* Battelle, Pacific Northwest Division, P.O. Box 999, Richland, Washington 99352; and
Department of Dermatology, P.O. Box 0989, University of California, San Francisco, California 94143
Received August 17, 2001; accepted October 30, 2001
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key Words: human; PBPK modeling; perchloroethylene; rat; soil matrix.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Environmental contaminants, such as PCE, have the potential to collect in soil; therefore, the assessment of dermal absorption from soil exposures is needed for a complete toxicokinetic exposure assessment (Sedman, 1989). PCE leaks into soil from storage tanks and can be deposited from contaminated air during rain. PCE is more soluble in soil than water and volatilization rates from soil are lower than from water. The EPA National Priorities List of hazardous waste sites identified PCE in at least 771 of 1430 sites (HAZDAT, 1996
). PCE is 31st on the 1999 CERCLA list of priority hazardous substances based on frequency of detection (detection limit of 1 to 5 µg/kg) in the environment and potential for human exposure (ATSDR, 2000
).
A noninvasive real-time mass spectrometric (MS/MS) breath analysis technique was employed to determine the circulating levels of PCE following dermal absorption. The percutaneous absorption of PCE from soil in rats and human volunteers and a physiologically based pharmacokinetic (PBPK) model that includes dermal absorption parameters to quantify the bioavailability of PCE were developed. In dermal absorption studies, bioavailability is often expressed either as an absolute amount or percentage of the dose absorbed. The percentage of the applied dose that is absorbed is dependent upon exposure conditions, such as length of exposure and exposure concentration, and is not useful in extrapolating beyond the experimental conditions. Of greater utility is the calculation of the percutaneous permeability coefficient (KP), which can be used in validated kinetic models to extrapolate beyond the experimental conditions.
A PBPK model was used to estimate the KP for dermal absorption of PCE in rats and humans. Equations for dermal absorption in the PBPK model were based upon Fick's first law of diffusion, as described by Jepson and McDougal (1997). Three different models that varied only in the description of the skin compartment were compared. The simplest model included a single homogenous dermal compartment. The other two models included two dermal compartments. The first, parallel model employed a compartment that was designed to mimic a follicular route of absorption and the remaining composite dermal compartment; the second, multilayered compartment divided the skin into a stratum corneum (sc) barrier and the remaining viable cutaneous tissue.
A number of different researchers have incorporated Fick's law equation into models to determine the transdermal flux under nonsteady state absorption conditions (Corley et al., 2000; Jepson and McDougal, 1997
; McDougal et al., 1986
; Poet et al., 2000b
). Integrating Fick's law into a PBPK model allows for an instantaneous estimation of KP and exposure concentration without requiring that measurements be taken at steady state or from an infinite source (Poet et al., 2000a
). The KP should be consistent regardless of exposure concentration and surface area for any given exposure site and chemical, but it can vary between exposure sites. Unless a mathematical model is used to account for changing exposure concentration, the calculation of flux or permeability coefficient must be assessed at steady state (Poet et al., 2000a
).
The intent of the research described in this article was to describe the absorption of PCE in rats and humans in a manner that will be applicable to risk assessment. The validity of different model descriptions of the dermal compartment are also compared. Like any biokinetic model, these models must be descriptive enough to obtain reasonable parameters while still being simple enough to have utility. The development of a model that adequately describes the data is integral to exposure assessment.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
HPLC grade (99.9% pure) PCE (CAS #127-18-4) and all other chemicals (reagent grade or better) were obtained from Sigma Chemical Co. (St. Louis, MO).
Dermal Exposures in Rats
Application techniques.
Dermal exposures to PCE were conducted in a soil matrix. 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 consisted of 30% sand, 18% clay, 52% silt, and it had an organic content of 1.3% and a pH of 6.8. This soil sample has been used in two previous studies involving the dermal absorption of methyl chloroform and trichloroethylene (Poet et al., 2000a,b
). Male F344 rats were anesthetized using a ketamine/xylazine mixture and the hair on the lower back clipper was shaved the day prior to exposure. Any rats that showed skin irritation or nicks were excluded from the study. Two soil loading volumes were compared for nonoccluded exposures. The low volume consisted of a target of 0.5 g of soil over 8 cm2 of skin (0.0625 g/cm2), and the high volume consisted of 5 g of soil over 5 cm2 (1 g/cm2). Occluded exposures employed 1 g/cm2. PCE was mixed with the soil the evening before exposures, sealed tightly with minimal headspace, and mixed overnight in a rotating mixer. The different exposure conditions are outlined in Table 1
.
|
A glass cell was used for the high volume exposures. Target concentrations of 15 or 50 mg/kg were placed inside a 2.5-cm diameter hand-blown glass cell (O.Z. Glass Co., Pinole, CA). The 2.5 cm deep cell was partitioned into two compartments by a vapor-permeable ceramic frit. The lower compartment contained the PCE-laden soil and the upper compartment contained activated charcoal to trap volatilized chemical. The glass cell only had space for the addition of 2 g of charcoal above the ceramic frit. Preliminary studies (as outlined above for the DuoDerm system) revealed that the 2 g of charcoal was not sufficient to block the escape of PCE into the chamber, so an additional charcoal patch was placed above the glass exposure cell. Following the addition of the charcoal patch, no PCE escaped into the chamber.
