Polysialylation controls immune function of myeloid cells in murine model of pneumococcal pneumonia

Abstract

Polysialic acid (polySia) is a post-translational modification of a select group of cell-surface proteins that guides cellular interactions. As the overall impact of changes in expression of this glycan on leukocytes during infection is not known, we evaluate the immune response of polySia-deficient ST8SiaIV^-/- mice infected with Streptococcus pneumoniae (Spn). Compared with wild-type (WT) mice, ST8SiaIV^-/- mice are less susceptible to infection and clear Spn from airways faster, with alveolar macrophages demonstrating greater viability and phagocytic activity. Leukocyte pulmonary recruitment, paradoxically, is diminished in infected ST8SiaIV^-/- mice, corroborated by adoptive cell transfer, microfluidic migration experiments, and intravital microscopy, and possibly explained by dysregulated ERK1/2 signaling. PolySia is progressively lost from neutrophils and monocytes migrating from bone marrow to alveoli in Spn-infected WT mice, consistent with changing cellular functions. These data highlight multidimensional effects of polySia on leukocytes during an immune response and suggest therapeutic interventions for optimizing immunity.

Keywords: CP: Immunology; Streptococcus pneumoniae; extracellular signal-regulated kinase ERK1/2; intravital microscopy; leukocyte migration; microfluidic migration chambers; monocytes; neutrophils; phagocytosis; polysialic acid; polysialyltransferase ST8SiaIV.

Figure 1.. ST8SiaIV^−/− mice are relatively resistant to pulmonary infection with Spn and clear the pathogen more effectively than WT mice

(A and B) Eight- to 12-week-old WT, ST8SiaIV^−/−, and ST8SiaII^−/− mice were infected intratracheally with 10⁴ colony-forming units (CFU) of Spn and were monitored daily for clinical signs of moribundity as described in STAR Methods (A). Pulmonary clearance of Spn was evaluated by quantitating CFU of Spn in the BAL fluid 24 h postinfection (B). Statistical significance of data from survival experiments in (A) was determined by log-rank test, and data in (B) are expressed as mean ± SEM (n = 3 mice per group) from each of four experiments. ***p ≤ 0.001; ****p ≤ 0.0001 (two-tailed t test).

Figure 2.. Pulmonary recruitment of myeloid cells from ST8SiaIV^−/− mice is impaired following infection with Spn

(A) Lung sections from uninfected and Spn-infected WT and ST8SiaIV^−/− mice 24 h postinfection were stained with H&E and imaged at 10× magnification, and lung injury score was determined as described in STAR Methods (white arrowheads indicate alveolar spaces; black arrowheads indicate bronchial walls). (B) Leukocytes were harvested from BAL fluid from WT and ST8SiaIV^−/− mice 24 h and 48 h postinfection with Spn and were analyzed by flow cytometry after staining with antibodies to CD45, CD11b, Ly6C, and Ly6G, and the percentage of neutrophils (CD11b, Ly6G: upper panels) and monocytes (CD11b, Ly6C: lower panels) in CD45 gated cells was determined (for gating strategy see Figure S2). (C) Lung homogenates from uninfected WT and ST8SiaIV^−/− mice and from mice 24 h and 48 h postinfection with Spn were prepared, and levels of IL-1β (left), IL-10 (second from left), KC (second from right), and MCP-1 (right) were measured by ELISA; WT (black bars), and ST8SiaIV^−/− (gray bars). Data in (B) and (C) represent values from individual mice from five independent experiments and were evaluated by t test with mean ± SEM. *p ≤ 0.05; **p ≤ 0.01; ns, not statistically significant. Scale bars, 100 µm.

Figure 3.. Absence of polySia in leukocytes from ST8SiaIV^−/− mice promotes phagocytosis of Spn

(A) Peritoneal macrophages were isolated from WT and ST8SiaIV^−/− mice, grown in tissue culture plates for 24 h, incubated with opsonized Spn, and lysed to quantitate the number of intracellular bacteria (CFU) as described in STAR Methods. (B) Leukocytes from WT mice (with or without treatment with endoN) and from ST8SiaIV^−/− mice were collected by BAL, incubated with opsonized Spn, and lysed to quantitate intracellular CFU as described in STAR Methods. (C) Cells that were collected by BAL from uninfected WT mice and from WT mice 24 h postinfection with Spn (with or without endoN treatment) were stained with antibody against CD45, CD11b, CD11c, Ly6C, Siglec F, and polySia, and polySia expression on gated AMs (CD11c⁺, Siglec F⁺) is shown in the dot plots. SSC, side scatter. (D) Cells that were collected by BAL from uninfected WT mice were stained with antibody against CD45, CD11c, and Siglec F and with annexin V and 7-AAD, andthe percentage of AMs displaying early apoptosis was analyzed by flow cytometry. Data in (A), (B), and (D) are expressed as mean ± SEM of each assay using cells from three mice per group from each of three experiments. **p ≤ 0.01; ***p ≤ 0.001 (two-tailed t test). Data in (C) are representative of three different experiments.

