Respiratory syncytial virus (RSV) is the leading cause of lower respiratory tract disease in children. Children previously vaccinated with a formalin-inactivated RSV vaccine experienced enhanced morbidity and mortality upon natural RSV infection. Histological analysis revealed the presence of eosinophils in the pulmonary infiltrate of the vaccinated children. Eosinophils are characteristic of Th2 responses, and Th2 cells are known to be necessary to induce pulmonary eosinophilia in RSV-infected BALB/c mice previously immunized with a recombinant vaccinia virus (vv) expressing the RSV G protein (vvG). Using IL-13-deficient mice, we find that IL-13 is necessary for eosinophils to reach the lung parenchyma and airways of vvG-immunized mice undergoing RSV challenge infection. IL-13 acts specifically on eosinophils as the magnitude of pulmonary inflammation, RSV G protein-specific CD4 T cell responses, and virus clearance were not altered in IL-13-deficient mice. After RSV challenge, eosinophils were readily detectable in the blood and bone marrow of vvG-immunized IL-13-deficient mice, suggesting that IL-13 is required for eosinophils to transit from the blood into the lung. Pulmonary levels of CCL11 and CCL22 protein were significantly reduced in IL-13-deficient mice indicating that IL-13 mediates the recruitment of eosinophils into the lungs by inducing the production of chemokines important in Th2 cell and eosinophil chemotaxis.
Respiratory syncytial virus (RSV)3 infection is the leading cause of lower respiratory tract disease in children in the United States resulting in over 100,000 hospitalizations and ∼1,000 deaths per year (1, 2, 3, 4). Infants infected with RSV present with symptoms of pneumonia or bronchiolitis (5, 6). After primary RSV infection, sufficient immunity to prevent re-infection is not induced (7, 8), although symptoms upon re-infection are less severe (5). In addition to young children, studies demonstrate that the elderly and immunocompromised are also at an increased risk for severe disease with RSV (9, 10). For these reasons, much effort has gone into the development of a safe and effective vaccine for RSV, though there has been no success to date. In the 1960s, a formalin-inactivated RSV (FI-RSV) vaccine administered to children caused increased morbidity and mortality upon subsequent natural RSV infection compared with children who received a control formalin-inactivated parainfluenza virus vaccine (11, 12, 13, 14). Histology performed on the deceased revealed an extensive mononuclear cell infiltrate and the presence of eosinophils (11) in contrast to the neutrophils usually observed after a primary RSV infection (15).
The attachment (G) protein of RSV is expressed on the surface of the virion and therefore represents an attractive candidate for inclusion in a vaccine aimed at eliciting RSV G protein-specific neutralizing Abs. The immune response induced by immunization with the G protein in mice has been well studied. Mice immunized with a recombinant vaccinia virus (vv) expressing the G protein of RSV (vvG) also exhibit pulmonary eosinophilia upon challenge RSV infection, mimicking the response of the children that received the FI-RSV vaccine (16). There is an immunodominant CD4 T cell response directed against RSV G183–195 (17) and no detectable RSV G protein-specific CD8 T cell response (18). The RSV G183–195-specific CD4 T cells predominately express the Vβ14 TCR chain and are comprised of a mixture of both Th1 and Th2 cells (17, 19). In mice immunized with vvG and challenged with RSV, depletion of either Vβ14+ T cells (19) or T1/ST2+ cells (a protein expressed on the surface of mouse Th2 cells) (20) prevents the development of pulmonary eosinophilia. These studies indicate that immunization of BALB/c mice with the G protein elicits a CD4 T cell response comprised of both Th1 and Th2 cells and that following challenge RSV infection the G protein-specific Th2 cells induce pulmonary eosinophilia.
Th2 cytokines exhibit multiple effects on eosinophils and may contribute to immunopathology in the lung. Murine studies have revealed that IL-5 promotes the maturation of eosinophils from bone marrow precursors (21), induces the recruitment of eosinophils (22), and promotes the survival of eosinophils (21, 22). In addition to the known roles for IL-5, studies have demonstrated that IL-13 is a chemoattractant for and prolongs the survival of human eosinophils (23). Recent murine studies have suggested that IL-13 may also play an important role in RSV-induced pulmonary injury (24, 25, 26, 27, 28). During primary RSV infection, IL-13 production in the lung triggers airway hyperreactivity and mucus production (25), but has also been shown to inhibit RSV-induced weight loss as well as limit virus replication (28). However, the role of individual Th2-associated cytokines in contributing to disease during secondary RSV infection has been less well defined.
