Assessment of the Percutaneous Absorption of Trichloroethylene in Rats and Humans Using MS/MS Real-Time Breath Analysis and Physiologically Based Pharmacokinetic Modeling

Torka S. Poet*,1, Richard A. Corley*, Karla D. Thrall*, Jeffrey A. Edwards*, Hanafi Tanojo{dagger}, Karl K. Weitz*, Xiaoying Hui{dagger}, Howard I. Maibach{dagger} and Ronald C. Wester{dagger}

* Chemical Dosimetry, Battelle, Pacific Northwest Division, Post Office Box 999, MSIN P7-59, Richland, Washington 99352; and {dagger} Department of Dermatology, University of California, San Francisco, California 94143.

Received January 2, 2000; accepted April 4, 2000


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The development and validation of noninvasive techniques for estimating the dermal bioavailability of solvents in contaminated soil and water can facilitate the overall understanding of human health risk. To assess the dermal bioavailability of trichloroethylene (TCE), exhaled breath was monitored in real time using an ion trap mass spectrometer (MS/MS) to track the uptake and elimination of TCE from dermal exposures in rats and humans. A physiologically based pharmacokinetic (PBPK) model was used to estimate total bioavailability. Male F344 rats were exposed to TCE in water or soil under occluded or nonoccluded conditions by applying a patch to a clipper-shaved area of the back. Rats were placed in off-gassing chambers and chamber air TCE concentration was quantified for 3–5 h postdosing using the MS/MS. Human volunteers were exposed either by whole-hand immersion or by attaching patches containing TCE in soil or water on each forearm. Volunteers were provided breathing air via a face mask to eliminate inhalation exposure, and exhaled breath was analyzed using the MS/MS. The total TCE absorbed and the dermal permeability coefficient (KP) were estimated for each individual by optimization of the PBPK model to the exhaled breath data and the changing media and/or dermal patch concentrations. Rat skin was significantly more permeable than human skin. Estimates for KP in a water matrix were 0.31 ± 0.01 cm/h and 0.015 ± 0.003 cm/h in rats and humans, respectively. KP estimates were more than three times higher from water than soil matrices in both species. KP values calculated using the standard Fick's Law equation were strongly affected by exposure length and volatilization of TCE. In comparison, KP values estimated using noninvasive real-time breath analysis coupled with the PBPK model were consistent, regardless of volatilization, exposure concentration, or duration.

Key Words: percutaneous absorption; PBPK modeling; real-time breath analysis; permeability coefficient.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Trichloroethylene (TCE) is a volatile organic compound that is used as a solvent and a degreaser. TCE is commonly found at many commercial and industrial settings and in the environment as a result of industrial effluents and disposal processes. High TCE exposures have resulted in typical solvent-induced CNS effects, hepatocarcinomas in mice, and nephrotoxicity in rats and mice (Bull et al., 1990Go; Dekant et al., 1990Go). Because TCE persists in water and soil, where it breaks down slowly (ATSDR, 1997Go), an assessment of the dermal absorption of TCE from both water and soil matrices is needed in order to fully assess health risks from typical environmental exposures.

In general, previous investigations of dermal absorption have employed conditions that do not mimic actual exposure situations and may significantly overestimate potential human body burdens. For example, dermal penetration methods using an impermeable occlusion technique can drastically alter skin hydration and temperature and create an artificially high exposure concentration due to the suppression of normal loss by volatilization (U.S. EPA, 1992Go). As a result, the use of occlusive techniques can affect bioavailability and will generally provide a worst-case dermal penetration scenario (highest uptake), which may be completely unrelated to realistic environmental exposure situations.

Erroneous estimates of percutaneous absorption may also be determined when standard Fick's law calculations of dermal flux are used without verifying that steady state was achieved. In comparison, a physiologically based pharmacokinetic (PBPK) model can be used to assess transdermal flux as a function of the permeability coefficient (KP), the area exposed (SA), and the concentration gradient across the skin ({Delta}C) during non–steady-state conditions. In addition, PBPK models can include mathematical descriptions of the test exposure system that account for loss of chemical through volatilization and the resulting changes in matrix concentration. Due to their fundamental biological basis (i.e., tissue characteristics, blood flow, metabolism rates, etc.), PBPK models have become increasingly popular for estimating kinetic parameters following dermal exposure as a function of the exposure matrix and duration (Corley et al., 1997Go; Corley et al., 2000Go; Jepson and McDougal, 1997Go; Jepson and McDougal, 1999Go; McDougal et al., 1986Go).

Percutaneous absorption of TCE has been investigated in rodents and in human skin in vitro (Bogen et al., 1992Go; Morgan et al., 1991Go; Nakai et al., 1999Go). Bogen et al. (1992) reported a Kp of 0.21 cm/h for trichloroethylene in hairless guinea pigs following exposures of most of the body surface in a water matrix. In partial thickness human skin in vitro, the KP was reported to be 0.12 cm/h (Nakai et al, 1999Go). Although some comparisons of the contributions of different routes of exposures have been made for humans while showering (Weisel and Jo, 1996Go), a human in vivo KP for TCE has not previously been reported.

