DEQ-CFW_00000420

Toxicology 340 (2016) 1-9
t Contents lists available at ScienceDirect '
10NIC0109
Toxicology
ELSEVIER journal homepage: www.elsevier.com/locate/toxicol
Absorption, distribution, metabolism, excretion, and kinetics of 2,3,3,3- CmssMark
tetrafluoro-2-(heptafluoropropoxy)propanoic acid ammonium salt
following a single dose in rat, mouse, and cynomolgus monkey
Shawn A. Gannon", William J. Fasanob, Michael R Mawnc, Diane L. Nabbb,
Robert C. Bucka, L. William Buxtona, Gary W. Jepsona, Steven R. Frameb
'The Chemours Company, Wilmington, DE, USA
bE L duPont de Nemours and Company, Haskell Global Centers for Health Fr Environmental Sciences, Newark, DE 19714, USA
`E L DuPont de Nemours and Company, Wilmington, DE, USA
ARTICLE INFO ABSTRACT
Article history:
Ammonium, 2,3,3,3-tetrafluoro-2-(heptafluoropropoxy)-propanoate has been developed as a processing
Received 23 September 2015
aid used in the manufacture of fluoropolymers. The absorption, distribution, elimination, and
Received in revised form 14 December 2015
distribution (ADME) and kinetic behavior of this substance has been evaluated in rats, mice, and
Accepted 23 December 2015
cynomolgus monkeys by oral and intravenous routes of exposure and studied in both plasma and urine.
Available online 29 December 2015
The test substance is rapidly and completely absorbed in both rats and mice and both in vivo and in vitro
experiments indicate that it is not metabolized. The test substance is rapidly eliminated exclusively in the
Keywords:
urine in both rats and mice, with rats eliminating it more quickly than mice (approximately 5 h
Fluoropolymer
elimination half-life in rats, 20 h half-life in mice). Pharmacokinetic analysis in monkeys, rats, and mice
ADME
Rodent
indicate rapid, biphasic elimination characterized by a very fast alpha phase and a slower beta phase. The
Monkey
beta phase does not contribute to potential accumulation after multiple dosing in rats or monkeys.
Pharmacokinetics
Comparative pharmacokinetics in rats, mice, and monkeys indicates that the rat is more similar to the
Toxicokinetics
monkey and is therefore a more appropriate rodent model for pharmacokinetics in primates.
© 2015 Elsevier Ireland Ltd. All rights reserved.
1. Introduction
The widespread presence of long -chain perfluoroalkyl acids
(PFAAs) such as perfluorooctane sulfonate (PFOS) and perfluor-
ooctanoic acid (PFOA) has spurred a move to alternative
fluorinated substances which have more favorable environmental
and biological properties, most notably rapid elimination from
living systems (Ritter, 2010; Buck et al., 2011; US EPA, 2006). Per -
and poly -fluorinated ether carboxylates have been developed as
alternative polymer processing aids for the aqueous emulsion
polymerization of tetrafluoroethylene (TFE) and other fluorinated
monomers in the synthesis of fluoropolymers (Buck, 2015; Buck
et al., 2011; Gordon, 2011). They have replaced ammonium
perfluorooctanoate which was historically used for this purpose
(Feiring, 1994).
* Corresponding author at: The Chemours Company, 1007 Market Street,
Wilmington, DE,19898, USA.
E-mail address: [e-mail] (SA. Gannon).
http://dx,doi.org/10,1016/j.tox.2015.12.006
0300-483X/© 2015 Elsevier Ireland Ltd. All rights reserved.
Ammonium, 2,3,3,3-tetrafluoro-2-(heptafluoropropoxy)-propa-
noate has been developed as a processing aid used in the
manufacture of fluoropolymers. The acid form is a liquid and the
ammonium salt is a solid at ambient temperature (20 oQ. Both are
infinitely soluble in water above the pKa (2.84) of the acid. An
aqueous solution is made and used in the fluoropolymer
manufacturing process. The processing aid is either captured for
re -use or thermally destroyed during fluoropolymer processing
(Brothers et al., 2009). Some fluoropolymer aqueous dispersions
that contain the processing aid are used to coat surfaces such as
metal for making non-stick cookwear. The fluoropolymer is sintered
onto the substrate surface at temperatures >265 " C. As shown in
Table 1, the processing aid decomposes at 150-160°C.
