Airway epithelial TSLP production of TLR2 drives type 2 immunity in allergic airway inflammation

Abstract

Epithelial cells (ECs)-derived cytokines are induced by different stimuli through pattern recognition receptors (PRRs) to mount a type-2-cell-mediated immune response; however, the underlying mechanisms are poorly characterized. Here, we demonstrated asthmatic features in both primary bronchial epithelial cells (pBECs) and mouse model using several allergens including ovalbumin (OVA), house dust mite (HDM), or Alternaria alternata. We found that toll-like receptor 2 (TLR2) was highly induced in ECs but not dendritic cells (DCs) by various allergens, leading to recruitment of circulating basophils into the lung via C-C chemokine ligand-2 (CCL2). TLR2 expression increased thymic stromal lymphopoietin (TSLP) production through the NF-κB and JNK signaling pathways to extend the survival of recruited basophils and resident DCs in the lung, predisposing a type-2-cell-mediated airway inflammation. Conversely, TLR2 deficiency impaired secretion of TSLP and CCL2, decreased infiltration of lung basophils, and increased resistance to Th2 response. Blocking TSLP also phenocopied these phenomena. Our findings reveal a pro-inflammatory role of airway ECs through a TLR2-dependent TSLP production, which may have implication for treating allergic asthma.

Keywords: TLR2; TSLP; basophils; signaling pathways; type 2 immune responses.

Figure 1

The expression levels of PRRs, cytokines, and chemokines in lung cells following allergen challenge. (A) Expression of PRRs in lung tissues after stimulation of mice with OVA, HDM, and fungi as determined by real‐time‐PCR. (B) Quantification of IL‐25, IL‐33, GM‐CSF, and TSLP mRNA in lung tissues. (C) mRNA quantification of chemokines and their receptors in lung tissues. (D) Expression (top) and gray analysis (bottom) of PRRs in BEAS‐2B cells after stimulation with different allergens was analyzed by western blot. (E) Calculation of cells in BALF. (F) Relative expression of cytokine genes in BEAS‐2B cells with OVA, HDM, and fungi stimulation. (G) Analysis of TSLP mRNA levels in BEAS‐2B cells after TLR2 siRNA treatment. (H) Expression of TLR2 mRNA and TSLP mRNA in BEAS‐2B with Pam3CSK4, OVA, and EF‐OVA. For RT‐PCR and calculation of cells in BALF, data are pooled from three independent experiments with five mice each. For analysis of TLR2 mRNA and TSLP mRNA in vitro, data are pooled from five independent experiments. For PCR Array, data are pooled from three independent experiments and one representative heat map of three independent experiments is shown. One representative western blot of five independent experiments is shown. Values are shown as mean ± SD and statistical analysis employed unpaired two‐tailed Student's t‐test. *p < 0.05, **p < 0.01, and ***p<0.001.

Figure 2

TLR2 signaling induces a type 2 immune response to OVA challenge. OVA‐immunized TLR2^−/−mice and WT mice treated with OxPAPC in sensitization phase (S‐OxPAPC) or in challenge phase (C‐OxPAPC) or in both phases. (A) Quantification of lung IL‐4, IL‐5, and IL‐13 by ELISA. (B) Quantification of serum OVA‐specific IgE (OVA‐sIgE) by ELISA. (C) Quantification of lung TSLP by ELISA. (D) Proportion of Th2 cells in MLNs was analyzed by flow cytometry. The gating scheme employed is depicted in Supporting Information Figure 1. (E) Proportion of eosinophils (CD11c⁻Siglec F⁺) in BALF. The gating scheme employed is depicted in Supporting Information Figure 2. (F) Lung tissue sections were analyzed by H&E, PAS, and Masson staining. Scale bar represents 200 μm. For ELISA, data are pooled from three independent experiments with five mice each. For flow cytometry in (D) and (E), one representative flow analysis of three independent experiments with five mice each is shown. For statistical analysis of flow cytometry, data are pooled from three independent experiments with five mice each. Values are shown as mean ± SD and statistical analysis employed unpaired two‐tailed Student's t‐test. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 3

The effect of TLR2 signaling on innate immune cells in the lung. OxPAPC‐treated mice or TLR2^−/− mice received OVA immunization. Lung cell suspension was subjected to flow cytometry analysis for the number and molecule expression of basophils and DCs in the lung. Alternatively, lung homogenates were subjected to protein analysis by ELISA. Gating strategy for basophils and DCs in blood by flow cytometry was shown in Supporting Information Figure 3. (A) Gating strategy for basophils and DCs in lung. (B) Quantification of lung c‐kit⁻ CD49b⁺ FcεRIa⁺ CD200R⁺ basophils and CD11c⁺ CD11b⁺ MHCII⁺ CD103⁻ DCs in OVA‐immunized TLR2^−/− and OxPAPC‐treated WT mice. (C) Quantification of basophils and DCsin peripheral blood. (D) Quantification of lung CCL2. (E) Determination of basophils in the lung after CCL2 blockade. (F) Expression levels of co‐stimulatory molecules on basophils and DCs in the lung. For flow cytometry, one representative flow analysis for basophils and DCs of three independent experiments with five mice each is shown. For statistical analysis of flow cytometry and ELISA, data are pooled from three independent experiments with five mice each. Values are shown as mean ± SD and statistical analysis employed unpaired two‐tailed Student's t‐test.*p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 4

