Exploring genetic variation that influences brain methylation in attention-deficit/hyperactivity disorder

Translational bioinformatics and data science for biomarker discovery in mental health: an analytical review

Translational bioinformatics and data science play a crucial role in biomarker discovery as it enables translational research and helps to bridge the gap between the bench research and the bedside clinical applications. Thanks to newer and faster molecular profiling technologies and reducing costs, there are many opportunities for researchers to explore the molecular and physiological mechanisms of diseases. Biomarker discovery enables researchers to better characterize patients, enables early detection and intervention/prevention and predicts treatment responses. Due to increasing prevalence and rising treatment costs, mental health (MH) disorders have become an important venue for biomarker discovery with the goal of improved patient diagnostics, treatment and care. Exploration of underlying biological mechanisms is the key to the understanding of pathogenesis and pathophysiology of MH disorders. In an effort to better understand the underlying mechanisms of MH disorders, we reviewed the major accomplishments in the MH space from a bioinformatics and data science perspective, summarized existing knowledge derived from molecular and cellular data and described challenges and areas of opportunities in this space.

DAT1 5'-Un-Translated-Region Methylation Patterns as Bio-Markers of ADHD Psycho-Pathology: Contribution to Disease Prognosis and to Monitoring of a Successful Therapy

Epigenetic modifications, such as changes in DNA methylation, have been linked to several diseases in recent years. The purpose of our study was to search for biomarkers that (using non-invasive techniques) could assist the clinician in the prognosis of infant/adolescent psychopathology. We previously showed that changes in methylation of the 5'-UTR in the DAT1/SLC6A3 gene can be used as a biomarker for the prognosis of initial severe ADHD: treatment-resistant severe ADHD children were characterized by methylated CpG 1 in particular, while methylated CpGs 2 and 6 were then found in children who improved after the therapy. Further, we confirmed these outcomes and provided the hypothesis that symptomatology might be influenced by the children's genotype and family environment. In particular, levels of CpG 3 methylation in the heterozygous ADHD children were associated with high paternal own risk or stress. Eventually, we found that the same biomarkers are more broadly useful in the field of internalizing or externalizing symptoms (when a certain vulnerability is already present in the child). In particular, it was seen how inheriting specific 9-repeat or 10-repeat VNTR alleles from the mother or from the father could modify the pattern of methylation at the 5'-UTR of the DAT1 gene. A specific pattern of methylations (with CpG 2 following either CpGs 1 + 3 or CpG 6 at the DAT1 5'-UTR) has been associated, therefore, with the likelihood of an internalizing or externalizing developmental trajectory entailing ADHD-like psycho-pathological characteristics. Since each individual responds differently to a specific treatment, we suggest that these methylation patterns may be used as biomarkers to monitor the outcome and/or predict the success of a given therapy (personalized medicine).

Child Developmental MRI (CDM) project: protocol for a multi-centre, cross-sectional study on elucidating the pathophysiology of attention-deficit/hyperactivity disorder and autism spectrum disorder through a multi-dimensional approach

INTRODUCTION

Neuroimaging studies on attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorder (ASD) have demonstrated differences in extensive brain structure, activity and network. However, there remains heterogeneity and inconsistency across these findings, presumably because of the diversity of the disorders themselves, small sample sizes, and site and parameter differences in MRI scanners, and their overall pathogenesis remains unclear. To address these gaps in the literature, we will apply the travelling-subject approach to correct site differences in MRI scanners and clarify brain structure and network characteristics of children with ADHD and ASD using large samples collected in a multi-centre collaboration. In addition, we will investigate the relationship between these characteristics and genetic, epigenetic, biochemical markers, and behavioural and psychological measures.

METHODS AND ANALYSIS

We will collect resting-state functional MRI (fMRI) and T1-weighted and diffusion-weighted MRI data from 15 healthy adults as travelling subjects and 300 children (ADHD, n=100; ASD, n=100; and typical development, n=100) with multi-dimensional assessments. We will also apply data from more than 1000 samples acquired in our previous neuroimaging studies on ADHD and ASD.

