Monogenic defects in Russian children with autism spectrum disorders

Abstract

BACKGROUND

Autism spectrum disorders (ASD) represent a substantial social problem affecting at least 1 in 100 children worldwide. These conditions are very often accompanied by intellectual disability (ID) and speech delay; thus, they can be considered within a clinical continuum of neurodevelopmental disorders. Given the high heterogeneity of ASD, the subjective nature of diagnostic criteria, and the presence of phenocopies, identifying genetic determinants of these disorders remains a challenge.

AIM

To investigate the spectrum and frequency of rare genetic variants in genes with proven association with ASD in Russian children.

METHODS

110 patients from 106 families were recruited into the study (mean age at diagnosis 6 years; boy-to-girl ratio 3:1. Most of the patients (84%) demonstrated a combination of ASD with developmental delay (DD) or ID. Patients with syndromic features were subjected to the chromosomal microarray analysis. The remained children underwent clinical exome sequencing aimed at identifying presumably monogenic causes of ASD. The study focused on rare (minor allele frequency ≤ 0.001) variants affecting high-confidence ASD-associated genes.

RESULTS

Pathogenic copy number variations were detected in three (7%) of the patients examined. Clinical exome sequencing revealed pathogenic/likely pathogenic variants in 12 of 105 cases (11%), indicating the presence of monogenic syndromes with established clinical significance (Pitt-Hopkins syndrome, ZTTK syndrome, syndromic X-linked ID of Billuart type, Snijders-Blok-Campeau, Helsmoortel-van der Aa, Coffin-Siris, Clark-Baraitser, Keefstra syndromes, etc.). In addition, 27 patients (26%) had 37 rare variants of unknown clinical significance in DSCAM, SHANK2, AUTS2, ADNP, ANKRD11, APBA2, ARID1B, ASTN2, ATRX, SCN1A, CHD2, DEAF1, EHMT1, GRIN2B, NBEA, NR4A2, TRIO, TRIP12, POGZ, EP300, FOXP1, PCDH19, GRIN2A, NCKAP1, and CHD8 genes. No specific variant was detected more than once in unrelated patients. Among the genes with rare variants found in 2 or more patients were TRIP12 (n = 4), AUTS2 (n = 3), ARID1B (n = 3), PCDH19 (n = 3), EP300 (n = 3), TRIO (n = 2), ASTN2 (n = 2), EHMT1 (n = 2), and CHD2 (n = 2). Of note, 5 male ASD/DD patients from 3 unrelated families had PCDH19 missense variants, confirming that at least some hemizygous males with non-mosaic PCDH19 variants may present with neurobehavioral abnormalities. These variants did not cause epilepsy restricted to females in patients’ mothers or sisters.

CONCLUSION

These data confirm a tremendous diversity of genetic causes of ASD. Clinical exome sequencing may serve as a reasonable alternative to whole-exome sequencing.

Key Words: Autism; Autistic spectrum disorders; Mutation; High-throughput sequencing; Intellectual disability; Neurodevelopmental disorders; Developmental delay; Mental retardation; PCDH19; Chromosomal microarray

Core Tip: While autism is a clinical diagnosis, genetic studies provide important clues on autism spectrum disorders (ASD) pathogenesis. From a practical point of view, DNA testing offers an opportunity to obtain valuable information on genetic risks and, sometimes, on the most effective treatment. The number of ASD-associated candidate genes exceeds 1000, yet often the causal role of a particular gene or allelic variant stays unproven. We utilized clinical exome sequencing for the DNA testing of ASD patients. The results obtained suggest that in children with ASD and developmental delay/mental retardation, the diagnostic yield of singleton clinical exome sequencing is comparable to that of singleton whole exome sequencing. Also, we assume that rare PCDH19 variants may play a role in causing autistic features in males.

