Advertisement

Differential attenuation of β2 integrin–dependent and –independent neutrophil migration by Ly6G ligation

Pierre Cunin, Pui Y. Lee, Edy Kim, Angela B. Schmider, Nathalie Cloutier, Alexandre Pare, Matthias Gunzer, Roy J. Soberman, Steve Lacroix, Eric Boilard, Craig T. Lefort and Peter A. Nigrovic

Data supplements

Article Figures & Data

Figures

  • Figure 1.

    Ly6G ligation but not deficiency attenuates IgG-mediated arthritis. (A-B) Male mice were treated with K/BxN serum 150 μL intraperitoneally on days 0 and 2 and assessed for arthritis over 8 days, measuring clinical scoring on a 0 to 12 scale (A) and change in ankle and wrist thickness (B). Results pooled from 2 identical experiments using a single pooled batch of K/BxN serum. Ly6G+/+ (WT, blue line; n = 4), Ly6G+/− (heterozygous [HET], green line; n = 8), and Ly6G−/− (knockout [KO], red line; n = 6). Clinical scoring and thickness change: P = not significant (ns). (C-D) WT and KO mice were treated with 5 μg of 2A3 or 1A8 1 day prior (day −1) and 2 days after (day +2) arthritis induction. (C) Clinical scoring: 2A3- vs 1A8-treated WT: P = .0013; 2A3- vs 1A8-treated KO: P = ns. (D) Thickness change: 2A3- vs 1A8-treated WT: P = .0058; 2A3- vs 1A8-treated KO: P = ns; n = 10-11 mice per group. **P ≤ .01.

  • Figure 2.

    Neutrophil recruitment to lung does not require β2 integrins and is unimpaired by Ly6G ligation. (A) PMNs were enriched from WT and CD18−/− marrow, stained with CVm (far-red) or PKH67 (green), and mixed at a 1:1 ratio (colors for WT and CD18−/− cells varied across experiments). Cells were injected IV into mice treated intranasally with IL-1β (25 ng) or LPS (1 μg) 2 hours previously. Four hours after cell injection, the presence of far-red PMNs (WT) and green PMNs (CD18−/−) in the BAL and in lung parenchyma was assessed by fluorescence-activated cell sorting (FACS). (B) Ratio of WT-to-CD18−/− PMNs in parenchyma and BAL fluid, n = 9-11 mice per group; mean ± SEM and deviation from 1:1 are: parenchyma IL-1β, 1.28 ± 0.11 (P = .03); parenchyma LPS, 1.41 ± 0.12 (P = .007); BAL IL-1β, 0.87 ± 0.10 (P = ns); BAL, 1.07 ± 0.18 (P = ns). (C) Mice were injected intraperitoneally with 5 μg of 1A8 or 2A3 and 18 hours later 25 ng of IL-1β was administrated intranasally. PMNs in the BAL and in lung parenchyma were counted by flow cytometry 6 hours later; n = 7-8 mice per group.

  • Figure 3.

    The dependence of neutrophils on β2 integrins for migration to peritoneum varies with stimulus. (A-B) WT and CD18−/− mice were injected intraperitoneally with E coli (107 CFU per mouse) or IL-1β (5 ng per mouse). Three hours later, PMNs in peritoneum lavage (A) and blood (B) were counted by FACS using counting beads (left histograms). (C) Percentage of Gr-1/Ly6C-high PMNs among CD45+ cells in WT and CD18−/− bone marrow (left), blood (middle), and peritoneum lavage (right) was evaluated by FACS. Representative of at least 5 mice. (D) Ratio of migrated-to-circulating PMNs, allowing an estimation of the relative PMN migration in WT and CD18−/− mice; n = 3-8 mice per group. (E) PMNs were enriched from WT and CD18−/− marrow, stained with PKH67 (green) or CVm (far-red), and mixed at a 1:1 ratio (colors for WT and CD18−/− cells varied across experiments). Cells were injected IV into mice treated with E coli or IL-1β intraperitoneally 1 hour previously. Three hours after cell injection, the presence of green PMNs (WT) and far-red PMNs (CD18−/−) in peritoneal lavage was assessed by FACS. (F) Calculation of the ratio of migrated WT-to-CD18−/− PMNs allows an estimation of the β2 integrin dependency in these 2 models of peritonitis; n = 13 mice per group. ***P ≤ .001. FSC, forward scatter; SSC, side scatter.

  • Figure 4.

    Ly6G ligation selectively impairs β2 integrin–dependent neutrophil migration into peritoneum. (A-B) Mice were injected intraperitoneally with 5 μg of anti-Ly6G (clone 1A8) or isotype control (clone 2A3) at day 0 and day 2. Five days later, mice were injected intraperitoneally with E coli (A) or IL-1β (B). Circulating (left panels) and migrated (right panels) PMNs were enumerated by flow cytometry; n = 10-18 mice per group. (C) Mice injected intraperitoneally with 1A8 or 2A3 at day 0 and day 2 were treated at day 5 with E coli or IL-1β prior injection of a mix of WT PMNs stained with PKH67 (green) and CD18−/− PMNs stained with CVm (far-red), as in Figure 3E. Three hours later, WT and CD18−/− PMNs in peritoneal lavage were counted by flow cytometry. (D) Calculation of the ratio of WT PMN-to-CD18−/− PMNs that migrate during peritonitis; n = 8 per group in E coli–treated mice and 14 per group in IL-1β–treated mice. *P ≤ .05; **P ≤ .01.

  • Figure 5.

    Ly6G ligation modulates the spatial relationship of Ly6G with β2 integrins. (A) Mice were treated with either E coli or IL-1β intraperitoneally 3 hours later, cells from peritoneal lavage were fixed and stained with Ly6G (AF594, red) and CD18 (FITC, green) monoclonal antibodies (mAbs). (Left) Representative photos from confocal microscope. (Right) Pearson coefficient in PMNs from E coli– or IL-1β–treated mice. (B) Peritoneal PMN from E coli– or IL-1β–treated mice were stained with FITC anti-CD18 (donor) and AF594 anti-Ly6G (acceptor) mAbs. Left, Representative images from each group displaying interacting τ1 lifetime in each pixel on a pseudocolor scale. Right, τ1 lifetime in at least 55 different pixels, representative of 3 independent experiments. Each image in panels A-B represents 12 × 12 μm. (C) PMNs were incubated with 10 μg/mL 1A8 or 2A3 and stimulated or not with 20 nM LTB4. After fixation, cells were stained with anti-CD18 (donor, “D”) without or with anti-Ly6G (acceptor, “D+A”) for FLIM. At least 55 different pixels per conditions were analyzed. Results are expressed as a percentage of τ1 lifetime in donor conditions and are representative of 2 independent experiments. *P ≤ .05; ***P ≤ .001.

  • Figure 6.

    Ly6G ligation impairs neutrophil attachment to endothelium. (A-B) Mice were treated with 5 µg of 2A3 or 1A8. Four hours later, leukocyte arrest in postcapillary venules of the cremaster muscle is induced by intra-arterial injection of 1 µg of CXCL1. (A) Duration of attachment of neutrophils on the endothelium in seconds. (B) Rolling flux was quantified as the number of leukocytes rolling past an arbitrary landmark, counted in 1-minute segments and then normalized to the leukocyte rolling flux prior to CXCL1 injection; n = 5 mice per group. ***P ≤ .001.