2024; 22(4): 635-645  https://doi.org/10.9758/cpn.24.1188
Epigenetic Insights into Autism Spectrum Disorder: DNA Methylation Levels of NR3C1, ASCL1, and FOXO3 in Korean Autism Spectrum Disorder Sibling Pairs
Miae Oh1, Nan-He Yoon2, Soon Ae Kim3, Hee Jeong Yoo4,5
1Department of Psychiatry, Kyung Hee University Hospital, Seoul, Korea
2Division of Social Welfare and Health Administration, Wonkwang University, Iksan, Korea
3Department of Pharmacology, School of Medicine, Eulji University, Daejon, Korea
4Department of Psychiatry, Seoul National University Bundang Hospital, Seongnam, Korea
5Department of Psychiatry, Seoul National University College of Medicine, Seoul, Korea
Correspondence to: Hee Jeong Yoo
Department of Psychiatry, Seoul National University Bundang Hospital, Seoul National University College of Medicine, 82 Gumi-ro 173beon-gil, Bundang-gu, Seongnam 13620, Korea
E-mail: hjyoo@snu.ac.kr
ORCID: https://orcid.org/0000-0003-0521-2718

Soon Ae Kim
Department of Pharmacology, School of Medicine, Eulji University, 77 Gyeryong-ro 771beon-gil, Jung-gu, Daejeon 34824, Korea
E-mail: sakim@eulji.ac.kr
ORCID: https://orcid.org/0000-0002-9831-0511
Received: March 22, 2024; Revised: May 13, 2024; Accepted: May 14, 2024; Published online: July 18, 2024.
© The Korean College of Neuropsychopharmacology. All rights reserved.

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
Objective: Previous research on autism spectrum disorder (ASD) in Koreans has primarily focused on genetic diversity because of its high heritability. However, the emerging recognition of transgenerational epigenetic changes has recently shifted research attention towards epigenetic perspectives.
Methods: This study investigated the DNA methylation patterns of the promoter regions of candidate genes such as NR3C1, ASCL1, and FOXO3 in blood samples from ASD probands and their unaffected siblings. The analysis included 54 families (ASD proband group: 54; unaffected biological sibling group: 63). The diagnostic process involved screening the probands and their siblings for ASD based on the Diagnostic and Statistical Manual of Mental Disorders 5th edition. Intelligence, social ability, and medical history were thoroughly assessed using various scales and questionnaires. Genomic DNA from blood samples was analyzed using a methylation-sensitive quantitative polymerase chain reaction to examine the DNA methylation status of candidate genes.
Results: Methylation levels in candidate gene promoter regions differed significantly between the proband and sibling groups for all candidate genes. Correlation analysis between the proband and sibling groups revealed strong and significant correlations in NR3C1 and ASCL1 methylation. Additionally, in the analysis of the relationship between DNA and ASD phenotypes, FOXO3 methylation correlated with social quotient in probands, and ASCL1 methylation was associated with nonverbal communication, and daily living skills as measured by the Korean Vineland Adaptive Behavior Scale. Notably, ASCL1 methylation was significantly associated with parental age at pregnancy.
Conclusion: This study proposes DNA methylation of NR3C1, ASCL1, and FOXO3 in peripheral blood samples is a potential epigenetic biomarker of ASD.
Keywords: Autism spectrum disorder; NR3C1; FOXO3; ASCL1; DNA methylation
INTRODUCTION

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by communication, social interaction, and the presence of repetitive and stereotyped behaviors, interests, and activities [1]. The Centers for Disease Control and Prevention reports that approximately one in 36 children in the United States is diagnosed with ASD [2]. The prevalence of ASD is increasing annually, suggesting a complex etiology involving both genetic and environmental factors. While numerous studies have identified genetic variations and rare de novo mutations associated with an increased risk of ASD, a clear genetic cause is identified in only about 10−20% of cases [3,4]. However, the increasing prevalence of ASD cannot be solely attributed to genetic factors. Conse-quently, many researchers are focusing on non-genetic or environmental factors, including advanced parental age, preterm birth, exposure to stress, pollutants, and environmental hormones, as well as prenatal exposure to alcohol, tobacco, or drugs and low socioeconomic status [5,6].

Epigenetic approaches are increasingly used to elucidate the growing incidence of neurodevelopmental disorders, which are thought to be influenced by evolving environmental factors [7]. Recent research indicates that, ASD could potentially involve an epigenetic component. These epigenetic variations, encompassing DNA methylation changes, histone modifications, chromatin remodeling, and the role of non-coding RNAs, have the potential to alter gene expression and are implicated in ASD development [8]. Investigations have particularly focused on valproic acid, a histone deacetylase inhibitor, in both animal and clinical research related to ASD. Animal studies have shown that exposure to external chemicals like valproic acid and trimethyltin can increase cortisol levels in the cortex. Moreover, it is reported that exposure to these substances during the perinatal period could lead to neurodevelopmental abnormalities [9,10].

Cortisol, a steroid hormone, plays a critical role in fetal organ and brain development, and its modulation is influenced by stress exposures and environmental conditions [11]. The NR3C1 gene is essential as it encodes the glucocorticoid receptor, a key regulator of the stress response, which significantly influences epigenetic alterations [12]. Early life stressors, including prenatal maternal stress, can lead to epigenetic modifications, such as DNA methylation changes in the NR3C1 gene. Importantly, NR3C1 methylation mediates the effects of early adversity on the development of internalizing behavioral problems and is associated with psychopathology in young children [13]. Additionally, a correlation between NR3C1 methylation and elevated morning cortisol levels has been observed in adolescents [14]. It has been suggested that excessive glucocorticoid exposure, especially late in gestation, may contribute to reduced early growth and affective dysfunction in adulthood. This effect is thought to occur through increased corticotrophin-releasing hormone levels in the amygdala and alterations in hippocampal corticosteroid receptor levels [15]. Studies have demonstrated that glucocorticoids activate FoxO3 transcription in murine myotubes through multiple glucocorticoid response elements, involving chromatin structural changes and DNA looping [16]. Changes in the constitutive expression of FoxO3 are linked to disrupted neural stem cell homeostasis, impaired primary neurosphere formation, and the loss of neural progenitor cells (NPCs) in the developing murine brain [17]. Additionally, FoxO3 has been reported to share common targets with the proneuronal basic helix-loop-helix transcription factor achaete-scute complex-like 1 (ASCL1) in NPCs. FOXO3 is known to repress the expression of a subset of ASCL1 neurogenic targets, thereby restraining neurogenesis both in vitro and in vivo [18]. Ascl1, a key regulator of neurogenesis in the mammalian brain, has been demonstrated in gain- and loss-of-function analyses to be essentials for promoting neurogenesis. It is widely used in protocols to reprogram somatic cells into induced neurons [19,20].