INTRODUCTION

Autism spectrum disorders (ASD) affect at least 1% of children worldwide[1]. Heritability of these conditions is very high, being within the range 80%-90%[2,3]. ASD diagnosis requires the presence of deficits in social communication/interaction and restricted, repetitive patterns of behavior, interests or activities[4]. Patients with ASD often suffer from many comorbidities like intellectual disability (ID), speech delay, epilepsy, attention-deficit hyperactivity disorder, anxiety, sleep disorders, gastrointestinal problems, etc.[5]. Genomic architecture of autism is extremely complex and includes both rare events with strong effects [single nucleotide variations and copy number variations (CNV)] and multiple common variants providing mild predisposition[6]. Moreover, these events may have an additive effect when inherited together[7]. Many autism genes are shared with other neurodevelopmental disorders reflecting common pathogenetic mechanisms. These genes have important functions in neurogenesis, chromatin remodeling, and metabolic pathways. Most rare ASD-related genetic variants arise de novo, although some population-specific peculiarities of their spectrum may exist. While many countries have already done large-scale projects recruiting thousands of ASD patients, Russian population remains relatively understudied. Diagnosis of ASD is based on behavioral patterns, but DNA testing may provide valuable information on genetic risks and, sometimes, on effective treatment. It should be noted that there is a growing demand from families for more accurate evaluation of ASD recurrence risk in offspring. In a real-world setting this practical query is often far from being answered, facing ambiguous candidate genes, variants of unknown significance, incomplete penetrance, etc.

The list of the genes potentially associated with ASD is constantly expanding, yet there is little consensus on which genes should be tested routinely. There are ongoing efforts for selecting the genes demonstrating sufficient evidence for their causal role in ASD[8-10]. To help this curation, Simons Foundation Autism Research Initiative (SFARI) database provides ranking systems [SFARI Gene Score and Evaluation of Autism Gene Link Evidence (EAGLE) Score]. These scores allow genes to be ranked into several categories according to the strength of evidence for their link with autism.

Here we report our experience of detection of rare variants in genes with high SFARI and EAGLE integral scores (i.e., high confidence ASD-related genes) in predominantly simplex ASD families. We opted for clinical exome sequencing as a diagnostic method due to cost considerations, and this approach enabled us to focus on a relatively small number of genes with strong links to ASD.

MATERIALS AND METHODS

The study included 110 children with ASD from 107 families. All patients had impaired social interaction and social communication, as well as stereotypical behavior, limited interests and hobbies. Patients were selected based on the results of consultation with a clinical geneticist and/or pediatric psychiatrist; the diagnosis was made in accordance with the standardized criteria provided by the American Psychiatric Association's Diagnostic and Statistical Manual, Fifth Edition[4].

The average age of patients was 6 years (range 2-18 years). 84 of the examined children (76.4%) were males, and 26 (23.6%) were females. Familial (multiplex) ASD cases included 6 patients from 3 families: One family had two affected monozygotic twins, one family had two siblings of different gender, and another one was characterized by two affected male first cousins. None of the parents reported consanguinity.

Slavic ethnic origin was reported by most of the study participants (96/110, 87%). In addition, this data set included Armenians (n = 3), Avars (n = 2), Dargins (n = 3), Chechens (n = 2), and Lezgins (n = 4).

17 patients (16%) had an isolated ASD, and 93 patients (84%) demonstrated a combination of ASD with developmental delay (DD, n = 82) or (ID, n = 11).

Informative facial dysmorphisms and/or other signs of a possible hereditary syndrome (micro- or macrocephaly, epilepsy, congenital heart disease, etc.) were observed in 18 patients (16%). Of those, Fragile X syndrome was suspected on clinical grounds in two male sibs; Fragile X Messenger Ribonucleoprotein 1 repeat analysis confirmed this diagnosis, therefore, these patients were excluded from further analysis.

Cytogenetic and molecular karyotyping studies

All syndromic patients underwent standard karyotyping, as well as chromosomal microarray analysis (CMA; Genoscan3000, Genomed, Russia) with minimum resolution of 100 kb. Another 31 ASD patients underwent CMA test prior to recruitment to the study, so they were already known to have no relevant cytogenetic abnormalities. The results have been re-examined by a board-certified cytogeneticist.

