Mi

Xa

b

c

d

e

f

g

h

a

ARRAA

KAIIO

1

bi2i2

GT

cw

0h

Toxicology 306 (2013) 59– 67

Contents lists available at SciVerse ScienceDirect

Toxicology

jou rn al hom epage: www.elsev ier .com/ locate / tox ico l

aternal exposure to airborne particulate matter causes postnatalmmunological dysfunction in mice offspring

inru Honga,b,c,∗, Chaobin Liud,∗∗, Xiaoqiu Chene, Yanfeng Songa, Qin Wangf, Ping Wangg, Dian Huh

Department of Obstetrics and Gynecology, Fuzhou General Hospital, Fujian, ChinaFuzhou Clinic Medical College, Fujian Medical University, Fujian, ChinaDongfang Affiliated Hospital of Xiamen University, No. 156 North Xi’erhuan Road, Fuzhou 350025, Fujian, ChinaDepartment of Obstetrics and Gynecology, Fujian Provincial Maternal and Child Health Hospital, No. 18 Daoshan Road, Fuzhou 350001, Fujian, ChinaFujian Central Station of Environmental Monitoring, No. 138 South Fufei Road, Fuzhou 350003, Fujian, ChinaFuzhou Product Quality Monitoring Bureau, No. 83 Yangnan Street, Fuzhou 350025, Fujian, ChinaThe Official Hospital of 92403 Unit of PLA, No. 34 Duihu Road, Fuzhou 350007, ChinaDepartment of Obstetrics and Gynecology, Changzheng Hospital, Second Military Medical University, No. 415 Fengyang Road, Shanghai 200003, China

r t i c l e i n f o

rticle history:eceived 30 November 2012eceived in revised form 15 January 2013ccepted 1 February 2013vailable online 14 February 2013

eywords:irborne particulate matter

n utero exposuremmunotoxicityffspring

a b s t r a c t

Evidence suggests that prenatal exposure to air pollution affects the ontogeny and development of thefetal immune system. The aim of this study was to investigate the effect of maternal exposure to airborneparticulate matter (PM) on immune function in postnatal offspring.

Pregnant female ICR mice were intralaryngopharyngeally administered with 30 �l of phosphatebuffered solution (the control group) or resuspended PM of Standard Reference Material 1649a at 0.09(low), 0.28 (medium), 1.85 (high) or 6.92 (overdose) �g/�l once every three days from day 0 to 18 ofpregnancy (n = 8–10). Offspring were sacrificed on postnatal day 30. Interleukin-4 and interferon-� levelsin plasma and splenocytes, splenic lymphocyte proliferation, and expressions of GATA-3 and T-bet mRNAin the spleen were tested. The spleen and thymus were histopathologically examined.

The offspring of the medium, high and overdose PM-exposed dams showed significantly suppressedsplenocyte proliferation. Decreased interferon-� and increased interleukin-4 levels in the blood and

splenocytes, and lowered T-bet and elevated GATA-3 mRNA expressions were found in the spleen inthe medium, high and overdose groups when compared with the control or low dose group (P < 0.05).Histopathology revealed prominent tissue damage in the spleen and thymus in the overdose group.

These results suggest that exposure of pregnant mice to PM modulates the fetal immune system, result-ing in postnatal immune dysfunction by exacerbation of Thl/Th2 deviation. This deviation is associatedwith altered T-bet and GATA-3 gene expressions.

. Introduction

As a result of extensive urbanization and an increasing num-er of vehicles, air pollution has become a severe problem

n many cities within economically emerging countries (Zhang,

011). Epidemiologic evidence has shown a significant increase

n immunologic diseases in severely polluted areas (Nastos et al.,010; Brauner et al., 2010). Additionally, the prevalence of

∗ Corresponding author at: Department of Obstetrics and Gynecology, Fuzhoueneral Hospital, No. 156 North Xihuan Road, Fuzhou 350025, Fujian, China.el.: +86 591 2285 9200; fax: +86 591 8789 3800.∗∗ Corresponding author. Tel.: +86 591 8727 9622.

E-mail addresses: hxr0812@163.com (X. Hong), lcb7073@163.com (C. Liu),henxq63@yahoo.com.cn (X. Chen), wqing70@163.com (Q. Wang),angpingtan@126.com (P. Wang), hudian0709@sina.com (D. Hu).

300-483X/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.tox.2013.02.004

© 2013 Elsevier Ireland Ltd. All rights reserved.

immunologic diseases in early life has been on the rise over thelast few years (Price et al., 2012). The exposure to air pollutionin utero has also been shown to be potentially devastating, as thefirst signs of some immune disorders, such as the infantile anaphy-lactic diseases, neonatal lupus erythematosus and some cancers,can develop in infants shortly after birth (Yu et al., 2006; Izmirlyet al., 2010; Kozyrskyj et al., 2011). Thus, the high prevalence ofimmunologic diseases in children may be putatively associated, atleast in part, with the consequent maldevelopment of the prena-tally impaired immune system (Dietert, 2011). If this is true, it willsuggest that exposure to air pollution in utero may well affect thedevelopment of the fetal immune system.

Airborne particulate matter (PM) is one of the main contributors

to ambient air pollution throughout the world, especially in devel-oping countries (Kan et al., 2009). PM is a common air pollutant thatcould have a significant impact on the immune system during bothpre- and postnatal periods of life, potentially resulting in system

6 icology

muiicMct(aht

cPderwiesboaGd(a

2

2

otwahwAtH

2

tpto(goiAp6((a

mbmmsow

f(fi

0 X. Hong et al. / Tox

alfunction in early life (Svendsen et al., 2007). In particular, partic-late matter that is <2.5 �m in diameter (PM2.5) has been shown to

nfluence fetal immune system development and appears to resultn altered immunoglobulin E (IgE) (Herr et al., 2011) and lympho-yte distributions (Herr et al., 2010) in cord blood among neonates.aternal exposure to air pollution before and during pregnancy

an also alter immunity in offspring, thus increasing a child’s risko develop health conditions like asthma and allergies later in lifeBaïz et al., 2011). In summary, data from epidemiologic surveysnd clinical observation suggests that the fetal immune system isighly vulnerable to environmental pollutants – much more so thanhe more developed systems of infants and older children.

Exposure to ambient levels of PM generally causes no appre-iable effects, but impalpable alteration that results from prenatalM exposure may come to exert a more evident impact on laterevelopment throughout childhood or even adulthood (Corsont al., 2010). Nonetheless, the results are not always consistentegarding the immune system, in spite of the fact that infantsho later develop immune disorders show some altered neonatal

mmune responses (Prescott, 2006). We hypothesized that in uteroxposure to PM may affect the development of the fetal immuneystem, resulting in its abnormal phenotype in early life followingirth. In the present work, we investigated in utero PM exposuren childhood immune function in a subacute murine model withdministration of resuspended standard PM. We tested T-bet andATA-3, both of which are the transcription factors that induce cellifferentiation of T helper cell subgroup 1 (Th1) and subgroup 2Th2), in addition to systemic and local biochemical measurementsnd morphological changes in the immune organs.

. Materials and methods

.1. Animals

Sixty females and 30 males specific pathogen free (SPF) ICR nonparous micef 6-week-old were purchased from Fujian Center for Disease Control and Preven-ion (Certificate number: SCXK2011-0001, Fuzhou, China), of which the female miceeighed 29 ± 2 g and the male mice 32 ± 2 g for copulation. The mice were housed in

climate controlled room at 23 ± 1 ◦C with 55 ± 7% humidity, in addition to a 12/12- light/dark cycle and standard rodent chow with water ad libitum. The experimentsere performed in accordance with the guidelines of the “Principles of Laboratorynimal Care” by the National Society for Medical Research in China. The experimen-

al protocol was approved by the Scientific Research Committee of Fuzhou Generalospital.

