Subtelomeric aberrations in Thai patients with idiopathic mental retardation and autism detected by multiplex ligation-dependent probe amplification

We detected 4 types of submicroscopic aberration in samples of 5 patients of 129 from the first screening. Three duplications at chromosome regions 15q11.2, Xp22.33, and 11p15.5, and 1 deletion at 1p36.33, were identified. All patients had been previously diagnosed with idiopathic MR. No abnormality was detected in any individuals diagnosed with autism. Two patients had the same duplication at Xp22.23. All initial results from the first screening were confirmed using different MLPA probes. The deletion of 1p36.33 was confirmed by real-time PCR. The results and the available clinical features of all affected patients are summarized in Table 1. Additionally, we also detected 3 samples with a 4q35.2 deletion in the initial MLPA screening, but these deletions were not confirmed by a different MLPA probe in the same region. Such false-positive results were noted in a previous study [9], which suggested that they were possibly caused by a single nucleotide polymorphism (SNP) in the MLPA screening probe for the 4q region. The MLPA analysis from all positive samples is shown in Figure 1.

Figure 1

All 5 families among the affected individuals with submicroscopic aberrations were invited to participate in this research. Unfortunately, 4 of these families declined to participate because of personal beliefs or other reasons, and only 1 family whose child had a duplication of 15q11.2 agreed to further investigation. With consent, the pedigree of the 15q11.2 duplication family is shown in Figure 2.

Figure 2

The subtelomeric regions at both ends of the human chromosome arms are gene-rich. Any unbalanced rearrangement involving genes in these regions can cause mental retardation and an abnormal phenotype. Unfortunately, standard karyotyping, even with high resolution chromosome banding (550 bands), cannot detect such aberrations. In this study, we screened samples from 129 patients with idiopathic MR and autism for any rearrangements with MLPA and detected 5 patients with positive screening (4%), which is within the range reported previously [2-5, 10, 16, 19, 20,25, 26].

The duplication of 15q11.2 detected in our patient was a de novo event. The major anomaly manifested in our affected individual was developmental delay without any dysmorphism. Although there were 2 other affected individuals in the family, the clinical symptoms of each individual seemed to be distinctive, and it is likely that individuals IV-9 and IV-10 are differently affected. The proximal region of chromosome 15q is a region prone to genomic rearrangement. This region contains several low-copy repeats (LCRs) that mediate various duplications and deletions via nonhomologous recombinations [27]. The most well-known syndromes caused by deletions in this region are the Prader–Willi and Angelman syndromes. Mapping our MLPA probes to the region showed that they were centromeric to the SNURF– SNRPN imprinted region, but this is still an area where rearrangements are known to frequently occur [23, 27]. Duplication of 15q11.2 can cause a variety of phenotypes, primarily manifesting neurological symptoms. Developmental delay, especially speech delay, is the most frequent problem detected [22, 23], and was also detected in our patient. Because duplication of 15q11-13 accounts for approximately 0.5%-3% of autistic spectrum disorder (ASD), ASD is another feature frequently observed in patients with 15q duplication [24]. Even though dysmorphism has been noted in some cases with proximal 15q duplication, no consistent physical anomaly has been reported [23]. As in our patient, developmental delay without dysmorphism seems to be the main feature of 15q duplication. Because our finding was de novo, this duplication is likely to be pathologic.