Increased IL-13 production in the lung has been reported in mice undergoing a challenge RSV infection after immunization with a secreted form of the RSV G protein (vvGs) (24, 27, 29, 30). Mice vaccinated with the vvGs exhibit increased disease and pulmonary eosinophilia during a challenge RSV infection as compared with mice immunized with the wild-type vvG. A role for IL-4 or IL-13 in RSV vaccine-enhanced disease in mice immunized with vvGs has been shown by demonstrating a lack of pulmonary eosinophilia in mice deficient for the IL-4Rα chain (27), a chain that is shared between IL-4R and IL-13R (31). Furthermore, vvGs-immunized IL-4-deficient mice and mice depleted of IL-4 develop pulmonary eosinophilia upon RSV challenge (24), indicating that IL-4 is not required for the development of pulmonary eosinophilia in vvGs-immunized mice undergoing a challenge RSV infection. The role of IL-13 in mediating pulmonary eosinophilia was investigated in this model using in vivo IL-13 depletion with an IL-13R antagonist (27). Eosinophilia was not significantly decreased when IL-13 was depleted either at the time of vvGs immunization or at the time of RSV challenge in wild-type mice. In these experiments, eosinophilia was only decreased when IL-13 was depleted immediately before and following vvGs immunization of IL-4-deficient mice. From these studies, it was concluded that IL-13 was sufficient for the development of pulmonary eosinophilia in vvGs-immunized BALB/c mice undergoing challenge RSV infection.
In the present study, we sought to determine the effect of IL-13 deficiency on pulmonary eosinophilia, inflammation, and viral clearance in mice previously vaccinated with the wild-type G protein. Using IL-13-deficient mice, we show that IL-13 is necessary for the development of pulmonary eosinophilia after challenge RSV infection of vvG-immunized mice. We demonstrate that the effects of IL-13 deficiency are specific to eosinophils as the magnitude of pulmonary inflammation, Ag-specific CD4 T cell responses and viral load are not altered in IL-4- or IL-13-deficient mice. We also show that wild-type and IL-13-deficient mice display a similar frequency of eosinophils in the peripheral blood and bone marrow demonstrating that IL-13 is not necessary for the development of eosinophils in vvG-immunized mice undergoing challenge RSV infection. We demonstrate that vvG-primed IL-13-deficient mice undergoing RSV challenge infection have significantly reduced levels of CCL11 and CCL22 in the lung as compared with wild-type controls. Our data suggest that IL-13 mediates the recruitment of eosinophils into the lung by the local production of chemokines that induce Th2 cell and eosinophil chemotaxis. These results have important implications for including the G protein in a potential RSV vaccine.
Materials and Methods
BALB/cAnNCr mice were purchased from the National Cancer Institute (Frederick, MD). IL-4- (BALB/c-Il-4tm2Nnt/J) and IL-4Rα chain-deficient (BALB/c-Il-4ratm1Sz/J) mice on the BALB/c background were purchased from The Jackson Laboratory. IL-13-deficient (32) and IL-4/13-deficient (33) BALB/c mice were provided by A. McKenzie (Medical Research Council Laboratory of Molecular Biology, Cambridge, U.K.). Female mice age 6–8 wk were used for all experiments. All experimental procedures were approved by the University of Iowa Animal Care and Use Committee.
Viruses and infection of mice
Recombinant vv stocks were a gift from G. W. Wertz and T. J. Braciale (University of Virginia, Charlottesville, VA) and J. L. Beeler (U.S. Food and Drug Administration, Bethesda, MD), and were grown in BSC-40 cells (American Type Culture Collection (ATCC)). Mice were infected with 3 × 106 PFUs of recombinant vv expressing either β-galactosidase (β-gal; vvβ-gal) or the G protein of RSV (vvG) by scarification with a 25-gauge needle at the base of the tail. RSV (A2 strain) was a gift from B. S. Graham (National Institutes of Health, Bethesda, MD) and was grown in HEp-2 cells (ATCC). After 3 wk, immunized mice were anesthetized with 30% halothane (Halocarbon Laboratories) in mineral oil (Fisher Scientific) and were given 1–3 × 106 PFUs of RSV intranasally.