.2. PM exposure

Airborne PM of Standard Reference Material (SRM) 1649a was purchased fromhe National Institute of Standards and Technology (Gaithersburg, MD, USA). Thisarticulate matter is an atmospheric material collected in an urban area with par-icle diameters ranging 6.7–100 �m (averaged 12.9 �m). Substances that depositn the surface of these particle cores include polycyclic aromatic hydrocarbonsPAHs), polychlorinated biphenyl (PCB) congeners and chlorinated pesticides, inor-anic salts, heavy metals, bioactive components such as lipopolysaccharides, andther trace elements, many of which, especially the organic components that dom-nate the deposits, are noxious to human health (Benner, 1998; Chiu et al., 2001;lbinet et al., 2006). Prior to exposure, SRM 1649a was resuspended in sterile phos-hate buffered saline (PBS) to obtain a final concentration of 0.09, 0.28, 1.85 or.92 �g/�l and was stored at 4 ◦C. The suspension, which contains the indissolublee.g., PAHs), poorly dissoluble (e.g., PCB and chlorinated pesticides) and dissolublee.g., inorganic constituents) components, was thoroughly mixed and sonicated fort least half an hour in an ultrasonic vibration instrument prior to administration.

Mice were allowed one week adjustment prior to the experiments. The femaleice were placed with the male mice (2:1) overnight and pregnancy was confirmed

y the presence of vaginal plug valve (referred to as gestation day 0). The pregnantice were weighed, numbered, and randomly divided into one of the five groups (12ice per group): experimental groups, which were administered with SRM 1649a

uspension at concentration of 0.09, 0.52, 1.85 or 6.92 �g/�l (low, medium, high,r overdose group), and the control group, which received the same volume of PBS

ithout particles.

The instillation procedure to deliver resuspended particles to the mice was per-ormed as described previously, by which 77.5–88.2% particles would enter the lungRao et al., 2003). The female mouse was briefly anesthetized in a glass containerlled with isoflurane. The mouse’s mouth was opened using rubber bands with the

306 (2013) 59– 67

tongue pulled gently aside by a pair of forceps. A 30-�l SRM 1649a suspension orthe same volume of PBS alone was pipetted to the base of the tongue, in which themouse was held steady for at least two deep breaths (less than 15 s). The particlesuspension was administered beginning from day 0 and repeated at day 3, 6, 9, 12,15 and 18 with a total of 18.9 (low), 109.2 (medium), 388.5 (high) or 1453.2 �g(overdose) of SRM 1649a per mouse. After the last SRM 1649a administration atday 18, the mice were housed individually and were allowed to deliver. Shortlyafter parturition, the pups, together with their dam in a cage, were transferred to anairborne particle-eliminated chamber for 30 days before the experiment was per-formed. Mice with the onset of delivery within day 20–22 of pregnancy and withnumber of pups within 8–12 per litter were allowed to enter the experiment.

2.3. Collection and storage of organ and blood samples

After preclusion of the dams that did not match the condition, there were 8–10dams (litters) enrolled in each group. Total of 40 fetal mice from the 8 to10 litterswith each litter 4–5 fetuses in a group were randomized, weighed, and decapi-tated after eutherization. Blood samples from the randomly selected fetal mice ineach litter, with each litter as one sample, were pooled into a heparin-containingtube (n = 8–10) and blood plasma was isolated and stored in −70 ◦C. The thoracicgland and spleen were identified, isolated and weighed. Ratios of organ weightssummed from the selected 4 to 5 fetal mice in each litter over their summed bodyweights were calculated and compared among the groups (n = 8–10). The randomlychosen 20 spleen organs and 20 thoracic gland organs from the total 40 organswith every litter 2–3 spleen or thoracic gland organs in a group were processed forhistopathological examination. Half of the remaining 20 spleen organs, with everylitter 2–3 organs pooled to form one sample (n = 8–10), were cut into small piecesabout 2–3 mm3 in ice-cold saline and stored in RNA storage liquid in −70 ◦C, andthe other half in 37 ◦C saline for preparation of spleen lymphocyte suspension.

2.4. Measurement of total immunoglobulin G1, G2a (IgG1, IgG2a) and IgEantibodies in plasma

Plasma samples were thermometrically balanced to room temperature. Total IgEantibodies in the plasma were measured by a sandwich technique using the ELISAkit according to the manufacturer’s protocols (Yamasa Co., Chiba, Japan), in whichtwo monoclonal rat anti-mouse IgE antibodies recognizing different epitopes on theFc�R fragment were used. Optical density at 450 and 550 nm was determined usinga microplate reader. Total IgG1 and IgG2a antibodies in the plasma were also mea-sured by ELISA. Anti-mouse Ig (BioLead Biology Sci & Tech Co. Ltd., Beijing, China)of 0.1 ml containing 10 �g/ml protein diluted by 0.05 mol/L coating buffer (pH = 9.6)was added to each reacting well in the microtiter plate and was incubated at 4 ◦Covernight. The plate was washed four times in PBS containing 0.1% Tween-20, addedby diluted plasma at 37 ◦C for 2 h, and washed sequentially by PBS and PBS con-taining 0.1% Tween-20 (PBST) each for two times. One hundred microliters dilutedhorseradish peroxidase-labeled anti-mouse IgG1 or IgG2a (1:1000, BioLead BiologySci & Tech Co. Ltd.) were added into each well, incubated at 37 ◦C for 1 h, washedfor 5 times, added by 100 �l substrate buffer, and then incubated away from light at37 ◦C for 30 min. The reaction was terminated by addition of 0.05 ml sulphuric acid(2 mol/L) and optical density was read at 450 nm with the microplate reader.

2.5. Spleen lymphocyte proliferation test

Spleen homogenates were filtered through a 200-screen mesh grit made ofstainless steel to prepare a single-cell suspension. The erythrocytes in the suspensionwere lyzed by addition of ammonium chloride. The lymphocytes were harvested bycentrifugation and re-suspended in RPMI 1640 (Sigma–Aldrich Co., St. Louis, MO,USA) containing 10% calf serum. The vial lymphocytes, confirmed by trypan bluestaining, were adjusted to a density of 1 × 106/ml, and inoculated in 96-well cultureplates with 200 �l in each well. Each sample was inoculated in 6 wells, 3 of whichwere added by 10 �l phytohemagglutinin and the other 3 were not to serve as thecontrol. The mixture was cultured at 37 ◦C for 72 h under 5% carbon dioxide andsaturated humidity, and was added by 10 �l tetramethylthiazole blue (Sigma, St.Louis, MO, USA) prior to the end of the culture. Supernatant of 100 �l was gentlyremoved from each well and replaced by the same amount of DMSO. The mixturewas then vibrated for 30 min and tested at 570 nm for optical density.

2.6. Assay for interferon-� (IFN-�) and interleukin (IL)-4 contents

The above spleen lymphocyte suspension preparation was inoculated in a 96-well culture plate, followed by 10 �l phytohemagglutinin per well, and cultured at37 ◦C for 48 h under 5% carbon dioxide and saturated humidity. The suspension wasthen centrifuged at approximately 3000 × g at 4 ◦C for 10 min and the supernatant

was used for the assay. Contents of IFN-� and IL-4 in the supernatant and plasmawere assayed by using a solid phase sandwich technique using a commerciallyavailable ELISA kit (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’sinstructions and were performed in duplicate. Results are expressed as pg/ml. Ratiosof IL-4/IFN-� were calculated.

X. Hong et al. / Toxicology 306 (2013) 59– 67 61

Table 1Primer sequences and fragment lengths of the targeted genes.

