Background
Asthma is a chronic airway inflammatory disorder associated with an aberrant immune response to allergens and tissue remodeling. Asthma symptoms include shortness of breath, coughing, and wheezing, which can vary in frequency and intensity over time [
1]. This condition affects more than 300 million people worldwide and poses a major challenge to healthcare costs [
2]. Asthma is thought to be caused by an abnormal inflammatory response to environmental agents in genetically susceptible individuals [
3,
4]. Although advancements in asthma management have been made in recent years, severe asthma that does not respond to corticosteroids remains a significant challenge, and asthma-related mortality still occurs. Hence, there is a need for a more comprehensive understanding of the underlying pathobiology of asthma to develop new and effective therapeutic options.
During allergen-induced airway inflammation, the initial inflammatory response involves the activation of T helper type 2 (T
H2) cells, which stimulate the influx of eosinophils, mast cells, and other leukocytes into the airways, along with increased serum IgE production. Subsequently, airway structural cells and resident immune cells release anti-inflammatory or pro-resolving mediators to control the extent of the immune response and resolve the inflammation and restore the airway back to its steady state. However, in asthmatic patients, the inflammatory process can become disrupted due to an overzealous inflammatory phase overproducing pro-inflammatory factors or the disruption of the anti-inflammatory phase. These can result in a loss of immune tolerance against common airborne allergens [
5,
6].
Isthmin-1 (ISM1) is a protein highly expressed in the trachea and lungs of both mice and humans. Its expression in the airway epithelium was significantly enhanced by airway instillation of lipopolysaccharide (LPS), a bacterial endotoxin [
7,
8]. ISM1 is also expressed by lung lymphocytes such as natural killer (NK) cells, NKT cells, and CD4
+ T cells [
9], suggesting a potential role in immune regulation. We previously identified ISM1 as a pro-apoptotic protein that functions through two cell surface receptors, αvβ5 integrin and cell-surface GRP78 (csGRP78) in endothelial cells [
10,
11] and induces lung vascular permeability via the csGRP78 receptor and Src activation [
12]. Recently, ISM1 has been shown to selectively induce apoptosis of pro-inflammatory alveolar macrophages (AMs) via the csGRP78 receptor, thereby maintaining lung homeostasis [
13].
House dust mite (HDM) extract is known to induce endoplasmic reticulum (ER) stress in both human and mouse bronchial epithelial cells, which triggers the upregulation of the ER chaperone GRP78 [
14]. This upregulation of GRP78 expression leads to increased csGRP78 in stressed cells [
15,
16]. In allergic asthma, the bronchial epithelium is often stressed and damaged [
17]. This airway epithelial damage is believed to initiate and orchestrate inflammatory responses by releasing chemokines and cytokines, which recruit and activate inflammatory immune cells. However, whether ISM1 plays a role in the pathogenesis of asthma remains unknown.
Herein, we investigated the role of ISM1 in allergic airway inflammation induced by HDM extract in mice. Our findings indicate that ISM1 deficiency exacerbates airway inflammation and airway hyperresponsiveness (AHR), whereas intratracheal (i.t.) delivery of recombinant ISM1 (rISM1) attenuates HDM-induced airway inflammation. RNASeq analyses of ISM1-deficient lungs revealed significant downregulation of lung adiponectin, an anti-inflammatory adipokine [
18,
19]. Furthermore, we discovered that ISM1 potently stimulates adiponectin secretion in cultured human type 2 alveolar epithelial cells (A549) and enhances adiponectin-facilitated apoptotic cell efferocytosis by AMs both in vitro and in vivo [
20]. Our work unveils a previously unknown link between ISM1, adiponectin, and allergic airway inflammation, providing novel molecular insights into the pathophysiology of allergic airway inflammation.
Methods
Mice
Wild-type (WT) C57BL/6J mice were purchased from InVivos Pte. Ltd., Singapore. ISM1 knockout (
ISM1−/−) mice were generated in-house on a C57BL/6J background as described previously [
8,
13]. Seven- to eight-week-old age-matched mice were used in this study. All mice were housed under pathogen-free conditions, with water and food pellets supplied
ad libitum.
