MicroRNA-223 Dampens Pulmonary Inflammation during Pneumococcal Pneumonia

Abstract

Community-acquired pneumonia remains a major contributor to global communicable disease-mediated mortality. Neutrophils play a leading role in trying to contain bacterial lung infection, but they also drive detrimental pulmonary inflammation, when dysregulated. Here we aimed at understanding the role of microRNA-223 in orchestrating pulmonary inflammation during pneumococcal pneumonia. Serum microRNA-223 was measured in patients with pneumococcal pneumonia and in healthy subjects. Pulmonary inflammation in wild-type and microRNA-223-knockout mice was assessed in terms of disease course, histopathology, cellular recruitment and evaluation of inflammatory protein and gene signatures following pneumococcal infection. Low levels of serum microRNA-223 correlated with increased disease severity in pneumococcal pneumonia patients. Prolonged neutrophilic influx into the lungs and alveolar spaces was detected in pneumococci-infected microRNA-223-knockout mice, possibly accounting for aggravated histopathology and acute lung injury. Expression of microRNA-223 in wild-type mice was induced by pneumococcal infection in a time-dependent manner in whole lungs and lung neutrophils. Single-cell transcriptome analyses of murine lungs revealed a unique profile of antimicrobial and cellular maturation genes that are dysregulated in neutrophils lacking microRNA-223. Taken together, low levels of microRNA-223 in human pneumonia patient serum were associated with increased disease severity, whilst its absence provoked dysregulation of the neutrophil transcriptome in murine pneumococcal pneumonia.

Keywords: Streptococcus pneumoniae; inflammation; microRNA-223; neutrophils; pneumonia.

Figure 1

miR-223 is differentially regulated in human serum, murine lung tissue and pulmonary neutrophils during pneumococcal pneumonia. (A) miR-223 was quantified in the serum of healthy subjects and CAP patients using qRT-PCR (n = 50 and n = 92, respectively). CAP patients were grouped by CURB-65 scores and the proportional odds model was utilized to analyze the correlation of serum miR-223 relative abundance and disease severity. Spearman correlation was performed to determine the relationship between CRP and serum miR-223 in CAP patients. (B) Expression of miR-223 in vitro and in vivo in WT naïve, sham- and S.pn. ST2-infected mice. Isolated BM-PMN were stimulated 2 and 6 hpi with S.pn. ST2 (n = 9) or sham (n = 5 and n = 7). WT mice were intranasally infected with S.pn. ST2 (n = 7; 24 hpi, n = 10; 48 hpi) or sham (n = 11; 24 hpi, n = 9; 48 hpi), followed by quantification of miR-223 in whole lungs. Expression of miR-223 was also determined in sorted lung PMN (n = 3; naïve, n = 6; 24 hpi S.pn. ST2). (A,B) Mann-Whitney U test, * p < 0.05, **** p < 0.0001. (A) Proportional odds model, p < 0.05, β = −0.91; Spearman correlation, p < 0.05, r = −0.2439. (B) 2-way ANOVA/Šidák’s multiple comparisons test, ** p < 0.01, *** p < 0.001. Data in (A,B) display individual values and median or mean. Error bars indicate (A) minimum to maximum values (CURB-65 plot) and SEM. CAP: community-acquired pneumonia, RQ: relative quantification, CRP: C-reactive protein, CURB-65: (confusion, urea nitrogen, respiratory rate, blood pressure, 65 years of age and older), S.pn. ST2: Streptococcus pneumoniae serotype 2, hpi: hours post-infection, PBS: phosphate-buffered saline.

