IL-4Rα on dendritic cells in neonates and Th2 immunopathology in respiratory syncytial virus infection

Elevated IL-4Rα expression on neonatal CD11b⁺ mDCs promotes Th2 biased responses to RSV infection.

Abstract

Respiratory syncytial virus (RSV) is one of the leading causes of bronchiolitis in children, and severe RSV infection early in life has been associated with asthma development. Using a neonatal mouse model, we have shown that down-regulation of IL-4 receptor α (IL-4Rα) with antisense oligonucleotides in the lung during neonatal infection protected from RSV immunopathophysiology. Significant down-regulation of IL-4Rα was observed on pulmonary CD11b⁺ myeloid dendritic cells (mDCs) suggesting a role for IL-4Rα on mDCs in the immunopathogenesis of neonatal RSV infection. Here, we demonstrated that neonatal CD11b⁺ mDCs expressed higher levels of IL-4Rα than their adult counterparts. Because CD11b⁺ mDCs mainly present antigens to CD4⁺ T cells, we hypothesized that increased expression of IL-4Rα on neonatal CD11b⁺ mDCs was responsible for Th2 - biased RSV immunopathophysiology. Indeed, when IL-4Rα was selectively deleted from CD11b⁺ mDCs, the immunopathophysiology typically observed following RSV reinfection was ablated, including Th2 inflammation, airway-mucus hyperproduction, and pulmonary dysfunction. Further, overexpression of IL-4Rα on adult CD11b⁺ DCs and their adoptive transfer into adult mice was able to recapitulate the Th2-biased RSV immunopathology typically observed only in neonates infected with RSV. IL-4Rα levels on CD11c⁺ cells were inversely correlated with maturation status of CD11b⁺ mDCs upon RSV infection. Our data demonstrate that developmentally regulated IL-4Rα expression is critical for the maturity of pulmonary CD11b⁺ mDCs and the Th2-biased immunopathogenesis of neonatal RSV infection.

Introduction

RSV causes acute lower respiratory tract infections in infants, and it is the leading cause of bronchiolitis in children under one year of age [1, 2]. RSV infection severe enough to require hospitalization during infancy is associated with increased risk of wheezing and development of asthma [3, 4]. Currently, no effective vaccines or therapeutics for RSV exist. The development of such products is partially hindered by our lack of knowledge of the infant immune system and its response to RSV infection.

RSV infection in infants induces a Th2-biased immune response, which is partly responsible for RSV pathogenesis [5, 6]. A study comparing cytokine levels in the serum of RSV- vs. influenza-infected infants found higher concentrations of Th2 cytokines (IL-4 and IL-5) in RSV-infected children, suggesting that the Th2-biased response is not simply an age-specific response but also a virus-specific response [7]. Animal studies have reported similar findings [8, 9]. Age at initial RSV infection is important in determining the Th2 bias [10]; infection of neonatal mice and reinfection upon adulthood induces a Th2-biased immune response leading to Th2 cytokines (IL-4, IL-5, and IL-13), airway remodeling, mucus hyperproduction, and airway hyperreactivity [9–12]. Further, polymorphisms in IL-4Rα have been associated with RSV disease severity [13]. We have previously shown that down-regulation of IL-4Rα with antisense oligonucleotide during primary RSV infection in neonatal mice mitigated Th2-biased immunopathophysiologies upon reinfection [12], which raises the intriguing possibility that IL-4Rα has a role in the immunopathogenesis of neonatal RSV infection.

DCs are APCs that initiate and dictate the ensuing adaptive immune response after exposure to a pathogen. DCs are functionally divided into two main classes: mDCs and pDCs. mDCs present Ags to T cells and help T cells to differentiate [14], whereas pDCs produce type I IFNs to combat viral infection [15]. mDCs are further classified into CD11b⁺ mDCs, which mainly present Ags to CD4⁺ T cells [16, 17], and CD103⁺ mDCs, which mainly cross-present Ags to CD8⁺ T cells [16]. Several functional and phenotypic differences have been shown between neonatal and adult DCs [18–20]. Additionally, RSV infection induces significantly less recruitment of pDCs and, thus, type I IFNs in the lung of neonatal vs. adult mice [21]; pDCs from cord blood express lower levels of type I IFNs compared with pDCs from adult peripheral blood in response to RSV stimulation ex vivo [22]. Neonatal mDCs express lower levels of costimulatory molecules (CD80 and CD86) and produce less IL-12 (a Th1 cytokine) in response to LPS compared with their adult counterparts [18, 19].

