https://orcid.org/0000-0001-9659-2503
Figures
Abstract
Autism spectrum disorder (ASD) is a set of genetically heterogenous neurodevelopmental disorders characterized by core symptoms including impaired social interaction, communication deficits, and restricted or stereotyped behaviors. While a significant number of cases are not explained by Mendelian inheritance, there is growing evidence for implication of non-coding RNAs (ncRNAs) in the development and inheritance of ASD. Transcriptional studies often face challenges due to patient-specific variations in gene expression and technical differences in preserving RNA integrity. We propose that isolating RNA from DNA/RNA hybrids provides a robust method to reliably capture transcriptional information. We performed a whole transcriptome analysis on blood samples from ASD patients and healthy controls to investigate transcripts associated with DNA/RNA hybrids. We identified 278,300 novel transcripts across 68,487 DNA/RNA hybrid loci, with significant enrichment in exonic and intronic regions. The novel long non-coding RNAs (lncRNAs) we found showed higher expression levels compared to known transcripts. Differential expression analysis revealed 301 significantly upregulated and 401 downregulated known transcripts in ASD samples compared to controls (|log2-fold change| > 1 and adjusted p-value < 0.05). Through qRT-PCR validation, we confirmed the significant upregulation of RN7SK and SMARCC2 associated with DNA/RNA hybrids in ASD patients. Pathway and enrichment analyses highlighted mitochondrial dysfunction and energy metabolism. Our results suggest that ncRNAs can form DNA/RNA hybrids that influence gene expression, providing preliminary insights into the mechanisms of transcriptional dysregulation in ASD.
Citation: Kianmehr L, MokhtarZadeh K, Yilmaz Z, Darzi Ramandi H, Mehmetbeyoglu Duman E, Funda Sener E, et al. (2025) DNA/RNA hybrid profiling in autistic patients: A focus on mRNA and non-coding RNA variations. PLoS One 20(11): e0326901. https://doi.org/10.1371/journal.pone.0326901
Editor: Asem Surindro Singh, University of Cincinnati College of Medicine, UNITED STATES OF AMERICA
Received: June 11, 2025; Accepted: October 7, 2025; Published: November 3, 2025
Copyright: © 2025 Kianmehr et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The datasets generated during the current study are available in the GEO repository, [https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE282063].
Funding: This study was supported by the TÜBİTAK 1002 project grant (project number 122S955), with S. Taheri as the principal investigator. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: NO authors have competing interests.
Introduction
Autism spectrum disorder (ASD) is a complex group of neurodevelopmental disorders characterized by impaired social interaction and communication, as well as the presence of restricted and repetitive behaviors [1]. The prevalence of ASD is increasing globally, yet its underlying mechanisms remain poorly understood. Both genetic and non-genetic factors as well as their interactions are implicated in the etiology of ASD, leading to altered brain development and neural activity [2,3].
While large-scale exome sequencing studies have identified hundreds of genes associated with ASD [4], no consistent, uniform genetic signature has emerged. Only a small fraction of cases is linked to monogenetic disorders such as Fragile-X syndrome and Rett syndrome [5]. For the majority, the genetic cause remains often uncertain because of complex genetic variations in human. Previous research has consistently highlighted aberrant gene expression patterns in various tissues of autistic individuals, including cerebral cortex and peripheral blood lymphocytes [6–8]. These studies reveal that transcriptional dysregulation in ASD often exhibits significant tissue-specific differences [6–8], such as directional changes in inflammatory genes observed between the brain and blood [6].
While transcriptional variation is a key feature in ASD, a major challenge is the variability observed among the patients. This inconsistency can partly stem from technical differences in how studies are performed. Since RNA can degrade or fragmented during standard sample preparation, a robust methodology is required to preserve transcriptional integrity. Isolating RNA specifically from DNA/RNA hybrids may offer a technique that better preserves this transcriptional information. DNA/RNA hybrids are present in normal cellular physiology, formed during transcription when the nascent RNA remains temporarily bound to its DNA template [9]. Dysregulation in their abundance and localization is associated with various human pathologies [10–13], including neurodevelopmental and neurodegenerative disorders. Specifically, the aberrant formation of these hybrids, particularly within repetitive genomic regions like telomeres and centromeres, has been linked to genomic instability [14,15]. However, their precise role in ASD remains largely unexplored.
Excessive DNA/RNA hybrid formation, particularly within repeated motifs in post-mitotic neurons, can potentially enhance transcription activity, leading to DNA damage and genomic instability. This can pose a physical block to transcription, leading to gene silencing, as seen in disorders associated with trinucleotide repeat expansions like Fragile-X syndrome [16–18]. Moreover, mutations in genes encoding proteins essential for resolving these structures, such as SETX (which encodes Senataxin) and RNaseH2, are linked to a global increase in hybrids and are implicated in neurodegenerative conditions [19–22]. We have also previously shown a heritable downregulation of miRNAs in autistic patients and their families, suggesting the involvement of non-Mendelian inheritance mechanisms in ASD [23].
Given the challenge of inconsistent results from traditional studies, we hypothesized that variations in DNA/RNA hybrid profiles could provide a better view of the transcriptional dysregulation observed in this disorder. To test this, we utilized a novel, antibody-independent method [24–26] to isolate and profile genomic regions associated with these hybrid structures. In this study, we applied this approach to blood samples from individuals with ASD and healthy controls. Using an integrated bioinformatics and molecular biology approach, we performed whole transcriptome analysis to identify differences in both known and novel transcripts associated with DNA/RNA hybrids. The results reveal altered DNA/RNA hybrid profiles in both messenger RNA (mRNA) and long non-coding RNA (lncRNA) regions in autistic patients, offering foundational perspectives on the potential molecular mechanisms underlying ASD.
Materials and methods
Study design
Blood samples were collected from both autistic patients and healthy controls following the acquisition of parental consent. Written informed consent was obtained from all parents before their children participated in the study. The cohort consisted of six patients diagnosed with ASD who presented to the School of Medicine Hospital of Erciyes University, along with six healthy individuals without known medical conditions, all of whom provided informed consent. The diagnosis of autism was confirmed by a multidisciplinary team, including an experienced child psychiatrist, a paediatric neurologist, and a genetic specialist. Diagnostic criteria were based on the Fourth and Fifth Editions of the Diagnostic and Statistical Manual of Mental Disorders according to DSM-IV-TR, and DSM-V criteria [27,28], utilizing the Childhood Autism Rating Scale (CARS) [29]. All participants underwent comprehensive screening for signs of infection, and individuals with acute illnesses were excluded to ensure the integrity of the blood samples and the validity of the research findings. The patient cohort included participants with a range of symptoms and family histories:
- Patient 1: A family history of schizophrenia in the paternal grandmother.
