Exogenous and Endogenous Triggers Differentially Stimulate Pigr Expression and Antibacterial Secretory Immunity in the Murine Respiratory Tract

Purpose Transport of secretory immunoglobulin A (SIgA) through the airway epithelial cell barrier into the mucosal lumen by the polymeric immunoglobulin receptor (pIgR) is an important mechanism of respiratory mucosal host defense. Identification of immunomodulating substances that regulate secretory immunity might have therapeutic implications with regard to an improved immune exclusion. Thus, we sought to analyze secretory immunity under homeostatic and immunomodulating conditions in different compartments of the murine upper and lower respiratory tract (URT&LRT). Methods Pigr gene expression in lung, trachea, and nasal-associated lymphoid tissue (NALT) of germ-free mice, specific pathogen-free mice, mice with an undefined microbiome, as well as LPS- and IFN-γ-treated mice was determined by quantitative real-time PCR. IgA levels in bronchoalveolar lavage (BAL), nasal lavage (NAL), and serum were determined by ELISA. LPS- and IFN-γ-treated mice were colonized with Streptococcus pneumoniae and bacterial CFUs were determined in URT and LRT. Results Respiratory Pigr expression and IgA levels were dependent on the degree of exposure to environmental microbial stimuli. While immunostimulation with LPS and IFN-γ differentially impacts respiratory Pigr expression and IgA in URT vs. LRT, only prophylactic IFN-γ treatment reduces nasal colonization with S. pneumoniae. Conclusion Airway-associated secretory immunity can be partly modulated by exposure to microbial ligands and proinflammatory stimuli. Prophylactic IFN-γ-treatment modestly improves antibacterial immunity in the URT, but this does not appear to be mediated by SIgA or pIgR. Supplementary Information The online version contains supplementary material available at 10.1007/s00408-021-00498-8.


Introduction
Airway epithelial cells (AECs) constitute the first line of defense against respiratory pathogens. They express transmembrane proteins, which form tight junctions that allow only small ions or water to traverse paracellularly [1]. Claudin and occludin are vital for epithelial defense [2] and altered claudin expression affects airway epithelial barrier function [3,4]. Furthermore, AECs constitutively secrete antimicrobial proteins, complement factors and cytokines, and rapidly mount antimicrobial immune responses upon inflammatory and infectious stimuli [5][6][7][8][9]. Importantly, AECs have a central function in antibodymediated mucosal immunity. Multimeric IgA and IgM are actively transported through AECs via the polymeric immunoglobulin receptor (PIGR) and are secreted into the mucosal lumen as secretory immunoglobulins (SIgs) [10]. Especially SIgA is known to prevent pathogen adhesion, thus averting microbial infiltration [11]. Moreover, SIgA plays a crucial role in the regulation of Streptococcus pneumoniae nasal colonization in mice [12].
In this context, Pigr deficiency manifests in susceptibility to mycobacterial respiratory infections [13] and development of a COPD-like phenotype driven by an altered lung microbiome and bacterial invasion of the airway epithelium [14]. The importance of SIgA for airway homeostasis is furthermore highlighted by the findings of SIgA deficiency in small airways of COPD patients, which is associated with persistent inflammation and airway wall remodeling [15]. Moreover, chronic airway diseases reduce PIGR expression in the bronchial epithelium resulting in increased disease severity (COPD) and impaired SIgA-mediated mucosal defense (asthma) [16,17]. While the key role of PIGR and secretory immunity for airway homeostasis is undisputable, knowledge on their expression and regulation in the airways is still fragmentary. Since targeted modulation of secretory immunity represents an interesting option to improve immune exclusion of respiratory pathogens, we here aimed to further dissect PIGR-mediated immunity in the airways with the specific focus on the applicability of exogenous and endogenous stimuli to regulate this aspect of humoral antimicrobial defense.

