Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 May 30;121(22):4555-66.
doi: 10.1182/blood-2012-09-459636. Epub 2013 Apr 2.

The CXCR1/2 ligand NAP-2 promotes directed intravascular leukocyte migration through platelet thrombi

Mehran Ghasemzadeh  1 , Zane S Kaplan, Imala Alwis, Simone M Schoenwaelder, Katrina J Ashworth, Erik Westein, Ehteramolsadat Hosseini, Hatem H Salem, Robyn Slattery, Shaun R McColl, Michael J Hickey, Zaverio M Ruggeri, Yuping Yuan, Shaun P Jackson
Affiliations

The CXCR1/2 ligand NAP-2 promotes directed intravascular leukocyte migration through platelet thrombi

Mehran Ghasemzadeh et al. Blood. .

Abstract

Thrombosis promotes leukocyte infiltration into inflamed tissues, leading to organ injury in a broad range of diseases; however, the mechanisms by which thrombi guide leukocytes to sites of vascular injury remain ill-defined. Using mouse models of endothelial injury (traumatic or ischemia reperfusion), we demonstrate a distinct process of leukocyte recruitment, termed "directed intravascular migration," specifically mediated by platelet thrombi. Single adherent platelets and platelet aggregates stimulated leukocyte shape change at sites of endothelial injury; however, only thrombi were capable of inducing directed intravascular leukocyte migration. Leukocyte recruitment and migration induced by platelet thrombi occurred most prominently in veins but could also occur in arteries following ischemia-reperfusion injury. In vitro studies demonstrated a major role for platelet-derived NAP-2 (CXCL-7) and its CXCR1/2 receptor in regulating leukocyte polarization and motility. In vivo studies demonstrated the presence of an NAP-2 chemotactic gradient within the thrombus body. Pharmacologic blockade of CXCR1/2 as well as genetic deletion of NAP-2 markedly reduced leukocyte shape change and intrathrombus migration. These studies define a distinct process of leukocyte migration that is initiated by homotypic adhesive interactions between platelets, leading to the development of an NAP-2 chemotactic gradient within the thrombus body that guides leukocytes to sites of vascular injury.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Platelet thrombi induce leukocyte recruitment and intravascular migration following intestinal IR injury. Spontaneous thrombus formation and leukocyte interactions in the mesenteric vasculature were examined after IR injury by using histology and real-time DIC, epifluorescence, and confocal microscopy. (A) The percentage of microvascular vessels with either partially or fully occlusive thrombi was quantified by using Carstairs staining of histologic sections (mean ± standard error of the mean [SEM]; Control group: n = 3 mice and 24 sections with 380 vessels counted; IR group: n = 5 mice and 40 histologic sections with 893 vessels counted). (B-C) Representative histologic sections of the small bowel vasculature demonstrating (B) occlusive fibrin-rich thrombi (dark red) and thrombi containing both platelets (*) and (C) fibrin with numerous leukocytes within the thrombus body after IR injury. (D) Representative DIC and fluorescence images illustrating polarized leukocytes (Gr-1 Ab, green) within the spontaneous thrombus body (DIC, demarcated) in an IR-injured mesenteric vein. (E) Representative DIC image and corresponding three-dimensional reconstruction of spontaneous platelet rich thrombi (GPIbβ Ab, blue) within the bowel wall microvasculature associated with leukocyte accumulation (Gr-1 Ab, red) after IR injury. (F) Representative images depicting polarized and spread leukocytes (Gr-1 Ab, green) in the presence [ii) Platelets] but not absence [i) Endothelium] of adherent platelets on the endothelium after IR injury. (G) The number of polarized/spread leukocytes per square millimeter on the surface of endothelium or spontaneous platelet thrombi following IR injury. *P < .05;***P < .001.
