ToxSci Advance Access originally published online on March 15, 2008
Toxicological Sciences 2008 103(2):241-259; doi:10.1093/toxsci/kfn054
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Tissue Exposures to Free and Glucuronidated Monobutylyphthalate in the Pregnant and Fetal Rat following Exposure to Di-n-butylphthalate: Evaluation with a PBPK Model
,1,2,3
* The University of North Carolina, Chappell Hill, North Carolina 27599
The Hamner Institutes for Health Sciences, Research Triangle Park, North Carolina 27709
1 To whom correspondence should be addressed at The Hamner Institutes for Health Sciences, 6 Davis Drive, Research Triangle Park, NC 27709. Fax: (919) 558-1300. E-mail: rclewell{at}thehamner.org.
Received November 21, 2007; accepted March 9, 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Human exposure to phthalic acid diesters occurs through a variety of pathways as a result of their widespread use in plastics. Repeated doses of di-n-butylphthalate (DBP) from gestation day (GD) 12 to 19 disrupt testosterone synthesis and male sexual development in the fetal rat. To gain a better understanding of the relationship of the target tissue (testes) dose to observed developmental effects, the pharmacokinetics of monobutyl phthalate (MBP) and its glucuronide (MBP-G) were examined in pregnant and fetal rats following single and repeated administration of DBP from GD 12–19. These data, together with results from previously published studies, were used to develop a physiologically based pharmacokinetic model for DBP and its metabolites in the male, pregnant and fetal rat. The model structure accounts for the major metabolic (hydrolysis, glucuronidation, oxidative metabolism) and transport processes (enterohepatic recirculation, urinary and fecal excretion, placental transfer). Extrapolation of the validated adult male rat model to gestation successfully predicts MBP and MBP-G levels in maternal plasma, placenta and urine, as well as the fetal plasma and testes. Sensitivity analysis indicates that plasma MBP kinetics are particularly sensitive to glucuronidation and enterohepatic recirculation: a decrease in the uridine 5'-diphospho-glucuronosyltransferase (UDPGT) capacity during gestation results in an increased MBP residence time, and saturation of UDPGT at the highest doses (> 100 mg/kg/day) causes a flattening out of the plasma time course data. Oxidative metabolism plays a significant role in elimination only at low doses (< 50 mg/kg DBP). Insights gained from modeling of the rat data will be used to support development of a human PBPK model for DBP.
Key Words: PBPK model; phthalate; DBP; glucuronide; gestation; rat; kinetics.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Phthalate esters are found in drinking water, food, personal care products and are used in the coating of some medications (ATSDR, 2001
; Heudorf et al., 2007; Wormuth et al., 2006). Although the relative source contributions have not been well characterized in the United States, a recent study in Europeans found that oral exposure through food sources accounted for 40–90% of the overall di-n-butylphthalate (DBP) exposure (Wormuth et al., 2006). Metabolites of several phthalates have been found in the urine of the U.S. population, including adults, children, infants, and pregnant women (Brock et al., 2002; Koch et al., 2005; Kohn et al., 2000; Latini et al., 2003; Silva et al., 2004), with the highest levels in women of child-bearing age. Some of these phthalates, including di-n-butylphthalate (DBP), are developmental toxicants in rats and mice (Ema et al., 1998, 2000; Fisher et al., 2003; Gray et al., 1999, 2000; Mylchreest et al., 1998, 1999). Exposure of pregnant rats to DBP impairs development of male reproductive tissues, as evidenced by reduced anogenital distance, nipple retention, hypospadias, delayed testes descent, and vaginal pouch development in male pups (Mylchreest et al., 2000, 2002). Although the mechanism of action of DBP has not been fully elucidated, these antiandrogenic effects result, at least in part, from the ability of the monoester metabolite (MBP) to inhibit testosterone production in the fetal rat testes (Akingbemi et al., 2004; Gray et al., 2000; Oishi and Hiraga, 1980; Parks et al., 2000). Repeated doses of DBP have been shown to reduce testosterone levels in the blood of adult male rats (Akingbemi et al., 2004; Kim et al., 2003, 2004; Parks et al., 2000) and in the testes of male fetuses from pregnant rats exposed to doses as low as 50 mg/kg/day from gestation day (GD) 12–19 (Lehmann et al., 2004).
In the environment, phthalates exist primarily as their dialkyl esters (ATSDR, 2001). After ingestion, they are rapidly hydrolyzed to the monoester and released into systemic circulation (Rowland et al., 1977; White et al., 1980). Monobutyl phthalate (MBP) is the most prevalent metabolite of DBP in rat and is thought to be the principal biologically active species (Mylchreest et al., 2000; Tanaka et al., 1978). MBP can be excreted unchanged in the urine, further metabolized via oxidation, or conjugated to glucuronic acid by uridine 5'-diphospho-glucuronosyltransferase (UDPGT). Oxidation of MBP occurs primarily through 
and -1 sidechain cleavage (Albro and Moore, 1974). The resulting oxidative metabolites may be further oxidized (with a terminal product of phthalic acid), conjugated by UDPGT, or excreted in urine (Albro and Moore, 1974). Glucuronidation appears to be a major route of clearance for MBP in the rat: MBP-glucuronide (MBP-G) accounted for greater than 20% of the total urinary metabolites and approximately 17% of the administered dose in rats after 1–10 mg/kg DBP iv (Payan et al., 2001). Both free MBP and MBP-G have been identified in the fetal rat plasma after maternal exposure to DBP (Fennell et al., 2004, Kremer et al., 2005; Saillenfait et al., 1998), though the processes driving fetal MBP kinetics (placental transport, glucuronide conjugation) are not well characterized. Although placental transfer of glucuronidated xenobiotics is typically inefficient (Dickinson et al., 1989; Fowler et al., 1988; Reynolds and Knott, 1989), the presence of MBP-G in the rat fetus suggests that it may be important to understanding fetal MBP exposure.
The combination of multiple nonlinear metabolic processes leads to rather complex behavior of DBP and its metabolites in vivo, making it difficult to predict a priori the pharmacokinetics of these chemicals at different doses. Inherent changes in physiology and biochemistry during pregnancy further compound this issue. Yet, in order to evaluate the risk associated with exposure to DBP, it is necessary to have a quantitative understanding of the dose–response of the active compound (MBP) at the target organ (fetal testis). The purpose of this research was twofold: (1) to gain a quantitative understanding of the relationship of fetal plasma and testes metabolite concentrations to external DBP doses, and (2) to identify the biochemical processes that are the major determinants of metabolite kinetics. In order to accomplish these goals, two in vivo studies were performed in which the distribution of MBP and MBP-G in the pregnant rat and fetus was measured after single or repeated doses of DBP. These data were then used to inform the development of a physiologically based pharmacokinetic (PBPK) model for DBP and its metabolites during gestation.
The structure of the PBPK model was determined based on three primary considerations: (1) its intended use, (2) the available data, and (3) previous investigations into DBP kinetics using PBPK models. The ultimate goal of our work was to develop a single model description that could be used with confidence to predict DBP, MBP, and MBP-G distribution and elimination in the pregnant and fetal rat under a variety of dosing scenarios, particularly those associated with observed developmental effects (repeated oral dosing). Although a few studies were available in the literature with information on DBP disposition during gestation (Calafat et al., 2006; Fennell et al., 2004; Kremer et al., 2005; Saillenfait et al., 1998), some significant data gaps still existed. In particular, urinary, fecal, and biliary excretion and the potential differences in kinetics after multiple exposures had not been characterized in the pregnant rat. Because more complete information on metabolite kinetics (including bile, urine, and feces) was available in the adult male than the pregnant rat, the model was initially parameterized for the adult male and then extended to gestation using the published and newly obtained kinetic data.
