DEQ-CFW_00004192

Tax co. gy and Applied Phannacology 250 (2011) €08-116
In vitro evaluation of the immunotoxic potential of p rfluorinated compounds (PFCs)
Errlarllaela COrsirli a'*, Anna Avogadro', Valentina Galbiati', Mario dell`A,gli b, Marina Marinovich
Corrado L. Gallia, Dori R. Germolec
a €_aboratory ofoiogy, neparirnent of Pharmacological Sciences, Clniversrrd degli Studi di Milano. Via Ral7aretti 9, 20133 Milano, Italy
Laboratory of Phoramognosy. Department of Pharmacological Sciences, University degli Stirdi di Milano,
Via Bolzareiti 9, 20133 Milano, Italy
National Toxicology Program, National Institute of Environmental Health Sciences, NIH, kTP, NC, USA
A R T I C L E I N F® A B S T R A C T
Article history: There is evidence frorn both epidemiology and laboratory studies that perfluorinated compounds may be
Received 27 September 2010 immunotoxic, affecting both cell -mediated and humoral immunity, The overall goal of this study was to
Revised 2 November 2010 investigate the mechanisms underlying the immunotoxic effects of perfluorooctane sulfonate (PFOS) and
Accepted 5 November 2010 erfluorooct ine acid (PFOA), using in vitro assays, The release of the pro-inflamrnator tokines IL-6, IL-8,
Avai:aEPle online 12 November 2010 p` " " �'" p �' �
and'rNF-y ",vas evaluated in lipolysaccharide (LPS)-stimulated human peripheral blood leukocytes and in the
Keywords: human prornyelocytic cell tine THP-1, while the release of IL-4, 11. 10 and IFN- / was evaluated in
Pe -fluorinated compounds phytohaemagglutinin (PHA) -stimulated peripheral blood leukocytes, PFOA and PFOS suppressed LPS-induced
Irnmunosunpression TNF-cx production in primary hurnan cultures and THP-I cells, while IL-8 was suppressed only if] THP-I cells,
PPAF-u receptor II.-6 release was decreased only by PFOS. Both PFOA and PFOS decreased T-cell derived, PHA -induced IL-4 and
cytoldne IL-10 release, while IFN-- -v release was affected only by PFOS. In all instances, PFOS was more potent than
LIMP-9 PFOA. Mechanistic investigations carried out in THP-1 cells demonstrated that the effect on cytokine release
Whole blood assay "Alas pre -transcriptional, as assessed by a reduction in LIPS -induced TNF-cx rnRNA expression. Using siRNA, a
role for PPAR-cv, could be demonstrated for PFOA-induced €mmunotoxictty, while an inhibitory effect on LPS-
induced I- r.B degradation could explain the immunomodulatory effect of PFOS. The dissimilar role of PPAR-ot
in PFOA and PFiOS-induced imrntrnotoxicity was consistent with the differing effects observed off LPS-induced
MMI P-9 release: PFOA, as the PPAR-(x agonist f'enofihrate, modulated the release, white PFOS had no effect,
Overall, these studies suggest that PFCs directly suppress cytokine secretion by immune cells, and that PFOA
and PFOS have different mechanisms of action.
® 2010 Elsevier Inc. All rights reserved.
Introduction
Perfluorinated compounds (PFCs), such as perfluorooctane sulfo-
nate 0"FOS), and perfluorooctane acid (PFOA), are an emerging class of
environmental contaminants commonly detected in blood samples of
both wildlife and humans (Stija et al., 2009; Rayne and Forest. 2009),
arising mainly from their use as surface treatment chemicals,
polymerization aids, and surfactants. Although the production of
PFOS and PFOA was phased out voluntarily beginning in 2000 by its
primary manufacturer, both PFOS and PFOA are persistent environ-
mental contaminants, These compounds are widely present in
surface, ground, marine, and drinking waters at concentrations that
vary from pg/I to Itg/L Some wastewaters contain PFCs at mg/I to low
g/I levels (Rayne and Forest, 2009). Blood samples of occupationally
exposed individuals and the general population in various countries
Abbreviations: PFCs, periluormated compounds; PFOA, pentadecailurooctanoic acid;
PFOA, perfiaorooctanoicsulfonamide; TNF-a, Clamor necrosis factor -a; IT interleuldnn
Corresponding author. Fax: [phone],
E-"nail address: erna�E. Co sini).
0041-008X 6 - see front matter K,'2010 Elsevier Inc. Al' rights reserved.
doi:10.1016; i.taap 2010.11.004
Were found to contain PFOS and PFOA at measurable levels, In the
United States, the mean serum concentration in the general
population was reported as 203 ng/ml for PFOS and as 17 ng/ml for
PFOA (Calafat et al., 2006). PFOA serum concentrations reported in
occupationally exposed humans were between 428 and 12,000 ng/ml
(Steenland et al., 2010). and between 145 and 8490 ng/ml for PFOS
(Olsen et al., 2007),
The health effects of perfluoroalkyl -compounds in humans remain
controversial, in spite of a number of experimental and epidemiolog-
ical studies "Butenhoff et al., 2004; Emmett et al., 2006; Steenland
et al., 2010). Data from experimental animals demonstrated that the
perfluorinated alkyl acids induce peroxisomal proliferation, hepato--
megaly, altered steroidogene sis, and body weight loss associated with
a wasting syndrome (Kennedy et al., 2004; Pastoor et al., 1987; Liu
et al., 1996b; Olsen et al., 1999; Biegel et al., 2001; Luau et aL, 2004).
Ammonium perfluorooctanoate, the ammonium salt of PFOA, pro-
duces increased aromatase activity and plasma estradiol levels, and
induces Leydig cell adenomas (Liu et al„ 1996a,b). Several studies also
indicated that PFOA suppressed antibody production, caused thymic
and splenic atrophy, and altered T-cell populations (fang et al„ 2001,
DEQ-CFW 00004192

E. Corsini et al. Toxicolog I
y and Applied Phorinacolfogl 250 i201 1) 108— 116
109
2002a,b; DeWitt et al„ 2008). Recent studies have also shown that
PFOS affects antibody production in the rodent immune system at
levels found in the general human population (Peden -Adams et al.,
2008). PFOS exposure suppressed immunity in mice resulting in a
significant increase in emaciation and mortality in response to
influenza A virus (Guruge et al., 2009), The pesticide sulfluramid,
which is rapidly metabolized to PFOS, has been demonstrated in mice
to target T-dependent, IgM antibody production at exposure levels
10-fold less than that observed with overt toxicity (Peden -Adams et
al_ 2007), further confirming the immune system as a sensitive target
of PFCs' toxicity,
The peroxisome pro li fera tor -activated receptors (PPAR) belong to
the nuclear hormone receptor superfamily, and there are three
primary subtypes: PPAR u, P, and -y, These receptors regulate
important physiological processes that impact lipid homeostasis,
inflammation, adipogeriesis, reproduction, wound healing, and carci-
nogenesis (Chinetti et al., 2000). Studies suggest that some of the
biological effects of the PFCs are mediated through PPAR and because
PFOA and PFOS both activate PPARcy, the role of I'll"ARut in PFOA and
PFOS immunotoxicity has been investigated. However, the specific
role of 11PARs in PFCs--immunotoxicity is still a matter of debate, and it
is unclear whether or not there is a direct affect on immune cells.
