The Relative Bioavailability and Metabolism of Bisphenol A in Rats Is Dependent upon the Route of Administration

Lynn H. Pottenger*, Jeanne Y. Domoradzki{dagger}, Dan A. Markham{dagger}, Steven C. Hansen{dagger}, Stuart Z. Cagen{ddagger} and John M. Waechter, Jr.{dagger},1

* Toxicology and Environmental Research and Consulting, Dow Europe, Horgen, Switzerland; {dagger} Toxicology and Environmental Research and Consulting, The Dow Chemical Company, Midland, Michigan 48674; and {ddagger} Health, Safety and Environment, Shell Chemical Company, Houston, Texas 77210

Received July 16, 1999; accepted October 12, 1999


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Bisphenol A (BPA) is used to produce polymers for food contact applications, thus there is potential for oral exposure of humans to trace amounts via the diet. BPA was weakly estrogenic in screening assays measuring uterine weight/response, although much higher oral doses of BPA were required to elicit a uterotropic response as compared to other routes of administration. The objective of this study was to determine if a route dependency exists in the pharmacokinetics and metabolism of 14C-labeled BPA following single oral (po), intraperitoneal (ip), or subcutaneous (sc) doses of either 10 or 100 mg/kg to Fischer 344 rats. Results indicated a marked route dependency in the pharmacokinetics of BPA. The relative bioavailability of BPA and plasma radioactivity was markedly lower following oral administration as compared to sc or ip administration. The major fraction of plasma radioactivity following oral dosing was the monoglucuronide conjugate of BPA (68–100% of plasma radioactivity). BPA was the major component in plasma at Cmax following sc or ip administration exceeded only by BPA-monoglucuronide in females dosed ip. Up to four additional unidentified metabolites were present only in the plasma of animals dosed ip or sc. One of these, found only following ip administration, was tentatively identified as the monosulfate conjugate of BPA. The monoglucuronide conjugate was the major urinary metabolite; unchanged BPA was the principal component excreted in feces. These results demonstrated a route dependency of BPA bioavailability in rats, with oral administration resulting in the lowest bioavailability, and offer an explanation for the apparent route differences in estrogenic potency observed for BPA.

Key Words: bisphenol A; bioavailability; absorption; metabolism; rats; route dependency.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Bisphenol A (4,4'-isopropylidene-2-diphenol, BPA) is a monomer used in the production of polycarbonate and epoxy resins. Because of the use of BPA in the production of materials used for food and potable water contact, as well as sealant and restorative materials used in dentistry, there could be potential for oral exposure of humans to trace amounts of BPA.

Generally, BPA has shown a low order of toxicity in extensive studies of its potential to produce adverse effects (BIBRA, 1989Go). Although it has been known for several decades that BPA produced weak responses in uterotropic assays (Bitman and Cecil, 1970Go; Dodds and Lawson, 1936Go), there has recently been renewed attention to the estrogenic potential of BPA. Specifically, Krishnan et al. (1993) found that BPA leaching from autoclaved polycarbonate flasks was confounding studies to determine if S cerevisiae produced estrogens. More recently, Gaido and coworkers (1997) confirmed the weak estrogenicity of BPA in vitro, showing BPA to be approximately 15,000 times less potent than 17ß-estradiol, and Kuiper et al. (1997) demonstrated that BPA could interact with both the {alpha}- and ß-estrogen receptors.

A review of the scientific literature revealed an apparent route dependency for the elicitation of estrogenic effects of BPA in pharmacologic screening assays, specifically uterine weight/response assays. Comparison of available data on the lowest observed effect level (LOEL) for in vivo effects of BPA on the uterus indicated that subcutaneous (sc) administration resulted in effects at doses considerably lower than that identified following either oral (po) or intraperitoneal (ip) administration of BPA (Bitman and Cecil, 1970Go; Bond et al., 1980Go; Dodds and Lawson, 1936Go; Morrissey et al., 1987Go). The chemical structure of BPA also suggested that first-pass metabolism and rapid elimination of parent would be probable following oral administration as is the case with phenolic compounds (Capel et al., 1972Go; Cassidy and Houston, 1984Go). Hence a route dependency in the bioavailability of BPA following oral exposure (defined as the amount of parent compound reaching the systemic circulation) as compared to the bioavailability of BPA following sc and ip dosing is likely be an important factor in determining the biologic activity of BPA.

This study investigated possible route dependency in the bioavailability of 14C-labeled BPA by determining the unchanged parent and radioactivity concentration-time courses in blood and plasma, respectively, following oral, ip, and sc administration of 14C-BPA at two dose levels. In addition, disposition and tissue distribution of BPA-derived radioactivity were investigated, along with elimination kinetics of BPA-derived radioactivity in urine and feces. Additional groups of animals were dosed to determine the number and structural identity (if possible) of 14C-BPA-derived plasma metabolites found following oral, ip, or sc administration of 14C-BPA using high performance liquid chromatography and mass spectroscopy.

The low dose level of 10 mg BPA/kg body weight was in the range reported to cause changes in uterine glycogen content by Bitman and Cecil (1970). The high dose of 100 mg BPA/kg body weight was selected as a dose level that elicited minimal symptoms of acute toxicity based on the results of a pilot study where BPA was administered ip at 100, 250, and 500 mg BPA/kg body weight. The high dose level in the pilot study was of the same order of magnitude as the dietary administration LOEL in mice reported by Reel et al., (1985; 0.25% in feed estimated equivalent to 438 mg BPA/kg bw/day).


    MATERIAL AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Study Design
Disposition/blood and plasma time course.
Groups of five rats/sex/dose/route implanted with indwelling jugular vein cannulae received a single dose of 10 or 100 mg 14C-BPA/kg bw by oral gavage (po), ip, or sc administration. Animals were placed in Roth-type metabolism cages immediately postdosing for a 7-day period, during which time blood and excreta were collected. Blood specimens were divided and a portion analyzed for BPA; the remainder was used to obtain plasma for analysis of radioactivity. Tissues were analyzed for radioactivity at terminal sacrifice. Excreta were analyzed for radioactivity and selected urine and fecal extracts were profiled by HPLC.

Plasma metabolites and/or parent compound.
Groups of three rats/sex/dose/route/time point received a single dose of 10 or 100 mg 14C-BPA/kg bw. Two sacrifice time points were selected for each dose and route, based onprevious plasma and blood time-course data (Table 1Go). The time postdosing at which the peak plasma radioactivity was reached (Cmax) was selected as the early sacrifice time. The later sacrifice time was to represent a time when only nonquantifiable (NQ) amounts of parent BPA remained in blood (NQ-Parent), and was selected for each dose group and route based on the time-to-NQ determined in the time-course study.


View this table:
[in this window]
[in a new window]
 
TABLE 1 Times of Sacrifice Postdosing for the Determination of Plasma Bisphenol A and Bisphenol A Metabolite Concentrations in Male and Female Fischer 344 Rats.
 
