ToxSci Advance Access originally published online on July 10, 2006
Toxicological Sciences 2006 93(2):432-442; doi:10.1093/toxsci/kfl056
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Development of a Physiologically Based Pharmacokinetic Model for Deltamethrin in the Adult Male Sprague-Dawley Rat
,
* Department of Environmental Health Science, College of Public Health, University of Georgia, Athens, Georgia 30602-2102; Department of Pharmaceutical and Biomedical Sciences, College of Pharmacy, University of Georgia, Athens, Georgia 30602-2354; Pharmacology Department, National Institute of Toxicological Research, Korea Food and Drug Administration, Seoul 122-704, South Korea; and U.S. Environmental Protection Agency, Office of Research and Development, National Exposure Research Laboratory, Research Triangle Park, North Carolina 27709
1 To whom correspondence should be addressed. Fax: 706-542-7472. E-mail: jwfisher{at}uga.edu.
Received February 20, 2006; accepted June 29, 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAAPPENDIXREFERENCES |
|---|
Deltamethrin (DLT) is a type II pyrethroid insecticide widely used in agriculture and public health. DLT is a potent neurotoxin that is primarily cleared from the body by metabolism. To better understand the dosimetry of DLT in the central nervous system, a physiologically based pharmacokinetic (PBPK) model for DLT was constructed for the adult, male Sprague-Dawley rat that employed both flow-limited (brain, gastrointestinal [GI] tract, liver, and rapidly perfused tissues) and diffusion-limited (fat, blood/plasma, and slowly perfused tissues) rate equations. The blood was divided into plasma and erythrocytes. Cytochrome P450–mediated metabolism was accounted for in the liver and carboxylesterase (CaE)-mediated metabolism in plasma and liver. Serial blood, brain, and fat samples were taken for DLT analysis for up to 48 h after adult rats received 2 or 10 mg DLT/kg po. Hepatic biotransformation accounted for
78% of these administered doses. Plasma CaEs accounted for biotransformation of 8% of each dosage. Refined PBPK model forecasts compared favorably to the 2- and 10-mg/kg po blood, plasma, brain, and fat DLT profiles, as well as profiles subsequently obtained from adult rats given 1 mg/kg iv. DLT kinetic profiles extracted from published reports of oral and iv experiments were also used for verification of the model's simulations. There was generally good agreement in most instances between predicted and the limited amount of empirical data. It became clear from our modeling efforts that there is considerably more to be learned about processes that govern GI absorption and exsorption, transport, binding, brain uptake and egress, fat deposition, and systemic elimination of DLT and other pyrethroids. The current model can serve as a foundation for construction of models for other pyrethroids and can be improved as more definitive information on DLT kinetic processes becomes available. Key Words: deltamethrin; PBPK modeling; male Sprague-Dawley rat; pyrethroids; toxicokinetics.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAAPPENDIXREFERENCES |
|---|
Pyrethroid insecticides are used extensively in agriculture, forestry, and public health (Casida and Quistad, 1998
; Soderlund et al., 2002). Their use in the United States has increased substantially during the past decade, due in part to the phasing out of many organophosphates. Traditionally, pyrethroids are divided into two classes (types I and II) based on their structure and toxicological actions. Type I compounds do not contain a cyano group while type II compounds do. Tremor and paresthesias are the major signs of poisoning by type I compounds. Salivation, hyperexcitability, choreoathetosis, and seizures are the major signs of type II poisoning (Lawrence and Casida, 1982; Soderlund et al., 2002). Deltamethrin (DLT), a type II compound, is one of the most potent pyrethroid insecticides (Ruzo et al., 1978; Wolansky et al., 2006). DLT is widely used in veterinary products, as well as in insecticide formulations (ATSDR, 2003; CALEPA, 2001).
The acute toxicity and mode of action of DLT have been studied extensively. Following administration of a single dose to rats (1.75 mg/kg iv, 7.5 or 10 mg/kg ip, or 15 or 50 mg/kg po), profuse salivation, coarse body tremors, biting or gnawing, lacrimation, choreoathetosis, and mortality were observed (Gray and Rickard, 1982; Rickard and Brodie, 1985; Soderlund et al., 2002). The marked neurotoxic effects of DLT have been attributed primarily to its binding to voltage-sensitive sodium channels (Choi and Soderlund, 2006; Soderlund et al., 2002). The parent compound is considered to be the proximate neurotoxicant, as demonstrated by correlation of brain DLT levels with the onset of signs of poisoning in rats (Rickard and Brodie, 1985). Intracerebral injection of DLT produced signs of neurotoxicity within a minute or less in mice (Lawrence and Casida, 1982). No toxicity studies were located in which DLT metabolites were administered to animals.
Toxicokinetic data for DLT and other pyrethroids are incomplete and in some cases contradictory. DLT is known to be rapidly absorbed when administered po or ip and to readily enter the central nervous system (CNS) (Anadon et al., 1996; Rickard and Brodie, 1985; Sheets et al., 1994). Anadon et al. (1996) reported much higher DLT concentrations in most brain regions than in plasma of orally dosed rats, whereas Rickard and Brodie (1985) reported relatively low brain levels in rats after ip injection. Hayes and Laws (1990) reported half-lives of 1–2 days for brain and 5 days for fat in rats administered 30 mg DLT/kg po.