HDM-induced allergic-like airway/lung inflammation
A well-established animal model of HDM-induced allergic-like airway/lung inflammation was adopted for this study [
21]. Mice were sensitized on days 1, 7, and 14 with 50 µg of HDM extract (
Dermatophagoides pteronyssinus) (Greer Laboratories) in 20 µL of normal saline via the intratracheal (
i.t.) route after short anesthesia with isoflurane. Control mice received an equal volume of saline. For rISM1 treatment, 25 µL (1 mg/mL) of recombinant mouse ISM1 dissolved in 20 mM Tris buffered saline (TBS, pH 7.6) was administered intratracheally on day 12, 13, and 1 h before HDM challenge on day 14, 15, and 16. The vehicle group received an equal volume of 20 mM TBS. All mice were euthanized on day 17, and blood was collected by cardiac puncture to measure total IgE and HDM-specific IgE in the serum. Bronchoalveolar lavage fluid (BALF) was collected by perfusing 0.5 mL of chilled PBS into the trachea through a 22-inch IV catheter, which was repeated three times to gather approximately 1.5 mL of BALF. The right lung was excised for protein and RNA extraction, whereas the left lung was fixed in 10% formalin and paraffin-embedded for histological sectioning. Serum, BALF, and the right lung were snap-frozen in liquid nitrogen and stored at -80 °C until use.
Assessment of airway hyperresponsiveness
An independent set of animals was allocated for the invasive airway hyperresponsiveness (AHR) measurement using a Buxco® FinePointe resistance and compliance system equipped with analysis software (Data Sciences International, Harvard Bioscience). AHR was induced with methacholine 3 days after the final HDM sensitization. Briefly, the mice were anesthetized with a combination of ketamine (75 mg/kg) and medetomidine (1 mg/kg). A tracheotomy was performed and the cannulated mice were transferred into a whole-body chamber and mechanically ventilated. Animals were acclimatized for 5 min before being aerosolized with PBS. The baseline lung resistance against the PBS challenge was recorded for 3 min. Then, an increasing concentration of methacholine (5, 10, 20, and 40 mg/ml) was aerosolized and lung resistance was recorded. The average lung resistance (RL) and dynamic compliance (Cdyn) values were used to demonstrate the changes in the lung function of the mice.
Enzyme-linked immunosorbent assay (ELISA)
Cytokines were quantified using LUNARIS™ mouse 12-plex cytokine kit (LMCY-20,120 S, Ayoxxa Biosystems) and MILLIPLEX® Mouse Th17 Magnetic Bead Panel (MTH17MAG-47 K, Millipore) according to the manufacturer’s protocol. Chemokines such as eotaxin-1, eotaxin-2, and monocyte chemoattractant protein-1 were measured by singleplex ELISA (Raybiotech). The serum levels of total IgE and HDM-specific IgE were measured using an OptEIA™ Mouse IgE ELISA kit (555,248, BD Biosciences) according to the manufacturer’s protocol. ISM1 was measured in mouse BALF using a LEGEND MAX™ Mouse Isthmin ELISA Kit (438,907, BioLegend). TGFβ1 in BALF and lung tissue lysates was measured using a TGFβ1 ELISA kit (DY1679, R&D Systems). Adiponectin levels were measured using a mouse adiponectin/Acrp30 Quantikine ELISA kit (R&D Systems).
Histopathologic evaluation of lung tissue
The left lung tissue section was stained with hematoxylin and eosin (H&E) and periodic acid-Schiff (PAS) to evaluate airway inflammation and goblet cell metaplasia, respectively. Arbitrary scoring was performed by three experienced observers who were blinded to the treatment group. The extent of peribronchial and perivascular cell infiltration and mucus-producing goblet cells in the lung tissue were quantified on a 0–4 scale as described previously [
22]. Inflammatory score: 0 = no infiltration; 1 = a few cells; 2 = a ring of cells with one-layer of cell depth; 3 = a ring of cells with two to four-layer of cell depth; 4 = a ring of cells with more than four layers of cell depth. The number of PAS-positive cells was counted under a light microscope, and the results were expressed as the percentage of PAS-positive cells over the total number of epithelial cells in the airway. PAS score: 0 = no PAS-positive cell; 1 = less than 25% positive cells; 2 = 50% positive cells; 3 = 75% positive cells; and 4 = more than 75% positive cells. The inflammatory and goblet cells were scored in at least six different fields for each lung section. Mean scores were obtained from five animals.
