key: cord-0888716-03kubcnx
authors: Zhang, Rui; Trower, Jessika; Wu, Tongde
title: Degradation of bacterial permeability family member A1 (BPIFA1) by house dust mite (HDM) cysteine protease Der p 1 abrogates immune modulator function
date: 2020-09-02
journal: Int J Biol Macromol
DOI: 10.1016/j.ijbiomac.2020.08.214
sha: c0cec17423fdd2a5f03ae6994294a3e6e3377cb1
doc_id: 888716
cord_uid: 03kubcnx

Bacterial permeability family member A1 (BPIFA1) is one of the most abundant proteins present in normal airway surface liquid (ASL). It is known to be diminished in asthmatic patients' sputum, which causes airway hyperresponsiveness (AHR). What is currently unclear is how environmental factors, such as allergens' impact on BPIFA1's abundance and functions in the context of allergic asthma. House dust mite (HDM) is a predominant domestic source of aeroallergens. The group of proteases found in HDM is thought to cleave multiple cellular protective mechanisms, and therefore foster the development of allergic asthma. Here, we show that BPIFA1 is cleaved by HDM proteases in a time-, dose-, and temperature-dependent manner. We have also shown the main component in HDM that is responsible for BPIFA1's degradation is Der p1. Fragmented BPIFA1 failed to bind E. coli lipopolysaccharide (LPS), and hence elevated TNFα and IL-6 secretion in human whole blood. BPIFA1 degradation is also observed in vivo in bronchoalveolar fluid (BALF) of mice which are intranasally instilled with HDM. These data suggest that proteases associated with environmental allergens such as HDM cleave BPIFA1 and therefore impair its immune modulator function.

Asthma is a common airway inflammatory disease that currently affects over 17.7 million adults (age ≥ 18) and 6.3 million children (age < 18) in the United States [1, 2]. Asthma poses a tremendous economic burden, with costs exceeding $56 billion annually [3] . The symptoms of asthma are characterized by intermittent airflow obstruction and airway inflammation, producing chest tightness, wheezing, and coughing. In genetically susceptible individuals, exposure to a wide variety of environmental stimuli, including allergens such as HDM, can induce and/or exacerbate the disease [4, 5] .

Structural and inflammatory changes throughout the airway wall lead to bronchial thickening and edema as well as increased mucus production and bronchoconstriction, all of which contribute to the episodic airflow obstruction typical of asthma [6] .

HDM, a major perennial allergen, is a significant cause of allergic rhinitis and allergic asthma-the most common type of asthma. HDM is ubiquitous in human habitats; therefore it is the most important source of indoor allergens [7, 8] . Over the past three decades, many important allergen groups from HDM have been identified and characterized at the molecular level [9] . HDM components that are thought to produce allergic inflammation include immunogenic epitopes, structural polysaccharide chitin from the exoskeleton of HDM, microbial adjuvant compounds, ligands originating from mite-associated compounds, and proteases [10, 11] . These components manifest diverse biological functions depending on their molecular structures. Protease activity is a common feature of many human allergens, including fungi, pollen, animal dander, and bee venom. HDMs and their fecal pellets contain several proteolytic enzymes [11] . To date, a handful of cysteine and serine protease allergens have been cloned, including Der p 1, Der p 3, Der p 6 and Der p 9 [12] . A significant body of evidence suggests that the proteolytic enzyme activities of these allergens enhance IgE synthesis and promote T H 2 inflammatory responses [13] . Therefore, the design of new and more effective therapeutic approaches for treating allergic asthma may rely on the development of drugs that suppress or confer resistance to HDM-mediated protein degradation.

Bactericidal/permeability-increasing (BPI) protein [21] , which binds and neutralizes endotoxic activity of lipopolysaccharides (LPS) from gram negative bacteria. Similar to BPI, the anti-microbial function of BPIFA1 has been supported by both in vitro and in vivo studies. Co-incubation with BPIFA1 can reduce the growth of a variety of Gramnegative bacteria [22] [23] [24] : BPIFA1 -/mice showed increased susceptibility to Pseudomonas aeruginosa (P. aeru) and significantly decreased survival rates compared to their wildtype (WT) littermate controls [25] . When human BPIFA1 was transgenically knocked into mouse airway epithelium, lung Mycoplasma pneumoniae load dropped significantly [22] . On the other hand, compared to BPI, BPIFA1 bears unique features that make it a multi-functional protein: 1) BPIFA1 has an additional helix (α6) in its C terminus that occludes the equivalent lipid-binding site in BPI. This helix interacts with an extended loop between β2 and β3. It has been speculated that the helix will move away upon lipid ligand binding [26] . (2) BPIFA1 contains an 18-residue domain (G22-A39, termed the "S18 region" in a previous publication [27] ) that is not present in BPI or other BPI-type proteins. Although this region was present in the 20-256 residue construct used for crystallization, it was disordered in the resolved structure and could not be visualized within the electron density maps [21] . By binding to the β subunit of the epithelial sodium channel (ENaC) via the S18 region, BPIFA1 negatively regulates sodium transport, maintaining proper airway surface liquid (ASL) volume. These data imply that BPIFA1 is a structurally diversified, multi-functional protein: elucidating the ultimate function of each domain may reveal a novel mechanism of action.

