Functional analysis of the impact of ORMDL3 expression on inflammation and activation of the unfolded protein response in human airway epithelial cells

Background The gene ORMDL3 was shown to be associated with early-onset asthma susceptibility in multiple independent genome-wide and candidate-gene association studies. Asthmatic patients have elevated expression levels of this gene. ORMDL3 encodes a transmembrane protein localized in the endoplasmic reticulum (ER) that may be involved in ER stress and inflammation. It is essential to validate the genetic associations linking ORMDL3 with asthma through functional studies that confirm the biological relevance of this gene in disease. We investigated the effects of manipulating ORMDL3 expression levels in vitro in airway cells on innate immune inflammatory responses, ER stress and activation of the unfolded protein response (UPR). Methods ORMDL3 expression levels were manipulated in airway cells using an overexpression plasmid and siRNA technologies. Successful modulation of ORMDL3 was confirmed at both the gene and protein level. The functional impact of modulation of ORMDL3 expression levels on inflammatory responses and activation of the UPR were quantified using complementary cellular and molecular immunology techniques. Results Cells with altered ORMDL3 levels responded equally well to innate immune stimuli and produced similar levels of pro-inflammatory cytokines compared to wild-type cells. Treatment with ER stress inducers, thapsigargin and tunicamycin, resulted in activation of the unfolded protein response (UPR). However, we observed no difference in UPR activation in cells with ORMDL3 knockdown compared to cells with normal ORMDL3 levels. Conclusions Our results suggest that ORMDL3 variation in the airway epithelium is unlikely to play a significant role in modulating innate immune responses and the UPR in the lung.


Introduction
Asthma and allergic diseases are rapidly becoming the most common chronic diseases in the developed world. Current asthma therapy treats symptoms of the disease, however it is ineffective in up to 25% of patients [1]. Asthma and allergic diseases are complex disorders caused by the interaction of various genetic and environmental factors [2][3][4].
Genome-wide association studies (GWAS) have been used to identify genes that may be involved in asthma pathogenesis [5]. Moffatt and colleagues first reported that multiple single nucleotide polymorphisms (SNPs) on chromosome 17q21 linked ORMDL3 (orosomucoid 1-like 3) to the risk of developing childhood asthma [6]. This association has since been reproduced in multiple independent studies [7][8][9][10][11][12][13][14]. However, little work has been done to elucidate the biological and functional relevance of this gene in asthma. The disadvantage of these association studies is that they cannot differentiate between true causal SNPs and non-causal variants simply in linkage disequilibrium with disease-causing genes. It is therefore imperative to validate GWAS data through functional studies that confirm the biological relevance of a gene in disease.
SNP variants have also linked ORMDL3 to inflammatory bowel disease (IBD) and Type I diabetes, suggesting that ORMDL3 may be involved in dysregulation of the immune system [15,16]. Association of ORMDL3 in both asthma and IBD is of interest because the lung and gut are composed of similar mucosal surface cells and these tissues are exposed to many potentially harmful antigens and allergens requiring tight regulation of the mucosal immune system [17]. This unique system is responsible for maintaining a delicate equilibrium between antigen responsiveness and tolerance and is therefore responsible for preventing hyper-reactivity [17]. Inappropriate immune responses to foreign components or commensal bacteria can lead to inflammation characteristic of asthma and IBD. Furthermore, the polymorphisms may be involved in regulation of mRNA expression of 17q21 locus genes, including ORMDL3 [6]. The expression of ORMDL3 was recently associated with elevated levels of IL-17 secretion [18] and ORMDL3 was expressed at higher levels in the peripheral blood of patients with recurrent wheeze compared to controls [19]. This correlation further supports the hypothesis that ORMDL3 is involved in immunity.
The ORMDL3 gene is a member of a family of conserved endoplasmic reticulum (ER)-localized transmembrane proteins [20]. The functions of the ORMDL proteins are currently unknown, but a recent study suggested that ORMDL3 is involved in ER-mediated Ca 2+ homeostasis and activation of the unfolded protein response (UPR) -ORMDL3 may inhibit sarco/endoplasmic reticulum Ca 2+ ATPase (SERCA) activity [21,22]. Disruptions to ER Ca 2+ concentrations can cause protein misfolding, and accumulation of these unfolded proteins can lead to ER stress [23,24]. UPR signaling cascades are initiated in response to this stress and have been shown to activate the JNK-AP-1 and NF-κB-IKK pathways [25][26][27]. The ER stress response and UPR, caused by changes in ORMDL3 expression, can initiate inflammation through induction of cytokine production. This mechanism may explain the role of ORMDL3 in asthma pathogenesis. Indeed, Miller et al. have shown that in mice ORMDL3 is an allergen and cytokine (IL-4 or IL-13) inducible ER gene expressed predominantly in airway epithelial cells, and that it activates the ATF6 pathway of the ER localized UPR regulating expression of metalloprotease, chemokine, and oligoadenylate synthetase genes [28].
Although the symptoms of asthma are largely driven by dysregulated T helper type 2 (T H 2) responses, innate immune responses are also involved in asthma pathogenesis [29,30]. Airway epithelia are central to host defense and immune regulation. These cells are among the first to encounter environmental insults and play an important role in shaping downstream immune responses. Any dysregulation of the innate immune response can result in hypersensitivity to environmental factors, leading to asthma symptoms.
Given the multiple lines of evidence suggesting that ORMDL3 is involved in immunity, we investigated the role of the gene in innate immune responses of airway cells. We hypothesized that elevated ORMDL3 levels result in heightened inflammatory responses that are associated with the asthmatic phenotype. Increased levels of ORMDL3 protein may in turn disrupt ER homeostasis, leading to ER overload and activation of the UPR, initiating inflammatory responses. Using an in vitro model, we manipulated ORMDL3 expression in airway cells to determine whether a difference in basal ORMDL3 expression affected inflammatory responses or activation of the UPR before and after stimulation.