ETHICS AND DISSEMINATION

The study protocol has been approved by the Research Ethics Committee of the University of Fukui Hospital (approval no: 20220601). Our study findings will be submitted to scientific peer-reviewed journals and conferences.

Control of centrosome distal appendages assembly and disassembly

Centrosomes are microtubule organizing centers involved in chromosome segregation, spindle orientation, cell motility and cilia formation. In recent years, they have also emerged as key modulators of asymmetric cell division. Centrosomes are composed of two centrioles that initiate duplication in S phase. The conservative nature of centriole duplication means that the two centrioles of a G1 cell are of different ages. They are also structurally different as only the older centriole carry appendages, an assembly of a subset of proteins primarily required for cilia formation. In a growing tissue, the non-motile, primary cilium acts as a mechano- and sensory organelle that influences cell behavior via modulation of signaling pathways. Here, we discuss the most recent findings about distal appendage composition and function, as well as cell cycle-specific regulation and their implications in various diseases.

Polygenic risk score for attention-deficit/hyperactivity disorder and brain functional networks segregation in a community-based sample

Neuroimaging studies suggest that brain development mechanisms might explain at least some behavioural and cognitive attention-deficit/hyperactivity disorder (ADHD) symptoms. However, the putative mechanisms by which genetic susceptibility factors influence clinical features via alterations of brain development remain largely unknown. Here, we set out to integrate genomics and connectomics tools by investigating the associations between an ADHD polygenic risk score (ADHD-PRS) and functional segregation of large-scale brain networks. With this aim, ADHD symptoms score, genetic and rs-fMRI (resting-state functional magnetic resonance image) data obtained in a longitudinal community-based cohort of 227 children and adolescents were analysed. A follow-up was conducted approximately 3 years after the baseline, with rs-fMRI scanning and ADHD likelihood assessment in both stages. We hypothesised a negative correlation between probable ADHD and the segregation of networks involved in executive functions, and a positive correlation with the default-mode network (DMN). Our findings suggest that ADHD-PRS is correlated with ADHD at baseline, but not at follow-up. Despite not surviving for multiple comparison correction, we found significant correlations between ADHD-PRS and segregation of cingulo-opercular networks and DMN at baseline. ADHD-PRS was negatively correlated with the segregation level of cingulo-opercular networks but positively correlated with the DMN segregation. These directions of associations corroborate the proposed counter-balanced role of attentional networks and DMN in attentional processes. However, the association between ADHD-PRS and brain networks functional segregation was not found at follow-up. Our results provide evidence for specific influences of genetic factors on development of attentional networks and DMN. We found significant correlations between polygenic risk score for ADHD (ADHD-PRS) and segregation of cingulo-opercular networks and default-mode network (DMN) at baseline. ADHD-PRS was negatively correlated with the segregation level of cingulo-opercular networks but positively correlated with the DMN segregation.

Catecholaminergic and cholinergic neuromodulation in autism spectrum disorder: A comparison to attention-deficit hyperactivity disorder

Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental disorder characterized by social impairments and restricted, repetitive behaviors. Treatment of ASD is notoriously difficult and might benefit from identification of underlying mechanisms that overlap with those disturbed in other developmental disorders, for which treatment options are more obvious. One example of the latter is attention-deficit hyperactivity disorder (ADHD), given the efficacy of especially stimulants in treatment of ADHD. Deficiencies in catecholaminergic systems [dopamine (DA), norepinephrine (NE)] in ADHD are obvious targets for stimulant treatment. Recent findings suggest that dysfunction in catecholaminergic systems may also be a factor in at least a subgroup of ASD. In this review we scrutinize the evidence for catecholaminergic mechanisms underlying ASD symptoms, and also include in this analysis a third classic ascending arousing system, the acetylcholinergic (ACh) network. We complement this with a comprehensive review of DA-, NE-, and ACh-targeted interventions in ASD, and an exploratory search for potential treatment-response predictors (biomarkers) in ASD, genetically or otherwise. Based on this review and analysis we propose that (1) stimulant treatment may be a viable option for an ASD subcategory, possibly defined by genetic subtyping; (2) cerebellar dysfunction is pronounced for a relatively small ADHD subgroup but much more common in ASD and in both cases may point toward NE- or ACh-directed intervention; (3) deficiency of the cortical salience network is sizable in subgroups of both disorders, and biomarkers such as eye blink rate and pupillometric data may predict the efficacy of targeting this underlying deficiency via DA, NE, or ACh in both ASD and ADHD.

Epigenetics and ADHD

There is robust evidence of genetic susceptibility to Attention-Deficit Hyperactivity Disorder (ADHD); however, there still remains significant variability that is not attributable to genetic factors. The emerging field of epigenetics is beginning to reveal how genotypic expression can be mediated by an array of variables including external environmental exposure, inter-individual developmental variation, and by the genome itself. Epigenetic modification plays a central role in neurobiological and developmental processes, and disturbances to these processes can have implications for a range of mental health problems. Although the field is still in its early days, this chapter will discuss the current standing of epigenetic research into ADHD. Firstly, key relevant epigenetic processes will be discussed. This will be followed by an overview of the key findings to date investigating the role of epigenetics in ADHD. Human studies have included the theory-driven approach of candidate-gene studies (CGS), as well as the increasingly popular exploratory approach of epigenome-wide association studies (EWAS). Overall, the findings are heterogeneous. However, it is possible that with more longitudinal studies and better characterised cohorts, both predictive and protective links between epigenetic processes and ADHD will be revealed.

The Role of Epigenetics in the Pathogenesis and Potential Treatment of Attention Deficit Hyperactivity Disorder

There is increasing evidence that dysregulated epigenetic mechanisms of gene expression are involved in the pathogenesis of attention deficit hyperactivity disorder (ADHD). This review presents a comprehensive summary of the current state of research on the role of epigenetics in the pathogenesis of ADHD. The potential role of epigenetic drugs in the treatment of ADHD is also reviewed. Several studies suggest that there are epigenetic abnormalities in preclinical models of ADHD and in ADHD patients. Regarding DNA methylation, many studies have reported DNA hypermethylation. There is evidence that there is increased histone deacetylation in ADHD patients. Abnormalities in the expression of microRNAs (miRNAs) in ADHD patients have also been found. Some currently used drugs for treating ADHD, in addition to their more well-established mechanisms of action, have been shown to alter epigenetic mechanisms of gene expression. Clinical trials of epigenetic drugs in patients with ADHD report favorable results. These data suggest that abnormal epigenetic mechanisms of gene expression may be involved in the pathogenesis of ADHD. Drugs acting on epigenetic mechanisms may be a potential new class of drugs for treating ADHD.

Identification of genetic variants influencing methylation in brain with pleiotropic effects on psychiatric disorders

Psychiatric disorders affect 29% of the global population at least once in the lifespan, and genetic studies have proved a shared genetic basis among them, although the underlying molecular mechanisms remain largely unknown. DNA methylation plays an important role in complex disorders and, remarkably, enrichment of common genetic variants influencing allele-specific methylation (ASM) has been reported among variants associated with specific psychiatric disorders. In the present study we assessed the contribution of ASM to a set of eight psychiatric disorders by combining genetic, epigenetic and expression data. We interrogated a list of 3896 ASM tagSNPs in the brain in the summary statistics of a cross-disorder GWAS meta-analysis of eight psychiatric disorders from the Psychiatric Genomics Consortium, including more than 162,000 cases and 276,000 controls. We identified 80 SNPs with pleiotropic effects on psychiatric disorders that show an opposite directional effect on methylation and gene expression. These SNPs converge on eight candidate genes: ZSCAN29, ZSCAN31, BTN3A2, DDAH2, HAPLN4, ARTN, FAM109B and NAGA. ZSCAN29 shows the broadest pleiotropic effects, showing associations with five out of eight psychiatric disorders considered, followed by ZSCAN31 and BTN3A2, associated with three disorders. All these genes overlap with CNVs related to cognitive phenotypes and psychiatric traits, they are expressed in the brain, and seven of them have previously been associated with specific psychiatric disorders, supporting our results. To sum up, our integrative functional genomics analysis identified eight psychiatric disease risk genes that impact a broad list of disorders and highlight an etiologic role of SNPs that influence DNA methylation and gene expression in the brain.