Beyond its well-established anti-microbial activity and more recently reported functions in airway physiology/pathology, BPIFA1 has been shown to function as an innate immunity modulator, as demonstrated in asthma, chronic obstructive pulmonary disease (COPD), and cystic fibrosis (CF) [28, 29] . Specifically, in the context of asthma, BPIFA1 deficiency has been linked to the severity and progression of the disease. For example T H 2 cytokine IL-13 was previously reported to significantly reduce BPIFA1 expression and secretion from human primary epithelial cells, rendering a host susceptible to M. pneumonia infection [30] . BPIFA1 -/mice exhibit enhanced eosinophilic inflammation the association between BPIFA1 deficiency and asthma pathology, but also raised new research questions regarding (1) why BPIFA1 levels decreased in asthmatic airways, apart from transcriptional regulation of IL-13, and (2) how lack of BPIFA1 contributes to the development of airway inflammation. Here, we therefore tested the hypothesis that HDM proteases cleave BPIFA1, causing loss-of-function, which, in turn, contributes to hyperactivation of Ca 2+ signal in airway epithelial cells and therefore increased secretion of pro-inflammatory cytokines.

Human epithelial cell line Calu-3 was purchased from ATCC. These cells were maintained in Minimum Essential Alpha (MEM α; Thermo Fisher) supplemented with 10% fetal bovine serum (FBS; Sigma Aldrich) and 0.1% Penicillin-Streptomycin (Thermo Fisher). HBECs were obtained from freshly excised bronchial specimens derived from normal and asthmatic subjects and were harvested as previously described under protocol #03-1396, approved by the University of North Carolina at Chapel Hill Biomedical Institutional Review Board [33] . Informed consent was obtained from all donors or authorized representatives of the donors. Briefly, HBECs were extracted from excess surgical pathology or autopsy specimens procured through cooperating surgeons and pathologists using protocols following relevant regulations. Airways were dissected by removing all excess connective tissue and cutting into 5-10 cm segments.

Specimens were dissociated over a period of 4-24 h, depending on the size of the tissue, in a 15-mL tube containing 9 mL tissue wash medium (Joklik Minimum Essential Medium <JMEM> supplemented gentamicin <50 μg/mL> and amphotericin <0.25 μg/mL>) plus 1 mL protease solution (1% Protease XIV with 0.01% DNase <10× stock>, Sigma using goat anti-hBPIFA1 or sheep anti-mBPIFA1 (both 1:2000; R&D Systems); Peroxidase-conjugated donkey anti-goat (1:3000; Santa Cruz Technologies), or donkey anti-sheep (1:3000; R&D Systems) IgG (H+L) from commercial sources.

Protease inhibitors aprotinin, E64D, phenylmethylsulfonyl fluoride (PMSF; Sigma-Millipore), and protease inhibitor cocktail (PIC; Roche) were purchased from indicated commercial sources.

Total mRNA was extracted from cells using RNaEasy reagent (Qiagen). Equal amounts of RNA (1 µg) were used for reverse transcription using a cDNA synthesis kit (BioRad). The following TaqMan probes were used: human BPIFA1 (assay ID: Hs00213177_m1) and human GAPDH (assay ID: Hs02786624_g1) (Thermo Fisher). qRT-PCR was performed on the Applied Biosystems 7500 fast machine (Thermo Fisher) as follows: one cycle of initial denaturation (95°C for 4 min), 45 cycles of amplification (95°C for 10s and 60°C for the 30s). Data are presented as relative mRNA levels normalized to the level of GAPDH; the value from the untreated cells was set as 1. PCR assays were performed three times with duplicate samples, which were used to determine the mean ± SEM.

Experiments were performed on 8-10-week C57BL/6J mice of both genders obtained from the Jackson Laboratory (Bar For HDM exposure, Dermatophagoides pteronyssinus lyophilized HDM extract (Greer Laboratories) was resuspended in DPBS at a concentration of 1 mg (dry weight)/ml. The HDM suspension was delivered in the morning between 9:00 and 12:00 AM by intranasal instillation in a single 20-µl volume while the mice were lightly anesthetized with isoflurane (2.5%). 24 h after instillation, mice were sacrificed and BALF was collected.

