First-Breath-Induced Type 2 Pathways Shape the Lung Immune Environment

Abstract

From birth onward, the lungs are exposed to the external environment and therefore harbor a complex immunological milieu to protect this organ from damage and infection. We investigated the homeostatic role of the epithelium-derived alarmin interleukin-33 (IL-33) in newborn mice and discovered the immediate upregulation of IL-33 from the first day of life, closely followed by a wave of IL-13-producing type 2 innate lymphoid cells (ILC2s), which coincided with the appearance of alveolar macrophages (AMs) and their early polarization to an IL-13-dependent anti-inflammatory M2 phenotype. ILC2s contributed to lung quiescence in homeostasis by polarizing tissue resident AMs and induced an M2 phenotype in transplanted macrophage progenitors. ILC2s continued to maintain the M2 AM phenotype during adult life at the cost of a delayed response to Streptococcus pneumoniae infection in mice. These data highlight the homeostatic role of ILC2s in setting the activation threshold in the lung and underline their implications in anti-bacterial defenses.

Keywords: S. pneumoniae; alarmin; alveolar macrophage; first breath; immune homeostasis; lung; newborn; pneumoniae.

Graphical abstract

Figure 1

Type 2 Alveolar Epithelial Cells Induce IL-33 at Birth (A) Whole-lung IL-33 quantification by ELISA at E19; postnatal days 1, 3, 5, 7, and 14 (P1–P14); and 3 and 4 weeks (3–4w) after birth. (B) qRT-PCR of pulmonary Il33 expression in WT mice at E19 and P1. (C) FACS analysis of viable citrine⁺ cells from Il33^Cit/+ mice at E19 exposed to vacuum or atmospheric pressure (control) for 6 hr. (D) Quantification of (C). (E) Whole-lung IL-33 quantification by ELISA of WT lungs at E19 exposed to vacuum or atmospheric pressure (control) for 6 hr. (F) FACS analysis of lung CD45 and citrine expression in Il33^Cit/+ reporter mice at the indicated time points (gates are set using WT as controls). (G) Percentage of Cit⁺CD45⁻ cells among lung cells, gated as in (F). (H) Flow cytometry of viable CD45⁻ lung cells from Il33^Cit/+ mice at P7, stained for EpCam and CD31. (I) Quantification of the Cit⁺ proportion of EpCAM⁺ cells between E19 and 8 weeks of age. (J) Micrographs of lung sections at E19, P1, and P3 from Il33^Cit/+ reporter mice. Red, surfactant protein C (SP-C); green, IL-33-driven citrine. Scale bars represent 75 μm. Data are representative of two independent experiments with three to five mice per time point, and graph bars represent mean ± SEM. ^∗∗p < 0.01 and ^∗∗∗∗p < 0.0001.

Figure 2

IL-33 Drives a Type 2 Immune Environment in Lungs of Newborns (A) Heatmap representation of cytokine levels in whole lung homogenates comparing WT and Il33^Cit/Cit mice at P7. Original values (see Figure S2A) were rescaled between zero and the maximum value detected for each cytokine and are presented as the fraction of maximum secretion. (B) Percentage of lung ILC2s (Lin⁻ ST2⁺ Thy1.2⁺ CD25⁺ ICOS⁺) analyzed by FACS at the indicated time points. (C) FACS analysis of lung ILC2s (Lin⁻ ST2⁺) in WT and Il33^Cit/Cit mice at P7, further gated for CD25⁺ and ICOS⁺ and quantified (right). (D) FACS analysis of lung ILC2s (Lin⁻ Thy1.2⁺) in WT and Il1rl1^−/− mice at P7, further gated for CD25⁺ and ICOS⁺ and quantified (right). (E) Percentage of lung eosinophils (F4/80⁺ CD11b⁺SiglecF⁺CD11c⁻) analyzed by FACS at the indicated time points. (F and G) FACS analysis of lung eosinophils (CD11b⁺SiglecF⁺CD11c⁻) and AMs (CD11b⁻SiglecF⁺CD11c⁺) at P7 in WT and Il33^Cit/Cit mice (F) and WT and Il1rl1^−/− mice (G). (H) Lung, blood, bone marrow, and spleen cells were analyzed by FACS for eosinophils (CD11b⁺SiglecF⁺F480⁺CD11c⁻) in P7, P14, P28, and adult (6–8 weeks) Il5^Cer/+ mice. Data are representative of one (A and H) or two (B–G) independent experiments with four mice per group. Graph bars represent mean ± SEM. ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001. For flow cytometry, all cells were pre-gated on viable, single, CD45⁺. E, embryonic; p, postnatal; w, week.

