Allergens stimulate store-operated calcium entry and cytokine production in airway epithelial cells

Aberrant immune responses to environmental allergens including insect allergens from house dust mites and cockroaches contribute to allergic inflammatory diseases such as asthma in susceptible individuals. Airway epithelial cells (AECs) play a critical role in this process by sensing the proteolytic activity of allergens via protease-activated receptors (PAR2) to initiate inflammatory and immune responses in the airway. Elevation of cytosolic Ca2+ is an important signaling event in this process, yet the fundamental mechanism by which allergens induce Ca2+ elevations in AECs remains poorly understood. Here we find that extracts from dust mite and cockroach induce sustained Ca2+ elevations in AECs through the activation of Ca2+ release-activated Ca2+ (CRAC) channels encoded by Orai1 and STIM1. CRAC channel activation occurs, at least in part, through allergen mediated stimulation of PAR2 receptors. The ensuing Ca2+ entry then activates NFAT/calcineurin signaling to induce transcriptional production of the proinflammatory cytokines IL-6 and IL-8. These findings highlight a key role for CRAC channels as regulators of allergen induced inflammatory responses in the airway.


Results
A screen of allergen extracts reveals insect allergens as activators of store-operated Ca 2+ entry in bronchial epithelial cells. Many studies have shown that allergen extracts from insects (HDM and cockroach) and fungi (Alternaria, Aspergillus) induce cytosolic Ca 2+ signals in AECs 8,13,22 . A multitude of factors including activation of PAR2 8 or other proteolytic receptors and ATP induction 23 have been implicated in the genesis of these Ca 2+ signals, yet it is not clear whether they primarily arise due to Ca 2+ release from internal stores or if they additionally involve Ca 2+ influx across the plasma membrane. We and others have previously shown that store-operated calcium entry (SOCE) is a major mechanism of Ca 2+ influx in bronchial epithelial cells and is stimulated by activation of PAR2 receptors 15,16 . However, whether allergens can activate CRAC channels in AEC is unknown. We therefore tested the ability of various allergen extracts to activate SOCE in bronchial BEAS-2B cells using fura-2 based Ca 2+ imaging (Table 1). Specific allergens including HDM, cockroach extracts, chitinase from Streptomyces griseus and fungal extracts from Alternaria and Aspergillus were applied in a Ca 2+free medium followed by re-addition of extracellular Ca 2+ to detect SOCE. This screen revealed that only a subset of allergens tested, limited to extracts from HDM and cockroach allergens, and, to a lesser extent, chitinase from Streptomyces griseus, activated store-operated Ca 2+ signals and Ca 2+ entry across the plasma membrane in the AECs (Table 1).

Cockroach extracts induce Ca 2+ signals in AEC by activating CRAC channels.
There is a strong correlation between sensitization and allergy to inhaled cockroach extracts and the incidence of acute asthmatic attacks 24,25 . In vitro and in vivo studies have shown that extracts from cockroach have proteinase activity and stimulate PAR2 receptors to mediate their inflammatory effects 9,10 . Induction of cytosolic Ca 2+ fluxes in response to cockroach extracts has been shown in alveolar A549 cells 26 , cultured human keratinocytes 27 and KNRK cells, a rat kidney cell line 10 . However, the pathways mediating these Ca 2+ fluxes are unknown. We found that administration of cockroach extract to BEAS-2B cells in a 2 mM Ca 2+ Ringer's solution produced a biphasic rise in cytoplasmic Ca 2+ : a rapid initial spike followed by sustained Ca 2+ signals that lasted more than 10 minutes (Fig. 1A). In most cells, the sustained component of the Ca 2+ response consisted of an elevated baseline with an oscillating component superposed on the baseline. The sustained signals elicited by cockroach extract were almost completely abolished in a Ca 2+ free Ringer's solution, suggesting that Ca 2+ influx across the plasma membrane was needed for this Ca 2+ signal (Fig. 1B). Moreover, the CRAC channel inhibitor, BTP2, significantly inhibited both the plateau Ca 2+ signals as well as the oscillating component (Fig. 1C), indicating that the sustained Ca 2+ signals arise from the opening of CRAC channels. Fig. 1A and Orai1 15 . Knockdown of STIM1 and Orai1 using siRNA reduced expression of these proteins ( Fig. 2A