BPIFA1 (250 µg/ml) was incubated with biotin-labeled HDM (100 µg/ml, Thermo Fisher) for 24 h at 37°C. The resulting solution was then centrifuged at 8000 x g for 5 min to remove any large debris and applied to a NeutroAvidin column pre-washed 3 times with 1x DPBS (Gibco). The sample volume was 200 µl and the protein-beads mixture was J o u r n a l P r e -p r o o f Journal Pre-proof incubated at room temperature for 30 min before being spun down at 500 g for 5 min. The flow-through was then applied to a second NeutroAvidin column to remove any residual biotin-HDM from BPIFA1 fragments, following the same procedure. BPIFA1 protein alone was used as control. Immunoblots against BPIFA1 and biotin-HDM were performed to monitor the quantity of protein before and after purification and to confirm the removal of HDM. Purified BPIFA1 or its fragments were then used for LPS binding assay or human whole blood incubation.

A modified, enzyme-linked immunosorbent assay (ELISA)-based LPS binding method was used to detect interactions between LPS and BPIFA1 as described previously [23] . Briefly, 96-well plates were coated overnight with purified LPS (400 ng) from Escherichia coli 055:B5 (Millipore Sigma). Wells were washed and blocked with 1% bovine serum albumin (BSA)-PBS for 1 h; 400 ng of purified BPIFA1 was then added to each well in triplicate. PBS was used as a control for this experiment. An antibody specific to human BPIFA1 (R&D Systems), diluted 1:5000 with BSA, was used to detect the LPS-bound BPIFA1. Horseradish peroxidase (HRP)-conjugated anti-goat antibody was used as the secondary antibody to detect binding recombinant BPIFA1. Enzyme activity was detected using a TMB Ultra 1-step assay (Pierce Biotechnology) and the reaction was stopped with H 2 SO 4 (Thermo Fisher). Absorbance was detected at OD 450nm in a

BioTek spectrophotometer (BioTek).

Human whole blood from healthy donors was purchased from a commercial source (VWR) and diluted 1:5 with RPMI 1640 (Thermo Fisher) containing 100 IU penicillin + 100 µg streptomycin/ml (Thermo Fisher) in polypropylene reaction vials (BD Biosciences). Incubation with BPIFA1 or BPIFA1 fragments in the presence of LPS was performed for 24 h at 37°C and 5% CO2 in humidified air; vials were shaken and centrifuged for 2 min at 400 g. The cell-free supernatants were stored at -80°C until cytokine measurement by ELISA.

TNFα and IL-6 were determined by ELISA (BioLegend) following the manufacturer's instructions.

Changes in cytosolic Ca 2+ concentration were measured using 5 μM fluo-4 dye (Thermo Fisher) loaded into Calu-3 cells in the presence of 1 mM probenecid and 1 x powerload (both from Thermo Fisher) for 1 h at 37°C. Cultures were then rinsed, and media were replaced with a standard Ringer's solution; fluorescence intensities were read every 15 or 30 s J o u r n a l P r e -p r o o f Journal Pre-proof using a Tecan Infinite Pro plate reader for fluo-4 (excitation: 494 ± 5 nm; emission: 516 ± 5 nm). A fluorescent baseline was established before the thapsigargin (TG, 5 µM, Thermo Fisher) was added to the wells, and changes in fluorescence were normalized to the baseline (F/F 0 ). ΔF/F 0 represents average peak fluorescent intensity changes of three independent experiments.

All data are presented as the mean ± SEM for n experiments. Differences between means were tested for statistical significance using paired or unpaired t-tests or their non-parametric equivalent as appropriate to the experiment.

Differences between groups were judged using ANOVA. From such comparisons, differences yielding P < 0.05 or smaller were judged to be significant. GraphPad Prism software was used for statistical analysis.

We previously measured sputum BPIFA1 levels in healthy donors, asthmatic patients, subjects with chronic obstructive pulmonary disease (COPD), and atopic individuals without asthma, with the latter two cohorts serving as disease controls.