In the present study, we demonstrated an age-dependent expression of IL-4Rα on pulmonary DCs. Neonatal CD11b⁺ mDCs (CD11c⁺ MHCII⁺ CD11b⁺) and pDCs (CD11c⁺ PDCA-1⁺) expressed higher levels of IL-4Rα compared with their adult counterparts, whereas neonatal CD103⁺ mDCs (CD11c⁺ MHCII⁺ CD103⁺) expressed less IL-4Rα than they did in adults. Because the age at initial infection is important in determining the Th2-biased immune response to RSV infection [9, 10, 23] and CD11b⁺ mDCs are mainly responsible for Ag presentation and activation of CD4⁺ T cells [16, 17], we hypothesized that the elevated expression of IL-4Rα on neonatal mDCs is responsible for the Th2-biased immunopathogenesis of neonatal RSV infection. To test that hypothesis, we infected mice deficient in IL-4Rα CD11b⁺ mDCs [24] and evaluated the Th2-biased immune response to RSV in vitro and in vivo. Furthermore, we adoptively transferred mDCs that overexpressed IL-4Rα into adult mice to confirm the importance of IL-4Rα in the induction of Th2-biased immune responses to RSV. Our data demonstrate that high levels of IL-4Rα, as occurs in infancy, inhibit the maturation of CD11b⁺ mDCs, and, for the first time to our knowledge, our data establish a critical role for IL-4Rα on CD11b⁺ mDCs in initiating the Th2-biased immunopathogenesis of neonatal RSV infection.

MATERIALS AND METHODS

Mice

BALB/c mice were purchased from Envivo (Cambridgeshire, United Kingdom) and were housed in a specific pathogen-free animal facility at the University of Tennessee Health Science Center (Memphis, TN, USA). CD11c-specific IL-4Rα^−/− mice (IL-4Rα^−/−DC) or their littermates (IL-4Rα^−/loxDC) were generated by crossing IL-4Rα^lox/lox mice with CD11c^cre IL-4Rα^−/− mice (gifts from Frank Brombacher, University of Cape Town, Cape Town, South Africa and were generously donated by Debroski Herbert, University of California, San Francisco, CA, USA) [24]. All mice were on BALB/c background. Breeders were time mated, and age-matched pups were used for all experiments. All animal experiments were performed according to Guidelines of Care and Use of Laboratory Animals and approved by University of Tennessee Health Science Center Animal Care and Use Committee.

Experimental design and RSV infection

Neonatal (5 d old) or adult mice (6 wk old) were infected intranasally with human RSV strain A-2 (2 × 10⁵ 50% tissue culture infective dose (TCID₅₀)/g body weight; Advanced Biotechnologies, Eldersburg, MD, USA) in 10 μl (neonates) or 50 μl (adults) of SFM (GE Healthcare Life Sciences, Little Chalfont, United Kingdom) or SFM alone (sham). A subset of the cohort was allowed to mature and was reinfected 4 wk later. At 6 d after reinfection, various endpoints were measured, including lung function and histopathology, BALF cytokines or cellularity, and T cell profile.

BMDCs transduction and transfer

Bone marrow cells were isolated from the tibia and femur of adult mice (5–7 wk old) mice, as previously described [25] and then cultured in DC media (RPMI 1640; GE Healthcare Life Sciences) supplemented with 10% heat-inactivated FBS (Thermo Fisher Scientific, Waltham, MA, USA), 100 U/ml penicillin (GE Healthcare Life Sciences), 100 mg/ml streptomycin (GE Healthcare Life Sciences) human FLT3 200 ng/ml (Tonbo Biosciences, San Diego, CA, USA) for 6 d at a density of 1 × 10⁶/ml in a 6-well plate [26]. BMDCs were then harvested and infected with third-generation lentiviral vectors (pLV-IL-4Rα-eGFP or control pLV-eGFP) (Cyagen Biosciences, Santa Clara, CA, USA) at a multiplicity of infection 20 in Opti-MEM media (Thermo Fisher Scientific) and 5 µg/ml of polybrene (Cyagen Biosciences) for 3 h. These cells were cultured in DC medium for 4 d. Transduced cells were then infected with RSV at a multiplicity of infection 50 overnight, and CD11b⁺mDCs that were positive for eGFP were sorted with a SH800 cell sorter (Sony Biotechnology, San Jose, CA, USA). These cells were adoptively transferred intranasally into the lungs of mice (1 × 10⁵ cells in 50 µl of SFM media). Control mice received 50 µl of SFM media.

RSV gene expression in the lung

RSV gene expression in the lungs was quantified using real-time RT-PCR that correlates well with a plaque assay or TCID₅₀ assay [9, 21, 27]. Lungs were isolated, and total RNA was extracted using the Total RNA extraction kit (Qiagen, Valencia, CA, USA) according to the manufacturer's instructions. cDNA was synthesized using Oligo (dT) and SuperScript III first-strand synthesis system (Thermo Fisher Scientific). Real-time PCR was performed using Platinum SYBR Green qPCR Supermix (Thermo Fisher Scientific) with the following primers: NS1 forward primer (5′-CACAACAATGCCAGTGCTACAA-3′) and NS1 reverse primer (5′-TTAGACCATTAGGTTGAGAGCAATGT-3′).