Plasma was obtained at the Cmax and NQ-Parent time points, and pooled plasma samples were analyzed by HPLC to compare metabolite profiles and to quantify parent and BPA-glucuronide.

Test Materials and Standards
Radiolabeled bisphenol A (4,4'-isopropylidene-2-14C-diphenol, 2,2-bis-(p-hydroxyphenyl)-2-14C-propane) was obtained from Wizard Laboratories (Davis, CA) with a specific activity of 55 mCi/mmole and a radiochemical purity of 99.3% (analyzed upon receipt by reversed-phase high-performance liquid chromatography (HPLC) with a UV detector in tandem with a flow-though radioactivity detector). Nonradiolabeled bisphenol A was supplied by The Dow Chemical Company (Freeport, TX) with a purity of 99.7%. The authentic standard of monoglucuronide of BPA was synthesized at The Dow Chemical Company. The structure was confirmed as the monoglucuronide of BPA by liquid chromatography/mass spectroscopy (LC/MS) and nuclear magnetic resonance (NMR) analysis.

Animals
Male and female Fischer 344 rats 8–9 weeks of age were purchased from the Charles River Breeding Laboratory (Raleigh, NC). Care and husbandry of animals were in accordance with the Guide for the Care and Use of Laboratory Animals (1985). Upon arrival in the laboratory (fully accredited by AAALAC International), the animals were examined by a veterinarian and found to be in good health, then acclimated to the laboratory environment for at least 1 week and to glass Roth-type metabolism cages for 2 days. Animal rooms were maintained at 21–23°C and 40–70% RH during 12-h light/dark cycles. Certified rodent chow (#5002; Purina Mills, Inc., St. Louis, MO) and municipal drinking water were provided ad libitum, except that all rats from all dose groups were fasted overnight with food withdrawn approximately 16 h prior to dosing with the radiotracer and returned approximately 4 h postdosing.

Dosing Solution Preparation and Administration
Radiotracer in acetonitrile (Fisher Scientific, Pittsburgh, PA, HPLC grade) stock solution was added to a weighed amount of unlabeled BPA, dissolved in acetone (Fisher Scientific, HPLC grade). A weighed amount of corn oil (Sigma Chemical Company, St. Louis, MO) was added while stirring at room temperature to permit evaporation of acetone and acetonitrile. Confirmation of the concentration of 14C-BPA in the dose solutions was analytically determined as described below. Stability analysis demonstrated that BPA concentration was essentially unchanged over a 7-week period. Each dosing day, prior to and following administration of 14C-BPA dose solution, the homogeneity of each dose solution used was confirmed by quantifying radioactivity. Rats were dosed with a target of 5- and 2.5-g dose solution/kg bw for the oral and ip, and the sc groups, respectively. The specific activities of the dose solutions ranged from 13–14 and from 23–26 µCi/g dose solution for the disposition/time-course oral and ip, and the disposition/time-course sc dose solutions, respectively; and from 68–71 and from 134–141 µCi/g dose solution for the plasma metabolite oral and ip, and the plasma metabolite sc dose groups, respectively.

Sample Collection and Preparation
Urine voided during successive 12-h intervals through 48 h and subsequent 24-h intervals through 168 h was collected in dry ice-chilled containers. Weighed aliquots of urine were mixed with Aquasol® liquid scintillation fluid (New England Nuclear, Boston, MA) and analyzed for radioactivity by liquid scintillation counting (LSC).

Feces were collected at 24-h intervals over 7 days postdosing in dry ice-chilled containers and stored frozen (–20°C). Aqueous homogenates (25% w/w) were prepared and weighed aliquots were solubilized using Soluene® 350 (Packard Instrument Company, Inc., Downers Grove, IL) and analyzed for radioactivity by LSC using Hionic-Fluor® (Packard Instrument Company, Inc., Downers Grove, IL).

Animals were sacrificed 7 days postdosing and were anesthetized with CO2 and exsanguinated via cardiac puncture. Tissues collected and analyzed for radioactivity were blood, brain, gonads, kidneys, liver, perirenal fat, skin, uterus, and remaining carcass. The blood, fat, ovaries, and uterus were directly solubilized, as was a representative sample of skin with hair, taken from the interscapular area. Skins from all sc dose groups were subsequently solubilized in entirety. Aqueous homogenates (25–33% w/w) were prepared from the remaining tissues and weighed aliquots were solubilized. The solubilized samples were analyzed for radioactivity by LSC. Following sacrifice and collection of the last cage rinse, a final cage wash (FCW) was collected for each animal. An aliquot of FCW was analyzed for radioactivity by LSC.

Blood specimens were collected at 0.083, 0.25, 0.5, 0.75, 1, 1.5, 2, 4, 8, 12, 18, 24, 48, 72, 96, 120, 144, and 168 h postdosing, via indwelling jugular vein cannulas (Harms and Ojeda, 1974Go). The cannula was rinsed with heparinized saline following each sample withdrawal. Blood specimens were divided and a weighed aliquot was used for parent analysis as described below. The remainder was placed in heparinized vials and centrifuged to obtain plasma. Plasma radioactivity was determined by LSC. Generally, blood sampling was continued through the absorption phase and continued until at least two consecutive time points resulted in nonquantifiable levels of plasma radioactivity.

Plasma metabolites.
Immediately following dosing, rats were placed either in polystyrene animal cages (postdosing sacrifice time of < 8 h) or in glass Roth-type metabolism cages (postdosing sacrifice time of >= 8 h). At the appropriate sacrifice time postdosing, animals were anesthetized with CO2 and exsanguinated via cardiac puncture. Plasma was obtained from the blood and pooled according to route, dose, sex, and sacrifice time, prior to HPLC analysis.

Analysis of Dose Solution Containing 14C-Bisphenol A
A chemical derivatization-mass spectrometry analytical method was developed employing a stable isotope (BPA substituted with 8 deuterium atoms, D8-BPA, Cambridge Isotope Laboratories, Andover, MA) internal standard to quantitate BPA in corn oil dosing solutions. A minimum of three replicate 10 mg aliquots (50 mg aliquots for low dose) was obtained from each corn oil dosing solution while the solution was mixing. Each aliquot was transferred to a 4 ml glass vial, fortified with a known amount of D8-BPA (~4 to 20 mg/g) internal standard, and diluted with 2 ml acetone followed by vortex mixing for ~1 min. A 50-µl aliquot of the acetone solution was transferred to clean 4 ml glass vials containing 1 ml of toluene. Fifty microliters of 5N NaOH plus 50 µl pentafluorobenzoyl chloride (PFBCl, Aldrich Chemical Company, Milwaukee, WI ) derivatization reagent were then added. The vials were sealed with Teflon-lined caps and vortex-mixed at 60°C for 30 min to complete derivization. A 2 ml aliquot of Milli-Q water (Millipore Corporation, Bedford, MA) was added to each vial and vortex-mixed for ~2 min to wash the toluene layer. The toluene layer was transferred to a glass autosampler vial and sealed with a Teflon-lined cap. Matrix standards (i.e., corn oil) and appropriate controls were prepared with each sample set. The samples, standards, and controls were analyzed using gas chromatography/negative chemical ionization/mass spectrometry (GC/NCI/MS) following the analysis conditions presented below. The level of BPA in corn oil was calculated based on the ratio of BPA to D8-BPA peak area response.

