Abstract
Neurodevelopmental disorders and sleep disturbances share genetic risk factors. DEAF1 genetic variants are associated with rare syndromes in which sleep disturbances are commonly reported, yet the specific sleep disorders in these patients, and the molecular mechanisms underlying this association, are unknown. We aimed to pinpoint specific biological processes that may be disrupted by pathogenic variants in this gene, comparing a list of DEAF1 regulatory target genes with a list of insomnia-associated genes, and using the intersect gene list as the input for pathway enrichment analysis. Thirty-nine DEAF1 regulatory targets were also identified as insomnia-associated genes, and the intersecting gene list was found to be strongly associated with immune processes, ubiquitin-mediated proteolysis pathways, and regulation of the cell cycle. This preliminary study highlights pathways that may be disrupted by DEAF1 pathogenic mutations and might be putative factors underlying the manifestation of insomnia in patients harboring such variants.
Citation:
Guerreiro P, Moysés-Oliveira M, Paschalidis M, et al. Sleep disturbances associated with DEAF1 pathogenic variants. J Clin Sleep Med. 2025;21(1):207–210.
Keywords: DEAF1, genetic variants, DAND, neurodevelopmental disorders, sleep, insomnia, circadian clock, circadian rhythm
INTRODUCTION
Sleep disorders are frequently reported in patients with neurodevelopmental disorders (NDDs), with prevalence rates reaching 80% in these individuals.1 Recently, we have identified pathways related to gene expression regulation as critical nodes of coalescence between the genetic architecture of sleep and rare NDDs.2 Prompted by this observation, we expanded our analysis to focus on chromatin regulators with known targets linked to rare NDDs that have a high prevalence of sleep disorders.2
A relevant example of an NDD-associated gene expression regulator is the Deformed Epidermal Autoregulatory Factor 1 (DEAF1) gene. Pathogenic variants on this transcription factor are associated with a phenotypic spectrum known as “DEAF1-associated NDDs” (DANDs).3 Sleep disturbances are frequent in DAND, with common reports of insomnia.3,4
To explore a potential link between pathogenic DEAF1 variants associated with DAND and sleep disturbances, we assessed whether DEAF1 target genes are related to insomnia. This study aims to identify specific biological processes related to sleep physiology that might be disrupted by DEAF1 mutations.
METHODS
Literature review of clinical data of patients with DAND
We reviewed the literature for clinical reports on patients with DAND, focusing on studies that described pathogenic DEAF1 variants and provided data on the presence or absence of sleep disturbances within their cohorts (Table 1). Additionally, we analyzed the prevalence of autism spectrum disorders and seizures in the reported cases, as they are highly prevalent DAND comorbidities and may also negatively impact sleep.1,3
Table 1.
Genotype of patients with DAND with sleep data described in Figure 1.
| Patient ID | Source | cDNA change | Protein change | Inheritance | Age |
|---|---|---|---|---|---|
| Individual 1 | (1) | c.683T>G | p.(Ile228Ser) | De novo | 9 |
| Individual 2 | (1) | c.791A>C | p.(Gln264Pro) | De novo | 7 |
| Individual 3 | (1) | c.670C>T | p.(Arg224Trp) | De novo | 10 |
| Individual 4 | (1) | c.762A>C | p.(Arg254Ser) | De novo | 10 |