Some data suggest that PPAR(Y, mediates many processes connected
with the immune system in an indirect fashion, by modulating lipid
levels leading to hepatotoxiciry and stress effects (Qazi et al., 2009),
Yang et al, (2002a) compared the ininiunorriodulating effects of PFOA
in wild -type and PPARa null mice: the reductions in spleen weight
and cellularity, in thymus weight and cellularity. and in mito-en--
induced lymphocyte proliferation caused by PFOA in wild -type mice
were not observed in 111--IAR(x null mice. indicating that I"PARcy, plays a
major role in the immunomodulation caused by PFOA. In contrast, as
demonstrated by Qazi et al, (2009), the immunotoxicitY of PFOS was
only partially dependent upon PPAR(Y, activation: the reduction in
thymus weight and in total number of thymocytes was only partially
attenuated in PPARa-riull animals, In vitro studies have demonstrated
that PFOA is a more potent agonist of marine I1PAR(x than PFOS
(rakacs and Abbott, 2007), which may explain the differences in the
immune responses in vivo. Effects independent of PPARcy, activation
may have greater potential impact on humans exposed to PFCs as
human hepatic PPAR(Y expression is only one -tenth that of rodents
,'Kennedy et al., 2004". Furthermore, the affinity of PFOA and PFOS for
iiI.11-flan Or marine PPARs is quite different, In vitro studies conducted
on Cos-1 cells transfected with mouse or human PPAR ot, 0/6, or -y'
reporter plasmids by Takacs and Abbott (2007) demonstrated that
PFOA (0.5 --100 l&I) significantly increased mouse and human 131"ARut
and mouse PPARF)/C) activity relative to vehicle, Whereas PFOS (1-
250 �M) significantly increased activation of mouse 1313ARoL and PPAR
f/() isoforms, no significant activation of mouse or human PPAR � was
observed with PFOA or PFOS. The mouse I1PAR(x appears to be more
sensitive to PFCs than the human PPAR(Y,, with PFOA having more
activity than PFOS with both the mouse and human PPAR isofornis,
further supporting the hypothesis of a different role for 11PARcX in
PFOA and PFOS toxicities,
The purpose of this study was to investigate whether PFCs can
directly affect immune cell function and to characterize the molecular
mechanisms underlying such effects.
Materials and methods
Cheniicals. Perfluorooctane sulfonate (PFOS;
PFOS: CAS# 1763--23-1; Lot#
T20G; Matrix Scientific)) and perflUOMOCtanOiC acid (PFOA; CAS# 335-
67-1; Lot# 02702EE; Sigma -Aldrich) were supplied through a
contract with the National Toxicology Program (Batelle, Columbus,
CH, USA), Chemical identity, purity, and stability were confirmed at
Batelle by infrared spectroscopy and nuclear magnetic resonance
(NMR) spectroscopy, Li pop olvsaccha ride from Escherichia coli sero--
type 0127:B8, antibodies against fisB arid 0-actin and all cell culture
reagents were from Sigma (St Louis, MO, USA), Antibodies against
phosphorylated-p65 were from Santa Cruz Biotechnology (Santa
Cruz, CA, USA), Phytohaemagglutinm (PHA) was from Invitrogen
(Paisley, UK), Reagents and primers for real tirne-PCR were from
Applied Biosystems (Foster City, CA, USA). Electrophoresis reagents
were from Bio-Rad (Richmond, CA, USA). All reagents were purchased
at the highest purity available.
Cells, For all experiments using the human prorriVelocytic cell line
THl_-1-1 (Istituto Zooprofilattico di Brescia, Brescia, Italy), cells were
diluted to 106 cells/ml in RPMI 1640 containing 2 111M L-&Itarnine,
0.1 mg/ml streptomycin. 1001U/ml penicillin, 50pM 2--mercapto--
ethanol, supplemented with 10% heated -inactivated fetal calf serum
(media) and cultured in 37 'C in 5% CO2 incubator. ForTNF-(X and IL-
8 releas'eO,5 x 10" cells were seeded in24-well plates, while forWestern
blot analysis and mRNA expression 4 x 106 cells were plated in 15 ml
polypropylene tubes, Cells were incubated with or without lipopoly-
saccharide(Ll"S) in the presence or absence of increasing concentrations
of PFCs, or dimethyl sulphoxide (DMSO; 0A11. final concentration) as
vehicle control, as described in the figure legends.
Whole blood assay, Healthy subjects, enrolled among colleagues Of
the researchers, were selected according to the guidelines of the
Italian Health authorities and to the Declaration of Helsinki principles,
Criteria for exclusion were abnormal laboratory values, medication
known to affect the immune system, i,e. steroids and nonsteroidal
anti- inflammatory drugs. or patients suffering from malignancies,
inflammations, and infections. All subjects signed an informed
consent and were informed about methods and aims of the study.
Blood samples (5 ml) were taken by venous puncture with sodium
citrate 0.5 M as anticoagulant. Sodium citrate was chosen instead of
heparin or EDTA as anticoagulant, since functional assays were
performed using the whole blood assay and heparin may be
contaminated with endotoxin, while ED'IA interferes with cell
activation. Blood was diluted 1:10 with RPMI 1640 cell Culture
riedium(Sig-tria, St Louis, USA) containing 2 rnM L-gliltanline, 0.1 mg/
ml streptomycin, 100 11-1/1111 penicillin. For the evaluation of cytokine
production, cultures were set up in 24-well plate containing I fril of
1:10 diluted whole blood, in medium alone or with increasing
concentrations of PFCs or DMSO as vehicle control (0.1% final
concentration) in the presence of I pg/ml LPS or 1.2 ltg/rnl PHA. For
IL-6, IL-8, and TNF-a release cells were incubated for 24 li, while for
ILA IL-10, and IFN-y release cells were incubated for 72 h, Cells were
incubated at 37 'C in a humidified 5% CO2 incubator,
Lactate dehydrogenase. Cell viability was assessed by lactate dehy--
drogeriase (LDH) leakage from damaged cells. LDH is a well known
indicator of cell membrane integrity and cell viability. LDH activity
was determined in cell -free supernatants using a commercially
available kit (Takara Bio Inc., Japan). Results are expressed as OD,
(ytoldne and NIMP-9 production, Cytokine and matrix metallopep-
tidase-9 (MMP-9) release was measured in cell -free supernatants
obtained by centrifugation at 1200 rpm for 5 min and stored at —80 'C
until measurement, Cytokine production and MMP-9 release were
assessed by commercially available sandwich 1`11SAs (11111111.1noTools
GrnbH, Friesoythe, Germany and R&D Systems, Minneapolis, MN, USA,
respectively), ELISAs were performed following the supplier's
instructions, Results are expressed in p,-,,'fnl. The litnit of detection was
15.6 pg/ml for all cytokines tested and 31,25 pg/nil for MMP-9,
Real tune RT-PCR, Total RNA was isolated using a Commercially
available kit JdReagent, Sigma) as described by the manufacturer.
For the synthesis of cDNA, 2.0 pg of total RNA was retro -tra n scribed
using a high -capacity cDNA archive kit from Applied Biosystems (Foster
DEQ-CFW-00004193

110 E. Corsini et al. / Toxicotogl and Applied PhannacolQgy 250 (2011) 108-116
City, CA, USA) following the supplier's instructions. TNF-cx gene
expression was evaluated by Real time reverse transcription --polymer--
ase chain reaction (Beal time-PCR). For PCR-analysis, Tact-Man_f,-PCR
technology was used, PCRs were performed in duplicate and according
to the standard protocol suggested by the manufacturer. For each PCR
reaction, 100 ng of the total RNA was used, The 18S ribosomal RNA
transcription was used as endogenous reference and the quantification
of the transcripts was performed by the AACr method.
