Document 4JNj9B79x29orLDg6E64z6q4R

OXICOLOGY AND APPLIED PHARMACOLOGY 99, 395-414(1989) Physiologically Based Pharmacokinetic Modeling of the Pregnant Rat: A Multiroute Exposure Model for Trichloroethylene and Its Metabolite, Trichloroacetic Acid Jeffrey W. Fisher, Temistocles A. Whittaker, Douglas H. Taylor,* Harvey J. Clewell III, and Melvin E. Andersen Hazard Assessment Branch, Toxic Hazards Division, Harry G. Armstrong Aerospace Medical Research Laboratory, Wright-Patterson Air Force Base, Ohio 45433-65 73, and*Miami University, Oxford, Ohio45056 Received July 18, 1988; accepted March 28,1989 Physiologically Based Pharmacokinetic Modeling of the Pregnant Rat: A Multiroute Expo sure Model for Trichloroethylene and its Metabolite, Trichloroacetic Acid. Fisher, J. W., Whittaker, T. a., Taylor, D. H., Clewell, H. J. Ill, and Andersen, M. E. (1989). Toxicol, Appl. Pharmacol., 99, 395-414. A physiologically based pharmacokinetic (PB-PK) model was developed to describe trichloroethylene (TCE) kinetics in the pregnant rat exposed to TCE by inhalation, by bolus gavage, or by oral ingestion in drinking water. The kinetics oftrichloro acetic acid (TCA), an oxidative metabolite of TCE, were described by a classical one-compart ment pharmacokinetic model. Among the required model parameters for TCE, partition co efficients (PCs) and kinetic constants for oxidation were determined by vial equilibration and gas uptake methods, respectively. The fat:blood PC was 33.9; the blood:air PC was 13.2; and the fetal tissue:fetal blood PC was 0.51. TCE was readily metabolized with high substrate affinity. In naive and pregnant female rats the maximum velocities of oxidative metabolism were 10.98 0.155 and 9.18 0.078 mg/kg/hr, while the estimated Michaelis constant for the two groups of rats was very low, 0.25 mg/liter. The first-order rate constant for oral absorption ofTCE from water was 5.4 0.42/fir"1 in naive rats. With TCA, the volume of distribution (0.618 liter/ kg) and the plasma elimination rate constant (0.045 0.0024/hour) were estimated both from intravenous dosing studies with TCA and from an inhalation study with TCE. By comparison of the two routes of administration, the stoichiometric yield of TCA from TCE was estimated to be 0.12 in pregnant rats. To develop a data base for testing the fidelity of the PB-PK model, inhalation and bolus gavage exposures were conducted from Day 3 to Day 21 of pregnancy and a drinking water exposure from Day 3 to Day 22 of pregnancy. Inhalation exposures with TCE vapor were 4 hr/day at 618 ppm. The TCE concentration in drinking water was 350 ng/ml and the gavaged rats received single daily doses of 2.3 mg TCE/kg, Time varying physiological parameters for compartment volumes and blood flows during pregnancy were obtained from the published literature. Using the kinetic parameters determined by experimentation, TCE concentrations in maternal and fetal blood and TCA concentrations in maternal and fetal plasma were predicted from the PB-PK model by computer simulation and compared favorably with limited data obtained at restricted time points during pregnancy for all three routes of exposure. On the basis of the PB-PK model, fetal exposure to TCE, as area-under-the-curve, ranged from 67 to 76% of maternal exposure. For TCA the fetal exposure was 63 to 64% of the maternal exposure. The fetus is clearly at risk both to parent TCE and its TCA metabolite. With further validation, PB-PK modeling ofexpected fetal exposure should prove helpful in the design and interpretation of teratology and reproductive toxicology studies with a variety of volatile chemicals. 1989 Academic Press. Inc. 395 0041-008X/89 $3.00 Copyright 1989 by Academic Press, Inc. Ail rights of reproduction in any form reserved. 396 FISHER ET AL. Trichloroethylene (TCE), a widespread envi ronmental contaminant, has been listed by the U.S. Environmental Protection Agency as a "possible human carcinogen" (U.S. EPA, 1987) and its oxidative metabolite, trichloro acetic acid (TCA), is a recognized carcinogen in rodents (Elcombe, 1985; Herren-Freud et al,, 1987). In addition to cancer, a large num ber of other toxic endpoints have been exam ined with this important commercial solvent (U.S. EPA, 1985, 1987). Teratogenic studies with rats have either been negative (Beliles et al, 1980) or demonstrated only slight devel opmental delays (Dormueller et al., 1979). In contrast, Taylor el al. (1985) observed irre versible deficits in locomotor activity in rat pups bom to dams that had been exposed to TCE in drinking water during pregnancy. Rat pups exposed to TCE in utero also exhibited reduced brain uptake of 2-deoxyglucose (Noland-Gerbec et al., 1986). These studies clearly indicated substantial effects ofTCE or its metabolites on the developing fetus. By themselves, however, these studies do not suffice to estimate fetal exposure in these par ticular dosing situations. Only the maternal dose regimens were specified and no mea surements of fetal concentrations of TCE and/or its metabolites were obtained. In an effort to more accurately predict fetal dosime try, we began development of generic phar macokinetic models for the pregnant and the lactating rat (Fisher et al,, 1987) and applied these models to the study of TCE and TCA, its stable, persistent oxidative metabolite. A complication encountered in pharmaco kinetic modeling in pregnancy is the time-de pendence of compartment volumes, blood flows and metabolic parameters at various stages of pregnancy. Most ofthe recent physi ologically based pharmacokinetic (PB-PK) models developed for other volatile chemi cals have been for animals where these physi ological parameters were considered to be constant over time. In constructing a PB-PK model for TCE in pregnancy, the increase in maternal body weight, in fetal and placental weight, and in blood flows to various regions during pregnancy can be found in the physio logical literature. Other mode! constants, re lated to tissue partitioning and rates of me tabolism of TCE, have to be measured di rectly by suitable experimentation. In the present work we have developed a PB-PK structure for TCE in pregnancy which has seven physiological compart ments, including placenta, fetus, and mam mary tissue. In addition, TCA kinetics were linked to TCE oxidation and were modeled by a hybrid approach which included the dam as a single compartment for TCA (Na tional Research Council, 1986). Blood flow carried TCA to the placenta where the TCA diffused into the fetal circulation. The overall model was developed from literature for physiological variables and from appropriate metabolic experiments in naive and pregnant female rats. The model was then used to pre dict maternal and fetal concentrations of TCE and TCA expected following three types of repeated exposures with the pregnant rats. These exposures were inhalation of 618 ppm TCE, ingestion of drinking water containing TCE, and daily single dose gavage with water containing TCE. The adequacy ofthe model ing approach to predict TCE and TCA dispo sition during pregnancy was inferred from the correspondence between prediction and observed concentrations in limited numbers of observations from pregnant rats and fe tuses killed at restricted times during gesta tion. METHODS Animats. Female, cesarean derived Fischer-344 rats (170-200 g), obtained from Charles River Breeding Lab oratory (Kingston, NY) were used for kinetic constant determinations. Pregnancy was initially determined by vaginal smears and by observing for the presence of sperm. Timed-pregnant female cesarean derived Fischer344 rats were delivered to the laboratory from Harlan Sprague Dawley (Indianapolis, IN) on Day 3 of preg nancy. These rats were used for