- Email: [email protected]
2J mice
Toxicology Letters, 42 (1988) 15 - 28 Elsevier
15
TXL 01984
A physiologically based pharmacokinetic model for 2,3,7, Stetrachlorodibenzo-p-dioxin in C57BL/6J and DBA/2J mice Hon-Wing Leung’, Robert H. Kul, Dennis J. Paustenbach’** and Melvin E. Andersen2 ‘Environmental Health and Safety, Syntex (U.S.A.) Inc.. Palo Alto, CA 94304, U.S.A., and ‘4921 Egret Court, Dayton, OH 45424, US.A. (Received 18 October 1987) (Revision received 3 February 1988) (Accepted 3 March 1988) Key words: 2,3,7,8-Tetrachlorodibenzo-p-dioxin; Mathematical simulation; Toxicokinetics; Body compartment; Tissue concentration; Binding protein; (Mouse)
SUMMARY A five-compartment physiologically based pharmacokinetic (PB-PK) model was developed to describe the time course of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the tissues of both C57BL/6J and DBA/ZJ mice. The PB-PK model included binding in blood and two hepatic binding sites, one in the cytosol and the other in the microsomes. First-order metabolism occurred in the liver. Model simulations were compared to literature results for the disposition of a single intraperitoneal dose of 10 pg/kg of [‘HITCDD, reported by Gasiewicz et al. [Drug Metab. Dispos. 11 (1983) 397-4031. In contrast to previous speculation, the greater accumulation of TCDD in the liver of the C57BL/6J mouse, as compared to the DBA/ZJ mouse, was not attributable to the higher fat content in the DBA/2J mouse. Instead, the disposition of TCDD in these mice was more dependent on the affinity of the microsomal binding proteins than on fat content. The microsomal dissociation constant in the C57BL/6J mouse estimated by the PB-PK model was about one-third its value in the DBA/2J mouse (20 versus 75 nM), i.e. there is more avid microsomal binding in the liver of the CS7BL/6J mouse. In the concentration range covered in these time-course studies, the cytosolic receptor, with its low capacity and very high affinity binding characteristics, does not play a major role in determining the overall tissue distribution pattern. The concentration and affinity of the microsomal binding protein in the liver appear to be primarily responsible for explaining the differences in the liver/fat concentration ratios between various strains and species of laboratory animals. Address for correspondence: Hon-Wing Leung, Ph.D., Environmental Health & Safety, Syntex (U.S.A.) Inc., 3401 Hillview Avenue, Palo Alto, CA 94303, U.S.A. *Present address: McLaren Environmental Engineering, ChemRisk Division, 11101 White Rock Road, Ranch0 Cordova, CA 95670, U.S.A. Abbreviations: PB-PK, physiologically based pharmacokinetic; TCDD, 2,3,7,8-tetrachlorodibenzo-pdioxin, AHH, aryl hydrocarbon hydroxylase. 0378-4274/88/$ 03.50 0 1988 Elsevier Science Publishers B.V. (Biomedical Division)
16
INTRODUCTION
2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is a contaminant produced during the manufacture of chlorophenols, hexachlorophene, and phenoxy herbicides. TCDD is one of the most extensively studied chemicals in recent years, and several reviews of its toxicology have been written [l-4]. Despite these investigative efforts, the exact mechanism of TCDD toxicity is still unclear. Among the many uncertainties, the diverse species difference in susceptibility to TCDD is one of the most intriguing. The acute oral LD5o ranges from 1 pg/kg in the guinea pig to 5 mg/kg in the hamster [4]. Although the lethal dose in humans has not been precisely identified, based on several incidents involving exposed persons, it is almost certainly greater than 100 pg/kg [5]. Another interesting species difference in TCDD toxicity is the specificity of pathological changes in the various organs. While thymic atrophy occurs in all animal species tested [6], other organ effects are characteristic for a given species, e.g. hepatic lesions are striking in mice, rats, and rabbits but relatively minor in guinea pigs, monkeys, cattle and horses. Mucosal changes in the gastrointestinal and urinary systems are observed in primates and cattle but not reported in the rat, mouse or guinea pig [6]. Chloracne, a hyperplastic epidermal change, is commonly observed in exposed human populations, but not in rats, hamsters or guinea pigs. Besides acute toxic effects, there is also marked species difference in the disposition of TCDD. The excretion half-life has been reported to be about 15 days in mice [7] and hamsters [S], 30 days in rats [9], 30-94 days in guinea pigs [lo-l 11, 455 days in monkeys [ 121, and 5-7 years in man [ 13,141. The differences in half-life do not appear to be correlated with the toxic potency of TCDD. In most animal species, TCDD accumulates in the liver and adipose tissue. However, the relative distribution between these two tissues is markedly different among various animal species and strains [ 151. One possible explanation for these differences may be the relative fat content in these tissues. Since TCDD is highly soluble in lipids, distribution to fatty depots in species and strains having a high body adipose content will significantly alter its disposition relative to those which do not. Differences in distribution may also be related to the affinity and capacity of receptor protein binding. In certain strains of mice, enzyme induction (e.g. aryl hydrocarbon hydroxylase or AHH) by TCDD has been shown to segregate with a single genetic locus, known as the Ah locus [16]. The non-reponsi\ve strains, typified by DBA/2J mice, appear to have an altered receptor with lower affinity for TCDD as compared to the responsive strains, typified by C57BL/6J mice. Poland et al. [171 have suggested that the greater accumulation of TCDD in the liver of C57BL/6J mice, as compared to the DBA/2J mice, is directly related to the strain differences in the affinity of the receptor proteins for TCDD. Gasiewicz et al. [71 have also suggested that other differences such as altered metabolism and excretion may also contribute to the difference in hepatic accumulation.
17
To examine the extent to which these factors affect the differential hepatic metabolism, we have developed a physiologically based pharmacokinetic model (PB-PK) for TCDD in the C57BL/6J and DBA/2J mice. Predictive models, such as the one described here, which incorporate known anatomical, physiological, thermodynamic, and transport characteristics, have been used to describe the behavior of a variety of pharmacological agents [18] and were recently applied to environmental toxicants such as styrene [19], methylene chloride [20], dieldrin [21], chlordecone [22], polychlorinated and polybrominated biphenyls [23,24], and tetrachlorodibenzofurans [25]. Successful use of this approach for scale-up from animals to humans has been demonstrated [19,20]. The present TCDD model has five compartments and includes binding in blood and two binding proteins in the liver. It provides quantitative descriptions of the time-course of elimination and levels of TCDD in various organs of C57BL/6J mice, an AHH-inducible strain with low body fat content, and DBA/2J mice, a non-AHH-inducible strain with high body fat content.
KAB .
QC
CA ;“CB
CA
i BLOOD PERITONEAL CAVITY; AP,
BMl ,KBl KA
CVL = CUPL CVL -
BM2,KB2
,\
I
.