These candidate genes have been proposed to be involved in neurodevelopment in cellular and animal models. However, since no disease-specific genetic variations have been reported in human studies on neurodevelopmental disorders, it is necessary to consider epigenetic changes and investigate the potential of biomarkers. Therefore, we hypothesized that there is a correlation between the methylation patterns of the promoter regions of candidate genes, including NR3C1, FOXO3, and ASCL1, which show expression changes in ASD that accompany clinical phenotypes. The primary aim of this preliminary study is to investigate the correlation between the DNA methylation patterns in the promoter regions of candidate genes and the clinical phenotypes of ASD. Additionally, the study seeks to explore the feasibility of utilizing these methylation patterns as epigenetic biomarkers for ASD.

METHODS

Participants

This study utilized genomic DNA samples obtained from a repository, with donors having consented to secondary research purposes. Participants include individuals diagnosed with ASD and their unaffected siblings, recruited from child and adolescent psychiatric clinics at Seoul National University Bundang Hospital. The diagnosis process involved screening based on the criteria outlined in the Diagnostic and Statistical Manual, 5th edition (DSM-5). Standardized diagnostic assessments were conducted using the Korean versions of the Autism Diagnostic Observation Schedule (K-ADOS-2), Autism Diagnostic Interview-Revised (ADI-R), and Korean version of Childhood Autism Rating Scale (K-CARS). To ensure quality and reliability, all administrations of the K-ADOS-2 and ADI-R were videotaped and reviewed by a certified ADOS-2/ADI-R researcher. Intelligence assessments were conducted using the Korean Wechsler preschool and primary scale intelligence, 4th edition (K-WPPSI-IV) and the Korean Wechsler Intelligence Scale for Children, 4th edition (K-WISC-IV). Social was abilities were evaluated using the Korean Vineland Adaptive Behavior Scales, 2nd edition (K-VABS), Social Communi-cation Questionnaire (SCQ), Social Responsiveness Scale, 2nd edition (SRS-2), and Social Maturity Scale (SMS). Comprehensive developmental histories of the participants, including prenatal history, pregnancy complications (e.g., vaginal bleeding, diabetes mellitus, hypothyroidism, and hyperthyroidism), postpartum issues, and medical history, were assessed. Exclusion criteria included neurological diseases, significant medical conditions, chromosomal anomalies, or non-Korean ethni-city.

Ethics Approval and Consent to Participate

The study procedure including informed consent, recruitment, and participation procedures was approved by the Institutional Review Board (IRB) of Seoul National University Bundang Hospital (IRB No. B-2210-785-302).

Methylation-sensitive HpaII/MspI Real-time Quantitative Polymerase Chain Reaction (qPCR)

Genomic DNA was isolated from the blood samples of ASD patients and their unaffected siblings, using the QIAGEN protocol. Each sample contributed 1,500 ng of genomic DNA, which was divided into three aliquots. Two aliquots underwent digestion: one with 10 units of HpaII (methylation-sensitive, New England Biolabs), and the other with MspI (methylation-insensitive, New England Biolabs), both in Cutsmart buffer. The third aliquot remained undigested (UD), serving as a background control. The digestion mixtures were incubated at 37°C for 1 hour, then stored at −20°C. Real-time qPCR, utilizing SYBR Green Super Mix (Bio-Rad Laboratories, Inc.), was conducted on a CFX96TM Real-Time System (Bio-Rad Laboratories, Inc.) to assess the DNA methylation status of the promoters of candidate genes such as NR3C1, ASCL1, and FOXO3. Primer sequences for the target gene regions are listed in Supplementary Table 1 (available online). ΔΔCT for each sample was calculated using the formula (ΔCT [MspI−UD] − ΔCT [HpaII−UD]) [21].

Clinical Assessment

Autism Diagnostic Observation Schedule-2 (ADOS-2)

The ADOS-2 is a standardized semi-structured instrument designed to diagnose ASD by observing communication and social behaviors through play and/or interviews applicable across all age groups [22]. The ADOS-2 consists of five modules, including toddler module and modules 1 to 4, arranged in developmental sequence. Each module features activities specifically tailored to the individual’s developmental age and expressive language skills. Each module is divided into two domains: social affect and restricted, repetitive behaviors (RRBs). ASD classification in ADOS-2 is based on scores that meet or exceed the predetermined algorithm threshold within these two domains. Notably, the ADOS-2 employs specific social activities, referred to as “presses,” designed to create stimulating and standardized contexts to elicit social communication behaviors and interactions [22].

Autism Diagnostic Interview-Revised (ADI-R)

The ADI-R is a semi-structured interview conducted with parents or primary caregivers to diagnose for evaluating the core symptoms of ASD [23]. It consists of 93 items, each rated on a scale from 0, indicating a socially appropriate level, to 3, representing very severe impairment. The ADI-R has four diagnostic domains: social interaction, communication, RRBs, and abnormalities of development evident at or before 36 months, with each domain having its own diagnostic criteria. The algorithm score primarily relies on parents’ descriptions of their child’s behaviors between the ages of four and five years. However, it also includes questions about behaviors observed at any time throughout the child’s life. In the case of children under four years old, the ADI-R ratings are based on their current behaviors.

Social Communication Questionnaires (SCQ)

The SCQ is a parent-reported scale consisting of 40 items, derived from the ADI-R. It evaluates reciprocal social interaction, language and communication, and repetitive/stereotyped behavior with “yes” or “no” responses [24]. The SCQ is available in two versions: the SCQ Lifetime Form and SCQ Current Form. The SCQ Current Form is used for children under five years, focusing on their behaviors observed in the past three months. In contrast, the SCQ Lifetime Form is appropriate for children aged 5 years and above, focusing on the child’s entire developmental history. A cutoff score of 15 or higher typically indicates the risk of ASD. In this study adjusted cutoff scores were used: a score of 10 for children below 47 months and 12 for children aged 48 months and older, based on a standardization study in Korea [25].

