Wild type C57BL/6 and Hsd11b1-deficient (Hsd11b1^-/-) mice were housed in specific pathogen-free (SPF) conditions in individually ventilated cages in the same room at the animal facility of the University of Konstanz. The generation of the Hsd11b1^-/- mice has been reported previously (17). Mice (male and female) from the same breeding stocks were used in experiments between 8-12 weeks of age. All experiments were conducted in accordance with the animal experimentation regulations of Germany and were approved by the Review Board of the Regional Council Freiburg i.B.

HEK293T cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) containing 5% steroid-free fetal bovine serum (FBS), 50 µg/ml gentamycin and 2,5 mM L-glutamine. E10 cells (murine alveolar epithelial type I cell line) were cultured in DMEM-F12 containing 10% FBS, 50 µg/ml gentamycin and 2,5 mM L-glutamine at 37°C, 5% CO₂. E10 cells were pre-treated for 1 h with buffer control or 1 µM corticosterone, and subsequently treated with 1 µg/ml lipopolysaccharide (LPS) (S. minnesota, Sigma, St. Louis, USA) or 10 ng/ml house dust mite (HDM) (Endotoxin: 1,61 EU/µg protein, Citeq Biologics, Groningen, Netherlands) extract. Supernatant and cells were harvested at indicated time points.

C57BL/6 and Hsd11b1^-/- mice were intraperitoneally (i.p.) injected with either 100 µl of phosphate buffered saline (PBS) (Sigma-Aldrich, St. Louis, USA), 100 µg LPS or 20 µg anti-CD3ε antibody (clone 145-2C11) in PBS. After 3 h, serum and lungs were collected and analyzed.

C57BL/6 and Hsd11b1^-/- mice were anesthetized with isoflurane and sensitized intranasally (i.n.) for 5 days with 100 µg (acute) or 5 days/week for 3 weeks with 25 µg (chronic) HDM extract in 50 µl PBS (25 µl/nostril). Control mice were treated with PBS. Mice were challenged with a single dose HDM extract 7 days after the last sensitization (acute) or for 5 continuous days (chronic). Mice were analyzed 24 h after the final challenge.

Euthanized mice were lavaged three time with 700 µl PBS by intratracheal flexible cannula (20G, DRIFTON, Hvidovre, Denmark). Cells were spun on glass slides or stained for flow cytometry. Protease inhibitor (Roche, Basel, Switzerland) was added to the fluid (BALF) for storage.

Lungs were cut into pieces and 200 mg were cultured in sfDMEM (5% steroid-free FBS, 50 µg/ml gentamycin, 2,5 mM L-glutamine) with or without 200 µg/ml metyrapone (MET) or 100 µM 11-dehydrocorticosterone (VWR, Radnor, USA), for 6 h (37°C, 5% CO₂). For LPS, anti-CD3ε antibody and HDM experiments, lungs were perfused, superior lung lobes were cut and 10 mg tissue was cultured as described before. Corticosterone levels measured in MET-treated samples were subtracted to exclude contamination by serum-derived GC.

Corticosterone levels in serum and ex vivo lung cultures was determined by a previously published GR-based luciferase reporter assay (16). In short, HEK293T cells were transfected with a glucocorticoid response element (GRE)-containing luciferase reporter construct (GRE2tk-LUC), a GR-expression plasmid (SVGR1) and a β-galactosidase-expression plasmid for normalization, before exposure to serum or ex vivo culture supernatant. Luciferase activity was determined after overnight incubation. GC concentrations were calculated using a corticosterone standard curve.

