Chromosome 15q13.3 Deletion Syndrome

A number sign (#) is used with this entry because it represents a contiguous gene deletion syndrome (chr15: 28.7-30.3 Mb, NCBI36). The 15q13.3 deletion is a recurrent 1.53-Mb deletion that resides between breakpoints BP4 and BP5 and includes 7 protein-coding genes. Specific genes implicated in the phenotype include CHRNA7 (118511) and OTUD7A (612024), both of which reside within the critical region. Most patients have heterozygous deletions, but some have homozygous deletions, which are associated with a more severe phenotype.

See also chromosome 15q11-q13 duplication syndrome (608636), which has been associated with overlapping features.

Susceptibility to idiopathic generalized epilepsy-7 (EIG7; 604827) has also has been linked to chromosome 15q13.3 deletion.

Description

Heterozygous deletion of chromosome 15q13.3 is associated with a highly variable phenotype, even within families segregating the same deletion. Individuals with the deletion may have mild to moderate mental retardation or learning difficulties, or may have no cognitive deficits. Some individuals have epilepsy. Various dysmorphic features have been described, but there is no consistent or recognizable phenotype (review by van Bon et al., 2009). Patients with homozygous deletions in this region have severe neurodevelopmental problems, with epileptic encephalopathy, hypotonia, and poor growth (Endris et al., 2010).

Clinical Features

Sharp et al. (2008) reported a recurrent microdeletion syndrome characterized by mental retardation, epilepsy, and variable dysmorphism of the face and digits. They described 9 affected individuals, including 6 probands: 2 with de novo deletions, 2 who inherited the deletion from an affected parent, and 2 with unknown inheritance. Features shared among 3 or more individuals included hypertelorism, upslanting palpebral fissures, prominent philtrum with full everted lips, short and/or curved fifth finger, and short fourth metacarpals. Skeletal and/or joint defects of the hand were observed in 7 of the 9 individuals. Seizures or abnormal electroencephalograms were reported in 7 of the 9 individuals. Sharp et al. (2008) recommended that testing for the 15q13.3 deletion syndrome should be considered in individuals with unexplained mental retardation, seizures, and mild dysmorphic features.

Ben-Shachar et al. (2009) identified 20 individuals, including 14 children and 6 parents from 12 families, with microdeletions of chromosome 15q13.3. Clinical features were variable, but 12 of 14 children had developmental delay, mental retardation, or borderline IQ. At least 6 children had symptoms within the range of autism spectrum disorder (ASD; 209850). Nine children demonstrated abnormal behavior, including aggressiveness, repeated head banging, and/or attention deficit hyperactivity disorder. Only 1 child had seizures. The facial appearance was variable and ranged from near normal to moderately dysmorphic. Common facial features included hypertelorism, short philtrum, and everted and thick upper lip. Mild digital aberrations, including brachydactyly and clinodactyly, were observed in 5 patients. Family studies showed that the deletion was inherited in 7 of 8 families where determinable. Two fathers with the deletion had learning disability and bipolar disorder (see 125480), whereas the 4 other parents with the deletion had no neurologic or psychiatric abnormalities, indicating incomplete penetrance. The last family had no parental samples available, but affected sibs suggested familial inheritance. Of note, 6 of the 14 children had been adopted, suggesting that adoption may have been related to cognitive, psychiatric, or social difficulties in their biologic parents who may have carried the deletion. Ben-Shachar et al. (2009) noted the wide range and heterogeneity of phenotypic expression reported for this deletion.

Miller et al. (2009) identified 5 unrelated patients with chromosome 15q13.2-q13.3 deletion involving breakpoints 4 and 5 (BP4-BP5) identified by array comparative genomic hybridization (CGH). All had subtle dysmorphic features, impaired language skills, and developmental delay. One patient had mental retardation, and 4 had below average to average intelligence, including 2 with significant learning disability. The patients often had oromotor dyspraxia with disarticulation. Some had mild motor delay. Most had a diagnosis of an autism spectrum disorder or autistic features, as well as difficulties with attention, hyperactivity, mood regulation, and impulsive behaviors. Two patients inherited the microdeletion from a mother with learning difficulties. The deletion sizes ranged from 1.50 to 1.93 Mb.