A previous PBPK model for DBP was also available in the literature (Keys et al. 2000) and was used to inform the current model structure. Keys et al. (2000) evaluated the ability of four different models (flow limited, diffusion limited, pH-trapping, and enterohepatic recirculation) to recapitulate blood levels of MBP after iv and oral dosing. The data available to Keys et al. (2000) and the goal of their work required that the models be kept relatively simple in structure (few adjustable parameters), precluding any description of the MBP metabolism or characterization of excreted products. In order to describe not only DBP and MBP, but also the glucuronide conjugate and oxidative metabolites of MBP in the plasma, urine, feces and developing fetus, a more comprehensive model is needed. By including all major metabolic (hydrolysis, oxidation, glucruonidation) and transport (biliary excretion, tissue partitioning, urinary and fecal excretion) processes, the current model provides both a means for investigating the processes driving kinetics and a platform for future use in risk assessment applications including extrapolation across doses, exposure routes and eventually species.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Pharmacokinetic Studies In Vivo
Pregnant Sprague–Dawley (Crl:CD(SD)) rats (Charles River Laboratories, Raleigh, NC) were housed in a temperature- and humidity-controlled, high efficiency particulate air-filtered environment on a 12-h light–dark cycle. Rats were provided NIH rodent diet (NIH-07, Zeigler Bros., Gardner, PA) and reverse-osmosis water ad libitum. Dosing solutions were prepared by mixing DBP and corn oil (Sigma-Aldrich, St Louis, MO). DBP concentrations in the dosing solutions were verified to be 48 ± 0.4, 89 ± 10, and 502 ± 9 mg/ml (mean ± SE) by gas chromatography with the method of Fennell et al. (2004)
. In the single dose study, GD 19 pregnant Sprague–Dawley rats (sperm positive on GD 0) were administered a single oral dose of DBP (0 or 500 mg/kg) in corn oil (1.0 ml/kg). Rats were euthanized with CO2 at 0.5-, 1-, 2-, and 24-h postdosing. In the repeated dose study, pregnant rats were given a daily dose of DBP (0, 50, 100, or 500 mg/kg) dissolved in corn oil (1.0 ml/kg) from GD 12 through 19. At 0.25, 0.5, 0.75, 1, 2, 4, 8, 12, 24, and 48 h after the final dose, rats were euthanized by CO2. Maternal blood was drawn by cardiac puncture. Fetal blood was collected using heparinized capillary tubes and pooled by litter in glass centrifuge tubes. Plasma was separated from whole blood by centrifugation at 1900 x g at 4°C and stored at –80°C. Placenta samples and amniotic fluid that was visibly clear of blood or other contaminants were pooled by litter. Testes from each male fetus were stored separately. All amniotic fluid and tissue samples were snap frozen in liquid nitrogen and stored at –80°C. Tissues were prepared and analyzed for MBP and MBP-G as described in the Supplementary Data.
Area under the curve (AUCt-inf), half-life (t1/2), and mean residence time (MRTt-inf) for MBP and MBP-G in the maternal and fetal plasma data from the repeated dosing study were calculated in WinNonlin (Pharsight, Mountain View, CA). Data were averaged by time point and dose group and modeled using a noncompartmental analysis with extravascular administration and no weighting. The data from the single dose study did not contain a sufficient number of time points to carry out a comparable analysis.
PBPK Model Structure
Keys et al. (2000) showed that a simple flow-limited model was sufficient for the description of DBP in the blood, whereas transport-limited tissue distribution and enterohepatic recirculation were important determinants of short-term MBP kinetics. These insights were used to inform the initial model structure. However, as the current model was intended to describe several aspects of DBP metabolite kinetics that were not addressed in the Keys model (MBP-G in plasma, fecal and urinary excretion products, and most importantly, fetal exposure), decisions regarding the current model structure were ultimately based on available disposition data. Defining the structure of the model was an iterative process, wherein the behavior of the data was used to develop an initial model, which was then tested by simulating additional data sets. Failure to recapitulate data led to revisions of the mechanistic hypotheses and model restructuring. For example, in the most preliminary version, only DBP, MBP, and MBP-G were specified in the model. However, when simulations were performed for the data of Payan et al. (2001), it was clear that urinary clearance and total plasma radioactivity could not be accurately described without including a description of the oxidative metabolites. Thus, the model was extended to include a simple description of the combined oxidative metabolites in the plasma and urine, which allowed the whole data set to be reproduced.
The final model (Fig. 1) contains four interconnected submodels, each with the necessary amount of detail to adequately describe its chemical species: DBP, MBP, MBP-G, and the combined oxidative metabolites (MBP-O). The individual submodels interact at sites of metabolism (hydrolysis of the diester, glucuronidation, hydrolysis of the glucuronide, and oxidation). The models for each chemical species in the adult rat are described below, followed by the modifications made to describe gestation. PBPK model simulations were performed using ACSL 11.0 (Advanced Continuous Simulation Language, AEgis Technologies Group, Inc., Huntsville, AL). Examples of model equations are shown in the Supplementary Data.
|
Adult Male Rat Model
Intact DBP.
Enzymes responsible for the hydrolysis of DBP are present in the intestinal mucosa, plasma and liver (Rowland et al., 1977
; Tanaka et al., 1978). Hydrolysis of DBP in the plasma (kbc) and liver (kc) is described as a first order rate as none of the tested doses were sufficient to overwhelm hydrolysis. Hydrolysis of the DBP in the upper GI (stomach + small intestine; GC1), on the other hand, is described as a saturable process based on the in vitro data of Rowland et al. (1977) and the apparent saturation of oral uptake at the highest doses (
500 mg/kg) (NIEHS, 1994). In vitro studies suggest that DBP is poorly absorbed in the gut, though a small amount may enter circulation intact, particularly at doses where hydrolysis is overwhelmed (White et al., 1980). Unabsorbed DBP may also be passed in to the lower intestine (GC2) and cleared in the feces. Oral absorption is described as a first order process (kad). Transport between the upper intestine (GC1) and lower intestine (GC2) compartments (kgic) is a first order rate. Fecal excretion (kfc) is described using a clearance rate (l/h). DBP that is taken up into the gut wall is passed to the liver via the portal blood where it is hydrolyzed, released into systemic circulation with the plasma or excreted into the bile (Tanaka et al., 1978). Biliary transfer of DBP into the duodenum is modeled as a clearance rate (kbdc) from the liver to the upper intestine (GC1). Efficient hydrolysis ensures that essentially 100% of the DBP is de-esterified before it can be taken up into the bile. However, competition between these processes (particularly after saturation of intestinal lipases at 
500 mg/kg doses) is included in the model in order to describe the wide range of published data. Transport of DBP into the tissues is assumed to be flow limited.
Free MBP.
Oral absorption is described as a first order process (kam). Movement through the GI and fecal excretion are clearance rates. Unlike DBP, MBP is readily absorbed in the gut wall and passed to liver via the portal blood. Glucuronidation and oxidation of free MBP in the liver are described using saturable kinetics. Free MBP may also be excreted into the bile or released into systemic circulation. Transfer of MBP in the bile is described in the same manner as DBP, using a clearance rate (kbmc) from the liver to the upper intestine (GC1). The MBP may then be reabsorbed or transported (kgic) into the lower intestine (GC2). Absorption may occur in both the upper and lower intestine compartments. Free MBP that is not absorbed in the intestine is cleared via the feces (kfc). Transport of MBP into the tissues from the plasma is modeled using diffusion-limitation. Secretion of free MBP into the urine is a saturable process, based on nonlinear behavior of excretion data at low doses (Payan et al., 2001).
MBP-glucuronide.
MBP-G formed in the liver (VmaxLc, KmL) may be excreted into the bile or released into systemic circulation. Biliary transfer of MBP-G is modeled as described for free MBP above. MBP-G then travels through the intestine (kgic) and is either hydrolyzed to MBP via β-glucuronidase (khydrc) in the lower intestine (GC2) or passed in the feces using first order clearance rates. In the model, the hydrolysis of MBP-G is the rate limiting step for reabsorption of conjugated MBP from the bile. Distribution of MBP-G into the tissues is modeled using flow limitation, assuming distribution with body water. Urinary excretion of MBP-G is modeled as a first order clearance rate from the plasma compartment.
Oxidative metabolites of MBP.
MBP-O formed by P450 metabolism (VmaxOc, KmO) in the liver is released into the body via the venous blood. A one-compartment volume of distribution model is used to describe the combined oxidative metabolites. Distribution is assumed to be the body water compartment. Urinary excretion is modeled as a first order clearance rate from the central compartment.
Modifications for Gestation
During gestation, both MBP and MBP-G are allowed to move freely between the arterial and placental plasma. Transfer of MBP between the placental plasma and the fetal plasma is described as diffusion-limited (Gentry et al., 2003). Versions of the model were tested which did or did not allow MBP-G to cross the placenta (described below). Based on the fit of the model simulations to available data, the final model does not include placental transfer of MBP-G. Based on fetal MBP kinetic data, as well as published data on UDPGT and β-glucuronidase activities in fetal tissues (Lucier et al., 1975; Lucier and McDaniel, 1977; Wishart, 1978), glucuronidation of MBP and hydrolysis of MBP-G are included in both dam and fetus. Transfer of MBP and MBP-G between the fetus and amniotic fluid are described as first order processes. Transfer between the fetal plasma and testes tissue is described using diffusion-limited transport.
