Comparison of Pharmacokinetic Interactions and Physiologically Based Pharmacokinetic Modeling of PCB 153 and PCB 126 in Nonpregnant Mice, Lactating Mice, and Suckling Pups

Sun Ku Lee, Ying C. Ou,1 and Raymond S. H. Yang,2

Quantitative and Computational Toxicology Group, Center for Environmental Toxicology and Technology, Department of Environmental and Radiological Health Sciences, Colorado State University, Fort Collins, Colorado 80523

Received June 20, 2001; accepted October 4, 2001


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Polychlorinated biphenyls (PCBs) are ubiquitous environmental contaminants that can induce neurological defects in infants and children via placental and lactational transfer. To investigate the lactational transfer of PCBs and compare pharmacokinetic interactions among nonpregnant, lactating mice and suckling pups, quantitative time-course measurements of PCB accumulation in tissues were performed. On postnatal day 1, nonpregnant and lactating C57BL/6 mice were exposed to PCB 153 (2,2',4,4',5,5'-hexachlorobiphenyl, 20 mg/kg) alone or a mixture of PCB 153 (20 mg/kg) and PCB 126 (3,3',4,4',5-pentachlorobiphenyl, 0.2 mg/kg) by oral gavage. At 1, 3, 6, and 13 days after treatment, PCB 153 and PCB 126 were determined in nonpregnant and maternal tissues as well as in neonatal tissues by gas chromatography (GC). Coadministration of PCB 153 and PCB 126 increased PCB 153 retention in the liver and decreased PCB 153 accumulation in the fat of nonpregnant mice. Lactational transfer was confirmed to be an efficient elimination mechanism for the lactating mice but a major source of exposure in the pups. However, little or no significant pharmacokinetic interactions were observed in lactating mice and suckling pups. To describe pharmacokinetic interactions between PCB 153 and PCB 126, a physiologically based pharmacokinetic model for PCB 153 disposition was developed. The effects of PCB 126 on the fat content in liver and a diffusion permeation constant in fat were incorporated into the physiologically based pharmacokinetic (PBPK) model. This model successfully describes PCB 153 disposition altered by PCB 126 in nonpregnant mice.

Key Words: PCB; lactational transfer; pharmacokinetic interactions; mixture; PBPK modeling; fat content; diffusion permeation constant.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Polychlorinated biphenyls (PCBs) are ubiquitous environmental contaminants causing a variety of toxicities in humans and animals (ATSDR, 1999Go). They are detected in numerous and diverse biological compartments including human serums and have become quantitatively the most important organochlorine residues in the environment due to their lipophilic character and their slow metabolism (Carpenter, 1998Go; Muhlebach et al., 1991Go). PCBs usually exist in a mixture form, and can comprise up to 209 different congeners (ATSDR, 1999Go). Although PCBs have been reported to induce a variety of toxic effects including immunologic, teratogenic, reproductive, carcinogenic, and neurological effects in both human and animal studies (Safe, 1989Go), the developing brain is particularly susceptible to PCB exposure (Schantz et al., 1996Go). PCBs have been reported to induce neurobehavioral deficits in children born at contaminated sites (Carpenter, 1998Go). Many of the defects were related to the nervous system, including abnormalities on behavioral assessment and increased activity, greater incidence of behavioral problems, lower IQ, and impaired visual recognition (Schantz, 1996Go). These neurological defects are primarily due to the exposure of the fetus through placental and/or lactational transfer (Buck, 1996Go; Seegal, 1996Go).

A major molecular structural factor in the determination of toxic properties and potencies in PCBs is the presence of chlorine atom on the ortho positions, influencing propensity to adapt planar conformation (Safe, 1994Go). Some non-ortho or mono-ortho chlorine-substituted PCBs have coplanar structures and high affinity for the Ah receptor. Their toxic effects are mediated by the Ah receptor, and their toxic potencies are determined by the affinity for the Ah receptor. In contrast, the di to tetra ortho-chlorinated PCBs have nonplanar structures and no measurable affinity for the Ah receptor. Their mechanisms of action are not completely understood (Kodavanti and Tilson, 2000Go; van der Burght et al., 1999Go). While both types of PCB congeners have been reported to cause neurodevelopmental deficits following in utero and/or lactational exposure (Hussain et al., 2000Go; Saghir et al., 2000Go), the mechanisms of action between planar and nonplanar PCBs differ significantly (Humphrey et al., 2000Go) and interactive effects between 2 types of PCBs are observed (van Birgelen et al., 1996Go; van der Plas et al., 1988Go).

The developing brain is different from the adult brain in both composition and function (Kalil et al., 2000Go), and the vulnerability of the developing brain to toxicants is critically dependent on timing and duration of exposure. Exposure to environmental toxicants coincident with the ontogeny is more likely to cause toxicities if they interfere with developmental processes within the critical period (Rice and Barone, 2000Go). Therefore, the characterization of exact pharmacokinetic profiles of environmental toxicants during the critical period of development is important both for understanding mechanisms of developmental neurotoxicities and in facilitating interpretation of the results from animal studies.