Bronchoalveolar lavage (BAL), lung, blood, and bone marrow collection
Peripheral blood was obtained from mice immunized with either vvβ-gal or vvG 7 days after RSV challenge. Blood was collected into 4% sodium citrate (Fisher Scientific), lysed two times with 0.84% NH4Cl, and then cytospun. BAL was performed by three successive washes with 1 ml of PBS to collect cells in the airspace of the lung. BAL cells were then counted and cytospun. Bone marrow was collected by washing femurs with RPMI 1640 medium (Invitrogen Life Technologies) supplemented with 10% bovine growth serum (HyClone Laboratories), 10 U/ml penicillin G, 10 μg/ml streptomycin sulfate, 2 mM l-glutamine (Invitrogen Life Technologies), 5 × 10−5 M 2-ME, 1 mM sodium pyruvate (Invitrogen Life Technologies), 0.1 mM nonessential amino acids (Invitrogen Life Technologies), and 10 mM HEPES (Invitrogen Life Technologies) and was cytospun. All cytospun cells were stained with Diff-Quik (Dade Behring) to determine the percentage and/or total number of eosinophils. The lung vascular bed was perfused by injecting 5 ml of RPMI 1640 medium, supplemented as described, into the right ventricle of the heart. The lungs were then dissected and single cell suspensions were prepared by pressing the lungs though a wire screen (Bellco Glass).
Intracellular cytokine staining and FACS analysis
Lung cells (2 × 106 cells/ml) were stimulated with 1 μM peptide in the presence of 10 μg/ml brefeldin A for 5 h at 37°C in RPMI 1640 medium supplemented as described. After incubation, FACS lysing solution (BD Biosciences) was used to fix the lung mononuclear cells and lyse the RBC. Cells were then washed with permeabilization buffer (staining buffer containing 0.5% saponin; Sigma-Aldrich), blocked with purified anti-FcγRII/RIII mAb (clone 93; eBioscience), and stained with anti-CD4 mAb (clone RM4-5; eBioscience), anti-Vβ14 mAb (clone 14-2; BD Pharmingen), anti-IFN-γ mAb (clone XMG1.2; eBioscience), anti-IL-13 mAb (clone eBio13A; eBioscience), or isotype controls. Cells were collected on a FACSCanto (BD Biosciences). Single color controls were used for compensation. Lymphocytes were gated based on forward and side light scatter properties and were then analyzed using FlowJo software (Tree Star). Background staining was determined using either isotype controls or no peptide controls.
Lungs were harvested from vvβ-gal- and vvG-immunized mice 4 days after RSV challenge. Lungs were homogenized in 1 ml of TRIzol (Invitrogen Life Technologies) and supernatants were collected. RNA was purified by sequential chloroform (200 μl/lung; Fisher Scientific) and isopropylalcohol (500 μl/lung; Fisher Scientific) extraction. Pellets were washed with 70% ethanol and air dried before resuspension in distilled water. RNA was cleaned using the RNeasy Plus Mini kit (Qiagen). cDNA was prepared using a SuperScript First-Strand Synthesis kit for RT-PCR (Invitrogen Life Technologies) and used as a template for real-time PCR. Real-time PCR to detect the nucleocapsid (N) gene of RSV were performed with TaqMan Universal PCR Master Mix (Applied Biosystems) on either an ABI 7000 or 7300 Real Time PCR System (Applied Biosystems) using universal thermal cycling parameters. Results were analyzed using Sequence Detection System Analysis Software (Applied Biosystems). N gene primers and probe were previously published (34) and purchased from Integrated DNA Technologies. The probe was synthesized to contain CFSE reporter dye and 3′-TAMRA quencher dye. Samples were compared with known standard dilutions of a plasmid containing the N gene of RSV. The number of N gene copies per lung was calculated based on the number of copies of the N gene in the sample and the total RNA isolated from the lung.