Western blot analysis, The activation NF+t B was assessed by mea-
suring cytosolic degradation of I+,B, nuclear translocation of p5O and
p65, and the phosphorylation of p65 (P-p65) by Western blot analysis.
Briefly, cells were allowed to acclimatize for 1 h at 37'C. and were
then treated with PFCs for 5 min, followed by LPS for 30 min, Cells
were then collected, washed once with PBS, centrifuged and lysed in
100l.d of homogenization buffer (50 mM TRIS, 450 rnM NaCl, 5 rnM
EDTA pH 7.5, 0,5%Triton X--100, 50 LaM PMSF, 2 f€g/nil aprotinin, 1 f€g/
nil pepstatin and 1 gg/ml leupeptin) and denatured for 10 min at
100 °C. The protein content of the cell lysate was measured using a
commercial kit (Bio-Rad), 40lig of extracted proteins was elec-
trophoresed into a 12 fl SDS-polyac_rylamide gel under reducing
conditions. The proteins were then transferred to PVDF membrane
(Ariersham, Little Chalfont, UK), The proteins were visualized using
prifriary art tibodies for lKB {1:2000), p5O (1:750), p65 (1:2000), 11-p65
(1:1500) and r)-actin (1:2000) and developed using enhanced
the€niluminescence (ECL, Arnersharn, Little Chalfont, UK), The image
of the blot was acquired with the Molecular Imager Gel Doc XR (Bio-
Rad). The optical density of the bands was calculated and analyzed by
means of the Image 1.47 program for digital image processing (Wayne
Rasband. Research Service Branch, NIMH, NIH, Bethesda, MD, USA).
NF-r,13 (p65; translocation. Nuclear extracts were prepared essen-
tially as described by Schreiber et al. (1989). Briefly, after treatment,
4 x 10t cells were resuspended in 0.4 ml of a hypotonic lysis buffer
(10 ruM Hepes, pH 7.8, 10 mM KCI, 2 rnM MgCl;,, 1 €nM dithiothreitol,
0A mM EDTA, 01 mM phenylmethylsulfcnlyl fluoride). Cells were
incubated o€1 ice for 15 min and then 25 fil of a 10% Nonidet 11-40
solution was added, and cells were mixed for 15 s, and then centrifuged
for 30 s at 12,000 rpm, Pe lleted nuclei were suspended in 50 0 of buffer
C (50 mM Hepes, pl-f 7.8, 50 mM KCI, 300 mM NaCl, 10% glycerol, 1 mM
dithiothreitol, 0A mM EDTA, 0A €nM phenylmethylsulforlyl fluoride),
mixed for 20 min, and centrifuged for 5 man at 12,000 rpm, The
supernatants represent the nuclear extracts. Protein concentrations in
both fractions were measured usinga commercial kit (Bio-Rad), Nuclear
extracts were used to assess NF-KB (p65) translocation using a
commercially available kit (Cayman Chemical, Ann Arbor, MI, USA)
that allows the detection of specific transcription factor activities in cell
extracts using air enzyme -finked imniunosorbent assay -based format.
Transient transfections. A luciferase reporter plasrnid with three NF-
rB sites from the E--selectin promoter as described previously (Brostjan
et at., 1997) and kindly provided by N. Marx (Department of Internal
Medicine 11-Cardiology, University of Ulm, Ulm, Germany) was used.
THI'-1 cells were transfecred by the DEAE-dextran method (Sarnbrook
and Russel, 2001), Briefly, cells were exposed to a mixture of DNA--
dextra i (750 fig/ml final concentration" for 30 min, using 700 ng NF-KB-
1 ac reporter plasmid/1.5 x 1O" cells. Cells were seeded in 12 well plates
at a concentration of 1.5 x 106 cells/ well and their incubated for 48 h in
complete medium (FCS supplemented RPMI-1640). Cells were treated
with increasing concentrations of PFOA and PFOS in the presence or
absence of LPS (0.1 fig/ml) for 3 h, At the end of the incubation, Britelite
Plus reagent (Perkin Elmer. Milan, Italy) was added (100 pi/well). The
luciferase assay was performed using a lununo€neter (Victor °r'iX3,
Perkin Elmer, Milan, Italy). Results are expressed as luciferase activity
normalized by protein content and represent the mean i SD values of
three experiments performed in triplicate, Parthenolide was used as an
inhibitor of NF-kB-driven transcription (82% inhibition at 20 pM, data
not shown),
Small interference RNA (siRP7A) for 1'PAR-ca We assessed the effect
of inducing RNA interference on PPAR-a using commercially available
reagents (Silencer Pre -designed siRNA from Ciiagen) following the
manufacturer's instructions. As control siRNA, a siRNA sequence that will
not cause the specific degradation of any cellular messages was used.
Forty-eight hours after siRNA transfection. PPAR-ce contents in whole
cell lysates was assessed by Western blotting using PPAR-tx antibody
(Sigma) to confirm the silencing of these protein. Cells were then
adjusted to 106/nil and treated with PFOA or PFOS (100 fig/nil) in the
presence of LPS (0.1 fig/ml) for 3 h to assess cytokine release and 48 h to
assess MMP-9 release, The PPAR-or, ago€list fenolibrate (50 .uM) was used
as reference compound in the MMP-9 experiment
Statistical analysis. All experiments were repeated at least three
times, with representative results shown. Data are expressed as €near
± standard deviation (SD), Statistical analysis was performed using
InStat software version 3.Oa (GraphPad Software, La Jolla, CA, USA).
Statistical differences were determined us€€rg ANOVA followed by a
multiple comparison test as indicated in the legends. Effects were
designated significant if p <_0.05,
Results
Effects of PFOA and PFOS on human, peripheral blood leukocytes
The whole blood assay was used to assess the immunornodLila tory
effects of PFOA and PFOS, Peripheral blood obtained from healthy
volunteers was diluted 1:10 and treated with increasing concentrations
of PFOA and PFOS (0.1--10 fig/ml) in the presence of LPS (I fig/ml) or
PHA (1.2 fig/ml) for 24 and 72 h, respectively. In Fig. 1 panels A, B, C the
effect of PFOA on LPS-induced release of TNF-cY,, IL-8, and IL-6 is
reported, while in panels D, E, F the effect of PFOS is shown . Both
PFCs induced a dose -related decrease in TNF-cx release (Figs, 1A and D),
IL-8 release was unaffected (Figs. 113 and E), while the release of IL.-6
was decreased only by PFOS and not by PFOA (Figs, I C and F). Regarding
T-cell derived cytokines, as reported in Table 1, both PFOA and PFOS
decreased PHA -induced 114 and 11. 10 release, while IFN--Y release was
affected only by PFOS, In all instances, PFOS was a more potent inhibitor
of cytokine production than PFOA, with effects appearing at lower
concentrations. At the concentration of 0A fag/rr11 (equivalent to 241 r1M
of PFOA and 200 nM of PFOS)), only PFOS was able to reduce the
secretion of TNF-(x, ILA IL-6, IL--10 and IFN- V.