| M2647 | (2) | c.700T>A | p.(Trp234Arg) | De novo | 4 |
| AD/1 | (3) | c.634G>A | p.(Gly212Ser) | De novo | 20 |
| AD/2 | (3) | c.634G>A | p.(Gly212Ser) | De novo | 7 |
| AD/3 | (3) | c.637A>C | p.(Thr213Pro) | De novo | 7 |
| AD/4 | (3) | c.640C>G | p.(Leu214Val) | De novo | 16 |
| AD/5 | (3) | c.641T>C | p.(Leu214Pro) | De novo | 3 |
| AD/6 | (3) | c.646A>G | p.(Lys216Glu) | De novo | 10 |
| AD/7 | (3) | c.648G>T | p.(Lys216Asn) | De novo | 9 |
| AD/8 | (3) | c.664 + 1G>T | p.(Pro174_Gly222del) | De novo | 5 |
| AD/9 | (3) | c.674G>A | p.(Gly225Glu) | De novo | 5 |
| AD/10 | (3) | c.683T>C | p.(Ile228Thr) | De novo | 29 |
| AD/11 | (3) | c.706A>G | p.(Ser236Gly) | De novo | 4 |
| AD/12 | (3) | c.757A>G | p.(Lys253Glu) | De novo | 38 |
| AD/13 | (3) | c.762_764delAAG | p.(Arg254del) | De novo | 25 |
| AD/14 | (3) | c.791A>C | p.(Gln264Pro) | De novo | 19 |
| AD/15 | (3) | c.815T>C | p.(Leu272Ser) | De novo | 6 |
| AD/16 | (3) | c.826G>C | p.(Ala276Pro) | De novo | 6 |
| AD/17 | (3) | c.826G>C | p.(Ala276Pro) | De novo | 14 |
| AR/1 | (3) | c.1104_1105dup/c.1617dup | p.(Asp369Alafs*51)/p.(Cys540Metfs*18) | Biallelic | 17 |
| AR/2 | (3) | c.130delA/c.1580_1582delTCT | p.(Arg44Glyfs*25)/p.(Phe527del) | Biallelic | 8 |
| AR/3 | (3) | c.701G>A/c.716A>G | p.(Trp234*)/p.(Glu239Gly) | Biallelic | 16 |
| AR/4 | (3) | c.671G>A/c.671G>A | p.(Arg224Gln) | Biallelic | 9 |
| AR/5 | (3) | c.671G>A/c.671G>A | p.(Arg224Gln) | Biallelic | 2 |
| Patient 1 | (4) | c.702G>T | p.(Trp234Cys) | De novo | 11 |
| Patient 2 | (4) | c.670C>T | p.(Arg224Trp) | De novo | 5 |
| Patient 3 | (4) | c.608T>C | p.(Leu203Pro) | De novo | 8 |
| Patient 4 | (4) | c.670C>T | p.(Arg224Trp) | De novo | 2 |
| Patient 5 | (4) | c.762A>C | p.(Arg254Ser) | De novo | 4 |
| Patient 6 | (4) | c.825C>T | p.(His275Gln) | De novo | 1 |
| Case 1 | (5) | c.662C>T | p.(Ser221Leu) | De novo | 8 |
| Case 2 | (5) | c.825C>G | p.(His275Gln) | De novo | 33 |
| Case 3 | (5) | c.712A>C | p.(Thr238Pro) | De novo | 8 |
| Case 4 | (5) | c.748A>G | p.(Lys250Glu) | De novo | 6 |
| Case 5 | (5) | c.754T>C | p.(Trp252Arg) | De novo | 14 |
| Case 6 | (5) | c.767T>G | p.(Ile256Ser) | De novo | 11 |
| Case 7 | (5) | c.880G>A | p.(Val294Leu) | De novo | 11 |
| Case 8 | (5) | c.890T>C | p.(Phe297Ser) | De novo | 10 |
| Case 9 | (5) | c.332A>C/c.563_1045del | p.(Asp111Ala)/p.(Ser130_Leu290del) | De novo | 21 |
| Case 10 | (5) | c.836G>C | p.(Cys279Ser) | De novo | 6 |
Sources: (1) Vulto-Van Silfhout et al, Am J Hum Genet. 2014;1;94(5):649–661; (2) Berger et al, Hum Genet. 2017;1;136(4):409–420; (3) Nabais Sá et al, Genet Med. 2019;21:2059–2069; (4) Chen et al, Clin Chim Acta. 2021;1;518:17–21; (5) McGee et al, Hum Mol Genet. 2023;13;32(3):386–401. cDNA = complementary deoxyribonucleic acid, DAND = DEAF1-associated neurodevelopmental disorder.
Identification of insomnia-associated DEAF1 target genes
We generated a list of DEAF1 target genes by mapping its binding sites, retrieved from “ENCODE” chromatin immuno-precipitation-sequencing datasets,5 and converting their genomic coordinates to gene promoters using “BioMart,”6 a data mining web tool accessed from the “Ensembl” genome database. We then sourced a list of insomnia-associated genes from a recent large-scale genome-wide association study for this sleep trait.7 Both lists were then compared to identify intersecting genes.