Parameterization in the Adult Male Rat PBPK Model
Physiological Parameters
Physiological parameters were obtained from measured values in the literature (Table 1). Adult male rat, body weight (BW), cardiac output, and fractional tissue volumes and blood flows were available from Brown et al. (1997). Fractional tissue volumes were scaled by BW and blood flows were scaled by BW0.75.
|
Kinetic Parameters
Final values for the kinetic parameters are given in Table 2. The majority of the data presented in this paper were obtained from Sprague–Dawley rats. However two of the studies in the adult male rat were performed in Wistar–Furth rats (NIEHS, 1994
, 1995). Although some variability may exist for parameters between these strains, it was not possible to tease out differences in biochemical parameters between rat strains based on the limited data currently available. Therefore kinetic parameters were kept the same for both strains and scaled allometrically as is typical for intra- and interspecies extrapolation (Dedrick, 1973). Permeability area cross product (PA), Vmax, and clearance constants were scaled by BW0.75. With the exception of metabolism parameters, fetal parameters were scaled in a similar manner to the maternal parameters: PAs were scaled by vfet0.75 and then multiplied by the total number of fetuses (n = 8) to obtain the value for the litter. Metabolism parameters (glucuronide conjugation and hydrolysis; VmaxLf and khydrfc) were scaled by adjusting the adult value by the ratio of the fetal:adult liver weight, because the fetal liver weight is not linearly correlated with BW. The value for the total litter was then calculated by multiplying by the number of fetuses (n = 8). Whenever possible parameters were taken from published values or calculated from in vitro studies. Nonetheless, the lack of specific tissue and metabolism data required that many of the model parameters be fitted to in vivo kinetic data.
|
Measured parameters.
Tissue:plasma partition coefficients were obtained from radiolabeled DBP studies (Williams and Blanchfield, 1975
), and as such, represent total phthalate rather than a specific compound. However, because free MBP is expected to be the primary chemical form in the tissues, the total radioactivity should be a reasonable predictor of MBP partitioning. Tissue:plasma ratios obtained from early time points (4 h postdosing) of radiolabeled studies were generally similar to the values obtained from vial equilibration studies performed by Keys et al. (2000) (partition coefficient for MBP in the richly perfused tissues (PMR) 1.3 vs. 1.2), with the exception of the partition coefficient for the slowly perfused tissue (muscle). Because their vial equilibration studies yielded questionable results for the muscle, Keys et al. (2000) used their published values for monoethylhexyl phthalate (MEHP) as surrogate for MBP partitioning in the slowly perfused tissue (PMS). Although muscle was not measured in Williams and Blanchfield (1975), another study indicated that muscle:blood ratios were similar to the other richly perfused tissues 24 h postdosing (Tanaka et al., 1978). Both the in vitro value for MEHP and the value described above for MBP in the richly perfused tissue at 4-h postdosing were tried as surrogates for the PMS by running the model against iv plasma data (NIEHS, 1994) and slowly perfused tissue data (Tanaka et al., 1978; not shown). The value for PMR (1.3) provided a better visual fit to the data and was therefore used in the model.
The affinity constant (Km) for hydrolysis in the gut was calculated from the data of Rowland et al. (1977). Rowland and coauthors spiked 1-ml aliquots of rat stomach contents with various concentrations of DBP and measured remaining DBP after 16 h. The Km for hydrolyzing enzymes (KmG = 350 mg/l) in the intestine was calculated by fitting a one-compartment model to the data in Berkeley Madonna (University of California, Berkeley, CA).
Fitted and calculated parameters.
For parameters that were not measured experimentally, the values were estimated by adjusting the parameters to obtain the best visual fit of the model to time course data. At this stage of model development and evaluation a statistical program was not used to estimate model parameters, although such an exercise would be particularly useful in conjunction with further applications (e.g., extrapolation to the human). For the duration of this paper, fitting of parameters refers to the manual process of determining one set of parameters that could consistently recapitulate a large base of diverse data. In order to minimize uncertainty in parameters, a sequential approach was followed so that the most pertinent data sets were used for each parameter. This approach to model parameterization is illustrated in Figure 2 and is described below.
|
Time course data on plasma MBP levels after a single iv dose (8 mg/kg) of the monoester (NIEHS, 1994
) were used to develop initial estimates of values for the parameters governing liver metabolism and diffusion-limitation in the richly and slowly perfused tissues (PARc and PASc). First approximations of these parameters were based on this iv data set because it represented the simplest possible dosing scenario, and was therefore dependent upon the least number of adjustable parameters. However, as this study measured only plasma MBP concentrations, additional data were needed to refine the parameters driving elimination and enterohepatic recirculation.
Payan et al. (2001) provided a more comprehensive data set, which was ideal for estimating elimination and metabolism parameters. Both iv and dermal exposure data were presented. However, as the dermal data could not be used in the model without significant changes to the structure that were not pertinent to the current work, only the iv data were used. The iv study included plasma DBP, MBP, MBP-G, and total 14C levels after a single iv dose of either 1 or 10 mg/kg 14C-DBP to adult male rats. Elimination of MBP, MBP-G, and total 14C was measured over 72 h in the urine. Bile cannulation or a sham operation (controls) was performed on a subset of rats dosed with 1 mg/kg 14C-DBP and the total amount of 14C excreted in the bile and/or feces and urine was measured at 30 h. Metabolism parameters (glucuronide conjugation, oxidation), and urinary clearance rates were adjusted to fit DBP, MBP, MBP-G and total 14C in the plasma and urine at the 1 mg/kg dose. Based on the increased water solubility of glucuronide conjugates, it was assumed for the sake of simplicity that MBP-G would be distributed with the water (volume of distribution (VD) = 0.65). The partition coefficient for MBP-G in the tissue (PGT) was then calculated from the VD, blood volume (VB), and combined tissue volume (VT) to be 0.3 [PGT = (VD – VB)/VT]. Bile cannula studies were simulated by turning off transfer of DBP, MBP, and MBP-G from the liver to the upper intestine and biliary transport parameters (kbg, kbm, kbd) were fit to the bile excretion data. The rates of MBP-G hydrolysis in the gut (khydrc) and fecal excretion (kfgc, kfdc, kfmc) and movement in the gut (kgic) were then adjusted to reproduce fecal excretion and plasma data in sham-operated rats.
With the above parameters constrained, the first order rate of MBP oral absorption (kam) was determined. Kam was fit to plasma MBP data obtained from adult male rats after a single oral bolus of 34 mg/kg MBP (NIEHS, 1994). Absorption of unchanged DBP was then calculated based on in vitro data using everted gut preparations (White et al., 1980). The measured rate for DBP absorption was 67-fold less than MBP. Using this measured ratio and the previously determined value for MBP absorption (20/h), a value of 0.3/h was calculated for kad.
PBPK Model Validation in the Adult Male Rat
Validation of the model structure and the final parameters was performed using separate data sets from those used for parameterization (NIEHS, 1995; 10 mg/kg from Payan et al., 2001). No parameters were changed to improve model fits to the "validation" data.
Application of Adult Male Rat DBP PBPK Model to Gestation
Physiological Parameters
During gestation, mammary gland (VM), and fat (VF) tissue growth were described as a linear processes based on the data of Hanwell and Linzell (1973), Knight and Peaker (1982), and Naismith et al. (1982), as described in Clewell et al. (2003). Placental volume (VPl) was described as the sum of the yolk sac and chorioallantoic placenta based on the model of O'Flaherty et al. (1992). Growth equations were available in the cited papers. The total BW of the dam was made equal to the initial BW plus the change in volume of the uterus, fat, mammary gland, placenta, and fetus. Fetal volume (Vfet) was described using the equations of O'Flaherty et al. (1992). Growth of fetal testes is proportional to the total BW, accounting for approximately 0.1% of the total fetal volume from GD 16 through the end of gestation (LaBorde et al., 1992; Naessany and Picon, 1982; Parks et al., 2000). Changes in amniotic fluid volume were described using a TABLE function in ACSL, with linear interpolation between data points (Park and Shepard, 1994; Wykoff, 1971).
Maternal cardiac output was described as the sum of initial cardiac output (Brown et al., 1997) and the change in blood flow to the placenta, mammary and fat tissues, per the approach of O'Flaherty et al. (1992). Changes in the fractional cardiac output to the mammary gland, fat and yolk sac were assumed to be proportional to changes in tissue volumes, with the exception of the chorioallantoic placenta which increased more rapidly than the tissue volume. Chemical transport within the fetus was modeled using diffusion, rather than blood flow limitation. Thus, no assumptions were made as to proportional blood flows to fetal tissues.
Kinetic Parameters
Two kinetic parameters were adjusted before using the model in the pregnant dam based on published in vitro studies: VmaxGc and VmaxLc. VmaxGc, the maximum capacity for DBP hydrolysis in the intestine, was decreased based on in vitro metabolism studies. Rowland et al. (1977) measured the disappearance of DBP in 1-ml aliquots of small intestine contents from adult male and female rats (33–40 days old). Their studies showed a sex difference in the ability to hydrolyze DBP, with the females metabolizing only 60% as much as the males. This sex difference was accounted for in the model by reducing the unscaled value for VmaxGc (90 vs. 150 mg/h/kg) for all simulations of female rats. The parameter was then scaled by BW0.75 to account for differences in BW.
