genetics Unit, Molecular Medicine Centre, Western General

Hospitals Trust, Edinburgh;

Department of Histopathology, Imperial College School of Medicine at St. Mary’s, London

Summary

Cherubism is a rare familial disease of childhood char-

acterized by proliferative lesions within the mandible

and maxilla that lead to prominence of the lower face

and an appearance reminiscent of the cherubs portrayed

in Renaissance art. Resolution of these bony abnor-

malities is often observed after puberty. Many cases are

inherited in an autosomal dominant fashion, although

several cases without a family history have been re-

ported. Using two families with clinically, radiologically,

and/or histologically proved cherubism, we have per-

formed a genomewide linkage search and have localized

the gene to chromosome 4p16.3, with a maximum mul-

tipoint LOD score of 5.64. Both families showed evi-

dence of linkage to this locus. Critical meiotic recom-

binants place the gene in a 3-cM interval between

D4S127 and 4p-telomere. Within this region a strong

candidate is the gene for ﬁbroblast growth factor recep-

tor 3 (FGFR3); mutations in this gene have been im-

plicated in a diverse set of disorders of bone devel-

opment.

Introduction

Cherubism (MIM 118400) is a rare, painless disﬁguring

disease primarily affecting bones of the jaw (ﬁrst de-

scribed by Jones [1933, 1938], with many subsequent

reports, including those by Peters [1979], Riefkohl et al.

[1985], Zachariades et al. [1985], Zohar et al. [1989],

Kaugars et al. [1992], Marck and Kudrick [1992], Vail-

Received January 15, 1999; accepted for publication April 28, 1999;

electronically published May 21, 1999.

Address for correspondence and reprints: Dr. Adrienne Flanagan,

Department of Histopathology, Imperial College School of Medicine

at St. Mary’s, Norfolk Place, London W2 1PG, United Kingdom. E-

mail: a.ﬂ[email protected]

*These authors contributed equally to this article.

0002-9297/99/6501-0020$02.00

lant et al. [1993], Penfold et al. [1993], Hitomi et al.

[1996], Valiathan and Prathanth [1997], and Southgate

et al. [1998]). The mandible and maxilla are usually

bilaterally enlarged, producing a full, round lower con-

tour of the face. The skin over the cheeks is stretched

and pulls down the lower eyelids. As a consequence, a

thin line of sclera is exposed beneath the iris and the

eyes appear to be raised heavenward in a manner rem-

iniscent of a the cherubs in Renaissance paintings (ﬁg.

1).

Radiologically, cherubism is characterized by multil-

ocular radiolucent areas that may be unilateral (Arnott

1978; Reade et al. 1984) (ﬁg. 1). The disease is usually

restricted to the mandibular and maxillary regions, al-

though, occasionally, radiological abnormalities have

been observed in the ribs. Histopathological evaluation

of the lesions shows proliferating ﬁbrous connective tis-

sue containing numerous multinucleated giant cells,

which are osteoclasts (Southgate et al. 1998). The clin-

ical and histological features of cherubism may some-

times present problems in diagnostic distinction from

giant-cell tumor, giant-cell granuloma, ossifying

ﬁbroma, ﬁbrous dysplasia of the jaw, and Paget disease

of bone (Kerley and Schow 1981; Flanagan et al. 1988;

Zohar et al. 1989; Kaugars et al. 1992; Penfold et al.

1993; Whitaker and Singh 1995).

Affected individuals are normal at birth. Usually, the

disease manifests in early childhood (at age 2–5 years)

and becomes more marked until puberty, at which time

the bony lesions begin to regress. However, the distortion

of the jaw during childhood leads to permanent dental

abnormalities. Other complications, such as visual dis-

turbance (due to lesions within the orbital bones) and

deafness, have also been reported but are relatively rare

(Hawes 1989; Marck and Kudrick 1992).