Cell staining and flow cytometry

To measure the expression of surface markers on pulmonary DCs, single-cell suspensions were prepared from mouse lungs using gentleMACS Octo Dissociator (Miltenyi Biotec, San Diego, CA, USA), as previously reported [21]. Single cells were then stained with fixable viability dyes, Abs to CD11c (N418), CD11b (M1/70; BioLegend, San Diego, CA, USA), CD103 (2E7), PDCA-1 (eBio 927; eBioscience, San Diego, CA, USA), and either IL-4Rα (mIL-4R-M1; BD Biosciences, San Jose, CA, USA) or MHCII (M5/114.15.2), CD80 (16-10A1), and CD86 (GL-1). All Abs and dyes were from eBioscience unless otherwise stated. The staining data were acquired on a FACS Canto II (BD Biosciences) flow cytometer and analyzed with FlowJo software (version 10; Tree Star; Ashland, OR, USA.). Because the MFI of IL-4Rα FMO controls were different among the mice at different ages (i.e., 1 d, 5 d, and 6 wk), the IL-4Rα MFI was normalized by subtracting the MFI of the age-matched FMO controls from the actual values.

To measure the T cell profile in the lung, single cells from lungs were isolated as mentioned and stimulated for 5 h at 37°C in RPMI 1640, containing 10% heat-inactivated FBS (Thermo Fisher Scientific), 100 U/ml penicillin (GE Healthcare), 100 mg/ml streptomycin (GE Healthcare), 5 ng/ml PMA (Sigma-Aldrich; St. Louis, MO, USA), 500 ng/ml ionomycin (Sigma-Aldrich), and protein transport inhibitor (1 µl/10⁶ cells; GolgiPlug, BD Biosciences). Cells were then stained with fixable viability dye, fixed with fixation buffer, permeabilized with permeabilization buffer, and labeled with Abs to CD3 (17A2), CD4 (RM4-5; BioLegend), IFN-γ (XMG1.2), and IL-4 (BVD6-24G2). All Abs, dyes, and buffers were from eBioscience unless otherwise stated. Flow data were then acquired and analyzed as described. Cells were gated on lymphocytes based on forward and side scatter light properties, singlets, and live cells and then on the CD3 and CD4 double-positive population.

BALF cellularity

BALF was isolated with 1 ml of PBS and 0.5% BSA; 20,000 cells from BALF were centrifuged onto slides and stained with a Hema-3 staining kit (Thermo Fisher Scientific). Cells were then differentiated and counted by one unbiased observer using an EVOS microscope (Thermo Fisher Scientific).

Pulmonary function test

Airway resistance was assessed with the FlexiVent FX system (Scireq, Montreal, QC, Canada) to increasing doses of methacholine (Sigma-Aldrich). Mice were anesthetized with ketamine/xylazine (180/10 mg/kg) and mechanically ventilated at a tidal volume of 10 ml/kg and a frequency of 2.5 Hz, using a computer-controlled piston ventilator. Resistance was calculated by recording pressure and volume in the airways using the single-compartment model. All data were normalized to their individual baseline resistance values [(Value − baseline)/Baseline] and plotted as normalized resistance.

Lung histopathology

After euthanasia, hearts of mice were perfused with PBS, and lungs were removed and gravity inflated with zinc formalin (Thermo Fisher Scientific) and then fixed for 24 h. Lung tissues were then embedded in paraffin, sectioned, and stained with H&E or PAS. Stained lung sections were visualized, and images acquired with an EVOS microscope. Inflammation was semiquantitatively scored in a blinded fashion based on the following scale: 1, very mild; 2, mild; 3, moderate; and 4, severe. PAS staining was scored by a quantitative method [28].

Cytokines in the lung

Cytokine levels, including IL-4, IL-12 (p40), and IFN-γ, were measured in the BALF with a Milliplex kit (EMD Millipore, Billerica, MA, USA) and a Luminex system (Luminex, Austin, TX, USA), per the manufacturer’s instruction. Each sample was analyzed in duplicate. IL-13 in the lung homogenates was quantified with a mouse IL-13 Platinum ELISA kit (eBioscience). The detection limits for IL-4, IL-12 (p40), IFN-γ, and IL-13 were 0.4, 4.9, 0.9, and 2.8 pg/ml, respectively.

In vitro T cell activation assay

Pulmonary CD11b⁺ mDCs (CD11c⁺MHCII⁺CD11b⁺) were isolated first with the CD11c Positive Selection Kit (StemCell Technologies, Vancouver, BC, Canada) and then with an SH800 cell sorter (Sony Biotechnology) from lungs of neonatal mice at 6 d after infection with RSV. CD4⁺ T cells were isolated from the spleen of naïve adult BALB/c mice with a CD4 positive selection kit (StemCell Technologies) and stained with 0.5 μM CFSE (Thermo Fisher Scientific). CFSE-stained CD4⁺ cells (1 × 10⁵) were then cocultured with CD11b⁺ mDCs (2 × 10⁴) in the presence of 1 μg/ml CD3 Ab (17A2; eBioscience) for 72 h. Cells were stimulated, stained, and analyzed by flow cytometry to determine the Th subpopulations. The purity of isolated CD11b⁺ mDCs or CD4⁺ cells was >90 percent.

Statistical analyses

Data were plotted as means ± sem and analyzed using GraphPad Prism 6 (GraphPad, La Jolla, CA, USA). Student’s t test or 2-way ANOVA with Bonferroni post hoc tests were used to compare the means among groups, where appropriate. Differences were considered statistically significant at P < 0.05.

