Professional Documents
Culture Documents
Toxicology
journal homepage: www.elsevier.com/locate/toxicol
a r t i c l e i n f o a b s t r a c t
Article history: Having found previously that high-dose, short-term dietary exposure of mice to perfluorooctanesulfonate
Received 30 April 2009 (PFOS) or perfluorooctanoate (PFOA) suppresses adaptive immunity, in the present study we characterize
Received in revised form 10 June 2009 the effects of these fluorochemicals on the innate immune system. Male C57BL/6 mice receiving 0.02%
Accepted 11 June 2009
(w/w) PFOS or PFOA in their diet for 10 days exhibited a significant reduction in the numbers of total white
Available online 21 June 2009
blood cells (WBC), involving lymphopenia in both cases, but neutropenia only in response to treatment
with PFOA. Moreover, both compounds also markedly reduced the number of macrophages (CD11b+ cells)
Keywords:
in the bone marrow, but not in the spleen or peritoneal cavity. The ex vivo production of tumor necrosis
Innate immunity
Lipopolysaccharide
factor-␣ (TNF-␣) and interleukin 6 (IL-6) by peritoneal macrophages isolated from animals treated with
Perfluorooctanesulfonate PFOA or PFOS was increased modestly. Moreover, both fluorochemicals markedly enhanced the ex vivo
Perfluorooctanoate production of these same cytokines by peritoneal and bone marrow macrophages stimulated either in
Tumor necrosis factor alpha vitro or in vivo with lipopolysaccharide (LPS); whereas there was no such effect on splenic macrophages.
Interleukin 6 The serum levels of these inflammatory cytokines observed in response to in vivo stimulation with LPS
were elevated substantially by prior exposure to PFOA, but not by PFOS. None of these parameters of
innate immunity were altered in animals receiving a dietary dose of these compounds that was 20-fold
lower (0.001%, w/w). These findings reveal that in addition to suppressing adaptive immunity, high-dose,
short-term exposure of mice to either PFOS or PFOA augments inflammatory responses to LPS, a potent
activator of innate immunity.
© 2009 Elsevier Ireland Ltd. All rights reserved.
0300-483X/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.tox.2009.06.010
208 M.R. Qazi et al. / Toxicology 262 (2009) 207–214
Peden-Adams et al., 2008; Zheng et al., 2008; DeWitt et al., 2. Materials and methods
2009; Dong et al., 2009; Qazi et al., 2009). For instance, 7–10
2.1. Animals
days of dietary or gavage exposure (approximately 40 mg/kg/day)
of mice to either PFOS or PFOA causes severe atrophy of the Male C57BL/6 (H-2b ) mice (6–8 weeks old at the beginning of each experi-
thymus (a 63–67% reduction in size) and spleen (a 34–50% reduc- ment) were obtained from the Microbiology and Tumorbiology Center at Karolinska
tion), two central organs of adaptive immunity (Yang et al., 2000, Institute (Stockholm, Sweden) or from B&K (Solna, Sweden). This mouse strain was
2001; Zheng et al., 2008; Qazi et al., 2009). At the same time, chosen because of its wide use and paradigmatic status in immunological research.
Animals were housed in the animal facilities at the Wenner-Gren Institute, Stock-
the populations of splenocytes and thymocytes of all phenotypes holm University, at 22 ◦ C with a 12-h light/12-h dark cycle, 50% humidity and access
are reduced dramatically in size (by approximately 23–78% and to the diets indicated below and tap water ad libitum. Upon arrival, each mouse was
48–99%, respectively). The most pronounced reduction in the case placed in a separate cage (for individual monitoring of food intake) and allowed
of the thymus involves immature cells, indicating that prolifer- to acclimatize to this environment for at least 5 days before starting the dietary
exposure to PFOS or PFOA. In order to mitigate the stress associated with single
ation and differentiation are disturbed. With both compounds,
housing, each cage was provided with white paper towels and toilet paper rolls. All
the reduction in splenic cellularity is associated with a signif- of the experiments performed on these animals were pre-approved by the North-
icant reduction in specific humoral immune responses against ern Stockholm Ethical Committee for Animal Experimentation (approval numbers
foreign antigens, e.g., horse or sheep red blood cells (Yang et al., N380/03, N299/05, and N300/05).
2002a; Zheng et al., 2008; Peden-Adams et al., 2008; Dong et al.,
2.2. Preparation of the diet
2009).
We have also observed that the immunotoxic, but not hep- PFOS (the tetraammonium salt, 98% purity) and PFOA (the free acid, 96% purity)
atotoxic effects of PFOA are largely reversed within 7–10 days were purchased from Sigma–Aldrich Sweden AB, Stockholm, Sweden. These sub-
after termination of exposure, revealing that no permanent dam- stances were dissolved in 20 ml acetone and mixed with 100 g powdered RMI (E) FG
SQC diet (containing 2.71% fat, 14.38% protein and 61.73% carbohydrate; SDS, Special
age to the thymus or spleen has occurred (Yang et al., 2001).
