Faculty Bibliography

Our previous studies demonstrated that prenatal bisphenol A (BPA) exposure affected the hepatic transcriptome and lipidome in rat offspring in a sex- and age-dependent manner. In this study, we investigated the effects of gestational exposure to BPA on the rat dams, after weaning period, and compared them with those of their offspring. Our results showed alterations in hepatic transcriptome related to insulin signaling, circadian rhythm, and infectious disease pathways in BPA-treated dams even 4 weeks after the exposure, whereas slight modifications on the lipid profile were found. Alterations in lipid and transcriptome profiles were more prominent in the prenatally BPA-exposed offspring at postnatal day (PND) 1 and 21 than those in the dams, suggesting that in utero exposure to BPA is more serious than exposure in the adulthood. Cryptochrome-1 (Cry1) and peroxisome proliferator-activated receptor delta (Ppard) were commonly altered in both dams and offspring. Nevertheless, the results of DIABLO (Data Integration Analysis for Biomarker discovery using Latent cOmponents), showed that multi-omics data successfully distinguished the exposed dams from the corresponding controls and their offspring with a high level of accuracy. The accuracy rates in BPA50 models (including control and 50 μg BPA/kg bw/day exposed groups) were smaller than those in BPA5000 models (control and 5000 μg BPA/kg bw/day exposed groups), suggesting dose-dependent severity in BPA effects. Palmitic acid and genes related to circadian rhythm, insulin responses, and lipid metabolism (e.g., 1-acylglycerol-3-phosphate O-acyltransferase 2 (Agpat2), B-cell CLL/lymphoma 10 (Bcl10), Cry1, Harvey rat sarcoma virus oncogene (Hras), and NLR family member X1 (Nlrx1)) were identified through DIABLO models as novel biomarkers of effects of BPA across two generations.

Perchlorate (ClO₄

OBJECTIVE:To determine if higher exposures measured in early childhood to environmental phenols, phthalates, pesticides, and/or trace elements, are associated with increased odds of having a diagnosis of Autism Spectrum Disorder (ASD), Developmental Delay (DD), or Other Early Concerns (OEC) compared to typically developing children (TD). METHODS:This study included 627 children between the ages of 2-5 who participated in the Childhood Autism Risks from Genetics and Environment (CHARGE) study. Urine samples were collected at the same study visit where diagnostic assessments to confirm diagnosis indicated during the recruitment process were performed. Adjusted multinomial regression models of each chemical with diagnosis as the outcome were conducted. Additionally, two methods were used to analyze mixtures: repeated holdout multinomial weighted quantile sum (WQS) regression for each chemical class; and a total urinary mixture effect was assessed with repeated holdout random subset WQS. RESULTS:Many urinary chemicals were associated with increased odds of ASD, DD or OEC compared to TD; however, most did not remain significant after false discovery rate adjustment. Repeated holdout WQS indices provided evidence for associations of both a phenol/paraben mixture effect and a trace element mixture effect on DD independently. In analyses adjusted for confounders and other exposures, results suggested an association of a pesticide mixture effect with increased risk for ASD. Results also suggested associations of a total urinary mixture with greater odds of both ASD and DD separately. CONCLUSION/CONCLUSIONS:Higher concentrations of urinary biomarkers were associated with ASD, DD, and OEC compared to TD, with consistency of the results comparing single chemical analyses and mixture analyses. Given that the biospecimens used for chemical analysis were generally collected many months after diagnoses were made, the direction of any causal association is unknown. Hence findings may reflect higher exposures among children with non-typical development than TD children due to differences in behaviors, metabolism, or toxicokinetics.

