Assessment of metabolic perturbations associated with exposure to phthalates among pregnant African American women

Abstract

Background: Phthalates have been linked with numerous harmful health effects. Limited data are available on the molecular mechanism underlying phthalate toxicity on human health. In this study, we measured urinary phthalate metabolites and used high-resolution metabolomics (HRM) to identify biological perturbations associated with phthalate exposures among pregnant African American (AA) women, who are disproportionately exposed to high phthalates levels.

Methods: We used untargeted HRM profiling to characterize serum samples collected during early (8-14 weeks gestation) and late (24-30 weeks gestation) pregnancy from 73 participants from the Atlanta AA Maternal-Child cohort. We measured eight urinary phthalate metabolites in early and late pregnancy, including Monoethyl phthalate (MEP), Mono(2-ethlyhexyl) phthalate (MEHP), and Mono (2-ethyl-5-hydroxyhexyl) phthalate (MEHHP), to assess maternal exposures to phthalates. Metabolite and metabolic pathway perturbation were evaluated using an untargeted HRM workflow.

Results: Geometric mean creatinine-adjusted levels of urinary MEP, MEHP, and MEHHP were 67.3, 1.4, and 4.1 μg/g creatinine, respectively, with MEP and MEHP higher than the mean levels of non-Hispanic blacks in the general US population (2015-2016). There were 73 and 1435 metabolic features significantly associated with at least one phthalate metabolite during early and late pregnancy (p < 0.005), respectively. Pathway enrichment analysis revealed perturbations in four inflammation- and oxidative-stress-related pathways associated with phthalate metabolite levels during both early and late pregnancy, including glycerophospholipid, urea cycle, arginine, and tyrosine metabolism. We confirmed 10 metabolites with level-1 evidence, which are associated with urinary phthalates, including thyroxine and thiamine, which were negatively associated with MEP, as well as tyramine and phenethylamine, which were positively associated with MEHP and MEHHP.

Conclusion: Our results demonstrated that urinary phthalate levels were associated with perturbations in biological pathways connected with inflammation, oxidative stress, and endocrine disruption. The findings support future targeted investigations on molecular mechanisms underlying the impact of maternal phthalates exposure on adverse health outcomes.

Keywords: High-resolution metabolomics; Metabolic perturbations; Phthalates; Urinary phthalate metabolites.

Fig. A.1

a. Manhattan plots of associations between log-transformed metabolic feature intensity and maternal phthalate metabolite level during early and late pregnancy in HILIC mode from MWAS models. X-axis denotes the retention time (in seconds), Y-axis denotes the negative log10 of the p-values calculated from the MWAS models. Significant features with p values less than 0.005 are colored. Blue triangles and red circles denote negative and positive associations, respectively. b. Manhattan plots of associations between log-transformed metabolic feature intensity and maternal phthalate metabolite level during early and late pregnancy in C18 mode from MWAS models. X-axis denotes the retention time (in seconds), Y-axis denotes the negative log10 of the p-values calculated from the MWAS models. Significant features with p values less than 0.005 are colored. Blue triangles and red circles denote negative and positive associations, respectively.

Fig. A.1

a. Manhattan plots of associations between log-transformed metabolic feature intensity and maternal phthalate metabolite level during early and late pregnancy in HILIC mode from MWAS models. X-axis denotes the retention time (in seconds), Y-axis denotes the negative log10 of the p-values calculated from the MWAS models. Significant features with p values less than 0.005 are colored. Blue triangles and red circles denote negative and positive associations, respectively. b. Manhattan plots of associations between log-transformed metabolic feature intensity and maternal phthalate metabolite level during early and late pregnancy in C18 mode from MWAS models. X-axis denotes the retention time (in seconds), Y-axis denotes the negative log10 of the p-values calculated from the MWAS models. Significant features with p values less than 0.005 are colored. Blue triangles and red circles denote negative and positive associations, respectively.

Fig. A.1

a. Manhattan plots of associations between log-transformed metabolic feature intensity and maternal phthalate metabolite level during early and late pregnancy in HILIC mode from MWAS models. X-axis denotes the retention time (in seconds), Y-axis denotes the negative log10 of the p-values calculated from the MWAS models. Significant features with p values less than 0.005 are colored. Blue triangles and red circles denote negative and positive associations, respectively. b. Manhattan plots of associations between log-transformed metabolic feature intensity and maternal phthalate metabolite level during early and late pregnancy in C18 mode from MWAS models. X-axis denotes the retention time (in seconds), Y-axis denotes the negative log10 of the p-values calculated from the MWAS models. Significant features with p values less than 0.005 are colored. Blue triangles and red circles denote negative and positive associations, respectively.

