Protecting DNA Is a Family Affair: Telomere Length and Cognition in Affected Individuals, Unaffected Siblings, and Parents

Background: Although not a diagnostic criterion, individuals with ASD commonly experience cognitive difficulties. One possible mechanism is shortened telomeres. Telomeres are repetitive non-coding DNA nucleotides that protect genes by capping chromosome ends and progressively shorten with age. Recently, two reports associated shortened telomere length (TL) with ASD or familial relation. Further, shortened telomeres have been associated with age-related cognitive decline. While previous studies found no relationship between TL and ASD core symptoms, the relationship between TL and cognitive function or sensory symptoms in individuals with ASD and family members is unknown.

Objectives: We aimed to replicate the finding of shortened TL in children with ASD compared to neurotypical (NT) controls, and add new findings concerning TL in unaffected siblings. We investigated relationships between TL, cognition, and ASD-related behaviors in affected individuals, unaffected siblings, and parents.

Methods: Our participants (n=380) included 69 male NT controls (7.1±2.3 years), 108 individuals with ASD (11 female; 8.3±8.4 years), 136 unaffected siblings (66 female; 10.3±7.3 years), and 67 parents (43 female; 38.7±8.4 years). TL of DNA derived from blood leukocytes was determined using an established quantitative polymerase chain reaction method. Cognitive function was measured via Stanford-Binet Intelligence Scale-5, core symptoms via Autism Diagnostic Observation Schedule-2, sensory symptoms via Sensory Profile, and ASD-related behaviors in parents via Broader Autism Phenotype Questionnaire (BAPq).

Results: Among male NT, ASD, and unaffected siblings, there was a significant ANCOVA [F(2, 232)=6.12, p=0.003], with NT males having longer TL than males with ASD (p=0.001) and unaffected siblings (p=0.05), controlling for age (Fig. 1a). In a mixed-model, males with ASD were not different from male unaffected siblings [F(1, 160)=2.48, p=0.12; Fig. 1b], controlling for age. However, when including both sexes (with sex covariate), unaffected sibling’s TL were longer than individuals with ASD [F(1, 232)=5.39, p=0.02], driven by females [F(1, 70)=4.77, p=0.03]. In individuals with ASD, TL was not related to core symptoms, but negatively related to sensory symptoms [vestibular: r(86)=-0.27; p=0.001 (Fig. 2a); visual: r(84)=-0.25; p=0.02; touch: r(80)=-0.23; p=0.03]. TL was positively related to cognition in parents only [knowledge: r(44)=0.40; p=0.006 (Fig. 2b); working memory: r(44)=0.29; p=0.05]. Parent’s TL was also positively related to the BAPq aloof domain [r(41)=0.43 p=0.004].

Conclusions: We replicated shortened TL in individuals with ASD compared to NT controls. This is the first study to demonstrate unaffected siblings’ TL is also reduced, but to a lesser degree than their affected siblings, which may be driven by females with ASD. We replicated findings of no relationship between TL and core ASD symptoms, but add new findings of relationships with sensory symptoms. Further, we demonstrate TL is more tightly coupled with cognition in parents, which is concerning for cognitive aging outcomes in affected individuals with reduced TL at young ages. We observed a surprising correlation between longer TL and greater aloof traits in parents, which may reflect a protective mechanism to social stress. Further research is warranted to determine if TL is both a biological mechanism of symptoms in individuals with ASD and a potential treatment target.