Figure 3

Lung ILC2s Expansion and Activation Coincides with AM Differentiation and M2 Polarization (A) Representative FACS profiles of expanding lung ILC2s (Lin⁻ ST2⁺) (top) and proportion of Tom⁺ cells (bottom) in Il13^Tom/+ mice between E19 and P10. (B) Quantification of absolute numbers of lung Lin⁻ ST2⁺ Thy1.2⁺ Tom^+/− cells at the indicated time points. (C) FACS plots illustrating percentages of AMs (F4/80⁺CD11b⁻CD11c⁺) at the indicated time points. (D) Absolute numbers of AMs gated as in (C) between E19 and P14. (E) AMs (F4/80⁺CD11b⁻CD11c⁺SiglecF⁺) were sorted on P7 from WT and Il13^Tom/Tom mice and M2 markers were assessed by RT-PCR. (F) AMs from WT and Il13^Tom/Tom (IL-13 deficient) mice on P7 were isolated as in (E) and cultured for 6 hr, and Cxcl1 and Tnf gene induction was assessed by RT-PCR. Values were normalized to Hprt and are expressed as fold change versus WT. (G) AMs from WT and Il13^Tom/Tom (IL-13 deficient) mice on P3, P7, P14, and P21 were isolated as in (E) and cultured for 6 hr, and spontaneous CXCL1 secretion was quantified by ELISA. Data are representative of three (A–D) or two (E–G) independent experiments with three or four mice per group. Values were normalized to Hprt and are expressed as fold change versus the indicated control. Bars represent mean ± SEM; ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001.

Figure 4

AMs from IL-13-Deficient Mice Present a Pro-inflammatory Phenotype and Improved Defenses against S. pneumoniae (A) AMs from adult WT and Il13^−/− mice were isolated by bronchoalveolar lavage and analyzed for expression of M2 polarization markers by RT-PCR. (B and C) AMs isolated as in (A) were in vitro stimulated with S. pneumoniae (MOI 100). The induction of Cxcl1 was quantified by RT-PCR (B), and supernatant protein levels were determined by ELISA (C). (D) CD45.1 WT monocytes were intra-tracheally transferred to WT and Il13^−/− CD45.2 recipients, and bronchoalveolar cells were harvested by lavage 2 weeks later. FACS-sorted recipient AMs and monocyte-derived AMs were analyzed for expression of M2 polarization markers by RT-PCR. (E–G) WT and Il13^−/− mice were i.n. infected with S. pneumoniae and sacrificed after 6 hr. PMN numbers in BALF were assessed on cytospins (E) and in lungs by FACS analysis (CD45⁺SSC^hiFSC^hiCD11b⁺ Ly6G⁺) (F). Lung CXCL1 was quantified by ELISA (G). (H–K) WT and Il13^−/− mice were i.n. infected with S. pneumoniae and sacrificed after 48 hr. CFU counts in lung homogenates (H) and blood (I). Lung CXCL1 was quantified by ELISA (J). H&E-stained lung sections were scored by a pathologist (see Experimental Procedures) (K, left). Representative H&E lung sections (K, right). Scale bars represent 180 μm. Data are representative of at least three independent experiments with four (A–C) and seven or eight (E–K) mice per group. Data in (D) are from a single experiment with six mice per group. PCR values were normalized to Hprt and expressed as fold change versus indicated control. Mean ± SEM are depicted; ^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001. BAL, bronchoalveolar lavage; CFU, colony-forming units; p.i., post-infection; PMN, polymorphonuclear cells.

Figure 5

Intranasal rmIL-13 Treatment Reversed the Inflammatory Phenotype of AMs in Il13^−/− Mice and the Responses to S. pneumoniae (A–C) WT and Il13^−/− mice were treated daily with rmIL-13 (6 ng in 50 μL NaCl) i.n., and AMs were isolated by bronchoalveolar lavage on day 3. M2 markers were assessed by RT-PCR (A). Cultured AMs were stimulated with S. pneumoniae (MOI 100), and fold induction of Cxcl1 was measured by RT-PCR (B). AMs were treated as in (B), and CXCL1 protein was quantified by ELISA (C). (D and E) WT and Il13^−/− mice were treated with mrIL-13 as in (A)–(C), infected i.n. with S. pneumoniae on day 3, and sacrificed after 48 hr. CFU counts in lung homogenates (D) and blood (E) are shown. Data in (A)–(C) are representative of two independent experiments with four mice per group. Data in (D) and (E) are from a single experiment with eight mice per group. Mean ± SEM are depicted; ^∗p < 0.05 and ^∗∗∗∗p < 0.0001. BALF, bronchoalveolar lavage fluid; CFU, colony forming units; i.n., intranasal; p.i., post-infection.