Shinawi et al. (2009) identified the same 680-kb deletion of chromosome 15q13.3 in 10 individuals from 4 families. One family of European ancestry contained 6 affected individuals. The proband was an 8-year-old boy with obesity, severe mental retardation, and mild facial dysmorphism, including epicanthal folds, anteverted nares, and a thin upper lip. Although he did not have seizures, EEG was abnormal. The deletion was present in his mother, 2 sibs, maternal aunt, and maternal grandmother. The mother and her sister had a history of mental retardation and epilepsy, and his sibs had global developmental delay. In a second family, the proband was a 21-month-old girl with impaired growth and severe global developmental delay. Her mother, who also carried the deletion, was reported to have normal intelligence, but had a history of epilepsy since age 5. Another unrelated patient with the deletion had mild mental retardation, attention deficit hyperactivity disorder, and aggressive behavior without seizures, and yet another had global developmental delay, hypotonia, and failure to thrive. In all, 4 of 10 individuals with the 680-kb deletion had seizures or EEG abnormalities, and 9 of 10 showed developmental delay and/or mental retardation.

Van Bon et al. (2009) reported 18 probands with heterozygous deletion of chromosome 15q13.1-q13.3. Sixteen of the patients who had a BP4-BP5 deletion were detected among a larger cohort of 6,624 persons referred for mental retardation and/or congenital defects, yielding a rate of 0.24%. One of the most severely affected individuals was a girl who had onset of seizures at age 1 month, showed delayed psychomotor development, and was severely mentally retarded with poor speech, apathetic behavior, and sleeping problems. She also had short stature and microcephaly. Dysmorphic features included asymmetric skull with bitemporal narrowing, bristly hair, synophrys, blepharophimosis, squint, bulbous nasal tip, and folded helices. Other features included tapering fingers, deep palmar creases, hypoplastic fourth and fifth toes, an external rotation of the feet, and multiple pigmented nevi. Brain MRI showed showed several abnormalities, such as an arachnoidal cyst, parenchymal hypoplasia, dislocation of cerebellar structures, and mild hypogenesis of the corpus callosum. Microarray analysis showed a deletion between BP4 and BP5, which was also present in the normal mother. At the other end of the phenotypic spectrum was a 9-year-old girl with normal cognitive development who presented at birth with tetralogy of Fallot and triphalangeal thumb. She had mild hypertelorism. Microarray analysis showed a BP4-BP5 deletion with uncertain inheritance. Overall, 16 of 18 probands had some degree of cognitive impairment varying from mild learning problems to all levels of mental retardation. Behavioral problems were frequent (59%), and comprised poor attention span, hyperactivity, and aggressive/impulsive behavior. Less common features were hypotonia (47%), prominent nasal tip (35%), short stature (24%), strabismus (18%), large ears (24%), cardiac defects (17%), fifth finger clinodactyly (24%), and pigmented nevi (18%). However, there were a total of 13 relatives who carried the same deletion as the proband, and all of these frequencies would be significantly lower if those individuals were included. Only 2 patients had a history of seizures. Two individuals had a different deletion of BP3-BP5, and 1 had a deletion of BP3-BP4. In the family of 1 proband with deletion of BP3-BP4, the deletion was also also present in an unaffected brother, father, and uncle, but not present in a mentally retarded brother, suggesting that it is not of clinical significance. Van Bon et al. (2009) concluded that BP4-BP5 deletion does not lead to a clinically recognizable syndrome, but that it likely plays a contributing role in the pathogenesis of conditions affecting the brain, including mental retardation. The variable phenotypic outcome of the deletion is likely to be determined in conjunction with other factors, such as a mechanism for overcoming primary embryologic defects.

Pagnamenta et al. (2009) reported 3 male sibs with autism associated with an approximately 2.0-Mb deletion of chromosome 15q13.3 involving BP4-BP5 that was inherited from their unaffected mother. All 3 boys had a history of language delay, and IQ levels ranged from 72 to 96. None had seizures, and none had dysmorphic features, although 2 had an increased head circumference (greater than 97th percentile).