CONTEXT/BACKGROUND:Phthalates may disrupt maternal-fetal-placental endocrine pathways, affecting pregnancy outcomes and child development. Placental corticotropin releasing hormone (pCRH) is critical for healthy pregnancy and child development, but understudied as a target of endocrine disruption. OBJECTIVE:To examine phthalate metabolite concentrations (as mixtures and individually) in relation to pCRH. DESIGN/METHODS:Secondary data analysis from a prospective cohort study. SETTING/METHODS:Prenatal clinics in Tennessee, USA. PATIENTS/METHODS:1018 pregnant women (61.4% non-Hispanic Black, 32% non-Hispanic White, 6.6% other) participated in the CANDLE study and provided data. Inclusion criteria included: low-medical-risk singleton pregnancy, age 16-40, and gestational weeks 16-29. INTERVENTION/METHODS:None. MAIN OUTCOME MEASURES/METHODS:Plasma pCRH at two visits (mean gestational ages 23.0 and 31.8 weeks) and change in pCRH between visits (ΔpCRH). RESULTS:In weighted quantile sums (WQS) regression models, phthalate mixtures were associated with higher pCRH at Visit 1 (β = 0.07, 95 %CI: 0.02, 0.11) but lower pCRH at Visit 2 (β = -0.08, 95 %CI: -0.14, -0.02). In stratified analyses, among women with gestational diabetes (n = 59), phthalate mixtures were associated with lower pCRH at Visit 1 (β = -0.17, 95 %CI: -0.35, 0.0006) and Visit 2 (β = -0.35, 95 %CI: -0.50, -0.19), as well as greater ΔpCRH (β = 0.16, 95 %CI: 0.07, 0.25). Among women with gestational hypertension (n = 102), phthalate mixtures were associated with higher pCRH at Visit 1 (β = 0.20, 95 %CI: 0.03, 0.36) and Visit 2 (β = 0.42; 95 %CI: 0.19, 0.64) and lower ΔpCRH (β = -0.17, 95 %CI: -0.29, -0.06). Significant interactions between individual phthalate metabolites and pregnancy complications were observed. CONCLUSIONS:Phthalates may impact placental CRH secretion, with differing effects across pregnancy. Differences in results between women with and without gestational diabetes and gestational hypertension suggest a need for further research examining whether women with pregnancy complications may be more vulnerable to endocrine-disrupting effects of phthalates.

While definitions vary, endocrine-disrupting chemicals (EDCs) have two fundamental features: their disruption of hormone function and their contribution to disease and disability. The unique vulnerability of children to low-level EDC exposures has eroded the notion that only the dose makes the thing a poison, requiring a paradigm shift in scientific and policy practice. In this review, we discuss the unique vulnerability of children as early as fetal life and provide an overview of epidemiological studies on programming effects of EDCs on neuronal, metabolic, and immune pathways as well as on endocrine, reproductive, and renal systems. Building on this accumulating evidence, we dispel and address existing myths about the health effects of EDCs with examples from child health research. Finally, we provide a list of effective actions to reduce exposure, and subsequent harm that are applicable to individuals, communities, and policy-makers. Expected final online publication date for the Annual Review of Pharmacology and Toxicology, Volume 62 is January 2022. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Fetal exposure to phthalates and bisphenols could be associated with kidney function. We aim to assess the association between maternal urine concentrations of phthalates and bisphenols during pregnancy and kidney function and size during childhood. In 1366 pregnant women from a prospective population-based cohort, we measured urine concentrations of phthalates, more specifically phthalic acid and metabolites of low molecular weight phthalates (LMWP) and high molecular weight phthalates (HMWP), with its subgroups of di-2-ethylhexylphthalate (DEHP) and di-n-octylphthalate (DNOP) metabolites, and bisphenol A, S and F during first, second and third trimester. We explored three methods of adjustment for maternal hydration status: creatinine standardization, covariate adjustment for creatinine and covariate-adjusted creatinine standardization plus covariate adjustment. We measured kidney size, calculated estimated glomerular filtration rate (eGFR) and the albumin/creatinine ratio in urine and assessed microalbuminuria at 6 years old. When applying creatinine standardization, we found some associations of higher maternal second trimester urine phthalic acid and overall mean phthalic acid and LMWP concentrations with higher eGFR. These associations were lessened when applying other methods of creatinine adjustment. The associations found when we applied the covariate adjustment for creatinine method were also lessened when applying other methods of creatinine adjustment. Only the association of higher second trimester phthalic acid maternal urine concentrations with higher eGFR at 6 years old remained significant irrespective of the method of creatinine adjustment. There were no consistent associations of maternal bisphenol A, S and F urine concentrations with childhood kidney function. There were no associations of maternal phthalate or bisphenol urine concentrations with kidney volume in children at 6 years old. Concluding, no consistent associations of maternal phthalate or bisphenol urine concentrations with childhood kidney function or volume could be found. Furthermore, the method of adjusting maternal urine phthalate and bisphenol concentrations for urinary dilution had a substantial effect on the associations with childhood kidney function, as it changed the conclusions about the directionality of the associations. Future studies including maternal kidney function are needed to further elucidate these association in humans.