Figure 6

ILC2 Are the Only Cells Producing IL-13 in the Lung at Homeostasis (A) IL-13 expression in ILC2s (Lin⁻ ST2⁺ ICOS⁺ Thy1.2⁺ CD25⁺) assessed by flow cytometry in adult, naive Il13^Tom/+ mice. Representative plots and percentage of tdTomato⁺ ILC2s are shown. (B and C) ILC2s and IL-13⁺ ILC2s were quantified by flow cytometry and intracellular staining for IL-13 in naive WT lungs. (B) Absolute numbers of total lung ILC2s (Lin⁻ ST2⁺ ICOS⁺ Thy1.2⁺ CD25⁺) and IL-13⁺ ILC2s. (C) Representative plots of IL-13⁺ ILC2s gated as in (A) and percentage of IL-13-producing ILC2s (right). (D) Lung cell populations were tested by FACS for IL-13 expression in healthy adult Il13^Tom/+ mice. Gating strategies are shown in Table S2. (E) IL-13 production by lung ILC2s in WT, Il1rl1^−/−, and Il13^−/− assessed by intracellular staining using flow cytometry (Iso, isotype control); representative plots and absolute numbers are depicted. (F and G) ILC2s were first expanded in lungs of Il13^Tom/+ mice via i.n. administration of rmIL-33 (0.5 μg/50 μL for 5 days), and then sorted Tom⁺ ILC2s were transferred intravenous to WT and Il13^−/− mice. (F) Representative FACS plots showing the homing of Il13^Tom/+ ILC2s in lungs 5 days after adoptive transfer. (G) AMs were isolated by bronchoalveolar lavage from WT and Il13^−/− recipients 5 days after adoptive transfer and in vitro stimulated with S. pneumoniae (MOI 100), and CXCL1 release was assessed by ELISA in supernatants. Data are representative of three (A–C), two (E–G), and one (D) independent experiments with four mice per group. Mean ± SEM are depicted; ^∗∗∗∗p < 0.0001. BAL, bronchoalveolar lavage; i.c., intracellular; i.v., intravenous; FSC, forward scatter.

Figure 7

Lung Resident ILC2s Polarize Tissue Resident AMs toward an M2 Phenotype and Dampen Early Inflammatory Responses against Bacteria (A) Flow cytometry plots of ILC2s in naive Il7r^CreRora^sg/fl mice and Il7r^CreRora^+/fl controls. (B) AMs isolated by flow cytometry from healthy Il7r^CreRora^sg/fl mice and controls at P7 and M2 markers evaluated by RT-PCR. (C and D) AMs isolated by bronchoalveolar lavage from healthy adult Il7r^CreRora^sg/fl mice and controls. (C) M2 markers evaluated by RT-PCR. (D) Primary AMs stimulated for 1 hr with S. pneumoniae (MOI 100) to assess the induction of Cxcl1 and Tnf. (E–J) Il7r^CreRora^sg/fl mice and controls were i.n. infected with S. pneumoniae (10⁵ CFUs) and sacrificed after 6 hr (E and F) or 48 hr (G–J). (E) PMN influx on BALF cytospins. (F) CXCL1 induction in whole-lung homogenate. (G and H) CFUs in lung (G) and in blood (H). (I) CXCL1 induction in whole-lung homogenate. (J) H&E-stained lung sections were scored by a pathologist (see Experimental Procedures) (J, left). Representative H&E lung sections (J, right). Scale bars represent 180 μm. (K–N) CD45.2 recipients were lethally irradiated and transplanted with WT or Rora^sg/sg bone marrow and sacrificed 8 months later. (K) Experimental setup. (L) Representative FACS plots of ILC2s in healthy WT/WT and WT/Rora^sg/sg bone marrow chimeras. (M) AMs isolated via bronchoalveolar lavage and assessed for M2 markers by RT-PCR or (N) stimulated with S. pneumoniae (MOI 100) to evaluate Cxcl1 and Tnf induction. (O and P) WT/WT and WT/Rora^sg/sg chimeras were infected with 10⁵ CFUs S. pneumoniae and sacrificed after 6 hr to assess (O) PMN influx and (P) Cxcl1 induction in lung tissue. Data are representative of at least two independent experiments with four (A–D, M, and N) and seven or eight (E–J, O, and P) mice per group. Data in (G) and (H) are pooled from two independent experiments. Mean ± SEM are depicted; ^∗p ≤ 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001. BALF, bronchoalveolar lavage fluid; CFU, colony-forming units; p.i., post-infection; PMN, polymorphonuclear cells.