Pathway enrichment analysis
The intersect gene list was then used as the input for pathway enrichment analysis using the “Enrichr” web tool,”8 considering “Gene Ontology” and “Reactome Pathway Database” terms for pathway representations. The Enrichr tool performs the Benjamini-Hochberg test adjusting for multiple comparisons with a significance threshold of adjusted P value < .05 and calculates the odds ratio (OR) for pathways’ overrepresentation.
RESULTS
Sleep and neurodevelopmental phenotypes associated with DAND
With the primary goal of investigating whether there exists a possible genotype-phenotype correlation between mutational mechanisms and/or affected proteins domains and sleep disorders, we first reviewed the available literature reporting de novo (Figure 1A) and biallelic variants (Figure 1B) in patients with DAND. In our literature review, we found that circa 81% of previously reported cases mention sleep disturbances, irrespective of the variants’ coordinates or mutational mechanisms involved. Circa 66% of biallelic and 78% of de novo variants affect the SAND domain of the DEAF1 protein, which is responsible for deoxyribonucleic acid binding and protein-protein interactions.3 Detailed clinical data, including descriptions of the patients’ sleep problems and comorbidities and diagnostic criteria were unavailable in most studies, limiting the establishment of accurate correlations between genotype and phenotype, as well as between comorbidities and sleep problems.
Figure 1. Genotypic and phenotypic information from patients with DAND and pathway enrichment analysis for insomnia-associated genes regulated by DEAF1.
(A), (B) Genetic and phenotypic data from patients with DAND with de novo (A) and biallelic variants (B). (C) Venn diagram of the intersection between DEAF1 regulatory targets and insomnia-associated genes. (D) Top 5 significantly enriched pathways among insomnia-associated genes regulated by DEAF1. + = present, − = absent, A.A.s = aminoacids, ASD = autism spectrum disorders, DNA = deoxyribonucleic acid, MHC = major histological complex, DAND = DEAF1-associated neurodevelopmental disorder, DEAF1 = Deformed Epidermal Autoregulatory Factor 1, M = male, F = female, U = unknown/unavailable.
Sleep and circadian-associated pathways disrupted by DEAF1 mutations
There were 39 overlapping genes between DEAF1 direct regulatory targets (1,179 genes) and insomnia-associated genes (569 genes) (Figure 1C). These 39 genes were found to be strongly associated with multiple pathways related to the regulation of transcription and the cell cycle, protein degradation and immune processes. The most statistically relevant pathways in each of these categories were: positive regulation of mitotic cell cycle (GO:0045931, adjusted P value = 6.75E-7, OR = 99.07), ubiquitination and proteasome degradation (R-HSA-983168, adjusted P value = 6.79E-3, OR = 11.88), class I Major Histological Complex-mediated antigen processing and presentation (R-HSA-983169, adjusted P value = 1.08E-02, OR = 9.57), and deoxyribonucleic acid-binding transcription factor binding (GO:0140297, adjusted P value = 1.55E-02, OR = 10.45) (Figure 1D).
DISCUSSION
Despite the limited sleep-related data in the reviewed literature, our analysis showed that the prevalence of sleep disturbances of patients with DAND is in line with the rates reported in previous studies regarding other NDD-associated phenotypes.1 Further clinical studies, focused on objective evaluations over patients’ sleep phenotypes, are required to establish an accurate genotype-phenotype correlation between variants and sleep, since most patients’ sleep problems were either not detailed enough or were not investigated at all. A better understanding of sleep in DAND can improve patient care and quality of life.
The data from our pathway enrichment analysis suggest that the loss of function of DEAF1 may disrupt important physiological processes linked to sleep and chronobiology, representing a putative etiology of sleep disturbances. For instance, both the cell cycle and immune function are known to be associated with the circadian clock.9 Ubiquitination is essential for the breakdown of proteins in both the intra and extracellular environments, including circadian and synaptic proteins, which respectively regulate the internal clock and have implications on neurodevelopmental rare syndromes.10 In the context of a DEAF1 pathogenic variant, the gene expression regulation exerted by the encoded transcription factor might be impaired, impacting these pathways and affecting sleep patterns in patients with DAND.