For occluded soil samples, target concentrations of 15 and 50 g/kg PCE in 5 g of soil were placed in a hand-blown glass cell (O.Z. Glass Co., Pinole, CA). The occluded cell had the same dimensions as the lower compartment of the nonoccluded cell and was completely enclosed. The glass cells were attached to the clipped area using a cyanoacrylate adhesive.
Dermal exposures.
To quantify total absorbed dose, the exact weight of administered soil was recorded and samples of exposure soil, the amount remaining at the end of the exposure, and the amount volatilized to the charcoal were analyzed using gas chromatography. Soil samples were analyzed using a Headspace Autosampler (Perkin-Elmer 40XL) linked to a Hewlett-Packard 5890 Series II GC (Hewlett-Packard, Avondale, PA). A Restek Rtx-Volatiles column (30 m x 0.32 mm x 1.5 µm) cross-bonded with phenylmethyl polysiloxane was used. The oven temperature was set at 90°C and the injection and FID detector temperatures were set at 120°C. Helium was the carrier gas at 8 psi. Charcoal from the nonoccluded patch system was extracted using toluene and PCE concentrations measured using similar GC conditions, except 2 µl of toluene was injected (splitless) onto the GC.
Immediately following dermal application, rats were individually placed in small off-gassing chambers as described by Gargas (1990) and a Teledyne Discovery II MS/MS equipped with an atmospheric sampling glow discharge ionization source sampled from the off-gassing chamber (representing exhalation from the animal) approximately every 5 s as described previously (Poet et al., 2000b). Breathing air was continually supplied to the rat through the lid of the off-gassing chamber at a measured rate (200 ml/min). Airflow rates were measured using flow meters from Sierra Instruments (Carmel Valley, CA). The ASGDI source derived reagent ions directly from the volatile chemicals in the sampled air. An electric potential was established by applying 400 V between two plates. Ions were then focused onto the MS/MS trap. Helium was used as a buffer and collision gas. The MS intensity data was converted to concentration (ppb) through the use of external standards prepared in Tedlar® bags. A standard curve was generated each day of experimentation. PCE was quantified by selective ion monitoring of the most abundant product m/z ratios 160164.
Dermal Exposures in Human Subjects
Study participants.
Two healthy female volunteers and one healthy male volunteer participated in the study; demographic data including gender, race, age, body weight, and height are given in Table 2. The studies were conducted under approval from both the University of California at San Francisco (Committee on Human Research) 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; no significant cardiovascular, hepatic, central nervous system, renal, hematological, or gastrointestinal diseases; and no dermatological problems. Since PCE is highly lipophilic and PCE pharmacokinetics are sensitive to the amount of body fat, the percent body fat for each subject was determined using a handheld near-infrared body fat analyzer (Futrex®, Gaithersburg, MD).
|
Subjects were provided clean breathing air via a facemask with a two-way nonrebreathing valve to prevent potential inhalation exposures. The exhaled breath was passed through a heated mixing chamber (1.3 l volume) from which the MS/MS drew a sample for analysis approximately every 5 s, as described previously (Poet et al., 2000b). Excess exhaled air was vented from the mixing chamber to a hood with negligible flow restriction via a large borehole exit tube. Background exhaled breath measurements were taken for 2 min before each exposure.
PBPK Model
The overall model structure was based on the flow-limited model for volatile chemicals initially described by Ramsey and Andersen (1984), as it was developed specifically for PCE by Reitz et al. (1996). The initial model structure was comprised of five compartments: fat, liver, rapidly perfused tissues, slowly perfused tissues, and skin linked by the systemic circulation. In a standard blood flow-limited model, the blood is assumed to equilibrate with exhaled breath. For rat exposures, an equation was also added to calculate 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 as it was drawn into the MS/MS, as described previously (Poet et al., 2000a).
