Toxicology 227 (2006) 156–164
Pharmacokinetic modeling of saturable, renal resorption of
perfluoroalkylacids in monkeys—Probing the
determinants of long plasma half-lives
Melvin E. Andersen a,∗ , Harvey J. Clewell III a , Yu-Mei Tan a ,
John L. Butenhoff b , Geary W. Olsen b
a
b
CIIT Centers for Health Research, Research Triangle Park, NC 27709-2137, United States
3M Company, Medical Department, 3M Center 220-06-W-08, St. Paul, MN 55144, United States
Received 2 June 2006; received in revised form 19 July 2006; accepted 1 August 2006
Available online 12 August 2006
Abstract
Perfluorooctanoate (PFOA) and perfluorooctanesulfonate (PFOS) compounds associated with surface protection product manufactures are distributed globally. The 3–5-year half-lives, reproductive and liver toxicity in animals, and lack of understanding
of the factors regulating retention in the body have led to a world-wide public concern for use of these materials. Using a novel
physiologically-motivated pharmacokinetic model for renal clearance, perfluoroalkylacid pharmacokinetics in monkeys was successfully described by renal resorption via high efficiency transporters for both intravenous and oral dosing. Intravenous dosing
with both PFOA and PFOS in Cynomolgus monkeys produced time course curves consistent with a two-compartment distribution.
Extending the PK model for intravenous dosing to examine blood and urine time course data for repeated oral dosing clearly
identified the saturable renal resorption. Resorption depends on kinetic factors for transport (TmC , transport maximum; KT , transport
affinity) and free fraction in plasma (fplasma ). For PFOA, these parameters were estimated to be 5 mg/(h kg) (TmC ), 0.055 mg/L (KT ),
and 0.02 (fplasma ). PFOS has longer half-life and had respective values of 13.6 mg/(h kg), 0.023 mg/L, and 0.025. PFOS appeared to
have a higher transport capacity and lower affinity than PFOA. Human kinetics indicates even higher resorption efficiency.
© 2006 Elsevier Ireland Ltd. All rights reserved.
Keywords: Perfluorooctanoate; Perfluorooctanesulfonate; Renal resorption; Physiologically-motivated pharmacokinetic modeling
1. Introduction
Perfluorooctanoate (PFOA) and perfluorooctanesulfonate (PFOS) are surfactants that have been used for var∗
Corresponding author. Tel.: +1 919 558 1205;
fax: +1 919 558 1300.
E-mail addresses: [email protected] (M.E. Andersen),
[email protected] (H.J. Clewell III), [email protected] (Y.-M. Tan),
[email protected] (J.L. Butenhoff), [email protected]
(G.W. Olsen).
ious applications including fluoropolymer production,
acid mist suppression, and firefighting foams/clothing.
PFOA is also a metabolic and environmental degradation product of fluorotelomer alcohols, which are used in
food packaging for heat- and grease-resistance purpose
(Kudo et al., 2005). PFOS may be present in the environment as a hydrolysis product of perfluorooctanesulfonyl fluoride and through metabolic and environmental degradation of N-alkyl-perfluorooctanesulfonamides
(Xu et al., 2004). Both these telomer and N-alkylperfluorooctanesulfonamides are used to make surface
0300-483X/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.tox.2006.08.004
M.E. Andersen et al. / Toxicology 227 (2006) 156–164
protection products that can be found in textiles, wax,
paints, cleaning products, clothing, and carpets. Today,
both PFOA and PFOS are widespread throughout the
global ecosystem. Both compounds, for example, are
found in wildlife samples from the Artic and remote
oceanic sites (Giesy and Kannan, 2001; Kannan et al.,
2001; Smithwick et al., 2005). In feeding studies in laboratory animals, toxic responses are seen in liver for
both perfluoroalkylacids (Butenhoff et al., 2002; Seacat
et al., 2002). In a 2-year chronic bioassay with PFOS
at dietary dose of 20 ␮g/g, a slight increase in hepatocellular adenoma was observed (p < 0.05) when adjusted
for survival (Unpublished data; 3M Corp.). Both compounds are peroxisomal proliferating receptor activators
in rats (Maloney and Waxman, 1999; Shipley et al.,
2004) and PFOA causes a constellation of rodent tumors
that have been noted with other peroxisomal proliferators (Biegel et al., 2001). In reproductive studies, PFOS
caused neonatal mortality in rat pups in the absence of
significant maternal toxicity (Luebker et al., 2005). In
developmental studies, PFOA caused delays in general
growth and development in animals (Butenhoff et al.,
2004; Lau et al., 2004, 2006).