the acute inhalation ex posure and the subchronic exposures. All rats were kept in separate cages and allowed access to commercial rat chow (Purina rat chow) and water ad libitum. PB-PK MODELING IN PREGNANCY 397 Chemicals. TCE (>99%) and TCA (98%) were ob tained from Aldrich Chemical Co. A TCA working stock for gas chromatographic standards, prepared by dissolv ing TCA in physiological saline prior to experimentation, was maintained as a working stock for 2 weeks. TCE stock solutions for gas chromatographic standards were :>t stored, but were prepared immediately prior to ex perimentation. TCE and TCA analyses. Heparinized blood samples for TCE analysis (0.1 ml) were dispensed into 1.5-ml glass Teflon-coated septa screw cap vials containing 1.0 ml of n-hexane. These vials were shaken for 1 hr on a Buchler Vortex evaporator at ambient temperature (2325*C) before gas chromatographic analysis ofthe TCE in n-hexane. A Model 5890A Hewlett-Packard gas chromatograph equipped with a polar DB-17 30 m capillary column ` i.25 mm i.d.), an electron capture detector, and an auto matic sampler was used for TCE analysis. The oven tem perature was 70"C, injector temperature 125'C, and de tector temperature 300C. The column split ratio was five and the argon/methane carrier flow was 0.50 ml/min through the column. Extraction efficiency ofTCE into nhexane from maternal and fetal blood was greater than 93%. The limit ofdetection for TCE was 0.03 Mg/ml and the retention time was 3.4 min. Heparinized maternal and fetal blood samples (0.2 ml each) for TCA analysis were placed in capillary blood se rum separators and centrifuged for 0.5-1.0 min in a Brinkman 3200 microcentrifuge. The supernatant (plasma) was collected and either frozen or prepared for immediate gas chromatographic analysis. Frozen sam ples were analyzed within 5 days. The TCA assay proce dure included a methylation step in order to use gas chro matography for detection. Fifty microliters ofeither fresh or thawed plasma was placed in a 1.5 glass vial containing 0.1 ml ofchilled 3N methanol ic HQ and then 1.0 ml of n-hexane was added. These samples were incubated for 30 min at 100'C prior to gas chromatographic analysis of the TCA methyl ester in n-hexane. Gas chromatographic conditions for TCA methyl ester analysis were the same as for TCE analysis except that the oven temperature was 100*C. Extraction ofTCA methyl ester from the fetal and maternal plasma into n-hexane was greater than 91%. The limit ofdetection for the TCA methyl ester was 0.07 Ug/ml and the retention time was 6.5 min. TCA kinetic constants by IV dosing. Four cannulated rats (Day 14-15 of pregnancy) were given 4.0 mg TCA/ kg in saline intravenously. Blood samples were collected from jugular cannulae at 1, 3, 10,22,28, and 47 hr post exposure. The plasma TCA elimination rate constant for the pregnant rat was calculated from log-linear regression of the TCA plasma concentration over time. Volume of distribution (liter/kg) was estimated by dividing the dose (mg/kg) ofTCA by the initial TCA plasma concentration (mg/liter, Y intercept at time zero). Single 4-hr TCE inhalation exposure. A single 4-hr in halation exposure to 600.4 ppm TCE (time weighted av erage, TWA) was conducted with six jugular cannulated Day 12 pregnant rats. The inhalation chamber, a six compartment ptexiglas container, and the cannulation procedure have been described by McDougal et al. (1985). Atmospheric levels of TCE in the chamber were monitored every 5 min with a 5880 Hewlett-Packard gas chromatograph equipped with a 6-ft J-in.-i.d. column containing 3% SE 30 on 80/100 mesh Supelcoport and an injector temperature of 200`C, oven temperature of 100'C, and FIDdetcctortemperature of250*C. The TCE retention time was 0.84 min. Blood samples (about 0.3 ml) were collected from the cannulae using 1.0 heparin ized syringes at 0.083,0.50, 1.0, 1.5, and 2.0 hr postexpo sure for TCE and TCA analyses as described earlier. Ad ditional samples (0.2 ml) were taken at 5.0, 21.0, 29.0, and 46.0 hr postexposure for TCA analysis. This expo sure permitted calculation ofthe yield ofTCA from TCE oxidation. Subchronic inhalation exposure: Seven pregnant rats (Day 3 of pregnancy) were exposed to 618 ppm TCE (TWA) in a 31 liter battery jar inhalation exposure sys tem (Leach, 1963; Andersen et al., 1984) 4 hr/day, 5 days/week for 3 weeks. In each exposure, atmospheric concentrations of TCE were analyzed every 5 min by a gas chromatograph (Model 5880A, Hewlett Packard) equipped with a flame ionization detector and an auto matic sampling valve. A 10-ft J-in.-o.d. stainless column was used containing 3% SE 30 on 80/100 mesh Supelco port. The injection temperature was I25*C, the detector temperature, 300*C, and the oven temperature, 80*C. The retention time for the TCE was 2.3 min. On Day 20 of pregnancy, maternal and fetal blood was collected for TCA analysis from three rats exposed to TCE on the pre vious day (20 hr postexposure). TCE and TCA analyses were also conducted on blood samples collected from four rats immediately after the 4-hr exposure. Pregnant rats were killed by cervical dislocation immediately after removal from the exposure chamber. The abdomen was opened and 0.3 ml of maternal blood was collected with a heparinized syringe from the inferior vena cava. Con currently, fetuses were decapitated and a total volume of 0.3 ml of fetal blood was obtained with heparinized capillary tubes from three to five fetuses. This exposure was used to estimate the transfer coefficients for TCA across the placenta. Subchronic drinking water exposure. Ten pregnant rats (Day 3 of pregnancy) were provided drinking water containing 350.0 9.5 (SE) pg of TCE/ml of water, 5 days per week for 3 weeks. Five pregnant rats received distilled water and served as controls. Fresh TCE-water solutions (three 2 liter flasks) were prepared daily, by add ing 1.0 ml of pure TCE to 2 liters of distilled water in a stoppered flask and stirring overnight. The TCE-water solutions from the flasks were combined in a 5 liter glass jar, stirred again for I hr, and dispensed into a 125-ml amber glass drinking water bottles, then capped with a nalgene stopper containing a sipping tube. SL 350H 398 FISHER ET AL. The initial TCE concentration in a portion ofthe water bottles was measured daily and averaged over 3 weeks. To determine the loss of TCE from the drinking water, the same drinking water bottles were sampled again 24 hr later. The average loss ofTCE from the drinking water bottles in the 24-hr period, on the basis of weekly aver ages, was slightly greater than 50% ofthe initial TCE con centration. TCE-water consumption was tabulated each day yielding an average consumption of 14.5 ml/day/rat for a 3-week period. Control rats drank an average of 16.9 ml distilled water/day/rat for the same 3-week period. In one case maternal blood was collected from four rats as early as Day 18 of pregnancy (1000-1100 hours) for TCA analysis. On Day 21 of pregnancy (1000-1100 hr) maternal and fetal blood were collected for TCE and TCA analyses from six rats. Subchronicgavage exposure. Nine pregnant rats (Day 3 of pregnancy) were dosed by bolus gavage with 2.3 mg TCE/kg in water (0.8 to 1.5 ml) 5 days/week for 3 weeks. Three pregnant rats were gavaged with distilled water and served as controls. Gavage animals were dosed with the TCE-water mixture used in the drinking water study. All aliquots of the TCE-water mixture were analyzed to quantify the amount of TCE in the water. On Day 20 of pregnancy, maternal and fetal blood was collected from four rats killed immediately after gavage for TCE and TCA analyses. Three hours after gavage another five rats were killed for TCA analysis. Drinking water and gavage animals were killed and blood was collected in the same manner as with the inhalation animals. Gas uptake. Gas uptake techniques have been used to assess in vivo kinetic constants for metabolism (Ftlser and Bolt, 1979; Andersen etal.. 1980; Gargas et al.. 1986a,b). Atmospheric concentrations of TCE in the gas uptake chamber were monitored by a gas chromatograph (Model 5890A, Hewlett-Packard) equipped with an auto matic sampling valve, flame ionization detector and a 10- ft j-in.