RESULTS

Expression of IL-4Rα on DCs was age dependent

In our previous study [12], we observed that the most significant down-regulation of IL-4Rα after antisense oligonucleotide treatment occurred in pulmonary DCs, and that down-regulation correlated with decreased Th2-biased immunopathologies during RSV reinfection, suggesting a role for IL-4Rα on DCs in RSV immunopathogenesis [12]. To explore that possibility, we first quantified IL-4Rα expression on various types of pulmonary DCs from mice at different ages (gating strategy in Supplemental Fig. 1). Specifically, we measured expression of IL-4Rα on pulmonary CD11b⁺ mDCs (CD11c⁺MHCII⁺CD11b⁺), CD103⁺ mDCs (CD11c⁺MHCII⁺CD103⁺), and pDCs (CD11c⁺PDCA-1⁺) from neonatal (1 or 5 d old) or adult mice (6 wk old) via flow cytometry (Fig. 1A). The expression of IL-4Rα on CD11b⁺ mDCs declined as age increased, with 1-d-old pups expressing the greatest amount (Fig. 1B). Interestingly, IL-4Rα expression on CD103⁺ mDCs increased with age, with adults expressing the greatest amount (Fig. 1C). Similar to CD11b⁺ mDCs, pDCs down-regulated IL-4Rα expression as age increased (Fig. 1D). These data suggest that the expression of IL-4Rα on pulmonary DCs is developmentally regulated and cell specific.

Figure 1. Expression of IL-4Rα on DCs was age dependent. Lung DCs from neonatal (1 or 5 d old) or adult (6 wk old) mice were analyzed by flow cytometry for the surface expression of IL-4Rα.

(A) Flow cytometric histogram graphs show a representative example of IL-4Rα expression on DC subsets. (B) IL-4Rα MFI on CD11b⁺ mDCs. (C) IL-4Rα MFI on CD103⁺ mDCs. (D) IL-4Rα MFI on pDCs. Shaded histograms represent FMO controls. Data are representative of 3 independent experiments with 4–5 mice/group. *P < 0.05.

Deletion of IL-4Rα on CD11c⁺ cells attenuated Th2-biased immune responses upon RSV reinfection

Having confirmed that neonatal CD11b⁺ mDCs express elevated levels of IL-4Rα, we further examined the role of IL-4Rα on CD11b⁺ mDCs in polarizing the Th2-biased immune response to RSV. We used a mouse model in which IL-4Rα is specifically deleted on CD11c⁺ cells (IL-4Rα^−/−DC) by crossing IL-4Rα^lox/lox mice with CD11c^Cre IL-4Rα^−/− mice [24]. In IL-4Rα^−/−DC mice, the expression of IL-4Rα is decreased on CD11b⁺ mDCs, CD103⁺ mDCs, and alveolar macrophages but not on T cells (Supplemental Fig. 2). The littermate controls (IL-4Rα^−/loxDC) have 1 copy of intact Il-4Rα. IL-4Rα^−/−DC and IL-4Rα^−/loxDC neonatal mice were infected with RSV (IL-4Rα^−/−DCRR and IL-4Rα^−/loxDCRR) or medium (IL-4Rα^−/−DCsham or IL-4Rα^−/loxDCsham) at 5 d of age and reinfected with RSV 4 wk later. At 6 d after reinfection, we analyzed the CD4⁺ T cell responses from the lungs of those mice. As expected, the IL-4Rα^−/loxDCRR mice that had one copy of intact Il-4Ra mounted a Th2-biased immune response upon RSV reinfection, although the magnitude of this Th2 bias was smaller than in BALB/c mice, as we previously published [12]. Importantly, we observed a significant decrease in the percentage of CD4⁺ IL-4⁺ T cells in the IL-4Rα^−/−DCRR mice compared with the IL-4Rα^−/loxDCRR mice (Fig. 2A). There was also reduction in CD4⁺ IFN-γ⁺ IL-4⁺ T cells in IL-4Rα^−/−DC RR mice vs. IL-4Rα ^−/loxDCRR mice (Fig. 2A). This reduction in Th2 cells was accompanied by a decrease in IL-13 (Table 1) in lung homogenates after RSV reinfection; in fact, IL-13 levels in lung homogenates were similar to uninfected groups (IL-4Rα^−/−DCsham or IL-4Rα^−/loxDCsham). IL-4 was very low in all groups and below the limit of detection in the uninfected groups. Although no difference was observed in the percentage of CD4⁺ IFN-γ⁺ T cells between the RSV-infected groups, we did observe an elevation in IL-12p40 levels in the BALF of IL-4Rα^−/−DCRR mice vs. IL-4Rα^−/loxDCRR (Table 1). These results were not due to a difference in relative viral gene expression of RSV (Fig. 2B) or baseline numbers of DCs during initial infection (Supplemental Fig. 3). These data indicate that IL-4Rα on CD11b⁺ mDCs has a role in the Th2-biased immune response to neonatal RSV infection in vivo.