Whole lungs with the heart attached were harvested from vvβ-gal- and vvG-immunized mice 7 days after RSV challenge. Lungs were placed in 10% formalin (Fisher Scientific) in a vacuum to increase the air to formalin exchange rate. Fixed lungs were processed and embedded in paraffin at the University of Iowa Comparative Pathology Laboratory. Paraffin blocks were sectioned at 5-μm thickness. Sections were H&E stained at the University of Iowa Central Microscopy Core. Eosinophils were detected by immunoperoxidase reaction for the eosinophil-specific major basic protein (MBP) using the avidin-biotin complex method. Sections were deparaffinized in xylenes, rehydrated in graded alcohols, and rinsed. Proteolytic digestion was accomplished using proteinase K (DakoCytomation) for 5 min at room temperature. Endogenous peroxidase activity was quenched using a 3% hydrogen peroxide solution for 8 min at room temperature. Endogenous biotin staining was blocked by application of 1.5% normal rabbit serum (DakoCytomation) for 30 min at room temperature. Sections were covered with rat anti-mouse monoclonal primary Ab, obtained from Drs. N. Lee and J. Lee (Mayo Clinic, Scottsdale, AZ) at a 1/500 dilution in 1.5% rabbit serum, whereas the negative control was rat IgG (Sigma-Aldrich) using a dilution of 1/2760 in 1.5% rabbit serum, incubated for 60 min at room temperature, and rinsed. Sections were then covered with biotinylated anti-rat secondary Ab (Vector Laboratories), using a dilution of 1/200 in 1.5% rabbit serum for 60 min at room temperature, rinsed, and then covered in avidin-biotin complex for 60 min at room temperature and rinsed again. Sections were incubated with diaminobenzidine chromagen for 5 min to demonstrate the signal of the primary Ab. A counterstain of Surgipath hematoxylin was applied for 30 s at room temperature. The sections were then blued in tap water, dehydrated, cleared, and mounted. All rinses were performed using 1× DakoCytomation buffer. Slides were blinded and scored by a board-certified veterinary pathologist (D. Meyerholz, University of Iowa, Iowa City, IA). H&E stained slides were scored from 0 to 5 on a graded scale in which 0 represents undetectable inflammation and 5 represents severe cellular inflammation. Perivascular eosinophils were scored on the following scale: 0, no to rare staining; 1, detectable staining; 2, small clusters; 3, moderately defined aggregates; and 4, well-defined aggregates. Quantification of interstitial eosinophils was calculated based on an average of five random high power fields.
Nunc-Immuno plates (Nalge Nunc International) were coated overnight at 4°C with 2 μg/ml of capture Ab (R&D Systems) in 0.1 M Na2HPO4 (pH 9.0). After washing with PBS-0.5% Tween 20 (Sigma-Aldrich), plates were blocked with RPMI 1640 medium supplemented as described for at least 2 h at room temperature. Whole lungs were harvested and were homogenized using glass douncers (Kontes Glass) in 1 ml of RPMI 1640 medium supplemented as described and containing a 1/200 dilution of protease inhibitor mix (Sigma-Aldrich). BAL and lung cells were centrifuged and 50 μl of supernatant was incubated overnight at 4°C. Recombinant murine CCL11 and CCL22 (R&D Systems) were diluted in PBS plus 10% FCS and used to calculate standard curves. Chemokine was detected by incubation with 0.1 μg/ml biotinylated anti-chemokine Ab (R&D Systems) for 2 h at room temperature. Avidin-peroxidase (1/400 dilution; Sigma-Aldrich) was added for 30 min before plates were developed with 3,3′,5,5′-tetramethylbenzidine dihydrochloride (Sigma-Aldrich). Reaction was stopped after 5 min with 2 N H2SO4 (Ricca Chemical). Plates were read at 450 nm using an ELx800 plate reader and analyzed using KC Junior software (both from Bio-Tek Instruments).
Statistical analyses were performed using GraphPad InStat. Experiments using five groups of mice with normal Gaussian distribution of the data were analyzed using an ANOVA. Data that did not have normal Gaussian distributions were analyzed using a Kruskal-Wallis nonparametric ANOVA with Dunn’s multiple comparisons post-test. Experiments of two groups of mice were analyzed using an unpaired t test, a Welsh corrected t test for instances in which significant differences in SD were found between the groups, or a Mann-Whitney U test in which Gaussian distribution was not observed.
IL-13 is produced by CD4 T cells in the lung
Previous work has suggested that the cytokines induced following RSV infection may contribute to disease pathogenesis (24, 26, 27, 35). Multiple studies have demonstrated increased levels of the Th2-associated cytokine IL-13 in the BAL and lung parenchyma of vvG-immunized mice undergoing challenge RSV infection (17, 19, 24, 27). To determine whether CD4 T cells are the source of the previously detected IL-13, lung cells from vvG-immunized mice were stimulated ex vivo and intracellularly stained. Fig. 1 shows that CD4 T cells are the source of IL-13 in vvG-immunized mice and that IL-13 is produced in an Ag-specific manner. Minimal IL-13 production was detected in vvβ-gal-immunized mice upon Ag stimulation.