Biallelic mutations in the death domain of PIDD1 impair caspase-2 activation and are associated with intellectual disability

PIDD1 encodes p53-Induced Death Domain protein 1, which acts as a sensor surveilling centrosome numbers and p53 activity in mammalian cells. Early results also suggest a role in DNA damage response where PIDD1 may act as a cell-fate switch, through interaction with RIP1 and NEMO/IKKg, activating NF-κB signaling for survival, or as an apoptosis-inducing protein by activating caspase-2. Biallelic truncating mutations in CRADD—the protein bridging PIDD1 and caspase-2—have been reported in intellectual disability (ID), and in a form of lissencephaly. Here, we identified five families with ID from Iran, Pakistan, and India, with four different biallelic mutations in PIDD1, all disrupting the Death Domain (DD), through which PIDD1 interacts with CRADD or RIP1. Nonsense mutations Gln863* and Arg637* directly disrupt the DD, as does a missense mutation, Arg815Trp. A homozygous splice mutation in the fifth family is predicted to disrupt splicing upstream of the DD, as confirmed using an exon trap. In HEK293 cells, we show that both Gln863* and Arg815Trp mutants fail to co-localize with CRADD, leading to its aggregation and mis-localization, and fail to co-precipitate CRADD. Using genome-edited cell lines, we show that these three PIDD1 mutations all cause loss of PIDDosome function. Pidd1 null mice show decreased anxiety, but no motor abnormalities. Together this indicates that PIDD1 mutations in humans may cause ID (and possibly lissencephaly) either through gain of function or secondarily, due to altered scaffolding properties, while complete loss of PIDD1, as modeled in mice, may be well tolerated or is compensated for.

Biallelic mutations in the death domain of PIDD1 impair caspase-2 activation and are associated with intellectual disability

PIDD1 encodes p53-Induced Death Domain protein 1, which acts as a sensor surveilling centrosome numbers and p53 activity in mammalian cells. Early results also suggest a role in DNA damage response where PIDD1 may act as a cell-fate switch, through interaction with RIP1 and NEMO/IKKg, activating NF-κB signaling for survival, or as an apoptosis-inducing protein by activating caspase-2. Biallelic truncating mutations in CRADD-the protein bridging PIDD1 and caspase-2-have been reported in intellectual disability (ID), and in a form of lissencephaly. Here, we identified five families with ID from Iran, Pakistan, and India, with four different biallelic mutations in PIDD1, all disrupting the Death Domain (DD), through which PIDD1 interacts with CRADD or RIP1. Nonsense mutations Gln863* and Arg637* directly disrupt the DD, as does a missense mutation, Arg815Trp. A homozygous splice mutation in the fifth family is predicted to disrupt splicing upstream of the DD, as confirmed using an exon trap. In HEK293 cells, we show that both Gln863* and Arg815Trp mutants fail to co-localize with CRADD, leading to its aggregation and mis-localization, and fail to co-precipitate CRADD. Using genome-edited cell lines, we show that these three PIDD1 mutations all cause loss of PIDDosome function. Pidd1 null mice show decreased anxiety, but no motor abnormalities. Together this indicates that PIDD1 mutations in humans may cause ID (and possibly lissencephaly) either through gain of function or secondarily, due to altered scaffolding properties, while complete loss of PIDD1, as modeled in mice, may be well tolerated or is compensated for.

