Background
Exposure to elevated levels of airborne particulate matter (PM) pollution has been associated with adverse health outcomes such as asthma and chronic obstructive pulmonary disease (COPD) [
1,
2]. Interestingly, several epidemiological studies have also recognized an association between exposure to elevated levels of PM and respiratory tract viral infections in infants and young children [
3‐
5]. Due to the heterogeneous nature of PM, the associated health effects of exposure to elevated levels of PM could not be attributed to a specific component. Recent studies by our collaborators identified the presence of “environmentally persistent free radicals” (EPFRs) in PM samples that were collected in various cities of United States [
6]. Though epidemiological studies suggest that infants and young children are susceptible to adverse health effects of PM exposure [
7], very limited mechanistic information is available to understand the impact of PM exposure in determining the outcome of respiratory tract infections in this immunologically immature subpopulation. Our previous studies demonstrated that acute early-life exposure to EPFR containing PM induces oxidative stress, inflammation and airway hyper-responsiveness in the lungs [
8]. These responses were accompanied by airway epithelial barrier dysfunction and pulmonary remodeling as evidenced by airway smooth muscle thickening as a result of epithelial to mesenchymal transition [
9]. We demonstrated that exposure to EPFRs induces a temporary immunosuppressive pulmonary environment in an oxidative stress dependent manner resulting in enhanced disease severity following influenza virus infection [
10,
11]. Specifically, we observed reduced adaptive T cell responses including reductions in protective T cell responses (i.e. Tc1 and Th1) following influenza infection [
10]. While we expected reductions in protective T cell responses due to the age of infection [
12], EPFR exposure further exacerbated disease as was evidenced by further reductions in protective T cell numbers and enhanced influenza morbidity and mortality.
In the current study, we sought to determine the mechanism of pulmonary immunosuppression observed during early-life exposure to EPFRs and its role in the exacerbation of the influenza disease severity. Since we had previously observed an increase in Tregs in the lungs of neonatal mice exposed to EPFRs [
11], we hypothesized that Tregs, and their effector cytokine IL10, were critical for EPFR-induced exacerbation of influenza disease severity in neonates. To address this hypothesis, we exposed neonatal mice deficient in Tregs to EPFRs and then infected them with influenza virus. Corollary experiments in which Tregs from EPFR-exposed neonatal mice were adoptively transferred into air-exposed neonatal mice were then performed to demonstrate the direct role of EPFR-induced Tregs in mediating the suppression of adaptive T cell responses and exacerbation of influenza disease severity. We also studied whether IL10 was the effector molecule responsible for EPFR-enhanced flu severity in neonatal mice. Our data demonstrates a mechanism whereby exposure to EPFRs alter pulmonary immune responses via the induction of Tregs and IL10 and modulate the protective immune responses against influenza infection.
Methods
Combustion derived PM
Radical containing combustion derived ultrafine PM with a mean aerodynamic diameter of ~0.2 μm was generated and characterized by our colleague Dr. Slawo Lomnicki at Louisiana State University as described earlier [
8,
13]. Briefly, particles are composed of 95% silica and 5% CuO and vapors of 1,2-dichlorobenzene are dosed as an adsorbate inside a custom made vacuum exposure system for 5 min at 10 torr at 230 °C. The surface area of these particles is 460 m
2/g and the density is 2.2 g/cm
3. Our previous studies determined the size of the particles in a monodispersion and the mean aerodynamic diameter (MAD) of these particles was found to be 0.195 μm. Analysis using flow cytometry and transmission electron microscopy (TEM) showed that particles once suspended were monodispersed with little or no aggregates [
14]. Further, our previous studies demonstrated that aerosol delivery of ultrafine particles under a constant rate could deposit the particles in the lower respiratory tract of neonatal rodents [
15]. For aerosol delivery, DCB particles (PM with chemisorbed EPFRs previously published as DCB230 [
10]) were suspended in 25 ml of irrigation saline containing 0.02% Tween-80 at a final concentration of 0.2 mg/ml. Particles were monodispersed by probe sonication immediately prior to exposure.