CVF = CFIPF
CVF
FAT
I
t
L--l cvs
I
1
cvs =
CSIPS 1
1
QS
I
MUSCLE/SKIN SLOWLY PERFUSED I
I QR
CVR = CRIPR
q
p
RICHLY PERFUSED
Fig. 1. Schematic representation of the five-compartment physiologically based pharmacokinetic for TCDD in mice. Abbreviations are defined in the Appendix.
model
18
MATERIALS
AND METHODS
Model structure The TCDD model (Fig. 1) was a modification of the general PB-PK model previously described [19]. It contained five compartments: blood, liver, fat, richly perfused tissues, and slowly perfused tissues. Since TCDD binds to protein receptors in the liver, the model was set up to provide for two hepatic binding sites, one corresponding to the non-inducible high affinity/low capacity cytosolic receptor [17], and the other to the inducible, low affinity/high capacity microsomal receptor [26]. In simulation of the intraperitoneal dose route used by Gasiewicz et al. [7], TCDD was assumed to be absorbed into the liver compartment by a first-order uptake process. The absorption rate constant (KA) of 0.02 h- ’ was chosen to fit the rise in tissue concentration with the C57BL/6J mice data. Bioavailability via this route of administration was assumed to be 100%. Mass balance differential equations for the model were similar to those used in the analysis of styrene data [19] except for the liver and blood. Definitions of the algebraic terms and an explanation of the liver and blood mass balance equations are given in the Appendix. Model parameters Three types of data are required to implement this physiological model: (1) partition coefficients; (2) physiological constants; and (3) biochemical constants. These data, which were obtained or calculated from the literature for each mouse strain being simulated, are summarized in Table I. Partition coefficient between fat and blood (Pfi was calculated from the data of Birnbaum [27]. The kidney was assumed to be representative of the richly perfused tissue and the partition coefficient (PR) was estimated from the data of Gasiewicz et al. [7]. The partition coefficient between liver and blood (PL) was assumed to be equal to that of PR. The slowly perfused tissue was assumed to be principally composed of skin and muscle. The partition coefficient of this compartment (Ps) was arbitrarily selected to fit the distribution data, but restricted to values intermediate between PR and PF. The model includes liver binding and incorporates binding constants and binding capacities that have been established by direct experimentation. BMZ and KBI of C57BL/6J mice were obtained or calculated from Gasiewicz and Rucci [28]. Teitelbaum [26] provided the value for BM2 and an estimate for KB2. Subsequent attempts at fitting the distribution data suggested that the best value for KB2 was almost exactly as Teitelbaum’s estimate. For the Ah-non-responsive DBA/2J mice, non-detectable levels of cytosolic receptor were reported [28]. However, it has been reported that the receptor is actually present, but the binding affinity may be low enough that it is below the experimental limit of detection [29]. For this reason, the BMI value of the DBA mouse was set to equal to that of the C57 mouse. The KB2 value was empirically fitted with the computer model. In addition, the model in-
19
eludes binding in the blood. Blood binding is described as a linear process with an effective equilibrium between bound and free TCDD given by the constant, KAB. This gives rise to a distribution of TCDD in blood between two forms, only one of which is exchangeable in the tissues. The fractional amount available is given by the ratio (l/(1 + KAB)).This apportioning in the blood gives rise to kinetic behavior TABLE I KINETIC CONSTANTS AND MODEL PARAMETERS BASED PHARMACOKINETIC MODEL FOR TCDD Parametera
Abbreviationsb
USED IN THE PHYSIOLOGICALLY Mice C57BL/6J
Weights (kg) Body
0.023 1
EW
Volumes ((I/obody weight) Liver Richly perfused Slowly perfused Fat Blood Cardiac output (l/h) Blood flow (W cardiac output) Liver Richly perfused Slowly perfused Fat Partition coefficients Liver/blood Richly perfused/blood Slowly perfused/blood Fat/blood Biochemical constants (nmol/liver)
DBA/ZJ 0.0239
VLC
5
5
VRC
4
4
VSC
76.1
70.5
VFC
5.9
VBC
5
5
QC
0.861
0.883
11.5
QLC QRC QSC QFC
25
25
51
51
17
17
7
7
PL
20
20
PR
20
20
PS
250
250
PF
350
350
BIUI
0.0042
@M)
KBI
0.29
(nmol/liver) (nM) (per hour) (per hour)
BM2
20
20
KB2
20
75
Dose (nmol/kg) “Scaled parameters:
2
KFC
3.25
1.75
KA
0.02
0.02
KAB
2.5
2.5
APO VR = 0.09 BW-VL;
0.0042
VS = 0.82 BW-VF;
bSee Appendix for definition of abbreviations
32 QR = 0.76 QC-QL;
for biochemical constants.
32 QS = 0.24 QC-QF.
20
similar to that observed for diffusion-limited uptake into tissues. Tissue partition coefficients used in the model are based on the distribution of free TCDD. Effective tissue:blood concentration ratios, that is the observed ratio of concentrations in various experiments, are related to total blood concentration. Thus, concentration ratios are given by Pi/(1 + KAB). There has not yet been any direct experimentation to establish this blood binding constant. The value used here (KAB = 2.5) gave good agreement with the disposition data when the literature values of the liver binding parameters were used in the simulation. Simulations The set of simultaneous differential equations was solved numerically with the aid of a Fortran-based simulation software package [30] run on an IBM PC/AT or an AT-compatible computer. RESULTS
Fig. 2 shows the simulated time-course of TCDD concentration in the liver and fat of C57BL/6J mice given 32 nmol TCDD per kg of body weight by intraperitoneal injection. The actual data sets were taken from the experimental study of Gasiewicz et al. [7]. There is good agreement between the simulated description generated by the model and the empirical data. Gasiewicz et al. [7] have observed that the liver/fat ratio of TCDD concentration in the C57BL/6J mice was about 3 times that in DBA/ZJ mice. Since DBA/2J mice had about twice the body fat content as compared to C57BL/2J mice, the difference
0
10
20
30
40
Days
Fig. 2. Time-course of 32 nmol/kg.
of TCDD disposition
in liver and fat of C57BL/6J
Lines represent the simulation by the model ( (0 fat; 0 liver) are those of Gasiewicz
mice given an intraperitoneal
fat; - - et al. [7].
dose
liver). Datum points
21
r
KB2 = IO
KB2 = 20
10 :
5 0
KB2 = 40 I
I
I
1
10
20
30
40
Days
Fig. 3. The influence of KB2 values on the time-course of TCDD concentration in the liver of C57BL/6J mice.
in hepatic concentration may be due to the greater capacity of the DBA/2J mouse to sequester the highly lipophilic TCDD in the adipose tissue. One of the strengths of physiological modeling is the ability to readily simulate the disposition of a chemical by altering certain biologic parameters that are difficult or impossible to manipulate under actual experimental conditions. Thus, we varied the fat content of the C57BL/6J mice from 3 to 12% of body weight in order to test the hypothesis that the distributional differences may have been influenced by body fat content. Simulations showed that TCDD concentration in the liver was relatively insensitive to body fat content (data not shown), indicating that this was not an important factor influencing the disposition of TCDD in the liver between the two strains of mice. Interestingly, the distributional behavior of TCDD in the C57BL/6J mice was profoundly influenced by the binding characteristics of the microsomal binding protein, especially the KB2 value. Fig. 3 shows the effect of varying the KB2 value on liver concentration/time data obtained in C57BL/6J mice following intraperitoneal administration of 32 nmol/kg TCDD. Fig. 4 shows that the model also gives good simulations of the TCDD excretion in both the C57BL/6J and DBA/2J mice. The model, however, did not give as satisfactory a simulation of the fat and liver TCDD concentrations in the DBA/2J mice if the input was set to be consistent with the uptake and elimination profile (Fig. 5). However, similar to the C57BL/6J mice, the disposition of TCDD in the DBA/2J mouse liver was also greatly influenced by the KB2 value but was rather insensitive to varying body fat content (data not shown). The best fit of the empirical data was obtained with a KB2 value of about 75 nM. Thus the TCDD binding affinity to the hepatic microsomal protein receptor in the DBA/2J mice was at least 3.5 times lower than that of C57BL/6J mice.