Quantitation in Rat Blood
A chemical derivatization-mass spectrometry analytical method was developed employing a D8-BPA stable isotope internal standard. A 4 ml glass sample vial containing 1 ml of toluene and a known amount of D8-BPA (~1 to 2 mg/g) internal standard was prepared for each time point. A portion (~150 µl) of each blood sample was added to its tared sample vial through the septum and vortex-mixed for ~1 min. If the blood samples were not analyzed immediately they were stored at –80°C until analysis. Multiple sets of matrix (rat blood) standards were prepared and stored with the samples to ensure that appropriate standards would be available for each analysis run. On the day of analysis, 50 µl of 5N NaOH plus 50 µl of PFBCl derivatization reagent was added to each vial. The vials were vortex-mixed at 60°C for 30 min to complete derivatization. The vials were centrifuged at 3000 rpm. for 10 min and the toluene layer was transferred to a glass autosampler vial and sealed with a Teflon-lined cap. The samples, standards, and controls were analyzed using GC/NCI/MS following the analysis conditions presented below. The level of BPA in rat blood was calculated based on the ratio of BPA to D8-BPA peak area response. The limit of quantitation (LOQ) for low and high dose level blood samples was 0.01 and 0.1 µg BPA/g blood, respectively.

Preparation of Pooled Plasma Samples for HPLC Radiochemical Analysis
Pooled plasma samples (Cmax plasma and NQ-Parent plasma) were allowed to thaw to laboratory temperature and mixed well. Approximately 700 µl weighed aliquots were mixed with ~300 µl weighed aliquots of acetonitrile (ACN, with 1% acetic acid). The mixture was vortexed and centrifuged (800 x g) for about 15 min. The top layer was removed and duplicate weighed aliquots (25–50 µl) were analyzed for radioactivity by LSC to determine extraction efficiency (98 ± 6%). The extracts were analyzed using HPLC/LSC following the conditions presented below. Aliquots of plasma extracts (100–200 µl) were injected onto the HPLC system and 20-s fractions were collected for analysis by LSC. The dpm obtained for the fractions were used to reconstruct a radiochemical HPLC profile. The radiolabeled peaks were integrated and peak areas were converted to µg-eq BPA. Recoveries from the HPLC system averaged 90 ± 13%.

Preparation of Pooled Plasma Samples for HPLC/–ESI/MS Analysis
Selected pooled plasma samples (female ip peak plasma, 100 mg/kg; and male sc NQ-parent plasma, 100 mg/kg) were also subjected to HPLC/ negative Electrospray Ionization/Mass Spectroscopy (HPLC/–ESI/MS) analysis. These samples were prepared as described above for HPLC analysis. Aliquots of pooled plasma (25 µl) were injected onto the HPLC system under the same conditions described for radiochemical analysis. Additional –ESI/MS analysis conditions are presented below.

Analytical Conditions
Typical analytical conditions are given below (although slight variations in conditions may have occurred over the course of the study).

Typical GC/MS analysis conditions for BPA dose confirmation and BPA blood levels

Typical HPLC/radioactivity analysis conditions

Typical HPLC/–ESI/MS analysis conditions

Selection of Doses
Preliminary studies demonstrated that a dose of 500 mg BPA/kg bw administered ip in corn oil was acutely toxic to F344 rats and resulted in death within 1 h postadministration (data not shown). An ip dose of 250 mg BPA/kg bw administered ip in corn oil resulted in shallow breathing and lateral recumbancy within 5 min postdosing, but was not lethal. Similar symptoms were observed following administration of a dose of 100 mg BPA/kg bw delivered ip, but the effects were resolved by 2–4 h postdosing. Overnight fasting did not appear to affect either the extent or the duration of these effects. During in-life conduct of this study, ip administration of the high dose resulted in observations of similar effects in some study animals; all such effects were resolved by 2–4 h postdosing. No such effects were noted following administration of either dose via the oral or sc routes, nor following administration of the low dose by any route.

Data Analysis
Means and standard deviations were calculated using Microsoft Excel with full precision. Area under the blood unchanged parent and plasma 14C concentration-time curves (AUC) were estimated based on the linear trapezoidal rule as described by Gibaldi and Perrier (1982) and calculated using WinNonLin noncompartmental analysis (Scientific Consulting, Inc., Cary, N.C.). AUC for both unchanged parent and total radioactivity were calculated based on the last quantifiable time point, and were not extrapolated to infinity, as extrapolation of AUC to infinity would place inappropriate emphasis on values that were at the limit of detection.

Good Laboratory Practices
This study was conducted in accordance with the EPA Toxic Substances Control Act (EPA-TSCA) Good Laboratory Practice Standards; the Organisation for Economic Co-Operation and Development (OECD, 1982) Principles of Good Laboratory Practice; and the FDA Good Laboratory Practice Regulations for Nonclinical Laboratory Studies, Final Rule (FDA-GLP, 1988).


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Metabolism and Routes of Excretion
The doses administered were within ± 12% of target for all groups (data not shown). Recoveries of administered dose for the disposition/time-course dose groups ranged from 84–98% of administered dose (Table 2Go). The somewhat lower recoveries obtained with the sc groups (84–95% of administered dose) prompted additional investigation. The entire pelt from all rats administered 14C- BPA via sc route was solubilized to account for any potential sequestration of a significant fraction of administered dose in the skin, or for any leakage onto the pelt from the injection site. This hypothesis was not supported, with an average of less than 1% of administered dose recovered from the pelts (data not shown).


View this table:
[in this window]
[in a new window]
 
TABLE 2 Disposition of 14C-BPA-Derived Radioactivity Following Administration to F344 Rats by Three Routes
 
Fecal elimination of 14C-BPA-derived radioactivity represented the major elimination pathway for all dose groups, comprising from 52–83% of administered dose (Table 2Go). HPLC analysis of fecal extracts from selected collection intervals (0–24, 24–48, and 48–72 h) allowed quantitation of the amount of BPA or BPA metabolites in feces, based on radioactivity; the identity of parent compound in feces was based on retention time match with an authentic standard of BPA. HPLC analysis revealed up to eight different unidentified radiolabeled peaks, most of which represented less than 5% of administered dose (data not shown). The major peak was identified as unchanged parent compound and extrapolation of analyzed fecal composition across the remaining intervals resulted in an estimate that 86–93% of total fecal radioactivity comprised parent compound across all routes and doses (data not shown).