PCB 153 (2, 4,5,2',4',5',-hexachlorobiphenyl) and PCB 126 (3,4,5,3',4',-pentachlorobiphenyl) were selected as a prototype of PCB mixture to understand the potentially complex pharmacokinetic interactions associated with PCB mixtures. These congeners are the ones that appear most prevalently both in the environment and in human serum (Humphrey et al., 2000Go). PCB 153 is a representative nonplanar congener that appears in the environment and mammalian tissues at the highest concentration (Muhlebach et al., 1991Go). PCB 126 is the most toxic PCB congener with coplanar structure (Safe, 1994Go) and binds to the Ah receptor with highest affinity. Both congeners have been suggested to induce neurobehavioral deficits via gestational and lactational transfer (Hussain et al., 2000Go). Pharmacokinetic interactions between coplanar PCBs or TCDD and nonplanar PCBs were previously reported in adult rat and mice (van Birgelen et al., 1996Go; van der Plas et al., 1988Go). However, no studies investigating pharmacokinetic interactions between these 2 congeners during the critical developmental period have been reported despite the potential importance of these interactions. In reality, PCBs are available to human beings as mixtures. Accordingly, the pharmacokinetic information on exposure to a single congener is not enough to describe pharmacokinetic profiles during the developmental period.

The objectives of our study were 3-fold. First, we wanted to investigate whether the coexposure to 2 different PCB congeners with distinct toxic mechanisms of action (PCB 153 and PCB 126) could change the pharmacokinetic profiles in mice. Second, we wanted to compare pharmacokinetic interactions among nonpregnant mice, lactating mice, and suckling pups because PCB congeners could be transferred to the fetus and suckling mice from their mothers (Orberg and Ingvast, 1977Go) and the extent of transfer was greater in lactation than through placenta (Vodicnik and Lech, 1980Go). Finally, physiologically based pharmacokinetic (PBPK) modeling, a simulation technique useful for species extrapolation and quantitative risk assessment (Dedrick and Bischoff, 1969Go), was used as a hypothesis-testing tool for deriving mechanistic insight for the observed pharmacokinetic data.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Chemicals.
PCB 153 and PCB 74 (2,4,4',5–tetrachlorobiphenyl) were purchased from Ultra Scientific (North Kingtown, RI). PCB 126 was obtained from Accustandard (New Haven, CT). The purities of all congeners used were over 98%, which was confirmed by both vendors. Pentane (HPLC grade), sea sand, and diethylether (> 99% purity) were purchased from VWR Scientific (Denver, CO). Sodium anhydrous sulfate and Florisil® (pesticide residue grade, 60–100 mesh) were purchased from Alltech Associates (Deerfield, IL).

Animals and treatment.
C57BL/6 nonpregnant and pregnant female mice were purchased from Harlan Sprague Dawley Laboratory (Indianopolis, IN) and housed individually in cages at Painter Center, Colorado State University, which is fully accredited by the American Association for Accreditation of Laboratory Animal Care (AAALAC). Eighteen nonpregnant mice and 24 pregnant mice were used in this study. The mice were maintained on a 12-h light/dark cycle at a constant temperature of 25°C and humidity of 55%. Diet (certified Teklad NIH-07 rodent diet) and tap water were provided ad libitum. The average number of pups from pregnant mice was 6. Because of the low fertility rate, we did not cull the litters and used all the pups.

On postnatal day (PND) 1, 12 lactating mice were exposed to PCB 153 alone (20 mg/kg bw) and the other 12 mice were exposed to a mixture of PCB 153 (20 mg/kg bw) and PCB 126 (0.2 mg/kg bw) through oral gavage. Corn oil was used as vehicle to dissolve both congeners. The volume of oral gavage was 5 ml/kg. The dosing time was between 0800 and 0900 h. At 1, 3, 6, and 13 days after oral gavage, 3 lactating mice and their pups in each group were anesthetized using isofluorane or CO2 gas and then euthanized. The tissues of lactating mice (liver, fat, skin, and brain) and pups (liver, brain, and the remaining carcass excluding GI tract) were collected from each animal. Nonpregnant mice were treated the same way as lactating mice and their tissues were collected at 1, 6, and 13 days after treatment. At each time point, 3 mice per group were sacrificed. All samples were frozen with liquid nitrogen and stored at -70°C until analysis.

Extraction.
Liver, fat, and brain samples were weighed (approximately 0.5 g/sample) and then transferred to stainless steel beakers. Sea sand and sodium sulfate anhydrous mixture (5 g/10 g) were added to the samples and then ground using a glass stirring rod until granular dry mass was obtained. PCB 74 (200 ng) was added to each of the ground samples as an internal standard. This congener has similar physicochemical characteristics but does not interfere with PCB 153 and PCB 126 in gas chromatography (GC) analyses. Twenty ml of pentane was added to each sample, which was then boiled on a hot plate with stirring. The boiling temperature was between 30°C and 40°C. The extracts were transferred to the filters and collected in glass tubes. For the extraction of the carcass from pups and skin tissues of adult mice, 10 ml of 50% sulfuric acid was added to each sample. After homogenization, 10 ml of pentane was added and vigorously shaken. When the layers separated, the upper layer was transferred to glass tubes. These processes were repeated 3 times. Collected extracts were concentrated to 2 ml with a nitrogen evaporator.

Clean up.
Before clean up, the adsorbents (sodium sulfate anhydrous and Florisil®) were layered (the ratio of sodium sulfate to Florisil® is 1 to 2) and packed into a Florisil® glass column (Allen Scientific Glass, Denver, CO) and then the column was rinsed with 100 ml of pentane. The concentrated extract was transferred to the column. The column was eluted with 100 ml of pentane/diethylether mixture (6% diethylether in pentane). The elution rate was adjusted to 20 drops per 10 s. Finally, the collected solution was concentrated to 1.0 ml with a nitrogen evaporator. The recovery rate of extraction is about 60%.