Social Responsiveness Scale (SRS)

The SRS is a 65-item questionnaire reported by parents or caregivers. Items are rated on a Likert scale from 1 (“not at all”) to 4 (“almost always”), assessing the severity of ASD symptoms [26]. It encompasses five subscales: social awareness, social cognition, social communication, social motivation, and autistic mannerisms. The SRS-2 is available in four versions: school-age (age 4−18 years), preschool (age 2.5−4.5 years), adult (age 19 years and older), and adult self-report form.

Childhood Autism Rating Scale (CARS) and CARS-2

The CARS consists of 15 items, informed by direct observation, parental reports, and chart reviews [27]. The scale utilizes a 7-point rating system, including a midpoint, ranging from 1 point (normal behavior) to 4 points (severely abnormal behavior). A total score of 30 or higher indicates a risk of ASD; scores between 30 and 36.5 suggest moderate risk, while scores of 37 or higher indicate a high risk. The CARS-2 [28] comprises the original CARS, now named the CARS Standard Version, and the CARS-2 High Functioning (CARS-2-HF), which is designed for children aged 6 years and older with an IQ of 80 or above. In the CARS-2-HF, a cutoff score of 28 is recommended for assessing the risk of ASD.

Korean version of the Vineland Adaptive Behavior Scale, 2nd edition (K-VABS)

The Korean version of the VABS parent/caregiver rating form was utilized to measure adaptive functioning in individuals from birth to 90 years of age [29,30]. Scored on a 3-point scale, it includes five domains: communication, daily living skills, socialization, motor skills, and maladaptive behavior. A higher score denotes more frequent utilization of skills. The five domains are combined to calculate the total adaptive behavior composite score, which is normatively set at an average of 100 with a standard deviation of 15 points.

Korean Vineland Social Maturity Scale (K-SMS)

The K-SMS is an interview and behavioral observation scale that assesses social competence and adaptive functioning [31]. It comprises 89 items categorized into behavioral milestones typically expected at each age. It is organized into eight subdomains: communication, general self-help, locomotion, occupation, self-direction, self-help eating, self-help dressing, and socialization skills to the K-SMS enables the calculation of an individual’s global social age and social quotient (SQ) based on their performance in these areas.

Statistical Analyses

Initially, we performed a set of independent t tests to compare the characteristics between the probands and their unaffected siblings. The mean methylation levels of NR3C1, FOXO3, and ASCL1 in the proband group were compared with those in the sibling group using paired t tests. Linear regression analyses were conducted to investigate the impact of changes in NR3C1, FOXO3, and ASCL1 methylation on the test results, including ADOS, ADI-R, CARS, SMS, full-scale intelligence quotient (FSIQ), VABS, SCQ, and developmental history. Logistic regression analyses were performed on the prenatal history, pregnancy complications, postpartum issues, medical history, and handedness according to the number of candidate genes. Additional analyses controlling for monthly age, FSIQ, and sex of the study population were conducted using the same analytical models. All statistical analyses were performed using SAS 9.4 (SAS Institute Inc.).

RESULTS

A total of 54 families (117 subjects) were included in the analysis. Among them, 9 families had two unaffected siblings, while the remaining 45 families each had one unaffected sibling. In the proband group, 92.59% were male (n = 50), in contrast to the sibling group, which consisted of 55.56% males (n = 35). The mean age of all participants was 86.70 ± 36.98 months, and the proband group having an average age of 84.83 ± 40.62 months. The FSIQ of the proband group was 66.13 ± 22.36, significantly lower than that of the sibling group (96.86 ± 18.88) (p < 0.001). The ADOS, ADI-R, K-CARS, and SCQ results showed statistically significant differences between the two groups, as did the adaptive skills measured by the SMS and VABS (p < 0.001). However, no significant differences were observed between the two groups in terms of pregnancy, postpartum, or neonatal factors (Table 1).

DNA methylation levels in gene promoter regions, assessed using paired t tests between the proband and sibling groups, showed statistically significant differences in NR3C1, FOXO3, and ASCL1 methylation (Table 2). Specifically, NR3C1 methylation was significantly higher in probands compared to their siblings, with a mean difference of 0.806, which was statistically significant (t = 3.200, p = 0.002). Conversely, FOXO3 methylation was significantly lower in probands compared to their siblings, with a mean difference of −0.392, also reflecting statistical significance (t = −2.170, p = 0.034). ASCL1 methylation was also significantly higher in probands than siblings by 0.543, with this outcome being statistically significant (t = 2.020, p = 0.048). In the correlation analysis of methylation levels between proband and siblings, NR3C1 methylation exhibited a strong significant correlation (correlation coefficient = 0.424, p = 0.001), and ASCL1 methylation showed a significant correlation, with a correlation coefficient of 0.283 (p = 0.026). However, the level of FOXO3 methylation did not demonstrate a significant correlation (Supplementary Fig. 1; available online).

Table 3 shows the associations between the methylation levels of candidate genes and clinical phenotypes. While no statistically significant correlations were observed between NR3C1 methylation and any clinical phenotype, FOXO3 methylation was associated with the SQ in the SMS (p = 0.037). ASCL1 methylation was correlated with the total scores of the ADOS (p = 0.029), nonverbal communication as measured by the ADI-R (p = 0.049), and total score and daily living skills as measured by the VABS (p = 0.033 and p = 0.014, respectively). Additionally, ASCL1 methylation was linked to parental age at the time of pregnancy (father, p = 0.024; mother, p = 0.046).

