Skap2 is required for β2 integrin-mediated neutrophil recruitment and functions

Abstract

Integrin activation is required for neutrophil functions. Impaired integrin activation on neutrophils is the hallmark of leukocyte adhesion deficiency (LAD) syndrome in humans, characterized by impaired leukocyte recruitment and recurrent infections. The Src kinase-associated phosphoprotein 2 (Skap2) is involved in integrin functions in different leukocyte subtypes. However, the role of Skap2 in β₂ integrin activation and neutrophil recruitment is unknown. In this study, we demonstrate the crucial role of Skap2 in regulating actin polymerization and binding of talin-1 and kindlin-3 to the β₂ integrin cytoplasmic domain, thereby being indispensable for β₂ integrin activation and neutrophil recruitment. The direct interaction of Skap2 with the Wiskott-Aldrich syndrome protein via its SH3 domain is critical for integrin activation and neutrophil recruitment in vivo. Furthermore, Skap2 regulates integrin-mediated outside-in signaling events and neutrophil functions. Thus, Skap2 is essential to activate the β₂ integrins, and loss of Skap2 function is sufficient to cause a LAD-like phenotype in mice.

Figure 1.

Pivotal role of Skap2 for neutrophil recruitment during sterile inflammation. (A–E) Neutrophil recruitment of WT and Skap2^−/− mice in response to focal hepatic necrosis. (A) Adhesion of neutrophils to the microvascular endothelium after injury or sham operation. n = 4 mice/group. (B) Adherent neutrophils within the indicated regions around foci of necrosis 4 h after injury. n = 4 mice/group. (C) Representative time-lapse images of neutrophils (green) and area of focal hepatic necrosis (red). (D and E) Serum levels of GOT (D) and GPT (E) 4 h after injury or sham operation. n = 4 mice/group. (F–J) Neutrophil recruitment of WT and Skap2^−/− mice in response to renal IRI. (F) Creatinine serum levels 24 h after IRI or sham operation. n = 4 mice/group. (G) Neutrophil numbers per kidney 24 h after IRI or sham operation. n = 4 mice/group. (H) Representative hematoxylin and eosin staining of kidney outer medulla 24 h after IRI or sham operation. (I) Acute tubular necrosis scores in kidneys 24 h after IRI or sham operation. n = 4 mice/group. (J) Neutrophil counts in blood 24 h after IRI or sham operation. n = 6 mice/group. *, P < 0.01; **, P < 0.001; Student's t test. See also Videos 1 and 2. Data are means ± SEM.

Figure 2.

Skap2 regulates TNF-mediated neutrophil recruitment. (A–C) IVM of TNF-inflamed postcapillary venules of WT and Skap2^−/− mice. (A) Cumulative histogram of rolling velocities. (Inset) Data in a bar graph. 175 data points for each genotype are shown. (B and C) Number of adherent cells (B) and number of extravasated cells (C) 2 h after TNF application. n = 4 mice/group. (D) Representative images of transmigrated cells 2 h after TNF application. (E–G) IVM of TNF-inflamed postcapillary venules of WT mice after injection of differentially ex vivo–labeled WT and Skap2^−/− bone marrow cells. Rolling velocities (E), number of adherent cells (F), and number of extravasated cells (G) 2 h after TNF application are shown. The percentage of total fluorescent cells for each genotype is shown. n = 3 mice/group. (H) Peripheral blood cell counts of WT (n = 7) and Skap2^−/− (n = 5) mice. (I and J) Expression of the indicated myeloid maturation markers and cell surface receptors on WT and Skap2^−/− neutrophils. Black lines show WT, gray dashed lines show Skap2^−/−, and black dotted lines show isotype control. Quantification is shown below. n = 3. *, P < 0.05; **, P < 0.01; Student's t test. Data are means ± SEM.

Figure 3.

