Csk-mediated Src family kinase regulation dampens neutrophil infiltration during pulmonary infection

Abstract

Neutrophil recruitment is crucial for pathogen elimination. However, precise control of the inflammatory response prevents overshooting reactions. Neutrophil activation initiates signaling, resulting in integrin β2 (Itgb2) activation and neutrophil arrest. Src family kinases are involved in multiple cellular processes and are negatively regulated by the C-terminal Src kinase (Csk). During this study, we investigated the mechanism by which Csk regulates integrin activation and neutrophil recruitment. Here, we demonstrated that Csk deficiency in murine neutrophils resulted in increased neutrophil adhesion to the endothelium along with decreased neutrophil transmigration into inflamed tissues compared with their littermate controls. In bacterial pneumonia, infected Csk-deficient mice showed higher bacterial burdens and decreased neutrophil recruitment, while other immune cell counts and cytokine levels were not significantly different compared to control. Analyses of Csk-deficient neutrophils revealed an increased Itgb2 affinity, leading to reduced migration and intravascular crawling. Mechanistically, elevated cAMP levels increased protein kinase A activity, which subsequently enhanced Csk activation. Csk, in turn, suppressed Src family kinase activation through phosphorylation (Y529). Hence, Csk-mediated regulation of neutrophil infiltration contributes to maintain a balanced immune response during bacterial pneumonia.

Keywords: Immunology; Inflammation; Integrins; Neutrophils.

Figure 1. Csk is crucial for neutrophil recruitment and bacterial clearance in K. pneumoniae–induced pneumonia.

(A–K) Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt mice were subjected to K. pneumoniae intratracheal injection or sham surgery. After 24 hours, bacterial burden regarding the colony forming units (CFUs) in lung (A), bronchoalveolar fluid (BALF) (B), blood (C), and spleen (D) and neutrophil recruitment into the lungs (E) and BALF (F) were determined. (G) Neutrophil recruitment (CD45⁺Gr-1⁺LyB.2⁺) in Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt mice in the interstitial, intravascular, and BALF compartments following intratracheal instillation of K. pneumoniae. Cell count of monocytes (MO; CD45⁺CD11b⁺CX3CR1⁺Ly6C^hiLy6G^–Gr-1^–) (H), alveolar macrophages (AM; CD45⁺CD64⁺F4/80⁺MARCO⁺SiglecF^hi) (I), natural killer cells (NK; CD45⁺CD27⁺CD335⁺) (J), and dendritic cells (DC; CD45⁺CD27^–CD24⁺CD11c⁺MHCII⁺) (K) in lung tissue 24 hours after lung infection. n as indicated, mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 by 1-way-ANOVA by Holm-Šídák multiple-comparison test (A–D), 1-way ANOVA with Tukey’s multiple-comparison test (E, F, and H–K), or 2-tailed Student’s t test (G).

Figure 2. Csk is crucial for neutrophil recruitment and bacterial clearance in S. aureus–induced pneumonia.

(A–F) Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt mice were subjected to S. aureus intratracheal injection or sham surgery. Bacterial burden regarding the colony forming units (CFUs) in lungs (A), bronchoalveolar fluid (BALF) (B), blood (C), and spleen (D) and neutrophil recruitment into the lungs (E) and BALF (F) were determined 24 hours after S. aureus injection. n = 3–10 mice per genotype, mean ± SEM. *P < 0.05; **P < 0.01; ****P < 0.0001 by 1-way ANOVA with Tukey’s multiple-comparison test.

Figure 3. Csk is not involved in the regulation of neutrophil effector functions.

(A and B) Phagocytosis of pHrodo-labeled K. pneumoniae, opsonized with serum (A) or IgG antibody (B), by Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt neutrophils. (C and D) Phagocytosis of phRodo-labeled K. pneumoniae, opsonized with serum (C) or IgG antibody (D), by Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt neutrophils treated with PP2, an SFK-specific inhibitor, or PP3 as negative control. (E and F) Adhesion-dependent oxidative burst of Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt neutrophils plated on fibrinogen alone or in the presence of serum-opsonized K. pneumoniae (E) or TNF (F). (G) NET formation of Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt neutrophils upon stimulation with untreated K. pneumoniae after 1 hour and 3 hours analyzed by flow cytometry. (H) Representative images of NET formation of Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt neutrophils after 3 hours of incubation with K. pneumoniae. DAPI staining is shown in blue color; H3Cit staining is visualized in orange color. Scale bars: 30 μm. (I) NET formation upon 1 hour or 3 hours of PMA stimulation analyzed by flow cytometry. n as indicated; n = 5 for E and F, mean ± SEM. *P < 0.05; **P < 0.01; ****P < 0.0001 by 2-way ANOVA with Tukey’s multiple-comparison test (A–D, G, and I) or 2-way ANOVA with Bonferroni’s multiple-comparison test (E and F).

Figure 4. Csk is involved in integrin-mediated neutrophil slow rolling, chemokine-induced arrest, and neutrophil recruitment in vivo.