The urinary elimination of 14C-BPA-derived radioactivity was consistently about 2-fold greater in females than in males, across all doses and routes, representing 13–16% and 21–34% of administered dose in males and females, respectively (Table 2Go). HPLC analysis of urine from selected collection intervals (0–12 or 12–24 h) allowed quantitation of BPA and the monoglucuronide conjugate of BPA, and the tentatively identified monosulfate conjugate of BPA, based on radioactivity; identity was based on matching retention times with authentic standard for BPA and BPA-glucuronide. HPLC analysis identified up to 14 different radiolabeled peaks in urine, with most peaks representing less than 3% of administered dose (data not shown). In addition, selected urine samples were subjected to –ESI/MS to structurally confirm the presence of BPA-glucuronide and to tentatively identify the monosulfate conjugate of BPA. Extrapolation of analyzed urinary composition across the remaining intervals resulted in an estimate of the monoglucuronide conjugate of BPA comprising 57–87% of total urinary radioactivity (Table 3Go).


View this table:
[in this window]
[in a new window]
 
TABLE 3 Distribution of Major Urinary Metabolites Following Administration of 14C-BPA to F344 Rats by Three Routes
 
Tissue radioactivity demonstrated some route differences, with orally dosed animals retaining from 0.03% to 0.26% of administered dose in tissues at 7-d postdosing terminal sacrifice, as compared with ip, 0.65–0.85%, and sc, 1.03–1.29% of administered dose (Table 2Go); however, no attempt was made to identify tissue metabolites.

There were no marked route-dependent or dose-dependent differences in the routes of excretion or in tissue distribution of 14C-BPA-derived radioactivity.

BPA Pharmacokinetics in Blood
The concentration-time profiles of BPA in blood are shown in Figure 1Go. BPA was not quantifiable in blood after the first 24 or 48 h (female high dose ip and sc only) postdosing (limit of quantitation, LOQ = 0.01 and 0.1 µg BPA/g blood for 10- and 100-mg/kg dose groups, respectively). Pharmacokinetic parameters estimated from these data are summarized in Table 4Go.




View larger version (25K):
[in this window]
[in a new window]
 
FIG. 1. Concentration-time profiles for 14C-BPA in blood of male and female F344 rats following administration of either 10 or 100 mg 14C-BPA/kg bw by oral, ip, or sc routes: (A) 100 mg/kg, males; (B) 100 mg/kg, females; (C) 10 mg/kg, males; (D) 10 mg/kg, females. BPA was no longer quantifiable after 48 h postdosing for any route. Data are the mean±standard deviation for three to five animals per dose group.

 

View this table:
[in this window]
[in a new window]
 
TABLE 4 Pharmacokinetic Parameters of BPA in Blood and of 14C-BPA-Derived Radioactivity in Plasma following Administration of 14C-BPA to F344 Rats by Three Routes
 
There were clear differences in the concentration-time profiles of BPA in blood, based on route of administration, dose, and sex. Time to achieve maximum blood concentrations (Tmax) for unchanged parent compound ranged from 0.083–4 h, with sc administration resulting in the latest Tmax postdosing, seen as a plateau starting at 0.75 h with the latest Tmax at 4 h (Fig. 1Go). The maximum blood concentration (Cmax) of unchanged parent compound was markedly dependent on route of administration. In addition, Cmax of BPA was dependent on dose and sex. Parent compound could not be quantified following oral administration of 10 mg BPA/kg bw to male rats, and was only transiently quantifiable following oral administration of this dose to female rats (Figs. 1C and 1DGoGo). Oral administration of 100 mg BPA/kg bw resulted in about a 10-fold increase in the concentrations of BPA in the blood of female rats compared with low dose and BPA was quantifiable in blood through the last sample analyzed at 12 h postdosing(Figs. 1B and 1DGo). BPA in the blood of males dosed with 100 mg BPA/kg bw was quantifiable for less than 0.75 h postdosing and the blood concentrations at Cmax were about 10-fold less than those found in female rats at the same dose (Figs. 1A and 1BGo). Oral administration resulted in Cmax values that were from 1 to 2 orders of magnitude lower than those found following ip or sc administration (Table 4Go). Administration via the ip route to females resulted in the highest Cmax for parent compound of any route of administration at both doses. The actual time required to reach nonquantifiable (NQ) concentrations of BPA in blood (time-to-NQ) was highly route, dose, and sex dependent. Following oral administration, parent BPA was no longer reproducibly quantifiable (i.e., not all animals in the same dose group had quantifiable values) in blood at 0.083 or 0.75 h postdosing in males, and at 1 or 18 h postdosing in females, at the low and high doses, respectively. IP administration resulted in time-to-NQ of 8 and 12 h for males and 24 and 72 h for females at the low and high doses, respectively. SC administration produced the longest time-to-NQ for both males and females (18 and 24 h, males; 48 and 72 h, females, at the low and high doses, respectively).

Pharmacokinetics of BPA appeared to be linear across dose levels for the same route of administration; the roughly dose-proportionate increases in Cmax and area-under-the-curve (AUC) support linearity for parent compound across the 10-fold dose range. The only sex differences found in blood BPA concentrations were following oral administration as mentioned previously.

The relative bioavailability of BPA was highly route-dependent, based on the AUC values listed in Table 4Go. Administration via the sc route resulted in dramatic differences in systemic bioavailability of unchanged parent BPA when compared with oral administration. Estimation of BPA AUC for sc administration resulted in values from 7-fold (high dose females) to 245-fold (high dose males) greater than for oral administration. Administration via the ip route also resulted in greater systemic bioavailability of parent BPA than oral administration, from 3-fold (low dose females) to 164-fold (high dose males). In addition, relative bioavailability of BPA was sex dependent following oral administration. AUC values increased roughly proportionately to the administered dose, supporting linearity of the pharmacokinetics of BPA across this dose range (Table 4Go).

Plasma Radioactivity Pharmacokinetics
The time course for 14C-BPA-derived plasma radioactivity concentration was measured over 168 h postdosing. Pharmacokinetic parameters estimated from the plasma time-course data are included in Table 4Go.

As was the case for unchanged parent compound, the levels of 14C-BPA-derived plasma radioactivity demonstrated marked route-dependent differences. In addition, 14C-BPA-derived plasma radioactivity was dependent to a lesser degree upon dose and sex. Tmax for total plasma radioactivity showed only minor differences across routes, ranging from 0.083 to 1 h postdosing (Table 4Go). Plasma radioactivity Cmax was consistently higher following ip administration, regardless of the dose, with the high dose ip female demonstrating the highest value of all groups (67.81 µg-eq BPA/g plasma). In particular, the Cmax for ip females at the 100 mg/kg dose was about 10-fold higher than for sc females at the same dose. The time-to-NQ for 14C-BPA-derived plasma radioactivity was also route dependent. For both sexes, the time-to-NQ was 72 h following oral administration, 72–120 h following ip administration, and 96–168 h following sc administration (LOQ <= 0.7 or 0.07 µg-eq BPA/g plasma for high and low dose, respectively).

