Genetic Variation in Melatonin Pathway Enzymes in Children with Autism Spectrum Disorder and Comorbid Sleep Onset Delay

Thursday, May 14, 2015: 5:30 PM-7:00 PM
Imperial Ballroom (Grand America Hotel)
O. J. Veatch1, J. S. Pendergast1, R. M. Leu2, C. H. Johnson1, S. H. Elsea3 and B. A. Malow4, (1)Vanderbilt University, Nashville, TN, (2)Pediatrics, Emory University, Atlanta, GA, (3)Baylor College of Medicine, Houston, TX, (4)Vanderbilt Kennedy Center, Vanderbilt University Medical Center, Nashville, TN

A large proportion of individuals with autism spectrum disorder (ASD) present with comorbid sleep disorders. Disrupted sleep not only affects cognitive functioning, but exacerbates ASD-related behavioral impairments. Previous studies established a strong influence of common genetic variation in the etiology of ASD. Genetic factors likely contribute also to comorbid expression of sleep disorders in ASD. Variations in genes whose products regulate endogenous melatonin modify sleep patterns in humans, and have also been implicated in ASD. However, there are major challenges identifying even modest, replicable effects related to this variation. It is possible that variable phenotypic expressivity in ASD, specifically expression of comorbidities, can be explained by underlying genetic differences. Stratifying ASD cases based on presentation of sleep disruption may help address issues related to phenotypic heterogeneity in genetic analyses and help reveal underlying relationships between genotype and phenotype.


Our hypothesis was that individuals with ASD and comorbid insomnia would harbor a greater load of variation in genes related to maintenance 


We studied a clinically unique ASD subgroup, consisting solely of children with comorbid expression of sleep onset delay. We evaluated variation in two melatonin pathway genes, acetylserotonin O-methyltransferase (ASMT) and cytochrome P450 1A2 (CYP1A2). We sequenced the protein coding region of ASMT, and genotyped seven predicted ‘dysfunctional’ single nucleotide polymorphisms (SNPs) in CYP1A2.


All patients were either homozygous, or heterozygous, for variants evidenced to alter ASMT transcript production. Compared to current estimates in ASD without reported evidence of sleep disturbances we observed higher frequencies (p≤0.04) for variants evidenced to decrease ASMT expression. We also observed substantially higher frequencies than currently reported in populations of European ancestry (p≤0.0007) for variant alleles in three SNPs that have been associated with decreased CYP1A2 enzymatic activity. Finally, we detected a relationship between genotypes in ASMT that are associated with lower gene expression, and genotypes in CYP1A2 that are associated with decreased enzymatic activity (r2=0.63).


Our results indicate that expression of sleep onset delay in individuals with ASD relates to melatonin pathway genes. Our findings suggest a mechanism connecting lower levels of ASMT transcript production with reduced CYP1A2 metabolic activity in children with ASD and comorbid sleep onset delay. We are currently in the process of attempting to replicate these genetic findings in a larger, clinically similar dataset.

See more of: Genetics
See more of: Genetics