Polycyclic aromatic hydrocarbons (PAHs) are ubiquitous endocrine-disrupting combustion by-products that have been linked to preterm birth. One possible mechanism is through disruption of placental corticotropin releasing hormone (pCRH), a key hormone implicated in parturition. As an extension of recent research identifying pCRH as a potential target of endocrine disruption, we examined maternal PAH exposure in relation to pCRH in a large, diverse sample. Participants, drawn from the CANDLE cohort, part of the ECHO-PATHWAYS Consortium, completed study visits at 16-29 weeks (V1) and 22-39 weeks (V2) gestation (n=812). Seven urinary mono-hydroxylated PAH metabolites (OH-PAHs) were measured at V1 and serum pCRH at V1 and V2. Associations between individual log-transformed OH-PAHs (as well as two summed PAH measures) and log(pCRH) concentrations across visits were estimated using mixed effects models. Minimally-adjusted models included gestational age and urinary specific gravity, while fully-adjusted models also included sociodemographic characteristics. We additionally evaluated effect modification by pregnancy complications, fetal sex, and maternal childhood trauma history. We observed associations between 2-OH-Phenanthrene (2-OH-PHEN) and rate of pCRH change that persisted in fully adjusted models (β=0.0009, 0.00006, 0.0017), however, positive associations with other metabolites (most notably 3-OH-Phenanthrene and 1-Hydroxypyrene) were attenuated after adjustment for sociodemographic characteristics. Associations tended to be stronger at V1 compared to V2 and we observed no evidence of effect modification by pregnancy complications, fetal sex, or maternal childhood trauma history. In conclusion, we observed modest evidence of association between OH-PAHs, most notably 2-OH-PHEN, and pCRH in this sample. Additional research using serial measures of PAH exposure is warranted, as is investigation of alternative mechanisms that may link PAHs and timing of birth, such as inflammatory, epigenetic, or oxidative stress pathways.

BACKGROUND:Fish and other seafood are an important dietary source of per- and polyfluoroalkyl substances (PFAS) exposure in many areas of the world, and PFAS were found to be pervasive in fish from the Great Lakes area. Few studies, however, have examined the associations between Great Lakes Basin fish consumption and PFAS exposure. Many licensed anglers and Burmese refugees and immigrants residing in western New York State consume fish caught from the Great Lakes and surrounding waters, raising their risk of exposure to environmental contaminants including PFAS. The aims of this study were to: 1) present the PFAS exposure profile of the licensed anglers and Burmese refugees and 2) examine the associations between serum PFAS levels and local fish consumption. METHODS:Licensed anglers (n = 397) and Burmese participants (n = 199) provided blood samples and completed a detailed questionnaire in 2013. We measured 12 PFAS in serum. Multiple linear regression was used to assess associations between serum PFAS concentrations and self-reported consumption of fish from Great Lakes waters. RESULTS:Licensed anglers and Burmese participants reported consuming a median of 16 (IQR: 6-36) and 88 (IQR: 44-132) meals of locally caught fish in the year before sample collection, respectively (data for Burmese group restricted to 10 months of the year). Five PFAS were detected in almost all study participants (PFOS, PFOA, PFHxS, PFNA and PFDA; 97.5-100%). PFOS had the highest median serum concentration in licensed anglers (11.6 ng/mL) and the Burmese (35.6 ng/mL), approximately two and six times that of the U.S. general population, respectively. Serum levels of other PFAS in both groups were generally low and comparable to those in the general U.S. POPULATION/METHODS:Among licensed anglers, Great Lakes Basin fish meals over the past year were positively associated with serum PFOS (P < 0.0001), PFDA (P < 0.0001), PFHxS (P = 0.01), and PFNA (P = 0.02) and the number of years consuming locally caught fish was positively associated with serum PFOS (P = 0.01) and PFDA (P = 0.01) levels. In the Burmese group, consuming Great Lakes Basin fish more than three times a week in the past summer was positively associated with serum PFOS (P = 0.004) and PFDA (P = 0.02) among the Burmese of non-Karen ethnicity, but not among those of Karen ethnicity, suggesting potential ethnic differences in PFAS exposure. CONCLUSIONS:Great Lakes Basin fish consumption was associated with an increase in blood concentrations of some PFAS, and especially of PFOS, among licensed anglers and Burmese refugees and immigrants in western New York State. In the Burmese population, there may be other important PFAS exposure routes related to residential history and ethnicity. Continued outreach efforts to increase fish advisory awareness and reduce exposure to contaminants are needed among these populations.