and Supplementary Fig. S1) and significantly reduced the amplitude of Ca 2+ elevation induced by cockroach extract (Fig. 2B). Both the average amplitude of Ca 2+ signal and the integrated Ca 2+ signal over time was significantly attenuated in siSTIM1 and siOrai1 treated cells (Fig. 2C,D). Knockdown of STIM1 by siStim1 showed good specificity and did not have any effect on STIM2 expression ( Supplementary Fig. S1). Interestingly, analysis of single cell Ca 2+ responses revealed that a proportion of cells in both siSTIM1 and siOrai1 treated samples showed Ca 2+ oscillations (Fig. 2F,G). It is likely that, given the incomplete knockdown of STIM1 and Orai1 by siRNA ( Fig. 2A), these oscillatory signals are mediated by the residual CRAC channel machinery. In contrast, cells treated with the siRNA control showed a more heterogeneous response with individual cells showing a sustained increase in Ca 2+ signal with oscillatory Ca 2+ signals riding on top of the elevated baseline, which accounted for the higher average Ca 2+ response (Fig. 2E). We also note that the average Ca 2+ elevation in response to cockroach extracts seen in siControl treated cells was lower than in untransfected control cells (Fig. 1D)   which in turn plays a critical role in mediating the inflammatory effects of HDM 8,30 . HDM has also been shown to activate Ca 2+ signals in airway epithelial cells in both primary epithelial cells and cell lines, and this is believed to occur through both PAR2-dependent and -independent mechanisms 8,30,31 . However, whether HDM can activate CRAC channels has not been studied. When administered in a Ca 2+ -free Ringer's solution, HDM induced only a transient Ca 2+ signal indicating that the extract causes Ca 2+ release from internal stores ( Supplementary Fig. S2).
In the presence of extracellular Ca 2+ , however, HDM extracts activated a sustained Ca 2+ signal in BEAS-2B cells that was inhibited by the CRAC channel inhibitor BTP2 ( Fig. 3A-C, Supplementary Fig. S2). Further, knockdown of the CRAC channel proteins STIM1 and Orai1 significantly abrogated the average sustained Ca 2+ signals seen in response to HDM ( Fig. 3D-F). These results indicate that HDM allergens mobilize cellular Ca 2+ elevations in bronchial epithelial cells by depleting ER Ca 2+ stores and activating CRAC channels encoded by STIM1 and Orai1.

Insect allergens mobilize Ca 2+ signals by activating PAR2 receptors. Both HDM and cockroach
allergens have been shown to activate PAR2 receptors on airway epithelial cells 1,9,30,31 . Whether this is the primary mechanism by which cellular Ca 2+ signals are generated remains a contentious issue, with evidence for both PAR2 dependent and independent mechanisms 8,30 . We therefore studied the effect of siRNA mediated PAR2 receptor knockdown on allergen induced Ca 2+ influx. Cytosolic Ca 2+ elevations in response to type IX trypsin, a well characterized PAR2 agonist, was strongly inhibited in the siRNA treated cells, confirming knockdown of PAR2 in these cells ( Fig. 4A-C). Importantly, Ca 2+ influx seen in response to cockroach allergens was also significantly inhibited in siPAR2 treated cells, indicating that cockroach extracts induced Ca 2+ signals by activating PAR2 receptors ( Fig. 4D-F). This conclusion is further supported by the Ca 2+ responses seen following paired application of the allergen and the PAR2 specific agonist, trypsin. Following application of the cockroach extract, administration of trypsin failed to elicit a Ca 2+ signal, suggesting that trypsin and cockroach extract activate the same signal transduction pathway, and prior desensitization of the PAR2 receptor or immediate downstream signaling attenuates the response to a second challenge to PAR2 (Fig. 4G). By contrast, Ca 2+ mobilization in response to P2Y receptor activation by UTP (therefore PAR2-independent) was unaffected following prior treatment with the allergen. Likewise, pre-application of the PAR2 agonist trypsin impaired a subsequent Ca 2+ response to cockroach allergen but not to UTP. (Fig. 4H). These results are consistent with the interpretation that cockroach allergens activate PAR2 receptors in AECs. However, in contrast to the effects of the cockroach extracts, knockdown of PAR2 elicited only modest inhibition of the Ca 2+ response to dust mite extract ( Fig. 4I-K). This result suggests that the Ca 2+ response to dust mite extracts is mediated by both PAR2-dependent as well as independent mechanisms.
