Central Hypoventilation Syndrome, Congenital

A number sign (#) is used with this entry because congenital central hypoventilation syndrome (CCHS) is most commonly caused by heterozygous mutation in the PHOX2B gene (603851) on chromosome 4p13.

The disorder is also rarely caused by mutation in several other genes, including RET (164761), GDNF (600837), EDN3 (131242), and ASCL1 (100790).

Evidence suggests that PHOX2B is the major disease-causing gene in isolated and syndromic CCHS (Amiel et al., 2003; Weese-Mayer et al., 2003).

Haddad syndrome, in which CCHS is associated with Hirschsprung disease, is caused by heterozygous mutation in the ASCL1 gene (100790).

Description

Idiopathic congenital central hypoventilation syndrome, also known as 'Ondine's curse' (Deonna et al., 1974), is a rare disorder characterized by abnormal control of respiration in the absence of neuromuscular, lung or cardiac disease, or an identifiable brainstem lesion. Patients breathe normally while awake, but hypoventilate with normal respiratory rates and shallow breathing during sleep; more severely affected patients hypoventilate both awake and asleep. These patients typically present in the first hours of life with cyanosis and increased carbon dioxide during sleep. A deficiency in autonomic control of respiration results in inadequate or negligible ventilatory and arousal responses to hypercapnia and hypoxemia (reviewed by Weese-Mayer et al., 1999).

Congenital central hypoventilation syndrome has been associated with several disorders classified as neurocristopathies, that is, aberrant phenotypes arising from a defect of migration or differentiation of neural crest cells. These include neuroblastoma (Haddad et al., 1978), ganglioneuroma (Swaminathan et al., 1989), and most frequently Hirschsprung disease (HSCR) which appears in 16% of CCHS patients. The association of CCHS and HSCR is referred to as Haddad syndrome.

Congenital central hypoventilation can be a feature of other developmental disorders, such as those caused by mutation in the MECP2 gene (300005).

Clinical Features

The syndrome of congenital central hypoventilation was first reported by Mellins et al. (1970). Cases were reported also by Deonna et al. (1974), Yasuma et al. (1987), O'Dell et al. (1987), Oren et al. (1987), and Weese-Mayer et al. (1988).

Haddad et al. (1978) described 3 patients of whom 2 were sisters. All 3 died in the first few months of life. They showed a combination of Ondine curse (failure of autonomic control of ventilation during sleep) and Hirschsprung disease (megacolon) (142623). Esophageal motility and control of heart rate were also markedly reduced. Neuropathologic studies postmortem showed no anatomic defect. The authors postulated a developmental defect of serotonergic neurons. Stern et al. (1980) also described a case in a male infant.

Minutillo et al. (1989) described a patient and pointed out possibly distinctive facial features (antimongoloid slanting eyes, triangular mouth, small nose, and low-set, posteriorly rotated ears). Familial depression of ventilatory response to hypoxia and hypercapnia (267480) and familial lethal sleep apnea (207720) are disorders of possibly related nature.

Folgering et al. (1979) found absence of the arcuate nucleus at autopsy in an infant with congenital central hypoventilation syndrome. Decreased muscarinic receptor binding in the arcuate nucleus has been implicated in the sudden infant death syndrome (SIDS; 272120) by Kinney et al. (1995).

Weese-Mayer et al. (1999) prepared a comprehensive statement concerning CCHS for the American Thoracic Society. They stated that approximately 100 cases had been reported. They estimated that 160 to 180 children with CCHS are living worldwide, but considered these numbers to be an underestimate. Extensive information was given concerning long-term comprehensive management. In addition to the clinical features of alveolar hypoventilation, patients with CCHS often manifest a spectrum of clinical symptoms reflecting dysfunction of the autonomic nervous system. These include Hirschsprung disease and/or severe constipation, feeding difficulty, decreased perception of discomfort, pupillary abnormalities, decreased perception of anxiety, profuse sweating, and decreased basal body temperature.

A heterogeneous group of patients with late-onset central hypoventilation syndrome (LO-CHS) was described by Katz et al. (2000).

Antic et al. (2006) reported 5 unrelated patients with onset of central hypoventilation after age 21 years. All survived into adulthood without artificial ventilatory support until the time of diagnosis, although all reported respiratory symptoms since childhood when a detailed history was taken. Presentations included unexpected hypoxemia during respiratory infection, sleep abnormalities, and seizures. Some of the patients had evidence of chronic symptoms, such as hypercarbia, polycythemia, and right heart changes. Two patients had cognitive impairment, which may have resulted from chronic hypoxemia. Molecular analysis identified an expansion of +5 alanine repeats in exon 2 of the PHOX2B gene (603851.0001). Some of the patients' children inherited the same expansion but showed onset within the first year of life, indicating incomplete penetrance associated with this relatively short expansion.