Masurel-Paulet et al. (2010) found that 16 (0.35%) of 4,625 patients tested for developmental delay had a microdeletion of chromosome 15q13.3 including the CHRNA7 gene. Twelve patients and 13 relatives had the common deletion between BP4 and BP5 ranging in size from 1.5 to 1.8 Mb. The phenotype was variable, and characterized by mild or no dysmorphic features and mild to moderate mental retardation. Two patients had epilepsy, 6 had anxiety disorder, 3 had phobias, 4 had inhibition, 2 were hyperactive, 1 self-mutilated, and 1 had autistic features. None had schizophrenia. In 6 families, the deletion was inherited from an apparently normal parent, indicating incomplete penetrance; in 4 families, the carrier parent reported learning disabilities. One patient with a severe phenotype including epileptic encephalopathy with retinopathy, autistic features, and choreoathetosis was homozygous for a 1.5-Mb BP4-BP5 deletion. Both of his parents, who carried the deletion, had mild mental retardation. In addition, 3 patients and 2 relatives had a smaller 500-kb deletion. A review of reported cases in the literature indicated that male carriers of a 15q13 deletion are more likely to be symptomatic.

Of 1,035 individuals carrying copy number variants associated with schizophrenia, Sahoo et al. (2011) identified 69 individuals with microdeletion at 15q13.3. Indications for study in these 69 individuals included developmental delay, autism, dysmorphic features, seizures, hypotonia, obsessive compulsive disorder, and congenital heart defect. The average age at diagnosis was 6.6 years with an age range of 0.1 to 19.7 years. Sahoo et al. (2011) concluded that these and other results from their study, the largest genotype-first analysis of schizophrenia susceptibility loci to that time, suggested that the phenotypic effects of copy number variants associated with schizophrenia are pleiotropic and imply the existence of shared biologic pathways among multiple neurodevelopmental conditions.

Hoppman-Chaney et al. (2013) identified 19 individuals, including 10 probands and 9 family members, with a heterozygous deletion of chromosome 15q13.3 including only the CHRNA7 gene and no neighboring genes, and 1 patient with a homozygous deletion of CHRNA7. The patients were ascertained from a database of over 15,000 individuals who underwent cytogenetic testing. Ten individuals were from 2 independent extended families, and the deletion segregated with the phenotype. Among 15 children with the deletion, 5 were under the age of 2 years and did not have obvious neurocognitive problems. The other children had variable developmental delay, cognitive disability, autistic features, and/or attention-deficit hyperactivity disorder. Five of 6 patients with available clinical information had dysmorphic features, such as upslanting palpebral fissures, dental malocclusion, hypertelorism, and macro- or microcephaly. Only 1 patient had seizures. One patient had a de novo deletion, and another with a more severe phenotype had a homozygous deletion. All 5 parents with the deletion had neurocognitive features of the disorder.

Lowther et al. (2015) identified a total of 246 cases (133 children, 113 adults) with deletions overlapping or within the 15q13.3 BP4-BP5 region, including 7 novel adult cases from local cohorts. No BP4-BP5 deletions were identified in 23,838 adult controls. Overall, 198 cases (121 children, 77 adults; 80.5%) had at least 1 neuropsychiatric diagnosis. Accounting for ascertainment, developmental disability/intellectual disability was present in 57.7%, epilepsy/seizures in 28.0%, speech problems in 15.9%, autism spectrum disorder in 10.9%, schizophrenia in 10.2%, mood disorder in 10.2%, and attention deficit hyperactivity disorder in 6.5%. By contrast, major congenital malformations, including congenital heart disease (2.4%), were uncommon. Placenta previa occurred in the pregnancies of 4 cases.

Using published and unpublished case subjects as well as clinical microarray data from 38,325 individuals with a range of neurodevelopmental disorders, Uddin et al. (2018) identified 156 individuals with 15q13.3 microdeletions impacting the typical BP4-BP5 region. Only 1 individual with this deletion was found among 22,241 control individuals (p less than 1.30 x 10(-29)). Eight of the 156 patients with deletions had a primary diagnosis of autism spectrum disorder (ASD). Analysis of the breakpoints found a minimal region of overlap associated with an atypical smaller deletion in 42 cases, which included only 2 genes: CHRNA7 and OTUD7A. OTUD7A was implicated in particular because a 5-year-old girl had a deletion encompassing only OTUD7A but not CHRNA7, and 3 probands with de novo heterozygous OTUD7A-specific mutations were identified in another cohort of over 6,000 ASD individuals studied.