We examined the effects of parental ages at pregnancy on DNA methylation levels, considering the ages of both the proband and sibling group. Our analyses revealed significant correlations. Specifically, the methylation levels of NR3C1 were associated with maternal age at pregnancy (p = 0.049), while ASCL1 methylation levels correlated with paternal age at pregnancy (p = 0.044). After adjusting for FSIQ, these associations remained consistent: NR3C1 methylation with maternal age (p = 0.019) and ASCL1 methylation with paternal age (p = 0.049) at pregnancy. Additionally, FOXO3 methylation also showed a significant association with maternal age at pregnancy (p = 0.032) when controlling for age and FSIQ. The link between ASCL1 methylation and ASD symptoms as measured by the ADI-R after adjusting for age was notable; ASCL1 methylation correlated with social interaction (p = 0.016) and nonverbal communication (p = 0.018). Moreover, daily living skills, as measured by the VABS, were related to ASCL1 methylation levels (p = 0.038). After adjusting for sex, significant associations emerged between ASCL1 methylation and social affect and RRBs on the ADOS (p = 0.045 and p = 0.043, respectively), daily living skills on the VABS (p = 0.027), and both maternal and paternal ages at pregnancy (p = 0.016 and p = 0.030, respectively). However, when clinical phenotypes were adjusted for age, FSIQ, and sex, no significant findings were observed (Supplementary Tables 2-6; available online).

DISCUSSION

Maternal prenatal chemical and psychological stress is associated with various physiological and adverse mental health outcomes in offspring. One notable impact is on the offspring’s hypothalamic-pituitary-adrenal (HPA) axis, which can be affected by maternal stress during pregnancy [32]. Offspring of mothers who experience stress during pregnancy tend to exhibit social impairment in adolescence and an increased propensity for anxiety and depression in adulthood [33,34]. There is a significant evidence indicating that adverse environmental factors during pregnancy, such as smoking, alcohol consumption, exposure to pollutants, and maternal stress, can lead to detrimental effects on offspring. These effects are associated with increased methylation at multiple cytosine-phosphate-guanine (CpG) sites of the NR3C1 gene. These epigenetic alterations have the potential to disrupt the function of the HPA axis and potentially predispose individuals exposed to early stress to a broad spectrum of psychiatric conditions in adulthood, such as major depression and borderline personality disorder [21,35,36]. Efforts to investigate the causes of ASD through DNA methylation are steadily increasing. For example, the SH3 and multiple ankyrin repeat domains protein 3 (SHANK3) gene has been recognized as a substantial contributor to ASD. The methylation status of SHANK3 can regulate the expression of its splice variants, implying that methylation status of candidate genes could be a potential predictor for ASD [37-39]. In this study, primers designed to amplify three specific gene promoter regions were carefully selected to exclude any genetic polymorphisms registered in the Database of Single Nucleotide Polymorphisms (dbSNP). This approach ensures that if genetic polymorphisms are discovered in the regions under investigation, considerations of genetic variability must be incorporated into the analysis of gene expression regulation, in addition to the epigenetic factors such as DNA methylation explored in this research.

Transcription factors such as glucocorticoid receptor, FOXOs and ASCL1 play critical roles in all major stages of the life of a granule neuron in the cerebellar cortex [40]. Consistent with findings from other studies on neurodevelopmental disorders and maternal stress [41,42], the results of this study also showed a significantly increased DNA methylation level of NR3C1 in the ASD group compared to the sibling group. In contrast, the methylation levels of FOXO3 were significantly higher in the unaffected siblings compared to the patients. Additionally, a significant correlation was observed between FOXO3 methylation and the SQ of the SMS. Considering the well-established association between elevated DNA methylation levels and suppression of gene expression, the increased FOXO3 methylation levels may induce changes in FOXO3 gene expression, potentially affecting the ASD phenotype. FOXO3 transcription factors are well-recognized for regulating gene expression patterns that governs stress resistance, cell cycle progression, autophagy, and apoptosis [43]. A previous study suggested a link between FoxO transcription factors, autophagic flux, and maturation of developing neurons in a genetic deficiency cell model [44]. Another study examining prenatal stress and FOXO gene methylation found behavioral changes and alteration in DNA methylation level in a mouse model prenatally exposed to trimethyltin. Altered FoxO signaling and neurodevelopmental pathways have also been identified by methylation binding domain sequencing, as well as increased FOXO3 and decreased ASCL1 expression in hippocampi of male mice subjected to prenatal trimethyltin exposure [45]. Although the role of the FOXO3 gene on neurodevelopment has been acknowledged in a knockout zebrafish model, direct evidence linking FOXO3 to social behavioral changes remains to be established [46].

An intriguing finding in this study was the association between ASCL1 and ASD phenotypes, such as daily living skills measured by the VABS, nonverbal communication measured by the ADI-R, and parental age at pregnancy. ASCL1, also known as MASH1, is a proneuronal factor that plays a crucial role in neurogenesis by competing with FOXO3 for binding to similar regions [18]. It is crucial in neurogenesis within the ventral telencephalon, and mutations in ASCL1 lead to a reduced number of cortical interneurons in mice [47,48]. Consequently, mutations in ASCL1 have the potential to cause an imbalance in the generation of neurons in the dorsal and ventral telencephalon, leading to disruptions in brain connectivity. Additionally, ASCL1 is essential for the specification and, presumably, the lateral migration of cortical interneurons [49]. Furthermore, mutations in ASD risk genes, such as contactin-associated protein-like 2, SHANK3, MET proto-oncogene, receptor tyrosine kinase, and Neuroligin3, result in abnormalities in interneurons in mice, affecting various aspects such as differentiation, migration, synaptogenesis, dendritic branching, and connectivity [38,50-52]. This suggests that malfunction in interneuron circuits is might be a common neurophysiological element in certain forms of ASDs. Notably, these associations persisted after adjusting; however, adjusting for intelligence, the previously observed associations between ASCL1 and the ASD phenotype were attenuated, with the exception of paternal age during pregnancy. Although there is currently no prior research on the relationship between ASCL1 and intelligence, these results suggest an association between ASCL1 and intelligence. Therefore, further large-scale studies are required.

The ages of both mothers and fathers have been associated with higher risk of pregnancy complications and adverse outcomes, such as slowed fetal growth, changes in placental structure, preeclampsia, and premature labor [53]. Offspring of older parents are also associated with an elevated risk of various neurodevelopmental and psychiatric disorders such as obsessive-compulsive disorders, schizophrenia, bipolar disorder, speech disorders, and ASD [54-57]. Some researchers suggest a link between these findings and DNA methylation. Parental age, especially maternal age, correlates with DNA methylation levels in offspring at birth. One study demonstrated a strong correlation between maternal age and DNA methylation levels of 142 genes in 168 newborns [58]. These findings are consistent with our results indicating an association between parental age at the time of pregnancy and NR3C1 and ASCL1 levels after adjusting for age or FSIQ.