The objective of this research was, therefore, to investigate the percutaneous absorption of TCE in rats and humans under realistic exposure conditions using a noninvasive technique (Poet et al. 2000Go). To evaluate the significance of dermal exposures to TCE in water and soil, the TCE PBPK model of Fisher et al. (1998) was modified by adding a skin compartment based on the design of Jepson and McDougal (1997) and used to define and compare skin permeability (KP) under various exposure scenarios. Occluded and nonoccluded dermal exposures were conducted in rats and humans using soil and water containing low levels of TCE. Exhaled breath was analyzed using an atmospheric sampling glow discharge ionization (ASGDI) source mass spectrometer (MS/MS) to follow the kinetics of TCE continuously throughout the absorption, distribution, and elimination phases according to Poet et al. (2000). Rats, which are commonly used in in vivo dermal absorption studies, were included for comparison with humans and for assessing multiple exposure concentrations.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Animals and Chemicals
Adult male F344 rats (200–240 g) were obtained from Charles River Inc. (Raleigh, NC). Prior to use, animals were housed in solid-bottom cages with hardwood chips, and were acclimated in a humidity- and temperature-controlled room with a 12-h light/dark cycle. Rodent feed (Purina rodent chow) and water were provided ad libitum.

HPLC-grade (99.5% pure) trichloroethylene (CAS #79-01-6) was obtained from Sigma Chemical Co. (St. Louis, MO.). All other chemicals were reagent grade or better and were obtained from Sigma Chemical Co. The soil sample was collected in Yolo County, California, and was prepared by passing it through a 40-mesh sieve and retaining it on an 80-mesh sieve. The soil was analyzed by an independent lab (DANR Analytical Laboratory, University of California) and consisted of 30% sand, 18% clay, 52% silt, an organic content of 1.3%, and a pH of 6.8. No volatile contaminants were detected in the soil using the MS/MS system described below.

Dermal Exposures in Rats
Application techniques.
Dermal exposures to TCE were conducted in both soil and water matrices using methods reported previously (Poet et al., 2000Go). Male F344 rats were anesthetized and the hair on the lower back clipper shaved the day prior to exposure. Groups of three rats were used at each concentration in occluded or nonoccluded water and soil exposures. An outline of the experimental design is given in Table 1Go.


View this table:
[in this window]
[in a new window]
 
TABLE 1 Experimental Design
 
For exposures in water, a 2.5-cm diameter hand-blown glass cell (O.Z. Glass Co., Pinole, CA), with a needle hole opening in the top to allow addition of the dosing solution, was attached to the clipped area using a cyanoacrylate adhesive. For exposures, 5 ml of the dosing solution at target TCE concentrations of 500 mg/l (0.05%) or 1500 mg/l (0.15%) in double-distilled water, which were approximately at or below the aqueous solubility limit for TCE, were injected into the glass cell. Immediately after filling the cell, the needle hole was sealed with silicone glue and tape.

For soil exposures, 1 g of soil at TCE target concentrations of 5000, 20,000, and 40,000 mg/kg was placed in the nonoccluded exposure system for 3 h. Preliminary studies with a nonoccluded system, patterned after the occluded glass cell but covered with a vapor permeable ceramic frit, were not successful at preventing leakage of TCE into the off-gassing chamber. Therefore, the nonoccluded exposure system used in previous work (Poet et al., 2000Go) was used for TCE soil exposures. Briefly, a ring of self-adhesive DuoDerm® (Bristol-Myers Squibb, Princeton, NJ) was attached to the clipped area on the back of each rat. The topical dose of TCE-spiked soil was applied to the 8-cm2 area bounded by the DuoDerm®, and covered with a transparent dressing (Bioclusive®, Johnson & Johnson, Arlington, TX) that allowed free passage of vapor. The patch was covered with an aerated weighboat and a muslin patch containing activated charcoal. The entire patch system was secured with self-adherent wrap. An 8-cm2 patch area representing approximately 3% of the total skin surface area was chosen to mimic limited exposure situations. For occluded soil samples, target concentrations of 5000 and 15,000 mg/kg TCE in 5 g of soil were placed in the occluded glass cell system as described for water exposures.

Dermal exposure conditions.
To quantify total absorbed dose, the exact weight of administered dose was recorded and samples of exposure media (soil or water), the amount remaining at the end of the exposure, and the amount volatilized to the charcoal were analyzed using gas chromatography as described previously (Poet et al., 2000Go).

Immediately following dermal application, rats were individually placed in 2-liter off-gassing chambers as described by Gargas (1990), and TCE was quantified in the chamber, representing exhaled breath. A Teledyne Discovery II MS/MS equipped with an ASGDI source sampled from the off-gassing chamber approximately every 5 s as described previously (Poet et al., 2000Go). TCE was quantified by selective ion monitoring of the m/z ratios 127–131. For occluded soil and water exposures, exhaled breath was monitored for 5 h. For nonoccluded soil exposures, exhaled breath levels declined more rapidly and the studies were terminated after 3 h.

Dermal Exposures in Human Subjects
Study participants.
Nine healthy volunteers participated in the study; medical histories and demographic data including, gender, race, age, body weight, and height are given in Table 2Go. The studies were conducted under approval from both the University of California at San Francisco human subjects Institutional Review Board (IRB), and the Pacific Northwest National Laboratory IRB in compliance with multiple project assurance number DOE.MPA.PNNL96-2000. Written consent was obtained from each subject prior to participation. Subjects reported no chronic conditions, significant cardiovascular, hepatic, central nervous system, renal, hematological, or gastrointestinal diseases, and no dermatological problems. As TCE is highly lipophilic and the pharmacokinetics of TCE in exhaled breath are sensitive to the amount of body fat, the percent body fat for each subject was determined using a hand-held near-infrared body fat analyzer (Futrex®, Gaithersburg, MD).


