Increased Mid-Gestational IFN-g, IL-4, and IL-5 in Mothers Giving Birth to a Child with Autism

Background: Some mothers of children with autism demonstrate immunological differences from mothers of typically developing children, which include autoimmune, allergic, and asthmatic phenomena.  Skewed maternal immune function during pregnancy can negatively impact fetal development.  Therefore, autism spectrum disorders may be related to prenatal exposures involving the maternal immune system.

Objectives:   To characterize mid-gestational serum cytokine profiles in mothers of children with autism compared to mothers of control children.      

Methods:   Blood samples were collected from pregnant women during 15-19 weeks of gestation.  The mothers included in this study gave birth to: 1) a child with autism (AU, n=84), 2) a child with a developmental delay but not autism (DD, n=49), or 3) a child with no known developmental disability (general population, GP; n=159).  Serum cytokine profiles were determined using Luminex technology.  Mean maternal cytokine levels were compared with a t-test.  Multivariable logistic regression was used to estimate odds ratios (OR_adj) and 95% confidence intervals (CI) for log-transformed cytokine values, adjusting for maternal age, race, ethnicity, birth country of mother, gestational age, and maternal weight at blood draw.

Results:   Mothers whose child was subsequently diagnosed with autism (M-AU) had significantly increased mid-gestational levels of serum IFN-γ (mean (pg/ml): 9.28 vs. 3.44, P=0.01; OR_adj= 1.52, 95% CI 1.19-1.93), IL-4 (mean: 10.31 vs. 5.16, P=0.01; OR_adj =1.51, 95% CI 1.12-2.03), and IL-5 (mean: 7.27 vs. 3.43, P=0.04; OR_adj=1.45, 95% CI 1.07-1.98) compared with mothers of children from the general population (M-GP).  Findings were similar for early onset and regressive autism, as well as autism with and without mental retardation.  Additionally, there was a significant positive correlation between IFN-γ, IL-4, and IL-5 levels, such that mothers with high levels of IFN-γ also had high levels of IL-4 and IL-5.  This correlation was most dramatic in the M-AU population (r = 0.72-95 in M-AU, r = 0.38-0.77 in M-GP). 

Conclusions:   We report elevated serum IFN- γ, IL-4, and IL-5 during pregnancy in women giving birth to a child with autism.  This cytokine profile is observed in individuals with allergy and asthma disorders.  Our findings in the peripheral blood therefore agree with previous epidemiological data, which indicate that mothers whose child was subsequently diagnosed with autism have a higher occurrence of mid-gestational allergy and asthma.  The maternal immune system is involved in many aspects of pregnancy and fetal development, and is uniquely regulated during gestation to maintain a favorable environment for the fetus.  Inappropriate maternal immune responses have been associated with various obstetrical and developmental complications.  Further investigation is needed to characterize the relationship between the observed maternal immunological phenotype and autism.