Although this study is the first to compare methylation levels of candidate genes including NR3C1, FOXO3, and ASCL1 in Koreans with ASD and their siblings, it has several limitations. Firstly, the proband group had a significantly higher proportion of males (92.59%) than females. Given the higher prevalence of ASD in males, this sex imbalance is an inherent limitation. Secondly, although the strength of this study lies in the analysis of siblings with the same genetic predisposition as the ASD group, the small sample size is a considerable limitation. Thirdly, there may be multiple DNA methylation sites in the gene promoter region, and that other functional DNA methylation sites might exist in different regions of the gene. In this study, measuring methylation levels at a single experimentally accessible site limited the comprehensive representation of the methylation status in the promoter region. Additionally, considering the diversity in DNA methylation levels among different tissues, extrapolating our findings from the peripheral blood to brain tissue may not be fully justified. Therefore, associations between candidate gene DNA methylation levels in the peripheral blood and ASD pathophysiology remain premature. Further research, involving larger populations and more comprehensive analyses of methylation site, is required to validate our study’s finding.

Acknowledgement

We thank Ms. Eun Hye Jang, who kindly assisted with experiments and data management.

Funding
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government MIST (No. 2020R1A2C1009499), the Bio & Medical Technology Development Program of the National Research Foundation (NRF), funded by the Korean government (MSIT) (No. 2021M3E5D9021878) and the Technology Innovation Program (No. 20023378, The human microbiome, which modulates serotonin and intestinal Th17 cells, ameliorates the severity of autism spectrum disorder) funded By the Ministry of Trade, Industry & Energy (MOTIE, Korea).
Conflicts of Interest

HJY receives royalties from Hakjisa for sales of the Korean versions of the Autism Diagnostic Observation Scale-2, Autism Diagnostic Interview-Revised, and Social Communication Questionnaire.

Author Contributions

Conceptualization: Hee Jeong Yoo, Soon Ae Kim. Data acquisition: Miae Oh, Soon Ae Kim. Formal analysis: Miae Oh, Nan-He Yoon, Soon Ae Kim. Investigation: Miae Oh, Nan-He Yoon, Soon Ae Kim. Methodology: Miae Oh, Soon Ae Kim, Hee Jeong Yoo. Project administration: Soon Ae Kim, Hee Jeong Yoo. Funding: Hee Jeong Yoo, Soon Ae Kim, Miae Oh. Supervision: Hee Jeong Yoo, Soon Ae Kim. Writing—original draft: Miae Oh, Soon Ae Kim. Writing—review & editing: Miae Oh, Soon Ae Kim.

Tables

Korean ASD sibling participant characteristics

Proband (n = 54) Sibling (n = 63) χ2 or t p value
Male (%) 50 (92.59) 35 (55.56) 20.074 < 0.001**
Age (mo) 84.83 ± 40.62 88.30 ± 33.80 0.500 0.615
FSIQ 66.13 ± 22.36 96.86 ± 18.88 7.590 < 0.001**
ADOS
Social affect 13.63 ± 3.72 5.16 ± 3.72 2.120 0.030*
RRB 3.15 ± 2.35 0.27 ± 0.63 −1.260 0.209
Total 16.78 ± 5.08 5.43 ± 3.87 −1.990 0.049*
CSS 7.39 ± 1.74 3.11 ± 2.06 2.120 0.030*
ADI-R
Social interaction 20.07 ± 5.15 4.19 ± 4.02 −18.710 < 0.001**
Communication total 13.92 ± 4.03 2.69 ± 3.22 −14.240 < 0.001**
Nonverbal communication 11.41 ± 3.03 1.63 ± 2.50 −8.140 < 0.001**
RRB 5.20 ± 1.93 0.57 ± 0.91 −16.180 < 0.001**
Abnormality before 36 months 4.13 ± 1.10 0.59 ± 1.19 −16.650 < 0.001**
K-CARS 35.15 ± 4.90 19.76 ± 3.55 −18.800 < 0.001**
SCQ
Current 14.44 ± 7.25 2.84 ± 4.65 −10.100 < 0.001**
Lifetime 16.85 ± 6.87 2.76 ± 3.98 −13.260 < 0.001**
SMS
Social quotient 71.57 ± 20.34 113.58 ± 17.21 11.890 < 0.001**
VABS (T score)
Communication 65.71 ± 15.98 100.57 ± 18.19 10.730 < 0.001**
Daily living skills 69.92 ± 17.36 96.89 ± 18.44 7.960 < 0.001**
Socialization 61.31 ± 14.01 93.10 ± 17.36 10.590 < 0.001**
Motor skills 72.55 ± 12.97 94.06 ± 12.75 6.690 < 0.001**
Total 61.35 ± 14.39 95.64 ± 17.79 11.140 < 0.001**
Handedness (%)
Right-handed 41 (75.93) 49 (79.03) 1.551 0.460
Left-handed 5 (9.26) 8 (12.90)
Ambidextrous 8 (14.81) 5 (8.06)
Age at pregnancy (yr)
Father 35.04 ± 5.55 34.05 ± 3.93 −1.090 0.277
Mother 32.13 ± 3.38 31.89 ± 2.93 −0.410 0.680
Pregnancy complications (%)
None 47 (87.04) 52 (83.87) 0.231 0.631
Yes 7 (12.96) 10 (16.13)
Gestational weeks 38.26 ± 4.73 38.83 ± 1.43 0.850 0.399
Neonatal anthropometry
Height 50.66 ± 2.37 50.07 ± 2.41 −1.270 0.207
Weight 3.79 ± 3.67 3.19 ± 0.45 −1.180 0.245
Head circumference 34.43 ± 1.32 33.82 ± 1.31 −2.230 0.029*

Values are presented as mean ± standard deviation or number (%).

Pregnancy complications: diabetes mellitus, vaginal bleeding, hypothyroidism/hyperthyroidism.

ASD, autism spectrum disorder; FSIQ, full-scale intelligence quotient; ADOS, Autism Diagnostic Observation Schedule; RRB, restricted, repetitive behavior; CSS, calibrated severity score; ADI-R, Autism Diagnostic Interview-Revised; K-CARS, Korean Childhood Autism Rating Scale; SCQ, Social Communication Questionnaire; SMS, Social Maturity Scale; VABS, Vineland Adaptive Behavior Scale.