VmaxLc, the maximum capacity for glucuronide conjugation in the liver, was also reduced in simulations of pregnant rats. In vitro studies in livers of nonpregnant female and pregnant rats showed that UDPGT activity for a variety of substrates (steroidal and nonsteroidal) was decreased by approximately 50% during gestation (GD 19–20) (Lucier et al., 1975; Luquita et al., 2001). This difference in activity was included in the model by reducing the unscaled value of VmaxLc by 50% (150 vs. 300 mg/h/kg) during gestation, which was scaled by BW0.75 to account for changes in BW. All other maternal parameters were scaled allometrically from the adult male rat.
Parameters describing transfer between the dam and fetus, and the fetus and amniotic fluid were fit to published placenta, fetal plasma, and amniotic fluid time course data (Fennell et al., 2004). Because no previous data were available for MBP concentrations in the placenta tissue or fetal testes, the partition coefficients (Pmpl, Pmft) for these tissues were fit to the data from the single dose data from the current studies (500 mg/kg).
Two possible explanations were considered for the presence of MBP-G in the fetus (Fennell et al, 2004): (1) MBP-G is not significantly transported in the placenta and fetal MBP-G must be formed in the fetus itself, or (2) maternally formed MBP-G is able to cross the placenta. Alternative versions of the model were developed and tested with the MBP-G time course data to determine which one described the data more accurately. In the first version, glucruonide conjugation and hydrolysis of MBP-G was described in the central compartment of the fetus (plasma). Fetal metabolism parameters were estimated from in vitro data described below. In the second version of the model, placental transfer of MBP-G was described in the same manner as free MBP, in addition to fetal UGT and β-glucuronidase activity. The partition coefficient and permeability area cross product were adjusted to achieve the best visual fit to the fetal plasma MBP-G time course data.
During fetal development, rat UDPGTs are classified into two distinct classes based on the preferred substrates. Steroidal UDPGTs are characterized by a low activity in the fetal liver, with a rapid increase to adult levels after birth. Nonsteroidal UDPGTs show a peak in activity on GD 19 at nearly adult levels, followed by a drop in activity at birth and a subsequent increase in adolescent rats (Lucier and McDaniel, 1977; Wishart, 1978). Although studies have not been performed specifically on glucuronidation of MBP in fetal livers, studies with a similar chemical, monoethylhexyl phthalate, showed that it is metabolized by steroidal UDPGTs (Sjoberg et al., 1999). Therefore, the fetal UDPGT activity in the model was estimated from the ratio of measured fetal:adult UDPGT activities toward several steroidal substrates from measured in vitro values. The in vitro activity averaged for the measured steroidal substrates was 0.6, 1.3, 2.5, 11.8, and 11.4% of the adult male rat value per mg microsomal protein on GD 17, 18, 19, 20, and 21 (Lucier and McDaniel, 1977; Wishart, 1978). These relative activities were used to calculate the value for maximum capacity of fetal UDPGT according to Equation 1. The amount of microsomal protein in the liver of the Sprague–Dawley fetus (7.3 mg/g liver) was obtained from Alcorn et al. (2007), as it was not reported in the Lucier and McDaniel (1977) or Wishart (1978) studies. The final values for VmaxLf (0.05, 0.12, 0.6, 3.0, and 4.4 mg/h) were coded into the model using a TABLE function, which employs linear interpolation to estimate parameter values between defined points. Fetal β-glucuronidase activity (khydrfc) was found to be 20% of the adult value based on measured activities per mg microsomal protein in vitro for various substrates (Lucier et al., 1975; Soucy et al., 2006). Khydrf was estimated from the adult male rat value in the same manner as VmaxLf.
 | (1) |
After setting the parameters, testing the model against the data of Fennell et al. (2004) and deciding on a final model for the fetal MBP-G, the gestation model was then tested with additional data sets from single dose studies on different days of gestation and alternate dose routes: current studies (GD 19, po); Saillenfait et al. (1998) (GD 14, po), and Kremer et al. (2005) (GD 19, iv). Saillenfait et al. (1998) treated GD 14 Sprague–Dawley rats with a single oral dose of 500 or 1500 mg/kg 14C-DBP in mineral oil. Total radioactivity, 14C-MBP and 14C-MBP-G were measured in the maternal plasma, placenta, amniotic fluid and whole fetus at 0.5-, 1-, 2-, 4-, 6-, 8-, 24-, and 48-h postdosing. Total radioactivity in the maternal urine and feces was also measured at 24- and 48-h postdosing. Kremer et al. (2005) performed a time course study in GD 19 Sprague–Dawley rats after a single iv dose of 10, 30, or 50 mg/kg MBP. Serial samples were collected from cannulated rats, and MBP and MBP-G plasma concentrations were measured at 0.08-, 0.25-, 0.5-, 0.75-, 1-, 1.5-, 2-, 4-, and 8-h postdosing.
Extrapolation of Acute DBP Gestation PBPK Model to Multiple Day Exposures
The ability to describe repeated dosing scenarios was tested by running the model against the repeated dosing data from the current studies without adjusting any of the model parameters determined from the single dose data. The model was also tested with the data of Calafat et al. (2006), who measured free MBP in the amniotic fluid of GD 18 Sprague–Dawley rats. Dams received an oral dose of 0, 100, or 250 mg/kg DBP once daily from GD 12–17. Amniotic fluid was pooled by litter and collected at the time of sacrifice (GD 18). Urine catch samples were also collected on GD 17 (approximately 6 h postdosing). However, because the urine samples were presented as concentrations and volumes were not included, they could not be used straightforwardly in the model.
Sensitivity Analysis of Model Parameters
A normalized sensitivity analysis was run on the gestation model to examine the relative influence of each of the parameters on model output. The model was run to determine the change in the average plasma MBP concentrations (24 h AUC) resulting from a 1% change in the value of each kinetic parameter. In an effort to determine the effect of metabolic saturation on relative parameter importance, the sensitivity analysis was performed at two nominal doses of DBP, representing unsaturated and saturated states (10 and 500 mg/kg, respectively). The following equation shows the calculation of the sensitivity coefficient.
 | (2) |
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
In order to relate tissue concentration of DBP metabolites to observed effects during fetal development, the distribution of MBP and MBP-G in the pregnant rat and fetus were determined after a single dose of 500 mg/kg DBP and repeated doses of DBP at a low (50 mg/kg), medium (100 mg/kg) and high (500 mg/kg) doses. In addition to allowing direct comparison of external and tissue dose, these data also supported the development of a PBPK model that was used to evaluate the relative importance of transport and metabolic processes in the pharmacokinetic behavior of DBP across doses, providing a robust platform for future risk assessment applications.
Pharmacokinetic Studies In Vivo
The repeated dosing study was designed to reproduce the experimental conditions of the published DBP effects studies in order to gain a better understanding of internal dose associated with observed effects. In this case, maternal and fetal plasma, placenta and amniotic fluid MBP levels may be used as surrogates for fetal dose. Concentration time course data collected in this present study appear in the Supplementary Data (Tables S1–4). Some key aspects of disposition of DBP and its metabolites were noted from the initial PK analysis of the data (Table S5).
Dose-dependent differences in AUC/D, Cmax, t1/2, and MRT illustrate the nonlinear behavior of MBP in the repeated dosing study (Supplementary Data), leading to some preliminary hypotheses about dose-dependent kinetics. For example, the fact that the apparent parameter differences in maternal plasma MBP do not take place at the same dose may suggest that more that more than one saturable process is driving MBP dose–response. Differences (two-fold) in Cmax/D, t1/2 and MRT between the 100 and 500 mg/kg/day groups, may be explained by saturation of DBP intestinal hydrolysis and prolonged absorption of MBP. AUC/D, on the other hand, increases from 2.6 to 4.8 between the 50 and 100 mg/kg/day groups, suggesting that a second saturable process (possibly clearance) may be involved. Nonetheless, the parameters are certainly affected by a combination of many factors and these simple PK models do not lend themselves to in-depth analysis of the processes driving dose-dependent kinetics. PBPK models that incorporate the various biochemical pathways into their description are better suited to mechanistic investigations. The model described in this paper was specifically designed for this purpose. The PBPK model also allows comparison of studies with fewer data points (i.e., the single dose study, testes data), where classical PK analysis cannot be performed.
