SKAP2 acts downstream of CD11b/CD18 and regulates neutrophil effector function

Abstract

Background: The importance of CD11b/CD18 expression in neutrophil effector functions is well known. Beyond KINDLIN3 and TALIN1, which are involved in the induction of the high-affinity binding CD11b/CD18 conformation, the signaling pathways that orchestrate this response remain incompletely understood.

Method: We performed an unbiased screening method for protein selection by biotin identification (BioID) and investigated the KINDLIN3 interactome. We used liquid chromatography with tandem mass spectrometry as a powerful analytical tool. Generation of NB4 CD18, KINDLIN3, or SKAP2 knockout neutrophils was achieved using CRISPR-Cas9 technology, and the cells were examined for their effector function using flow cytometry, live cell imaging, microscopy, adhesion, or antibody-dependent cellular cytotoxicity (ADCC).

Results: Among the 325 proteins significantly enriched, we identified Src kinase-associated phosphoprotein 2 (SKAP2), a protein involved in actin polymerization and integrin-mediated outside-in signaling. CD18 immunoprecipitation in primary or NB4 neutrophils demonstrated the presence of SKAP2 in the CD11b/CD18 complex at a steady state. Under this condition, adhesion to plastic, ICAM-1, or fibronectin was observed in the absence of SKAP2, which could be abrogated by blocking the actin rearrangements with latrunculin B. Upon stimulation of NB4 SKAP2-deficient neutrophils, adhesion to fibronectin was enhanced whereas CD18 clustering was strongly reduced. This response corresponded with significantly impaired CD11b/CD18-dependent NADPH oxidase activity, phagocytosis, and cytotoxicity against tumor cells.

Conclusion: Our results suggest that SKAP2 has a dual role. It may restrict CD11b/CD18-mediated adhesion only under resting conditions, but its major contribution lies in the regulation of dynamic CD11b/CD18-mediated actin rearrangements and clustering as required for cellular effector functions of human neutrophils.

Keywords: CD11b/CD18 integrin; Src kinase associated phosphoprotein 2 (SKAP2); adhesion; antibody-dependent cellular cytotoxicity (ADCC); filamentous actin; neutrophils; phagocytosis.

Figure 1

KINDLIN3 interactome in NB4 cells under steady state reveals potential proteins involved in CD11b/CD18 signaling. (A) Representative western blot analysis of total NB4 wild-type (WT) cell lysates or cells lacking KINDLIN3 or reconstituted with BirA-KINDLIN3. Lysates were immunoblotted for KINDLIN3, HA-tag, and α-TUBULIN. (B) NADPH oxidase activity of NB4 BirA-KINLDIN3-expressing neutrophils in the presence or absence of biotin after stimulation with unopsonized or serum-treated (STZ) zymosan yeast particles (three independent experiments). (C) Adhesion of NB4 neutrophils in the presence or absence of biotin stimulated with dithiothreitol (DTT) or phorbol 12-myristate 13-acetate (PMA) (four independent experiments). (D) Representative streptavidin affinity purification of NB4 WT or BirA-KINDLIN3-expressing neutrophils after incubation with biotin. Lysates were immunoblotted for streptavidin. (E) Label-free quantification (LFQ) intensity plot depicts the levels of biotinylated FERMT3 gene (KINDLIN3 protein) identified in each condition after LC-MS/(MS) analysis. (F) Venn diagram depicts the relation of the protein populations after comparison of NB4 WT neutrophils in the presence of biotin or NB4 BirA-KINLDIN3 neutrophils in the absence of biotin, with NB4 BirA-KINDLIN3 neutrophils in the presence of biotin. In red, the selected population of interest. (G) Volcano plot depicts the notable protein differences between NB4 neutrophils expressing BirA-KINDLIN3 and WT cells in the presence of biotin. Red circles indicate KINDLIN3 (FERMT3 gene), SKAP2, ILK, and TALIN1 (TLN1 gene). (H) Gene ontology (GO) term analysis for KINDLIN3. (I) label-free quantification (LFQ) intensity plot depicts the levels of SKAP2 biotinylation in each condition after LC-MS/(MS) analysis. Bars show mean ± SEM. Statistics: one-way ANOVA with Tukey correction: ns, nonsignificant; *p < 0.05; ***p < 0.001, ****p < 0.0001.