PFOA and PFOS are now found in human blood from
the general population (Calafat et al., 2006; Kannan et
al., 2004; Olsen et al., 2004a,b, 2003). Serum elimination half-lives in retired production workers for PFOA
and PFOS and, on average, are approximately 4–5 years
(Olsen et al., 2005). Elimination half-lives in laboratory
animals are considerably shorter than those observed in
humans and, particularly in the case of PFOA, show
interspecies variation and gender differences within
rats (Butenhoff et al., 2004; Kudo and Kawashima,
2003). In animals, the primary route of elimination is
in urine (Butenhoff et al., 2004; Kudo and Kawashima,
2003).
Because of the widespread environmental distribution, long half-lives, and toxicity, these perfluoroacids
have received intense public scrutiny and high visibility in the lay press in regard to risks that might be
posed to human and wildlife by exposures to these compounds (Burros, 2005; Cortese, 2005). Risk assessments
and interpretation of human biomonitoring studies with
PFOA and PFOS are currently impeded by the absence
of PK models that are able to explain the physiological
basis for extended biopersistence and species differences
in elimination. In order to identify the biological determinants of the long half-lives for PFOS and PFOS in
Cynomolgus monkeys and humans, a physiologicallymotivated pharmacokinetic model is developed to evaluate quantitative evidence for a role of renal resorption
in controlling persistence.
157
2. Materials and methods
A PB-PK model for renal resorption was developed and
used to examine time course data on plasma and urine concentrations of PFOA and PFOS for two dosing situations in
Cynomolgus monkeys: (i) intravenous administration at a volume of 2 mL/kg and (ii) oral dosing studies in which the animals
were dosed daily with the acids in a capsule for 26 weeks, followed by an extended observation period after the cessation
of daily dosing. For intravenous dosing, three male and three
female monkeys received a single intravenous dose of potassium PFOA at 10 mg/kg (Butenhoff et al., 2004) or potassium
PFOS at 2 mg/kg (Noker and Gorman, 2003) into a superficial
arm or leg vein. For oral dosing, groups of four to six male
monkeys each were given daily (7 days per week) oral doses
(by capsule) of 0, 3, 10, or 20 mg/kg PFOA (30 mg/kg for 12
days, reduced to 20 mg/kg when dosing reinitiated on test day
22) for 26 weeks (Butenhoff et al., 2002). Groups of four to
six male and female monkeys were given daily oral doses of
0, 0.03, 0.15, or 0.75 mg/kg PFOS for 26 weeks (Seacat et al.,
2002).
In the oral dosing study, urine was collected by moving
monkeys to a metabolism cage overnight, so only a portion of
the voided PFOA is captured (Butenhoff et al., 2004). Thus, the
urine concentration in the model was calculated by dividing
the excretion rate into the urine (Qfil × Cfil ; in mg/h) by the
estimated hourly urine flow for monkeys of a particular body
weight. The ratio mg/h divided by L/h gave the instantaneous
urine concentration in mg/L.
The PK model for resorption was based on the original
description for renal resorption of glucose from the classic
work by Shannon et al. (Hostetter and Meyer, 2004; Shannon
and Fisher, 1938) where resorption takes place due to an apical
transporter with a tubular maximum (Tm ; mg/h) and a transporter dissociation constant (KT ; mg/L). The Tm for each animal
was scaled allometrically to body weight at a given constant,
TmC (mg/(h kg)) body weight).
In this model (Fig. 1), the intravenous bolus entered the
central compartment directly at time zero. Oral uptake was
zero-order over 2 h duration after intubation. Free concentration of chemical (C1 ) in the central compartment either moved
into the tissue compartment, based on inter-compartmental rate
constants (i.e. k12 and k21 ) or was filtered by the kidney (i.e.
filtrate compartment). The flow filtered was 10% of cardiac
output (Qfilc ), which equals to half of the blood flow to kidney (20% of cardiac output) with full plasma clearance. This
filtration rate could be scaled allometrically to body weight
to give the flow for each individual animal (Qfil ). The filtrate
compartment had inflow by filtration and outflow to urine and
resorption. Flow of filtrate out of the proximal tubules to distal
parts of the nephron was set equal to the input flow of filtered
plasma (Qfil ). While the model originally had a filtrate volume
(Vfil ) set at 10 ml, the simulations were essentially insensitive
to the value selected for this parameter. The estimate of this
parameter could be further refined with improved understanding of the filtrate volume resident in the proximal tubules at
any time. The kinetic model description for intravenous dos-
158
M.E. Andersen et al. / Toxicology 227 (2006) 156–164
Therefore, clearance can be described as the following equation:
Qfil
clearance = fplasma Qfil
(2)
Qfil + (Tm /KT )
The minimum, low-concentration clearance (as Cfil approaches
0.0) then depends on four parameters—fplasma , Tm , KT , and Qfil .