Immunohistochemistry and immunofluorescence staining
Lung Sections (5 μm) were deparaffinized and rehydrated and antigen was retrieved by heating in sodium citrate buffer (10 mM sodium citrate, 0.05% Tween 20, pH 6.0). After quenching the endogenous peroxidase activity with 3% hydrogen peroxide, the sections were blocked with 3% BSA, stained with the respective primary and secondary antibodies and reacted with diaminobenzidine substrate. Nuclei were counterstained with hematoxylin. Immunofluorescence-stained slides were counterstained with DAPI. Images were captured with a Zeiss Axio Imager M2 microscope and analyzed using ImageJ (NIH).
Flow cytometry
Total and differential cell counts in the BALF were determined by Cytoflex LX flow cytometer with CytExpert software ver.2.3 (Backman Coulter). Forward and side scatter plots were used to exclude cell debris and clumps. Total leukocytes were identified as CD45+, AMs as CD11c+ Siglec-F+, eosinophils as CD11c− Siglec-F+, neutrophils as GR-1+ CD11b+, and dendritic cells as CD11c+ CD11b+. All antibodies used for flow cytometry were purchased from Miltenyi Biotec. A sequential gating strategy was performed to differentiate among different cell types.
Western blotting
Frozen lung tissues were ground in chilled 8 molar urea lysis buffer containing protease and phosphatase inhibitor cocktail (Roche). On the other hand, protein in the cell culture conditioned media was precipitated using cold acetone (-20 °C). Palleted protein was then resolubilized in electrophoresis sample buffer. Protein lysates (30 µg protein) were electrophoretically separated on either 10% SDS PAGE glycine gels or 15% tricine gels and transferred to a nitrocellulose membrane. After blocking with 5% BSA, the membranes were sequentially incubated with the primary antibody and secondary antibody. Finally, the protein bands were analyzed using the LI-COR Odyssey Imaging system. The primary antibodies used were as follows: rabbit anti-cleaved TGFβ1 (V) (sc-146, SCBT), mouse anti-RIP3 (B-2) (sc-374,639, SCBT), rabbit anti-pMLKL (D6E3G, CST), rabbit anti-caspase 8 (D35G2, CST), rabbit anti-cleaved caspase 3 (D175, CST), rabbit anti-PPARγ (16,643-AP, Proteintech), rabbit anti-adiponectin (21,613-AP, Proteintech), goat anti-adiponectin (AF1119, R&D Systems), and mouse anti-β-actin (C4) (sc47778, SCBT) was used as the loading reference. The secondary antibodies used were as follows: donkey anti-mouse IRDye 680, donkey anti-rabbit IRDye 800, and donkey anti-goat IRDye 800 (all from LI-COR Biosciences). For the antibody neutralization experiment, the following antibodies were used: mouse anti-GRP78 (A10) (sc-376,768, SCBT), mouse anti-αvβ5 (P1F76, SCBT).
RNA sequencing
Total RNA was isolated from perfused whole lungs using TRIzol reagent (Invitrogen) according to the manufacturer’s recommended protocol. Strand-specific mRNA-seq libraries for the Illumina NovaSeq 6000 platform were generated and sequenced by Novogene, Singapore. Briefly, high-quality total RNA with an RNA integrity number (RIN) > 8 was subjected to poly-A enrichment and a directional mRNA library was prepared. The library was then subjected to paired-end sequencing of 150 bp read length and sequence depth of 40 million reads per sample. The raw sequencing data were filtered to remove low-quality reads and adaptor sequences. Clean reads were mapped onto the reference genome Mus musculus (GRCm38/mm10) using STAR software. The number of reads that mapped a certain gene or transcript was measured to calculate the gene expression level and the raw counts obtained were normalized to FPKM (fragments per kilobase of transcript sequence per million base pairs sequenced), which took sequence depth and gene length into account. Pearson correlation and principal component analysis (PCA) were performed using normalized read counts.