Figure 2. Deletion of IL-4Rα on CD11c⁺ cells attenuated Th2-biased immune responses upon RSV reinfection.

IL-4Rα^−/−DC or IL-4Rα^−/loxDC mice were infected at 5 d old and reinfected with RSV 4 wk later (IL-4Rα^−/−DCRR and IL-4Rα^−/loxDCRR, respectively). Control mice were sham infected (IL-4Rα^−/−DCsham and IL-4Rα^−/loxDCsham). Pulmonary Th profile was measured via flow cytometry at 6 d after reinfection, and virus gene expression was calculated via real-time RT-PCR at 4 d after primary infection. (A) Percentage (upper panel) and numbers (middle panel) of pulmonary CD4⁺ T cells expressing IFN-γ, IL-4, or both IFN-γ and IL-4. The bottom panel displays representative flow plots. (B) Relative expression of the RSV NS1 gene in the lung at 4 d after primary infection. Data are representative of 3 independent experiments with 4–5 mice/group. *P < 0.05.

TABLE 1.

Deletion of IL-4Rα on CD11c⁺ cells attenuated Th2-biased cytokine responses in the lung upon RSV reinfection

Cytokines	IL-4Rα−/loxDCsham	IL-4Rα−/−DCsham	IL-4Rα−/loxDCRR	IL-4Rα−/−DCRR
BALF (pg/ml)
IFN-γ	B.D.	B.D.	51.9 ± 15.9	88.9 ± 14.4
IL-4	B.D.	B.D.	1.1 ± 0.2	1.2 ± 0.2
IL-12p40	B.D.	B.D.	9.2 ± 3.4	21.0 ± 4.3*
Lung homogenates (pg/ml)
IL-13	21.0 ± 2.1	19.3 ± 4.5	27.7 ± 1.6†	18.9 ± 2.1*

IL-4Rα^−/loxDC or IL-4Rα^−/−DC mice were infected at 5 d old and reinfected with RSV 4 wk later (IL-4Rα^−/loxDCRR or IL-4Rα^−/−DCRR, respectively). Control mice received serum-free viral growth medium (IL-4Rα^−/loxDCSham or IL-4Rα^−/−DCSham). Cytokines were measured in BALF (IFN-γ, IL-4, and IL-12p40) or lung homogenates (IL-13) at 6 d after reinfection and are reported in picograms per milliliter. n = 4–5 mice/group. B.D., below the detection limit.

^*P < 0.05 vs. IL-4Rα^−/loxDCRR.

^†P < 0.05 vs. IL-4Rα^−/loxDCSham or IL-4Rα^−/−DCSham.

Deletion of IL-4Rα on CD11c⁺ cells protected mice from immunopathophysiology upon RSV reinfection

In addition to the pulmonary CD4⁺ T cell responses, we also examined pathophysiologic endpoints, including BALF cellularity and pulmonary function and pathology in IL-4Rα^−/−DCRR and IL-4Rα^−/loxDCRR mice. We observed substantially reduced inflammatory responses in the airways of IL-4Rα^−/−DCRR mice, as evidenced by reduced total cells, macrophages, and eosinophils in the BALF compared with IL-4Rα^−/loxDCRR mice (Fig. 3A). As shown in Fig. 3B, RSV-induced airway hyperreactivity in IL-4Rα^−/loxDCRR mice, whereas airway resistance in IL-4Rα^−/−DCRR mice was no different than that of the sham controls. No significant differences in resistance were detected between groups of sham-infected mice. Peribronchiolar and perivascular inflammation was significantly reduced in IL-4Rα^−/−DCRR vs. IL-4Rα^−/loxDCRR mice (Fig. 3C). Strikingly, there was little-to-no mucus production in the airways of IL-4Rα^−/−DCRR, whereas significant amounts of airway mucus were obvious in IL-4Rα^−/loxDCRR mice (Fig. 3D). That observation is consistent with the fact that airway resistance in IL-4Rα^−/−DCRR mice was no different from that of sham-infected mice. The decrease in airway inflammation, airway hyperreactivity, and mucus hyperproduction in IL-4Rα^−/−DCRR mice suggests that elevated IL-4Rα expression on CD11b⁺ mDCs is partially responsible for the immunopathophysiology upon RSV reinfection.

Figure 3. Deletion of IL-4Rα on CD11c⁺ cells protected mice from immunopathophysiology upon RSV reinfection.

IL-4Rα^−/−DC or IL-4Rα^−/loxDC mice were infected and reinfected with RSV, as previously described. BALF cellularity, airways resistance in response to methacholine (MeCh) challenge, and histology were performed at 6 d after reinfection. (A) BALF cellularity. Macs, macrophages; Lymph, lymphocytes; Neutro, neutrophils; Eos, eosinophils. (B) Airway resistance. (C) H&E-stained lung sections and inflammation score. Arrowheads point to cellular infiltrates. (D) PAS-stained lung sections and PAS score. Arrowheads point to mucus. Scale bar, 400 μm in the upper panels. The bottom panels are images taken under higher magnification of the selected area in the upper panel (inset box). Scale bar, 100 μm in the bottom panels. Data are representative of 2 independent experiments with 4–5 mice/group. *P < 0.05.