Gas chromatographic analyses.
A HP-5890 Series II Plus gas chromatograph (Hewlett Packard, Wilmington, DE) with ECD detector was used to determine PCB amount. The analysis condition was based on a previous method (Mills et al., 1963Go). Briefly, DB-5 (crosslinked 5% phenylmethylsilicone, 30 m x 0.32 mm, Agilent Technologies, Palo Alto, CA) capillary column was used. The initial temperature was 60°C, programmed to 200°C at 30°C/min for 6 min, subsequently to 250°C at 3°C/min for 15 min. The flow of carrier gas, helium, was 5 ml/min. The make-up gas was pure nitrogen at a flow rate of 70 ml/min. The injector temperature was 225°C and the detector temperature was 320°C. The volume of injection was 2 µl per sample.

The samples were quantitated using internal standard method as previously reported (Storr-Hansen, 1991Go). Standard solutions for PCB 153 and PCB 126 were prepared in the range from 5 ng/ml to 500 ng/ml. To each solution, PCB 74 (200 ng/ml) was added as an internal standard. Calibration curve was created and fitted using quadratic regression equation. The detection limit of GC was 5 ng/ml.

Statistics.
Differences of tissue concentration between samples from various treatment groups and time points were tested for significance by two-way ANOVA, followed by Fisher's multiple comparison test. All analyses were performed with the statistical software, Minitab (p < 0.01; Windows version 12.0).

Construction of an interaction PBPK model.
To explain pharmacokinetic interactions between PCB congeners, a PBPK model for the combination effects of PCB 153 and PCB 126 was developed. The previous PBPK model for PCBs (Lutz et al., 1977Go) was composed of 5 lumped compartments and successfully described individual PCB disposition including PCB 153 using flow-limited transfer. However, this model could not explain the change of PCB 153 disposition by coexposure to PCB 126.

In order to describe pharmacokinetic changes of PCB 153 upon coexposure to PCB 126, the PBPK model of Lutz et al. was modified as follows:

First, we incorporated time-dependent increase of partition coefficient in the liver. This was supported by several experimental observations. It has been shown that some coplanar PCBs (e.g., PCB 126) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) induce fat accumulation in the liver in a time-dependent manner (Koga et al., 1990Go). Furthermore, PCB 126 could increase lipid content by 50% in murine liver 7 days after oral exposure (Nobuyuki et al., 1990). In general, the partition coefficient in tissues depends on the n-octanol/water coefficient, Kow, and profiles for water and lipid contents in each tissue (De Jongh et al., 1997Go). Since PCB 153 is a lipophilic compound, its partition coefficient in tissues is highly proportional to lipid content in tissues. Therefore, the partition coefficient in the liver (PL) was modified to a time-dependent equation, PL = 10 + {alpha} x TIME where {alpha} is the coefficient for time-dependent increase of partition coefficient and TIME is the minutes after PCB 153 exposure.

Second, the fat compartment was described as diffusion-limited, and a diffusion permeation constant (PAFC) from fat-blood to fat tissues was added. It has been suggested that PCB 153 circulates in the body in association with lipoproteins. To be absorbed in fat tissues, PCB 153 was suggested to be dissociated with lipoproteins by lipoprotein lipases (Noren et al., 1999Go). Because TCDD and coplanar PCBs can inhibit lipoprotein lipase in the fat via Ah receptor activation (Olsen et al., 1998Go), it is likely that a TCDD-like PCB congener such as PCB 126 can also inhibit lipoprotein lipase, thus decreasing the uptake of PCB 153 in the fat. Both in vivo and in vitro studies showed that lipoprotein lipase in adipocytes was inhibited as soon as 1 h after TCDD administration (Olsen et al., 1998Go). Accordingly, diffusion permeation constant from fat-blood to fat tissues (PAFC) was assumed to be decreased by coexposure to PCB 126 and PCB 153.

Third, brain compartment was added as a major target organ because PCBs can induce neurotoxicities in both adult and young mice.

A schematic diagram for this PBPK model is presented in Figure 1Go. Other physiological and biochemical parameters were obtained from a report by Brown et al. (1997) and a previous article (Lutz et al., 1977Go) and are shown in Table 1Go.



View larger version (27K):
[in this window]
[in a new window]
 
FIG. 1. Schematic diagram of the physiologically based pharmacokinetic model used to describe pharmacokinetic interaction on PCB 153 disposition in the nonpregnant mice. All compartments except fat are flow-limited. Fat tissue is described as a diffusion-limited compartment. The abbreviations are given in Table 1Go.

 

View this table:
[in this window]
[in a new window]
 
TABLE 1 PBPK Model Parameters
 
All PBPK model construction, simulation, and parameter estimations were performed using the Berkeley Madonna software package (version 8.01 for Windows, Kagi Shareware, Berkeley, CA).


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Effect of Coexposure to PCB 126 and PCB 153 on Body Weight Gain in Growing Pups
PCB 126 has been previously reported to suppress normal weight gain in growing rats (Van Birgelen et al., 1994Go), so the body weights of nonpregnant mice, lactating mice, and suckling pups were monitored after treatment. At the dosing regimen applied, no differences in body weights were observed between the group exposed to PCB 153 alone and the group coexposed to PCB 153 and PCB 126 (data not shown).