Inheritance

CCHS caused by mutation in the PHOX2B gene is transmitted in an autosomal dominant pattern with incomplete penetrance (summary by Parodi et al., 2010).

Rutishauser and Feldges (1977) reported the disorder in a mother and daughter. Khalifa et al. (1988) reported affected monozygotic twins.

Hamilton and Bodurtha (1989) described black brother and sister with the association of Hirschsprung disease and CCHS. Curiously, they were half sibs (same father, different mothers). Since there was no known relationship of the mothers, the possibility of autosomal dominant inheritance with reduced penetrance (or paternal gonadal mosaicism) might be considered.

From segregation analyses, Weese-Mayer et al. (1993) concluded that multifactorial and major-locus models are almost equally likely in CCHS.

Hypothesizing that CCHS is the most severe manifestation of general autonomic nervous system dysfunction (ANSD), Marazita et al. (2001) investigated the genetics of ANSD. They performed major locus segregation analysis utilizing regressive models in 52 probands with CCHS and 52 age-, race-, and gender-matched controls. CCHS probands were assumed to be affected; controls and relatives were designated as affected if they had 2 or more relevant symptoms. They found that case families were consistent with transmission of a major effect; control families were not. In the total data set, the best-fitting model was considered to be codominant mendelian inheritance of a major gene for ANSD.

In a case-control family study of autonomic nervous system dysfunction in idiopathic CCHS, Weese-Mayer et al. (2001) found that, under each of 2 arbitrary definitions of ANSD affection, 16% of CCHS sibs had the ANSD phenotype with 2 or more symptoms, compared to 4% of control sibs. Aunts and uncles of CCHS cases were also significantly more likely to have 2 or more ANSD symptoms than were aunts and uncles of the controls.

Silvestri et al. (2002) reported offspring born to 4 women with idiopathic CCHS. One of the children was diagnosed with CCHS, 1 had recurrent and apparently life-threatening events, 1 was born prematurely with severe chronic lung disease and diminished ventilatory responses to carbon dioxide, and 1 was apparently healthy with no clinical manifestations suggestive of disordered respiratory control at 25 months of age. Thus, 2 and possibly 3 of these patients illustrated transmission of altered respiratory control from a CCHS mother.

Parodi et al. (2010) performed polysomnography in 3 apparently healthy parents of a child with CCHS due to an expanded alanine allele in the PHOX2B gene. One mother, with a 5-alanine expansion, was found to have significant sleep breathing control anomalies. The other 2 parents had normal sleep breathing patterns and were found to be somatic mosaic for 6-alanine expansions. These findings confirmed variable expressivity and incomplete penetrance of PHOX2B mutations, and suggested that polysomnography and assessment of the degree of somatic mosaicism should be conducted in asymptomatic mutation carriers.

Among 45 unrelated Japanese families in which a child had congenital central hypoventilation syndrome, Meguro et al. (2012) found that 1 patient (2%) inherited a 5-alanine expansion mutation from a parent with late-onset central hypoventilation syndrome. In this family, the maternal grandmother and maternal uncle, who carried the same expansion, also had late-onset central hypoventilation syndrome. In addition, 9 patients (20%) inherited a 5- to 7-alanine expansion mutation from apparently asymptomatic parents with somatic mosaicism. The level of mosaicism in these parents ranged from 5 to 62%, and 3 patients had affected sibs. Meguro et al. (2012) postulated that the expansion in somatic mosaicism results from a replication error, rather than unequal crossing over, since contracted alleles were not identified. These findings had implications for genetic counseling.

Molecular Genetics

Mutation in the PHOX2B Gene

In mice, the development of reflex circuits of the autonomic nervous system is dependent on the paired-like homeobox gene Phox2b (603851). For that reason, Amiel et al. (2003) investigated the human ortholog, PHOX2B, as a candidate gene in CCHS. They found heterozygous de novo mutations in PHOX2B in 18 of 29 affected individuals. Most mutations consisted of 5 to 9 alanine expansions within a 20-residue polyalanine tract probably resulting from nonhomologous recombination (603851.0001). They showed that PHOX2B is expressed in both the central and the peripheral autonomic nervous system during human embryonic development.

Sasaki et al. (2003) studied 7 patients with isolated CCHS and 3 CCHS patients with Hirschsprung disease. In 4 patients they detected polyalanine expansions in the PHOX2B gene and in 1 patient a novel frameshift mutation in PHOX2B. They could not reject the possibility that mutations in the RET (164761), GDNF, PHOX2A, and HASH1 (100790) genes may also be involved in the pathogenesis of CCHS.

In 65 of 67 CCHS probands (97%), Weese-Mayer et al. (2003) found heterozygosity for the exon 3 polyalanine expansion mutation in PHOX2B. There was an association between repeat mutation length and severity of the CCHS/ANSD phenotype. Of the 2 probands who did not carry the expansion mutation, one had a nonsense mutation in exon 3 that truncated the protein and the other had no mutation in PHOX2B but had a previously reported EDN3 frameshift point mutation.