- Patient 2: A family history of autism and a male sibling with the disorder.
- Patient 3: A family history of Asperger syndrome, an uncle with schizophrenia, and a family history of speech delay.
- Patient 4: A history of intellectual disability, a sibling with autism, and epilepsy/seizures, with consanguinity present in the family.
- Patient 5: A history of intellectual disability and a sister with similar neurodevelopmental features. The patient’s development halted after a febrile seizure at 8 months, and there is third-degree consanguinity in the family.
- Patient 6: A small cerebellum, mild intellectual disability, and a paternal cousin with a history of seizures and autism. Consanguinity is also present in the family.
Blood samples were obtained from six patients and six healthy individuals between January 2009 and December 2010. All participants were from the same ethnic background, which was explicitly stated as Turkish families from the Anatolian region. Clinical characteristics of the samples are summarized in S1 Table. This study was approved by Ethics Committee of the Erciyes University School of Medicine (Committee No: 2011/10, approval date: 09-20-2011).
DNA/RNA hybrid preparation
We used a previously developed method to isolate DNA/RNA hybrids from total nucleic acids without relying on antibodies, which minimizes potential artifacts and enhances reproducibility [29–31]. This approach modifies the classical TRIzol-chloroform extraction protocol to specifically fractionate and purify DNA/RNA hybrids without the need for antigenic recognition or RNaseH treatment.
Following the enzymatic removal of proteins, total nucleic acids were isolated and then fractionated using the standard TRIzolTM protocol [30]. The two primary fractions, the aqueous phase (containing free RNA) and the chloroform-water interphase (containing genomic DNA and DNA/RNA hybrids), were separately precipitated with ethanol. The interphase material, containing the target hybrid, was subsequently purified using Zymo-SpinTM columns (Zymo Research Corp, Irvine, CA, USA) or chloroform extraction. To resolve the hybrid structures from residual proteins, the ethanol-precipitated interface material was treated overnight at 56°C with a buffer (20 mM Tris (pH 8), 50 mM EDTA, 0.5% SDS, 20 µM dithiothreitol, and 400 µg/mL Proteinase K). The resulting DNA/RNA hybrid-containing fraction was then subjected to DNase digestion to remove unbound genomic DNA, followed by additional column purification. This multi-step process isolated the RNA component specifically bound within the hybrid structures, separating it from the free RNA fraction. Finally, the presence of isolated DNA/RNA hybrid fraction was confirmed by testing the sensitivity of extracts to RNaseH (an enzyme specific for degrading the RNA strands of a DNA/RNA hybrid) and DNase (to complete removal of remaining genomic DNA fragments).
RNA sequencing
A quantity of 10–100 ng of RNA was recovered from blood samples after DNase digestion. RNA libraries were prepared from biological replicates of the DNA/RNA hybrid-bound RNA from the autistic patients and healthy controls. High-throughput sequencing was performed using the Illumina HiSeq 2500 or MiSeq platform at Eurofins Medigenomix GmbH (Ebersberg, Germany).
Bioinformatics analysis
Quality control of the raw sequencing data was performed using FastQC (version 0.11.9) and Trimmomatic (version 0.39) to remove low-quality reads and adapter sequences [31]. This process involved removing 12 nucleotides from the 5′ end of each read and trimming bases with a Phred score below 20 from both 5′ and 3′ ends. Reads shorter than 35 base pairs after trimming were discarded. Trimmed reads were aligned to the human reference genome (GRCh38) using HISAT2 [32] with the command: ‘hisat2 --dta --rna-strandness’. Samtools was used for sorting and indexing alignment files.
Transcriptome assembly
Potential ribosomal RNAs and reads not aligned to the reference genome were filtered out using RseQC tool (version 4.0.0) [33]. Genome-guided transcriptome assembly was performed using StringTie (version 2.1.1) [34], using the command `stringtie –rf` with the annotation file in GFF3 format (gencode.v39.chr_patch_hapl_scaff.annotation.gff3) obtained from the GENCODE database (https://www.gencodegenes.org/human). For each sample, an Individual transcriptome was generated and merged using the StringTie merge command to create a reference transcriptome assembly for quantification. Strand-specific settings were applied throughout both the alignment and assembly processes. Gene expression quantification was conducted using HTSeq-count (version 0.11.2) [35]. For visualization, the Integrative Genome Viewer (IGV, version 2.12.0) was used [36]. To facilitate comparative visualization, reads were normalized across samples to 107 reads per sample.
Differential gene expression (DGE) analysis
DGE analysis was performed using DESeq2 (version 1.36.0) [37]. To account for multiple testing, p-values were adjusted for multiple comparisons using the Benjamini-Hochberg method. Transcripts with an absolute log2-fold change greater than 1 and an adjusted p-value of less than 0.05 were considered significantly differentially expressed. Gene ontology (GO) and KEGG pathway over-representation analyses were performed using the Cluster Profiler package (version 4.4.4) [38] and the org.Hs.e.g.,db annotation package (version 3.15.0).
Detection of novel lncRNAs
Novel transcripts were identified using GffCompare (version 0.12.6) [39], focusing on unannotated transcripts originating from intergenic and antisense regions with a minimum length of 200 nucleotides. These transcripts were subsequently screened for coding potential. Nucleotide sequences were extracted using SeqKit [40] and GffRead. The coding potential of each transcript was evaluated using the Coding Potential Assessment Tool (CPAT), which employs a logistic regression model to predict coding potential from nucleotide sequences [41]. The highest coding potential was determined across all identified open reading frames (ORFs) for each transcript. Transcripts with a coding probability score below a threshold of 0.364 were classified as novel lncRNAs.
Peak finding
Peak identification was performed using HOMER (version 4.11) [42] to detect significantly enriched peaks, which represents genomic region-specific associations of DNA/RNA hybrids. The analysis was conducted using the commands: ‘findpeaks –o auto’ and ‘annotatePeaks.pl’. A False Discovery Rate (FDR) threshold of 0.001 was applied. Peaks were then annotated to genomic features, including promoter-TSS (Transcription Start Site), 5’ UTR (Untranslated region), exon, intron, 3’ UTR, TTS (Transcription termination site), intergenic, and non-coding regions, based on the GRCh38 reference genome.
cDNA synthesis and qRT-PCR
Total RNA samples were reverse transcribed into complementary DNA (cDNA) using the Evoscript Universal cDNA Master Kit (Roche, Mannheim, Germany, Cat No: 07912439001), in a final reaction volumes of 20 µL following the manufacturer’s protocol. cDNA samples were stored at −80 °C until further analysis. qRT-PCR validation was performed on six significantly differentially expressed transcripts using the LightCycler 480 II high-throughput Real-Time PCR system (Roche, Mannheim, Germany). Prior to amplification, cDNA samples were diluted in 1:5 in nuclease-free water. SYBR Green Master Mix (Roche, Mannheim, Germany, Cat No: 04707516001) was used to quantify transcript levels of lncRNAs (SLC12A5-AS1, RN7SK), and protein-coding genes (SLC16A3, NLGN3, SMARCC2 and ADAMTSL4), using GAPDH as the reference gene. The reaction mix was prepared according to the manufacturer’s instructions. Relative gene expression changes were calculated using the 2−ΔΔCt across all experimental groups [43].
