A SPLUNC1 Peptidomimetic Inhibits Orai1 and Reduces Inflammation in a Murine Allergic Asthma Model

Abstract

Orai1 is a plasma membrane Ca²⁺ channel that mediates store-operated Ca²⁺ entry (SOCE) and regulates inflammation. Short palate lung and nasal epithelial clone 1 (SPLUNC1) is an asthma gene modifier that inhibits Orai1 and SOCE via its C-terminal α6 region. SPLUNC1 levels are diminished in asthma patient airways. Thus, we hypothesized that inhaled α6 peptidomimetics could inhibit Orai1 and reduce airway inflammation in a murine asthma model. To evaluate α6–Orai1 interactions, we used fluorescent assays to measure Ca²⁺ signaling, Förster resonance energy transfer, fluorescent recovery after photobleaching, immunostaining, total internal reflection microscopy, and Western blotting. To test whether α6 peptidomimetics inhibited SOCE and decreased inflammation in vivo, wild-type and SPLUNC1^−/− mice were exposed to house dust mite (HDM) extract with or without α6 peptide. We also performed nebulization, jet milling, and scanning electron microscopy to evaluate α6 for inhalation. SPLUNC1^−/− mice had an exaggerated response to HDM. In BAL-derived immune cells, Orai1 levels increased after HDM exposure in SPLUNC1^−/− but not wild-type mice. Inhaled α6 reduced Orai1 levels in mice regardless of genotype. In HDM-exposed mice, α6 dose-dependently reduced eosinophilia and neutrophilia. In vitro, α6 inhibited SOCE in multiple immune cell types, and α6 could be nebulized or jet milled without loss of function. These data suggest that α6 peptidomimetics may be a novel, effective antiinflammatory therapy for patients with asthma.

Clinical Relevance

Nonsteroidal alternatives are needed for the treatment of asthma. Here, we show that addition of a SPLUNC1 α6 peptidomimetic reduces inflammation in a murine house dust mite asthma model, suggesting that lack of SPLUNC1 is proinflammatory and that the α6 peptide may serve as a novel therapeutic for the treatment of asthma.

Asthma is a common chronic disease affecting more than 300 million people worldwide (1). Asthma is typified by chronic airway inflammation, bronchoconstriction, mucus accumulation, and airway hyperresponsiveness (2). In allergic asthma, inflammation is initiated by type 2 helper T lymphocytes (Th2), which release proinflammatory mediators such as IL-13. This leads to inflammatory infiltrates composed largely of eosinophils, but also containing mast cells, neutrophils and other immune cells, which release proinflammatory mediators that promote epithelial injury and contraction of airway smooth muscle (3–6). Reduction of airway inflammation is therefore vital for the effective treatment of asthma patients. Currently, asthma is managed with inhaled corticosteroids, long-acting β₂-adrenergic receptor agonists, muscarinic antagonists, and leukotriene receptor antagonists. However, approximately 5–10% of patients with asthma do not respond well to corticosteroids, and their symptoms can be life threatening, suggesting that new asthma therapies are required (7–9).

Orai1 is a plasma membrane Ca²⁺ channel that is involved in store-operated Ca²⁺ entry (SOCE), a fundamental process that is typically activated by G protein-coupled receptors (10). During SOCE, Ca²⁺ leaves the endoplasmic reticulum (ER), which then stimulates a second, amplifying wave of Ca²⁺ influx through Orai1 (11, 12). SOCE and Orai1 are upstream of transcription factors such as nuclear factor of activated T cells, which facilitates the onset of inflammation. Increases in cytoplasmic Ca²⁺ also stimulate mucin, protease, and cytokine secretion (13). Thus, inhibition of Orai1 is an attractive target for novel antiinflammatory therapies owing to its proximal role in multiple convergent proinflammatory pathways (14).

Short palate lung and nasal epithelial clone 1 (SPLUNC1, gene name BPIFA1) is a multifunctional protein that is expressed in airway epithelia and secreted into the lung lumen. SPLUNC1 regulates airway surface liquid hydration (15), smooth muscle contraction (16) and has antimicrobial properties (17). SPLUNC1 is also generating interest for its antiinflammatory properties (18–20) and has been identified as a gene modifier for cystic fibrosis (21) and asthma (22). Additionally, patients with cystic fibrosis with low sputum SPLUNC1 concentrations experience pulmonary exacerbations more frequently and have higher sputum IL-1β levels than patients with higher SPLUNC1 levels (23). SPLUNC1 expression is negatively regulated by IL-13, and sputum SPLUNC1 levels are reduced in patients with asthma (16, 24). Moreover, SPLUNC1 deficiency promoted allergic or eosinophilic inflammation in mice (19), indicating that SPLUNC1 may play a role in modulating the Th2 immune response of the lung.