Gene Primer sequence Fragmentlength (bp)

T-bet Forward: 5′-GGT GTC TGG GAA GCT GAG AG-3′ 374Reverse: 5′-TCT GGG TCA CAT TGT TGG AA-3′

GATA-3 Forward: 5′-TCT GGA GGA GGA ACG CTA ATG G-3′ 408Reverse: 5′-GAA CTC TTC GCA CAC TTG GAG ACT C-3′

2

daAedfZu

2

tGpGatTmtTTfnacts

2

wOmNtfc

3

3

oglsi5rgr3i

Fig. 1. Levels of total IgG1 and IgG2a, and IgE antibodies in the plasma in the mouseoffspring with their dams receiving intralaryngopharyngeal SRM 1649a during the

with the medium or high dose group (P < 0.05). Levels of IL-4 andratios of IL-4/IFN-� were significantly higher in the medium, highand overdose groups when compared with the control or the lowdose group, and were significantly higher in the overdose group

Fig. 2. Values of optical density in the supernatants from the spleen lymphocyte

GAPDH Forward: 5′-GAA GGG CTC ATG ACC ACA G-3′ 166Reverse: 5′-GGA TGC AGG GAT GTT C-3′

.7. Histological examination

The spleen and thoracic gland tissue samples were fixed with paraformaldehyde,ehydrated, embedded in paraffin, sliced at 5 �m, and stained with hematoxylinnd eosin (H&E) for light microscopic examination (400×, Olympus BX-41, Japan).n experienced histopathologist who was blinded with the experimental designxamined all the slides with reference to the pathological scoring standards, and theegree of each histopathological abnormality was evaluated in a semi-quantitativeashion and graded numerically from 0 (normal) to 3 (severest) (Xiping et al., 2007;hang et al., 2009). Ten micrographs from each specimen were taken from consec-tive squares to define the degree of the anatomic abnormality.

.8. T-bet and GATA-3 mRNA expressions

Copy number of T-bet and GATA-3 mRNA in the spleen was estimated by rela-ive quantification using real-time quantitative polymerase chain reaction (TaqManene Expression assay) (Heid et al., 1996; Oomizu et al., 2006). The assay waserformed with iCycler iQ real-time polymerase chain reaction detector (BioRad,ermany). The spleen samples from each litter of the young mice were poolednd homogenized by TRIzol Reagent (Invitrogen) methodology. The sequence ofhe primers and the length of the fragments for T-bet and GATA-3 were shown inable 1. Reverse transcription was performed to synthesize the first-strand comple-entary DNA (cDNA) from 1 �g of total RNA using oligod (dT)18 primers according

o the manufacturer’s instructions (Invitrogen). Applied Biosystems supplied theaqMan minor groove binder probes for T-bet and GATA-3 genes, as well as theaqMan Rodent GAPDH Control Reagents (Applied Biosystems, Carlsbad, CA, USA)or quantification of the internal standard glyceraldehyde-3-phosphate dehydroge-ase (GAPDH) RNA. The thermocycler program was set as 2 min at 50 ◦C and 10 mint 95 ◦C for one cycle, and 15 s at 95 ◦C and 1 min at 60 ◦C for 60 cycles. The geneopies were measured and calculated by correlating the polymerase chain reactionhreshold cycle obtained from the spleen tissue sample to the amplicon-specifictandard curve.

.9. Statistical analysis

Normally distributed data are expressed as mean ± standard deviation (SD)hile abnormally distributed data expressed as median (interquartile range).ne litter was considered one experimental unit in the statistic analysis. Nor-ally distributed data were assessed by one-way ANOVA among the groups, andewman–Keuls post hoc testing was applied for pair-wise comparisons between

he means. Abnormally distributed data were analyzed by Kruskal–Wallis H testor non-parametric tests of multiple comparisons and Mann–Whitney U test foromparisons between two groups.

. Results

.1. Organ weight

All maternal mice survived throughout the experiment. Weightsf the pregnant mice before the SRM 1649a challenge among theroups and the pooled weights of the selected fetuses from aitter (as one sample) were not significantly different (data nothown). For the controls, low, medium, high and overdose groupsn the offspring, ratios of spleen weight over the body weight were.52 ± 0.30, 5.57 ± 0.29, 5.48 ± 0.27, 5.54 ± 0.21 and 5.38 ± 0.32,espectively, and were not significantly different among the fiveroups (F = 1.084, P > 0.05). Similar results were seen in ratios of tho-

acic gland weight over their body weight, which were 2.94 ± 0.19,.03 ± 0.14, 3.01 ± 0.14, 2.93 ± 0.16 and 2.94 ± 0.20, correspond-

ngly (F = 1.493, P > 0.05).

pregnancy. n = 8–10. CT: control, LD: low dose, MD: medium dose, HD: high dose,and OD: overdose. *P < 0.05 compared to the controls; #P < 0.05 compared to lowdose; +P < 0.05 compared to overdose.

3.2. Total IgG1, IgG2a and IgE antibodies

In the plasma, levels of IgG1 and IgE antibodies were signifi-cantly higher in the medium, high and overdose groups comparedwith the control or low dose group (P < 0.05). Levels of IgG2a anti-bodies were significantly lower in the overdose group comparedwith the control, low, medium or high dose group (P < 0.05) (Fig. 1).

3.3. Spleen lymphocyte proliferation

Values of optical density from the spleen lymphocyte prolifera-tion test was significantly lower in the medium, high and overdosegroups compared with the control group, and significantly lower inthe high and overdose groups compared with the low dose group(P < 0.05). No significant difference was found between the controlgroup and the low dose group (P > 0.05) (Fig. 2).

3.4. IFN-� and IL-4 levels

In the plasma and supernatant of the cultured spleen cells, levelsof IFN-� were significantly lower in the medium, high and overdosegroups when compared with the control or low dose group, andwere significantly lower in the overdose group when compared

proliferation test in the mouse offspring with their dams receiving intralaryngopha-ryngeal SRM 1649a during the pregnancy. n = 8–10. CT: control, LD: low dose, MD:medium dose, HD: high dose, and OD: overdose. *P < 0.05 compared to the controls;#P < 0.05 compared to low dose.

62 X. Hong et al. / Toxicology

Fig. 3. Interleukin-4 (IL-4), interferon-� (IFN-�) and IL-4/IFN-� in the plasma andspleen tissue in the mouse offspring with their dams receiving intralaryngopha-rm#

w(

3

slIw(rpdspiwTwdt

yngeal SRM 1649a during the pregnancy. n = 8–10. CT: control, LD: low dose, MD:edium dose, HD: high dose, and OD: overdose. *P < 0.05 compared to the controls;

P < 0.05 compared to low dose; †P < 0.05 compared to overdose.

hen compared with the medium or the high dose group (P < 0.05)Fig. 3).

.5. Morphology of spleen and thoracic gland tissues

The spleen tissue from the control group exhibited an intacttructure of splenic corpuscle with intensive and hyperchromaticymphocytes around its central arteries under light microscopy.n the splenic pulp area, distinct splenic cord and splenic sinus

ith sparsely distributed lymphocytes and phagocytes were seenFig. 4a). On the contrary, the spleen tissue from the overdose groupevealed an unclear boundary between the white pulp and redulp, where extrusive hyperplasy in the white pulp and laterallyeviated central arteries were observed. The periarterial lymphaticheaths with indistinct boundaries and heavy infiltration of lym-hocytes were volumetrically increased 3–4 times larger than those

n the control group, in addition to the locally hyalinized vesselall that appeared to be much thicker in the splenic arterioles.

he splenic sinus in the red pulp was dilated, congested, presentedith heavily proliferated phagocytes, and surrounded by unevenlyistributed and locally agglomerated lymphocytes (Fig. 4b–f). Thehoracic gland tissue manifested intensive, evenly distributed and

306 (2013) 59– 67

deep-color lymphocytes in its cortex, and distinct blood vesselsand reticulocytes with dispersively distributed lymphocytes in itsmedulla in the control group (Fig. 5a). In the overdose group, thethymic cortex was broadened with plenty of proliferated immaturemonocytoid cells whereas the thymic medulla was coarctate with alarger amount of apoptotic debris and apoptotic bodies. The bound-aries between the cortex and medulla were unclear (Fig. 5b–d). Theoverall histopathology scores indicate that the histopathologicalchanges were most severe in the spleen and thoracic gland tissuesin the overdose group, moderate in the high dose group, and minorin the medium and low dose groups when compared with the con-trol group. A detailed analysis of histological scores is shown inTable 2.

3.6. T-bet and GATA-3 mRNA expressions

The medium, high and overdose groups had a significantly lowerlevel of T-bet mRNA expression and a significantly higher level ofGATA-3 expression compared to the control group (P < 0.05). Lev-els of GATA-3 and T-bet mRNA were not significantly differentbetween the low dose group and the control group (P > 0.05) (Fig. 6).