Statistical analysis and visualization
Data visualization and statistical analyses were performed using R software (version 4.3.3). Heatmaps were generated with the pheatmap package (https://cran.r-project.org/package=pheatmap) [44], while additional visualizations were created using ggplot2 [45] and ggrepel packages. Data were normalized to a standardized scale ranging from 0 to 1 before generating heatmaps. Various statistical tests─including the t-test, chi-square test, and Fisher’s Exact test─were utilized for variance analysis, with the specific test selected based on the characteristics of the data and the research objectives. All statistical analyses were executed within the RStudio environment, with statistical significance established at an adjusted p-value < 0.05.
Results
We investigated the transcriptome profiles of DNA/RNA hybrids in blood samples from individuals with ASD and healthy controls to reveal the potential transcriptional alterations in the disorder. Applying our previously developed methodology for analyzing DNA/RNA hybrid fractions, we established a robust framework for transcriptome analysis. Specifically, we sought to identify differences in hybrid formation that may contribute to ASD pathology.
A total of 3.5 × 10⁸ total paired-end reads were generated. Due to quality control and read coverage requirements, the final analysis was conducted using five samples (three ASD patients and two healthy controls). Genomic mapping of RNA-seq reads was performed using the human GRCh38 genome assembly, as summarized in the schematic pipeline (Fig 1). Following quality control and mapping, uniquely mapped paired-ends reads were used for genome-guided transcriptomes assembly.
The pipeline illustrates the bioinformatic analysis steps: 1. Quality control and alignment: raw FASTQ files undergo quality control (FastQC) and trimming (Trimmomatic). Processed reads are aligned to the human reference genome (GRCh38) using HISAT2 to generate BAM files. 2. Transcriptome assembly and quantification: Ribosomal RNA contamination is filtered (RSeQC). Genome-guided transcriptome assembly is performed (StringTie), and gene expression levels are quantified (HTSeq-count). 3. DGE analysis is conducted (DESeq2). Functional enrichment analysis, including Gene Ontology and KEGG, is performed (ClusterProfiler). 4. Novel transcript identification: Assembled transcripts are compared to reference annotations (GFFCompare) to identify novel transcripts, which are then screened for coding potential (CPAT) to identify novel lncRNAs. 5. Peak finding and visualization: Genomic region-specific associations of DNA/RNA hybrids are identified (peak finding) using HOMER. RNA signals are visualized using IGV.
Identification of novel transcripts associated with DNA/RNA hybrids
Genome-guided transcriptome assembly identified 278,300 transcripts distributed across 68,487 DNA/RNA hybrid loci. The high support scores for the generated transcripts reflect the high sensitivity and accuracy of the assembly process (S2 Table). This assembly revealed 9,501 novel exons and 2,930 novel introns spanning 5,196 genomic loci. The biotype analysis of known transcripts showed 49% were protein-coding, 36% were lncRNAs, and 15% were pseudogenes-derived transcripts (S1 Fig).
Enrichment of DNA/RNA hybrids in genomic regions of ASD blood cells
Genome-wide distribution analysis using HOMER evaluated the distribution of DNA/RNA hybrids across genomic features, including the promoter-TSS, 5’ UTR, exons, introns, 3’ UTR, TTS, intergenic, and non-coding regions. DNA/RNA hybrids were enriched across all genomic regions; however, with the most significant enrichment observed in exonic and intronic regions (Fig 2). When comparing patient groups, the ASD blood cells exhibited a pronounced and significant accumulation of DNA/RNA hybrids within the exonic and intronic regions compared to control blood cells.
Genomic regions annotated include the promoter-TSS (Transcription Start Site), 5’ UTR (Untranslated Region), exon, intron, 3’ UTR (Untranslated Region), TTS (Transcription Termination Site), intergenic, and non-coding regions. Peaks are notably enriched in the exonic and intronic regions of ASD patients compared to controls. Error bars represent the mean ± Standard Deviation (SD). Statistical comparisons were performed for each genomic region using One-way ANOVA followed by Duncan’s multiple comparison test (p-value < 0.05).
Differential expression and identification of novel lncRNAs associated with DNA/RNA hybrids
Transcriptome profiling of DNA/RNA hybrids-derived transcripts identified 13,305 known transcripts across all samples after filtering out lowly expressed transcripts. Novel lncRNA transcripts exhibited higher expression level compared to both annotated protein-coding genes and known lncRNAs (Fig 3a).
(A) Distribution of expression levels of lncRNAs, novel lncRNAs and protein-coding transcripts across all samples. (B) Heatmap of the most significantly differentially expressed transcripts comparing ASD patients to healthy controls. Color intensity represents expression level (Z-score of log-normalized counts), with yellow/orange indicating high expression (log2-fold change >1, adjusted p-value <0.05) and blue indicating low expression (log2-fold change <−1, adjusted p-value <0.05). (C) Expression levels of selected candidate transcripts. Jitter plots represent the variance stabilized transformed (VST) normalized counts for key selected DE transcripts: protein-coding genes (ADAMTSL4, NLGN3, SLC16A3, and SMARCC2) and lncRNAs (RN7SK, SLC12A5-AS1). Each dot represents an individual sample and mean ± standard error (SE) is shown. Statistical significance is indicated as follows: p < 0.001(***), p < 0.01(**), p < 0.05(*). All selected transcripts exhibited significantly higher expression in ASD patients than in controls.
Differential expression analysis (using |log2-fold change| > 1 and adjusted p-value < 0.05) identified 702 differentially expressed transcripts: 301 transcripts were significantly upregulated and 401 transcripts were significantly downregulated in ASD patients compared to healthy controls (S3 Table). The most significantly upregulated transcripts (based on log2-fold change and baseline expression) are illustrated (Fig 3b, S2 and S3 Figs). Among these, several transcripts, including protein-coding transcripts (ADAMTSL4, NLGN3, SLC16A3, and SMARCC2), and lncRNAs (RN7SK, SLC12A5-AS1) exhibited higher expression levels in ASD samples compared to healthy controls (Fig 3c). ADAMTSL4, SMARCC2, and SLC12A5-AS1 demonstrated high significance (p-value < 0.001). SLC1A3 and NLGN3 were significantly different (p-value < 0.01), and RN7SK showed moderate significance (p-value < 0.05).
