Functional Insights into Chromatin Remodelling from Studies on CHARGE Syndrome

Abstract

CHARGE syndrome is a rare genetic syndrome characterised by a unique combination of multiple organ anomalies. Dominant loss-of-function mutations in the gene encoding chromodomain helicase DNA binding protein 7 (CHD7), which is an ATP-dependent chromatin remodeller, have been identified as the cause of CHARGE syndrome. Here, we review recent work aimed at understanding the mechanism of CHD7 function in normal and pathological states, highlighting results from biochemical and in vivo studies. The emerging picture from this work suggests that the mechanisms by which CHD7 fine-tunes gene expression are context specific, consistent with the pleiotropic nature of CHARGE syndrome.

Keywords: CHARGE syndrome; CHD7; chromatin remodelling; congenital disease; epigenetic mechanisms.

Figure 1

Overview of Reported Chromodomain Helicase DNA Binding Protein 7 (CHD7) Mutations. Mutations reported in CHD7 aligned along a schematic representation of the CHD7 protein (E): splice site (A), missense (B), frame shift (C) and nonsense mutations (D), and partial gene deletions [red lines in (F)] and duplication [green line in (F)]. The mutations are spread throughout the CHD7 gene, but missense mutations occur only in the middle of the gene. Key: chromodomain; helicase N; DEXDc; Helicase C; SANT domain; BRK domain. Adapted from . An overview of mutations and polymorphic variants of the CHD7 gene can be found in the CHD7 locus-specific databaseⁱ.

Figure 2

Proposed Model of Reported Chromodomain Helicase DNA Binding Protein 7 (CHD7) Action at Enhancers. (A) CHD7 is recruited to poised cell type-specific enhancers through interactions with cell type-specific pioneer transcription factors (pTF) and histone modifications, such as H3 mono methyl K4 (H3K4me1) (yellow dot). pTFs can access their DNA-binding sites without a requirement for additional chromatin remodelling. Upon recruitment, CHD7 catalyses nucleosome translocation along the DNA, thereby revealing additional TF-binding sites (green rectangle). (B) Recruitment of additional TFs, perhaps associated with co-activators such as p300 (not shown), results in further modification (H3K27ac, green dot) and remodelling of chromatin, recruitment of RNA Polymerase II (Pol II) and transcription. Enhancer-associated RNA transcripts (eRNA), together with interactions with TFs, may facilitate association with subunits of the Mediator complex and chromatin looping to allow long-range enhancer–promoter interactions. Mediator association with core transcriptional machinery (such as TFIID) facilitates transcriptional initiation at the transcriptional start site (TSS). Additional interactions, for example the recruitment of CHD1 by Mediator and Pol II, enhance transcriptional elongation by remodelling chromatin in front of the elongating Pol II. (C) CHD7-mediated chromatin remodelling may also result in inhibition of enhancer activity and reduced gene expression. (i) Nucleosome repositioning might promote the association of TFs that are unable to recruit transcriptional co-activators, or TFs complexed with co-repressors (CoR) leading to changes in chromatin modification (e.g., H3K27me3, red dot) and structure. (ii) Alternatively, CHD7 might remodel chromatin to interfere with pTF recruitment, thereby effectively shutting off enhancers in specific contexts. The net effect will be the failure to initiate effective long-range enhance–promoter association and enhancement of gene expression. Future experiments will be required to test this model.

Figure 3

Key Figure: Developmental and Disease-associated Pathways Regulated by Chromodomain Helicase DNA Binding Protein 7 (CHD7). CHD7 can affect the activity of several signalling pathways that control development. Established connections are in black and hypothetical interactions or associations are in blue. CHD7 regulates bone morphogenetic protein (Bmp4) expression or interacts with SMAD1 to control BMP pathway genes implicated in heart development . CHD7 can either positively regulate orthodenticle homeobox 2 (Otx2) expression during ear development , or repress Otx2 expression during early cerebellar development . The latter results in reduced fibroblast growth factor 8 (Fgf8) expression and cerebellar vermis hypoplasia. CHD7 can antagonize retinoic acid signalling in neuronal progenitors, and retinoic acid pathway inhibition can rescue inner ear defects, implying hyperactive retinoic acid signalling as a cause of inner ear defects . CHD7 interacts with SRY (sex determining region Y)-box 2 (SOX2) in neural stem cells, and co-regulates the expression of the genes encoding Jagged 1 (Jag1), GLI family zinc finger 3 (Gli3), and v-myc avian myelocytomatosis viral oncogene neuroblastoma derived homolog (Mycn) , which might explain some overlapping phenotypes with Alagile, Feingold, and Pallister Hall syndromes, although this has not been proven. Noteworthy, autosomal dominant de novo mutations in SOX2 result in a phenotype that shares features with CHARGE syndrome such as eye and outer ear malformations, oesophageal atresia, hearing loss, and hypogonadotropic hypogonadism . CHD7 loss is associated with p53 pathway hyperactivation . The underlying mechanisms might include both direct effects of CHD7 on p53 gene expression and effects on rDNA transcription, leading to defects in ribosome biogenesis . CHD7 controls sema domain, immunoglobulin domain (Ig), short basic domain, secreted, (semaphorin) 3E (sema3e) expression, which may underlie defects in neural crest cell migration and olfactory development as well as being responsible for the clinical overlap with Kallmann syndrome . CHARGE and 22q11del syndromes show phenotypic overlap . The presumed shared developmental pathways have not been identified, but are likely to include shared CHD7 and T-box 1 (TBX1) target genes. CHARGE and Kabuki syndromes also show significant clinical overlap. The genes mutated in Kabuki syndrome, encoding lysine (K)-specific methyltransferase 2D (KMT2D) and lysine (K)-specific demethylase 6A (KDM6A), encode histone modification enzymes that together might control CHD7 recruitment to H3K4me1-marked enhancer regions. Therefore, CHD7, KMT2D, and KDM6A are expected to regulate the same target genes .