Animals
C57BL/6NHsd breeders were originally purchased from Harlan (Indianapolis, IN). IL10 knock out mice (B6.129P2-Il10tm1Cgn/J) (stock number-002251) were purchased from Jackson laboratories (Bar Harbor, ME). These mice were homozygous for the Il10
tm1Cgn
targeted mutation. Neonatal mice were not identified for sex and both sexes were used for experiments. Separate cohorts of mice were used for independent experimental endpoints. All mice were housed in ventilated cages and supplied with filtered air in a specific pathogen free environment with free access to food and water under controlled conditions with 12 h light/dark cycle, temperature and humidity. Mice were time-mated to obtain maximum number of birth cohorts. All animal protocols were prepared and followed in accordance with Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee.
Neonatal exposure to EPFRs and influenza infection
Three day old neonatal mice were exposed to DCB at 200 μg/m
3 or filtered air for 30 min/day for seven consecutive days. Modeling data (MPPD v2.0) was used to produce an equivalent particle deposition per alveolus compared to that of an infant human being exposed over a 24 h time period to an average of 35 μg/m
3 of PM
2.5. The physiological parameters that were incorporated in the algorithm and the calculations used to determine air flow in the chamber were described previously [
11]. Recent development of computational methods to quantify the transport and deposition of particles in human respiratory tract suggests the deposition of ultrafine particles in the lower respiratory tract [
16]. Since respiratory mechanical parameters in mice were found to be similar to rats [
17], mouse modeling data was obtained from literature and default rat lung parameters were overwritten to correct for physiological parameters with the assistance of Drs. Price and Asgharian (MPPD support). At four days post exposure (dpe), neonates were infected intranasally (i.n.) with mouse-adapted human influenza A strain (PR/8/34 H1N1) (Advanced Biotechnologies, MD) at 1.25 TCID
50/neonate diluted in 10 μl Dulbecco’s Phosphate-Buffered Saline (DPBS) or sham infected with the same volume of DPBS.
Pulmonary viral load
Neonatal mice were sacrificed at 4 days post-infection (dpi) and lungs were isolated. Lungs were homogenized and peak pulmonary viral loads were determined by using Reed-Muench calculator [
18]. Briefly, Madin-Darby Canine Kidney (MDCK) epithelial cells were seeded at a density of 1.2× 10
4 cells/well in a 96 well plate. Lungs isolated at 4dpi were homogenized in 1 ml DPBS on ice and centrifuged at 2000xg for 10 min at 4 °C. Cells were inoculated with a 10-fold dilution series of the supernatant obtained from the homogenized lungs and incubated at 37 °C with 5% CO
2 for five days. Signs of influenza-induced cytopathic effect was determined daily and the scoring for cytopathic effect was finalized on day five and TCID
50 was calculated.
Depletion of Tregs
Tregs were depleted in the lungs of neonatal mice using the purified anti-mouse CD25 antibody (PC61 clone) (BioLegend, San Diego, CA). Briefly, 5 μg of anti-CD25 antibody in 10 μl volume was administered i.n. to neonatal mice every two days on days 0 and 3, and 6 dpe at 2 h prior to the exposure to DCB. Control mice received similar amount of rat IgG isotype control antibody (10 μl). All mice were exposed to DCB from day 3 to day 9 of age for seven consecutive days. Treg depletion efficiency was determined at 4 dpe to verify whether Tregs were depleted before infection and also at 10 dpe to verify whether depletion is maintained until 10 dpe. Lungs were isolated on days 4 dpe and 10 dpe and a single cell suspension was prepared to stain for markers CD3, CD4, CD25, and transcription factor FoxP3. Percentage of Tregs in the lung was determined using flow cytometry and cells positively stained for surface markers CD3, CD4, CD25 and transcription factor Foxp3 were considered Tregs.