Cloning ORMDL3 cDNA into pEGFP-N1 vector
The ORMDL3 gene was amplified from cDNA using forward primer 5 0 -CTAAGAATTCATGAATGTGGGCAC AGCGCAC-3 0 and reverse primer 5 0 -TACTGGTACCCC GTACTTATTGATTCCAAAAATCCGGACT-3 0 , introducing EcoRI and KpnI restriction endonuclease sites, respectively. The ORMDL3 PCR product was then inserted into a pEGFP-N1 eukaryotic expression vector (Clontech). ORMDL3 and eGFP are in frame and produce a fusion protein with eGFP expressed at the C-terminus of ORMDL3. The construct was verified by sequencing and is denoted as pEGFP-ORMDL3. Protein is denoted as ORMDL3-eGFP.

Cell transfection
A549 and 1HAE cell lines were transfected with pEGFP-ORMDL3, scramble (non-specific) or ORMDL3-specific siRNA (pre-designed by Qiagen) using Amaxa W Cell Line Nucleofector W Kit T (Lonza). Two ORMDL3-specific siRNAs were used. Concentrations used for transfection represent pooled siRNA concentration. Cells were seeded into a 24-well plate (BD Biosciences) at a density of 2x10 5 cells/well for A549 cells or 1x10 5 cells/ well for 1HAE cells.

ER stress induction and UPR activation
Cells were stimulated with tunicamycin (200 μg/mL) (Calbiochem) or thapsigargin (10 μM) (Sigma) for 2 or 4 hours to activate the UPR. For ORMDL3 knockdown cells, stimulation was performed 24 hours post-transfection. RNA was extracted and expression of genes XBP-1u, XBP-1s, and CHOP were then quantified as markers of UPR activation. For measurement of p-eIF2α levels by Western blot, lysates from unstimulated cells with ORMDL3 knockdown were collected 24 hours post-transfection. Experiments were repeated three times (n = 3).

RNA isolation and reverse transcription
RNA was extracted from lysates using E.Z.N.A. W Total RNA Kit (Omega Bio-Tek) according to the manufacturer's protocol. Extracted RNA was reverse transcribed into cDNA using the SuperScript W VILO ™ cDNA Synthesis Kit (Invitrogen). Complement DNA was diluted to 200 ng/μl prior to quantification of gene expression by qPCR. This method was followed for all samples, unless otherwise stated (see PCR Array).