To identify novel lncRNAs associated with DNA/RNA hybrids, we analyzed 1,755 previously unannotated transcripts. The CPAT classified these transcripts using a coding probability threshold of 0.364, resulting in the identification of 1,048 novel lncRNAs and 707 protein-coding transcripts (S4 and S5 Figs, S4 Table). All 1048 novel lncRNAs were confirmed as unannotated as they were absent from the GENCODE v39 reference transcriptome. However, none of these novel lncRNAs met the defined threshold for differential expression in this cohort. Analysis of their exon structure revealed that the majority (58.1%) consisted of a single exon, while 36.1% contained two or more exons (S6 Fig). Principal component analysis (PCA) demonstrated separation of variance between the ASD patients and healthy controls (S7 Fig). Furthermore, hierarchical clustering identified a module containing 45 lncRNAs that were consistently enriched across all ASD patient samples (Fig 4).
Hierarchical clustering reveals a distinct module containing 45 novel lncRNAs (blue box) which represents a group of transcripts consistently enriched in the DNA/RNA hybrid fractions of the ASD patients samples (right).
Integrative Genomics Viewer (IGV) visualization
To further explore the functional relevance of differentially expressed transcripts, we cross-referenced our results with the SFARI database (https://gene.sfari.org/), a comprehensive resource cataloging autism-related genetic factors, identifying several transcripts previously reported as associated with autism, including GIGYF1, SMARCC2, EPHA1, IL1R2, KCNJ15, KMT2E, MYH10, NLGN3, SBF1, and SGSM3 (S5 Table).We further examined the mRNA expression levels of these selected transcripts across various tissues using the GTEx portal (https://gtexportal.org/home/) (S8 Fig). For the IGV visualization, reads from five ASD patients and three healthy controls were utilized. This analysis confirmed strong expression signals in ASD samples compared to healthy controls for key genes such as SLC16A3 and GIGYF1 (Fig 5). Furthermore, strong signals observed for other transcripts associated with DNA/RNA hybrids, including RN7SK and mitochondrial genes (MT-ND5 and MT-CYB), all of which exhibited significant signal differences in ASD samples compared to healthy controls (Fig 5).
ASD patient samples are represented by red tracks and healthy controls by blue tracks. The green arrow indicates the specific genomic loci selected for qRT-PCR validation.
Functional enrichment and KEGG pathway analysis
To explore the functional roles of differentially expressed transcripts associated with DNA/RNA hybrids in ASD patients (using |log2-fold change| ≥ 1 and p-value < 0.05), we conducted GO and KEGG pathway analyses.
The most significantly enriched GO terms were categorized for Biological Process (BP), Molecular Function (MF), and Cellular Component (CC) (Fig 6). The results demonstrated that the transcripts in ASD patients were primarily associated with mitochondrial function, the electron transport chain (ETC), OXPHOS, and energy metabolism (Fig 6). Other relevant GO terms within BP category included neuronal structure and development, such as regulation of plasma cell projection assembly and neuronal active transporters (Fig 6). Additionally, KEGG pathway over-representation analysis revealed a significant enrichment in pathways related to oxidative phosphorylation and multiple neurodegenerative disorders, including Alzheimer’s, Parkinson’s, Huntington’s disease, Amyotrophic Lateral Sclerosis (ALS), and prion disease (Fig 7).
The plots display the most enriched GO terms across BP, MF, and CC. Enrichment is primarily associated with mitochondrial function, the ETC and energy metabolism. The Y-axis represents the enriched GO terms, and the X-axis represents the gene ratio.
The size of each circle reveals the number of genes involved in the pathway. The color scale represents the adjusted p-value.
Upon further investigation, key mitochondrial genes were highlighted among these enriched GO terms (S9 Fig), including ND1 and ND5 (from ETC complex I), CYTB (from ETC complex III), COX1, COX2, and COX3 (from ETC complex IV), and ATP5F1E (from ETC complex V). Consistent with these results, IGV visualization also showed a high expression signal of mitochondrial genes, including ND5 and CYB in ASD patients compared to controls (Fig 5).
Validation of RNA-seq results by qRT-PCR analysis
To validate RNA-seq results, we isolated DNA/RNA hybrids-bound RNA and total RNA were from the blood samples of ASD patients and healthy controls. Six transcripts, identified as significantly differentially expressed in the RNA-seq analysis, were selected for validation: ADAMTSL4, NLGN3, RN7SK, SLC12A5-AS1, SLC16A3, and SMARCC2. Relative expression levels were quantified using qRT-PCR with GAPDH as a reference gene (primers listed in S6 Table).
In the DNA/RNA hybrids-bound RNA, qRT-PCR results were consistent with the RNA-seq, confirming the increased expression for all six transcripts (Fig 8). Specifically, RN7SK and SMARCC2 showed statistically significant upregulation in ASD patients compared to controls (p-value < 0.01). The remaining four transcripts–ADAMTSL4, NLGN3, SLC12A5-AS1, and SLC16A3–displayed a similar trend of increased expression but did not reach statistical significance.
Transcripts validated include the lncRNAs RN7SK and SLC12A5-AS1, and the protein-coding transcripts ADAMTSL4, NLGN3, SLC16A3, and SMARCC2. Data are presented as means ± SEM. Statistical significance was determined using a two-tailed, unpaired Student’s t-test (* p ≤ 0.05 and ** p ≤ 0.01).
In contrast, analysis of total RNA fraction showed significant differential expression (means± SEM) for four transcripts: ADAMTSL4, NLGN3, RN7SK, and SLC16A3 (p-value < 0.01). NLGN3 and RN7SK maintained elevated expression in ASD patients. However, ADAMTSL4 and SLC16A3 were downregulated relative to healthy controls (Fig 9). SMARCC2 and SLC12A5-AS1 did not show significant differential expression in the total RNA fraction.
Transcripts validated include the lncRNAs RN7SK and SLC12A5-AS1, and the protein-coding transcripts ADAMTSL4, NLGN3, SLC16A3, and SMARCC2. Data are presented as means ± SEM. Statistical significance was assessed by two-tailed, unpaired Student’s t-test (* p ≤ 0.05 and ** p ≤ 0.01).
Discussion
The aberrant formation of DNA/RNA hybrids can lead to genomic instability, potentially contributing to neurodevelopmental disorders. However, the mechanisms underlying genome dysregulation in autistic patients remains unclear. In this study, we compared the transcriptomic profiles of DNA/RNA hybrids in ASD patients and healthy controls, hypothesizing that variations in hybrid regions could provide evidence of the molecular mechanisms of disorder.