Cytogenetics

Among the 9 individuals with deletions of 15q13 reported by Sharp et al. (2008), the proximal breakpoint of the largest deletion was contiguous with breakpoint 3 (BP3) of the Prader-Willi and Angelman syndrome region (see 105830), extending 3.95 Mb distally to BP5. The smaller 1.5-Mb deletion had a proximal breakpoint within the larger deletion (BP4) and shares the same distal BP5. This recurrent 1.5-Mb deletion contains 6 genes, including a candidate gene for epilepsy, CHRNA7 (118511), that the authors suggested was responsible for the seizure phenotype observed in this syndrome. The BP4-BP5 region undergoes frequent inversion, suggesting a possible link between this inversion polymorphism and recurrent deletion. The frequency of these microdeletions in mental retardation cases is approximately 0.3% (6/2,082 tested), a prevalence comparable to that of Williams-Beuren (194050), Angelman, and Prader-Willi (176270) syndromes.

Makoff and Flomen (2009) noted that BP4-BP5 region has many segmental duplications and inverted repeats, and suggested that the BP4-BP5 inversion polymorphism identified by Sharp et al. (2008) probably did not predispose to the observed chromosome 15q13.3 microdeletions, since nonallelic homologous recombination between duplicons in the same orientation would produce a small deletion removing CHRFAM7A (609756) but leaving CHRNA7 intact. Makoff and Flomen (2009) suggested that a polymorphic inversion of the CHRFAM7A gene (Flomen et al., 2008) would be more likely to predispose to the observed microdeletions, including deletion of the CHRNA7 gene.

In the study by Ben-Shachar et al. (2009), the most common BP4-BP5 deletion of 1.6 Mb corresponded to chromosome 15q13.3, and was seen in 11 of 12 families. One family had a larger BP3-BP5 deletion of 3.4 Mb, corresponding to 15q13.1q13.3. In all of the families, the BP4-BP5 critical region was deleted.

The common 680-bp deletion in the 10 patients studied by Shinawi et al. (2009) lay within the 1.5-Mb deletion of 15q13.3 and encompassed the entire CHRNA7 gene and the first exon of 1 of the isoforms of OTUD7A (612024). The 680-bp deletion was always accompanied by a 90-kb duplication within BP4. Detailed analysis allowed narrowing of the precise breakpoints, and indicated that the deletion likely resulted from nonallelic homologous recombination between low-copy repeats within the FAM7A1 and FAM7A2 genes. Shinawi et al. (2009) concluded that haploinsufficiency for CHRNA7, and not for OTUD7A, underlay the phenotype. In their population, Shinawi et al. (2009) found a significantly higher number of similar-sized duplications, suggesting that they are comparatively benign.

Sahoo et al. (2011) analyzed 38,779 individuals referred to the diagnostic laboratory for microarray testing for the presence of copy number variants encompassing 20 putative schizophrenia susceptibility loci. They also analyzed the indications for study for individuals with copy number variants overlapping those found in 6 individuals referred for schizophrenia. After excluding larger gains or losses that encompassed additional genes outside the candidate loci (e.g., whole-arm gains/losses), Sahoo et al. (2011) identified 1,113 individuals with copy number variants encompassing schizophrenia susceptibility loci and 37 individuals with copy number variants overlapping those present in the 6 individuals referred for schizophrenia. Of these, 1,035 had a copy number variant of 1 of 6 recurrent loci: 1q21.1 (612474, 612475), 15q11.2 (608636), 15q13.3, 16p11.2 (611913), 16p13.11 (610543, 613458), and 22q11.2 (192430, 608363). Sahoo et al. (2011) identified the 15q13.3 microdeletion in 69 individuals; 5 were de novo, 12 maternally inherited, 6 paternally inherited, and 46 of unknown inheritance. The microdeletion was seen in 69 of 23,250 cases referred to their laboratory for a frequency of 0.30% and not at all in 5,674 controls for a p value of less than 0.0001 (see Itsara et al., 2009). This microdeletion was previously reported in 0.2% of schizophrenia patients compared to 0.017% controls by Kirov et al. (2009). The frequency reported in a case-control comparison in a variable neurodevelopmental deficit population was 0.48% of cases, with a control frequency of 0.2%, as reported by Vassos et al. (2010). Sahoo et al. (2011) concluded that the results from their study, the largest genotype-first analysis of schizophrenia susceptibility loci to that time, suggested that the phenotypic effects of copy number variants associated with schizophrenia are pleiotropic and imply the existence of shared biologic pathways among multiple neurodevelopmental conditions.