Clinical exome sequencing

Massively parallel sequencing was performed on a Genolab M device (GeneMind, China) with at least 100-fold coverage. The sample preparation process included enzymatic DNA fragmentation and adapter ligation. For enrichment, the Clinical Exome Sequencing probe set (TruSight™ One Sequencing Panel, Illumina, 4813 genes) was utilized. Sequencing was performed using paired reads of 150 bp. Sequencing accuracy was assessed by the number of reads with a high-quality score (Phred Quality Score, or Q score). The BWA (Burrows-Wheeller Aligner) algorithm was used for the alignment procedure to the reference genome. The resulting alignment files in “.bam” format were processed using the GATK v.2.7 software suite (Genome Analysis Tool Kit and Picard v.1.86) to identify all differences from the reference sequence (variants). Annotation of the identified variants was performed using the Annovar resource (http://www.openbioinformatics.org/annovar/).

Variant prioritization

Our analysis was intentionally limited to a list of highly confident ASD-associated genes. Gene selection was based on Simons Foundation Autism Research Initiative (SFARI) and EAGLE integral scores (https://gene.sfari.org/database/; latest Release 2024 Q4) reflecting the degree of evidence for association with ASD. When assessing the significance of variants, we considered information from publicly available databases (Franklin, OMIM, ClinVar). Combined annotation dependent depletion (CADD) score ≥ 20 was taken as a threshold for the prediction of functional significance of missense variants.

Variant prioritization was based on the following criteria: (1) Population frequency of ≤ 0.001 or not reported; (2) Variant affects ASD-associated gene and is classified as pathogenic or likely pathogenic according to the American College of Medical Genetics and Genomics (ACMG) and/or ClinVar criteria; (3) Missense variants affecting any of the 92 genes with EAGLE score of > 12 (high confidence) included into the diagnostic panel; (4) Likely gene damaging (LGD) event affecting any ASD-associated gene; and (5) Missense variants affecting any ASD-associated gene with integral CADD score of ≥ 30.

In addition, the possibility of a relationship between the phenotype and the presence of pathogenic/Likely pathogenic variants in genes that were not previously associated with ASD, was taken into account.

Whenever possible, the presence of the identified variants was analyzed in children`s parents by Sanger sequencing. Several possible mechanisms of ASD heritability were considered. In the most common scenarios, the disease is either associated with de novo variants absent in healthy parents or fits into an autosomal recessive or X-linked inheritance model. Alternatively, autosomal dominant inheritance with incomplete penetrance also cannot be completely excluded: Indeed, the presence of a rare heterozygous variant in a presumably healthy parent does not necessarily contradict with its potential causative role[11].

RESULTS

Key findings of the study are summarized in the Figure 1. Molecular cytogenetic analysis revealed pathogenic CNV in three patients (7% of 41 syndromic cases studied). A 2 years 7 months boy (karyotype 46, XY) and an 8-year-old girl [45, XX r(22), ring chromosome 22] were found to have heterozygous deletion 22q13.33, which includes the SHANK3 Locus and is associated with Phelan-McDermid syndrome. Along with severe DD and ASD, these patients had multiple minor congenital anomalies (deformed ears, short sunken nasal bridge, hypertelorism, epicanthus, fifth fingers clinodactyly, etc.). Muscular hypotonia and a high pain threshold for injuries were also noted. In addition, the girl had pronounced hypertrichosis of the back and limbs, and persistent self-harming behaviour (biting her wrists and fingers, resulting in chronic traumatization).

Figure 1 Key findings of the study. ASD: Autism spectrum disorders; CMA: Chromosomal microarray analysis; CES: Clinical exome sequencing; LP: Likely pathogenic; P: Pathogenic; CNV: Copy number variations; VUS: Variants of unknown clinical significance; FMR1: Fragile X Messenger Ribonucleoprotein 1; FRAXA: Fragile X syndrome.

A 16-year-old girl was found to have a combination of a pathogenic microduplication 1p36.33-p.36.32 of 2.6 Mbp with a 11p15.5 microdeletion of 424080 bp of unknown clinical significance. Multiple facial dysmorphisms were accompanied by autistic behavior and learning difficulties (difficulty expressing thoughts, attention deficit, poor memory, underdeveloped abstract thinking).

The remaining 105 patients underwent clinical exome sequencing. The results of the molecular testing are presented in Table 1.

Table 1 Pathogenic/Likely pathogenic variants.