Cherubism is a familial disease in which the trait is

transmitted in an autosomal dominant fashion (Peters

1979; Zohar et al. 1989; Southgate et al. 1998), al-

though several sporadic cases have been described (Kau-

gars et al. 1992; Ayoub and El-Mofty 1993). The pen-

etrance is high (Peters 1979), but the precise estimate

will depend on whether clinical or radiological diag-

152 Am. J. Hum. Genet. 65:151–157, 1999

Figure 1 Photographs and x-ray of 19-year-old girl with cherubism who is from family CH1. A, Anterior-posterior view, demonstrating

characteristic facial disﬁgurement of cherubism, including visible sclera below iris because of overgrowth of maxilla. B, lateral view, emphasizing

protruding mandible and proptosis of eyes. C, X-ray, showing that pathology is restricted to bones of skull that are formed by endochondrial

ossiﬁcation.

nostic criteria are used. Males are more commonly and

more severely affected than females. Cherubism has been

associated with other genetic disorders, including a num-

ber of cases both of a Noonan-like syndrome (known

as the “Noonan-like/multiple giant-cell lesion syn-

drome”) (MIM 163955; Cohen et al. 1974; Dunlap et

al. 1989; Cohen and Gorlin 1991; Betts et al. 1993)

typiﬁed by short stature; low–normal intelligence; ocular

hypertelorism; prominent posterorly angulated ears; gi-

ant-cell lesions of the bones, joints, and soft tissues; pec-

tus excavatum; and pulmonic stenosis) and of Ramon

syndrome (MIM 266270; Ramon et al. 1967; Pina-Neto

et al. 1986), which is typiﬁed by short stature, mental

retardation, epilepsy, gingival ﬁbromatosis, and hyper-

trichosis. Cherubism has also been described in an in-

dividual with fragile-X mental retardation (Quan et al.

1995).

The gene underlying cherubism has not been mapped.

We have therefore obtained samples from two large fam-

ilies with the disease and have performed a genomewide

linkage search.

Subjects and Methods

Families were interviewed and samples were collected

with full informed consent of the patients involved. The

studies were performed with the approval of the Re-

search Ethics Committee of St. Mary’s NHS Trust,

London.

DNAs were extracted from peripheral venous blood

samples by conventional methods. Polymorphic micro-

satellite markers, predominantly from the Ge´ne´thon set,

were PCR ampliﬁed (with one primer of each pair either

radio- or ﬂuorescently labeled) and were electrophoresed

through 4.5% denaturing polyacrylamide gels on either

an ABI377 DNA sequencer or a conventional gel rig, as

appropriate.

Two-point and multipoint linkage analyses were per-

formed by the MLINK suite of programs. A model was

employed in which disease alleles were given a frequency

of .0001, the penetrance was set at 90%, and no spo-

radic cases were allowed. Because of the variable severity

of the disease, strict diagnostic criteria were employed

in the assignment of disease status. The assignment of

affected status required evidence of a characteristic ra-

diological appearance (mandibular and/or maxillary

multilocular radiolucency) and/or typical histological

appearance in a biopsy. Individuals were classiﬁed as

“likely” to be affected if they had clinical features of the

disease (as indicated by information obtained from

childhood photographs or from the reports of family

members) in the absence of histological or radiological

conﬁrmation. Linkage analyses were performed, classi-

fying “likely” affected cases as unknown. Patients

thought, on the basis of pedigree information, to be ob-

ligate gene carriers, and for whom histological or ra-

diological conﬁrmation was not available or who were

clinically normal, were classiﬁed as affected for the pur-

poses of linkage. Allele frequencies were determined

from 24 unrelated chromosomes in families CH1 and

CH2.

Results

Families with Cherubism

Two families were used in a genomewide linkage

search for the cherubism gene (ﬁg. 2). Details of the

Mangion et al.: Cherubism Gene on Chromosome 4p 153

Figure 2 Pedigrees of cherubism families CH1 (A) and CH2 (B). Blackened symbols denote that individual is deﬁnitely affected, on the

basis of conﬁrmation by radiology or histology; stippled symbols denote that individual is likely affected, on the basis of clinical information,

but with no conﬁrmation by radiology or histology; unﬁlled symbols denote unaffected individuals and obligate gene carriers for whom

information is not available or who were phenotypically normal.

clinical status of individuals within these families are

summarized in table 1. Three cases (403, 408, 409) in

family CH2 were known to have been treated with ra-

diotherapy during childhood, to mitigate the effects of

the disease. Of these cases, one (409) subsequently de-

veloped an osteosarcoma of the jaw, in the area which

had been irradiated, and died. None of the affected in-

dividuals in family CH1 was treated by radiotherapy.

