**9. Chromatin remodeling**

type of study [120, 150]. Generally these two themes (CNV and miRNAs) are investigated

A few studies have investigated the transcriptome in ASD samples of postmortem brains and some of them used mRNA from peripheral blood of patients [1, 54, 71, 110]. More recently, disruption of miRNA expression has been repeatedly reported in microarray studies and it is believed to be linked to the pathogenesis of autism (Sarachana et al., 2010; [28, 55]. However, lymphoblastoid cells are not representative of neural tissue and very few miRNAs exhibit

According to [93], miRNA loci are underrepresented in highly polymorphic and wellvalidated CNV regions. One study investigated the pathogenic role of miRNAs in autism by checking associations with *de novo* CNVs. Twenty-four miRNA genes likely to play a patho‐ genic role in autism were identified on chromosomes 1, 2 and 22. Two, *mir-HSA-4436b-1 and 4436b-HSA-mir-2*, appear to be strong candidates. Unfortunately, the targets of these miRNAs were not identified [94]. The difference in penetrance of the deleted/duplicated genes may be explained by a variety of factors including: (i) pre-natal exposure to environmental risk factors; (ii) the presence/absence of functional SNPs in genes that encode proteins related to suscept‐ ibility for autism; (iii) epistasis; (iv) epigenetic factors and (v) the number and type of genes encoding proteins co-existing in different CNVs and overlapping in the same miRNA.

The *HEY1, SOX9*, *miR-486* and *miR-181b* are some candidate genes. All of these are involved in the development and function of the nervous system, and some, such as *HEY1*, are involved in Notch signaling networks [55]. However, a systematic analysis of CNV-miRNAs based on their interactions with target genes identified other miRNAs such as hsa-miR-590-3p, hsamiR-944, miR-HSA-570, hsa-miR-34a, hsa-miR-124, hsa-miR-548f, hsa-miR-429, miR-HSA-200b, hsa-miR-195 and miR-497-HSA. Moreover, the miRNAs related with CNVs can explain the difference in levels of the important genes that are controlled by them. These CNVmiRNAs can also harm the overall biogenesis and processing of all miRNAs by targeting key

On the other hand, dysfunction of neuronal miRNAs can result in a number of neuropatho‐ logical conditions. It has been reported that neural miRNAs and their target mRNAs are coexpressed, suggesting their participation in feedback mechanisms to connect the transcriptional activation with the control of local dendritic protein synthesis [145]. Interest‐ ingly, the functions attributed to miRNAs overlap with growth abnormalities, delays and the disruption of neuronal maturation observed in the brains of autistic individuals. Aberrations in the translational control of multiple mRNAs mediated by targets of each miRNA may lead to the difference in phenotypes observed in ASD. Moreover, multiple miRNAs may target the same mRNA leading to phenotypes resulting from the converging of several loci in CNVs. Thus, a change in expression or level of miRNA will affect the expression of target genes and might have a pleiotropic effect that would produce a more severe autistic phenotype. However, one can not underestimate the clinical relevance of the deregulation of a single or a subset of CNV-miRNAs. Based on this, it is clear that the characterization of this relationship may illustrate the complexity of the underlying neuronal development, function and dysfunction

that will eventually help in the understanding and treatment of autism [150].

separately.

consistent deregulation between studies.

324 Autism Spectrum Disorder - Recent Advances

molecules in the miRNA pathway [150].

Chromatin is defined simply and collectively as genomic DNA associated to proteins within the nucleus. There is a vast assortment of chromatin factors dedicated to the DNA packaging and the enzymatic functions involved in changing chromatin states. Nucleosomes are the primary unit of chromatin organization with a histone core (H2A/B, H3, and H4) and linking subunit H1. They keep DNA condensed and regulated by only releasing genes into the open conformation when their accessibility is needed [81]. Interestingly, disruptions in chromatin regulator genes are frequently the cause of neurodevelopmental and neuropsychiatric disorders. Chromatin regulators are widely expressed in the brain, yet symptoms suggest that specific circuits are altered when they mutate [143]

Chromatin regulator genes are also altered by *de novo* mutations in a small proportion of ASD cases. They are involved in various cellular processes such as transcriptional regula‐ tion, cell cycle regulation, genomic stability and DNA damage repair, but it is still unclear how mutations in chromatin regulators lead to behavioral phenotypes. Following deletion models, it is suggested that mutations that affect chromatin regulators can lead to ASD because they are "global regulators" that are placed on top of a hierarchy of regulators. The regulators interact with many genes and pathways, and thus disruption can simultaneous‐ ly affect multiple target genes. Furthermore, it is proposed that these genes may also act as phenotypic capacitors, protecting processes of genetic development and environmental perturbations [138].

Methylation of the genome in certain areas appears to remodel chromatin and consequently the genes that are in this region. The regulation of each histone modification requires specific enzymes that add or remove the methyl or acetyl group. Several mutated genes associated to ASD encode histone demethylases, including *KDM5C*, a demethylase of histone *H3K4* implicated in gene repression and *JMJD1C*, a demethylase of histone H3K9 implicated in hormone-dependent transcriptional activation. Furthermore other genes involved in specific chromatin remodeling, such as *SMARCC1, SMARCC2*, *ARID1A*, *ARID1B*, *CHD8*, *CHD1*, *CHD3*, *CHD75*, and *ATRX*, have been described as mutated in rare cases [66, 77, 81, 105, 114, 111].
