Genetics of Craniosynostosis

Published on 13/03/2015 by admin

Last modified 13/03/2015

Print this page

This article have been viewed 1224 times

CHAPTER 181 Genetics of Craniosynostosis

Craniosynostosis, or the premature fusion of calvarial sutures, occurs in approximately 1 in 2500 live births ¹ and has traditionally been classified as being either nonsyndromic or syndromic based on phenotypic descriptions. The children considered nonsyndromic often had single-suture synostosis without other abnormalities, whereas those with syndromic disorders were more likely to have multiple sutures fused, harbor other skeletal anomalies, and have a strong family history. Although phenotypic classification has been the “gold standard” for some time, there has been a recent movement toward genotypic classification as the genetic basis of craniosynostosis has slowly been revealed. This movement toward defining the genetic basis of the disease started in 1993 with discovery of the first mutation identified in synostosis, an MSX2 gene mutation in Boston-type synostosis.

As the genetic story unfolds, it is likely that all forms of synostosis will be revealed to have a genetic origin, and therefore the traditional “syndromic” cases will probably be the ones in which the affected gene has more protean manifestations. Investigations have revealed that craniosynostosis can be caused by mutations in multiple genes, including those for fibroblast growth factor receptors (FGFR1 to FGFR3), transcription factors (TWIST, MSX2, RUNX2), transmembrane ligands for Eph receptor tyrosine kinases (EFNB1), structural proteins (FBN1), cell regulatory proteins (TGFBR1 and TGFBR2), proteins involved in electron transfer from the reduced form of nicotinamide adenine dinucleotide phosphate (POR), small guanosine triphosphatases (RAB23), and a RecQ helicase family protein (RECQL4).²^–¹⁵ Some genes (RBP4, GPC3, C1QTNF3, IL11RA, PTN, POSTN, WIF1, ANXA3, CYFIP2) have also been identified as potentially playing a role in premature calvarial fusion, but these genes must still be studied in more detail.¹⁶ The objective of this review is to outline the different types of syndromic and nonsyndromic craniosynostosis and summarize the major genetic mechanisms behind their pathogenesis.

Syndromic Craniosynostosis

Despite the genetic underpinnings of craniosynostosis, it is likely that most clinicians will continue to use the traditional syndromic and phenotypic nomenclature for the majority of patients. The most common forms of syndromic craniosynostosis include Muenke’s syndrome, Saethre-Chotzen syndrome, Crouzon’s syndrome, Pfeiffer’s syndrome, Apert’s syndrome, and Crouzon’s syndrome with acanthosis nigricans. It is probable that as time goes on, we will see a combination of the phenotypic and genotypic nomenclature, for instance, saying either Pfeiffer’s syndrome FGFR1 or Pfeiffer’s syndrome FGFR2 to convey a true sense of the disease and its underlying etiology.

Inheritance Patterns of Syndromic Synostosis

Many of the syndromic conditions follow a familial inheritance pattern, although spontaneous mutations are possible. The inheritance pattern in familial cases tends to be autosomal dominant (with the exception of craniofrontonasal syndrome, which is X linked). Therefore, there is a 50% transmission rate to offspring, with variable penetrance. Perhaps more interesting is that in sporadic cases, many of the syndromes have been found to be associated with advanced paternal age. This indicates a certain fragility of the male genomic substrate.

Homology in Genetic Mutation

As our understanding of the processes improve, it is becoming clear that there are broad similarities in the genetic basis of the syndromes. Certain genes, especially those involved with the fibroblast growth factor receptor (FGFR), are able to broadly affect the processes of bone growth and development. This can explain the other common features often seen in some of these syndromes, including bony anomalies of the midface, base of the skull, and digits. Two genes, TWIST, which encodes a transcription factor, and FGFR, appear to be responsible for the vast majority of syndromic synostoses. Two other genes, MSX1, which encodes a transcription factor, and FBN1, which encodes the structural protein fibrillin, are lesser players in the genetic basis of synostosis.