22
25
Fig. 4. Kinetics of TCDD excretion in mice. Computer simulation ( C57BL/6J; - - - DBA/ZJ); empirical data (0 C57BL/6J; 0 DBA/ZJ). Description of the way metabolism and excretion are simulated by the computer model is given in the Appendix.
DISCUSSION
We have constructed a flow-limited physiological model for TCDD to explain the pharmacokinetic differences in tissue distribution between C57BL/6J and DBA/2J mice. The latter is characterized as an Ah-non-responsive strain and has twice as much fatty tissue stores as the former strain. DBA/W mice accumulate less TCDD in their livers than C57BL/6J mice. Storage of highly lipophilic substances such as TCDD in the adipose tissue usually has a marked effect on the pharmacokinetics [31]. However, the PB-PK model for TCDD predicted that the size of the fat compartment alone could not explain the lower accumulation of TCDD in the liver of the DBA/%J mice compared with the C57BL/6J mice. Instead, the disposition of TCDD in the mouse was profoundly affected by the affinity of the microsomal binding protein. The differential accumulation of TCDD in the livers of the C57BL/6J and DBA/2J mice was apparently due to the differing affinity characteristics of the microsomal TCDD binding proteins. There have been several previous attempts to construct physiologically based pharmacokinetic models for polyhalogenated biphenyls and polychlorinated dibenzofurans [24,25]. In these studies the disposition of these persistent hydrocarbon chemicals was described by simple partitioning between the blood and the various
23
0
10
20
30
40
Days
Fig. 5. Time-course of TCDD disposition in liver and fat of DBA/2J mice. Computer simulation ( ~ fat; - - - liver); empirical data (0 fat; 0 liver).
tissues with first-order metabolism in the liver. The partition coefficients used in these models were estimated by determining blood and tissue levels in animals killed at various times after dosing. These models did not account for the role that extensive tissue binding to particular cellular proteins might play in determining the overall disposition of the xenobiotic. In contrast, the present description attempts to provide a biochemical basis for the observed tissue distribution. Thus, the model structure has terms for partitioning which describe the general tissue solubility of TCDD and clearly specify that TCDD has higher solubility in fat (PF = 350) than in liver (PL = 20).The tissue concentration ratio that is achieved in any dosing study, though, is dependent both on solubility and specific binding. Liver concentrations in C57BL/6J mice are higher than fat concentrations (Fig. 2), not due to some intrinsically higher solubility, but because of the high affinity tissue binding in the liver in this concentration range. Gasiewicz et al. [7] did not determine muscle concentrations of TCDD and only examined skin content on day 25 in DBA/2J mice. Since no data were available for these tissues, the assignment of a high value for PS (250)was made only to give a good fit to the liver and fat curves. To assure ourselves that conclusions on the importance of microsomal binding were not an artifact to model structure, we formulated an alternative model that maintained a low value of PS (20)and maintained that fit to the liver and fat concentrations by adjusting the intraperitoneal bioavailability to 67%. This model also gave a good description of the time-course data. More importantly, however, the behavior of this alternative formulation with changes in fat compartment volumes was nearly identical to the model shown in this paper. The liver/fat ratios were still most sensitively dependent on the microsomal binding affinities, not on the amount of fat.
25
APPENDIX
Definitions of the algebraic constants and mass balance equations for the liver and blood compartment * The liver mass balance equation contained terms for influx and efflux of TCDD in arterial and venous blood and first order metabolism:
F
= QL(CA - CVL) - q
+ (y)
/
(1 + KM?)
For describing the data from Gasiewicz et al. [7], the initial dose is the amount injected in the peritoneal cavity (APO). The instantaneous rate of absorption (dAP/dt) is first-order with respect to the amount remaining to be absorbed (AP): $!/
= KA . AP = KA ’ APO . ebKA ’ t
Total mass of TCDD in the liver at any time is the time integral of the rate equation: AL = INTEGRAL
(dAL/dt,
0)
In contrast to the description of styrene disposition [19], TCDD in the liver was distributed between two binding proteins and free TCDD: AL
I/~.~vL.P~
=
+
Bhf1-a~ + Bkf..*CVL KB2 + CVL KBI + CVL
This equation can be rearranged CVL = AL/[(VL.PL)+
*Abbreviations: of TCDD
AL?, amount
metabolized
be absorbed; BMI, microsomal protein; of total TCDD tion of TCDD TCDD
in venous
from
peritoneal
constant;
of TCDD
by microsomes
protein;
in blood;
AL, amount
in liver; AP, amount
TCDD binding capacity to cytosolic BW, body weight; CA, concentration
(free and bound)
concentration
cytosolic
BMI/(KBZ + CVL)+ BM2/(KB2+
in blood;
in venous blood leaving
CVR,
in terms of CVL:
of TCDD blood leaving
cavity
PL, liver/blood
slowly perfused
into liver; KAB,
KB2, TCDD
binding
partition
flow through liver compartment; slowly perfused tissue; t, time;
CL, concentration blood TCDD
constant
coefficient;
leaving
in liver tissue; AM,
in peritoneal
amount
cavity remaining
to
protein; BM2, TCDD binding capacity to of free TCDD in blood; CB, concentration of TCDD
fat; CVL, concentration
in venous
of TCDD
of TCDD
CVL)]
richly
tissue; DOSE, binding QF, blood
in venous blood leaving liver;
perfused
tissue;
dose of TCDD;
to blood;
to microsomal
in liver tissue; CVF, concentra-
of TCDD
KBI,
protein;
flow through
CVS,
concentration
KA, absorption
TCDD
binding
KFC, first-order fat compartment;
of
constant constant
metabolic
to rate
QL, blood
QR, blood flow through richly perfused tissue; QS, blood flow through VB, volume of blood; VL, volume of liver.