Immunoblot analyses indicated significantly decreased BPIFA1 protein levels in asthmatic patients' samples compared to the other donors [32] . In the same vein, treatment of healthy HBECs for 24 h apically with 100 µg/ml HDM extract resulted in decreased BPIFA1 protein levels in apical lavages and basal lateral media-but not in the lysates of these cells ( Fig. 1A and 1B) . These findings suggest that only secreted BPIFA1 protein is affected by HDM treatment. We then examined BPIFA1 mRNA levels after HDM treatment. Up to 48 h post-treatment, no significant change was observed ( Fig. 1C) . This evidence suggests that allergens such as HDM that are associated with protease activity may partially contribute to the decrease of BPIFA1 protein levels in the airways of allergic asthma patients. To confirm this speculation, we performed time-and dose-course studies by incubating purified human recombinant BPIFA1 with HDM at the indicated time and dose ranges. As expected, HDM cleaves BPIFA1 in both time-and dose-dependent manners (Fig. 1D-1G ). The half-life of BPIFA1 in the presence of 100 µg/ml HDM is roughly 0.83 h, whereas the EC 50 of HDM at 4 h is 32.4 µg/ml. Taken together, these data indicate that reduction of BPIFA1 protein levels in the asthmatic airway is likely due to HDM protease activity and that this proteolytic effect is time-and dose-dependent without changing BPIFA1 mRNA levels. Because HDM is a major perennial allergen that causes allergic rhinitis and allergic asthma, we wanted to test whether HDM exerts its function at different temperature conditions. We incubated 10 µg/ml of recombinant BPIFA1 with 0, 30, and 100 µg/ml HDM at 4, 25, and 37 ºC for 4 h, respectively. The immunoblot results show that HDM cleaves more than 50% of BPIFA1 at 100 µg/ml regardless of temperature, whereas a lower concentration of HDM requires higher temperature (25 ºC and 37 ºC) to work (Fig.4) . A study has shown that most US homes have detectable levels of dust mite allergen in a bed; further, levels previously associated with allergic sensitization and asthma are common in US bedrooms [34] . Our data corroborate this finding, suggesting that HDM-mediated BPIFA1 degradation is temperature-insensitive and that, as long as a sufficient amount of HDM is present, protein substrate such as BPIFA1 can be cleaved at low as 4 ºC, which likely leads to the perennial allergy problems.

More than 20 groups of HDM allergens have been identified to date, at least four of which are known to express catalytically competent proteinase activity [13, 35, 36] . These components exert profound effects on airways, which promotes allergic sensitization [13, 37] . Identifying key components in HDM that are responsible for BPIFA1 degradation could enable us to further elucidate the potent allergenicity of HDM and inform specific treatment strategies. Based on the known protease components in HDM and their properties, we carried out a "short gun" protease inhibitor screening using protease inhibitors against a broad range of important proteases of well-defined activity. We incubated 10 µg/ml of recombinant BPIFA1 and HDM with or without PIC (1X), E64D, PMSF, and aprotinin respectively, at 37 ºC for 4 h, and subjected the products to immunoblot. PIC, which is a broad-spectrum protease inhibitor cocktail, partially inhibited HDM-mediated BPIFA1 degradation at an arbitrary 1X working concentration. A similar effect was seen with E64D (5 µM), a pan cysteine protease inhibitor, whereas serine protease inhibitor PMSF (5 µM) and trypsin inhibitor aprotinin (5 µM) both failed to prevent HDM mediated BPIFA1 degradation ( Fig. 3A and 3B ). These data point to possible component(s) in HDM that could underly BPIFA1 degradation to a group of cysteine proteases. To further confirm the specificity of cysteine protease inhibitor E64D for blocking HDM-mediated BPIFA1 degradation, we performed an E64D dose-course experiment with a fixed amount of recombinant BPIFA1 protein (10 µg/ml) and HDM (100 µg/ml). The inhibitory effect was detectable at a concentration of 3 µM, with nearly complete block of BPIFA1 occurring at 30 µM J o u r n a l P r e -p r o o f Journal Pre-proof ( Fig. 3C and 3D) . Therefore, these results support the conclusion that BPIFA1 proteolysis can be attributed to an HDMassociated cysteine protease(s).

The various HDM allergens of different mite species have been classified according to structurally similar homologs [13] . Many of these groups in HDM have been identified as having protease activity (groups 1, 3, 6, and 9). Group 1 HDM allergens exhibit cysteine protease activity, and are responsible for more than 50% of sensitization of HDM-allergic subjects [38] . Because our findings suggested that HDM-mediated BPIFA1 degradation is blocked by a pan cysteine protease inhibitor, E64D (Fig.3) , we further examined which group 1 allergen is responsible for BPIFA1 degradation. 10 µg/ml of recombinant BPIFA1 protein was incubated with or without 100 µg/ml of HDM, Der p1, and Der f1 for 24 h, respectively. Immunoblot results showed that both HDM and Der p1 cleaved BPIFA1 completely, whereas Der f1 had no effect on BPIFA1 at all (Fig. 4A and 4B) . Moreover, the cleavage pattern of Der p1 also resembled that of HDM, suggesting that Der p1 is the active component in HDM that degrades BPIFA1. To further confirm the specificity of Der p1 cleavage, we again performed an E64D dose-course experiment with a fixed amount of recombinant BPIFA1 protein (10 µg/ml) and Der p1 (100 µg/ml). E64D showed a dose-dependent effect on blocking Der p1-mediated BPIFA1 degradation, with the lowest effective dose at 0.3 µM (Fig. 4C and 4D ). Taken together, these results suggest that Der p1 is the protease component of HDM that is responsible for BPIFA1 degradation.