We also had 1 patient with 11p15.5 duplication and 2 patients with Xp22.33 duplication in this study. The 11p15.5 region is known to contain a cluster of imprinted genes that together play an important role in the control of fetal growth. An abnormality in the imprinting mechanism of genes in this region or a uniparental disomy of chromosome 11 can cause either Beckwith–Wiedemann syndrome (BWS) or Silver– Russell syndrome (SRS) [28]. Paternal duplication of the 11p15.5 region has been reported in a few patients with BWS and maternal duplication of the same region can cause SRS [28, 29]. This duplication detected in BWS/SRS can arise as a de novo event or as a result of chromosomal rearrangement from inherited translocation or inversion. In addition to abnormal growth, psychomotor retardation is often detected in this group of patients [28-30]. From the medical records of our patient, in addition to MR, she also had several dysmorphic features (choanal atresia, coloboma of the right upper and lower eyelids, right cleft lip) (Table 1), but there was no record of postnatal overgrowth suggesting BWS. The patient also had no apparent evidence of intrauterine growth retardation or postnatal growth delay indicating SRS. Therefore, our patient is unlikely to fulfill the diagnostic criteria for either syndrome, although a thorough clinical examination should be performed to confirm these observations. However, considering the several dysmorphic features she clearly exhibits, this 11p15.5 duplication, if pathological, is more likely to be a complex submicroscopic chromosomal rearrangement than a simple duplication. Further investigation would be needed to elucidate the precise pathology.

We detected 2 patients with an Xp22.3 duplication in this study, both male with developmental delay (2/129 = 1.6%). Xp22.3 is a gene-rich region, and deletion of any of several genes in this region can cause a variety of abnormal phenotypes such as X-linked ichthyosis with developmental delay, attention deficit hyperactivity disorder, autism, and social communication deficit [31]. However, the clinical significance of the Xp22.3 duplication is still uncertain and has been classified as a normal variant [32, 33], a pathologic change [34-36] or an unclassified rearrangement [37]. The major gene thought to contribute to abnormal phenotypes in this region is the gene for steroid sulfatase (STS). A previous study of patients with developmental delay and Xp22.3 duplication suggested a possible association of the duplication with abnormal phenotypes, especially those involving STS [37]. The same study suggested the duplication might be a risk factor or a modifier of associated intellectual disability [37]. Mapping our MLPA probes to the Xp22.3 region showed that our probes laid approximately 7 Mb telomeric to the STS locus. Because of limitations of the MLPA method, the actual duplication size and STS involvement could not be determined in this study. Because parental DNA was not available, the significance of this duplication could not be examined, and thus remains uncertain. Therefore, there is still a possibility that this duplication may be a rare variant, and further screening of this aberration in a normal population should be performed.

Terminal deletion of the short arm of chromosome 1 (1p36) is generally difficult to identify by conventional cytogenetics due to the light-staining G-negative bands, but this deletion is thought to be a contiguous gene syndrome [38]. This deletion can be derived from an unbalanced translocation, complex rearrangement or can be merely a simple deletion. A previous study found no correlation between the deletion size and abnormal phenotypes [38]. Common clinical phenotypes of 1p36 deletion include global developmental delay, hypotonia, hearing impairment, heart defects, and distinct dysmorphic features including microcephaly or brachycephaly, midface hypoplasia, deep-set eyes, flat nasal bridge, large anterior fontanelle, and a prominent forehead [38-40]. Because the family of the patient with the 1p36 deletion refused to be part of any further study, we could not perform additional tests that would indicate more conclusively that this deletion derived from any familial chromosomal rearrangement. However, the patient was noted to have frontal bossing with a flat head, which could have been brachycephaly, feature frequently found in patients with 1p36 deletion syndrome. Therefore, the 1p36.33 deletion detected in our study was likely to be pathologic.

Our study shows that implementing subtelomeric MLPA screening in Thai patients with idiopathic mental retardation and autism, who have a normal karyotype, can increase the diagnostic yield. Because some subtelomeric defects may be caused by unbalanced chromosomal rearrangements resulting in a complex submicroscopic aberration, MR and dysmorphism are frequently detected in this group of patients. Currently, a recommended test for MR patients with normal karyotype is microarray analysis, which, although being a useful tool for whole genome screening, is also costly and requires an array facility, which may not be widely available, especially in developing countries. MLPA is an effective alternative way to detect subtelomeric rearrangement of all chromosomes in a single reaction, and is also cheaper and less laborious than FISH. Therefore, we recommend MLPA screening in patients with idiopathic MR with dysmorphism and normal karyotype in diagnostic laboratories where a microarray facility is unavailable.