To investigate whether DNA/RNA hybrid regions are altered in ASD, we employed a robust, previously established methodology [24,25] that enables the recovery of RNA attached to genomic DNA within these hybrid structures. Transcriptome analysis of the resulting DNA/RNA hybrids from blood samples led to identification of over 1,000 previously unannotated/novel lncRNAs. Our results, highlight these novel transcripts associated with DNA/RNA hybrids, which originate from antisense and intergenic regions. Our comprehensive transcriptome assembly included 278,300 transcripts, of which 13,305 were known transcripts expressed across all samples. DGE analysis subsequently revealed 401 significantly downregulated and 301 upregulated known transcripts in ASD patients compared to controls. The pronounced accumulation of DNA/RNA hybrids observed in exonic and intronic regions of ASD blood cells suggests a direct link to altered transcriptional activity in both coding (exonic) and non-coding (intronic) sequences.
The role of lncRNAs has been contentious due to their initially unclear functions. However, accumulating evidence now showing that lncRNAs play critical roles in chromatin remodeling and enhancer regulation during development [46]. This underscores their importance in diverse biological processes. Indeed, dysregulation of ncRNAs has been implicated in the phenotypic changes of various complex disorders, including autism [47], suggesting a significant role for ncRNAs in transcriptional regulation within ASD. Consistent with this, our previous study reported decreased levels of six miRNAs in the immediate family members of autistic patients [23], and other studies have explored ncRNA expression in postmortem brain tissues and mouse models of ASD [48–50], further supporting their involvement in pathogenesis of ASD.
Our results align with the growing recognition that lncRNAs exhibit tissue-specific expression [51] and that structural alterations in DNA/RNA hybrids are important contributors to human diseases [13]. Consistent with this, our prior work has identified variations in DNA/RNA hybrids at telomeric regions and other ncRNAs across multiple pathologies [25,26]. This phenomenon is evident in two ASD-related conditions like Prader-Willi syndrome (PWS) and Angelman syndrome (AS), where complex epigenetic mechanisms underlie disease pathology [52]. This study extends these observations by focusing on dynamic RNA-DNA interactions, which could advance our understanding of cellular and developmental biology, as well as gene-environment interactions.
Recent research highlights the interplay between heredity and de novo mutations in autism. While over 100 susceptibility genes have been identified, they account for only a fraction of cases, indicating that they may not fully elucidate the disease aetiology [53]. Increasingly, non-Mendelian inheritance, characterized by predominant variations in ncRNAs, is being recognized as a key contributor. This mechanistic evidence highlights a potential avenue for autism intervention strategies aimed at mitigating risks associated with non-Mendelian inheritance.
Evaluating transcription levels within hybrid regions offers a novel approach to identifying biomarkers and biological pathways that could potentially restore normal gene expression. In this regard, we identified several potential biomarkers, including small nuclear ncRNA RN7SK and a subunit of the BAF chromatin remodeling complex SMARCC2. While information on RN7SK is limited, it has been linked to neuronal development and neurodevelopmental disorders [54,55,56]. Furthermore, SMARCC2, is considered a high-confidence ASD candidate [57,56]. A significant enrichment of rare missense variants in SMARCC2 among individuals with ASD, suggesting a potential genetic susceptibility [58].
Functional enrichment analysis indicated that the majority of enriched GO terms were associated with mitochondrial function and energy metabolism. Mitochondrial dysfunction is associated with various clinical and biochemical abnormalities in individuals with ASD, contributing to the diverse symptoms and comorbidities observed [59]. For example, deficiency of Cytochrome-c oxidase, a key enzyme in the ETC directly impairs energy production [60]. Additionally, dysregulation of plasma cell projection assembly, suggested by our GO analysis indicates defects in cell projections, including axon and dendrite formation, essential for neuronal polarity and circuit formation [61]. These processes may impact synapse formation and neuronal connectivity in ASDs. Other relevant GO terms linked to neuronal active transporters suggest that their impairment disrupts neurotransmitter uptake and ion homeostasis, potentially leading to synaptic dysfunction in neurodevelopmental pathologies [62].
The KEGG pathway over-representation analysis revealed enriched pathways in neurodevelopmental disorders, notably shared common mechanisms with autism. Mitochondrial dysfunction represents a central feature in the development and progression of several neurodegenerative conditions, including Alzheimer’s [63,64], Parkinson’s [65,66], Huntington’s disease, ALS [67], and prion disease [68]. This dysfunction broadly encompasses impaired energy production through perturbed OXPHOS and increased oxidative stress, which contributes to neuronal and synaptic damage, cognitive decline, and overall neuronal degeneration [69]. This shared pathology makes mitochondrial dysfunction and its key component, OXPHOS disruption, a promising target for novel diagnostic and therapeutic strategies across these disorders.
While these results, supported by preclinical proof-of-concept studies, hold promise, additional family-based research is warranted to clarify the role of non-Mendelian inheritance in autism. Phenotypic and disease variations across generations may arise from multigenic influences or non-Mendelian heritable mechanisms, necessitating further investigation under controlled experimental conditions. A key innovation of this work is its focus on “hybrid marks” and the exploration of pathways connecting DNA to RNA. Although the precise pathogenic role of DNA/RNA hybrids remains to be fully elucidated, deciphering their mechanisms may offer novel therapeutic opportunities. Comprehensive profiling of these hybrid regions is essential for future research, particularly in understanding their potential contribution to the transgenerational transmission of neurodevelopmental disorders. This mechanistic view underscores the need for caution in therapeutic applications, particularly given the potential for patient-specific variations in gene expression. To our knowledge, this study provides novel insights into variations of DNA/RNA hybrid-associated ncRNAs as heritable signals in autism, potentially serving as markers for ASD susceptibility.
Limitations of the study
The primary limitation of this study is the small sample size, which restricted the statistical power needed to ascertain subtle biological differences and limits the conclusive inference of the results. Future validation must be conducted in significantly larger cohort studies to enhance the robustness of results. A second limitation lies in the choice of biological matrix. While the analysis of DNA/RNA hybrid fractions in peripheral blood samples provided a foundational transcriptomic signature, this approach does not fully capture the complexities of non-Mendelian transmission, which are mediated in germline tissues. To more conclusively delineate the precise role of DNA/RNA hybrids in non-Mendelian inheritance, subsequent research must focus on examining these alterations in sperm, egg, or embryo samples. While such experiments present considerable ethical challenges in human subjects, further research incorporating preclinical models, dedicated clinical studies, and family-based experimental designs will be crucial for a comprehensive evaluation of non-Mendelian inheritance mechanisms in autism.
