Detection of Clinically Relevant Genetic Variants in Children with Autism Spectrum Disorder by Whole Exome Sequencing

Autism spectrum disorder (ASD) is a neurodevelopment disorder which is related with a good number of genetic mutations. There is a lack of data regarding the genetic framework of children with ASD in Bangladesh. Hence, this study was conducted to detect clinically relevant genetic variants of ASD. A total of13 children with ASD aged between 2-10 years were included in this study, DSM-5 was used to confirm the diagnosis of ASD and to exclude other neurodevelopment, emotional, and behavioral disorders. Whole Exome Sequencing was done under the supervision of a geneticist. Mean age of participants was 5.54±3.13 years with male predominance (61.5℅). Positive family history of neurodevelopment disorder was present in 46.2%. The average loss of acquired skill of study participants was 26.45±14.50 months. Most of the patients had delayed development of vocalization (92.3%), body gesture (61.5%), and spontaneous phrases (69.2%). Attention deficit hyperactivity disorder (ADHD) was the most common (46.2%) co-morbidity Majority patients (76.92%, n = 10) had presence of gene mutation, wherein 38.46% (n = 5) had variants of uncertain clinical significance (VOUS), 30.77% (n = 4) had likely pathogenic, and only 1 patient (7.69%) had pathogenic gene mutation. In this study, a good number of pathogenic genes related to ASD namely TENM4, ASXL1, CHD3, TSC2 and CACNA1H were detected. Larger multicenter study is recommended.


Introduction
Autism spectrum disorder (ASD) is a group of highly heterogeneous neurodevelopmental disorders affecting 1 in 59 children aged 8 years. Boys are four times more likely to be affected than girls (Baio et al., 2018). It is characterized by impaired reciprocal social interaction and communication, as well as restricted repetitive interests and behaviors (American Psychiatric Association, 2013). The symptoms could develop gradually from early childhood, affecting daily functioning and persisting throughout one's life (Stefanatos, 2008). Given the variety of phenotypes and severity, it's believed that genetic factors play pivotal role in the pathogenesis of ASD, in combination with developmental and environmental factors (Anagnostou et al., 2014). The rapid progression of genomics technologies, coupled with expanding cohort sizes, have led to significant progress in characterizing the genetic architecture of complex disorders (Schizophrenia Working Group of the Psychiatric Genomics Consortium, 2014; Sanders et al., 2015;de Lange et al., 2017;Deciphering Developmental Disorders Study, 2017;Marshall et al., 2017).
This study was undertaken to analyze the genetic framework of children with ASD. This will highlight the complexity of the genetic landscape of the disease and also would shed light on some of the biological pathways at risk in ASD.

Materials and methods
This was a descriptive cross sectional study conducted from July 2019 to June 2020 in the Department of Pediatric Neurology, Bangabandhu Sheikh Mujib Medical University (BSMMU). Thirteen children aged 2-10 years diagnosed as ASD were enrolled for the study. Formal ethical approval was taken from the Institutional Review Board (IRB) of BSMMU. Informed written consent was taken from the parents or care givers of the study participants.
The parents of the children were interviewed face to face using a structured questionnaire. The socio-demographic information, presenting complaints, history of present illness, birth history, developmental history, family history, history of comorbidities like seizure, hyperactivity, sleep and feeding pattern were recorded. General examination and neurological examination was done. Diagnosis of ASD was done according to DSM-5 criteria. (Chiche, 2016).
For genetic study, 3 mL of blood sample was taken from anti-cubital vein with full aseptic precaution from each child. Serum was separated from blood samples at collection place. Sample was stored at -20 0 C temperature until transported to genetic laboratory, Neurogen Technologies Limited (A genetic lab) in Bangladesh for Whole Exome Sequencing. A geneticist supervised the whole process. The statistical analysis was carried out by using the Statistical Package for Social Sciences version 24.0 for Windows.