View this table:
[in this window]
[in a new window]
 
TABLE 2 Human Subjects
 
Dermal exposure conditions.
Human volunteers were dermally exposed to TCE in soil or water using either a patch system or by immersing a hand in a container with 4 liters water or 4 kg soil. Human subject information and surface areas of exposure are given in Table 2Go. The occluded water and soil exposure cells were identical to the rat systems described above, with the exception that the human exposure cells were larger (4.0 cm x 2.2 cm) and contained 20 ml of water or 20 g of soil in each. Two exposure cells were placed on each arm of the volunteer (four cells/volunteer) for a total of 80 ml of water or 80 g of soil and a total exposed surface area of 50.2 cm2. TCE target exposure concentrations were 1000 mg/l in water and 5000 mg/kg in soil. Samples were collected before and after the exposure for GC analysis. For whole-hand exposures, TCE was prepared in 4 l water or 4 kg soil in a separate room and covered with plastic wrap prior to use. Samples (water or soil) were collected from the container prior to and during the exposure at 0.5, 1, 1.5, and 2 h for GC analysis of TCE concentrations in the exposure media over time.

Subjects were provided clean breathing air via a face mask with a two-way nonrebreathing valve to prevent inhalation exposure. The exhaled breath was passed through a heated mixing chamber (1.3-l volume) from which the MS/MS withdrew a sample for analysis approximately every 5 s as described previously (Poet et al., 2000Go). Excess exhaled air was vented from the mixing chamber to a hood with negligible flow restriction via a large bore hole exit tube. Background exhaled breath measurements were taken for 4 min before each exposure.

PBPK Model
A previously established PBPK model for TCE (Fisher et al., 1998Go) was modified to include a separate skin compartment to describe the uptake of TCE following dermal exposure as described by Jepson and McDougal (1997). For rat exposures, an equation was also added to describe the amount of chemical in the off-gassing chamber in terms of input from the exhaled breath and removal from the chamber either by rebreathing or to the MS/MS:

(1)

where ACH is the amount of TCE in the chamber (mg), QALV is the alveolar ventilation rate (l/h), CEXH is the concentration of TCE in the exhaled breath (mg/L), CCH is the concentration of TCE in the chamber (mg/L), and FCH is the air flow through the chamber (l/h).

The dermal PBPK model incorporated specific descriptions for richly and poorly perfused tissues, fat, liver, lung, and skin (Fig.1Go). TCE-specific parameters, including partition coefficients and metabolism rates, were taken from the literature (Fisher et al., 1998Go; Fisher et al., 1991Go; Mattie et al., 1994Go). Media (soil and water) partition coefficients were determined using the method of Gargas et al. (1989). Briefly, quadruplicate samples of 1 g soil or 2 ml water were placed in sealed vials and incubated for 2 h with TCE. The amount of TCE in the headspace of vials containing soil or water was compared to empty reference vials containing only TCE. The skin:soil and skin:water partition coefficients were calculated by dividing the skin:air partition coefficient by the soil:air or water:air partition coefficient, respectively. Parameters for the rat and human PBPK models are given in Tables 3 and 4GoGo.



View larger version (24K):
[in this window]
[in a new window]
 
FIG. 1. Schematic of the PBPK model used to describe the disposition of TCE in either rats or humans during dermal exposures. The PBPK model of Fisher et al. (1998) was modified to include dermal absorption as described in Materials and Methods section.

 

View this table:
[in this window]
[in a new window]
 
TABLE 3 Parameters Used in the PBPK Model for Rats and Humans
 

View this table:
[in this window]
[in a new window]
 
TABLE 4 Tissue Partition Coefficients for Rats and Humans
 
The uptake of TCE through the skin was described by incorporating Fick's Law of diffusion into the PBPK model. Equations to describe the rate of change of chemical concentration across the skin (Jepson and McDougal, 1997Go) and evaporation to patch components were

(2)

(3)
where Csk is the TCE concentration in the skin (mg/l), SA is the surface area exposed (cm2), Csurf is the concentration of TCE at the skin surface (mg/l), Pskm is the skin:matrix (soil or water) partition coefficient (unitless), Cart is the arterial blood TCE concentration (mg/l), Qsk is the blood flow to the skin (l/h), Cven is the venous blood TCE concentration (mg/l), Ccoal is the concentration in the charcoal from evaporative loss (mg/l), Vs represents the volume of the dosing solution (kg), and Kloss is the rate of evaporation loss from the media to the charcoal (h–1).

For human subjects, the observed delay before the appearance of TCE in exhaled breath was estimated in an iterative manner from the exhaled breath data upon optimization of Kloss and Kp. The delay was estimated for each individual in each exposure by manual adjustment of the time before exposure onset until the best visual fit to experimental exhaled breath concentrations was achieved. The permeability coefficient and Kloss were estimated for each individual animal or human subject by a least-squares fit of the model to the data using the SimusolvTM (Dow Chemical Co.) optimization subroutine. The percent variability explained for the optimized values using a simple well-stirred skin compartment in the PBPK model was always >= 80%.

Data Analysis
Values are presented as the mean ± SD of n = 3–4 exposures, with the exception of human soil hand immersion, which was preformed using two volunteers. A Student's t-test or an ANOVA followed by the Student-Newman-Keuls test was used for statistical comparisons of parameters for different exposure conditions, where appropriate, at p < 0.01.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Dermal Exposures of Rats to TCE
Rats were exposed to TCE in water using a fully occluded patch system at actual concentrations of either 600 ± 40 or 1600 ± 300 mg/l, as measured by GC headspace analysis. Peak chamber concentrations, representing exhaled breath levels, of approximately 300 and 610 ppb, respectively, were reached within 2 h of dermal exposure and declined slowly thereafter (Fig. 2Go).