PBPK Model Validation in the Adult Male Rat
The results of model development and parameterization in the adult male rate are provided in the Supplementary Data (Figs. S1 and S2, Table S7). The final model differed significantly in structure from the previous model of Keys et al. (2000), primarily as a result of the current need to describe additional metabolites in tissues and urine and to facilitate extension to gestation. Differences between the two models are summarized in the Supplementary Data (Table S6). Before extending the model to gestation, the final model structure and parameter values were tested by simulating data in the male rat that were not used for parameterization. No parameters were altered to fit the validation data. Plasma and urine metabolite kinetics were well-described by the model for iv doses of 10 mg/kg 14C-DBP (Fig. 3). Predicted fecal and urinary excretion were 9 and 89% of the total dose compared with the measured values of 10.9 ± 1.1 and 85.8 ± 2.4%, respectively (Payan et al., 2001). The model also reproduced plasma MBP kinetics after a single iv dose of 20 mg/kg DBP or an oral dose of 50 or 200 mg/kg DBP (NIEHS, 1995) (Fig. 4).
|
|
Application of Adult Male Rat DBP PBPK Model to Gestation
The reduced value for VmaxLc based on in vitro data improved the fit of the model to maternal plasma simulations of the data of Fennell et al. (2004)
(Fig. 5). Increasing the value for khydrc would provide similar benefits to model fit. However, literature suggests that β-glucuronidase activity is not increased in the maternal liver during gestation (Isaksson et al., 1988). Therefore, the value for khydrc was not changed in the gestation model. The reduced capacity for diester hydrolysis in the female versus the male (90 vs. 150 mg/h/kg) did not affect plasma appreciably at these dose levels (50–250 mg/kg DBP). No additional parameters were changed to obtain the fits shown in Figure 6A.
|
|
Fetal plasma MBP data from this same study (Fennell et al., 2004
) were used to fit the permeability area cross product and partition coefficient for placental–fetal transfer of MBP (PAFetc, PMFet), and first order transfer rates between the fetal plasma and amniotic fluid (ktransm1c, ktransm2c) (Figs. 6B and 6C). Two descriptions of fetal MBP-G kinetics were tested with the model. The first version assumed fetal MBP-G is formed and hydrolyzed in situ. Using this description, together with the literature-derived parameters for glucuronide conjugation and hydrolysis, the model successfully described the slower appearance of MBP-G in the fetal plasma as compared with MBP (Fig. 6B). The second version of the model included the possibility of diffusion-limited transfer of MBP-G between the mother and fetus. Addition of placental transfer did not improve the fit of the model to the published data over that which assumes all MBP-G is formed in the fetus. In fact, any attempt to include transfer to or from the fetus caused the model to overpredict MBP-G uptake or clearance in fetal plasma, respectively (not shown). Thus, the placental transfer was not included in the final version of the model. This description is supported by the ability of the model to reproduce the nonlinear behavior of fetal MBP-G across doses, as well as the changes in fetal MBP-G levels across different days of fetal development. Nonetheless, additional studies (i.e., ex vivo placenta perfusion) are needed to definitively rule out placental transport of the glucuronide.
The utility of the gestation model for extrapolation across doses, routes and gestational age was tested against a combination of data from our laboratory and published studies: single oral DBP dose on GD 19 (current study; Fig. 7) single oral DBP dose on GD14 (Saillenfait et al., 1998; Fig. 8), and single iv MBP dose on GD 19 (Kremer et al., 2005; Fig. 9). Model simulations were consistent with single dose kinetics in the maternal and fetal plasma, and placenta across dose routes (iv vs. po) and dose levels (10 mg/kg MBP and 500 mg/kg DBP) without changing any of the previously determined parameters. The success in simulating the trend of the data from GD 14 to 20 supports the approach used to estimate parameters that are changing throughout gestation. In particular, maternal glucuronide levels in gestation (reduced vs. predictions in male rat) and the increase in fetal glucuronidation capacity with gestational age are reproduced using the Vmax values calculated from the in vitro activity measured for steroidal substrates.
|
|
|
The model adequately captures the kinetic behavior of MBP and MBP-G in the amniotic fluid at the lower doses (50–250 mg/kg) 19–20. However, they were not as concordant for GD 14 (not shown). The data of Saillenfait et al. (1998)
suggested that clearance of both MBP and MBP-G is much faster in the GD 14 fetus. It was not possible to fit that particular data set without making large changes (>10-fold) in the fetal/amniotic fluid transport parameters. Without additional support for a time-dependent difference in amniotic fluid transfer, this change was not made to the model. In contrast, amniotic fluid MBP and MBP-G levels from the current single dose studies indicated reduced clearance at the 500 mg/kg dose on GD 19 (Fig. 7C). The model is able to predict amniotic fluid after doses of 50–250 mg/kg DBP, but underpredicts amniotic fluid levels at 500 mg/kg. The reason for this nonlinear relationship between the fetal plasma and amniotic fluid is not currently known. Fetal testes parameters were fitted to the data from the current single dose study (Fig. 7D), as no previous data were available for this tissue. Both MBP and MBP-G appear to be distributed with body water. Partition coefficients were 0.3 and permeability area cross products were 1 l/h/kg, suggesting rapid equilibration with the plasma.
Model predictions for 24 h urinary (77% of dose) and fecal (15% of dose) excretion in the GD 20 rat were similar to the measured values in the naïve female rat (77 ± 8 and 7 ± 3.5% of dose, respectively) after a 100 mg/kg oral DBP dose (Fennell et al., 2004). Model-predicted excretion did not differ between the nonpregnant and GD 20 rat.
Extrapolation of Acute DBP Gestation PBPK Model to Multiple Day Exposures
In general, the model was able to predict maternal and fetal plasma and placenta kinetic behavior at the 50, 100, and 500 mg/kg/day doses from the current repeated dose studies (Figs. 10 and 11) without changing any kinetic parameters. Although the model falls outside of the standard deviations for some time points, it consistently captures the kinetic behavior of MBP and MBP-G, including the slowed uptake and clearance of MBP at high doses (Fig. 10A) and the apparent accumulation of MBP-G in the fetal plasma and amniotic fluid over time (Fig. 11). The majority of the fetal testes samples had MBP and MBP-G concentrations that were below the method limit of quantitation, with the exception of the 0.5- to 2-h time points in the 100 mg/kg/day group and the 0.5- to 12-h time points in the 500 mg/kg/day groups. Model simulations based on flow-limited transfer of MBP and MBP-G between the fetal plasma and testes are shown versus the data in Figure 11B. The lack of analytical sensitivity precludes testing of the model at lower doses. However, the ability to predict uptake and peak concentrations within a factor of two from the mean at 100 and 500 mg/kg/day, suggests that fetal testes concentrations are linearly correlated with fetal plasma levels. Additionally, the ability of the flow-limited model to describe single and repeated dose data (Figs. 7B and 11B) suggests that MBP does not accumulate in the fetal testes over time. In fact, the testes concentrations are consistently lower than fetal plasma levels (Pmft = 0.3), which are linearly correlated with maternal plasma MBP concentrations.
|
|
The repeated dosing data clearly exhibit dose-dependent kinetics (Figs. 10 and 11; Table S5). This behavior was seen consistently across data sets (NIEHS, 1994
; Saillenfait et al., 1998). The model successfully simulated the dose–response trend in the plasma, indicating that the model structure has components consistent with the biological determinants of metabolite kinetics. Visual examination of the model fit with varying parameter values showed that saturation of glucuronidation in the maternal liver is primarily responsible for the reduced clearance at DBP doses 500 mg/kg. Intestinal DBP hydrolysis is also saturated at the high dose, leading to slower absorption and a prolonged period of uptake, evidenced by the increased Tmax. Reduced absorption also leads to increased presystemic loss through fecal excretion in the model, which has a negative effect on plasma AUC. The model prediction of both MBP and MBP-G is within a factor of two of most of the measured amniotic fluid concentrations at the 50 and 100 mg/kg/day doses (Fig. 11C). However, as was noted in the single dose simulations, amniotic fluid concentrations are severely underpredicted by the model at the 500 mg/kg/day dose. Measured MBP and MBP-G in the amniotic fluid of the 500 mg/kg/day dose group are approximately five times higher than would be expected based on the lower dose groups assuming a linear dose–response. It is possible that a dose-dependent biokinetic difference (saturable transport/hydrolysis within the amniotic fluid, transporter mediated uptake in the fetus, etc.) could be responsible for this observed behavior. Additional studies are needed to better define the mechanisms of chemical transport between fetus and amniotic fluid in order to better understand observed kinetics across doses.
The data of Calafat et al. (2006) were simulated (not shown here) by providing a daily maternal oral dose of 100 or 250 mg/kg DBP from GD 12–17. The simulation was allowed to run to the reported day of sacrifice (GD 18). The model-predicted amniotic fluid MBP concentrations (1.6 and 6.5 mg/l) were within the range of the measured values for both the 100 (0.3–2.4 mg/l) and the 250 mg/kg/day (3.8–22.3 mg/l) dose groups.