Results and discussion
The phenotype and genotype of 13 children with ASD was obtained. Mean age of the participants was 5.54±3.13 years. There was a male predominance (61.5℅). Maximum children were from urban area (69.2%) and belonged to middle class (69.2%). Most of the subject never went to school (69.2%). Father and mother of study participants were mostly literate (92.3 and 100%). Most fathers (84.6%) were service  (Table 3). Attention deficit hyperactivity disorder (ADHD) was most common (46.2%) comorbidity (Table 4). Genetic study by WES of the studied participants showed that majority (76.92%) had presence of gene mutation, wherein 30.77% had likely pathogenic, 38.46% had variants of uncertain clinical significance (VOUS) and only 1 patient (7.69%) had pathogenic genetic mutation. (Table 5). The genetic mutations found in the WES are showed in Table 6.
ASD is a neurodevelopmental disorder with clinical and genetic heterogeneity. Although environmental factors play some roles but genetic factors are the key cause of ASD. With the advent of genetic test, in the last few decades significant advancement was observed in the genetic test of ASD (Tan et al., 2017). The developed country had made significant progress in WES diagnosis of ASD, while due to financial constrain countries like Bangladesh did not have genetic profile of children with ASD (Anagnostou et al., 2014;Al-Mubarak et al., 2017). With this purpose, this study was done to explore the genetic spectrum of children with ASD in Bangladesh.
In this descriptive cross sectional study mean age of participants was 5.54±3.13 years with male predominance (61.5℅). Most of the patients were from urban areas and from middle class. Due to language and communication impairment, majority of the study subject were not school going (69.2%). However, 15.4% of the study subjects went to normal school and another 15.4% went to special school. As most of the study participants were from urban area, most of the parents were literate in this study group and holding service was the main profession of most of the father (84.6%).
Epigenetic deregulation is an important factor causing autism. (Gardener, Spiegelman and Buka, 2009). In this study, the risk factors of the study subjects were evaluated. Regarding the family history, most significant risk factor was positive family history of other neurodevelopmental disorder (NDD) which was found in 64.2% of the subject. Other risk factors identified were family history of ASD, consanguinity of parents, maternal diabetes and maternal infection. Similar finding was found in related study where positive family history of NDD and maternal diabetes were risk factor of ASD (Gardener, Spiegelman and Buka, 2009   Regarding the clinical features, delayed onset of vocalization was observed in about 92.3% of the study subject. Other important parameter were delayed polysyllabic babbling, abnormal body gesture in 61.5%, delayed onset of phrases ( 69.2%), delayed onset of sentence making (69.2%) and inability to tell his/her own name (69.2%). In related studies, the core symptoms mentioned were reciprocal social deficit, communication impairment, rigid ritualistic interests. Thus, the clustering of symptoms are very importing for diagnosis of ASD along with the suggestive age (Nazeer and Ghaziuddin, 2012).
Comorbid disorders are common in cases of ASD. In this study, 46.2% of the children had ADHD, 38.5% had sleep disturbances and 30.8% had ID and 7.7% had epilepsy. In a large population based study by Mohammadi MR et al,86% of the children with ASD had comorbidities. The most common one was ID observed in 70.3%. Other comorbid conditions were epilepsy, enuresis and ADHD with prevalence rates of 29.7%, 27% and 21.62%, respectively (Mohammadi et al., 2019). WES is an efficient diagnostic tool to detect the genetic mutation of ASD (Du et al., 2018). In this study, interestingly, in more than two third of the subjects genetic test were positive. The pattern of positivity was as follows: likely pathogenic 30.77%, variant of uncertain clinical significance (VOUS) (38.46%) and pathogenic 7.69%. The pathogenic and likely pathogenic genetic mutations we found here were TENM4, ASXL1, CHD3, TSC2, CACNA1H etc.
One important gene related to ASD in this study was CHD3. Chromodomain helicase DNA binding protein 3 (CHD3) participates in the remodeling of chromatin by deacetylating histones. Two de novo missense variants and one de novo in-frame deletion variant were identified in the CHD3 gene in ASD probands following WES (Iossifov et al., 2014;Yuen et al., 2017). In the case in this study, missense heterozygous mutation was detected which was likely pathogenic. ASXL1 gene has been reported in previous studies in link with ASD (De Rubeis et al., 2014).
One case with missense mutation of TSC2 gene was identified here. Tuberous sclerosis is caused by mutations in one of two tumor suppressor genes: TSC1 (9q34) and TSC2 (16p13.3) (Maheshwar et al., 1997). One case of CACNA1C pathogenic mutationwas found here. This gene is rarely associated with ASD. In a previous study, this mutation was identified in 6 cases out of 461 individuals with ASD (Splawski et al., 2006). One case of mutation of teneurintransmembrane protein 4 (TENM4) gene was identified in this study. This geneis expressed primarily in the brain and has been detected in several transcriptomes derived from neuronal tissues. The link to ASD of this gene is very rare (Hor et al., 2015).
Other VOUS genetic mutations found in this study were HECW2, CACNA1A, ATP2C2, SH3TC2, KANK1, FOXP1 etc. One of our case had HECW2 mutation. This gene mutation is associated with ID, epilepsy, visual impairment apart from ASD (Berko et al., 2017). While, CACNA1A gene is another gene related to ASD found in one of the cases in this study. Previous study detected that a single nucleotide polymorphism (SNP) in CACNA1A confers risk to ASD (Li et al., 2015). The ATP2C2 gene is closely related to language impairment and dyslexia, some links with ASD has also been observed (Eicher and Gruen, 2015). Two cases of ATP2C2 mutation were also detected here where both had VOUS and missense mutation. There was one case of SH3TC2 gene mutation in this study. This gene encodes SH3 domain and tetratricopeptide repeats-containing protein 2 and is linked to peripheral neuropathy (Senderek et al., 2003).
Again, there was one case with KANK1 mutation in this study. This gene has potential impact on neurodevelopment. Vanzoet et al reported a case series of ASD with KANK1 mutation (Vanzo et al., 2019). Another mutation which was observed in this case series was in FOXP1. Haploinsufficiency of the forkhead-box protein P1 (FOXP1) gene leads ASD traits, intellectual disability, language impairment, and psychiatric features (Siper et al., 2017).

Conclusion
This is probably the first case series of children with ASD with whole exome sequencing. Although, the number of cases are small, but it will highlight the genetic profile of ASD children of this zone and initiate further studies in this field. The pathogenic genes detected in relation to ASD were TENM4, ASXL1, CHD3, TSC2, and CACNA1H.