Systemic disease during Streptococcus pneumoniae acute lung infection requires 12-lipoxygenase-dependent inflammation

Abstract

Acute pulmonary infection by Streptococcus pneumoniae is characterized by high bacterial numbers in the lung, a robust alveolar influx of polymorphonuclear cells (PMNs), and a risk of systemic spread of the bacterium. We investigated host mediators of S. pneumoniae-induced PMN migration and the role of inflammation in septicemia following pneumococcal lung infection. Hepoxilin A3 (HXA3) is a PMN chemoattractant and a metabolite of the 12-lipoxygenase (12-LOX) pathway. We observed that S. pneumoniae infection induced the production of 12-LOX in cultured pulmonary epithelium and in the lungs of infected mice. Inhibition of the 12-LOX pathway prevented pathogen-induced PMN transepithelial migration in vitro and dramatically reduced lung inflammation upon high-dose pulmonary challenge with S. pneumoniae in vivo, thus implicating HXA3 in pneumococcus-induced pulmonary inflammation. PMN basolateral-to-apical transmigration in vitro significantly increased apical-to-basolateral transepithelial migration of bacteria. Mice suppressed in the expression of 12-LOX exhibited little or no bacteremia and survived an otherwise lethal pulmonary challenge. Our data suggest that pneumococcal pulmonary inflammation is required for high-level bacteremia and systemic infection, partly by disrupting lung epithelium through 12-LOX-dependent HXA3 production and subsequent PMN transepithelial migration.

Fig. 1

Diverse serotypes of S. pneumoniae elicit robust PMN migration in vitro in a dose, time and capsule-dependent manner. A, H292 monolayers were infected with the indicated doses of S. pneumoniae or B. subtilis.1×10⁶ PMNs were added to the basolateral side and PMN migration to the apical side was quantified by MPO assay. Shown is a representative of two experiments. B, Apical sides of H292 monolayers were infected with 1×10⁷ S. pneumoniae TIGR4. 1×10⁶ PMNs were added to the basolateral side, and allowed to migrate for the indicated time periods. PMN migration to the apical side was quantified by MPO assay. Shown is a representative of two experiments. C, H292 monolayers were infected with 1×10⁷ S. pneumoniae TIGR4, its capsular mutant (Δcps), S. pneumoniae D39, or its capsular mutant R6. Basolateral-to-apical PMN migration was quantified by MPO assay. Positive and negative control wells received formylated-Met-Leu-Phe (“fMLP”) and HBSS+Ca/Mg, respectively. Shown is a representative of two experiments.

Fig. 2

Live bacteria and de novo protein synthesis by the host cells are required for S. pneumoniae-induced PMN migration. A, Apical sides of H292 monolayers were infected with either live or heat-killed S. pneumoniae TIGR4. PMNs were added basolaterally and PMN migration to the apical side was quantified by MPO assay. fMLP and HBSS+Ca/Mg were used as positive and negative controls, respectively. Binding of live and heat-killed bacteria to H292 monolayers, when assessed by viable counts or microscopically, were statistically indistinguishable. Shown is a representative of two experiments. B, Protein synthesis in H292 monolayers was blocked by treatment with cycloheximide before infection with S. pneumoniae TIGR4 (1×10⁷/monolayer), and PMN migration was assessed as described in Materials and Methods. Shown is a representative of two experiments.

Fig. 3

12-lipoxygenase is induced upon pneumococcal infection of pulmonary epithelial monolayers and is required for PMN transmigration. A, Mock-treated H292 monolayers, or monolayers pretreated with 1μM baicalein or 50 μM CDC, were infected with TIGR4. After infection, lipid extracts were prepared from the conditioned buffer, as stated in Materials and Methods. PMN migration in response to extracted lipids was measured as stated before. Shown is a representative of two experiments; B, Semi-quantitative RT-PCR analyses of ALOX12 and ALOX15 levels were performed using total RNA from infected or uninfected H292 cells, using - specific primers. GAPDH was used as the internal control. Shown is a representative gel picture from three independent experiments. Band intensities normalized to the band intensities of GAPDH, which was taken as 1, are presented below each band. Data represent means from three independent experiments. C, Lysates of H292 monolayers infected with TIGR4 were analyzed by immunoblotting using anti- ALOX12, ALOX15 or GAPDH antibodies. Band intensities were normalized to the band intensities of GAPDH and their values are given below each blot. Shown are the means from two independent experiments. D, H292 monolayers were pretreated with the indicated doses (μM) of CDC, baicalein (“Bcn”), or caffeic acid (“CA”) prior to infection with TIGR4. PMNs were added to the basolateral side and migration was quantified by MPO assay. fMLP, HBSS+Ca/Mg or DMSO were used as positive, negative and solvent controls, respectively. Shown is a representative of two experiments.

