Ambient Air Pollution Exposure is Associated with the Infant Gut Microbiota
Maximilian Bailey1, Zachariah E. M. Morgan1, William B. Patterson1, Justin P. Shaffer2, Noopur Naik1, Roshonda B. Jones3, Paige K. Berger3, Kelsey Schmidt3, Hilary A. Minor4, Frederick Lurmann4, Michael I. Goran3, Jeremy Sarnat5, Donghai Liang5, Howard H. Chang5, Tanya L. Alderete1
1Department of Integrative Physiology, University of Colorado Boulder, Boulder, CO, USA; 2Department of Pediatrics, University of California San Diego, La Jolla, CA, USA; 3Department of Pediatrics, The Saban Research Institute, Children’s Hospital of Los Angeles, University of Southern California, Los Angeles, CA, USA; 4Sonoma Technology, Inc., Petaluma, CA, USA; 5Department of Environmental Health, Rollins School of Public Health, Emory University, Atlanta, GA
Background: Exposure to ambient air pollutants (AAP) is associated with the composition and function of the gut microbiome in adults; however, this relationship has not been studied in infancy, a critical period for the development of the gut microbiome. Alterations to the developing gut microbiome have the potential to alter important physiological systems, such as host metabolism as well as the immune, endocrine, and nervous systems. This study aimed to examine the relationships between AAP and the infant gut microbiota.
Methods: This study included 105 infants from the Southern California Mother’s Milk Study. Time weighted averages of exposure to particulate matter (PM2.5, PM10) and nitrogen dioxide (NO2), were estimated during the first 6 months of life. The infant gut microbiota was characterized using 16S rRNA sequencing at 6 months of age. First, to account for excess zeroes and overdispersion, common within infant gut microbial sequencing data, associations between AAP and the composition of the gut microbiota were examined using a Zero Inflated Negative Binomial Regression (ZINBR), where zero truncated incidence risk ratios (IRR) and FDR adjusted p-values are reported using the Benjamini-Hochberg method. Further, associations between AAP and metrics of microbial alpha- and beta-diversity measures were examined via multivariate linear regression. Next, compositionally aware, multinomial models were used to describe associations between AAP and the abundance of each gut microbe, and new methods (Songbird) for visualizing those associations were used to identify an important subset of associated taxa. Based on a directed acyclic graph, models adjusted for sex, breastfeeding frequency, socioeconomic status, birthweight and infant age as potential confounders.
Results: PM10, PM2.5 and NO2 were associated with eleven, eight and eleven different gut bacterial genera respectively. For example, PM10 was associated with the abundance of Dialister (IRR=1.22, p<0.01) and PM2.5 was associated with the abundance Actinomyces (IRR=1.43, p<0.01). Further, NO2 was associated with the abundance of Actinomyces (IRR=1.17, p<0.01). Multinomial analysis also revealed that PM10 exposure was associated with the composition of the gut microbiota based on the log-ratio of differentially ranked taxa sub-groups (R2=0.28, p<0.001). For example, all microbial genera positively associated with PM10 via ZINBR were also classified as being in the top 35% of differentially ranked taxa positively associated with PM10 exposure. Measures of AAP exposure were not observed to be associated with alpha- or beta-diversity measures.
Conclusion: Early postnatal exposure to AAP was found to be associated with the composition of the infant gut microbiota at 6-months of age. This analysis identified several microbial genera and microbial profiles associated with AAP exposure over the first 6 months of life. These results suggest that early life AAP exposure may impact the developing gut microbiome, which may have important implications for infant development. For example, genera associated with AAP have been found to be associated with adolescent obesity (e.g., Actinomyces). Our future studies will examine the longitudinal associations between AAP with infant growth trajectories as well as the gut microbiome and fecal metabolome in the first 2 years of life.