Chemokine- and adhesion-dependent survival of neutrophils after transmigration through cytokine-stimulated endothelium

Abstract

We examined the fate of neutrophils following transmigration through an endothelial monolayer cultured on "Transwell" membrane filters. Treatment of human umbilical vein endothelial cells (HUVEC) with increasing doses of tumor necrosis factor-alpha increased the efficiency of transmigration and markedly reduced apoptosis among the transmigrated neutrophils in a dose-dependent manner. Apoptosis was also inhibited after transmigration of neutrophils through HUVEC stimulated with interleukin (IL)-1beta but not so effectively after chemotaxis through unstimulated HUVEC driven by IL-8 added below the filter. Inhibition of beta2-integrin binding after transmigration or coating the lower chamber with a nonadhesive polymer (polyhydroxyl-ethyl-methacrylate) abrogated neutrophil survival. Although integrin engagement during migration itself was not essential to inhibit apoptosis, activation of neutrophils through CXC chemokine receptors was necessary. Quite brief exposure to the HUVEC (30-120 min) was effective in reducing subsequent apoptosis, although if coincubation with the HUVEC were prolonged, neutrophil apoptosis was reduced further. Neutralization of granulocyte macrophage-colony stimulating factor inhibited this additional effect. Thus, a complex interplay between migration- and activation-dependent signals and adhesive interaction in tissue may combine to effectively prolong the survival of neutrophils recruited during inflammation.

Fig. 1

Percentage of neutrophils undergoing transmigration and subsequent apoptosis after incubation with HUVEC on Transwell filters. HUVEC were treated with different concentrations of TNF for 4 h and washed free of TNF, and then neutrophils were added for 2 h before removal of the filter. (A) Percentage of added neutrophils migrating into the lower chamber of the Transwell. (B) Percentage of transmigrated neutrophils undergoing apoptosis after 24 h, judged by nuclear morphology. Control = Freshly isolated neutrophils incubated for 24 h in matching plates. (C) Percentage of transmigrated neutrophils showing decreased mitochondrial transmembrane potential (i.e., low level of fluorescence of DiOC₆). (D) Photomicrographs of stained neutrophils used to assess apoptosis from nuclear morphology: Fresh = Freshly isolated cells; 100 U/ml = neutrophils 24 h after transmigration through HUVEC treated with 100 U/ml TNF; 0 U/ml = neutrophils 24 h after transmigration through untreated HUVEC. A = Examples of apoptotic cells; R = red blood cell found in the fresh isolate. Data are the mean ± sem from seven to 10 independent experiments. ANOVA showed a significant effect of TNF concentration on migration and on apoptosis assessed by either method (P<0.01 in each case).

Fig. 2

Time course of apoptosis of neutrophils, which have transmigrated through HUVEC. Comparisons were made between freshly isolated neutrophils (▲) and neutrophils that had migrated for 2 h through unstimulated HUVEC (□) or through HUVEC that had been treated with 1 U/ml TNF (△) or 100 U/ml TNF (■). Data are the mean ± sem from three of more independent experiments, except at 72 h, where n = 2. ANOVA showed significant effects of time and treatment on apoptosis (P<0.01 in each case).

Fig. 3

Effects of function-blocking mAb or agents against β1-, β2-, or β3-integrins or mAb against VCAM-1 on (A) transmigration of neutrophils or (B) apoptosis of transmigrated cells. HUVEC were treated with 100 U/ml TNF for 4 h and washed free of TNF, and then neutrophils were added for 2 h before removal of the filter. Antibodies against integrins (mAb13=anti-β₁; R6.5E=anti-β₂; SZ21=anti-β₃) or CT7010 were added to the neutrophils, or mAb against VCAM-1 (1G11) were added to HUVEC for 30 min prior to addition of neutrophils to the HUVEC. Data are the mean ± sem from four experiments. ANOVA showed a significant effect of treatment on transmigration (P<0.05). *, P < 0.05, compared with untreated control (None) by Dunnett test.

Fig. 4

Effects of adding function-blocking antibodies or agents against β1- or β2-integrins to the lower chamber on apoptosis of transmigrated cells. HUVEC were treated with 100 U/ml TNF for 4 h and washed free of TNF, and then neutrophils were added for 2 h before removal of the filter. Antibodies or CT7010 were added to the lower chamber immediately before addition of neutrophils. Data are the mean ± sem from four independent experiments. ANOVA showed a significant effect of treatment on apoptosis (P<0.01). **, P < 0.01, compared with untreated control (None) by Dunnett test.

Fig. 5

Effects of function-blocking antibodies against CXCR1 and/or CXCR2 on (A) transmigration of neutrophils or (B) apoptosis of transmigrated cells. HUVEC were treated with 100 U/ml TNF for 4 h and washed free of TNF, and then neutrophils were added for 2 h before removal of the filter. Antibodies against CXCR1, CXCR2, or both were added to the neutrophils for 30 min prior to addition of neutrophils to the HUVEC. Data are the mean ± sem of three to seven experiments. ANOVA showed a significant effect of treatment on transmigration and on apoptosis (P<0.01 in each case). *, P < 0.05; **, P < 0.01, compared with untreated control (None) by Dunnett test.

Fig. 6

Percentage of neutrophils undergoing transmigration and subsequent apoptosis after incubation with HUVEC on Transwell filters for different times. HUVEC were unstimulated (open bars) or treated with 100 U/ml TNF (solid bars) for 4 h and washed free of TNF, and then neutrophils were added for 0.5, 2, or 24 h before removal of the filter. (A) Percentage of added neutrophils migrating into the lower chamber of the Transwell. (B) Percentage of transmigrated neutrophils undergoing apoptosis after 24 h, judged by nuclear morphology. Data are the mean ± sem from four to eight independent experiments. ANOVA showed a significant effect of TNF concentration and of time of coincubation on migration and on apoptosis (P<0.01 in each case).

Fig. 7

Effect of neutralizing mAb against GM-CSF on apoptosis of neutrophils, which had transmigrated through HUVEC for different periods. HUVEC were unstimulated (open bars) or treated with 100 U/ml TNF (solid bars) for 4 h and washed free of TNF, and then neutrophils were added and incubated for 2 or 24 h before removal of the filter. Apoptosis was evaluated after a total of 24 h incubation. Antibody was added to the lower chamber prior to the addition of the neutrophils. Data are the mean ± sem from four independent experiments. ANOVA showed a significant effect of TNF dose (P<0.01) but not antibody on apoptosis for 2-h incubations and a significant effect of TNF dose and antibody on apoptosis for 24-h incubations (P<0.01 in each case). *, P < 0.05, compared with control without antibody at the same time of incubation by Dunnett test.