For initial differential expression analysis, the read counts from the gene expression level analysis were input into DESeq2 with the following thresholds: Padj ≤ 0.05, log2FC ≥ 0. For subsequent analysis, the biomaRt R package was used to convert the ensemble ID to a mouse gene symbol and to obtain the gene types. Protein-coding genes were chosen for further analysis. Genes with mean counts per million (CPM) across samples less than 2 were filtered out using the CPM function from the edgeR package. Only protein-coding genes with CPM > 2 were input into DESeq2 for differential expression analysis with the following parameters: log2FC ≤ -1 or log2FC ≥ 1 & Padj < 0.05 to remove low expressing genes and to identify highly differentiated genes. Deseq2-normalized expression values were plotted as heatmaps and volcano plots using the heatmap R package and volcano R package respectively. For pathway enrichment analysis, the Fgsea R package was used with the mouse gene ortholog version of the Kyoto Encyclopedia of Genes (KEGG) and REACTOME gene sets from MsigDB database V5.2. Pathways with a Benjamini-Hochberg Padj < 0.05 were considered significantly deregulated pathways for each gene set. Gene Ontology (GO) and KEGG pathway analyses of DEGs were performed using DAVID software. The p-value was calculated by the Benjamini-Hochberg correction procedure for multiple hypothesis testing.
qRT-PCR
Total RNA (1 µg) from lung tissues was reverse transcribed using a Maxima First Strand cDNA Synthesis kit (Thermo Scientific) according to the manufacturer’s instructions. For qPCR, SsoAdvanced Universal SYBR Green Supermix (Bio-Rad) was used. The primer sequences used are listed in Additional file 1: Table
S1. The relative quantification of gene expression was determined using the comparative Cq (ΔΔCq) method and represented as the Log
2 fold change.
Efferocytosis Assay
In vitro efferocytosis assay was performed according to a previously described protocol with minor modifications [
23]. Jurkat T cells were exposed to UV irradiation (254 nm) for 30 min and incubated in complete RPMI media supplemented with 10% FBS for 2 h to induce early apoptosis. Apoptotic Jurkat T cells were then labeled with Phrodo™ Red, succinimidyl ester (P36600, Thermo Scientific) for 1 h according to the manufacturer’s protocol. Labeled-apoptotic Jurkat T cells (1 × 10
6 cells) were treated with recombinant adiponectin (HY-P7129, MedChemExpress) or A549 conditioned media treated with (CM
+ ISM1) or without rISM1 (CM
− ISM1) for 1 h and then added to MH-S cells at a 1:10 ratio (Jurkat: MH-S) in Opti-MEM I reduced serum medium. The serum-free condition was used to nullify the contribution of adiponectin in serum. Efferocytosis was detected and measured using the IncuCyte ZOOM live cell imaging system (Sartorius) [
24]. ISM1-induced adiponectin in A549 conditioned media was then neutralized using an anti-human adiponectin/Acrp30 polyclonal antibody (AF1065, RnD Systems) prior to the efferocytosis assay. In some experiments, apoptotic Jurkat T cells were labeled with CellTracker™ Green CMFDA (5-chloromethyl fluorescein diacetate) (Invitrogen). Quantifications were carried out using six microscopic fields per well, and efferocytosis capacity was expressed as a percentage of cells containing fluorescent-labeled apoptotic bodies over the total number of MH-S cells in the field.