Our published data indicate that SPLUNC1, via its α6 region, regulates Orai1 and Ca²⁺ signaling (16). For example, SPLUNC1-containing secretions from healthy primary human bronchial epithelial cultures (HBECs) were capable of inhibiting SOCE, but those from donors with asthma contained diminished SPLUNC1 concentrations and did not regulate SOCE (16). As SPLUNC1 concentrations are diminished in patients with asthma (16), we hypothesized that a peptidomimetic of SPLUNC1’s α6 region would inhibit Orai1 to reduce pulmonary inflammation. Here, we investigated the effect of α6 treatment on inflammation in a murine model of allergic asthma and studied α6’s ability to inhibit Orai1 in vitro.

Methods

Full methods are available in the data supplement.

Mouse Husbandry

Wild-type (WT) and SPLUNC1^−/− mice were exposed to house dust mite extract (HDM), and BAL was obtained for cell counting as described (25, 26). In all experiments, equal numbers of male and female mice were used. All procedures were approved by University of North Carolina-Chapel Hill’s Institutional Animal Care and Use Committee and performed according to the principles outlined by the National Institutes of Health guidelines for the care and use of animals in biomedical research.

α6 Peptide

The α6 peptide (DITLVHDIVNMLIHG) was made by Fmoc solid-state phase peptide synthesis at University of North Carolina and by Shanghai Royobiotech. Where noted, TAMRA was conjugated to its amino terminus. Powdered peptides were stored at −80°C, and stocks were made up fresh in DMSO or saline as needed.

BAL Cell Counts and Immunofluorescence

Collected BAL samples were affixed to slides using a Rotofix 32 A cytology centrifuge (Hettich). BAL was fixed in 4% paraformaldehyde. For cell counts, samples were stained with Kwik Diff (ThermoFisher) per manufacturer’s instructions. Individual immune cell populations were determined microscopically and were counted manually. Monocytes and macrophages could not be distinguished and are combined as “macrophages.” For immunofluorescence, samples were permeabilized in 0.1% triton-X100 in PBS, blocked overnight in blocking solution (PBS + 0.1% triton-X100 + 10% FBS + 5% normal goat serum). Cells were incubated in rabbit anti-Orai1 primary antibody (Sigma Aldrich) overnight at 4°C, washed in PBS, and stained for 3 hours at 4°C in 1:3000 goat antirabbit IgG Dylight-633 secondary antibody (ThermoFisher Scientific). Cells were mounted in Vectashield with DAPI (Vector Labs) and imaged on a Leica SP8 confocal microscope. Cell types were identified by brightfield microscopy.

Binding Assay

HEK293T cells were seeded at a density of 5,000 cells/well in a 96-well plate and cultured for 24 hours. Cells were exposed to a range of concentrations of α6-TAMRA for 1 hour before three washes in PBS. α6-TAMRA florescence intensity was then read using a multiplate reader.

Orai1 Imaging and Western Blotting

Fluorescence recovery after photobleaching (FRAP), Förster resonance energy transfer (FRET), and total internal reflection (TIRF) microscopy were performed as previously described (27). To measure Orai1 protein levels, HEK293T cells were transfected with Orai1-YFP and Western blotting was performed as described (16).

Confocal Microscopy and Deconvolution

HEK293T cells were transfected with Orai1-YFP. Cells were exposed to vehicle control (0.1% DMSO) or 10 μM α6-TAMRA for the times indicated. Cells were washed in PBS, fixed in 4% paraformaldehyde, immunostained as described above for RAB5 (rabbit anti-RAB5 [Cell Signaling, 2143S] and goat antirabbit IgG DyLight 633 [Invitrogen 35562]) and imaged using a Leica SP8 confocal microscope. Multiple images were taken in Z-stack, and resolution was enhanced to 35 nm²/pixel using Leica’s Lightning deconvolution software per the manufacturer’s instructions.

