Estimation of placental and lactational transfer and tissue distribution of atrazine and its main metabolites in rodent dams, fetuses, and neonates with physiologically based pharmacokinetic modeling
Introduction
Atrazine [ATR; 2-chloro-4-(ethylamino)-6-(isopropylamino)-s-triazine, CAS# 1912-24-9] is a chlorotriazine herbicide used extensively on crops to control broadleaf weeds (EPA, 2003). Due to its widespread use, relative persistence in water (ATSDR, 2003) and extreme persistence in soil (Jablonowski et al., 2009), ATR is ubiquitous in the environment (Battaglin et al., 2009).
Possible exposure sources for ATR include contaminated air (dust), food, and drinking water (García et al., 2012, Lozier et al., 2012, Mosquin et al., 2012). Surface and drinking water ATR concentrations (up to 224 and 34 μg/L, respectively) in places with heavy ATR use, such as the Midwestern U.S., substantially exceed current maximum contaminant levels (MCL), i.e., 3 and 0.1 μg/L in U.S. and Europe, respectively (ATSDR, 2003, Mosquin et al., 2012). According to the EPA guidelines for acute exposure risk assessment of ATR, the lowest observed adverse effect level (LOAEL), no observed adverse effect level (NOAEL), reference dose (RfD) and population adjusted dose (PAD) are 70,000, 10,000, 100 and 10 μg/kg/day, respectively (EPA, 2003). For the general population in the U.S., the estimated acute and chronic dietary exposures to ATR are 0.234–0.857 and 0.046–0.286 μg/kg/day, respectively, which is relatively low (Gammon et al., 2005). On the other hand, the exposure levels could reach up 151,000 μg per work shift for ATR manufacturing workers, indicating a much higher occupational overexposure risk (Catenacci et al., 1993).
ATR and/or its metabolites have been frequently detected in spot urine samples from pesticide applicators (ATR equivalents [ATR and up to 8 identifiable metabolites, of which DE, DIP and DACT, abbreviations defined below, account for > 85%]: 100–510 μg/L; detected in every sample; Barr et al., 2007), their families (ATR mercapturate: 0.024–4.9 μg/L, 27% positive samples) and the general population (ATR mercapturate: ≤ 3.8 μg/L [creatinine normalized]; 14% positive samples, Curwin et al., 2007), including pregnant women and young children (Chevrier et al., 2011, Curwin et al., 2007). Quantifiable levels of ATR (> 0.05 μg/L) or one of its metabolites (ATR mercapturate: > 0.02 μg/L) in pregnant women's first-morning-void urine have been associated with adverse birth outcomes, such as fetal growth restriction (Chevrier et al., 2011).
ATR and its metabolites have also been detected in plasma, urine and multiple tissues (including the brain, liver and kidney) of ATR-treated rodents (Fraites et al., 2011, Ross et al., 2009), in fetuses and in the milk of orally exposed rat dams (Fraites et al., 2011), in human umbilical cord plasma samples from residentially exposed, low risk, urban population (Whyatt et al., 2003), and in breast milk samples collected from a general population in France (Balduini et al., 2003). Thus, in utero and lactational exposures may be important routes for ATR to reach the developing fetus or neonate.
In the body, ATR is metabolized by several hepatic P450s (e.g., CYP2B1, CYP2D1, and CYP2E1; Hanioka et al., 1998a) fairly rapidly to desethylatrazine (DE; 2-chloro-4-amino-6-isopropylamino-s-triazine, CAS# 6190-65-4) and desisopropylatrazine (DIP; 2-amino-4-chloro-6-ethylamino-s-triazine, CAS# 1007-28-9), which, in turn, are metabolized to didealkylatrazine (DACT; 2-chloro-4,6-diamino-1,3,5-triazine, CAS# 3397-62-4), the major in vivo metabolite of ATR in mice (Ross and Filipov, 2006, Ross et al., 2009), rats (Brzezicki et al., 2003), and, apparently, humans (Barr et al., 2007; Fig. 1). During gestational and/or lactational stages, ATR is also extensively metabolized following a similar pattern (Fraites et al., 2011). Emerging evidence suggests that the metabolism of ATR is auto-inducing and is physiological stage-independent, i.e., short-term ATR exposure increases its own metabolism and/or the expression of ATR-metabolizing P450 isoforms in peripubertal (Pogrmic-Majkic et al., 2012), adult (Hanioka et al., 1998b), and pregnant and/or lactating rats (Fraites et al., 2011). However, there is still much unknown about the pharmacokinetic behavior of ATR in dams, fetuses, and neonates. Of note, while some such pharmacokinetic data are available in the rat (Fraites et al., 2011), there is no gestational of lactational pharmacokinetic study in the mouse, which is important for species comparison and extrapolation.