View larger version (25K):
[in this window]
[in a new window]
 
FIG. 2. PBPK model prediction (line) and exhaled breath data (points averaged every 1-min interval) for representative rats exposed to 600 (diamonds) and 1600 mg/l (circles) TCE in 5 ml of water (occluded) over a 5 cm2 area of the back.

 
For occluded soil exposures, actual dose concentrations were 5300 ± 850 and 15,600 ± 2200 mg/kg, as measured by GC headspace analysis. Peak exhaled breath levels for these exposures were approximately 1500 and 6000 ppb, respectively (Fig. 3Go). At these occluded exposure concentrations, peak exhaled breath levels were reached within 1–2 h and remained relatively steady over the 5-h exposure, reflecting the reservoir of TCE available for continual absorption.



View larger version (19K):
[in this window]
[in a new window]
 
FIG. 3. Concentration-time plots for TCE exposures in occluded soil. Points are exhaled breath data (averaged every 1-min interval); lines are PBPK simulations of representative rats exposed to 5300 (diamonds), and 15,600 mg/kg (circles) in 5 g of soil over a 5-cm2 area of the back.

 
Actual nonoccluded soil exposure concentrations were to 5000 ± 1900, 20,300 ± 7700, or 40,600 ± 3400 mg/kg, as measured by GC headspace analysis. Peak chamber concentrations from exhaled breath after nonoccluded dermal exposures were approximately 200, 2200, and 4400 ppb for the low, middle, and high exposure concentrations, respectively (Fig. 4Go). Peak exhaled breath concentrations were reached within about 1 h of dermal application, and declined more rapidly than observed with the occluded exposures, reflecting volatilization of the TCE to the charcoal patch.



View larger version (23K):
[in this window]
[in a new window]
 
FIG. 4. Concentration-time plots for TCE exposures in nonoccluded soil. Points are exhaled breath data (averaged every 1-min interval); lines are PBPK simulations of representative rats exposed to 5000 (squares), 20,300 (diamonds), and 40,600 mg/kg (circles) in 1 g of soil over an 8-cm2 area of the back.

 
TCE concentrations remaining in the exposure system at the end of the exposure varied depending on the exposure matrix, volume of media, and whether the system was occluded (Fig. 5Go). For example, at 3 h of exposure, 1.6 ± 1.2% of the original dosing concentration of TCE remained in the nonoccluded soil exposure cell, and 60.7 ± 8.1% of the original exposure dose was recovered in the charcoal patch. In contrast, at 5 h of exposure, more than 60% of the original concentration of TCE remained in the occluded soil exposure cell, and 20% remained in the occluded water exposure cell. By percentage, twice as much TCE was absorbed from occluded water than soil; as there was no loss of the TCE from the exposure cell, this difference can be attributed entirely to the effect of the vehicle on dermal bioavailability.



View larger version (12K):
[in this window]
[in a new window]
 
FIG. 5. Percentage of TCE in the exposure media (occluded soil, nonoccluded soil, or occluded water) and charcoal (nonoccluded) from rat studies. Lines are PBPK estimations; points are concentrations measured by headspace analysis for amount in occluded soil (diamonds; n = 6), occluded water (squares; n = 6), nonoccluded soil (circles; n = 9), and charcoal (triangles; n = 9). Values are percentage calculated for all exposure concentrations ± SD

 
The modified PBPK model of Fisher et al. (1998) was used to estimate the permeability coefficient (KP) for dermal absorption of TCE under the different exposure scenarios for each rat (Table 5Go). A KP of 0.31 ± 0.01 cm/h provided a good fit to all of the data sets for occluded water exposures. In comparison, the estimated KP for both occluded and nonoccluded soil exposures was approximately 0.09 cm/h, over 3-fold lower than the KP estimated for water exposures (Table 5Go).


View this table:
[in this window]
[in a new window]
 
TABLE 5 PBPK Model Estimates for the Dermal Absorption of TCE in Rats
 
For exposures to TCE in nonoccluded soil, the rate of loss of TCE from the soil to the charcoal (Kloss) was an integral factor in modeling the dermal uptake of TCE. Kloss was optimized for each data set in an iterative manner by the least-squares fit of the PBPK model to the concentrations of TCE remaining in the media at the end of the exposure and to the fit of the exhaled breath data using the optimization subroutine of SimusolvTM (Dow Chemical Co). The TCE lost to the patch system was relatively constant for all samples and a Kloss of 0.76 ± 0.06 hr–1 provided a good fit to all of the data (Fig. 5Go).

A cyclical pattern in the exhaled breath profile was noted for most rat exposures (see especially Fig. 3Go). This wave in the exhaled breath levels was likely associated with patterns of activity and was not compensated for by the PBPK model. In the model, the pulmonary rate was held constant over the exposure period, whereas breathing rate would be expected to change with activity level and cause the variations in exhaled breath levels that were observed.

Dermal Exposures of Humans to TCE
Three human volunteers were exposed to TCE for 2 h by placing a hand in 4 liters of water containing actual exposure concentrations ranging from 810 to 1300 mg/l, as measured by GC analysis. A 0.1- to 0.55-h delay in the appearance of TCE in exhaled breath occurred in all human volunteers. The delay in the dermal absorption of chemicals in humans is commonly reported and likely due to stratum corneum loading of the chemical (Roberts et al., 1999Go). No delay was observed in any of the rat exposures. The delay was estimated for each individual exposure by manual adjustment of the time for the onset of exposure in the PBPK model. Peak exhaled breath TCE levels associated with the whole-hand water exposures ranged from 750 to 1200 ppb for the three volunteers over these exposure concentrations. A representative exhaled breath profile is given in Fig. 6Go.