Sensitivity Analysis
Sensitivity analysis performed at 10 and 500 mg/kg DBP revealed a dose-dependent difference in model sensitivity to some of the chemical specific parameters (Fig. 12). At 10 mg/kg, the plasma MBP concentrations are more sensitive to parameters governing oxidative metabolism (KmO, VmaxOc) of MBP. At the higher dose (500 mg/kg), where oxidative metabolism is saturated, the model predictions of plasma MBP levels are more sensitive to parameters describing elimination and recirculation of the glucuronide conjugate. Oral absorption also becomes important at 500 mg/kg/day; VmaxGc and kad are positively correlated to plasma levels and kgic is negatively correlated. This suggests that upon saturation of hydrolysis (VmaxGc), inefficient absorption of the diester and clearance via fecal excretion increases presystemic loss, and reduces maternal exposure. As expected, fetal plasma MBP levels are influenced by the same parameters governing maternal plasma concentrations. Pmfet, the parameter governing partitioning between maternal and fetal plasma, had the largest effect on fetal AUC, with a sensitivity coefficient of 0.9. The AUC for MBP in the fetal testes showed the same dependence on maternal parameters as illustrated in Figure 12, in addition to Pmfet and the testes:plasma partition coefficient (Pmft). Maternal plasma levels are not affected by changes in Pmfet or Pmft.
|
Using the Multiple Exposure Model to Interpret Effects Data
Several studies have shown that exposure to DBP during the period of sexual development in the fetal rat may lead to delayed development of the male reproductive tract. Dose–response examinations showed several overt effects in the male offspring from dams exposed to
500 mg/kg/day from GD 12–21, including hypospadias (cleft penis), nipple retention, reduced anogenital distance (AGD), cryptorchidism, seminiferous tubule degeneration, testis interstitial cell proliferation, and malformed epididymis, seminal vesicle, vas deferens and ventral prostate (Mylchreest et al., 2000). Of these effects, only nipple retention and reduced AGD have been found lower doses (100 and 250 mg/kg/day, respectively). Delayed preputial separation has been noted in some studies at doses 100 mg/kg/day, but not in others at doses up to 500 mg/kg/day (Mylchreest et al., 1999, 2000). All of these adverse effects occur in concert with decreased testes testosterone concentration, which may provide a more sensitive marker for disruption of androgen-dependent development (Gray et al., 2000; Lehmann et al., 2004; Mylchreest et al., 2002). Although these effects have been well-studied with regard to external DBP dose, there is currently no information on the internal (fetal) dose expected from these exposures. In order to illustrate how this PBPK model may be used to relate external DBP to internal MBP dose in multiple day studies, the model was run with the repeated exposure parameters at 16 different dose levels, ranging from 1 to 550 mg/kg/day. The AUC for the last day of dosing was then used to calculate the average concentration in the maternal and fetal plasma, and fetal testes (daily AUC/24 h). The predicted dose–response is illustrated in Figure 13, together with reported lowest observed adverse effect levels (LOAELs) for several landmarks of male sexual development. Model predictions suggest that the average daily MBP concentration in the fetal rat testes must reach levels of approximately 1, 2, and 18 mg/l in order to cause noticeable changes in testosterone production, nipple retention or more overt effects (i.e., reduced AGD), respectively. Using the model, these testes concentrations may also be correlated to fetal and maternal plasma, which are often used as surrogates for fetal dose in the human. Maternal plasma MBP concentrations associated with reduced testosterone, nipple retention and reproductive tract malformations in the rat fetus are 5, 11, and 60 mg/l.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Few data are available in which controlled doses of phthalates have been administered to human subjects (Anderson et al., 2001
). In order to interpret such data and develop realistic estimates of fetus and/or fetal testes exposure, models such as the one provided here must be able to relate external dose to plasma, tissue and urine kinetics. PBPK models use animal and in vitro data to build a platform that can be used to predict human kinetics by accounting for differences in physiology and, where necessary, biochemistry (i.e., enzyme activities). In order to maximize the utility of this model in future applications to humans, this model included descriptions of MBP, MBP-G, and oxidative metabolites in the plasma and tissues of interest as well as the urine, feces and bile. Detailed radiolabeled studies allowed parameterization to be performed against specific data. The resulting model described metabolite distribution and excretion in adult male rats over doses spanning three orders of magnitude (1 mg/kg to > 500 mg/kg), regardless of the form of the chemical used (MBP, DBP) or the route of exposure (iv, po). Successful extrapolation of the male rat model to describe kinetics in the pregnant rat suggests that—with the exception of glucuronidation and intestinal hydrolysis—the metabolic and chemical transport processes do not change significantly as a result of pregnancy. Reduced hydrolysis of the diester in the intestine of the female rat had little effect on maternal kinetics other than the initial uptake at the highest dose (500 mg/kg). The reduced capacity for glucuronide conjugation does affect plasma kinetics, causing a notable increase in free MBP residence time in both maternal and fetal plasma. This dose-dependent behavior could have important implications for perinatal toxicity. As glucuronidation plays a major role in the elimination of many xenobiotics, a decrease in capacity during gestation could result in increased fetal exposure to a variety of chemicals. This finding emphasizes that understanding the determinants of chemical kinetics in the adult male will not be sufficient for assessing exposure to the pregnant female or fetus.
The comprehensive nature of this model allows a thorough investigation into the transport and metabolic processes that are responsible for the overall observed kinetics. Because all of the major metabolic (hydrolysis, oxidation, glucruonidation) and transport (biliary excretion, tissue partitioning, urinary and fecal excretion) processes are included in the model, it is possible to examine the effect of these individual processes on chemical behavior. This was accomplished both through a quantitative sensitivity analysis and through manual manipulation of model parameters coupled with observation of the effect on fits to the data. Model analysis indicates that saturation of diester hydrolysis is responsible for the prolonged absorption observed at doses > 100 mg/kg, and reduced phthalate exposure due to the poor absorption and subsequent presystemic loss of DBP. In fact, saturation of diester hydrolysis may mitigate the effect of reduced clearance on the plasma AUC to some extent. Saturation of glucuronidation appears to be responsible for reduced clearance of MBP the higher doses (100–500 mg DBP/kg/day). This conclusion is particularly interesting in light of the available toxicity data, as many of the more serious developmental effects have been seen only at doses at or above 500 mg/kg. Reduced anogenital distance, for example, which is a source of concern due to its indication of impaired sexual development in both the rat and human (Fisher et al., 2003) is only seen at doses 250 mg/kg/day. This effect may in fact be a result of reduced clearance of MBP due to saturation of glucuronidation. Saturation of the oxidation pathway and renal uptake of MBP occur at much lower doses than saturation of glucuronide conjugation (< 50 mg/kg). Thus, these pathways do not appear to be responsible for slowed clearance at high doses.
In the attempt to describe plasma kinetics at low doses (1 mg/kg DBP), it became clear that the oxidative metabolites are more important to overall kinetics than has previously been assumed based on high dose toxicity studies. In fact, these metabolites are predicted to make up approximately 40% of the excreted metabolites at the 1 mg/kg dose. The oxidative metabolites of monophthalates may still be active and could play a significant role in human toxicity, as they have been shown to constitute the largest fraction of the urinary metabolites after human exposure to DEHP, a phthalic acid diester with a similar toxicological profile to DBP (Koch et al., 2005). Although the current model attempts to account for these metabolites by including a generic compartment and urinary clearance of "other" metabolites, the description could only be tested by comparing MBP, MBP-G, and total radioactivity measurements simultaneously. The lack of data for validation of oxidative metabolites remains an important gap in the pool of information available for phthalate risk assessment efforts.
In addition to saturable metabolism and renal uptake, enterohepatic recirculation has a significant effect on the retention of MBP and MBP-G in the plasma. Greater than 40% of a given dose is transferred into the bile, and nearly 90% of those bile metabolites are then reabsorbed in the intestine and re-released into circulation. When reuptake is absent in the model, clearance of both MBP and MBP-G from the plasma is increased and the free MBP and oxidative metabolites make up only a small fraction of urinary excretion. Thus, extensive enterohepatic recirculation in the rat increases internal exposure to the active metabolite (free MBP).
Insights provided by the model into the processes driving MBP kinetics (enterohepatic recirculation, glucuronidation) are not only helpful to understanding DBP dose–response, but may also help to improve accuracy of model extrapolation to other chemicals and other species through the design of targeted experiments. The successful description of UDPGT-dependent MBP kinetics across life-stages based on in vitro data bodes well for further applications of this model. A similar approach may be useful with other phthalates (i.e., DEHP) in rats and, more importantly, in other species. In vitro metabolism studies require less time, money, and animals than traditional in vivo studies and are much more feasible in the human than in vivo dosing.