ID	Gender	Age at diagnosis	Phenotype	Mutation detected	Pathogenicity/disease	Segregation analysis	EAGLE gene score	SFARI gene score	Other rare variants detected
2639	Female	2 years	ASD, DD; frontal bossing, flat nasal bridge, strabismus, arachnodactily	TCF4, NM_001083962.2 c.1841C>T, (p.Ala614Val), rs1568303352	P (# 610954 PITT-HOPKINS SYNDROME)	De novo	13.5	1s	-
3942	Female	17 years	ASD, severe ID, epilepsy, CHD (pulmonary stenosis), smooth philtrum, broad nasal bridge; insensitivity to pain	ADNP NM_001282531.3 c.2157C>G (p.Tyr719*) rs587777526	P (# 615873 HELSMOORTEL-VAN DER AA SYNDROME)	De novo	41.5	1s	-
4027	Male	3 years 10 months	ASD, DD; mild dysmorphic features	TRIP12 NM_001348323.3 c.3038_3041dup (p.Leu1014Phefs*37)	LP (# 617752 CLARK-BARAITSER SYNDROME)	De novo	27	1s	-
4071	Male	2 years 1 month	ASD, DD; congenital hypothyroidism, strabismus, muscular hypotonia, tall forehead, scaphocephaly, flat nasal bridge, short nose with anteverted nostrils, low-set ears	SON NM_138927.4 c.5753_5756 del (p.Val1918Glufs*87)	P (# 617140 ZTTK SYNDROME)	De novo	11	1s	-
4072	Male	3 years 9 months	ASD, DD	SLC9A9 NM_173653.4 c.8_9del (p.Arg3Thrfs*10) rs750792945	LP (# 613410 AUTISM, SUSCEPTIBILITY TO, 16; AUTS16)	ND	-	2	-
4217	Female	3 years 9 months	ASD, DD	ARID1B NM_001374828.1 c.1293_1311del (p.Gly434Alafs*12) rs943407609	P (# 135900 COFFIN-SIRIS SYNDROME 1)	De novo	34.75	1s	TRIP12 NM_001348323.3 c.4426G>A (p.Val1476Met) rs781311402
4232	Female	7 years 3 months	ASD, DD; strabismus, low-set ears, micrognathia, low anterior hairline	SERPINI1 NM_001122752.2 c.553T>G (p.Tyr185Asp)	LP (# 604218 ENCEPHALOPATHY, FAMILIAL, WITH NEUROSERPIN INCLUSION BODIES)	ND	-	-	DIP2A NM_015151.4 c.2626C>T (p.Arg876Cys) rs199807759
4259	Male	6 years 6 months	ASD, ID; cerebellar vermis hypoplasia, tetraventricular dilatation	TUBB3 NM_006086.4 c.1172G>A (p.Arg391His) rs886039497	P/LP (# 614039 CORTICAL DYSPLASIA, COMPLEX, WITH OTHER BRAIN MALFORMATIONS 1)	ND	-	-	CACNA1D NM_001128840.3 c.5017G>A (p.Glu1673 Lys) rs147973409 CACNA1D NM_001128840.3 c.5377C>T (p.Arg1793Trp) rs555675934
4386	Male	6 years	ASD, ID; epilepsy, hydrocephalus.Triangular face, beaked nose, protruding ears, thin upper lip	OPHN1 NM_002547.3 c.644_645del (p.Val215Glyfs*35) rs1569244467 hemizygous	P (# 300486 INTELLECTUAL DEVELOPMENTAL DISORDER, X-LINKED, SYNDROMIC, BILLUART TYPE)	Maternal	-	2	-
4436	Male	16 years	ASD, ID; hypertelorism	EHMT1 NM_024757.5 c.732del (p.Phe244 Leufs*38)	LP (# 610253 KLEEFSTRA SYNDROME 1)	De novo	13.5	1s	EP300 NM_001429.4 c.4256T>C (p.Ile1419Thr) rs1278019392 paternal. RAI1 NM_030665.4 c.4340G>C (p.Arg1447Thr) rs767484843 maternal
KM	Male	15 years	ASD, ID; macrocephaly, frontal bossing hypertelorism	CHD3 NM_001005273.3 c.5642G>T (p.Arg1881 Leu) rs1567877108	P/LP (#618205 SNIJDERS-BLOC-CAMPEAU SYNDROME)	De novo	-	1s	-
ShV	Male	5 years	ASD, DD; macrocephaly, epilepsy	PIK3CA NM_006218.4 c.23G>A (p.Gly8Asp)	LP (# 615108 COWDEN SYNDROME 5)	De novo	-	3	-

ASD: Autism spectrum disorders; DD: Developmental delay; ID: Intellectual disability; LP: Likely pathogenic; P: Pathogenic; ND: No data.

