How to Cite this Article: McDonald-McGinn DM, Feret H, Nah H-D, Bartlett SP, Whitaker LA, Zackai EH. 2010. Metopic craniosynostosis due to mutations in GLI3: A novel association. Am J Med Genet Part A 152A:1654–1660.
Metopic craniosynostosis due to mutations in GLI3: A novel association†
Article first published online: 25 JUN 2010
Copyright © 2010 Wiley-Liss, Inc.
American Journal of Medical Genetics Part A
Volume 152A, Issue 7, pages 1654–1660, July 2010
How to Cite
McDonald-McGinn, D. M., Feret, H., Nah, H.-D., Bartlett, S. P., Whitaker, L. A. and Zackai, E. H. (2010), Metopic craniosynostosis due to mutations in GLI3: A novel association. Am. J. Med. Genet., 152A: 1654–1660. doi: 10.1002/ajmg.a.33495
- Issue published online: 25 JUN 2010
- Article first published online: 25 JUN 2010
- Manuscript Accepted: 4 APR 2010
- Manuscript Received: 13 JAN 2010
- premature fusion of the sutures;
- Sonic hedgehog;
We report on the novel association of trigonocephaly and polysyndactyly in two unrelated patients due to mutations within the last third (exon 14) and first third (exon 6) of the GLI3 gene, respectively. GLI3 acts as a downstream mediator of the Sonic hedgehog signal-transduction pathway which is essential for early development; and plays a role in cell growth, specialization, and patterning of structures such as the brain and limbs. GLI3 mutations have been identified in patients with Pallister–Hall, Grieg cephalopolysyndactyly syndrome (GCPS), postaxial polydactyly type A1, preaxial polydactyly type IV, and in one patient with acrocallosal syndrome (ACLS). Furthermore, deletions including the GLI3 gene have been reported in patients with features of GCPS and ACLS. To date, trigonocephaly has not been associated with abnormalities of GLI3 and craniosynostosis is not a feature of GCPS. However, Hootnick and Holmes reported on a father with polysyndactyly and son with trigonocephaly, polysyndactyly, and agenesis of the corpus callosum, considered GCPS thereafter. Guzzetta et al. subsequently described a patient with trigonocephaly, polysyndactyly, and agenesis of the corpus callosum postulating a diagnosis of GCPS, later considered ACLS. In retrospect, these two patients, evaluated prior to mutational analysis, and our patients, with confirmed mutations, likely fall within the GLI3 morphopathy spectrum and may provide a bridge to better understanding those patients with overlapping features of GCPS and ACLS. Based on this observation, we suggest GLI3 studies in patients presenting with this constellation of findings, specifically metopic craniosynostosis with polysyndactyly, in order to provide appropriate medical management and genetic counseling. © 2010 Wiley-Liss, Inc.
In this timely article McDonald-McGinn et al. document mutations of GLI3 in two patients with metopic craniosynostosis and polysyndactyly of the limbs. The authors propose that this pattern expands the phenotypes caused by GLI3 mutations.
I would like to draw the reader's attention to a related paper in the June issue of the Journal: Kini et al. describe a series of patients with metopic synostosis demonstrating quite convincingly the causal heterogeneity of this defect. Notably these authors describe a patient in the collection with metopic synostosis and a mutation of GLI3 that they call Greig cephalopolysyndactyly syndrome.
The combination of these articles provides new insights into metopic craniosynostosis.
