Epileptic Encephalopathy, Early Infantile, 9

A number sign (#) is used with this entry because early infantile epileptic encephalopathy-9 (EIEE9), also known as epilepsy and mental retardation restricted to females (EFMR), is caused by mutation in the gene encoding protocadherin-19 (PCDH19; 300460) on chromosome Xq22.

Description

Epileptic encephalopathy-9 is an X-linked disorder characterized by seizure onset in infancy and mild to severe intellectual impairment. Autistic and psychiatric features have been reported in some individuals. The disorder affects heterozygous females only; transmitting males are unaffected (summary by Jamal et al., 2010).

For a general phenotypic description and a discussion of genetic heterogeneity of EIEE, see EIEE1 (308350).

Clinical Features

Juberg and Hellman (1971) reported a family in which 15 females, related either as sisters or first cousins through their fathers, had a grand mal convulsive disorder of early onset associated with mental retardation. In the sibship of the fathers, 6 of 9 males had affected daughters. Their mother and her mother were said to have a convulsive disorder. Female-limited autosomal dominant inheritance was proposed. In an update on this family, Fabisiak and Erickson (1990) reported that 4 brothers of affected females had had 5 unaffected daughters, while 4 affected women had had 1 unaffected and 4 affected daughters and 2 unaffected sons. This unusual transmission pattern was discussed in terms of germline imprinting, neuronal sexual differentiation, and the generally higher risk of seizures observed when the mother, rather than the father, is affected.

Ryan et al. (1997) restudied the family reported by Juberg and Hellman (1971) and Fabisiak and Erickson (1990). Additional births were recorded and molecular confirmation of paternity provided. Affected infants appeared normal until 4 to 18 months of age, when they began to have partial and generalized afebrile convulsions that gradually increased in frequency and were accompanied by developmental regression. In most females, the frequency of seizures declined dramatically by the age of 2 to 3 years, but cognitive development remained markedly impaired in most of the affected subjects. Some were profoundly mentally retarded and required chronic institutional care, and 3 women with only mild cognitive impairment had had disabling psychiatric symptoms. Only 2 of 20 affected persons had grossly normal intellectual and emotional function. Two severely affected women showed occasional stereotypic, purposeless hand movements, including flapping and wringing, with some apparent loss of fine motor abilities. Other features of Rett syndrome (312750), such as episodic hyperventilation, bruxism, and scoliosis, were lacking. Spasticity, dystonia, and hyperreflexia were also absent, and all patients had remained ambulatory.

Scheffer et al. (2008) reported 4 additional unrelated families, 2 of Australian and 2 of Israeli origin, with a phenotype similar to that reported by Juberg and Hellman (1971). There were 27 affected females with a mean age of seizure onset of 14 months (range 6 to 36 months). All had convulsive attacks at some stage, associated with fever in 17 (63%). Seizure types included tonic-clonic (26), tonic (4), partial (11), absence (5), atonic (3) and myoclonic (4). In most patients, seizures ceased at a mean age of 12 years. Developmental delay and intellectual disability was highly variable. Seven patients had normal development, whereas 4 had delay from birth, and 12 had developmental regression. However, most (67%) had intellectual disability or borderline intellect. Psychiatric features included autistic, obsessive, and aggressive features. Two individuals had episodes of schizophreniform psychosis in their early to mid-twenties that required hospitalization. A striking clinical feature of 5 obligate male carriers was the presence of obsessive traits and interests, as well as controlling, rigid, inflexible personalities.

Depienne et al. (2009) identified mutations in the PCDH19 gene (see, e.g., 300460.0006; 300460.0007) in 11 of 45 unrelated females with epileptic encephalopathy of infancy who were negative for mutations in the SCN1A gene (182389). Clinical features included seizure onset before 12 months of age, multiple seizure types, often associated with fever, mild to severe mental retardation with poor language development, and ataxia. Depienne et al. (2009) emphasized the phenotypic overlap with Dravet syndrome (607208), but noted that patients with PCDH19 mutations have comparatively fewer myoclonic and absence seizures, fewer episodes of status epilepticus, and less photosensitivity compared to those with Dravet syndrome.