Pulmonary eosinophilia is dependent upon IL-13 and signaling through the IL-4Rα chain
RSV infection of BALB/c mice previously immunized with vvG results in lung inflammation, pulmonary eosinophilia, and systemic disease (16, 29). Depletion of Th2 cells has been shown to ablate pulmonary eosinophilia in vvG-immunized mice (20). In this experiment, we sought to determine the role of the Th2-associated cytokines IL-4 and IL-13 in mediating RSV vaccine-enhanced pulmonary eosinophilia and inflammation. Wild-type, IL-4-, IL-13-, IL-4/13-, and IL-4Rα-deficient BALB/c mice were immunized with either vvβ-gal or vvG and challenged 3 wk later with RSV. Cells in BAL were examined 7 days after RSV challenge for the presence of eosinophils. IL-13-, IL-4/13-, and IL-4Rα-deficient mice immunized with vvG exhibited significantly decreased (p < 0.05) percentages (Fig. 2,A) and total numbers (Fig. 2 B) of eosinophils in the BAL compared with wild-type mice. IL-4-deficient mice tended to have a lower frequency and total number of eosinophils in the BAL as compared with immunized wild-type mice, but the difference was not significant (p > 0.05). As expected, mice undergoing a primary RSV challenge (vvβ-gal-immunized) did not exhibit pulmonary eosinophilia. No significant differences (p > 0.05) in total BAL cell counts were detected between any of the mouse strains immunized with vvG (data not shown). These data demonstrate that IL-13 is required for pulmonary eosinophilia in vvG-immunized mice undergoing a challenge RSV infection.
Virus load, the extent of RSV-induced pulmonary inflammation, and the magnitude of the memory CD4 T cell response is unaltered by IL-4 or IL-13 deficiency in vvG-immunized mice
To examine the impact of IL-4 or IL-13 deficiency on pulmonary inflammation, we determined the total number of cells in the lung 7 days after RSV challenge of vvβ-gal- or vvG-immunized mice. No differences in total lung cellularity were detected (Fig. 3,A). To confirm that individual T cell populations within the lung were not altered, we examined both the CD4+ and CD8+ cell compartments. We observed no differences in the total number of CD4+ (Fig. 3 B) or CD8+ cells (data not shown) in vvβ-gal- or vvG-immunized wild-type, IL-4-, IL-13-, IL-4/13-, or IL-4Rα-deficient mice at day 7 after challenge RSV infection.
The CD4 T cell response to the RSV G protein is known to be highly restricted with the majority of the RSV G183–195-specific cells expressing the Vβ14 TCR (17). The RSV G183–195-specific CD4 T cell response consists of both Th1 and Th2 cells (19) with most cells exhibiting a Th1 phenotype based on the production of IFN-γ following ex vivo peptide stimulation (36). Therefore, we next sought to determine the impact of IL-4 or IL-13 deficiency on the RSV G183–195-specific Th1 response. Lung mononuclear cells harvested from vvβ-gal- and vvG-immunized mice 7 days after RSV challenge were stimulated in vitro with RSV G183–195 peptide and assayed for IFN-γ production via intracellular cytokine staining. Fig. 3,C demonstrates that the overall RSV G183–195-specific Th1 response is unaltered in mice deficient in IL-4 or IL-13. In addition, the IFN-γ production by the Vβ14+ subpopulation of CD4 cells in response to RSV G183–195 peptide stimulation was also unaltered (Fig. 3 D).
To determine the effect of IL-4 or IL-13 deficiency on viral clearance, we examined the lungs from wild-type, IL-4-, IL-13-, IL-4/13-, and IL-4Rα-deficient mice immunized with vvβ-gal or vvG and challenged with RSV for viral load using real-time PCR. Viral load was determined by measuring the number of copies of the N gene in the lungs 4 days after RSV challenge, which is the time at which peak virus titers are generally obtained as measured by plaque assay in mice undergoing a primary RSV infection. We found that the amount of detectable N gene in the vvG-immunized IL-4-, IL-13-, IL-4/13-, or IL-4Rα-deficient mice did not significantly differ (p > 0.05) from the amount observed in vvG-immunized wild-type mice (Fig. 4). The vvβ-gal-immunized cytokine-deficient mice did not exhibit significantly (p > 0.05) different copy numbers of the N gene compared with the vvβ-gal-immunized wild-type mice. As expected, Fig. 4 also shows that the mice vaccinated against RSV (vvG) have decreased viral load compared with mice that are undergoing a primary RSV infection (vvβ-gal). These results indicate that virus levels are unaffected by the presence or absence of IL-4 and/or IL-13.