Effect of PFOA and PFOS on LPS-indicted cytokine production in THP--1
cells
Similar to the findings in peripheral blood leukocytes, in the human
promyelocytic cell line THE-1, PFOA and PFOS were able to modulate in a
dose -dependent manner LPS-irrduced TNF-ce release (Figs. 2A and Q,
with PFOS being the more potent inhibitor of cytokine production. In
contrast to the lack of effect on I1-8 in peripheral blood leukocytes, PFOS
suppressed LPS-induced IL-8 release in THE-1 cells in a dose -dependent
manner (Fig. 2D). PFOA also suppressed LPS-induced IL-8 release, but
only at the highest dose of 100 f€g/ml that was not examined in
peripheral blood leukocytes (Fig. 2B), LPS-induced only a modest
release of IL-6 in THP-1 cells, Following 24 h of treatment the a€nou€rt of
IL-6 released was 122 -l_ 21 pg/ml in LI'S treated cells, Consistent with
the results obtained in peripheral blood leukocytes no difference from
control values was observed in cel is treated with 10 gg/ml PFOA and LPS
(121-1_-- 12 pg/ml), while a decrease was observed in the cells treated
with 10 fag/ml PFOS and LPS (89 ± 25, p <0.05 ). The immunomodulatory
effects observed, both in peripheral blood leukocytes as well as i€rTHP-1
cells, were not due to cvtotoxicity, as no cvtotoxiciqr was observed
DEQ-CFW 00004194

F. Corsirti et aC. /Toxicology anti Applied Phormacoto�,l 250 (2011) 108-116
ME
A 2s0a
2000
1500
1000
z
M 500
0
l 1,
Alt,
t 20000
Is 15000
2 10000
A
f5'+
5000
0
600
600
400
200
0
LIPS V 1 10
PFOA (lagiml)
LPS 0A 1 10
PFOA(lA9tml)
LIPS 01 1 10
PFOA (lcgfml)
D 2500
200D
150a NA Md1
1000 xm
500. 6
0
LIPS GA 1 10
PFOS (9e%W)
E 30000
25000
20000
15000
2 10000
5000
0
Mi*
LPS 0.1 1 10
PFOS (lt9W)
LPS 0.1 1 10
PFOS (ing'M)
Fig. 1. Fffect of PFOA and PFOS on LPS-induced release of'IS'I u, L S, and 1L-6 in pei r4'heral blood teuRocytes. The whole blood: was diLrted 1:10 in culture me.i:xn7 an('! treated with
increasing concentrations o' PFOA (panels A, B, C), PFOS (Panels D, F, F) or DMSO � 0.1`E final concentration' vehicle control in the presence of I µgyrnl LPS for 241). Each value
represents the mean --sr') of triplicate wells of as ine donor. Statistical istica l analysis was performed with Dunnett's :rnALlp:e comparison test with * p ,0.05 and h <0.01 vs LPS treated
cells. Data is representative of three independent donors.
following treatment with 100 ftg/ml of either PFOA and PFOS for 48 h, as
assessed by LDH leakage (Table 2).
PFDt1 and PFOS inhibited LPS-induced TlslF-ce inP.NA expression
Having established that the TNF-ca release data obtained from
THP-1 cells was similar to the data obtained with human peripheral
Table I
Effect of PFOA and PFOS on PHA -induced cytofdne release in peripheral blood
leukocytes.
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Treatment I1.-4 1pg;inI) IL-10 (pghml) lFDd-y (pg;ml)
DMSO vehicle 88 f 6 1949 ± 8.3 36,920 i 1907
PFOA (p31ml)
0.1 86 f 9 1581 f 108 33,228 ±4719
1.0 73 f 2 1331 f 187 28,968 ± 3261
10.0 69±10` 1264_-21* 29,205_<-9001
PFOS (pg/rrd)
0.1 57f22' 1135±252** 20,306-L1606*
1.0 54f4** 1131±178** 14,9.39±773**
10.0 45f7** 855±76— 13,064±904**
whole blood was dilute:: 1:10 in culture nediuni and treated with increasing
concen r ations of PFOA and PFOS or DMS0 (0.1% final concentration) as vehicle
control ur tsae presence of PHA 1 7 } g;;r; for 72 It. F.ach tisflue represents the ±SD
of triplicate wells of a single donor, with * p<0.05 and **p<0.01 vs Pr[A treated cells.
Data is representative of three independent donors.
leukocytes, we believe the THP-1 cells are an appropriate in vitro
model to study the molecular mechanisms underlying PFCs modula-
tion of cytokine production, In order to investigate if PFCs acted at the
pre- or post -transcriptional level, LPS-induced I'NF-ax and IL-8 €nRNA
levels were evaluated by ITT-PCIa. THP-1 cells were treated with PFOA
and PFDS (100 f[g/ml) or DMSO as vehicle control (0 1 % final), and LPS
(0.1 pg/ml). TNF-cx mRNA levels were evaluated after 1 h, since in
previous experiments we found that in THP-1 cells TNF--tz mRNA
peaks at 1 h and declines thereafter (28), As shown in Fig, 3A, in LPS
treated cells a significant increase in TNF--cx mRNA level was detected.
In contrast, in cells treated with either PFOA or PFOS, LPS-induced
TNF-cx mRNA levels were significantly reduced, indicating that the
PFCs acted at the pre -transcription level, PFOA and PFOS did not
modulate constitutive TNF--cx mRNA expression, To further verify that
modulation of cytokine secretion occurred at the pre -transcription
level, we quantified the levels of LPS-induced IL--8 mRNA expression
following PFOA or PFOS exposure, Similar to the results observed with
TNF-oc, IL-8 mRNA expression was also reduced following treatment
with PFOS and PFOA (data not shown).
Effect of PFOA and PFOS on LA'S -induced NF-t<li activation i€1 THP-I cells
It is known that LPS-induced TNF-cY production is dependent on
NF-rB activation, Thus, we investigated the effect of PFOA and PFOS
on LPS--induced NF-rB activation. Cells were treated with PFC1A
and PFOS (100 gg/ml) or D€ ISD as vehicle control (01 % final) and
LPS (0,1 ug/ml) for 30 rain. NF-rB activation was assessed by
DEQ-CFW 00004195

112 E. Corsird et al. / Toxfcolfo�,l and Applied Pharmacology 250 (2011) 108-116
A 10000
8000
6000
4000
LL
2000
10000
ME
LPS 0.1 1 10 100
:PFOA (pg/Ml)
LPS 0.1 1.0 10 100
:PFOA (pglml)
C10000
6
'ai 6000
a
4000
z
2000
15000
10000
svelte'
LIPS 0.1 1 10 100
PfOS (Pg/ml)
LPS 0-1 1:0 10 100
PFOS (pglml)
Fig. 2. Effect of PFOA and PFOS on LPS-nduced release ofTNF-uand IL-8 inTHP-1 cells. Cells (106/inl) were treated with increasing concentrations of PFOA (panels Aand 9), PFOS
(panels Candl)) orDMSO (0.l`E final concentration) vel)i(:Iecoat i)tiritliepl,e.seri(:ei)[O.J jAg,/rnlLPS for h. Eachvollje analysis
was performed wrth Dunneh.'s inuttlple compartson test with * p,0.05 and **t O.OJ vs UPS treated cells.
measuring I-r;B degradation, NF-KB nuclear translocation (data not
shown) and p65 phosphorylation by Western blot analysis. and NF-rB
p65 binding to DNA by EUSA. As shown in Fig. 313, only PFOS
prevented LPS--induced I-rB degradation, 1--KB degradation is required
for NF-KB translocation from the cytosol to the nuclei, indeed,
PFOS prevented Nl`-KB p5O,'p65 nuclear translocation (not shown)
and binding of NF-i<B to DNA (Fig. 3Q. PFOA did not affect LPS-induced
1--KB degradation (Fig. 313), NF-KB nuclear translocation or binding to
DNA (Fig. 3Q
To determine if other specific regulatory factors involved in NF-FB--
mediated transcription were attenuated by PFCs, THP-1 cells were
transiently transfected with a luciferase reporter plasmid CO11StrUCt
containing three NF-isB sites, and the effect of PFOA and PFOS On
luciferase activity, as indicator of the promoter activity, was
measured. Transfected cells showed a significant dose -dependent
decrease in LPS-induced luciferase activity relative to control after
treatment with both PFOA and PFOS (Fig. 4A), confirming NF-rB as an
intracellular target of PFCs. This is consistent with the ability of PFOS
to inhibit I-vB degradation, NF--KB nuclear translocation and DNA
Table 2
Effect of 13' FOA, PFOS and fenofibrate on LPS-induced MMP-9 release and LDH leakape.
