IL-37 Ameliorating Allergic Inflammation in Atopic Dermatitis Through Regulating Microbiota and AMPK-mTOR Signaling Pathway-Modulated Autophagy Mechanism

Abstract

Interaction between eosinophils and dermal fibroblasts is essential for provoking allergic inflammation in atopic dermatitis (AD). In vitro co-culture of human eosinophils and dermal fibroblasts upon AD-related IL-31 and IL-33 stimulation, and in vivo MC903-induced AD murine model were employed to investigate the anti-inflammatory mechanism of IL-1 family cytokine IL-37 in AD. Results showed that IL-37b could inhibit the in vitro induction of AD-related pro-inflammatory cytokines IL-6 and TNF-α, and chemokines CXCL8, CCL2 and CCL5, increase autophagosome biogenesis-related LC3B, and decrease autophagy-associated ubiquitinated protein p62 by regulating intracellular AMP-activated protein kinase (AMPK) and mammalian target of rapamycin (mTOR) signaling pathway. In CRISPR/Cas9 human IL-37b knock-in mice, IL-37b could significantly alleviate MC903-stimulated ear tissue swelling, itching sensation and the level of circulating cytokine IL-6 and ear in situ expression of AD-related TNF-α, CCL5 and transforming growth factor-β. Moreover, IL-37b could significantly upregulate Foxp3+ regulatory T cells (Treg) in spleen and ear together with significantly increased serum Treg cytokine IL-10, and decrease eosinophil infiltration in ear lesion. IL-37b knock-in mice showed a distinct intestinal microbiota metabolic pattern upon MC903 stimulation. Furthermore, IL-37b restored the disordered gut microbiota diversity, through regulating the in vivo autophagy mechanism mediated by intestinal metabolite 3-methyladenine, adenosine monophosphate, 2-hydroxyglutarate, purine and melatonin. In summary, IL-37b could significantly ameliorate eosinophils-mediated allergic inflammation via the regulation of autophagy mechanism, intestinal bacterial diversity and their metabolites in AD. Results therefore suggest that IL-37 is a potential anti-inflammatory cytokine for AD treatment.

Keywords: IL-37; allergic inflammation; atopic dermatitis; autophagy; eosinophils; microbiota.

FIGURE 1

Effect of IL-37b on the induction of autophagy by analyzing the expression of p62, LC3B, AMPK and mTOR in human eosinophil-dermal fibroblast co-culture upon IL-31 (100 ng/mL) and IL-33 (100 ng/mL) stimulation. Cq: chloroquine (20 μM), autophagy negative control. Rapa: rapamycin (10 ng/mL), autophagy positive control. To verify the up-regulatory effect of IL-37b on AMPK, dorsomorphin (20 μM, Compound C/C.C), an AMPK inhibitor, together with IL-37b were applied in the co-culture. (A) Western blots analysis and (B) Quantification of AMPK, mTOR, p62 levels normalized to β-actin and LC3II/LC3I ratio (100 ng/mL IL-37b treatment for 24 h). (C) Western blot analysis. The Cq band for β-actin has been spliced from the same gel, due to an experimental error. Please see the Supplementary Material for the original blot image. (D) Quantification of AMPK, mTOR, p62 levels normalized to β-actin and LC3II/LC3I ratio in co-culture treated with IL-31, IL-33, and IL-37b (100 ng/mL), chloroquine and rapamycin treatment for 16 h. (E) Representative images of LC3 puncta (green) in co-culture treated with IL-31, IL-33, and IL-37b (100 ng/mL), chloroquine and rapamycin treatment for 16 h. (F) Representative images of LC3 puncta (green) in co-culture treated with IL-31, IL-33, and IL-37b (100 ng/mL), chloroquine and rapamycin treatment for 1, 2, and 4 h. Nuclei were visualized by DAPI (blue) staining. (G) Western blot analysis. (H) Quantification of AMPK and mTOR levels normalized to β-actin in co-culture treated with Compound C and IL-37b for 16 h. Chloroquine (Cq) pretreatment for 60 m prior to IL-37b treatment; rapamycin (Rapa) pretreatment for 10 m prior to IL-31 and IL-33 treatment. Bar charts are shown as mean ± SEM of triplicate independent experiments. *P < 0.05, **P < 0.01, and ***P < 0.001 when compared between the denoted groups.