To determine whether BPIFA1 cleavage occurs in vivo, we next tested whether BPIFA1 is cleaved in mouse BALF when exposed to an increased dose of HDM. We instilled 1 µg, 10 µg, and 50 µg HDM into mice airways respectively. 24 h after installation, BALF was collected and subject to immunoblot. Similar to what we observed in our in vitro studies, HDM decreased the amount of full-length BPIFA1 in a dose-dependent manner (Fig. 5A, upper panel; Fig 5B) , without significant impact on total protein amount (Fig. 5A, lower panel) . However, unlike human BPIFA1, we did not detect lower molecular weight fragments in these mice BALF samples. We speculate that this could be due to the sequence difference between human and mouse BPIFA1 protein. Previous research showed that human and mouse BPIFA1 proteins share ~72% identity [39] . Mouse BPIFA1 has an extra 24 amino acid sequence inserted in the corresponding N-terminal "S18" region in human protein [27] , with other gaps and amino acid changes spread along the sequence. These residue differences may affect BPIFA1 being recognized and bound by HDM and therefore lead to a different cleavage pattern.

A decrease in full-length BPIFA1 protein in vivo can be attributed to transcriptional downregulation rather than posttranslational degradation. To exclude the interference of transcriptional regulation, we adopted an ex vivo method using cell-free BALF from mice and incubate it with an increased amount of HDM extract for 24 h at 37 °C. Consistent with the results from in vivo studies, increasing doses of HDM resulted in a decrease trending of full-length BPIFA1 (Fig. S2) , with no small fragments. Although this does not rule out transcriptional regulation of BPIFA1 by HDM, it indicates that HDM-mediated protein degradation also plays a role in the downregulation of BPIFA1.

To assess the effect of HDM cleavage on BPIFA1, we first purified HDM-cleaved BPIFA1 using neutravidin and biotin-labeled HDM. Immunoblot results showed that biotin-HDM cleaves BPIFA1 in the same pattern as untagged HDM, and total BPIFA1 or BPIFA1 fragment amounts did not change after two rounds of purification (Fig. S1 , upper panel), whereas biotin-HDM was retained on neutravidin beads after the first round of incubation (Fig. S1, lower panel) .

We then examined intact and cleaved BPIFA1 for their ability to bind to LPS from E. coli. and found that when cleaved by HDM, BPIFA1 binding to LPS was significantly reduced (Fig. 6A ). Since previous research shows that BPIFA1 binds to the α4 region of LPS [24] , we conclude that HDM impairs this ability by cutting into the α4 helix of BPIFA1.

We also evaluated whether HDM cleavage affects BPIFA1's ability to regulate Ca 2+ influx, in light of our previous finding that BPIFA1 inhibits Ca 2+ influx in airway smooth cells by binding to Orai1 channel [32] . When incubated with Calu-3 cells, intact BPAIF1 protein decreased the TG-induced cytosolic Ca 2+ increase by half, whereas HDM-cleaved BPIFA1 completely lost this inhibitory effect ( Fig. 6B and S2) , suggesting that HDM also cleaves BPIFA1 in its cterminal α6 region.

A method that is widely used to study the activation of the innate immune system is in vitro stimulation of human whole blood using LPS. LPS-stimulated whole blood generally exhibits massive activation of a battery of pro-inflammatory cytokines, such as tumor necrosis factor (TNF)-α, Interleukin (IL)-1β, IL-6, IL-10, and interferon (IFN)-γ; it is therefore considered a credible tool to study sepsis and endotoxemia ex vivo [40] . Because our LPS binding assay suggests that HDM cleavage impairs BPIFA1's ability to bind to LPS, we incubated LPS-stimulated human whole blood in the presence of intact or HDM-cleaved BPIFA1 to investigate its effects on cytokine secretion pattern. As shown in Fig. 6C and 6D, 100 ng/ml LPS induced surges of TNFα and IL-6 secretion, which was inhibited by co-treating human whole J o u r n a l P r e -p r o o f Journal Pre-proof blood with intact BPIFA1. This inhibitory effect was partially (TNFα) or completely (IL-6) abolished with HDM-cleaved BPIFA1, suggesting that when cleaved by HDM, BPIFA1 lost its function to modulate the innate immune response.