John C. Carey
Premature fusion of the metopic suture and its resultant abnormal head shape is also referred to as trigonocephaly because of the triangular appearance of the forehead when examined from above. Trigonocephaly is etiologically heterogeneous as it may occur as an isolated malformation or as part of a multiple anomaly syndrome [Lajeunie et al., 1998]. Examples of syndromic metopic craniosynostosis include: valproic acid embryopathy [Lajeunie et al., 2001]; chromosomal aneuploidy such as deletions of 9p and 11q [Azimi et al., 2003; Johnston et al., 2005]; and several previously described malformation syndromes whose molecular basis has yet to be defined including Frydman syndrome [Frydman et al., 1984], Opitz C trigonocephaly syndrome [Opitz, 1969; Azimi et al., 2003], Say Meyer trigonocephaly syndrome [Say and Meyer, 1981; Azimi et al., 2003], and autosomal dominant trigonocephaly [Hennekam and Van Den Boogaard, 1990]. In addition, trigonocephaly has occasionally been observed in association with well described craniosynostosis conditions such as Crouzon syndrome due to a Ser267Pro mutation in FGFR2 [Tartaglia et al., 1999]; Muenke syndrome due to the Pro250Arg mutation in FGFR3 [van der Meulen et al., 2006]; and Saethre–Chotzen syndrome, prior to the availability of mutational analysis/microdeletion testing [Cristofori and Filippi, 1992; and Hunter et al., 1976 as suggested by Cohen and MacLean, 2000]. Moreover, a child with nonsyndromic trigonocephaly was recently reported with an unusual mutation in the IgIII loop domain of FGFR1 (Ile300Trp) [Kress et al., 2000]. However, no mutations were subsequently identified in 81 patients with both syndromic and nonsyndromic trigonocephaly screened for this unusual FGFR1 mutation [Jehee et al., 2006].
Craniosynostosis syndromes have classically been characterized by the type of sutural involvement and the absence or presence and severity of the extremity findings. For example, bicoronal craniosynostosis with symphalangism of the hands and feet is typically associated with Apert syndrome; whereas medially deviated broad thumbs and great toes are seen in Pfeiffer syndrome; as compared with Saethre–Chotzen syndrome where laterally deviated duplicated/triplicated great toes are frequently observed; all in contrast with Crouzon syndrome where there is usually no extremity involvement. Here we report, for the first time, on two unrelated patients with metopic craniosynostosis and limb involvement due to mutations in GLI3.
A male presented at 2½ weeks of age due to a history of metopic craniosynostosis (Fig. 1) and four limb postaxial polydactyly (Fig. 2). He was the former 3,900 g (>90%) product of a 37-week gestation born to a 27-year-old G1P0-1 mother via caesarian secondary to maternal cholestasis. Pregnancy and family history were noncontributory. On physical exam, his length was 54 cm (90th centile) and head circumference was 37.5 cm (85th centile). The trigonocephaly was appreciable as were upslanting palpebral fissures. Interpupillary distance was 4 cm (50th centile). Full digit postaxial polydactyly of all four extremities was noted. In addition, the distal phalanges of both thumbs appeared broad (Fig. 3). Premature fusion of the metopic suture was confirmed by 3-D CT scan (Fig. 4). The lateral and third ventricles were noted to be top normal to mildly prominent in size but no structural brain anomalies were identified.
Follow-up developmental history at 14 months revealed normal motor, speech, and cognitive milestones with walking at age 13 months, babbling present, and appropriate social interaction.
Presuming the patient had two separate entities (metopic suture fusion and postaxial polydactyly) we performed the following laboratory studies: SNP array and TWIST mutational analysis (Saethre–Chotzen testing) due to the trigonocephaly, as well as GLI3 sequencing in light of the limb findings. Both the array and TWIST studies were normal; however, the GLI3 analysis revealed a novel frameshift mutation in exon 14 (c.4542_4545delCCAC) resulting in premature protein termination (p.His1515ProfsX3). Although not reported previously, this type of mutation is expected to be pathogenic. Parental studies were normal.
A male presented initially 13 years ago, at 4 months of age, due to a history of metopic craniosynostosis (Fig. 5) and four limb anomalies (Fig. 6). He was the former 2,665 g (25%) product of a 37-week gestation born to a 33-year-old G1P0-1 mother by spontaneous vaginal delivery. Pregnancy and family history were noncontributory. On physical examination at that time, his height was 62 cm (25th centile) and head circumference was 42.5 cm (60th centile). He had apparent trigonocephaly and relative hypertelorism with an interpupillary distance of 4.9 cm (97th centile). His extremities were notable for bilateral complete cutaneous syndactyly of the third and fourth fingers; duplication of the great toe on the right with soft tissue syndactyly of toes two and three; and medial deviation of the great toe on the left. Premature fusion of the metopic suture was confirmed by 3-D CT scan (Fig. 7). Mild prominence of the subarachnoid spaces and ventricular systems were noted at that time but no structural brain anomalies were identified.
