Chronic lung inflammation primes humoral immunity and augments antipneumococcal resistance

Abstract

Airway epithelial cells (AECs) display remarkable plasticity in response to infectious stimuli and their functional adaptations are critical for antimicrobial immunity. However, the roles of AECs and humoral mediators to host defense in non-communicable lung inflammation remain elusive. We dissected pulmonary defense against Streptococcus pneumoniae in hosts with pre-existing inflammatory conditions (SPC-HAxTCR-HA mice). Lung tissue transcriptomics and bronchoalveolar lavage fluid (BALF) proteomics revealed an induction of humoral defense mechanisms in inflamed lungs. Accordingly, besides antibacterial proteins and complement components being overrepresented in inflamed lungs, elevated polymeric immunoglobulin receptor (pIgR)-expression in AECs correlated with increased secretory immunoglobulin (SIg) transport. Consequently, opsonization assays revealed augmented pneumococcal coverage by SIgs present in the BALF of SPC-HAxTCR-HA mice, which was associated with enhanced antipneumococcal resistance. These findings emphasize the immunologic potential of AECs as well as their central role in providing antibacterial protection and put forward pIgR as potential target for therapeutic manipulation in infection-prone individuals.

Figure 1

Transcriptional profile of whole lung tissue from SPC-HAxTCR-HA vs. SPC-HA mice. RNA from lung tissue of SPC-HA and SPC-HAxTCR-HA mice (n = 3/group) was isolated and samples were individually analyzed on whole transcriptome microarrays. Differential expression was analyzed by one-way ANOVA with p-value cutoff p < 0.05 and fold change cutoff FC > ±2 comparing SPC-HAxTCR-HA vs. SPC-HA. Fold changes were calculated based on group-average of signal intensities. (a) Normalized log₂ signal intensities of significantly regulated transcripts from each individual microarray replicate were clustered according to k-means clustering (cluster 1–4). Color code represents z-score. Gene symbols of selected prominent transcripts within each cluster are stated. (b) Volcano scatter plot of significantly regulated transcripts with cluster assignment referring to (a). Gene symbols of selected prominent transcripts are stated. SxT = SPC-HAxTCR-HA; FC = fold change. (c) Gene Set Enrichment Analysis (GSEA) for canonical hallmark gene sets from the Molecular Signature Database (MSigDB) of microarray data comparing SPC-HAxTCR-HA vs. SPC-HA (FDR < 5%).

Figure 2

Chronic lung inflammation alters the BALF proteome. (a) Venn diagram comparing LC-MS/MS-identified proteins from bronchoalveolar lavage fluid (BALF) of SPC-HA vs. SPC-HAxTCR-HA mice. Samples were pooled from 5 mice/group. (b) Venn diagram comparing genes with ANOVA p-value p < 0.05 and a fold change FC > ±2 in lungs of SPC-HAxTCR-HA mice vs. genes (identified by their respective products) only detected in BALF from SPC-HAxTCR-HA mice.

Figure 3

Increased mucosal transport of secretory antibodies in inflamed lungs. (a) anti-pIgR and anti-Gapdh immunoblot of 20 µg protein from whole lung homogenates of SPC-HA (n = 3) and SPC-HAxTCR-HA (n = 3) mice. Densitometric quantification of protein bands is stated in arbitrary units above each lane. Relative pIgR quantity was calculated normalizing densitometric pIgR value to the corresponding Gapdh value and subsequently comparing normalized pIgR values of the SPC-HAxTCR-HA group to the SPC-HA group. Data are representative for at least two individual experiments with similar results. (b) Lung tissue sections were stained with anti-pIgR (green), representative alveolar structures from n = 3/group are depicted. White circles illustrate representative densitometrically quantified tissue areas. Calculated total cell fluorescence (CTCF) was determined as: Integrated density of fluorescence-positive cell – (Area of fluorescence-positive cell × mean fluorescence intensity of background signal). Median CTCF of quantified areas in representative images are depicted as white numbers. IgA and IgM levels in (c) bronchoalveolar lavage fluid (BALF) and (d) serum of SPC-HA and SPC-HAxTCR-HA mice were determined by ELISA. (e) Relative secretory IgA concentrations in serial dilutions of BALF samples were determined by ELISA. Results are expressed as the mean optical density (OD) at 450 nm ± SEM, *p < 0.05, ** p < 0.01 (n = 6–7/group).

Figure 4

Increased pneumococcal binding capacities by lung mucosal fluid in inflamed lungs. Pneumococci were co-incubated with bronchoalveolar lavage fluid (BALF) supernatants from SPC-HA and SPC-HAxTCR-HA mice. Bacteria were stained with anti-mouse IgA or anti-IgM antibodies and analyzed by flow cytometry (FACS). (a) Representative FACS plots of IgA+ pneumococci incubated with BALF from SPC-HA or SPC-HAxTCR-mice; control samples (CTRL) were stained with anti-IgA without prior incubation with BALF. (b) Percentages of IgA+ pneumococci and relative fluorescence intensities (c) of IgA+ pneumococci. (d) Representative FACS plots of IgM+ pneumococci incubated with BALF from SPC-HA or SPC-HAxTCR-mice; control samples (CTRL) were stained with anti-IgM without prior incubation with BALF. (e) Percentages of IgM+ pneumococci and relative fluorescence intensities (f) of IgM+ pneumococci. Relative fluorescence intensities are calculated by the ratio of the MFI of each individual sample over the mean MFI of the SPC-HA control group. Data are pooled from 2 independent experiments with similar results. *p < 0.05 **p < 0.01, ***p < 0.001.

Figure 5

Improved bacterial clearance is not associated with increased AM numbers nor with increased intra-pulmonary neutrophil recruitment in SPC-HAxTCR-HA mice. Mice were oropharyngeally inoculated with ~1.5 × 10⁴ CFU S. pneumoniae TIGR4 or were mock-treated (PBS). At 24 h p.i. mice were sacrificed and phagocytic leukocyte subsets in the BALF (a,b) were quantified by flow cytometry. (c) BALF supernatants were anylyzed by ELISA for levels of complement component C5a. Data were pooled from at least 2 independent experiments (n = 5–8/group). **p < 0.01.

Figure 6

Improved antipneumococcal resistance in pre-diseased SPC-HAxTCR-HA mice. Mice were oropharyngeally inoculated with ~1.5 × 10⁴ CFU S. pneumoniae TIGR4 and were sacrificed at the indicated time points. The bacterial burden in bronchoalveolar lavage fluid (BALF) (a) and lung tissue homogenates (b) was determined. BALF and lung homogenates were taken from the same mice. Data were pooled from 2 independent experiments. Dashed lines indicate the detection limit. *p < 0.05. (c) Mice were oropharyngeally inoculated with ~1.5 × 10⁶ CFU S. pneumoniae TIGR4 and survival was followed over a period of 14 days. *p < 0.05 (n = 15/group, 2 independent experiments).