Hynes et al. (2010) reported 2 sisters with early-onset seizures associated with a PCDH19 mutation (N557K; 300460.0008). One was a 25-year-old librarian who had onset of febrile and afebrile seizures at 18 months of age. At age 10 years, she was diagnosed with Asperger syndrome (608638) with poor visuospatial skills, but completed school and got a university degree. Her 23-year-old sister developed hemiclonic seizures at age 2 years. Her seizures recurred throughout childhood and were highly refractory until age 16 years. She was diagnosed with attention-deficit/hyperactivity disorder and showed borderline intelligence with good preservation of visuospatial function. Both sisters had EEG abnormalities. Hynes et al. (2010) also reported an unrelated 7-year-old Vietnamese girl with a de novo mutation in the PCDH19 gene. She presented at age 1 year with a seizure and later had clusters of afebrile generalized tonic-clonic seizures including status epilepticus. At age 4 years, she showed moderate intellectual disability and was diagnosed with autism spectrum disorder at age 7.

Marini et al. (2010) reported 13 unrelated girls with pathogenic mutations in the PCDH19 gene. The mean age at seizure onset was 8.5 months. Eight (62%) patients presented with febrile seizures, 4 (31%) with cluster of focal seizures, and 1 with de novo status epilepticus. Subsequent seizure types included afebrile tonic-clonic, febrile, and afebrile status epilepticus, absences, myoclonic, and focal seizures. Seven (54%) patients had a clinical diagnosis consistent with Dravet syndrome. Seizures were particularly frequent at onset in most patients, manifesting in clusters and becoming less frequent with age. Mental retardation was present in 11 of the 13 patients, ranging from mild (7; 64%) to moderate (1; 9%) to severe (3; 27%). Six patients (46%) had autistic features in association with mental retardation.

Jamal et al. (2010) reported 3 unrelated girls with EFMR due to de novo mutations in the PCDH19 gene. All had classic features of the disorder, with seizure onset in infancy, intellectual impairment, and autistic features. Seizures were difficult to control; all 3 patients were on multiple antiepileptic medications. Of note, each patient presented soon after routine childhood vaccination. Jamal et al. (2010) emphasized the lack of family history in these patients.

Inheritance

Although autosomal dominant inheritance was tentatively suggested in reports of the family by Juberg and Hellman (1971), with suggestions of special mechanisms to explain the limitation to females, evidence of X-linkage was presented by Ryan et al. (1997): transmission of the abnormal allele from 1 male to another was not observed in the pedigree. Because the abnormal phenotype was limited to females, however, recognition of male-to-male transmission would require that an affected woman pass the disorder to a female descendant through at least 2 generations of carrier males. In this kindred, the sons of carrier males had had 4 daughters, none of them affected. In addition, there was a large proportion of affected females among the daughters of transmitting males. Of the 17 daughters of transmitting males, 16 were affected, a gross distortion from the 1 to 1 segregation ratio expected of an autosomal dominant with complete penetrance. This observation suggested X-linked inheritance with disease expression limited to females, in whom penetrance was high but incomplete, as indicated by the 1 asymptomatic subject.

Ryan et al. (1997) considered 4 hypothetical models for X-linked dominant inheritance with male sparing in EFMR. The first model posited a functional homolog of the disease locus on the Y chromosome that is protected in males. It would be assumed that the X-linked EFMR locus is silenced in the usual fashion on the inactive X of females, and that the mutant allele is recessive to its normal, Y-linked homolog. The second model, 'metabolic interference' (Johnson, 1980), proposed a locus with a wildtype allele A and mutant allele A-prime such that homozygosity (or hemizygosity) for either allele has no phenotypic consequences, but the heterozygous state produces a harmful effect due to 'metabolic interference' between the protein products of the 2 alleles. In the third model, a disturbance in the X-inactivation process downstream from the expression of XIST was hypothesized. In this model, the female phenotype results from 'functional disomy' for the genes that fail to inactivate. This is similar to the situation in mentally retarded, dysmorphic females with a ring chromosome which remains active because it lacks a functional X-inactivation center. A fourth model suggested that the EFMR locus encodes a protein that is not required for development of the male brain. Alternatively, the male brain may be protected from the adverse effects of the EFMR mutation by fetal androgens. Page (1997) discussed these possibilities. He mused that 'if this X-Y hypothesis proves correct (it is highly speculative conjecture at present) then the X-to-Y transposition that occurred during hominid evolution was an act of prophylactic gene therapy, protecting males..., three million years later, from mental retardation and epilepsy.'