Left lung lobes were digested and single cell suspensions were obtained as previously described (18). Cells were stained with different antibody panels ( Table S2 ) as follows: cells were washed twice with PBS and incubated with a fixable viability dye ( Table S2 ) for 30 min at 4°C. Afterwards, cells were washed twice with staining buffer (5% BSA, 2 mM EDTA, 2 mM NaN₃ in PBS) and incubated in blocking buffer (TruStain FcX™ Plus antibody (BioLegend, San Diego, CA, USA) in staining buffer) for 30 min at 4°C. Cells were then incubated with the respective antibodies in staining buffer for 30 min at 4°C, followed by two washing steps with staining buffer. For cytokine analysis, cells were stimulated with phorbol myristate acetate (PMA, 20 ng/ml), Ionomycin (1 µg/ml) and Brefeldin A (5 µg/ml) for 6 h prior to antibody staining. For intracellular staining, cells were fixed and permeabilized with Transcription Factor Staining Buffer Set (eBioscience, San Diego, USA) for 30 min. Samples were analyzed on a LS Fortessa (Becton Dickinson (BD), Franklin Lakes, USA). Compensation, cleaning and pre-gating for single cell, live, CD45⁺ populations was performed in FlowJo (v10.8.1). Data was exported and analyzed using an adapted R script based on Nowicka et al., 2017, Brummelman et al., 2019 and Ingelfinger et al., 2021 (19–21). FCS data were transformed (inverse hyperbolic arcsinh function) and downsampled to 2500 cells/FCS file for normalization. Combined data were applied to Uniform Manifold Approximation and Projection (UMAP) analysis (umap package v0.2.8.0) and applied to unsupervised FlowSOM clustering (FlowSOM package v2.0.0 and ConsensusClusterPlus v1.56.0). 25 to 30 clusters were, based on their marker expression, manually annotated to different myeloid und lymphoid immune cell subsets. Frequencies and total cell numbers were calculated, exported and plotted in GraphPad Prism (v.8.0; La Jolla, USA).

RNA isolation was performed with RNAsolv (Omega Bio-tek, Norcross, USA) or SVtotal RNA Isolation System (PROMEGA, Madison, USA) according to the manufacturer’s protocols. SYBRGreen-based RT-qPCR was performed with StepOnePlus Real-Time PCR system/QuantStudio™ 3 System (Applied Biosystems, Waltham, USA) (Primers: Table S4 ). Expression was normalized to β-Actin.

Formalin-fixed and paraffin-embedded murine lung tissue sections were deparaffinized, rehydrated and hematoxylin and eosin (H&E) stained, Sirius red stained or periodic acid-Schiff (PAS) stained. For immunofluorescence, heat-induced epitope retrieval (sodium citrate pH 6,0) was performed prior to antibody staining with AlexaFluor 488-labelled anti-11β-HSD1 antibody, or isotype control ( Table S1 ) (22). Cytospins were stained with Hemacolor® Rapid staining Kit (Merck, Darmstadt, Germany) according to the manufacturer’s protocol.

Interleukin (IL)-4, IL-6, tumor necrosis factor (TNF) and IgE concentrations in serum, BALF or lung homogenates were quantified by ELISA according to the manufacturer’s protocols ( Table S3 ).

E10 cells were harvested and lysed with radioimmunoprecipitation assay (RIPA) buffer ( Table S3 ) and 40 µg protein (determined by Pierce™BCA Protein Assay Kit, Thermo Fisher, Waltham, USA) were loaded on a 12% SDS-PAGE. After electrophoretic size separation, proteins were transferred to polyvinylidene difluoride membranes (Roche). Membranes were blocked, incubated o/n at 4°C with primary antibodies ( Table S1 ), and for 1 h at RT with the corresponding secondary antibody ( Table S1 ) prior to development with ECL reagent ( Table S5 ). Imaging was performed with ImageQuant LAS4000 (GE Healthcare, Chicago, USA).

Analyses were performed using GraphPad Prism (v.8.0; La Jolla, USA). Normal distribution tests: D’Agostino-Pearson omnibus, Shapiro Wilk, Kolmogorov-Smirnov test. Details are indicated in the Figure legends.

Our previous studies on extra-adrenal GC synthesis in murine lungs suggested a possible role of 11β-HSD1-mediated GC reactivation during acute immunological stress, but this was not investigated in a genetic model (16). Based on our observation that Hsd11b1 is presumably expressed in all different epithelial cell types of the lung, we investigated a complete Hsd11b1 knockout in this study ( Figure 1A ; Supplementary Figure 1A ) (23). Firstly, we characterized constitutive lung GC synthesis in Hsd11b1^-/- (KO) mice in comparison with wild type (WT) animals. In KO mice, 11β-HSD1 expression was completely absent on mRNA and protein level ( Figures 1B, E ), and ex vivo corticosterone synthesis as well as 11-dehydrocorticosterone (11-Dhc) conversion were significantly reduced ( Figure 1C ). In contrast, serum GC levels remained unchanged ( Figure 1D ). Although 11β-HSD1 has been described to be important during lung development, we did not detect histological differences in the lung tissue ( Figure 1B ). Interestingly, we found a trend of enhanced expression of the steroidogenic enzymes Cyp11a1, Cyp11b1 and the cholesterol transporter Star, which are important components of the de novo GC synthesis pathway ( Figure 1E ). The immune cell composition in lungs and bronchoalveolar lavage (BAL) of KO mice was largely unchanged ( Figures 1F-J ; Supplementary Figures 1B-I ), however, minor differences in the frequencies of IFNγ⁺ CD8⁺, as well as IL-17A⁺ CD4⁺ T cells and natural killer T (NKT) cells indicate a certain bias towards a Th1/Th17-type differentiation in KO mice under steady-state conditions ( Figures 1H-J ).