TicatnlenL '01) H T ) LD , , MMP-9`ijg/n.l)
Control
0.626 A-- 0.041
15.4 A-- 3.7
LPS
0.778 -0.070
7257.8 1015
Fenofibrate +.PS
0.6 74 0.060
49,530.2 11.5",
PFCA + UPS
0.665 0.027
4387.2 163,5**
PFCS 4- LPS
0.586 0.040
678 7.4 214,8
THP-1 cellW3/lnll) wen-, treated with PFOA (100pp/irfl), PFOS (100pp/nil-,
fenofibrate (501,m) or DMSO (O.J% final concentrotion) as vehicle control in the
presence orabsenceof LPS 0.1 pg/rnl for 48 1). No cytotoxiclt.y was observed using this
exposure regjnaen as evidenced by the lack of effectsvalue
or, TDH release. Fach i
represents the mean ± SD, n=3, with "p,0.01 VS TpStreated cells.
binding. However, for PFOA other mechanisms are likely to be
responsible of the lack of NF-KB transcriptional activity. We could
indeed demonstrate that PFOA, as well as PFOS, prevented p65
phosphorylation at Ser276, which is required for optimal NF--KB-
mediated transcription (Fig, 4B). The lack of p65 phosphorylation
can explain the inhibitory effect of PFOA on luciferase activity.
Taken together our data indicate that PFCs, interfering with LPS-induced
NF--KB transactivation, prevented transcription and translation of'I'NF-(X,
resulting in a decrease in the release of this cytokine in monocytes.
D�fferences in, the role of PRARcy in PFCs-induced immunotoxicirly
Although their exact role in PFCs-immunotoxicity is still a matter
of debate, IIIIARs are relevant to the study of the biological effects of
these compounds as they can activate PPARcy, In vitro studies have
demonstrated that PFOA is a more potent agonist of routine PPARCY
than PFOS (Takacs and Abbott, 2007). By silencing PPAR(Y,, a differing
role of 131"ARut in PFOA and PFOS-induced immunotoxicity could be
demonstrated. As shown in Fig. 5, under experimental conditions
resulting in a 32% decrease in PPARcy expression (Fig. 5A), only
the PFOA-induced inhibition of cytokine release was reversed
(Figs. 5B and Q. In contrast. the immunomodulatory effect of PFOS
on LPS-induced TNF-a and IL-8 release was not affected by PPARot
silencing, as a similar decrease was observed in both mRNAsi and
control ofigo treated cells in comparison to naive cells,
A difference in the role of PPAR.cy in PFOA and PFOS-induced
immunotoxicity is further supported by their disparate effects on LPS-
induced MMP-9 release, Agorrists for PPAR.cy have been shown to
reduce LPS-induced secretion of MMP--9 in human monocytes,
Suggesting that PPAR(Y may play a beneficial role in inflammation
and tissue remodeling (Shu et al- 2000; Delavre- Orthez et al., 2005)
As shown in Table 2, only PFOA and the PPARu agonist fenofibrate
were able to modulate MMV-9 release, PFOS did not modulate Ll--'S-
induced MMP-9 release. This effect was mediated by PPAR(Y
activation, as it was recovered by PPAR(x silencing (Fig. 5D),
DEQ-CFW-00004196

E. Corsini et al. Toxicolog I
y and Applied Phorinacologl 250 i201 1) 108- 116
113
Wo
z 1'0
I:]
M$
100
90
80
10
60
R9
C wo
600
Sro
oo
4W
t 7j 300
6 M -
U 200
4L �R
2 100
LPS PFOA+ LPS PFOS + LIPS
P-acan
Cant PFOA Pros
LPS PFOA10+PFOA IOOPFO$I+PFOSIO+PFOS IDO
LPS + LPS LPS LPS + LPS
H& 3. Effecl, of PFOA and PFOS on LPS-induced tnRNA expression and NY-0
activation in 'F4P--1 ce1ls.,A)TN.F--(x nRNA expresslOn. Cells (10l3/mll were treated with
PFOA (100lg/lill), PFOS (100pg/rnfl, or DMSO (0.1% final concentration) as vehicle
control in the -presence of0.1 Pg/nal UPS for 1 h. (B) in the same experimental conditions,
cells were treated for 30 min to assess r­KB degradation. While in (C), NF-KB p65 D!.,;A
binding was evaluated in cells exijosed. to PFOA (10-100 Pg/rfll) and PFOS (1-100 ltg/nil)
in the presence of 0.1 p/nil LPS for 30 in.n. Each value represents the mean -.4- SD of 3-4
independent experiments. Statistical analysis was performed with Dunnert's inultijiJe
comparison test with *p<0.05 and **p,0.01 vs LPS treated cells.
Discussion
Using primary hurnan leukocytes and theTHP-1 cell line exposed to
PFOA and PFOS in vitro, we demonstrated that PFCs directly affected
immune cell activation and reduced cytokine production {both pro- and
anti-inflammatory) through different mechanisms, Finding similar
suppression of cytokine release in both models, we examined the
mechanisms underlying the suppression ofpro-inflammatory cytokines
in THFl- 1 cells. PFOS exerted effects independent from 1113ARCY activation
to reduce LPS-induced I-rB activation, Nl`-rB binding to DNA, p65
phosphorylation, and transcription. PFOA, acting through PPARcy,,
A 120
100
80
CL
9-3
60
ecLe-
40
z
W
cont LPS 01 1 10 100 coM LPS 0.1 1 10 100
PFOS i, LPS PFOA + LPS
LIPS PFOA + LPS PFOS + LPS
Fig. 4. Effect of PFOA and PFOS on UPS -induced. NF-KB p"Ornoter activiq, and NF-KB p65
phosphorylation in THP-1 cells. (A) Cells ('10'Aird) were treated with increasing
concentrations of PFOA. PFOS or OMSO (0.1% final concentration) vehicle control in the
'0.1 �lg/in' LPS 0A pg/.ril for 3 h forNF-r.13 promoter activity.[3) Cells (106/111)
presence o I I
were treated with PFOA" 100 pg/rfil), PFOS (100 kg/nil), or DMSO (0.1% final concenLra-
hon) as vehicle control in the presence SejjCe Of LPS S 0.1 ].gj'rnl for Mann for p65
I
phosphorylabon. Each value represents the inean ± SD oi 3­4 independent experiments.
statisucal 3 na[ysiswas performed wil-11 Dunnett's parison tesi, with *t? , 0.05
and P<0.01 vs UPS treated cells.
prevented p65 phosphorylation and NF-isB-mediated transcription, In
both cases, an anti-inflarnmatory effect was observed. In both primary
human cells and in THP-1 cells, PFOS was the more potent compound,
acting at lower (loses and affecting more endopints. Cells involved in
both innate and specific immune responses, namely monocytes and T
lyn-iphocvtes, were affected as both LPS and PHA -induced cytokine
production were decreased following PFOA and PFOS exposure, These
results were consistent with previous studies that reported immuno-
modulation in experimental animals exposed to PFOA and PFOS,
including altered inflammatory responses, cytokine production, re-
duced lymphoid organ weights and decreased antibody production
DeWitt et al., 2009).