Metabolism of DLT in mice and rats has been known for some time to involve esterase-mediated cleavage of the ester linkage and ring hydroxylation by cytochrome P450s (CYPs) (Ruzo et al., 1978, 1979; Soderlund and Casida, 1977). Metabolic rate constants for these pathways were recently determined in vitro for the adult, male Sprague-Dawley (SD) rat (Anand et al., 2006). Van Dijk and Burri (1993) reported that 27% of a 2.4-mg/kg iv dose of DLT was excreted in the feces of female rats, possibly as metabolites. Ruzo et al. (1978) observed that 13–20% of DLT was eliminated in feces of rats within 2–4 days after oral bolus dosing with 0.9–1.6 mg DLT/kg, suggesting partial systemic uptake or exsorption of the parent compound.
Physiologically based pharmacokinetic (PBPK) models have been useful tools in dose-response analysis and risk assessment of a variety of toxic chemicals. There is, however, no published PBPK model for DLT or other pyrethroids. The objective of this research was to develop a PBPK model for DLT in the adult, male SD rat, as an initial step in the construction of PBPK models for DLT and other pyrethroids in immature rats and humans. Kinetic time-course data for blood and plasma, brain (the target organ), and fat (the primary storage site) were collected to characterize the disposition of orally administered DLT. Initial and refined PBPK model predictions of DLT dosimetry in the adult rat are presented and compared to published laboratory data.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAAPPENDIXREFERENCES |
|---|
Chemicals
DLT [(S)-alpha-cyano-3-phenoxybenzyl-(1R,cis)2,2-dimethyl-3-(2,2-dibromovinyl)-cyclopropane carboxylate] was obtained from ChemService (Westchester, PA). It was 98% pure. Methanol (high-performance liquid chromatography [HPLC] grade) and acetonitrile (ACN) (HPLC grade) were procured from Burdick and Jackson (Muskegon, MI). Glycerol formal was obtained from Sigma-Aldrich (St Louis, MO). Sulfuric acid and deionized water (HPLC grade) were purchased from J.T. Baker (Phillipsburg, NJ).
Animals
Adult, male SD Crl:CD rats were purchased from Charles River Laboratories (Raleigh, NC). They weighed 327–490 g (mean body weight [BW] = 414 g) and were 90–110 days old at the time of experimentation. Timed pregnant SD rats were purchased from the same source. The protocol for this study was approved by the University of Georgia Animal Care and Use Committee, and all animals were housed in an Association for Assessment and Accreditation of Laboratory Animal Care–approved animal facility with a 12-h light/dark cycle at 21°C and 50 ± 10% humidity. Purina Rat Chow 5001 (Purina, St Louis, MO) and tap water were available ad libitum. The adult rats were allowed a minimum acclimation period of 2 weeks before experiments were initiated. The females were housed in shoebox cages while they delivered and nursed their offspring. All dosings commenced between 0900 h and 1000 h.
Experiment
DLT was dissolved in glycerol formal prior to dosing. Adult rats were fasted for 12 h before being dosed with 2 or 10 mg DLT/kg po in a total volume of 1 ml/kg. Groups of five rats were sacrificed 1, 2, 6, 12, 24, and 48 h after dosing. Arterial blood, whole brain, and perirenal fat were collected at each time point after euthanizing the animals with CO2. An additional group of adult animals was utilized for an iv experiment. Each of these rats was anesthetized in our laboratory by im injection of 0.1 ml/100 g BW of a "cocktail" consisting of ketamine hydrochloride (100 mg/ml), acepromazine maleate (10 mg/ml), and xylazine hydrochloride (20 mg/ml) (3:2:1, vol:vol:vol). A cannula (PE50 polyethylene tubing) was surgically inserted into the right carotid artery and securely ligated. The cannula was passed under the skin and exteriorized at the nape of the neck, so the animals could move about freely following their recovery. Water was provided, but food was withheld during the 24-h postsurgical recovery period before dosing. These subjects received 1 mg DLT/kg iv via a caudal vein. Serial micro blood samples were then taken from these unanesthetized animals via the carotid cannula. Serial sacrifices were performed in a separate experiment to obtain blood, liver, and skeletal muscle specimens from 10-day-old SD rats given 2 mg DLT/kg po. At sacrifice of the 10-day-old rats, each animal was decapitated at 0.5, 1, 2, 6, 12, 24, and 48 h after dosing. Blood samples were collected in heparinized tubes, and skeletal muscle and liver samples were analyzed for DLT content. These liver and skeletal muscle time-course data were used to calculate liver:plasma and slowly perfused tissue (muscle):plasma partition coefficient (PC) values. The immature rat tissue data will be presented elsewhere. Each of the biological samples was placed into a 2-ml screw cap vial (National Scientific Co., Scottsdale, AZ), sealed, and frozen at – 80°C until analysis. Blood samples were centrifuged (13,000 rpm for 5 min) for plasma separation. Blood, plasma, and RBCs were assayed separately for their DLT concentration. The distribution of DLT between plasma and RBC did not vary appreciably with time of sampling after dosing (data not shown).
Analysis
DLT concentrations were measured by the HPLC method of Kim et al. (2006). Tissues were homogenized in 50% ACN in water (1:4). DLT was extracted from tissues, blood and plasma by vortexing each with twice its volume of ACN. An aliquot of the supernatant was injected directly onto a C18 Ultracarb 5 ODS column (Phenomenex, Torrance, CA) in a Shimadzu HPLC system (Shimadzu, Canby, OR). Our limits of quantitation and detection were 0.05 and 0.01 µg/ml, respectively.