26
and solved by approximation. For comparison with Gasiewicz et al. [7], total liver concentration plotted:
(CL) was
CL = AL/V-L
The rate of metabolism dAM dt
= KFC/(BFvp
(nmol/h)
* CVL
. VL
The total amount metabolized AM
= INTEGRAL
(dAM/dt,
at any time is:
(AM) is the time integral of this equation: 0)
It is assumed that metabolized TCDD is rapidly eliminated from the body. The elimination plots as (70 excreted are derived as follows: 9’0 EXCRETED
= (AM/DOSE)
. 100
In the blood, TCDD equilibrates between free and bound forms with an equilibrium constant, KAB. The amount of TCDD in blood (AB) is the sum of the two: AB
= VB * (CA
+ KAB
* CA)
The combined concentration of TCDD in blood (CB), the parameter determined experimentally, is linearly related to free TCDD in blood (CA), which is the exchangeable TCDD at any time: CB = +j
= CA * (1 + KAB)
The absorbed TCDD enters the portal blood and is distributed between the free and bound pools of TCDD. Only the free form is available for distribution into the liver and the liver mass balance equation must include the input term: liver input = (9)/l
+ KAB)
The remainder of the absorbed material passes through to the venous blood giving a slightly modified blood mass balance equation: @/
= QF . CVF + QL . CVL dAP dt’
KAB
(1 + KAB)
+ QS + CVS
+ QR . CVR
-
QC * CA +
27 REFERENCES 1 International Agency for Research on Cancer (1977) Some Fumigants, the Herbicides 2,4-D and 2,4,5,-T, Chlorinated Dibenzodioxins and Miscellaneous Industrial Chemicals, Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, World Health Organization, Lyons, Vol. 15, pp. 41-102. 2 Environmental Protection Agency (1985) Health Assessment Document for Polychlorinated Dibenzo-pdioxins, EPA-600/S-84014F, Environmental Protection Agency, Cincinnati, OH. 3 National Research Council of Canada (1981) Polychlorinated Dibenzo-p-dioxins: Criteria for their Effects on Man and his Environment, NRCC 18574, National Research Council of Canada, Ottawa, Ont. 4 American Medical Association (1984) The Health Effects of ‘Agent Orange’ and Polychlorinated Dioxin Contaminants: An Update, American Medical Association Press, Chicago, IL. 5 Tschirley, F.H. (1986) Dioxin. Sci. Am. 254, 29-35. on the 6 Vos, J.G., Moore, J.A. and Zinkl, J.G. (1973) Effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin immune system of laboratory animals. Environ. Health Perspect. 19, 149-162. 7 Gasiewicz, T.A. Geiger, T.A., Rucci, G. and Neal, R.A. (1983) Distribution, excretion, and metabolism of 2,3,7,8-tetrachlorodibenzo-p-dioxin in C57BL/6J, DBA/ZJ and B6D2F,/J mice. Drug Metab. Dispos. 11, 397-403. 8 Olson, J.R., Gasiewicz, T.A. and Neal, R.A. (1980) Tissue distribution, excretion and metabolism of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the Golden Syrian hamster. Toxicol. Appl. Pharmacol. 56, 78-85. 9 Neal, R.A., Olson, J.R., Gasiewicz, T.A. and Geiger, L.E. (1982) The toxicokinetics of 2,3,7,8-tetrachlorodibenzo-p-dioxin in mammalian systems. Drug Metab. Rev. 13, 355-385. 10 Olson, J.R. (1986) Metabolism and disposition of 2,3,7,8-tetrachlorodibenzo-p-dioxin in guinea pigs. Toxicol. Appl. Pharmacol. 85, 263-273. 11 Gasiewicz, T.A. and Neal, R.A. (1979) 2,3,7,8-Tetrachlorodibenzo-p-dioxin tissue distribution, excretion, and effects on clinical chemical parameters in guinea pigs. Toxicol. Appl. Pharmacol. 51, 329-339. 12 Bowman, R.E., Schantz, S.L., Weerasinghe, N.C.A., Meehan, J. Pan, J., Gross, M.L., Boehm, K.M. Van Miller, J.P. and Welling, P.G. (1987) Clearance of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) from body fat of rhesus monkeys following chronic exposure. Toxicologist 7, 158. 13 Poiger, H. and Schlatter, C. (1986) Pharmacokinetics of 2,3,7,8-TCDD in man. Chemosphere 15, 1489-1494. 14 Pirkle, J.L., Wolff, W.H., Patterson, Jr., D.G., Needham, L.L., Michael, J.E. Miner, J.C. and Peterson, M.R. (1987) Estimates of the half-life of 2,3,7,8-tetrachlorodibenzo-p-dioxin in ranch hand veterans. In: Seventh International Symposium on Chlorinated Dioxins and Related Compounds, University of Nevada, Las Vegas, NV, October 4-8, Abstract HT-08. 15 Byard, J.L. (1987) The toxicological significance of 2,3,7,8-tetrachlorodibenzo-p-dioxin and related compounds in human adipose tissue. J. Toxicol. Environ. Health 22, 381-403. 16 Gielen, J.E., Goujon, F.M. and Nebert, D.W. (1972) Genetic regulation of aryl hydrocarbon hydroxylation induction. J. Biol. Chem. 247, 1125-1137. 17 Poland, A., Glover, E. and Glende, AS. (1976) Stereo-specific, high affinity binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin by hepatic cytosol. J. Biol. Chem. 251, 4936-4946. 18 Himmelstein, K.J. and Lutz, R.J. (1979) A review of the applications of physiologically-based pharmacokinetic model. J. Pharmacokinet. Biopharm. 7, 127-145. 19 Ramsey, J.C. and M.E. Andersen (1984) A physiologically-based description of the inhalation pharmacokinetics of styrene in rats and humans. Toxicol. Appl. Pharmacol. 73, 159-175. 20 Andersen, M.E., Clewell, III, H.J., Gargas, M.L., Smith, F.A. and Reitz, R.H. (1987) Physiologically based pharmacokinetics and the risk assessment process for methylene chloride. Toxicol. Appl. Pharmacol. 87, 185-205.
28 21 Lindstrom, F.T., Gillett, J.W. and Rodecap, S.E. (1974) Distribution of HEOD (dieldrin) in mammals. I. Preliminary model. Arch. Environ. Contam. Toxicol. 2, 9-42. 22 Bungay, P.M. Dedrick, R.L. and Matthews, H.B. (1981) Enteric transport of chlordecone in the rat. J. Pharmacokinet. Biopharm. 9, 309-341. 23 Tuey, D.B. and Matthews, H.B. (1980) Distribution and excretion of 2,2’,4,4’,5,5’-hexabromobiphenyl in rats and man: pharmacokinetics model predictions. Toxicol. Appl. Pharmacol. 53, 420-43 1. 24 Tuey, D.B. and Matthews, H.B. (1980) Use of a physiological compartmental model for the rat to describe the pharmacokinetics of several chlorinated biphenyls in the mouse. Drug Metab. Dispos. 8, 397-403. 25 King, F.G., Dedrick, R.L., Collins, J.M., Matthews, H.B. and Birnbaum, L.S. (1983) A physiological model for the pharmacokinetics of 2,3,7,8-tetrachlorodibenzofuran in several species. Toxicol. Appl. Pharmacol. 67, 390-400. 26 Teitelbaum, P.J. (1978) The Hepatic Uptake of 2,3,7,8-Tetrachlorodibenzo-P-Dioxin in the Mouse, Ph.D. Dissertation, University of Rochester, Rochester, N.Y. 27 Birnbaum, L.S. (1986) Distribution and excretion of 2,3,7,8-tetrachlorodibenzo-p-dioxin in congenic strains of mice which differ at the Ah locus. Drug Metab. Dispos. 14, 34-40. 28 Gasiewicz, T.A. and Rucci, G. (1984) Cytosolic receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin. Evidence for a homologous nature among various mammalian species. Mol. Pharmacol. 26, 90-98. 29 Mason, M.E. and Okey, A.B. (1982) Cytosolic and nuclear binding of 2,3,7,8-tetrachlorodibenzo-pdioxin in the Ah receptor in extra-hepatic tissues of rats and mice. Eur. J. Biochem. 123, 209-215. 30 Agin, G.L. and Blau, G.E. (1982) Application of DACSL (Dow Advanced Continuous Simulation Language) to the Design and Analysis of Chemical Reactors, Am. Inst. Chem. Eng. Symp. Ser. No. 214. Vol. 78, pp. 108-118. 31 Paustenbach, D.J., Carlson, G.P., Christian, J.E. and Born, G.S. (1986) A comparative study of the pharmacokinetics of carbon tetrachloride in the rat following repeated inhalation exposures of 8 and 11.5 hr/day. Fundam. Appl. Toxicol. 6, 484-497. 32 Leung, H.W., Andersen, M.E., Ku, R.H. and Paustenbach, D.J. (1988) A physiological pharmacokinetic description of the tissue distribution and enzyme induction of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Toxicologist 8, 155.