The exact role of Pl'AlZs in PFCs-immuno toxicity is likely to be
complex with some effects resulting from a IIIIAR--mediated mechanism
(as in the reduced cytokine release associated with PFOA exposure from
the current study), while others occur via a PPAR -independent
mechanism (as observed in the reduced cytokine secretion associated
with PFOS exposure in the Current study). There is mounting
experimental animal data demonstrating PPARot independence of
some immune effects (DeWitt eta L, 2009). As human PPARcy expression
is significantly less than that of rodents, the identification of effects
independentof PPAR.cy activation is important for evaluating human risk
from experimental animal studies of PFCs. We demonstrated a role of
PPAR(Y, in PFOA-induced effects on cytokine secretion in hUrnan cells
DEQ-CFW-00004197

114 E. Corsini et al. / Toxfcolfo�,l and Applied Pharmacology 250 (2011) 108-116
AftmVe MRNAM Op cont
PPAR
..........
0.81 0.55 0.80 radD
100
so
V
4
60
40
6 20
$ 0
V 100
60
40
a.
®s 20
MMP-9 (0/6 of LIPS -treated cells)
Treatment
NalVe
73,0*11"
56*1.5**
PPAFta mRNAsl
86,1*3.9-,§§
66*1,4--,§§
011go control
74.0zt5.7**
31d:6.2**
PFOArktiva PFOS naive PFOA PFOS PFOA ollgo PFOS ottgo
mRNAsl mRNAsi control control
PFOA fw?ve PFOS natty PfOA PFOS PFOAobp PFOSolgo
mRNAst MRNAM control cww
Fig. 5. E-"*ecL o'*PP.AR-a silencing on the inhibition of PFOA and PFOS of*LPS-induced cytokine release in THP-I cells. PPAR-uvvas silenced as described in the Material an niefhods
section, cells 10"/nil1 where then treated. with PFOA (100 1.g/nffl, PFOS '100 kg/nil), fenofibrate (70 PA), or DMISO Al% final concentration) vehicle control in the presence of
0.1 l.g/nll LPS for 3 h for cytokine release and 48 it for MMP-9 release. (A) PPARce ininiunoreactivity following PPARa silencing; (B) TNF-a release: (C) IT 8 release; (D) MMP-9
release. Each value represents the rnean ± SE) of three samples. Statistical onolysis was perioroted with T'ahey's multiple comparison test with ­p<0.01 vs LPS treated celts and
§§p,0A vs naive cells or cells treated with the control olip.
exposed in vitro. In contrast, our data demonstrated that the inhibitory
effect of PFOS on in vitro cytokine production by human leukocytes
occurs independently of PPARa, and involves inhibition Of I -KB
degradation,
NF-isB activation plays a key role in inflammation, immunity, cell
proliferation, apoptosis and cytokine production in both T cells and
monocytes/macrophages (Crabtree and Clipstone. 1994; Viatour et al.,
2005). A detailed analysis of NF-KB in THP-1 cells showed that at
concentrations that did not produce cellular cvtotoxici both PFOA and
PFOS inhibit the signaling pathways that regulate NF-KB activation, Both
chemicals inhibited p65 phosphorylation and NF--KB promoter activity;
�,;
however, while PFOS acted upstream inhibiting I -KB degradation, PFOA
did not interfere with LPS-induced 1--KB degradation or NF-rB nuclear
translocation and DNA binding, indicating a downstream effect.
Decreased PPAR activity is closely associated with increased levels of
inflammatory mediators (Chinetti et al., 2000)". We demonstrated, using
RNAsi, a role of PPAR(Y in PFOA--induced immunotoxicity, but not in
PFOS inhibition of cytokine production. The phosphorylation of p65,
which is required for optimal NF-I-B dependent gene transcription
(Sclunitz et al., 2001", appears to be an important target of PFOA-
induced PPAR(Y, activation, which may account for the defective
cytokine production observed.
The onset of inflammatory gene expression is driven by the
transcription factor NF-isB, whose transcriptional activity is regulated
at multiple levels (Vanden Berghe et aL, 2003). NF-isB activity is
initially regulated by cytoplasmic degradation of the I-rB inhibitor
and nuclear translocation. Following this, transactivation of nuclear
p65 can be further influenced by posttranslational modifications, such
as phosphorylation and/or acetylation. The p65 phosphorylation is a
process highly regulated by both cell- and stifflUILls-dependent
activating kinases, Ser276 phosphorylation has a crucial role in the
interaction with and the engagement of the cofactor CBP/p300, being
therefore important in order to establish gene activation (Saccani
et al., 2002; VermeLflen et al., 2002, 2003).
The pathways responsible for the anti-inflammatory role of 111--IAR(Y
ligands are not yet clear, and a number of different mechanisms have
been proposed (Cunard, 2005). One mechanism by which PPAR(Y
negatively interferes with inflammatory gene expression is by up -
regulation of the cytoplasmic inhibitor molecule F-KB alpha, thus
establishing an autoregUlatOFY loop (Vanden Berghe et aL, 2003). In
our experimental conditions, PFCs and LPS were added almost
simultaneously (5 min) and I -KB expression was evaluated 30 min
later. Under these conditions, we did not observe any change in 1--KB
expression: 0.96 —i 0A 8 I-KB/0-actin relative densitometric analysis in
vehicle treated cells, 0.93±0.18 in PFOA and 0,82±0.17 in PFOS
treated cells. We found that PFOA, a PPARet agonist, does not affect
LPS-induced NF-rB nuclear translocation or DNA binding, but it
inhibits p65 phosphorylation and NF-kB promoter activity, interfering
with the mRNA transcription of TNF-(Y, and its release. It is possible
that PPARa, forming heterodirners with RXRot, may act as negative
transcription factor for major pro -inflammatory cytokines such as
TNF-a, by binding to peroxisome prolifLrator-responsive elements in
the regulatory region of the cytokine gene. Alternatively, as also our
data supports, PPARct may influence activation of other transcription
factors, i.e. NF-KB, involved in the regulation of inflammatory cytokine
expression (Devchand et al., 1996),
DEQ-CFW-00004198

E. Corsirui at aC. /Toxicology anti Applied Pharmacoto�,y 250 i2011) 108-116
RM
The different roles of PPARa in PFOA and PFOS-induced immuno-
toxicity, are also demonstrated by their divergent effects on LPS--
induced MMP-fJ release. MMP-fJ has been shown to be important
in IL--8 induced mobilization of inflammatory cells, and its modulation
has been associated with the anti-inflammatory effect of PPARcc
agonists (Shu et al., 2000). Fenoff9brate significantly down -
regulated LPS-induced secretion of iy'IMP-9, while failing to modulate
11.-8 release, Suggesting that of PPARu may modulate only a subset
of pro -inflammatory genes (Shu et al., 2000). The downregulation
of MMP-0 has been shown important for the beneficial anti-
inflammatory effect of PPARct agonists in the treatment of ather-
oscerosis (Bouhlel et al., 2008; Tsimihodi€nos et al., 2009), This is
consistent with our results, as PFOA, as well as the well --defined
PPAR-ty agonist fenoflbrate, reduced MMP-0 release induced by L PS
in THP- 1 cells. while PFOS was infective. This inhibitory effect was
reverted by PPARa silencing, confirming the direct role of 11PARcti
activation in LPS-induced MM11-9 release.