Treg Adoptive Transfer
Lungs were isolated from DCB exposed neonatal mice at 10 dpe. Single cell suspension of the isolated lungs was pre-enriched for CD4+ T cells using biotinylated anti-mouse CD4 antibody and Cellection Biotin Binder kit (Life Technologies, Grand Island, NY), following manufacturer’s instruction. Subsequently, CD4+ enriched fraction was used to isolate CD4+CD25+ cells (Tregs) using Easysep Mouse CD4+CD25+ Regulatory T cell isolation kit (Stem cell Technologies, Vancouver, Canada), following manufacturer’s instructions. The amount of Tregs transferred into each mouse was equivalent to the amount we observed in the lungs of DCB treated mice plus 30% extra to account for losses during transfer. Thus, since we observed ~5.0*10e5 CD25+ Tregs in the lungs of EPFR exposed neonatal mice, we transferred ~6.4*10e5 CD4+CD25+ Tregs in 10 μl of serum free media to air-exposed neonatal mice i.n. 2 h prior to infection with influenza virus. All mice were sacrificed on 6 dpi (10 dpe) and analyzed for Th1 and Tc1 responses by means of intracellular cytokine staining and flow cytometry.
Treatment with recombinant IL10 (rIL10)
To understand the effect IL10 in EPFR-induced exacerbation of influenza disease severity, air exposed WT neonatal mice were treated with rIL10 (BioLegend, San Diego, CA) (150 pg per mice via i.n. route) at 4 dpe. Mice were subsequently infected with influenza virus (1.25 TCID50per mouse) at 2 h-post transfer. The amount of IL10 administered was equivalent to IL10 observed in EPFR-exposed mice at 4dpe and total amount instilled was adjusted for transfer efficiency (i.e. +30%).
Flow cytometry and intracellular staining
Mice were sacrificed and blood was gently cleared from the lungs using retrograde vascular perfusion using Hank’s Balanced Salt Solution (HBSS). Lungs were isolated and a single cell suspension was prepared as described earlier [
10,
19]. Briefly, lungs were dissected and collected into 2.4 ml ice-cold HBSS containing 1 mg/ml collagenase I (Invitrogen NY) and 150 ng/ml DNase (Sigma Aldrich, NY) per neonatal lung and mechanically dissociated using Octodissociator (Miltenyi, Germany). Dissociated lungs were incubated at 37 °C for 30 min with continuous shaking (200 rpm). Following incubation, lungs were dissociated with Octodissociator for a second time and remaining cell clumps were filtered through 40 μm cell strainer to obtain a single cell suspension. The resulting cell suspension was treated with RBC lysis buffer (BioLegend, San Diego, CA) to lyse residual RBCs. For intracellular staining, cells were stimulated for five hours at 37 °C in RPMI 1640 media containing 5% heat-inactivated fetal bovine serum, 5 ng/ml phorbol 12-myristate 13-acetate, 500 ng/ml Ionomycin (Sigma-Aldrich) in the presence of a protein transport inhibitor (GolgiPlug, BD Biosciences, San Jose, CA). T helper-1 cells (Th1) subsets were identified as CD3
+, CD4
+, and IFNγ
+ and cytotoxic T cell-1 (Tc1) subsets were identified as CD3
+, CD8
+, and IFNγ
+. Tregs were identified as CD3
+, CD4
+ CD25
+, and intracellular Foxp3
+ cells. The antibodies used for intracellular staining were: eFluor450-CD3 (clone- 17A2; eBioscience, CA), FITC- CD8 (clone- 53–6.7; eBioscience), PerCP-CD4 (clone- RM4-5; BioLegend, CA), and PE-IFNγ (clone- XMG1.2; eBioscience). For Treg staining, following antibodies were used eFluor450-CD3 (clone-17A2; eBioscience), PerCP-CD4 (clone - RM4-5; BioLegend), PE-CD25 (clone- PC61.5; eBioscience), and FITC-Foxp3 (clone- FJK-16 s; eBioscience). Fixable live/dead marker eFluor 780 (eBioscience, San Diego, CA) was used to exclude dead cells from analysis. Following antibody staining, cells were analyzed with FACS Canto II (BD Biosciences) and flow cytometry data was analyzed using FlowJo Software v.10.1 (Tree Star, OR).