Quartz Crystal Microbalance with Dissipation Assay

A semipurified, native mucus preparation was placed on the quartz crystal microbalance with dissipation assay (QCM-D) as described (28), and then α6 and poly-L-lysine were added.

Statistical Analysis

Statistical analysis was performed using GraphPad Prism and Instat software using the tests indicated in figure legends. All experiments were repeated on at least three separate occasions. Curve fitting was performed in GraphPad Prism using y = 100/[1 + 10^(LogEC50-X)] for the dose–response and y = Bmax × X/(K_d + X) where Bmax is maximal binding capacity and K_d is the dissociation constant for peptide binding. Data were represented as mean ± SEM, and P < 0.05 was considered significant.

Results

SPLUNC1^−/− Mice Display Exaggerated Eosinophilic Inflammation in a Murine Model of Allergic Asthma

We investigated the role of SPLUNC1 in allergy-induced asthma by exposing WT and SPLUNC1^−/− mice to HDM. After sensitization and challenge with HDM, BAL was collected and stained, and cell counts were performed (Figure 1A). Consistent with an inflammatory response, HDM challenge increased BAL cell counts for both WT and SPLUNC1^−/− mice (Figure 1B). In the absence of HDM exposure, both WT and SPLUNC1^−/− BAL was comprised largely of macrophages (Figure 1C). HDM exposure induced a decrease in macrophage numbers in SPLUNC1^−/− but not WT mice (Figure 1C), as well as a significant increase in eosinophil and neutrophil counts in both WT and SPLUNC1^−/− BAL (Figures 1D and 1E). HDM exposure resulted in a significantly greater increase in eosinophil counts in SPLUNC1^−/− mice than in WT littermates, suggesting a hyperinflammatory response (Figure 1D).

Figure 1.

Short palate lung and nasal epithelial clone 1 (SPLUNC1^−/−) mice exhibit increased eosinophilic inflammation after house dust mite (HDM) exposure. (A) Representative images of BAL collected from wild-type (WT) and SPLUNC1^−/− mice in the absence (naive) and presence (HDM) of HDM treatment. Scale bar, 30 μm. (B) Total immune cell counts from collected BAL as well as (C) macrophage, (D) eosinophil, and (E) neutrophil counts. N = 15–17 mice per group. *P < 0.05, **P < 0.01, and ****P < 0.0001; one-way ANOVA with Tukey’s post-test.

Orai1 Expression Is Upregulated in BAL Immune Cells from HDM-exposed Mice and Can Be Rescued by Treatment with α6

Previously, we identified SPLUNC1 as an inhibitor of the Orai1 plasma membrane Ca²⁺ channel and SOCE, which was conferred by SPLUNC1’s sixth α-helical region (α6) (16). We and others have shown that compounds including peptides can be reliably delivered to murine lungs via the nose (29, 30). Accordingly, we generated a peptide based on this region and tested whether it could reduce Orai1 levels in HDM-exposed mice after intranasal delivery. Mice were sensitized to HDM and then challenged with HDM with or without two intranasal doses of 1.35 mg/kg α6 peptide, and BAL immune cells were fixed and stained for Orai1. HDM treatment resulted in a significant increase in Orai1 expression in SPLUNC1^−/− but not WT littermate controls (Figures 2A–2C). In both WT and SPLUNC1^−/− mice, intranasal α6 reduced Orai1 expression in HDM-treated BAL immune cells (Figures 2A–2C). These data indicate that Orai1 expression in immune cells may play a role in the sensitization of inflammatory responses to allergens under chronic exposure conditions and that this effect may be regulated by SPLUNC1.

Figure 2.

HDM-induced upregulation of Orai1 in BAL immune cells is normalized by inhaled α6 peptide. (A) Representative confocal microscopy images of BAL macrophages from wild-type and SPLUNC1^−/− mice with or without HDM treatment with or without α6 treatment stained for Orai1 (red) and DAPI (blue). Scale bar, 10 μm. Orai1 fluorescence intensity from BAL from each treatment group for (B) wild-type and (C) SPLUNC1^−/− mice. Data points are values from regions of interest encompassing multiple cells. N = 6–9 mice per group. **P < 0.01 and ***P < 0.001; one-way ANOVA with Tukey’s post-test.