Pharmacokinetics of 14C-BPA-derived plasma radioactivity appeared to be proportionate to dose for the same route and sex, except for females following ip administration, as noted before (Table 4Go). The approximately dose-proportionate increases in Cmax, AUC, and plasma radioactivity concentrations across time for the other dose groups support linearity for 14C-BPA-derived plasma radioactivity across this 10-fold dose range.

The relative systemic bioavailability of 14C-BPA-derived plasma radioactivity was clearly route dependent, based on the area-under-the-curve (AUC) values summarized in Table 4Go; the high dose data are depicted in Figure 2Go. Administration via either the sc or the ip route resulted in a relative 14C-BPA-derived plasma radioactivity bioavailability that was from 2- to 3-fold greater than for oral administration, based on AUC values. In addition, bioavailability of 14C-BPA-derived plasma radioactivity was dependent to a lesser degree on sex (< 2-fold), whereas bioavailability increased proportionate to administered dose. The relative systemic bioavailability estimated for the 10 mg/kg dose groups demonstrated similar route-dependent effects (Table 4Go).



View larger version (40K):
[in this window]
[in a new window]
 
FIG. 2. Relative bioavailability of BPA and 14C-BPA–derived radioactivity following a single 100 mg/kg dose to Fischer 344 rats by three routes.

 
Plasma Metabolites
The 14C-BPA-derived plasma radioactivity values found in the animals dosed to determine plasma metabolites were similar to the values reported for the same time points from the previous disposition/time-course data (data not shown), demonstrating the reproducibility of 14C-BPA plasma pharmacokinetics in F344 rats. In addition, the Cmax levels of parent BPA quantified by MS in blood corresponded very well with those quantified by flow-through radiochromatography in plasma. This suggested that parent BPA was not sequestered in the formed elements of blood and that the blood BPA values would be comparable to the values obtained in plasma.

Representative HPLC radiochromatograms of the plasma metabolites are presented in Figure 3Go. Radiochemical peaks were numbered consistently across the different samples, with a total of 11 peaks with different retention times resolved, and one peak region. This unresolved area, eluting in the same region as the peaks P-2, P-3, and P-4, and comprising at least 3 peaks, was termed PR-2,3,4, for peak region 2,3,4. No one sample contained all 11 peaks.



View larger version (29K):
[in this window]
[in a new window]
 
FIG. 3. Typical reconstructed radiochromatogram of male rat plasma extract at either Cmax or NQ-Parent time point following administration of 100 mg 14C-BPA/kg bw by three routes: (A) oral male, NQ-Parent; (B) ip male, Cmax; (C) sc male, NQ-Parent.

 
The major peaks found in plasma were two well-resolved peaks, P-9 and P-5. Plasma peak P-9 from the HPLC profile had the same retention time as the authentic standard of BPA when analyzed using separate injections and when employing HPLC co-chromatography (data not shown). Plasma peak P-5 from the HPLC profile was identified as the monoglucuronide conjugate of BPA using HPLC/–ESI/MS. The –ESI/MS spectrum (Fig. 4AGo) showed a base peak at m/z = 403, corresponding to the M – H, and an acetic acid adduct (+ 60 amu) at m/z = 463. This spectrum also showed a peak at m/z = 227, which corresponds to a loss of 177 (glucuronide = 177 amu), M – C6H9O6-. The –ESI/MS spectrum of P-5 was consistent with the spectrum of the authentic standard of the monoglucuronide conjugate of BPA (Fig. 4BGo). In addition to the mass spectral confirmation, P-5 had the same retention time as the authentic standard when analyzed using separate injections and when employing HPLC co-chromatography (data not shown).



View larger version (15K):
[in this window]
[in a new window]
 
FIG. 4. (A) Typical ESI mass spectrum of Peak P-5 in rat plasma, which was identified as the BPA-monoglucuronide conjugate. (B) Typical HPLC/–ESI/MS mass spectrum of an authentic standard of BPA-monoglucuronide conjugate.

 
Plasma peak P-8 from the HPLC profile was tentatively identified as the monosulfate conjugate of BPA using HPLC/–ESI/MS. The –ESI/MS spectrum (Fig. 5Go) showed a base peak at m/z = 307, corresponding to the M – H. This spectrum also showed a peak at m/z = 227, which corresponds to a loss of 81 (sulfate = 81 amu), M – HO3S. An authentic standard of the monosulfate conjugate of BPA was not available for this work; therefore, comparison of mass spectrum and retention time with an authentic standard was not possible.



View larger version (11K):
[in this window]
[in a new window]
 
FIG. 5. Typical ESI mass spectrum of Peak P-8 in rat plasma, which was tentatively identified as the BPA-monosulfate conjugate.

 
The unresolved peak region PR-2,3,4 comprised a major fraction of plasma radioactivity for ip and sc routes only at the NQ-Parent time point, when total plasma radioactivity was from 2- to 50-fold less than Cmax values. No evidence of BPA glucuronide or sulfate conjugates was observed in control rat plasma analyzed using the same preparation and analysis conditions (data not shown).

Figures 6 and 7GoGo depict the quantitative results from the radiochemical HPLC profiling of plasma obtained following administration of 14C-BPA at the Cmax and NQ-Parent time points, respectively. The composition of 14C-BPA-derived plasma radioactivity clearly demonstrated route-, sex-, dose-, and time-dependent differences, both qualitative and quantitative. Specifically, following oral administration at all times and doses investigated, the large majority of 14C-BPA-derived plasma radioactivity was P-5, the monoglucuronide conjugate of BPA (68–100% of plasma radioactivity) (data not shown; values calculated from data depicted in Figs. 6 and 7GoGo). This was true both for males and females, even as early as 0.083 h postdosing. By administering more total radioactivity to the plasma metabolite dose groups, and by increasing the total sample size, the detection limits were more sensitive than in the disposition/time-course dose groups (0.01 and 0.07 µg-eq BPA/g plasma, for low dose plasma metabolite and disposition/time-course groups, respectively). Thus, it was possible to detect unchanged parent compound in the low dose oral male plasma metabolite dose group samples, whereas unchanged parent BPA was never quantifiable in the corresponding disposition/time-course dose group. In addition, it was possible to determine a time-to-NQ for the low dose oral female group of 18 h postdosing.



View larger version (31K):
[in this window]
[in a new window]
 
FIG. 6. Quantitative comparison of BPA and BPA metabolites in plasma at the Cmax time postdosing from male and female rats following administration of either 10 or 100 mg BPA/kg bw by oral, ip or sc routes: (A) 100 mg/kg, males; (B) 100 mg/kg, females; (C) 10 mg/kg, males; (D) 10 mg/kg, females. Data derived from aliquots of plasma pooled from three animals/sex/route.