Chemical-specific partition coefficients and metabolism rates were obtained from a previously established PBPK model for PCE (Reitz et al., 1996). In this study, human blood/air partition coefficients were determined experimentally, and human tissue/blood partition coefficients were estimated by dividing the rat tissue/air by the human blood/air partition coefficient (Reitz et al., 1996
). Reitz et al. (1996) also conducted a sensitivity analysis that showed that the blood/air partition coefficient had the most impact on the model estimation of amount of PCE metabolized. However, this was for inhalation exposures, and blood/air partitioning would affect the transfer of PCE from inhaled air. Soil partition coefficients were determined using the method of Gargas et al. (1989), with modifications. Quadruplicate samples of 1 g soil were placed in sealed vials and incubated for 2 h with PCE at 32°C (the approximate temperature at the skin surface) with vigorous shaking in a heating block. PCE was added as vapor from Tedlar® bags containing approximately 10,000 ppm PCE. The amount of PCE in the headspace of vials containing soil was compared to empty reference vials containing PCE only. The soil:skin partition coefficient was calculated by dividing the soil:air partition coefficient by the skin:air partition coefficient. The skin:air partition coefficient was determined by Mattie et al. (1994) in male F344 rats. Parameters for the rat and human PBPK models are given in Tables 3 and 4
.
|
|
![]() | ((1)) |
![]() | ((2)) |
Kloss and Kp were estimated for each individual animal or human subject by a least-squares fit of the model to the exhaled breath, charcoal, and soil data (where appropriate) using the SimusolvTM optimization subroutine (Dow Chemical Co.). For human subjects, the observed delay before the appearance of PCE in exhaled breath was estimated by visually extrapolating the appearance of PCE in the exhaled breath back to the x-axis. Thus, each parameter was not only optimized to the exhaled breath concentrations, but also to the chemical concentrations in exposure media and escaping into the charcoal patches.
When it became apparent that the single homogenous dermal compartment model would not adequately describe the human exhaled breath data, two additional model structures with dual dermal compartments were established. These two dermal models consisted of a parallel dermal compartment, and a dual layered compartment structure, as described previously by Bookout et al. (1996, 1997). The parallel model is predicated on the theory that the follicles in the skin serve as a shunt for chemical absorption (Kao et al., 1988). Therefore, the skin is divided into a follicle compartment that contains the follicles and sweat glands and a composite dermal compartment that contains the remaining layers of the skin. The area of the follicular compartment was assumed to be 1/100th of the rest of the skin and the depth was assumed to be 388/560th that of the entire skin (Grabau et al., 1995
).
The permeability coefficients through the two parallel compartments are related to the KP through the skin as a whole such that the combined permeability coefficients must match the single compartment coefficient following an area correction, as described in Bookout et al. (1997). Therefore, the composite KP was first optimized for the one compartment model, then fixed in the two compartment model, and the two new KPs were optimized to the data but constrained by their relationship to the composite KP. The equations used in the PBPK model to describe the dual parallel compartments and the relationship between the composite KP and the KPs for the dual compartments are given in the Appendix.
The second alternative dermal model described by Bookout et al. (1996) separates the stratum corneum from the underlying viable cutaneous tissue. The barrier function of the skin primarily resides in the stratum corneum (Scheuplein, 1978). In this dual layered model, the dermis and viable epidermis are combined into the viable cutaneous compartment. Blood exchange occurs only in the viable cutaneous compartment, and the stratum corneum imparts a barrier between the surface exposure and the viable cutaneous tissue. The stratum corneum compartment was assumed to be 11/560th of the total skin volume (Bookout et al., 1996
).
As for the parallel model, the total permeability coefficient for both compartments is related to the permeability coefficient for a single well-stirred compartment. For a layered model, the KPs for each new compartment are inversely related to the composite KP for the single compartment (Bookout et al., 1996). Therefore, the KP first optimized for the one compartment model was also used as the composite KP for the dual-layered model. A diagrammatic representation of the two alternative compartment models is given in Figure 1
, and the equations used in the PBPK model to describe each model are given in the Appendix. Since only the sc compartment is in contact with the media and only the viable cutaneous compartment is involved with blood exchange, the partition coefficients for the sc compartment:media (PSC/media) and the viable cutaneous compartment:blood (PVC/blood) were calculated using the media/air and blood/air partition coefficients, respectively. The permeability coefficient for the viable cutaneous compartment (KPVC) was not optimized directly, but was based on the inverse relationship between the overall KP and KP for the sc compartment (KP SC), as described by Bookout et al. (1996). The two new partition coefficients (PSC, PVC) were calculated in the same fashion as for the parallel model. PSC and KPSC were optimized using the log likelihood function (LLF) of SimusolvTM, and PVC and KPVC were optimized indirectly using the equation in the Appendix as described by Bookout et al. (1996).
|
Data Analysis
Data are presented as the mean ± SD of n = 35 exposures. 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. Statistical significance was considered at p < 0.01.
Dermal compartment evaluation.