Diets Services, Essex, UK) to obtain a concentration of 0.02% (w/w) (the dose that
Furthermore, it has been shown that the developing adap- elicits maximal responses, as demonstrated in our previous experiments (Yang et
tive immune system is also sensitive to the effects of PFOS, al., 2000, 2001, 2002a; Qazi et al., 2009)) or 0.001% (w/w). This chow was subse-
which results in functional defects in specific humoral antibody quently dried for 12–24 h in a ventilated hood, after which no smell of acetone was
responses detectable in adulthood (Keil et al., 2008). Moreover, detectable. Control food was prepared in the same manner, except that no xenobiotic
was added. Thereafter, the chow was shaped into cakes to facilitate quantitation of
both PFOS and PFOA appear to exert their effects on the murine
food consumption (since mice scatter powdered food all over their cages). All diets
adaptive immune system at least partially via the peroxisome were stored at 4 ◦ C prior to use.
proliferator-activated receptor-alpha (Yang et al., 2002b; Qazi et al.,
2009). 2.3. Dietary exposure to either PFOS or PFOA with and without induction of
In addition to their adaptive immunity, higher vertebrates endotoxemia
possess an innate branch of the immune system that responds In the first series of experiments, our objective was to characterize the direct
rapidly and non-specifically to invading microorganisms, lacks effects of PFOS and PFOA on the unactivated innate immune system. In this case,
memory, and involves a small number of germ-line encoded groups of four mice each received chow containing 0.02% or 0.001% PFOS or PFOA or
receptors (collectively referred to as Toll-like receptors, TLR) that a control diet (without added xenobiotic) for 10 consecutive days.
Subsequently, we examined whether exposure to PFOS or PFOA influences innate
recognize conserved molecular patterns associated with microbial
immune responses to LPS. For this purpose, the mice were again treated as described
pathogens (PAMPs) (e.g., bacterial lipopolysaccharide (LPS), viral above and, thereafter, on day 10, each of the three groups was divided randomly into
single- and double-stranded RNA, and CpG motifs in viral DNA two subgroups, one of which was injected through the tail vein with 0.1 ml sterile
(Medzhitov and Janeway, 1998; Kawai and Akira, 2005; Borghesi physiological saline containing 300 g LPS (Escherichia coli 055:B5; Sigma–Aldrich
and Milcarek, 2007). Previous studies on the effects of perfluo- Sweden, Stockholm), the optimal dose for induction of acute inflammation (Redl et
al., 1993), while the other group received the same volume of the vehicle alone via
rinated compounds on this branch of the immune system have the same route.
focused primarily on the effects of PFOS on the function of Nat-
ural Killer (NK) cells, which spontaneously recognize and destroy 2.4. Collection of blood, organs and peritoneal cells
both virus-infected and tumor cells (Cerwenka and Lanier, 2001).
High doses of this compound decrease, while low doses enhance Either directly following the period of exposure (in the first series of studies) or
2 h after administration of LPS (in the second series), the mice were bled by retroor-
the activity of these cells (Dong et al., 2009; Peden-Adams et bital puncture under light isoflurane anesthesia and thereafter sacrificed by cervical
al., 2008; Zheng et al., 2008). It has also been shown that ges- dislocation. Blood samples for preparation of plasma or serum were collected in cap-
tational exposure to PFOS can suppress the function of NK cells illary blood collection tubes (Microtainer BD Bioscience, NJ, USA) containing either
in adult offspring, suggesting that the development of NK activ- dipotassium EDTA or no additive. The blood samples collected in the latter tubes
were allowed to clot at 4 ◦ C and then centrifuged in order to obtain serum, which
ity is susceptible to the effects of this compound (Keil et al.,
was stored at −20 ◦ C until being assayed for cytokines. The liver, epididymal fat,
2008). spleen and thymus were collected and weighed.
The present study was designed to determine whether expo- Resident peritoneal exudate cells (primarily macrophages; see Section 3) were
sure to PFOS and/or PFOA also influences other cells of the innate collected by flushing the peritoneal cavities of the mice immediately following sac-
immune system. To this end, we examined potential alterations in rifice with 10 ml Dulbecco’s Modified Eagle Medium (D-MEM), pH 7.0, containing
4500 mg glucose and 110 mg sodium pyruvate/l and 4 mM l-glutamine (Invitrogen
the numbers of the different types of circulating innate immune AB, Stockholm, Sweden) and supplemented with 0.02% (w/v) sodium bicarbonate
cells, as well as the functional capacity of the macrophages (key and 100 IU penicillin and 100 g streptomycin/ml. These cells were washed once
cellular components of innate immune responses) present in with culture medium prior to subsequent manipulations.
the peritoneal cavity, bone marrow and spleen to produce the
pro-inflammatory cytokines tumor necrosis factor-alpha (TNF-␣) 2.5. Preparation of cell suspensions from the spleen and bone marrow
and interleukin-6 (IL-6), the key mediators of inflammation (Van Spleens were dissected out aseptically and thereafter gently teased apart
Amersfoort et al., 2003). In addition, these same parameters were with forceps in RPMI 1640 medium, pH 7.0, containing Glutamax (Invitrogen
monitored in association with activation of these innate immune AB, Stockholm, Sweden) supplemented with 15 mM HEPES, 100 IU penicillin and
cells by bacterial LPS, both in vitro and in vivo. Our present find- 100 g streptomycin/ml, and 0.02% (w/v) sodium bicarbonate. The spleen cells thus
obtained were washed twice with culture medium prior to subsequent manipula-
ings demonstrate that in contrast to their suppressive effects on the
tions.
adaptive branch of the murine immune system, high-dose, short- Bone marrow cells were flushed out of the tibia with sterile PBS, using a syringe
term exposure to either PFOS or PFOA activates certain components with a 27-gauge needle. These cells were also washed twice with RPMI 1640 medium
of innate immunity in mice. prior to use.