Perinatal protein malnutrition results in genome-wide disruptions of 5-hydroxymethylcytosine at regions that can be restored to control levels by an enriched environment

Maternal malnutrition remains one of the major adversities affecting brain development and long-term mental health outcomes, increasing the risk to develop anxiety and depressive disorders. We have previously shown that malnutrition-induced anxiety-like behaviours can be rescued by a social and sensory stimulation (enriched environment) in male mice. Here, we expand these findings to adult female mice and profiled genome-wide ventral hippocampal 5hmC levels related to malnutrition-induced anxiety-like behaviours and their rescue by an enriched environment. This approach revealed 508 differentially hydroxymethylated genes associated with protein malnutrition and that several genes (N = 34) exhibited a restored 5hmC abundance to control levels following exposure to an enriched environment, including genes involved in neuronal functions like dendrite outgrowth, axon guidance, and maintenance of neuronal circuits (e.g. Fltr3, Itsn1, Lman1, Lsamp, Nav, and Ror1) and epigenetic mechanisms (e.g. Hdac9 and Dicer1). Sequence motif predictions indicated that 5hmC may be modulating the binding of transcription factors for several of these transcripts, suggesting a regulatory role for 5hmC in response to perinatal malnutrition and exposure to an enriched environment. Together, these findings establish a role for 5hmC in early-life malnutrition and reveal genes linked to malnutrition-induced anxious behaviours that are mitigated by an enriched environment.

Obsessive-compulsive disorder and attention-deficit/hyperactivity disorder: distinct associations with DNA methylation and genetic variation

BACKGROUND

A growing body of research has demonstrated associations between specific neurodevelopmental disorders and variation in DNA methylation (DNAm), implicating this molecular mark as a possible contributor to the molecular etiology of these disorders and/or as a novel disease biomarker. Furthermore, genetic risk variants of neurodevelopmental disorders have been found to be enriched at loci associated with DNAm patterns, referred to as methylation quantitative trait loci (mQTLs).

METHODS

We conducted two epigenome-wide association studies in individuals with attention-deficit/hyperactivity disorder (ADHD) or obsessive-compulsive disorder (OCD) (aged 4-18 years) using DNA extracted from saliva. DNAm data generated on the Illumina Human Methylation 450 K array were used to examine the interaction between genetic variation and DNAm patterns associated with these disorders.

RESULTS

Using linear regression followed by principal component analysis, individuals with the most endorsed symptoms of ADHD or OCD were found to have significantly more distinct DNAm patterns from controls, as compared to all cases. This suggested that the phenotypic heterogeneity of these disorders is reflected in altered DNAm at specific sites. Further investigations of the DNAm sites associated with each disorder revealed that despite little overlap of these DNAm sites across the two disorders, both disorders were significantly enriched for mQTLs within our sample.

CONCLUSIONS

Our DNAm data provide insights into the regulatory changes associated with genetic variation, highlighting their potential utility both in directing GWAS and in elucidating the pathophysiology of neurodevelopmental disorders.

Involvement of the 14-3-3 Gene Family in Autism Spectrum Disorder and Schizophrenia: Genetics, Transcriptomics and Functional Analyses