E-selectin–mediated slow rolling and GPCR-dependent adhesion are regulated by Skap2. (A and B) E-selectin–mediated slow rolling of WT and Skap2^−/− neutrophils. (A) In vivo rolling velocities in TNF-inflamed postcapillary venules treated with pertussis toxin and blocking anti–P-selectin antibody. (Inset) Cumulative histogram data in a bar graph are shown. n = 4 mice/group. (B) In vitro rolling velocities in autoperfused flow chambers coated with E-selectin or E-selectin and ICAM-1. n = 4 mice/group. (C) Knockdown of Skap2 in HL-60 cells. Quantification is shown on the right. n = 3. (D) Adhesion of control or Skap2 knockdown HL60 cells on flow chambers coated with E-selectin and an isotype, anti–β₂ integrin conformation reporter (KIM127), or anti–β₂ integrin (TS2/4) antibody. n = 3. (E and F) Adhesion of neutrophils in postcapillary venules of WT and Skap2^−/− mice after i.v. injection of CXCL1 (E) or LTB₄ (F). n = 4 mice/group. (G and H) IVM of postcapillary venules of WT and Skap2^−/− mice before or 1 h after superfusion with CXCL2. Number of adherent cells (G) and number of extravasated cells (H) are shown. n = 4 mice/group. (I) Adhesion of WT and Skap2^−/− neutrophils on autoperfused flow chambers coated with P-selectin/ICAM-1 or P-selectin/ICAM-1 and CXCL1 or PMA. n = 3 mice/group. (J) Soluble ICAM-1 binding of CXCL1-, PMA-, or Mn²⁺-stimulated WT and Skap2^−/− neutrophils. n = 3. (K) Adhesion of control or Skap2 knockdown HL60 cells on flow chambers coated with P-selectin, IL-8, and an isotype, anti–β₂ integrin conformation reporter (mAb24), or anti–β₂ integrin (TS2/4) antibody. n = 3. (L) Soluble ICAM-1 binding of IL-8–stimulated control or Skap2 knockdown HL-60 cells. n = 3. (M and N) LFA-1 clustering on WT and Skap2^−/− neutrophils. (M) Representative microscopy images. (N) Percentage of cells showing LFA-1 clusters after CXCL1 stimulation in solution or after plating on E-selectin with shear. 50 cells/experiment were analyzed. n = 3. (O–Q) WT and Skap2^−/− neutrophils were stimulated with CXCL1 for 3 min in solution, and lysates were immunoblotted with anti–p-ERK1/2 and anti-ERK1/2 (O), anti–p-p38 and anti-p38 (P), or anti–p-Akt, anti-Akt, and anti–α-tubulin antibody (Q). Quantification is shown on the right. n = 3. (R) Concentration of intracellular calcium measured in Indo-1–labeled WT and Skap2^−/− neutrophils before and after CXCL1 stimulation. n = 3. *, P < 0.05; **, P < 0.001; ***, P < 0.001; ^#, P < 0.05 versus all time points; Student's t test. Data are means ± SEM. Ctrl, control; E-Sel., E-selectin; P-sel., P-selectin.

Figure 4.