PBPK Model Development
AcslXtreme v1.3.2 (Aegis Technologies Group, Inc., Huntsville, AL) simulation software was used to implement the PBPK model for DLT. The structure for the model is presented in Figures 1A and 1B.
|
Oral absorption of DLT into the systemic circulation is a nonlinear (dose dependent) process, resulting in less than complete systemic uptake and apparent dose-dependent fecal excretion of orally administered DLT. Instead of assuming direct uptake of oral DLT into the liver, the model was configured with a gastrointestinal (GI) tract (and lumen), from which DLT reached the liver via the portal vein (Fig. 1A). The rate of absorption of DLT from the lumen into the blood supply of the GI tract was described empirically using two compartments (Fig. 1B and Tables 1 and 2). These oral uptake kinetic constants are referred to as the gastric and intestinal absorption rate constants (i.e., KS and KI, respectively) and the gastric-emptying rate constant KSI. The constant values were determined by visual fitting of predicted plasma concentrations of DLT with observed concentrations for the 10-mg DLT/kg po group.
|
|
Fecal excretion of unabsorbed DLT from the intestinal compartment (Fig. 1B) was simply described with an empirically calibrated Michaelis-Menten equation (VmaxFec and KmFec, Tables 1 and 2) by fitting limited fecal excretion data (three data points) after oral dosing with DLT. These fecal elimination data were obtained from Bosch (1990)
, who administered 0.55 mg DLT/kg po to rats and reported that 24–28% of the dose was excreted in the feces, and from Ruzo et al. (1978), who reported similar findings of 21 and 14% fecal elimination following oral bolus doses of 0.9 and 1.6 mg/kg of DLT, respectively. The fraction of oral DLT that is excreted in feces for doses less than 0.6 mg/kg is unknown. An in vivo experiment was conducted, in which SD rats were given 10 mg DLT/kg po. Blood samples were taken 2 and 6 h postdosing and centrifuged at 13,000 rpm for 5 min for plasma separation. Blood, plasma, and RBCs were assayed for their DLT concentrations. (These data are to be reported elsewhere.) This experiment revealed that 93% of DLT in the blood was present in the plasma. Experimental plasma and tissue time-course data were used to estimate in PC values for DLT. The tissue/plasma PCs (otherwise known as distribution ratios) were calculated as the ratio of the area under the tissue DLT concentration versus time curve (AUC) to the plasma AUC. Time-course data from the experiment in which adult rats received 2 mg DLT/kg po were used to derive fat:plasma and brain:plasma PC values. The skeletal muscle (i.e., slowly perfused) and liver PC values were based on tissue data from the analogous experiment with 10-day-old rats (data described elsewhere). The rapidly perfused:plasma and GI:plasma PCs were set equal to the liver:plasma ratio.
Preliminary PBPK model.
A preliminary model was constructed with seven flow-limited compartments to explore the kinetics of DLT. The PBPK model compartments (Fig. 1A) consisted of the blood (BL) and plasma (P), brain (B), liver (L), fat (F), GI tract, and lumped tissues (rapidly [R] and slowly [S] perfused tissues). Fat was included in the model structure because DLT is very lipophilic. The brain is the primary target organ (ATSDR, 2003). Liver and plasma were included in the model structure in order to describe hepatic and extrahepatic metabolism. The GI tract was chosen for anatomical completeness. Blood was represented by two subcompartments, namely erythrocytes and plasma. The arterial input and venous output of each tissue compartment were assumed to consist solely of plasma. Metabolic constants are described below.
The initial model was created and exercised by simulating the DLT iv and po tissue time-courses. Attempts to simulate the empirical data for plasma and tissues were initially undertaken using only flow-limited equations (Figs. 2A–2C). The most obvious disagreements between the model predictions and observations were with adipose tissue (Fig. 2C).
|
Refined PBPK model.
A revised model was constructed that included both flow- and diffusion-limited compartments (Fig. 1). Visual inspection of the experimental data revealed that clearance of DLT from fat was much slower than clearance from plasma. In the refined model, three compartments were considered to be diffusion limited (plasma/erythrocytes [E], fat and slowly perfused tissues). An initial estimate of the permeability area (PA) value for fat was determined by visual inspection after fitting the 10-mg/kg po data sets. The PA value for slowly perfused tissues was arbitrarily set to 0.7 (l/h). The same chemical-specific input parameters were utilized in both the preliminary and refined models, with the exception of the PA terms (Table 1).
DLT is largely metabolized in rats by liver CYPs and carboxylesterases (CaEs) and by plasma CaEs. Anand et al. (2006) assessed the relative contribution of each in vitro by monitoring the rates of disappearance of the parent compound upon incubation of a series of concentrations of DLT with adult SD rat plasma and liver microsomes. The Km and Vmax values for each metabolic pathway were incorporated into the PBPK model without any modifications. Physiological input parameters (i.e., tissue blood/plasma flows, blood/plasma volume fractions, and tissue volumes) are included in Table 2. Tissue volumes (Schoeffner et al., 1999) and tissue blood flows (Delp et al., 1991) have previously been measured in adult, male SD rats in our laboratories. Plasma volume fractions (PVFs) for fat and slowly perfused tissues were calculated using the whole-blood PVF (PVFBL) and their respective blood volume fractions (BVFs), as shown in the Appendix. The muscle BVF value was taken from Brown et al. (1997), and the fat BVF was obtained from Andersen et al. (1993).