Conclusions
This work provides novel evidence suggesting a role for DNA/RNA hybrid variations in the molecular mechanisms underlying ASD. Whole transcriptome analysis demonstrated significant alterations of transcripts associated with DNA/RNA hybrids between ASD patients and controls, advancing mechanistic perspective of ASD. This highlights how alterations in hybrid regions, particularly those involving ncRNAs, may modulate gene expression (either enhancing or inhibiting transcription) and may contribute to the pathology of the disorder. The observation of key pathways like mitochondrial dysfunction and signals suggestive of non-Mendelian heritability underscore that dysregulation of DNA/RNA hybrid formation is likely a critical and complex molecular feature contributing to ASD susceptibility.
Supporting information
S1 Fig. RNA biotypes of known transcripts expressed in all blood samples.
The left pie chart shows RNA biotypes of all expressed transcripts, while the right pie chart displays RNA biotypes of differentially expressed transcripts.
https://doi.org/10.1371/journal.pone.0326901.s001
(TIF)
S2 Fig. Number of the most significantly differentially expressed transcripts between autistic patients and healthy controls.
The graph is based on normalized transcript counts. Blue and red bars represent autistic patients and healthy controls, respectively.
https://doi.org/10.1371/journal.pone.0326901.s002
(TIF)
S3 Fig. Volcano plot of differentially expressed transcripts.
The plot displays the relationship between log₂ fold change and statistical significance (adjusted p-value). The x-axis represents the log₂ fold change in transcript expression, while the y-axis shows the –log₁₀ of the adjusted p-value. Transcripts with significant differential expression (Padj < 0.05) are highlighted in red for those upregulated in autistic patients (log₂ fold change > 2) and in blue for those downregulated (log₂ fold change < –2).
https://doi.org/10.1371/journal.pone.0326901.s003
(TIF)
S4 Fig. Distribution of coding probability.
Graph shows number of transcripts per coding probability. The dashed red line shows the threshold (0.364). Coding probability below this amount considered non-coding transcripts.
https://doi.org/10.1371/journal.pone.0326901.s004
(TIF)
S5 Fig. Box plot showing the expression level of known transcripts through all samples.
Expression values are log2- normalized counts. The top of each bars shows the number of each transcript per genomic regions.
https://doi.org/10.1371/journal.pone.0326901.s005
(TIF)
S6 Fig. Exonic structure of novel lncRNAs.
94.2% of the lncRNA transcripts had either 1 or 2 exons.
https://doi.org/10.1371/journal.pone.0326901.s006
(TIF)
S7 Fig. Principal Component Analysis (PCA) plot of samples showing the separation between ASD patients and control based on gene expression profiles.
The first principal component (PC1) explains 68.8% of the variance, while the second component (PC2) explains 20.6%. Blue points represent ASD samples (S_03, S_05, S_06), and red points represent control samples (S_07, S_08), indicating distinct clustering between the two conditions.
https://doi.org/10.1371/journal.pone.0326901.s007
(TIF)
S8 Fig. Bulk tissue gene expression profiles of SMARCC2 and GIGYF1 across various tissue types, based on data from the GTEx project.
Notably, both genes show significant expression in whole blood.
https://doi.org/10.1371/journal.pone.0326901.s008
(TIF)
S9 Fig. Genes associated with enriched GO terms in biological processes.
Various genes of different mitochondrial electron transport chain complexes are present. The size of each circle representing GO terms indicates the number of genes associated with that term. Fold change (log2) is based on the normalized counts of each gene.
https://doi.org/10.1371/journal.pone.0326901.s009
(TIF)
S1 Table. Clinical characteristics of patients and healthy controls, including age and gender distribution.
https://doi.org/10.1371/journal.pone.0326901.s010
(PDF)
S2 Table. Sensitivity and precision percent across multiple levels of transcripts.
https://doi.org/10.1371/journal.pone.0326901.s011
(PDF)
S3 Table. Differential expression analysis2 revealed 702 differentially expressed transcripts, with 301 transcripts significantly upregulated, exhibiting a higher density of DNA/RNA hybrids in ASD patients (log2-fold change > 1, adjusted P < 0.05), while 401 transcripts were significantly downregulated (log2-fold change < −1, adjusted p < 0.05).
https://doi.org/10.1371/journal.pone.0326901.s012
(XLSX)
S4 Table. Novel lncRNAs based on coding probability < 0.364 from CPAT tools.
(Column 1 and 2: seq_ID): provides the identifier of the sequence being analyzed, (Column 3: mRNA): Sequence length of the mRNA or transcript, (Column 3: ORF): Indicates a specific Open Reading Frame (ORF) within the sequence, (Column 4: strand): shows the strand (+ forward, – reverse) of the transcript where the ORF is located, (Column 5: frame): indicates the reading frame (1, 2, or 3) of the ORF, (Column 6: ORF_start): specifies the starting position of the ORF, (Column 7: ORF_end): specifies the ending position of the ORF, (Column 8: ORF), (Column 9,10: ORFFickettn and Hexamer): contain scores related to nucleotide composition bias (Fickett) and hexamer usage bias, (Column 11: Coding_prob): presents the final probability score (between 0 and 1) of the ORF being protein-coding.
https://doi.org/10.1371/journal.pone.0326901.s013
(XLSX)
S5 Table. List of ASD-related genes among the DEGs, previously reported in the SFARI database as being associated with autism.
The cause of the genetic association of each gene with ASD and the number of reports for each gene in relation to ASD is also present.
https://doi.org/10.1371/journal.pone.0326901.s014
(PDF)
S6 Table. List of primers sequencing nucleotides forward and reverse for Real-Time PCR.
https://doi.org/10.1371/journal.pone.0326901.s015
(PDF)
Acknowledgments
We are grateful to the Betul Ziya Eren Genome and Stem Cell Center (Genkök) at Erciyes University for providing the facilities and resources necessary for this research.