M.R. Qazi et al. / Toxicology 262 (2009) 207–214 209
As expected (Yang et al., 2000, 2002a; Qazi et al., 2009), dietary 3.3. The inflammatory responses of macrophages isolated from
administration of 0.02% PFOS or PFOA to mice for 10 days signif- the peritoneal cavity, bone marrow and spleen of mice upon
icantly enhanced the weight of their livers, whereas the weights exposure to lipopolysaccharide (LPS) in vitro are altered by prior
of the whole body, thymus, spleen and epididymal fat depots were in vivo treatment with PFOS or PFOA
reduced (not shown). Moreover, food consumption by the animals
exposed to PFOS or PFOA was reduced by approximately 25% (to To examine whether PFOS and/or PFOA might also influence
approximately 3 g/day) and 35% (2.6 g/day), respectively, in com- the functional properties of peritoneal, bone marrow and splenic
210 M.R. Qazi et al. / Toxicology 262 (2009) 207–214
Table 2
Effects of a 10-day dietary exposure of male C57BL/6 mice to 0.02% (w/w) of PFOA or PFOS on the numbers and relative proportions of macrophages (CD11b+ cells) among
the cells isolated from the peritoneal cavity, bone marrow and spleen.a .
Compartment Treatment
Total cells (106 ) Macrophages % of total Total cells (106 ) Macrophages % of total Total cells (106 ) Macrophages % of total
(106 ) (106 ) (106 )
Peritoneal cavity 3.7 ± 0.5 2.9 ± 0.4 78.1 ± 1.9 3.3 ± 0.6 2.8 ± 0.5 84.2 ± 1.2* , b 2.3 ± 0.5 2.0 ± 0.4 85.6 ± 1.0*
Bone marrow 8.9 ± 0.7 1.8 ± 0.2 20.1 ± 1.5 7.0 ± 0.4 0.8 + 0.1* 12.1 + 1.4* 4.9 ± 0.6* 0.6 ± 0.1* 12.2 ± 0.9*
Spleen 81.4 ± 6.9 10.7 ± 2.4 12.7 ± 1.7 50.5 ± 1.5* 7.3 + 0.7 14.4 + 1.2 36.2 ± 4.0* 7.9 ± 0.1 21.5 ± 1.5* , c
a
Peritoneal, bone marrow and spleen cells were isolated and incubated with a monoclonal antibody directed towards CD11b, a selective cell-surface marker for macrophages.
Subsequently, the numbers and percentages of macrophages present were determined by flow cytometry.
b
All values are means ± SE (standard error) for eight animals in each group. Differences were analyzed for statistical significance employing the U-test.
c
Significantly different (p < 0.05) from the corresponding values for both control and PFOS-treated animals (ANOVA test).
*
Statistically significant as compared with the corresponding control group.
macrophages, we determined the ability of these cells to produce As expected, in vitro stimulation of cells isolated from the
TNF-␣ and IL-6 ex vivo with and without stimulation with LPS peritoneal cavity, bone marrow and spleen of control mice with
(100 ng/ml). Since the isolation of macrophages would introduce LPS potently enhanced the production of both TNF-␣ and IL-6
uncertainties (i.e., concerning the extent of recovery and the func- (Fig. 1a–f). Moreover, in vivo exposure to PFOS potentiated these ex
tional state of the isolated cells) into the analysis and since in vivo responses of peritoneal and bone marrow cells to LPS (Fig. 1a,
response to LPS macrophages are the primary producers of TNF- b, d, and e), while, at the same time, attenuating the correspond-
␣ and IL-6 (Freudenberg et al., 1986; Salkowski et al., 1995), we ing production of TNF-␣ by spleen cells (Fig. 1c). A similar pattern
performed these experiments on crude cell suspensions. As shown was observed following treatment with PFOA, except that in this
in Fig. 1, in the absence of this potent activator of macrophages, case, the ex vivo production of TNF-␣ and IL-6 by spleen cells in
peritoneal cells isolated from mice treated with either compound response to in vitro stimulation with LPS was also significantly
produced significantly enhanced levels of both of these cytokines enhanced (Fig. 1a–f). This latter enhancement can be accounted for
(Fig. 1a and d); the bone marrow cells from animals administered partially by the increase in the proportion of macrophages among
PFOA secreted elevated levels of TNF-␣ (Fig. 1b); whereas the pro- the splenocytes. These findings indicate that in vivo treatment
duction of this latter cytokine by spleen cells was reduced by both with either of these fluorochemicals alters the ex vivo responses
PFOS and PFOA (Fig. 1c and f). of macrophages located in all three of these compartments to LPS.
Fig. 1. High-dose, short-term exposure of mice to PFOS or PFOA alters the ex vivo production of TNF-␣ and IL-6 by cells isolated from the peritoneal cavity, bone marrow and
spleen of mice, both with and without subsequent stimulation in vitro by lipopolysaccharide (LPS). Male C57BL/6 mice received a normal (control) diet or diet containing
0.02% PFOS or PFOA for 10 days. Peritoneal, bone marrow and spleen cells were then collected and cultured in D-MEM or RPMI 1640 medium for 18 h as described in Section
2. Thereafter, culture media were collected, fresh medium containing 100 ng LPS/ml added and the cells then incubated for an additional 18 h, following which the culture
media were harvested. The levels of TNF-␣ (a–c) and IL-6 (d–f) in all of the culture media collected (both prior to and after activation by LPS) were determined by ELISA
procedures. The values presented are means ± SE (standard errors) for a total of eight mice in each group, examined in two independent experiments. For clarity, the actual
concentrations of TNF-␣ or IL-6 for unstimulated cells are shown in parentheses above the corresponding bars. *Statistically significant as compared with the corresponding
control group.