Quantification of ORMDL3 mRNA expression
Gene expression was calculated relative to GAPDH or PPIA (encoding cyclophilin A) and was quantified by SYBR Green chemistry (PerfeCTa ™ qPCR SuperMix, Quanta Biosciences) using a 7300 Real Time PCR System (Applied Biosystems). Reactions were performed in triplicate using the following cycling conditions: 50°C for 2 mins, 95°C for 10 mins, [95°C for 15 s, 60°C for 1 min] x 40. The relative expression of the measured gene was calculated by the Pfaffl method [35]. The primers used are listed in Table 1.

Western blot analysis
Cells were lysed in 50 μl RIPA Buffer + 1x HALT ™ protease inhibitor (Thermo Scientific). Cell debris were removed by centrifugation: 18,000 x g for 10 min at 4°C. Proteins were analyzed by standard Western blotting protocols where they were transferred onto Immobilon W -FL transfer membrane (Millipore). Antibodies used for Western blot

Statistical analysis
Data are shown as mean ± SEM of three separate experiments. Results were analyzed using one-way ANOVA with Bonferroni post-test. Statistical analysis was performed using GraphPad Prism5 (GraphPad Software, Inc.). Differences with p < 0.05 were considered significant.

ORMDL3 modulation in airway cells
In order to determine functional impact of ORMDL3 modulation, knockdown of the gene was performed using siRNAs. A549 and 1HAE cells were transfected with scramble (non-specific) or ORMDL3-specific siRNA. Modeling variation expected to occur in the human population, we achieved 40-70% knockdown of ORMDL3 gene expression using siRNA concentrations of 50 nM-500 nM. We also confirmed that ORMDL3-specific siRNA did not affect expression of genes in the same family, ORMDL1 or ORMDL2 (Additional file 2). Sequences of primers used for qPCR are listed in Table 1. ORMDL3 has 84% and 80% protein sequence homology to ORMDL1 and ORMDL2, respectively. This presented a challenge for confirming knockdown of ORMDL3 protein because commercially available antibodies detect all three proteins. Therefore, we were unable to show ORMDL3 protein knockdown in cells transfected with siRNA alone. To overcome this limitation, we constructed a fusion protein, where ORMDL3 is tagged with an eGFP protein. Co-transfection of this overexpression plasmid with ORMDL3-specific siRNA enabled us to knockdown ORMDL3 and detect changes at the protein level using an anti-GFP antibody.
Airway cells were co-transfected with both pEGFP-ORMDL3 and siRNA (scramble or ORMDL3). Protein and mRNA were analyzed for gene knockdown. At the transcript level, we observed ( Figure 1) a small but significant increase (p < 0.05) in ORMDL3 expression in the cells transfected with pEGFP-ORMDL3 and scramble siRNA compared to cells alone, confirming successful overexpression of the ORMDL3 gene. Transfection with pEGFP-ORMDL3 and increasing concentrations of ORMDL3 siRNA resulted in a titration effect of increasing ORMDL3 knockdown. One advantage to constructing a fusion protein, ORMDL3-eGFP, is that eGFP is only expressed with expression of ORMDL3. Therefore, when cells are cotransfected with pEGFP-ORMDL3 and siRNA, knockdown of ORMDL3 protein could subsequently be detected by immunoblot analysis using the anti-GFP antibody. Expression of the ORMDL3-eGFP protein was confirmed by Western blot of whole cell lysate collected from the cells in each condition. Figure 1 shows knockdown of ORMDL3-eGFP protein, confirming that ORMDL3 siRNA affects protein expression. The double band may be explained by either variation in mRNA splicing or posttranslational modifications to the fusion protein, such as acetylation, methylation, myristylation, phosphorylation, or glycosylation.
ORMDL3 knockdown does not affect IL-6 or IL-8 production following innate immune activation 1HAE cells transfected with pEGFP-ORMDL3 and scramble or ORMDL3-specific siRNA were stimulated 24 hours post-transfection and supernatants were collected 48 hours post-transfection. Stimuli used were TNF-α, IL-1β, LPS, and flagellin. TNF-α and IL-1β were chosen because both are early response cytokines that perpetuate acute inflammatory processes. LPS and flagellin, in contrast, are common microbial antigens recognized by the innate immune system. Two classic and biologically-relevant NF-κB-induced cytokines with important roles in innate immunity, interleukin-6 (IL-6) and interleukin-8 (IL-8), were measured by ELISA. Despite confirmation of ORMDL3 mRNA and protein knockdown, we did not observe any impact on IL-6 or IL-8 production after stimulation as shown in Figure 2A-B. Although the cells have low baseline responsiveness to LPS and flagellin, our results indicate that ORMDL3 knockdown does not enhance sensitivity to these stimuli. Similar results were obtained in A549 cells, as well as 1HAE cells transfected with siRNA alone (Additional file 2).

