Duplication of 9p24.3 in three unrelated patients and their phenotypes, considering affected genes, and similar recurrent variants

Abstract Background Recent studies suggest that duplication of the 9p24.3 chromosomal locus, which includes the DOCK8 and KANK1 genes, is associated with autism spectrum disorders (ASD), intellectual disability/developmental delay (ID/DD), learning problems, language disorders, hyperactivity, and epilepsy. Correlation between this duplication and the carrier phenotype needs further discussion. Methods In this study, three unrelated patients with ID/DD and ASD underwent SNP aCGH and MLPA testing. Similarities in the phenotypes of patients with 9p24.3, 15q11.2, and 16p11.2 duplications were also observed. Results All patients with ID/DD and ASD carried the 9p24.3 duplication and showed intragenic duplication of DOCK8. Additionally, two patients had ADHD, one was hearing impaired and obese, and one had macrocephaly. Inheritance of the 9p24.3 duplication was confirmed in one patient and his sibling. In one patient KANK1 was duplicated along with DOCK8. Carriers of 9p24.3, 15q11.2, and 16p11.2 duplications showed several phenotypic similarities, with ID/DD more strongly associated with duplication of 9p24.3 than of 15q11.2 and 16p11.2. Conclusion We concluded that 9p24.3 is a likely cause of ASD and ID/DD, especially in cases of DOCK8 intragenic duplication. DOCK8 is a likely causative gene, and KANK1 aberrations a modulator, of the clinical phenotype observed. Other modulators were not excluded.


| INTRODUCTION
The overlapped dedicator of cytokinesis 8 (DOCK8, * 611432) gene and the KN motif and ankyrin repeat domains 1 (KANK1, * 607704) gene are both found on chromosome 9 at locus 9p24.3. The DOCK8 gene encodes a member of the DOCK protein family that participates in the intracellular signaling network. The product of the DOCK8 gene is important for proper immune cell migration, synapse formation, and signal transduction (Gadea & Blangy, 2014;Su, 2010), and is also expressed in the brain, mainly in the caudate nucleus, which is engaged in proper cognitive functioning (Lin et al., 2017;Su, 2010). The KANK1 gene encodes a member of the Kank protein family which contains multiple ankyrin repeat domains. The product of the KANK1 gene functions in cytoskeleton formation by regulating actin polymerization (Gee et al., 2015).
Genetic duplications of 9p24.3, and therefore the DOCK8 and KANK1 genes, are associated with autism spectrum disorders (ASD), intellectual disability/developmental delay (ID/DD), attention deficit hyperactivity disorder (ADHD), epilepsy, learning problems, and language disorders (Engelhardt et al., 2009;Glessner et al., 2017;Griggs et al., 2008;Krgovic et al., 2018;Oikonomakis et al., 2016;Ruiter et al., 2007;Su, 2010). Loss of function of the DOCK8 and KANK1 genes, through deletion or sequence changes, causes an autosomal-recessive form of the hyper-immunoglobulin-E (or hyper-IgE) syndrome, cerebral palsy spastic quadriplegia type 2, and central nervous system (CNS) development disorders (Gee et al., 2015;Glessner et al., 2017;Lin et al., 2017). ' Population studies have shown that copy number variants (CNVs) of DOCK8 or KANK1 are not frequently detected in the general population (0.11%), although duplications of these genes are more common than deletions in the records held by the Database of Genomic Variants (DGV), the International Standards for Cytogenomic Arrays (ISCA), and the Database of Genomic Variation and Phenotype in Humans Using Ensemble Resources (DECIPHER).
Here we report findings of duplications of chromosome location 9p24.3 (i.e., the DOCK8/KANK1 genes) in three unrelated patients, which provide evidence of the pathogenic impact of this duplication in patients with ID/DD, ASD, ADHD, and macrocephaly. We also compare our results with the previously published phenotypes of patients carrying 9p24.3, 15q11.2, and 16p.11.2 duplications.

| Editorial policies and ethical considerations
According to the Code of Ethics of the World Medical Association, informed consent was obtained from the patients' parents prior to clinical data collection from the specialist and testing and processing of biological materials (Declaration of Helsinki, Supplemental Information S1, S2).

| Materials
Patients with ID/DD and ASD were referred to genetics consultation at the Department of Medical Genetics, University Hospital Olomouc by specialists from the Department of Psychiatry and Pediatrics, Olomouc University Hospital. Additionally, tests and biological material from patients meeting the clinical criteria for the testing of congenital metabolic defects (e.g., eating habits) were processed in the Laboratory for Inherited Metabolic Disorders, University Hospital and Palacký University in Olomouc.