M.R. Qazi et al. / Toxicology 262 (2009) 207–214 211
Fig. 2. Exposure of mice to PFOS or PFOA prior to induction of endotoxemia by intravenous injection of lipopolysaccharide (LPS) potentiates the ex vivo production of TNF-␣
and IL-6 by cells isolated from the peritoneal cavity and bone marrow, but not from the spleen. Male C57BL/6 mice received a normal (control) diet or diet containing 0.02%
PFOS or PFOA for 10 days. On day 10, half of the mice receiving each treatment were injected (inj.) through the tail vein with 300 g lipopolysaccharide (LPS) and 2 h later
peritoneal, bone marrow and spleen cells were collected from all of the animals and cultured as described in Section 2. The levels of TNF-␣ (a–c) and IL-6 (d–f) in the culture
media collected were then determined by ELISA. The values presented are means ± SE (error bars) for a total of eight mice in each group, examined in two independent
experiments. *Statistically significant as compared with the corresponding control group.
3.4. In vivo cytokine responses associated with els of both cytokines, as expected; and pretreatment with either
lipopolysaccharide-induced inflammation are also altered by PFOS or PFOA potentiated these responses to LPS, although not to
high-dose, short-term exposure to PFOS or PFOA a statistically significant extent in the case of PFOS (Table 3). Thus,
the enhancement by these fluorochemicals of the ex vivo respon-
The observations documented above raised the question as to siveness of peritoneal and bone marrow macrophages to LPS is,
whether exposure to PFOS and/or PFOA renders mice more sus-
ceptible to inflammatory responses associated with infection. As
Table 3
shown in Fig. 2a–f, when, as a positive control, mice not exposed to Prior 10-day dietary exposure of male C57BL/6 mice to 0.02% (w/w) of PFOS or PFOA
a perfluorochemical were challenged with an intravenous injection potentiates the elevations in circulating levels of the pro-inflammatory cytokines,
of LPS (300 g/mouse) 2 h prior to sacrifice, the ex vivo produc- tumor necrosis factor-alpha (TNF-␣) and interleukin-6 (IL-6) that occur in response
tion of both TNF-␣ and IL-6 by peritoneal and bone marrow cells to challange with bacterial lipopolysaccharide (LPS).a .
and splenocytes was, as expected, enhanced markedly (compare Treatment Serum levels (pg/ml)
all bars labelled Control + LPS inj. in Fig. 2a–f to all Control bars in
TNF-␣ IL-6
Fig. 1a–f). At the same time, in essential agreement with our in
None (control) 1.3 ± 0.6 2.0 ± 0.6
vitro findings (see above), pretreatment with PFOS or PFOA prior
PFOS 2.0 ± 0.5 24 ± 10* , b
to the challenge with LPS further potentiated production of these PFOA 2.1 ± 0.6 5.0 ± 2.0*
cytokines by both peritoneal and bone marrow cells (Fig. 2a, b, d None (control) + LPS injection 4900 ± 1640 212,000 ± 23,000
and e), while attenuating the ex vivo responses of splenocytes to PFOS + LPS injection 6810 ± 1430 251,000 ± 29,000
LPS-induced inflammation (Fig. 2c and f). PFOA + LPS injection 28,100 ± 6830* 407,000 ± 51,000*
indeed, reflected in changes in the circulating levels of these two to either of these compounds, both the relative proportion and total
inflammatory cytokines in vivo. number of macrophages in the bone marrow are reduced, whereas
no such effects are observed in the peritoneal cavity and spleen.
This organ specificity may reflect the heterogeneity of macrophage
3.5. Dietary exposure to a 20-fold lower dose (0.001%, w/w) of
populations, which contain subpopulations that may be derived
PFOS or PFOA does not alter any of these parameters of innate
from different precursor cells (Naito et al., 1996; Gordon, 2002;
immunity in mice
Gordon and Taylor, 2005). For instance, macrophages derived from
promonocytes in the bone marrow are short-lived and do not pos-
When the dietary dose of PFOS or PFOA to which mice were
sess the capacity to proliferate; whereas the resident macrophages
exposed was reduced 20-fold, no alterations in organ weights, cir-
in peripheral tissues, derived from granulocyte/macrophage colony
culating numbers of blood cells or plasma levels of cytokines or
forming cells are long-lived and can replenish themselves through
the ex vivo production of cytokines by cells isolated from the peri-
proliferation (Naito et al., 1996; Gordon, 2002; Gordon and Taylor,
toneal cavity, bone marrow or spleen, with or without stimulation
2005). Thus, it is certainly possible that bone marrow macrophages
by LPS, were observed (data not shown). Thus, the effects of short-
are susceptible to, e.g., induction of necrosis and/or apoptosis by
term treatment with PFOS or PFOA on murine innate immunity
PFOS and PFOA, whereas the resident macrophages in the peri-
described here appear to be a high-dose phenomenon.
toneal cavity and spleen are resistant.
Our third significant finding is that cells isolated from the peri-
4. Discussion toneal cavity and bone marrow, but not from the spleen of mice
subjected to high-dose, short-term exposure to either PFOS or
We demonstrate here that dietary treatment of mice with 0.02% PFOA produce enhanced levels of the pro-inflammatory cytokines
PFOS or PFOA for 10 days results in significant alterations in a num- TNF-␣ and IL-6 in response to stimulation by LPS in vitro. Since
ber of parameters associated with the innate immune system. Our macrophages are the major producers of these cytokines in these
initial observation that both of these treatments lead to severe three compartments, we conclude that exposure to either PFOS or
leukopenia combined with lymphopenia was expected, since it is PFOA alters the functionality of these cells.