Genetic-Linkage Analysis

A framework genomic-linkage map of polymorphic

microsatellite markers spaced at 15–20-cM intervals was

used to perform the genomewide search. Initial evidence

of linkage to cherubism was obtained by use of D4S412,

for which an LOD score of 3.57 was obtained at a re-

combination fraction (v) of .05 (table 2). Subsequently,

LOD scores obtained by use of additional markers an-

alyzed in the vicinity of D4S412 conﬁrmed linkage to

this region, in both families. The highest two-point LOD

score obtained was 5.06 at v 5 0 from D4S3360. With

D4S3360 and D4S2936, maximum three-point LOD

scores of 1.16 and 4.48 (total 5.64) were obtained from

families CH1 and CH2. Linkage was conﬁrmed by a

haplotype of marker alleles segregating with the disease

in both pedigrees (ﬁg. 2). The two individuals (405 and

408) in family CH2 previously designated as “likely”

affecteds (table 1) carried the haplotype linked to the

disease in this family. When these were included as def-

154 Am. J. Hum. Genet. 65:151–157, 1999

Table 2

Total Two-Point LOD Scores for Cherubism Families CH1 and CH2

at Values of Theta from Markers on Chromosome 4p

LOD S

CORE AT

v 5

ARKER

0 .05 .1 .15 .2 .25 .3 .35 .4

D4S2936 4.05 3.61 3.17 2.72 2.26 1.81 1.37 .95 .57

D4S3038 3.18 2.83 2.48 2.12 1.76 1.41 1.06 .74 .44

PDE6B 2.51 2.18 1.86 1.55 1.25 .97 .72 .49 .29

D4S3360 5.06 4.55 4.03 3.5 2.95 2.40 1.86 1.32 .82

D4S1614 2.86 2.53 2.20 1.86 1.54 1.22 .92 .64 .38

D4S43 2.62 2.29 1.96 1.64 1.33 1.03 .76 .52 .31

D4S127 22.27 .76 .82 .77 .67 .55 .44 .32 .21

D4S3034 .91 .79 .67 .56 .45 .35 .26 .18 .11

D4S412 .66 3.57 3.36 3.03 2.63 2.20 1.75 1.30 .85

Table 1

Clinical Information and Disease Status for Families CH1 and CH2

Individual Information on Disease Status

Disease Status

CH1-101 Photo, PMH Obligate carrier, likely affected

CH1-103 Photo, PMH Obligate carrier, likely affected

CH1-201 Clinical, radiology, histology Deﬁnitely affected

CH1-202 Clinical, radiology Deﬁnitely affected

CH1-203 Clinical, radiology Deﬁnitely affected

CH1-204 Clinical, radiology Deﬁnitely affected

CH2-202 No information, DNA unavailable Obligate carrier

CH2-204 No information, DNA unavailable Obligate carrier

CH2-301 No evidence of cherubism Obligate carrier

CH2-303 No information, DNA unavailable Obligate carrier

CH2-308 No evidence of cherubism Obligate carrier

CH2-403 PMH, radiotherapy Obligate carrier, likely affected

CH2-405 PMH Likely affected

CH2-406 PMH Obligate carrier, likely affected

CH2-408 Radiotherapy, photo Likely affected

CH2-409 Radiotherapy, photo Likely affected

CH2-410 PMH Obligate carrier, likely affected

CH2-412 No information, DNA unavailable Obligate carrier

CH2-414 PMH Obligate carrier, likely affected

CH2-501 Clinical, histology, radiology Deﬁnitely affected

CH2-503 Clinical, histology, radiology Deﬁnitely affected

CH2-504 Histology Deﬁnitely affected

CH2-510 Radiology Deﬁnitely affected

CH2-511 Clinical, radiology Deﬁnitely affected

CH2-601 Clinical, radiology Deﬁnitely affected

Photo 5 suggestive childhood photograph; Clinical 5 clinical examination during child-

hood; PMH 5 patient told in the past by medical professional he or she was affected.

initely affected cases in the linkage analysis, the maxi-

mum total two-point LOD score was 6.01 at v 5 0 from

D4S3360, and, with D4S3360 and D4S2936, a three-

point LOD score of 5.51 was obtained from family CH2

(total LOD score, from both families, was 6.67). Three

individuals who previously had been phenotypically

classiﬁed as “unaffected,” who were not obligate gene

carriers, and who had a 50% chance of inheriting the

linked haplotype, did not inherit it, suggesting that the

penetrance of the disease is relatively high, as assumed

in the model (data not shown, to preserve con-

ﬁdentiality).