Baseline Pharmacokinetic Interactions of Coexposure to PCB 126 and PCB 153: Distribution of PCB 153 in the Tissues of Nonpregnant Mice
Pharmacokinetic interactions indeed occurred in nonpregnant mice when PCB 153 and PCB 126 were given to them as a mixture. The most significant findings are: (1) time-dependent retention of PCB 153 in the liver when coexposed to PCB 126, and (2) a reduction in the rate of PCB 153 accumulation in the fat when coexposed to PCB 126. The details in relation to figures and tables are given below.

Figure 2AGo shows the change of PCB 153 distribution in the liver between the group exposed to PCB 153 alone and the group coexposed to PCB 153 and PCB 126. At 1 day after treatment, there were no differences in the concentration of PCB 153 between the 2 groups. However, the concentration of PCB 153 in the liver of animals exposed to both congeners was more than 2 times higher than that in the liver of mice exposed to PCB 153 alone at 13 days after treatment.



View larger version (21K):
[in this window]
[in a new window]
 
FIG. 2. The concentration of PCB 153 in the tissues of nonpregnant mice as a function of time. The nonpregnant C57BL/6 female mice were exposed to PCB 153 (20 mg/kg bw) or PCB 153 (20 mg/kg bw) + PCB 126 (0.2 mg/kg bw) through oral gavage. One, 6, and 13 days after oral gavage, the nonpregnant mice were sacrificed by CO2 and their tissues were collected. Values are expressed as ng PCB 153/g tissue. Data points represent mean ± SEM. **Significantly different from the corresponding group exposed to PCB 153 alone (p < 0.01).

 
Similar results were shown in the change of PCB 153 distribution in the brain between the group exposed to PCB 153 alone and the group coexposed to PCB 153 and PCB 126 (Fig. 2BGo). At 13 days posttreatment, the concentration of PCB 153 in the group coexposed to PCB 153 and PCB 126 was higher than that in the group exposed to PCB 153 alone, but this difference was not statistically significant.

Coexposure of PCB 153 and PCB 126 changed the distribution pattern of PCB 153 in the fat. As shown in Figure 2CGo, the group exposed to PCB 153 alone showed that the concentration of PCB 153 in fat increased rapidly until reaching plateau (6 days after treatment). On the contrary, the concentration of PCB 153 in the group coexposed to PCB 153 and PCB 126 increased slowly during that time, suggesting a reduction of the rate of uptake.

Skin is the major deposit for PCBs (Vodicnik and Lech, 1980Go). Figure 2D Goshows the change of PCB 153 distribution in skin between the group exposed to PCB 153 alone and the group coexposed to PCB 153 and PCB 126. There were no significant differences in PCB 153 concentration between the group exposed to PCB 153 alone and the group coexposed to PCB 153 and PCB 126.

Postnatal Pharmacokinetics: Comparison of PCB 153 Distribution on PND 2 and 14 among Nonpregnant, Maternal, and Pup Tissues
The pharmacokinetic interactions described above for nonpregnant mice after coexposure to PCB 153 and PCB 126 did not occur in lactating mice and suckling pups. However, there was strong evidence to confirm earlier studies that lactation is an efficient mechanism to transfer of PCB 153 to the pups. Consequently, the lactating mice had the lowest PCB 153 concentrations in the liver and brain among 3 groups of mice and the pups had the highest concentrations in the liver in early lactation period.

PCB 153 has been reported to be accumulated mostly in fat and skin tissues and transferred readily from lactating mice to their offspring (Vodicnik and Lech, 1980Go). We investigated differences in lactational transfer between the group exposed to PCB 153 alone and in combination with PCB 126. Also, we attempted to compare pharmacokinetic interactions among nonpregnant mice, lactating mice, and their pups. The tissue concentrations of PCB 153 in mice of different stages were compared at PND 2 and 14 after oral gavage at PND 1. Figure 3 Goshows the differences of PCB 153 concentration in the liver among nonpregnant mice, lactating mice, and their pups. Interestingly, the concentrations of PCB 153 in neonatal livers were much higher than nonpregnant and lactating mice on PND 2. However, PCB 153 in neonatal livers was rapidly decreased; thus, there were no longer any differences in PCB 153 concentrations of livers between nonpregnant mice and suckling pups by PND 14. It seems that rapid growth of neonatal liver could be a confounding factor for this result. However, constant concentration of PCB 153 in neonatal carcass was observed during the 2 weeks after birth, despite the rapid increase of body weight. It suggests that such rapid growth of tissue and body weight may not be a significant confounding factor. Coexposure to PCB 126 increased the retention of PCB 153 in the liver of nonpregnant mice but not in the liver of lactating mice and their pups. Figure 4 Goshows the concentrations of PCB 153 in the brains of nonpregnant mice, lactating mice, and suckling pups. At PND 2, the concentrations of PCB 153 were higher in the brains of nonpregnant mice than those of lactating mice and suckling pups. By PND 14, there was no difference between nonpregnant mice and suckling pups. In contrast, the concentrations of PCB 153 in the brain of lactating mice were significantly lower than those brains of suckling pups and nonpregnant mice at PND 14. Coexposure to PCB 126 did not change the concentration of PCB 153 in the brains of lactating mice and pups. Figure 5 Goshows the difference of PCB 153 concentration in fat tissues between nonpregnant and lactating mice. In nonpregnant mice, PCB 153 concentration increased at PND 14 from at PND 2, whereas, PCB 153 did not change or slightly decreased at PND 14 in lactating mice. Coexposure of PCB 126 did not change the disposition of PCB 153 in fat tissues at PND 14. In both groups, PCB 153 tissue concentration was higher in nonpregnant mice than that in lactating mice at PND 14.