well established that this regimen of exposure induces pronounced Together, these findings suggest that PFOS and PFOA may sen-
atrophy of the thymus and spleen (Yang et al., 2000, 2001; Zheng et sitise macrophages to external insults, in particular by infectious
al., 2008; Dong et al., 2009; Qazi et al., 2009). It is noteworthy, how- agents. In fact, this proposal receives strong support from two
ever, that only PFOA causes neutropenia under these experimental additional observations described here. First, exposure of mice
conditions, which indicates that there are differences in the mech- to PFOS or PFOA prior to the induction of endotoxemia through
anisms by which these compounds exert their biological effects. intravenous injection of LPS significantly enhances the ex vivo pro-
Moreover, it implies that at least certain fluorochemicals can reduce duction of TNF-␣ and IL-6 by the cells subsequently isolated from
the number of circulating neutrophils, the cellular mediators of the peritoneal cavity and bone marrow, but does not alter the cor-
innate immunity that first encounter infectious agents and play a responding production by splenocytes.
pivotal role in limiting the spread of microbial infections (Nathan, This lack of potentiation of inflammatory responses in the spleen
2006). Circulating neutrophils have a half-life of only 11.4 h in mice is consistent with the general concept that macrophages adapt
and must constantly be produced in large numbers in the bone mar- functionally to the requirements of the tissues in which they inhabit
row (von Vietinghoff and Ley, 2008). PFOA may elicit neutropenia (Naito et al., 1996; Takahashi et al., 1996; Gordon and Taylor, 2005).
through direct effects on the bone marrow itself. This suggestion The spleen functions as a secondary immune organ that responds to
gains some support from our observation that exposure to PFOA, blood-born antigens and, at the same time, serves as a graveyard for
but not PFOS reduces the total number of cells in the bone mar- aging erythrocytes. These functions are performed by several spe-
row. cific subpopulations of macrophages, e.g., red pulp (F4/80 positive),
At the same time, it is possible that PFOA induces neutropenia marginal metallophilic (MOMA-1-positive), marginal zone (ER-TR-
indirectly by the induction of severe food restriction. Indeed, our positive) and white pulp (MOMA-2-positive) macrophages (Naito
studies have demonstrated that dietary exposure of mice to 0.02% et al., 1996; Takahashi et al., 1996).
(w/w) PFOS or PFOA reduces their food intake to approximately Of considerable interest in this context is the observation that
65% and 75%, respectively, of the amount consumed by untreated CD14, a 55-kDa glycoprotein which, together with TLR-4 and MD-
animals (Qazi et al., 2009; and present study). Severe caloric restric- 2, acts as a co-receptor for the detection of LPS (Kitchens, 2000),
tion is known to substantially reduce the numbers, chemotaxis and is expressed constitutively on the surface of bone marrow and
phagocytic activity of neutrophils in the bone marrow pool, as well peritoneal, but not splenic macrophages (Matsuura et al., 1994).
as in the peripheral blood of rats (Suda et al., 1976; Ogawa et al., Moreover, LPS can induce the expression of this molecule by splenic
1985; Garcia and Barbieri, 1986; Levin et al., 1993; Slapnickova and macrophages (Matsuura et al., 1994). We are presently testing
Berger, 2002). In addition, moderate caloric restriction is known the hypothesis that PFOS and PFOA prevent or, at least, attenu-
to reduce thymic and splenic weight and cellularity, as well as ate the expression of the CD14 receptor by splenic macrophages,
the numbers of CD4+ CD8+ thymocytes and splenic B-lymphocytes thereby inhibiting their ability to produce inflammatory cytokines
(Hishinuma et al., 1990; Poetschke et al., 2000; Savino, 2002; Martin in response to activation by LPS.
et al., 2008) and to alter the nucleolar structure and function in The second and more significant indication that PFOS and PFOA
lymphocytes (Berger et al., 2005). Together, these observations can sensitize macrophages to external insults is our finding that the
indicate that while moderate food restriction affects primarily lym- increases in circulating levels of TNF-␣ and IL-6 caused by an in vivo
phocytes, severe food restriction can exert adverse effects on both challenge with LPS can be enhanced even further by prior exposure
lymphocytes and neutrophils. We are now examining the poten- to either of these perfluorochemicals. In fact, this observation sug-
tial contribution of caloric restriction to the reduced number of gests that not only macrophages in many different organs (with
circulating neutrophils observed in mice exposed to PFOA. the exception of the spleen), but also possibly the entire mononu-
Our second major finding is that high-dose, short-term expo- clear phagocyte system (MPS) is sensitized by such exposure. This
sure to PFOS or PFOA alters the absolute and relative sizes of the proposal could be tested by determining the LD50 for LPS in mice
macrophage populations in the peritoneal cavity, bone marrow and with and without exposure to PFOS or PFOA, but this approach is
spleen in an organ-dependent manner. For example, upon exposure ethically unacceptable.
M.R. Qazi et al. / Toxicology 262 (2009) 207–214 213
The mechanism(s) via which high-dose, short-term expo- mone products influence immune responses). Investigations of this
sure of mice to PFOS or PFOA enhances cytokine production by nature are currently being performed in our laboratory.
macrophages in response to LPS remains to be elucidated. How-
ever, one possibility is that this phenomenon may reflect elevated Conflicts of interest
levels of apoptosis and/or necrosis induced by high doses of PFOS or
PFOA (Martin et al., 2007; Cui et al., 2009). Under normal physiolog- None of the authors have any conflicting interests.
ical conditions, macrophages employ various cell surface molecules
and receptors to internalise apoptotic cells and thereby remove Acknowledgments
them from the system without causing inflammation (Krysko et
al., 2006). In contrast, in connection with their internalization This study was financed by an unrestricted research grant from
and clearance of necrotic cells, which involves macropinocytosis, the 3M Company (St. Paul, Minnesota, USA). We would also like to
macrophages are provoked to increase their production of pro- thank Jarl Olsson for his skilful technical assistance in diet prepara-
inflammatory cytokines (Krysko et al., 2006). In addition, release tion and animal care.
of heat-shock proteins and/or large amounts of uric acid by the
necrotic cells themselves can elicit pro-inflammatory responses in
References
macrophages via both TLR2 (the receptor that recognizes bacterial
lipoproteins) and TLR4 (which recognizes LPS) (Krysko et al., 2006). Andersen, M.E., Butenhoff, J.L., Chang, S.C., Farrar, D.G., Kennedy Jr., G.L., Lau,
This possibility that clearance of excessive numbers of necrotic or, C., Olsen, G.W., Seed, J., Wallace, K.B., 2008. Perfluoroalkyl acids and related
perhaps, even apoptotic cells by macrophages during high-dose chemistries—toxicokinetics and modes of action. Toxicol. Sci. 102, 3–14.