Thus, differences in these parameters from one compound to
another or from one species to another would lead to differences
in urinary clearance and in plasma half-lives. The ratio Tm /KT
(L/h) approximates the clearance from the renal filtrate back
into the plasma compartment.
The model was coded and fitted to the measured data in
the computing software, Berkeley Madonna® (Berkeley, CA).
Solutions were obtained using the Rosenbrock algorithm in the
software package. Curve fitting were done by unweighted leastsquares in the software’s curve fitting utility. All simulations
were run on a computer equipped with dual 3.0 GHz Pentium©
processors and the Windows XP operating system.
Fig. 1. A schematic for a physiologically-motivated renal resorption
pharmacokinetic model.
ing of PFOA (same model structure for PFOS) is provided
in Appendix A. The parameters and their values are listed in
Tables 1 and 2.
The rate of loss of compound to urine is related to the volumetric flow of plasma filtered (Qfil ), the fraction PFOA free in
plasma (fplasma ), the concentration of PFOA in the filtrate (Cfil ),
and the proportion of compound in filtrate that moves unresorbed through to urine, as shown in the following equation:
unresorbed proportion =
Qfil
Qfil + (Tm /KT )
(1)
3. Results
In the monkeys, the plasma time course of PFOA after
intravenous dosing was consistent with a two compartment PK model (Fig. 2). PFOA had an initial volume
of distribution of 0.1–0.2 L/kg with further, though limited, distribution into other tissues. With the exception of
the transport maximum (TmC ) for individual monkeys,
intravenous time course curves for individual animals
were described with a single set of model parameters.
The range for TmC (mg/(h kg body weight)) required
to fit the intravenous time course plasma data from
Table 1
Parameters used in the PFOA model for intravenous (iv) or oral simulations
Parameters
Values
Sources
Body weight (BW)
3.5–6.5 kg
Cardiac output (Qcc)
Volume of renal filtrate (Vfil )
15 L/(h kg)
0.01 L
Renal plasma filtration rate (Qfilc )
0.1
Volume of distribution (Vdc )
Transport maximum (TmC )
0.11 L/kg (iv), 0.14 L/kg (oral)
4.73 mg/(kg h) (iv), 5 mg/(kg h) (oral)
Transport affinity constant (KT )
Proportion of PFOA free in blood
(fplasma )
Compartmental transfer rate constant
from central to tissues (k12 )
Compartmental transfer rate constant
from tissues back to central (k21 )
0.01 mg/L (iv), 0.055 mg/L (oral)
0.018 (iv), 0.02 (oral)
Experimental design (Butenhoff et al., 2002,
2004)
Corley et al. (1990)
Assumption; the model simulations were not
sensitive to this parameter
Assumption; equals to half of the blood flow to
kidney (20%)a
Optimized
Optimized; average values from TmC that gives
the best fit for individual monkeys
Optimized
Optimized
3.4 h−1 (iv), 3.3 h−1 (oral)
Optimizedb
0.13 h−1 (iv), 0.1 h−1 (oral)
Optimizedb
a No data on monkeys were found. Based on the blood flow rates to kidneys reported for humans (17.5%) and dogs (17.3%) (Brown et al., 1997),
20% was chosen for the monkeys.
b For the iv simulations, k and k were obtained by fitting the model to the data from the three monkeys with slower renal clearance.
12
21
M.E. Andersen et al. / Toxicology 227 (2006) 156–164
159
Table 2
Parameters used in the PFOS model for intravenous (iv) or oral simulations
Parameters
Values
Sources
Body weight (BW)
3.5–6.5 kg
Cardiac output (Qcc)
Volume of renal filtrate (Vfil )
15 L/(h kg)
0.01 L
Renal plasma filtration rate (Qfilc )
0.1
Volume of distribution (Vdc )
Transport maximum (TmC )
0.13 L/kg (iv), 0.22 L/kg (oral)
113.3 mg/(kg h) (iv)b , 13.6 mg/(kg h) (oral)
Transport affinity constant (KT )
Proportion of PFOA free in blood
(fplasma )
Compartmental transfer rate constant
from central to tissues (k12 )
Compartmental transfer rate constant
from tissues back to central (k21 )
0.19 mg/L (iv), 0.023 mg/L (oral)
0.042 (iv), 0.025 (oral)
Experimental design (Butenhoff et al.,
2002, 2004)
Corley et al. (1990)
Assumption; the model simulations were
not sensitive to this parameter
Assumption; equals to half of the blood
flow to kidney (20%)a
Optimized
Optimized; average values from TmC that
gives the best fit for individual monkeys
Optimized
Optimized
21.6 h−1 (iv), 3.3 h−1 (oral)
Optimized
0.87 h−1 (iv), 0.1 h−1 (oral)
Optimized
a
No data on monkeys were found. Based on the blood flow rates to kidneys reported for humans (17.5%) and dogs (17.3%) (Brown et al., 1997),
20% was chosen for the monkeys.