View larger version (14K):
[in this window]
[in a new window]
 
FIG. 6. PBPK model prediction (line) and exhaled breath data (points averaged every 1 min) for a representative human subject (subject 8) exposed to TCE for 2 h by immersing a hand in 4 l of water. See Tables 2 and 6GoGo for subject-specific exposure and model parameters.

 
Two volunteers immersed their hands in 4 kg of soil with TCE concentrations of 4000 and 4200 mg/kg, as measured by GC analysis. Peak exhaled breath concentrations of approximately 1200–1600 ppb were achieved within 0.5 h following the characteristic delay for the two subjects, and remained constant over the 2-h exposure period (Fig. 7Go). Although the peak exhaled breath concentrations were similar following exposures to TCE in soil and water, the exposure concentration in soil exposures were five times higher than for water (Table 6Go).



View larger version (12K):
[in this window]
[in a new window]
 
FIG. 7. PBPK model prediction (line) and exhaled breath data (points averaged every 1 min) for a representative human subject (subject 7) exposed to TCE for 2 h by immersing a hand in 4 kg of soil. See Tables 2 and 6GoGo for subject-specific exposure and model parameters.

 

View this table:
[in this window]
[in a new window]
 
TABLE 6 PBPK Model Estimates for Human Dermal Absorption
 
Three volunteers were exposed to TCE in water using the patch system at actual exposure concentrations ranging from 850 to 1000 mg/l TCE in a total of 80 ml of water (20 ml/patch) over a surface area of 50.2 cm2. As with the hand immersion exposures, there was a delay in the appearance of TCE in the exhaled breath of these subjects; peak exhaled breath concentration of 80–200 ppb were reached within 0.5 h after the delay (Fig. 8Go).



View larger version (16K):
[in this window]
[in a new window]
 
FIG. 8. PBPK model prediction (line) and exhaled breath data (points averaged every 1 min) for a representative human subject (subject 5) exposed to TCE for 6 h by placing four glass cells on the forearms containing 80 ml of water over 50.2 cm2. See Tables 2 and 6GoGo for subject-specific exposure and model parameters.

 
The exhaled breath profile following exposure of 4 volunteers to occluded soil patches containing 3200–21000 mg/kg in 80 g of soil was similar to that seen following exposures in water. For subjects exposed to less than 9000 mg/kg, peak exhaled breath concentrations ranging from 70–150 ppb were reached within 1 h of exposure (Fig. 9Go).



View larger version (16K):
[in this window]
[in a new window]
 
FIG. 9. PBPK model prediction (line) and exhaled breath (data points averaged every 1 min) for a representative human subject (subject 4) exposed to TCE for 6 h by placing four glass cells on the forearms containing 80 g of soil over 50.2 cm2. See Tables 2 and 6GoGo for subject-specific exposure and model parameters.

 
The PBPK model estimate of the amount of TCE remaining in the soil or water exposure media from the hand immersion studies over time compared to the amount in the samples collected every 30 min is represented in Fig. 10Go. Because the concentration of TCE in the containers of water or soil declined over the 2-h exposure period, as measured by GC analysis, the loss rates (Kloss) of TCE from the exposure systems were optimized in an iterative manner, as described in the rat studies. Kloss was not significantly different between exposure conditions at 0.24 ± 0.01 h–1 for hand immersion exposures in both water and soil (Table 6Go). In spite of this loss, over 50% of the original concentration remained in the media from the hand immersion studies at the end of the 2-h exposure.



View larger version (10K):
[in this window]
[in a new window]
 
FIG. 10. Percentage of TCE in the exposure media (occluded soil, nonoccluded soil, or occluded water) from human studies. The lines are PBPK estimations; points are concentrations measured by headspace analysis for amount in the 4 kg of soil (dotted line, diamonds; n = 2), 4 liters of water (dotted line, squares; n = 3, ± SD), patches containing soil (dashed line, circles; n = 3, ± SD), and patches containing water (solid line, triangles; n = 3, ± SD). The PBPK model predictions for 4 liters of water and 4 kg of soil overlap.

 
The GC analysis of water and soil media sampled from the patch exposure system at the end of the exposures indicated that a nonspecific loss occurred over the 6-h exposure period, even though an occluded patch system that was effective in previous studies with methyl chloroform was used (Poet et al., 2000Go). Because each human volunteer wore a face mask with a two-way nonrebreathing valve, no inhalation exposure occurred. While 4 and 0.6% of the TCE was absorbed through the skin for water and soil, respectively, less than 30% of the original TCE concentration remained in either media at the end of the 6-h exposure. A Kloss that was not significantly different from hand immersion studies fit both the exhaled breath data from each subject and the GC data for patch media concentrations (Fig. 10Go). To prevent this nonspecific loss with rat exposures, an extra layer of tape was added to seal the silicone that filled the needle hole. No nonspecific loss was observed in the rat patch studies (see above and Fig. 5Go).