Aringer, M., Smolen, J.S., 2003. SLE—complex cytokine effects in a complex autoim-
exposure to PFOS or PFOA renders the macrophages highly respon- mune disease: tumor necrosis factor in systemic lupus erythematosus. Arthritis
sive to LPS obviously requires further investigation. Although the Res. Ther. 5, 172–177.
potential consequences of such sensitization for the animal itself Berger, J., Machackova, M., Berger, Z., 2005. Effects of feed restriction on the nucle-
olar structure and function in lymphocytes. Basic Clin. Pharmacol. Toxicol. 97,
remain to be elucidated, potentiation of the production of TNF-␣ 236–237.
and IL-6 in response to LPS by pretreatment with PFOS or PFOA Borghesi, L., Milcarek, C., 2007. Innate versus adaptive immunity: a paradigm past
suggests that these compounds might render the animal more sus- its prime? Cancer Res. 67, 3989–3993.
Calafat, A.M., Wong, L.Y., Kuklenyik, Z., Reidy, J.A., Needham, L.L., 2007. Polyfluo-
ceptible to disease conditions (e.g., systemic autoimmune diseases)
roalkyl chemicals in the U.S. population: data from the National Health and
in which TNF-␣ and IL-6 play a pivotal role (Ishihara and Hirano, Nutrition Examination Survey (NHANES) 2003–2004 and comparisons with
2002; Aringer and Smolen, 2003). NHANES 1999–2000. Environ. Health Perspect. 115, 1596–1602.
Canadian Government Department of the Environment, 2008. Pefluorooctanesul-
One general observation here is that following short-term
fonate and its salts and certain other compounds regulations, vol. 142. Canada
dietary administration to mice at a dose of 0.02% (w/w), in almost Gazette, Part II, pp. 1322–1325.
all respects the effects of PFOS on the innate branch of the immune Cerwenka, A., Lanier, L.L., 2001. Natural killer cells, viruses and cancer. Nat. Rev.
system are less pronounced than those of PFOA. Furthermore, we Immunol. 1, 41–49.
Cui, L., Zhou, Q.F., Liao, C.Y., Fu, J.J., Jiang, G.B., 2009. Studies on the toxicological
have recently demonstrated that with this same short-term, high- effects of PFOA and PFOS on rats using histological observation and chemical
dose regimen of exposure, PFOA causes more severe deleterious analysis. Arch. Environ. Contam. Toxicol. 56, 338–349.
effects on the adaptive immune system than does PFOA (Qazi et al., DeWitt, J.C., Shnyra, A., Badr, M.Z., Loveless, S.E., Hoban, D., Frame, S.R., Cunard, R.,
Anderson, S.E., Meade, B.J., Peden-Adams, M.M., Luebke, R.W., Luster, M.I., 2009.
2009). Together, these findings indicate that with this regimen of Immunotoxicity of perfluorooctanoic acid and perfluorooctane sulfonate and
treatment, PFOS is less potent than PFOA in causing immunotoxi- the role of peroxisome proliferator-activated receptor alpha. Crit. Rev. Toxicol.
cological effects in mice. 39, 76–94.
Dong, G.H., Zhang, Y.H., Zheng, L., Liu, W., Jin, Y.H., He, Q.C., 2009. Chronic effects
Finally, we found that dietary exposure of mice to a 20-fold lower of perfluorooctanesulfonate exposure on immunotoxicity in adult male C57BL/6
dose (i.e., 0.001%, w/w) of PFOS or PFOA for 10 days does not alter mice. Arch. Toxicol. (April 3 [Epub ahead of print]).
any of the parameters examined here. Clearly, as is also the case Freudenberg, M.A., Keppler, D., Galanos, C., 1986. Requirement for
lipopolysaccharide-responsive macrophages in galactosamine-induced
with respect to adaptive immunity (Qazi et al., 2009), the short-
sensitization to endotoxin. Infect. Immun. 51, 891–895.
term effects of these compounds on the innate immune system of Garcia, P.B., Barbieri, D., 1986. Influence of protein malnutrition on the phagocytic
mice is a high-dose phenomenon. With respect to environmental function of neutrophils in rats. Arch Latinoam Nutr. 36, 662–677.
Gordon, S., 1998. The role of the macrophage in immune regulation. Res. Immunol.
relevance, it should be noted that the levels of PFOS or PFOA in our
149, 685–688.
mice receiving a dietary dose of 0.001% are about 1000 times higher Gordon, S., 2002. Pattern recognition receptors: doubling up for the innate immune
than those observed in general human populations (Lau et al., response. Cell 111, 927–930.
2007) and, furthermore, several recent biomonitoring studies have Gordon, S., Taylor, P.R., 2005. Monocyte and macrophage heterogeneity. Nat. Rev.
Immunol. 5, 953–964.
demonstrated that the serum concentrations of these compounds Hishinuma, K., Nishimura, T., Konno, A., Hashimoto, Y., Kimura, S., 1990. Augmenta-
in the general population have declined significantly in recent years tion of mouse immune functions by dietary restriction: an investigation up to 1
(Calafat et al., 2007; Olsen et al., 2008; Spliethoff et al., 2008). Thus, year of age. Ann. Nutr. Metab. 34, 76–84.