Absence of IL-4Rα on CD11b⁺ mDCs dampened Th2 responses to RSV infection in vitro

Although the expression of IL-4Rα decreased in neonatal CD11b⁺ mDCs in the IL-4Rα^−/−DC mice, it was also decreased in CD103⁺ mDCs and alveolar macrophages (Supplemental Fig. 2). To examine the role of IL-4Rα specifically on CD11b⁺ mDCs to induce a Th2-biased immune response to RSV, we performed in vitro T cell–activation assays. Pulmonary CD11b⁺ mDCs were isolated from RSV-infected neonatal IL-4Rα^−/−DC or IL-4Rα^−/loxDC mice. Those cells were subsequently cocultured for 3 d with splenic CD4⁺ T cells isolated from naïve adult BALB/c mice, and T cell phenotypes were then measured by flow cytometry. As expected, neonatal CD11b⁺ mDCs from IL-4Rα^−/loxDC mice induced a substantial amount of Th2 cells (CD4⁺IL-4⁺), and mimicking our in vivo data, CD11b⁺ mDCs from IL-4Rα^−/−DC mice induced significantly less Th2 cells and a decreased ratio of Th2/Th1 (CD4⁺IL-4⁺/ CD4⁺IFN-γ⁺) cells (Fig. 4). These data suggest that IL-4Rα expression on CD11b⁺ mDCs promotes Th2-biased immune responses to RSV.

Figure 4. Absence of IL-4Rα on CD11b⁺ mDCs dampened Th2 responses to RSV infection in vitro.

Neonatal IL-4Rα^−/−DC and IL-4Rα^−/loxDC were infected with RSV. CD11b⁺ mDCs were isolated from those mice at 6 d after infection and cultured with CD4⁺ T cells purified from the spleens of naïve, adult BALB/c mice. Th profile was measured after 72 h coculture using flow cytometry. (A) Percentage of pulmonary CD4⁺ T cells expressing IFN-γ. (B) Percentage of pulmonary CD4⁺ T cells expressing IL-4. (C) Ratio of pulmonary CD4⁺ T cells expressing IL-4 and IFN-γ. Data are representative of 2 independent experiments with 3 samples/group. *P < 0.05.

Loss of IL-4Rα expression on CD11b⁺ mDCs enhanced their maturation response to RSV

To delineate the mechanism by which IL-4Rα on CD11b⁺ mDCs induced Th2-biased responses to RSV in neonates, we measured the expression of maturation markers (CD80 and CD86) on CD11b⁺ mDCs. Wild-type neonates and adults or IL-4Rα^−/−DC neonates (IL-4Rα^−/−DCNR) were infected with RSV, and the expression of CD80 and CD86 was measured on pulmonary CD11b⁺ mDCs at 14 d after infection by flow cytometry (Fig. 5). Consistent with previous studies [18, 19], neonatal CD11b⁺ mDCs were less mature (i.e., reduced expression of CD80 and CD86) compared with their adult counterparts during RSV infection. In the absence of IL-4Rα, the expression of CD80 and CD86 was significantly enhanced on CD11b⁺ mDCs (IL-4Rα^−/−DCNR compared with neonatal mice infected with RSV). These data suggest that elevated expression of IL-4Rα on neonatal CD11b⁺ mDCs suppresses the ability of those cells to mature in response to RSV infection, resulting in a Th2-biased, adaptive immune response.

Figure 5. Loss of IL-4Rα on CD11b⁺ mDCs enhanced their maturation response to RSV.

Neonatal or adult wild-type mice and neonatal IL-4Rα^−/−DC mice were infected with RSV. Maturation markers (CD80 and CD86) and MHC-II were measured on lung CD11b⁺ mDCs via flow cytometry. (A) MFI of CD80 on CD11b⁺ mDCs. (B) MFI of CD86 on CD11b⁺ mDCs. (C) MFI of MHCII on CD11b⁺ mDCs. Data are representative of 3 independent experiments with 4–5 mice/group. *P < 0.05.

Overexpression of IL-4Rα on adult CD11b⁺ mDCs exacerbated Th2 responses to RSV infection in adult mice

Expression of IL-4Rα on CD11b⁺ mDCs declines significantly with age. To further confirm the pathogenic role of IL-4Rα, we overexpressed IL-4Rα on adult mDCs. Specifically, we transduced adult BMDCs with IL-4Rα (pLV-IL-4Rα-eGFP) or eGFP (pLV-eGFP) expressing lentivirus. Successful transduction was verified by flow cytometry (Supplemental Fig. 4). CD11b⁺ mDCs that were positive for eGFP were then sorted and adoptively transferred to the lungs of naïve adult mice; 1 wk later, those mice were infected with RSV. At 6 d after infection, we analyzed pulmonary CD4⁺ T cell responses. Overexpression of IL-4Rα on adult CD11b⁺ mDCs (pLV-IL-4Rα-eGFP) resulted in a lower percentage but no significant change in the number of CD4⁺IFN-γ⁺ T cells (Fig. 6A). A higher percentage and number of CD4⁺IL-4⁺ T cells were observed when IL-4Rα was overexpressed on adult CD11b⁺ mDCs compared with the control groups (Fig. 6A and B). There was no change in the percentage of CD4⁺IFN-γ⁺IL-4⁺ cells with overexpression of IL-4Rα on adult CD11b⁺ mDCs, but there was an increase in the number of those cells (Fig. 6B).