-o.d. stainless steel column (3% SE30 on 80/100 mesh Supelcoport). Injection temperature was 125*C, detector temperature, 300*C, and oven temperature, 100`C. Atmospheric TCE samples were taken at 5 min after injection ofTCE into the chamber atmosphere and then every 10 min for the duration of exposure (1.5-6.0 hr). The retention time for TCE was 2.5 min. Experimental exposures for kinetic constant determi nations were conducted with nonpregnant, and Day 13- 14 pregnant rats. Nonpregnant and pregnant rats (four per concentration) were exposed to initial TCE concen trations of 5075, 2200, 1100, and 111 ppm; and 2050, 1005, and 103 ppm respectively. the maximum rate of metabolism (mg/hr) for a 1.0-kg animal (allome- trically scaled) was estimated by computerized nonlinear least-squares techniques (Simusolv, Dow Chemical Co.). The pregnancy model (Fig. 1; Appendix II) for Day 14 of pregnancy was used to obtain statistical best fit estimates of while the four-compartment model structure (Ramsey and Andersen, 1984) was used for fitting TCE i real Fig. 1. Physiologically based pharmacokinetic preg nancy model used to describe the disposition of trichlorethylene (TCE) and the compartmental model used to describe the disposition of trichloroacetic acid (TCA). TCE enters the body by inhalation or by oral ingestion (gavage or drinking water). TCA is formed from metabo lism ofTCE in the liver. Details ofthe pregnancy model are found in the Appendices. with the naive rats. Physiologic parameters for the preg nancy model (Fig. 1) are discussed later under Methods (Physiological parameters)and reported in Table 1, while TCE kinetic constants are reported in Table 2. Appendi ces I and II contain pregnancy model nomenclature and mathematical equations. First-order oral uptake rate constant for TCE. Three adult naive rats were prepared with jugular cannulas and gavaged with 7.6 mg TCE/kg in a total volume of4 ml of water. Heparinized blood samples were collected at 2, 5, 10,20, and 30 min post-gavage. The TCE-water solution for the oral uptake study was prepared by mixing TCE and distilled water overnight in a 25-ml Teflon sealed glass vial with a stirbar and magnetic mixer. The firstorder rate constant (K,,) for oral absorption TCE was de termined by using Simusolv to optimize the naive rat model within the constraints of the model. This calcu lated K,, value for TCE was then used to describe repeated gavage dosing ofthe pregnant rats. Partition coefficients. TCE tissue/air partition coeffi cients were determined for blood, muscle, fat, and liver WSL 035012] -> `4'"' wm PB-PK MODELING IN PREGNANCY 399 using virgin females and for blood, placenta, fetal blood, and fetal tissue using near term pregnant females. The TCE mammary tissue/air partition coefficient was deter mined using an early postpartum lactating rat The vialequilibration method (Sato and Nakajima, 1979; Gargas et at.. 1989) was employed to measure tissue solubilities because of its simplicity and the large existing data base for other volatile organics developed using this method (Gargas et at.. 1989). Tissue/air TCE partition coeffi cients were divided by the blood/air partition coefficient to obtain the tissue/blood partition coefficient. A PerkinElmer 3920 gas chromatograph equipped with a 6-ft Jin.-o.d. stainless steel column containing 3% SP 2100 was used for TCE analyses. The oven temperature was 100'C, the injector temperature 175"C, and the detector, 250*C. The retention time for TCE under these conditions was 0.34 min. TCA is a nonvolatile chemical unsuited for vial equili bration methods. Tissue/saline TCA partition coeffi cients were determined for blood and placenta using a centrifugation technique (Jepson, 1986). The placenta/ saline TCA partition coefficient was divided by the blood/saline partition coefficient to estimate the pla centa/blood partition coefficient. Physiological parameters. Certain time dependent physiological changes that occur during pregnancy (Ta ble 1) were incorporated into the pregnancy model by using the table function ofthe computer simulation soft ware, Advanced Continuous Simulation Language (ACSL) (Mitchell and Gauthier, 1981). All tissues were assumed to have unit density. Physiologic constants that vary with time included dam body weight gain and growth of the placenta, fetus, mammary, and fat tissues. Accompanying increases in maternal blood supply to the placenta and mammary tissues and fetal blood supply to the placenta tissue were also taken into account. Fetal growth from Days 14 to 22 ofgestation was described by FW (fetal weight, grams) = [a{t - 13)]\ (1) where a is a specific growth constant (0.17) and (is gesta tion time in days (Huggett and Widdas, 1951). Fetal weight before Day 13 was considered to be insignificant. At birth, pup weight (3.5-4.1 g) agreed favorably with the predicted pup weight of 3.6 g. The (t~ 11) time function reported by these authors was adjusted to (t - 13) to ac count for an increase in gestation time of 2 days for the rats used in this study. Placental growth from Days 14 to 22 of gestation was described by Lpu (placenta) - 0.102 exp[0.217(r - 13)] (ml), (2) where t is the gestation time in days (Olanoffand Ander son, 1980). Placenta weights (detached disks, 0.45 to 0.50 g) on Day 20 of pregnancy compared favorably to the calculated weight. Maternal blood flow to the placenta on Days 14 to 22 of gestation was described as a function of placenta weight as shown by Qtu (maternal placental blood flow) = Fpu-75.0 (ml/hr), (3) where 75,0 is the placenta flow rate constant (ml/hr/g) (Olanoffand Anderson, 1980). Fetal blood flow to the placenta 2rel) on Days 14 to 22 of gestation was 25% of the maternal blood flow to the placenta (Eq. 3). This estimate of fetal blood flow was derived from the pregnancy model developed by Olanoff and Anderson (1980). The fetal growth (Eq. 1), placental growth (Eq. 2) and maternal blood flow to the placenta (Eq. 3) were calcu lated for each day of gestation (Days 14-22) and multi plied by a factor of 7.0 to reflect the observed average litter size for these studies. Placenta weight (volume) was expressed as a percentage of maternal body weight and ranged from 0.4% on Day 14 ofgestation to 2.9% ofbody weight on Day 22 ofgestation. The calculated fetal litter weight for seven fetuses ranged from 0.034 g on Day 14 ofgestation to 25.1 g on Day 22 ofgestation. Fetal weight and was not scaled to maternal body weight. The effect oflitter size on TCE and TCA kinetic profiles in the dam and fetus is discussed under Discussion. Maternal blood flow to the placenta (for the litter) was expressed as a per centage of cardiac output and ranged from 1.5% of car diac output on Day 14 ofgestation to 9.9% on Day 22 of gestation. Fetal blood flow (for the litter) to the placenta was expressed as a percentage (25%) of maternal placen tal blood flow and ranged from 0.37% of maternal car diac output on Day 14 ofgestation to 2.5% on Day 22 of gestation. Mammary tissue was assumed to reach 21 % ofits total growth by Day 10 of pregnancy and 59% by Day 20 of pregnancy (Griffith and Tumer, 1961). Mammary tissue weight, expressed as a percentage ofbody weight, was as signed values of 2, 2.7, 3.2, and 4.4% at Days 3, 12, 14, and 22 of pregnancy (Knight et al.. 1984). Maternal blood flow to the mammary tissue was described as a lin ear increase (1 to 9% of cardiac output) from Day 3 to Day 22 ofgestation (Hanwell and Linzell, 1973). Fat accumulation during pregnancy was described ac cording to Naismith et al. (1982). Body fat values, ex pressed as a percentage of body weight, were 6,6, 8, and 12% on Days 3,9, 16, and 22 ofgestation, respectively. Maternal weight gain during pregnancy was estimated by subtracting the calculated placental and fetal weights from the measured weight of the pregnant rat. Maternal weight gain was described as a linear function increasing by 10% of the initial weight (Day 3 of gestation) by Day 22 of gestation. Volume of the liver, slowly perfused and richly perfused tissue groups were given values of 4%, 70.6 to 59.7%, and 8%, of body weight. Blood flow to the liver, fat, slowly perfused and richly perfused tissue groups were respectively 25, 9, 15, and 48.1 to 29.6% of cardiac output. st* 400 FISHER ET AL. TABLE 1 Physiological Constants Used in the PB-PK Models for the Naive and Pregnant Rat Naive dam Pregnant dam Body weights (kg) Acute inhalation Subchronic inhalation Single gavage Subchronic gavage Subchronic drinking water -- -- 0.170 -- -- 0.192 0.187-0.206" -- 0.144-0.158" 0.168-0.185" Percentage ofbody weight Liver Richly perfused Slowly perfused Fat Mammary tissue Maternal/fetal placenta'' Fetal litter* Flows (liter/hr) Alveolar ventilation Cardiac output 4.0 5.0 76.0 6.0 -- -- -- 4.0 4.0 70.6-59.7 6.0-12.0 2.0-4.4 0.4-2.9 0.034-25. Ig 14.0* body wt074 14.0-body wt074 19.9-body wt074 14.0-body wt0,74 Percentage ofcardiac output Liver Richly perfused Slowly perfused Fat Mammary tissue Maternal placenta* Fetal placenta* 25.0 51.0 15.0 9.0 -- -- -- 25.0 48.1-29.6 15.0 9.0 1.0-9.0 1.5-9.9 0.37-2.5 " Measured initial body weight on Day 3 of pregnancy and predicted maternal body weight on Day 22 ofgestation (less fetuses and placentas). * Litter size equals seven. RESULTS Partition coefficients. The blood:air parti tion coefficient (PC) was lower in the preg nant rat (13.2 0.34) than in the naive fe male rat (15.0 0.35). It was lower still for fetal blood (9.6 0.33). TCE partitioned into tissues as expected for a chemical that is mod erately lipophilic (Table 2). Rates ofmetabolism. When estimating the kinetic constants from the gas uptake experi ments (Figs. 2A and 2B), both and Km for TCE metabolism were initially allowed to vary. The maximum rate of metabolism was significantly higher in naive rats (11.5 0.04 mg/kg/hr) than in pregnant rats (9.36 0.14 mg/kg/hr). The estimated Km values in these two groups of rats were, respectively, 0.37 0.02 and 0.50 0.11 mg/liter. To obtain a consistent description of the gas uptake curves with naive female rats it was necessary to include a first-order metabolic component with a rate constant of 3.6 0.2 hr-1 (Gargas et al., 1986a). The low values of Km indicate that TCE metabolism is flow limited in these rats (Andersen, 1981) and, as expected, Km had relatively little impact on the simulations when set at values below about 0.5 mg/liter. PB-PK MODELING IN PREGNANCY 401 TABLE2 Kinetic Constants for Modeling Trichloroethylene and Trichloroacetic Acid in the Naive and Pregnant Rat Naive dam Pregnant dam Partition coefficients TCE Blood/air Liver/blood Rapidly perfused/blood Slowly perfused/blood Fat/blood Mammary tissue/blood Placenta/blood Fetal blood/air Fetal tissue/fetal blood TCA Placenta/matemal blood Metabolic constants TCE V(mg/hr) " K,, (mg/liter) K(hr')* (fetal litter, ml) PO (unitless) TCA Pd (maternal, liter)e Pd (fetal litter, ml) AT (hr-1)4' PI (liter/hr) P2 (liter/hr) 15.00 1.46 1.46 0.46 29.83 -- -- -- -- -- 3.02 0.25 7.08 -- -- -- -- -- -- -- 13.20 1.66 1.66 0.52 33.90 4.57 0.52 9.60 0.51 0.74 2.53 0.25 0.0 0.034-25.1 0.12 0.185 0.034-25.1 0.076 0.0013 0.0020 * Body weight fixed at 0.175 kg, P'mtt (pregnant rat) = 9.18- body wt0-74 and PTM, (naive rat) " 10.98 body wt74 b Body weight fixed at 0.175 kg, K = 4.2/body wt.0-3 ' Body weight fixed at 0.175 kg, VA - 0.618 body wt. * Body weight fixed at 0.175 kg, K - 0.045/body wt.3 For the sake ofconsistency, the Km value was set to 0.25 mg/liter for all groups of rats and the gas uptake results re-optimized. The new Fma, values were 10.98 0.155 and 9.18 0.078 mg/kg/hr for the naive and pregnant rats, with a first-order rate constant of 4.2 0.08 hr-1 in the naive rats. These metabolic constants were used for all subsequent simu lations. Oral uptake rate constant for TCE. Peak TCE blood concentrations in naive female rats were obtained within 4 min after bolus intubation of 7,6 mg TCE/kg in a water solu tion (Fig. 3). The simulated blood time course curve was lit to experimental data by using a four-compartment physiological model (Ramsey and Andersen, 1984) and op timizing for only Ka, the first-order uptake rate constant, until a best least squares fit to the data was obtained. The resulting value for Ka, 5.4 0.42 hr-1, was then used in simulat ing the rate of gastrointestinal absorption of TCE during pregnancy. Pharmacokinetics ofintravenously admin istered TCA. The plasma TCA elimination after intravenous dosing with 4 mg TCA/kg SL 035015 402 FISHER ET AL. Fig. 2. The uptake of TCE from a closed recirculated atmosphere. The smooth curves were generated by the computer model and the squares represent experimentally determined atmospheric levels of TCE. (A) Naive female rats. The initial chamber TCE concentrations were 111, II00,2200, and 5075 ppm. Four rats were used for each exposure. (B) Pregnant rats (Day 13-15 of pregnancy). The initial chamber TCE concentrations were 103, 1005, and 2050 ppm. Four rats were used for each exposure. in a saline solution (not shown) was ade quately described by a one-compartment model with an elimination rate constant and 95% confidence interval of 0.046 (0.0390.053) hr-1. The estimated volume of distri bution for TCA in the pregnant rat was 0.508 (0.444-0.675) liters/kg. Acute 4-hr TCE inhalation exposure. The PB-PFC pregnancy model was used to predict the TCE blood time course in Day 12 preg nant rats exposed to 618 ppm TCE for 4 hr (Figs. 4A, 4B). In addition to acting as a vali dation of the TCE portion of the simulation model (Fig. 4A), this experiment allowed an PB-PK MODELING IN PREGNANCY 403 Fig. 3. Oral uptake of TCE after bolus intubation of a water solution with 7.6 mg TCE/kg. Mean TCEblood concentrations with standard deviation bars are depicted with a smooth line generated by computer model (n - 3). estimation ofthe yield ofTCA from TCE me tabolism and of the pharmacokinetic charac teristics ofTCA formed during the inhalation exposure. TCE oxidation produces trichloroicetaldehyde which is reduced to trichloroethanol or oxidized to TCA. The proportion of TCE oxidized to TCA, designated PO, was estimated to be 0.12 in this exposure situa tion (Fig. 4B). The volume of distribution for TCA was set at 0.618 liter/kg and the opti mized elimination rate constant was 0.045 0.0025 hr-1 for a 1-kg animal. These ki netic constants for TCA distribution and elimination were used for all subsequent sim ulations. Subchronic TCE inhalation exposure. The acute inhalation exposure of the pregnant rat at Day 12 of gestation produced information on the kinetics ofTCA in the dam. Fetal TCA exposure characteristics had to be inferred from experiments later in pregnancy when the fetuses were of sufficient size for TCE and TCA blood analyses. Consequently, the sub chronic inhalation exposure was used (1) to test the fidelity of the PB-PK model both for predicting TCE concentrations in the dam and fetus (Fig. 5A) and for predicting TCA concentrations in the dam (Fig. 5B), and (2) to estimate the transfer