(663a) Environmental Chemical Diethylhexyl Phthalate Alters Intestinal Microbiota Community Structure and Metabolite Profile in Mice

Ming Lei^a,1, Rani
Menon^b,1, Sara Manteiga^a,†, Nicholas Alden^a,
Carrie Hunt^d, Robert C. Alaniz^d, Kyongbum Lee^a,*^,#, and Arul Jayaraman^b,c,d,*

^a. Department
of Chemical and Biological Engineering, Tufts University, Medford, MA 02155

^b. Artie
McFerrin Department of Chemical Engineering, Texas A&M University, College
Station, Texas 77843

^c.
Department of Biomedical Engineering, Texas A&M University, College
Station, TX 77843

^d. Department
of Microbial Pathogenesis and Immunology, College of Medicine, Texas Health
Science Center, Texas A&M University, College Station, TX 77843

^e.
Artie McFerrin Department of Chemical Engineering, Texas A&M University,
College Station, TX 77843a

Exposure to environmental
chemicals during windows of development is a potentially contributing factor in
gut microbiota dysbiosis, and linked to chronic diseases and developmental
disorders. We used a community-level model of microbiota metabolism to investigate
the effects of diethylhexyl phthalate (DEHP), a
ubiquitous plasticizer implicated in neurodevelopmental disorders, on the
composition and metabolite outputs of gut microbiota in young mice.
Administration of DEHP by oral gavage increased the abundance of Lachnoclostridum, while decreasing Akkermansia, Odoribacter, and Clostridium
sensu stricto (Figure 1).

Figure 1. Metagenomic (16S
rRNA) analysis of fecal microbiota from DEHP-exposed mice. (A) PCA on OTU
counts. The percentage represents the percent variance explained by each axis.
(B) Alpha diversity and (C) LefSE analysis of fecal
microbiota OTU counts.

Addition of DEHP to in vitro cultured cecal
microbiota increased the abundance of Alistipes, Paenibacillus, and Lachnoclostridium (Figure 2A). Untargeted metabolomics showed
that DEHP broadly altered the metabolite profile in the culture (Figure 2B).
Notably, DEHP enhanced the production of p-cresol,
while inhibiting butyrate synthesis (Figure 2C). Metabolic model-guided
correlation analysis indicated that the likely sources of p-cresol are Clostridium species (Figure 3). Our results
suggest that DEHP can directly modify the microbiota to affect production of
bacterial metabolites linked with neurodevelopmental disorders.

Figure 2.
Significant microbial and metabolite changes in in vitro cultured cecal luminal contents with DEHP. (A) LefSe
analysis of genus-level microbiota changes induced by DEHP. (B) Scatter plot of
first two PC scores from PCA of metabolite features detected in the cecal cultures (positive mode IDA data). (C) Dose-dependent
changes in p-cresol and butyric acid with DEHP on day 7. *: -value<0.05 when
compared to day 7 culture without DEHP addition (two-tailed t-test).

Figure 3. Model
of metabolic reactions in in vitro culture of cecal
luminal contents. (A) Fraction of genus-level OTU counts represented by the
metabolic model. (B) Hierarchical clustering of genera and metabolites in the
model. (C) Correlation network showing significant Pearson correlations between
genera (circles) and metabolites (squares). Fold-change from day 1 to 7 is
indicated by red (decrease) and green (increase) colors. Solid edges between
nodes indicate that the genus has at least one species capable of metabolizing
the connected metabolite (per database annotation of the genome), while dotted
lines indicate a purely empirical correlation.

Several previous studies have
pointed to environmental chemical exposure during windows of development as a
contributing factor in neurodevelopmental disorders, and correlated these
disorders with microbiota dysbiosis, little is known about how the chemicals
specifically alter the microbiota to interfere with development. The findings
reported here unambiguously establish that a pollutant linked with
neurodevelopmental disorders can directly modify the microbiota to promote the
production of a potentially toxic metabolite (p-cresol) that has also been correlated with neurodevelopmental
disorders. Further, we use a novel modeling strategy to identify the
responsible enzymes and bacterial sources of this metabolite. Our results suggest
that specific bacterial pathways could be developed as diagnostic and
therapeutic targets against health risks posed by ingestion of environmental
chemicals.