3.7. Correlations between levels of IFN-� and IL-4 andexpressions of T-bet and GATA-3

IFN-� levels in the plasma and spleen were positively corre-lated with T-bet mRNA expression (rPlasma = 0.768, rSpleen = 0.757,P < 0.05) and were negatively correlated with GATA-3 mRNAexpression (rPlasma = −0.797, rSpleen = −0.824, P < 0.05), whereasIL-4 levels were negatively correlated with T-bet mRNA expres-sion (rPlasma = −0.755, rSpleen = −0.812, P < 0.05) and were posi-tively correlated with GATA-3 mRNA expression (rPlasma = 0.801,rSpleen = 0.847, P < 0.05).

4. Discussion

Particulate air pollution has been shown to be associated withthe induction of immune response (Miyata and van Eeden, 2011).This study intended to present findings on the capabilities andpotentials of maternal exposure to different doses of standard PM tomodulate immune responses in the offspring. Systemic and organicimmune responses were observed in young offspring whose damswere exposed to standard PM SRM 1649a. Our data demonstratedthat maternal exposure to PM resulted in modulation of Th1 andTh2 immune response in the postnatal offspring, as evidencedby decreased total IgG2a (Th1-dependent) and increased totalIgG1 (Th2-dependent) levels in the plasma, decreased IFN-� andincreased IL-4 levels in the plasma and the spleen, and decreasedexpression of transcription factor T-bet (Th1-related) and increasedexpression of GATA-3 (Th2-related) in the spleen in an exposuredose-dependent manner. Inhibited splenic lymphocyte prolifera-tion suggested compromised cell immunity. Mounting evidencealready exists for the involvement of airborne particulate pollut-ants in the exacerbation of immune response in the young andadults (Zhang and Smith, 2007). Our results may imply that alteredimmune response may exist in offspring that are prenatally chal-lenged by airborne particles as well.

PM alters the normal developmental pattern for the immunesystem that is constantly changing during pre- and postnatalgrowth. Susceptibility of the fetus to PM may be attributable toeither direct or indirect hits on the immune system to alter cell dif-ferentiation, proliferation, maturation and/or possible secretion. T

helper cells are the regulatory and effector cells in immune reaction.Th1 and Th2 are functionally balanced under normal conditionsand the imbalance in Th1/Th2 is associated with immunofunctionaldisorders. Th2 induces differentiation of B cells and production

X. Hong et al. / Toxicology 306 (2013) 59– 67 63

Fig. 4. Representative photomicrographs of pathological changes in the spleen tissue in the young mice. Panel a shows normal morphological appearance in the spleen tissuefrom the control group. Panels b–f show various pathological alterations in the samples from overdose group. Opaque boundaries between white pulp and red pulp withdeviated central arteries were observed in panel b. In panel c, periarterial lymphatic sheaths were enlarged and a large number of lymphocytes were present. In panel d,the vessel wall in the splenic arterioles was thickened with local hyalinization. In panel e, the splenic sinus in the red pulp was dilated and congested with agglomeratedlymphocytes. In panel f, the splenic sinus exhibited massively increased macrophages. Arrows point to typical histopathological changes. Scale bars represent 2000 �m (a,b), 400 �m (c, d, f) or 800 �m (e). (For interpretation of the references to color in this figu

Table 2Comparison of histopathological severity scores for spleen and thoracic gland,M(QR).

Groups Spleen Thoracic gland

Control 0.0 (0.0) 0.0 (0.0)Low dose 0.0 (1.0) 0.0 (0.0)Medium dose 0.5 (1.0) 0.0 (1.0)High dose 1.5 (2.0)* 1.0 (1.5)*

Overdose 2.0 (1.0)*,# 2.0 (2.0)*,#

H 11.585 9.647P <0.05 <0.05

Note:* P < 0.05 versus control or low dose group.# P < 0.05 versus low dose group.

re legend, the reader is referred to the web version of the article.)

of immunoglobulins by secretion of cytokines including IL-4. Theenhancement of Th2 cellular response includes increased IL-4 and -5 production, elevated serum levels of total IgE, IgG1, and decreasedlevels of IgG2a (Zhu et al., 2004). An activity and Th2-skewingcapacity of airborne particles have been shown in different ani-mal models and in humans (Porter et al., 2007; He et al., 2010;Zhao et al., 2012). Results from the present study suggested thatintrauterine exposure to PM exaggerates Th2 immune response inthe offspring postnatally, as indicated by elevated plasma levels ofIgG1 and IgE, and plasma and spleen levels of IL-4 in the offspringof the PM-administered dams. Maternal exposure to overdose PMreduced the level of plasma total IgG2a in the offspring compared

with the controls; the immune response induced by the presenceof PM dosed in the other three doses was, though less powerful tothe overdose, of the Th2 type. On the other hand, decreased IFN-� production in plasma and spleen may imply a suppressed Th1

64 X. Hong et al. / Toxicology 306 (2013) 59– 67

Fig. 5. Representative photomicrographs of pathologic changes in the thoracic gland in the young mice. Panel a shows normal morphological appearance in the thoracicgland tissue from the control group. Broaden thymic cortex and narrowed thymic medulla were seen (b) with opaque boundaries between cortex and medulla (c). Relicd hange

fm

isltaapoPtl

Fwno

ebris in the cytoplasm was present (d). Arrows point to typical histopathological c

unction. These results imply that maternal exposure to PM possiblyodulates both Th1- and Th2-responses in the offspring.The existence of discrete developmental processes within the

mmune system is likely to influence differential developmentalusceptibility to toxicants, which may result in lifelong toxico-ogical changes (Dietert et al., 2000). Hamada et al. (2007) foundhat prenatal exposure to aerosolized leachate of residual oil flysh resulted in an increasing susceptibility to the development ofsthma in early life, in which the offspring were challenged by theollutant prenatally and postnatally coupled with sensitization byvalbumin. In the present study, the offspring were challenged by

M prenatally, which may further imply that the prenatal effect onhe immune system could be “memorized” and presented later inife.

ig. 6. T-bet and GATA-3 mRNA expressions in the spleen tissue in the young miceith their dams receiving intralaryngopharyngeal SRM 1649a during the pregnancy.

= 8–10. CT: control, LD: low dose, MD: medium dose, HD: high dose, and OD:verdose. *P < 0.05 compared to the controls.

s. Scale bars represent 2000 �m (a, b), 800 �m (c) or 200 �m (d).

In contrast to the extensive reports on induction of Th2responses, such as IgG1 and IgE antibody and IL-4 cytokine by expo-sure to environmental chemicals, few have referred to transcriptionfactor involvement. T-bet and GATA-3 are two transcription fac-tors that have been found to play a critical role in differentiation ofTh1 or Th2 cells. T-bet promotes Th1 cell production by inducingthe expression of cytokines, which are required for Th1 function,such as IFN-�, and suppressing the expression of GATA-3 (Agnelloet al., 2003). Mice with an inactivation of the T-bet gene possessan increased number of Th2 cells and an absence of Th1 cells,which makes them susceptible to Th2-mediated immune disor-ders (Szabo et al., 2002). GATA-3, whose expression is modulatedby IL-4, affects the gene locus of IL-4 and promotes Th2 differenti-ation (Zhou and Ouyang, 2003). Studies regarding differentiationof helper T cells reveal that GATA-3 functions as the transacti-vator of the IL-4 gene and as the determinant for Th2 lineage(Yoh et al., 2003; Pai et al., 2004). By exposing adult Wistar–kyotorats to PM2.5, Zhao et al. reported decreased T-bet and increasedGATA-3 expressions in the heart after PM exposure, suggest-ing a link between PM exposure and cardiac injury (Zhao et al.,2012). Our results in a maternal–fetal–filial model showed similarreversely changed expressions of T-bet and GATA-3 mRNA in anexposure dose-dependent manner. Although the reasons remainunclear, the present data may indicate that prenatal exposure toPM causes modulation of both Th1 and Th2 related-transcriptionfactors in young mice. Significantly decreased T-bet and increasedGATA-3 expressions in the spleen, coupled with suppressed pro-duction of IFN-� and exaggerated production of IL-4 in plasma and

the spleen support both Th1- and Th2-response paradigm in theoffspring. This was further supported by a mathematic correlationbetween changes in T-bet and GATA-3 gene expressions and IFN-� and IL-4 cytokine levels. The effector cytotoxic T cells revealed

icology

agoitTeiil

tbspBigmp

acdairmppaugscariaofisa

sCalaeaaonoTith

rimais

X. Hong et al. / Tox

decreased IFN-� production in all except low dose PM-treatedroups, whereas a higher level of IL-4 is parallel to a larger dosef exposed PM. Therefore, the IL-4/IFN-� ratio in T helper cells inntrauterinely PM-exposed animals was markedly changed duringhe immune response, further suggesting a deviation of Th1/Th2.here are physiological coherences among these observed param-ters, e.g., altered T-bet and GATA-3 gene expressions resultingn cytokine disequilibrium, triggering the Th2 pathway, resultingn immune responses by IgG1, IgG2a, IgE, and histopathologicalesions.