Figure 6. Overexpression of IL-4Rα on adult CD11b⁺ mDCs exacerbated Th2 responses to RSV infection in adult mice.

BMDCs were isolated from adult mice, transduced with pLV-IL-4Rα-eGFP or pLV-eGFP control, and infected with RSV. CD11b⁺ mDCs that were positive for eGFP were sorted and adoptively transferred into the lungs of adult mice. Those mice were then infected with RSV, and 6 d later, pulmonary Th subsets were quantified. (A) Percentage of CD4⁺ T cells expressing IFN-γ or IL-4 or both IFN-γ and IL-4. (B) Number of CD4⁺ T cells expressing IFN- γ, IL-4, or both IFN-γ and IL-4. Data are representative of 2 independent experiments with 3–4 mice/group. *P < 0.05.

DISCUSSION

Infants are particularly vulnerable to severe RSV infection; both human and animal studies implicate a Th2-biased immune response in the pathogenesis of RSV in infants [5, 6, 9–11]. In particular, single-nucleotide polymorphisms in the IL-4Rα gene that increase the receptor response to its cytokines (IL-4 or IL-13) are associated with increased IgE production, hospitalization rates from RSV infection, and susceptibility to asthma [13, 29]. Pharmacologic down-regulation of IL-4Rα in the lung is protective against RSV disease in neonatal mice and is effective in preventing Th2-biased immunopathogenesis [12]. In the current study, we observed increased expression of IL-4Rα on pulmonary CD11b⁺ mDCs in neonatal, compared with adult, mice. Deletion of IL-4Rα on neonatal CD11c⁺ cells (including CD11b⁺ mDCs and other cells) resulted in a significant reduction in Th2 polarization, RSV-associated airway hyperreactivity, inflammation, and eosinophilia after RSV reinfection in vivo. This was accompanied by an increase in IL-12 production. In vitro T cell–activation assays demonstrated the specificity of IL-4Rα on neonatal CD11b⁺ mDCs in the induction of Th2 responses to RSV. Because adult CD11b⁺ mDCs express significantly less IL-4Rα, complimentary experiments in which adult CD11b⁺ mDCs were made to overexpress IL-4Rα were also performed. Overexpression of IL-4Rα on adult CD11b⁺ mDCs with the transfer of those cells into the lungs of adult mice before RSV infection was able to recapitulate the Th2-biased immunopathogenesis observed in neonatal mice. Mechanistically, we demonstrated that pulmonary CD11b⁺ mDCs were less mature in neonates compared with adults in response to RSV infection and that deletion of IL-4Rα on CD11b⁺ mDCs enhanced their maturation, altering Th2/Th1 immune responses to RSV.

The maturation status of DCs is important to their function, and the costimulatory molecules CD80 and CD86 are thought to be essential as a second signal for T cell activation. Here, we demonstrated that the levels of CD80 and CD86 on CD11b⁺ mDCs were associated with the axis of CD4⁺ T cell differentiation (i.e., Th1 vs. Th2). Neonatal CD11b⁺ mDCs expressed less CD80 and CD86 than their adult counterparts did (Fig. 5) and induced fewer Th1 cells in response to RSV infection [9]. Interestingly, CD103⁺ mDCs, which mainly activate CD8⁺ T cells, have also been shown to be immature in neonates compared with adults, and the maturation status of CD103⁺ mDCs influences the epitope hierarchy of CD8⁺ T cells in response to RSV infection [30]. Thus, it is exciting to speculate that maturation status of CD11b⁺ mDCs may also dictate epitope hierarchy of CD4⁺ T cells in response to RSV infection—a concept that we are currently exploring. We further demonstrated that the expression of CD80 and CD86 on CD11b⁺ mDCs was partially regulated by the expression of IL-4Rα because deletion of IL-4Rα on CD11b⁺ mDCs restored some expression of CD80 and CD86 after RSV infection. The mechanism or mechanisms by which IL-4Rα is developmentally regulated and how IL-4Rα regulates the expression of CD80 and CD86 are currently unclear and warrant further investigation.

The ratio of Th2/Th1 cells decreased by 50% in the IL-4Rα^−/−DC mice compared with IL-4Rα^−/loxDC mice both in vivo and in vitro (Fig. 2 and 4). Although the reduction was not massive, it was significant. The reduction in Th2/Th1 ratio was accompanied by improved lung function and significantly decreased lung pathologies. We observed this phenomenon in our previous publication [12]. A subtle reduction in IL-4Rα on Th cells and DCs significantly improved lung pathology and abolished lung dysfunction after RSV reinfection. Our data indicate that the pathophysiology is determined by the balance of Th1/Th2 immune responses and that a small shift from the Th2-biased immune response is all that is necessary to improve lung pathophysiology.