The marker order and the sex-averaged genetic dis-

tances within this interval from the Ge´ne´thon map are

4pter– D4S2936, D4S3038–1.8 cM – D4S1614–

1.9 cM–D4S3034, D4S412. On the basis of both ad-

ditional information from the Genetic Location Da-

tabase and mapping data from this study, we have in-

ferred the following marker order: 4pter–D4S2936,

D4S3038–PDE6B– D4S3360–D4S1614– D4S43–

D4S127–D4S3034, D4S412. Critical meiotic recom-

binants in family CH2 place the gene between D4S127

and 4p-telomere. D4S127 is physically located within

band 4p16.3, the terminal cytogenetic band on the

short arm of chromosome 4. All published polymorphic

short tandem repeats located within the region of link-

age deﬁned by the recombinant at D4S127 in family

CH2 have been analysed in families CH1 and CH2 and

in ﬁve additional families with cherubism. There is no

clear evidence of a segregating haplotype common to

more than one family with cherubism.

Discussion

We have provided strong evidence for the localization

of the gene for cherubism to chromosome 4p16.3, be-

tween D4S127 and 4p-telomere, an interval of ∼3 cM.

Both families analyzed show linkage to this region, so,

on the basis of this small series, there is no evidence for

Mangion et al.: Cherubism Gene on Chromosome 4p 155

genetic heterogeneity. These families have classic cher-

ubism, on the basis of strict, objective clinical criteria.

In this form, the disease is restricted to the bony tissues

of the lower face. However, aspects of the cherubism

phenotype are present in other diseases. For example,

some features of cherubism have been reported in as-

sociation with a Noonan-like syndrome (Dunlap et al.

1989; Betts et al. 1993). The classic form of Noonan

syndrome has previously been mapped to chromosome

12. If cases of the Noonan-like/multiple giant-cell lesion

syndrome are due to the Noonan syndrome gene on

chromosome 12, the genetic basis of this form of cher-

ubism must be different from that of the classic cher-

ubism present in families CH1 and CH2. There are no

published data pertaining to this issue. However, it is

also possible that Noonan syndrome with cherubism ei-

ther is allelic with respect to the cherubism gene on chro-

mosome 4 (perhaps as a result of a contiguous-gene syn-

drome) or is due to an entirely different gene. Cherubism

has also been reported as part of Ramon syndrome (Ra-

mon et al. 1967; Pina-Neto et al. 1986), which has not

yet been genetically mapped. However, the published

pedigrees are consistent with autosomal recessive inher-

itance. If this interpretation is correct, the genetic basis

of Ramon syndrome is also likely to differ from that of

classic cherubism, which is clearly due to an autosomal

dominant trait. It is therefore possible that the bony

lesions that characterize cherubism constitute a pheno-

typic picture common to a number of disease processes

that arise from multiple, distinct, initiating pathogenetic

events, and it would not be surprising if genetic heter-

ogeneity in classic cherubism is ultimately encountered.

Within the currently deﬁned interval there is one ma-

jor plausible candidate for the cherubism gene, ﬁbroblast

growth factor receptor 3 (FGFR3) (MIM 134934).

FGFR3 is composed of three glycosylated extracellular

immunoglobulin-like domains, a transmembrane do-

main, an intracellular kinase domain that is split, and a

carboxyl terminus. Mutations in FGFR3 are known to

cause a remarkably diverse set of diseases associated

with disordered growth of cartilage and bone. Achon-

droplasia is predominantly due to a single mutation in

the transmembrane domain (Rousseau et al. 1994;

Shiang et al. 1994). Hypochondroplasia is attributable

to missense mutations in the tyrosine kinase domain

(Bellus et al. 1995). Thanatophoric dysplasia type 1 is

caused by missense mutations which create cysteine res-

idues in the linker region between between immuno-

globulin domains 2 and 3 (Tavormina et al. 1995), than-

atophoric dysplasia type 2 by a mutation in the kinase

domain (Tavormina et al. 1995). Crouzon syndrome in

association with acanthosis nigricans is caused by mis-

sense mutations within the transmembrane domain

(Meyers et al. 1995). A mutation in the extracellular

domain causes non-syndromic craniosynostosis

(Muenke et al. 1997). Many of these mutations consti-

tutively activate the kinase activity of FGFR3, which, in

turn, is believed to inhibit normal development of car-

tilage. By contrast, homozygous disruption of FGFR3

in mice results in both deafness and bony abnormalities

associated with overgrowth of long bones (Colvin et al.