Eosinophilia is decreased in the lung parenchyma of IL-13-deficient mice, but the magnitude of pulmonary inflammation is not altered
Our results demonstrate that IL-13 is required for airway (i.e., BAL) eosinophilia in vvG-immunized mice undergoing challenge RSV infection (Fig. 2). Thus, we sought to determine whether eosinophilia was also altered in IL-13-deficient mice in the lung parenchyma. Whole lungs were harvested from vvβ-gal- or vvG-immunized wild-type or IL-13-deficient mice 7 days after RSV challenge. Eosinophils were detected in lung sections by staining with a mAb against the MBP. Eosinophils were detectable in the lung parenchyma of vvG-immunized wild-type mice, both in perivascular (Fig. 5,A) and interstitial (Fig. 5,B) regions. In vvG-immunized IL-13-deficient mice, eosinophilia in both perivascular and interstitial regions was significantly decreased (p < 0.05) compared with wild-type mice. Representative perivascular sections of vvβ-gal-immunized wild-type mice (Fig. 5,C), vvG-immunized wild-type mice (Fig. 5,D), vvβ-gal-immunized IL-13-deficient mice (Fig. 5,E), and vvG-immunized IL-13-deficient mice (Fig. 5,F) demonstrate the eosinophil-specific staining and lack of background staining. Examples of eosinophil detection in interstitial regions of vvG-immunized wild-type (Fig. 5,G) and IL-13-deficient (Fig. 5 H) mice are also shown. Interstitial areas of vvβ-gal-immunized wild-type and IL-13-deficient mice were similar to vvG-immunized IL-13-deficient sections (data not shown).
Because of the differences in eosinophilia that we observed in the various regions of the lung parenchyma, we wanted to determine whether the magnitude of inflammation was altered in any region of the lung in IL-13-deficient mice. No significant differences (p > 0.05) were found in the airway (Fig. 6,A), alveolar (Fig. 6,B), perivascular (Fig. 6,C), or peribronchiolar (Fig. 6,D) regions of the lung between vvG-immunized wild-type and IL-13-deficient mice. Representative sections are shown for vvβ-gal-immunized wild-type (Fig. 6,E) and IL-13-deficient mice (Fig. 6,G), as well as vvG-immunized wild-type (Fig. 6,F) and IL-13-deficient mice (Fig. 6 H). These data indicate that IL-13 specifically affects the influx of eosinophils into the lungs of vvG-immunized mice undergoing RSV challenge infection.
Eosinophils are detectable in the blood and bone marrow of IL-13-deficient mice
The difference in eosinophils detected in the BAL and lung parenchyma of IL-13-deficient mice led us to question whether vvG-immunized IL-13-deficient mice could produce eosinophils upon RSV challenge. Blood and femoral bone marrow were collected from vvβ-gal- and vvG-immunized wild-type and IL-13-deficient mice 7 days after RSV challenge. Eosinophils were detectable at similar levels in the blood (Fig. 7,A) and bone marrow (Fig. 7 B) of IL-13-deficient and wild-type mice. These results demonstrate that vvG-primed mice generate eosinophils from the bone marrow and that the eosinophils make it into the blood in the absence of IL-13. However, IL-13 is required for the eosinophils to enter the lung or for survival of eosinophils within the lung.
CCL11 and CCL22 are decreased in the lungs of IL-13-deficient mice
IL-13 has previously been shown to induce production of CCL22 (37, 38), a chemoattractant for Th2 cells (38) that are known to be necessary for the development of pulmonary eosinophilia in vvG-immunized mice undergoing challenge RSV infection (20). In addition, mice immunized with vvG and challenged with RSV have increased levels of pulmonary CCL11 (39), a chemoattractant for eosinophils (40). To determine whether IL-13 leads to the recruitment of eosinophils into the lung through production of eosinophil and Th2 cell chemotactic factors, we examined the presence of CCL11 and CCL22 in the lungs of vvβ-gal- and vvG-immunized wild-type and IL-13-deficient mice. Lung supernatants harvested 3 days after RSV challenge were tested directly ex vivo in an ELISA. IL-13-deficient mice immunized with vvG had significantly decreased (p < 0.05) amounts of CCL11 protein in the lung (Fig. 8,A) as compared with vvG-immunized wild-type mice. Significantly decreased (p < 0.05) amounts of CCL22 protein were also detected in the lungs (Fig. 8,B) of vvG-immunized IL-13 deficient mice. In contrast, IFN-γ protein levels were not decreased in IL-13-deficient mice (Fig. 8 C), supporting our intracellular cytokine staining data and demonstrating that the decrease of CCL11 and CCL22 in IL-13-deficient mice is not due to an overall decrease in cytokine and chemokine production. These data suggest that IL-13 recruits eosinophils to the lung through a CCL11- and/or CCL22-dependent mechanism.