Inclusion of several possible biomarkers (plasma, urine, amniotic fluid MBP) in the model allows evaluation of their usefulness as fetal dose surrogates. Failure of the model to describe amniotic fluid metabolites across doses suggests that more complex processes are involved in uptake and clearance of MBP and MBP-G in the amniotic fluid and fetus than simple passive diffusion. In fact, the tendency to concentrate MBP-G speaks to potentially limited clearance. A better understanding of amniotic transport and/or metabolism is needed before it could be used with confidence as a surrogate for fetal dose. With the current model, however, it is possible to determine fetal MBP exposure from a variety of surrogates, including external dose, and maternal plasma and urine. Thus, this model provides a means of estimating fetal dose from easily obtained biological samples.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
This paper illustrates the extension of a model for DBP in the adult male rat to describe gestation through the inclusion of maternal and fetal growth and a simple description of placental transport, made possible by newly generated data. By incorporating important biochemical processes, it was possible to test how the changes in metabolism during pregnancy may affect fetal exposure. The use of realistic, data-validated kinetic and physiological parameters also reduces uncertainty in use of the model to predict target tissue exposure across doses, routes, developmental stages and even species. This validated model can now serve as a basis for extrapolation across species, life-stages, and exposure scenarios (acute v. repeated dosing) to provide quantitative measures of target tissue dosimetry in the population of interest.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Supplementary data are available online at http://toxsci.oxfordjournals.org/.
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
American Chemistry Council funded the project as part of the Long Range Research Initiative.
|
NOTES |
|---|
2 Current address: Boston Scientific Corporation, Maple Grove, MN 55311.
3 Current address: Integrated Laboratory Systems, Inc., Research Triangle Park, NC 27707.
|
ACKNOWLEDGMENTS |
|---|
We would like to thank Dr Deborah Keys for supplying her model code and preliminary discussions on previous modeling efforts. Model code will be made available upon request.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Akingbemi BT, Ge R, Klinefelter GR, Zirkin BR, Hardy MP. Phthalate-induced Leydig cell hyperplasia is associated with multiple endocrine disturbances. Proc. Natl. Acad. Sci. U. S. A. (2004) 101:775–780.
Albro PW, Moore B. Identification of the metabolites of simple phthalate diesters in rat urine. J. Chromatogr. (1974) 94:209–218.[CrossRef][Web of Science][Medline]
Alcorn J, Elbarbry FA, Allouh MZ, McNamara PJ. Evaluation of the assumptions of an ontogeny model of rat hepatic cytochrome P450 activity. Drug Metab. Dispos. (2007) 35:2225–2231.
Altman PL, Dittmer DS. (1971) Volume of blood in tissue: Vertebrates. Ch. 148. In Respiration and Circulation, pp. 383–387. Federation of American Societies for Experimental Biology, Bethesda, MD.
Anderson WAC, Castle L, Scotter MJ, Massey RC, Springall C. A biomarker approach to measuring human dietary exposure to certain phthalate diesters. Food Addit. Contam. (2001) 18:1068–1074.[CrossRef][Web of Science][Medline]
ATSDR. Agency for Toxic Disease Registry. In: Toxicological Profile for Di-n-butyl Phthalate (2001) Altlanta, GA: ATSDR.
Brock JW, Caudill SP, Silva MJ, Needham LL, Hilborn ED. Phthalate monoesters levels in the urine of young children. Bull. Environ. Contam. Toxicol. (2002) 68:309–314.[CrossRef][Web of Science][Medline]
Brown RP, Delp MD, Lindstedt SL, Rhomberg LR, Beliles RP. Physiological parameter values for physiologically based pharmacokinetic models. Toxicol. Ind. Health. (1997) 13:407–484.
Buelke-Sam J, Holson JF, Nelson CJ. Blood flow during pregnancy in the rat: II. Dynamics of and litter variability in uterine flow. Teratology (1982) 26:279–288.[CrossRef][Web of Science][Medline]
Calafat AM, Brock JW, Silva MJ, Gray LE Jr., Reidy JA, Barr DB, Needham LL. Urinary and amniotic fluid levels of phthalate monoesters in rats after the oral administration of di(2-ethylhexyl) phthalate and di-n-butyl phthalate. Toxicology (2006) 217:22–30.[CrossRef][Web of Science][Medline]
Clewell RA, Merrill EA, Yu KO, Mahle DA, Sterner TR, Mattie DR, Robinson PJ, Fisher JW, Gearhart JM. Predicting fetal perchlorate dose and inhibition of iodide kinetics during gestation: A physiologically-based pharmacokinetic analysis of perchlorate and iodide kinetics in the rat. Toxicol. Sci. (2003) 73:235–255.
Dedrick RL. Animal scale–Up. J. Pharmacokinet. Biopharm. (1973) 1:435–461.[CrossRef][Web of Science][Medline]
Dickinson RG, Fowler DW, Kluck RM. Maternofetal transfer of phenytoin, p-hydroxy-phenytoin and p-hydroxy-phenytoin-glucuronide in the perfused human placenta. Clin. Exp. Pharmacol. Physiol. (1989) 16:789–797.[CrossRef][Web of Science][Medline]
Ema M, Miyawaki E, Kawashima K. Further evaluation of developmental toxicity of di-n-butyl phthalate following administration during late pregnancy in rats. Toxicol. Lett. (1998) 98:87–93.[CrossRef][Web of Science][Medline]
Ema M, Miyawaki E, Kawashima K. Critical period for adverse effects on development of reproductive system in male offspring of rats given di-n-butyl phthalate during late pregnancy. Toxicol. Lett. (2000) 111:271–278.[CrossRef][Web of Science][Medline]
Fennell TR, Krol WL, Sumner SC, Snyder RW. Pharmacokinetics of dibutylphthalate in pregnant rats. Toxicol. Sci. (2004) 82:407–418.
Fisher JS, Macpherson S, Marchetti N, Sharpe RM. Human ‘testicular dysgenesis syndrome’: A possible model using in-utero exposure of the rat to dibutyl phthalate. Hum. Reprod. (2003) 18:1383–1394.
Fowler DW, Eadie MJ, Dickinson RG. Transplacental transfer and biotransformation studies of valproic acid and its glucuronide(s) in the perfused human placenta. J. Pharmacol. Exp. Ther. (1988) 249:318–323.[Web of Science]
Gentry PR, Covington TR, Clewell HJ 3rd. Evaluation of the potential impact of pharmacokinetic differences on tissue dosimetry in offspring during pregnancy and lactation. Regul. Toxicol. Pharmacol. (2003) 38:1–16.[CrossRef][Web of Science][Medline]
Gray LE Jr, Ostby J, Furr J, Price M, Veeramachaneni DN, Parks L. Perinatal exposure to the phthalates DEHP, BBP, and DINP, but not DEP, DMP, or DOTP, alters sexual differentiation of the male rat. Toxicol. Sci. (2000) 58:350–365.
Gray LE Jr, Wolf C, Lambright C, Mann P, Price M, Cooper RL, Ostby J. Administration of potentially antiandrogenic pesticides (procymidone, linuron, iprodione, chlozolinate, p,p'-DDE, and ketoconazole) and toxic substances (dibutyl- and diethylhexyl phthalate, PCB 169, and ethane dimethane sulphonate) during sexual differentiation produces diverse profiles of reproductive malformations in the male rat. Toxicol. Ind. Health. (1999) 15:94–118.
Hanwell A, Linzell JL. The time course of cardiovascular changes in the rat. J. Physiol. (1973) 233:99–109.
Heudorf U, Mersch-Sundermann V, Angerer J. Phthalates: Toxicology and exposure. Int. J. Hyg. Environ. Health. (2007) 210:623–634.[CrossRef][Web of Science][Medline]
Isaksson A, Hultberg B, Bergenfeldt M. Lysosomal enzymes in pregnant and steroid treated rats. Horm. Metab. Res. (1988) 20:274–277.[Web of Science][Medline]
Keys DA, Wallace DG, Kepler TB, Conolly RB. Quantitative evaluation of alternative mechanisms of blood disposition of di(n-butyl) phthalate and mono(n-butyl) phthalate in rats. Toxicol. Sci. (2000) 53:173–184.
Kim HS, Kim TS, Shin JH, Moon HJ, Kang IH, Kim IY, Oh JY, Han SY. Neonatal exposure to di(n-butyl) phthalate (DBP) alters male reproductive-tract development. J. Toxicol. Environ. Health A (2004) 67:2045–2060.[CrossRef][Web of Science][Medline]
Kim HS, Saito K, Ishizuka M, Kazusaka A, Fujita S. Short period exposure to di-(2-ethylhexyl) phthalate regulates testosterone metabolism in testis of prepubertal rats. Arch. Toxicol. (2003) 77:446–451.[CrossRef][Web of Science][Medline]
Knight CH, Peaker M. Mammary cell proliferation in mice during pregnancy and lactation in relation to milk yield. Q. J. Exp. Physiol. (1982) 67:165–177.