The α6 Peptide Inhibits Store-operated Ca²⁺ Entry in a Range of Cell Types

Orai1 is widely expressed throughout the body, including in immune cells (31, 32). To see if α6 could inhibit Ca²⁺ signaling in human immune cells, we stimulated SOCE with thapsigargin (1 μM) in the continuous presence of extracellular Ca²⁺ in a range of cell types pretreated with α6 (100 μM; 3 h) or vehicle control. Cytosolic [Ca²⁺] was measured using the fluorescent Ca²⁺ indicator, fluo-4-AM. Inhibitory effects of α6 were observed in Jurkat T cells, as well as in alveolar macrophages, peripheral blood-derived neutrophils and T cells, and umbilical cord-derived mast cells (Figure 3A). HEK293T cells have been extensively used as a model to study Orai1 and SOCE (33). We therefore used this cell type to specifically evaluate α6’s inhibitory effects on SOCE. Cells were pretreated with a range of α6 concentrations or vehicle (1% DMSO for 3 hours). Thapsigargin (100 nM) was added in the absence of extracellular Ca²⁺ to stimulate Ca²⁺ emptying from the ER (see Methods) (34). After ER emptying, extracellular Ca²⁺ was returned and SOCE measured. The α6 peptide had no effect on ER Ca²⁺ release, but reduced SOCE by approximately 50% at 100 μM concentration (Figure 3B). We then generated a dose–response curve from our data (Figure 3C). α6 displayed an IC₅₀ of 142 nM.

Figure 3.

α6 inhibits Ca²⁺ signaling. (A) Thapsigargin (Tg; 1 μM)-induced Ca²⁺ response in a range of immortalized and human primary cell types with or without α6 (100 μM; 3 h) in the continuous presence of extracellular Ca²⁺. Data are normalized to vehicle control for each cell type. N = 6–11 per cell type. (B) Mean change in fluo-4 over time as an indicator of cytosolic [Ca²⁺]. HEK293T cells were pretreated with α6 (100 μM; 3 h) or vehicle and then stimulated with 100 nM thapsigargin in the absence of extracellular Ca²⁺. After the first peak, 1.2 mM Ca²⁺ was added extracellularly to stimulate Ca²⁺ influx. N = 12 wells from 3 plates per data point. (C) α6 dose–response curve of peak store-operated Ca²⁺ entry measured by fluo-4. Data are normalized to vehicle control. IC₅₀ = 142 nM; R² = 0.234. N = 9–21 wells from 3–6 plates. Blue-lined circles = vehicle; red-lined squares = α6 peptide. *P < 0.05, **P < 0.01, ***P = 0.001, and ****P < 0.0001; Student’s t test. IC₅₀ = half-maximal inhibitory concentration; veh = vehicle.

α6 Is an Allosteric Regulator of Orai1

To probe α6 interactions, we created a fluorescently tagged α6 (α6-TAMRA). We measured α6 binding kinetics by exposing cells to a range of α6-TAMRA concentrations for 1 hour before washing off unbound peptide and measuring fluorescence intensity (Figure 4A). α6-TAMRA bound extracellularly with a K_d of 2.21 μM. We have previously shown that SPLUNC1’s N-terminus binds extracellularly to epithelial sodium channel (ENaC) and induces conformational changes to inhibit the channel (16). To see if α6 similarly induces allosteric changes in Orai1, we used FRET to investigate the possibility of α6-induced conformational changes in Orai1. Extracellular α6 (30 μM, 10 min) resulted in a significant increase in intracellular CFP-Orai1-YFP-Orai1 FRET, indicating that α6 induced allosteric changes in Orai1 (Figure 4B). Next, we exposed HEK293T cells to 10 μM α6-TAMRA for a range of incubation times, washed off unbound peptides, and measured intracellular fluorescence by confocal microscopy. Intracellular α6-TAMRA fluorescence increased rapidly after its addition to cells (Figure 4C). To investigate whether internalization was mediated by Orai1, intracellular accumulation was compared between cells with endogenous Orai1 expression and those transfected with Orai1-YFP. Overexpressing cells internalized α6-TAMRA to a greater extent than cells with endogenous expression (Figure 4C). Superresolution confocal microscopy of Orai1-YFP-transfected HEK293T cells revealed that α6-TAMRA binds to and internalizes with Orai1 into RAB5-containing vesicles (Figure 4D). To investigate the effect of α6 on plasma membrane Orai1-YFP densities, TIRF microscopy was used. A significant reduction in plasma membrane Orai1 was observed within 1 hour of exposure to 30 μM α6, with an almost complete depletion of Orai1-YFP from the plasma membrane by 3 hours. Consistent with these data, shorter incubations with α6 did not inhibit SOCE (n = 16; P < 0.999). Moreover, α6 did not reduce the plasma membrane expression of ANO1-mCherry, suggesting that this effect may be specific to Orai1 (Figure 4E). Western blot analysis of whole cell lysates from vehicle (1% DMSO, 3 h) and α6 (30 μM, 3 h)-exposed HEK293T cells transfected with Orai1-YFP revealed that α6 treatment resulted in a 20% reduction in Orai1 protein concentrations (Figure 4F), indicating that Orai1 internalized by α6 is degraded. After ER Ca²⁺ depletion, STIM1 relocates to ER–plasma membrane junctions to interact with and activate Orai1 (35). To investigate whether α6 binding prevented the colocalization of STIM1-mCherry and Orai1-YFP, we measured their colocalization by confocal microscopy in the presence and absence of thapsigargin and/or α6 treatment. In vehicle-treated cells, thapsigargin caused a significant increase in Orai1 and STIM1 colocalization (Figures 4G and 4H). This effect was significantly attenuated by α6 pretreatment (30 μM, 3 h). Similarly, FRAP of Orai1-YFP revealed that thapsigargin treatment reduced Orai1 plasma membrane motility (Figure 4I), most likely owing to increased interactions with ER-bound STIM1. This effect was also significantly blocked by α6 pretreatment (30 μM, 1 h). Representative FRAP, FRET, and TIRF images and the full Western blots are available in the data supplement (Figures E1–E4).