| Methods
Genetic testing of ID/DD and ASD patients was conducted according to international recommendations (Ho et al., 2016;Richards et al., 2015). Large chromosomal aberrations (≥ 5 Mb) and aberrant repeats in the promotor of the fragile X mental retardation protein translational regulator 1 (FMR1), resulting from the testing, were excluded by conventional karyotyping and fragment analysis performed according to Zhou, 2006. CNVs were tested using single nucleotide polymorphism (SNP) array for comparative genomic hybridization (aCGH; Zhou, 2006). A Cytoscan HD instrument (Affymetrix, Santa Clara, CA, USA) was used for SNP aCGH analysis according to the manufacturer's protocols. The CHAS v1.2.2 program (Affymetrix, Santa Clara, CA, USA) was used to call CNVs. Data from this analysis were stored in the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/) under number GSE11 4870 and GSE13 2453. Identified CNVs were clinically classified using curated genetics databases (ISCA, DECIPHER, SFARI, DGV) according to established guidelines (Kearney et al., 2011;Schaefer & Mendelsohn, 2013). Additionally, multiplex ligation-dependent probe amplification (MLPA) with Probemix P385-A2 was used to more accurately determine the extent of CNV revelation by SNP aCGH in the DOCK8 gene. MLPA reactions were set up according to the manufacturer's protocols (MRC-Holland, Amsterdam, Netherlands).
All PCR products were determined by capillary electrophoresis with an ABI 3130 genetic analyzer and the Gene Mapper software (Applied Biosystems, Foster City, CA) and the Coffalyser program (MRC-Holland, Amsterdam, Netherlands) was used to call CNVs (MRC-Holland, Amsterdam, Netherlands). Tests for connexin 26 and the Prader-Willi syndrome (PWS, 15q11-13) were conducted

| RESULTS
Three unrelated patients with ID/DD and ASD carried a duplication of chromosome locus 9p24.3, which includes the DOCK8 and KANK1 genes. CNV analysis proved the inherited origin of the 9p24.3 duplication in one patient and his sibling. This observation corresponds to the 1.75% frequency of this duplication in patients with ID/DD and ASD, and no 9p24.3 duplication was detected in patients without ID/DD in the Department of Medical Genetics, University Hospital Olomouc. Normal karyotypes and numbers of CGG triplets in the FMR1 promoter were observed in all three patients. Aberration of Connexin 26 was excluded in one patient with hearing impairment and no genomic and methylation changes were shown in the 15q11-13 region for two patients. Table 1 summarizes the clinical features of these patients.

| Patient 1
Patient 1 was male, Caucasian, and of nonconsanguineous parents; the mother and father were 34 and 31 years old, respectively. The mother had been treated for epilepsy, with no reports of other monogenic diseases or neurodevelopmental disorders (ID/DD and ASD). A clinical picture of the father is unavailable. Patient 1 was the mother's third pregnancy.
The pregnancy was uneventful; delivery was spontaneous during week 36 of gestation; birth weight was 2,500 g (50th percentile), length 48 cm (60th percentile).
The boy had been treated for bronchial asthma and underwent an orchiectomy at 1 year of age. He was achieving developmental milestones normally until age 18 months when speech delay, hyperactivity, and sleep disorder were observed. Hearing impairment and ADHD were diagnosed at age 3 years. He underwent genetic testing at age 4 years for hearing impairment, ADHD, and obesity. The diagnosis of ASD with mild intellectual impairment was confirmed by a psychiatrist at age 5. Growth parameters at age 7 years were as follows: height 116 cm (3rd percentile), weight 38.6 kg (>97th percentile), BMI (body mass index) 28.7 (obese), and head circumference 54 cm (88th percentile; Figure 1). Apart from a short philtrum and broader nasal bridge, the boy exhibited no dysmorphic features (Table 1).

| Siblings of Patient 1
The boy had two older sisters. The oldest was by a different father and was 22 years old and healthy at the time of the study. She was not genetically tested because of her good health and different father. The second sister shared the same father as Patient 1 and was 11 years old at the first visit to the Department of Medical Genetics. She was born in week 38 of pregnancy; birth weight 2300 g (10th percentile), length 50 cm (50th percentile). Apart from dyslalia, her psychomotor development was uneventful up to age 4 years when ADHD and compulsive behaviors were diagnosed by a psychiatrist. At age 11 years, she had the following parameters: height 136 cm (4th percentile), weight 45.5 kg (>97th percentile), BMI 25 (overweight), and head circumference 53 cm (50th percentile; Figure 3). She was overweight and exhibited facial features slightly like those of her brother. The second sister of Patient 1 inherited the same 9p24.3 duplication (Human GRCh37/hg19 chromosome 9: 271533-440683, 169 kb) as her brother.

| Patient 2
Patient 2 was male, Caucasian, and the first child of nonconsanguineous parents; the mother and father were 29 and 38 years old, respectively. The mother had latent tetany but was being treated with medication; she also had a family history of Bekhterev's disease and Raynaud syndrome. The father had been treated for Crohn's disease. No other monogenic diseases or neurodevelopmental disorders were reported in the families of either parent. Patient 2 had no siblings, and neither parent had ID/DD and ASD. Patient 2 was the mother's second pregnancy. The mother experienced syncope and olfactory aura during the third trimester of her pregnancy, perhaps the equivalent of an epileptic seizure causing olfactory aura and syncope but without convulsions. Delivery was in gestational week 37 by Caesarean section (mother's choice); birth weight was 2950 g (60th percentile), length 50 cm (75th percentile).
Patient 2 was achieving developmental milestones normally until age 15 months when speech delay, stereotyped movements, negativistic and aggressive behaviors, and hyperactivity were observed. The diagnosis of ASD with ADHD was confirmed by a psychiatrist at age 3 years and the child was referred to genetic counselling. Growth parameters at age 3.5 years were as follows: height 102 cm (50th percentile), weight 16 kg (50th percentile), BMI 15.4 (normal weight), head circumference 53 cm (>97th percentile). The boy was suspected of having hypoacusis (not investigated at that time). He also met the clinical criteria for testing congenital metabolic disorders which were excluded (amino acid and fatty acid metabolism and organic aciduria).