Ca 2+ responses to fungal and bacterial allergens. The ability of cockroach and dust mite extracts to stimulate SOCE led us to next consider whether this Ca 2+ influx pathway might be a common feature of other allergenic pathways. Chitinase enzyme from fungal and insect sources has been implicated in airway inflammation and elevated expression of a mammalian chitinase enzyme has been noted in mouse models of asthma and in allergic asthma in humans 22 . Moreover, Hong et al. have noted that Chitinase stimulates Ca 2+ flux in airway epithelial cells through a mechanism likely involving PAR2 receptors 22 . We found that chitinase extracts from Streptomyces griseus produced oscillatory Ca 2+ signals in a significant fraction of BEAS-2B cells (Fig. 5A). These Ca 2+ signals were inhibited by exposing cells to BTP2, suggesting that, like cockroach and HDM extracts, chitinase activates CRAC channels in AECs (Fig. 5A-C).
Allergens derived from the fungus Alternaria alternata have been shown to trigger a Th2 type response through the release of IL-33 from airway epithelial cells in a Ca 2+ dependent manner 32,47 . Alternaria also induces the production of IL-6, IL-8 and GM-CSF from AEC 13 . We found that following treatment with Alternaria extracts, BEAS-2B cells showed a slow but progressive increase in their [Ca 2+ ] i levels at concentrations of 30 μ g/mL or above (Fig. 5D). However, this increase was not affected by pre-treatment with BTP2, ruling out involvement of CRAC channels in this process (Fig. 5E,F). Furthermore, Alternaria extracts did not evoke release of Ca 2+ from internal stores (Fig. 5G). We did not observe concomitant reduction of fura 340 and 380 signal following addition of the fungal extracts, ruling out the possibility of proteolytic cell damage. These results indicate that the slow [Ca 2+ ] i rises evoked by Alternaria extracts do not involve CRAC channels.
Another fungus that is commonly associated with inflammatory lung diseases including asthma, allergic sinusitis, bronchopulmonary aspergillosis, and chronic eosinophilic pneumonitis, is Aspergillus fumigatus 33 . Although some studies have implicated cross-talk between TLRs and PAR2 receptors in the response to this fungus, the basic mechanisms by which Aspergillus fumigatus triggers airway inflammation remain largely unknown. In our tests, neither low nor high molecular weight fractions of extracts from Aspergillus fumigatus induced Ca 2+ signals in bronchial BEAS-2B cells (Fig. 5H,I). Thus, the ability of Aspergillus fumigatus extracts to modulate signaling pathways in AECs including inhibition of Jak-Stat signaling 33 is likely not mediated by CRAC channels. Overall, these results suggest that the activation of Ca 2+ influx through CRAC channels is confined to a specific subset of insect allergens that include cockroach, dust mites, and chitinase enzyme.