Figure 4.

α6 is an allosteric regulator that induces Orai1 internalization and degradation. (A) α6-TAMRA binding (1 h incubation) at a range of concentrations. All N = 8. R² = 0.73; K_d = 2.21 μM. (B) Förster resonance energy transfer (FRET) efficiency of Orai1-YFP and Orai1-CFP interactions in HEK293T cells with or without α6 (30 μM; 3 h). N = 15 cells from 3 coverslips per group. Representative images are provided in Figure E1 in the data supplement. (C) HEK293Tcells (blue solid circles) or HEK293T cells expressing Orai1-YFP (red solid circles) were exposed to 30 μM α6-TAMRA over time, and intracellular fluorescence normalized to DAPI fluorescence was measured. N = 15 from 3 coverslips per group. (D) Representative image of Orai1-YFP (green), RAB5 (blue), and vehicle or α6-TAMRA (10 μM; 10 min; red). Scale bar, 2 μm. (E) Total internal reflection (TIRF) microscopy images of plasma membrane Orai1-YFP (red) and Anoctamin-1 (ANO1) (blue) in HEK293T cells exposed to α6 over time. N = 24–30 cells per group from 3 coverslips. Representative images provided in Figure E2. (F) Western blots of whole-cell lysate from HEK293T cells transfected with Orai1-YFP with or without α6 (30 μM; 3 h), and graph showing paired Orai1-YFP/GAPDH densitometry. N = 6 blots. (G) Representative images of Orai1-YFP (green) and Stromal interaction molecule 1 (STIM1)-mCherry (red). Scale bars, 10 μm. (H) Percent colocalization between Orai1-YFP and STIM1-mCherry in transfected HEK293T cells in the absence (basal) and presence (Tg) of thapsigargin (2 μM; 10 min) with or without α6 (30 μM, 3 h). N = 15 cells from 3 coverslips. (I) Fluorescence recovery after photobleaching of Orai1-YFP in HEK293T cells with or without thapsigargin (1 μM) with or without α6 (30 μM, 1 h). N = 30–41 cells from 9 coverslips. Representative images are in Figure E2. *P < 0.05, **P = 0.01, ***P < 0.001, and ****P < 0.0001; (B and F) Student’s t test; (C) two-way ANOVA; (E, H, and I) one-way ANOVA with Tukey’s post-test. CFP = cyan fluorescent proteins; TAMRA = tetramethylrhodamine; YFP = yellow fluorescent proteins.

α6 Reduces Eosinophilic Inflammation in HDM-exposed Mice

Since SPLUNC1^−/− mice displayed exaggerated inflammatory responses to HDM exposure (Figure 1), and α6 reduced Orai1 expression in BAL immune cells (Figure 2), we next evaluated the ability of α6 to reduce airway inflammation following an HDM challenge. After HDM sensitization, intranasal administration of α6 dose-dependently reduced BAL eosinophil and neutrophil levels, without affecting macrophage levels. Indeed, at higher concentrations, α6 treatment returned each population of immune cells to concentrations seen in naive mice (Figures 5A–5E).