The exhaled breath data from each individual volunteer was analyzed using the PBPK model with physiological parameters set for humans (Table 3Go). The body weight, total body surface area, body surface area exposed, and percent body fat were individually set for each subject, as given in Table 2Go. For human exposures, a KP of 0.017 ± 0.003 cm/h provided a good fit to all of the data sets for water exposures. The estimated KP for both occluded and nonoccluded soil exposures averaged over 3-fold lower (0.005 ± 0.002 cm/h) than the KP values estimated for water exposures, a difference equivalent to that observed in the rat exposures (Table 6Go).


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The dermal pathway constitutes a potentially significant route of exposure for many chemicals. For a volatile chemical, such as TCE, dermal uptake will depend upon several factors, including the exposure concentration, exposure duration, exposure site, vehicle, chemical volatility, and the permeability coefficient for the passage of the chemical across the skin. A combination of PBPK modeling and real-time breath analysis was used to investigate the importance of exposure media (water or soil), volatility (occluded or nonoccluded), regional variations in absorption (hand or forearm), and species (rat or human) on the dermal bioavailability of TCE. The differences in bioavailability were reflected in the estimated KP values for each of these exposure scenarios.

Exposures via the aqueous matrix lead to higher dermal bioavailabilities in both rats and humans. In the studies described here, the KP for TCE in water was 3.5-fold higher than for soil in both rats and humans. A comparison of soil and water patch exposures showed that in humans, six times more TCE was absorbed from water than soil. A similar finding was observed in rats, where twice as much applied TCE was absorbed from water exposures compared to soil.

A patch system that permitted the volatilization of TCE was used for some of the rat exposures, and a loss term was included in the PBPK model to facilitate the assessment of dermal uptake during changing exposure concentrations as a result of evaporation. The volatility of TCE played an important role in the amount available to be absorbed through skin. Preliminary studies in humans confirmed that exhaled breath TCE concentrations were below the MS/MS level of detection with nonoccluded patch exposure conditions due to volatilization of the TCE.

In rats, there was no difference in KP between occluded and nonoccluded soil exposures once volatilization (Kloss) was accounted for in the PBPK model. Although the percentage absorbed in occluded and nonoccluded exposures was the same, TCE was still present in the soil in the occluded exposure system at the end of exposure, but not in the nonoccluded system. According to these results, nonocclusion (allowing the chemical to volatilize) will have little effect on total absorption over short exposure periods, depending on loss of chemical from the exposure site, but could have significant impact over longer exposures. Although the percent absorbed was the same between occluded and nonoccluded soil exposures following these 3–5 h exposures, longer exposures to nonoccluded soil would not be likely to result in a higher percentage absorbed.

Studies have shown that the dermal absorption of numerous compounds is greater in rats than in humans (Bronaugh, 1998Go; Jepson and McDougal, 1997Go; McDougal et al., 1990Go; U.S. EPA, 1992Go; Wester and Maibach, 1986Go). However, the vast majority of pharmacokinetic studies are conducted using rats, and the impact of multiple factors can be more easily and economically investigated using nonhuman models. In this study, both rat and human dermal exposures were conducted to allow direct comparison of TCE dermal absorption between rats and human subjects and to validate the noninvasive technique for use in human studies. Using the methods described here, the KP for the percutaneous absorption of TCE from a water matrix was 0.31 cm/h in rats and 0.015 cm/h in humans following hand immersion studies. This value for KP in rats is simular to those reported in other rodents. The value estimated for humans is lower than rats and in vitro human skin (Nakai et al., 1999Go). Although the human KP was expected to be less than the rat KP, the magnitude of the difference was likely due to site-specific anatomical differences in percutaneous absorption, which are well known (U.S. EPA, 1992Go). The stratum corneum is thicker on the hand than on the back (Scheuplein and Blank, 1971Go), and Maibach et al. (1971) found that the percutaneous absorption of pesticides varied with site of application in humans.

A number of predictive models have been proposed for estimating the rate of chemical transport across the skin. The U.S. Environmental Protection Agency (U.S. EPA) generic dermal bioavailability equation, based on Fick's Law (U.S. EPA, 1992Go) is

(4)