Treatment with Immunomodulating Substances
BALB/c, C57BL/6 J, and C57BL/6JRj mice were treated intranasally (i.n.) with LPS (Sigma-Aldrich, Germany), (10 µg/ 25 µl PBS or solvent alone) or recombinant murine IFN-γ (Peprotech, Germany), (1 µg/ 20 µl ddH 2 O with 5% BSA or solvent alone). BALB/c and C57BL/6 J mice were sacrificed 1 or 2 days post-treatment. Lung, trachea, NALT, BAL, NAL, and serum were collected. Organs were used for RNA isolation and qPCR. Fluids were used for ELISA analysis. Blood was collected by cardiac puncture. BAL fluid was collected by flushing the lungs with 1 ml PBS via the trachea. The nasopharynx was flushed with 1 ml PBS via the trachea and NAL fluid was collected at the nostrils.

Enzyme-Linked Immunosorbent Assay (ELISA)
Relative IgA levels (including monomeric IgA, dimeric IgA, and SIgA molecules) in BAL, NAL, and serum were determined by ELISA using a monoclonal rat anti-mouse IgA capture antibody (Southern Biotech, USA) in combination with a polyclonal rabbit anti-mouse IgA secondary antibody (Abcam, UK) and a polyclonal swine anti-rabbit, HRPOlinked detection antibody (Dako, UK).

Statistical Analysis
Statistical analyses were performed either by two-tailed, unpaired t test (Gaussian distribution, two groups), Mann-Whitney test (no Gaussian distribution, two groups), one-way ANOVA (Gaussian distribution, more than two groups, post-test: Bonferroni's Multiple Comparison Test), or Kruskal-Wallis test (no Gaussian distribution, more than two groups, post-test: Dunn's Multiple Comparison Test) using GraphPad Prism software (GraphPad Software Inc., USA, Version 5.04).

Results
In order to determine differences in Pigr gene expression in the upper (URT) and lower respiratory tract (LRT) and its dependency on genetic background, sex, and age, we initially compared Pigr expression patterns in commonly used mouse strains and different sexes. BALB/c (Fig. 1a) as well as C57BL/6 J mice ( Basal Pigr expression in the airways. RNA from lung tissue, trachea, and nasal-associated lymphoid tissue (NALT) of a BALB/c mice (n = 5) and b C57BL/6 J mice (n = 14) was isolated and reversely transcribed into cDNA. Pigr expression was assessed by qPCR. Actb served as reference gene (data for individual mice are graphed; mean is indicated by horizontal line; ** for p ≤ 0.01; *** for p ≤ 0.001) expression is age-dependent (Fig. S2a-c) but found no such correlations. Together, these data demonstrate marked tissue-specific differences in Pigr gene expression, which were however independent of genetic background, sex, and age.
To determine whether Pigr expression and secretory immunity in the airways were influenced by microbial exposure we compared germ-free (GF) mice (no microbial exposure), SPF mice (IVCs, exposure to a limited microbial flora), and mice with an undefined microbiome (open cage maintenance, highest degree of exposure to airborne microorganisms). While similar Pigr expression levels were observed in the LRT of all three experimental groups, we detected significantly lower Pigr expression in the NALT of mice with an undefined microbiome compared to SPF mice (Fig. 2a). In contrast to the unaltered (lung, trachea) or even reduced (NALT) Pigr expression levels in mice with an undefined microbiome, we detected significantly increased IgA concentration in the LRT (Fig. 2b) and URT (Fig. 2c) in this group, which was associated with a systemic IgA increase (Fig. 2d).
Since in the intestine IgA production and Pigr expression are induced by microbial components [20][21][22], we investigated whether intranasal (i.n.) treatment of mice with LPS would affect airway Pigr expression and IgA levels. While Pigr expression in nose and trachea was not affected, we detected significantly increased expression in lung tissue 48 h after LPS treatment (Fig. 3a). Interestingly and in discordance with the increased pulmonary Pigr expression IgA levels in BAL and NAL fluid significantly decreased after LPS treatment. The amount of IgA in serum was however not affected (Fig. 3b). We thus speculated that LPS altered the barrier function of AECs, resulting in a decreased epithelial leakage. Therefore, we analyzed Cldn gene expression in lung and NALT. However, no significant differences in lung or NALT became apparent upon LPS treatment (Fig. 3c, d).
Next to LPS, Interferon-γ (IFN-γ) was shown to regulate human PIGR gene expression [23] and we confirmed the Pigr-inducing potential of this cytokine in murine AECs (Fig. S3). To assess possible effects of IFN-γ on PIGR-mediated secretory immunity in vivo, we analyzed airway secretory immunity in IFN-γ-treated mice. While IFN-γ did not affect Pigr expression after 48 h (Fig. 4a), the amount of pulmonary IgA increased after IFN-γ treatment. IgA levels in NAL and serum were however unaffected (Fig. 4b). We tested whether epithelial leakage might underlie the increased IgA levels by measuring Cldn gene expression in lung and NALT. Cldn gene expression was however not affected by single IFN-γ treatment (Fig. 4c, d) and NALT. Nevertheless, airway IgJ expression was unaltered upon cytokine treatment (Fig. 4e, f), indicating that increased airway IgA levels are most likely not due to an IFN-γ-mediated increase of Ig-producing cells.
We finally tested whether modulation of airway secretory immunity by LPS-and IFN-γ treatment would ultimately affect antimicrobial defense. To this end, mice were i.n. treated with a single dose of LPS or IFN-γ. Two days post-treatment mice were inoculated with a colonizing strain of S. pneumoniae serotype 19F and airway bacterial burden was assessed (Fig. 5a). As expected, no pneumococci were detected in the lung tissue. While LPS treatment did not affect pneumococcal colonization (Fig. 5b), IFN-γ treatment led to significantly decreased nasal bacterial burden (Fig. 5c). These results demonstrate that at least in the URT mucosal immunity can be augmented by prophylactic IFN-γ treatment.