Further dissecting the molecular interactions between DEAF1 and its insomnia-related target genes, as well as their role in triggering and maintaining sleep, might provide therapeutic approaches tailored to patients with DAND. The data leveraged by this preliminary study paves the way for patient-specific care and future studies focused on novel therapies that aim to modulate the affected biological processes.
DISCLOSURE STATEMENT
Each author has read and approved this manuscript. Institution where work was performed: Sleep Institute, Associação Fundo de Incentivo à Pesquisa, São Paulo, Brazil. Funding: P.G.: Associação Fundo de Incentivo à Pesquisa; M.M.O.: Associação Fundo de Incentivo à Pesquisa, Fundação de Amparo à Pesquisa do Estado de São Paulo; M.P.: Fundação de Amparo à Pesquisa do Estado de São Paulo; A.K.: Associação Fundo de Incentivo à Pesquisa; L.C.: Fundação de Amparo à Pesquisa do Estado de São Paulo; T.B.D.: none; A.C.M.: Fundação de Amparo à Pesquisa do Estado de São Paulo; B.P.M.: Associação Fundo de Incentivo à Pesquisa; L.N.G.A.: Fundação de Amparo à Pesquisa do Estado de São Paulo; M.L.A.: Conselho Nacional de Desenvolvimento Científico e Tecnológico, Fundação de Amparo à Pesquisa do Estado de São Paulo; S.T.: none. The authors report no conflicts of interest.
ABBREVIATIONS
- DAND
DEAF1-associated neurodevelopmental disorder
- DEAF1
Deformed Epidermal Autoregulatory Factor 1
- NDD
neurodevelopmental disorder
- OR
odds ratio
REFERENCES
- 1. Angriman M, Caravale B, Novelli L, Ferri R, Bruni O . Sleep in children with neurodevelopmental disabilities . Neuropediatrics. 2015. ; 46 ( 3 ): 199 – 210 . [DOI] [PubMed] [Google Scholar]
- 2. Moysés-Oliveira M, Paschalidis M, Souza-Cunha LA, et al . Genetic basis of sleep phenotypes and rare neurodevelopmental syndromes reveal shared molecular pathways . J Neurosci Res. 2023. ; 101 ( 7 ): 1058 – 1067 . [DOI] [PubMed] [Google Scholar]
- 3. Nabais Sá MJ, Jensik PJ, Mcgee SR, et al . De novo and biallelic DEAF1 variants cause a phenotypic spectrum . Genet Med. 2019. ; 21 ( 9 ): 2059 – 2069 . [DOI] [PubMed] [Google Scholar]
- 4. Vulto-Van Silfhout AT, Rajamanickam S, Jensik PJ, et al . Mutations affecting the SAND domain of DEAF1 cause intellectual disability with severe speech impairment and behavioral problems . Am J Hum Genet. 2014. ; 94 ( 5 ): 649 – 661 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Landt SG, Marinov GK, Kundaje A, et al . ChIP-seq guidelines and practices of the ENCODE and modENCODE consortia . Genome Res. 2012. ; 22 ( 9 ): 1813 – 1831 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Smedley D, Haider S, Ballester B, et al . BioMart - biological queries made easy . BMC Genomics. 2009. ; 10 ( 1 ): 22 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Watanabe K, Jansen PR, Savage JE, et al. ; 23andMe Research Team . Genome-wide meta-analysis of insomnia prioritizes genes associated with metabolic and psychiatric pathways . Nat Genet. 2022. ; 54 ( 8 ): 1125 – 1132 . [DOI] [PubMed] [Google Scholar]
- 8. Xie Z, Bailey A, Kuleshov MV, et al . Gene set knowledge discovery with enrichr . Curr Protoc. 2021. ; 1 ( 3 ): e90 . [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Farshadi E, van der Horst GTJ, Chaves I . Molecular links between the circadian clock and the cell cycle . J Mol Biol. 2020. ; 432 ( 12 ): 3515 – 3524 . [DOI] [PubMed] [Google Scholar]
- 10. Moyses-Oliveira M, Yadav R, Erdin S, Talkowski ME . New gene discoveries highlight functional convergence in autism and related neurodevelopmental disorders . Curr Opin Genet Dev. 2020;65:195–206. [DOI] [PubMed]