coefficients for TCA from the maternal plasma into fetal plasma (Fig. 5B). On the basis of subchronic inhalation re sults, the TCA transfer coefficients were ad justed to give an adequate representation of the fetal plasma TCA concentration at Day 20 of gestation (Fig. 5B). To produce corre spondence with the data, PA 1 and PA2 had to be set at different values (see Appendix II). PA 1, the maternal to fetal transfer coefficient, was 0.0013 liter/hr, while the fetal to mater nal coefficient was estimated to be 0.0020 li ter/hr. With the estimation of these transfer coefficients, all parameters of the PB-PK pregnancy model were available and the time-courses of both TCE and TCA could be predicted for the subchronic drinking water and gavage exposure regimens. It bears em phasis that with the exception of the fitted fe tal plasma TCA concentrations in the inhala tion studies, all subchronic results were pre dicted from the PB-PK pregnancy model. Trichloroethylene exposure during preg nancy. The experiments during pregnancy were not intended to be large scale examina- SL 035017 404 FISHER ET AL. TIME (HOURS) Fig. 4. Comparison of predicted (solid continuous lines) and experimental concentrations of TCE in venous blood (A) and TCA in plasma (B) of pregnant rats following a 4-hr 600.4 ppm inhalation exposure on Day 12 ofgestation. Data points are means standard deviation (n - 6). tions of the time course behavior of TCE and TCA at multiple times during gestation. The guiding philosophy in this study was to de velop a PB-PK model for the pregnant rat with minimal reliance on in vivo experimen tation and then to predict expected maternal and fetal TCE and TCA exposure from the model. The in vivo experiments, which were still tedious and required killing significant numbers ofdams and fetuses, were purposely limited in scope and were designed to probe the ability of this modeling approach to pro vide a reasonable representation of the ob served results with both TCE and TCA by three routes of administration at several re stricted times during gestation. The maternal PB-PK MODELING IN PREGNANCY 405 GESTATION DATS Fig. S. Inhalation. Comparison ofa portion ofthe predicted (solid continuous lines) and the experimen tal concentrations of TCE in maternal and fetal blood (A) and TCA in maternal and fetal plasma (B). Pregnant tats were exposed by inhalation to 618 ppm TCE, 4 hr/day, 5 days/week, for 3 weeks. Maternal blood (n = 3) and pooled fetal blood were collected on Day 20 ofgestation, 20 hr after exposure the previous day for TCA analysis and on Day 20 of pregnancy immediately after exposure (n = 4) for TCE and TCA analyses. and fetal concentrations of TCE observed for the inhalation (Fig. 5A) and gavage rats (Fig, 6A) compared very favorably with prediction and none were off by more than a factor of 2. The greatest discrepancy was observed with the blood TCE levels in the dams dosed by gavage. For comparison, the peak TCE con centration at the end of the 4-hr inhalation at SL 0350ig 406 FISHER ET AL. GESTATION DAYS Fig. 6. Gavage. Comparison ofa portion ofthe predicted (solid continuous lines) and the experimental concentration ofTCE in maternal and fetal blood (A) and TCA in maternal and fetal plasma (B). Pregnant dams were given bolus intubation (2.3 mg TCE/kg) in water, 5 days/week, for 3 weeks. Maternal (n = 4) and pooled fetal blood was collected for TCE and TCA analyses after dosing on Day 20 of pregnancy and three hours post exposure (n = 5) on the same day for TCA analysis. 600 ppm was nearly 24 fig TCE/ml blood, while the gavaged rats had a measured concen tration less than 0.3 fig TCE/ml blood at 5 min after dosing. TCE was not found at measurable concentrations in maternal or fetal blood ofthe rats exposed by ingestion of TCE in drinking water. Consistent with this observation, the peak TCE blood concentrations predicted from the model were below the limit of detection (0.03 fig TCE/ml blood). PB-PK MODELING (N PREGNANCY 407 Fig. 7. Drinking water. Comparison of a portion of the predicted (solid continuous lines) and the mean experimental concentrations ofTCA in maternal and fetal plasma. Rats were provided daily drinking water containing an initial mean TCE concentration of 350 pg TCE/ml water, 5 days/week, for 3 weeks. TCEwater consumption remained constant throughout pregnancy. Maternal blood (n = 4) was collected be tween 0900 and 1000 hr ofDay 18 ofpregnancy for TCA analysis. Maternal blood (n = 6) and pooled fetal blood were also collected at the same time ofday on Day 21 ofpregnancy for TCA and TCE analyses. Trichloroacetic acid exposure during preg nancy. TCA concentrations in the fetuses of dams exposed by inhalation were used to ad just the transfer coefficients and do not repre sent predictions from the PB-PK pregnancy model. The maternal TCA plasma concen tration (Fig. 5B), predicted on the basis ofthe parameters obtained by limited experimenta tion and by review of the literature regarding physiological parameters in pregnant rats, was in good agreement with the observed val ues. For the drinking water and single dose gavage portions of the study there was good agreement between observed and predicted TCA concentrations in the maternal plasma (Figs. 6B and 7). The model tended to slightly under predict fetal TCA concentrations. The largest discrepancy was observed with the ga vage rats where the 3-hr postgavage fetal plasma TCA concentration was almost twice the predicted value (Fig. 6B). In general, there was very good agreement between the pre dicted and observed behavior of TCA over the three dose routes in both the dam and fe tus. Maximum TCA concentrations in ma ternal plasma were 13,0.7, and 2.8 pg TCA/ ml plasma for the inhalation, gavage, and drinking water rats, respectively. DISCUSSION The Kmax values obtained for TCE in fe male rats and its dependence on the physio logical state of the rat were consistent with other literature. Turcan et al. (1980, 1981), for instance, reported that the cytochrome P450 monooxygenase system activity is re duced during pregnancy, a decrease that has been linked to altered steroid hormones (Neims et ai, 1976). Our pregnant female rats had a Kmax (9.18 mg/kg/hr) value signifi cantly lower than the Fmax (10.98 mg/kg/hr) in naive female rats. In addition, TCE metab olism has also been examined by gas uptake methods in male rats (Andersen et ai, 1987). SL 035021 408 FISHER ET AL. The mate rat Fmax, 11.0 mg/kg/hr, was slightly higher than the l/mai in naive, female rats. On a comparative basis, TCE has a higher rate of oxidative metabolism than other chloroethylenes. For male rats the Vmax values for vinyl chloride, vinylidene chloride, /raws-dichloroethylene and c/s-dichloroethylene were respectively, 2.5, 7.5, 3.0, and 3.0 mg/kg/hr (Gargas et al., 1988). Similar com parative studies ofthe entire family ofchloro ethylenes have not been conducted in female rats. The techniques for PB-PK modeling in pregnancy developed in this paper could eas ily be expanded for application with these other ethylenes, as well as a variety of other volatile chemicals, once a suitable data base is established for female tissue partition co efficients and metabolic rates for these other chemicals. Determining the exposure profile ofthe de veloping fetus to xenobiotics as a result of maternal xenobiotic exposure is of consider able toxicological interest for teratology and reproductive studies. Using the rat as a model, physiologically based mathematical models have been developed for evaluating the disposition of drugs in the pregnant dam and developing fetus. Olanoff and Anderson (1980) constructed a time-dependent physio logical model for pregnancy. Seven maternal compartments and seven fetal