Discussion
Previous studies on airway Pigr gene and PIGR protein expression mainly employed in vitro approaches [24,25] or utilized tissue from patients with chronic respiratory diseases [16,[26][27][28]. Murine studies were either analyzing respiratory Pigr expression in the context of interleukin treatment [29,30] or exposure to pathogen-associated molecules (Cholera toxin, amoeba lysates) [31]. These studies revealed that PIGR/Pigr expression is considerably influenced by exogenous and endogenous stimuli present in the airway microenvironment.
To our knowledge, we are first to report compartmentspecific and sex-independent differences in basal airway Pigr gene expression levels in vivo. We observed that BALB/c as well as C57BL/6 J mice showed highest Pigr expression in trachea, followed by NALT and lung. We speculated that this originated from inherent differences in microbial density. Due to their anatomical localization, trachea and nasal cavities are more frequently exposed to microbial stimuli compared to the lung [32]. It is possible that a higher abundance of microbial ligands in the murine URT provide more signals triggering Pigr gene expression compared to the LRT, as well. This in turn might prevent bacterial spread from URT to LRT contributing to the relatively low bacterial density in the lung. Contrary to our expectations, we could not detect any alterations regarding Pigr expression between germ-free and microbially colonized mice, which suggests that Pigr expression in the murine respiratory tract is in fact unaffected by microbial colonization. However, as we assessed whole-tissue Pigr gene expression, we cannot fully exclude the possibility that signal dilution effects (e.g., from leukocytes) affect compartment-specific Pigr expression in our analyses.
Commensal intestinal bacteria induce the production of IgA in mice [20]. Furthermore, it is known that lymphocyte numbers in nasal mucosa are dependent on housing conditions and exposure to microbial stimuli [33]. While those findings clearly highlight the impact of the microbiota on lymphocyte-associated mucosal immunity, the relationship between airway-associated secretory immunity and the level of microbial exposure is largely unknown. Our experiments revealed that Pigr expression was lower in NALT of mice with an undefined microbiome (highest microbial exposure) compared to mice maintained under SPF conditions. At the same time, airway and systemic IgA levels were increased in these mice. As fecal IgA levels depend on the composition of the intestinal microflora [34], it is conceivable that a similar effect might be present in murine airways. However, the fact that Pigr expression in NALT is reduced while IgA levels are increased indicates that there is likely no correlation between Pigr expression and IgA abundance at the wholetissue level. However, we did not analyze Pigr expression exclusively in stromal cells but in whole tissue. Since it was shown that the microbial environment shapes cell composition in the mucosa, it is possible that an accumulation of leukocytes in mice with a high microbial exposure reduces the overall Pigr signal.
Previous studies revealed that human and murine intestinal epithelial cells exhibit increased Pigr/PIGR gene expression after LPS stimulation in vitro [22,35,36]. In line with this, we detected significantly increased pulmonary Pigr expression after LPS treatment. However, we also found that LPS did not alter overall Pigr expression in trachea or NALT. This might arise from the fact that bacterial colonization-and therefore exposure to e.g., LPS-is more pronounced in the URT [32], resulting in a lower sensitivity of URT airway stromal cells to LPS. In contrast to this result, airway IgA levels were decreased in LPS-treated mice, while systemic IgA levels were unaffected. We hypothesized that decreased IgA levels resulted from decreased epithelial leakage and tested this by determining Cldn18 and Cldn7 expression. Claudins are major proteins that maintain epithelial barrier function and altered claudin expression results in altered AEC barrier function in the LRT (Cldn18) and URT (Cldn7) [2][3][4]. However, Cldn18 and Cldn7 expression were not affected by LPS, which disconfirmed our hypothesis.