The data presented in this study demonstrate that IL-13 is required for the development of pulmonary eosinophilia in vvG-immunized mice undergoing challenge RSV infection. Previous work examined the role of IL-13 following immunization with a vv encoding only the secreted form of the RSV G protein (vvGs) (27). Wild-type BALB/c vvGs-immunized mice depleted of IL-13 at either the time of immunization or at the time of challenge did not exhibit any change in pulmonary eosinophilia (27). However, pulmonary eosinophilia was significantly decreased when IL-13 depletion was performed at the time of vvG-immunization of IL-4-deficient mice (27). These studies indicated that IL-13 is sufficient for the development of pulmonary eosinophilia.
Our results using mice genetically deficient in IL-13 that are immunized with vvG demonstrate that IL-13 is necessary for the development of pulmonary eosinophilia. Several experimental differences may explain the discrepancy between our current data and previous studies. The depletion of IL-13 in the previous investigation was done either at the time of vaccination or at the time of challenge (27), whereas our current study used IL-13-deficient mice. This difference suggests that in wild-type mice, the presence of IL-13 at either immunization or challenge is sufficient for the development of pulmonary eosinophilia and only in the complete absence of IL-13 is pulmonary eosinophilia inhibited. Another experimental difference between our studies and those previously published is the form of the G protein used for immunization. In this study, we use vvG, which encodes for both the secreted and membrane-bound forms of the RSV G protein, whereas the previous work used vvGs, which only encodes for the secreted form (27). Previous studies have established that priming with vvGs elicits more eosinophilia than priming with vvG (24, 29). Therefore, it is formally possible that vvGs and vvG have unique cytokine requirements necessary for the induction of pulmonary eosinophilia. For example, the presence of the membrane form of the G protein may promote the development of eosinophilia through a mechanism that relies solely on IL-13 rather than both IL-13 and IL-4.
We further demonstrate that IL-13 has specific effects on eosinophils as the magnitude of the lung inflammation (Figs. 3,A and 6) and total BAL infiltrates (data not shown) are not altered in IL-13-deficient mice. This supports previous work in vvGs-immunized mice in which no difference in bronchovascular, perivenous, or interstitial inflammation was observed after IL-13 depletion or in mice deficient for IL-4 (27). It is interesting, however, that the previous study found that the interstitial inflammation was significantly decreased when IL-13 was depleted from IL-4-deficient mice before and following vvGs immunization (27). In contrast, we demonstrate that IL-4 and IL-13 deficiency does not alter the magnitude of the inflammation in the lung (Figs. 3,A and 6). The minor discrepancy between our findings and that of Graham and colleagues (27) may relate to the differences in pulmonary inflammation induced during challenge RSV infection of mice previously immunized with vvG vs vvGs as described. Our data demonstrate that eosinophilia induced by prior vaccination with the wild-type G protein occurs via an IL-13-dependent mechanism, whereas the mechanisms that control the extent of the pulmonary inflammation and injury are IL-13-independent.
The role of IL-13 during primary RSV infection has recently been examined in the murine model. IL-13 is induced following a primary infection with a clinical isolate of RSV (25, 26). Treatment of acutely infected mice with an anti-IL-13 Ab results in decreased airway hyperreactivity, decreased mucus production, and decreased levels of RSV Ag (25). Together, these data suggest that IL-13 may also be a mediator of specific aspects of RSV-induced disease during primary RSV infections. Our current study focused on the role of IL-13 during a secondary response to RSV in mice previously immunized with the G protein. During a secondary response to RSV, we did not observe a role for IL-13 in clearance of RSV as was previously observed during primary RSV infection of mice (25). However, it is important to point out that the previous study used a clinical isolate of RSV that has recently been shown to induce significantly higher production of IL-13 in the lung than the strain of RSV used in our current study (41).
Recent work using IL-5 transgenic mice has suggested that eosinophils decrease viral titer during an acute infection with RSV (42). Interestingly, an increase in viral titer has previously been observed during primary RSV infection of IL-13-deficient mice on a C57BL/6 background (28). Conversely, over-expression of IL-13 in C57BL/6 mice was shown to protect RSV infected mice from increased viral load (28). Taken together, these data suggest that eosinophils or IL-13 may help to accelerate the clearance of RSV during a primary infection. In this study, during RSV challenge of vvG-immunized BALB/c mice, we observe no difference in the number of N gene copies in IL-13-deficient mice compared with wild-type mice (Fig. 4 B), which is consistent with previous results obtained in the vvGs immunization system (27). Our results suggest that during a secondary response to the RSV G protein, eosinophils and/or IL-13 do not contribute to virus clearance in BALB/c mice.