View larger version (28K):
[in this window]
[in a new window]
 
FIG. 3. Comparison of concentrations of PCB 153 in the liver of nonpregnant mice, lactating mice, and suckling pups. On PND 1, the nonpregnant and lactating mice were exposed to PCB 153 (20 mg/kg bw) or PCB 153 (20 mg/kg bw) + PCB 126 (0.2 mg/kg bw) via oral gavage. On PND 2 and PND 14, the nonpregnant mice, lactating mice, and suckling pups were sacrificed and their livers were collected. Bars represent mean and SEM. Statistic differences were evaluated with Tukey's multiple comparison after 2-way ANOVA (a, statistically different from nonpregnant mice exposed to PCB 153 only at same date; b, statistically different from nonpregnant mice exposed to PCB 153 + PCB 126 at same date; c, statistically different from lactating mice exposed to PCB 153 only at same date; d, statistically different from lactating mice exposed to PCB 153 + PCB 126 at same date; p < 0.01).

 


View larger version (29K):
[in this window]
[in a new window]
 
FIG. 4. Comparison of concentrations of PCB 153 in the brain of nonpregnant mice, lactating mice, and suckling pups. The nonpregnant and lactating mice were exposed to PCB 153 (20 mg/kg bw) or PCB 153 (20 mg/kg bw) + PCB 126 (0.2 mg/kg bw) via oral gavage on post natal day 1 (PND 1). On PND 2 and PND 14, nonpregnant, lactating mice and suckling pups were sacrificed and their brains were collected. Bars represent mean and SEM. Statistic differences were evaluated with Tukey's multiple comparison after 2-way ANOVA (a, statistically different from nonpregnant mice exposed to PCB 153 only at same date; b, statistically different from nonpregnant mice exposed to PCB 153 + PCB 126 at same date; c, statistically different from lactating mice exposed to PCB 153 only at same date; d, statistically different from lactating mice exposed to PCB 153 + PCB 126 at same date; p < 0.01).

 


View larger version (34K):
[in this window]
[in a new window]
 
FIG. 5. Comparison of concentrations of PCB 153 in the fat of nonpregnant mice and lactating mice. The nonpregnant and lactating mice were exposed to PCB 153 (20 mg/kg bw) or PCB 153 (20 mg/kg bw) + PCB 126 (0.2 mg/kg bw) via oral gavage on PND 1. On PND 2 and PND 14, the nonpregnant and lactating mice were sacrificed and their fat tissues were collected. Bars represent mean and SEM. Statistic differences were evaluated with Tukey's multiple comparison after 2-way ANOVA (a, statistically different from nonpregnant mice exposed to PCB 153 only at same date; b, statistically different from nonpregnant mice exposed to PCB 153 + PCB 126 at same date, p < 0.01).

 
Time-Course Distribution of PCB 153 in Suckling Pups with or without Coexposure to PCB 126
Figure 6A Goshows the change of PCB 153 distribution in pup liver between the groups exposed to PCB 153 alone and the group coexposed to PCB 153 and PCB 126. In both groups, the concentration of PCB 153 in neonatal liver decreased over time. There were no significant differences in PCB 153 concentration at each time point between 2 groups. This result was different from that of nonpregnant mice, which showed that coexposure of PCB 126 and PCB 153 increased the retention of PCB 153 in the liver (Fig. 2AGo).



View larger version (15K):
[in this window]
[in a new window]
 
FIG. 6. The concentration of PCB 153 in the tissues of suckling pups as a function of time. The lactating mice were exposed to PCB 153 (20 mg/kg bw) or PCB 153 (20 mg/kg bw) and PCB 126 (0.2 mg/kg bw) through oral gavage on PND 1. At each time, the suckling pups were sacrificed by CO2 and their tissues were collected. Values are expressed as the amount of PCB 153 per 1 g of tissue. Data points represent mean ± SEM. PND, postnatal day. (A) Liver, (B) brain, (C) carcass.

 
There were no significant differences in PCB 153 distribution in pup brain between the 2 groups (Fig. 6BGo). The concentration of PCB 153 in the brain of both groups has decreased after treatment.

Fat tissues in neonatal mice were not significant enough to be reliably collected and analyzed. Therefore, PCB 153 was measured in pup carcass without liver, brain, and gastrointestinal (GI) tract. As shown in Figure 6CGo, the concentration of PCB 153 in the carcass increased until 3 days after treatment and then decreased slightly in the group exposed to PCB 153 alone, whereas it remained at almost the same level in the group coexposed to PCB 153 and PCB 126. There were no significant differences in PCB concentration at each time point between the 2 groups.

Mechanistic Consideration for Pharmacokinetic Interactions between PCB Congeners and PBPK Modeling
The preceding results suggested that coexposure to PCB 126 increased the retention of PCB 153 in the liver and decreased the absorption of PCB 153 in the fat of nonpregnant mice, but not in those of lactating mice and nursing pups. As indicated earlier, the mechanistic basis for coexposure to PCB 126-induced pharmacokinetic interactions was hypothesized to be: (1) time-dependent fat accumulation in the liver, and (2) the decrease of PCB 153 uptake in the fat due to the lack of dissociation of PCB 153:lipoprotein complex through the inhibition of lipoprotein lipase. In order to address these mechanistic considerations and quantitatively analyze the effects of PCB 126 on tissue distribution of PCB 153, an interaction PBPK model for PCB 153 was developed.