Koch HM, Bolt HM, Preuss R, Angerer J. New metabolites of di(2-ethylhexyl) phthalate (DEHP) in human urine and plasma after single oral doses of deuterium-labelled DEHP. Arch. Toxicol. (2005) 79:367–376.[CrossRef][Web of Science][Medline]
Kohn MC, Parham F, Masten SA, Portier CJ, Shelby MD, Brock JW, Needham LL. Human exposure estimates for phthalates. Environ. Health Perspect. (2000) 108:A440–A442.[Web of Science][Medline]
Kremer JJ, Williams CC, Parkinson HD, Borghoff SJ. Pharmacokinetics of monobutylphthalate, the active metabolite of di-n-butylphthalate, in pregnant rats. Toxicol. Lett. (2005) 159:144–153.[CrossRef][Web of Science][Medline]
LaBorde JB, Hansen DK, Young JF, Sheehan DM, Holson RR. Prenatal dexamethasone exposure in rats: Effects of dose, age at exposure, and drug-induced hypophagia on malformations and fetal organ weights. Fundam. Appl. Toxicol. (1992) 19:545–554.[CrossRef][Web of Science][Medline]
Latini G, De Felice C, Presta G, Del Vecchio A, Paris I, Ruggieri F, Mazzeo P. In utero exposure to di-(2-ethylhexyl) phthalate and duration of human pregnancy. Environ. Health Perspect. (2003) 111:1783–1785.[Web of Science][Medline]
Lehmann KP, Phillips S, Sar M, Foster PM, Gaido KW. Dose-dependent alterations in gene expression and testosterone synthesis in the fetal testes of male rats exposed to di (n-butyl) phthalate. Toxicol. Sci. (2004) 81:60–68.
Lucier GW, McDaniel OS. Steroid and non-steroid UDP glucuronyltransferase: Glucuronidation of synthetic estrogens as steroids. J. Steroid Biochem. (1977) 8:867–872.[CrossRef][Web of Science][Medline]
Lucier GW, Sonawane BR, McDaniel OS, Hook GE. Postnatal stimulation of hepatic microsomal enzymes following administration of TCDD to pregnant rats. Chem. Biol. Interact. (1975) 11:15–26.[CrossRef][Web of Science][Medline]
Luquita MG, Catania VA, Pozzi EJ, Veggi LM, Hoffman T, Pellegrino JM, Ikushiro SI, Emi Y, Iyanagi T, Vore M, et al. Molecular basis of perinatal changes in UDP-glucuronosyltransferase activity in maternal rat liver. J. Pharmacol. Exp. Ther. (2001) 298:49–56.
Mylchreest E, Cattley RC, Foster PMD. Male reproductive tract malformations in rats following gestational and lactational exposure to di(n-butyl) phthalate: An antiandrogenic mechanism? Toxicol. Sci. (1998) 43:47–60.
Mylchreest E, Sar M, Cattley RC, Foster PMD. Disruption of androgen-regulated male reproductive development by di(n-butyl) phthalate during late gestation in rats is different from flutamide. Toxicol. Appl. Pharmacol. (1999) 156:81–95.[CrossRef][Web of Science][Medline]
Mylchreest E, Sar M, Wallace DG, Foster PM. Fetal testosterone insufficiency and abnormal proliferation of Leydig cells and gonocytes in rats exposed to di(n-butyl) phthalate. Reprod. Toxicol. (2002) 16:19–28.[CrossRef][Web of Science][Medline]
Mylchreest E, Wallace DG, Cattley RC, Foster PMD. Dose-dependent alterations in androgen-regulated male reproductive development in rats exposed to di(n-butyl) phthalate during late gestation. Toxicol. Sci. (2000) 55:143–151.
Naessany S, Picon R. Onset of a feedback inhibition by testosterone in male rat fetuses. Biol. Neonate. (1982) 41:234–239.[Web of Science][Medline]
Naismith DJ, Richardson DP, Pritchard AE. The utilization of protein and energy during lactation in the rat, with particular regard to the use of fat accumulated during pregnancy. Br. J. Nutr. (1982) 48:433–441.[CrossRef][Web of Science][Medline]
NIEHS. National Toxicology Program (NTP). In: Prestart toxicokinetic study report: Di-n-butyl phthalate in rodent plasma, June 30, 1994 (1994) Research Triangle Park, NC: NIEHS. NIH Contract NO: N01-ES-15307. NTP Chem Task No: HEM00324.
NIEHS. (1995) National Toxicology Program (NTP). toxicokinetic study report: The toxicokinetics and metabolism of di-n-butyl phthalate March 29, 1995.NIEHS, Research Triangle Park, NC. NIH Contract NO: N01-ES-15307. NTP Chem Task No: CHEM00324.
O'Flaherty EJ, Scott W, Schreiner C, Beliles RP. A physiologically based kinetic model of rat and mouse gestation: Disposition of a weak acid. Toxicol. Appl. Pharmacol. (1992) 112:245–256.[CrossRef][Web of Science][Medline]
Oishi S, Hiraga K. Effects of phthalic acid monoesters on mouse testes. Toxicol. Lett. (1980) 6:239–242.[CrossRef][Web of Science][Medline]
Park HW, Shepard TH. Volume and glucose concentration of rat amniotic fluid: Effects on embryo nutrition and axis rotation. Teratology (1994) 49:465–469.[CrossRef][Web of Science][Medline]
Parks LG, Ostby JS, Lambright CR, Abbott BD, Klinefelter GR, Barlow NJ, Gray LE Jr. The plasticizer diethylhexyl phthalate induces malformations by decreasing fetal testosterone synthesis during sexual differentiation in the male rat. Toxicol. Sci. (2000) 58:339–349.
Payan JP, Marty JP, Fabry JP, Beydon D, Boudry I, Ferrari E, Canel F, Grandclaude MC, Vincent CM. In vivo and in vitro percutaneous absorption of [(14)C]di-n-butylphthalate in rat. Drug Metab. Dispos. (2001) 29:843–854.
Reynolds F, Knott C. Pharmacokinetics in pregnancy and placental drug transfer. Oxf. Rev. Reprod. Biol. (1989) 11:389–449.[Medline]
Rowland IR, Cottrell RC, Phillips JC. Hydrolysis of phthalate esters by the gastro-intestinal contents of the rat. Food Cosmet. Toxicol. (1977) 15:17–21.[CrossRef][Web of Science][Medline]
Saillenfait AM, Payan JP, Fabry JP, Beydon D, Langonne I, Gallissot F, Sabate JP. Assessment of the developmental toxicity, metabolism, and placental transfer of Di-n-butyl phthalate administered to pregnant rats. Toxicol. Sci. (1998) 45:212–224.
Silva MJ, Barr DB, Reidy JA, Malek NA, Hodge CC, Caudill SP, Brock JW, Needham LL, Calafat AM. Urinary levels of seven phthalate metabolites in the U.S. population from the National Health and Nutrition Examination Survey (NHANES) 1999-2000. Environ. Health Perspect. (2004) 112:331–338.[Web of Science][Medline]
Sjoberg P, Egestad B, Klasson-Wehler E, Gustafsson J. Glucuronidation of mono(2-ethylhexyl)phthalate. Some enzyme characteristics and inhibition by bilirubin. Biochem. Pharmacol. (1999) 41:1493–1496.[CrossRef]
Soucy NV, Parkinson HD, Sochaski MA, Borghoff SJ. Kinetics of genistein and its conjugated metabolites in pregnant Sprague-Dawley rats following single and repeated genistein administration. Toxicol. Sci. (2006) 90:230–240.
Tanaka A, Matsumoto A, Yamaha T. Biochemical studies on phthalic esters. III. Metabolism of dibutyl phthalate (DBP) in animals. Toxicology (1978) 9:109–123.[CrossRef][Web of Science][Medline]
Thompson RC, Hollis OL. Irradiation of the gastrointestinal tract of the rat by ingested ruthenium-96. Am. J. Physiol. (1958) 194:308–312.
White RD, Carter DE, Earnest D, Mueller J. Absorption and metabolism of three phthalate diesters by rat small intestine. Food Cosmet. Toxicol. (1980) 18:383–386.[CrossRef][Web of Science][Medline]
Williams DT, Blanchfield BJ. The retention, distribution, excretion, and metabolism of dibutyl phthalate-7-14 C in the rat. J. Agric. Food Chem. (1975) 23:854–858.[CrossRef][Web of Science][Medline]
Wishart GJ. Functional heterogeneity of UDP-glucuronosyltransferase as indicated by its differential development and inducibility by glucocorticoids. Demonstration of two groups within the enzyme's activity towards twelve substrates. Biochem. J. (1978) 174:485–489.[Web of Science][Medline]
Wormuth M, Scheringer M, Vollenweider M, Hungerbuhler K. What are the sources of exposure to eight frequently used phthalic acid esters in Europeans? Risk Anal. (2006) 26:803–824.[CrossRef][Web of Science][Medline]
Wykoff MH. Weight changes of the developing rat conceptus. Am. J. Vet. Res. (1971) 32:1633–1635.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Sun and K. S. Pang Disparity in Intestine Disposition between Formed and Preformed Metabolites and Implications: A Theoretical Study Drug Metab. Dispos., January 1, 2009; 37(1): 187 - 202. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||