 


View larger version (28K):
[in this window]
[in a new window]
 
FIG. 7. Quantitative comparison of BPA and BPA metabolites in plasma at the NQ-Parent time postdosing from male and female rats following administration of either 10 or 100 mg BPA/kg bw by oral, ip, or sc routes: (A) 100 mg/kg, males; (B) 100 mg/kg, females; (C) 10 mg/kg, males; (D) 10 mg/kg, females. Data derived from aliquots of plasma pooled from three animals/sex/route.

 
In contrast to the large fraction of plasma radioactivity represented by BPA-monoglucuronide in plasma from orally dosed rats, following ip or sc administration, BPA-monoglucuronide represented a much smaller fraction of plasma radioactivity: 27–69% and 17–49% of plasma radioactivity, respectively (data not shown; values calculated from data depicted in Figs. 6 and 7GoGo). Both ip and sc administration resulted in larger fractions of plasma radioactivity comprised of unchanged parent compound, corresponding with the differences in BPA AUC demonstrated for the disposition/time-course data. Unchanged parent comprised 27–51% and 65–76% of total plasma radioactivity following ip and sc administration, respectively, whereas oral administration resulted in only 2–8% of total plasma radioactivity comprised of unchanged parent compound (data not shown; values calculated from data depicted in Figs. 6 and 7GoGo).

Qualitative route differences were evident also, as is shown in Figures 6 and 7GoGo. Both ip and sc administration resulted in the formation of unique metabolite profiles, with radiochemical peaks that were not observed in samples following oral administration of 14C-BPA. Examples included PR-2,3,4, which was only observed at the late time point following ip or sc administration, but comprised the majority of plasma radioactivity for that time point; P-2, which eluted early and was only observed at Cmax and only following ip or sc administration; P-6, which was observed only following ip or sc administration; P-8, which was observed only following ip administration, and was tentatively identified as the monosulfate conjugate; and P-11, which eluted after the unchanged parent compound peak and was found only following ip or sc administration. Other than the PR-2,3,4 peak, these route-dependent metabolites accounted for a minor amount of total plasma radioactivity.

Sex differences in plasma metabolites were mostly quantitative in nature. Females demonstrated much higher Cmax values, which corresponded with increased individual metabolites in plasma. In particular, following ip administration of high dose 14C-BPA, females demonstrated about 5-fold more monoglucuronide conjugate at Cmax than males, and about 2-fold more unchanged parent compound than males. This corresponded well with the increased urinary elimination of monoglucuronide demonstrated by females compared with males from the disposition/time-course data described earlier. There was an apparent qualitative sex difference with P-6 detected only in plasma from males (high ip and low sc; Figs. 6A and 7CGoGo). However, P-6 could have been present in females at levels below the LOQ (0.01–0.21 µg-eq/g plasma), so this apparent difference may not represent a true sex difference in metabolism.

Finally, plasma metabolite profiles were time dependent, especially for the ip and sc administration routes. Oral administration demonstrated slight changes in proportion of plasma radioactivity represented by monoglucuronide over the time frame investigated, but the majority of plasma radioactivity was always the monoglucuronide conjugate. This was not the case for ip and sc administration, where the early time point comprised a significant amount of unchanged parent compound; the later time point profile consisted mainly of the unknown poorly resolved peak region described earlier, PR-2,3,4.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The results from male rats dosed orally in the current study generally confirm the findings of previous investigators (Knaak and Sullivan, 1966Go) but the addition of females and the comparison with other routes of administration considerably extends the information available on the pharmacokinetics and metabolism of BPA. However, some notable differences were found between the results of the current investigation and previously published work. Knaak and Sullivan reported from 54% to 60% of the administered dose in feces by 8 days postdosing whereas generally around 80% of the dose was excreted via the feces in the current study regardless of the route of administration. In the current study for both orally administered BPA and other routes, a smaller fraction of the dose was excreted by male rats via the urine (~13–16%) than was reported previously (26–31% of the administered dose by 8 days postdosing). The larger fraction found in feces, with a concomitant decrease in the fraction found in the urine, is probably a reflection of the differences in the doses administered between the two studies. The previous investigators gave a very high dose of 800 mg BPA/kg bw, whereas the doses in this study were 10 and 100 mg/kg bw. Thus, the higher fraction excreted via the urine in the work of the previous investigators was likely due to the saturation of the metabolic or excretory mechanisms responsible for the elimination of BPA via the fecal route. Further evidence of metabolic saturation at 800 mg/kg dose used by Knaak and Sullivan was the presence of a hydroxylated metabolite of BPA found in feces which comprised ~35% of the fecal radioactivity. As Pfeiffer and Metzler (1998) have reported an absence of type I binding of BPA in rat microsomes at concentrations up to 20 nM, there is apparently no binding to the oxidative cytochrome P450s at lower tissue concentrations and thus an absence or very low degree of oxidative BPA metabolism at lower doses. However, the work of Knaak and Sullivan suggests oxidative metabolism can occur in vivo at high doses, perhaps only after saturation of other metabolic pathways as has been seen for other phenolic compounds (Smith et al., 1998Go). Alternately, the different strains of rats used may explain the differences from study to study in the fraction of the dose excreted via a given route. In the current study, analysis of fecal extracts through 72 h postdosing found the majority (86–99%) of fecal radioactivity to be free BPA based on retention time match with an authentic BPA standard, with up to seven minor unidentified metabolites present in feces. No metabolites were found to represent more than 7% of the administered dose. Thus, at the doses in the current study, there was no evidence that a significant fraction of the fecal radioactivity was a hydroxylated metabolite of BPA as reported by Knaak and Sullivan.

The results of this study clearly demonstrated that the pharmacokinetics of parent BPA in blood and 14C-BPA-derived radioactivity in plasma (representing principally the monoglucuronide conjugate of BPA) were dependent upon the route of administration and, to a lesser degree, the sex of the animal and the dose. The relative bioavailability, that is, the concentration of BPA in the systemic blood and the relative bioavailability of 14C-BPA-derived radioactivity (i.e., parent plus BPA metabolites) was clearly dependent upon the route of administration, with oral administration resulting in markedly less bioavailability than either ip or sc administration. Evidence supporting this route-dependent disposition of BPA includes the large differences in the AUCs for both parent compound in blood concentration-time profiles and plasma radioactivity concentration-time profiles, the differences in blood and plasma Cmax values between parent compound and 14C-BPA-derived radioactivity (principally the monoglucuronide conjugate of BPA), and differences in the time-to-NQ values for both parent compound and 14C-BPA-derived radioactivity. It should be noted that the AUC for both BPA blood concentration and plasma radioactivity were the greatest following sc administration, despite the fact that this route of administration had the lowest overall recoveries of the total administered dose, suggesting that some of the administered dose may have been lost from the dosing site.