Figure 5.

α6 dose-dependently reduces pulmonary inflammation in HDM-exposed mice. (A) Representative images of collected BAL. Scale bar, 100 μm. (B) Total immune cell counts (IC₅₀ = 0.2617 mg/kg; R² = 0.19), (C) macrophage counts, (D) eosinophil counts (IC₅₀ = 0.012 mg/kg; R² = 0.16), and (E) neutrophil counts (IC₅₀ = 0.19 mg/kg; R² = 0.18) are shown. N = 6–15 mice per group.

α6 Does Not Interact with Mucus or Cross the Epithelial Barrier and Remains Effective after Jet Milling and Nebulization

Increased mucus secretion is a feature of the asthmatic lung, and mucus and/or mucins have been shown to interact with a number of proteins and peptides (36) that may impede their delivery via inhalation. We therefore evaluated whether α6 could interact with mucus. We have previously used QCM-D to measure mucus–peptide interactions (28). Although the positive control, poly-L-lysine, induced changes in frequency and dissipation that were indicative of increased mucin cross-linking and decreased hydration, respectively, QCM-D analysis demonstrated that α6 does not interact with mucus (Figure 6A). As Orai1 is ubiquitously expressed, we then investigated whether α6-TAMRA could cross the epithelial barrier in HBECs. After 3 hours of exposure, α6-TAMRA remained in HBEC’s mucosal compartment and was not detected serosally indicating that it does not cross the epithelial barrier (Figure 6B).

Figure 6.

α6 does not interact with mucus and can be formulated for inhalation. (A) Quartz crystal microbalance with dissipation traces showing the change in mucus rheology after addition of α6, followed by poly-L-lysine (positive control). Frequency shift (blue) and dissipation shift (red). (B) Three-dimensional rendering of human bronchial epithelial cultures labeled with calcein-AM (green) with or without α6-TAMRA (10 μM, 3 h, red). (C) Scanning electron micrograph of jet-milled α6 showing consistent particle size of approximately 5 μm. (D) Peak thapsigargin-induced Ca²⁺ signaling in HEK293T cells treated with unprocessed, jet-milled, or nebulized α6, or vehicle for 1 hour. Scale bars, 20 μm. Data are mean n = 3–6. ***P < 0.001 and ****P < 0.0001; one-way ANOVA with Tukey’s post-test. AM = acetoxymethyl ester.

Finally, we investigated the suitability of using α6 as an inhaled therapy for asthma. Dry powered inhalers are commonly used for delivery of therapeutics for obstructive airway diseases, and thus we investigated the suitability of α6 for use in these devices. Dry powder inhalers require particle small and consistent particle sizes for effective delivery to the lower airways, which is commonly achieved via jet milling. Scanning electron microscopy revealed that jet-milled α6 displayed a consistent particle size that was suitable for inhalation to the lower airways (1–5 μm) (Figure 6C), making it a good candidate for dry powder inhalation (37). Using the cytosolic Ca²⁺ indicator fluo-4, the efficacy of α6 as an inhibitor of thapsigargin-induced Ca²⁺ signaling was assessed after jet milling (required for dry power inhalers) and nebulization. Neither preparation had a significant effect on α6 activity (Figure 6D).

Discussion

SPLUNC1 expression is decreased by Th2 cytokines such as IL-13 in normal primary human airway epithelia, is spontaneously diminished in HBECs derived from patients with asthma, even after a month in culture, and is reduced in sputum of patients with asthma (16). Recently, SPLUNC1 was shown to be an asthma gene modifier (22). Indeed, patients with asthma with the CC allele had elevated serum IgE levels and higher exhaled NO levels (22) compared with patients with CT or TT alleles. These researchers also found that IL-13–induced CCL26 (eotaxin-3) concentrations were higher in nasal epithelial cultures derived from patients with asthma expressing the SPLUNC1 CC allele than those with other alleles. These data are consistent with SPLUNC1^−/− mice studies, in which the lack of SPLUNC1 resulted in exaggerated eosinophilic inflammation in ovalbumin-challenged mice (19) and with our finding that HDM challenge causes more eosinophilic inflammation in SPLUNC1^−/− mice (Figure 1).