Dust mite and cockroach allergens induced generation of IL-6 and IL-8 through activation of CRAC channels. An important consequence of allergen sensing by the airways is the induction of
pro-inflammatory mediators such as IL-6 and IL-8, which leads to the recruitment of various immune cells to the airway 9,10,12 . IL-8 plays an important role in the recruitment of neutrophils to the site of airway injury whereas IL-6 is a pleiotropic cytokine that is critical for B cell differentiation as well as T cell activation 34 . We found that exposure of BEAS-2B cells to cockroach allergens or HDM extracts resulted in the induction of IL-6 and IL-8 both at 6 and 24 hour time points (Fig. 6A-H). The induction of these cytokines was abolished by the CRAC channel antagonist, BTP2, indicating that Ca 2+ entry through CRAC channels is essential for the generation of these cytokines (Fig. 6A-H). This result is consistent with our previous report demonstrating that activation of PAR2 receptors leads to induction of IL-6 and IL-8 in a CRAC channel dependent manner 15 . Moreover, cyclosporine A, a calcineurin inhibitor, impaired the generation of IL-6 and IL-8 following challenge by allergens, indicating that calcineurin/NFAT signaling is critical for the induction of IL-6 and IL-8 by allergens (Fig. 6A-H) 35 .   Together, these results indicate that insect allergens stimulate the production of IL-6 and IL-8 via NFAT dependent gene expression that is driven by Ca 2+ entry through CRAC channels.

Discussion
Interactions between common allergens found in ambient air such as insect (house dust mites, cockroach) or fungal (Aspergillus, Alternaria) allergens and the airway epithelium underlie the development of airway inflammation seen in allergic diseases like asthma 36,37 . The biological effect of these allergens is mediated, in part, by proteolytic activity contained within them, which produces epithelial cell damage and activates protease activated receptors to trigger signaling cascades that lead to production of several key inflammatory mediators from AECs 5 .
A key signaling event in allergen induced cell signaling is the elevation in cytosolic Ca 2+ , which has been proposed to occur through both PAR2 dependent and independent pathways [10][11][12]30 . However, the specific Ca 2+ entry pathways that mediate allergen-evoked Ca 2+ signals have not been determined. In this study, we show that insect allergens derived from cockroach and dust mite extracts activate CRAC channels in bronchial epithelial cells. CRAC channel activation occurs, at least in part, through stimulation of PAR2 receptors and mediates an important role in the induction of the inflammatory modulators IL-6 and IL-8. Our primary finding is that cockroach and dust mite allergens induce sustained and oscillatory Ca 2+ signals in bronchial BEAS-2B cells by activating CRAC channels. Both pharmacological inhibition by BTP2 and knockdown of the canonical CRAC channel proteins STIM1 and Orai1 significantly abrogated insect allergen induced Ca 2+ signals (Figs 2 and 3). For both allergens, while the sustained component of the Ca 2+ signal was completely inhibited in Ca 2+ free buffer (Fig. 1B and data not shown), some residual Ca 2+ influx persisted in cells in which STIM1 and Orai1 were knocked down (Figs 2F and 3D). This is most likely due to the incomplete knockdown of CRAC channel proteins as seen in the Western blot ( Fig. 2A). However, we cannot rule out that the possibility that other CRAC channel proteins (STIM2, Orai2, 3) also make some contribution to the allergen-induced Ca 2+ signals. Interestingly, while the average [Ca 2+ ] i rise following stimulation with cockroach extracts was lower in siSTIM1 and siOrai1 treated cells, a fraction of these cells showed Ca 2+ oscillations (Fig. 2F,G). The specific nature of Ca 2+ signals is often determined by complex interactions between the agonist, agonist receptor, IP 3 receptors and Ca 2+ channels 38 . It is possible that the reduced Ca 2+ influx in the siSTIM1 and siOrai1 treated cells fundamentally affected feedback to IP 3 receptors and changed the nature of Ca 2+ signals to the oscillatory type. It would be interesting to test if the sustained Ca 2+ signals seen in response to cockroach and dust mite allergens become oscillatory in nature when the concentration of the allergens is reduced in the external media, as has been shown for other agonists like ATP 39 . If true, this could have important implications for downstream signaling. For example, depending on the concentration of inhaled allergens in the airway, AEC might produce either oscillatory or sustained Ca 2+ signals, with each producing a distinct biological response.