The objective of the present study was to determine the
absorption, distribution, metabolism and elimination (ADME)
profile following oral and/or intravenous dosing of ammonium,
2,3,3,3-tetrafluoro-2-(heptafluoropropoxy)-propanoate, [Fig. 1,
CF3CF2CF20CF(CF3)COOH•NH3, CAS# 62037-80-3] in the rat,
mouse and cynomolgus monkey. The chronic toxicology and
carcinogenicity of this test substance was recently described (Rae
et al., 2015), as was the aquatic toxicology (Hoke et al., 2015).
DEQ-CFW 00000420

S.A. Gannon et al./Toxicology 340 (2016) 1-9
Forty-five male and forty-five female Cr1:CD1 (ICR) mice per
dose level were dosed orally with 10 or 30 mg/kg of the test
substance in water with a dose volume of 4 mL/kg. Blood was
collected via serial sacrifice (n=3 mice per time point) prior to
dosing and at 0.25, 0.5, 1, 2, 4, 8, 12, and 24 h after dosing.
Additional blood samples were collected once daily on test days 2-
7. Samples were centrifuged to generate plasma and then frozen at
-20 °C until analyzed by HPLC-MS.
2.5. Rat and cynomolgus monkey intravenous plasma
pharmacokinetics
A total of 6CrI:CD SD rats (3 males and 3 females) per dose level
were assigned to the study. The animals were fasted overnight
prior to dosing and through the first 2 h of blood collection. The low
dose group received a single 10 mg/kg intravenous (via tail vein)
bolus of the test substance formulated in sterile phosphate
buffered saline, pH 7.6 at a dose volume of 1 mL/kg. The high
dose group received a single 50 mg/kg dose of the test substance
formulated in the same manner as the low dose. Blood
(approximately 0.1 mL per sample) was collected from the tail
vein prior to dosing and at 0.083, 0.25, 0.5, 1, 2, 4, 8, 12, and 24 h
after dosing. Additional blood samples were collected once daily on
test days 2-7. Samples were centrifuged to generate plasma and
then and then frozen at -20°C until analyzed by HPLC-MS.
A total of 6 non -naive cynomolgus monkeys (3 male and
3 female) were assigned to the study. The animals were fasted
overnight prior to dosing and through the first 4 h of blood sample
collection. All monkeys received a single 10 mg/kg intravenous
bolus of the test substance delivered via a peripheral vein. The dose
was formulated in sterile phosphate buffered saline, pH 7.6 at a
dose volume of 2 mL/kg. Blood (approximately 0.5 mL per sample)
was collected from the femoral vessel prior to dosing and at 0.083,
0.167, 0.25, 0.5, 1, 2, 4, 8, 12, and 24 h post dose. Additional blood
samples were collected once daily on test days 3-21. Samples were
centrifuged to generate plasma and then frozen at -20 °C and until
analyzed by HPLC-MS.
2.6. Absorption, distribution, and elimination in the rat and mouse
The conduct of this study was designed to comply with the Tier
1 requirements of U.S. EPA, OPPTS 870.7485-Metabolism and
Pharmacokinetics, Health Effects Test Guidelines (1998). Because
the test substance was not metabolized in rat hepatocytes nor were
any metabolites observed in other studies, a radiolabeled form of
the test substance was not required. Rodents were housed
individually in glass metabolism units during the study. Rats
were fasted overnight and through the first 2 h of blood collection.
Mice were not fasted. Five rodents of each sex and species received
the test substance orally at 3 (mice) or 30 (rats) mg/kg. Rodents
were returned to individual metabolism units following dosing.
Urine and feces were collected on dry ice for 24 h prior to dosing,
and then at 0-6 h, 6-12 h,12-24 h, and every 24 h until 168 h after
dosing. At the end of the experiment (168h post dose), rodents
were sacrificed by CO2 asphyxiation followed by exsanguination.
The following tissues were collected: liver, fat, gastrointestinal
tract, gastrointestinal tract contents, kidney, spleen, and whole
blood. The carcass was retained. Tissues and carcass were saved for
analysis in the event that the dose was not recovered in the excreta.
Individual cages were rinsed with deionized water.
To determine if metabolism had occurred, mass spectra data
from the urine samples were screened for suspected metabolites
manually and automatically for unexpected metabolites using the
IntelliExtractTm (ACD/Labs, Toronto, Ontario, Canada) control -
comparison data processing tool.