Kaminsky et al. (2011) presented the largest copy number variant case-control study to that time, comprising 15,749 International Standards for Cytogenomic Arrays cases and 10,118 published controls, focusing on recurrent deletions and duplications involving 14 copy number variant regions. Compared with controls, 14 deletions and 7 duplications were significantly overrepresented in cases, providing a clinical diagnosis as pathogenic. The 15q13.2-q13.3 (BP4-BP5) deletion was identified in 46 cases and no controls for a p value of 1.44 x 10(-10) and a frequency of 1 in 342 cases.

Antonacci et al. (2014) showed that recurrent deletions in the chromosome 15q13.3 region are likely caused by genomic instability due to segmental duplications and inversion events. They demonstrated that the inversion breakpoints map near GOLGA8 (616180) and are likely caused by a duplicated sequence that increases the probability of unequal crossovers.

Inheritance

The chromosome 15q13.3 deletion is inherited in approximately 75% of cases, but has also been found in apparently normal individuals, consistent with incomplete penetrance (summary by Hoppman-Chaney et al., 2013).

Masurel-Paulet et al. (2010) reported a patient with a severe phenotype including epileptic encephalopathy with retinopathy, autistic features, and choreoathetosis who was homozygous for a 1.5-Mb BP4-BP5 deletion. Both of his parents, who carried the deletion, had mild mental retardation.

Endris et al. (2010) reported 2 additional unrelated German patients with homozygous deletions of 15q13.3 associated with severe epileptic encephalopathy. One patient was a 13-year-old boy who presented at birth with severe hypotonia and developed therapy-resistant seizures with hypsarrhythmia at age 4 months. At age 13, he had severe psychomotor retardation, poor growth, and dystonic tetraparesis. Brain MRI showed mildly reduced volume of the frontal lobes and dilated extraaxial spaces. He was compound heterozygous for the recurrent 1.5-Mb deletion inherited from his father and a smaller 680-kb deletion inherited from his mother. Neither parent was affected. The second patient had axial hypotonia at birth and developed seizures at age 10 months. Brain MRI showed complex anomalies, including reduced hemispheric volume, abnormal white matter signals, thin corpus callosum, dysplastic cerebellum, and retrocerebellar arachnoid cysts. Optic nerve atrophy was suspected. At age 20 months, he had no motor development, no eye contact, and poor growth. Genetic analysis found compound heterozygosity for the 1.5-Mb deletion inherited from the unaffected mother and an overlapping 3.4-Mb deletion inherited from the unaffected father. The common deleted region in all 4 alleles included the CHRNA7 gene.

Among 110 individuals with deletions overlapping or within the 15q13.3 BP4-BP5 region for whom data on transmission was available, Lowther et al. (2015) found that 94 (85.4%) were inherited. Significantly more were of maternal than of paternal origin (69 vs 25 cases; p less than 0.0001).

Molecular Genetics

Hoppman-Chaney et al. (2013) identified 9 probands with heterozygous deletions of chromosome 15q13.3 including only the CHRNA7 gene (118511) and no neighboring genes, who had a variety of neurocognitive defects consistent with the larger chromosome 15q13.3 deletion syndrome. One of the patients with a more severe phenotype was homozygous for the deletion, and 1 additional patient carried a de novo deletion. The findings provided further evidence that deletion of the CHRNA7 gene is responsible for the clinical features of this syndrome.

