Cord Blood Metabolomics and Autism Spectrum Disorder

INTRODUCTION: Autism spectrum disorder (ASD) has increased dramatically in prevalence and incidence in recent years (1:59 in the US, CDC). With prenatal and/or genetic origins, early detection and intervention are critical to improve outcomes. Few studies have assessed perinatal biomarkers for later development of ASD. We hypothesize that cord blood plasma-based metabolomics will differentiate infants who would later be diagnosed with ASD and typically developing (TD) controls. METHODS: Banked cord blood plasma samples and clinical data from the Iowa Maternal Fetal Tissue Bank were used. Samples were analyzed via gas chromatography and mass spectrometry (GC-MS) and relative MS signal intensity of a maximum of was quantified ratiometrically. All metabolites were normalized to internal controls (succinate) to account for effects of extraction, derivatization, and/or loading. Ratiometric levels of each metabolite were compared by t-test between ASD and control groups. Results P<0.01 are reported given the pilot nature of this small study. Metabolite set enrichment analysis (MSEA) was performed against human libraries. Significantly altered metabolites were compared to a tertiary sample of cord blood from individuals later diagnosed with attention deficit hyperactivity disorder (ADHD) to determine specificity for ASD. RESULTS: 91 metabolites were detected and analyzed. Homocysteine (fold change [FC]: 1.411; p=0.017), Myristic acid (FC: 1.221; p=0.084), Pentadecanoic acid (FC: 1.256; p=0.093) were increased in ASD samples, while L-Isoleucine (FC: 0.858; p=0.047), L-Threonine (FC: 0.7131; p=0.068), O-Phosphoethanolamine (FC: 0.704; p=0.070), and 2-Hydroxybutyric acid (FC: 0.615; p=0.073) were decreased. Of these, only pentadecanoic was also significantly changed in ADHD samples (FC: 1.514; p=0.002). MSEA revealed enrichment of homocysteine degradation (p=0.017), catecholamine biosynthesis (p=0.048), threonine and 2-Oxobutanoate degradation (p=0.068), Phosphatidylcholine Biosynthesis (p=0.070), and Phosphatidylethanolamine Biosynthesis (p=0.070) in ASD cord blood samples. CONCLUSION: We found that cord blood plasma metabolomics reveals intriguing and specific differences between ASD and control samples. While some of differences are supported by evidence, these alterations have not previously been extended to the prenatal period. Future work will investigate concordance with maternal blood metabolomics to determine maternal-fetal pathways.

Duke Scholars

Author Brandon Schickling