In vvG-immunized mice challenged with RSV, IL-13 is produced by RSV G183–195-specific CD4 T cells (17) (Fig. 1). The Vβ14+ subset of CD4 T cells is thought to be the major source of IL-13 in vvG-immunized mice (19). Depletion of Vβ14+ cells from mice immunized with either vvG or vvGs reduces the amount of IL-13 produced (43). In vvG-immunized mice undergoing an RSV challenge infection, the vast majority of Vβ14+ cells produce IFN-γ following ex vivo peptide stimulation. In our IL-13-deficient mice, we show that the Vβ14+ IFN-γ response is not altered (Fig. 3 D). These data suggest that, although depletion of all Vβ14+ cells can decrease pulmonary eosinophilia, it appears to be the relatively small subset of Vβ14+ cells that produce IL-13 that actually mediate the development of pulmonary eosinophilia during RSV challenge of vvG-immunized mice.
RSV challenge of FI-RSV-immunized mice (44) and macaques (45) also leads to IL-13 production. FI-RSV induced IL-13 expression is G protein-independent as mice immunized with FI-RSV lacking either the RSV G183–195 epitope or the whole G protein do not have altered IL-13 expression (46). It has been shown, however, that IL-4 deficiency or depletion of IL-13 in FI-RSV-immunized mice is sufficient to prevent the development of pulmonary eosinophilia upon RSV challenge (27). In addition, pulmonary eosinophilia in FI-RSV-immunized mice is also independent of Vβ14+ cells as depletion of Vβ14+ cells does not reduce pulmonary eosinophilia after RSV challenge (43). These data highlight some of the important differences between FI-RSV and vvG vaccination and suggest that IL-13 may contribute to RSV vaccine-enhanced disease in both vaccination models through independent mechanisms.
In IL-13-deficient mice, we did not detect eosinophils in the lung parenchyma (Fig. 5). Eosinophils were detected, however, at wild-type levels in both the blood and the bone marrow (Fig. 7). This demonstrates that IL-13-deficient mice are capable of generating eosinophils upon vvG-immunization and RSV challenge, and suggests that IL-13 is necessary for the recruitment of eosinophils from the blood into the lung in this model. One possible mechanism of recruitment could be IL-13-mediated up-regulation of VCAM-1 on epithelial cells that would allow tethered eosinophils to extravasate into the lung tissue (47, 48). It also remains possible that IL-13, which is a known survival factor for eosinophils (23), is necessary for the survival of eosinophils within the lung. We believe this mechanism is unlikely as we were unable to detect apoptotic bodies of eosinophils with the anti-MBP staining shown in Fig. 5. An additional possibility is that IL-13 is necessary for the increased production of chemokines that contribute to the recruitment of eosinophils. IL-13 induces the production of CCL22 (37, 38), a chemoattractant for Th2 cells (38). Fig. 8,B demonstrates that CCL22 protein is significantly decreased in the lungs of IL-13-deficient mice. The necessity of Th2 cells for the development of pulmonary eosinophilia in vvG-immunized mice undergoing challenge RSV infection has been established (20). In this context, our data suggest that the recruitment of Th2 cells by IL-13 is important for the downstream recruitment of eosinphils. CCL11 is known to be produced during RSV challenge of vvG-immunized mice (39), and depletion of CCL11 significantly decreases eosinophilia (49). We show that vvG-immunized IL-13-deficient mice undergoing challenge RSV infection have significantly decreased amounts of CCL11 protein in the lung as compared with wild-type mice (Fig. 8 A). Together, these data suggest that IL-13 recruits eosinophils to the lung through mechanisms that are dependent upon the production of CCL11 and CCL22.
IL-13 is an important mediator of RSV-induced disease during a primary infection. In this study, we demonstrate that IL-13 is necessary for pulmonary eosinophilia in mice previously immunized with the RSV G protein. We also show that other measures of RSV vaccine-enhanced disease occur independently of IL-13. Our data suggest that IL-13 acts to specifically recruit eosinophils from the blood into the lung through a CCL11- and CCL22-dependent mechanism.
We thank Stanley Perlman, Kevin Legge, and John Harty for critically reviewing the manuscript; Kathryn Chaloner and Jeffrey Dawson for assistance with statistical analyses; and Stacey Hartwig for excellent technical assistance.
The authors have no financial conflict of interest.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work was supported by funding from the American Heart Association Predoctoral Fellowship (to E.M.C.), University of Iowa Department of Pathology (to D.K.M.), and National Institutes of Health Grant AI 063520 (to S.M.V.).
Abbreviations used in this paper: RSV, respiratory syncytial virus; FI-RSV, formalin-inactivated RSV; vv, vaccinia virus; β-gal, β-galactosidase; BAL, bronchoalveolar lavage; MBP, major basic protein; N, nucleocapsid protein.