In vivo efferocytosis assay was performed as described earlier [
25]. Briefly, Phrodo™ Red-labeled early apoptotic Jurkat T cells were instilled into WT and
ISM1−/− mouse lungs via intratracheal delivery. BALF was collected 90 min after the mouse had woken up from anesthesia while dosing with the apoptotic cells. The macrophage population in the BALF was stained with FITC-conjugated CD68 antibody (Clone FA-11, Biolegend). Efferocytosis capacity was assessed by flow cytometry, where CD68-Phrodo double-positive cells were identified as phagocytotic macrophages.
Statistical analysis
Results are presented as the mean ± SD. Data were analyzed with Student’s t test or one-way ANOVA, followed by a post hoc multiple comparison test with GraphPad Prism version 9 software. Error bars represent the standard deviation of the mean. A P-value < 0.05 was considered statistically significant. The levels of statistical significance were denoted at *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, and n.s. non-significant.
Discussion
We demonstrated here for the first time the role of ISM1 in regulating airway inflammation in response to aeroallergens. ISM1 deficiency exacerbated HDM-induced AHR in mice, while intratracheal administration of rISM1 suppressed HDM-induced allergic-like airway/lung inflammation (Figs.
1,
2,
3 and
7). Moreover, we showed that
ISM1−/− lungs harbored significantly lower adiponectin, while rISM1 dose-dependently stimulated adiponectin secretion in cultured AT2 alveolar epithelial cells (Figs.
5 and
6). We demonstrated that ISM1-induced adiponectin acts as an opsonizing agent in enhancing AM efferocytosis of apoptotic cells both in vitro and in vivo. Reduced AM efferocytosis under ISM1 deficiency likely led to increased secondary necroptosis, resulting in more severe inflammation in
ISM1−/− mice in HDM-induced allergic asthma. Indeed, necroptosis in
ISM1−/− lungs under HDM challenge was significantly elevated, as demonstrated by upregulated pMLKL and RIP3 levels in lung lysates and increased LDH release in BALF. Immunofluorescence staining further revealed that at least some AMs were undergoing necroptosis. Overall, our findings reveal a novel anti-inflammatory mechanism in airways mediated through the ISM1-adiponectin signaling axis.
Importantly, we found that
Ism1 mRNA is consistently downregulated in allergic airways from both mice and humans through in silico analysis (Fig.
1). This observation aligns with the work of Roffel et al., who identified higher expression of miR-223-3p in asthma patients. This miRNA was predicted to negatively regulate ISM1 expression (Supplementary Table
S3) [
52]. The downregulation of ISM1 expression upon HDM sensitization shown by qRT-PCR in the present study validated the RNA sequencing data. Interestingly, in contrast to asthma conditions, our earlier study found that ISM1 expression was upregulated in an LPS-induced acute lung injury mouse model [
53]. Similarly, ISM1 levels increased in the mouse bronchial epithelium following cigarette smoke exposure [
13]. Notably, both LPS and cigarette smoke induce a T
H1-biased inflammatory response primarily through the activation of TLR-4 and downstream NFκB signaling pathways. A recent study by Rivera-Torruco et al. reported an increase in ISM1
+ lung hematopoietic progenitor cells during
P. aeruginosa infection in mice, which is also a T
H1-type inflammatory response [
54]. In contrast, HDM sensitization is associated with a T
H2-type inflammatory response. These data together suggest that different inflammatory disease models may impact ISM1 expression or ISM1 + cells differently.
We wish to point out that there was no significant increase in immune cell numbers in
ISM1−/− mice under saline challenge (Fig.
2C), although a less pronounced version of this phenotype was observed in a different set of experiments (Fig.
7B). In our initial observations, we reported that naïve
ISM1−/− lungs presented spontaneous low-grade inflammation with higher leukocyte numbers than WT lungs [
8]. In addition to the possibility that saline instillation into the airway triggers a certain level of immune response, we would like to emphasize that the phenotype of higher leukocyte number appeared to be weaker in
ISM1−/− mice generated from the C57BL/6J background than in those generated from the FVB/N background [
13]. Additionally, it is crucial to acknowledge that phenotypes can diminish over time due to possible adaptation through prolonged and repeated breeding in mice. Nevertheless, we demonstrated that ISM1-deficient mice displayed an enhanced inflammatory response following HDM challenge (Fig.