ORMDL3 knockdown does not enhance UPR activation upon stimulation
We next explored the effects of ORMDL3 expression on activation of the UPR. Initiation of the UPR is mediated by one or more of the ER-membrane protein sensors: PKR-like eukaryotic initiation factor 2α kinase (PERK), inositol requiring enzyme 1 (IRE1), and activating transcription factor-6 (ATF6) [23]. Activation of any of the three pathways initiates signaling cascades that mediate changes to relieve ER stress. The gene XBP-1 is a substrate for IRE1 ribonuclease [24]. Upon activation of the IRE1 pathway, the IRE1 ribonuclease removes a 26-bp intron from the unspliced variant, XBP-1u, which results in the spliced variant, XBP-1s [36]. This spliced variant is the active form of the gene that contributes to ER stress responses. CHOP transcription, in contrast, can be induced by the PERK and ATF6 pathways [24]. Phosphorylated eIF2α (p-eIF2α) is an early marker of PERK pathway activation and is upstream of CHOP induction [23]. Expression changes in these three genes that signify UPR activation, XBP-1u, XBP-1s, and CHOP, were determined by qPCR. We also evaluated phosphorylation of eukaryotic initiation factor 2α (eIF2α) by Western blot, as modulation of ORMDL3 expression has been reported to influence eIF2α phosphorylation [21]. Figure 2 Cytokine production in cells with ORMDL3 knockdown. A) Secreted IL-6 and B) IL-8 levels after cell stimulation. 1HAE cells were stimulated with TNF-α, LPS, flagellin, or IL-1β for 24 hours. Data represent the mean ± SEM of three experimental repeats. Statistical analysis was performed using one-way ANOVA with Bonferroni post-test.

A B
As positive controls for UPR activation, we stimulated 1HAE cells with tunicamycin and thapsigargin. Both are inducers of ER stresstunicamycin inhibits N-linked glycosylation and thapsigargin inhibits the SERCA pump causing ER calcium stores to be depleted [37]. Quantitative measurement of transcript levels showed that both XBP-1s and CHOP increased, while XBP-1u decreased upon stimulation with either tunicamycin or thapsigargin ( Figure 3A-B). This confirmed the utility of measuring these genes to monitor UPR activation.
At baseline (unstimulated cells), variation in ORMDL3 expression did not induce UPR activation in A549 or 1HAE cells (Additional file 2). 1HAE cells with ORMDL3 knockdown were also stimulated with tunicamycin or thapsigargin ( Figure 3C-D). In both conditions, ORMDL3 knockdown did not show increased UPR activation compared to the negative control. Furthermore, levels of phosphorylated-eIF2α were indistinguishable between ORMDL3 knockdown cells and controls ( Figure 3E).

