OBJECTIVERett syndrome (RTT) is a severe childhood neurodevelopmental disorder mainly affecting females. The pathogenic gene is located at Xq28, which codes for the methyl-CpG-binding protein 2. MECP2 gene is affected by X chromosome inactivation (XCI). The different XCI patterns of females could affect the expression ratios of pathogenic gene, causing changes in clinical symptoms. In order to understand the XCI patterns in RTT patients and the relationship between XCI pattern, genotype and phenotype, the XCI patterns in patients with RTT and their mothers, the parental origin of the priority inactive X chromosome in RTT, and the relations of XCI patterns with genotype and phenotype in RTT cases were analyzed.

METHODSGenomic DNA was extracted from peripheral blood of 55 cases with RTT (52 with MECP2 mutations, 3 without mutations), 53 mothers of RTT cases and 48 normal female controls. DNA was digested with methylation sensitive restriction endonuclease Hpa II. Then the undigested and digested DNAs were amplified via PCR for the first exon of human androgen receptor (AR) gene. PCR products were analyzed by Genescan.

RESULTSThe heterozygotic rates of AR gene were 82%, 77% and 83% in RTT patients, mothers and controls, respectively. XCI distribution pattern of RTT was different from that of the mothers and control, P < 0.05. More mothers and controls than RTT patients were in the area of XCI 50:50 - 59:41. The differences between them were statistically significant (P < 0.05). No significant difference in XCI distribution patterns between mothers and the control groups was found (P > 0.05). Non-random XCI rates in the areas of XCI >or= 65:35 and >or= 80:20 were 53.35% and 17.8%, respectively, in RTT patients, compared with the mothers group (36.6%, 7.3%) and control group (35%, 10%), it was higher in RTT patients, but the difference was not statistically significant (P > 0.05). In 18 of 21 cases with XCI >or= 65:35, the priority inactive X chromosome was of paternal origin (85.7%). Variable XCI patterns were observed in the same gene mutation patients. The highly skewed XCI as well as the random XCI were found in patients with mild, severe and typical phenotype. The rate of highly skewed XCI in atypical patients was higher than that in typical RTT patients. The rate of highly skewed XCI in T158M was higher than the other type mutations. No highly skewed XCI was observed in cases with R133C mutation.

CONCLUSIONThe XCI distribution pattern of RTT patients was different from that of RTT mother and control groups. There was no significant difference in XCI distribution patterns between mothers and the control groups. It was not a main genetic pattern in RTT that mothers as the carriers to transmit the pathogenic gene to the patients. Non-random XCI was not the main XCI pattern in RTT patients. The priority inactive X chromosome was mainly of paternal origin. XCI could modify the clinical phenotype of RTT, but had limitations in explaining all the phenotypes manifested in RTT cases.