Skap2 is required for Mac-1–dependent intravascular crawling, extravasation, and chemotaxis of neutrophils. (A–D) Intravascular crawling of Ly-6G–labeled neutrophils in postcapillary venules of WT and Skap2^−/− mice during superfusion with CXCL2. (A) Percentage of adherent cells that crawled (A), crawling velocity (B), crawled distance (C), and representative images (D) are shown. The arrow indicates direction of movement. n = 4 mice/group. (E–G) Crawling of CXCL2-stimulated WT and Skap2^−/− neutrophils on serum-coated parallel plate flow chambers in vitro. Crawling velocity (E) and accumulated (F) and Euclidian (G) crawled distance before (preflow), during (flow), and after (postflow) applying flow at 2 dyn/cm² are shown. 80 cells/experiment were analyzed. n = 3. (H) Soluble fibrinogen binding of CXCL1-stimulated WT and Skap2^−/− neutrophils. n = 4. (I) Adhesion of control or Skap2 knockdown HL60 cells on flow chambers coated with P-selectin, IL-8, and an isotype, anti–Mac-1 activation reporter (CBRM1/5), or anti–Mac-1 (M1/70) antibody. n = 3. (J) Soluble fibrinogen binding of IL-8–stimulated control or Skap2 knockdown HL60 cells. n = 3. (K) Transmigration of WT and Skap2^−/− neutrophils through a TNF-stimulated bEnd.5 cell layer in response to a soluble gradient of CXCL1. n = 3. (L–P) Chemotaxis of WT and Skap2^−/− neutrophils in response to a soluble CXCL1 gradient in vitro. Representative trajectory plots (L), migration velocity (M), accumulated (N) and Euclidean distance (O), and forward migration index of chemotaxing neutrophils (P) are shown. 75 cells/experiment were analyzed. n = 4. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ^#, P < 0.001 versus all other groups; Student's t test (A–C, H–K, and M–P) or one-way ANOVA (E–G). Data are means ± SEM. Ctrl, control. See also Videos 3 and 4.

Figure 5.

Cross talk of Skap2 and WASp is indispensable for β₂ integrin activation and neutrophil recruitment. (A–C) IVM of TNF-inflamed postcapillary venules of WT, Was^−/−, and Skap2^−/−Was^−/− mice. Rolling velocities (A), number of adherent cells (B), and number of extravasated cells (C) 2 h after TNF application are shown. n = 4 mice/group. (D) Adhesion of neutrophils in postcapillary venules of WT, Was^−/−, and Skap2^−/−Was^−/− mice after i.v. injection of CXCL1. n = 4 mice/group. (E and F) Soluble ICAM-1 binding (E) and soluble fibrinogen binding (F) of CXCL1-stimulated WT, Was^−/−, and Skap2^−/−Was^−/− neutrophils. n = 3. (G) Immunoprecipitation of Skap2 in unstimulated WT neutrophils or neutrophils plated on E-selectin with shear or stimulated with CXCL1 in solution. Immunoprecipitates were immunoblotted with anti-WASp and anti-Skap2 antibody. Input was with anti–α-tubulin antibody. n = 3. (H) WT neutrophils were left unstimulated, plated on E-selectin with shear, or stimulated with CXCL1 in solution. Lysates were incubated with GST alone (control), GST fusion proteins of the Skap2 CC, PH, or SH3 domains, or full-length Skap2. Precipitates were immunoblotted with anti-WASp, and input was with anti–α-tubulin antibody. n = 3. (I and J) In vitro co-purification of His-WASp by different Skap2 GST fusion proteins. Precipitates were immunoblotted with anti-His or anti-GST, and input controls were with anti-GST antibody. n = 3. (K–N) WT and Skap2^−/− or Was^−/− neutrophils were left unstimulated, plated on E-selectin with shear, or stimulated with CXCL1 in solution. Lysates were immunoprecipitated with anti-WASp (K and L) or anti-Skap2 (M and N) antibody followed by immunoblotting with anti-phosphotyrosine (4G10), anti-WASp, or anti-Skap2 antibody. Input was immunoblotted with anti–α-tubulin antibody. Quantification is shown on the right. n = 3. *, P < 0.05; ^#, P < 0.05 versus all time points; Student's t test (A–D and K–N) or one-way ANOVA (E and F). Data are means ± SEM. Ctrl, control; E-Sel., E-selectin; IP, immunoprecipitate; PD, precipitate.

Figure 6.