where SA is the surface area (cm2), C is the starting concentration (mg), and T is the time (h). Equation 4 is used when total absorption can be determined and steady state is either achieved or assumed not to be a factor (Bogen et al., 1992Go; Kezic et al., 1997Go; Nakai, et al., 1999Go; U.S. EPA, 1992Go). In the work described here, a PBPK model was used to represent dermal absorption characteristics, describe experimental data, and determine dermal absorption parameters. A PBPK model offers several advantages, including the ability to estimate permeability constants, describe dosimetry, and integrate factors such as chemical volatility, and the ability to incorporate non–steady-state exposure conditions. The rapid collection of kinetic data such as those obtained using the noninvasive real-time breath analysis system provided the opportunity to assess the uptake, distribution, and elimination phases of dermal absorption in both rats and humans.In exposure conditions where the dose is finite (or a volatile chemical is allowed to evaporate from the exposure matrix), the simple Equation 4 will rarely provide an accurate prediction of Kp. The generic Fick's Law Equation does not take into account the change in the exposure matrix due to volatilization or uptake. At later time points, the generic equation will underpredict KP values as exposure concentration declines due to volatilization or uptake. In contrast, shorter exposures that do not reach steady state would result in estimated KP values that are higher than the actual value. A comparison of KP values estimated using the PBPK model and calculated based on Equation 4 for an occluded rat exposure is shown in Fig. 11Go. In the occluded system, there is no evaporative loss of compound, and rats show no delay in the appearance of compound in the exhaled breath. Under these conditions, KP is overpredicted using Equation 4 until peak exhaled breath concentrations are reached (around 1 h), then underpredicted as exhaled breath concentrations decline as the exposure concentration decreases solely due to uptake. Equation 4 results in permeability values that are estimated from absorption averages, whereas the PBPK model-estimated KP is instantaneous. Thus, the PBPK model provides a consistent KP value by considering the loss of TCE from the exposure system by integrating concentration over time (dC/dt).For human water hand immersion exposures, there is essentially an inexhaustible amount of TCE to be absorbed, and thus Equation 4 does not result in a substantial underprediction of KP at later time points (Fig. 12Go). Equation 4, however, fails to adjust time for the apparent delay observed in absorption of compounds through human skin. Thus, the average KP determined using Fick's law equation underpredicts for short exposures and slowly rises, whereas the PBPK model estimates a consistent KP value. Using Equation 4, the KP for TCE in humans would be predicted to range from 0.0062 to 0.028 cm/h. This range results in a 4.5-fold difference in the amount of TCE expected to be absorbed.In conclusion, breath analysis and PBPK modeling were used to ascertain percutaneous permeability coefficients for TCE absorption in rats and humans. The breath analysis provided real-time insight into the lag time in absorption of TCE in human subjects and a simple noninvasive way of quantifying absorption. Furthermore, PBPK modeling proved to be a reliable method of determining KP, relative to the generic Fick's law equation, under various exposure conditions. The KP values reported for TCE absorption in hairless guinea pigs (Bogen et al., 1992Go) and human skin in vitro (Nakai, et al., 1999Go) are in the same range as the rat value determined here, whereas the human in vivo KP was much lower. Differences in the estimation of Kp can have a significant impact on predictions of dermal bioavalability. A PBPK model can be used to correct for non–steady-state exposure conditions and other factors, including volatility and individual physiological variations, and thus provide an accurate estimation of Kp.



View larger version (8K):
[in this window]
[in a new window]
 
FIG. 11. A comparison of predicted KP values using a representative rat exposed to 1500 mg/l TCE in an occluded water patch using the default EPA Equation 4 and the PBPK model. In rats, no delay in TCE absorption was observed and no volatilization (Kloss = 0) occurred in the occluded water exposures. Therefore, the only factors affecting the estimate of KP are steady state and the changing matrix concentration due to absorption. The PBPK model predicted a consistent KP regardless of the length of exposure (diamonds). By using the PBPK model to calculate the total absorbed over exposure lengths of 0.1–4 h and then inserting total absorbed into Equation 4, the calculated KP is high for the early time points and below predicted by the PBPK model after longer exposures (squares).

 


View larger version (9K):
[in this window]
[in a new window]
 
FIG. 12. A comparison of predicted KP values for a whole-hand immersion in water (subject 8: see Fig. 6Go) using the default EPA Equation 4 and the PBPK model. The PBPK model predicted a consistent KP, regardless of the length of exposure (diamonds). By using the PBPK model to calculate the total TCE absorbed over exposure lengths of 0.4–2 h and then inserting total absorbed into Equation 4, the calculated KP is low, but increases with exposure time (squares). Under conditions of no volatilization (Kloss = 0) the PBPK model predicts more total TCE would be absorbed, and thus KP would also increase when the total absorbed is used in Equation 4 (circles). Finally, when the 0.3-h delay that was observed in this subject is subtracted from the length of exposure (T) for Equation 4 and Kloss is set to 0, the KP obtained using Equation 4 matches the PBPK model prediction, but only after equilibrium is reached (triangles).

 


    ACKNOWLEDGMENTS
 
The authors thank Annette Barnes for her assistance in conducting the rat exposures.


    NOTES
 
This research was supported in full under Cooperative Agreement DE-FG07-97ER62509, Environmental Management Science Program, Office of Science and Technology, Office of Environmental Management, United States Department of Energy (DOE). However, any opinions, findings, conclusions, or recommendations expressed herein are those of the authors and do not necessarily reflect the views of the DOE.

1 To whom correspondence should be addressed. Fax: (509) 376-9064. E-mail: torka.poet{at}pnl.gov. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
ATSDR (Agency for Toxic Substances and Disease Registry). (1997). Toxicological profile for trichloroethylene (update). ATSDR, Atlanta, GA.

Bogen, K. T., Colston, B. W. Jr., and Machicao, L. K. (1992). Dermal absorption of dilute aqueous chloroform, trichloroethylene, and tetrachloroethylene in hairless guinea pigs. Fundam. Appl. Toxicol. 18, 30–39.[ISI][Medline]

Bronaugh, R. L. (1998). Current issues in the in vitro measurement of percutaneous absorption. In Dermal Absorption and Toxicity Assessment. (M. S. Roberts and K. A. Walters, Eds.), pp.155–159. Marcel Dekker, New York.