The two major genetic mechanisms are loss-of-function mutations and gain-of-function mutations. It is theorized that TWIST is normally an upstream inhibitor in the process of suture formation, and a loss-of-function mutation in the TWIST gene leads to lack of inhibition of bone formation and resultant sutural fusion. In contrast, FGFRs are downstream of TWIST, and overexpression of them will lead to gain of function and sutural fusion. Interestingly, different genes at different chromosome locations may give rise to the same syndromic disorder, thus indicating that different genes can result in a similar end point. In addition, mutations in the same gene can lead to different disorders. Clearly, there remains much we do not understand about the genetic basis of disease.

The homology of mutations across the syndromes seems to be well preserved. Muenke’s syndrome (FGFR3 mediated), Apert’s syndrome (FGFR2 mediated), and Pfeiffer’s syndrome type I (FGFR1 mediated) have mutations in the same immunoglobulin (Ig) II-III linker regions on their respective chromosomes, all leading to altered affinity for the FGFR ligand. In Crouzon’s syndrome, Pfeiffer’s syndrome types II and III, Jackson-Weiss syndrome, and select cases of Saethre-Chotzen syndrome, the mutations are in FGFR2 in the Ig III domain and lead to constitutive activation of the receptors. In both cases, the result is overactivation of FGFR. Therefore, either overactivation of FGFRs or loss of function of TWIST can lead to a syndrome-type synostosis. The numerous mutations that have been implicated in these syndromes are detailed in the following sections.

Muenke’s Syndrome

Muenke’s syndrome has been one of the most instructive genetic conditions in the field of synostosis because of its variable clinical features and its seemingly high genetic prevalence in patients with craniosynostosis. From a cranial perspective, it is often associated with unicoronal or bicoronal craniosynostosis and may be associated with a range of other abnormalities, including brachydactyly, thimble-like middle phalanges, coned epiphyses, carpal and tarsal fusions, sensorineural hearing loss, and developmental delay,¹⁷ although perhaps the most striking finding is the phenotypic variability of the syndrome. It is believed that many cases of “nonsyndromic” synostosis, including many previously thought to be isolated unilateral coronal synostosis, are actually variants of Muenke’s syndrome. It is inherited in an autosomal dominant fashion and is caused by a point mutation in Pro250Arg in the Ig II-III linker region of the FGFR3 gene on chromosome 4p16.3.¹⁷ This mutation is highly associated with familial cases of Muenke’s syndrome, and affected females have a more severe phenotype.^18,¹⁹ In patients with sporadic Muenke’s syndrome, it has been suggested that advanced paternal age might result in de novo Pro250Arg mutations.²⁰ Numerous studies of patients with Muenke’s syndrome and the Pro250Arg mutation have shown that the clinical findings are highly variable.^18,²¹^–²⁶ In fact, some patients with this mutation have been found to have congenital bilateral sensorineural deafness without symptomatic craniosynostosis.²⁷ Because of the high variability in phenotype and the association of coronal synostosis with other craniosynostosis syndromes, genetic testing for the Pro250Arg mutation in FGFR3 is necessary to establish the diagnosis of Muenke’s syndrome.¹⁷

Apert’s Syndrome

Apert’s syndrome is characterized by coronal synostosis, severe syndactyly in the fingers and toes, symphalangism, radiohumeral fusion, and mental retardation.^28,²⁹ Other clinical features include midface hypoplasia, a short nose with a depressed nasal bridge, hypertelorism, proptosis, a trapezoidal-shaped mouth, and a variety of dermatologic manifestations, including severe acne.³⁰^–³² As with the other craniosynostosis syndromes, there is a wide range of severity of deformities.³³ The syndrome follows autosomal dominant inheritance and is caused by a number of different mutations in FGFR2 on chromosome 10q26, including two missense mutations (Ser252Trp, Pro253Arg) and two Alu insertions.³⁴^–³⁶ Interestingly, the Pro253Arg mutation in Apert’s syndrome is in the homologous Ig II-III linker region of FGFR2, similar to the Pro250Arg mutation in FGFR3 in Muenke’s syndrome. Patients with Ser252Trp tend to have a high frequency of cranial and cleft palate anomalies, but less syndactyly than in those with the Pro253Arg genotype. Sporadic cases have been associated with advanced paternal age.^37,³⁸