Fig. 4

12-lipoxygenase activity is required for alveolar neutrophil recruitment even during absence of S. pneumoniae infection. Wild type C57BL/6 mice, with or without CDC treatment, or Alox15^−/− mice were sacrificed and PMNs present in BALF or lungs were analyzed by flow cytometry as described in Materials and Methods. Shown is a representative of two independent experiments.

Fig. 5

12-lipoxygenase is induced in vivo upon pneumococcal infection. Wild type C57BL/6 mice (with or without CDC pretreatment) or Alox15^−/− mice were intratracheally challenged with 2×10⁵ S. pneumoniae TIGR4. Control mice received only PBS. Mice were euthanized, lung homogenates were immunoblotted and probed using anti-ALOX12, ALOX15 or tubulin antibodies. Shown is a representative figure from two independent experiments. Corresponding molecular weight markers are denoted.

Fig. 6

12-lipoxygenase activity is required for pulmonary inflammation during S. pneumoniae infection of mice. Untreated [“+TIGR4”; panel A(ii)] or CDC-treated [“+CDC, +TIGR4”; panel A(iii)] wild type (“WT”) C57BL/6 mice, or Alox15^−/− [panel A(iv)] mice were inoculated intratracheally with TIGR4, or with PBS alone [panel A(i)], as described in Materials and Methods. Mice were sacrificed, and hematoxylin and eosin stained lung sections (at 20× magnification) were examined by light microscopy. Inset: lung sections at 40× magnification; arrowheads denote PMNs. Figures are representative of two independent experiments. Bronchoalveolar lavage fluid (BALF) was isolated and cytospin preparations were prepared. PMNs were counted from Wright-Giemsa stained preparations by observation at 100×. Average number of PMNs per 5 fields of observation is noted in the corresponding histology figures. For flow cytometry, BALF and lungs from infected WT (with or without CDC treatment) or Alox15^−/− mice were collected. Cells present in the digested lungs and BALF were stained with relevant MAbs and the fluorescence intensities of the stained cells were determined by flow cytometry. Collected data were analyzed to determine the numbers of PMNs, macrophages (Mɸ), dendritic cells or T cells in the lungs (B, top) and in the BALF (B, bottom).

Fig. 7

S. pneumoniae-elicited PMN migration compromises epithelial barrier integrity and facilitates pneumococcal transepithelial migration. A, Compromise in barrier integrity due to S. pneumoniae -elicited PMN migration was measured using the migration of fluorescein isothiocyanate (FITC) labelled-dextran (MW 40kDa, Sigma). Infection and PMN migration times were 2½h each. Control monolayers received either bacteria or PMNs. Shown is a representative of two experiments. B, Apical sides of H292 monolayers were infected with TIGR4, and PMNs, or HBSS+Ca/Mg alone, were added the basolateral side. Transepithelial migration of bacteria was quantitated as stated in Materials and Methods. Shown is a representative of two experiments. C, Apical sides of H292 monolayers were infected with TIGR4, and PMNs were added to either basolateral or apical side of the monolayers. Control monolayers received HBSS+Ca/Mg instead of PMNs. Transepithelial migration of bacteria was quantitated as stated in Materials and Methods. Shown is a representative of two experiments. D, H292 monolayers were pretreated with CDC, and the apical side was infected with TIGR4. PMNs, or HBSS+Ca/Mg alone, were added the basolateral side. Bacterial migration was determined as stated in Material and Methods. Shown is a representative of two experiments.

Fig. 8

12-LOX mediated inflammation promotes bacteremia and lethality during S. pneumoniae infection of mice. Mock-treated wild type C57BL/6 mice (“WT”), CDC-treated wild type mice, or Alox15^−/− mice were intratracheally inoculated with TIGR4 (see Materials and Methods). A control group of wild type mice received PBS. Six mice per group were used for the experiment. A, The cfu in the murine blood was determined at the specified time points. Shown is a representative of two experiments. B, The survival of mice was monitored over a 7 day post-infection period. Shown is a representative of two experiments.