Figure 2
Figure 2
Microvascular platelet thrombi induce directed intravascular leukocyte conveyance in response to localized endothelial injury. (A-F) GFP or C57Bl/6 mouse mesenteric veins were subjected to needle injury with local microinjection of thrombin, and the thrombus formation and leukocyte recruitment were monitored by confocal, epifluorescence, or DIC microscopy. (A) Representative DIC and fluorescence images of thrombi (red) and leukocyte recruitment (green) in mesenteric veins of C57Bl/6 mice following repetitive injury at the indicated time postinjury. (B) Time course of leukocyte recruitment to thrombi expressed as number per unit volume (mean ± SEM; n = 4), and quantified as described in “Materials and methods” and supplemental Methods, Quantitative analysis of leukocyte recruitment and adhesion to thrombi in vivo. (C) Left panel: representative image depicting migrating leukocytes (Gr-1 Ab, green) at different positions between the margin and the center (the site of vascular injury) of a thrombus (3,3′-dihexyloxacarbocyanine iodide [DiOC6], red) 10 minutes postinjury; the inset demonstrates the polarized morphology of individual migrating leukocytes. Right panel: migration paths (dotted lines) of individual leukocytes moving from the margins of the thrombus toward the site of vascular injury (thrombus core, red shaded area). (D) Migration velocity (µm/min) of individual leukocytes determined by using ImageJ software (n = 30). (E) Images depicting the presence of leukocytes (Gr-1 Ab, green) within the three-dimensional thrombus body (upper left), taken from front, back, and lateral perspectives 30 minutes postinjury. (F) Images depicting leukocyte migration (Gr-1 Ab, green) through the Top, Middle, and Base of a representative thrombus (red, 30-mm height as schematically depicted on the upper right) at the indicated time postinjury.
Figure 3
Figure 3
Leukocyte recruitment by platelet thrombi in injured arteries. (A-D) C57Bl/6 mice were administered an anti–Gr-1 Ab and DiOC6 prior to needle injury of arteries. Subsequent leukocyte thrombus interactions were monitored by DIC and fluorescence microscopy. The number of (A) stably adherent leukocytes and (C) migrating leukocytes to the site of vascular needle injury were quantified in sham-operated C57Bl/6 mice (Control), and mice subjected to IR injury (IR) (mean ± SEM; Control group: n = 7 mice with 13 injuries; IR group: n = 8 mice with 14 injuries). (B) Representative DIC images of thrombi induced by needle injury in sham-operated mice (Control) or mice after IR injury (IR). Note the (B) significant leukocyte recruitment (pseudo-colored green) and (D) migration (Gr-1 Ab, green) to thrombi after IR injury. *P < .05; ***P < .001.
Figure 4
Figure 4
Platelet granule release induces neutrophil polarization and motility. Neutrophils (2 × 106/mL) were incubated with platelet releasate or shed proteins generated by the indicated agonist. Neutrophils were then assessed for shape change and Mac-1 activation, as described in “Materials and methods.” (A) Representative DIC images demonstrating neutrophil shape change following incubation with platelet releasate (29 µg/mL) for 20 minutes at 37°C (red arrow, red blood cells; yellow arrow, neutrophils). (B-C) Relative potency of platelet releasate derived from ionophore A23187-stimulated platelets: (B) line graph demonstrating the dose-dependent effects of the releasate on neutrophil shape change and Mac-1 activation, and (C) representative fluorescence-activated cell sorter (FACS) profiles of Mac-1 activation for each indicated dose of releasate (protein per milliliter of neutrophils) and formyl methionyl leucyl