Allergens associated with HDM (D. pteronyssinus and D. farinae) are a major cause of allergic diseases worldwide.

Prevalence data for HDM allergen sensitization as many as 130 million people in the general population worldwide to roughly 50% among asthmatic patients [9] . Given its high prevalence, it is necessary to understand the mechanisms involved in orchestrating the allergic response. Research from the past two decades of studying HDM and its components has suggested that allergenic effects in HDM allergy are triggered and perpetuated through two main routes: the IgEdependent allergic response and the innate immune system [41, 42] . The protease activity of some allergens has been shown to primarily contribute to their allergenicity via the IgE-dependent route [43] . This is best exemplified by the major HDM allergen Der p 1, which elicits IgE responses in more than 80% of patients who are sensitive to D. pteronyssinus and is the most immunodominant allergen involved in the expression of IgE-mediated dust mite hypersensitivity [44] .

Some of the substrates of Der p1 include human cell surface molecules such as the low-affinity IgE receptor (CD23/FcεRII), the loss of which from IgE-secreting B cells is thought to promote and enhance IgE immune responses by ablating an important feedback inhibitory mechanism that normally limits IgE synthesis [45, 46] . CD40 on dendritic cells (DCs) is also cleaved by Der p1, resulting in decreased IL-12 production, thus making DCs less responsive to stimulation through the CD40L-CD40 pathway and inducing generation of DCs that favor a T H 2 response [47] .

Despite a growing body of evidence concerning possible HDM/Der p1 substrates, only a handful has been identified and characterized. Besides the previously discussed CD23 and CD40, studies performed with protein substrates such as α1antitrypsin [48] and insulin β-chain [49] have shown that Der p 1 has a preference for small hydrophobic residues (Val/Ala) in the S 2 position, charged residues (Glu/Arg) in the S 1 position, and small hydrophilic/hydrophobic residues (Ser/Ala) in the S 1 ' position. Emerging in silico methods-in particular, a novel bioinformatics tool called prediction of protease specificity (PoPS; http://pops.csse.monash.edu.au/home.html)-have helped identify two salient molecules of immunological interest: DC-SIGN (CD209) and DC-SIGNR (CD299: L-SIGN) [50] . Otherwise, the lack of effective ways to screen for or predict potential protease substrates remains a bottleneck to our understanding of HDM/Der p1 allergenicity. To the best of our knowledge, the ability of HDM/Der p1 to degrade BPIFA1 has not been previously reported. However, neutrophil elastase, which is abundant in the lungs of cystic fibrosis (CF) patients, has been shown to J o u r n a l P r e -p r o o f Journal Pre-proof degrade BPIFA1 [51] . This degradation contributes to increased ENaC activity and ASL dehydration in CF HBECs [19] .

Our findings add new evidence that the substrates of HDM/Der p1 proteases are not limited to cell surface molecules.

Secretory proteins, such as BPIFA1, are also subject to proteolytic effects, which implies potential unidentified candidate proteins, as well as underlying regulatory mechanisms.

The data presented in this study demonstrate that BPIFA1 is subject to proteolytic degradation by HDM, leading to its loss of function in modulating Ca 2+ influx, E.coli LPS binding and the E.coli LPS-induced TNFα and IL-6 production in whole human blood cells. Further, our data suggest that HDM degrades BPIFA1 protein at a broad range of doses and various ambient temperatures ( Fig.1 and Fig.2) , which reflects the diverse HDM amount and environmental conditions that people are exposed to daily basis. Although it is experimentally challenging to reproduce these variable exposure conditions to reliably predict the quantity of HDM or associated components that end up in the airway, evidence shows that HDM is found in the sputum and even deep lungs [52] , which supports our hypothesis that HDM directly interacts and degrades BPIFA1. We have also demonstrated that the main component responsible for HDM-mediated BPIFA1 degradation is Der p1, a cysteine protease ( Fig. 3 and Fig. 4) . Future studies should to be carried out to identify the cleavage sites of HDM/Der p1 within BPIFA1, thereby enabling the generation of HDM/Der p1-resistant mutants. Reintroducing these mutant proteins into patients' airways presents a possible avenue for a novel asthma treatment strategy.