Figure 2

miR-223^−/− mice exhibit aggravated lung inflammation 48 hpi S.pn. ST2. (A) Physiological parameters, including changes in body temperature and body weight post-S.pn. infection in WT and miR-223^−/− mice. (B) Bacterial burden recorded in lungs, BAL, spleen and blood in WT and miR-223^−/− mice 48 hpi (n = 10–12). (C) Histopathological analyses of murine lungs 48 hpi. H&E staining indicates predominant neutrophilic and lymphocytic cellular infiltrates in lungs of miR-223^−/− and WT mice, respectively. Pictures are representative of overall inflammation in WT and miR-223^−/− mice. * denotes predominantly neutrophilic infiltrates coupled to pronounced edema, whilst # denotes perivascular lymphocytic cuff formation coupled to marginal edema. Pleuritis, steatitis, perivascular edema and hemorrhage scores collectively represent the overall inflammation score exhibited in the lungs of WT and miR-223^−/− mice 48 hpi (n = 4). Scale bar indicates 100 μm. (D) Survival curves of S.pn. ST2-infected WT and miR-223^−/− mice over the duration of 120 hpi. (A) 2-way ANOVA/Šidák’s multiple comparisons test, * p < 0.05. (C) Mann-Whitney U test was performed to analyze statistical significance in the overall inflammation score; * p < 0.05. Data in (A–C) display individual and mean values, while data in (D) displays censored subjects only. hpi: hours post-infection, CFU: colony-forming unit, BAL: bronchoalveolar lavage.

Figure 3

Absence of miR-223 leads to an enhanced and prolonged PMN response in the lungs and BAL of miR-223^−/− mice post-S.pn. ST2 infection. (A) PMN numbers, frequencies and representative dot plots in BAL and lungs 48 hpi (n = 10–12). (B) Frequencies and representative dot plots of BAL and lung PMN undergoing early and late apoptosis 48 hpi (n = 5–7). (C) CXCL1, CXCL2, CXCL5 chemokines quantified in the BAL of mice through ELISA 48 hpi (n = 10–14). (A,C) Unpaired t-test; * p < 0.05. (B) 2-way ANOVA/Šidák’s multiple comparisons test; *** p < 0.001. Data display individual values and means, error bars represent SEM. PMN: polymorphonuclear neutrophil, BAL: bronchoalveolar lavage, S.pn. ST2: Streptococcus pneumoniae serotype 2.

Figure 4

miR-223^−/− mice exhibit increased capacity of pro-inflammatory cytokine and chemokine production following S.pn. ST2 infection. Pro-inflammatory cytokines were quantified in the BAL of WT (n = 11) and miR-223^−/− (n = 13) mice 48 hpi using the LEGENDPlex Mouse Inflammation Panel (BioLegend, San Diego, CA, USA), whilst MPO was quantified using the MPO Mouse ELISA kit (Hycult Biotech, Uden, Netherlands). Mann-Whitney U test was performed to analyze statistical significance. * p < 0.05, ** p < 0.01; data display individual values and means, error bars represent SEM. BAL: bronchoalveolar lavage, S.pn. ST2: Streptococcus pneumoniae serotype 2, MPO: myeloperoxidase.

Figure 5

scRNA-seq of murine lungs 24h following S.pn. ST2 infection. (A) UMAP plot of identified cell populations in the lungs of WT and miR-223^−/− mice following S.pn. ST2 (WT, miR-223^−/−; n = 3) or sham (PBS Ctrl, WT; n = 2) infection. (B) Number of significant differentially expressed genes (minimum fold change 1.3) in miR-223^−/− mice (relative to WT mice) infected with S.pn. (C) Dot plot of significantly differentially expressed genes involved in inflammation, granulocyte maturation, enzymatic digestion and antibacterial defense in WT versus miR-223^−/− mice. (D) Pathway enrichment analysis of miR-223^−/− versus WT lung cells sequenced following S.pn. infection. Coloration and point sizes indicate log₂-transformed fold changes and mean expression levels, respectively. AT1: alveolar epithelial cells type I, AT2: alveolar epithelial cells type II, AM: alveolar macrophages, SMC: smooth muscle cells, MΦ & DC: interstitial and inflammatory macrophages/monocytes and dendritic cells, ly. Endothelial cells: lymphatic endothelial cells, NK cells: natural killer cells, GO:BP: Gene Ontology:Biological Process, KEGG: Kyoto Encyclopedia of Genes and Genomes, UMAP: uniform manifold approximation and projection.