Bezemer et al. (2011) found that respirable pollutants activatedhe innate immune response with Th2-immune responsiveness,ut exposure of mice to environmental PM dampened IFN-�ecretion. Our results indicated a significantly suppressed IFN-�roduction, which seems to be more pronounced than that ofezemer’s. Although there is difference regarding PM doses, admin-

stration route, animal strains, and measurement parameters, thereater vulnerability of the fetal immune system to PM exposureay elevate susceptibility to immunologically mediated disorders

ostnatally.The spleen and thoracic gland are the most important peripheral

nd central immune system organs involved in T and B lympho-yte differentiation and maturation. The spleen and thoracic glandevelop largely during the prenatal and early postnatal periodsnd their development is highly sensitive to immunotoxic chem-cal substances (Holladay and Smialowicz, 2000). Little is knownegarding the effects of environmental toxicants on the develop-ent of immune organs during the prenatal period. We exposed

regnant mice to PM for the whole gestational duration. Althoughrenatal exposure to PM did not alter the immune organ weightnd its ratio over the body weight in the offspring, the eval-ation of histopathological changes in the spleen and thoracicland reflected dose-related small to severe lesions. Biological mea-urements, including spleen lymphocyte proliferation, indicated aompromised lymphocyte proliferation capacity in medium, highnd overdose groups, suggesting an impaired cellular immuneesponse in the spleen. Splenic lymphocyte proliferation plays anmportant role in immunological regulation, anti-viral infectionnd anti-oncogenesis. Thus, immunologic damage in the offspringf PM-treated mothers was morphologically and functionally con-rmed in the present study. Significant immunologic damage mayuggest a greater sensitivity and susceptibility to immune disorderst a young stage of development.

A predominance of Th2 in early life has been associated withome disorders including allergic diseases in children and later life.linical evidences revealed that a high circulating Th2/Th1 ratiot birth, as reflected by Th1- and high Th2-associated chemokineevels, was associated with the development of eczema, wheezend sensitization in children (Sandberg et al., 2009; Abrahamssont al., 2011). Rothers et al. (2011) reported that total IgE levels andctive asthma through age 5 years correlate with IL-4, IL-5, IFN-�nd other cytokines regulations in early life. Using a murine modelf childhood asthma, Siegle et al. reported that administration ofeutralizing antibodies against IL-4 or IL-25 prevents developmentf some key features in asthma, suggesting an association betweenh2 response and immune disease during neonatal period or latern childhood (Siegle et al., 2011). We may suggest a raised suscep-ibility to the immune-related diseases in the offspring mice withigher Th2/Th1 ratio resulting from prenatal exposure to PM.

Heterogeneity of the monocytes/macrophages has long beenecognized and, in part, is a result of the specialization, especiallyn function, of tissue macrophages in particular microenviron-

ents (Mantovani et al., 2005, 2008). “M1” cells, or classicallyctivated macrophages, are inflammatory cells and may benitiated by stimulating their progenitors with inflammatorytimuli, such as lipopolysaccharides and granulocyte macrophage

306 (2013) 59– 67 65

colony-stimulating factor. “M2” cells, or alternatively activatedmacrophages, display a more regulatory function and may be gen-erated when progenitors are cultured with IL-4 and IL-13. However,the heterogeneity of the mononuclear phagocyte system is poorlyrecapitulated, and some models and experiments do not repre-sent a model to study the specialized functions of the diverse celltypes that are present in vivo, or the regulation of their develop-ment and functions by the tissue microenvironment (Geissmannet al., 2010). Due to the complexity of the mechanisms involvedin the heterogeneity of the tissue macrophages and the limitedscope of our work, we intentionally did not show any data aboutthe phenotypic changes in M1/M2 at both circulatory and tis-sues levels in order to prevent any misleading or misinterpreteddata. The role and its heterogeneity of tissue macrophages in airpollution exposure-induced diseases have been studied by otherresearchers recently. Sun et al. (2009) found that PM2.5 expo-sure increased systemic increased the numbers of F4/80+ adiposetissue macrophages (ATMs), increased markers of classically acti-vated (M1) ATMs, and decreased markers of alternatively activated(M2) ATMs in rodent models. In an early exposure rodent model,Xu et al. (2010) found that PM2.5 exposure led to a significantincrease in the expression of proinflammatory genes (M1) tumornecrosis factor-�, nitric oxide synthase-2 and IL-6, no changes inintegrin �X expression, and downregulation of IL-10 expression,demonstrating that PM2.5 exposure downregulated genes corre-lated with an anti-inflammatory M2 phenotype while inducing aproinflammatory phenotype. However, this categorization shouldnot be simply extrapolated in studies of macrophage polarizationstates as more details and factors should be considered in termsof in vivo versus in vitro models and animal versus human stud-ies (Geissmann et al., 2010). To assess a Th2 response to PM2.5,Deiuliis and his team measured IL-4 expression in CD4+ T cells iso-lated from the lung and found no difference compared with thecontrols, suggesting that PM2.5 exposure has little physiologicaleffect on Th2 populations, while a Th1 response may dominate(Deiuliis et al., 2012). In the same study, three major CXCR3-binding chemokines (CXCL9, CXCL10, and CXCL11) were induced ina variety of cells, including lung macrophages and dendritic cells,in response to other cytokine mediators such as IFN-�, provid-ing an amplification loop for Th1 immune responses by attractingmore CXCR3-expressing Th1 cells that was evidenced by a markedincrease in CXCR3+CD4+ cells in response to PM2.5 exposure. There-fore, both macrophages and dendritic cells have crucial roles in airpollution exposure-induced immunological response and disease,which merits further mechanistic investigation.

In the present study, the exposure model was prepared inpregnant mice whose offspring were experimentally processed atpostnatal day 30, which is equivalent to childhood around 5–10years of age in humans (Downing et al., 2009). PM doses usedin our study were calculated with reference to the U.S. NationalAmbient Air Quality Standard 26–534 �g/m3 for environmentalinhalable particulate matters with the airborne particulate matterswith aerodynamic diameters less than 10 �m (PM10). Particularemphasis was given to real world scenarios, with PM10 in largecities and industrial regions of developing countries like China aver-aging around 80 �g/m3 (Xie et al., 2005; Kan et al., 2007; Xu et al.,2008; van Donkelaar et al., 2010; The U.S. Embassy in Beijing, 2011;Yang et al., 2012), and with a peak value of up to 2000 �g/m3 underextreme conditions (Wang et al., 2009; Kan et al., 2012). As thevital capacity for an adult mouse is 24 ml/min, the total calculatedamount of air inhaled into the body over 24 h is about 0.035 m3.Moreover, the daily inhaled amount of PM10 for a mouse is about

0.9 �g and 18.5 �g according to the standard 26 and 534 �g/m3,which is the reference for the low dose and high dose we used.The medium dose and overdose amounts were determined withreference to 80 �g/m3 and 2000 �g/m3, respectively. We did not

6 icology

iwtrb

t2dtaatpcps2ctitsSrwiectrtamimemlii(aftvm

eissaam(esWw2

C

6 X. Hong et al. / Tox

ncrease the administered PM doses with increasing murine bodyeight during the experimental period because we were aiming

o investigate the effect of PM under analogous environmentallyelevant levels, which were not associated with the increase in theody weight.