compartments were used to describe the kinetics of tetracy cline in the pregnant rat. Temporal changes in maternal and fetal blood flows to the pla centa and growth of the placenta and fetal compartments were described as gestation progressed. In other work, Gabrielsson and Paalzow (1983) developed a physiologically based pregnancy model for morphine using six maternal compartments and one fetal compartment. More recently Gabrielsson et al. (1985) developed a physiologically based pregnancy model for methadone in the preg nant rat using 11 maternal compartments and one fetal compartment. Our PB-PK pregnancy model was developed to describe the uptake, disposition and elimination of TCE, a well metabolized, volatile chemical. as well as TCA, one of its nonvolatile metabo lites. The acute and subchronic TCE inhalation studies both served useful purposes in devel oping the compartmental model for TCA in the pregnant rat. The acute 4-hr inhalation study provided information for determining the amount ofoxidized TCE that is converted to TCA and for refining the compartmental kinetic constants (AT and Vd) for TCA in the dam. The estimate of the proportion of TCE converted to TCA (12%) is consistent with other literature. DeKant et al. (1986) re ported that about 17% of a 2 mg/kg dose of TCE was excreted in the urine as TCA in adult female Wistar rats. Ogata et al. (1979) reported that female Wistar rats excreted a 2.3:1 ratio of trichloroethanol to TCA, which is about 30% TCA ifall the metabolized TCE is accounted for by the alcohol and acid. The clearance term, PA1 (Fig. 1), was 0.0013 liter/hr. PA2 (Fig. 1), a composite rate constant consisting of transfer rate and tissue solubility information was given a value of 0.0020 liter/hr (see Appendix II). The actual litter size for the dams ranged from 3 to 12 pups per litter, and the average was 7. To better understand the effect oflitter size on the disposition of TCE and TCA in the dam and fetus, the pregnancy model was exercised assuming 1, 3, or 12 fetuses per dam. This was done by adjusting fetal litter growth (Eq. 1) and placental growth (Eq. 2) to reflect the appropriate number of fetuses. Total blood flow to the placenta for the whole litter was still assumed to be 25% of adjusted maternal placental blood flow. TCA placen tal transfer coefficients, PA 1, PA2, were ad justed, accordingly (e.g., PA 1 /7 equals PA 1 for litter size equal to 1). No appreciable changes were observed in simulated dam or fetus TCE-blood or TCA-plasma time courses as a result of adjusting litter size within the model structure. There are several advantages in conducting inhalation exposures to develop physiologi cal models with volatile organics and their metabolites for multiple routes of exposure. PB-PK. MODELING IN PREGNANCY 409 Inhalation exposure provides a means for producing a substantial chemical body bur den which facilitates examination of tissue concentration time course data. Our inhala tion exposures with the pregnant rats pro vided estimates of the kinetic constants for TCA and the yield of TCA from oxidative metabolism of TCE. Equally important, the TCE blood concentration time course could be predicted over a fairly wide range as TCE was taken up and cleared from the body. These results set the stage for the other expo sure routes. The modeling results of the sub chronic drinking study were very promising. Although individual drinking patterns of the rats were greatly simplified (zero order rate of intake, 1200 to 0600 hours), the stable me tabolite, TCA, was detected in the dam and fetal plasma near model predicted plasma concentrations (Fig. 7). Modeling results of gavage dosing were also promising for TCE, and only slightly less successful for TCA. Peak TCE blood levels after a single gavage dose (7.6 mg TCE in water/kg) occurred within a similar time period (4 min) as re ported by other investigators. D'Souza et al. (1985) reported that peak TCE blood concen trations occurred within 6 to 10 min after us ing gavage dosing with PEG 400 as vehicle. Withey et al. (1983) reported that peak TCE blood levels occurred within 5 to 6 min after gavage dosing using water as vehicle. Using a first-order rate constant of 5.4 hr-1 to describe the gastrointestinal absorption of TCE (Fig. 3), the model adequately described the up take and clearance of TCE from the blood af ter a bolus gavage dose. For the repeated ga vage dosing, the maternal and fetal blood TCE levels were in close agreement with the model predictions (Fig. 6A). However, TCA in the fetal plasma was not detected in two of four litters whose mothers had just been dosed by gavage (Fig. 6B), and 3 hr after dos ing the model underpredicted the fetal plasma levels by a factor of 2. The pregnancy model had several simplify ing assumptions. For example, the fetal and placenta compartments were described only from Days 14 of gestation, and there was no fetal metabolism of TCE, Fetuses have a lim ited P450 metabolic capacity near term (Neims et al., 1976), The fetal compartment was described as just one compartment for both the TCE and TCA disposition. Despite these simplifications, the model attempted to account for the major physiologic, biochemi cal, and physical-chemical processes which govern the disposition of TCE and TCA in the pregnant rat and generally good agree ment was observed between predicted and measured blood TCE and plasma TCA concentrations (Figs. 5-7). Withey and Karpinski (1985) exposed pregnant SpragueDawley rats (Day 17 of pregnancy) for 5 hr to several chlorinated hydrocarbon vapors. Ma ternal blood and fetal tissue were collected immediately following exposure for hydro carbon analyses. To describe these data within our PB-PK pregnancy model, body weights were set to those reported by Withey and Karpinski (1985) and simulations were conducted for exposures at 302, 1040, 1559, and 2088 ppm TCE. The model was config ured to predict maternal blood and fetal tis sue concentrations at 5 min postexposure. To obtain an optimal fit with these SpragueDawley rats, Fmai was increased to 12.0 mg/ kg/hr and blood:air PC was reduced to 10.0. These constants compare to the respective values of 9.2 mg/kg/hr and 13.2 in the Fischer rats. The overall correspondence be tween the observed and predicted data with only minor changes in model parameters was very good (Fig. 8), increasing our confidence in the applicability of this PB-PK approach for analyzing the disposition of TCE in preg nant rats under a variety of exposure condi tions. One use of the physiologically based preg nancy model is to compare TCE and TCA exposure profiles of the dam with the expo sure profiles of the fetus as a result of mater nal exposure to TCE. In addition, route ofex posure comparisons can also be conducted. One measure of chemical exposure is the cu mulative blood-TCE or plasma-TCA con- 410 FISHER ET AL. Fig. 8. PB-PK model predicted and experimentally determined TCE levels in maternal blood and fetal tissue directly after 5-hr inhalation exposures at 302,1040, 1559, and 2088 ppm TCE. Experimental data were taken from Withey and Karpinski (1985). was set to 12,0 mg/kg/hr and the blood/air PC to 10.0. centration area under the curve (AUC) for the dam and fetus. The AUC blood or plasma value provides a measure of the availability of the TCE or TCA for transport to various target tissues. Table 3 shows the model de rived AUC blood-TCE and plasma-TCA val ues for the fetus and dam during gestation. Table 3 also includes predicted amounts of TCE metabolized and exhaled for the three routes of exposure. The fetal blood AUC val ues for TCE were 67,69, and 76% of the ma ternal venous blood AUC values for the gavage, drinking water, and inhalation animals, respectively, for