Angelman syndrome (AS) is a neurodevelopmental disorder characterized by severe DD/ID, gait ataxia, microcephaly, seizures, and a happy demeanor that includes frequent laughing, smiling, and excitability [15]. AS is caused by deficient expression of the maternally inherited copy of the UBE3A (ubiquitin-protein ligase E3A) gene which can result from one of several different genetic abnormalities involving the proximal part of the long arm of chromosome 15 (15q11–q13). This two megabase (Mb) region of chromosome 15 contains multiple imprinted genes, meaning that their expression is dependent on the parent of origin. Within this region of chromosome 15, the MKRN3, MAGEL2, NECDIN, and SNURF–SNRPN genes, as well as a cluster of C/D box small nucleolar RNAs (snoRNAs) are paternally expressed, while the UBE3A and ATP10C genes are maternally expressed. Imprinting of genes in this domain is coordinately controlled by a bipartite imprinting center (IC) which overlaps the SNRPN promoter and extends approximately 35 kb upstream. Flanking this imprinted region are several low-copy repeats that predispose the region to chromosomal rearrangement by unequal crossing over.

Imprinted UBE3A expression is tissue specific, displaying predominantly maternal expression in the human fetal brain and adult frontal cortex but biparental expression in other tissues. UBE3A encodes a protein that is involved in the ubiquitination pathway which targets certain proteins for degradation [16]. Four known genetic mechanisms cause loss of UBE3A expression, and explain 85–90 % of AS cases [17]. A 4 Mb deletion of the maternal chromosome 15q11–q13 region occurs sporadically in 65–70 % of cases and is due to unequal crossing over. Paternal uniparental disomy (UPD) of chromosome 15 is detected in about 7 % of cases and is likely to be postzygotic in origin. Approximately 3 % of AS cases involve an imprinting defect caused by microdeletions of the IC and result in lack of expression of the maternally inherited UBE3A gene in the brain. Point mutations within UBE3A (mostly truncating mutations) are found in approximately 11 % of cases [17]. Currently, approximately 10 % of patients with a clinical diagnosis of AS have no identifiable chromosomal or molecular abnormality, most likely due to an undetected abnormality affecting the UBE3A gene or a mutation in another gene within the ubiquitination pathway.

While the AS phenotype results from loss of expression of maternally expressed UBE3A, another developmental disorder, Prader-Willi syndrome (PWS), is due to loss of expression of paternally expressed genes within this 15q11–q13 region. PWS is characterized by infantile hypotonia, hypogonadism, dysmorphic appearance, small hands and feet, hyperphagia and obesity, DD, and ID. Approximately 70 % of PWS cases involve a 4 Mb deletion of the paternal chromosome 15q11–13, while 20 % of PWS cases are due to maternal UPD of chromosome 15. Another 1–5 % of cases are thought to be due to an imprinting defect. Recent evidence suggests that loss of expression of one or more of the C/D box snoRNAs, including SNORD116, encoded within the SNRPN locus may cause the PWS phenotype [18, 19]. The following section reviews the available diagnostic testing for AS, as well as PWS, since the primary testing method used to diagnose AS also will diagnose PWS.

Infants with AS commonly present with non-specific findings such as DD and/or seizures. These findings may result in a broad differential diagnosis, which can include inborn errors of metabolism, mitochondrial encephalopathy, cerebral palsy, and other neurodevelopmental syndromes including Rett syndrome, Christianson syndrome, and Pitt-Hopkins syndrome, which will be reviewed later in this chapter. Analysis of parent-specific DNA methylation imprints in the 15q11–q13 chromosome region is typically the first test that is ordered as it will detect approximately 80 % of individuals with AS, including those with a deletion, paternal UPD, or an IC defect. This analysis also is used to diagnose PWS and will detect nearly 100 % of those cases. Methylation analysis for AS/PWS typically investigates the methylation status of CpG dinucleotide sites within the promoter region of the SNRPN gene. Once the diagnosis of AS (or PWS) is established through an abnormal DNA methylation pattern, the specific genetic mechanism involved can be investigated for the purposes of genetic counseling. Fluorescent in situ hybridization (FISH) or array-based comparative genomic hybridization (aCGH) analysis can be used to detect a deletion, while DNA microsatellite analysis can be performed to investigate the presence of UPD. For patients with an AS phenotype but normal DNA methylation, sequence analysis of the UBE3A gene can be used for further evaluation of the diagnosis. Collectively, molecular testing for AS through DNA methylation studies and sequencing of UBE3A will identify an alteration in approximately 90 % of individuals with clinical AS/PWS.