15
TXL 01984
A physiologically based pharmacokinetic model for 2,3,7, Stetrachlorodibenzo-p-dioxin in C57BL/6J and DBA/2J mice Hon-Wing Leung’, Robert H. Kul, Dennis J. Paustenbach’** and Melvin E. Andersen2 ‘Environmental Health and Safety, Syntex (U.S.A.) Inc.. Palo Alto, CA 94304, U.S.A., and ‘4921 Egret Court, Dayton, OH 45424, US.A. (Received 18 October 1987) (Revision received 3 February 1988) (Accepted 3 March 1988) Key words: 2,3,7,8-Tetrachlorodibenzo-p-dioxin; Mathematical simulation; Toxicokinetics; Body compartment; Tissue concentration; Binding protein; (Mouse)
SUMMARY A five-compartment physiologically based pharmacokinetic (PB-PK) model was developed to describe the time course of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the tissues of both C57BL/6J and DBA/ZJ mice. The PB-PK model included binding in blood and two hepatic binding sites, one in the cytosol and the other in the microsomes. First-order metabolism occurred in the liver. Model simulations were compared to literature results for the disposition of a single intraperitoneal dose of 10 pg/kg of [‘HITCDD, reported by Gasiewicz et al. [Drug Metab. Dispos. 11 (1983) 397-4031. In contrast to previous speculation, the greater accumulation of TCDD in the liver of the C57BL/6J mouse, as compared to the DBA/ZJ mouse, was not attributable to the higher fat content in the DBA/2J mouse. Instead, the disposition of TCDD in these mice was more dependent on the affinity of the microsomal binding proteins than on fat content. The microsomal dissociation constant in the C57BL/6J mouse estimated by the PB-PK model was about one-third its value in the DBA/2J mouse (20 versus 75 nM), i.e. there is more avid microsomal binding in the liver of the CS7BL/6J mouse. In the concentration range covered in these time-course studies, the cytosolic receptor, with its low capacity and very high affinity binding characteristics, does not play a major role in determining the overall tissue distribution pattern. The concentration and affinity of the microsomal binding protein in the liver appear to be primarily responsible for explaining the differences in the liver/fat concentration ratios between various strains and species of laboratory animals. Address for correspondence: Hon-Wing Leung, Ph.D., Environmental Health & Safety, Syntex (U.S.A.) Inc., 3401 Hillview Avenue, Palo Alto, CA 94303, U.S.A. *Present address: McLaren Environmental Engineering, ChemRisk Division, 11101 White Rock Road, Ranch0 Cordova, CA 95670, U.S.A. Abbreviations: PB-PK, physiologically based pharmacokinetic; TCDD, 2,3,7,8-tetrachlorodibenzo-pdioxin, AHH, aryl hydrocarbon hydroxylase. 0378-4274/88/$ 03.50 0 1988 Elsevier Science Publishers B.V. (Biomedical Division)
16
INTRODUCTION
2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is a contaminant produced during the manufacture of chlorophenols, hexachlorophene, and phenoxy herbicides. TCDD is one of the most extensively studied chemicals in recent years, and several reviews of its toxicology have been written [l-4]. Despite these investigative efforts, the exact mechanism of TCDD toxicity is still unclear. Among the many uncertainties, the diverse species difference in susceptibility to TCDD is one of the most intriguing. The acute oral LD5o ranges from 1 pg/kg in the guinea pig to 5 mg/kg in the hamster [4]. Although the lethal dose in humans has not been precisely identified, based on several incidents involving exposed persons, it is almost certainly greater than 100 pg/kg [5]. Another interesting species difference in TCDD toxicity is the specificity of pathological changes in the various organs. While thymic atrophy occurs in all animal species tested [6], other organ effects are characteristic for a given species, e.g. hepatic lesions are striking in mice, rats, and rabbits but relatively minor in guinea pigs, monkeys, cattle and horses. Mucosal changes in the gastrointestinal and urinary systems are observed in primates and cattle but not reported in the rat, mouse or guinea pig [6]. Chloracne, a hyperplastic epidermal change, is commonly observed in exposed human populations, but not in rats, hamsters or guinea pigs. Besides acute toxic effects, there is also marked species difference in the disposition of TCDD. The excretion half-life has been reported to be about 15 days in mice [7] and hamsters [S], 30 days in rats [9], 30-94 days in guinea pigs [lo-l 11, 455 days in monkeys [ 121, and 5-7 years in man [ 13,141. The differences in half-life do not appear to be correlated with the toxic potency of TCDD. In most animal species, TCDD accumulates in the liver and adipose tissue. However, the relative distribution between these two tissues is markedly different among various animal species and strains [ 151. One possible explanation for these differences may be the relative fat content in these tissues. Since TCDD is highly soluble in lipids, distribution to fatty depots in species and strains having a high body adipose content will significantly alter its disposition relative to those which do not. Differences in distribution may also be related to the affinity and capacity of receptor protein binding. In certain strains of mice, enzyme induction (e.g. aryl hydrocarbon hydroxylase or AHH) by TCDD has been shown to segregate with a single genetic locus, known as the Ah locus [16]. The non-reponsi\ve strains, typified by DBA/2J mice, appear to have an altered receptor with lower affinity for TCDD as compared to the responsive strains, typified by C57BL/6J mice. Poland et al. [171 have suggested that the greater accumulation of TCDD in the liver of C57BL/6J mice, as compared to the DBA/2J mice, is directly related to the strain differences in the affinity of the receptor proteins for TCDD. Gasiewicz et al. [71 have also suggested that other differences such as altered metabolism and excretion may also contribute to the difference in hepatic accumulation.
17
To examine the extent to which these factors affect the differential hepatic metabolism, we have developed a physiologically based pharmacokinetic model (PB-PK) for TCDD in the C57BL/6J and DBA/2J mice. Predictive models, such as the one described here, which incorporate known anatomical, physiological, thermodynamic, and transport characteristics, have been used to describe the behavior of a variety of pharmacological agents [18] and were recently applied to environmental toxicants such as styrene [19], methylene chloride [20], dieldrin [21], chlordecone [22], polychlorinated and polybrominated biphenyls [23,24], and tetrachlorodibenzofurans [25]. Successful use of this approach for scale-up from animals to humans has been demonstrated [19,20]. The present TCDD model has five compartments and includes binding in blood and two binding proteins in the liver. It provides quantitative descriptions of the time-course of elimination and levels of TCDD in various organs of C57BL/6J mice, an AHH-inducible strain with low body fat content, and DBA/2J mice, a non-AHH-inducible strain with high body fat content.
KAB .
QC
CA ;“CB
CA
i BLOOD PERITONEAL CAVITY; AP,
BMl ,KBl KA
CVL = CUPL CVL -
BM2,KB2
,\
I
.