2.7. Sample processing and analysis
2.7.1. HPLC-MS analysis of urine and plasma
The urine samples were prepared for analysis by pipetting
950 µL of 15% acetonitrile in HPLC grade water solvent and 50 µL of
sample. The samples were then mixed with a vortex mixer for
1 min prior to analysis. As necessary, additional sample dilutions
were performed using the 15% acetonitrile in HPLC grade water
solvent to ensure that the sample responses were within the
calibration curve.
The plasma samples were prepared for analysis by pipetting
150 µL acetonitrile into a 1.7 mL microcentrifuge tube, and
pipetting 50 µL of sample. The sample tubes were then vortexed
for 1 min and centrifuged at 14,000 RCF for 30 min. After
centrifugation, 100 µL of sample supernatant was placed into a
HPLC vial and 400 µL of HPLC grade water was added and mixed.
As necessary, additional sample dilutions were performed using
the 15% acetonitrile in HPLC grade water solvent to ensure that the
sample responses were within the calibration curve. An internal
standard (13C-PFOA) was added to all prepared samples and
standards at a concentration of 5 ng/mL. The LOQ was established
from the lowest calibration standard at 1 ng/mL multiplied by the
sample dilution factor. The sample dilution factor for all plasma
samples was 20x which corresponds to an LOQ of 20 ng/mL. The
lowest calibration standard had a response of at least 5 times the
blank.
Samples were chromatographically separated on an Agilent
Model 1200HPLC using a Zorbax® RX-C18 2.1 x 150 mm ID,
5 micron particle size and analyzed using an Applied Biosystems
API 4000 mass spectrometer equipped with a Turbo Spray ion
source set to negative mode. Mobile phase A was 0.15% acetic acid
in water and mobile phase B was 0.15% acetic acid in acetonitrile.
The separation was isocratic at 70:30A:B. The separation was
followed by a 2.5 min wash at 100% B and equilibration at initial
conditions for 2.5 min. The flow rate was 600 µL. The mass
spectrometer mode was MRM monitoring the test substance
(329.1 - 284.9 m/z) and the internal standard (415 370 m/z) with
a dwell time of 150 ms.
The urine samples were analyzed using the method described
above but with a Phenomenex® Luna C8 150 x 2 mm column,
5 micron particle size and the addition of 0.15% triethylamine to the
mobile phase. An internal standard was not used for the urine
analysis or the rat single oral dose experiment. The plasma samples
from the rat single oral dose experiment were analyzed using
essentially the same method but utilizing a Micromass Quattro
Micro mass spectrometer using an Agilent Model 1100HPLC.
2.7.2. Pharmacokinetic calculations
Pharmacokinetic parameters were calculated using PK Plus, a
module of GastroPlus (Simulations Plus, Lancaster, California,
USA). The data was entered and the goodness -of -fit was calculated
for both one and two compartment models. In all cases the two -
compartment model provided a better fit to the observed data. The
absorption rate constant, the alpha elimination rate constant, the
beta elimination rate constant, and the volumes of distribution of
both the central and peripheral compartment were calculated for
the two -compartment model. These pharmacokinetic constants
were also used to model the effects of repeated exposure using
GastroPlus.
3. Results
Analyzed concentrations of the test substance are reported as
nanograms of the solubilized anion 2,3,3,3-tetrafluoro-2-(hepta-
fluoropropoxy)-propanoate per milliliter of plasma or urine. All
data are presented as mean and standard deviation.
DEQ-CFW 00000421

S.A. Gannon et al./Toxicology 340 (2016) 1-9
5
1,000,000
100,000
10,000
1,000
100
J 10
E
5 1
c
O
cti 1,000,000
c
U
U100,000
10,000
1,000
100
10
1
Rat, IV
Monkey, IV
Rat, Oral
Mouse, Oral
0 48 96 144 192 240 288 336 384 0 48 96 144 192 240 288 336 384
Time (hr)
Fig. 2. Concentration of the test substance (ngimL) in plasma from rats following a 10 mg/kg intravenous dose; rats following a 10 mg/kg oral dose; cynomolgus monkeys
following a 10 mgikg intravenous dose; and mice following a 10 mg/kg oral dose. The last blood collection time was 168 h in rodents and 504 h (21 days) in monkey. The
analyte was not detected after the 384 h (16 day) collection in monkey. Data points without error bars indicate that the analyte was detected in less than three of the individual
samples at that time point. In the mouse dataset, the standard deviation was approximately equal to the average for the 96 h female and the 168 h male point. These points are
labeled with a horizontal bar (-) above the symbol. The limit of detection was 20 ng/mL for oral data sets and 1 ng/mL for the intravenous data sets.