2). These results align with our earlier report indicating that ISM1 deficiency intensified acute lung inflammation in response to LPS challenge, and the inflammatory phenotypes were reversed by exogenous supplementation of rISM1 [
8]. Collectively, these findings suggest that ISM1 may be critical in limiting airway inflammation in asthma.
Compared with WT mice,
ISM1−/− mice exhibited higher levels of serum IgE upon HDM challenge but similar levels of IL-4, IL-5, and IL-13 (Fig.
2B A). These discrepancies could be due to the T
H1 leaning immune responses of the C57BL/6 strain as compared to BALB/c, a more commonly used strain for asthma studies with T
H2-prone immune responses [
55,
56]. In addition, the time of cytokine detection could also be a contributing factor. It has been shown that T
H2 cytokines peak at 2 weeks post-HDM challenge and subside from 3 weeks onward, although AHR persists [
57]. Furthermore, the production and regulation of IgE antibodies involve multiple factors and complex interactions involving various cytokines and cellular signaling pathways. According to a previous study, while IgE synthesis exhibited a positive correlation with the concentration of IL-4 and IL-13, a robust inverse correlation was also observed between the amount of IgE produced and the concentration of IFNγ, and a potent IgE response can be induced with relatively low levels of IL-4 and IL-13, provided that the levels of IFNγ are also low [
58]. Our results align with these findings, where the IFNγ level was significantly lower in HDM-sensitized
ISM1−/− mice (Fig.
3B). Moreover, TGFβ
1, whose cleaved active form was found to be highly expressed in
ISM1−/− mice upon HDM challenge, could also be a potential contributing factor (Fig.
2I). TGFβ has been demonstrated to inhibit T
H2 development by suppressing IL-4 production [
59]. Meanwhile, IL-21 is also known to be a negative regulator of IgE class switch recombination in the geminal center [
60]. A higher level of IL-21 in BALF under HDM challenge in
ISM1−/− mice, while serum IgE is higher, suggests that other regulatory factors may also be involved. Nevertheless, further investigation is necessary to understand the specific role that ISM1 plays in regulating the adaptive immune response.
Adiponectin is an anti-inflammatory adipokine mainly secreted in adipocytes but is also expressed in the lungs. Studies have shown that low adiponectin levels are associated with type 2 diabetes and obesity [
61,
62]. Obesity is an important comorbidity in asthmatic patients [
63]. In obese-asthma patients, macrophage efferocytosis was 40% lower than that of non-obese subjects [
64]. Moreover, macrophages derived from diet-induced obese mice presented impaired efferocytosis [
65,
66]. Similarly, macrophages from diabetic mice also showed significant impairment in efferocytosis [
67]. Furthermore, obese HDM mice have been reported to present reduced plasma adiponectin and higher macrophage infiltration into the lungs and BALF [
68]. ISM1 was recently identified as an adipokine secreted by adipocytes, that promotes glucose uptake in an insulin-like fashion [
36]. However, contrary to
ISM1−/− lungs, adiponectin expression was shown to have no significant alteration in
ISM1−/− adipocytes [
36]. We demonstrate here that
Adipoq mRNA, which encodes for the adiponectin protein, was significantly downregulated in
ISM1−/− lungs (Figs.
5 and
6). ISM1 dose-dependently stimulated adiponectin secretion in lung AT2 cells and promoted AM efferocytosis (Fig.
7). Consistently, adiponectin-deficient mice reported an impaired clearance of apoptotic cells, while adiponectin supplementation showed a reversal effect [
23]. Adiponectin-deficient mice presented heightened allergic airway inflammation and spontaneous emphysema, phenotypes similar to those of ISM1-deficient mice [
19,
38]. Systemic adiponectin application has been reported to attenuate allergen-induced airway inflammation and hyperresponsiveness in mice [
18]. Hence, ISM1 may suppress allergic airway inflammation by stimulating lung adiponectin production and promoting adiponectin-mediated AM efferocytosis. Note that our previous report suggested no obvious alteration in efferocytosis of AMs obtained from
ISM1−/− mice in high-serum media [
13]. We speculate that the high serum culture media used might have masked the efferocytosis deficiency since a high amount of adiponectin is known to be present in serum [
69].