*p < 0.05, **p < 0.01.

The differences of DNA methylation level for candidate gene between proband and sibling group

Gene Proband Sibling Proband - sibling mean difference 95% confidence level t p value
NR3C1 2.16 ± 1.84 1.44 ± 1.99 0.806 0.302 1.309 3.200 0.002**
FOXO3 0.36 ± 0.73 0.74 ± 2.05 −0.392 −0.754 −0.031 −2.170 0.034*
ASCL1 1.14 ± 1.39 0.75 ± 1.17 0.543 0.005 1.081 2.020 0.048*

Values are presented as mean ± standard deviation.

*p < 0.05, **p < 0.01.

Correlation between the NR3C1, FOXO3, and ASCL1 DNA methylation level and clinical phenotypes of ASD sibling pairs

NR3C1 FOXO3 ASCL1



Coefficient SE p value Coefficient SE p value Coefficient SE p value
FSIQ −0.056 1.800 0.975 3.033 4.555 0.509 −3.163 2.382 0.191
ADOS
Social age −0.087 0.280 0.758 −0.191 0.705 0.788 0.686 0.357 0.061
RRB −0.038 0.178 0.833 −0.060 0.448 0.895 0.390 0.229 0.095
Total −0.124 0.379 0.745 −0.251 0.957 0.795 1.076 0.479 0.029*
CSS −0.002 0.133 0.990 0.291 0.332 0.384 0.188 0.173 0.283
ADI-R
Social interaction −0.208 0.389 0.595 0.952 0.975 0.334 0.966 0.498 0.058
Communication total −0.323 0.357 0.372 −0.388 1.005 0.702 0.260 0.551 0.640
Nonverbal communication 0.102 0.372 0.787 0.380 0.938 0.690 0.793 0.376 0.049*
RRB 0.061 0.147 0.679 −0.131 0.371 0.725 0.275 0.191 0.157
Abnormality before 36 mo −0.035 0.084 0.682 0.017 0.211 0.935 0.151 0.109 0.171
K-CARS −0.088 0.381 0.819 −0.667 1.013 0.514 0.905 0.537 0.099
SMS
Social quotient 0.878 1.584 0.582 8.711 4.053 0.037* −4.197 2.223 0.065
VABS
Communication 0.437 1.220 0.722 3.693 3.072 0.235 −2.558 1.602 0.117
Daily living skills 0.168 1.322 0.899 −0.097 3.373 0.977 −4.247 1.671 0.014*
Socialization −0.119 1.061 0.911 1.920 2.692 0.479 −2.664 1.375 0.058
Motor skills 0.746 1.342 0.582 4.148 3.117 0.193 −1.388 1.771 0.440
Total 0.128 1.098 0.908 2.293 2.783 0.414 −3.092 1.409 0.033*
SCQ
Current −0.476 0.546 0.388 −0.937 1.382 0.501 1.132 0.710 0.117
Lifetime −0.345 0.522 0.512 1.066 1.313 0.421 0.975 0.680 0.158
Age at pregnancy
Father −0.16 0.425 0.709 0.223 1.072 0.836 1.248 0.535 0.024*
Mother 0.490 0.247 0.053 0.717 0.638 0.267 0.667 0.326 0.046*
Gestational weeks 0.538 0.361 0.143 0.782 0.917 0.398 0.044 0.496 0.930
Neonatal anthropometry
Height 0.058 0.186 0.754 −0.307 0.464 0.511 0.291 0.240 0.231
Weight −0.408 0.277 0.148 0.225 0.723 0.757 0.431 0.370 0.249
Head circumference −0.187 0.105 0.081 −0.145 0.276 0.602 0.177 0.143 0.222

ASD, autism spectrum disorder; SE, standard error; FSIQ, full-scale intelligence quotient; ADOS, Autism Diagnostic Observation Schedule; RRB, restricted, repetitive behavior; CSS, calibrated severity score; ADI-R, Autism Diagnostic Interview-Revised; K-CARS, Korean Childhood Autism Rating Scale; SMS, Social Maturity Scale; VABS, Vineland Adaptive Behavior Scale; SCQ, Social Communication Questionnaire.

*p < 0.05.