Through parameter optimization of 2 coefficients ({alpha}, PAFC), the change of pharmacokinetic profiles of PCB 153 was simulated successfully (Fig.7Go). The result of parameter optimization is shown in Table 2Go. Thus, a PBPK model incorporating fat accumulation in the liver and inhibition of lipoprotein lipase activity in the fat adequately described the change of PCB 153 disposition induced by PCB 126.



View larger version (13K):
[in this window]
[in a new window]
 
FIG. 7. Model simulations (solid and dashed line) of PCB 153 in the tissues of nonpregnant mice. Modifications of partition coefficient in the liver and diffusion permeation constant in the fat successfully simulated the change of PCB 153 disposition induced by coexposure of PCB 126 + PCB 153.

 

View this table:
[in this window]
[in a new window]
 
TABLE 2 Modified PBPK Parameters to Fit Pharmacokinetic Changes of PCB 153 Affected by PCB 126
 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Our most significant results can be summarized as follows: First, coexposure to PCB 153 and PCB 126 increased the retention of PCB 153 in the liver and decreased the rate of PCB 153 accumulation in the fat of nonpregnant mice. Second, the incorporation of a time-dependent increase of partition coefficient in the liver and the decrease of diffusion permeation constant in the fat into a PBPK model could adequately account for pharmacokinetic changes of PCB 153 in nonpregnant mice. Third, coexposure to PCB 126 and PCB 153 may not affect the extent of lactational transfer of PCB 153. While lactational transfer is an efficient elimination mechanism for the lactating mice, it is a major source of exposure in the pups. Fourth, interactive effects of PCB 126 on the disposition of PCB 153 are different among nonpregnant mice, lactating mice, and their pups.

The present study demonstrated that coexposure of C57BL/6 female mice to a coplanar PCB congener (PCB 126) together with a nonplanar congener (PCB 153) affects the redistribution of nonplanar congener. In general, PCB 126 is assumed to exert its biological or toxicological actions through an Ah receptor-mediated mechanism (Safe, 1984). TCDD, which is also assumed to bind and transduce biological action via interactions with the Ah receptor, was previously shown to increase the retention of PCB 153 in B6C3F1 mice (van Birgelen et al., 1996Go). Therefore, the effect of PCB 126 on the increase of PCB 153 retention may be due to the action through Ah receptor mediated process. TCDD is also able to induce fat accumulation in the liver (van Birgelen et al., 1996Go). Accordingly, the increase of fat content in the liver may increase PCB 153 retention due to the highly lipophilic properties of PCB 153, which confers the highest affinity for fat among PCB congeners. Recently it was also reported that PCB 126 could reduce essential fatty acid and thus resulted in the increase of lipid content in the liver (Matsusue et al., 1999Go). These studies suggest that TCDD and dioxin-like PCB congeners can increase the retention of nonplanar lipophilic congeners in the liver through the increase of lipid content in liver. In our studies, we adopted this mechanism because it is consistent with our experimental results. Contrarily to our findings, some previous studies showed different results (De Jongh et al., 1993Go). They showed that coexposure of coplanar PCB with nonplanar PCB increased the retention of coplanar PCB in the liver. It seems that the difference is due to the dose of PCBs. The dose of nonplanar PCB they used was 100 mg/kg, which is 5 times higher than the dose we used.

Since PCBs have high lipophilicity and low metabolism, the adipose tissue is the major storage site of most PCB congeners except dioxin-like congeners (Muhlebach et al., 1991Go). However, our results showed that there was a decreased uptake of PCB 153 into the fat of nonpregnant mice coexposed to PCB 153 and PCB 126. One possible mechanism is the decrease of lipoprotein lipase activity in adipose tissue by TCDD or coplanar PCBs. In support of this possibility, Olsen et al. (1998) showed that TCDD and coplanar PCBs induced a statistically significant time- and dose-dependent decrease in lipoprotein lipase activity in preadipocyte cell line and suggested that different PCB and dioxin congeners could reduce lipoprotein lipase activity through the Ah receptor activation. Lipoprotein lipase on the membrane of adipocyte hydrolyzes lipoproteins into free fatty acids and proteins. PCB 153 could bind to lipoproteins in the liver and be transported to other tissues via lipoproteins (Noren et al., 1999Go). In fat tissues, PCB 153 is dissociated from lipoproteins by lipoprotein lipase and released into the fat (Gallenberg and Vodicnik, 1987Go). Therefore, the inhibition of lipoprotein lipase activity by coplanar PCB in the fat could result in the decrease of PCB 153 uptake by adipose tissues.

Through the incorporation of effects of lipid accumulation in the liver and inhibition of lipoprotein lipase in the fat, the observed pharmacokinetic changes of PCB 153 in nonpregnant mice could be successfully simulated using PBPK modeling. It should be noted that the interaction PBPK model in this article is still preliminary in nature. The time-dependent change of partition coefficient for the liver compartment provided an adequate fit to the data collected during the experimental period of 14 days. However, if the experimental period is very long, the changes of partition coefficient in the liver may become unrealistically large. Of course, such long-term accumulation of fat must be verified with experimental results if subsequent PBPK model refinement is to be attempted. Further, as the effects of PCB 126 may be related to receptor binding (i.e., Ah receptor), the concentration of PCB 126 may become an important factor regarding the fat accumulation in the liver in multiple dose level studies. Refinement of this interaction PBPK model is in progress in our laboratory.