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

Bull, R. J., Sanchez, I. M., Nelson, M. A., Larson, J. L., and Lansing, A. J. (1990). Liver tumor induction in B6C3F1 mice by dichloroacetate and trichloroacetate. Toxicology 63, 341–359.[ISI][Medline]

Corley, R. A., Gordon, S. M., and Wallace, L. A. (2000). Physiologically based pharmacokinetic modeling of the temperature-dependent dermal absorption of chloroform by humans following bath water exposures. Toxicol. Sci. 53, 13–23.[Abstract/Free Full Text]

Corley, R. A., Markham, D. A., Banks, C., Delorme, P., Masterman, A., and Houle, J. M (1997). Physiologically based pharmacokinetics and dermal absorption of 2-butoxyethanol vapor by humans. Fundam. Appl. Toxicol. 39, 120–130.[ISI][Medline]

Dekant, W., Vamvakas, S., Koob, M., Kochling, A., Kanhai, W., Muller, D., and Henschler, D. (1990). A mechanism of haloalkene-inducted renal carcinogenesis. Environ. Health Perspect. 88, 107–110.[ISI][Medline]

Fisher, J. W., Gargas, M. L., Allen, B. C., and Andersen, M. E. (1991). Physiologically based pharmacokinetic modeling with trichloroethylene and its metabolite, trichloroacetic acid, in the rat and mouse. Toxicol. Appl. Pharmacol. 109, 183–195.[ISI][Medline]

Fisher, J. W., Mahle, D., and Abbas, R. (1998). A human physiologically based pharmacokinetic model for trichloroethylene and its metabolites, trichloroacetic acid and free trichloroethanol. Toxicol. Appl. Pharmacol. 152, 339–359.[ISI][Medline]

Gargas, M. L. (1990). An exhaled breath chamber system for assessing rates of metabolism and rates of gastrointestinal absorption with volatile compounds. J. Am. Coll. Toxicol. 9, 447–453.[ISI]

Gargas, M. L., Burgess, R. J., Voisard, D. E., Cason, G. H., and Andersen, M. E. (1989). Partition coefficients of low-molecular-weight volatile chemicals in various liquids and tissues. Toxicol. Appl. Pharmacol. 98, 87–99.[ISI][Medline]

Jepson, G. W., and McDougal, J. N. (1997). Physiologically based modeling of nonsteady state dermal absorption of halogenated methanes from an aqueous solution. Toxicol. Appl. Pharmacol. 144, 315–324.[ISI][Medline]

Jepson, G. W., and McDougal, J. N. (1999). Predicting vehicle effects on the dermal absorption of halogenated methanes using physiologically based modeling. Toxicol. Sci. 48, 180–188.[Abstract]

Kezic, S., Mahieu, K., Monster, A. C., and de Wolff, F. A. (1997). Dermal absorption of vaporous and liquid 2-methoxyethanol and 2-ethoxyethanol in volunteers. Occup. Environ. Med. 54, 38–43.[Abstract]

Maibach, H. I., Feldman, R. J., Milby, T. H., Serat, W. F. (1971). Regional variation in percutaneous penetration in man. Pesticides. Arch. Environ. Health 23, 208–211.[ISI][Medline]

Mattie, D. R., Bates, G. D. Jr., Jepson, G. W., Fisher, J. W., and McDougal, J. N. (1994). Determination of skin:air partition coefficients for volatile chemicals: Experimental method and applications. Fundam. Appl. Toxicol. 22, 51–57.[ISI][Medline]

McDougal, J. N., Jepson, G. W., Clewell, H. J. III, Gargas, M. L., and Andersen, M. E. (1990). Dermal absorption of organic chemical vapors in rats and humans. Fundam. Appl. Toxicol. 14, 299–308.[ISI][Medline]

McDougal, J. N., Jepson, G. W., Clewell, H. J., MacNaughton, M. G., and Andersen, M. E. (1986). A physiologically pharmaockinetic model for dermal absorption of vapors in the rat. Toxicol. Appl. Pharmacol. 85, 286–294.[ISI][Medline]

Morgan, D. L., Cooper, S. W., Carlock, D. L., Sykora, J. J. Sutton, B., Mattie, D. R. and McDougal, J. N. (1991). Dermal absorption of neat and aqueous volatile organic chemicals in the Fischer 344 rat. Environ. Res. 55, 51–63.[ISI][Medline]

Nakai, J. S., Stathopulos, P. B., Campbell, G. L., Chu, I. Li-Muller, A., and Aucoin, R. (1999). Penetration of chloroform, trichloroethylene, and tetrachloroethylene through human skin. J. Toxicol. Environ. Health. 58, 157–170.[ISI]

Poet, T. S., Thrall, K. D., Corley, R. A., Hiu, X., Edwards, J. A., Weitz, K. K., Maibach, H. I., and Wester, R. C. (2000). Utility of real time breath analysis and physiologically based pharmacokinetic modeling to determine the percutaneous absorption of methyl chloroform in rats and humans. Toxicol. Sci. 54, 42–51.[Abstract/Free Full Text]

Roberts, M. S., Anissimov, Y. G., and Gonsalvez, R. A. (1999). Mathematical models in percutaneous absorption. In Percutaneous Absorption: Drugs-cosmetics-mechanisms-methodology (R.L.. Bronaugh and H.I. Maibacc, Eds.) pp. 3–64. Marcel Dekker, New York.

Scheuplein, R. J., and Blank, I. H. (1971). Permeability of the skin. Physiol. Rev. 51, 702–747.[Free Full Text]

U.S. EPA. (1986). Exposure Factors Handbook. EPA/600/P-95/002Ba. National Center for Environmental Assessment, Washington, DC.

U.S. EPA. (1992). Dermal exposure assessment: Principles and applications. EPA/600/8–91/011B. National Center for Environmental Assessment, Washington, DC.

Weisel, C. P., and Jo, W-K. (1996). Ingestion, inhalation, and dermal exposures to chloroform and trichloroethylene from tap water. Environ. Health Perspect. 104, 48–51.[ISI][Medline]

Wester, R. C and Maibach, H. W. (1986) Dermatopharmacokinetics: A dead membrane or a complex multifunctional viable process. Prog. Drug. Metab. 9, 95–109.[ISI]