Systemic blood and or plasma concentrations of BPA were very low (from nonquantifiable to only 2.29 µg-eq/g at peak for females) following oral administration of 10 or 100 mg/kg bw, as BPA was metabolized principally to its monoglucuronide conjugate via first-pass metabolism. As the extent of absorption has been defined as the amount of parent compound reaching the systemic circulation (Gibaldi and Perrier, 1982Go), the low systemic blood concentrations of BPA following oral administration as compared to other routes of administration can be referred to as incomplete absorption of BPA. However, several factors beyond incomplete absorption from the gastrointestinal tract probably contributed to the low systemic bioavailability observed for BPA following oral administration, including first-pass metabolism by the intestine and/or liver as well as intestinal secretion. In the current study, the time required for the appearance of radioactivity in the feces was longer than the typical transit time for the movement of GI tract contents in the rat (data not shown). This indicates that first-pass metabolism, not incomplete absorption from the intestine, is the likely explanation for the low systemic bioavailability of BPA. This high metabolic clearance by first-pass metabolism is likely to be mediated by both intestinal and hepatic enzymes as the capacity of the intestinal conjugating enzymes toward phenol has been reported to exceed the contribution of the hepatic enzymes (Cassidy and Houston, 1984Go). The likely contribution of intestinal UDP-glucuronosyl transferase (UGT) is supported by the significantly greater systemic blood concentrations of BPA following ip administration as compared to oral administration (Fig. 1Go, Table 4Go). The rapid and markedly higher systemic blood concentrations of BPA following sc or ip dosing (as compared to oral) indicates a considerably higher bioavailability of BPA following dosing by these routes. Thus, extrapolation of the effects observed following ip or sc administration of BPA to other routes of administration more relevant to potential human exposure (i.e., oral) is tenuous, as the bioavailability of both parent BPA and plasma radioactivity were clearly route-dependent.

This route dependency in the relative bioavailability likely offers a pharmacokinetic explanation for apparent route differences in estrogenic potency (i.e., uterine changes) that have been observed for BPA in female rats. Specifically, recent studies have demonstrated a clear route dependency in the magnitude of the uterotrophic response to BPA, where significantly larger oral doses were required to produce an estrogenic response than those required subcutaneously (Laws and Carey, 1997Go; Twomey, 1998aGo and bGo). The route dependency in parent compound bioavailability explains the route dependency in the uterotrophic response to BPA based on the assumption that free parent is responsible for the estrogenic effects of BPA. As the difference across routes of administration was less for the plasma concentration of radioactivity (shown to be principally the monoglucuronide conjugate of BPA), any estrogenic activity expressed by this major metabolite may erode the previous conclusion. Recently, however, it has been reported that the monoglucuronide of BPA was not a ligand in vitro for either the {alpha} (human) or ß (mouse) estrogen receptor, nor did it have estrogenic activity as measured in a reporter cell assay that measures gene expression from chimeric estrogen receptors transiently transfected into MCF-7 cells (Matthews and Zacharewski, 1999Go).

The metabolic fate of BPA in rats was also found to be dependent upon the route of administration, as both quantitative and qualitative differences were observed for the circulating metabolites in blood. Oral administration resulted in the largest fraction of plasma glucuronide and sc administration the smallest fraction of plasma glucuronide (68–100% and 17–49% of plasma radioactivity for oral and sc administration, respectively). The fraction of plasma radioactivity represented by unchanged parent compound was also route dependent, with oral administration resulting in the smallest fraction of parent material at Cmax and sc administration the largest fraction of parent material at Cmax (2–8% and 65–76% of plasma radioactivity for oral and sc administration, respectively). Qualitatively, the monosulfate of BPA was present in both plasma and urine only following ip administration, and an unresolved area of peaks (43–71% of NQ-parent plasma radioactivity) was present only following ip or sc dosing. However, these route-specific metabolites, with unknown potential for biologic activity, represented less than 1% of the administered dose at the postdosing time these specimens were collected.

The fraction of the administered dose excreted via the urine or feces was dependent upon sex, without, however, any marked sex differences in the urinary profiles of metabolites (data not shown). One possible explanation is that female rats, with a larger fraction of the administered dose in the systemic circulation (Fig. 2Go and Table 4Go), had a larger amount of the administered dose available for renal excretion. Alternately, sex differences in the conjugation of BPA is a possible explanation, as such differences have been reported for other compounds (Mulder, 1986Go). Examples include hepatic phenolic UGT (male activity is twice that of females; Zhu et al., 1996) and rat hepatic 17ß-estradiol glucuronidation (female activity greater than male; ibid). Another potential mechanism for the sex difference in kinetics could be sex differences in extrahepatic glucuronidation, either by the intestine or kidney, as has been previously reported (Rush et al., 1983Go).

Following ip administration, the mean Cmax values for 14C-BPA-derived radioactivity in the plasma appeared to be dependent upon the dose administered in both sexes, as evidenced by a disproportionate increase in Cmax values at the 100 mg/kg dose (Table 4Go). These data suggest that hepatic metabolism was not rate-limiting to the elimination of plasma radioactivity, as nonlinearity was not as apparent for parent compound (Table 4Go). One possible explanation for this route-dependent nonlinearity across doses might be saturation of biliary excretion. Thus, once the rapidly absorbed ip dose was metabolized in the liver, the excess metabolites (principally the monoglucuronide of BPA) were transferred into the systemic circulation, rather than being eliminated via bile, causing the disproportionate rise in Cmax.

Once the initial phase of absorption and elimination were completed, it appears that 14C-BPA-derived radioactivity demonstrated similar kinetics across these routes of administration, doses, and sex. By about 18–24 h postdosing, all the 100 mg/kg animals showed similar levels of plasma radioactivity, as did all low dose groups, decreasing at similar rates via an apparent first-order process (data not shown). Either binding to plasma proteins or enterohepatic circulation of 14C-BPA-derived radioactivity may explain the longer elimination phase of plasma radioactivity as compared to parent compound.

In summary, the bioavailability of both 14C-BPA-derived plasma radioactivity and BPA demonstrated a clear dependence on the route of administration with oral administration resulting in the lowest bioavailability of both parent compound and 14C-BPA-derived radioactivity. To a lessor extent, the pharmacokinetics of BPA also appeared dependent upon the sex of the animal and dose administered. Hence, extrapolation of the effects observed following ip or sc administration of BPA to other routes of administration more relevant to potential human exposure (i.e., oral) is tenuous, as both BPA bioavailability and metabolism were clearly route dependent.


    ACKNOWLEDGMENTS
 
The authors would like to thank K. Gibson, C. Thornton, J. Ormand, H. Denome, J. Whalen, T. Miller, and S. Luptowski for excellent technical assistance and M. Dryzga for application of his exceptional surgical skills and laboratory supervisory responsibilities. J. Lacher's veterinary skills were invaluable in the completion of this study. This study was sponsored by the Bisphenol A Global Industry Group, organized through The Society of the Plastics Industry, Inc., Washington, D.C., USA, and the European Chemical Industry Council, Brussels, Belgium.