1996; Deng et al. 1996). A further candidate for cher-

ubism is MSX1 (also known as “Hox7”; MIM 142983).

A mutation in MSX1 is responsible for an autosomal

dominant syndrome of agenesis of second premolars and

third molars (MIM 106600; Vastardis et al. 1996), and

mice rendered homozygous for a nonfunctioning MSX1

gene show cleft palate with other facial and dental ab-

normalities (Satokata and Maas 1994). On the basis of

published genetic and physical maps, MSX1 is located

centromeric to the currently deﬁned cherubism region,

but minor changes in the map order would include it.

A polymorphic microsatellite marker located within

MSX1 was not informative in family CH2 (data not

shown).

It is interesting that the pedigrees in both families

show some clinical evidence of anticipation. This is par-

ticularly intriguing in family CH2, in which at least two

branches (301, 403, and 501 and 308, 410, and 504)

provide reasonable evidence for progressive worsening

of the phenotype over three generations. However, in

view of the age-dependent phenotype in this disease, it

is difﬁcult to assess this formally. There have been no

previous reports of anticipation in cherubism, and this

observation may represent a bias of ascertainment of the

younger—and thus obviously affected—individuals.

Although an increased risk of frank neoplastic change

associated with cherubism has been alluded to in the

published literature (Caballero Herrera et al. 1998), this

is clearly rare. Of three individuals in our series of fam-

ilies who were known to have been treated with radi-

otherapy to mitigate the effects of the disease, one de-

veloped an osteosarcoma of the mandible. It is highly

likely that the radiation treatment was causally impli-

cated in the development of this tumor. Whether the

presence of cherubism increased the likelihood of neo-

plastic change in response to the radiation dose is less

clear. However, it is plausible, given the proliferative na-

ture of the disease, the rarity of osteosarcoma (even after

irradiation), and the fact that other diseases character-

ized by bone remodeling (such as Paget disease or ﬁbrous

dysplasia) are also associated with an elevated risk of

osteosarcoma. Indeed, the existence of unilateral and

asymmetric cases of cherubism, the localized nature of

the changes within bone even when it is bilateral, and

the histological similarity to other presumed neoplastic

conditions (such as giant-cell tumor) suggest that the

proliferative lesions of cherubism may themselves con-

stitute multiple neoplastic clones. On the other hand, a

striking feature of the disease is its resolution after pu-

156 Am. J. Hum. Genet. 65:151–157, 1999

berty, which is more suggestive (although not deﬁnitive)

of a hyperplastic than of an autonomous neoplastic pro-

cess. Although we currently have no insight into the

mechanism of action of the cherubism gene, it should

now be possible to evaluate some of these hypotheses

further—for example, by examinination of biopsy sam-

ples from the bony lesions of cherubism for evidence of

loss of heterozygosity on chromosome 4p, and ulti-

mately by structural and functional analyses of the cher-

ubism gene itself.

Acknowledgments

We would like to thank the families with cherubism, for

their enthusiastic help and encouragement, and the following

medical practitioners, for their help in contacting the families

and obtaining the samples; Dr. J. A. Jones, The Health Centre,

Penarth, Vale of Glamorgan, United Kingdom; Mr. J. C.

Lowry, Royal Inﬁrmary, Blackburn, Lancs., United Kingdom;

Mr. M. C. Gregory, Royal Gwent Hospital, Newport, Gwent,

United Kingdom; Mr. G. D. D. Roberts, York District Hospital,

York, United Kingdom; Miss B. Jones, West Hill Hospital,

Dartford, Kent, United Kingdom; Dr. M. Ireland, Royal Vic-

toria Inﬁrmary, Newcastle upon Tyne; Mr W. Peters, Poole

General Hospital, Poole, Dorset, United Kingdom. This work

was supported by the Cancer Research Campaign, the Medical

Research Council, and the Institute of Cancer Research. N.R.

is an MRC Clinical Research Fellow.