As the model becomes progressively complicated with the increase in the number of dermal compartments, there is a possibility that more parameters than necessary to explain the data have been added. In order to discriminate the goodness of fit of the dual compartment models versus the one-compartment model, likelihood ratio tests were performed. Models are considered to be nested when the basic model structures are identical except for the addition of complexity, such as the added dermal compartments. In this case, the model with the single well-stirred dermal compartment was nested within the dual dermal compartment models. Under these conditions, the likelihood ratio can be used to statistically compare the relative ability of the models to describe the same data, as described in the "Reference Guide for Simusolv" (Steiner et al., 1990). The hypothesis that one model is better than another is calculated using the likelihood functions evaluated at the maximum likelihood estimates. Since the parameters are optimized in the model using the maximum log likelihood function (LLF), the resultant LLF is used for the statistical comparison of the models. The equation states that two times the log of the likelihood ratio follows a chi-square (
2) distribution with r degrees of freedom:
![]() | ((3)) |
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
PCE exhaled breath levels in rats exposed to PCE in 0.5 g of soil (low volume exposures) under nonoccluded conditions peaked in less than 1 h (Fig. 2). The rate of loss (Kloss) was concurrently estimated by fitting to the exhaled breath data, the concentration remaining in the soil, and the amount volatilizing to the patch. The parameters KP and Kloss were optimized in an iterative manner until all three endpoints (soil concentration, charcoal concentration, and exhaled breath profile) were described by the model using the LLF function of SimusolvTM. An example of rat chamber concentration data and PBPK model predictions is given in Figure 2
. Model predictions of charcoal and soil concentrations over time are compared with actual soil and charcoal concentrations as determined using GC analysis in Figure 3
. The KP and Kloss values that resulted in the best fit to the data for the nonoccluded low volume exposures were 0.087 ± 0.013 cm/h and 0.791 ± 0.065/h, respectively (Table 5
). Exposures to PCE in 5 g of soil resulted in similar exhaled breath profiles (Fig. 4
). As for the low volume exposures, the peak exhaled breath concentration was reached under an hour and steadily declined. The KP was not significantly different between any of the exposure conditions (Table 5
).
|
|
|
|
The percentage (and amount) absorbed was considerably higher for the high volume occluded exposure over the nonoccluded exposure. For the occluded exposures, the exhaled breath concentrations reached a plateau and remained steady between 1 and 5 h whereas in the nonoccluded exposures, the exhaled breath concentrations began to decline in less than 1 h (Figs. 2, 4, and 5). The PBPK model predictions indicated that the exposure concentration declines to less than 10% of the original concentration in the low volume exposures and less than 30% in the high volume exposures by 1 h, about the time that exhaled breath concentrations peaked (Fig. 3
). In the occluded exposures, the percentage absorbed would have continued to increase for exposures longer than 5 h since 45% of the original concentration remained in the soil. In contrast, there was less than 1% of the original concentration of PCE remaining in either of the nonoccluded exposure systems to be absorbed at 5 h.
A cyclical pattern in the exhaled breath profile was noted for most rat exposures. This same pattern was observed previously for similar exposures to trichloroethylene (Poet et al., 2000a). These waves in the exhaled breath levels were associated with patterns of activity where the rats moved around in the chamber. 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.
Human Exposures
To track PCE soil concentrations over time, aliquots of soil were taken from the exposure container every 30 min. The large exposure volume (4 kg) and surface area of the container resulted in a slow decline of PCE from the soil as it was volatilized and absorbed into the body (Fig. 6). The rate of loss (Kloss) was concurrently estimated by fitting to the exhaled breath data and the concentration remaining in the soil over time. The parameters (KP and Kloss) were optimized in an iterative manner using the LLF function of SimusolvTM until both endpoints (soil concentration and exhaled breath profile) were described by the model. The Kloss (rate of volatilization from the container) that resulted in the best fit of the data was 0.851 ± 0.054/h (Table 7
). This Kloss is remarkably similar to that determined for the rat exposures (0.953 ± 0.076/h: Tables 5 and 7
).
|
|
|
Four new parameters were added to each of the dual compartment models. To fit the parallel compartment model to the data, the overall KP was set equal to the one compartment KP and two new KP parameters were introduced, KPFO for the follicular compartment and KPCD for the remaining composite dermal compartment. KPFO and KPCD were related to KP as described by Bookout et al. (1997) using the equation given in the Appendix. Likewise, the two new partition coefficients for the follicular subcompartment (PFO) and the composite dermal subcompartment (PCD) were related to the partition coefficient of the homogenous skin, which was measured using a vial equilibrium technique by Mattie et al. (1994) following a volume correction. PFO and KPFO were optimized using the LLF of Simusolv, and PCD and KPCD were optimized indirectly using the equation in the Appendix as described by Bookout et al. (1997). Additional partition coefficients for follicle/media follicle/blood, and composite/media, composite/blood were calculated by dividing by the media and blood partition coefficients, respectively. Since the new permeability parameters and partition coefficients for the two dual dermal compartment models were related to the one compartment KP, the volatilization rate and total amount absorbed were held constant, equal to the results from the single compartment model. As expected, the parallel model resulted in a slightly steeper slope for the initial rise in PCE breath concentrations, and a concomitantly faster decline. Visually, the apparent fit to the data was not improved using the parallel model (Fig. 7).