There are several limitations in the present study. First, althoughhe exposure model is frequently used in similar studies (Riva et al.,011; Musah et al., 2012) for its ease of manipulation and repro-ucibility (Rao et al., 2003), the route of instillation is not equivalento inhalation. The SRM 1649a particles we used in the present studyre total suspended particulate matters with many possessing largeerodynamic diameters that cannot enter the alveoli of the lung orranslocation across the blood-gas barrier in terms of the entirearticles as a whole but it is still possible, and, it is likely, that thehemical components from those PM10 or bigger particles couldenetrate into the circulation and reach remote organs and tis-ues (Hertz-Picciotto et al., 2008) and on the fetus (Ravindra et al.,001), so as the fine components (Mills et al., 2009). Although wealculated the doses based on real world exposure levels in loca-ions where humans may be exposed (especially in major citiesn developing countries like China) the lack of a modern, sophis-icated exposure system that allows inhalation exposure in ourtudy limits scientifically the data interpretation and significance.econd, the variability of the particles’ recovery does not preciselyeflect the local situation. Although SRM 1649a is a standardized,ell-studied particle reference that has been used in several exper-

mental observations (Sun et al., 2008; Ulrich et al., 2010; Wongt al., 2011), it is different in property from other ambient parti-les like diesel exhaust particles (Wong et al., 2011). We believehat using particles collected from the local area would best rep-esent the impact on the residents who live there. To best servehis purpose, an exposure system that has a capability to suck thembient air into the system, and allows certain sizes of particulateatters (such as PM2.5) into the chambers is under construction

n our laboratory. Third, although the current study investigated atultiple levels, especially associated with the T-bet and GATA-3

xpressions, there is a lack of evidence and pathways regarding theechanisms by which PM exposure ultimately induces immuno-

ogic responses. Besides, locally deposited substances may affectndirectly the body by inducing local biological reaction such asnflammation, oxidative stress, profibrotic and allergic processesvan Berlo et al., 2012), with the mechanisms of hemodynamiclterations in placental blood flow and reduction of nutrient trans-er into the fetus (Proietti et al., 2012). It is reasonably postulatedhat SRM 1649a may affect the immune function in the offspringia both direct and indirect pathways in the present exposureodel.In summary, the present study demonstrated that intrauterine

xposure to PM resulted in a deviation of Th1/Th2 that skewedmmune responses toward the Th2 phenotype by exacerbatingystemic IgG1, IgE and IL-4 productions and splenic GATA-3 expres-ion, and by attenuating systemic IgG2a and IFN-� productionsnd splenic T-bet expression. The components in PM are complex,nd the attached substances that are present on PM, which consistostly of polyaromatic hydrocarbons exerting as proallergic effects

Lubitz et al., 2010), allergens, and some bacterial components likendotoxin attaching on the particles (Inoue et al., 2007), are oftenuggested for being responsible for the side-effect of PM on health.

hether the attached substances of PM are involved in this ske-ing toward the Th2 side, as shown by other reporters (Braun et al.,

010; Lubitz et al., 2010), remains to be further elucidated.

onflict of interest statement

There are no conflicts of interest.

306 (2013) 59– 67

Acknowledgements

This work was supported by research grants from the ChineseNational Natural Science Fund (#81172677), Fujian Science andTechnology Bureau Unode Project (#2007Y0017), and Medical Sci-ence Technology Project of Nanjing Command (#07M093), China.The authors would like to thank Prof. Hongyu Yu for his histopatho-logical evaluation and Mr. Geoffrey Gatts for his critical reading andediting of the manuscript.

References

Abrahamsson, T.R., Sandberg, A.M., Forsberg, A., Björkstén, B., Jenmalm, M.C., 2011.A Th1/Th2-associated chemokine imbalance during infancy in children devel-oping eczema, wheeze and sensitization. Clin. Exp. Allergy 41, 1729–1739.

Agnello, D., Lankford, C.S., Bream, J., Morinobu, A., Gadina, M., O’Shea, J.J., Frucht,D.M., 2003. Cytokines and transcription factors that regulate T helper cell dif-ferentiation: new players and new insights. J. Clin. Immunol. 23, 147–161.

Albinet, A., Leoz-Garziandia, E., Budzinski, H., Viilenave, E., 2006. Simultaneous anal-ysis of oxygenated and nitrated polycyclic aromatic hydrocarbons on standardreference material 1649a (urban dust) and on natural ambient air samples bygas chromatography–mass spectrometry with negative ion chemical ionisation.J. Chromatogr. A 1121, 106–113.

Baïz, N., Slama, R., Béné, M.C., Charles, M.A., Kolopp-Sarda, M.N., Magnan, A.,Thiebaugeorges, O., Faure, G., Annesi-Maesano, I., 2011. Maternal exposure toair pollution before and during pregnancy related to changes in newborn’s cordblood lymphocyte subpopulations. The EDEN study cohort. B. M. C. PregnancyChildbirth 11, 87.

Benner, B.A., 1998. Summarizing the effectiveness of supercritical fluid extraction ofpolycyclic aromatic hydrocarbons from natural matrix environmental samples.Anal. Chem. 70, 4594–4601.

Bezemer, G.F., Bauer, S.M., Oberdörster, G., Breysse, P.N., Pieters, R.H., Georas, S.N.,Williams, M.A., 2011. Activation of pulmonary dendritic cells and Th2-typeinflammatory responses on instillation of engineered, environmental dieselemission source or ambient air pollutant particles in vivo. J. Innate. Immun.2, 150–166.

Braun, A., Bewersdorff, M., Lintelmann, J., Matuschek, G., Jakob, T., Göttlicher, M.,Schober, W., Buters, J.T., Behrendt, H., Mempel, M., 2010. Differential impact ofdiesel particle composition on pro-allergic dendritic cell function. Toxicol. Sci.113, 85–94.

Brauner, E.V., Andersen, C.E., Andersen, H.P., Gravesen, P., Lind, M., Ulbak, K., Her-tel, O., Schüz, J., Raaschou-Nielsen, O., 2010. Is there any interaction betweendomestic radon exposure and air pollution from traffic in relation to childhoodleukemia risk? Cancer Cause Control 21, 1961–1964.

Chiu, C.H., Turle, R., Poole, G., Thibert, B., Brubaker Jr., W.W., Schantz, M.M., Wise, S.A.,2001. Determination of polychlorinated dibenzo-p-dioxin and dibenzofurancongeners in air particulate and marine sediment standard reference materials(SRMs). Fresenius J. Anal. Chem. 369, 356–363.

Corson, L., Zhu, H., Quan, C., Grunig, G., Ballaney, M., Jin, X., Perera, F.P., Factor, P.H.,Chen, L.C., Miller, R.L., 2010. Prenatal allergen and diesel exhaust exposure andtheir effects on allergy in adult offspring mice. Allergy Asthma Clin. Immunol. 6,7.

Deiuliis, J.A., Kampfrath, T., Zhong, J., Oghumu, S., Maiseyeu, A., Chen, L.C., Sun, Q.,Satoskar, A.R., Rajagopalan, S., 2012. Pulmonary T cell activation in responseto chronic particulate air pollution. Am. J. Physiol. Lung Cell. Mol. Physiol. 302,L399–L409.

Dietert, R.R., 2011. Maternal and childhood asthma: risk factors, interactions, andramifications. Reprod. Toxicol. 32, 198–204.

Dietert, R.R., Etzel, R.A., Chen, D., Halonen, M., Holladay, S.D., Jarabek, A.M., Landreth,K., Peden, D.B., Pinkerton, K., Smialowicz, R.J., Zoetis, T., 2000. Workshop to iden-tify critical windows of exposure for children’s health: immune and respiratorysystems work group summary. Environ. Health. Perspect. 108, 483–490.