The mechanism(s) underlying altered SPLUNC1 levels, allergic mouse phenotypes, and disease severity in patients with asthma are not well understood. However, our previous observation that SPLUNC1 inhibits the Orai1 Ca²⁺ channel may shed light on this matter (16). Orai1 is an essential component of SOCE in multiple cell types (38), and appropriate changes in SOCE are needed to mount an efficient inflammatory response and for naive T cells to become activated. Too little SOCE is immunosuppressive, and too much SOCE is proinflammatory (38). SOCE helps control gene expression, the secretion of macromolecules, including cytokines, and the contraction of smooth muscle (39, 40). Polymorphisms in Orai1 have recently been shown to be associated with allergic dermatitis disease severity (41). Of note, Th2 cytokines such as IL-13 are significantly involved in both allergic dermatitis and asthma, suggesting that Orai1 function may also influence asthma disease severity. Despite this wealth of knowledge, beyond its role in airway smooth muscle, Orai1 expression and function in the lung is poorly understood. However, our working hypothesis is that SPLUNC1 is a negative regulator that acts as a brake on Orai1 activity to prevent excessive inflammation. Consistent with this hypothesis, SPLUNC1^−/− mice exhibited increased Orai1 concentrations in BAL immune cells after HDM challenge, whereas the WT littermate controls did not show altered Orai1 levels after HDM exposure (Figure 2). Thus, we posit that increased Orai1 levels facilitated the increased inflammation seen in SPLUNC1^−/− mice by increasing Ca²⁺ influx. Importantly, these data suggest that SPLUNC1 replacement and/or Orai1 inhibition are attractive targets for the treatment of asthma.

We have previously demonstrated that SPLUNC1’s N-terminal S18 region regulates the epithelial Na⁺ channel ENaC, whereas the C-terminal α6 helix regulates Orai1 (16, 30, 42). Interestingly, the N-terminal S18 region is wholly absent from murine SPLUNC1 (42), suggesting that SPLUNC1’s effects in allergic mice models (Figure 1) are owing to changes in Orai1 and not to altered ENaC levels. We previously found that SPLUNC1’s ability to inhibit Orai1 was fully abrogated by deletion of SPLUNC1’s α6 helix (16, 42). Thus, we hypothesized that an α6 helix peptidomimetic would be a viable therapy for treating allergic inflammation in the lung. Indeed, α6 had a similar IC₅₀ for inhibiting Ca²⁺ signaling as SPLUNC1 (142 nM [Figure 3B] vs. ∼100 nM [16], respectively), and both directly affected Orai1 (Figure 4 and [16]). Small peptides can be readily made by solid-state synthesis and are often regulated as small molecules rather than biologics by the U.S. Food and Drug Administration. Additionally, many small peptides are stable, do not need to be refrigerated, and are often too small to induce an immunogenic response (43). Moreover, as α6 is a mimetic of a protein that is naturally expressed, it would be predicted to be well tolerated. Indeed, we have previously administered a peptide based on SPLUNC1’s N-terminus to normal healthy volunteers and patients with cystic fibrosis without any adverse effects (44), suggesting that α6 peptidomimetics may be safe to administer to patients with asthma after the appropriate U.S. Food and Drug Administration-approved toxicology studies.

Consistent with Orai1 being ubiquitously expressed, α6 inhibited the thapsigargin-induced Ca²⁺ response in human peripheral blood-derived T cells, neutrophils, umbilical cord-derived mast cells, and alveolar macrophages to a similar extent as seen in HEK293T and Jurkat cells (Figure 3). We also found that intranasal delivery of α6 reduced eosinophil and neutrophil numbers in a dose-dependent manner in mice (Figure 5). It is important to note that this decrease occurred 24 hours after two doses of α6, suggesting that α6 has a long duration of action. Sharma and McNeill stated that “there is no consensus on the best method for selecting a first dose in humans” (45). However, using their allometric model (multiply by 0.081 to convert from mouse to human) and based on the murine dose of 0.1 mg/kg, a 70-kg human would require approximately 0.56 mg of inhaled α6. This is well within the output range of both metered dose inhalers and dry powder inhalers, suggesting that the doses of α6 used to reduce inflammation in HDM-exposed mice are therapeutically relevant.