Figure 2

SKAP2 interacts or is associated with CD11b/CD18 at steady state. (A) CD18 precipitation in primary neutrophils from two donors under steady state. Lysates were immunoblotted for CD18 and SKAP2. (B) Representative microscopy images of NB4 WT or SKAP2 knockout (SKAP2KO) cells, fixed, permeabilized, and stained for SKAP2 (in purple) or nuclear staining DAPI (in cyan). (C) Representative histograms show the geometric mean fluorescent intensity (GeoMFI) for the expression of FcγRIIIb, IIa or I, or CD11b in NB4 scrambled (Scr), CD18KO, KINDLIN3KO, and SKAP2KO neutrophils. Histograms are shown as normalized to mode. (D) Representative CD18 precipitation in NB4 Scr, KINDLIN3KO, or SKAP2KO neutrophils under steady state. Lysates were immunoblotted for CD18, KINDLIN3, TALIN1, and SKAP2. Immunoprecipitation, IP; total cell lysate, TCL.

Figure 3

SKAP2 deficiency impairs CD18-dependent effector responses. (A) NADPH oxidase activity for NB4 neutrophils either untreated or after stimulation with zymosan yeast particles, serum-treated zymosan (STZ), or PMA (five independent experiments). (B) Representative frames from live cell imaging depict the phagocytosis of unopsonized zymosan (in green) by NB4 Scr or SKAP2KO cells (white arrows indicate phagocytosis, white circles indicate zymosan binding). (C, D). Kinetics of the percentage of phagocytic cells over time toward unopsonized (C) or serum-opsonized (D) zymosan using flow cytometry (five independent experiments). (E) Antibody-dependent cellular cytotoxicity (ADCC) of NB4 WT, Scr, or SKAP2KO neutrophils toward EGFR-expressing A431 cells in increasing target–effector ratios from 1 in 25 to 1 in 200, in the presence or absence of monoclonal antibody cetuximab (12 independent experiments). Bars show mean± SEM. Statistics: A, mixed one-way ANOVA with Tukey correction; (C–E), unpaired t-test. ns, nonsignificant; *p < 0.05; **p < 0.001; ***p < 0.001; ****p < 0.0001. cetuximab, Cmab.

Figure 4

SKAP2 orchestrates the interplay between actin cytoskeleton rearrangements and integrin-mediated adhesion. (A) Adhesion of NB4 neutrophils in the presence or absence of blocking antibodies against CD11b and CD18 after stimulation with DTT or PMA. No coating was included (≥4 data points from four independent experiments). (B) NB4 WT neutrophils unstimulated or stimulated with N-formyl-Met-Leu-Phe (fMLF; fMLP) on fibronectin-coated glass. The cells were stained for filamentous actin (F-actin, green), nuclear stain DAPI (cyan), and SKAP2 (magenta). Zoomed images highlight the distribution of F-actin and co-localization with SKAP2. (C) Adhesion of NB4 neutrophils after preincubation with vehicle control or latrunculin B and stimulation with DTT or PMA (≥6 data points from five independent experiments). Bars show mean ± SEM. Statistics: A, mixed one-way ANOVA with Tukey correction; C, one-way ANOVA with Sidak correction. ns, nonsignificant; **p < 0.001; ***p < 0.0001. Latrunculin B, LatB.

Figure 5

CD18 cluster formation and mediated cell avidity are altered in the absence of SKAP2. (A) Representative microscopy images from NB4 WT or SKAP2KO cells on fibronectin-coated glass and stimulation with DTT. Cells were stained with CD18 (in red). Zoomed images highlight CD18 clustering. (B) Quantification of CD18 clustering after analysis of the mean fluorescent intensity (MFI) in the presence or absence of DTT. Unstimulated condition was used as a cut-off value for cluster formation. The results are normalized to the expression of CD18 in absence of stimulus, per cell line (four independent experiments). Bars show mean ± SEM. Statistics: unpaired t-test. **p < 0.001.