IL10 and TNFα ELISA
Lungs were isolated from neonatal mice following retrograde vascular perfusion. Isolated lungs were flash frozen in liquid nitrogen and stored at -80 oC until processed further for cytokine analysis. Briefly, lungs were homogenized in 1 ml of Tissue Protein Extract Reagent (T-PER) (Life Technologies, Grand Island, NY) containing protease inhibitor cocktail (Thermo Scientific, Waltham, MA). Homogenates were centrifuged at 10,000xg for 5 min and supernatants were collected for analysis. IL10 levels were determined by mouse IL10 quantikine ELISA kit following manufacturer’s (R&D systems, Minneapolis, MN) instructions. TNFα levels were measured using Legend Max mouse TNFα ELISA kit (BioLegend, San Diego, CA) following manufacturer’s instructions. Data was normalized to the total protein content in the supernatants as determined by BCA protein assay (Life Technologies, Grand Island, NY).
Immunohistochemistry
Neonatal mice were sacrificed and lungs were excised and inflated at a constant fluid pressure of 25 cm with Zinc-formalin and fixed overnight as described earlier [
11]. Subsequently, the lungs were transferred to 70% ethanol until ready for embedding. Lungs were dehydrated, paraffin-embedded and sectioned into 5 μm thick sections and mounted onto glass slides. Immunohistochemical staining was performed incubating with rabbit polyclonal anti-human H1N1 influenza virus antibody (Clontech Laboratories, Mountain View, CA) at a concentration 5 μg/ml in PBS containing 1% BSA. Tissue sections were washed with Tris Buffered Saline containing 0.5% Tween 20 (TBST) and incubated with anti-rabbit IgG (ImmPRESS anti-rabbit IgG kit, Vector Laboratories, Burlingame, CA). Following incubation, sections were washed in TBST and incubated with peroxidase substrate solution until desired stain intensity was developed. Sections were further counter stained with hematoxylin and observed for lung areas positively stained for influenza virus.
Statistics
All results were expressed as mean ± SEM and were analyzed using GraphPad Prism 6 (GraphPad Software Inc., Version 6.0.7, La Jolla, CA). One-way ANOVA or Two-way ANOVA with Tukey’s multiple comparison’s test with Holm-Sidak’s correction for multiple comparisons, and unpaired t-tests were performed to determine the difference between groups. p values less than 0.05 were considered statistically significant.
Discussion
Exposure to PM from combustion sources has been shown to affect pulmonary immunologic homeostasis and exacerbate respiratory health effects such as asthma [
8,
20,
21]. The association between exposure to elevated levels of PM and enhanced susceptibility to respiratory tract infections in children has been documented. [
3,
5,
22‐
25]. Although studies have indicated this association, very limited information is available on the potential mechanisms and biological rationale associated with these phenomena. Yet, it is important to understand how exposure to elevated levels of PM increase susceptibility to respiratory tract infections. Particle pollutants such as diesel exhaust particles were shown to reduce host defense in adult mice models of influenza infection [
26], however studies from our group and other groups have demonstrated that infant responses to infectious agents and PM are markedly different from that of adults [
12,
27‐
29]. These differences could be due to several factors including age and strain of the mice, the composition and dose of PM, the type of the infectious agent, timing of the exposure to PM (before and during vs. after infection), etc. Effects of exposure to PM are particularly detrimental in infants and young children; and children are particularly at risk to respiratory infections compared to adults. Our previous studies with both infectious and non-infectious agents showed that adults and neonates exhibit marked differences in immunological responses [
11,
12,
30]. Since infancy is a time of rapid pulmonary and immunological development, insults to the respiratory and immune system during this critical developmental period may result in exaggerated manifestation of disease and/or long-term health effects.
Since infants are particularly susceptible to PM pollution and their immune systems are often characterized as immature, it is important to understand how early life-exposure to EPFR containing PM increases severity of respiratory tract infections such as influenza. Our previous studies demonstrated that early-life exposure to PM alters airway barrier function [
9], and our more recent data demonstrates that such exposure in neonates also impairs protective T effector cell responses towards influenza virus infection [
10]. Like others, we demonstrated a role for PM-induced oxidative stress [
27]; and specifically, we found that particles containing radicals and inducing oxidative stress were necessary for increased viral load and enhanced morbidity and mortality following influenza infection [
10]. Apart from realizing that oxidative stress was necessary for this immunosuppressive environment, the mechanism responsible for PM-induced immunosuppression was unclear. Using a neonatal mice model of PM exposure and infection, we have successfully recapitulated the enhanced disease due to respiratory tract viral infection following exposure to elevated levels of PM [
10] as reported in several epidemiological studies [
3,
5,
22‐
24]. In the current study, using early-life exposure, we explored the mechanism of EPFR-induced pulmonary immunosuppression and subsequent increase in influenza disease severity.
