Interleukin-10 Produced by Myeloid-Derived Suppressor Cells Provides Protection to Carbapenem-Resistant Klebsiella pneumoniae Sequence Type 258 by Enhancing Its Clearance in the Airways

Abstract

Carbapenem-resistant Klebsiella pneumoniae sequence type 258 (CRKP-ST258) can cause chronic infections in lungs and airways, with repeated episodes of bacteremia. In this report we addressed whether the recruitment of myeloid cells producing the anti-inflammatory cytokine interleukin-10 (IL-10) modulates the clearance of CKRP-ST258 in the lungs and establishes bacterial persistence. Our data demonstrate that during pneumonia caused by a clinical isolate of CRKP-ST258 (KP35) there is an early recruitment of monocyte-myeloid-derived suppressor cells (M-MDSCs) and neutrophils that actively produce IL-10. However, M-MDSCs were the cells that sustained the production of IL-10 over the time of infection evaluated. Using mice unable to produce IL-10 (IL-10^-/-), we observed that the production of this cytokine during the infection caused by KP35 is important to control bacterial burden, to prevent lung damage, to modulate cytokine production, and to improve host survival. Importantly, intranasal transfer of bone marrow-derived M-MDSCs from mice able to produce IL-10 at 1 day prior to infection improved the ability of IL-10^-/- mice to clear KP35 in the lungs, decreasing their mortality. Altogether, our data demonstrate that IL-10 produced by M-MDSCs is required for bacterial clearance, reduction of lung tissue damage, and host survival during KP35 pneumonia.

Keywords: Klebsiella pneumoniae ST258; interleukin-10; monocytic-myeloid-derived suppressor cells; neutrophils.

FIG 1

Neutrophils, M-MDSCs, and Ly6C⁻ monocytes are the main IL-10 producers during KP35 infection. (A to C) IL-10/GFP VertX mice were intranasally infected with 1 × 10⁸ CFU of KP35, and IL-10 production by myeloid cells was evaluated by flow cytometry at 12, 24, 48, 72, 96, and 240 hpi on neutrophils, M-MDSCs, and Ly6C⁻ monocytes, as indicated. For each cell population, data are presented as the number of total lung cells, the numbers of lung GFP⁺ (IL-10⁺) cells, the percentage of GFP⁺ (IL-10⁺) cells relative to the total number of cells, and the fold change in mean fluorescence intensity (MFI) of GFP in the total cell population. WT uninfected mice were included to establish the autofluorescence level (data not shown). *, P < 0.05, for comparison of each group with the uninfected control (by one-way ANOVA with a Tukey posttest). FITC, fluorescein isothiocyanate.

FIG 2

IL-10 production is required for KP35 clearance and host survival. WT and IL-10^−/− mice were intranasally infected with 1 × 10⁸ CFU of KP35. (A) Bacterial burden was evaluated in lung, BALF, and spleen at 24 and 48 hpi. (B to D) Survival rate, severity of the disease, and body weight change were evaluated over 7 days († represents the number of dead mice at 4 dpi). For bacterial burden, * P < 0.05 (Mann-Whitney U test comparing results between both groups at each time point); for the survival curve, ***, P = 0.0005 (log rank test); for disease severity (clinical score), *, P < 0.05 (two-way ANOVA with a Tukey posttest comparing results for both groups at each time point). ns, not significant.

FIG 3

IL-10 production modulates lung damage during KP35 infection. WT and IL-10^−/− mice were intranasally infected with 1 × 10⁸ CFU of KP35, and different parameters of lung damage were evaluated, including lung histopathology by H&E staining (A), lung histopathological score (B), total BALF proteins (C) and total lung cell apoptosis by TUNEL assay (D) and TUNEL-positive area (E). (F) Neutrophil activation was measured by the quantification of MPO activity in the BALF. Scale bar, 2 mm. *, P < 0.05, for a comparison of results between groups at each time point or between the infected group and the respective uninfected (Unf) group (two-way ANOVA with a Tukey posttest).

FIG 4

IL-10 production does not influence immune cell recruitment to the airways of KP35-infected mice. Recruitment of neutrophils (A), M-MDSCs (B), and Ly6C⁻ (C) monocytes was measured in the lung and BALF of uninfected and infected WT and IL-10^−/− mice at 24 and 48 hpi. *, P < 0.05, for a comparison of results between groups at each time point or between the infected group and the respective uninfected (Unf) group (two-way ANOVA with a Tukey posttest).