Figure 4.. Adoptively transferred neutrophils and monocytes from ST8SiaIV^−/− mice migrate less efficiently than WT cells to lungs of recipient Spn-infected WT mice

(A) Schematic representation of adoptive transfer experiment performed on Spn-infected WT mice. (B) Total lung was harvested from Spn-infected WT mice 24 h after receiving an intra-orbital injection of differentially labeled WT and ST8SiaIV^−/− BM-derived neutrophils or monocytes, and the number of labeled donor neutrophils (left) and monocytes (right) was determined by flow cytometry. (C and D) Cells from BM, PB, and spleen of adoptively transferred mice were harvested, and the number of labeled donor neutrophils (C) and donor monocytes (D) was determined by flow cytometry. Data in (B), (C), and (D) are expressed as mean ± SEM (n = 2–3 mice per group) from each of four experiments. **p ≤ 0.01; ns, not significant (non-parametric Wilcoxon signed-rank test).

Figure 5.. Migration in vitro of ST8SiaIV^−/− neutrophils toward CXCL1 and fMLP is compromised compared with WT neutrophils

(A and B) Chemotaxis of neutrophils from WT and ST8SiaIV^−/− mice was evaluated in a microfluidic-based microchannel assay, as described in STAR Methods. Cell migration toward fMLP (A) and CXCL1/KC (B) gradients was visualized and recorded by time-lapse live microscopy. The speed, velocity, and persistence of cells was computed from individual cell tracks (n = 191 and n = 234 from WT and ST8SiaIV^−/− mice, respectively). Data are expressed as mean ± SD using cells from three mice per group from each of four experiments. ***p ≤ 0.001; ****p ≤ 0.0001; ns, not significant (Welch’s t test). (C) Bone marrow neutrophils from WT and ST8SiaIV^−/− mice were isolated and stained for CD11b, Ly6G and CXCR1 and were analyzed by flow cytometry. Dot plots represent data from five mice of each strain. (D) Bone marrow neutrophils were isolated and incubated with CXCL1/KC (100 ng/mL), washed five times with PBS, lysed in RIPA buffer, and analyzed for boundCXCL1/KC by ELISA. Data are expressed as mean ± SEM (n = 3 mice per group) from each of three experiments. ****p ≤ 0.0001 (two-tailed t test). (E) Bone marrow neutrophils were isolated and incubated with 1 µM fMLP for the indicated times and collected and lysed in RIPA buffer, and equal amounts of protein (25 µg) from each lysate was separated on SDS-PAGE and analyzed for the amount of phosphorylated and total ERK as indicated in STAR Methods. Histogram shows the ratio of phosphorylated to total ERK at each time point. The immunoblot is representative of results obtained from three independent experiments.

Figure 6.. ST8SiaIV^−/− neutrophils are more adherent to and display reduced extravasation through the cremasteric microvasculature compared with WT cells

ST8SiaIV^−/− (magenta) and WT (green) BM neutrophils were prepared and injected into WT recipient mice. (A–D) Adhesion to (A), rolling along (B), and extravasation through (C and D) the cremasteric vascular bed of labeled donor cells were observed by intravital microscopy as noted in STAR Methods. To induce recruitment of donor neutrophils in recipient WT mice, the cremasteric tissue was locally injected with TNF-α prior to cell transfusion. The number of adherent neutrophils (A) was determined by counting the stationary cells in a single 20- to 50-µm diameter venule per microscopic field over 30-s frames. The number of rolling cells (B) comprised cells per main venule per microscopic field migrating up to 100 µm per 30-s frame. A representative intravital microscopic image is shown in (C). (D–F) The number of emigrated extravascular donor WT and STSia8IV^−/− mouse neutrophils was measured per microscopic field from 0.5 to 2 h post cell transfusion (D). The distance traveled from the nearest blood vessel by these cells was measured using Imaris software (E). A representative intravital microscopic image of extravasated neutrophils is shown in (F). Data acquired from 4–6 vascular beds from three different recipient WT mice are expressed as mean ± SEM. *p ≤ 0.05; **p ≤ 0.01; ****p ≤ 0.0001 (two-tailed t test). Scale bars, 100 µm.

Figure 7.. PolySia is progressively lost from the surface of myeloid cells as they migrate from bone marrow to alveoli

Neutrophils and monocytes were isolated from BM, PB, and alveolar spaces of Spn-infected WT and ST8SiaIV^−/− mice 24 h and 48 h postinfection, stained with antibodies against CD45, CD11b, Ly6G, Ly6C, and polySia, and analyzed by flow cytometry. Histograms from individual mice (left panels) showing absolute cell counts as well as composite data from multiple mice showing percentage of cells polysialylated (middle panels) and MFI of staining (right panels) are presented for CD45-gated neutrophils (Ly6G/CD11b) (A) and monocytes (Ly6C/CD11b) (B) harvested at 24 h and 48 h postinfection (middle and right columns). Data represent mean ± SEM (n = 4–9 mice) from four independent experiments. ***p ≤ 0.001 (two-tailed t test).