Peripheral blood leukocyte counts. (A) Total leukocyte counts. Blood was collected from the tail vein of 6 male mice of each genotype. The total leukocytes were counted on a hemocytometer. (B) Leukocyte subset counts. Flow cytometry was used to determine the fraction of each subset. Neutrophils were determined as Gr-1^highF4/80^low cells, eosinophils as Gr-1^intermediateF4/80^high cells (these cells are also very high in SSC), and monocytes as Gr-1^highF4/80^high and Gr-1^-F4/80^high cells. Values are means plus or minus SEM; *P < .05; **P < .01; ***P < .001.

Neutrophil migration in thioglycolate-induced peritonitis. WT, L-selectin^−/−, PSGL-1^−/−, and DKO mice given an intraperitoneal injection of thioglycolate. Absolute neutrophil counts (A,B) and macrophage counts (C,D) in the peritoneal exudates were determined 4 hours (A,C) and 8 hours (B,D) after the thioglycolate injection. Each circle represents the value of an individual mouse. Bars express the mean. *P < .05; **P < .01; ***P < .001.

Leukocyte rolling flux fraction in TNF-α–treated cremaster muscle venules. (A,B) Leukocyte rolling flux fraction in 2.5-hour (A) and 6-hour (B) TNF-α–treated cremaster muscle venules of WT, L-selectin^−/−, PSGL-1^−/−, and DKO mice. (C) Effect of anti–E-selectin and anti–L-selectin mAbs on leukocyte rolling in 2.5-hour TNF-α–treated cremaster muscle venules of PSGL-1^−/− mice. Leukocyte rolling flux fractions were determined in PSGL-1^−/− mice that had received the anti–E-selectin mAb 9A9, the F(ab′)₂ fragments of the anti–L-selectin mAb MEL-14, or isotype controls.

Serum levels of G-CSF, IL-17, and GM-CSF are elevated in mice deficient in both L-selectin and PSGL-1. Serum levels of G-CSF (A), GM-CSF (C), and IL-3 (D) were determined by a Bio-Plex cytokine assay. Serum levels of IL-17 (B) were determined by ELISA. Data are expressed as means plus or minus SEM from 8 mice per group. *P < 0.05; **P < 0.01; ***P < .001 versus WT.

Competitive migration of PSGL-1^−/− and DKO neutrophils into the inflamed peritoneum. Peripheral blood leukocytes were isolated from PSGL-1^−/− and DKO mice and labeled with CFSE and TRITC, respectively. The mixture of these cells was injected via the tail vein into recipient DKO mice that had just received thioglycolate intraperitoneally. After 4 and 8 hours, the blood and peritoneal exudate cells were collected, stained with anti–Gr-1-APC, and analyzed by flow cytometry. (A) Representative plots showing CFSE-labeled PSGL-1^−/− neutrophils and TRITC-labeled DKO neutrophils in the input, and the peritoneal exudates and peripheral blood collected 4 hours after injection. (B) The ratio of DKO neutrophils to PSGL-1^−/− neutrophils in the peritoneal exudates and peripheral blood normalized by the input ratio. Data are expressed as means plus or minus SEM from 3 mice. Results represent one of 3 similar experiments. *P < .05; ***P < .001.

Cremaster muscle endothelial cells bind L-selectin. (A) L-selectin–IgG binding to neutrophils. Peripheral blood leukocytes from WT, PSGL-1^−/−, and FucT-IV^−/−/VII^−/− mice treated with or without an intrascrotal injection of TNF-α were stained with L-selectin–IgG complexed with PE-labeled goat F(ab′)₂ anti–human IgG with (shaded histograms) or without (open histograms) EDTA, and then stained with anti–Gr-1-APC. Cells were gated for neutrophils (SSC^highGr-1^high). (B) Expression of endothelial markers on cremaster muscle endothelial cells. Cremaster muscle was removed from WT mice treated with or without an intrascrotal injection of TNF-α. Single-cell suspensions enriched for endothelial cells by Percoll were stained with the indicated mAbs (open histograms) or isotype controls (shaded histograms) followed by secondary reagents together with anti–CD31-APC and anti–CD45-FITC or -PE-Cy5. Cells were gated on CD31⁺CD45⁻, as shown in the top panel. (C-E) L-selectin–IgG binding to cremaster muscle endothelial cells. Single-cell suspensions from the cremaster muscle were stained with L-selectin–IgG complexed with PE-labeled goat F(ab′)₂ anti–human IgG and then stained with anti–CD31-APC and anti–CD45-FITC. Cells were gated on CD31⁺CD45⁻. (C) L-selectin–IgG binding to cremaster muscle endothelial cells from WT, PSGL-1^−/−, and FucT-IV^−/−/VII^−/− mice treated with or without an intrascrotal injection of TNF-α. The cells were stained with (shaded histograms) or without (open histograms) EDTA. Percentages of L-selectin–binding cells are given. The numbers in parentheses indicate mean fluorescence intensity values. (D) Inhibition of L-selectin–IgG binding to cremaster muscle endothelial cells by the anti–L-selectin mAb MEL-14. The L-selectin–IgG secondary antibody complex was incubated with MEL-14 (dotted line) or isotype control (solid line) before added to the cells. (E) L-selectin–IgG binding to cremaster muscle endothelial cells treated with heparitinase. The cells were treated with heparitinase before being analyzed for L-selectin–IgG binding. The cells were stained with (shaded histograms) or without (open histograms) EDTA. The cells were also stained with (open histograms) or without (shaded histograms) the anti–Δ-heparan sulfate mAb 3G10 followed by FITC-labeled donkey anti–mouse IgG. Results represent one of 3 to 5 similar experiments.