References
- 1. Lakoucheva LM, Muotri AR, Sebat J. Getting to the cores of autism. Cell. 2019;178(6):1287–98. pmid:31491383
- 2. Auerbach BD, Osterweil EK, Bear MF. Mutations causing syndromic autism define an axis of synaptic pathophysiology. Nature. 2011;480(7375):63–8. pmid:22113615
- 3. Liu L, Lei J, Sanders SJ, Willsey AJ, Kou Y, Cicek AE, et al. DAWN: a framework to identify autism genes and subnetworks using gene expression and genetics. Mol Autism. 2014;5(1):22. pmid:24602502
- 4. Satterstrom FK, Kosmicki JA, Wang J, Breen MS, De Rubeis S, An J-Y, et al. Large-scale exome sequencing study implicates both developmental and functional changes in the neurobiology of autism. Cell. 2020;180(3):568–584.e23. pmid:31981491
- 5. Miles JH. Autism spectrum disorders--a genetics review. Genet Med. 2011;13(4):278–94. pmid:21358411
- 6. He Y, Zhou Y, Ma W, Wang J. An integrated transcriptomic analysis of autism spectrum disorder. Sci Rep. 2019;9(1):11818. pmid:31413321
- 7. Voineagu I, Wang X, Johnston P, Lowe JK, Tian Y, Horvath S, et al. Transcriptomic analysis of autistic brain reveals convergent molecular pathology. Nature. 2011;474(7351):380–4. pmid:21614001
- 8. Alter MD, Kharkar R, Ramsey KE, Craig DW, Melmed RD, Grebe TA, et al. Autism and increased paternal age related changes in global levels of gene expression regulation. PLoS One. 2011;6(2):e16715. pmid:21379579
- 9. García-Muse T, Aguilera A. R loops: from physiological to pathological roles. Cell. 2019;179(3):604–18. pmid:31607512
- 10. Crossley MP, Bocek M, Cimprich KA. R-loops as cellular regulators and genomic threats. Mol Cell. 2019;73(3):398–411. pmid:30735654
- 11. Marnef A, Legube G. R-loops as Janus-faced modulators of DNA repair. Nat Cell Biol. 2021;23(4):305–13. pmid:33837288
- 12. Niehrs C, Luke B. Regulatory R-loops as facilitators of gene expression and genome stability. Nat Rev Mol Cell Biol. 2020;21(3):167–78. pmid:32005969
- 13. Petermann E, Lan L, Zou L. Sources, resolution and physiological relevance of R-loops and RNA-DNA hybrids. Nat Rev Mol Cell Biol. 2022;23(8):521–40. pmid:35459910
- 14. Kabeche L, Nguyen HD, Buisson R, Zou L. A mitosis-specific and R loop-driven ATR pathway promotes faithful chromosome segregation. Science. 2018;359(6371):108–14. pmid:29170278
- 15. Liu W, Ma W, Yuan Y, Liu L, Zhang Y, Sun S. R-loops coordinate with transcription and replication to maintain genome stability in budding yeast. Nucleic Acids Res. 2021;49(11):6213–26. pmid:34048296
- 16. Groh M, Lufino MMP, Wade-Martins R, Gromak N. R-loops associated with triplet repeat expansions promote gene silencing in Friedreich ataxia and fragile X syndrome. PLoS Genet. 2014;10(5):e1004318. pmid:24787137
- 17. Haeusler AR, Donnelly CJ, Periz G, Simko EAJ, Shaw PG, Kim M-S, et al. C9orf72 nucleotide repeat structures initiate molecular cascades of disease. Nature. 2014;507(7491):195–200. pmid:24598541
- 18. Richard P, Manley JL. R loops and links to human disease. J Mol Biol. 2017;429(21):3168–80. pmid:27600412
- 19. Coquel F, Neumayer C, Lin Y-L, Pasero P. SAMHD1 and the innate immune response to cytosolic DNA during DNA replication. Curr Opin Immunol. 2019;56:24–30. pmid:30292848
- 20. Lima WF, Murray HM, Damle SS, Hart CE, Hung G, De Hoyos CL, et al. Viable RNaseH1 knockout mice show RNaseH1 is essential for R loop processing, mitochondrial and liver function. Nucleic Acids Res. 2016;44(11):5299–312. pmid:27131367
- 21. Mackenzie KJ, Carroll P, Martin C-A, Murina O, Fluteau A, Simpson DJ, et al. cGAS surveillance of micronuclei links genome instability to innate immunity. Nature. 2017;548(7668):461–5. pmid:28738408
- 22. Skourti-Stathaki K, Kamieniarz-Gdula K, Proudfoot NJ. R-loops induce repressive chromatin marks over mammalian gene terminators. Nature. 2014;516(7531):436–9. pmid:25296254
- 23. Ozkul Y, Taheri S, Bayram KK, Sener EF, Mehmetbeyoglu E, Öztop DB, et al. A heritable profile of six miRNAs in autistic patients and mouse models. Sci Rep. 2020;10(1):9011. pmid:32514154
- 24. Kianmehr L, Khazali H, Rajabi-Maham H, Sharifi-Zarchi A, Cuzin F, Rassoulzadegan M. Genome-wide distribution of nascent transcripts in sperm dna, products of a late wave of general transcription. Cells. 2019;8(10):1196. pmid:31623338
- 25. Rassoulzadegan M, Sharifi-Zarchi A, Kianmehr L. DNA-RNA Hybrid (R-Loop): from a unified picture of the mammalian telomere to the genome-wide profile. Cells. 2021;10(6):1556. pmid:34205454
- 26. Mehmetbeyoglu E, Kianmehr L, Borlu M, Yilmaz Z, Basar Kılıc S, Rajabi-Maham H, et al. Decrease in RNase HII and Accumulation of lncRNAs/DNA hybrids: a causal implication in psoriasis? Biomolecules. 2022;12(3):368. pmid:35327560
- 27.
American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-IV-TR. 4th ed. text rev. ed. Washington (DC): American Psychiatric Association; 2000.
- 28.
American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-5. 5th ed. Washington (DC): American Psychiatric Association; 2013.
- 29. 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(1):91–103. pmid:6927682
- 30. Rio DC, Ares M Jr, Hannon GJ, Nilsen TW. Purification of RNA using TRIzol (TRI reagent). Cold Spring Harb Protoc. 2010;2010(6):pdb.prot5439. pmid:20516177
- 31. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20. pmid:24695404
- 32. Kim D, Paggi JM, Park C, Bennett C, Salzberg SL. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol. 2019;37(8):907–15. pmid:31375807
- 33. Wang L, Wang S, Li W. RSeQC: quality control of RNA-seq experiments. Bioinformatics. 2012;28(16):2184–5. pmid:22743226
- 34. Pertea M, Pertea GM, Antonescu CM, Chang T-C, Mendell JT, Salzberg SL. StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat Biotechnol. 2015;33(3):290–5. pmid:25690850
- 35. Anders S, Pyl PT, Huber W. HTSeq--a Python framework to work with high-throughput sequencing data. Bioinformatics. 2015;31(2):166–9. pmid:25260700
- 36. Robinson JT, Thorvaldsdóttir H, Winckler W, Guttman M, Lander ES, Getz G, et al. Integrative genomics viewer. Nat Biotechnol. 2011;29(1):24–6. pmid:21221095
- 37. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. pmid:25516281
- 38. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2021;2(3):100141. pmid:34557778
- 39. Pertea G, Pertea GM. GFF Utilities: GffRead and GffCompare. F1000Res. 2020;9:1087. pmid:32489650
- 40. Shen W, Le S, Li Y, Hu F. SeqKit: a cross-platform and ultrafast toolkit for FASTA/Q file manipulation. PLoS One. 2016;11(10):e0163962. pmid:27706213
- 41. Wang L, Park HJ, Dasari S, Wang S, Kocher J-P, Li W. CPAT: coding-potential assessment tool using an alignment-free logistic regression model. Nucleic Acids Res. 2013;41(6):e74. pmid:23335781
- 42. Heinz S, Benner C, Spann N, Bertolino E, Lin YC, Laslo P, et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell. 2010;38(4):576–89. pmid:20513432
- 43. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4):402–8. pmid:11846609
- 44. Kolde R, Kolde MR. Package ‘pheatmap’. R package. 2015;1:790.