phenylalanine (fMLP) (2 µM). (D) Neutrophils (2 × 106/mL) were incubated in the absence (Rest) or presence of platelet releasates generated by activating platelets with either ionophore (Iono, 2 μM), thrombin + collagen (Thr-collagen, 1 U/mL thrombin and 10 μg/mL collagen), thrombin alone (Thr, 1 U/mL), collagen alone (10 μg/mL), or adenosine diphosphate (ADP; 10 μM) for 20 minutes at 37°C as described in supplemental Methods. Platelet releasate-induced neutrophil shape change was quantified as described in “Materials and methods” (mean ± SEM; n = 3). (E) The neutrophil shape change activity in platelet releasate and surface-shed proteins was quantified as described in “Materials and methods” (mean ± SEM; n = 3). (F) The effect of shed proteins and platelet releasate on Mac-1 activation was also examined, as describe in supplemental Methods and compared with that achieved in response to fMLP (2 µM) (mean ± SEM; n = 3). FITC, fluorescein isothiocyanate; ns, not significant, P > .05; **P < .01; ***P < .001.
Figure 5
Figure 5
Purification, identification, and characterization of NAP-2 as the major platelet-derived chemokine inducing neutrophil shape change and polarization. (A) Platelet releasate (3.42 mg, 4,000 units activity) was applied to a Heparin HP Column, bound proteins (gray line)and eluted with an NaCl gradient (0-2 M; green line); active fractions (shaded area) were identified by using the neutrophil shape change assay. Eighty-three percent of the original activity was recovered in two fractions co-eluted with NAP-2 as demonstrated by immunoblot analysis using anti–NAP-2/CTAP-III and anti-PF4 Abs. (B) Platelet releasate was subjected to 2 rounds of NAP-2 immunodepletion by using the anti–NAP-2/CTAP-III Ab, as detailed in supplemental Methods. The activity in the starting material (SM) and NAP-2–depleted releasate (first and second IP) was assessed as described above (mean ± SEM; n = 3; P < .01). Samples were also subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis and western blotting analysis for NAP-2 (inset). (C) Neutrophils (2 × 106/mL) were first treated for 10 minutes with buffer (Control), serine protease inhibitor leupeptin (10 μg/mL), His-EB22.4 (10 μg/mL), or phenylmethanesulfonylfluoride (PMSF; 1 mM), and then incubated with platelet releasate (15 μg/mL) for 20 minutes at 37°C. The percentage of cells undergoing shape change was analyzed by DIC microscopy (percentage of total cells per field; mean ± SEM; n = 3). (D) Dose-dependent effects of recombinant NAP-2 (NAP-2Rec) on neutrophil shape change. (E-F) Regulation of Mac-1 activation through platelet–leukocyte cross-talk. (E) Relative potency of intact activated platelets (activated platelet + releaseate) versus MP-free releasate or MP-rich platelet releasate at inducing increased Mac-1 expression on the surface of isolated neutrophils. Intact activated platelets, MP-rich releasates, and MP-free releasates were prepared as described in supplemental Methods, equalized for volume, and incubated with washed neutrophils (2 × 106/mL) at the indicated platelet:neutrophil ratios for 30 minutes at 37°C. Mac-1 expression was assessed by FACS using an anti-human CD11b (ICRF44) Ab (mean ± SEM; n = 3). Representative FACS histogram profiles of Mac-1 expression at a platelet:neutrophil ratio of 30:1 are depicted on the right. (F) Neutrophils (2 × 106/mL) were treated with dimethylsulfoxide (Control), CXCR1/2 antagonist MSGA 8-73 (5 μM; CXCR1/2 Antag), or PAF antagonist CV-3988 (10 μM; PAF Antag) for 10 minutes at 37°C. The effect of PAF and CXCR1/2 antagonists on activated platelets induced Mac-1 activation and was assessed by using the CBRM1/5-FITC Ab (platelet:neutrophil ratio of 10:1) (mean ± SEM; n = 3). ns, not significant, P > .05; **P < .01; ***P < .001.