b For the iv simulations, only data from female monkeys were evaluated in this study.
each of three male and three female monkeys varied
from 2 to 7 mg/(h kg), giving an estimate of an average value of 4.73 ± 2.51 mg/(h kg). In the monkeys with
faster renal clearances, it was more difficult to unam-
biguously evaluate the inter-compartmental transfer rate
constants (i.e. k12 and k21 ) from curve fitting. Therefore, these two parameters for the three monkeys with
faster clearance were estimated as the averaged k12
Fig. 2. Time course of PFOA in plasma of Cynomolgus monkeys following intravenous dosing at 10 mg PFOA/kg body weight. Data (Butenhoff
et al., 2004) are represented in symbols; simulations are represented in lines. The inset shows the same plot on a linear scale.
160
M.E. Andersen et al. / Toxicology 227 (2006) 156–164
Fig. 3. Time course of PFOA in plasma and urine of Cynomolgus monkeys dosed daily with 20 mg PFOA/kg body weight in a capsule. Data
(Butenhoff et al., 2004) are represented in symbols; simulations are represented in lines. The inset shows the simulations of time course of PFOA
in plasma and urine of Cynomolgus monkeys dosed daily with 0.1 mg PFOA/kg body weight are also shown to represent situations of daily dosing
that do not lead to saturation of the putative transporter.
(3.4 h−1 ) and k21 (0.13 h−1 ) from the three monkeys
with slower overall clearance of PFOA from blood. The
values for the other parameters were: KT = 0.055 mg/L;
Vdc = 0.11 L/kg; fplasma = 0.018; Qfilc = 0.10 (Table 1).
The same model structure and a common set of
parameters, except for TmC , were used to describe the
plasma and urine time course data after cessation of
daily dosing with 20 mg PFOA/kg. The values of TmC
required to fit the plasma and urine time course for
each of the 3 monkeys (shown by symbols in Fig. 3)
were between 4 and 6 mg/(h kg). The averaged TmC of
5 mg/(h kg) was then used to generate the simulated time
course of PFOA in plasma and urine (shown by the
plotted lines in Fig. 3). During the 20 mg/(kg day) oral
dosing regimen with PFOA, the inferred filtrate concentrations saturated resorption and produced very differently shaped elimination curves for plasma and urine
(Fig. 3). Importantly, the simulated plasma time course
accurately captured the rapid approach to steady state at
the higher daily doses and the slower terminal half-life.
The plasma and urine time courses for a daily dosing that
do not lead to saturation of the putative transporter (e.g.
0.1 mg/(kg day)) were also simulated and the urine and
plasma curves were uniformly parallel during and after
exposure (Fig. 3; inset). For these simulations, the value
for parameters other than TmC were optimized after setting TmC to 5 mg/(h kg) (average value from individual
fits): KT = 0.055 mg/L; Vdc = 0.14 L/kg; fplasma = 0.020;
Qfilc = 0.10; k12 = 3.3 h−1 ; k21 = 0.1 h−1 . The ratio Tm /KT
that approximates the PFOA clearance from the renal filtrate back into the plasma compartment was 90.9 L/h.
As with PFOA, the saturable, resorption PK model
was used to investigate the long half-life of PFOS in monkeys after intravenous dosing (Noker and Gorman, 2003)
and oral dosing (Seacat et al., 2002). The intravenous
time course data (single intravenous dose of 2 mg/kg)
indicated that some time and/or dose-dependent changes
occur in distribution of PFOS between the blood and tissue compartments. The time-dependent changes were
less noticeable in female than in male monkeys (Noker
and Gorman, 2003). Therefore, for the present work,
only female data were compared with predictions from
a model structurally similar to that for PFOA, which
did not have a time-dependent component (Fig. 4).
The values for model parameters were: KT = 0.19 mg/L;
TmC = 113.3 mg/(h kg); Vdc = 0.13 L/kg; fplasma = 0.042;
Qfilc = 0.10; k12 = 21.6; k21 = 0.87 (3.8 kg monkeys).