The four new parameters for the layered dermal model were KPSC and PSC for the sc compartment and KPVC and PVC for the viable cutaneous compartment. The sc barrier imparted a slower absorption and elimination from the skin, and a substantially improved fit to the data, by visual inspection (Fig. 7). The rate-limiting KPSC was much lower than the KPVC (Table 8
). KPVC was very similar to the KP estimated for the absorption of PCE in the rats.
|
|
|
The chi-square (2) value for 4 degrees of freedom at a confidence level of 0.99 is 13.28 and the hypothesis is that the more complicated models result in better fits to the data. The comparison of the LLF for the rat single versus parallel model using Equation 3
gives:
![]() |
Since 281 < 9.488 the dual parallel model does not give a statistically better fit than the single dermal compartment model. The remaining statistical model comparisons are given in Table 10. Using the log likelihood ratio, only the dual layered model for human dermal absorption of PCE fits the null hypothesis and gives a statistically improved fit to the data; this verifies the visual assessment of the data.
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The exhaled breath profile for PCE was similar to previous results with trichloroethylene in rats, but PCE showed a slower absorption and elimination than trichloroethylene in humans. PCE is the least volatile and the most lipophilic of the three chemicals. Also, at 108.5, the soil/air partition coefficient is 13-fold greater than for methyl chloroform and 100-fold greater than for trichloroethylene. The permeability coefficient for PCE in rats (0.111 cm/h) was between methyl chloroform (KP = 0.15) and TCE (KP = 0.088), although in practical terms, the KP for these three chlorinated solvents were very similar.
Rat dermal exposures to PCE in soil were used to assess the effect of exposure concentrations and soil loading volume. Rats were exposed to 15 and 50 mg PCE/kg soil at loading volumes of either 0.0625 or 1 g/cm2. The fraction absorbed decreased with increasing load (Table 6); this is in agreement with model predictions of McKone (1990). The increased fractional absorption is related to the lower amount of PCE applied and the increased surface area/soil volume that allowed for increased absorption on a percent basis. However, the amount absorbed and KP were essentially the same for the low and high volume exposures. Trapping PCE at the skin surface in the occluded exposure system also resulted in an increased percentage absorbed, but did not alter the KP (Table 6
). By definition, KP should be independent of concentration, loading volume, volatilization, and time of exposure, although each of these variables can alter the percentage absorbed (Poet et al., 2000a
; U.S. EPA, 1992
).
Absorption parameters (KP, fraction absorbed) and loss rate (Kloss) did not vary across doses (Table 6). The high volume exposure systems (5 g of soil) demonstrate the importance of loss of chemical, when PCE is permitted to volatilize away from the exposure system; only
10% is absorbed, much less than the
50% absorbed from the occluded system. In a real world exposure scenario, it seems likely that PCE concentration in the soil will be low due to evaporation and movement through the soil. In addition, chemical soil sorption follows a process of aging by which the chemical contact with the soil shifts from a reversible interaction on the surface of the soil particle to deeper sites within the soil particles (Pignatello and Xing, 1996
). In such aged soils, dermal bioavailability may decrease even further than observed in the exposures conducted in this study. The process of aging takes varying length of time depending upon the soil type, temperature, microbial content, etc. Thus, this study focused on the impact of reversible binding of PCE in a specific soil matrix and volatility. The solubility of PCE in water is 0.01, and at concentrations below this, no PCE was detected in exhaled breath of either rats exposed to 5 ml of water or a single human subject exposed to 4 l of water using the hand immersion method (data not shown).
The low KP and the expected volatilization of PCE from soil would indicate that dermal exposures would not result in significant uptake of PCE. In a PBPK model for PCE developed by Rao and Brown (1993), the octanol/water partition coefficient was used to calculate a maximal dermal permeability coefficient of 0.125 cm/h (U.S. EPA, 1992). Applying this coefficient to their model, they estimated that approximately 78% of the total dose of PCE would be contributed by the dermal absorption from the combined inhalation and dermal uptake while showering. Bogen et al. (1992) calculated a KP of 0.37 cm/h in hairless guinea pigs from a dilute aqueous exposure, and Nakai et al. (1999) determined a PCE permeability coefficient of 0.018 cm/h in human skin in vitro. The permeability of PCE determined here for rats (0.111 cm/h) was similar to the KP of 0.125 cm/h estimated from the octanol/water partition coefficient (Rao and Brown, 1993
; U.S. EPA, 1992
). However, the human KP for PCE absorption through the hand was considerably lower than that estimated for human abdominal and breast skin in vitro reported by Nakai et al. (1999). Absorption through human abdominal and breast skin is expected to be less than through the much thicker hand (U.S. EPA, 1992
), and the difference between rat and human absorption is consistent with previous studies that have shown that percutaneous absorption can range from 120-fold higher in rats compared to humans (Bronaugh, 1998
; U.S. EPA, 1992
). In the previous studies with trichloroethylene and methyl chloroform, the KP in humans were approximately 20- and 40-fold higher, respectively. The difference in KP for PCE between the rats and humans was over 100-fold. Some of these differences are likely due to the different exposure sites. The thickness of the epidermis on the human hand ranges from
85 µm on the back of the hands to
370 µm on the fingertips (International Commission on Radiological Protection, 1992
). The epidermis of the F344 rat is in the range of 18 µm (Grabau et al., 1995
). Even more important is the difference between the stratum corneum layer between the hands and feet and the rest of the body. Over most of the body, the stratum corneum is uniform and ranges from 1316 µm, but on the palms of the hands and the soles of the feet, the stratum corneum can be more than 600 µm thick (International Commission on Radiological Protection, 1992
). The stratum corneum rats has been estimated at 613 µm in frozen sections from different sites on Sprague-Dawley rats (Monteiro-Riviere et al., 1990
) and 11.2 ± 5.1 µm in the back of F344 rats (Grabau et al., 1995
).