FIG 5

IL-10 modulates proinflammatory cytokine production during KP35 pneumonia. (A to F) The production of TNF-α, IL-6, IL-1β, IL-12p40, IL-23p19, and IFN-γ in BALF, as indicated, from WT and IL-10^−/− mice was analyzed by Luminex technology at 24 and 48 hpi. *, P < 0.05, for a comparison of results of both groups at each time point (two-way ANOVA with a Tukey posttest).

FIG 6

IL-10 produced by BM-MDSCs is critical for restoring bactericidal capacity in IL-10^−/− mice. BM-MDSCs from IL-10^−/− and IL-10/GFP VertX mice were generated in vitro. One day before the infection (day −1), 1 × 10⁶ BM-MDSCs were transferred intranasally into IL-10^−/− mice. At day 0 each group was intranasally infected with 1 × 10⁸ CFU of KP35, and at 48 h postinfection, mice were euthanized. (A to C) Bacterial burden in lung tissue, spleen, and BALF was determined. (D and E) IL-10 production by MDSCs was identified by flow cytometry in the lung tissue. (F and G) Cytokine production in BALF was quantified by Luminex assay. *, P < 0.05, for a comparison of results for each parameter (by one-way ANOVA with a Tukey posttest). UT, untransferred.

FIG 7

IL-10 produced by BM-MDSCs modulates lung injury during KP35 pneumonia. BM-MDSCs from IL-10^−/− and IL-10/GFP VertX mice were generated in vitro. One day before the infection (day −1), 1 × 10⁶ BM-MDSCs were transferred intranasally into IL-10^−/− mice. At day 0 each group was intranasally infected with 1 × 10⁸ CFU of KP35; at 48 h postinfection, mice were euthanized and lung histopathology was assessed through H&E staining (A). Lung injury was quantified through a histopathological score (B) and through the quantification of total proteins in the BALF (C). *, P < 0.05; ns, not significant, for a comparison of results for each parameter (by one-way ANOVA with a Tukey posttest).

FIG 8

IL-10 produced by BM-MDSCs is critical for host survival during KP35 pneumonia. (A to C) BM-MDSCs from IL-10^−/− and IL-10/GFP VertX mice were generated in vitro. One day before the infection (day −1), 1 × 10⁶ BM-MDSCs were transferred (T) intranasally into receptor IL-10^−/− mice. At day 0 each group was intranasally infected (I) with 1 × 10⁸ CFU of KP35, and survival rate, body weight, and disease severity, as indicated, were followed over 7 days. Survival rate was evaluated with a log rank test. Disease severity levels between IL-10^−/− mice (untransferred [UT]) and IL-10^−/− mice that received IL-10/GFP BM-MDSCs by transfer were compared by two-way ANOVA with a Tukey posttest.

FIG 9

Graphical summary of major differences of the immune responses and the role of IL-10 during KP35 and KPPR1 pneumonia. Major differences in the immune responses against sublethal doses of KPPR1 and KP35 have been described in WT and IL-10^−/− mice. (A) The immune response against KPPR1 is characterized by a rapid bacterial clearance and a rapid infiltration of neutrophils but a much slower infiltration of M-MDSCs at 72 hpi. WT mice infected with KPPR1 produce TNF-α and IL-6 from 48 hpi to at least 96 hpi. (B) During KP35 pneumonia the immune response is characterized by early recruitment of neutrophils and M-MDSCs able to produce IL-10 and other suppressive effectors. As a consequence, these mice present weaker production of proinflammatory cytokines and a high but decreasing bacterial burden in the lung tissue. Incredibly, these mice are very resistant to KP35 and CRKP-ST258. (C) In the absence of IL-10 the immune response against KPPR1 is more efficient, presenting enhanced bacterial clearance in the first 72 hpi, increased neutrophil activation and proinflammatory cytokine production, and enhanced survival compared to levels in mice able to produce IL-10. (D) On the other hand, in the absence of IL-10, KP35 infection leads to a diminished bacterial clearance capacity in the first 48 hpi and equivalent numbers of neutrophils and M-MDSCs but impaired neutrophil activation and increased proinflammatory cytokine production during the first 48 hpi, with around 90% mortality at 4 dpi.