Developmental history showed normal acquisition of motor, language, and cognitive milestones with walking at 13 months and speech emergence by one year of age. Follow-up evaluation at age 4 years 6 months, including formal developmental testing using the WPPSI-R, revealed a FSIQ score of 99. At 13 years of age he attends regular school with high academic achievement.
We initially viewed the metopic synostosis and extremity findings as two separate entities and performed high resolution karyotype and sequence analysis of FGFR 1 and 2 and TWIST. These studies were normal with the exception of identifying a paternally inherited 21 base pair duplication between mutations 243 and 277 of the TWIST gene resulting in an in-frame insertion of 7 amino acids N terminal to the DNA binding domain. This finding has subsequently been classified as a polymorphism [Elanko et al., 2001]. Thereafter the identification of the GLI3 mutation in our Patient 1 prompted us to perform similar testing on Patient 2 which showed a novel frameshift mutation in exon 6 of the GLI3 gene (c.1018delA). As in Patient 1, this too is predicted to result in premature protein termination (p.Ser340ValfsX7) and therefore would be expected to be pathogenic. Parental studies are unavailable at this time.
The GLI3 gene, located on chromosome 7p13, encodes for a protein that is a zinc finger bifunctional transcription factor, having both a transcriptional repressor and activation effect. It is a downstream mediator of the Sonic hedgehog (SHH) signal-transduction pathway which is essential for early development as it plays a role in cell growth, cell specialization, and the patterning of structures such as the brain and limbs. Mutations in the GLI3 gene have been identified in patients with Pallister–Hall syndrome (PHS), Grieg cephalopolysyndactyly syndrome (GCPS), postaxial polydactyly type A1, preaxial polydactyly type IV and in one patient with acrocallosal syndrome (ACLS) [Elson et al., 2002]. In addition, deletions including the GLI3 gene have been discovered in patients with overlapping features of GCPS and ACLS [Johnston et al., 2003].
Truncating mutations in the middle third of the gene are seen in patients with Pallister–Hall syndrome. They generate a constitutive gain of function of the GLI3 repressor protein that is likely to be independent of SHH controlled post-translational regulation. Radhakrishna et al. 1999 has used the term GLI3 morphopathy to lump the remaining entities associated with GLI3 mutations in the first and last third of the gene resulting in haploinsufficiency (GCPS, postaxial polydactyly type A1, preaxial polydactyly type IV), as the border of the phenotypic characterizations amongst the diagnoses is somewhat blurred.
Biesecker 2004 defined patients with GCPS (OMIM 17500) as having true hypertelorism, macrocephaly, and extremity involvement including preaxial polydactyly with cutaneous syndactyly of at least one limb or mixed pre and postaxial polydactyly. Johnston et al. 2005, however, suggested that patients presenting with features in the GCPS spectrum may not have manifestations amenable to phenotype diagnoses because of variable severity and nonspecific clinical features of GCPS. Furthermore they proposed relaxed clinical criteria in selecting patients for molecular analysis.
Returning to the key feature of metopic craniosynostosis in our patients, we note that there have been no reports of trigonocephaly in a patient with a GLI3 mutation to date. Moreover, although craniosynostosis has been listed as an occasional feature in GCPS, we were unable to confirm the association on a review of the literature with the following exceptions: Hootnick and Holmes 1972 reported on a father with polysyndactyly and his son with trigonocephaly, polysyndactyly, and agenesis of the corpus callosum, thereafter considered to have GCPS by Gorlin et al. 2001; and Guzzetta et al. 1996 described a patient with trigonocephaly, polysyndactyly (fusion of the 3rd and 4th rays on both hands, bifid thumbs, and a supernumerary 2nd ray on both feet), and agenesis of the corpus callosum postulating a diagnosis of GCPS or Carpenter syndrome (now known to be due to RAB23 mutations). In commenting on the Guzzetta et al. article, Fryns et al. 1997 suggested ACLS as another diagnostic possibility, harkening back to two previous patients whom he had described with ACLS, as well as a 6-month-old reported on by Turolla et al. 1990, all with mild manifestations of ACLS.