Days 14 to 22 of gestation. The fetal plasma AUC values for TCA were 64,63, and 64% of the maternal AUC values for the gavage, drinking water and inhalation animals, respectively, for Days 14 to 22 of gestation (Table 3). Thus, the fetus should be susceptible to both TCE and TCA insult as a result of the maternal TCE exposure. While TCE is not a potent rodent teratogen, subtf biochemical or neurological affects may oc cur in the developing fetus as a consequents of maternal exposure to TCE. Recently Tay lor et at (1985) has shown irreversible behav ioral deficits (locomotor activity) in rat pup indirectly exposed to TCE and its metabolii by-products from maternal exposure to TCf in the drinking water. In the drinking watei exposure the predominant tissue exposure was from TCA (585 vs 2.7 (mg-hr)/liter), B) inhalation the AUC for TCE is greatly in creased to 1864 (mg-hr)/liter, with a 10-folc increase in TCA exposure (585 vs 5853 (mg hr)/liter). It would be interesting to examine the in utero toxicity of TCE by inhalation ex posures. The use of PB-PK pregnancy models for examining fetal dosimetry is important for establishing more rigorous dose-response re lationships for these behavioral teratology SL 035024 >. v r. ;v PB-PK MODELING IN PREGNANCY 411 TABLE 3 Model Derived Exposure Estimates For TCE and TCA during 3 Weeks of Pregnancy Route ofexposure Gavage Drinking water Inhalation Jam TCE metabolized (mg) TCE exhaled (mg) TCE in maternal blood (AUC) (mg/hr/liter) TCA in maternal plasma (AUC) Fetus TCE in fetal blood (AUC) TCA in fetal plasma (AUC) 5.30 0.24 0.72 (0.30) 106.43 (44.84)" 0.20 28.95 32.81 1.01 2.69(1.05)" 584.52 (261.57) 0.72 164.10 311.97 873.02 1864.05(725.00)" 5201.20(2298.79)' 552.35 1473.66 " AUC value in parentheses corresponds to modelled fetal development period (Day 14 to the end of Day 22 of pregnancy). This AUC value was used to compare the maternal exposure to TCE and TCA with fetal exposure to these chemicals. studies. With further validation of fetal expo sure, PB-PK modeling may provide an im portant framework for interpreting a variety of teratology and reproductive toxicology ex periments. APPENDIX I: PREGNANCY MODEL NOMENCLATURE TCE P Pi Pi Q, C,, C, Cv, At Maternal blood/air partition co efficient (liter blood/liter air) Fetal blood/air partition coefficient (liter blood/liter air) Tissue (/)/blood partition coeffi cient (liter blood/liter tissue) Blood flow to zth tissue (liter/hr) Arterial blood concentration (mg/ liter) Concentration in zth tissue (mg/li- ter) Venous blood concentration leav ing z th tissue (mg/liter) Amount in zth tissue (mg) TCA ^tgai Vd Cjcai K PI, Amount in zth tissue (mg) Volume ofdistribution (liter) Concentration in zth tissue (mg/li ter) Elimination rate constant (/hr) Tissue (z')/blood partition coeffi cient (liter blood/liter tissue) Subscripts z TCE and TCA pla placenta fet fetus p plasma 1 liver APPENDIX II: MODEL COMPARTMENTS TCE Equations. The tissue mass balance equations for TCE were developed as out lined by Ramsey and Andersen (1984), with slight elaboration to include both saturable and first-order metabolic processes in the liver (Gargas el ai, 1988). The rate of change in the amount of TCE in the placenta consists of two terms, one re- SL 035025 412 FISHER ET AL. lated to the maternal circulation to the pla centa and the other, to the fetal circulation to the placenta dApiJdt = Qpia(Cj -- Cpia/Ppia) -- dAfcJdt. (4) The second term is identical to the rate of change ofTCE in the fetus dAfcJdt -- Qia(C^JPpia' P\/P ~ Cfex/Pfel)- (5) The TCE concentration in the fetal drainage leaving the placenta includes a ratio of parti tion coefficients accounting for differences in partitioning of TCE between maternal and fetal blood (Table 2). TCA equations. The rate of TCA produc tion for the pregnant rat is expressed as a pro portion (PO) of the rate of TCE metabolism (PO = 0.12). A stoichometric conversion fac tor (SC equals 163.4/131.4) accounts for the molecular weight increase which occurs as a result of the enzymatic conversion of TCE to the oxidized TCA metabolite dATCJdt - (PO) (SC). (6) The rate of change in the amount of TCA in the maternal plasma is described by the TCA production term (Eq. (6)), a first-order plasma elimination term, and a term describ ing the flow of blood to the placenta dAjcAp/dt = dATCJdt -- Va K- Ctcap ~ (2pla(Ctcap " C-TCApla/PIpla)- (7) The concentration of TCA in the maternal plasma is Ctcap - ^TCAp/fd) and ^tcap is the integral of Eq. (7): (8) Ajcap -- I dArcApJo (9) In the placenta, TCA was considered to be flow-limited with respect to the maternal blood supply, but transplacental movement was modeled as a diffusion process. The equation dAfCAp\Jdt -- (?p|a( Ctcap -- CxcApia/PIpij -- PA 1 CjCApla/PIpla + PA 1 CjCAfet/PIfel* (10) contains placenta/matemal blood (PIpiJ and fetal tissue/matemal blood (PIfrt) partition coefficients. PI^ was estimated by the ratio of the placenta/saline and the maternal blood/ saline partition coefficients. No estimate of Plfe, was obtained in these present studies. Thus the equation used for placental TCA was simplified to dATCA(tiddt = (2pu(Ctcap -- CTCAp|a/PIpla) -- PA1 CrcApla/PIpla + PA2 CrCAfet- 0 0 PA2 then has information on the permea tion coefficient and tissue solubility. For fit ting the fetal TCA concentrations, PA1 and PA2 were estimated as independent parame ters. The last two terms in Eq. (11) are for de scribing the rate of change in the amount of TCA in the fetus. The concentration in the fetus is the integration of these terms divided by the calculated fetal volume. Exposures. Three routes of TCE exposure were considered; gavage, drinking water, and inhalation. The gavage TCE exposure is in corporated into the pregnancy model by as suming that the amount of TCE absorbed from the gastrointestinal (GI) tract into liver of the dam is first order dAnaJdt = Ka-Ap. (12) The amount of TCE remaining in the ma ternal GI tract (Agj) is: A* = U*o)exp[-A;(f)]. (13) where Agl) equals the amount ofTCE (mg) ad ministered per gavage, Ka equals 5.4/hr-1, and time t is 0 * t < 24 hr. The drinking water TCE exposure is de scribed by assuming that rats drink the TCE water solution for 6 hr at a zero-order rate SL 035026 PB-PK MODELING IN PREGNANCY 413 (jtdw equals 2.42 ml/hr) from 2400 to 0600. The rate of change in the amount of TCE orally ingested by drinking TCE in water is time course ofenzyme induction for inhaled styrene in rats based on arterial blood;inhaled air concentration ratios. Toxicol. Appl. Pharmacol 73, 176-187. Beules, R. P., Brusick, D. J., and Mecler, F. J. dAiJdt = A^w-Cdw. (14) (1980). Teratogenic-Mutagenic Risk of Workplace Contaminants: Trichloroethylene, Perchloroethylene, Because of the nonspecific loss of TCE and Carbon Disulfide. U.S. Department of Health, Ed ,'in the water in the bottles, a first-order loss ucation and Welfare, contract number 210-77-0047. i rate (ks) was determined (0.031/hr, t\ = 22.4 Dekant, W., Schulz, A., Metzler, m., and HenSCHLEr, D. (1986). Absorption, elimination and me hr). TCE concentration in the drinking water tabolism oftrichloroethylene: A quantitative compari bottle is: son between rats and mice. Xenobiotica 16, 143-152. Cdw = (Cdwi)exp[--,(/)]. (15) Dorfmueller, M. a., Henne, S. P,, York, R. G., Bornschein, R. L., and Manson, J. M. (1979). Eval Cdw, is the initial concentration of TCE in the drinking water bottle (mg/liter) and time, ms 0 * t s 24 hr. The zero-order drinking rate uation of teratogenicity and behavioral toxicity with inhalation exposure of maternal rats to trichloroethyl ene. Toxicology 14, 153-166. D'Souza, R. W,, Bruckner, J. V., and Feldman, S, incorporated into the model by adding the input rate equation to the liver compartment of the dam. (1985). Oral and intravenous trichloroethylene phar macokinetics in the rat, J. Toxicol. Environ. Health 15,587-601. Elcombe, C. R. (1985). Species differences in carcinoge nicity and peroxisome proliferation to trichloroethyl ACKNOWLEDGMENTS ene: A biochemical human hazard assessment. Arch. Toxicol (Suppl.) 