The Skap2 SH3 domain mediates WASp localization and integrin activation. (A and B) Subcellular localization of Skap2 and WASp in unstimulated or CXCL1-stimulated WT or Skap2^−/− neutrophils. (A) Representative microscopy images. (B) Statistics of WASp plasma membrane localization. 75 cells/experiment were analyzed. n = 3. (C) Expression of Skap2 in WT, Skap2^−/−, and Skap2^−/− neutrophils reconstituted with Skap2, vector control, Skap2 R140M, or Skap2ΔSH3. Lysates were immunoblotted with anti-Skap2 and anti–α-tubulin antibody. n = 2. (D) Subcellular localization of Skap2 and WASp in unstimulated or CXCL1-stimulated Skap2^−/− neutrophils reconstituted with Skap2, Skap2ΔSH3, or Skap2 R140M. n = 3. (E) Statistics of WASp plasma membrane localization. 75 cells/experiment were analyzed. n = 3. (F) Phosflow analysis of pAKT in Skap2^−/− neutrophils reconstituted with the vector control, Skap2, or Skap2 R140M after stimulation with CXCL1. Fold-increase in pAKT compared with the unstimulated control is shown. n = 2. (G–I) IVM of TNF-inflamed postcapillary venules of WT mice after bone marrow transplantation of Skap2^−/− hematopoietic stem cells retrovirally transduced with the vector control, Skap2, or Skap2ΔSH3. Rolling velocity (G), number of adherent cells (H), and number of extravasated cells (I) 2 h after TNF application are shown. n = 3 mice/group. (J) Soluble ICAM-1 binding of CXCL1-stimulated transduced Skap2^−/− neutrophils. n = 3. (K) Expression of Skap2 in WT, Skap2^−/−, and Skap2^−/− neutrophils reconstituted with Skap2, vector control, or Skap2 W336K. Lysates were immunoblotted with anti-Skap2 and anti–α-tubulin antibody. n = 2. The same experiment as in C is shown. (L–N) IVM of TNF-inflamed postcapillary venules of WT mice after bone marrow transplantation of Skap2^−/− hematopoietic stem cells retrovirally transduced with the vector control, Skap2, or Skap2 W336K. Rolling velocity (L), number of adherent cells (M), and number of extravasated cells (N) 2 h after TNF application are shown. n = 3 mice/group. (O) Soluble ICAM-1 binding of CXCL1-stimulated reconstituted Skap2^−/− neutrophils. n = 3. *, P < 0.05; **, P < 0.01; Student's t test (B), one-way ANOVA (E–I and L–N), or two-way ANOVA (J and O). Data are means ± SEM. Ctrl, control; MSCV, murine stem cell virus.

Figure 7.

The Skap2/WASp complex triggers actin polymerization, thereby enabling β₂ integrin activation by recruitment of talin-1 and kindlin-3. (A and B) F-actin polymerization of WT, Skap2^−/−, Was^−/−, and Skap2^−/−Was^−/− neutrophils (A) and Skap2^−/− neutrophils reconstituted with vector control, Skap2, or Skap2 W336K (B) after stimulation with CXCL1 in solution. n = 3. (C and D) Soluble ICAM-1 binding (C) and LFA-clustering (D) of CXCL1-stimulated WT, Skap2^−/−, and Was^−/− neutrophils after pretreatment with DMSO or Lat. A. For clustering, 50 cells/experiment were analyzed. n = 3. (E) Knockdown of WASp in HL-60 cells. Quantification is shown on the right. n = 3. (F–I) Control, Skap2, or WASp knockdown HL60 cells were pretreated with DMSO or Lat. A and left unstimulated, plated on E-selectin with shear, or stimulated with IL-8 in solution. Lysates were incubated with GST fusion proteins of the β₂ integrin cytoplasmic domain (F and G) or immunoprecipitated with anti–β₂ integrin antibody (H and I). Precipitates were immunoblotted with anti–talin-1 and anti–kindlin-3 or anti–β₂ integrin antibody, respectively. Input was immunoblotted with anti-GAPDH antibody. Quantifications are shown below. n = 3. *, P < 0.001; ^#, P < 0.05 versus all other groups; one-way ANOVA (A–B and F–I), two-way ANOVA (C and D), or Student's t test (E). Data are means ± SEM. Ctrl, control; E-Sel., E-selectin; IP, immunoprecipitate; PD, precipitate.