Impact of ORMDL3 knockdown on the expression of multiple genes involved in inflammation, asthma & allergy
To expand our search for immune functions potentially altered by ORMDL3, the expression of 168 genes was determined at two time points (2 and 24 hours) following stimulation with TNF-α. We performed PCR arrays to profile expression of cytokines, chemokines, and key genes involved in asthma and allergy (a complete list of genes that were studied can be found in Additional file 1). Gene expression was compared between 1HAE cells with high (plasmid + 500 nM scramble siRNA) and low (plasmid + 500 nM of ORMDL3 siRNA) ORMDL3 expression that were stimulated for 24 hours with TNF-α. Stimulation with TNF-α induced a robust inflammatory response in the cells, enabling us to observe whether variation in basal ORMDL3 levels impacts the immune response. These arrays identified eight genes (CCL2, TSLP, CSF2, CCL5, VEGFA, ADRB2, IL1RL1, and IL13RA1), shown in Figure 4, that appeared to be differentially regulated by more than 1.5 fold and that were expressed at relatively high levels (average threshold cycle <30). However, upon replication to validate these results, we determined that the differences were not statistically significant.
The same arrays were performed on cells stimulated for 2 hours with TNF-α. The same conditions for "high" and "low" ORMDL3 expression were used. From our results, we identified only one gene, IL23A, that was differentially regulated by more than 1.5 fold (fold regulation −2.37) and amplified at cycle <30. Replication and comparison of IL23A expression between knockdown conditions yielded no significant difference. No other genes, as identified in the previous arrays, were found to be differentially regulated at this time point. These results suggest that ORMDL3 variation does not have a meaningful impact on expression of a large panel of immune-related genes in airway epithelial cells.