8,6-17. Filser, J. G., and Bolt, H. M. (1979). Pharmacokinet The authors thank Dr. Jim Cooper for the animal ics of halogenated ethylenes in rats. Arch. Toxicol. 42, breeding support; Dr. Michael Gargas for the technical 123-136. advise on determining partition coefficients and meta llic constants for volatile chemicals; Mr. Ken Collier tor typing the manuscript; Mr. Carlyle Flemming for the statistical support; Captain Gary Jepson for the helpful discussion and suggestions on partition coefficients de terminations for nonvolatile chemicals and for the train ing in gas chromotography; and Ms Ellen Goldey for help with the subchronic exposures. Fisher, J. W., Andersen, M. E., Clewell, H. J. Ill, and Taylor, D. H. (1987). Modeling lactational transfer oftrichloroethylene (TCE) and its metabolite, trichloroacetic acid (TCA) in the rat. Toxicologist 7, 139 (abstract). Gabrielsson, j. l., Johansson, P,, Bondesson, U., and Paalzow, L. K. (1985). Analysis of methadone disposition in the pregnant rat by means of a physio logical flow model. J. Pharmacokinet. Biopharm. 13, REFERENCES 355-372. Gabrielsson, J. L., and Paalzow, L. K. (1983). A physiological pharmacokinetic model for morphine Andersen, M. E. (1981). a physiologically based toxicokinetic description of the metabolism of inhaled gases and vapors: Analysis at steady-state. Toxicol Appl. Pharmacol. 60,509-526. Andersen, M. E., Gargas, M. L., Clewell, H. J. Ill, and Severyn, K. M. (1987). Quantitative evaluation of the metabolic interactions between trichloroethyl ene and 1,1-dichloroethylene in vivo using gas uptake disposition in the pregnant rat. J. Pharmacokinet. Bio pharm. 11, 147-163. Gargas, M. l., Andersen, M. E., and Clewell, H. J., Ill (1986a). A physiologically based simulation approach for determining metabolic constants from gas uptake data. Toxicol. Appl. Pharmacol. 86, 341-- 352. Gargas, M. L., Clewell, H. J. Ill, and Andersen, methods. Toxicol. Appl Pharmacol 89,149-157. M. E. (1986b). Metabolism ofinhaled dihalomethanes Andersen, M. E., Gargas, M. L., Jones, R. A., and Jenkins, L. J., Jr. (1980). Determination of the ki netic constants for metabolism of inhaled toxicants in vivo using gas uptake measurements. Toxicol. Appl. Pharmacol. 54,100-116. Andersen, M. ., Gargas, M. L., and Ramsey, J. C. in vivo: Differentiation of kinetic constants for two in dependent pathways. Toxicol. Appl Pharmacol. 82, 211-223. Gargas, M. L., Burgess, R. J., Voisard, D. E., Ca son, G. H., and Andersen, M. E. (1989). Partition coefficients of low-molecular-weight volatile chemi (1984). Inhalation pharmacokinetics: Evaluating sys cals in various liquids and tissues. Toxicol. Appl. Phar temic extraction, total in vivo metabolism, and the macol 97,87-99. 035027 SL 414 FISHER ET AL. Gargas, M. L., Seybold, P, G., and Andersen, M, E. (1988). Modeling the tissue solubilities and metabolic rate constant (Vmax) of halogenated methanes, eth anes and ethylenes. Toxicol, Lett. 43,235-256. Griffith, D. R., and Turner, C. W. (1961). Normal growth of rat mammary glands during pregnancy and early lactation. Soc. Exp. Biol- Med. Proc. 106, 448450. Hanwell, A., and Linzell, J. L. (1973). The time course ofcardiovascular changes in lactation in the rat. J. Physiol. 233,93-109. Herren-Freund, S. L., Pereira, m. a., Khoury, M. D., and Olson, G. (1987). The carcinogenicity of trichloroethylene and its metabolites, trichloroacetic acid and dichloroacetic acid, in mouse liver. Toxicol. Appl. Pharmacol. 90,183-189. Huggett, A, St G., and Widdas, W. F. (1951). The relationship between mammalian foetal weight and conception age. J. Physiol. 114,306-317. Jepson, G. W. (1986). A Kinetic Modelfor Acetylcholin esterase Inhibition by Diisopropyffluorophosphate in Crude Rat Brain Homogenate. Master's thesis, Wright State University, Dayton, OH. Knight, C. H., Docherty, a. H,, and Peaker, M. (1984). Milk yield in rats in relation to activity and size of the mammary secretory cell population. J. Dairy Res. 51,29-35. Leach, L. J. (1963). A Laboratory Test Chamber for Studying Airborne Materials, pp. 1-12, AEC Research and Development Report UR-629. McDougal, J. N., Jepson, G. W., Clewell, H. J. Ill, AND Andersen, M. E. (1985). Dermal absorption of dihalomethanc vapors. Toxicol. Appl. Pharmacol. 79, 150-158. Mitchell and Gauthier, Assoc., Inc. (1981). Advanced Continuous Simulation Language (ACSL). Mitchell and Gauthier. Associates, 3 ed,, Concord. MA. (Vaismith, D. J., Richardson, D. P., and Pritchard, A. E. (1982). The utilization ofprotein and energy dur ing lactation in the rat, with particular regard to the use of fat accumulated in pregnancy. J. Nutr. 48,433441. National Research Council (1986). Dose route extrapo lations: Using inhalation toxicity data to set drinking water limits. In Drinking Water and Health (R. D. Thomas, Ed.), Vol. 6, pp. 185-236. Natl. Acad. Press, Washington, DC. Neims, a. H., Warner, M, W., Loughman, P. M.. and Aranda, J. V. (1976). Developmental aspects of the hepatic cytochrome P450 monooxygenase system. Annu. Rev. Pharmacol. Toxicol. 16,427-445. Noland-Gerbec, E. A., Pfohl, R. J., Taylor, D. H., and Bull, R. J. (1986). 2-deoxyglucose uptake in the developing rat brain upon pre- and postnatal exposure to trichloroethylene. Neurotoxicology 7, 157-164. Ogata, M., Norichika, K., Shimada, Y., and MeGURO, T. (1979), Differences in urinary trichloroethyl ene metabolites of animals. Acta Med. Okayama 33, 415-421. Olanoff, L. S., and Anderson, J. M. (1980). Con trolled release oftetracycline-III: A physiological phar macokinetic model of the pregnant rat. J. Pharmaco- kinet. Biopharm. 8, 599-620. Ramsey, J. C., and Andersen, M. E. (1984). a physio logically based description of the inhalation pharma cokinetics of styrene in rats and humans. Toxicol. Appl. Pharmacol. 73,159-175. Sato, A,, and Nakajima, T, (1979). Partition coeffi cients of some aromatic hydrocarbons and ketones in water, blood, and oil, Brit. J. Ind. Med. 36,231 -234. Taylor, D. H., Lagory, K. E., Zaccaro, D. J., Pfohl, R. J., and Laurie, R. D. (1985). Effect of tri chloroethylene on the exploration and locomotor ac tivity of rats exposed during development. Sci. Total Environ. 47,415-420. Turcan, R. G., Symons, A. M., and Parke, D. V, (1980). Factors affecting hepatic drug metabolism dur ing pregnancy in the rat. J. Endocrinology 87,23-24. Turcan, R. G., Tamburini, P. P., Gibson, G. G., Parke, D. V., and Symons, A. M. (1981). Drug me tabolism, cytochrome P450 spin-state, and phospho lipid changes during pregnancy in the rat. Biochem. Pharmacol. 30,1223-1225. United States Environmental Protection Agency (U.S. EPA) (1985). Health Assessment Document for Tri chloroethylene. Final Report E-600/8-82-006B, Wash ington DC. United States Environmental Protection Agency (U.S. EPA) (1987). Addendum to the Health Assessment Document for Trichloroethylene: Updated carcinoge nicity assessment for trichloroethylene. Draff: EPA 1600/8-22/006FA, Office of Health and Environmen tal Assessment, Washington DC. Withey, J. R., Collins, B. T., and Collins, P. G. (1983). Effect of vehicle on the pharmacokinetics and uptake of four halogenated hydrocarbons from the gastrointestinal tract of the rat. / Appl. Toxicol. 3, 249-253. Withey, J. R., and Karpinski, K. (1985). The fetal dis tribution of some alaphatic chlorinated hydrocarbons in the rat after vapor phase exposure. Biol. Res. Preg nancy 6,79-88. SL 035028