As with our previous studies [9, 12], we observed a CD4⁺ T cell population expressing both IFN-γ⁺ and IL-4⁺ after RSV reinfection. Initially, those cells were thought to be pre-Th1/Th2 cells that were in the intermediate stage during the commitment process of Th1/Th2 cells [31, 32]. However, recent data demonstrate that this phenotype is acquired after full differentiation and that these cells support both type 1 and type 2 inflammatory responses [33]. IL-4Rα deletion on CD11b⁺ mDCs decreased the percentage of those cells in the lungs of mice after RSV reinfection. The exact role of those cells in RSV pathogenesis remains elusive, but the reduction of those cells has consistently coincided with reduced pulmonary immunopathophysiology upon RSV reinfection here and in our previously published work [9, 12].

Although deletion of IL-4Rα on CD11b⁺ mDCs did not alter the percentage of pulmonary Th1 (CD4⁺ IFN-γ⁺) cells after RSV reinfection, there was an increase in the level of IL-12p40 (a Th1 cytokine) in the BALF (Table 1). The elevated IL-12 (p40) level may result from the decreased IL-13 because IL-13 has been previously shown to inhibit IL-12 production via IL-4Rα signaling [34].

A limitation of our in vivo studies was that IL-4Rα was not specifically deleted on CD11b⁺ mDCs. In fact, IL-4Rα was deleted on CD11c^+/hi cells, including AMs, CD11b⁺ mDCs, and CD103⁺ mDCs, and on CD11c^+/lo cells, including pDCs, NK cells, B cells, and CD8⁺ T cells [35]. However, our in vitro data (T cell activation assays with neonatal CD11b⁺ mDCs), our in vivo data (adoptive transfer of adult CD11b⁺ mDCs that overexpress IL-4Rα), and data from the literature (summarized below) argue for the specific role of IL-4Rα on CD11b⁺ mDCs in RSV immunopathogenesis. IL-4Rα was deleted on CD103⁺ mDCs in our model; however, we speculated that the deletion of IL-4Rα on CD103⁺ mDCs most likely has minimal effect on the Th2-biased response to RSV infection for the following reasons. CD103⁺ mDCs mainly activate CD8⁺, not CD4⁺, T cells [16, 17]. Additionally, neonatal CD103⁺ mDCs express significantly less IL-4Rα compared with their adult counterparts, whereas the Th2 bias is only observed in neonates. In the Cryptococcus-infection model, deletion of IL-4Rα specifically on AMs (LysM^CreIL-4Rα^−/lox) was unable to alter the Th2 responses [36], suggesting that IL-4Rα expression on AMs has no effect on Th2 polarization. As to the cells expressing low levels of CD11c, we and others [24] have observed an inefficient Cre function and, thus, inefficient deletion of IL-4Rα in those cells, including B [24] and T cells (Supplemental Fig. 2).

In summary, we have demonstrated a novel, age-dependent mechanism (IL-4Rα on CD11b⁺ mDCs) responsible for the Th2-biased immunopathogenesis during neonatal RSV infection. We have shown that IL-4Rα expression on CD11b⁺ mDCs is developmentally regulated, and greater expression of IL-4Rα is associated with a less-mature phenotype of the CD11b⁺ mDCs in response to RSV infection, resulting in Th2-biased immune responses. The deletion of IL-4Rα on CD11b⁺ mDCs significantly decreased the Th2-biased immunopathogenesis during RSV infection by allowing enhanced maturation of CD11b⁺ mDCs. Our data suggest that modulating IL-4Rα signaling on CD11b⁺ mDCs may be a strategy for developing RSV therapeutics and that vaccine strategies should investigate such signaling in determining efficacy.

AUTHORSHIP

S.A.C. was the principal investigator and designed the study, analyzed the data, and wrote the manuscript. B.S. designed and performed most of the experiments, analyzed the data, and wrote the manuscript draft. D.Y. designed the study, analyzed the data, assisted in the Flexivent experiment, and wrote the manuscript. J.S. assisted in the in vitro T cell–activation assay and the FlexiVent experiment. D.T.S. performed the IL-13 ELISA experiment. S.J., G.I.L. and A.A.S. helped in flow cytometry experiments, J.N.H. assisted in lentiviral transduction experiments.

Glossary

AM

alveolar macrophages

BALF

bronchoalveolar lavage fluid

BMDC

bone marrow–derived dendritic cell

DC

dendritic cells

IL-4Rα^−/−DC

mice in which IL-4Rα is specifically deleted on CD11c⁺ cells

FMO

fluorescence minus one

IL-4Rα^−/loxDC

littermate control mice in which 1 copy of IL-4Rα is present on CD11c⁺ cells

mDC

myeloid dendritic cell

MFI

mean fluorescence intensity

PAS

periodic acid–Schiff

pDC

plasmacytoid dendritic cell

RR

mice infected with RSV as neonates and reinfected as adults

RSV

respiratory syncytial virus

SFM

serum-free medium

sham

mice treated with medium

DISCLOSURES

The authors have no conflicts of interest.