(A–D) Intravital microscopy of postcapillary venules in the murine cremaster muscle 2 hours after intrascrotal TNF injection. (A) Rolling velocities of neutrophils from Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt mice. (B) Adherent cells per square millimeter and (C) the number of extravasated cells per 1.5 × 10⁴ μm² tissue area surrounding postcapillary venules. (D) Representative reflected light oblique transillumination microscopy photographs. White circles represent transmigrated neutrophils within the tissue, boxes indicate the analyzed tissue area. Scale bars: 50 μm. (E and F) Chemokine-induced arrest of neutrophils in postcapillary venules of Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt mice before and following CXCL1 injection. (E) Number of adherent cells per mm². (F) Representative images of postcapillary venules of Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt mice following CXCL1 injection. White circles represent adherent neutrophils within the vessel. Scale bars: 50 μm. (G–I) Carotid cannulas were placed in Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt mice and connected to autoperfused flow chambers. Average rolling velocity of neutrophils on (G) E-selectin and E-selectin/ICAM-1 and (H) P-selectin and P-selectin/ICAM-1. (I) Number of adherent cells on P-selectin/ICAM-1– and P-selectin/ICAM-1/CXCL1–coated flow chambers. n as indicated; n = 3 for E–H, mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001 by 2-tailed Student’s t test (A–C), 2-way-ANOVA with Šídák’s multiple-comparison test (E), or 2-way ANOVA with Tukey’s multiple-comparison test (G–I).

Figure 5. Csk is required for intravascular crawling and migration of leukocytes.

(A–C) Intravascular crawling of leukocytes in venules of the murine cremaster muscle during superfusion with CXCL2. Presented are the mean crawling velocity of adherent cells (A), mean distance crawled by adherent cells (B), and the percentage of adherent cells that crawled (C) in Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt mice. (D–H) Chemotaxis of isolated Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt neutrophils on fibronectin in response to a soluble CXCL1 gradient in vitro. Migration velocity (D), Euclidian and accumulated distance (E and F), and forward migration index (G) of chemotaxing neutrophils. (H) Representative trajectory plots of Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt neutrophils. n = 3 mice per genotype, mean ± SEM. *P < 0.05; **P < 0.01; ***P < 0.001 by 2-tailed Student’s t test.

Figure 6. Csk is involved in CD11a and CD11b activation regulation.

(A and B) Csk protein levels in HL-60 cells after lentiviral transduction with scrambled shRNA or shRNA against Csk. (A) Representative Western blots of HL-60 scrambled or HL-60 Csk-knockdown lysates, immunoblotted against total Csk (tCsk) and Vinculin. (B) Quantification of tCsk levels by Western blot. (C–E) HL-60 cells were analyzed using a flow chamber adhesion assay with E-selectin and either an Ab specific for the intermediate conformation of CD11a (KIM127) (C) or P-selectin, IL-8 and an Ab specific for the full open conformation of CD11a (mAb24) (D), or an Ab specific for the activation epitope of CD11b (CBRM1/5) (E), or a control IgG Ab. Adherent cells per field of view were counted. Analysis of (F) ICAM-1 binding and (G) CD11b-dependent fibrinogen binding in unstimulated and CXCL1-stimulated Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt neutrophils, measured by flow cytometry. n as indicated, mean ± SEM. **P < 0.01; ***P < 0.001; ****P < 0.0001 by 2-tailed Student’s t test (B) or 1-way ANOVA with Tukey’s multiple-comparison test (C–G).

Figure 7. Csk regulates the activity of Src kinases through a cAMP-dependent pathway.

(A) cAMP levels in cell lysates of murine bone marrow–derived WT neutrophils after stimulation with CXCL1 (2 minutes) or E-selectin (5 minutes). (B and C) Blood-perfused flow chambers coated with E-selectin or E-selectin/ICAM-1 were used to analyze rolling velocities. Rolling velocity of human neutrophils isolated from whole blood (B) and neutrophils isolated from Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt mice (C) after incubation with different concentrations of 8-CPT-cAMP, a cAMP analog and selective activator of the cAMP-dependent PKA. (D) Chemokine-induced arrest of neutrophils in postcapillary venules of Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt mice before and following CXCL1 injection, 30 minutes after intraarterial injection of the specific Src family kinase inhibitor PP2 or the inactive control PP3. Significant differences within the Csk^fl/flLyz2^wt/wt between PP2 and PP3 are marked in black; significant differences within the Csk^fl/flLyz2^cre/wt between PP2 and PP3 are marked in gray. (E–G) Bone marrow–derived neutrophils were left untreated or were stimulated with CXCL1 for 1 minute, E-selectin for 5 minutes, or serum-opsonized K. pneumoniae for 1 minute. Lysates were immunoblotted with an Ab against total Src (tSrc) and p-Src Y416 or Y529. (E–G) Representative Western blots of total lysates of Csk^fl/flLyz2^wt/wt and Csk^fl/flLyz2^cre/wt neutrophils showing the phosphorylation of Src Y416 and Y529 and total amounts of Src. n as indicated; n = 3–4 for D, mean ± SEM. *P < 0.05; **P < 0.01; ****P < 0.0001 by 1-way ANOVA with Tukey’s multiple-comparison test (A) or 2-way ANOVA with Šídák’s multiple-comparison test (B–D).