Uddin et al. (2018) performed whole-genome or whole-exome sequencing in over 6,000 individuals with a neurodevelopmental disorder who did not have a 15q13.3 microdeletion and detected 8 de novo mutations in genes within the breakpoint interval (BP4-BP5), 3 of which occurred in the OTUD7A gene. In 2 sibs (proband is case 3) with autism spectrum disorder (ASD), Uddin et al. (2018) identified a de novo heterozygous 9-bp in-frame deletion in the OTUD7A gene (Asn492_Lys494del). In vitro studies showed that the mutation resulted in normal levels of mutant protein production, but that mutant protein was not able to rescue the abnormal dendritic phenotype of mouse neurons with heterozygous deletion of chromosome 15q13 (Df(h15q13)+/- mouse mutants; see ANIMAL MODEL). Expression of mutant OTUD7A in cultured wildtype neurons significantly reduced dendritic spine length, consistent with a mild dominant-negative effect. Studies of patient cells were not performed. Uddin et al. (2018) also identified 2 additional unrelated patients with ASD associated with de novo heterozygous intronic variants in the OTUD7A gene (c.-223+11014A-G and c.1150+935del); functional studies of these variants and studies of patient cells were not performed.

Animal Model

Fejgin et al. (2014) found that mutant mice with a heterozygous 15q13.3 microdeletion (Df(h15q13)/+) showed marked changes in neuronal excitability in acute seizure assays, with increased propensity to develop myoclonic and absence-like seizures, but decreased propensity for clonic and tonic seizures. Mutant mice had impaired long-term spatial reference memory and a decreased theta frequency in hippocampus and prefrontal cortex, as well as auditory processing deficits similar to those observed in schizophrenia. The neurologic abnormalities in these mice recapitulated some of the phenotypic features observed in humans with 15q13.3 deletions.

Using RNA sequencing, Uddin et al. (2018) found that Df(h15q13)/+ mice had significantly decreased expression of several genes involved in forebrain cortical development, including Chrna7 and Otud7a. This decreased expression was associated with a small, but significant, reduction in dendritic spine density, mature mushroom-shaped spines, dendritic length, and dendritic arborization in the frontal cortex compared to wildtype. Isolated cortical neurons from heterozygous mutant mice showed similar abnormalities. Expression of wildtype OTUD7A was able to rescue the abnormalities in dendritic spine density, length, and the proportion of mushroom and stubby spines, but expression of CHRNA7, KLF13 (605328), or FAN1 (613534), other genes within the candidate region, was unable to rescue these defects. However, CHRNA7 was able to rescue dendrite outgrowth, suggesting some potential overlap in function. Uddin et al. (2018) concluded that OTUD7A is a critical gene in the 15q13.3 microdeletion syndrome.

Yin et al. (2018) found that homozygous Otud7a-null mice had preweaning growth delay, delayed motor milestones, increased seizure-like activity, and impaired vocalization, although memory and learning appeared to be normal as measured by fear conditioning and novel object recognition tests. Homozygous and heterozygous-null mice showed impaired acoustic startle response compared to wildtype, and female mutant mice had reduced prepulse inhibition, all of which may represent features of schizophrenia. Primary cortical neurons derived from Otud7a-null mice showed a significant reduction in dendritic spine density compared to controls, and this defect could be rescued by expression of wildtype Otud7a, although there were no apparent defects in dendritic growth or dendritic complexity. Primary neurons from mutant mice also showed a decrease in functioning excitatory synapses. Yin et al. (2018) concluded that OTUD7A deficiency largely accounts for the human phenotypes associated with 15q13.3 deletion syndrome, and that OTUD7A is important in the regulation of dendritic spine density and activity.

Using 12 behavioral assessments and electroencephalogram recordings with freely-moving mice, Yin et al. (2017) found that loss of Chrna7 expression was not sufficient to cause statistically significant social, behavioral, or neuropsychiatric-like changes. Yin et al. (2017) also found no evidence for electrophysiologic phenotypic difference between Chrna7 -/- and wildtype mice. They concluded that knockout of Chrna7 in mice does not recapitulate phenotypes observed in humans with chromosome 15q13.3 microdeletion syndrome.