Birch pollen allergen-induced dsDNA release activates cGAS-STING signaling and type 2 immune response in mice

Abstract

Detecting cytoplasmic or extracellular DNA from host or pathogen origin by DNA sensor cyclic GMP-AMP synthase (cGAS) and stimulator of interferon genes (STING) triggers immune responses with secretion of type I interferons and inflammatory cytokines. However, STING agonists function as type-2 adjuvant promoting allergic asthma. Here, we asked how cGAS/STING signaling pathway influences allergen-induced type-2 immune responses in models of allergic airway diseases induced by birch pollen extract, house dust mite, or ovalbumin plus Alum. We report increased extracellular dsDNA in the airways, together with cGAS and STING gene expression, following allergen challenge in these models, correlating dsDNA and type-2 cytokine IL-4, IL-5, and IL-13 release. Allergen-induced type-2 immune responses were reduced in cGAS- or STING-deficient mice. Further, blocking cGAS function with the specific inhibitor RU.521 protected mice from birch pollen allergen-induced airway inflammation and type-2 immune responses. Thus, DNA sensing by cGAS contributes to type-2 immune responses and may represent a therapeutic target for allergic lung inflammation.

Keywords: Immune response; Immunology; Molecular biology.

Graphical abstract

Figure 1

Allergen exposures trigger dsDNA release and cGAS-STING pathway mobilization (A–I) Wild-type (WT) mice were exposed to birch pollen extract (BPE; A–C), house dust mite (HDM; D–F), and ovalbumin with alum (OVA; G–I) and DNA sensing pathway characterized. Mice were immunized with BPE (10 μg/mouse i.p. on day 0 and 14) and challenged daily from day 21–24 (10 μg/mouse i.n.). Alternatively, mice were sensitized with HDM (25 μg i.n.) on day 0 and 7, followed by three HDM challenges (10 μg/mouse i.n.) on day 14–16. Immunization with OVA (20 μg/mouse emulsified in 2 mg Alum gel i.p.) on day 0, 7, and 14 was followed by challenges on day 21–24 (OVA 10 μg/mouse i.n.). Control mice received saline solution. (A–I), Extracellular dsDNA concentration in the acellular fractions of bronchoalveolar lavage fluid (BALF), Tmem173 and Mb21d1 transcripts in the lungs normalized to RN18s or Hprt1 expression. (J–L) Correlation of dsDNA and IL-4, IL-5, and IL-13, respectively, in BALF of BPE exposed mice. Data are representative of two independent experiments. Graph data in (A–I) are presented as mean ± SEM with n = 5–10 mice/group in the saline vehicle control and 5–10 mice/group for the allergen-exposed group. Each point represents an individual mouse. Statistics were performed by non-parametric Mann-Whitney test (A–I), and by Spearman for correlation analysis (J–L). p value < 0.05 (∗); p value < 0.01 (∗∗), p value < 0.001 (∗∗∗), p value < 0.0001 (∗∗∗∗).

Figure 2

Birch pollen induces a cGAS/STING-dependent lung inflammation (A) WT, cGAS^−/−, and STING^−/− mice were sensitized twice with i.p. injections of birch pollen extract (BPE, equivalent to 10 μg Bet v 1) on days 0 and 14, followed by four intranasal BPE challenges (10 μg Bet v 1) daily on day 21–24, and analyzed on day 25 (A). Control mice were immunized and challenged with saline alone. (B and C) Total cell and eosinophil counts in BALF determined by May-Grünwald Giemsa staining of BAL cell cytospins. (D–G) IL-33, IL-4, IL-5, and IL-13 in BALF. (H–J) CCL11, CCL24, AREG concentrations in BALF. (K–N) Representative FACS dot plots with frequency and absolute number of CD4⁺IL-4⁺ and CD4⁺IL-13⁺ cells in lung. (O) Type I IFNβ concentration in BALF. Data are pooled from two independent experiments and are presented as mean ± SEM with n = 5–10 mice/group in the saline vehicle control and 7–11 mice/group for the BPE-exposed group. Each point represents an individual mouse. Statistics were performed by using one-way ANOVA test followed by Sidak multiple comparison test (A–C, F, I, J, and M–O) or non-parametric Kuskal-Wallis test followed by Dunn multiple comparisons test (E, G, and H). p value < 0.05 (∗); p value < 0.01 (∗∗), p value < 0.001 (∗∗∗), p value < 0.0001 (∗∗∗∗). See also Figure S1.

Figure 3

cGAS/STING-dependence of birch pollen-induced airway resistance WT, cGAS^−/−, and STING^−/− mice were treated as in Figure 2 and analyzed 24 h after the last intranasal birch pollen extract (BPE) allergen challenge or saline as vehicle control. (A) Airway resistance was measured with increasing dose of methacholine using a FinePointe system (Buxco). (B–D) Lung tissue histology with Periodic acid Schiff (PAS) staining of goblet cells for mucus production (B) (scale bar 1 mm for low magnification and 100 μm for ×20 higher magnification), with pathology scoring of the peribronchial infiltration of inflammatory cells (C) and mucus production (D). (E and F) Gob5 and Muc5ac gene expression were measured in lung tissue. Data are presented as mean values ± SEM of n = 9–10 animals in the saline group and 9–11 animals for BPE group. Statistics were performed by two-way ANOVA test followed by Tukey multiple comparisons test (A), one-way ANOVA test followed by Sidak multiple comparison test (C–E) or non-parametric Kuskal-Wallis test followed by Dunn multiple comparisons test (F). ns: not significant. p value < 0.05 (∗); p value < 0.01 (∗∗), p value < 0.001 (∗∗∗), p value < 0.0001 (∗∗∗∗).

Figure 4

Selective inhibition of cGAS mediated signaling by RU.521 reduces type 2 immune response and lung inflammation WT mice were sensitized twice with i.p. injections of birch pollen extract (BPE, using doses equivalent to 10 μg Bet v 1) on day 0 and 14, followed by four daily challenges on day 21–24 of intranasal BPE (equivalent to 10 μg Bet v 1) under Isoflurane anesthesia (3%). (A) cGAS inhibitor (RU.521; at the doses of 5, 15, and 50 μg/kg Invitrogen) was administered to mice by i.n. route during sensitization and challenge. Twenty-four hours after the last BPE challenge, differential cell counts were determined in BAL. (B) Eosinophil counts in BAL. (C–G) Concentrations of CCL24 and IL-33 in lung homogenates, and IL-4, IL-5, and IL-13 in BALF. (H–J) PAS stained lung sections of goblet cells for mucus production (H) (scale bar 1 mm for low magnification and 100 μm for ×20 higher magnification), with pathology scoring of the peribronchial infiltration of inflammatory cells (I) and mucus production (J). Data are presented as mean values ±SEM of n = 5 animals in the saline group and 7–8 animals for BPE group. Statistics were performed by one-way ANOVA test followed by Sidak multiple comparison test (B–D and J) or non-parametric Kuskal-Wallis test followed by Dunn multiple comparisons test (G and I). p value < 0.05 (∗); p value < 0.01 (∗∗), p value < 0.001 (∗∗∗), p value < 0.0001 (∗∗∗∗). See also Figure S2.