References
  1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-5. American Psychiatric Association;2013.
    CrossRef
  2. Centers for Disease Control and Prevention (CDC). Autism spectrum disorder (ASD) [Internet]. CDC; 2021 [cited at 2023 Oct 30]. https://www.cdc.gov/autism/index.html.
  3. Betancur C. Etiological heterogeneity in autism spectrum disorders: more than 100 genetic and genomic disorders and still counting. Brain Res 2011;1380:42-77.
    Pubmed CrossRef
  4. Christian SL, Brune CW, Sudi J, Kumar RA, Liu S, Karamohamed S, et al. Novel submicroscopic chromosomal abnormalities detected in autism spectrum disorder. Biol Psychiatry 2008;63:1111-1117.
    Pubmed KoreaMed CrossRef
  5. Lyall K, Croen L, Daniels J, Fallin MD, Ladd-Acosta C, Lee BK, et al. The changing epidemiology of autism spectrum dis-orders. Annu Rev Public Health 2017;38:81-102.
    Pubmed KoreaMed CrossRef
  6. Bai D, Yip BHK, Windham GC, Sourander A, Francis R, Yoffe R, et al. Association of genetic and environmental factors with autism in a 5-country cohort. JAMA Psychiatry 2019;76:1035-1043.
    Pubmed KoreaMed CrossRef
  7. Argentieri MA, Nagarajan S, Seddighzadeh B, Baccarelli AA, Shields AE. Epigenetic pathways in human disease: the impact of DNA methylation on stress-related pathogenesis and current challenges in biomarker development. EBioMedicine 2017;18:327-350.
    Pubmed KoreaMed CrossRef
  8. Grafodatskaya D, Chung B, Szatmari P, Weksberg R. Autism spectrum disorders and epigenetics. J Am Acad Child Adolesc Psychiatry 2010;49:794-809.
    Pubmed CrossRef
  9. Cho SH, Chai JH, Chang SY, Kim SA. Acute valproate exposure induces sex-specific changes in steroid hormone metabolism in the cerebral cortex of juvenile mice. Neurochem Res 2020;45:2044-2051.
    Pubmed CrossRef
  10. Lee JH, Cho SH, Jang EH, Kim SA. Sex-specific changes in brain estrogen metabolism induced by acute trimethyltin exposure. In Vivo 2021;35:793-797.
    Pubmed KoreaMed CrossRef
  11. Moisiadis VG, Matthews SG. Glucocorticoids and fetal programming part 1: outcomes. Nat Rev Endocrinol 2014;10:391-402.
    Pubmed CrossRef
  12. Mansell T, Vuillermin P, Ponsonby AL, Collier F, Saffery R, Ryan J. Maternal mental well-being during pregnancy and glucocorticoid receptor gene promoter methylation in the neonate. Dev Psychopathol 2016;28(4pt2):1421-1430.
    Pubmed CrossRef
  13. Parade SH, Ridout KK, Seifer R, Armstrong DA, Marsit CJ, McWilliams MA, et al. Methylation of the glucocorticoid receptor gene promoter in preschoolers: links with internalizing behavior problems. Child Dev 2016;87:86-97.
    Pubmed KoreaMed CrossRef
  14. Raffetti E, Melas PA, Landgren AJ, Andersson F, Forsell Y, Lavebratt C, et al. DNA methylation of the glucocorticoid receptor gene predicts substance use in adolescence: longitudinal data from over 1000 young individuals. Transl Psychiatry 2021;11:477.
    Pubmed KoreaMed CrossRef
  15. Welberg LA, Seckl JR, Holmes MC. Prenatal glucocorticoid programming of brain corticosteroid receptors and corticotrophin-releasing hormone: possible implications for behaviour. Neuroscience 2001;104:71-79.
    Pubmed CrossRef
  16. Kuo T, Liu PH, Chen TC, Lee RA, New J, Zhang D, et al. Transcriptional regulation of FoxO3 gene by glucocorticoids in murine myotubes. Am J Physiol Endocrinol Metab 2016;310:E572-E585.
    Pubmed KoreaMed CrossRef
  17. Schmidt-Strassburger U, Schips TG, Maier HJ, Kloiber K, Mannella F, Braunstein KE, et al. Expression of constitutively active FoxO3 in murine forebrain leads to a loss of neural progenitors. FASEB J 2012;26:4990-5001.
    Pubmed CrossRef
  18. Webb AE, Pollina EA, Vierbuchen T, Urbán N, Ucar D, Leeman DS, et al. FoxO3 shares common targets with ASCL1 genome-wide and inhibits ASCL1-dependent neurogenesis. Cell Rep 2013;4:477-491.
    Pubmed KoreaMed CrossRef
  19. Vierbuchen T, Ostermeier A, Pang ZP, Kokubu Y, Südhof TC, Wernig M. Direct conversion of fibroblasts to functional neurons by defined factors. Nature 2010;463:1035-1041.
    Pubmed KoreaMed CrossRef
  20. Wilkinson G, Dennis D, Schuurmans C. Proneural genes in neocortical development. Neuroscience 2013;253:256-273.
    Pubmed CrossRef
  21. Niknazar S, Nahavandi A, Peyvandi AA, Peyvandi H, Roozbahany NA, Abbaszadeh HA. Hippocampal NR3C1 DNA methylation can mediate part of preconception paternal stress effects in rat offspring. Behav Brain Res 2017;324:71-76.
    Pubmed CrossRef
  22. Lord C, Rutter M, DiLavore PC, Risi S, Gotham K, Bishop SL, et al. Autism diagnostic observation schedule, second edition (ADOS®-2). Western Psychological Services;2012..
  23. Rutter M, LeCouteur A, Lord C. Autism diagnostic interview–revised (ADI®-R). Western Psychological Services;2003.
    CrossRef
  24. Rutter M, Bailey A, Lord C. Social communication questionnaire (SCQ). Western Psychological Services;2003.
  25. Kim JH, Sunwoo HJ, Park SB, Noh DH, Jung YK, Cho IH, et al. A validation study of the Korean version of Social Communi-cation Questionnaire. J Korean Acad Child Adolesc Psychiatry 2015;26:197-208.
    CrossRef
  26. Constantino JN, Gruber CP. Social responsiveness scale, second edition (SRS™-2). Western Psychological Services;2012.
  27. Schopler E, Reichler RJ, DeVellis RF, Daly K. Toward objective classification of childhood autism: childhood autism rating scale (CARS). J Autism Dev Disord 1980;10:91-103.
    Pubmed CrossRef
  28. Schopler E, Van Bourgondien ME, Wellman GJ, Love SR. Childhood autism rating scale, second edition (CARS™2). Western Psychological Services;2010.
  29. Cicchetti DV, Carter AS, Gray SAO. Vineland adaptive behavior scales. In: Volkmar FR, editor. Encyclopedia of autism spectrum disorders. 2nd ed. Springer;2013.
  30. Hwang S, Kim J, Hong S. Korean vineland adaptive behavior scales-II. Korea Psychology Co.;2014.
  31. Kim SK, Kim OK. Social maturity scale. Joong Ang Juck Sung;1985.
  32. Brunton PJ, Russell JA. Prenatal social stress in the rat programmes neuroendocrine and behavioural responses to stress in the adult offspring: sex-specific effects. J Neuroendocrinol 2010;22:258-271.