Cardinal features of ACLS generally include hallux duplication, postaxial polydactyly, absence of the corpus callosum, and developmental delay/mental retardation. When discussing the almost universal feature of severe mental retardation in ACLS, Gorlin et al. 2001 pointed to both the aforementioned Fryns et al. and Guzzetta et al. patients as exceptions to this rule. Moreover, as the child reported by Guzzetta et al. 1996 also had trigonocephaly, Gorlin et al. 2001 introduced the possibility of a distinct yet pathogenetically related condition. Perhaps in retrospect, and in light of our patients described herein, the individual described by Guzzetta et al. 1996, as well as the family reported by Hootnick and Holmes 1972, all evaluated prior to the availability of mutational analysis, fall within the GLI3 morphopathy spectrum. If so, they would provide a bridge to better understanding those patients with overlapping features of GCPS and ACLS.
This clinical overlap of GCPS and ACLS is noteworthy and historically has been the subject of much debate [Schinzel, 1982; Legius et al., 1985]. Note in particular the following: patients with severe GCPS, caused by deletions of GLI3, have been reported with manifestations overlapping ACLS including anomalies of the corpus callosum [Johnston et al., 2003]; Bonatz et al. 1997 reported on a child with ACLS whose father had GCPS; and as mutations in GLI3 have been identified in one patient with ACLS [Elson et al., 2002]. In contrast many authors consider ACLS to be heterogeneous with the majority having autosomal recessive inheritance due to the presence of parental consanguinity [Gorlin et al., 2001; Johnston et al., 2003], and linkage to GLI3 has been excluded in at least one family with ACLS [Brueton et al., 1988].
Thus, based on this historical information, it seems plausible that the previous cases reported with the GCPS/ACLS phenotypic overlap actually fall under the GLI3 morphopathy umbrella [Radhakrishna et al., 1999], whereas those families with only siblings affected with ACLS, and where the parents have no manifestations of the disorder, do in fact have an alternative etiology. Likewise, it is feasible that the patients described by Hootnick and Holmes 1972 and Guzzetta et al. 1996, both with metopic suture synostosis, polysyndactyly and normal development, as well as our two patients with trigonocephaly, polysyndactyly, and normal intelligence, due to confirmed GLI3 mutations, expand the GLI3 morphopathy spectrum and may be considered to have an atypical form of GCPS. This is supported by the biology of GLI3.
Given the essential role of SHH-GLI3 signaling in antero-posterior specification of limb and digit formation, even a slight disturbance in the GLI3 expression pattern is expected to produce a certain degree of limb malformation. Studies of a genetic mouse model for GCPS, extra toes (Xtj), which carries a Gli3 null allele, show that heterozygous (Gli3+/−) mice display mild preaxial polydactyly, while homozygous (Gli3−/−) mice present severe polydactyly marked by complete loss of digit identity [Hui and Joyner, 1993]. The greater severity exhibited in homozygous null mice underscores the importance of the expression level of GLI3 protein. Based on phenotypic similarity with Gli3+/− mice, the patients described in this report are likely to express functional GLI3 protein at a reduced level. However, the biochemical fate of the mutant RNAs and proteins has yet to be determined for their translatability, stability, and function.
The polydactyly in Gli3−/− mouse mutants has been postulated to involve Shh. Shh is normally expressed only in the zone of polarizing activity (ZPA) located in the posterior part of the limb bud to organize antero-posterior pattering of the limb. In the Gli3−/− mutant limb bud, Shh is ectopically expressed on the anterior side opposite the ZPA, inducing the mirror image duplication of the limb and thus explaining the polydactyly [Büscher et al., 1997]. In addition to determining digit identity and numbers, Gli3 is involved in digit separation by inducing apoptosis in the mesenchyme through downstream effectors, Msx2 and possibly BMP4 [Bastida et al., 2004; Lallemand et al., 2009]. It is also involved in induction or maintenance of cell death by limiting the number of cells expressing Fgf8 [Aoto et al., 2002]. Gli3 normally functions to repress Fgf8. In Gli3−/− mice, Fgf8 is upregulated in the apical ectodermal ridge twofold, with a reduction in apoptosis in the interdigital ridges, explaining the syndactyly.