In total, pathogenic/Likely pathogenic variants were found in 12 of 105 (11%) patients assessed by clinical exome sequencing. Notably, 4 patients also carried additional rare alleles of unknown significance in high-confidence ASD genes (Table 1). Pathogenic/Likely pathogenic variants were detected in the majority of syndromic (9 of 13, 69%) children compared to only 3 of 92 (3%) nonsyndromic patients (P = 0.0000).

In 6 cases, the detected variants were clearly considered pathogenic by ACMG criteria, which made it possible to confidently establish the diagnosis of a monogenic condition (Pitt-Hopkins syndrome, ZTTK syndrome, syndromic X-linked ID, Billuart type, Snijders-Blok-Campeau, Helsmoortel-van der Aa, and Coffin-Siris syndrome).

In another 6 cases, likely pathogenic variants in TRIP12, SLC9A9, SERPINI1, TUBB3, EHMT1, and PIK3CA genes were detected. In four patients (4027, 4072, 4259, 4436) molecular findings corresponded well to the phenotype, leading to a diagnosis of Clark-Baraitser syndrome, Autism susceptibility type 16, Complex cortical dysplasia with other brain malformations type 1, and Kleefstra syndrome type 1, respectively. A 5-year-old boy (Patient ShV) with macrocephaly and epilepsy had germline PIK3CA variant. Analysis of the DNA obtained from his parents revealed that this variant arose de novo; this fact, together with clinical findings, let us reclassify this variants with unknown clinical significance (VUS) to LP variant and make a diagnosis of Cowden syndrome type 5. Of note, PIK3CA gene has SFARI gene score 3 (Suggestive Evidence).

A 7-year-old girl (Patient 4232) carried likely pathogenic allele c.553T>G in the SERPINI1 gene. The associated neurodegenerative condition, familial encephalopathy with neuroserpin inclusion bodies, typically manifests from the third to fifth decade of life; thus, presymptomatic diagnosis of this condition cannot be excluded. SERPINI1 defects have never been described in association with ASD.

Segregation analysis was performed in 9 families, in 8 of which the de novo status of the mutation was proven, and in one case (Patient 4386) an X-linked variant was inherited from a healthy mother.

In another 27 patients (27/105, 26%), 37 VUS in DSCAM, SHANK2, AUTS2, ADNP, ANKRD11, APBA2, ARID1B, ASTN2, ATRX, SCN1A, CHD2, DEAF1, EHMT1, GRIN2B, MBD5, NBEA, NR4A2, TRIO, TRIP12, POGZ, EP300, FOXP1, PCDH19, GRIN2A, NCKAP1, and CHD8 genes were detected. 8 patients carried two, and 1 patient had 3 rare VUS. Details of the identified variants are provided in Supplementary Table 1.

Three LGD variants, which are not currently classified as pathogenic/Likely pathogenic, were identified. These variants affected YY1, ATRX and GRIN2A genes, being found in 1 patient each. 23 missense variants in genes with EAGLE score of > 12 with and integral CADD predictive score of ≥ 20 were detected in 18 patients. 6 missense variants with CADD score of ≥ 30 detected in 6 patients; they involved SCN1A, APBA2, ASTN2, NR4A2, and NBEA genes.

No specific variant was detected more than once in unrelated patients. Among the genes with rare variants found in 2 or more patients were TRIP12 (n = 4), AUTS2 (n = 3), ARID1B (n = 3), PCDH19 (n = 3), EP300 (n = 3), TRIO (n = 2), ASTN2 (n = 2), EHMT1 (n = 2), and CHD2 (n = 2).

Interestingly, non-mosaic hemizygous missense variants of PCDH19 were identified in five male ASD patients (mean age at diagnosis 6 years). In particular, such variants were detected in two unrelated families (two first cousins and two monozygotic twins, respectively) as well as in one simplex case (Patient 4123). All these patients had both ASD and DD. All PCDH19 candidate variants were maternally inherited and located within extracellular (EC) domains EC3 or EC5 (exon 1).