The fit of the PBPK model to the data using a single, well-stirred skin compartment was poor for the human exposures, with less than 75% of the variability of the data explained by optimizing KP using the LLF using SimusolvTM. The slower than predicted elimination might be thought to involve a lower rate of metabolism than predicted. Estimated human metabolic rate constants for PCE have varied from 6.1732.9 mg/h and 2.664.66 mg/l for Vmax and KM, respectively (Hattis et al., 1993; Reitz et al., 1996
). However, as is typical for flow limited models, simulations of the low exposures were insensitive to changes in the metabolic rate constants.
Since the concentrations of PCE in blood and liver were well below the KM, the most likely cause of the poor fit to the human data was the use of the overly simplistic well-stirred dermal compartment. Because the stratum corneum is so prominent on the hand, and it has been proposed to function as a barrier to percutaneous absorption, a model that separated out the stratum corneum and the viable epidermis into two compartments was used. The stratum corneum compartment represented a barrier to absorption with a rate-limiting Kp and the viable epidermis compartment contained the rest of the skin and the blood exchange. The optimized Kp for the stratum corneum compartment (KpSC) was not different from the Kp estimate for the one compartment model. However, the optimized Kp for the transfer of PCE from the stratum corneum to the viable cutaneous compartment (KpVC) was very similar to the Kp estimated for the dermal absorption of PCE in rats (Tables 5 and 8).
The second prevalent theory for the different rate of absorption between laboratory animals and humans is the existence of skin structures, such as follicles, that act as shunts (Kao et al., 1988; McKone, 1990
; Scheuplein and Blank, 1971
). Therefore, an additional two-compartment model that might explain the increased rat absorption over the human was compared. For the parallel model, both subcompartments were in contact with the exposure dose and the blood exchange. However, the utilization of this parallel model form did not improve the fit of the data (Fig. 7
).
Since the depth of the stratum corneum and the number of follicles can vary greatly between species (Grabau et al., 1995), the two different dual skin compartment models were also tested against the rat exhaled breath data (Fig. 8
). No improvement in the fit of the model was observed using either of the dual dermal compartment models (Table 10
). The permeability coefficients for the composite dermal (KPFO) in the parallel model and the subcutaneous barrier (KPSC) in the layered model were not different form the aggregate KP for the one, homogenous compartment, indicating that, in these models, the sc barrier drives the absorption rate, not the existence of skin appendages such as follicles.
Researchers in this laboratory group have previously conducted dermal absorption studies with other volatile organic compounds. For temperature-dependent absorption of chloroform from whole-body bathwater exposures, hand or forearm exposures to trichloroethylene, or arm-only exposures to 2-butoxyethanol vapor a single, well-stirred skin compartment provided a good fit to the data (Corley et al., 1997, 2000
; Poet et al., 2000a
). For methyl chloroform, however, a reasonable fit to the data was only obtained after an initial delay was added and the length of the exposure was artificially adjusted to account for the delay at the end of exposure. In this way, exposures that were continued for 2 h were modeled as if they were extended to up to 3.3 h (Poet et al., 2000b
). This assumes that the skin is acting as a depot that continues to release chemical after the cessation of the actual exposure. Since the dual-layered dermal structure seems to describe the dermal absorption of PCE in these very similar exposures, the same model structure was used to describe the methyl chloroform exposures in three human subjects exposed to methyl chloroform by immersing a hand in a bucket of water containing 1.2 ± 0.276 g/l or soil containing 7.1 ± 0.896 g/kg (Fig. 9
), as reported previously. The updated model structure employed a more empirical model scenario with two differences. First, the lag time in the model was described by extrapolating the delay in the appearance of methyl chloroform in the exhaled breath back to the x-axis, instead of optimizing to the fit of the data. Second, no time was added to the model describing the exposure beyond the actual exposure time, thus, the end of the exposure described in the model was 2 h, the time the hand was removed from the water. The fit using the dual layered model was significantly improved over the simple well-stirred compartment, using the likelihood ratio test as described in the Materials and Methods. A similar improvement of fit was demonstrated for the exposures to methyl chloroform in water (Table 11
).
