There have been a few case reports from India identifying isolated sequence variants and deletions in MECP2 gene [4,5]. In a large Indian study, the common and novel sequence variations in MECP2 gene in Rett syndrome patients were identified, but large deletions in MECP2 gene were not studied [6]. In this study, we present the sequencing variations and deletional forms of MECP2 gene identified in Rett syndrome patients, classified into classical and atypical forms using the using the Rett syndrome revised diagnostic criteria and nomenclature, 2010 [3].

This cross-sectional study included girls with a clinical suspicion of Rett syndrome. Ethical clearance was obtained from Institutional Ethics committee of SGPGI and AIIMS. Informed written consent was taken from parents/guardians of all participants. All the girls included in the study had history of neuroregression (prior to five years of life) and postnatal microcephaly with normal MRI findings. The girls were divided into two groups. Group I included girls with a clinical diagnosis of classical or atypical Rett syndrome based on the Rett syndrome revised diagnostic criteria and nomenclature, 2010 [3]. Group II included all girls who did not fulfil all the criteria listed in the revised diagnostic criteria and nomenclature, 2010 but had neuroregression, postnatal microcephaly (with normal MRI) and other features suggestive of Rett syndrome were observed like autistic features, poor social interaction and eye contact, isolated play, inappropriate laughter, stereotypical movements, delayed speech etc. In both the groups, Sanger sequencing of MECP2 gene (two, three and four exons) was performed followed by Multiplex ligation dependent probe amplification (MLPA) (P0245- covering MECP2 exon 1 upstream and exons 3 and 4) in MECP2 mutation negative cases to look for partial and whole gene deletions of MECP2 gene. The Sanger sequencing was done using five sets of primers, one each for exon 2 and 3 and three overlapping fragments for exon 4; exon 1 was not sequenced [7].

A total of 36 girls, with postnatal microcephaly (with normal MRI) and history of neuroregression, prior to five years of age, were studied.

Group I consisted of 19 girls – 14 classical Rett syndrome and five atypical Rett syndrome girls, and Group II included 17 girls. These patients were tested as some features suggestive of Rett syndrome (other than neuroregression and postnatal microcephaly) were observed. The mean age at diagnosis in the three groups was 4 year, 3 year and 4.5 years, respectively. All the cases were of sporadic occurrence. The age of girls ranged from 1.6 to 10 years. The mean age of neuroregression was significantly higher in group II (3 years) as compared to group I (1.2 years).

The Sanger sequencing of MECP2 gene in group I identified sequence variations in 11 out of 14 classical Rett syndrome girls (78%) and one out of five atypical Rett syndrome girls (20%). In group I, MLPA detected large exonic deletions in two classical Rett syndrome girls (14%) (Web Fig. 1). The total yield of molecular testing (Sanger sequencing and MLPA was 92.8% (n=13) in classical Rett syndrome girls and 20% (n=1) in atypical Rett syndrome girls. The total yield of molecular testing in group I was around 74%. Molecular testing did not identify any MECP2 sequence variation or copy number change in group II (17 girls). Additional molecular testing were done and revealed pathogenic copy number variations in two out of 17 group II girls (11%).

Out of the 12 sequence variations identified in MECP2 gene, 11 were known pathogenic mutations and one was a novel mutation (Table I). The novel missense mutation- c.427G>A was identified in group I, in classical Rett syndrome girl. The missense variant was predicted as likely pathogenic by mutation prediction softwares (Mutation taster, SIFT (Sorting Intolerant From Tolerant) and Polyphen 2).

TABLE I Mutations Identified in MECP2 Gene

A known deletion was identified, c.1153_1190del 38 in group I classical Rett syndrome girl in the homozygous form. The finding was confirmed by repeat testing. Apart from being a homozygous deletion; it can be a heterozygous deletion with probable allele drop out or selective amplification of shorter fragment. Uniparental disomy was ruled out by SNP microarray. Karyotype was normal 46, XX. Whole exonic or whole deletion of MECP2 was ruled out by MLPA (P0245).

The novel mutation identified in this study most likely leads to loss of function of the MBD domain of MECP2 protein, which is essential to bind MECP2 to methylated DNA and hence, this might affect transcriptional regulation. Interestingly, all the sequence variations in MECP2 gene were located in the exon 4 and most were C>T transitions. No genotype phenotype correlation was observed as severity of clinical manifestations was same – being classical Rett phenotype in patients with different mutations: missense, and small and large deletions. The general restriction of Rett syndrome to females and the very low recurrence rate appears to be the result of a large proportion of these mutations arising via spontaneous deamination of 5-methylcytosine to thymine in the heavily methylated male germ cells [13].

Large exonic deletions are found in approximately 8-10% of Rett syndrome (including both classical and atypical forms) [14,15]. As compared to previous studies which looked at deletions of MECP2 gene with specific probe set (e.g. P015), we used the MLPA-P0245 probe set. Though it is commonly used and very useful, it is not specific for MECP2 gene. This probes set, with a limited number of probes for each specific chromosome region, is meant to designed routinely to detect the common deletion/duplication involving different chromosomes, in a cost effective way. We hereby put forward the importance of using the commonly used probe set P0245 for detection of large exonic deletions of MECP2 gene.

In Group I cases where molecular testing of MECP2 did no show mutation other genes like CDKL5 and FOXG, can be tested [16,17]. In group II patients with no MECP2 mutations, other causes of neuroregression and postnatal microcephaly should be sought for. In two of these 17 patients, cytogenetic microarray detected pathogenic copy number variations.

All the cases in the given study were sporadic. The MECP2 sequencing of mothers of girls with mutations in MECP2 was not done. Rett syndrome is of sporadic occurrence in 99 % of cases. In the remaining one percent there is either gonadal mosaicism in mother or there is highly skewed X-chromosome inactivation leading to no phenotype of Rett syndrome in the mother [18]. The risk of recurrence of Rett syndrome in next pregnancy is low, but it cannot be totally excluded as germline mosaicism has been documented in Rett syndrome. Hence, prenatal diagnosis can be provided in the next pregnancy.

As per our study, revised clinical diagnostic criteria are useful guide for clinical and molecular diagnosis of Rett syndrome. MECP2 mutations are more likely to be identified in classical than atypical Rett syndrome. Around 10% mutations in Rett syndrome patients are large exonic deletions of MECP2 gene and hence, MLPA should be done when mutation is not identified by Sanger sequencing of MECP2 gene. MLPA with the commonly used probe set, P0245, is found to be effective in detection of large deletions in girls with classical Rett syndrome.

Acknowledgements: Naveen Sankhyan, Pediatric Neurology, PGIMER, Chandigarh; Rekha Gupta, Medical Genetics, MGMCH, Jaipur: Viswanathan V, Pediatric Neurologist, Kanchi Kamakoti CHILDS Trust and Apollo Children’s Hospitals, Chennai; and Leena Deshpande, Dr Yewale Hospital, Mumbai.

Contributors: ML,KM,SRP: conception and design of the study, the acquisition of data, and the analysis and interpretation; AR, PS: conducting mutation analysis; KM, SRP, NG,MK: drafting and critical review.

Funding: Indian Council of Medical Research (63/8/2010/BMS); Competing interests: None stated.