We previously described that immune cell-induced GC synthesis in the lungs was abolished in adrenalectomized mice, suggesting that lung GC synthesis depends on adrenal precursors, and that the lung tissue converts serum-derived 11-Dhc to corticosterone in an 11β-HSD1-dependent manner (16). To test this hypothesis, we challenged WT and Hsd11b1^-/- (KO) mice intraperitoneally (i.p.) with lipopolysaccharide (LPS), anti-CD3 antibody or PBS as control and analyzed them after 3 h ( Figure 2A ). While we could measure an immune cell activation-induced increase in serum GC levels in both WT and KO mice ( Figure 2B ), lung GC synthesis was induced only in ex vivo lung cultures of WT but not of KO mice ( Figure 2C ). Both, anti-CD3 antibody and LPS injection did not result in significantly enhanced Hsd11b1 expression but promoted, depending on the trigger, enhanced mRNA expression of inflammatory cytokines in lung tissue, i.e. Ifng, Tnf and Il1b ( Figures 2D, E ). Interestingly, despite short and indirect immune cell stimulation, we observed significantly more IL-6 protein in the bronchoalveolar lavage fluid (BALF) but not in liver and spleen of KO mice ( Figure 2F ; Supplementary Figures 2A, B ). Furthermore, we observed augmented leukocyte adherence to the vasculature and extravasation, indicating initial immune cell infiltration into the lung tissue ( Figure 2G ). Overall, the results confirm the previously suggested importance of 11β-HSD1-dependent reactivation of local GC in an acute model of lung inflammation.

To further study the role of lung GC synthesis in the regulation of airway inflammation we continued to analyze the relevance of 11β-HSD1-mediated GC reactivation in an acute model of house-dust mite (HDM)-induced airway hypersensitivity ( Figure 3A ). Interestingly, while exposure of WT mice to HDM extract did not result in significant immune cell infiltration of lung tissue, KO mice showed significantly enhanced immune cell numbers in the lung tissue ( Figures 3B, C ; Supplementary Figure 3A ), and the airway space, as quantified by an increased cell count in the BAL ( Figures 3D, E ; Supplementary Figure 3B ). While total protein levels in lung homogenates of HDM treated mice remained unchanged, IL-4 was significantly enhanced in KO mice compared to control-treated animals ( Figure 3F ; Supplementary Figure 3G ). Total serum IgE levels remained unchanged at this time point ( Supplementary Figure 3F ). Strikingly, while Ccl11 (eosinophil chemoattractant) expression was enhanced in both WT and KO mice, in line with the observed eosinophil infiltration, elevated Cxcl1 (neutrophil chemoattractant) expression and associated neutrophil infiltration was only observed in lungs and BAL of KO mice ( Figures 3G, H ; Supplementary Figures 3C-E ). Interestingly, serum GC levels remained unchanged and ex vivo lung GC synthesis was not induced ( Supplementary Figures 3H, I ). These data indicate that the lack of basal GC reactivation in the lung of 11β-HSD1-deficient mice results in an exacerbation of acute HDM-induced allergic immune responses with a strong neutrophilic component.

In order to determine the effects of missing GC reactivation during long-term airway inflammation, mice were chronically exposed to HDM extract ( Figure 4A ). Similar to the acute model, but more pronounced, we observed a strongly aggravated phenotype of the KO animals. The immune cell infiltration into the lungs and airway space was drastically enhanced and collagen deposition and mucus production were more pronounced in HDM-treated KO mice compared to WT ( Figures 4B-E ; Supplementary Figures 4A-E ). Additionally, total protein and IL-4 in lung homogenates and IL-4 and IL-5 mRNA expression were significantly enhanced in KO mice compared to WT mice ( Figure 4F ; Supplementary Figure 4G ). Moreover, there was a tendency of elevated total serum IgE levels, especially in the KO mice ( Supplementary Figure 4H ). At this time point after chronic allergen exposure, Ccl11 expression in KO lungs was significantly enhanced compared to those in WT lung, which is in line with the observed trend of enhanced eosinophil infiltration. While an allergen-induced neutrophilic infiltration was still not observed in WT lung, despite a significant increase in Cxcl1 expression, lungs of KO mice showed a remarkable increase in neutrophil infiltration ( Figures 4G, H ; Supplementary Figure 4F ). Similar to the acute model, serum GC levels were not increased upon HDM exposure and ex vivo lung GC synthesis was not induced ( Supplementary Figures 4I, J ). From these data, we concluded that the lack of basal GC reactivation results in an aggravated inflammatory phenotype and retains the characteristics of a mixed eosinophilic and neutrophilic inflammation in KO mice over time.