CVF = CFIPF
CVF
FAT
I
t
L--l cvs
I
1
cvs =
CSIPS 1
1
QS
I
MUSCLE/SKIN SLOWLY PERFUSED I
I QR
CVR = CRIPR
q
p
RICHLY PERFUSED
Fig. 1. Schematic representation of the five-compartment physiologically based pharmacokinetic for TCDD in mice. Abbreviations are defined in the Appendix.
model
18
MATERIALS
AND METHODS
Model structure The TCDD model (Fig. 1) was a modification of the general PB-PK model previously described [19]. It contained five compartments: blood, liver, fat, richly perfused tissues, and slowly perfused tissues. Since TCDD binds to protein receptors in the liver, the model was set up to provide for two hepatic binding sites, one corresponding to the non-inducible high affinity/low capacity cytosolic receptor [17], and the other to the inducible, low affinity/high capacity microsomal receptor [26]. In simulation of the intraperitoneal dose route used by Gasiewicz et al. [7], TCDD was assumed to be absorbed into the liver compartment by a first-order uptake process. The absorption rate constant (KA) of 0.02 h- ’ was chosen to fit the rise in tissue concentration with the C57BL/6J mice data. Bioavailability via this route of administration was assumed to be 100%. Mass balance differential equations for the model were similar to those used in the analysis of styrene data [19] except for the liver and blood. Definitions of the algebraic terms and an explanation of the liver and blood mass balance equations are given in the Appendix. Model parameters Three types of data are required to implement this physiological model: (1) partition coefficients; (2) physiological constants; and (3) biochemical constants. These data, which were obtained or calculated from the literature for each mouse strain being simulated, are summarized in Table I. Partition coefficient between fat and blood (Pfi was calculated from the data of Birnbaum [27]. The kidney was assumed to be representative of the richly perfused tissue and the partition coefficient (PR) was estimated from the data of Gasiewicz et al. [7]. The partition coefficient between liver and blood (PL) was assumed to be equal to that of PR. The slowly perfused tissue was assumed to be principally composed of skin and muscle. The partition coefficient of this compartment (Ps) was arbitrarily selected to fit the distribution data, but restricted to values intermediate between PR and PF. The model includes liver binding and incorporates binding constants and binding capacities that have been established by direct experimentation. BMZ and KBI of C57BL/6J mice were obtained or calculated from Gasiewicz and Rucci [28]. Teitelbaum [26] provided the value for BM2 and an estimate for KB2. Subsequent attempts at fitting the distribution data suggested that the best value for KB2 was almost exactly as Teitelbaum’s estimate. For the Ah-non-responsive DBA/2J mice, non-detectable levels of cytosolic receptor were reported [28]. However, it has been reported that the receptor is actually present, but the binding affinity may be low enough that it is below the experimental limit of detection [29]. For this reason, the BMI value of the DBA mouse was set to equal to that of the C57 mouse. The KB2 value was empirically fitted with the computer model. In addition, the model in-
19
eludes binding in the blood. Blood binding is described as a linear process with an effective equilibrium between bound and free TCDD given by the constant, KAB. This gives rise to a distribution of TCDD in blood between two forms, only one of which is exchangeable in the tissues. The fractional amount available is given by the ratio (l/(1 + KAB)).This apportioning in the blood gives rise to kinetic behavior TABLE I KINETIC CONSTANTS AND MODEL PARAMETERS BASED PHARMACOKINETIC MODEL FOR TCDD Parametera
Abbreviationsb
USED IN THE PHYSIOLOGICALLY Mice C57BL/6J
Weights (kg) Body
0.023 1
EW
Volumes ((I/obody weight) Liver Richly perfused Slowly perfused Fat Blood Cardiac output (l/h) Blood flow (W cardiac output) Liver Richly perfused Slowly perfused Fat Partition coefficients Liver/blood Richly perfused/blood Slowly perfused/blood Fat/blood Biochemical constants (nmol/liver)
DBA/ZJ 0.0239
VLC
5
5
VRC
4
4
VSC
76.1
70.5
VFC
5.9
VBC
5
5
QC
0.861
0.883
11.5
QLC QRC QSC QFC
25
25
51
51
17
17
7
7
PL
20
20
PR
20
20
PS
250
250
PF
350
350
BIUI
0.0042
@M)
KBI
0.29
(nmol/liver) (nM) (per hour) (per hour)
BM2
20
20
KB2
20
75
Dose (nmol/kg) “Scaled parameters:
2
KFC
3.25
1.75
KA
0.02
0.02
KAB
2.5
2.5
APO VR = 0.09 BW-VL;
0.0042
VS = 0.82 BW-VF;
bSee Appendix for definition of abbreviations
32 QR = 0.76 QC-QL;
for biochemical constants.
32 QS = 0.24 QC-QF.
20
similar to that observed for diffusion-limited uptake into tissues. Tissue partition coefficients used in the model are based on the distribution of free TCDD. Effective tissue:blood concentration ratios, that is the observed ratio of concentrations in various experiments, are related to total blood concentration. Thus, concentration ratios are given by Pi/(1 + KAB). There has not yet been any direct experimentation to establish this blood binding constant. The value used here (KAB = 2.5) gave good agreement with the disposition data when the literature values of the liver binding parameters were used in the simulation. Simulations The set of simultaneous differential equations was solved numerically with the aid of a Fortran-based simulation software package [30] run on an IBM PC/AT or an AT-compatible computer. RESULTS
Fig. 2 shows the simulated time-course of TCDD concentration in the liver and fat of C57BL/6J mice given 32 nmol TCDD per kg of body weight by intraperitoneal injection. The actual data sets were taken from the experimental study of Gasiewicz et al. [7]. There is good agreement between the simulated description generated by the model and the empirical data. Gasiewicz et al. [7] have observed that the liver/fat ratio of TCDD concentration in the C57BL/6J mice was about 3 times that in DBA/ZJ mice. Since DBA/2J mice had about twice the body fat content as compared to C57BL/2J mice, the difference
0
10
20
30
40
Days
Fig. 2. Time-course of 32 nmol/kg.
of TCDD disposition
in liver and fat of C57BL/6J
Lines represent the simulation by the model ( (0 fat; 0 liver) are those of Gasiewicz
mice given an intraperitoneal
fat; - - et al. [7].
dose
liver). Datum points
21
r
KB2 = IO
KB2 = 20
10 :
5 0
KB2 = 40 I
I
I
1
10
20
30
40
Days
Fig. 3. The influence of KB2 values on the time-course of TCDD concentration in the liver of C57BL/6J mice.
in hepatic concentration may be due to the greater capacity of the DBA/2J mouse to sequester the highly lipophilic TCDD in the adipose tissue. One of the strengths of physiological modeling is the ability to readily simulate the disposition of a chemical by altering certain biologic parameters that are difficult or impossible to manipulate under actual experimental conditions. Thus, we varied the fat content of the C57BL/6J mice from 3 to 12% of body weight in order to test the hypothesis that the distributional differences may have been influenced by body fat content. Simulations showed that TCDD concentration in the liver was relatively insensitive to body fat content (data not shown), indicating that this was not an important factor influencing the disposition of TCDD in the liver between the two strains of mice. Interestingly, the distributional behavior of TCDD in the C57BL/6J mice was profoundly influenced by the binding characteristics of the microsomal binding protein, especially the KB2 value. Fig. 3 shows the effect of varying the KB2 value on liver concentration/time data obtained in C57BL/6J mice following intraperitoneal administration of 32 nmol/kg TCDD. Fig. 4 shows that the model also gives good simulations of the TCDD excretion in both the C57BL/6J and DBA/2J mice. The model, however, did not give as satisfactory a simulation of the fat and liver TCDD concentrations in the DBA/2J mice if the input was set to be consistent with the uptake and elimination profile (Fig. 5). However, similar to the C57BL/6J mice, the disposition of TCDD in the DBA/2J mouse liver was also greatly influenced by the KB2 value but was rather insensitive to varying body fat content (data not shown). The best fit of the empirical data was obtained with a KB2 value of about 75 nM. Thus the TCDD binding affinity to the hepatic microsomal protein receptor in the DBA/2J mice was at least 3.5 times lower than that of C57BL/6J mice.