Here, we demonstrate that Tregs are responsible for creating and maintaining the immunosuppressive environment in the lungs of EPFR-exposed mice. In fact, depletion of Tregs in EPFR-exposed neonatal mice resulted in increased protective T cell responses (i.e. Tc1 and Th1) following influenza infection. Further, adoptive transfer of Tregs from EPFR-exposed neonates to air-exposed neonates suppressed protective T cell responses toward influenza in the recipient air-exposed neonatal mice. We also observed an increase in IL10 in the lungs of EPFR-exposed neonates. Interestingly, in EPFR-exposed neonates IL10 expression was bimodal (i.e., increasing immediately after exposure to DCB, returning to baseline and then peaking at 6 dpi). Administration of rIL10 to air-exposed mice prior to influenza infection recapitulated many of the effects observed in EPFR-exposed mice infected with influenza including increased pulmonary viral load and decreased weight gain. Mice deficient in IL10 exhibited decreased viral load and morbidity compared to WT neonates exposed to EPFRs and infected with influenza virus. Interestingly, Treg and IL10 induction occurred in the absence of influenza infection indicating that early life exposure to EPFRs activates a non-specific immunosuppressive environment in the lungs. Thus, immunosuppression of adaptive T cell responses against influenza appears to be a “bystander” effect that allows for enhanced severity of respiratory tract viral infections.
In addition to Tregs, we observed a significant increase in IL10 expression in the total lung homogenates of neonatal mice exposed to EPFRs. IL10 is an immunoregulatory cytokine produced by a wide range of immune cell types including Tregs and plays a role in regulating the growth and differentiation of various immune cells [
31] and in maintaining immune homeostasis by regulating the adaptive immune responses [
32,
33]. It inhibits the activity of Th1 cells, NK cells, and macrophages - all of which are required for pathogen clearance during infection [
34]. Our data demonstrated that IL10 is sufficient to exacerbate the morbidity and pulmonary viral load associated with influenza infection (Fig.
4b, c, and d, respectively). While influenza infection alone is known to induce increased expression of IL10 at the time of peak adaptive T cell response [
35], we also observed an increase in IL10 levels in the lungs of EPFR exposed neonates at 1 dpe (Fig.
1b), prior to influenza infection. Although beyond the scope of this manuscript, early IL10 is most likely produced by monocytes and may play a role in the expansion of inducible Tregs [
36].
Our results demonstrating increases in Tregs and IL10 in EPFR-exposed mice are not dissimilar to results from patients infected with the 2009 pandemic influenza A H1N1 virus (pH1N1). Even though these studies looked at different compartments (i.e. peripheral blood versus pulmonary), they also observed an increase in the frequency of Tregs in infected patients that correlated with decreased frequencies of CD3
+, CD4
+ and CD8
+ T cells [
37]. And in a separate study, elevated levels of IL10 in sera were positively correlated to severity of disease in patients infected with pH1N1 [
38]. It is interesting to postulate that severe lung injury resulting from EPFR exposure in neonatal mice or following pH1N1 infection is a common mechanism responsible for the induction of Tregs and reduction of protective T cells responses and we are currently exploring these possibilities.
Finally, we have previously shown very different immunological responses to similar acute EPFR exposure protocols in adult mice. In fact, Th17 mediated immune responses, which was accompanied by significant pulmonary neutrophilia, predominated in the lungs of adult mice exposed to EPFRs [
21]. The reasons for such differences are currently unclear, but they could be a result of differences in maturation status of the immune and/or pulmonary systems.
Acknowledgements
Dr. Slawo Lomnicki and the Superfund materials core supplied DCB230. Authors would like to thank Dr. Farhana Hasan of Louisiana State University for her assistance in synthesis and characterization of combustion derived PM.