BACKGROUND:Animal and epidemiological studies suggest that prenatal exposure to polycyclic aromatic hydrocarbons (PAHs) may negatively impact toddler neurodevelopment. METHODS:We investigated this association in 835 mother-child pairs from CANDLE, a diverse pregnancy cohort in the mid-South region of the U.S. PAH metabolite concentrations were measured in mid-pregnancy maternal urine. Cognitive and Language composite scores at ages 2 and 3 years were derived from the Bayley Scales of Infant and Toddler Development, 3rd edition (Bayley-3). Behavior Problem and Competence scores at age 2 were derived from the Brief Infant and Toddler Social Emotional Assessment (BITSEA). We used multivariate linear or Poisson regression to estimate associations with continuous scores and relative risks (RR) of neurodevelopment delay or behavior problems per 2-fold increase in PAH, adjusted for maternal health, nutrition, and socioeconomic status. Secondary analyses investigated associations with PAH mixture using Weighted Quantile Sum Regression (WQS) with a permutation test extension. RESULTS: = 0.05). All other estimates were consistent with null associations. CONCLUSION/CONCLUSIONS:In this large southern U.S. population we observed some support for adverse associations between PAHs and neurodevelopment.

Volatile organic compounds (VOCs) are ubiquitous environmental pollutants that are associated with birth defects, leukemia, neurocognitive deficits, reproductive impairment and cancer in humans exposed to these compounds. Exposure to VOCs can be assessed by measuring their metabolites in urine. Little is known, however, about the temporal variability in urinary VOC metabolite (VOCM) concentrations within and between individuals. In this study, we determined the variability in the concentrations of 38 VOCMs in urine samples collected from 19 healthy individuals across a period of 44 days. We also measured seven biomarkers of oxidative stress (lipid, protein and DNA damage) in urine to assess the relationship of VOC exposure to oxidative stress. Seventeen VOCMs had detection frequencies (DFs) of >60% in urine, and we limited further data analysis to those compounds. The creatinine-adjusted geometric mean concentrations of VOCMs ranged from 2.70 μg/g to 327 μg/g in spot and 2.60 μg/g to 551 μg/g in first morning void (FMV) urine samples. Calculation of the intra-class correlation coefficients (ICCs) for 17 VOCM concentrations to assess their predictability and repeatability in urinary measurements showed ranges of 0.080-0.425 in spot and 0.050-0.749 in FMV urine samples, revealing notable within-individual variability. Our results suggest that taking only single measurements of VOCM concentrations in urine in epidemiological investigations may lead to exposure misclassification. In addition, VOCM concentrations were significantly and positively correlated with oxidative stress biomarkers. This study thus provides important information for formulating sampling strategies in the biomonitoring of VOC exposure in human populations.