Model Validation
Several published kinetic data sets were utilized to evaluate the ability of our PBPK model to predict DLT behavior. Anadon et al. (1996) dosed male Wistar rats orally with 26 mg/kg of DLT dissolved in sesame oil and measured DLT concentrations in plasma, anoccygeus muscle, and other tissues for up to 48 h. The DLT kinetic data for muscle were used for model development. Gray and Rickard (1982) conducted serial analyses of radioactivity levels in plasma, cerebrum, cerebellum, spinal cord, and liver of female LAC:Porton rats given 1.75 mg 14C-DLT/kg iv. Marei et al. (1982) dosed male SD rats orally with 3 mg DLT/kg in glycerol formal and monitored fat for DLT persistence over 3 weeks. Haines et al. (2004) obtained limited plasma and brain DLT profiles from male Long-Evans rats given 30 mg DLT/kg po in corn oil. Time-course data from each of these studies were used to evaluate the performance of our refined PBPK model. The same chemical-specific and physiological parameters we utilized for the male SD rat were employed to model the kinetics of DLT in each of the rat strains/sexes used by the other investigators.
Sensitivity Analysis
Sensitivity analysis was performed to identify model parameters that primarily influenced the peak blood and peak brain DLT concentrations. Each parameter was increased by 0.1% and dose metrics of interest computed. Normalized Sensitivity Coefficients (NSCs) were calculated using Equation 1 (Evans and Andersen, 2000).
 | (1) |
r is the change in the response variable, p is the value of the parameter of interest (e.g., volume of fat), and p is the change in the parameter value. Each parameter was changed 0.1% (i.e., p ÷ p = 0.001). To maintain mass balance, BW was recomputed after changing a parameter value of compartmental volume. Similarly, plasma flow (QP) was recomputed after changing a parameter value of compartmental flow. NSC values were calculated for 43 model parameters for the two dose metrics following oral dosing with 2 mg/kg DLT. The time to peak concentration was 1.6 h for blood and brain. |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAAPPENDIXREFERENCES |
|---|
Model Development and Evaluation
Preliminary and refined PBPK model predictions of plasma, blood, and fat DLT concentrations were compared to measured concentrations in Figures 2A–2C, respectively. The plasma/blood and fat profiles were obtained by serial sacrifices of groups of rats following oral administration of 10 mg DLT/kg in glycerol formal. The preliminary model employed flow-limited equations for each of its compartments. This initial model modestly underpredicted the maximum plasma and blood DLT levels at 1 and 2 h (Figs. 2A and 2B). The revised model more accurately forecast this early rise, while each model adequately predicted the decline in plasma and blood levels from 6 to 48 h postdosing. Flow-limited assumptions failed to describe the kinetics of DLT in adipose tissue (Fig. 2C). Diffusion-limited uptake and clearance of DLT from fat was adequately described by using a visually fitted PA value of 4.0 x 10–3. The PA term for exchange between erythrocytes and plasma was set to a value of 1.0 l/h because plasma and whole-blood profiles followed similar patterns (Figs. 2A and 2B). The PA term for slowly perfused tissues was set to a value of 0.7 l/h to describe the relatively slow clearance of DLT from muscle (Fig. 6). This provided adequate agreement between measured (Anadon et al., 1996
) and predicted concentrations of DLT for the first 12 h after dosing. The clearance of DLT at later times may be somewhat slower than what the model predicted. Additional empirical muscle data would be helpful. The use of diffusion-limited equations resulted in dramatic improvement in prediction of DLT deposition in fat (Fig. 2C). Fecal excretion of DLT was adequately predicted using VmaxFec and KmFec values of 0.03 mg/kg/h and 0.45 mg/l (Fig. 2D). Interestingly, these limited data suggest that the fecal excretion of DLT may be greater at lower oral doses.
|
) and plasma (
) of male Wistar rats given 26 mg DLT/kg po by Anadon et al. (1996)Good agreement was obtained between model-predicted and measured plasma DLT concentrations during the distribution and terminal elimination phases in SD rats given 1 mg/kg iv (Fig. 3). Figure 4 shows model-predicted and DLT concentrations in plasma, blood, brain, and fat measured following 2 and 10 mg DLT/kg po. The model simulations matched the kinetics of DLT quite well in each biological medium. Simulations of DLT kinetics revealed that for po doses of 2 and 10 mg/kg, the majority of DLT was metabolized by the liver (
80%). The P450 pathway accounted for 60% of the hepatic metabolism. Hepatic and plasma CaEs accounted for 26% of total DLT metabolism based on the simulations. No evidence of metabolic saturation was manifest in this dosage range. Details of the above simulations are presented in Table 3.
|
|
|
Model Validation
There was generally good agreement between the refined model's blood and liver simulations and empirical time-course data from female LAC:Porton rats given 1.76 mg 14C-DLT/kg iv by Gray and Rickard (1982)
(Fig. 5A). Peak liver concentrations were accurately predicted, but peak blood concentrations were higher than predicted. Blood and liver DLT levels diminished somewhat more rapidly than forecast. Model-simulated whole-brain DLT levels were plotted against levels measured in the cerebellum, cerebrum, and spinal cord in Figure 5B. There was adequate concordance during the redistribution and terminal elimination phases. DLT levels from 30 min to 1 h were underpredicted, as were intermediate blood time points in our own iv data set (Fig. 3).
|
), and spinal cord (Predicted plasma levels of DLT were substantially greater than levels reported by Anadon et al. (1996)
for the first 36 h following 26 mg/kg po (Fig. 6). The reason for this discrepancy is unknown, but could be related to several issues including kinetic differences between strains of rats, the severe toxicity experienced at this high dosage level, and/or the oil dosage vehicle utilized.