Dibbens et al. (2011) reported 2 unrelated families in which 2 sisters in each family had EFMR confirmed by genetic analysis. The mutations were not found in the parents' peripheral blood cells by routine analysis, but low levels of the mutation were detectable on follow-up screening in each of the mothers' cells, indicating somatic, and presumably, germline mosaicism for the mutations. The mother of 1 of the sister pairs had a history of refractory clusters of seizures in childhood, but had normal intellect. Dibbens et al. (2011) noted the implications of the findings for genetic counseling, since these mothers are at risk of having further affected daughters.

Pathogenesis

Dibbens et al. (2008) proposed a hypothesis to explain the clinically unaffected status of male carriers with hemizygous PCDH19 mutations. Since the PCDH19 gene is subject to X inactivation, hemizygous transmitting males likely have a homogeneous population of PCDH19-negative cells, whereas affected females are likely to be mosaics comprising PCDH19-negative and PCDH19-wildtype cells. This tissue mosaicism in females may scramble cell-cell communication, which then manifests clinically as EFMR. In males, the authors postulated functional rescue by a related but nonparalogous Y-chromosome protocadherin gene, PCDH11Y (400022), which is expressed in human brain. PCDH11Y has an X chromosome paralog, PCDH11X (300246), that shows strong sequence similarity. However, the differences in brain expression patterns between these 2 genes may account for differential ability to compensate for absence of PCDH19.

Depienne et al. (2009) reported 1 of 28 males with a similar phenotype to EFMR who was found to be somatic mosaic for a deletion encompassing the PCDH19 gene. The findings suggested a pathogenic mechanism by which 2 populations of cells, mutant and wildtype, are required for disease to occur, whereas a homogeneous cell population, either mutant or wildtype, does not lead to disease. A male patient with mosaic PCDH19 would have 2 populations of cells, like a heterozygous female. The authors termed this mechanism 'cellular interference.'

Mapping

Ryan et al. (1997) mapped the EFMR locus to chromosome Xq22, suggesting X-linked inheritance. Cytogenetic analysis of an EFMR female and a male carrier showed no abnormalities. The EFMR locus was proximal to the fragile X locus (FMR1; 309550) located at Xq27.3, and distal to XIST (314670) which lies in Xq13.2. Studies of X-chromosome inactivation in leukocytes of EFMR subjects by analysis of methylation patterns within the androgen receptor gene (AR; 313700) showed that the proportion of subjects with prominent skewing did not differ significantly from controls.

By combined linkage analysis of 4 unrelated families with EFMR, Scheffer et al. (2008) found linkage to Xq22 (maximum 2-point lod score of 3.5 at DXS990).

Molecular Genetics

In female patients with the syndrome of female-restricted epilepsy and mental retardation (EIEE9), Dibbens et al. (2008) identified 6 different mutations in the PCDH19 gene (e.g., 300460.0001-300460.0005). One of the families had been reported by Juberg and Hellman (1971).

Hynes et al. (2010) identified 2 different mutations in the PCDH19 gene (see, e.g., 300460.0008) in 2 (2.3%) of 86 female probands with epilepsy with or without mental retardation. One of the probands had been reported in the study of Dibbens et al. (2008). Hynes et al. (2010) did not identify PCDH19 mutations in a cohort of 42 females with Rett syndrome (RTT; 312750) who did not have identifiable RTT-associated mutations or in 57 females with autism spectrum disorder (209850), indicating that mutations in the PCDH19 gene are not associated with these disorders, despite some clinical overlap of these features in patients with PCDH19 mutations.

Marini et al. (2010) identified 13 different mutations in the PCDH19 gene in 13 (11%) of 117 female patients with febrile seizures and a wide spectrum of epilepsy phenotypes. Eleven mutations were novel. Mutations were inherited in 3 probands: 2 from apparently unaffected fathers and 1 from a mother with a history of generalized epilepsy in childhood. There were no apparent genotype/phenotype correlations or correlations with X-inactivation status.