FIGURE 2

In vivo effect of IL-37 on the production of cytokine and chemokine in AD mice. Serum level of (A) IL-6, (B) IL-5, (C) IL-13, (D) IL-10, and mRNA expression of (E) IL-4, (F) TGF-β, (G) CCL2, (H) CCL5, and (I) TNF-α of wild type and IL-37b Tg mice upon MC903 stimulation. Bar charts are shown as mean ± SEM (n = 6 mice). *P < 0.05 and **P < 0.01 when compared between the denoted groups.

FIGURE 3

Histological assessment of AD-like skin lesions and immune cell recruitment in ear lesion tissue and spleen. (A) Ear swelling and H&E staining of (B) Sham, (C) Wild type, (D) IL-37 Tg in AD-like skin lesions, (E) Snatching time of wild type and IL-37b Tg mice upon MC903 stimulation. mRNA expression of (F) Foxp3 and (G) CCR3 of wild type and IL-37b Tg mice upon MC903 stimulation. Immunohistofluorescence analysis of eosinophils and dermal fibroblasts by co-staining of dermal fibroblasts and eosinophils using the anti-Vimentin (green) and anti-MBP (red; white triangle pointed) antibodies, respectively. Nuclei were visualized by DAPI (blue) staining. (H) Sham; (I) Wild type mice with MC903 stimulation; (J) IL-37b Tg mice with MC903 stimulation. (K) % of Treg cells in splenic CD4+ Th cells of different groups. Bar charts are shown as mean ± SEM (n = 6 mice). *P < 0.05 and **P < 0.01 when compared between the denoted groups.

FIGURE 4

Restoration of gut microbiota by IL-37b in AD. (A) Species abundance heatmap of dominant 35 genera. (B) Principal Component Analysis (PCA). (C) Non-Metric Multi-Dimensional Scaling (NMDS). When the value of Stress factor is less than 0.2, it is considered that NMDS is reliable. (D) Unweighted Pair-group Method with Arithmetic Means (UPGMA) cluster tree based on Unweighted Unifrac distance (1) and Weighted Unifrac distance (2). (E) Histogram of LDA scores. (F) Bar chart of the relative content of representative pathogenic bacteria between W2 and T2. W1, stool samples of wild type mice at day 1 without MC903 stimulation; T1, stool samples of IL-37b Tg mice at day 1 without MC903 stimulation; W2, stool samples of wild type mice at day 17 with MC903 inducing AD model; T2, stool samples of IL-37b Tg mice at day 17 with MC903 inducing AD model. Bar charts are shown with triplicate independent experiments (n = 1 mouse in each experiment).

FIGURE 5

Restoration of metabolic patterns by IL-37b in AD. (A) Metabolite volcano map. (B) Metabolite clustering heat map. (C) Relative abundance of differential metabolites among different groups with Log₁₀ of peak intensity; p: positive ion mode, n: negative ion model. A1, stool samples of wild type mice at day 1 without MC903 stimulation; B1, stool samples of IL-37b Tg mice at day 1 without MC903 stimulation; A2, stool samples of wild type mice at day 15 with MC903 inducing AD model; B2, stool samples of IL-37 Tg mice at day 15 with MC903 inducing AD model. N = 6 mice.

FIGURE 6

Autophagy mechanisms induced by IL-37b in AD. Autophagy related representative metabolites between different groups with Log10 of peak intensity. (A) Adenosine 5’-monophosphate. (B) 3-methyladenine. (C) Melatonin. (D) Purine. (E) D-α-Hydroxyglutaric acid. A2, stool samples of wild type mice at day 15 with MC903 inducing AD model; B2, stool samples of IL-37 Tg mice at day 15 with MC903 inducing AD model. (F) Western blot analysis of autophagy related proteins in ear tissue of wild type mice and IL-37b Tg mice, upon MC903 stimulation for 16 days and (G) Quantification of AMPK, mTOR, IL-37b levels normalized to β-actin and LC3II/LC3I ratio. (1: sham group, wild type mice without MC903 stimulation; 2: IL-37b non-AD group, IL-37b mice without MC903 stimulation; 3: wild type AD group, wild type mice with MC903 stimulation; 4: IL-37b Tg AD group, IL-37b Tg mice with MC903 stimulation). (H) Immunostaining assessment of LC3, AMPK, mTOR, and IL-37 expression in ear tissue of wild type mice and IL-37b Tg mice, upon MC903 stimulation for 16 days. Bar charts are shown as mean ± SEM (n = 6 mice). *P < 0.05 and **P < 0.01 when compared between the denoted groups.