Family 1. Monozygotic twin brothers (patients 4448, II-1, and 4449, II-2), aged 5 years, were diagnosed with ASD and DD. Patient II-2 also had a history of epileptiform activity on EEG. Genetic testing revealed a hemizygous PCDH19 variant NM_001184880.2 c.841G>A (p.Val281Ile) classified as a VUS. Their mother was a healthy heterozygous carrier of the same variant. Their younger male sibling also carried the hemizygous PCDH19 c.841G>A variant but exhibited no neurological or psychiatric symptoms (Figure 2). He was considered as having low ASD risk upon M-CHAT-R testing. The variant has not been reported in gnomAD.

Figure 2 Family trees. A: Pedigrees of families 1; B: 2 with rare PCDH19 variants. Individuals with autism spectrum disorders depicted as black boxes.

Family 2. Two male first cousins (Patients 3897, III-1 and 3851, III-2), aged 5 and 6 years, respectively, presented with ASD and DD. Both carried rare hemizygous PCDH19 variant NM_001184880.2 c.1552C>G (p.Leu518Val) classified as a VUS. Their mothers (II-1 and II-3) and the younger sister of the patient 3851, III-3, were heterozygous carriers but exhibited no neurological or psychiatric symptoms (Figure 2). This variant is reported in gnomAD Exomes with frequency of 0.0000055 (one heterozygote found in 114184 females; no hemizygotes reported).

Simplex case 4123. Patient 4123, an 8-year-old male, was diagnosed with moderate ID, structural epilepsy, and ASD-like symptoms. Genetic analysis identified a previously undescribed PCDH19 variant c.1651G>T (p.Val551Phe) in hemizygous state. The variant was also detected in the patient’s mother and sister, none of whom ever exhibited neurological or psychiatric abnormalities. The patient also carried EP300 c.3031G>A (p.Glu1011 Lys) variant of unknown significance inherited from his mother.

Besides the abovementioned two multiplex families with PCDH19 variants, the study included only one family with two siblings affected. Discordance between the siblings was found: The 5-year boy had MBD5 c.83G>A (p.Arg28His) variant, which was absent in his 7-year-old sister.

DISCUSSION

This is the first systematic next generation sequencing-based study of the genetic causes of ASD in Russian patients. It confirms a striking diversity of ASD-associated alleles and suggests a role for the PCDH19 gene in causing autistic features in males.

The trajectory of the genetic testing of subjects with ASD is a subject of debate. While most professional societies suggest starting analysis of ASD genetics from CMA[12,13], some researchers advocate using whole exome sequencing (WES) as a first-tier test[14]. The efficiency of detecting genetic causes of ASD by this method falls within 8%-26%[15]. Diagnostic efficiency may be substantially increased with the use of WES in the trio format[16], however, this approach is associated with additional costs.

The diagnostic yield of DNA testing varies greatly depending on the characteristics of the patients. Overall, causative genetic defects are more often detected in patients with a combination of ASD and ID[17,18]. While the probability of finding a disease-causing variant is generally higher in WES compared with targeted panels, recent meta-analysis suggests that this difference is not significant[19].

Although the total number of ASD candidate genes is rapidly growing and approaches up to 1000[20], reliable data on ASD-causality are often lacking. So we decided to limit our search by a relatively humble list of high-confidence ASD genes that may be used in clinical rather than research settings. This restriction was aimed at more accurate risk assessment and making genetic counseling more understandable for the families.

Our results confirm that causative variants are more likely to be detected in syndromic than in nonsyndromic ASD, as demonstrated by others[21,22]. The yield of clinical exome sequencing in our study (11% of pathogenic/Likely pathogenic variants) may be directly compared with the one obtained by Chérot et al[23] who detected pathogenic/Likely pathogenic variants in 4% of ASD patients using the same enrichment kit. Moreover, our data demonstrate approximately the same yield as singleton exome sequencing[22,24].

