Mutation analysis of the MECP2 gene in patients with Rett syndrome

To the Editor:

Rett syndrome (RTT) is a childhood neurodevelopmental disorder almost exclusively affecting females. Classically, it is characterized by normal development for the first 6 months followed by developmental regression with loss of acquired skills, especially purposeful hand movements. Many girls develop characteristic “hand washing” or “hand wringing” stereotypes and abnormal breathing patterns with periods of hyperventilation and apnea. The prevalence of the disease ranges from 1 in 10,000 to 1 in 20,000 females [Hagberg, 1985; Kozinetz et al., 1995]. Besides the classic form of RTT, a number of variants have been described: the infantile seizure onset type, the congenital form, the “forme fruste”, the preserved speech variant (PSV), and the late childhood regression form [Hagberg and Skjeldal, 1994].

A gene for RTT (MECP2) has been identified on the X chromosome in q28 [Amir et al., 1999] that binds to single methylated CpG base pairs throughout the genome and “silences” other genes [Nan et al., 1998].

To date, MECP2 mutations have been documented in up to 72% [Hoffbuhr et al., 2001] of sporadic and ∼50% of familial cases [Amir and Zoghbi, 2000]. Furthermore, screening of RTT patients for MECP2 mutations has shown random X inactivation in over 90% of RTT females [Hoffbuhr et al., 2000], the paternal origin of de novo MECP2 mutations [Kondo et al., 2000] and the maternal inheritance of mutations in familial forms, with a skewed X-inactivation pattern in circulating white blood cells of healthy female carriers [Schanen et al., 1998].

To provide further insights into the distribution and spectrum of mutations at the MECP2 locus, 21 patients from Sicily were analyzed: 14 with classical RTT syndrome and seven with variant forms (Table I). All these patients were diagnosed according to Hagberg and Skjeldal [1994] criteria.

The PCR products for exons 2, 3, and 4 were sequenced directly and mutations in MECP2 were found in 10 of 14 classical RTT patients (71.4%) and in three of seven cases with a variant form (42.8%). The spectrum of the mutations is summarized in Table II.

Three new variants were also detected: the substitutions K12N (36G→C), 1554C→G, and 1626C→G in the 3′UTR region. These nucleotide changes were also found in the patients' fathers. However, these variations were not found in more than 200 normal chromosomes, so they represent very rare polymorphisms.

Two novel mutations were found in a patient with classical RTT and in another patient with a possible congenital form: P376S (1126 C→T) and S388P (1162 C→T), respectively. These mutations have not been described previously and were not found in more than 100 normal chromosomes, so we hypothesize that these are disease-causing mutations and not polymorphisms.

Direct sequencing of DNA samples from both parents of 11 individuals with a MECP2 mutation, revealed that all but one (P376S) were de novo mutations.

The novel missense mutation (S388P) was found in a patient with a possible congenital form. She has had severe psychomotor delay since the first months of life, and infantile spasms from the age of 5 months. Moreover, at age 5 years she suddenly presented with renal failure.

The other novel mutation P376S was found in a nine-year-old girl. Her clinical history included developmental delay and seizures with onset at 10 months of age. She shows ataxia, lower limb hypertonia, hyperventilation, and scoliosis, and she is no longer able to walk. This missense mutation was inherited from her asymptomatic mother.

In agreement with previous studies, no clear correlation could be detected when the type of mutation was compared with the clinical features of the patient, even if some indication for the difference in the phenotype between missense and truncating mutations was reported [Wan et al., 1999; De Bona et al., 2000]. Amir et al. [2000] suggested that the X-inactivation pattern is one important modulator of the phenotype in RTT patients. In any condition where there is at least a partial protein product from the mutated allele, the severity of disease may depend almost entirely on the degree of X inactivation rather than on the mutation itself [Hampson et al., 2000].

To determine the XCI contribution in phenotypic variability of RTT patients, we evaluated XCI patterns in the carrier mother and in 13 patients with MECP2 mutations using the AR methylation assay [Pegoraro et al., 1994]. Ten out of 13 were informative for the locus: seven of these had a random pattern of XCI, two patients had moderately skewed XCI and only one patient (R1) showed a non-random XCI pattern (85:15). The carrier mother has a completely biased XCI pattern (90:10). This finding is consistent with the fact that she is clinically normal.

In conclusion, these results confirm that MECP2 is the major locus in classical RTT and that mutations in the MECP2 gene can be found in the majority of Sicilian RTT patients.

The two novel mutations described in this study lie in the region of the gene encoding the C-terminal of the protein, which enhances binding of the MECP2 to nucleosomal DNA [Chandler et al., 1999]. It remains to be shown in which way this kind of mutation and its location are involved: most likely it decreases interactions between MECP2 and components of the nucleosome. However, in eight out of 21 subjects of the present study no mutations were found: four of them fulfill the criteria for classical RTT, indicating that RTT syndrome could be caused by other defects that affect the same system of gene silencing.