Maternal Cholesterol and Autism

Background: Several lines of research point to a role for cholesterol metabolism in the pathogenesis of autism spectrum disorders (ASD). First, children with Smith-Lemli-Opitz syndrome (SLOS), a genetic defect in cholesterol synthesis, have a high prevalence of ASD. In addition, a recent study found plasma cholesterol levels were below the 5^thpercentile in 19% of children with ASD, showing an increased risk for ASD in children with low plasma cholesterol levels but without SLOS. The association between cholesterol and autism is not surprising considering that cholesterol is important for many processes in the developing brain including patterning of the forebrain, neuronal growth and survival, as well as synapse formation. Several studies in animals and humans have shown that maternal cholesterol contributes to the fetal cholesterol pool and fetal steroidogenesis, at least early in embryonic development. Further, low maternal cholesterol levels are associated with adverse birth outcomes and maternal sterol gene variation predicts preterm delivery. Finally, preterm delivery is a risk factor for ASD, and maternal sterol genes predict cholesterol concentration in newborns. Thus perturbations in maternal cholesterol metabolism, especially cholesterol deficiency, may be a risk factor for ASD in the offspring.

Objectives: The objective of the study was to determine if maternal plasma cholesterol concentration is a predictor of ASD in the offspring with the hypothesis that cholesterol deficiency is a risk factor for ASD.

Methods: The study was conducted at 3 institutions: Oregon Health & Science University, Kennedy Krieger Institute and University of Colorado Medical Center. Study participants were recruited among mothers of children with ASD participating in the Autism Treatment Network, excluding individuals with treated/untreated dyslipidemia, diabetes mellitus or other major chronic illnesses. Blood was collected in the morning after an overnight fast and plasma cholesterol concentration was measured. Dietary intake was assessed using the Adult Block food frequency questionnaire.

Results: Eighty five (n=85) subjects were recruited (36.4±6.6 years of age, mean ± SD; ranging from 20 to 52). BMI was typical of U.S. women’s (27.1 ± 6.3 on average with 42% normal; 31% overweight, 27% obese). Dietary intake was also typical with 1,719±746 cal/day (15% protein, 38% fat, and 47% carbohydrates), relative nutritional deficiencies in calcium and iron (80% and 81% below DRI for calcium and iron respectively) as well as folic acid (98% were below DRI). Cholesterol intake was 218±107 mg/day with only 20% of the participants above 300 mg/day. The average plasma cholesterol concentration was 174.6±33.3 mg/dl. However 20% of the participants (mothers of children with ASD) were below the 5^thpercentile for cholesterol levels. No significant relationship was observed between maternal cholesterol concentration or dietary cholesterol intake and any of the offspring ’s ADOS scores (Communication, Social Interaction, Stereotypy/Aggression).

Conclusions: Mothers of children with ASD have low plasma cholesterol concentrations. These data suggest that maternal cholesterol deficiency is a risk factor for autism.