Limited plasma and brain DLT time-course data from experiments with male Long-Evans rats were provided by Haines et al. (2004) for model validation. The considerable intersubject variability evident in Figures 7A and 7B may have resulted from injury of some animals by the high (30 mg/kg) oral dose. Plasma DLT concentrations were slightly overpredicted, though the simulation of brain concentrations remained within the range of empirical levels for the 6-h measurements made.
|
The simulations for adipose tissue time-course data of Marei et al. (1982)
were also acceptable (Fig. 8). These researchers dosed male SD rats with 3 mg DLT/kg po in glycerol formal and monitored DLT levels in the animals' adipose tissue for 21 days but did not sample during the uptake phase.
|
Sensitivity Analysis
Sensitivity coefficients (NSC) were calculated for 43 model parameters for two dose metrics, peak brain and peak blood concentrations. The NSC coefficients ranged in value from – 0.61 to 0.99 for the brain-related dose metric, and from – 0.61 to 0.65 for the blood-related dose metric. Of the 43 model parameters tested for the 2-mg/kg oral dose, 14 were found to have |NSC| > 0.1 (Fig. 9). These two sets of coefficients were highly correlated with each other (r = 0.99 and n = 13), except for the NSCs for the brain:plasma PC (PCB), which was logically influential on the brain-related dose metric (NSC = 0.99) but not the blood-related dose metric (NSC = – 0.01). Blood and brain peak concentrations were sensitive to liver: plasma PC (PCL), the liver CYP metabolic constants (Km1 and Vmaxc1), and the liver CaE constants (Km2 and Vmaxc2). These dose metrics were only moderately influenced by metabolic clearance in the blood (NSC [Vmaxc3] = – 0.05 and NSC [Km3] = 0.05). Peak blood and brain concentrations were also positively sensitive to the changes in the first-order GI transfer rate (KSI) and intestinal absorption rate (KI). This pattern was expected, since DLT absorption occurs primarily in the intestine in our model. The dose metrics were sensitive to the capacity constant for fecal excretion VmaxFecc (NSC = – 0.14, for both) but not to KmFec (NSC = 0.02, for both). Both parameters were influential at low dosage (0.5 mg/kg po), and neither was influential at higher dosage (10 mg/kg po).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAAPPENDIXREFERENCES |
|---|
Relatively little has been reported on the pharmacokinetics of DLT or other pyrethroids in humans and other mammals, despite the insecticides' widespread agricultural and household use. A PBPK model for DLT was constructed for the adult SD rat. The model utilized both flow- and diffusion-limited kinetics to describe the time-course of DLT in blood and tissues. DLT exhibited pronounced diffusion-limited behavior in the fat. Systemic clearance of DLT was assumed to occur by metabolism as well as fecal excretion. As the biotransformation of DLT involves two metabolic pathways (CyPs, CaE) and two primary metabolizing tissues (liver and blood), the metabolic constants were determined in vitro (Anand et al., 2006
). Oral bioavailability was assumed to be dose dependent. This first-generation hybrid PBPK model for DLT was generally successful in describing the kinetics of DLT in several tissues, including the target organ (brain). Both the acute neurotoxicity of DLT and our analytical limit of detection precluded collection of data over a wider range of doses. Despite these limitations, we were able to obtain comprehensive DLT time-course data sets for rats administered single oral bolus doses of 2 and 10 mg/kg po and 1 mg/kg iv for model development. The preliminary model was refined and subsequently validated by comparing its predictions with published data from experiments of other researchers involving single po DLT doses of 3, 26, and 30 mg/kg and a single iv dose of 1.75 mg/kg.
PBPK models, including the present one for DLT in the adult rat, reveal gaps in knowledge and often indicate areas for further research. To that end, specific kinetics issues for DLT are discussed below. Anadon and his coworkers (1996) conducted the most comprehensive kinetic study of DLT. DLT was found to be absorbed rapidly (Tmax = 1.83 h) from the GI tract of fasted rats. Our own blood/plasma profiles and PBPK model simulations for fasted rats given 2 and 10 mg DLT/kg po show maximal or near-maximal DLT concentrations at the first two sampling times (i.e., 1 and 2 h postdosing). Nevertheless, it is unclear how extensively DLT and other pyrethroids are absorbed from the GI tract, as published data are very limited. Anadon et al. (1996) concluded that DLT was only partially absorbed, as reflected by a long GI residence time and low bioavailability in fasted rats. Ruzo et al. (1978) and Bosch (1990) have apparently reported the only fecal excretion data. In their experiments, rats received oral doses ranging from 0.6 to 1.6 mg 14C-DLT/kg. The 14C in feces was primarily DLT. Three of their data points were included in our Figure 2D. Surprisingly, our modeling with these data indicated that fecal excretion of DLT would vary inversely with the orally administered dose. This implies that bioavailability of low DLT doses are substantially less than that of relatively high doses. This behavior may also indicate transporter-mediated disposition of DLT in rat.
A major elimination pathway for lipophilic chemicals such as hexachlorobenzene is passive diffusion from blood into the lumen of the lower GI tract (i.e., exsorption) (Rozman et al., 1985). Evidence to date that DLT or other pyrethroids are P-glycoprotein (P-gp) substrates is very limited. Cypermethrin, a structurally related pyrethroid, is reported to be a P-gp substrate in two insect species (Buss et al., 2002; Srinivas et al., 2004). P-gp, an ATP-binding cassette (ABC) transporter, is abundantly expressed in the biliary cannalicular membrane of the hepatocytes (Endres et al., in press; Leslie et al., 2005). Therefore, DLT could also be returned to the GI tract via bile, though Crawford et al. (1981) found no parent compound in the bile within 4–5 h of giving anesthetized rats using 1.7–2.5 mg cypermethrin/kg po. Focused experiments are necessary to better understand these complex absorption and exsorption processes, so that they can be quantified and incorporated into an improved PBPK model for DLT and other pyrethroids.