Discussion
Asthma is a complex disease affecting many individuals in the developed world. Genome-wide association studies have recently been used to identify genetic causes for such complex diseases. One particular gene, ORMDL3, is of interest because of its association with asthma, IBD, and Type I diabetesall of which are caused by immune-mediated pathology [6,10,22,38,39]. The gene ORMDL3 is an ER-membrane protein and is potentially involved in Ca 2+ -signaling in the ER and sphingolipid synthesis [20,21,40]. It has also been correlated to activation of the UPR, though the mechanisms remain unclear [21]. Activation of the UPR may be biologically relevant, as ER stress, the UPR, and inflammation have all been linked [23]. However, the functional role ORMDL3 in the pathogenesis of asthma has yet to be elucidated.
Airway epithelial cells play an important role in innate immunity and in the development of asthma. Current findings in literature indicate that ORMDL3 is involved in immunity and that asthmatics have higher expression of the gene than non-asthmatics [18,21,22]. A recent study by Miller et al. also investigated the role of ORMDL3 in airway epithelial cells. They reported that in vitro overexpression of ORMDL3 activated the ATF6 pathway of the UPR and induced expression of several genes with potential importance in the pathogenesis of asthma [28]. Our investigation, in contrast, focuses on the effect of variation of ORMDL3 expression levels, at baseline, on the innate immune responsiveness of airway epithelial cells. By manipulating ORMDL3 expression in vitro to mimic differences in gene expression established between asthmatics and healthy individuals, we aimed to understand the role of ORMDL3 on the innate immune response and UPR activation status in airway epithelial cells. This method ensured control and the confidence that any effect on the innate immune response was in fact correlated with a change in ORMDL3 expression levels. If the same experiments were performed on ex vivo airway cells of patients, genetic and other differences between individuals could have affected the results.
After knockdown of ORMDL3 in vitro, cells were stimulated with cytokines (TNF-α, IL-1β) or common microbial components (LPS, flagellin). We monitored production of interleukin-6 (IL-6) and interleukin-8 (IL-8) (alias CXCL8), two pro-inflammatory cytokines produced by airway cells that are relevant in asthma pathogenesis. Specifically, IL-6 is elevated in individuals with asthma [41] and is also regulated by ATF6 during activation of the UPR [42]. Similarly, transfection of ORMDL3 into human airway epithelial cells triggers ATF6 activation and IL-8 secretion [28]. However, in our experimental system, variation in ORMDL3 expression levels did not affect NF-κB-induced innate immune production of IL-6 and IL-8 in airway epithelial cells.
We next explored the effects of ORMDL3 expression on activation of the UPR. UPR signaling cascades are initiated in response to ER stress, and restoration of homeostasis is achieved by attenuating translation, restoring protein folding, or degrading misfolded proteins [24]. Although often associated with abnormal physiological conditions, the UPR plays a central beneficial role in normal physiology; as illustrated by the role of the UPR in terminal B cell differentiation which requires a massive increase in the biosynthetic capacity to synthesize antibodies in response to infection [43]. However, the ER stress response and UPR can also initiate inflammation through induction of cytokine production or activation of transcriptional regulators of inflammatory genes. Cytokines IL-6 and IL-8 are examples of genes that may be induced by UPR activation [23]. ER stress and the UPR have been implicated in many immune-related diseases including IBD, diabetes, chronic obstructive pulmonary disease (COPD), arthritis, and neurodegenerative inflammatory diseases [44]. It is poorly understood whether ER stress is an underlying cause of disease or if its induction is a result of chronic inflammation. Indeed, it is possible that environment factors such as infection or inhalation of smoke particles can activate the UPR, triggering the onset of lung disease in genetically predisposed individuals [45]. However, it is also possible that ER stress is exacerbated by inflammation and contributes to the perpetuation of the disease.
Cantero-Recasens et al. previously reported that ORMDL3 overexpression activated the PERK pathway, but did not affect the IRE1 pathway of the UPR [21]. In contrast, Miller et al. reported that ORMDL3 overexpression activated the ATF6 pathway, but not the PERK or IRE1 pathways [28]. In our study, we chose four markers of UPR activation: XBP-1u, XBP-1s, CHOP, and p-eIF2α. With activation of the UPR, we expect downregulation of XBP-1u and upregulation of XBP-1s and CHOP. However, our results demonstrate that knockdown of ORMDL3 does not activate the UPR, in either unstimulated or stimulated cells. Immunoblot analysis also showed no change in p-eIF2α levels with ORMDL3 knockdown. Furthermore, downstream markers of UPR activation, IL-6 and IL-8 cytokines, were produced at similar levels in unstimulated cells with varying ORMDL3 levels. This further supports our results that ORMDL3 does not activate the UPR. Differences in our results compared to previous work might be due to the different types of cells, conditions, or markers used. It is possible that the effects of variation in ORMDL3 expression are a cell typedependent phenomenon. While no effect on the inflammatory response was detected in airway cells, other cells types such as dendritic cells or T cells may be affected by altered ORMDL3 expression. Observations made by Lluis et al. suggest that the 17q21 locus may potentially play a role in T-cell development [18].
Taking a broader approach, PCR arrays looking at expression of 168 common immunity genes were performed. We reasoned that although ORMDL3 levels may not affect the production of IL-6 or IL-8 cytokines, perhaps they were impacting gene expression of other important immune genes, such as IL-33, IL-25 and TSLP, which have all been implicated in asthma pathogenesis [46]. Verification of differential expression of these genes at a transcript level, however, did not show any significant changes between the ORMDL3 knockdown conditions. This suggests that altering ORMDL3 expression does not have a profound effect on the expression of innate immune genes upon stimulation in the airway epithelia. However, there may be other genes that are affected that were not investigated in this study. Pfeifer et al. recently showed that IL-17C cytokine is expressed by human bronchial epithelial cells and is induced by bacterial infection [47]. It may be worthwhile in future experiments to investigate a broader range of immune-related genes. Interestingly, we did not observe changes to expression of the genes reported by Miller et al., MMP-9, CCL-20, CXCL-10, CXCL-11, or IL-8. This variance may be explained by differences in experimental conditions. Our study examined outcomes in gene expression after stimulation of cells co-transfected with ORMDL3 and ORMDL3-specific siRNA, while the other study used a different experimental approach.
Although this study focused exclusively on the potential role of ORMDL3 in asthma pathogenesis, it is possible that neighboring genes such as GSDML contribute to disease susceptibility at this locus. Many groups consider ORMDL3 as an 'asthma gene'; however, it should be acknowledged that the identified SNPs associating this gene to asthma susceptibility are not located in the gene itself. Even so, these polymorphisms have been consistently correlated with increased odds of asthma risk, highlighting the importance of this locus in disease susceptibility [6][7][8][9][10][11]13,14].
Our data show that variation in ORMDL3 expression is not correlated with differential innate immune responses to stimuli or activation of the UPR in vitro in airway epithelial cells. Taken together, our results are biologically relevant because they suggest that normal human variation of ORMDL3 expression is not likely an important factor in increasing the innate immune response of airway cells we observe in asthmatics. Despite these results, this