PBPK models have been increasingly useful mechanistic tools to describe and estimate the disposition of chemicals in biological systems (Dedrick and Bischoff, 1969Go). The purpose for the construction of the PBPK model in this study is to quantitatively account for observed interactive effects between PCB 153 and PCB 126 at different stages of development. As in any modeling work in toxicology, the ultimate goal is to predict target tissue dosimetry following the establishment of a validated model. In this regard, the PBPK model for lactational transfer of PCB 153 is now being developed in our laboratory to simulate pharmacokinetic differences between lactating mice and nonpregnant mice.


    ACKNOWLEDGMENTS
 
We are grateful to Dr. John Tessari, Mr. Brian Cranmer, and Dr. Ivan Dobrev for their technical assistances in the setup of gas chromatography for the measurement of PCB 153. Dr. Stephen Safe, Texas A&M University, kindly provided samples of PCB 153 and PCB 126. We also appreciate Dr. Melvin E. Andersen's stimulating discussions. This study was supported by NIEHS research grant R03 ES10116-01.


    NOTES
 
Portions of this data were presented at the 40th annual meeting of the Society of Toxicology, March 2001, San Francisco, CA.

1 Present address: Otsuka Maryland Research Institute, 9900 Medical Center Drive, Rockville, MD 20850. Back

2 To whom correspondence should be addressed. Fax: (970) 491-8304. E-mail: raymond.yang{at}colostate.edu. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
ATSDR (1999). Toxicological Profile for Polychlorinated Biphenyls. Agency for Toxic Substances and Disease Registry, U. S. Department of Health and human Services, Public Health Service, Atlanta, GA.

Brown, R. P., Delp, M. D., Lindstedt, S. L., Rhomberg, L. R., and Beliles, R. P. (1997). Physiological parameter values for physiologically based pharmacokinetic models. Toxicol. Ind. Health 13, 407–484.[ISI][Medline]

Buck, G. M. (1996). Epidemiologic perspective of the developmental neurotoxicity of PCBs in humans. Neurotoxicol. Teratol. 18, 239–241, discussion 271–276.[ISI][Medline]

Carpenter, D. O. (1998). Polychlorinated biphenyls and human health. Int. J. Occup. Med. Environ. Health 11, 291–303.[Medline]

Dedrick, R. L., and Bischoff, K. B. (1969). Pharmacokinetics in applications of the artificial kidney. Chem. Eng. Prog. Symp. Ser. 64, 32–44.

De Jongh, J., Verhaar, H. J., and Hermens, J. L. (1997). A quantitative property-property relationship (QPPR) approach to estimate in vitro tissue-blood partition coefficients of organic chemicals in rats and humans. Arch. Toxicol. 72, 17–25.[ISI][Medline]

De Jongh, J., Wondergem, F., Seinen, W., and Van den Berg, M. (1993). Toxicokinetic interactions between chlorinated aromatic hydrocarbons in the liver of the C57Bl/6 mouse: I. Polychlorinated biphenyls (PCBs). Arch. Toxicol. 67, 453–460.[ISI][Medline]

Gallenberg, L. A., and Vodicnik, M. J. (1987). Potential mechanisms for redistribution of polychlorinated biphenyls during pregnancy and lactation. Xenobiotica 17, 299–310.[ISI][Medline]

Humphrey, H., Gardiner, J., Pandya, J., Sweeney, A., Gasior, D., McCaffrey, R., and Schantz, S. (2000). PCB congener profile in the serum of humans consuming Great Lakes fish. Environ. Health Perspect. 108, 167–172.[ISI][Medline]

Hussain, R., Gyori, J., DeCaprio, A., and Carpenter, D. O. (2000). In vivo and in vitro exposure to PCB 153 reduces long-term potentiation. Environ. Health Perspect. 108, 827–831.[ISI][Medline]

Kalil, K., Szebenyi, G., and Dent, E. W. (2000). Common mechanisms underlying growth cone guidance and axon branching. J. Neurobiol. 44, 145–158.[ISI][Medline]

Kodavanti, P. R., and Tilson, H. A. (2000). Neurochemical effects of environmental chemicals: In vitro and in vivo correlations on second messenger pathways. Ann. N.Y. Acad. Sci. 919, 97–105.[Abstract/Free Full Text]

Koga, N., Beppu, M., and Yoshimura, H. (1990). Metabolism in vivo of 3,4,5,3',4'-pentachlorobiphenyl and toxicological assessment of the metabolite in rats. J. Pharmacobiodyn. 13, 497–506.[Medline]

Lutz, R. J., Dedrick, R. L., Matthews, H. B., Eling, T. E., and Anderson, M. W. (1977). A preliminary pharmacokinetic model for several chlorinated biphenyls in the rat. Drug Metab. Dispos. 5, 386–396.[Abstract]

Matsusue, K., Ishii, Y., Ariyoshi, N., and Oguri, K. (1999). A highly toxic coplanar polychlorinated biphenyl compound suppresses Delta5 and Delta6 desaturase activities which play key roles in arachidonic acid synthesis in rat liver. Chem. Res. Toxicol. 12, 1158–1165.[ISI][Medline]