Although the simulation captured the overall time course
behavior, it did not provide good correspondence with
the rapid loss from plasma and the apparent rise in plasma
M.E. Andersen et al. / Toxicology 227 (2006) 156–164
161
doses, 2 and 10 mg PFOS/(kg body weight day), were
also simulated, even though no data to compare with,
to show the different time course behavior at high
dose region, especially at 10 mg/(kg day) (Fig. 5). The
values for model parameters were: KT = 0.023 mg/L;
TmC = 13.6 mg/(h kg); Vdc = 0.22 L/kg; fplasma = 0.025;
Qfilc = 0.10; k12 = 3.3; k21 = 0.1 (6.5 kg monkeys). Timeand dose-dependent changes in distributional parameters
may explain the systematic fitting errors with the averaged TmC values. The 0.15 mg/(kg day) simulation tends
to be uniformly lower; the 0.75 mg/(kg day) simulation
tends to be higher than the data. The 0.75 mg/(kg day)
results are best represented with a TmC of 16.
4. Discussion
Fig. 4. Time course of PFOS in blood of Cynomolgus monkeys following intravenous dosing at 2 mg PFOA/kg body weight. Data (Noker
and Gorman, 2003) are represented in symbols; simulations are represented in lines.
concentrations over the first 20 days (Fig. 4). A more
complete evaluation on alternative model structures is
underway to better characterize the time-dependent time
course data for both male and female monkeys and to
account for the discrepancies in the present description.
These discrepancies may well be associated with control
of synthesis of PFOA/PFOS binding proteins. Genes of
this type are known to be induced by PFOA and PFOS
in rats (Guruge et al., 2006).
Three daily dose rates used in the oral study:
0.75, 0.15 and 0.03 mg PFOS/(kg body weight day)
(Butenhoff et al., 2002) were simulated and compared to the measured data (Fig. 5). Two additional
Fig. 5. Time course of PFOS in blood of Cynomolgus monkeys during
and after cessation of oral dosing at 0.75, 0.15 and 0.03 mg PFOS/(kg
body weight day). Data (Butenhoff et al., 2002) are represented in symbols; simulations are represented in lines. Two additional doses, 2 and
10 mg PFOS/kg body weight, were also simulated.
For convenience, the pharmacokinetics (PK) of PFOA
have been described with one-compartment or simple
first-order models (EPA, 2005; Kemper, 2003); however,
the observed PK are more complicated. During studies with daily oral dosing and extended post-exposure
observation periods, Cynomolgus monkeys have a rapid
approach to steady-state plasma concentrations together
with a very much slower terminal half-life (Butenhoff
et al., 2002). These differences in apparent elimination
rates with increasing dose were clear indications that
capacity-limited, saturable processes must be involved
in the kinetic behavior of these compounds. Kinetic analysis of the data from monkey exposures, together with
mechanistic data on the transporters that are responsible
for the observed kinetics, makes it possible to identify
the key determinants of PFOA/PFOS kinetics, and hence
risk in humans. For example, once the transporter that
is responsible for the retention of PFOA/PFOS at low
exposure levels has been identified, information on the
activity of that transporter in the human can be used to
determine whether there is a potential for susceptible
sub-populations due to polymorphic variation.
With PFOA, the majority of excretion in laboratory
animals is via urine (Butenhoff et al., 2004; Kemper,
2003) and renal processing in rats is likely to involve
transporter proteins. As expected from the half-lives,
urinary excretion in human volunteers is much slower
than in rats (Harada et al., 2005). Regardless of species,
urinary excretion for PFOS is much slower than it is
for PFOA (Harada et al., 2005). The lower elimination rate in humans compared to laboratory animals
appears to be due to reduced renal clearance, which
could reflect differences in renal transport mechanisms.
Compared to PFOA, PFOS had a slower terminal halflife and more rapid approach to steady-state associated
with repeated dose oral administration (Fig. 5). The
162
M.E. Andersen et al. / Toxicology 227 (2006) 156–164
longer-half life of PFOS can be explained through higher
capacity and higher affinity transporter characteristics.
The monkey resorption clearance (Tm /KT ) calculated for
PFOS, 587 L/h, was nearly nine-fold greater than that for
PFOA. With the current model structure, it is straightforward to scale the parameters from 4 to 7 kg monkeys
to a standard 70 kg human, and assess the transporter
characteristics that would lead to any specific human
half-life. If the free fraction in plasma is assumed to be
similar between monkey and human (i.e. 2%), the 4–5year human half-life can be accounted for by doubling
TmC to approximately 25 mg/(h kg).
Transporter proteins may participate in either secretion from blood to the urinary filtrate or in resorption
from the urinary filtrate back into blood. If resorption is
the dominant process involved in kidney processing of
these acids, the slower human urinary clearance would
reflect more efficient renal resorption in humans. An
inverse correlation is reported between levels of mRNA
of oatp1 (Slco1a1), a transporter thought to be responsible for resorption of anions from the renal filtrate, and
renal clearance of PFOA in male rats (Kudo et al., 2002).