Several studies have established a role for PAR2 receptors in the induction of Ca 2+ signals in response to cockroach and dust mite allergens 9,10,12,40 . While components of dust mite allergens such as Der p3 and Der p5 induce Ca 2+ signals in kidney and alveolar epithelial cell lines through PAR2 receptors, other components such as Der p1 do not activate Ca 2+ signals 8,30 . We found that knockdown of PAR2 receptors significantly inhibited Ca 2+ signals in response to both cockroach and dust mites, though the inhibition of Ca 2+ signal was incomplete, likely due to incomplete knockdown of the PAR2 protein (Fig. 4). However, given that the dust mite extract likely contains a combination of several serine proteases, and possibly many other undefined proteins, it is possible that additional PAR2 independent mechanisms also mediate the observed elevation of cellular Ca 2+ signals. Future studies that examine the Ca 2+ responses to specific purified or recombinant dust mite proteases (e.g., Der p1, 3, 5 and 9) will help to discern the contributions of precise components of the dust mite extracts to the observed Ca 2+ signals.
Interestingly, we failed to detect the involvement of CRAC channels in the Ca 2+ elevation evoked by extracts of the Alternaria fungus. A previous study that used a high concentration of Alternaria extracts (200 μ g/mL) found that the extracts cause Ca 2+ elevations acting through the autocrine stimulation of purinergic receptors by ATP 23 . Here, we employed a lower concentration of the extract (30 μ g/mL) and found that while the extract induced Ca 2+ elevations in the presence of extracellular Ca 2+ , no response occurred in Ca 2+ -free Ringer's buffer, arguing against activation of purinergic receptors, at least in the concentration range we tested. Moreover, the observed Ca 2+ influx seen in the presence of extracellular Ca 2+ was not dependent on CRAC channels as the CRAC channel inhibitor BTP2 had no effect. These results indicate that Alternaria evokes Ca 2+ influx through other Ca 2+ influx pathways whose identity remains to be determined.
In conclusion, we provide evidence showing that bronchial epithelial cells sense cockroach and dust mite allergens through the activation of cell surface PAR2 receptors, which in turn leads to the opening of store-operated CRAC channels. The ensuing Ca 2+ signal is known to play an important role in the induction of the cytokines IL-6 and IL-8 through an NFAT dependent mechanism. These results demonstrate that CRAC channels may have a central role as effectors of allergen signaling in the airway epithelium.   where R is the F 340 /F 380 fluoresce intensity ratio and R max (9.645) and R min (0.268) were determined by in-vitro calibration of FURA2 42 . β (20.236) was determined from the F min /F max ratio at 380 nm and K d is the apparent dissociation constant of fura-2 binding to Ca 2+ (135 nM).
Western blots. BEAS-2B cells were cultured in 6-well plates. At 70% confluency, cells were washed with cold PBS and lysed in a solution containing 150 mM NaCl, 50 mM Tris, 1% Triton-X-100, 0.1% SDS and 1x Protease Inhibitor Cocktail (Sigma) for 45 minutes. Cell lysates were obtained using a cell scraper, lysates were spun down at 4 °C for 30 minutes and supernatants were collected and stored at − 80 °C. For Western blotting, samples were heated to 99 °C for 5 minutes in Laemmli Sample Buffer (Bio-Rad) containing 0.1% β -mercaptoethanol, run on 10% SDS-PAGE gels, and transferred to nitrocellulose membrane. Orai1, STIM1, and STIM2 proteins were detected using an affinity purified polyclonal antibodies and peroxidase labelled secondary antibodies 43,44 . Analysis of cytokine secretion. BEAS-2B cells were cultured on 24 well plates in DMEM/F12 media with 5% FBS. 24-48 hours later, cells were pre-treated with CRAC channel inhibitor BTP2 (500 nM) or calcineurin inhibitor Cyclosporin A (500 nM) for 45-60 min before being stimulated with dust mite and cockroach allergens for 6 or 24 hours. Supernatants were collected and stored at − 80 deg. C. Levels of inflammatory mediators IL-6 and IL-8 was measured using ELISA kits (RayBiotech for IL-6, and LifeTechnologies for IL-8).
Data analysis. Average cytosolic Ca 2+ traces and bar graphs summarizing the data are reported as mean ± SEM. For data sets involving more than two groups, initial statistical analysis was performed using ANOVA with a confidence interval of 5%. This was followed by two-tailed paired student t-test for comparing different treatment conditions within the set.