As plasma levels of PFOA and PFOS ranging between 3.7 and
12,000 ng/ml have been found in the human population (Fromme et
al., 2009), the concentrations (0.1-100lig/€nl) used in our i€1 vitro
studies are likely to be relevant for the human situation for PFOS,
which was able to significantly reduce cytokine release at concentra-
tion as low as 100 ng/ml, For PFOA, concentrations higher than
1000 rig/ail were necessary to significantly reduce cytokine release,
and these levels are unlikely to be observed i€1 the hu€nan population,
It is, however, important to consider that humans are exposed to
several endogenous and exogenous PPAR ligands. Therefore, we
cannot exclude a possible role of PFOA in the total ligand contribution
to receptor occupancy and activities, and Subsequent influence o€1
patho--physiological processes involving PPARet activation. While the
anti-inflammatory effects of PPAR ligands may be of therapeutic value
with controlled administration for chronic inflammatory disorders,
they are likely to be detrimental in the case of infections, as clearly
shown by Guruge et al. (2009). which demonstrated that PFOS
exposure i€1 mice resulted i€1 a significant increase i€1 emaciation and
mortality in response to influenza A virus. Further studies are needed
to elucidate the potential contribution of chemicals that chronically,
activate PPARtx in the susceptibility to infections, cancer and in the
development and persistence of chronic inflammatory diseases.
Overall, time have demonstrated that in vitro exposure to PFOA and
PFOS directly inhibited cytokine production in cultured hu€nan
leukocytes, Furthermore, we demonstrated that the observed effects
of PFOA were dependent upon PPARa activation, while the effects
of PFOS were independent of PPARtx activation, The decrease in LPS-
induced cytokine production by PFOS may be explained by the
observed inhibition of I-rsB degradation associated with PFOS
exposure.
Conflict of interest
Authors declare not having any financial or personal interest. nor
having a€1 association with any individuals or organizations that could
have influenced inappropriately the submitted work,
Acl(nowlerfgments
We would like to thank Dr, Chad Blystone and Dr, Andrew Rooney
for their review and helpful suggestions. This research was supported
in part bythe Intramural Research Program of the National Institute of
Environmental Health Sciences, and the National Institutes of Health.
This article may be the work product of an employee or group of
employees of the National Institute of Environmental Health Sciences
(NIEFIS), and the National Institutes of Health (Nil-1), however, the
statements. opinions or conclusions contained therein do not necessar-
ily represent the statements, opi€lions or Conclusions of NIEHS, NIFI or
the United States government.
References
Riegel, L.P r-lu tt, ndi I .. Frame, S.R., O'Connor, J.C., Cook, J.C.. 2001. Mechanisms of
extrahepatic tumor induction by peroxisome prolifeiators in male CD rats. Toxicol.
Sc.. 60, 44 5.
Rouhlel, M A, Staels, B., Chinetti-Gbaguidi, G., 2008. Peroxisome proliferator-activated
receptors -from active regulators of macrophage biology to pharmacological
targets in the treatment of cardiovascular disease. J. Intern. Med. 263, 28-42.
Brostjan, C., Anratber, J., Csizanadia, V., Natarajan, G., Winkler, H.,1997. Glucocorticofds
inhibit E-setectin expression by targeting NF-kappaB and not ATF/ c-Jun. J.
Immunok. 158, 3836-3844.
Butenhof. J I C y or D.W., Moore, J.A., Olsen G.W., Rode c ks, J. Mande:, J.H Zobel, "..P ,
2004. Characterization of risk for general population exposure o perfluouooc anoate.
ReguL Toxicol. Pharmacot. 39, 363--380.
( nafat A.M., Ye, X. Silva, k4j Ku klehyi:�, Z., Needham, L.L., 2006. I-€uman exposure
assessrneni: to envv�onmeaatail cbearuivals using bioaruonitoring, lnt J. AardraL 166--171.
Chinetu, G., Fruch,rt, J.('., S:aels, B., 2000 Peroxisome Lvraliferatur-activateu: receptors
(PPAURs): nuclear receptors at the crossroads between lipid metabolism and
inflammation. fnflamm. Res. 49, 497- 505.
Crabtree, G.R., Clipstone, N.A, 1994. Signal transmission between the plasma
membrane and nucleus of'T lymphocytes. Annu. Rev. Biochem. 63, 1045-1083.
Cunard, R., 2005. The potential use of PPARalpba agonists as immunosuppressive
agents. Corr. Opin. Invest. Drugs 6, 467-472.
Delayre-Orthez, C., Becker, J., Guenon, L. Lagente, V., Auwerx. J., Frossard, N., Pons, F.,
2005. PPARalpha downregutates airway inflammation induced by lipopokysaccha-
ride in the mouse. Respir. Res. 6, 91.
Devchand, P.R., Keller, H., Peters, J.M., Vazquez, M., Gonzalez, F.J., Wahli, W., 1996. The
1'FAR u fv a o--le+xkotr:ene B4 pathway to inflammation and control. Nature 384.
39--43.
DeWitt, J.C., Copeland, C.B.. Strynar. M.J., Luebke, PW., 2008. Perfluorooctanoic acid-
iauiuced imrncuaomocadation in adrult C57BL/6,1 or C'S7BL/6 N female trice, Eruviroaa.
Health Perspect. 116, 644--650.
DeWitt, J.C., Copeland, C.B., Luebke, PW., 2009. Suppression of Immoral immunity by
perfluorooctanoic acid is independent of elevated serum corticosterone concen-
tration in mice. Tcsxacol. Sci. 109, 106-112.
Emmett, E.A, Zhang, H., Shoier, F.S., Freeman, D., Rodway, N.V., Desai, C., Shaw, L.M.,
2006. Community exposure to perfluorooctanoate: relationships between serum
levels and certain health parameters. J. Occup. Environ. Med. 48, 771-779.
Fromme, H., Tittlelmier, S.A., Votkei, W., Wilhelm, M., Twardella. D., 2009. Perfluorfnated
compounds -exposure assessment for the general population in Western countries.
lit J. Hyg. Environ. Health 212, 239-270.
Gur uge, K S Ibkono, H Shi.nad a Mutraa7c m7i, K., Hasegawaj. 4 eunu LVi., Y rnanaha,
N., Yamashita, N., 2009. Effect o: pefEuorooctane sulfonate �P OSI on influenza A
virus-ind u:_ed m yu ality in fern,ale B6C 3F1 :rcce. I to Nicol. Sc . 34, 687--691.
[Kennedy Jr., G.L., Butenhoff, (.L., Olsen, G W. ('Connor. J.C„ Seat -'at. AndL, Perkins, R.G.,
Biegel, L.B.. Morp by, S,ft., Farrar, D.G., 2004. Tire toxicology of perfluorooctanoate-.
Cnt Rev. Toxicol. 34, 351-384.