- 45.
Wickham H. ggplot2: Elegant Graphics for Data Analysis. New York: Springer; 2016.
- 46. Mattick JS, Amaral PP, Carninci P, Carpenter S, Chang HY, Chen L-L, et al. Long non-coding RNAs: definitions, functions, challenges and recommendations. Nat Rev Mol Cell Biol. 2023;24(6):430–47. pmid:36596869
- 47. Qureshi IA, Mehler MF. Long non-coding RNAs: novel targets for nervous system disease diagnosis and therapy. Neurotherapeutics. 2013;10(4):632–46. pmid:23817781
- 48. Parikshak NN, Swarup V, Belgard TG, Irimia M, Ramaswami G, Gandal MJ, et al. Genome-wide changes in lncRNA, splicing, and regional gene expression patterns in autism. Nature. 2016;540(7633):423–7. pmid:27919067
- 49. Velmeshev D, Magistri M, Faghihi MA. Expression of non-protein-coding antisense RNAs in genomic regions related to autism spectrum disorders. Mol Autism. 2013;4(1):32. pmid:24007600
- 50. Koç B, Fucile G, Schmucki R, Giroud N, Bergauer T, Hall BJ. Identification of natural antisense transcripts in mouse brain and their association with autism spectrum disorder risk genes. Front Mol Neurosci. 2021;14:624881. pmid:33716665
- 51. Statello L, Guo C-J, Chen L-L, Huarte M. Gene regulation by long non-coding RNAs and its biological functions. Nat Rev Mol Cell Biol. 2021;22(2):96–118. pmid:33353982
- 52.
Powell WT. Investigation of transcription regulation in the Prader-Willi/Angelman syndrome locus in neurons and the role of long, non-coding RNAs. Davis: University of California; 2013.
- 53. Cirnigliaro M, Chang TS, Arteaga SA, Pérez-Cano L, Ruzzo EK, Gordon A, et al. The contributions of rare inherited and polygenic risk to ASD in multiplex families. Proc Natl Acad Sci U S A. 2023;120(31):e2215632120. pmid:37506195
- 54. Bazi Z, Bertacchi M, Abasi M, Mohammadi-Yeganeh S, Soleimani M, Wagner N, et al. Rn7SK small nuclear RNA is involved in neuronal differentiation. J Cell Biochem. 2018;119(4):3174–82. pmid:29091296
- 55. Schneeberger PE, Bierhals T, Neu A, Hempel M, Kutsche K. De novo MEPCE nonsense variant associated with a neurodevelopmental disorder causes disintegration of 7SK snRNP and enhanced RNA polymerase II activation. Sci Rep. 2019;9(1):12516. pmid:31467394
- 56. Ben-David E, Shifman S. Combined analysis of exome sequencing points toward a major role for transcription regulation during brain development in autism. Mol Psychiatry. 2013;18(10):1054–6. pmid:23147383
- 57. Nussinov R, Yavuz BR, Arici MK, Demirel HC, Zhang M, Liu Y, et al. Neurodevelopmental disorders, like cancer, are connected to impaired chromatin remodelers, PI3K/mTOR, and PAK1-regulated MAPK. Biophys Rev. 2023;15(2):163–81. pmid:37124926
- 58. Machol K, Rousseau J, Ehresmann S, Garcia T, Nguyen TTM, Spillmann RC, et al. Expanding the Spectrum of BAF-Related Disorders: De Novo Variants in SMARCC2 Cause a Syndrome with Intellectual Disability and Developmental Delay. Am J Hum Genet. 2019;104(1):164–78. pmid:30580808
- 59. Siddiqui MF, Elwell C, Johnson MH. Mitochondrial dysfunction in autism spectrum disorders. Autism Open Access. 2016;6(5):1000190. pmid:27928515
- 60. Srinivasan S, Avadhani NG. Cytochrome c oxidase dysfunction in oxidative stress. Free Radic Biol Med. 2012;53(6):1252–63. pmid:22841758
- 61. Takano T, Funahashi Y, Kaibuchi K. Neuronal polarity: positive and negative feedback signals. Front Cell Dev Biol. 2019;7:69. pmid:31069225
- 62. Mermer F, Poliquin S, Rigsby K, Rastogi A, Shen W, Romero-Morales A, et al. Common molecular mechanisms of SLC6A1 variant-mediated neurodevelopmental disorders in astrocytes and neurons. Brain. 2021;144(8):2499–512. pmid:34028503
- 63. Perez Ortiz JM, Swerdlow RH. Mitochondrial dysfunction in Alzheimer’s disease: role in pathogenesis and novel therapeutic opportunities. Br J Pharmacol. 2019;176(18):3489–507. pmid:30675901
- 64. Ashleigh T, Swerdlow RH, Beal MF. The role of mitochondrial dysfunction in Alzheimer’s disease pathogenesis. Alzheimers Dement. 2023;19(1):333–42. pmid:35522844
- 65. Bose A, Beal MF. Mitochondrial dysfunction in Parkinson’s disease. J Neurochem. 2016;139 Suppl 1:216–31. pmid:27546335
- 66. Chen C, Turnbull DM, Reeve AK. Mitochondrial dysfunction in Parkinson’s Disease-cause or consequence? Biology (Basel). 2019;8(2):38. pmid:31083583
- 67. Alqahtani T, Deore SL, Kide AA, Shende BA, Sharma R, Dadarao Chakole R, et al. Mitochondrial dysfunction and oxidative stress in Alzheimer’s disease, and Parkinson’s disease, Huntington’s disease and amyotrophic lateral sclerosis - an updated review. Mitochondrion. 2023;71:83–92. pmid:37269968
- 68. Keller G, Binyamin O, Frid K, Saada A, Gabizon R. Mitochondrial dysfunction in preclinical genetic prion disease: A target for preventive treatment? Neurobiol Dis. 2019;124:57–66. pmid:30423473
- 69. Ayala-Peña S. Role of oxidative DNA damage in mitochondrial dysfunction and Huntington’s disease pathogenesis. Free Radic Biol Med. 2013;62:102–10. pmid:23602907