Matera et al. (2004) screened the PHOX2B gene in 27 patients with CCHS, including 3 with associated Hirschsprung disease and 3 with late-onset CCHS, and identified 3 heterozygous frameshift mutations and 22 polyalanine expansions ranging from 5 to 13 residues. The authors noted that phenotype severity increased with increasing polyalanine expansion size. Polyalanine triplet expansions were also detected in the affected sibs of 2 familial cases and in 2 asymptomatic parents. Matera et al. (2004) concluded that their findings demonstrated autosomal dominant inheritance with reduced penetrance.

Trochet et al. (2005) reported the clinical and molecular assessments of a cohort of 188 probands with CCHS, either isolated or associated with Hirschsprung disease and/or tumors of the sympathetic nervous system (TSNS). The mutation detection rate was 92.6% (174/188), and the most prevalent mutation was an in-frame duplication leading to an expansion of +5 to +13 alanines in the 20-alanine stretch of the carboxy terminal of the PHOX2B protein (603851.0001). Analysis of genotype-phenotype interactions strongly supported the contention that patients with CCHS who develop malignant TSNS harbor either a missense or a frameshift heterozygous mutation of the PHOX2B gene. These data pointed to another link between congenital malformations and tumor predisposition when a master gene in development is mutated.

De Pontual et al. (2006) genotyped the RET locus in 143 patients with CCHS who were known to have mutations in the PHOX2B gene. The odds ratios of HSCR for patients heterozygous and homozygous for the nonsyndromic HSCR-predisposing RET haplotype (ATA), which contained the hypomorphic intron 1 allele (164761.0050), were 2.39 and 4.74, respectively; 16 patients with a PHOX2B alanine expansion and no predisposing RET haplotype also had HSCR. De Pontual et al. (2006) concluded that there are both RET-dependent and RET-independent HSCR cases and suggested that at least 1 more modifier gene must be involved.

Mutations in Other Genes

To test the hypothesis that CCHS and HSCR share a common molecular pathology, Bolk et al. (1996) conducted mutation analysis of potential candidate genes in pediatric patients with CCHS and HSCR or isolated CCHS. They identified a frameshift mutation in the gene encoding endothelin-3 (131242.0003).

Bolk et al. (1996) used SSCP analysis to study mutations of the RET gene in 14 patients with CCHS. All detected nucleotide changes in the RET gene were classified as polymorphic variants. Cytogenetic study did not reveal chromosomal abnormalities (except a familial inv(2)(p11.2q13) in 1 case). Amiel et al. (1998) reviewed the mutations in the RET-GDNF signaling pathway in Ondine curse. They failed to detect EDNRB or EDN3 mutations in their series. By contrast, screening the coding sequence of the RET (164761) and GDNF (600837) genes in 5 unrelated cases of isolated CCHS and in 2 cases of CCHS-HSCR association, they found mutations in children with isolated CCHS (1 in 7) and the CCHS-HSCR association (1 in 7), respectively. Identification of mutations in RET-GDNF pathway and the endothelin pathway in Ondine curse shed light on the genetic bases of this life-threatening condition and supports the view that CCHS is a neural crest cell disorder. Nevertheless, mutations had been reported in a minority of patients tested to that time. Amiel et al. (1998) stated that the involvement of at least 3 genes belonging to distinct signaling pathways, the incomplete penetrance of the mutation in carrier parents, and the variable expression of the respiratory control defect observed in Ret -/- homozygous mice exposed to hypercapnia (Burton et al., 1997) support the view that an interactive polygenic inheritance is involved in Ondine curse.

In 2 patients with CCHS and 1 with Haddad syndrome, de Pontual et al. (2003) identified mutations in the ASCL1 gene (100790.0001-100790.0003).

Associations Pending Confirmation

Weese-Mayer et al. (2002) studied 19 children with CCHS, 5 of whom also had Hirschsprung disease, for mutations in the brain-derived neurotrophic factor gene. They identified a mutation (113505.0001) in the BDNF gene in 1 child with isolated CCHS as well as in his father, who did not have CCHS but had symptoms of postural hypotension and vasovagal syncope.

Animal Model

Shirasawa et al. (2000) disrupted the Rnx (604640) locus in mouse embryonic stem cells. The phenotype of Rnx-deficient mice resembled that of congenital central hypoventilation.

De Pontual et al. (2006) generated doubly heterozygous (Phox2b +/- and Ret +/-) mice and observed that the intestine of the mutant mice was indistinguishable from wildtype littermates; the authors concluded that a greater than 50% loss of function for each gene must be necessary in the mouse for an enteric phenotype to occur.