Fig. A.1

a. Manhattan plots of associations between log-transformed metabolic feature intensity and maternal phthalate metabolite level during early and late pregnancy in HILIC mode from MWAS models. X-axis denotes the retention time (in seconds), Y-axis denotes the negative log10 of the p-values calculated from the MWAS models. Significant features with p values less than 0.005 are colored. Blue triangles and red circles denote negative and positive associations, respectively. b. Manhattan plots of associations between log-transformed metabolic feature intensity and maternal phthalate metabolite level during early and late pregnancy in C18 mode from MWAS models. X-axis denotes the retention time (in seconds), Y-axis denotes the negative log10 of the p-values calculated from the MWAS models. Significant features with p values less than 0.005 are colored. Blue triangles and red circles denote negative and positive associations, respectively.

Fig. A.2.

The extracted ion chromatograph of identified chemicals. The metabolites were considered to be acceptable for chemical identification that had one or multiple pure peaks.

Fig 1.. Metabolic pathways associated with ≥4 phthalate metabolites models at the first clinical visit.

Cells were colored according to the strength (i.e. p-value) of the association between each of metabolic pathways and significant features (p-value< 0.005) that were associated with phthalate metabolites. Pathways are ranked according to the total number of the significant pathway-phthalate metabolites associations (p < 0.05) in the HILIC column and the C18 column. For HILIC chromatography column with positive ion mode, the metabolic features with following adducts were captured and considered in our pathway analysis and chemical annotation: M[+], M + H[+], M-H2O + H[+], M + Na[+], M + K[+], M + 2H[2+], and M(C13) + 2H[2+] ; For C18 chromatography column with negative ion mode, the metabolic features with following adducts were considered in our pathway analysis and chemical annotation: M-H[−], M + Cl[−], M + ACN-H[−], M + HCOO[−], M(C13)-H[−], M-H2O-H[−], and M + Na-2H[−]” ^Average number of metabolites within the specific metabolic pathway on significant models. #Number of metabolic features in the samples with m/z matched to the metabolites within the specific metabolic pathway.

Fig 2.. Metabolic pathways associated with ≥4 phthalate metabolites models at the second clinical visit.

Cells were colored according to the strength (i.e. p-value) of the association between each of metabolic pathways and significant features (p-value< 0.005) that were associated with phthalate metabolites. Pathways are ranked according to the total number of the significant pathway-phthalate metabolites associations (p < 0.05) in the HILIC column and the C18 column. ^ Average number of metabolites within the specific metabolic pathway on significant models. For HILIC chromatography column with positive ion mode, the metabolic features with following adducts were captured and considered in our pathway analysis and chemical annotation: M[+], M + H[+], M-H2O + H[+], M + Na[+], M + K[+], M + 2H[2+], and M(C13) + 2H[2+] ; For C18 chromatography column with negative ion mode, the metabolic features with following adducts were considered in our pathway analysis and chemical annotation: M-H[−], M + Cl[−], M + ACN-H[−], M + HCOO[−], M(C13)-H[−], M-H2O-H[−], and M + Na-2H[−]” #Number of metabolic features in the samples with m/z matched to the metabolites within the specific metabolic pathway.

Fig 3.. Venn diagram of metabolic pathways related to phthalate exposure in pregnant African American women during early and late pregnancy.

Fig 4.. The potential mechanism of phthalate leading to adverse health outcome.

Red arrow represents positive association and green arrow represents negative association. Acronym: T4, Thyroxine; GDM, Gestational diabetes mellitus; LDH, Lactate dehydrogenase; PDH, Pyruvate dehydrogenase; TPP, Thiamine pyrophosphate; α-KGDH, α-ketoglutarate dehydrogenase; //, Metabolic steps blocked in thiamine deficiency; UPS, ubiquitin- proteasome system; ?, Further studies are required to confirm the biological function of Bilirubin.

Fig 5.. Confirmed metabolites with Level I Confidence in the Tyrosine Metabolism pathway associated with maternal phthalate exposure:

positive association (red) and negative association (green). Acronym: TPO, Thyroid peroxidase; AADC, aromatic l-amino acid decarboxylase; TH, tyrosine hydroxylase; SULT, Sulfotransferase; COMT, Catechol-O-methyltransferase