TLR2 signaling promotes TSLP production to activate basophils and DCs leading to airway inflammation. For in vivo study, TLR2^−/− mice or OxPAPC‐treated WT mice received OVA sensitization and OVA challenge. Lung cells were subjected to flow cytometry analysis for determination of the number and molecule expression of basophils and DCs. Alternatively, lung cells were subjected to basophil or DC isolation by flow cytometry followed by RNA extraction for RT‐PCR. For some experiments, lung homogenates were used to determine protein levels by ELISA or histology. Gating strategy for basophils and DCs in lung by flow cytometry was shown in Supporting Information Figure 3. (A) Quantification of TSLP and CCL2 in the culture supernatant of pBECs by ELISA. pEBCs were isolated from WT or TLR2^−/− mice followed by stimulation with OVA. Culture supernatant was collected for determination of cytokines. (B) Quantification of the lung TSLP level was determined by ELISA. (C) Expression of TSLPR on lung basophils and DCs was analyzed by flow cytometry. (D) Quantification of the number and IL‐4 mRNA expression of lung basophils and DCs by qPCR. (E) Expression of co‐stimulatory molecules on lung basophils and DCs by flow cytometry. (F) Determination of eosinophils in BALF by flow cytometry and quantification of IL‐4, IL‐5, and IL‐13 in lung homogenates by ELISA. (G) Histological analysis of lung tissues by H&E, PAS, and Masson staining. Scale bar represents 200 μm. For ELISA, RT‐PCR, calculation of cells in BALF, and flow cytometry, data are pooled from three independent experiments with five mice each. One representative flow analysis for TSLP receptor expression on basophils of three independent experiments with five mice each is shown. Values are shown as mean ± SD and statistical analysis employed unpaired two‐tailed Student's t‐test.*p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 5

TLR2–NF‐KB/JNK signaling pathways mediate TSLP production to prevent apoptosis of basophils and DCs. (A) OVA‐immunized mice received TSLP or anti‐TSLP were subjected to flow cytometry for determination of apoptosis in basophils and DCs in the lung. (B) Lung cell suspensions of OVA‐immunized mice were stimulated with TSLP or anti‐TSLP in vitro and were subsequently subjected to flow cytometry for the analysis of apoptosis in basophils and DCs. (C) Expression level of anti‐apoptosis protein Bcl‐XL in bone marrow‐derived basophils (BMBas) and BMDCs after TSLP treatment at different time points. (D) Analysis of p‐p65, p‐JNK, and p‐p38 in BEAS‐2B cells after OVA, HDM, and fungi stimulation in BEAS‐2B cells by western blot. (E) Expression level of TSLP mRNA in BEAS‐2B cells treated with inhibitors of NF‐κB, JNK, and p38 followed by OVA, HDM, and fungi stimulation by RT‐PCR. (F) Analysis of NF‐κB and p‐JNK in OxPAPC‐pretreated BEAS‐2B cells with OVA, HDM, and fungi stimulation by western blot. (G) Analysis of NF‐κB and JNK signals in OVA‐stimulated pBECs from WT or TLR2^−/− mice by western blot. For statistical analysis of apoptosis, data are pooled from three independent experiments with five mice each. For RT‐PCR and western blot, data are pooled from five independent experiments. One representative western blot of five independent experiments is shown. Values are shown as mean ± SD and statistical analysis employed unpaired two‐tailed Student's t‐test. *p < 0.05, **p < 0.01, and ***p < 0.001.

Figure 6

NF‐κB and JNK signaling pathways participate in TLR2‐derived TSLP production and activation of basophil and DC in vivo. OVA‐immunized mice were intraperitoneally injected with BAY11‐7082 or SP600125 to inhibit the NF‐κB or JNK pathway, respectively. (A) Analysis of p‐p65, p‐IКBα, and p‐JNK in the lung by western blot. (B) Quantification of TSLP and CCL2 in the lung by ELISA. (C) Number of basophils and DCs in the lung by flow cytometry. (D) Expression levels of co‐stimulatory molecules on basophils and DCs in the lung analyzed by flow cytometry. (E) Quantification of IL‐4, IL‐5, and IL‐13 in the lung by ELISA. (F) Histological analysis of lung tissues by H&E, PAS, and Masson staining. Scale bar represents 200 μm. For statistical analysis of western blot, data are pooled from three independent experiments with five mice each and one representative western blot is shown. For statistical analysis of ELISA and flow cytometry, data are pooled from three independent experiments with five mice each. Values are shown as mean ± SD and statistical analysis employed unpaired two‐tailed Student's t‐test. *p < 0.05, **p < 0.01, and ***p < 0.001.