22
25
Fig. 4. Kinetics of TCDD excretion in mice. Computer simulation ( C57BL/6J; - - - DBA/ZJ); empirical data (0 C57BL/6J; 0 DBA/ZJ). Description of the way metabolism and excretion are simulated by the computer model is given in the Appendix.
DISCUSSION
We have constructed a flow-limited physiological model for TCDD to explain the pharmacokinetic differences in tissue distribution between C57BL/6J and DBA/2J mice. The latter is characterized as an Ah-non-responsive strain and has twice as much fatty tissue stores as the former strain. DBA/W mice accumulate less TCDD in their livers than C57BL/6J mice. Storage of highly lipophilic substances such as TCDD in the adipose tissue usually has a marked effect on the pharmacokinetics [31]. However, the PB-PK model for TCDD predicted that the size of the fat compartment alone could not explain the lower accumulation of TCDD in the liver of the DBA/%J mice compared with the C57BL/6J mice. Instead, the disposition of TCDD in the mouse was profoundly affected by the affinity of the microsomal binding protein. The differential accumulation of TCDD in the livers of the C57BL/6J and DBA/2J mice was apparently due to the differing affinity characteristics of the microsomal TCDD binding proteins. There have been several previous attempts to construct physiologically based pharmacokinetic models for polyhalogenated biphenyls and polychlorinated dibenzofurans [24,25]. In these studies the disposition of these persistent hydrocarbon chemicals was described by simple partitioning between the blood and the various
23
0
10
20
30
40
Days
Fig. 5. Time-course of TCDD disposition in liver and fat of DBA/2J mice. Computer simulation ( ~ fat; - - - liver); empirical data (0 fat; 0 liver).
tissues with first-order metabolism in the liver. The partition coefficients used in these models were estimated by determining blood and tissue levels in animals killed at various times after dosing. These models did not account for the role that extensive tissue binding to particular cellular proteins might play in determining the overall disposition of the xenobiotic. In contrast, the present description attempts to provide a biochemical basis for the observed tissue distribution. Thus, the model structure has terms for partitioning which describe the general tissue solubility of TCDD and clearly specify that TCDD has higher solubility in fat (PF = 350) than in liver (PL = 20).The tissue concentration ratio that is achieved in any dosing study, though, is dependent both on solubility and specific binding. Liver concentrations in C57BL/6J mice are higher than fat concentrations (Fig. 2), not due to some intrinsically higher solubility, but because of the high affinity tissue binding in the liver in this concentration range. Gasiewicz et al. [7] did not determine muscle concentrations of TCDD and only examined skin content on day 25 in DBA/2J mice. Since no data were available for these tissues, the assignment of a high value for PS (250)was made only to give a good fit to the liver and fat curves. To assure ourselves that conclusions on the importance of microsomal binding were not an artifact to model structure, we formulated an alternative model that maintained a low value of PS (20)and maintained that fit to the liver and fat concentrations by adjusting the intraperitoneal bioavailability to 67%. This model also gave a good description of the time-course data. More importantly, however, the behavior of this alternative formulation with changes in fat compartment volumes was nearly identical to the model shown in this paper. The liver/fat ratios were still most sensitively dependent on the microsomal binding affinities, not on the amount of fat.
25
APPENDIX
Definitions of the algebraic constants and mass balance equations for the liver and blood compartment * The liver mass balance equation contained terms for influx and efflux of TCDD in arterial and venous blood and first order metabolism:
F
= QL(CA - CVL) - q
+ (y)
/
(1 + KM?)
For describing the data from Gasiewicz et al. [7], the initial dose is the amount injected in the peritoneal cavity (APO). The instantaneous rate of absorption (dAP/dt) is first-order with respect to the amount remaining to be absorbed (AP): $!/
= KA . AP = KA ’ APO . ebKA ’ t
Total mass of TCDD in the liver at any time is the time integral of the rate equation: AL = INTEGRAL
(dAL/dt,
0)
In contrast to the description of styrene disposition [19], TCDD in the liver was distributed between two binding proteins and free TCDD: AL
I/~.~vL.P~
=
+
Bhf1-a~ + Bkf..*CVL KB2 + CVL KBI + CVL
This equation can be rearranged CVL = AL/[(VL.PL)+
*Abbreviations: of TCDD
AL?, amount
metabolized
be absorbed; BMI, microsomal protein; of total TCDD tion of TCDD TCDD
in venous
from
peritoneal
constant;
of TCDD
by microsomes
protein;
in blood;
AL, amount
in liver; AP, amount
TCDD binding capacity to cytosolic BW, body weight; CA, concentration
(free and bound)
concentration
cytosolic
BMI/(KBZ + CVL)+ BM2/(KB2+
in blood;
in venous blood leaving
CVR,
in terms of CVL:
of TCDD blood leaving
cavity
PL, liver/blood
slowly perfused
into liver; KAB,
KB2, TCDD
binding
partition
flow through liver compartment; slowly perfused tissue; t, time;
CL, concentration blood TCDD
constant
coefficient;
leaving
in liver tissue; AM,
in peritoneal
amount
cavity remaining
to
protein; BM2, TCDD binding capacity to of free TCDD in blood; CB, concentration of TCDD
fat; CVL, concentration
in venous
of TCDD
of TCDD
CVL)]
richly
tissue; DOSE, binding QF, blood
in venous blood leaving liver;
perfused
tissue;
dose of TCDD;
to blood;
to microsomal
in liver tissue; CVF, concentra-
of TCDD
KBI,
protein;
flow through
CVS,
concentration
KA, absorption
TCDD
binding
KFC, first-order fat compartment;
of
constant constant
metabolic
to rate
QL, blood
QR, blood flow through richly perfused tissue; QS, blood flow through VB, volume of blood; VL, volume of liver.