Additionally, castration of male rats increases clearance
while decreasing oatp1 mRNA and treatment of castrated rats with testosterone increases oatp1 mRNA and
decreases clearance. These observations indicate a role
for tubular resorption via oatp1 in the longer half-lives
in male rats. But, to our knowledge, the possibility that
PFOA has saturable renal resorption is not given serious
consideration.
This present study provides the first description of
a PK model for a xenobiotic based on saturable, highaffinity renal resorption of a filtered compound. Since
similar high affinity uptake processes exist for renal
resorption of glucose (Shannon and Fisher, 1938) and
amino acids (Roby and Segal, 1989), we expected that
this model structure may become important for other
products of intermediary metabolism, xenobiotics, and
drugs moved by various transporter proteins into and
out of the renal filtrate (Lee and Kim, 2004; Sekine et
al., 2006). Based on the results in rats, further studies
of oatp1 in rodents and other species may be important
to establish the molecular carriers determining reuptake
(Kudo et al., 2002). High affinity long chain fatty acid
transporter proteins (FATPs) have been identified in multiple tissues including kidney (Stahl, 2004) and might
be considered as possible transporters involved in fluoroacid resorption based on the similarities in structures
for fatty acids and these perfluoroacids. In general, the
lower chain length fatty acids (i.e. C-8) are not good substrates for FATPs (Abumrad et al., 1998); nonetheless,
PFOA and PFOS could be examined to see if they com-
pete for transport with longer chain fatty acids. It is also
possible that PFOA and PFOS resorption occurs by a
protein completely different from the conventional renal
transporters. Multi-drug transporters also can transport
fatty acid analogs. EmrD (an E. coli multitransporter)
has a hydrophobic core in the transporter channel and is
known to transport sodium dodecyl sulfate (Yin et al.,
2006).
The successes of our PK model are obviously not,
by themselves, direct proof that renal resorption controls serum elimination half-lives with PFOA and PFOS.
Nonetheless, the behaviors of the model are entirely consistent with (i) the premise that renal resorption explains
the kinetics across species, (ii) the dose-dependencies
with PFOA and PFOS, and (iii) the gender related differences in oatp1 in rats. Furthermore, saturable renal
excretion cannot be a viable explanation, because PK
models based on this process predict higher clearances at
lower concentrations (Sauerhoff et al., 1977), the reverse
of the observed behavior with these acids. Overall, we
believe strongly that the quantitative agreement of the
model simulations with the larger body of kinetics data is
extremely compelling evidence that saturable, high affinity resorption processes govern the kinetics of these acids
in multiple species. Our physiologically-motivated pharmacokinetic model for renal resorption can be applied
to evaluate the broader body of fluoroalkylacids disposition studies in various species and determine which
re-uptake transporters have kinetic properties consistent
with half-lives seen in different species. Also, any future
development of a more complete physiologically-based
pharmacokinetic model should include the renal resorption descriptions presented in the current study.
Acknowledgements
The authors thank several colleagues, Drs. David Dorman, Shelia Collins, Russell Thomas, William Greenlee,
and Joseph Rodricks for careful review and many helpful
comments in the preparation of this manuscript.
Appendix A. Supplementary data
Supplementary data associated with this article
can be found, in the online version, at doi:10.1016/
j.tox.2006.08.004.
References
Abumrad, N., Harmon, C., Ibrahimi, A., 1998. Membrane transport of
long-chain fatty acids: evidence for a facilitated process. J. Lipid
Res. 39, 2309–2318.
M.E. Andersen et al. / Toxicology 227 (2006) 156–164
Biegel, L.B., Hurtt, M.E., Frame, S.R., O’Connor, J.C., Cook, J.C.,
2001. Mechanisms of extrahepatic tumor induction by peroxisome
proliferators in male CD rats. Toxicol. Sci. 60, 44–55.
Brown, R.P., Delp, M.D., Lindstedt, S.L., Rhomberg, L.R., Beliles,
R.P., 1997. Physiological parameter values for physiologically based pharmacokinetic models. Toxicol. Ind. Health 13,
407–484.
Burros, M., 2005. Eating Well: Is There an Extra Ingredient in Nonstick Pans? In The New York Times, New York.
Butenhoff, J., Costa, G., Elcombe, C., Farrar, D., Hansen, K., Iwai,
H., Jung, R., Kennedy Jr., G., Lieder, P., Olsen, G., Thomford, P.,
2002. Toxicity of ammonium perfluorooctanoate in male Cynomolgus monkeys after oral dosing for 6 months. Toxicol. Sci. 69,
244–257.
