Modification of Associations Between Arsenic Exposure and Social Behavioral Development by the Infant Microbiome

• In Event: The Effects of Early Life Environmental Toxicants on Neurodevelopment From Early to Middle Childhood

Abstract

Introduction: Arsenic is a common drinking water contaminant that is associated with children’s health outcomes, including impaired neurodevelopment. Some studies have described an adverse association between arsenic exposure and autism-related social behaviors, but others found no association. The gut microbiome – the bacteria, fungi, and viruses within the gastrointestinal tract – is at the interface of environmental toxicants and the host system, and may alter host susceptibility to arsenic, especially in early life when host metabolism mechanisms are not fully developed.
Hypotheses: We hypothesized 1) that common features of the infant gut microbiome would alter the association between arsenic exposure and autism-related social behavior in a population-based cohort and 2) this may be partially due to metabolism of arsenic by the gut microbiome to arsenicals that have differential health effects.
Study Population: The New Hampshire Birth Cohort Study (NHBCS) is a longitudinal pregnancy cohort that has enrolled over 2500 participants, with ongoing recruitment. Participants are followed throughout childhood with questionnaires and in-person visits.
Methods: At six-weeks (6W) and one-year (1Y) postpartum, caregivers of NHBCS participants provided toenail clippings (6W only), and urine and stool samples. Arsenic was quantified in toenails and urine using inductively coupled plasma mass spectrometry. We measured total arsenic in toenails and in urine we quantified arsenic species [inorganic arsenic (iAs), monomethylarsonic acid (MMA), and dimethylarsinic acid (DMA)]. Fecal DNA was extracted and underwent metagenomic sequencing. Raw sequences were processed using the BioBakery pipeline and bacterial species and gene pathways were annotated using MetaCyc and KEGG Orthology (KO) databases. When participants were three years old, caregivers completed the Social Responsiveness Scale (SRS-2), which assesses children’s social behavior on a continuous scale. To address the first hypothesis, we selected the 10 most abundant bacterial species and MetaCyc pathways at 6W and 1Y. We modeled SRS-2 T-scores as a function of toenail arsenic, microbiome features dichotomized at their median, and an interaction between exposure and the microbiome in 78 participants, adjusting for confounders. To address the second hypothesis, we selected KOs involved in arsenic metabolism. We linearly regressed the percent of each arsenic species (iAs, MMA, DMA) against KO relative abundance (80-100% prevalence) or detection (10-80% prevalence), with separate models at 6W (n=186) and 1Y (n=190).
Results: The relative abundance of certain bacterial pathways at 6W modified the association between toenail arsenic and SRS-2 T-scores. Among participants with relative abundance of branched-chain amino acid (BCAA) biosynthesis pathways below the median, arsenic was associated with more autism-related social behaviors [β=1.67 (95% CI: 0.3, 3.04)], whereas there was no association among participants with BCAA biosynthesis pathway relative abundance above the median. When examining the association between arsenic metabolism KOs and arsenic speciation, we found several KOs were associated with different arsenic profiles. K16509, an arsenate reductase transcription regulator, was associated with lower %MMA and higher %DMA at both 6W and 1Y, suggesting a consistent contribution to arsenic metabolism. Together our findings suggest that the infant microbiome may play a role in shaping resilience or susceptibility to certain environmental exposures.

Authors

• Hannah Laue, Dartmouth College
  Presenting Author

• Yuka Moroishi, Dartmouth

• Juliette C Madan, Dartmouth

• Margaret R. Karagas, Dartmouth College Geisel School of Medicine,