    NOTES
 
1 To whom all correspondence should be addressed at Toxicology and Environmental Research and Consulting, The Dow Chemical Company, Building 1803, Midland, MI 48674. Fax: (517) 638–9863. E-mail: jwaechter{at}dow.com. Back

Presented in part at the 1997 and 1998 Society of Toxicology Annual Meetings, abstract numbers 729 and 1386, respectively.


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIAL AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
BIBRA (1989). Toxicology Profile Bisphenol A. The British Industrial Biological Research Association, Carshalton, Surrey, U. K.

Bitman, J., and Cecil, H. C. (1970). Estrogenic activity of DDT analogs and polychlorinated biphenyls. J. Agric. Food Chem. 18: 1108–1112.

Bond, G. P., McGinnis, P. M., Cheever, K. L., Harris, S. J., Platnick, H. B., and Neimeier, R. W. (1980). Reproductive effects of bisphenol A. Toxicologist 19, A23 (Abstract only).

Capel, I. D, French, M. R., Millburn, P., Smith, R. L., and Williams, R. T. (1972). The fate of 14C-phenol in various species. Xenobiotica 2, 25–34.[ISI][Medline]

Cassidy M. K., and Houston, J. B. (1984). In vivo capacity of hepatic and extrahepatic enzymes to conjugate phenol. Drug Metab. Disp. 12, 619–624.[Abstract]

Dodds E. C. and Lawson W. (1936). Synthetic estrogenic agents without the phenenthene nucleus. Nature 137, 996.

Gaido, K. W., Leonard, L. S., Lovell, S., Gould, J. C., Babai, D., Portier, C. J., and McDonnell, D. P. (1997). Evaluation of chemicals with endocrine modulating activity in a yeast-based steroid hormone receptor gene transcription assay. Toxicol. Appl. Pharmacol. 143, 205–212.[ISI][Medline]

Gibaldi, M., and Perrier, D. (1982). Pharmacokinetics, Second Edition, Revised and Expanded, Marcel Dekker, Inc. New York, New York. pp. 167–169.

Guide for Care and Use of Laboratory Animals (1985). US Dept. of Health and Human Services, National Institutes of Health, Publ. No. 85–23.

Harms, P.G., and Ojeda, S.R. (1974). A rapid and simple procedure for chronic cannulation of the rat jugular vein. J. Appl. Physiol. 36, 391–392[Free Full Text]

Knaak, J. B. and Sullivan, L. J. (1966). Metabolism of bisphenol A in the rat. Toxicol. Appl. Pharmacol. 8, 175–184.[ISI][Medline]

Krishnan, A. V., Stathis, P., Permuth, S. F., Tokes, L., and Feldman, D. (1993). Bisphenol A: An estrogenic substance is released from polycarbonate flasks during autoclaving. Endocrinology 132, 2279–2286.[Abstract]

Kuiper, G. G. J., Carlsson, B., Grandien, K., Enmark, E., Haggblad, J., Nilsson, S., and Gustafsson, J-A. (1997). Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors {alpha} and ß. Endocrinology 138(3), 863–870.

Laws, S. C., and Carey, S. A. (1997). Comparison of the estrogenic activity of 4-tertoctylphenol, nonylphenol, bisphenol A and methoxychlor in Long Evans female rats. Toxicologist 36, 359.

Matthews, J. B., and Zacharewski, T. R. (1999). Comparison of the interactions of bisphenol A and bisphenol A glucuronide with estrogen receptor {alpha} and ß. International Symposium on Environmental Endocrine Disruptors '99, Kobe, Japan. Abstract B18.

Morrissey, R. E., George, J. D., Price, C. T., Tyl, R. W., Marr, M. C., and Kimmel, C. A. (1987). The developmental toxicity of bisphenol A in rats and mice. Fundam. Appl. Toxicol. 8, 571–582.[ISI][Medline]

Mulder, G. J. (1986). Sex differences in drug conjugation and their consequences for drug toxicity. Sulfation, glucuronidation and gluthatione conjugation. Chem. Biol. Interact. 57, 1–15.[ISI][Medline]

Organisation for Economic Co-Operation and Development (OECD), The OECD Principles for Good Laboratory Practice as specified by the European Economic Community, Council Directive 87/18 EEC.

Pfeiffer, E., and Metzler, M. (1998). In vitro metabolism of bisphenol A. Toxicologist 42, 279.

Reel, J. R., George, J. D., Lawton, A. D., Myers, C. B., and Lamb, J. C. (1985). Unpublished report entitled "Bisphenol A: Reproduction and Fertility Assessment in CD-1 Mice When Administered in the Feed," dated May 31, 1985, for the National Toxicology Program from Research Triangle Institute, RTP, NC [NTIS PB86–103207].

Rush, G. F., Newton, J. F., and Hook, J. B. (1983). Sex differences in the excretion of glucuronide conjugates: the role of intrarenal glucuronidation. J. Pharmacol. Exp. Ther. 227, 658–662.[Abstract]

Smith, R. A., Christenson, W. R., Bartels, M. J., Arnold, L. L., St. John, M. K., Cano, M., Garland, E. M., Lake, S. G., Wahle, B. S., McNett, D. A., and Cohen, S. M. (1998). Urinary physiologic and chemical metabolic effects on the urothelial cytotoxicity and potential DNA adducts of o-phenylphenol in male rats. Toxicol. Appl. Pharmacol. 150, 402–413.[ISI][Medline]

Twomey, K. (1998a). Bisphenol A: Uterotropic assay in immature rats (subcutaneous dosing). Report to the Bisphenol A Global Industry Group from Central Toxicology Laboratory, Cheshire, U. K. [CTL Report No. CTL/P/5943].

Twomey, K. (1998b). Bisphenol A: Uterotropic assay in immature rats (oral dosing). Report to the Bisphenol A Global Industry Group from Central Toxicology Laboratory, Cheshire, U. K. [CTL Report No. CTL/P/6029].

U.S. Department of Health and Human Services, Food and Drug Administration-GLPS. Title 21 CFR, Part 58-Good Laboratory Practice Regulations for Nonclinical Laboratory Studies, Final Rule.

U.S. Environmental Protection Agency-TSCA GLPS, Title 40 CFR, Part 792-Toxic Substances Control Act (TSCA); Good Laboratory Practice Standards, Final Rule.

Zhu, B. T., Suchar, L. A., Huang, M-T., and Conney, A. H. (1996). Similarities and differences in the glucuronidation of estradiol and estrone by UDP-glucuronosyltransferase in liver microsomes from male and female rats. Biochem. Pharmacol. 51, 1195–1202.[ISI][Medline]