Upregulation of Fgf8 was seen in the anterior neural ridge, the isthmus, and the facial primordia in Gli3−/− mutant embryos [Aoto et al., 2002]. FGF8 not only inhibits apoptosis but also stimulates Runx2 expression and osteoblast differentiation [Zhou et al., 2000], which may provide an explanation for premature ossification of a midline suture. In fact, increasing evidence points to an inhibitory role of Gli3 repressor (Gli3-R) in skeletal development and osteogenesis. In vitro analyses have shown that Gli3-R inhibits expression and activity of two essential osteogenic factors, BMP2 and Runx2, respectively [Garrett et al., 2003; Ohba et al., 2008]. Thus, the combination of a decrease in cell apoptosis and an increase in osteogenic differentiation may explain the occurrence of metopic synostosis as a part of the wide spectrum of nonoverlapping clinical features in GLI3 morphopathies, which likely is a result of loss of GLI3-R function.
In summary, our two unrelated patients, with an apparently novel collection of findings including trigonocephaly and polysyndactyly due to mutations within the last third (exon 14) and first third (exon 6) of the GLI3 gene, respectively, expand the phenotype associated with GLI3 mutations to include metopic craniosynostosis. Furthermore, these patients, and in retrospect the patients described by Hootnick and Holmes 1972 and Guzzetta et al. 1996, evaluated prior to the availability of mutational analysis, provide a bridge to better understanding those patients previously described with overlapping features of GCPS and ACLS. Thus, based on this new association, we would suggest GLI3 mutational analysis and/or deletion studies in those individuals with metopic suture fusion and limb anomalies and in patients with Carpenter syndrome found to be negative for RAB23 mutations, in order to provide appropriate medical management and genetic counseling.
The authors would like to thank: Laurie Lunny, Sarah MacDougall, Kristin D'Aco, M.D., Brandon Calderon, Von Bartz, Patricia Schultz, M.S.N., Rosa Casasanto, and Rosa Soto for their expert assistance in the preparation of this manuscript; Prevention Genetics for performing the clinical molecular testing; and the families of the two children presented here for allowing us to share their medical information and photographs.
- 2002. Mouse GLI3 regulates Fgf8 expression and apoptosis in the developing neural tube, face, and limb bud. Dev Biol 251: 320–332. , , , .
- 2003. Clinical and genetic aspects of trigonocephaly: A study of 25 cases. Am J Med Genet Part A 117A: 127–135. , , , , , , .
- 2004. Levels of Gli3 repressor correlate with Bmp4 expression and apoptosis during limb development. Dev Dyn 231: 148–160. , , , , , .
- 2004. What can you learn from one gene: GLI3. J Med Genet 43: 465. .
- 1997. Acrocallosal syndrome: A case report. J Hand Surg 22A: 492–494. , , , .
- 1988. Am J Med Genet 31: 799–804. , , , .
- 1997. Evidence for genetic control of Sonic hedgehog by Gli3 in mouse limb development. Mech Dev 62: 175–182. , , , .
- 2000. Craniosynostosis—Diagnosis, evaluation and management, 2nd edition, Oxford University Press. NY. , .
- 1992. Saethre-Chotzen syndrome with trigonocephaly. Am J Med Genet 44: 611–614. , .
- 2001. A survey of TWIST for mutations in Craniosynostosis Reveals a variable length polyglycine tract in asymptomatic individuals. Hum Mut 18: 535–541. , , , , , , , .
- 2002. De novo GLI3 mutation in acrocallosal syndrome: Broadening the phenotypic spectrum of GLI3 defects and overlap with murine models. J Med Genet 39: 804–806. , , , , .
- 1984. Trigonocephaly: A new familial syndrome. Am J Med Genet 18: 55–59. , , .