Mills, P. A., Onley, J. H., and Gaither, R. A. (1963). Rapid method for chlorinated pesticide residues in nonfatty foods. J. Assoc. Offic. Agric. Chem. 46, 186–191

Muhlebach, S., Wyss, P., and Bickel M. (1991). The use of 2,4,5,2',4',5'-hexachlorobiphenyl (6-CB) as an unmetabolizable lipophilic model compound. Pharmacol. Toxicol. 69, 410–415.[ISI][Medline]

Noren, K., Weistrand, C., and Karpe, F. (1999). Distribution of PCB congeners, DDE, hexachlorobenzene, and methylsulfonyl metabolites of PCB and DDE among various fractions of human blood plasma. Arch. Environ. Contam. Toxicol. 37, 408–414.[ISI][Medline]

Olsen, H., Enan, E., and Matsumura, F. (1998). 2,3,7,8-Tetrachlorodibenzo-p-dioxin mechanism of action to reduce lipoprotein lipase activity in the 3T3–L1 preadipocyte cell line. J. Biochem. Mol. Toxicol. 12, 29–39.[Medline]

Orberg, J., and Ingvast, C. (1977). Effects of pure chlorobiphenyls (2,4',5-trichlorobiphenyl or 2,2',4,4',5,5'-hexachlorobiphenyl) on the disappearance of 14C from the blood plasma after intravenous injection of 4-14C-progesterone and on the hepatic drug metabolizing system in the female rat. Acta Pharmacol. Toxicol. (Copenh.) 41, 11–17.[Medline]

Rice, D. C., and Barone, S. (2000). Critical periods of vulnerability for the developing nervous system: Evidence from humans and animal models. Environ. Health Perspect. 108(Suppl.), 511–533.[ISI][Medline]

Safe, S. (1989). Polychlorinated biphenyls (PCBs): Mutagenicity and carcinogenicity. Mutat. Res. 220, 31–47.[ISI][Medline]

Safe, S. (1994). Polychlorinated biphenyls (PCBs): Environmental impact, biochemical and toxic responses, and implications for risk assessment. Crit. Rev. Toxicol. 24, 87–149.[ISI][Medline]

Saghir, S., Hansen, L., Holmes, K., and Kodavanti, P. (2000). Differential and non-uniform tissue and brain distribution of two distinct 14C-hexachlorobiphenyls in weanling rats. Toxicol. Sci. 54, 60–70.[Abstract/Free Full Text]

Schantz, S. L. (1996). Developmental neurotoxicity of PCBs in humans: What do we know and where do we go from here? Neurotoxicol. Teratol. 18, 217–227, discussion 229–276.[ISI][Medline]

Schantz, S. L., Seo, B. W., Moshtaghian, J., Peterson, R., and Moore, R. (1996). Effects of gestational and lactational exposure to TCDD or coplanar PCBs on spatial learning. Neurotoxicol. Teratol. 18, 305–313.[ISI][Medline]

Seegal, R. F. (1996). Can epidemiological studies discern subtle neurological effects due to perinatal exposure to PCBs? Neurotoxicol. Teratol. 18, 251–254, discussion 271–276.[ISI][Medline]

Storr-Hansen, E. (1991). Comparative analysis of thirty polychlorinated biphenyl congeners on two capillary columns of different polarity with non-linear multi-level calibration. J. Chromatogr. 558, 375–392.

van Birgelen, A. P., Ross, D. G., DeVito, M. J., and Birnbaum, L. S. (1996). Interactive effects between 2,3,7,8-tetrachlorodibenzo-p-dioxin and 2,2`,4,4`,5,5`-hexachlorobiphenyl in female B6C3F1 mice: Tissue distribution and tissue-specific enzyme induction. Fundam. Appl. Toxicol. 34, 118–131.[ISI][Medline]

Van Birgelen, A. P., Van der Kolk, J., Fase, K. M., Bol, I., Poiger, H., Brouwer, A., and Van den Berg, M. (1994). Toxic potency of 3,3',4,4',5-pentachlorobiphenyl relative to and in combination with 2,3,7,8-tetrachlorodibenzo-p-dioxin in a subchronic feeding study in the rat. Toxicol. Appl. Pharmacol. 127, 209–221.[ISI][Medline]

van der Burght, A. S., Clijsters, P. J., Horbach, G. J., Andersson, P. L., Tysklind, M., and van den Berg, M. (1999). Structure-dependent induction of CYP1A by polychlorinated biphenyls in hepatocytes of cynomolgus monkeys (Macaca fascicularis). Toxicol. Appl. Pharmacol. 155, 13–23.[ISI][Medline]

van der Plas, S. A., de Jongh, J., Faassen-Peters, M., Scheu, G., van den Berg, M., and Brouwer, A. (1988). Toxicokinetics of an environmentally relevant mixture of dioxin-like PHAHs with or without a non-dioxin-like PCB in a semi-chronic exposure study in female Sprague Dawley rats. Chemosphere 37, 1941–1955.

Vodicnik, M. J., and Lech, J. J. (1980). The transfer of 2,4,5,2',4',5'-hexachlorobiphenyl to fetuses and nursing offspring. I. Disposition in pregnant and lactating mice and accumulation in young. Toxicol. Appl. Pharmacol. 54, 293–300.[ISI][Medline]