Of three families with multiplex ASD patients, two, including one with monozygotic twins, demonstrated concordance of genetic events; brother and sister from another family were discordant. Contrary to conventional wisdom, non-sharing of putatively causative SNVs between affected siblings has been reported by several researchers[25,26]. For example, data obtained by Yuen et al[26] demonstrated that 69% of affected siblings did not share the same ASD-related variant(s).

Rare hemizigous variants of the PCDH19 were found in two monozygous twin boys and two male first cousins with ASD as well as in one simplex case. Although in one family a hemizygous boy was unaffected, this finding remains interesting. The PCDH19 gene encodes protocadherin-19, a protein primarily expressed in the brain. It plays a crucial role in cell adhesion through homophilic binding, i.e., interacting only with other protocadherin-19 molecules on adjacent cells. This process is crucial for the proper formation of neuronal connections and the stability of neural networks. Conversely, dysfunction of PCDH19 can lead to the disorganization of neural networks[27]. Heterozygous mutations in PCDH19 are known to cause severe X-linked epilepsy restricted to females (EFMR). This condition follows an unusual X-linked inheritance pattern, where clinical manifestations are observed in heterozygous females while hemizygous males are described as “asymptomatic carriers”[28]. Severe phenotypic expression in females is mainly attributed to cellular interference resulting from random X-chromosome inactivation. A similar effect occurs in mosaic males, where both mutant and wild-type cells coexist in the brain causing symptoms resembling those in females[29]. Hemizygous non-mosaic males were long thought to be unaffected, but recent findings suggest that rare PCDH19 variants may be associated with ASD[30–32] and intellectual deficiency/mental retardation[33]. Moreover, Pcdh19 knock-out mouse models exhibited autism-like behaviors in males, including impaired social interaction, repetitive behaviors, and heightened anxiety[34].

Previously, variants in non-mosaic affected males were found both in EC[30] and in the cytoplasmic domain[31,32]. All the variants detected in affected boys included in this study were located in the EC domains EC3 and EC5 of exon 1. Notably, most pathogenic PCDH19 variants associated with epilepsy are also found in exon 1[35,36], almost half of those being located in the EC3 and EC4 domains[37].

Our findings confirm the observations that at least some hemizygous males with non-mosaic PCDH19 variants may present with neurobehavioral abnormalities. Evidently, such variants do not cause EFMR phenotype in patients’ mothers or sisters.

Our study has some limitations. Firstly, a targeted singleton approach has been used instead of whole-exome trio sequencing, thus restricting the opportunity for discovery of new ASD candidates and putting aside the majority of non-coding variants.

Secondly, CNV analysis was performed only in a minor fraction of our ASD patients; thus, we focused our efforts exclusively on rare monogenic/oligogenic events. In contrast, the only study of Russian ASD patients available in PubMed[38] is based on case-control study of CNV burden and common SNP. We are aware of the possibility of CNV and SNV co-occurrence in the same individual, providing additive effect[39]. Data on rare SNVs detected in Russian patients are currently limited to a single disease entity[40] or to a clinically special subgroup, e.g., autism with regression[41]. Thirdly, the study involved a substantial number of hospitalized patients, i.e. subjects with severe clinical manifestations of the disease. As a result, there was a recruitment bias towards patients with both ASD and ID, which may have affected the diagnostic yield.

It is highly likely that in the near future, whole-genome trio sequencing will largely replace other approaches to the genetic examination of patients with suspected monogenic disease. However, this approach is not compatible with resources currently available, and the search for cost-efficient alternatives remains a valuable avenue for clinical investigations.

Even the use of relatively small gene panels reveals many VUS which can rarely be reliably reclassified by bioinformatics tools. It is unlikely that every VUS can be subjected to a meaningful functional study; so, case-control studies as well as segregation analysis will remain the mainstream. Analysis of different ethnic groups is also of potential value. This study mainly included ethnic Russians and, like many others, did not reveal significant population-specific features of ASD. However, analysis of residents of more isolated ethnic communities, such as some of the large autonomous regions of the North Caucasus, could potentially provide new insights.

CONCLUSION

Clinical exome sequencing may serve as a more accessible alternative to WES in diagnosis of well-established genetic causes of ASD. Mutation-negative patients could be further analyzed by WES or whole-genome sequencing to reveal new ASD candidates.