The observed erythrocyte:plasma PC (PCE) was quite low, relative to the tissue:plasma PC values (Table 1). Although, RBCs express very low levels of P-gp, Abraham et al. (2001) described the presence of substantial levels of other ABC proteins (i.e., MRP1 and CFTR) in RBC membranes, which could potentially participate in DLT efflux.
Adipose tissue played a major role in the systemic disposition of DLT. Near-maximal concentrations in perirenal fat, reached by 6 h after oral dosing in the current experiments, substantially exceeded concentrations in blood and other tissues. As anticipated, elimination from the fat was much slower. Marei et al. (1982) reported similar findings with DLT and several other pyrethroids in rats. As mentioned previously, it was necessary to modify the fat compartment in our PBPK model to be diffusion limited, in order to predict the slow release of DLT.
Our experimental brain DLT time-course data and most published brain profiles (Gray and Rickard, 1982; Haines et al., 2004) were predicted reasonably well. The compound's very high lipophilicity (log KOW = 6.1) favors its diffusion through endothelial, pericyte, and astrocyte membranes. DLT's molecular weight of 505.2 would not be expected to hinder its passage through the blood:brain barrier (BBB). The progressive decline in brain DLT levels predicted by the PBPK model was consistent with our experimental findings and those of Gray and Rickard (1982) and Marei et al. (1982).
The PBPK model assumes uniform distribution and no binding of DLT in the brain and other body compartments. Gray and Rickard (1982) and Rickard and Brodie (1985) observed only minor regional differences in DLT brain concentrations in rats administered DLT by iv and ip injections. We also found relatively little variability in DLT from one brain region to another in orally dosed immature rats (data not shown). In contrast, Anadon et al. (1996) reported marked differences in amounts of DLT present in six parts of the brain of adult Wistar rats that received 26 mg/kg po. Six hours after dosing, the mean DLT concentration in the hypothalamus was 72 times higher than that in the medulla oblongata, the region with the lowest concentration. It may be worthy of note that the BBB is much less prominent in the hypothalamus than in the other brain regions analyzed by Anadon and his coworkers (1996).
Our experiments revealed that DLT levels in the brain were substantially lower than in blood. It is unclear why CNS levels are so low. Rat brain homogenate was found to slowly metabolize DLT to a very limited extent (Rickard and Brodie, 1985). Substantial P-gp expression in the BBB (Leslie et al., 2005) may serve as an effective efflux transporter for DLT, but this has not been confirmed. The largely phospholipid composition of the brain may be an important factor in the limited partitioning of the highly lipophilic compounds. The solubilities of organic chemicals in phospholipids are significantly lower than in neutral lipids (Poulin and Krishnan, 1995).
Empirical and model-predicted blood and plasma DLT concentrations were substantially higher than brain concentrations. Mean blood:brain ratios we measured in groups of five rats 1, 2, and 6 h after the 10 mg/kg po administration were 7.0, 6.2, and 3.4, respectively. Gray and Rickard (1982), Haines et al. (2004), and Rickard and Brodie (1985) also observed high DLT blood:brain ratios, that progressively diminished with time due to a relatively rapid drop in blood levels. Anadon et al. (1996) also reported a progressive decrease in the DLT blood:brain ratio over time, but DLT concentrations they measured in most regions of the brain usually far exceeded blood concentrations. Anadon and coworkers' animals exhibited severe neurotoxicity, including whole-body tremors and choreoathetosis. It is not clear how pharmacokinetic determinants (e.g., regional blood flow, metabolism, and membrane transport processes) may be altered under these circumstances.
In conclusion, this first-generation PBPK model for DLT represents the initial step toward being able to reliably predict the internal dosimetry of a pyrethroid insecticide for different dosage and exposure scenarios. There was reasonable concordance between the model simulations and most of the empirical data, despite frequent differences in rat strains, dosage vehicles, and analytical techniques. Crofton et al. (1995), for example, reported that both dosage route and vehicle had pronounced effects on the neurotoxic potency of DLT in rats. Models for common pyrethroids can be utilized to predict the likelihood of cumulative brain deposition and toxic effects.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAAPPENDIXREFERENCES |
|---|
Equations and abbreviations used in the present model as well as comparison of the kinetic behavior of DLT with some other lipophilic chemicals are available online at http://toxsci.oxfordjournals.org/.
|
APPENDIX |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAAPPENDIXREFERENCES |
|---|
Calculation of PVF of a tissue based on BVF and whole-blood PVF (PVFBL) in the tissue (BVF values were taken from Brown et al. (1997)
; BVF value for the slowly perfused was set equal to that of muscle):
 |
Calculation of erythrocytes: Plasma PC (PCE) based on PVF of the whole-blood (PVFBL) and blood:plasma PC (PCBL) (details shown in the online supplement):
 |
|
NOTES |
|---|
Disclaimer: The contents do not necessarily reflect the views and policies of the U.S. Environmental Protection Agency nor does mention of trade names or commercial products constitute endorsement or recommendation for use.