Downing, C., Balderrama-Durbin, C., Hayes, J., Johnson, T.E., Gilliam, D., 2009. Noeffect of prenatal alcohol exposure on activity in three inbred strains of mice.Alcohol. Alcohol. 1, 25–33.

Geissmann, F., Gordon, S., Hume, D.A., Mowat, A.M., Randolph, G.J., 2010. Unravellingmononuclear phagocyte heterogeneity. Nat. Rev. Immunol. 10, 453–460.

Hamada, K., Suzaki, Y., Leme, A., Ito, T., Miyamoto, K., Kobzik, L., Kimura, H., 2007.Exposure of pregnant mice to an air pollutant aerosol increases asthma suscep-tibility in offspring. J. Toxicol. Environ. Health A 70, 688–695.

He, M., Ichinose, T., Yoshida, S., Nishikawa, M., Mori, I., Yanagisawa, R., Takano,H., Inoue, K., Sun, G., Shibamoto, T., 2010. Urban particulate matter in Beijing,China, enhances allergen-induced murine lung eosinophilia. Inhal. Toxicol. 22,709–718.

Heid, C.A., Stevens, J., Livak, K.J., Williams, P.M., 1996. Real time quantitative PCR.Genome Res. 6, 986–994.

Herr, C.E., Dostal, M., Ghosh, R., Ashwood, P., Lipsett, M., Pinkerton, K.E., Sram, R.,

Hertz-Picciotto, I., 2010. Air pollution exposure during critical time periods ingestation and alterations in cord blood lymphocyte distribution: a cohort oflivebirths. Environ. Health 9, 46.

Herr, C.E., Ghosh, R., Dostal, M., Skokanova, V., Ashwood, P., Lipsett, M., Joad, J.P.,Pinkerton, K.E., Yap, P.S., Frost, J.D., Sram, R., Hertz-Picciotto, I., 2011. Exposure

icology

H

H

I

I

K

K

K

K

L

M

M

M

M

M

N

O

P

P

P

P

P

R

R

R

R

S

X. Hong et al. / Tox

to air pollution in critical prenatal time windows and IgE levels in newborns.Pediatr. Allergy Immunol. 22, 75–84.

ertz-Picciotto, I., Park, H.Y., Dostal, M., Kocan, A., Trnovec, T., Sram, R., 2008. Prena-tal exposures to persistent and non-persistent organic compounds and effectson immune system development. Basic Clin. Pharmacol. Toxicol. 102, 146–154.

olladay, S.D., Smialowicz, R.J., 2000. Development of the murine and humanimmune system: differential effects of immunotoxicants depend on time ofexposure. Environ. Health Perspect. 108, 463–473.

noue, K., Takano, H., Yanagisawa, R., Sakurai, M., Ueki, N., Yoshikawa, T., 2007. Effectsof diesel exhaust particles on cytokine production by splenocytes stimulatedwith lipopolysaccharide. J. Appl. Toxicol. 27, 95–100.

zmirly, P.M., Kim, M.Y., Llanos, C., Le, P.U., Guerra, M.M., Askanase, A.D., Salmon,J.E., Buyon, J.P., 2010. Evaluation of the risk of anti-SSA/Ro-SSB/La antibody-associated cardiac manifestations of neonatal lupus in fetuses of mothers withsystemic lupus erythematosus exposed to hydroxychloroquine. Ann. Rheum.Dis. 69, 1827–1830.

an, H., Chen, B., Hong, C., 2009. Health impact of outdoor air pollution in China:current knowledge and future research needs. Environ. Health Perspect. 117,A187.

an, H., Chen, R., Tong, S., 2012. Ambient air pollution, climate change, and popula-tion health in China. Environ. Int. 42, 10–19.

an, H., London, S.J., Chen, G., Zhang, Y., Song, G., Zhao, N., Jiang, L., Chen, B., 2007. Dif-ferentiating the effects of fine and coarse particles on daily mortality in Shanghai,China. Environ. Int. 33, 376–384.

ozyrskyj, A.L., Bahreinian, S., Azad, M.B., 2011. Early life exposures: impact onasthma and allergic disease. Curr. Opin. Allergy Clin. Immunol. 11, 400–406.

ubitz, S., Schober, W., Pusch, G., Effner, R., Klopp, N., Behrendt, H., Buters, J.T., 2010.Polycyclic aromatic hydrocarbons from diesel emissions exert proallergic effectsin birch pollen allergic individuals through enhanced mediator release frombasophils. Environ. Toxicol. 25, 188–197.

antovani, A., Romero, P., Palucka, A.K., Marincola, F.M., 2008. Tumour immunity:effector response to tumour and role of the microenvironment. Lancet 371,771–783.

antovani, A., Sica, A., Locati, M., 2005. Macrophage polarization comes of age.Immunity 23, 344–346.

ills, N.L., Donaldson, K., Hadoke, P.W., Boon, N.A., MacNee, W., Cassee, F.R., Sand-ström, T., Blomberg, A., Newby, D.E., 2009. Adverse cardiovascular effects of airpollution. Nat. Clin. Pract. Cardiovasc. Med. 6, 36–44.

iyata, R., van Eeden, S.F., 2011. The innate and adaptive immune response inducedby alveolar macrophages exposed to ambient particulate matter. Toxicol. Appl.Pharmacol. 257, 209–226.

usah, S., Dejarnett, N., Hoyle, G.W., 2012. Tumor necrosis factor-� mediates inter-actions between macrophages and epithelial cells underlying proinflammatorygene expression induced by particulate matter. Toxicology 299, 125–132.

astos, P.T., Paliatsos, A.G., Anthracopoulos, M.B., Roma, E.S., Priftis, K.N., 2010. Out-door particulate matter and childhood asthma admissions in Athens, Greece: atime-series study. Environ. Health 9, 45.

omizu, S., Onishiw, N., Suzukiz, H., Ueda, K., Mochizukiz, M., Morimotoz, K.,Kawamoto, S., Ono, K., Kameyoshiz, Y., Hide, M., 2006. Oral administration of pul-verized Konjac glucomannan prevents the increase of plasma immunoglobulin Eand immunoglobulin G levels induced by the injection of syngeneic keratinocyteextracts in BALB/c mice. Clin. Exp. Allergy 36, 102–110.

ai, S.Y., Truitt, M.L., Ho, I.C., 2004. GATA-3 deficiency abrogates the development andmaintenance of T helper type 2 cells. Proc. Natl. Acad. Sci. U. S. A. 101, 1993–1998.

orter, M., Karp, M., Killedar, S., Bauer, S.M., Guo, J., Williams, D., Breysse, P., Geo-ras, S.N., Williams, M.A., 2007. Diesel-enriched particulate matter functionallyactivates human dendritic cells. Am. J. Respir. Cell Mol. Biol. 37, 706–719.

rescott, S.L., 2006. Maternal allergen exposure as a risk factor for childhood asthma.Curr. Allergy Asthma Rep. 6, 75–80.

rice, K., Plante, C., Goudreau, S., Boldo, E.I., Perron, S., Smargiassi, A., 2012. Riskof childhood asthma prevalence attributable to residential proximity to majorroads in Montreal, Canada. Can. J. Public Health 103, 113–118.

roietti, E., Röösli, M., Frey, U., Latzin, P., 2012. Air pollution during pregnancy andneonatal outcome: a review. J. Aerosol. Med. Pulm. Drug Deliv. (Epub ahead ofprint) PMID: 22856675.

ao, G.V.S., Tinkle, S., Weissman, D.N., Antonini, J., Kashon, M., Salmen, R., Battelli, L.,Willard, P., Hubbs, A., 2003. Efficacy of a technique for exposing the mouse lungto particles aspirated from the pharynx. Toxicol. Environ. Health 66, 1441–1452.

avindra, Mittal, A.K., van Grieken, R., 2001. Health risk assessment of urbansuspended particulate matter with special reference to polycyclic aromatichydrocarbons: a review. Rev. Environ. Health 16, 169–189.

iva, D.R., Magalhães, C.B., Lopes, A.A., Lanc as, T., Mauad, T., Malm, O., Valenc a, S.S.,Saldiva, P.H., Faffe, D.S., Zin, W.A., 2011. Low dose of fine particulate matter(PM2.5) can induce acute oxidative stress, inflammation and pulmonary impair-ment in healthy mice. Inhal. Toxicol. 23, 257–267.

others, J., Halonen, M., Stern, D.A., Lohman, I.C., Mobley, S., Spangenberg, A.,Anderson, D., Wright, A.L., 2011. Adaptive cytokine production in early life dif-

ferentially predicts total IgE levels and asthma through age 5 years. J. AllergyClin. Immunol. 128, 397–402.

andberg, M., Frykman, A., Ernerudh, J., Berg, G., Matthiesen, L., Ekerfelt, C., Nilsson,L.J., Jenmalm, M.C., 2009. Cord blood cytokines and chemokines and develop-ment of allergic disease. Pediatr. Allergy Immunol. 20, 519–527.