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, PC., Hahn, C., Hurtt, M.E., 1996a. The direct effect of hepatic peresxisome
pioliferators on rat Leydfg cell function in vitro. Fundam. Appi. Toxicol. 30,
102-108.
Liu, R.C., Hurtt, M.E., Cook, J.C., Biegel, L.B.,199bb. Effect of the pemxisome proliferator,
ammonium perfluorooctanoate (C8), on hepatic aromatase activity in adult male
CrI:CI)BR (CD) rats. Fundam. Appl. Toxicol. 30, 220-228.
Olsen, G.W., Burris, J.M., Mandel, J.H.. Zabel, L.R.,1999. Serum perfluorooctane sulfonate
and bepaic and lipid clue -:ail chen_istry tests in fluouochernical production
employees. J. Occup. Eriviroli. Meci. 41, 799--806,
Olsen, G.W., Burris, JM, Eirresman, D,J.. lryehlicb, J.W., Seacat, A.M., Butenhoff, J7.1
Zobel, LP, 2007. Half --life of serurn elim notion of per:luorooc an�sru:faruate,
perfluorohexanesulfonate, and perfluorooctanoate in retired fluorochemical
production workers. Environ. Health Perspect. 115, 1298-1305.
Pastoor, T.P., Lee, K.P., Perri, M.A.. Giilies, P.j., 1987. Biochemical and morphological
studies of ammonium perfluorooctanoate-induced hepatomegaly and peroxisome
proliferation. Exp. MoL Pathol. 47, 98-109.
Peden -Adams, M.M., FuDaky, J.G., Dabra. S, EuDaly, A., Heesemann, I.., Smythe, J., IKeil, D.E.,
2007. Suppression of humoral immunity following exposure to the perfluorinated
insecticide sulfluramid. J. Toxicol. Env ron. Health A 70, 1130-1141.
Peden -Adams, M M., ]Zeller, J.M., Fudaly, J.G., Berger, I., G'Ikeson, G.S., Keil, D.E., 2008.
Suppression of hmnorak immunity in mice following exposure to perfluorooctane
sulfonate. Tox col. Scf. 104, 144--154.
Qazf, M,ft., Xia, f , Bogcianstca, J.. Chang, .S.C., Ebresman, D.J.. Butenhoff, J,L Nelson, B.D
liePierre, (`f'., Abedi-Valugerdt, M., 2009. The atrophy and changes in the cellular
conmosiiaons of the thYrnos and spleen observed in mice subjected t7 short; -tern:
exposure to periluorooct<mesruifonate are high -close phenomena mediated in part by
peroxisome prokiferaLm-activated receptor -alpha (PPARalpha ). Toxicology 260, 68-76.
Rayne, S., Forest, K., 2009. Perfluo:oa_kyl su"f'mrc and carboxylic acids: a critical review
of physicochemical properties, levels and patterns in waters and wastewaters, and
treatment methods. J. Environ. Sci. Health A Toxic/Hazard. Subst. Environ. Eng. 44,
114 5-1199.
Saeeani, S., Pantano, S., Natoli, G.,'2002. p38-Dependent marking of inflammatory genes
for increased NF-kappa B recruitment. Nat Imhlunok. 3, 69-75.
Sambrook, J., Russell, D.W., 2001. Molecular Cloning: A Laboratory Manual, 3rd ed. Cold
Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
rc hrintz, ndl.L., Bacher, S., :Kracht, M., 2001, _11 kappa pa B-mclependeat control of NF- kappa B
activity by modulatory Phospbarylad.ians.'Frends Biocheari. Sci. 26,186-190,
DEQ-CFW 00004199

116 E. Corshd et al. / Toxicoto�,v and Applied Phannacotogy 250 (2011) 108-116
Schreiber, E.P., Matthias, M.M., Schaffner, W., 1989. Rapid detection of octameric
binding, proteins with mirn ex:ractprepared from a small number of cells, Nuc;ew
Acids Res. 17, 6419--6422.
Shu, H.. Wong, B., zhou, G., D, Y., Berger. J., Woods, JAM, Weight, S.D.. Cat, 2000.
Act:va ion of PPARalpha or ga nnra :educes secretion ofmatrix rnetaalloproteimse 9
but not interleukin 8 from human rnonocytic'I'_ 1P i cells. Biochem. Biophys, Res.
Commun. 267, 345-349.
Steenland, K., Fletcher, T., Savitz, D.A., 2010. Epidemiolo& evidence on the health
effects of perfluorooctanoic acid (PFOA). Environ. Health Perspea. 118,
1100-1108.
Suja, F., Pramanik, B.K., Zain, S.M., 2009. Contamination, Noaccmnulation and toxic
effects of perfluorinated chemical (PFCs) in the water environment: a review
paper. Water Sci. Technol. 60, 1533-1544.
Takacs, M.L., Abbott, B.D., 2007. Activation of mouse and human peroxisome
proliferator-activated receptors (alpha, beta/delta, gamma) by perfluorooctanoic
acid and perfluorooctane s elfonate.'Iox:col. Sci. 95, 108-117.
1'sirnihodimos, V., Liberopouios. F.... Ehsaf, M., 2009. Pleiotropic effeci.s of fenofibr re.
Cum Pharin. Ices 15. 517--r28.
Vanden Be: ghe, W., Ve: meulen, L., Delerive, P., De Bosscher, K., St aeIs, B., Haegeman, G'.,
2003. A Paradigm for gene regu:arion: inflammation, NF-kappaB and PPAR, Adv,
Exp. Med. Biol. 544, 181-196.
Verneulen, L., De Wilde, G., Notebaert, S., Vanden Bergee, W., Haegeman, G., 2002,
Regulation of sae transcriptional activity of the nuclear factor ,3ppaB p65 subunit.
Blodhem. Pharntacol. 64, 963--970.
Vermeulen, L., 7e Wilde, G., Van Danmre, P., Vanden Be:ghe, VJ..'Haegeman, G., 2003.
Transcriptional nscriptional activation of the NF kappaB p65 subunit by Tit ogee- am: stress
activated protein k nose-1 1,4SK1). EMBO J. 22, 1313 1324.
Viatour, P., Men- Alle, M.P., Bours, V., Chariot, A, 2005. Phosphorylation of NF-kappaB
and IkappaB proteins: implications in cancer and inflammation. Trends Biocirem.
Sc.. 30, 43-52.
Yang, Q., Me, Y., Enksson, A.M., Nelson, B.D., DePierre, J.vv., 2001. Furtner evidence for
the involvement of inhibition of cell proliferation and development in thymic and
splenic atrophy induced by the peroxisome proliferator perfluoroctarimc acid in
mice. Biochem. Pharmacol. 62, 1133-1140.
Yang, Q., Xie, Y., Alexson, S.E., Nelson, B.D., DePierre, .W., 2002a. Involvement of the
peroxisome proLferator-activated receptor alpha in the immunomodulat:on
caused by peroxisome prolifecal ors :n once. Bioche:ri. Pharnnacol. 63,
1893--1900.
Yang, Q., Abedi Va1ugeid:, M.. Xie, 4„ Zhao, X.Y„ Mbl:e,, G., Nelson, > D, DePierre, (W.,
20021). Potent suppression n of rhe adaptive immune response in mice upon dherary
exposure l:o the potent peroxsore p o:i.eraror, perfiuorooctanwc acid. [n1-
Immunopharmacol. 2, 389-397.
DEQ-CFW 00004200