|
ACKNOWLEDGMENTS |
|---|
Special thanks go to Mr Srinivasa Muralidhara and Dr Jerry Campbell for their help in the laboratory. This project was supported by U.S. Environmental Protection Agency STAR grant R830800.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAAPPENDIXREFERENCES |
|---|
Abraham, E. H., Shrivastav, B., Salikhova, A. Y., Sterling, K. M., Johnston, N., Guidotti, G., Scala, S., Litman, T., Chan, K. C., Arceci, R. J., et al. (2001). Cellular and biophysical evidence for interactions between adenosine triphosphate and P-glycoprotein substrates: Functional implications for adenosine triphosphate/drug cotransport in P-glycoprotein overexpressing tumor cells and in P-glycoprotein low-level expressing erythrocytes. Blood Cells Mol. Dis. 27, 181–200.[CrossRef][ISI][Medline]
Anadon, A., Martinez-Larranaga, M. R., Fernandez-Cruz, M. L., Diaz, M. J., Fernandez, M. C., and Martinez, M. A. (1996). Toxicokinetics of deltamethrin and its 4'-HO-metabolite in the rat. Toxicol. Appl. Pharmacol. 141, 8–16.[ISI][Medline]
Anand, S. S., Bruckner, J. V., Haines, W. T., Muralidhara, S., Fisher, J. W., and Padilla, S. (2006). Characterization of deltamethrin metabolism by rat plasma and liver microsomes. Toxicol. Appl. Pharmacol. 212, 156–166.[CrossRef][ISI][Medline]
Andersen, M. E., Mills, J. J., Gargas, M. L., Kedderis, L., Birnbaum, L. S., Neubert, D., and Greenlee, W. F. (1993). Modeling receptor-mediated processes with dioxin: Implications for pharmacokinetics and risk assessment. Risk Analysis 13, 25–36.[CrossRef][ISI][Medline]
Agency for Toxic Substances and Disease Registry (ATSDR) (2003). Toxicological Profile for Pyrethrins and Pyrethroids. ATSDR, Atlanta, GA. Available at: http://www.atsdr.cdc.gov/toxprofiles/tp155.html. Accessed July 20, 2006.
Bosch, A. (1990). Metabolism of 14C-deltamethrin in rats. Unpublished report no. 187108 from Hazleton Laboratories America, Inc., Madison, WI, AgrEvo document A70824. Submitted to WHO by Hoechst Schering AgrEvo, Frankfurt-am-Main, Germany (GLP). Available at: http://www.inchem.org/documents/jmpr/jmpmono/v00pr04.htm. Accessed July 20, 2006.
Brown, R. P., Delp, M. D., Lindstedt, S. L., Rhomberg, L. R., and Beliles, R. P. (1997). Physiological parameter values for physiologically based pharmacokinetic models. Toxicol. Industr. Health 13, 407–484.[ISI][Medline]
Buss, D. S., McCaffery, A. R., and Callaghan, A. (2002). Evidence for p-glycoprotein modification of insecticide toxicity in mosquitoes of the Culex pipiens complex. Med. Vet. Entomol. 16, 218–222.[CrossRef][ISI][Medline]
California Environmental Protection Agency (CALEPA) (2001). Department of Pesticide Regulation Database. Available at: http://www.cdpr.ca.gov/dprdatabase.htm. Accessed July 20, 2006.
Casida, J. E., and Quistad, G. B. (1998). Golden age of insecticide research: Past, present, or future? Ann. Rev. Entomol. 43, 1–16.[CrossRef][ISI][Medline]
Choi, J. S., and Soderlund, D. M. (2006). Structure-activity relationships for the action of 11 pyrethroid insecticides on rat Nav 1.8 sodium channels expressed in Xenopus oocytes. Toxicol. Appl. Pharmacol. 211, 233–244.[CrossRef][ISI][Medline]
Crawford, M. J., Croucher, A., and Hutson, D. H. (1981). Metabolism of cis-cypermethrin and trans-cypermethrin in rats. Balance and tissue retention study. J. Agric. Food Chem. 29, 130–135.[CrossRef][ISI][Medline]
Crofton, K. M., Kehn, L. S., and Gilbert, M. E. (1995). Vehicle and route dependent effects of a pyrethroid insecticide, deltamethrin, on motor function in the rat. Neurotoxicol. Teratol. 17, 489–495.[CrossRef][ISI][Medline]
Delp, M. D., Manning, R. O., Bruckner, J. V., and Armstrong, R. B. (1991). Distribution of cardiac output during diurnal changes of activity in rats. Am. J. Physiol. 261, H1487–H1493.
Endres, C. J., Unadkat, J. D., Chung, F. S., and Peng, P. 2006. The role of transporters in drug interactions. Eur. J. Pharm. Sci. 27, 501–517.[CrossRef][ISI][Medline]
Evans, M. V., and Andersen, M. E. (2000). Sensitivity analysis of a physiological model for 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD): Assessing the impact of specific model parameters on sequestration in liver and fat in the rat. Toxicol. Sci. 54, 71–80.
Gray, A. J., and Rickard, J. (1982). The toxicokinetics of deltamethrin in rats after intravenous administration of a toxic dose. Pest. Biochem. Physiol. 18, 205–215.[CrossRef]
Haines, W., Marshall, R. S., Hunter, D. L., and Padilla, S. (2004). Pharmacokinetic differences may explain the age-related sensitivity of deltamethrin, a pyrethroid insecticide in rats. Toxicologist 78, 1109.
Hayes, W. J., and Laws, E. R., Eds. (1990). Handbook of Pesticide Toxicology, General Principles. Academic Press, Inc., New York. Available at: http://pmep.cce.cornell.edu/profiles/extoxnet/carbaryl-dicrotophos/deltamethrin-ext.html. Accessed July 20, 2006.