306 (2013) 59– 67 67

Siegle, J.S., Hansbro, N., Dong, C., Angkasekwinai, P., Foster, P.S., Kumar, R.K., 2011.Blocking induction of T helper type 2 responses prevents development of diseasein a model of childhood asthma. Clin. Exp. Immunol. 165, 19–28.

Sun, Q., Yue, P., Deiuliis, J.A., Lumeng, C.N., Kampfrath, T., Mikolaj, M.B., Cai, Y.,Ostrowski, M.C., Lu, B., Parthasarathy, S., Brook, R.D., Moffatt-Bruce, S.D., Chen,L.C., Rajagopalan, S., 2009. Ambient air pollution exaggerates adipose inflamma-tion and insulin resistance in a mouse model of diet-induced obesity. Circulation119, 538–546.

Sun, Q., Yue, P., Kirk, R.I., Wang, A., Moatti, D., Jin, X., Lu, B., Schecter, A.D., Lippmann,M., Gordon, T., Chen, L.C., Rajagopalan, S., 2008. Ambient air particulate mat-ter exposure and tissue factor expression in atherosclerosis. Inhal. Toxicol. 20,127–137.

Svendsen, E.R., Yeatts, K.B., Peden, D., Orton, S., Alexis, N.E., Creason, J., Williams, R.,Neas, L., 2007. Circulating neutrophil CD14 expression and the inverse associa-tion of ambient particulate matter on lung function in asthmatic children. Ann.Allergy Asthma Immunol. 99, 244–253.

Szabo, S.J., Sullivan, B.M., Stemmann, C., Satoska, A.R., Sleckman, B.P., Glimcher, L.H.,2002. Distinct effects of T-bet in Th1 lineage commitment and IFN-� productionin CD4 and CD8 T cells. Science 295, 338–342.

The U.S. Embassy in Beijing, 2011. Beijing Air Chaoyang District, Beijing. MetOneBAM 1020 and Echotech EC9810 monitors, reporting PM2.5 and ozonereadings. Format for each: pollutant type; concentration; AQI; definition.http://twitter.com/#!/BeijingAir (accessed 29.11.11).

Ulrich, K., Wölfle, S., Mayer, A., Heeg, K., Braunbeck, T., Erdinger, L., Bartz, H., 2010.Extractable organic matter of standard reference material 1649a influencesimmunological response induced by pathogen-associated molecular patterns.Environ. Sci. Pollut. Res. Int. 17, 1257–1267.

van Berlo, D., Hullmann, M., Schins, R.P., 2012. Toxicology of ambient particulatematter. E.X.S. 101, 165–217.

van Donkelaar, A., Martin, R.V., Brauer, M., Kahn, R., Levy, R., Verduzco, C., Villeneuve,P.J., 2010. Global estimates of ambient fine particulate matter concentrationsfrom satellite-based aerosol optical depth: development and application. Envi-ron. Health Perspect. 118, 847–855.

Wang, W., Primbs, R., Tao, S., Simonich, S., 2009. Atmospheric particulate mat-ter pollution during the 2008 Beijing Olympics. Environ. Sci. Technol. 43,5314–5320.

Wong, S.S., Sun, N.N., Fastje, C.D., Witten, M.L., Lantz, R.C., Lu, B., Sherrill, D.L., Gerard,C.J., Burgess, J.L., 2011. Role of neprilysin in airway inflammation induced bydiesel exhaust emissions. Res. Rep. Health Eff. Inst. 159, 3–40.

Xie, S., Yu, T., Zhang, Y., Zeng, L., Qi, L., Tang, X., 2005. Characteristics of PM10, SO2,NO(x) and O3 in ambient air during the dust storm period in Beijing. Sci. TotalEnviron. 345, 153–164.

Xiping, Z., Li, C., Miao, L., Hua, T., 2007. Protecting effects of dexamethasone onthymus of rats with severe acute pancreatitis. Mediators Inflamm. 2007, 72361.

Xu, H., Wang, X., Pöschl, U., Feng, S., Wu, D., Yang, L., Li, S., Song, W., Sheng, G., Fu, J.,2008. Genotoxicity of total and fractionated extractable organic matter in fineair particulate matter from urban Guangzhou: comparison between haze andnonhaze episodes. Environ. Toxicol. Chem. 27, 206–212.

Xu, X., Yavar, Z., Verdin, M., Ying, Z., Mihai, G., Kampfrath, T., Wang, A., Zhong, M.,Lippmann, M., Chen, L.C., Rajagopalan, S., Sun, Q., 2010. Effect of early particulateair pollution exposure on obesity in mice: role of p47phox. Arterioscler. Thromb.Vasc. Biol. 30, 2518–2527.

Yang, C., Peng, X., Huang, W., Chen, R., Xu, Z., Chen, B., Kan, H., 2012. A time-stratifiedcase-crossover study of fine particulate matter air pollution and mortality inGuangzhou, China. Int. Arch. Occup. Environ. Health 85, 579–585.

Yoh, K., Shibuya, K., Morito, N., Nakano, T., Ishizaki, K., Shimohata, H., Nose, M.,Izui, S., Shibuya, A., Koyama, A., Engel, J.D., Yamamoto, M., Takahashi, S., 2003.Transgenic overexpression of GATA-3 in T lymphocytes improves autoimmuneglomerulonephritis in mice with a BXSB/MpJ-Yaa genetic background. J. Am.Soc. Nephrol. 14, 2494–2502.

Yu, Z., Loehr, C.V., Fischer, K.A., Louderback, M.A., Krueger, S.K., Dashwood, R.H.,Kerkvliet, N.I., Pereira, C.B., Jennings-Gee, J.E., Dance, S.T., Miller, M.S., Bailey, G.S.,Williams, D.E., 2006. In utero exposure of mice to dibenzo[a,l] pyrene produceslymphoma in the offspring: role of the aryl hydrocarbon receptor. Cancer Res.66, 755–762.

Zhang, J.J., 2011. Air pollution in China, with Junfeng (Jim) Zhang by Ashley Ahearn.Environ. Health Perspect, doi:10.1289/ehptrp060811.

Zhang, J.J., Smith, K.R., 2007. Household air pollution from coal and biomass fuelsin China: measurements, health impacts, and interventions. Environ. HealthPerspect. 115, 848–855.

Zhang, R.P., Zhang, X.P., Ruan, Y.F., Ye, S.Y., Zhao, H.C., Cheng, Q.H., Wu, D.J.,2009. Protective effect of Radix Astragali injection on immune organs of ratswith obstructive jaundice and its mechanism. World J. Gastroenterol. 15,2862–2869.

Zhao, J., Xie, Y., Qian, C., Li, L., Jiang, R., Kan, H., Chen, R., Song, W., 2012. Imbalanceof Th1 and Th2 cells in cardiac injury induced by ambient fine particles. Toxicol.Lett. 208, 225–231.

Zhou, M., Ouyang, W., 2003. The function role of GATA-3 in Th1 and Th2 differenti-ation. Immunol. Res. 28, 25–37.

Zhu, F., Kandimalla, E.R., Yu, D., Tang, J.X., Agrawal, S., 2004. Modulation ofovalbumin-induced Th2 responses by secondgeneration immunomodulatoryoligonucleotides in mice. Int. Immunopharmacol. 4, 851–862.