The 14-3-3 protein family are molecular chaperones involved in several biological functions and neurological diseases. We previously pinpointed YWHAZ (encoding 14-3-3ζ) as a candidate gene for autism spectrum disorder (ASD) through a whole-exome sequencing study, which identified a frameshift variant within the gene (c.659-660insT, p.L220Ffs*18). Here, we explored the contribution of the seven human 14-3-3 family members in ASD and other psychiatric disorders by investigating the: (i) functional impact of the 14-3-3ζ mutation p.L220Ffs*18 by assessing solubility, target binding and dimerization; (ii) contribution of common risk variants in 14-3-3 genes to ASD and additional psychiatric disorders; (iii) burden of rare variants in ASD and schizophrenia; and iv) 14-3-3 gene expression using ASD and schizophrenia transcriptomic data. We found that the mutant 14-3-3ζ protein had decreased solubility and lost its ability to form heterodimers and bind to its target tyrosine hydroxylase. Gene-based analyses using publicly available datasets revealed that common variants in YWHAE contribute to schizophrenia (p = 6.6 × 10-7), whereas ultra-rare variants were found enriched in ASD across the 14-3-3 genes (p = 0.017) and in schizophrenia for YWHAZ (meta-p = 0.017). Furthermore, expression of 14-3-3 genes was altered in post-mortem brains of ASD and schizophrenia patients. Our study supports a role for the 14-3-3 family in ASD and schizophrenia.

Biochemical characteristics of the chondrocyte-enriched SNORC protein and its transcriptional regulation by SOX9

Snorc (Small NOvel Rich in Cartilage) has been identified as a chondrocyte-specific gene in the mouse. Yet little is known about the SNORC protein biochemical properties, and mechanistically how the gene is regulated transcriptionally in a tissue-specific manner. The goals of the present study were to shed light on those important aspects. The chondrocyte nature of Snorc expression was confirmed in mouse and rat tissues, in differentiated (day 7) ATDC5, and in RCS cells where it was constitutive. Topological mapping and biochemical analysis brought experimental evidences that SNORC is a type I protein carrying a chondroitin sulfate (CS) attached to serine 44. The anomalous migration of SNORC on SDS-PAGE was due to its primary polypeptide features, suggesting no additional post-translational modifications apart from the CS glycosaminoglycan. A highly conserved SOX9-binding enhancer located in intron 1 was necessary to drive transcription of Snorc in the mouse, rat, and human. The enhancer was active independently of orientation and whether located in a heterologous promoter or intron. Crispr-mediated inactivation of the enhancer in RCS cells caused reduction of Snorc. Transgenic mice carrying the intronic multimerized enhancer drove high expression of a βGeo reporter in chondrocytes, but not in the hypertrophic zone. Altogether these data confirmed the chondrocyte-specific nature of Snorc and revealed dependency on the intronic enhancer binding of SOX9 for transcription.

Methylphenidate for Attention-Deficit and Hyperactivity Disorder in Adult Patients With Substance Use Disorders: Good Clinical Practice

Attention-deficit and hyperactivity disorder (ADHD) is a widespread neurodevelopmental disorder in children and adolescents, persisting into adulthood in a majority of them. ADHD and substance use disorders (SUDs) commonly co-occur in the clinical adult population. The higher-than-normal prevalence rates of SUDs in people with ADHD indicate increased risk for developing SUD. This narrative review deals with the question of whether or not adults with both disorders should be treated with methylphenidate (MPH), addressing specific issues surrounding this form of treatment. MPH is considered as first-line pharmacotherapy for ADHD. However, because of its stimulant-like reinforcing properties, MPH has a significant addictive potential to which persons with SUDs are especially susceptible. Appropriate treatment is therefore complex. Because of concerns about misuse and diversion of MPH medication, clinicians may be reluctant to use MPH to manage ADHD symptoms in these patients. However, it is essential to diagnose and treat ADHD adequately as appropriate therapy reduces the impairments, as well as the risk of developing comorbid disorders and poor treatment response. MPH should not be deprived of these patients because of the risk for misuse, especially as several strategies can be applied to minimize this risk. To conclude, carefully applied guideline-based diagnostics to clarify the potential presence of ADHD as well as a responsible prescription practice in a well-defined therapeutic setting with reliable monitoring of medication intake and regular consultations are essential conditions for a safe and proficient MPH treatment of ADHD in patients with SUD.