The recurrence risk for AS in a family and the type of prenatal testing that is available will vary according to the chromosome or molecular defect that is ultimately detected in the proband. For this reason, prenatal diagnosis should be undertaken only after the specific genetic mechanism in the proband has been determined and the parents have received genetic counseling. In cases where de novo deletions or UPD have been demonstrated, recurrence risk will be low, although prenatal testing may be offered for reassurance purposes. For cases of IC and UBE3A mutations, recurrence risk can be as high as 50 %. Since germline mosaicism has been reported in mothers of AS patients with UBE3A mutations, prenatal testing should be offered even if a mutation is not identified in the mother. Because methylation status within the promoter region of the SNRPN gene is established in early embryonic development, DNA methylation analysis on cells obtained from CVS is theoretically possible; however, most clinical molecular laboratories currently performing prenatal methylation testing prefer to use amniocytes for testing due to the relative hypomethylation of cells derived from the placenta [15]. Prenatal testing is appropriate for families without a previous child with AS if a deletion of chromosome 15q11–q13 is suspected on chromosome analysis of CVS or amniotic fluid, or if trisomy 15 is noted on CVS but a normal karyotype is found on amniotic fluid, which would suggest that a trisomy rescue event may have occurred (possibly resulting in UPD). Prenatal testing also would be indicated if either a de novo chromosome 15 translocation or a supernumerary marker chromosome is found by karyotype analysis.

If the methylation pattern or methylation-specific amplification is characteristic of only paternal inheritance in the individual tested, then a diagnosis of AS is confirmed (Fig. 7.3, lane 8). Conversely, a methylation pattern characteristic of only maternal inheritance is diagnostic for PWS (Fig. 7.3, lane 7). Methylation assays detect virtually all cases of PWS and AS that are caused by large deletions, UPD, and IC defects; however, they will not detect rare small deletions that do not involve the SNRPN locus or sequence alterations within the UBE3A gene. Therefore, methylation analysis will detect approximately 80 % of AS cases. Sequencing of the UBE3A gene will detect another 11 % of cases. Approximately 10 % of AS cases will not be detected with the currently available tests.

The SNRPN probe for Southern blot analysis is available from American Type Culture Collection (http://​www.​atcc.​org). Caution should be exercised when interpreting MS-PCR results, as allele dropout has been reported [27]. Because of the risk of allele dropout, multiple MS-PCR primer sets should be considered to minimize the possibility of a false-positive result. An MS-MLPA kit is available from MRC-Holland (http://​www.​mrc-holland.​com). A pyrosequencing kit to evaluate methylation status of the AS/PWS critical region is available from Qiagen (http://​www.​qiagen.​com). There are many microsatellite markers available for testing of the 15q11–q13 region, but caution should be exercised because some previously used markers are now considered problematic, such as DS15S113 and D15S817 [28, 29].

Rett syndrome is an X-linked neurodevelopmental disorder that almost exclusively affects females. In the classic form of the disease, affected girls appear to develop normally until the age of 6–18 months followed by a characteristic pattern of regression, which includes deceleration of head growth leading to acquired microcephaly, autistic features, loss of speech and purposeful hand use, irregular breathing patterns, stereotypical hand wringing, and seizures [30]. There are well-documented criteria for the clinical diagnosis of Rett syndrome which include the clinical features that are considered as either necessary or supportive for the diagnosis [31]. The frequency of classic Rett syndrome is approximately 1 in 10,000 females [32]. In 1999, mutations in the methyl-CpG-binding protein 2 (MECP2) gene located at Xq28 were reported to be the underlying cause for Rett syndrome [33]. Since that time, a broader range of clinical phenotypes have also been associated with mutations in MECP2 [34]. This section focuses on MECP2 variants causing Rett and atypical-Rett syndromes which are X-linked dominant conditions. The other MECP2-related phenotypes primarily affect males and are considered to be X-linked recessive disorders.