|
|
As models become more complex, there is a risk of increasing the number of parameters beyond that which are needed. When it becomes necessary to add complexity to the model, additional unknown parameters are also added and it is essential to justify their use, both biologically and practically. The use of the layered dermal model to describe human percutaneous absorption is justified by a visual assessment, statistically, and physiologically. The ability of the dual-layered model to drastically improve the fit of the human PCE exposure data and the inability of the parallel model to improve the fit for either the rat or human exposures indicates that the stratum corneum plays an important role in the barrier function of the skin toward absorption of PCE, particularly when exposure is to the hands. That the estimation of a Kp through the viable cutaneous tissue so closely resembles the rat Kp may suggest that the role of the stratum corneum is very important in species differences in dermal absorption, but the role of skin appendages such as hair follicles may be minimal.
The impact of the two-compartment dermal model on risk assessment for PCE will be most evident in species comparisons. Currently, it is difficult to extrapolate from rodent to human exposure assessments. For PCE, it appears that taking into account skin morphology into species-specific models may facilitate such extrapolations in the future. The thick stratum corneum on the hand may have influenced the human KP determined in these experiments, as indicated by the effect of separating out the skin in the PBPK model to a dual layered compartment, and exposures to other body sites may alter the assessment of KP for PCE.
![]() |
APPENDIX |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Parallel Subcompartment Model
The parallel subcompartment model partitioned the skin compartment into a follicular (FO) and the remaining composite dermal (CD). Both compartments were in contact with the PCE-laden soil and were involved in blood exchange (Fig. 1).
Permeability constants.
The permeability coefficient (KP) for the single well-stirred dermal compartment times the total surface area (SA) equals the sum of the KP times the SA for each subcomartment. The permeability for the CD compartment was determined by the relationship between the optimized KP for the FO compartment and the overall KP for the single well-stirred compartment once the equation was rearranged to the following:
![]() | ((4)) |
![]() | ((5)) |
Partition coefficients.
According to Bookout et al. (1996), the partition coefficients measured experimentally for the single skin compartment is equal to the sum of the partition coefficients for each of the subcompartments following a volume correction. The partition coefficient for the CD compartment was determined based on the optimized partition coefficient for the FO compartment and the partition coefficient measured experimentally for whole skin by Mattie et al. (1994).
Mass-balance equations for dermal absorption.
Equations for the transfer of PCE through the FO and CD compartments replace the single equation (Equation 1) for the well-stirred dermal compartment model.
![]() | ((6)) |
![]() | ((7)) |
Layered Subcompartment Model
The layered subcompartment model partitioned the skin compartment into a subcutaneous (SC) and the remaining viable cutaneous (VC) compartments. Only the SC compartment was in contact with the PCE-laden soil and only the VC compartment was involved in blood exchange (Fig 1).
Permeability constants.
With the layered model, the KPs have an inverse relationship to each other. Therefore, 1/KP equals the sum of the recripocals of the subcompartment KPs. The permeability for the viable cutaneous (VC) compartment was determined by the relationship between the optimized KP for the subcutaneous (SC) compartment and the overall KP for the single well-stirred compartment.
![]() | ((8)) |
Partition coefficients.
The partition coefficients measured experimentally for the single skin compartment is equal to the sum of the partition coefficients for each of the subcompartments following a volume correction. The partition coefficient for the VC compartment was determined based on the optimized partition coefficient for the SC compartment and the partition coefficient measured experimentally.
![]() | ((9)) |
Mass-balance equations for dermal absorption.
Equations for the transfer of PCE through the SC and VC compartments replace the single equation (Equation 1) for the well-stirred dermal compartment model.
![]() | ((10)) |
![]() | ((11)) |
![]() |
ACKNOWLEDGMENTS |
---|
![]() |
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.
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
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, 3039.[ISI][Medline]
Bookout, R. L., Jr., McDaniel, C. R., Quinn, D. W., and McDougal, J. N. (1996). Multilayered dermal subcompartments for modeling chemical absorption. SAR QSAR Environ. Res. 5, 133150.[Medline]
Bookout, R. L., Jr., Quinn, D. W., and McDougal, J. N. (1997). Parallel dermal subcompartments for modeling chemical absorption. SAR QSAR Environ. Res. 7, 259279.[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. 155159. 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, 407484.[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, 1323.
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, 120130.[ISI][Medline]
Devore, J. L. (1995). Tests of hypotheses. In Probability and Statistics for Engineering and the Sciences, pp. 304345. Wadsworth, Belmont, CA.
