Urinary Metabolomic Profile in Autism Spectrum Disorder: A Comparison between Patients, Siblings and Healthy Controls.

Background: Despite the progress in understanding the neurobiology of Autism Spectrum Disorder (ASD), the causes remain still unknown. A complex relationship between genetic, epigenetic and environmental factors contributes to ASD etiopathogenesis and it is responsible of the clinical heterogeneity. Metabolomics explores the molecular complexity of ASD and the relationships among phenotypes related to external agents. As an emerging tool of network medicine, metabolomics provides a direct functional read-out of the phenotype by the detection, identification, and quantification of metabolites in biological fluids in order to recognize metabolic alteration between comparative samples. Recent evidences show a different urinary metabolomic profile between ASD children and their unaffected siblings. More in details high level of mammalian-microbial cometabolites, alteration in nicotinic-acid metabolism and mitochondrial dysfunction and a dysregulation of antioxidant status and amino acid metabolism have been observed in ASD individuals.

Objectives: to identify urinary metabolic pathways involved in the etiopathogenesis of ASD in 3 different samples: ASD children, siblings and control group (typical development). To describe a specific urinary metabolomic profile related to the clinical phenotype (intelligence quotient, functioning, core symptoms, disruptive behaviours, medical comorbidity such as sleep and gastrointestinal problems).

Methods: we enrolled 105 children, (age range 18 months–11 years): 38 ASD, 35 unaffected siblings, 32 typical developing children without familiarity for ASD. Morning urine samples were collected for all participants. Urinary metabolites were analysed and quantified by Gas chromatography–mass spectrometry (GC-MS). Standardized neuropsychological assessment was performed for ASD children and their affected siblings.

Results: Our preliminary results on 45 children (26 ASD and 20 unaffected siblings) show two distinct urinary metabolomics profiles in the ASD population compared to their unaffected siblings (OPLS-DA R₂ 0.5, Q₂ 0.02, with a valid permutation test=0.049). The loading plot analysis reveals the different clusterization among the two groups (Figure 1). The main metabolic findings in ASD children include high concentrations of mammalian-microbial metabolites (lyxose, xylose, glucose, arabitol, sorbitol, threitol, fucose, p-cresol) and an alteration of tryptophan pathway (kynurenic acid, xanturenic acid, quinic acid, ribonic acid) (Figure 2). The metabolites most notably changed were compared with information obtained by Human Metabolome Database.

Conclusions: High level of mammalian-microbial metabolites, suggesting gastrointestinal (GI) dysbiosis, can be related to the GI symptoms described in autistic children. In particular, the high urinary level of p-cresol, can be due to proliferation of gut bacteria, constipation and increased oral exploration of objects. Low concentrations of kynurenic acid are associated with dysregulation of the tryptophan pathway, increased oxidative stress, excitatory/inhibitory imbalance.Our preliminary findings suggest a potential role of gastrointestinal dysbiosis, perturbation of antioxidant status, in the etiopathogenesis of ASD comorbidity such as gastrointestinal disorders, epilepsy, sleep disorders.