Butenhoff, J.L., Kennedy Jr., G.L., Hinderliter, P.M., Lieder, P.H., Jung,
R., Hansen, K.J., Gorman, G.S., Noker, P.E., Thomford, P.J., 2004.
Pharmacokinetics of perfluorooctanoate in Cynomolgus monkeys.
Toxicol. Sci. 82, 394–406.
Calafat, A., Kuklenyik, Z., Caudill, S., Reidy, J., Needham, L., 2006.
Perfluorochemicals in pooled serum samples from United States
residents in 2001 and 2002. Environ. Sci. Technol. 40, 2128–
2134.
Corley, R.A., Mendrala, A.L., Smith, F.A., Staats, D.A., Gargas, M.L.,
Conolly, R.B., Andersen, M.E., Reitz, R.H., 1990. Development
of a physiologically based pharmacokinetic model for chloroform.
Toxicol. Appl. Pharmacol. 103, 512–527.
Cortese, A., 2005. Inside the News: Will Environmental Fear Stick to
DuPont’s Teflon? The New York Times, New York.
EPA, U.S., 2005. Draft Risk Assessment of the Potential Human Health
Effects Associated with Exposure to Perfluorooctanoic Acid and
its Salts. U.S. EPA Office of Pollution Prevention and Toxics. Risk
Assessment Division.
Giesy, J., Kannan, K., 2001. Global distribution of perfluorooctane
sulfonate. Environ. Sci. Technol. 35, 1339–1342.
Guruge, K.S., Yeung, L.W.Y., Yamanaka, N., Miyazaki, S., Lam,
P.K.S., Giesy, J.P., Jones, P.D., Yamashita, N., 2006. Gene expression profiles in rat liver treated with perfluorooctanoic acid (PFOA).
Toxicol. Sci. 89, 93–107.
Harada, K., Inoue, K., Morikawa, A., Yoshinaga, T., Saito, N.,
Koizumi, A., 2005. Renal clearance of perfluorooctane sulfonate
and perfluorooctanoate in humans and their species-specific excretion. Environ. Res. 99, 253–261.
Hostetter, T.H., Meyer, T.W., 2004. The development of clearance
methods for measurement of glomerular filtration and tubular reabsorption. Am. J. Physiol. Renal Physiol. 287, F868–F870.
Kannan, K., Corsolini, S., Falandysz, J., Fillmann, G., Kumar, K.S.,
Loganathan, B.G., Mohd, M.A., Olivero, J., Van Wouwe, N., Yang,
J.H., Aldoust, K.M., 2004. Perfluorooctanesulfonate and related
fluorochemicals in human blood from several countries. Environ.
Sci. Technol. 38, 4489–4495.
Kannan, K., Koistinen, J., Beckmen, K., Evans, T., Gorzelany, J.,
Hansen, K., Jones, P., Helle, E., Nyman, M., Giesy, J., 2001.
Accumulation of perfluorooctane sulfonate in marine mammals.
Environ. Sci. Technol., 35.
Kemper, R., 2003. Perfluorooctanoic acid: Toxicokinetics in the rat.
In Project ID: DuPont 7473. USEPA public docket, administrative
record AR-226-1499.
Kudo, N., Kawashima, Y., 2003. Toxicity and toxicokinetics of perfluorooctanoic acid in humans and animals. J. Toxicol. Sci. 28,
49–57.
Kudo, N., Iwase, Y., Okayachi, H., Yamakawa, Y., Kawashima, Y.,
2005. Induction of hepatic peroxisome proliferation by 8-2 telomer
163
alcohol feeding in mice: formation of perfluorooctanoic acid in the
liver. Toxicol. Sci. 86, 231–238.
Kudo, N., Katakura, M., Sato, Y., Kawashima, Y., 2002. Sex hormoneregulated renal transport of perfluorooctanoic acid. Chem. Biol.
Interact. 139, 301–316.
Lau, C., Butenhoff, J.L., Rogers, J.M., 2004. The developmental toxicity of perfluoroalkyl acids and their derivatives. Toxicol. Appl.
Pharmacol. 198, 231–241.
Lau, C., Thibodeaux, J.R., Hanson, R.G., Narotsky, M.G., Rogers,
J.M., Lindstrom, A.B., Strynar, M.J., 2006. Effects of perfluorooctanoic acid exposure during pregnancy in the mouse. Toxicol. Sci.
90, 510–518.
Lee, W., Kim, R., 2004. Transporters and renal drug elimination. Annu.
Rev. Pharmacol. Toxicol. 44, 137–166.