Pfeiffer’s Syndrome

Pfeiffer’s syndrome is characterized by coronal synostosis with or without premature fusion of other calvarial sutures, broad fingers and toes, and partial syndactyly of the fingers and toes.³⁹ Patients may also have maxillary hypoplasia, a small nose with a low nasal bridge, hypertelorism, shallow orbits, proptosis, strabismus, limb malformation, and radiohumeral synostosis.^39,⁴⁰ In 1993, Cohen suggested categorizing Pfeiffer’s syndrome into three different types based on the severity of deformities, ranging from type I (classic Pfeiffer’s syndrome, a mild entity with autosomal dominant inheritance) to types II and III (more severe, with early death and sporadic inheritance).⁴¹ Similar to Muenke’s and Apert’s syndromes, mutation of Pro252Arg in the Ig II-III linker region of FGFR1 on chromosome 8p11.2-p11 has been identified,⁴² thus suggesting that all three genes for FGFR play a similar role in suture development and lead to distinct craniosynostosis syndromes.⁴³ In fact, each syndrome is associated with altered ligand-binding specificity for the FGFs. Genetic analysis has revealed that FGFR1 mutations cause type I Pfeiffer’s syndrome whereas FGFR2 mutations have been associated with all three types of Pfeiffer’s syndrome.^36,³⁹ Clearly, those associated with FGFR2 mutations have a much more extreme phenotypic expression. As a result, it has recently been proposed that Pfeiffer’s syndrome caused by FGFR2 mutations should not be classified as Pfeiffer’s syndrome but rather as Crouzon’s syndrome.⁴³ This is obviously at odds with traditional nomenclature inasmuch as limb deformities have generally been considered pathognomonic of Pfeiffer’s syndrome.^39,⁴⁴

Crouzon’s Syndrome

Crouzon’s syndrome is an autosomal dominant disorder characterized by the classic triad of coronal synostosis, midfacial hypoplasia, and exopthalmos.^45,⁴⁶ Other manifestations may include involvement of other calvarial sutures, brachycephaly, hypertelorism, Chiari I malformation, hydrocephalus, and mental retardation.⁴³ Numerous missense mutations in the Ig III domain of the FGFR2 gene have been implicated, many of which involve the gain or loss of a cysteine residue.^4,⁴⁷^–⁵³ The effect of the mutation is constitutive activation of FGFR2, which is believed to be due to the unpaired cysteine residues forming intermolecular disulfide bonds.^54,⁵⁵ In cases in which the mutation does not involve a cysteine residue but still results in constitutive activation of FGFR2, it is believed that the mutation also plays an important role in increasing the amount of intermolecular disulfide bonding by disrupting the normal intramolecular disulfide bonds.⁵⁶ In addition to the missense mutations, one study reported a patient suffering from Crouzon’s syndrome with an in-frame insertion.⁵⁷ Sporadic cases of Crouzon’s syndrome have been associated with germinal mosaicism and advanced paternal age.^46,⁵⁸^–⁶¹ Chang and coworkers thus suggest that genetic testing for Crouzon’s syndrome should include sequencing of the Ig III region of FGFR2.⁵¹ In another article, Preston and associates suggested that mutations in PAX2 (on chromosome 10q25) might also be involved in Crouzon’s syndrome.⁵³ It is important to note that Crouzon’s syndrome with acanthosis nigricans is a different disorder; it is caused by an A391E mutation in FGFR3.^62,⁶³