26
and solved by approximation. For comparison with Gasiewicz et al. [7], total liver concentration plotted:
(CL) was
CL = AL/V-L
The rate of metabolism dAM dt
= KFC/(BFvp
(nmol/h)
* CVL
. VL
The total amount metabolized AM
= INTEGRAL
(dAM/dt,
at any time is:
(AM) is the time integral of this equation: 0)
It is assumed that metabolized TCDD is rapidly eliminated from the body. The elimination plots as (70 excreted are derived as follows: 9’0 EXCRETED
= (AM/DOSE)
. 100
In the blood, TCDD equilibrates between free and bound forms with an equilibrium constant, KAB. The amount of TCDD in blood (AB) is the sum of the two: AB
= VB * (CA
+ KAB
* CA)
The combined concentration of TCDD in blood (CB), the parameter determined experimentally, is linearly related to free TCDD in blood (CA), which is the exchangeable TCDD at any time: CB = +j
= CA * (1 + KAB)
The absorbed TCDD enters the portal blood and is distributed between the free and bound pools of TCDD. Only the free form is available for distribution into the liver and the liver mass balance equation must include the input term: liver input = (9)/l
+ KAB)
The remainder of the absorbed material passes through to the venous blood giving a slightly modified blood mass balance equation: @/
= QF . CVF + QL . CVL dAP dt’
KAB
(1 + KAB)
+ QS + CVS
+ QR . CVR
-
QC * CA +
27 REFERENCES 1 International Agency for Research on Cancer (1977) Some Fumigants, the Herbicides 2,4-D and 2,4,5,-T, Chlorinated Dibenzodioxins and Miscellaneous Industrial Chemicals, Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans, World Health Organization, Lyons, Vol. 15, pp. 41-102. 2 Environmental Protection Agency (1985) Health Assessment Document for Polychlorinated Dibenzo-pdioxins, EPA-600/S-84014F, Environmental Protection Agency, Cincinnati, OH. 3 National Research Council of Canada (1981) Polychlorinated Dibenzo-p-dioxins: Criteria for their Effects on Man and his Environment, NRCC 18574, National Research Council of Canada, Ottawa, Ont. 4 American Medical Association (1984) The Health Effects of ‘Agent Orange’ and Polychlorinated Dioxin Contaminants: An Update, American Medical Association Press, Chicago, IL. 5 Tschirley, F.H. (1986) Dioxin. Sci. Am. 254, 29-35. on the 6 Vos, J.G., Moore, J.A. and Zinkl, J.G. (1973) Effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin immune system of laboratory animals. Environ. Health Perspect. 19, 149-162. 7 Gasiewicz, T.A. Geiger, T.A., Rucci, G. and Neal, R.A. (1983) Distribution, excretion, and metabolism of 2,3,7,8-tetrachlorodibenzo-p-dioxin in C57BL/6J, DBA/ZJ and B6D2F,/J mice. Drug Metab. Dispos. 11, 397-403. 8 Olson, J.R., Gasiewicz, T.A. and Neal, R.A. (1980) Tissue distribution, excretion and metabolism of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in the Golden Syrian hamster. Toxicol. Appl. Pharmacol. 56, 78-85. 9 Neal, R.A., Olson, J.R., Gasiewicz, T.A. and Geiger, L.E. (1982) The toxicokinetics of 2,3,7,8-tetrachlorodibenzo-p-dioxin in mammalian systems. Drug Metab. Rev. 13, 355-385. 10 Olson, J.R. (1986) Metabolism and disposition of 2,3,7,8-tetrachlorodibenzo-p-dioxin in guinea pigs. Toxicol. Appl. Pharmacol. 85, 263-273. 11 Gasiewicz, T.A. and Neal, R.A. (1979) 2,3,7,8-Tetrachlorodibenzo-p-dioxin tissue distribution, excretion, and effects on clinical chemical parameters in guinea pigs. Toxicol. Appl. Pharmacol. 51, 329-339. 12 Bowman, R.E., Schantz, S.L., Weerasinghe, N.C.A., Meehan, J. Pan, J., Gross, M.L., Boehm, K.M. Van Miller, J.P. and Welling, P.G. (1987) Clearance of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) from body fat of rhesus monkeys following chronic exposure. Toxicologist 7, 158. 13 Poiger, H. and Schlatter, C. (1986) Pharmacokinetics of 2,3,7,8-TCDD in man. Chemosphere 15, 1489-1494. 14 Pirkle, J.L., Wolff, W.H., Patterson, Jr., D.G., Needham, L.L., Michael, J.E. Miner, J.C. and Peterson, M.R. (1987) Estimates of the half-life of 2,3,7,8-tetrachlorodibenzo-p-dioxin in ranch hand veterans. In: Seventh International Symposium on Chlorinated Dioxins and Related Compounds, University of Nevada, Las Vegas, NV, October 4-8, Abstract HT-08. 15 Byard, J.L. (1987) The toxicological significance of 2,3,7,8-tetrachlorodibenzo-p-dioxin and related compounds in human adipose tissue. J. Toxicol. Environ. Health 22, 381-403. 16 Gielen, J.E., Goujon, F.M. and Nebert, D.W. (1972) Genetic regulation of aryl hydrocarbon hydroxylation induction. J. Biol. Chem. 247, 1125-1137. 17 Poland, A., Glover, E. and Glende, AS. (1976) Stereo-specific, high affinity binding of 2,3,7,8-tetrachlorodibenzo-p-dioxin by hepatic cytosol. J. Biol. Chem. 251, 4936-4946. 18 Himmelstein, K.J. and Lutz, R.J. (1979) A review of the applications of physiologically-based pharmacokinetic model. J. Pharmacokinet. Biopharm. 7, 127-145. 19 Ramsey, J.C. and M.E. Andersen (1984) A physiologically-based description of the inhalation pharmacokinetics of styrene in rats and humans. Toxicol. Appl. Pharmacol. 73, 159-175. 20 Andersen, M.E., Clewell, III, H.J., Gargas, M.L., Smith, F.A. and Reitz, R.H. (1987) Physiologically based pharmacokinetics and the risk assessment process for methylene chloride. Toxicol. Appl. Pharmacol. 87, 185-205.
28 21 Lindstrom, F.T., Gillett, J.W. and Rodecap, S.E. (1974) Distribution of HEOD (dieldrin) in mammals. I. Preliminary model. Arch. Environ. Contam. Toxicol. 2, 9-42. 22 Bungay, P.M. Dedrick, R.L. and Matthews, H.B. (1981) Enteric transport of chlordecone in the rat. J. Pharmacokinet. Biopharm. 9, 309-341. 23 Tuey, D.B. and Matthews, H.B. (1980) Distribution and excretion of 2,2’,4,4’,5,5’-hexabromobiphenyl in rats and man: pharmacokinetics model predictions. Toxicol. Appl. Pharmacol. 53, 420-43 1. 24 Tuey, D.B. and Matthews, H.B. (1980) Use of a physiological compartmental model for the rat to describe the pharmacokinetics of several chlorinated biphenyls in the mouse. Drug Metab. Dispos. 8, 397-403. 25 King, F.G., Dedrick, R.L., Collins, J.M., Matthews, H.B. and Birnbaum, L.S. (1983) A physiological model for the pharmacokinetics of 2,3,7,8-tetrachlorodibenzofuran in several species. Toxicol. Appl. Pharmacol. 67, 390-400. 26 Teitelbaum, P.J. (1978) The Hepatic Uptake of 2,3,7,8-Tetrachlorodibenzo-P-Dioxin in the Mouse, Ph.D. Dissertation, University of Rochester, Rochester, N.Y. 27 Birnbaum, L.S. (1986) Distribution and excretion of 2,3,7,8-tetrachlorodibenzo-p-dioxin in congenic strains of mice which differ at the Ah locus. Drug Metab. Dispos. 14, 34-40. 28 Gasiewicz, T.A. and Rucci, G. (1984) Cytosolic receptor for 2,3,7,8-tetrachlorodibenzo-p-dioxin. Evidence for a homologous nature among various mammalian species. Mol. Pharmacol. 26, 90-98. 29 Mason, M.E. and Okey, A.B. (1982) Cytosolic and nuclear binding of 2,3,7,8-tetrachlorodibenzo-pdioxin in the Ah receptor in extra-hepatic tissues of rats and mice. Eur. J. Biochem. 123, 209-215. 30 Agin, G.L. and Blau, G.E. (1982) Application of DACSL (Dow Advanced Continuous Simulation Language) to the Design and Analysis of Chemical Reactors, Am. Inst. Chem. Eng. Symp. Ser. No. 214. Vol. 78, pp. 108-118. 31 Paustenbach, D.J., Carlson, G.P., Christian, J.E. and Born, G.S. (1986) A comparative study of the pharmacokinetics of carbon tetrachloride in the rat following repeated inhalation exposures of 8 and 11.5 hr/day. Fundam. Appl. Toxicol. 6, 484-497. 32 Leung, H.W., Andersen, M.E., Ku, R.H. and Paustenbach, D.J. (1988) A physiological pharmacokinetic description of the tissue distribution and enzyme induction of 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat. Toxicologist 8, 155.