Figure 6
Figure 6
NAP-2 regulation of leukocyte recruitment and migration at sites of vascular injury. (A) C57Bl/6 (Control) mice were administered pertussis toxin (PTX; 4 µg per mouse, 2 hours prior to experiment), PAF antagonist (WEB2086 or CV-3988, 10 mg/kg), or the CXCR1/2 antagonist MSGA 8-73 (10 mg/kg, CXCR1/2 Antag) prior to needle injury of mesenteric veins. The number of firmly adherent leukocytes was quantified by using real-time DIC imaging 20 minutes postinjury (mean ± SEM; n = 3 to 4). (B) Recombinant NAP-2 (50 μg/mL) or saline (Control) were injected 20 to 30 μm upstream from the thrombus 10 minutes postinjury, and the number of firmly adherent leukocytes was quantified (mean ± SEM; n = 3). Representative DIC images following saline or NAP-2 injection are depicted (red, firmly adherent cells; yellow, rolling cells). (C) An anti–NAP-2 Ab (Alexa-546—labeled) was injected at the site of thrombus formation 15 minutes and 30 minutes postinjury. Images depict representative overlay images of NAP-2 staining (red) localized near the base of thrombi (blue outline) adjacent to the site of vascular injury. (D) The effects of PAF antagonist CV-3988 (PAF Antag), CXCR1/2 antagonist MSGA 8-73 (CXCR1/2 Antag), and inactive control (monocyte chemotactic protein1 9-76) on leukocyte directional migration at the thrombus base were examined over a 30-minute time frame by fluorescence microscopy (mean ± SEM; n = 3 to 4). The surface area of thrombi was also quantified. ns, not significant, P > .05; **P < .01.
Figure 7
Figure 7
Leukocyte migration to the site of endothelial injury is greatly attenuated in NAP-2−/− mice. (A-C) C57Bl/6 wild-type (WT) and NAP-2−/− mice were subjected to mesenteric venous needle injury with local microinjection of thrombin. Thrombus formation and leukocyte recruitment at the thrombus base were monitored for 20 minutes by using fluorescence and DIC microscopy. (A) The number of leukocytes migrating toward the site of vascular injury was quantitated in C57Bl/6 (WT) and NAP-2−/− mice (mean ± SEM; WT group, n = 15 injuries; NAP-2−/− group, n = 15 injuries). The surface area of thrombi was also quantified. (B) Representative fluorescence images depicting leukocyte–thrombus interactions (leukocytes: Gr-1 Ab, green; platelets: DiOC6, red) in C57Bl/6 (WT) and NAP-2−/− mice; the insets demonstrate the marked reduction in shape-changed and migrating leukocytes in NAP-2−/− mice. (C) The number of leukocytes interacting with the thrombus margin was quantitated 20 minutes postinjury in C57Bl/6 (WT) and NAP-2−/− mice. (D) Platelet lysates from C57Bl/6 (WT) and NAP-2−/− mice were subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis and western blotting analysis for NAP-2, CXCL1, and CXCL5, and representative blots from 1 of 3 independent experiments are shown. ns, not significant; ***P < .001.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Eltzschig HK, Eckle T. Ischemia and reperfusion—from mechanism to translation. Nat Med. 2011;17(11):1391–1401. - PMC - PubMed
    1. Moens AL, Claeys MJ, Timmermans JP, Vrints CJ. Myocardial ischemia/reperfusion-injury, a clinical view on a complex pathophysiological process. Int J Cardiol. 2005;100(2):179–190. - PubMed
    1. Gawaz M. Role of platelets in coronary thrombosis and reperfusion of ischemic myocardium. Cardiovasc Res. 2004;61(3):498–511. - PubMed
    1. Mao Y, Zhang M, Tuma RF, Kunapuli SP. Deficiency of PAR4 attenuates cerebral ischemia/reperfusion injury in mice. J Cereb Blood Flow Metab. 2010;30(5):1044–1052. - PMC - PubMed
    1. Kuroda T, Shiohara E, Homma T, Furukawa Y, Chiba S. Effects of leukocyte and platelet depletion on ischemia—reperfusion injury to dog pancreas. Gastroenterology. 1994;107(4):1125–1134. - PubMed

Publication types

MeSH terms

-