    Pubmed CrossRef
  33. Morley-Fletcher S, Rea M, Maccari S, Laviola G. Environmen-tal enrichment during adolescence reverses the effects of prenatal stress on play behaviour and HPA axis reactivity in rats. Eur J Neurosci 2003;18:3367-3374.
    Pubmed CrossRef
  34. Talge NM, Neal C; Glover V; Early Stress, Translational Re-search and Prevention Science Network: Fetal and Neonatal Experience on Child and Adolescent Mental Health. Anten-atal maternal stress and long-term effects on child neurodevelopment: how and why? J Child Psychol Psychiatry 2007;48:245-261.
    Pubmed KoreaMed CrossRef
  35. Chiu YH, Coull BA, Cohen S, Wooley A, Wright RJ. Prenatal and postnatal maternal stress and wheeze in urban children: effect of maternal sensitization. Am J Respir Crit Care Med 2012;186:147-154.
    Pubmed KoreaMed CrossRef
  36. Niknazar S, Nahavandi A, Peyvandi AA, Peyvandi H, Zare Mehrjerdi F, Karimi M. Effect of maternal stress prior to conception on hippocampal BDNF signaling in rat offspring. Mol Neurobiol 2017;54:6436-6445.
    Pubmed CrossRef
  37. Durand CM, Betancur C, Boeckers TM, Bockmann J, Chaste P, Fauchereau F, et al. Mutations in the gene encoding the synaptic scaffolding protein SHANK3 are associated with autism spectrum disorders. Nat Genet 2007;39:25-27.
    Pubmed KoreaMed CrossRef
  38. Moessner R, Marshall CR, Sutcliffe JS, Skaug J, Pinto D, Vincent J, et al. Contribution of SHANK3 mutations to autism spectrum disorder. Am J Hum Genet 2007;81:1289-1297.
    Pubmed KoreaMed CrossRef
  39. Gauthier J, Spiegelman D, Piton A, Lafrenière RG, Laurent S, St-Onge J, et al. Novel de novo SHANK3 mutation in autistic patients. Am J Med Genet B Neuropsychiatr Genet 2009;150B:421-424.
    Pubmed CrossRef
  40. Nakatani H, Martin E, Hassani H, Clavairoly A, Maire CL, Viadieu A, et al. Ascl1/Mash1 promotes brain oligodendrogenesis during myelination and remyelination. J Neurosci 2013;33:9752-9768.
    Pubmed KoreaMed CrossRef
  41. Bromer C, Marsit CJ, Armstrong DA, Padbury JF, Lester B. Genetic and epigenetic variation of the glucocorticoid receptor (NR3C1) in placenta and infant neurobehavior. Dev Psychobiol 2013;55:673-683.
    Pubmed KoreaMed CrossRef
  42. Berretta E, Guida E, Forni D, Provenzi L. Glucocorticoid receptor gene (NR3C1) methylation during the first thousand days: environmental exposures and developmental outcomes. Neurosci Biobehav Rev 2021;125:493-502.
    Pubmed CrossRef
  43. Salih DA, Brunet A. FoxO transcription factors in the maintenance of cellular homeostasis during aging. Curr Opin Cell Biol 2008;20:126-136.
    Pubmed KoreaMed CrossRef
  44. Schäffner I, Minakaki G, Khan MA, Balta EA, Schlötzer-Schrehardt U, Schwarz TJ, et al. FoxO function is essential for maintenance of autophagic flux and neuronal morphogenesis in adult neurogenesis. Neuron 2018;99:1188-1203.e6.
    Pubmed KoreaMed CrossRef
  45. Kim SA, Chai JH, Jang EH. Prenatal trimethyltin exposure induces long-term DNA methylation changes in the male mouse hippocampus. Int J Mol Sci 2021;22:8009.
    Pubmed KoreaMed CrossRef
  46. Peng K, Li Y, Long L, Li D, Jia Q, Wang Y, et al. Knockdown of FoxO3a induces increased neuronal apoptosis during embryonic development in zebrafish. Neurosci Lett 2010;484:98-103.
    Pubmed CrossRef
  47. Casarosa S, Fode C, Guillemot F. Mash1 regulates neurogenesis in the ventral telencephalon. Development 1999;126:525-534.
    Pubmed CrossRef
  48. Abashkin DA, Karpov DS, Kurishev AO, Marilovtseva EV, Golimbet VE. ASCL1 is involved in the pathogenesis of schizophrenia by regulation of genes related to cell proliferation, neuronal signature formation, and neuroplasticity. Int J Mol Sci 2023;24:15746.
    Pubmed KoreaMed CrossRef
  49. Liu YH, Tsai JW, Chen JL, Yang WS, Chang PC, Cheng PL, et al. Ascl1 promotes tangential migration and confines migratory routes by induction of Ephb2 in the telencephalon. Sci Rep 2017;7:42895.
    Pubmed KoreaMed CrossRef
  50. Alarcón M, Abrahams BS, Stone JL, Duvall JA, Perederiy JV, Bomar JM, et al. Linkage, association, and gene-expression analyses identify CNTNAP2 as an autism-susceptibility gene. Am J Hum Genet 2008;82:150-159.
    Pubmed KoreaMed CrossRef
  51. Campbell DB, Sutcliffe JS, Ebert PJ, Militerni R, Bravaccio C, Trillo S, et al. A genetic variant that disrupts MET transcription is associated with autism. Proc Natl Acad Sci U S A 2006;103:16834-16839.
    Pubmed KoreaMed CrossRef
  52. Jamain S, Quach H, Betancur C, Råstam M, Colineaux C, Gillberg IC, et al. Mutations of the X-linked genes encoding neuroligins NLGN3 and NLGN4 are associated with autism. Nat Genet 2003;34:27-29.
    Pubmed KoreaMed CrossRef
  53. Sartorius GA, Nieschlag E. Paternal age and reproduction. Hum Reprod Update 2010;16:65-79.
    Pubmed CrossRef
  54. King MD, Fountain C, Dakhlallah D, Bearman PS. Estimated autism risk and older reproductive age. Am J Public Health 2009;99:1673-1679.
    Pubmed KoreaMed CrossRef
  55. Sipos A, Rasmussen F, Harrison G, Tynelius P, Lewis G, Leon DA, et al. Paternal age and schizophrenia: a population based cohort study. BMJ 2004;329:1070.
    Pubmed KoreaMed CrossRef
  56. Menezes PR, Lewis G, Rasmussen F, Zammit S, Sipos A, Harrison GL, et al. Paternal and maternal ages at conception and risk of bipolar affective disorder in their offspring. Psychol Med 2010;40:477-485.
    Pubmed CrossRef
  57. Comings DE, MacMurray JP. Maternal age at the birth of the first child as an epistatic factor in polygenic disorders. Am J Med Genet B Neuropsychiatr Genet 2006;141B:1-6.
    Pubmed CrossRef
  58. Adkins RM, Thomas F, Tylavsky FA, Krushkal J. Parental ages and levels of DNA methylation in the newborn are correlated. BMC Med Genet 2011;12:47.
    Pubmed KoreaMed CrossRef


This Article

Close ✕


Cited By Articles
  • CrossRef (0)
  • Scopus (0)
  • Download (214)

Author ORCID Information

Funding Information

Services
Social Network Service

e-submission

Archives