Figure 8.

Integrin activation is dependent on WASp-mediated de novo actin polymerization but also on PIP₂ synthesis. (A–C) F-actin polymerization (A), soluble ICAM-1 binding (B), and LFA-1 clustering (C) of Was^−/− neutrophils reconstituted with WASp, vector control, or a WASp mutant lacking the VCA domain (WASpΔVCA) after stimulation with CXCL1 in solution. n = 3. (D and E) PIP₂ (D) and PIP₃ (E) levels of WT, Skap2^−/−, and Was^−/− neutrophils after CXCL1 stimulation in solution or plating on E-selectin with shear. n = 2. (F) Subcellular localization of talin-1 in unstimulated or CXCL1-stimulated WT neutrophils pretreated with DMSO, PAO, or Lat. A. Representative images and statistics of talin-1 plasma membrane localization are shown. 40 cells/experiment were analyzed. n = 3. (G) Immunoprecipitation of β₂ integrin in IL-8– or E-selectin–stimulated HL-60 cells after pretreatment with DMSO or PAO. Precipitates were immunoblotted with anti–talin-1 and anti–kindlin-3 or anti–β₂ integrin antibody. Input was immunoblotted with anti-GAPDH antibody. Quantification is shown below. n = 3. *, P < 0.05; one-way ANOVA (A and D–F), two-way ANOVA (B and C), or Student's t test (G). Data are means ± SEM. Ctrl, control; E-sel., E-selectin; IP, immunoprecipitate; MSCV, murine stem cell virus.

Figure 9.

Skap2 regulates integrin-mediated outside-in signaling. (A–D) Neutrophils from WT or Skap2^−/− mice were plated on pRGD and adhesion (A) and spreading (B and C) were analyzed. (D) Representative images of spread neutrophils. 75 cells/experiment were analyzed. n = 3. (E–G) Adhesion-dependent oxidative burst of WT and Skap2^−/− neutrophils plated on pRGD alone (E) or ICAM-1 (F) and fibrinogen (G) in the presence of TNF. n = 4. (H) Oxidative burst of WT and Skap2^−/− neutrophils stimulated with PMA in solution. n = 4. (I and J) WT and Skap2^−/− neutrophils were plated on pRGD for 10 min, and lysates were immunoblotted with anti–p-ERK1/2 and anti-ERK1/2 (I) or anti–p-Akt and anti-Akt (J) and anti–α-tubulin antibody. Quantification is shown on the right. n = 4. *, P < 0.05; **, P < 0.01; ^#, P < 0.05 versus all other time points; Student's t test. Data are means ± SEM. Ctrl, control.

Figure 10.

Skap2 regulates TNF-mediated integrin activation and outside-in signaling as well as neutrophil functions. (A) Soluble ICAM-1 binding of TNF-stimulated WT and Skap2^−/− neutrophils. n = 3. (B) Oxidative burst of WT and Skap2^−/− neutrophils stimulated with TNF in solution. n = 3. (C and D) Surface expression of CD11b (C) and CD18 (D) of TNF-stimulated WT or Skap2^−/− neutrophils. Quantification is shown below. n = 3. (E–G) Phagocytosis of pHrodo E. coli particles by WT and Skap2^−/− neutrophils. Phagocytosis (E), percentage of phagocytic cells (F), and representative images (G) of phagocytic neutrophils are shown. n = 3. (H) In vitro NET formation of WT and Skap2^−/− neutrophils stimulated with resting or TRAP-activated platelets. n = 3. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ^#, P < 0.05 versus all other time points; Student's t test. Data are means ± SEM. Ctrl, control.