Modulation of respiratory dendritic cells during Klebsiella pneumonia infection

Abstract

Background: Klebsiella pneumoniae is a leading cause of severe hospital-acquired respiratory tract infections and death but little is known regarding the modulation of respiratory dendritic cell (DC) subsets. Plasmacytoid DC (pDC) are specialized type 1 interferon producing cells and considered to be classical mediators of antiviral immunity.

Method: By using multiparameter flow cytometry analysis we have analysed the modulation of respiratory DC subsets after intratracheal Klebsiella pneumonia infection.

Results: Data indicate that pDCs and MoDC were markedly elevated in the post acute pneumonia phase when compared to mock-infected controls. Analysis of draining mediastinal lymph nodes revealed a rapid increase of activated CD103+ DC, CD11b+ DC and MoDC within 48 h post infection. Lung pDC identification during bacterial pneumonia was confirmed by extended phenotyping for 120G8, mPDCA-1 and Siglec-H expression and by demonstration of high Interferon-alpha producing capacity after cell sorting. Cytokine expression analysis of ex vivo-sorted respiratory DC subpopulations from infected animals revealed elevated Interferon-alpha in pDC, elevated IFN-gamma, IL-4 and IL-13 in CD103+ DC and IL-19 and IL-12p35 in CD11b+ DC subsets in comparison to CD11c+ MHC-class IIlow cells indicating distinct functional roles. Antigen-specific naive CD4+ T cell stimulatory capacity of purified respiratory DC subsets was analysed in a model system with purified ovalbumin T cell receptor transgenic naive CD4+ responder T cells and respiratory DC subsets, pulsed with ovalbumin and matured with Klebsiella pneumoniae lysate. CD103+ DC and CD11b+ DC subsets represented the most potent naive CD4+ T helper cell activators.

Conclusion: These results provide novel insight into the activation of respiratory DC subsets during Klebsiella pneumonia infection. The detection of increased respiratory pDC numbers in bacterial pneumonia may indicate possible novel pDC functions with respect to lung repair and regeneration.

Figure 1

Dissection of respiratory leukocyte and DC subsets (MoDC, CD11b^hi DC, CD103⁺ DC, pDC) during Klebsiella pneumonia infection. Respiratory leukocyte subsets were dissected by flow cytometry as described in Materials and methods (A-C). Relative frequencies and absolute numbers of respiratory DC subsets were quantitated at indicated timepoints and compared to mock-infected control animals (B, C). Mean ± SEM; n ≥ 4; *p < 0.05; **p < 0.01; ***p < 0.001 versus mock-infected controls.

Figure 2

Respiratory DC subsets during Klebsiella pneumonia infection. Dotplot staining and gating of respiratory DC subsets (CD103⁺ DC, CD11b⁺ DC, MoDC, pDC) in Klebsiella-infected animals and mock-infected controls. Fluorescence minus one (FMO) isotype control for proper identification of CD64⁺ MoDC.

Figure 3

Significant lung pathology in Klebsiella pneumonia infected animals. Histopathology of mouse lungs two (upper row) and five days (bottom row) after mock (left column) and K. pneumoniae (right column) infection: No changes were present in mock infected mice. In contrast, extensive interstitial infiltration by predominantly neutrophils and macrophages was present at two days after infection of mice with K. pneumoniae. After five days, additional, predominantly perivascular, lymphocytic infiltrates were seen. Hematoxylin & Eosin stained lung sections at 200 × magnification.

Figure 4

Confirmation of respiratory pDC identity during Klebsiella infection (d5 p.i.) by extended phenotyping and cell sorting. Identity of pDC during Klebsiella pneumonia was confirmed by additional phenotyping of CD11c + 120 g8+ CD11b^neg cells for mPDCA-1 and Siglec-H in parallel with isotype controls (A). Functional ability of respiratory pDC from Klebsiella infected animals to produce IFN-alpha was confirmed by cell sorting and CPG ODN stimulation in comparison to control populations (B). Respiratory pDC during Klebsiella pneumonia are CD11b^neg 120G8⁺ Siglec-H⁺ mPDCA-1⁺ cells (A, B) and exhibit the capacity to produce large amounts of IFN-α (B). Mean ± SEM; n ≥ 3.

Figure 5

Relative mRNA cytokine expression of respiratory CD11b⁺ DC, CD103⁺ DC, pDC from Klebsiella pneumoniae-infected animals. Semi-quantitative mRNA cytokine expression was analysed by real time RT-PCR in sorted respiratory DC subsets from d5-infected animals. Mean fold regulation in relation to MHC-II^low CD11c⁺ cells; n ≥ 3.

Figure 6

Respiratory DC-subset specific expression of MHC-class II, CD86 and CD274 at different timepoints during acute Klebsiella pneumonia infection. Surface MHC-class II, CD86 and CD274 on total respiratory DC (DC) and DC subsets (pDC, CD103⁺ DC, MoDC, CD11b⁺ DC) were analyzed by flow cytometry during acute and post-acute Klebsiella pneumonia (Kp) infection (48 h and d5 p.i.) and compared to mock-infected control (ctr) animals (A, B). MFI is median fluorescence intensity; mean ± SEM; n ≥ 3; *p < 0.05; **p < 0.01; ***p < 0.001 versus mock-infected controls (A-B).

Figure 7

Different capacity of purified Klebsiella pneumoniae-stimulated respiratory DC subsets to activate OVA TCR-transgenic naive CD4⁺ responder T cells in a CFSE-dilution assay. Respiratory DC subsets were sorted to high purity (>98%), pulsed with the model antigen OVA and matured with Klebsiella pneumoniae lysate (Kp). Naïve CD4⁺ responder T cells were sorted from OVA TCR transgenic animals and labeled with CFSE. Negative controls included DC without Klebsiella lysate (control), responder T cells cultured without DC (no DC) and DC-T cell co-cultures without OVA (no OVA). Proliferation was quantitated by CFSE-dilution of responder T cells (A,B). Mean ± SEM; n ≥ 3.

Figure 8

Quantitative and phenotypic analysis of DC subsets in the MLN at different time points during Klebsiella pneumonia infection. Relative and absolute CD103⁺ DC, CD11b⁺ DC, MoDC and pDC subset numbers (A, B) and phenotypic markers (C) were analyzed at indicated time points in the MLN and compared to mock-infected controls by flow cytometry as described in Materials and methods. Mean ± SEM; n ≥ 3; *p < 0.05; **p < 0.01; ***p < 0.001 versus mock-infected controls.