Table 3
Material balance of the test substance following oral dosing.
Rat (30 mg/kg) Mouse (3 mg/kg)
Male
Female
Male
Female
Mean (%)
SD
Mean
SD
Mean
SD
Mean
SD
(%)
(%)
(%)
(%)
(%)
(%)
(%)
Urine
103
2.7
100
6.4
90
6.9
92
6.0
Feces
1
1.0
1
0.6
2
1.0
2
0.6
Cage wash
1
0.5
5
5.1
10
4.0
6
3.2
Total
105
2.2
106
1.4
101
3.2
99
3.2
screened for suspected metabolites manually and automatically
for unexpected metabolites using the lntelliExtracJm (ACD/Labs,
Toronto, Ontario, Canada) control -comparison data processing
tool. In all cases, there was no evidence of metabolism observed in
any of the samples by either method. Only the anionic form of the
test substance was detected. This finding, taken with near
complete recovery of the dose in urine confirms that the test
substance was rapidly absorbed and eliminated without metabo-
lism in urine following oral dosing in rodents.
4. Discussion
The kinetic behavior of the test substance has been evaluated in
multiple species by oral and intravenous routes of exposure and
repeated measurements in both plasma and urine. The test
substance is rapidly and completely absorbed as evidenced by the
recovery of 100% of the dose in the urine of both rats and mice over
the range of oral doses tested (3 to 30 mg/kg). In vitro hepatocyte
metabolism indicates that this test material is not metabolized,
DEQ-CFW 00000422

S.A. Gannon et al./Toxicology 340 (2016) 1-9
and this is further supported by analysis of urine from rats and
mice in which no metabolites were detected. This was expected
since structurally similar perfluorinated carboxylic acids like
perfluorohexanoic and perfluorooctanoic acid are also not
metabolized (Gannon et al., 2011; Kennedy et al., 2004). Since
this resistance to metabolism is intrinsic to the strength of the
carbon -fluorine bond (Key et al., 1997), it is doubtful that
metabolism if this material would occur in any mammal, including
humans.
The volume of distribution in the central compartment is low
(less than 0.2 L/kg) in rats, mice, and monkeys, indicating that the
test substance is most likely confined primarily to the blood
volume and does not preferentially partition into the tissues. This
is also supported by the observation that total body clearance
occurred rapidly in rats and mice. In rats, both plasma kinetics and
urinary elimination kinetics indicate that the nearly the entire dose
is eliminated within 12-24 h. Because the urinary elimination rate
is very rapid, the slight sex difference observed in the plasma
kinetics was not readily apparent in the urine kinetics. The
elimination rate is slower in mice than in rats and this is observed
in both the plasma kinetics and the urinary elimination kinetics.
The pharmacokinetics following an intravenous dose in rats or
monkeys also shows rapid elimination, consistent with the data
presented above. Blood samples from monkeys were collected for
21 days following dosing, but the test substance was not detected
after Day 16. This lack of detection after Day 16 (384 h) is consistent
with the average calculated beta elimination half-life of —72 h
since at 384 h, between 5 and 6 half-lives have passed and only 1-
40
30
20
10
J
75
c
050
io
c
m
25
c
0
U
50
40
30
20
10
3% of the initial beta phase plasma concentration is expected to
remain.
The pharmacokinetics of this test material differ between rats
and mice, the two most commonly used animal models in
toxicology. To determine which rodent species might better model
the pharmacokinetics in primates, including humans, the phar-
macokinetics of the test substance was also examined in
cynomolgous monkey. The results described above indicate that
the pharmacokinetics in rats are more similar to the monkey. The
mouse shows somewhat slower alpha and beta elimination phases
than the rat or monkey, and the two compartment nature of the
pharmacokinetic curve in rats and monkey is less apparent in
mouse. Because of these findings, the rat is a more representative
model for pharmacokinetics in primates than the mouse.