Ishihara, K., Hirano, T., 2002. IL-6 in autoimmune disease and chronic inflammatory
although experiments analogous to the ones described here, but proliferative disease. Cytokine Growth Factor Rev. 13, 357–368.
involving low-dose, long-term exposure are obviously warranted, Kawai, T., Akira, S., 2005. Pathogen recognition with Toll-like receptors. Curr. Opin.
it appears unlikely that the effects on the innate immune system Immunol. 17, 338–344.
Keil, D.E., Mehlmann, T., Butterworth, L., Peden-Adams, M.M., 2008. Gestational
of mice documented here represent an immediate environmental exposure to perfluorooctane sulfonate suppresses immune function in B6C3F1
concern. mice. Toxicol. Sci. 103, 77–85.
The present findings demonstrate for the first time that at least Kissa, E., 2001. Fluorinated Surfactants and Repellents. Marcel Dekker, New York.
Kitchens, R.L., 2000. Role of CD14 in cellular recognition of bacterial lipopolysaccha-
certain perfluorinated compounds can exert adverse effects on the
rides. Chem. Immunol. 74, 61–82.
innate branch of the immune system, an important defence system Krysko, D.V., D’Herde, K., Vandenabeele, P., 2006. Clearance of apoptotic and necrotic
found in all plants and animals. The mechanisms underlying the cells and its immunological consequences. Apoptosis 11, 1709–1726.
Lau, C., Anitole, K., Hodes, C., Lai, D., Pfahles-Hutchens, A., Seed, J., 2007. Perfluo-
immunotoxic effects of both PFOS and PFOA on both the adaptive
roalkyl acids: a review of monitoring and toxicological findings. Toxicol. Sci. 99,
and innate branches of the immune system require further elu- 366–394.
cidation. Toxicokinetic investigations will help elucidate whether Lehmler, H.J., 2005. Synthesis of environmentally relevant fluorinated surfactants—a
these compounds exert their immunotoxic effects directly (e.g., by review. Chemosphere 58, 1471–1496.
Levin, S., Semler, D., Ruben, Z., 1993. Effects of two weeks of feed restriction on
being accumulated in primary and/or secondary immune organs) some common toxicologic parameters in Sprague–Dawley rats. Toxicol. Pathol.
or indirectly (e.g., by acting on the endocrine organs whose hor- 21, 1–14.
214 M.R. Qazi et al. / Toxicology 262 (2009) 207–214
Martin, M.T., Brennan, R.J., Hu, W., Ayanoglu, E., Lau, C., Ren, H., Wood, C.R., Cor- Savino, W., 2002. The thymus gland is a target in malnutrition. Eur. J. Clin. Nutr. 56
ton, J.C., Kavlock, R.J., Dix, D.J., 2007. Toxicogenomic study of triazole fungicides (Suppl. 3), 46–49.
and perfluoroalkyl acids in rat livers predicts toxicity and categorizes chemicals Slapnickova, M., Berger, J., 2002. Rat neutrophil phagocytosis following feed restric-
based on mechanisms of toxicity. Toxicol. Sci. 97, 595–613. tion. Comp. Clin. Path. 11, 172–177.
Martin 2nd., L.B., Navara, K.J., Bailey, M.T., Hutch, C.R., Powell, N.D., Sheridan, J.F., Sohlenius, A.K., Andersson, K., Bergstrand, A., Spydevld, O., DePierre, J.W., 1994.
Nelson, R.J., 2008. Food restriction compromises immune memory in deer mice Effects of perfluorooctanoic acid—a potent peroxisome proliferator in rat—on
(Peromyscus maniculatus) by reducing spleen-derived antibody-producing B cell Morris hepatoma 7800C1 cells, a rat cell line. Biochim. Biophys. Acta 1213, 63–
numbers. Physiol. Biochem. Zool. 81, 366–372. 74.
Matsuura, K., Ishida, T., Setoguchi, M., Higuchi, Y., Akizuki, S., Yamamoto, S., 1994. Sohlenius, A.K., Eriksson, A.M., Högström, C., Kimland, M., DePierre, J.W., 1993.
Upregulation of mouse CD14 expression in Kupffer cells by lipopolysaccharide. Perfluorooctane sulfonic acid is a potent inducer of peroxisomal fatty acid beta-
J. Exp. Med. 179, 1671–1676. oxidation and other activities known to be affected by peroxisome proliferators
Medzhitov, R., Janeway Jr., C.A., 1998. Innate immune recognition and control of in mouse liver. Pharmacol. Toxicol. 72, 90–93.
adaptive immune responses. Semin. Immunol. 10, 351–353. Sohlenius, A.K., Lundgren, B., DePierre, J.W., 1992. Perfluorooctanoic acid has persis-
Naito, M., Umeda, S., Yamamoto, T., Moriyama, H., Umezu, H., Hasegawa, G., Usuda, H., tent effects on peroxisome proliferation and related parameters in mouse liver.
Shultz, L.D., Takahashi, K., 1996. Development, differentiation, and phenotypic J. Biochem. Toxicol. 7, 205–212.
heterogeneity of murine tissue macrophages. J. Leukoc. Biol. 59, 133–138. Spliethoff, H.M., Tao, L., Shaver, S.M., Aldous, K.M., Pass, K.A., Kannan, K., Eadon, G.A.,
Nathan, C., 2006. Neutrophils and immunity: challenges and opportunities. Nat. Rev. 2008. Use of newborn screening program blood spots for exposure assessment:
Immunol. 6, 173–182. declining levels of perfluorinated compounds in New York State infants. Environ.