| Patient 3
Patient 3 was male, Caucasian, and the first child of nonconsanguineous parents aged 30 (mother) and 31 (father) years. The mother was treated for cyclic bleeding disorders. No other monogenic diseases or neurodevelopmental disorders (ID/DD and ASD) were reported for the parents. Patient 3 was an only child and was fetus B of a multiple pregnancy. Fetus A underwent fetal death spontaneously. The delivery was without complications. His birth weight and length were unavailable.
Newborn adaptation was normal. At age 13 months, Patient 3 underwent neurologist counselling for delays in all developmental areas. Significant developmental delays, intellectual disability, mild paleocerebellar aphasia, and incontinence were observed at age 4.5 years. ASD was diagnosed by a specialist in the psychiatric department based on autistic features such as the absence of speech and lack of understanding of communication as well as avoidance of social contact. At this age, the boy weighed 14 kg (3rd percentile) and had a head circumference of 50 cm (25th percentile). No facial dysmorphic features were observed ( Figure 5). He was also referred to urology for testicular retention and to orthopedics for flat feet. Negative results were observed in magnetic resonance imaging of his brain as well as congenital metabolic disorders screening.

| DISCUSSION
Recent reports indicate that CNVs of 9p24.3, including DOCK8 and KANK1, have clinical importance (Engelhardt et al., 2009;Glessner et al., 2017;Griggs et al., 2008;Krgovic et al., 2018;Oikonomakis et al., 2016;Ruiter et al., 2007;Su, 2010). However, carriers of these CNVs presented with different phenotypes, especially between deletions and duplications (Glessner et al., 2017;Jing et al., 2014;Krgovic et al., 2018;Qin et al., 2016;Singh et al., 2017). Duplications of 9p24.3 are two times more frequent than deletions in the general population (0.07% vs 0.03%) and in subpopulations of patients with ASD and ID/DD (0.46% vs 0.23%), but these ratios are not significant (p = 0.09, p > 0.05; Vanzo et al., 2019;Krgovic et al., 2018). Deletion of 9p24.3 is accompanied by more serious phenotypes than duplication; microcephaly and ADHD, specifically, appeared more frequently in carriers of  Table 3; DECIPHER). In our study, we observed one patient and his sister who inherited 9p24.3 duplication from their father, but the father was unavailable for psychiatric examination and no evidence about his health exists. Inheritance information was also unavailable for the remaining patients we observed.
We further analyzed the genetic material from our patients using a more detailed method (MLPA) since SNP aCGH cannot detect intragenic changes. MLPA revealed partial duplication of DOCK8 in all patients observed. We hypothesized that partially overlapping duplications of DOCK8 might interrupt a reading frame. Only five cases of intragenic duplication of DOCK8 were described in the DECIPHER v9.21 database, and these patients' phenotypes included ID/ DD (3/5, 60.00%), ASD (1/5, 20.00%), and no macrocephaly, microcephaly or ADHD (DECIPHER v9.21).
Indeed, other modulatory pathways were not excluded and some of these may be similar to 15q11.2 and 16p11.2 duplications, for example, dosage-sensitive pathways, the presence of a protective allele, sequence changes, and the impact of duplicated noncoding genes (Benítez-Burraco et al., 2017;Jing et al., 2014;Lee et al., 2015;Ulfarsson et al., 2017;Zwaag et al., 2009). We emphasize the importance of future analyses, particularly messenger ribonucleic acid/protein expression studies and more detailed methods for revealing the start and end of duplications of 9p24.3 and other loci.

| CONCLUSION
This study provides evidence of the pathogenic role of the 9p24.3 duplication. Additionally, we observed similar clinical phenotypes in carriers of the 9p24.3, 15q11.2 and 16p11.2 duplications in population frequency, frequency in ASD and ID/DD patients, microcephaly, and macrocephaly. We also observed a stronger association of the ID/DD phenotype with the 9p24.3 duplication than with the 15q11.2 and 16p11.2 F I G U R E 6 Graphical display of significant differences in phenotype between 9p24.3, 15q11.2, and 16p11.2 duplications and 9p24.3, 15q11.2, and 16p11.2 deletions. The data for comparison are presented in Table 3 and were analyzed using the Fisher exact test with a p value of 0.05 representing statistical significance duplications. Moreover, we accede to the idea that DOCK8 is a likely causative gene of the clinical phenotypes observed and concluded that KANK1 aberrations may modulate these phenotypes. Finally, we emphasize the need for detailed genetic testing in patients with duplications of 9p24.3 as well as of 15q11.2 and 16p11.2.