(222e) An Exposome Connectivity Paradigm for the Mechanistic Assessment of the Effects of Prenatal and Early-Life Exposure to Metals on Neurodevelopment

The introduction of exposome studies to research started to make clear that the timing of exposure to environmental factors could be even more important than the dose. Given the complexity of the individual lifetime exposome assessment, a broad array of technologies must be employed. The connectivity approach to environmental health, builds upon the connectivity across different biological scales in a systems biology approach to elucidate the mechanisms underlying the environmental burden of disease. Adopting a data-driven paradigm that is guided by systems biology principles this approach couples comprehensiveness in exposome and health associations and biological plausibility. Thus, it can unravel causal links between environmental exposures, genetic predisposition and early biological markers of adverse health outcomes without needing to revert to very large population sizes. In the frame of HEALS (Health and Environment-wide associations via Large population Studies) we have been applying the exposome connectivity paradigm on several pre-existing cohorts re-analysing biobanked samples using a full multi-omics platform coupling epigenetics, SNPs, metabolomics and adductomics, and applying integrative bioinformatics and exposome-wide association algorithms to draw links between combined exposures to metals and endocrine disrupters and metabolic pathway dysregulation, as well as between metabolic pathway perturbations and clinically observed phenotypes of neurodevelopmental disorders such as problems in linguistic, motor development and cognitive capacity.

The current study, called PHIME, presents an integrated methodology about the investigation of the effects of prenatal and postnatal exposure to metals on child neurodevelopment. Heavy metals are well known neurotoxicants and exposure to heavy metals has been extensively associated with neurodevelopmental disorders. The links between in utero exposure to metals, metabolic pathway deregulation, and clinically observed phenotypes of neurodevelopmental disorders were drawn through a urinary and plasma untargeted metabolomics analysis using both nuclear magnetic resonance (NMR) spectroscopy and liquid chromatography-mass spectrometry (MS), followed by integrative bioinformatics and exposome-wide association algorithms. Heavy metals prenatal exposure was determined by measuring mercury in hair samples collected at birth, while cord blood and breast milk samples were analysed for mercury, cadmium, lead, and arsenic, as well as for essential elements (selenium, zinc, copper). Additional factors considered as exposure and effects modifiers included in the study were sociodemographic and anthropometric parameters, as well dietary factors. Cognitive function, language, fine and gross motor development were assessed in children at the age of 18 months by the Bayley Scale for Infant Development (Bayley-III) development tool. The Wechsler Intelligence Scale for Children - Fourth Edition was used for the assessment of the Full-Scale Intelligence Quotient (FSIQ). Finally, individual metabolome profiles have been derived analyzing biobanked urine and plasma samples taken at birth, and urine and plasma samples taken from children at the age of 8, using two different analytical techniques, LC-MS and NMR. These metabolic profiles were used for metabolic pathway analysis at the individual level and the identified perturbed pathways were associated with both exposure and outcome parameters using EWAS.

In order to generate a list of candidate biomarkers for the stressors of interest, which were identified by untargeted analysis of metabolic profile, a literature search was performed including combinations of the following terms: ‘’exposure’’, ‘’metals’’, ‘’neurodevelopment disorders’’, ‘’biomarkers’’ and ‘’underlying mechanisms’’. The query was perform for both levels of analysis (molecular biomarkers and pathway analysis) for all the experiments. It was designed not only to determine candidate biomarkers based on their known biological function, but also to avoid the exclusion of a biomarker, which is associated with the occurrence of a stressor, even though there is yet paucity of knowledge regarding its exact biological function.