OECD, 2002. Hazard Assessment of Perfluorooctane Sulfonate (PFOS) and its Salts. Sci. Technol. 42, 5361–5367.
Paris, France. Suda, A.K., Mathur, M., Deo, K., Deo, M.G., 1976. Kinetics of mobilization of neu-
OECD, 2007. Report No. 21. Lists of PFOS, PFAS, PFOA, PFCA, Related Compounds and trophils and their marrow pool in protein-calorie deficiency. Blood 48, 865–
Chemicals that May Degrade to PFCA. Organisation for Economic Co-operation 875.
and Development, Paris. Takahashi, K., Naito, M., Takeya, M., 1996. Development and heterogeneity of
Ogawa, Y., Matsumoto, K., Kamata, E., Ikeda, Y., Kaneko, T., 1985. Effect of feed restric- macrophages and their related cells through their differentiation pathways.
tion on the peripheral blood and bone marrow cell counts of Wistar rats. Jikken Pathol. Int. 46, 473–485.
Dobutsu 34, 407–416. United States Environmental Protection Agency, 2002. Rules and Regulations. United
Olsen, G.W., Burris, J.M., Ehresman, D.J., Froehlich, J.W., Seacat, A.M., Butenhoff, J.L., States Federal Register, vol. 67, pp. 72854–72867.
Zobel, L.R., 2007. Half-life of serum elimination of perfluorooctanesulfonate, United States Environmental Protection Agency, 2006. Proposed rules. United States
perfluorohexanesulfonate, and perfluorooctanoate in retired fluorochemical Federal Register, vol. 71, pp. 12311–12324.
production workers. Environ. Health Perspect. 115, 1298–1305. Van Amersfoort, E.S., Van Berkel, T.J., Kuiper, J., 2003. Receptors, mediators, and
Olsen, G.W., Mair, D.C., Church, T.R., Ellefson, M.E., Reagen, W.K., Boyd, T.M., Herron, mechanisms involved in bacterial sepsis and septic shock. Clin. Microbiol. Rev.
R.M., Medhdizadehkashi, Z., Nobiletti, J.B., Rios, J.A., Butenhoff, J.L., Zobel, L.R., 16, 379–414.
2008. Decline in perfluorooctanesulfonate and other polyfluoroalkyl chemicals von Vietinghoff, S., Ley, K., 2008. Homeostatic regulation of blood neutrophil counts.
in American Red Cross adult blood donors, 2000–2006. Environ. Sci. Technol. 42, J. Immunol. 181, 5183–5188.
4989–4995. Xia, Z., Triffitt, J.T., 2006. A review on macrophage responses to biomaterials. Biomed.
Peden-Adams, M.M., Keller, J.M., Eudaly, J.G., Berger, J., Gilkeson, G.S., Keil, D.E., 2008. Mater. 1, R1–R9.
Suppression of humoral immunity in mice following exposure to perfluorooc- Yang, Q., Xie, Y., Depierre, J.W., 2000. Effects of peroxisome proliferators on the
tane sulfonate. Toxicol. Sci. 104, 144–154. thymus and spleen of mice. Clin. Exp. Immunol. 122, 219–226.
Poetschke, H.L., Klug, D.B., Perkins, S.N., Wang, T.T., Richie, E.R., Hursting, S.D., 2000. Yang, Q., Xie, Y., Eriksson, A.M., Nelson, B.D., DePierre, J.W., 2001. Further evidence for
Effects of calorie restriction on thymocyte growth, death and maturation. Car- the involvement of inhibition of cell proliferation and development in thymic
cinogenesis 21, 1959–1964. and splenic atrophy induced by the peroxisome proliferator perfluoroctanoic
Qazi, M.R., Xia, Z., Bogdanska, J., Chang, S.-C., Ehresman, D.J., Butenhoff, J.L., Nel- acid in mice. Biochem. Pharmacol. 62, 1133–1140.
son, B.D., DePierre, J.W., Abedi-Valugerdi, M., 2009. The atrophy and changes in Yang, Q., Abedi-Valugerdi, M., Xie, Y., Zhao, X.-Y., Möller, G., Nelson, B.D., DePierre,
the cellular compositions of the thymus and spleen observed in mice subjected J., 2002a. Potent suppression of the adaptive immune response in mice upon
to short-term exposure to perfluorooctanesulfonate are high-dose phenomena dietatry exposure to the potent peroxisome proliferator, perfluorooctanoic acid.
mediated in part by peroxisome proliferator-activated receptor-alpha (PPAR␣). Int. Immunol. Pharmacol. 2, 389–397.
Toxicology 260, 68–76. Yang, Q., Xie, Y., Alexson, S.E., Nelson, B.D., DePierre, J.W., 2002b. Involvement of
Redl, H., Bahrami, S., Schlag, G., Traber, D.L., 1993. Clinical detection of LPS and animal the peroxisome proliferator-activated receptor alpha in the immunomodulation
models of endotoxemia. Immunobiology 187, 330–345. caused by peroxisome proliferators in mice. Biochem. Pharmacol. 63, 1893–1900.
Salkowski, C.A., Neta, R., Wynn, T.A., Strassmann, G., van Rooijen, N., Vogel, S.N., 1995. Zheng, L., Dong, G.H., Jin, Y.H., He, Q.C., 2008. Immunotoxic changes associated with
Effect of liposome-mediated macrophage depletion on LPS-induced cytokine a 7-day oral exposure to perfluorooctanesulfonate (PFOS) in adult male C57BL/6
gene expression and radioprotection. J. Immunol. 155, 3168–3179. mice. Arch. Toxicol. (October 21 [Epub ahead of print]).