Functional studies using full-length and cleaved BIPFA1 protein indicate that HDM cleavage impairs several functions of BPIFA1, including LPS binding, Ca 2+ regulation, and cytokine secretion (Fig. 6) . These findings suggest the existence of multiple HDM/Der p1 cleavage sites throughout the BIPFA1 protein; HDM/Der p1 will "hit" all of these sites given sufficient time and dosage. It will be critical to establish the sequence of these events so that we can gain more knowledge on the HDM-BPIFA1 dynamic and identify the most "vulnerable" part of BPIFA1 protein. Another question that stems from these functional studies of BPIFA1 is whether the LPS binding and cytokine modulating are essentially the same function. It is commonly believed that apically secreted BPIFA1 normally 'mops up' excess bacterial LPS; following allergen exposure, reduced BPIFA1 levels increases LPS, leading to exaggerated inflammatory cytokine secretion and allergic response [31] . In our study, when incubated with whole human blood in the absence of E.coli LPS, full-length BPIFA1 did not significantly affect either TNFα or IL-6 secretion. However, in the presence of LPS, full-length-but not HDM-cleaved-BPIFA1 effectively inhibited the secretion of these cytokines (Fig. 6C and Fig. 6D ). We can only speculate from these data that HDM simultaneously impairs the domains in BPIFA1 that are responsible for binding to J o u r n a l P r e -p r o o f Journal Pre-proof LPS and regulating cytokine secretion. To further delineate the relationship between the structure and function of BPIFA1, we will need to test whether BPIFA1 Δα4 -a BPIFA1 mutant that lacks the LPS binding α4 region-exerts the same immune-modulating functions as the full-length protein, under both basal and LPS challenged conditions. The findings of this study provide substantial in vitro and ex vivo evidence that HDM-especially its cysteine protease component Der p1-cleaves BPIFA1 and thereby negatively impacts a host of BPIFA1-mediated antimicrobial and antiinflammatory functions. We also detected a dose-dependent decrease of full-length BPIFA1 protein in the presence of HDM in mouse BALF without significant change in total protein amount, indicating the substrate specificity of HDM.

There was a discrepancy in HDM cleavage patterns between human and mouse BPIFA1 proteins, which we attribute to the amino acid sequence differences between these two species and possible HDM preference for amino acid residues around cutting sites. Our data will expand the knowledge regarding environmental allergen-induced airway diseases and potentially lead to novel protein/peptide-based biologics that are resistant to the highly proteolytic environment typical of many airway diseases.

Our finding that BPIFA1 reduces LPS induced IL-6 and TNFα production are timely in light of the current pandemic, since the fatal outcome observed with coronavirus 2019 (COVID-19) infection often results from diffuse alveolar damage, desquamation of pneumocytes, and hyaline membrane formation, indicating acute respiratory distress syndrome (ARDS) and multi-organ failure-both of which are associated with the hyperproduction of cytokines, also known as a cytokine storm or cytokine release syndrome. Clinical reports show that both mild and severe forms of disease result in changes in circulating leukocyte subsets and cytokine secretion, particularly IL-6, IL-1β, IL-10, TNF, Granulocyte-macrophage colony-stimulating factor (GM-CSF), IFN-induced protein 10 (IP-10), IL-17, Chemokine (C-C motif) ligand 7 (CCL7), and interleukin-1 receptor antagonist (IL-1ra) [53] [54] . Therefore, anti-cytokine interventions, such as blocking IL-6 production/secretion, appear to be potentially efficacious and proactive strategies against the adverse clinical outcomes caused by COVID-19. Our data showed that BPIFA1 reduces LPS induced IL-6 and TNFα production in whole human blood cells and that this inhibitory effect was abolished by HDM protease (Fig.6C and 6D) . Such results raise more questions for our future studies: First and foremost, can we develop BPIFA1 and its derived peptides into biologics to reduce IL-6-initiated inflammation in COVID-19 infected lungs and thereby prevent further damage? Based on current results, we speculate that peptides will be more preferable since they will be less susceptible to protease degradation, which is commonly seen in highly inflamed airways [55] . Secondly, since BPIFA1 (and other lung surfactant J o u r n a l P r e -p r o o f Journal Pre-proof proteins) is subject to protease degradation in diseased airways, is there a correlation between the severity of the disease/infection and residual BPIFA1 amount? In other words, what is the interaction between COVID-19, BPIFA1, environmental proteases, and host immune response? Answering these questions will provide comprehensive information on how BPIFA1 and its derivative peptides could potentially be used for treating COVID-19-induced inflammation. These answers could also potentially open up a new avenue of research into airway inflammation, with a focus on developing alternative anti-inflammatory drugs based on proteins composing our immune defense mechanisms.

The authors declare that they have no competing interests. 

Asthma's Impact on the Nation. Data from the CDC National Asthma Control Program

Dendritic cells and epithelial cells: linking innate and adaptive immunity in asthma

The role of allergen exposure and avoidance in asthma, Adolescent medicine: state of the art reviews

The cell biology of asthma

House dust mite interactions with airway epithelium: role in allergic airway inflammation

How is the indoor environment related to asthma?: literature review

Respiratory allergy caused by house dust mites: What do we really know?