Although the data shows a biphasic elimination, the alpha
phase elimination is so fast that the beta phase is negligible and
does not contribute to accumulation (Toutain and Bousquet-
Melou, 2004) in either the rat or monkey. In both of these species,
the concentration is typically 1000 times less than the Cn,dX by 24 h
(Fig. 2). When the pharmacokinetics of repeated dosing is
modeled, the steady state plasma concentration is achieved
following the first dose. In contrast, the slower elimination rate
in the mouse means that the mouse requires approximately
4 consecutive doses before reaching steady state in plasma (Fig. 5).
The physiological reason for the biphasic elimination is not known.
Some perfluorinated carboxylic acids were shown to be substrates
for organic anion transporters, a class of transporters responsible
for renal reabsorption (Yang et al., 2010). A mechanism by which
3 24 48 72 96 120 144 168
Time (hr)
3
0
x
m
Fig. 5. 2-compartment model of seven consecutive oral doses of 10 mg/kglday dosing in rats, mice, and monkey.
DEQ-CFW 00000423

S.A. Gannon et al./Toxicology 340 (2016) 1-9
Lau, C., Anitole, K., Hodes, C., Lai, D., Pfahles-Hutchens, A., Seed, J., 2007.
Perfluoroalkyl acids: a review of monitoring and toxicological findings. Toxicol.
Sci. 99, 366-394,
Lou, L, Wambaugh, J.F., Lau, C., Hanson, R.G., Lindstrom, A.B., Strynar, M.J., Zehr, R.D.,
Setzer, R.W., Barton, H.A., 2009. Modeling single and repeated dose
pharmacokinetics of PFOA in mice. Toxicol. Sci. 107, 331-341.
Olsen, G.W., Burris, J.M., Ehresman, D.J., Froehlich, J.W., Seacat, A.M., Butenhof'f, J.L.,
Zobel, L.R., 2007. Half-life of serum elimination of perfluorooctanesulfonate,
perfluorohexanesulfonate, and perfluorooctanoate in retired fluorochemical
production workers. Environ. Health Perspect. 115, 1298-1305.
Rae, J.M., Craig, L., Slone, T.W., Frame, S.R., Buxton, L.W., Kennedy, G.L., 2015.
Evaluation of chronic toxicity and carcinogenicity of ammonium 2,3,3,3-
tetra fluoro-2-(heptafluoropropoxy)-propanoate in Sprague-Dawley rats.
Toxicol. Rep. 2, 939-949.
Ritter, S.K., 2010, Fluorochemicals go short. Chem. Eng. News 88, 12-17.
Seglen, P.O.,1976. Preparation of isolated rat liver cells. Methods Cell Biol. 13,29-83.
Toutain, P.L., Bousquet-Melou, A., 2004. Plasma terminal half-life. J, Vet. Pharm.
Ther. 27, 427-439.
U.S. Environmental Protection Agency (EPA), 2006. The 2010/2015 PFOA
Stewardship Program. EPA-HQ-2003-0012-1071, see EPAwebsite: http://www.
epa.gov/assessing-and-managing-chemicals-under-tsca/20102015-pfoa-
stewards h ip-p rograrn.
Yang, C.-H., Glover, K.P., Han, X., 2010, Characterization of cellular update of
perfluorocatanoate via organic anion -transporting polypeptide 1A2, organic
anion transporter 4, and mate transporter 1 for their potential roles in
mediating human renal reabsorption of perfluorocarboxylates. Toxicol. Sci. 117
(2), 294-302.
DEQ-CFW 00000424

Table 2-2. Uncertainty/safety factors for various reference values
Reference value
UFa
FQPAb
UA
UH
UL
UD
ARE
1, 3, 10
1, 3, 10
1, 3, 10
ND
NA
AEGL
1, 3, 10
1, 3, 10
Y
ND'
NA
OPP acute and
intermediate RfDs
10
10
3,10
NDe
10+
OW HAs
1, 3, 10
1, 3, 10
1, 3, 10
case -specific
NA
ATSDR MRLs
1, 3, 10
1, 3, 10
1, 3, 10
ND
NA
a Uncertainty factors: UA = animal -to -human; U1, = within -human variability;
UL = LOAEL-to-NOAEL; UD = database deficiency.
b Additional safety factor required under FQPA.
Endpoint = lethality, not really a LOAEL-to-NOAEL adjustment in this case.
d Database deficiencies considered, and a factor may be included for intermediate Rf )s if, for
example, there is no reproduction and fertility study.
Overlaps with the FQPA safety factor (see U.S. EPA, 2002b)
ND = not done
NA = not applicable
2-12
DEQ-CFW 00000425