Since being associated with MECP2, the field of Rett syndrome research has focused attention on determining the function of the MECP2 protein to better understand the underlying pathophysiology of the condition. The current perspective of MECP2 function is complex and is based on its apparent involvement as both a transcriptional repressor and activator [35]. MECP2 binds preferentially to methylated DNA via its methyl-CpG-binding domain, and silences transcription by recruiting corepressor complexes through its transcriptional repression domain. More recent studies have demonstrated that loss of MECP2 leads to reduced expression of numerous genes, implying a role in upregulating gene expression, although the mechanisms involved are not well understood [36]. This is an area of great interest given that Rett syndrome is not associated with any sort of gross anatomic abnormalities of the brain. MECP2 clearly affects transcriptional regulation more dramatically in certain regions of the brain than other regions. The physiologic significance of these differences needs to be understood in order for individuals affected by Rett syndrome to benefit from promising new therapeutic options in the future.

Independent studies have confirmed MECP2 as the major causative gene for Rett syndrome by the identification of pathogenic mutations in approximately 95 % of classic cases. Approximately 85 % of classic Rett patients have either point mutations or small insertions/deletions within the MECP2 coding region, while larger deletions have been identified in approximately 10 % of classic cases [37]. Nearly 50 % of Rett patients have one of eight recurrent point mutations, while most of the remainder has one of the many other described pathogenic alterations. It should be noted that greater than 99 % of mutations in individuals with classic Rett syndrome are de novo, with the large majority occurring on the paternally inherited allele [38]. Mutations also have been found in atypical mild variant cases and in severe early-onset cases of Rett syndrome. In addition, MECP2 mutations have been documented in patients with an Angelman-like presentation and in individuals with autistic phenotypes. Affected males with variable phenotypes ranging from lethal neonatal encephalopathy to uncharacterized ID have been shown to be caused by MECP2 mutations. The presentation of Rett syndrome also has been documented in males with Klinefelter (47, XXY) syndrome, as well as those with mosaic MECP2 mutations [39]. Definitive genotype-phenotype correlations have not been consistent, although several findings related to specific mutations appear to be reproducible [40]. Some studies have investigated multiple individuals with the same mutation but variable degrees of clinical severity, demonstrating that additional factors beyond the specific MECP2 alteration play significant roles in the presentation of Rett syndrome [41]. In males, larger duplications of Xq28 that include MECP2 have a consistent set of clinical features that include hypotonia, severe ID, absence of speech, seizures, and recurrent respiratory infections that often lead to death at an early age [42]. Given the spectrum of neurodevelopmental phenotypes associated with MECP2 mutations, the clinical utility of molecular testing is significant.

The MECP2 gene is composed of four exons, which give rise to two distinct MECP2 isoforms [43]. Diagnostic testing for Rett syndrome should begin with DNA sequence analysis of the entire MECP2 coding region given the profile of known mutations. In the past, mutation scanning using denaturing high performance liquid chromatography (DHPLC) was used, but very few laboratories still utilize this method given the relative ease of sequencing MECP2. Sanger sequencing with capillary electrophoresis continues to be considered the gold standard for point mutation detection. Approximately 85 % of classic Rett syndrome patients have mutations that are detectable by sequencing of MECP2, which is performed by numerous clinical molecular laboratories (http://​www.​genetests.​org/​).

To increase the overall mutation detection for the MECP2 gene, additional testing is available for larger MECP2 dosage alterations, present in approximately 10 % of classic Rett patients. Deletions and duplications involving all or part of the gene have been identified by dosage-sensitive testing methods such as multiplex ligation-dependent probe amplification (MLPA), quantitative real-time PCR analysis, and targeted aCGH.