The imbalance between the cellular reactive oxygen species (ROS), which may be an effect of exposure to metals, and the inability of the cell to detoxify them, leads to oxidative stress. According to literature, abnormalities in the citric acid cycle, urea cycle, and amino acid metabolism play a key role in the pathogenesis of oxidative stress. Formate, 2-oxoglutarate, isocitrate, glycerol, carnitine, glutathione, methionine, cysteine, pyruvate, N-acetylglutamic acid, β-alanine, serine, arginine, citrulline, tryptophan, alpha-D-glucose, (S)-lactate and acetate which have been detected in samples from PHIME cohort study, could play the role of candidate biomarkers for the neurodevelopmental disorders due to oxidative stress. In addition, the perturbations of the identified pathways, for the homeostatic operation of which the presence of the above biomarkers is crucial must be examined as a putative underlying mechanism. Some of these identified pathways are: S-methyl-5-thio-Alpha-D-ribose 1-phosphate degradation, folate metabolism, serotonin degradation, taurine biosynthesis, citrulline-nitric oxide cycle etc. Henceforth, dysfunctions in carnitine metabolism may affect calcium homeostasis, which is involved in the oxidation of phosphorylation, leading to neurodevelopmental disorders. Biochemical markers directly or indirectly related to mitochondrial dysfunction, found to participate in dysregulated metabolic pathways were the carnitine, alanine, lactate, pyruvate, lysine and the acylcarnitine.

According to EWAS analysis, the concentration of lead (Pb) found in mother’s plasma was negatively related to the metabolic pathways related to the effects of nitric oxide and the differentiation pathway. Cadmium (Cd) concentration levels showed a negative correlation with the degradation (III) of melatonin and dopamine, and the presence of mercury (Hg) with the metabolic pathway tyrosine degradation I. Selenium (Se) concentration (Se) found to be negatively correlated with the tetrapyrrole biosynthesis II and heme biosynthesis pathways, while copper (Cu) with caffeine and theobromine metabolism. Finally, manganese (Mn) appears to adversely affect the metabolic pathways of fatty acid biosynthesis, glycolysis/gluconeogenesis, amino sugar and nucleotide sugar metabolism, porphyrin and chlorophyll metabolism. With respect to children's samples, postnatal exposure to cadmium (Cd) is negatively related to the pathways of fatty acid biosynthesis, amino sugar and nucleotide sugar metabolism, porphyrin and chlorophyll metabolism and glycolysis/gluconeogenesis. Selenium (Se) levels was negatively associated with dopamine degradation.

Besides the negative association of the aforementioned pathways with neurodevelopment, EWAS analysis indicated that nutrition, as well as social factors, play an important role. In particular, the frequency of fish consumption during the pregnancy and the first 8 years of life, and the frequency of mushrooms consumption by children showed negative correlation with neurodevelopment. Other factors negatively correlated with neurodevelopment were the partner's age, maternal weight, and the source of drinking water.

From the cohort participants, one child has been diagnosed with Autism spectrum disorder (ASD). The exposure levels to As and Hg were above the 95^th percentile, and according to untargeted metabolomics analysis, the levels of L-arginine, and L-tryptophan were above the 95^th percentile, while the (S)-Lactate was below 5^th percentile. The aforementioned results suggest major disturbances to cells biochemistry, which resulted in the impairment of antioxidant defense mechanisms leading to the clinically observed results in linguistic, motor development and cognitive capacity.

Two are the main characteristics that stand out from this study. The fact that we have chosen to consider factors such as dietary habits or socio-economic status as additional parameters associated with child neurodevelopment. Moreover, we were particularly interested in capturing the overall metabolome perturbation and associate it with (i) the totality of environmental exposure factors we could quantify; and (ii) clinically observed neurodevelopmental disorders. The scope of the study was to investigate how the perinatal and early-life exposome affected child neurodevelopment. Exposome-based associations of clinical or sub-clinical health outcomes with the early-life external and internal exposome supports the elucidation of the mechanisms through which xenobiotics interact with and eventually perturb cell metabolism to induce specific pathways of toxicity in infants and young children. However, in vitro testing coupled to targeted metabolomics on metabolically active relevant cell lines could be a viable way forward towards providing the mechanistic evidence of the observed exposome-wide associations.