In eosinophil-dominant type-2 allergic asthma, apoptotic eosinophils are cleared by AMs through efferocytosis [
70]. Clearance of apoptotic cells by lung AMs has been reported to prevent the development of HDM-induced allergic airway inflammation [
45]. It has also been reported that efferocytosis is impaired in non-eosinophilic asthma [
71]. Delayed clearance of apoptotic cells in
ISM1−/− mice could trigger secondary necroptosis, leading to heightened airway inflammation. Meanwhile, our RNASeq also identified that genes associated with damage-associated molecular patterns (DAMPs) such as S100a8 and S100a9, were highly upregulated (Table
1). Upregulation of DAMPs also suggests the involvement of increased necroptosis in inflammation [
72]. These findings together support a mechanism whereby ISM1 restrains airway inflammation and hyperresponsiveness by promoting adiponectin production in the airways. Nevertheless, the possibility that ISM1 also reduces eosinophil infiltration into the airways cannot be excluded.
Alveolar epithelial cells are composed of two major cell types, alveolar type 1 (AT1) and alveolar type 2 (AT2) cells, which cover 99% of the internal surface of the lungs. Besides functioning as a barrier separating the internal and external environments, epithelial cells are also immunologically active which sense changes in the airway environment and interact with immune cells [
73]. AT2 cells are responsible for repairing processes and modulating immune responses upon lung injury [
74]. AT2 cell apoptosis has been implicated in the pathogenesis of idiopathic pulmonary fibrosis [
75]. In asthma, AT2 cells protect the lungs against allergen-induced airway inflammation by secreting TGFβ
1, which stimulates regulatory T cell (Treg) development [
76]. The Treg-secreted anti-inflammatory cytokine IL-10 is involved in the suppression of T
H17-mediated inflammation [
77]. Indeed, lung tissue from
ISM1−/− mice produced significantly less IL-10 than that of WT mice under HDM sensitization. Furthermore, while HDM treatment markedly reduced BALF IL-21 in WT mice, ISM1 deficient mice maintained a relatively high level of BALF IL-21 post-HDM sensitization (Fig.
3). Previous studies have reported that IL-21 and its receptor are necessary for the development of eosinophilic airway inflammation [
78,
79]. In human asthma, elevated IL-21 expression correlates with disease severity [
79]. IL-21 has been reported to have an immunoregulatory function via its ability to induce IL-10 production [
80] and directly induce Treg apoptosis [
30]. Nevertheless, there was no increase in IL-21 in
ISM1−/− lungs after HDM challenge compared to saline treated
ISM1−/− lungs, hence IL-21 could not be responsible for the reduced IL-10 in
ISM1−/− lungs under HDM sensitization. How ISM1 deficiency leads to a reduction in lung IL-10 requires further investigation. These findings together suggest that ISM1 could exert its anti-inflammatory function in the airways by promoting the production of anti-inflammatory cytokines such as adiponectin and IL-10.
AMs are the major immune cells in the alveolar space, and they often adhere to the alveolar epithelium and play a predominantly immune-suppressive role in maintaining homeostasis. Lung epithelial cells are known to secrete anti-inflammatory proteins such as CD200 to maintain AM in a quiescent state and limit inflammatory amplitude [
81]. In conjunction with our previous report that
ISM1−/− lungs harbor more activated pro-inflammatory AMs and trigger spontaneous emphysema [
8,
13], our findings here showed that HDM challenge generated a more severe allergic airway inflammation with pre-existing emphysema.
ISM1−/− mice may thus be useful in studying asthma-COPD overlap syndrome, whose molecular mechanisms remain poorly studied [
82,
83]. Future studies using lung-specific
ISM1−/− mouse lines would help to further clarify the roles of ISM1 in the lungs.