Mucin hypersecretion is a common phenotype in patients with asthma, and whether or not α6 interacts with mucus could affect its delivery to the lung. Positively charged peptides and proteins can interact with mucus, and our positively charged control peptide, poly-L-lysine, interacted with mucus, causing changes in QCM-D frequency and dissipation that were indicative of increased cross-linking and altered hydration, respectively. In contrast, α6 did not interact with mucus (Figure 6A), suggesting that it will be able to diffuse through increased mucus layers after inhalation to reach its target cells. We fluorescently labeled α6 with TAMRA and used this to probe its permeability across epithelia. Indeed, all α6-TAMRA was retained apically and did not cross the epithelial barrier (Figure 6B), suggesting that it will have minimal systemic exposure. Similarly, we have previously found that a peptidomimetic based on SPLUNC1’s S18 region also did not cross the epithelial barrier (28). However, unlike S18, which did not internalize, we observed that α6-TAMRA accumulated intracellularly in HEK293T cells after binding to and internalizing cell surface Orai1 (Figure 4). This intracellular accumulation was not observed in HBECs (Figure 6B). Orai1 expression in the airway remains poorly characterized; however, receptor-mediated Ca²⁺ entry in HBECs has been observed to occur exclusively through the basolateral membrane (46), supporting the hypothesis that Orai1 expression is polarized in these cells and diminished at the plasma membrane. We hypothesize that the absence of α6 binding and internalization in HBEC cultures is owing to the absence of Orai1 expression in the apical membrane of airway epithelia. Intriguingly, our observation that α6 neither interacts with nor penetrates the epithelial barrier suggests that α6 imparts its antiinflammatory effects solely through interactions with immune cells already recruited to the airway lumen. We hypothesize that α6 may be reducing Ca²⁺-dependent cytokine release from these cells and/or reducing the residence time of eosinophils in the airway lumen. However, further mechanistic studies to fully understand the extent of α6 penetration into the lung, and its target cell types, will be required.

In cardiac myocytes, atrial natriuretic peptide has previously been shown to regulate L-type Ca²⁺ channels (47), and exchanger inhibitory peptide (XIP) regulated a Na⁺/Ca²⁺ exchanger (48). However, to the best of our knowledge, we are the first to identify a peptide that can regulate Orai1. Our previous studies indicated that SPLUNC1 only bound to Orai1-expressing airway smooth muscle (16). Similarly, we found that α6 bound extracellularly and cleared Orai1 from the plasma membrane (Figure 4). We expressed Orai1 constructs with C-terminally linked, cytoplasmic CFP and YFP and measured FRET. We found that extracellular α6 induced significant changes in intracellular FRET (Figure 4B), indicating that α6 induced allosteric changes in Orai1. After depletion of ER stores, STIM1 aggregates at the plasma membrane, where it colocalizes with and activates Orai1 (10). We found that α6 abolished this colocalization (Figure 4C) and also prevented the thapsigargin-induced decrease in Orai1 mobility (Figure 4D). Moreover, α6 decreased Orai1 at the plasma membrane and reduced total Orai1 protein levels (Figures 4F–4H). This allosteric internalization and degradation of Orai1 differentiates α6 from “traditional” small molecule antagonists such as CM4620, which occlude the channel pore (49), and suggests that α6 may have a long duration of action as the effects (i.e., degradation of Orai1) would persist after washout. Indeed, α6 reduced Orai1 levels and reduced inflammation 24 hours after α6 treatment (Figures 2 and 5).

Conclusions

We have generated a novel peptide that reduced Orai1 levels in the lung, leading to reduced inflammation in a murine model of asthma. SPLUNC1 replacement has been proposed as a therapeutic strategy for the treatment of patients with asthma (22). Protein-based therapies are a viable approach, but working with full-length proteins comes with various challenges, including synthesis and storage. However, peptidomimetics can perform similar functions as full-length proteins, and they can be manufactured in a straightforward and cost-effective manner using solid-state synthesis. Small molecules are often rapidly cleared from the lungs and can exert systemic, off-target effects. In contrast, aerosolized peptides often have extremely poor systemic bioavailability and hence minimal systemic effects (44, 50). This observation is pertinent to α6 because Orai1 is ubiquitously expressed, and systemic bioavailability may lead to significant side effects. Therefore, we propose that α6 will be therapeutically delivered by inhalation directly to the site of action and will likely be retained in the lung, minimizing systemic exposure.