Innate immune responses in house dust mite allergy

Orchestrating house dust mite-associated allergy in the lung

Proteases as Th2 adjuvants

Allergens with Protease Activity from House Dust Mites

Differential display identification of plunc, a novel gene expressed in embryonic palate, nasal epithelium, and adult lung

Identification of human whole saliva protein components using proteomics

PLUNC (palate, lung and nasal epithelial clone) proteins in human nasal lavage fluid

Molecular cloning and characterization of spurt, a human novel gene that is retinoic acid-inducible and encodes a secretory protein specific in upper respiratory tracts

Purification and characterization of PLUNC from human tracheobronchial secretions

PLUNC is a secreted product of neutrophil granules

Shotgun proteomic analysis of human-induced sputum

Molecular basis for pH-dependent mucosal dehydration in cystic fibrosis airways

SPLUNC1 promotes lung innate defense against Mycoplasma pneumoniae infection in mice

Multifunctional role of human SPLUNC1 in Pseudomonas aeruginosa infection

Structural Features Essential to the Antimicrobial Functions of Human SPLUNC1

Structural characterization of the pulmonary innate immune protein SPLUNC1 and identification of lipid ligands

Identification of the SPLUNC1 ENaC-inhibitory domain yields novel strategies to treat sodium hyperabsorption in cystic fibrosis airway epithelial cultures

The Role of BPIFA1 in Upper Airway Microbial Infections and Correlated Diseases

Bactericidal/Permeability-increasing protein fold-containing family member A1 in airway host protection and respiratory disease

Function and regulation of SPLUNC1 protein in Mycoplasma infection and allergic inflammation

SPLUNC1 deficiency enhances airway eosinophilic inflammation in mice

House dust mite allergen in US beds: results from the First National Survey of Lead and Allergens in Housing

Recombinant house dust mite allergens

Hierarchy and molecular properties of house dust mite allergens

House dust mite major allergens Der p 1 and Der p 5

activate human airway-derived epithelial cells by protease-dependent and protease-independent mechanisms

Effect of mite allergenic components on innate immune response: Synergy of protease (Group 1 & 3) and non-protease (Group 2 & 7) allergens

Characterisation of the human plunc gene, a gene product with an upper airways and nasopharyngeal restricted expression pattern

Human endotoxemia: a model for mechanistic insight and therapeutic targeting

The innate immune response in house dust mite-induced allergic inflammation

House dust mite allergy: Its innate immune response and immunotherapy

A mite subversive: cleavage of CD23 and CD25 by Der p 1 enhances allergenicity

Mite faeces are a major source of house dust allergens

Recombinant Der p 1 and Der f 1 with in vitro enzymatic activity to cleave human CD23, CD25 and alpha1-antitrypsin, and in vivo IgE-eliciting activity in mice

A major house dust mite allergen disrupts the immunoglobulin E network by selectively cleaving CD23: innate protection by antiproteases

The proteolytic activity of the major dust mite allergen Der p 1 conditions dendritic cells to produce less interleukin-12: allergen-induced Th2 bias determined at the dendritic cell level

The house dust mite allergen

Der p1 catalytically inactivates alpha 1-antitrypsin by specific reactive centre loop cleavage: a mechanism that promotes airway inflammation and asthma

Heterogeneous proteolytic specificity and activity of the house dust mite proteinase allergen Der p I

The protease allergen Der p 1 cleaves cell surface DC-SIGN and DC-SIGNR: experimental analysis of in silico substrate identification and implications in allergic responses

Human neutrophil elastase degrades SPLUNC1 and impairs airway epithelial defense against bacteria

Dust mites living in human lungs--the cause of asthma?

Pathological findings of COVID-19 associated with acute respiratory distress syndrome

Cytokine storm and leukocyte changes in mild versus severe SARS-CoV-2 infection: Review of 3939 COVID-19 patients in China and emerging pathogenesis and therapy concepts

B are mean±s.e.m., and were analyzed using one-way ANOVA followed by Tukey post-hoc analysis

Calu-3 cells were incubated with intact BPIFA1 or HDM-cleaved BPIFA1 proteins, and changes in the fluo-4 emission ratio over time were recorded. ΔF/F 0 represents the average peak fluorescence intensity increase after TG addition (n=5). (C) and (D) are ELISA readouts of TNFα and IL-6 produced by human whole blood cells in the presence or absence of BPIFA1 (n=5). Data are mean±s.e.m., and were analyzed using one-way ANOVA followed by Tukey post-hoc analysis

We thank CF Center Histology, Tissue and Molecular Cores and Dr. Robert Tarrant