Kim, K.-B., Bartlett, M. G., Anand, S. S., Bruckner, J. V., and Kim, H. J. (2006). Rapid determination of the synthetic pyrethroid insecticide, deltamethrin, in rat plasma and tissues by HPLC. J. Chromatogr. B 834, 141–148.
Lawrence, L. J., and Casida, J. E. (1982). Pyrethroid toxicology: Mouse intracerebral structure toxicity relationships. Pest. Biochem. Physiol. 18, 9–14.
Leslie, E. M., Deeley, R. G., and Cole, S. P. C. (2005). Multidrug resistance proteins: Role of P-glycoprotein, MRP1, MRP2, and BCRP (ABCG2) in tissue defense. Toxicol. Appl. Pharmacol. 204, 216–237.[CrossRef][ISI][Medline]
Marei, A. E.-S. M., Ruzo, L. O., and Casida, J. E. (1982). Analysis and persistence of permethrin, cypermethrin, deltamethrin, and fenvalerate in the fat and brain of treated rats. J. Agric. Food Chem. 30, 558–562.[CrossRef][ISI][Medline]
Poulin, P., and Krishnan, K. (1995). A biologically-based algorithm for predicting human tissue:blood partition coefficients of organic chemicals. Human Exp. Toxicol. 14, 273–280.[ISI][Medline]
Rickard, J., and Brodie, M. E. (1985). Correlation of blood and brain levels of the neurotoxic pyrethroid deltamethrin with the onset of symptoms in rats. Pest. Biochem. Physiol. 23, 143–156.[CrossRef]
Rozman, T., Scheufler, E., and Rozman, K. (1985). Effect of partial jejunectomy and colectomy on the disposition of hexachlorobenzene in rats treated or not treated with hexadecane. Toxicol. Appl. Pharmacol. 78, 421–427.[CrossRef][ISI][Medline]
Ruzo, L. O., Engel, J. L., and Casida, J. E. (1979). Decamethrin metabolites from oxidative, hydrolytic, and conjugative reactions in mice. J. Agric. Food Chem. 27, 725–731.[CrossRef][ISI][Medline]
Ruzo, L. O., Unai, T., and Casida, J. E. (1978). Decamethrin metabolism in rats. J. Agric. Food Chem. 26, 918–925.[CrossRef][ISI][Medline]
Schoeffner, D. J., Warren, D. A., Muralidhara, S., Bruckner, J. V., and Simmons, J. E. (1999). Organ weights and fat volume in rats as a function of strain and age. J. Toxicol. Environ. Health A 56, 449–462.[CrossRef][ISI][Medline]
Sheets, L. P., Doherty, J. D., Reiter, L. W., and Crofton, K. M. (1994). Age-dependent differences in the susceptibility of rats to deltamethrin. Toxicol. Appl. Pharmacol. 126, 186–190.[CrossRef][ISI][Medline]
Soderlund, D. M., and Casida, J. E. (1977). Effects of pyrethroid structure on rates of hydrolysis and oxidation by mouse-liver microsomal enzymes. Pest. Biochem. Physiol. 7, 391–401.[CrossRef]
Soderlund, D. M., Clark, J. M., Sheets, L. P., Mullin, L. S., Piccirillo, V. J., Sargent, D., Stevens, J. T., and Weiner, M. L. (2002). Mechanisms of pyrethroid neurotoxicity: Implications for cumulative risk assessment. Toxicology 171, 3–59.[CrossRef][ISI][Medline]
Srinivas, R., Udikeri, S. S., Jayalakshmi, S. K., and Sreeramulu, K. (2004). Identification of factors responsible for insecticide resistance in Helicoverpa armigera. Compar. Biochem. Physiol. Toxicol. Pharmacol. 137, 261–269.[CrossRef]
Van Dijk, A., and Burri, R. (1993). [14C-Benzyl]-deltamethrin: Distribution, kinetics and excretion after single intravenous administration to female rats. Unpublished report no. 299160 from RCC Itingen, Switzerland, 5 August 1993, AgroEvo document A51513. Submitted to WHO by Hoechst Schering AgrEvo, Frankfurt-am-Main, Germany (GLP). Available at: http://www.inchem.org/documents/jmpr/jmpmono/v00pr04.htm. Accessed July 20, 2006.
Wolansky, M. J., Gennings, C., and Crofton, K. M. (2006). Relative potencies for acute effects of pyrethroids on motor function in rats. Toxicol. Sci. 89, 271–277.
Yu, K. O., Narayanan, L., Mattie, D. R., Godfrey, R. J., Todd, P. N., Sterner, T. R., Mahle, D. A., Lumpkin, M. H., and Fisher, J. W. (2002). The pharmacokinetics of perchlorate and its effect on the hypothalamus/pituitary-thyroid axis in the male rat. Toxicol. Appl. Pharmacol. 182, 146–159.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K.-B. Kim, S. S. Anand, H. J. Kim, C. A. White, and J. V. Bruckner Toxicokinetics and Tissue Distribution of Deltamethrin in Adult Sprague Dawley Rats Toxicol. Sci., February 1, 2008; 101(2): 197 - 205. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Godin, J. A. Crow, E. J. Scollon, M. F. Hughes, M. J. DeVito, and M. K. Ross Identification of Rat and Human Cytochrome P450 Isoforms and a Rat Serum Esterase That Metabolize the Pyrethroid Insecticides Deltamethrin and Esfenvalerate Drug Metab. Dispos., September 1, 2007; 35(9): 1664 - 1671. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||