Developmental exposure of laboratory animals to higher levels of ATR (35–200 mg/kg) results in various adverse effects ranging from suppression of postnatal development to full-litter resorption (Narotsky et al., 2001, Rayner et al., 2005, Rooney et al., 2003). Of note, perinatal exposure of rodents to environmentally-relevant low doses of ATR causes neurobehavioral deficits (≥ 1 μg/kg) and structural brain changes (100 μg/kg) in the offspring (Belloni et al., 2011, Giusi et al., 2006), suggesting that the developing brain might be particularly sensitive to ATR. In terms of the effects of ATR's main metabolite DACT on the developing nervous system, in vivo studies do not exist at this point. However, our recent in vitro study suggested that DACT is less potent than ATR. Nevertheless, high concentrations of DACT disrupt dopaminergic neuron morphological differentiation (Lin et al., 2013). The studies described above highlight the potential of adverse effects of ATR overexposure on the developing fetus and neonate, the brain in particular. However, these studies do not correlate adverse effects with estimations of fetal or neonatal target tissue concentrations of ATR or its metabolites.
Risk assessment of ATR in sensitive subpopulations, including fetuses and infants, is limited by the scarcity of human pharmacokinetic data. Physiologically based pharmacokinetic (PBPK) models in rodents are useful tools that can aid the process because they can perform route-to-route, species, and dose extrapolations, as well as dose–response analysis. Fetal and/or neonatal rodent PBPK models have been developed for several other xenobiotics (Corley et al., 2003, Lu et al., 2012); these models facilitate the risk assessment of developmental exposure to these chemicals. At present, PBPK models for ATR are available for adult male rats (McMullin et al., 2003, McMullin et al., 2007, Timchalk et al., 1990) and mice (Lin et al., 2011), but not for rodent dams, fetuses, or neonates. In order to improve our understanding of potential adverse effects due to developmental ATR exposure by providing fetal and neonatal tissue dosimetry for ATR and its metabolites, while taking advantage of very recent pharmacokinetic data from rat dams exposed to ATR during gestation and/or lactation (Fraites et al., 2011), we sat out to develop PBPK models for ATR describing its kinetic behavior in rodent dams, fetuses, and neonates.
Section snippets
Data source for model calibration
The data used to calibrate the gestational and lactational models for ATR are from two independent studies, performed by the same group (Fraites et al., 2011). In study 1, pregnant Sprague Dawley rats were treated with ATR (5 or 25 mg/kg) by daily oral gavage from gestational day (GD) 14 to GD20. Two hours after the last dosing on GD20, maternal plasma, tissues (the brain and 4th mammary gland), and fetuses were collected. Fetuses were analyzed on a per whole fetus basis. In study 2, Sprague
Gestational model
Model predictions of maternal plasma, maternal tissue and fetal concentrations of ATR and its metabolites at 2 h after the last dosing on GD20 were compared to measured data in pregnant rats dosed orally with 5 mg/kg ATR from GD14 to GD20 (Fraites et al., 2011). Results for ATR and its major metabolite DACT are shown in Fig. 3A, while results for the intermediate metabolites (DE and DIP) are provided in Appendix A (Fig. A2). Overall, the model slightly overestimated (within a factor of 2)
Discussion
The present PBPK models properly describe placental/lactational transfer and tissue distribution of ATR and its metabolites in pregnant, lactating, fetal and neonatal rats; rat-based models also extrapolate well to mice. Model predictions indicate that (1) the fetus is exposed to both ATR and its major metabolite DACT at levels similar to their maternal plasma levels, with DACT's levels being much higher than ATR, (2) the nursing neonate is exposed primarily to DACT at levels around one-third
Conflict of interest statement
The authors declare that there are no conflicts of interest.
Disclaimer
The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the U.S. Food and Drug Administration. Mention of trade names is not an endorsement of any commercial product.
Acknowledgments
The experimental data used for model calibration was based on Fraites et al. (2011). We thank all the authors because this work was essential for our publication. We also wish to acknowledge Drs. K. Barry Delclos, Xiaoxia Yang, and Frederick Beland, National Center for Toxicological Research, Food and Drug Administration, Jefferson, AR, for critically reviewing this manuscript. Our interactions with Dr. Sheppard A. Martin, Neurotoxicology Branch, Toxicity Assessment Division, National Health
References (129)
- et al.
Characterization of human cytochrome P450 induction by pesticides
Toxicology
(2012) - et al.
Amino acids responsible for decreased 2,3-biphosphoglycerate binding to fetal hemoglobin
Blood
(1997) - et al.
Regional hepatic CYP1A1 and CYP1A2 induction with 2,3,7,8-tetrachlorodibenzo-p-dioxin evaluated with a multicompartment geometric model of hepatic zonation
Toxicol. Appl. Pharmacol.
(1997) - et al.
Triazinic herbicide determination by gas chromatography-mass spectrometry in breast milk
J. Chromatogr. B Anal. Technol. Biomed. Life Sci.
(2003) - et al.
Early exposure to low doses of atrazine affects behavior in juvenile and adult CD1 mice
Toxicology
(2011) - et al.
Endosulfan induces CYP2B6 and CYP3A4 by activating the pregnane X receptor
Toxicol. Appl. Pharmacol.
(2010) - et al.
Biological monitoring of human exposure to atrazine
Toxicol. Lett.
(1993) - et al.
Development and application of a physiologically based pharmacokinetic model for triadimefon and its metabolite triadimenol in rats and humans
Toxicol. Lett.
(2011) - et al.
Effects of atrazine on cytochrome P450 enzymes of zebrafish (Danio rerio)
Chemosphere
(2009) - et al.
Proteomic analysis of diaminochlorotriazine (DACT) adducts in three brain regions of Wistar rats
Toxicol. Lett.
(2010)