Luebker, D.J., York, R.G., Hansen, K.J., Moore, J.A., Butenhoff, J.L.,
2005. Neonatal mortality from in utero exposure to perfluorooctanesulfonate (PFOS) in Sprague-Dawley rats: dose–response, and
biochemical and pharamacokinetic parameters. Toxicology 215,
149–169.
Maloney, E.K., Waxman, D.J., 1999. Trans-activation of PPARalpha
and PPARgamma by structurally diverse environmental chemicals.
Toxicol. Appl. Pharmacol. 161, 209–218.
Noker, P., Gorman, G., 2003. A pharmacokinetic study of potassium
perfluorooctanesulfonate in the Cynomolgus monkey. In Southern Research Institute Study ID: 9921.6. USEPA public docket,
administrative record AR-226-1356.
Olsen, G., Church, T.R., Hansen, K., Burris, J.M., Butenhoff, J., Mandel, J.H., Zobel, L.R., 2004a. Quantitative evaluation of perfluorooctanesulfonate (PFOS) and other fluorochemicals in the serum
of children. J. Children’s Health 2, 53–76.
Olsen, G., Ehresman, D., Froehlich, J., Burris, J.M., Butenhoff, J.,
2005. Evaluation of the half-life (T1/2) of elimination of perfluorooctanesulfonate (PFOS), perfluorohexanesulfonate (PFHS),
and perfluorooctanonate (PFOA) from human serum. http://www.
chem.utoronto.ca/symposium/fluoros/pdfs/
TOX017Olsen.pdf. Access date: 26 March 2006.
Olsen, G.W., Church, T.R., Larson, E.B., van Belle, G., Lundberg,
J.K., Hansen, K.J., Burris, J.M., Mandel, J.H., Zobel, L.R., 2004b.
Serum concentrations of perfluorooctanesulfonate and other fluorochemicals in an elderly population from Seattle, Washington.
Chemosphere 54, 1599–1611.
Olsen, G.W., Hansen, K.J., Stevenson, L.A., Burris, J.M., Mandel, J.H.,
2003. Human donor liver and serum concentrations of perfluorooctanesulfonate and other perfluorochemicals. Environ. Sci. Technol.
37, 888–891.
Roby, K.A., Segal, S., 1989. Cystine and lysine reabsorption in
the isolated perfused rat kidney. Am. J. Physiol. 256, F901–
F908.
Sauerhoff, M.W., Braun, W.H., LeBeau, J.E., 1977. Dose-dependent
pharmacokinetic profile of silvex following intravenous administration in rats. J. Toxicol. Environ. Health 2, 605–618.
Seacat, A.M., Thomford, P.J., Hansen, K.J., Olsen, G.W., Case, M.T.,
Butenhoff, J.L., 2002. Subchronic toxicity studies on perfluorooctanesulfonate potassium salt in Cynomolgus monkeys. Toxicol.
Sci. 68, 249–264.
Sekine, T., Miyazaki, H., Endou, H., 2006. Molecular physiology of
renal organic anion transporters. Am. J. Physiol. Renal Physiol.
290, F251–F261.
Shannon, J., Fisher, S., 1938. The renal tubular reabsorption of glucose
in the normal dog. Am. J. Physiol. 122, 765–774.
Shipley, J.M., Hurst, C.H., Tanaka, S.S., DeRoos, F.L., Butenhoff, J.L.,
Seacat, A.M., Waxman, D.J., 2004. Trans-activation of PPARalpha
164
M.E. Andersen et al. / Toxicology 227 (2006) 156–164
and induction of PPARalpha target genes by perfluorooctane-based
chemicals. Toxicol. Sci. 80, 151–160.
Smithwick, M., Muir, D.C., Mabury, S.A., Solomon, K.R., Martin,
J.W., Sonne, C., Born, E.W., Letcher, R.J., Dietz, R., 2005. Perflouroalkyl contaminants in liver tissue from East Greenland polar
bears (Ursus maritimus). Environ. Toxicol. Chem. 24, 981–986.
Stahl, A., 2004. A current review of fatty acid transport proteins
(SLC27). Pflug. Arch. 447, 722–727.
Xu, L., Krenitsky, D.M., Seacat, A.M., Butenhoff, J.L., Anders,
M.W., 2004. Biotransformation of N-ethyl-N-(2-hydroxyethyl)perfluorooctanesulfonamide by rat liver microsomes, cytosol, and
slices and by expressed rat and human cytochromes P450. Chem.
Res. Toxicol. 17, 767–775.
Yin, Y., He, X., Szewczyk, P., Nguyen, T., Chang, G., 2006. Structure
of the multidrug transporter EmrD from Escherichia coli. Science
312, 741–744.