The immune response in HDM-induced airway hypersensitivity is complex and involves various cell types of the innate and adaptive immune system (24). Thus, we next expanded the immune cell analysis to additional cell types, revealing major differences in the immune cell composition between WT and KO lungs exposed to HDM extract ( Figures 5A-C ). While the number of alveolar macrophages decreased upon HDM treatment, monocytes, conventional dendritic cells type 1 (cDCs1), natural killer (NK) cells, plasmacytoid DC (pDCs) and type 1 and 2 interstitial macrophages (iMs1, 2) remained largely unchanged in both, WT and KO lungs ( Figures 5D, E ; Supplementary Figures 5A, B ). However, the number of monocyte-derived macrophages (moMs) and monocyte-derived DCs (moDCs) drastically increased in lungs of acutely treated KO mice, indicating stronger inflammation compared to WT mice ( Figure 5D ). While this effect on moMs was less pronounced in chronically treated mice, the increase of moDCs seemed to be slightly enhanced over time, with a consistently greater increase in KO mice ( Figures 5D, E ). A similar observation was made for cDCs2, an important DC subset that contributes to Th2 and Th17 polarization of CD4⁺ T cells ( Figures 5D, E ). Taken together, these results confirm a dysregulated immune response in lungs of mice lacking GC reactivation, which is in line with the drastic phenotype of Hsd11b1-deficient mice in both models.

Thus far, our results demonstrate that the absence of 11β-HSD1 and associated GC reactivation substantially contributes to deregulated immune cell activation and the aggravated inflammatory phenotype during HDM-induced airway hypersensitivity. The remarkable differences in neutrophil and eosinophil infiltration into the lungs of KO mice compared to WT mice indicate specific differences in their respective immune response towards HDM extract. To further characterize the underlying mechanism, we performed in vitro assays with an alveolar epithelial cell type 1 cell line (E10). Exposure of E10 cells to HDM extract resulted in significantly enhanced IL-6 secretion and Cxcl1 expression ( Figures 6A, B ). This induction of inflammatory cytokines and chemokines could be suppressed when cells were pre-treated with corticosterone ( Figures 6A, B ), indicating an important role of local GC in regulating the expression of epithelial-derived cytokines and chemokines that potentially drive a Th17-type immune response. In line with this, E10 cells treated with corticosterone, prior to LPS exposure (an important component of HDM extract), showed decreased c-Jun-N-terminal kinase (JNK) and extracellular signal-regulated kinase (ERK) phosphorylation, as well as IκBα (nuclear factor of kappa light polypeptide gene enhancer in B cells inhibitor α) degradation, indicative of reduced activator protein 1 (AP-1) and NFκB (nuclear factor kappa-light-chain-enhancer of activated B cells) transcriptional activity ( Figure 6C ). In this context, we continued to analyze lymphoid immune cell subsets and their cytokine expression to determine the type of immune response in our chronic model of airway inflammation ( Figures 6D-J ; Supplementary Figures 6A-F ). Similar to some innate immune cells, the number of CD4⁺ and CD8⁺ T cells, and NKT cells increased more drastically in KO lungs compared to WT lungs after chronic exposure to HDM extract ( Figures 6D-F ). While there was a trend for increased numbers of IL-4⁺ CD4⁺ T cells after HDM allergen exposure, no significant differences between WT and KO could be demonstrated ( Figure 6G ). In contrast, a moderate difference could be observed in IFNγ⁺ CD4⁺ T cells, but not in IFNγ⁺ or IL-17A⁺ CD8⁺ T cells. ( Figure 6H ; Supplementary Figure 6E ). More strikingly, the analysis of IL-17A⁺ CD4⁺ T cells and IL-17A⁺ NKT cells revealed not only a massive increase of these cells in HDM-treated KO lungs, but also significant lower numbers in WT lungs ( Figures 6I, J ). In conclusion, our results demonstrate that 11β-HSD1-mediated GC reactivation regulates T cell priming and that the lack of local GC reinforces a Th17-type immune response that drives neutrophilic inflammation and associated tissue damage in HDM-induced airway hypersensitivity.