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, 8799.[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. Amer. Coll. Toxicol. 9, 447453.[ISI]
Grabau, J. H., Dong, L., Mattie, D. R., Jepson, G. W., and McDougal, J. N. (1995).Comparison of anatomical characteristics of the skin for several laboratory animals. TAL/OE-R-1995-066. Toxciology Division, Armstrong Laboratory, Wright-Patterson Air Force Base, OH.
Hattis, D., White, P., and Koch, P. (1993). Uncertainties in pharmacokinetic modeling for perchloroethylene: II. Comparison of model predictions with data for a variety of different parameters. Risk Anal. 13, 599610.[ISI][Medline]
HAZDAT (1996). Agency for Toxic Substances and Disease Registry, Atlanta, GA.
International Commission on Radiological Protection (1992). Report of the Task Group on Reference Man. Pergamon Press, New York.
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, 315324.[ISI][Medline]
Kao, J., Hall, J., and Helman, G. (1988). In vitro percutaneous absorption in mouse skin: Influence of skin appendages. Toxicol. Appl. Pharmacol. 94, 93103.[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, 5157.[ISI][Medline]
McDougal, J. N., Jepson, G. W., Clewell, H. J., III, MacNaughton, M. G., and Andersen, M. E. (1986). A physiologically based pharmacokinetic model for dermal absorption of vapors in the rat. Toxicol. Appl. Pharmacol. 85, 286294.[ISI][Medline]
McKone, T. E. (1990). Dermal uptake of organic chemicals from a soil matrix. Risk Anal. 10, 407419.[ISI][Medline]
Monster, A. C. and Houtkooper, J. M. (1979). Estimation of individual uptake of trichloroethylene, 1,1,1-trichloroethane and tetrachloroethylene from biological parameters. Int. Arch. Occup. Environ. Health 42, 319323.
Monteiro-Riviere, N. A., Bristol, D. G., Manning, T. O., Rogers, R. A., and Riviere, J. E. (1990). Interspecies and interregional analysis of the comparative histologic thickness and laser Doppler blood flow measurements at five cutaneous sites in nine species. J. Invest. Dermatol. 95,
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 A 58, 157170.[ISI][Medline]
National Cancer Institute (1986). Carcinogenenesis Bioassay of Tetrachloroethylene. National Toxicology Program.
Pignatello, J. J., and Xing, B. (1996). Mechanisms of slow sorption of organic chemicals to natural particles. Environ. Sci. Technol. 30, 111.[ISI]
Poet, T. S., Corley, R. A., Thrall, K. D., Edward, J. A., Tanojo, H., Weitz, K. K., Maibach, H. I., and Wester, R. C. (2000a). Assessment of the percutaneous absorption of trichloroethylene in rats and humans using MS/MS real-time breath analysis and physiologically based pharmacokinetic modeling. Toxicol. Sci. 56, 6172.
Poet, T. S., Thrall, K. D., Corley, R. A., Hui, X., Edwards, J. A., Maibach, H. I., Weitz, K. K., and Wester, R. C. (2000b). 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, 4251.
Ramsey, J. C., and Andersen, M. E. (1984). A physiologically based description of the inhalation pharmacokinetics of styrene in rats and humans. Toxicol. Appl. Pharmacol. 73, 159175.[ISI][Medline]
Rao, H. V., and Brown, D. R. (1993). A physiologically based pharmacokinetic assessment of tetrachloroethylene in groundwater for a bathing and showering determination. Risk Anal. 13, 3749.[ISI][Medline]
Reitz, R. H., Gargas, M. L., Mendrala, A. L., and Schumann, A. M. (1996). In vivo and in vitro studies of perchloroethylene metabolism for physiologically based pharmacokinetic modeling in rats, mice, and humans. Toxicol. Appl. Pharmacol. 136, 289306.[ISI][Medline]
Scheuplein, R. J. (1978). Permeability of the skin: A review of major concepts. Curr. Probl. Dermatol. 7, 172186.[Medline]
Scheuplein, R. J., and Blank, I. H. (1971). Permeability of the skin. Physiol. Rev. 51, 702747.
Sedman, R. M. (1989). The development of applied action levels for soil contact: A scenario for the exposure of humans to soil in a residential setting. Environ. Health Perspect. 79, 291313.[ISI][Medline]
Steiner, E. C., Rey, T. D., and McCroskey, P. S. (1990). Reference Guide for Simusolv. The Dow Chemical Company, Midland, MI.
U.S. EPA (1992). Dermal exposure assessment: Principles and applications. EPA/600/8-91/011B. United States Environmental Protection Agency, Washington, DC.
U.S. EPA. (1996). Exposure Factors Handbook. EPA/600/P-95/002Ba. United States Environmental Protection Agency, Washington, DC.