Electronic-Database Information

Accession numbers and URLs for data in this article are as

follows:

Ge´ne´thon map, http://waldo.wi.mit.edu/ftp/distribution/

human_STS_releases/july97/genmap/Chr4.genmap (for ge-

netic map)

Genetic Location Database, http://cedar.genetics.soton.ac.uk/

public_html/ (for physical and genetic marker maps)

Online Mendelian Inheritance in Man (OMIM), http://www

.ncbi.nlm.nih.gov/Omim (for Cherubism [MIM 118400],

Noonan-like syndrome [MIM 163955], Ramon syndrome

[MIM 266270], FGFR3 [MIM 134934], MSX1 [MIM

142983], autosomal dominant syndrome of agenesis of sec-

ond premolars and third molars [MIM 106600])

References

Arnott DG (1978) Cherubism—an initial unilateral presen-

tation. Br J Oral Surg 16:38–46

Ayoub AF, El-Mofty SS (1993) Cherubism: report of an ag-

gressive case and review of the literature. J Oral Maxillofac

Surg 51:702–705

Bellus GA, McIntosh I, Smith EA, Aylsworth AS, Kaitila I,

Horton WA, Greenhaw GA, et al (1995) A recurrent mu-

tation in the tyrosine kinase domain of ﬁbroblast growth

factor receptor 3 causes hypochondroplasia. Nat Genet 10:

357–359

Betts NJ, Stewart JC, Fonseca RJ, Scott RF (1993) Multiple

central giant cell lesions with a Noonan-like phenotype. Oral

Surg Oral Med Oral Pathol 76:601–607

Caballero Herrera R, Vinals Iglesias H (1998) Cherubism: a

study of three generations. Medicina Oral 3:163–171

Cohen MM, Gorlin RJ (1991) Noonan-like multiple giant-cell

lesion syndrome. Am J Med Genet 40:159–166

Cohen MM, Ruvalcaba RHA, Graham CB, Harrison MT,

Morgan AF (1974) A new syndrome simulating the Noonan

syndrome, the Leopard syndrome and hyperparathyroidism.

Syndrome Ident 2:14–17

Colvin JS, Bohne BA, Harding GW, McEwen DG, Ornitz DM

(1996) Skeletal overgrowth and deafness in mice lacking

ﬁbroblast growth factor receptor 3. Nat Genet 12:390–397

Deng C, Wynshaw-Boris A, Zhou F, Kuo A, Leder P (1996)

Fibroblast growth factor receptor 3 is a negative regulator

of bone growth. Cell 84:911–921

Dunlap C, Neville B, Vickers RA, O’Neil D, Barker B (1989)

The Noonan syndrome/cherubism association. Oral Surg

Oral Med Oral Pathol 67:698–705

Flanagan AM, Tinkler SMB, Horton MA, Williams DM,

Chambers TJ (1988) The multinucleate giant cells in giant

cell granulomas of the jaw are osteoclasts. Cancer 62:

1139–1145

Hawes MJ (1989) Cherubism and its orbital manifestations.

Ophthal Plast Reconstr Surg 5:133–140

Hitomi G, Nishide N, Mitsui K (1996) Cherubism: diagnostic

imaging and review of the literature in Japan. Oral Surg

Oral Med Oral Pathol Oral Radiol Endod 81:623–628

Jones WA (1933) Familial multilocular cystic disease of the

jaws. Am J Cancer 17:946-950

——— (1938) Further observations regarding familial mul-

tilocular cystic disease of the jaws. Br J Radiol 11:227–241

Kaugars GE, Niamtu J III, Svirsky JA (1992) Cherubism: di-

agnosis, treatment, and comparison with central giant cell

granulomas and giant cell tumors. Oral Surg Oral Med Oral

Pathol 73:369–374

Kerley TR, Schow CE Jr (1981) Central giant cell granuloma

or cherubism: report of a case. Oral Surg Oral Med Oral

Pathol 51:128–130

Marck PA, Kudryk WH (1992) Cherubism. J Otolaryngol 21:

84–87

Meyers GA, Orlow SJ, Munro IR, Przylepa KA, Jabs EW

(1995) Fibroblast growth factor receptor 3 (FGFR3) trans-

membrane mutation in Crouzon syndrome with acanthosis

nigricans. Nat Genet 11:462–464

Muenke M, Gripp KW, McDonald-McGinn DM, Gaudenz K,

Whitaker LA, Bartlett SP, Markowitz RI, et al (1997) A

unique point mutation in the ﬁbroblast growth factor re-

ceptor 3 gene (FGFR3) deﬁnes a new craniosynostosis syn-

drome. Am J Hum Genet 60:555–564

Penfold CN, McCullagh P, Eveson JW, Ramsay A (1993) Giant

cell lesions complicating ﬁbro-osseous conditions of the

jaws. Int J Oral Maxillofac Surg 22:158–162

Peters WJ (1979) Cherubism: a study of twenty cases from

one family. Oral Surg Oral Med Oral Pathol 47:307–311

Pina-Neto JM, Moreno AF, Silva LR, Velludo MA, Petean EB,

Ribeiro MV, Athayde-Junior L (1986) Cherubism, gingival

ﬁbromatosis, epilepsy, and mental deﬁciency (Ramon syn-

Mangion et al.: Cherubism Gene on Chromosome 4p 157

drome) with juvenile rheumatoid arthritis. Am J Med Genet

25:433–441

Quan F, Grompe M, Jakobs P, Popovich BW (1995) Sponta-

neous deletion in the FMR1 gene in a patient with fragile-

X syndrome and cherubism. Hum Mol Genet 4:1681–1684

Ramon Y, Berman W, Bubis JJ (1967) Gingival ﬁbromatosis

associated with cherubism. Oral Surg Oral Med Oral Pathol

24:435–448

Reade PC, McKellar GM, Radden BG (1984) Unilateral man-

dibular cherubism: brief review and case report. Br J Oral

Maxillofac Surg 22:189–194

Riefkohl R, Georgiade GS, Georgiade NG (1985) Cherubism.

Ann Plast Surg 14:85–90

Rousseau F, Bonaventure J, Legeai-Mallet L, Pelet A, Rozet

JM, Maroteaux P, Le Merrer M, et al (1994) Mutations in

the gene encoding ﬁbroblast growth factor receptor-3 in

achondroplasia. Nature 371:252–254

Satokata I, Maas R (1994) Msx1 deﬁcient mice exhibited cleft

palate and abnormalities of craniofacial and tooth devel-

opment. Nat Genet 6:348–355.

Shiang R, Thompson LM, Zhu YZ, Church DM, Fielder TJ,

Bocian M, Winokur ST, et al (1994) Mutations in the trans-

membrane domain of FGFR3 cause the most common ge-

netic form of dwarﬁsm, achondroplasia. Cell 78:335–342

Southgate J, Sarma U, Townend JV, Barron J, Flanagan AM

(1998) Study of the cell biology and biochemistry of cher-

ubism. J Clin Pathol 52:831–837

Tavormina PL, Shiang R, Thompson LM, Zhu YZ, Wilkin DJ,

Lachman RS, Wilcox WR, et al (1995) Thanatophoric dys-

plasia (types I and II) caused by distinct mutations in ﬁbro-

blast growth factor receptor 3. Nat Genet 9:321–328

Valiathan A, Prashanth VK (1997) Cherubism: presentation

of a case. Angle Orthod 67:237–238

Vaillant JM, Romain P, Divaris M Cherubism. (1989) Findings

in three cases in the same family. J Craniomaxillofac Surg

17:345–349

Vastardis H, Karimbux N, Guthua SW, Seidman JG, Seidman

CE (1996) A human MSX1 homeodomain missense muta-

tion causes selective tooth agenesis. Nat Genet 13:417–421.

Whitaker SB, Singh BB (1995) Intraoral giant cell lesions: the

peripheral and central forms of these entities. Pract Perio-

dontics Aesthetic Dent 7:41–47

Zachariades N, Papanicolaou S, Xypolyta A, Constantinidis

I (1985) Cherubism. Int J Oral Surg 14:138–145

Zohar Y, Grausbord R, Shabtai F, Talmi Y (1989) Fibrous

dysplasia and cherubism as an hereditary familial disease:

follow-up of four generations. J Craniomaxillofac Surg 17:

340–344