- 1997. Polysyndactyly and trigonocephaly with partial agenesis of corpus callosum: An example of the variable clinical spectrum of the acrocallosal syndrome? Clin Dysmorphol 6: 285–286. , , .
- 2003. Selective inhibitors of the osteoblast proteasome stimulate bone formation in vivo and in vitro. J Clin Invest 111: 1771–1782. , , , , , , , , , , , .
- 2001. Syndromes of the head and neck, 4th edition, New York: Oxford University Press. , , .
- 1996. Letter to the Editors: Polysyndactyly and trigonocephaly with partial agenesis of corpus callosum. Clin Dysmorphol 5: 179–182. , , , , .
- 1990. Autosomal dominant craniosynostosis of the sutura metopica. Clin Genet 38: 374–377. , .
- 1972. Familial polysyndactyly and craniofacial anomalies. Clin Genet 3: 128–134. , .
- 1993. A mouse model of Greig cephalopolysyndactyly syndrome: The extra-toes mutation contains an intragenic deletion of the Gli3 gene. Nat Genet 3: 241–246. , .
- 1976. Trigonocephaly and associated minor anomalies in mother and son. J Med Genet 13: 77–79. , , .
- 2006. Mutational screening of FGFR1, CER1, and CDON in a large cohort of trigonocephalic patients. Cleft Palate-Cranio J 43: 148–151. , , , , , , , , , , .
- 2005. Molecular and clinical analyses of Greig cephalopolysyndactyly and Pallister-Hall syndromes: Robust phenotype prediction from the type and position of GLI3 mutations. Am J Hum Genet 76: 609. , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , .
- 2003. Clinical and molecular delineation of the Greig cephalopolysyndactyly contiguous gene deletion syndrome and its distinction from acrocallosal syndrome. Am J Med Genet Part A 123A: 236–242. , , , , , , , , , , .
- 2000. An unusual FGFR1 mutation (fibroblast growth factor receptor 1 mutation) in a girl with nonsyndromic trigonocephaly. Cytogenet Cell Genet 91: 138–140. , , , .
- 1998. Syndromal and nonsyndromal primary trigonocephaly: Analysis of a series of 237 patients. Am J Med Genet 75: 211–215. , , , .
- 2001. Craniosynostosis and fetal exposure to sodium valproate. J Neurosurg 95: 778–782. , , , , , .
- 2009. Msx genes are important apoptosis effectors downstream of the Shh/Gli3 pathway in the limb. Dev Biol 331: 189–198. , , , .
- 1985. Schinzel acrocallosal syndrome: A variant example of Greig syndrome. Ann Genet 28: 239–240. , , , , .
- 2008. Patched1 haploinsufficiency increases adult bone mass and modulates Gli3 repressor activity. Dev Cell 14: 689–699. , , , , , , , , , , , , , , .
- 1969. The C syndrome of multiple congenital anomalies. Birth Defects 5: 161–166. , , , .
- 1999. The phenotypic spectrum of GLI3 morphopathies includes autosomal dominant preaxial polydactyly type-IV and postaxial polydactyly type-A/B; no phenotype prediction from the position of GLI3 mutations. Am J Hum Genet 65: 645–655. , , , , , , , , , , , .
- 1981. Familial trigonocephaly associated with short stature and developmental delay. Am J Dis Child 135: 711–712. , .
- 1982. Acrocallosal syndrome. Am J Med Genet 12: 201–203. .
- 1999. Fibroblast growth factor receptor mutational screening in newborns affected to metopic synostosis. Child's Nerv Syst 15: 389–394. , , , , , , , .
- 1990. How wide is the clinical spectrum of the acrocallosal syndrome? Report of a mild case. J Med Genet 27: 516–518. , , .
- 2006. Clinical report trigonocephaly in Muenke syndrome. Am J Med Genet Part A 140A: 2493–2494. , , .
- 2000. A Pro250Arg substitution in mouse Fgfr1 causes increased expression of Cbfa1 and premature fusion of calvarial sutures. Hum Mol Genet 9: 2001–2008. , , , , , .