Rett syndrome is commonly considered in females with DD, making the volume of MECP2 testing performed worldwide significant. Molecular confirmation of Rett syndrome is straightforward when one of the eight common MECP2 mutations is present. Interpreting other less common alterations can be assisted by use of available mutation databases. RettBASE, the database dedicated to MECP2 (http://​mecp2.​chw.​edu.​au/​), and the Human Gene Mutation Database (HGMD) are currently the most comprehensive resources. Some of the changes reported in the databases are not clinically correlated, leaving their pathogenicity uncertain. When other novel variants are identified, evaluation of additional family members to determine the segregation pattern may allow the clinical significance of an alteration to be interpreted with greater certainty. For females with variants of unknown clinical significance, the most common strategy is to test both parents. Alterations that are shown to be de novo are normally accepted as pathogenic mutations. In some rare cases, two de novo alterations are present with uncertainty which change is pathogenic, or whether their collective impact on protein function is causative of the symptoms. A variant of unknown significance inherited from the father is interpreted as not being clinically significant given the expectation that the variant would have an even greater phenotypic consequence in a male with a single X chromosome. When variants are inherited from the mother, the interpretation becomes more challenging and typically requires additional testing which often includes both X-inactivation testing of the mother and variant analysis of additional family members. If the proband and her mother carry the same alteration and both have normal/random X-inactivation patterns, the variant typically is interpreted as a benign variant with no clinical significance. Testing of extended family members has the potential to identify the change in a phenotypically normal male which also confirms that the change is not clinically relevant. When testing males, the interpretation issues change to some degree and the strategy for these will be discussed later in the X-linked intellectual disability section of this chapter.

One of the issues encountered in sequence-based clinical testing is the interpretation of novel sequence variations, particularly alleles of uncertain pathogenic significance. A valuable resource for laboratories and referring clinicians is provided by the American College of Medical Genetics, which has issued updated recommended standards and guidelines for interpretation of sequence variations [44]; the recommendations are available at http://​www.​acmg.​net. Another current issue pertaining to molecular testing for Rett syndrome is the value of two-tier testing (sequencing followed by dosage analysis) to provide comprehensive mutation analysis of MECP2. Identification of the mutation in the proband facilitates prenatal testing in subsequent pregnancies, although the majority of MECP2 point mutations are de novo and of paternal origin, with extremely low rates of recurrence. Prenatal testing should be offered to females who are known to carry a point mutation or large rearrangement in MECP2. In cases where the mother is not a carrier, prenatal testing may be sought for parental reassurance due to rare reports of germline mosaicism [45]. The current focus on point mutation analysis within the coding region by sequence analysis does not rule out potential mutations in regulatory elements or other important noncoding regions of MECP2. Equally important to consider are the other conditions that phenotypically overlap with Rett syndrome and should be given further consideration when clinical testing for MECP2 is negative. Several of these syndromes and their respective gene associations are discussed in the following section.

AS and Rett syndromes are clinically similar conditions that often present with numerous overlapping phenotypic features. Clinicians with significant clinical experience typically can distinguish the two conditions, but in many cases clinical diagnosis can be challenging. As discussed earlier, AS and Rett syndromes are caused by disruption or loss of normal gene function of the UBE3A and MECP2 genes, respectively. Testing for these conditions is common given their prevalence and the number of individuals with DD suggestive of one or both of these diagnoses. Accordingly, many patients have clinical presentations that mimic AS and Rett syndrome and have normal clinical test results for both disorders. In these instances, additional testing of genes causing diseases with overlapping phenotypes may provide the molecular diagnosis needed. Some of the most common conditions that should be considered in the differential diagnosis for AS and Rett syndrome for which diagnostic testing is available are discussed in this section. The syndromes detailed in this section are not meant to be an all-inclusive list, but rather as a starting point for second tier testing for patients with features in this clinical spectrum of ID/DD.

The molecular confirmation of a clinical diagnosis allows the physician to provide recurrence risks and genetic counseling, as well as prenatal diagnosis, to the proband and additional family members. Once the presence of a causative mutation has been established in an individual, targeted analysis of fetal samples is available for prenatal testing, as well as for other family members.