Since IgA binds LPS [37], it is possible that the administered LPS was already bound to IgA in the mucosal lumen. This might reduce the amount of detectable IgA, as the ELISA detects free IgA molecules with the highest functionality. However, the fact that Pigr expression in lung was increased while IgA in BAL was decreased and Pigr expression in NALT was unaltered while IgA in NAL was increased highlights that there is no consistent correlation between the two molecules in whole-tissue analysis of LPS-treated mice.
As LPS, IFN-γ induces PIGR expression in human epithelial cells [23,38,39]; however, its effect on Pigr and secretory immunity in vivo have not been addressed before. Despite no effect on airway Pigr expression, IFN-γ treatment increased pulmonary IgA levels. As Cldn expression was unaltered, we hypothesized that increased IgA concentrations after IFN-γ treatment might arise from mucosal B cells. Yet, airway IgJ expression was not affected, which suggests that activated B cells are most likely not the cause of increased airway IgA levels following IFN-γ stimulation. As mentioned before, we analyzed Pigr gene expression in tissues and not exclusively in stromal cells. It is known that intradermal IFN-γ injection stimulates intradermal lymphocyte migration in rats [40]. Thus, it is conceivable that i.n. IFN-γ treatment leads to the accumulation (and activation) of lymphocytes in the airways as well, which might reduce net Pigr expression. This might explain why there is no consistent correlation between Pigr expression and IgA levels in whole-tissue analysis of IFN-γ-treated mice.
As IgA is crucial for antimicrobial defense [15,[41][42][43], we investigated whether altered IgA levels upon LPS-and IFN-γ treatment correlated with altered antimicrobial immunity. We have chosen S. pneumoniae for experimental colonization of mice as it is one of the most relevant respiratory pathogens [44] and IgA is vital for antagonizing pneumococcal colonization and infection in vivo and in vitro [12,[45][46][47]. Indeed, prophylactic IFN-γ treatment significantly reduced nasal pneumococcal counts indicating improved antibacterial immunity. Since IFN-γ treatment led to increased IgA levels only in BAL and not in NAL, the effect IFN-γ has on colonization is most likely not mediated by its effect on IgA. Since IFN-γ triggers antibacterial activity in pulmonary macrophages [48] and macrophages are present in the murine NALT [49], it is conceivable that IFN-γ induces antibacterial activity in these cells as well. IFN-γ triggers the production of antibacterial molecules (e.g., β-defensins) [50]. Future studies will clarify whether IFN-γ-stimulated production of these antibacterial factors underlies the improved mucosal immunity in the URT.
In conclusion, our study demonstrates that secretory immunity in URT and LRT is differentially regulated by endogenous as well as exogenous stimuli. Further studies are needed to elucidate the underlying molecular frameworks as well as possible avenues for, e.g., prophylactic enhancement of airway mucosal immunity in infection-prone individuals.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.