FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Topham, M. K.
Right arrow Articles by Zimmerman, G. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Topham, M. K.
Right arrow Articles by Zimmerman, G. A.
(The FASEB Journal. 1998;12:733-746.)
© 1998 FASEB

RESEARCH COMMUNICATION

Human endothelial cells regulate polymorphonuclear leukocyte degranulation

Matthew K. Tophama,b,c, Holly J. Carvethc, Thomas M. Mcintyrea,c,d, Stephen M. Prescotta,c,e, and Guy A. Zimmermana,c,1

a Nora Eccles Harrison Cardiovascular Research and Training Institute, University of Utah School of Medicine, Salt Lake City, Utah, 84112–5000, USA
b The Eccles Program in Human Molecular Biology and Genetics, University of Utah School of Medicine, Salt Lake City, Utah, 84112–5000, USA
c Department of Medicine, University of Utah School of Medicine, Salt Lake City, Utah, 84112–5000, USA
d Department of Pathology, University of Utah School of Medicine, Salt Lake City, Utah, 84112–5000, USA
e Department of Biochemistry, University of Utah School of Medicine, Salt Lake City, Utah, 84112–5000, USA


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 EXPERIMENTAL PROTOCOLS
 RESULTS
 DISCUSSION
 REFERENCES
 
Neutrophil degranulation is an important event in inflammatory responses. We examined the regulation of neutrophil (PMN) degranulation by resting and activated human endothelial cells. Whereas PMNs adherent to endothelial cells that were stimulated to express P-selectin and platelet-activating factor did not release the specific granule marker lactoferrin or the primary granule enzyme, elastase, PMNs adherent to endothelial cells stimulated with interleukin-1 (IL-1) or tumor necrosis factor secreted both. PMN degranulation was dependent on the time of incubation of endothelial cells with the cytokine, its concentration, and the time of incubation of the PMNs with endothelial cells. Degranulation of PMNs and their adhesion to stimulated endothelial cells are correlated events, but they could be dissociated by blocking the tethering molecules used by the endothelial cells and neutrophils under these conditions. This suggested that paracrine signaling molecules that induce PMN degranulation are produced by cytokine-stimulated endothelial cells. We found that endothelial cells stimulated with IL-1 release newly synthesized degranulating factors that require transcription and translation. IL-8 was synthesized, released, and signaled granular secretion by PMNs. However, experiments with blocking antibodies indicated the presence of an additional degranulating factor not accounted for by IL-8. These experiments demonstrate that human endothelial cells regulate degranulation of neutrophils by generating signaling factors that are expressed differentially depending on the endothelial agonist and other features. Active modification of neutrophil granular secretion by endothelial cells can influence physiologic acute inflammatory responses but may also contribute to pathologic vascular and tissue damage.—Topham, M. K., Carveth, H. J., McIntyre, T. M., Prescott, S. M., Zimmerman, G. A. Human endothelial cells regulate polymorphonuclear leukocyte degranulation. FASEB J. 12, 733–746 (1998)


Key Words: monoclonal antibodies • neutrophils • PMN adhesion • cytokine • tumor necrosis factor


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 EXPERIMENTAL PROTOCOLS
 RESULTS
 DISCUSSION
 REFERENCES
 
THE INTERACTION OF endothelial cells and polymorphonuclear leukocytes (PMNs; neutrophils)2 is critical in physiologic acute inflammatory responses to bacterial invasion and tissue injury, and in many syndromes of inflammatory tissue damage. Human endothelial cells express tethering (adhesion) molecules and also synthesize signaling molecules that induce functional changes in PMNs. The pattern of tethering and signaling factors varies in a time- and agonist-dependent fashion (1). Within minutes of stimulation with thrombin, histamine, or leukotriene C4, human endothelial cells coexpress P-selectin and platelet-activating factor (PAF), leading to adhesion and juxtacrine activation of PMNs (1–4). Within hours of stimulation with interleukin-1 (IL-1), tumor necrosis factor {alpha} (TNF-{alpha}), or bacterial lipopolysaccharide (LPS), endothelial monolayers express E-selectin and IL-8, which mediate both adhesion of PMNs and their activation (5–14). Local activation of tethered PMNs by signaling molecules generated by stimulated endothelial cells confers significant biologic advantages for the spatial regulation of physiologic inflammatory responses and has been demonstrated in vivo (1, 15). This mechanism of information transfer is also used in endothelial interactions with other classes of leukocytes (16).

Degranulation by PMNs is a critical activation response (17). Human PMNs contain primary `azurophilic') and secondary (`specific') granules as well as gelatinase-containing granules and a population of secretory vesicles (17–22). Mobilization of granular constituents is dependent on the degree of activation of the PMN, with secretory vesicles most readily trans~located, followed by gelatinase-containing, secondary, and primary granules (17, 18, 21).

Granular translocation and secretion are important in host defense. Neutrophil subcellular granules are reservoirs for chemotactic factor receptors (18, 20, 23), for copies of {alpha}m/ß2 integrin as well as other integrins and adhesive molecules that recognize matrix proteins (20, 22, 24, 25), and for lactoferrin, which has pleiotropic effects in inflammation (26–29). PMN degranulation, plasma membrane modification, and polarized cellular migration are closely related events (17, 18, 30–33) that are further linked to adhesive interactions of leukocytes (27, 28, 34, 35). Recruitment of {alpha}m/ß2 integrin and matrix adhesion receptors to the plasma membrane as a consequence of degranulation influences binding of neutrophils to specific vascular structures, and secretion of granular proteases facilitates movement though the subcellular matrix (17, 24, 25, 34–39). Thus, PMN degranulation is critical for most initial steps in the physiologic acute inflammatory response (32). However, neutrophil granular proteases and hydrolases also induce vascular and tissue injury when released in an unregulated fashion (40–43).

Granular secretion by PMNs is influenced profoundly by interaction of the cells with surfaces (17). Yet little is known about the influence of endothelial cells on degranulation and comprehensive reviews contain no information on this point (17, 18, 21), even though the endothelium is the first surface that PMNs encounter in acute inflammation. Here we report experiments demonstrating that human endothelial cells regulate the neutrophil degranulation response.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
EXPERIMENTAL PROTOCOLS
 RESULTS
 DISCUSSION
 REFERENCES
 
Reagents
We used gelatin (type II) from Fisher (Pittsburgh, Pa.), Hank's balanced salt solution (HBSS) from Whittaker Bioproducts (Walkersville, Md.), and human albumin (25% USP) from Miles, Inc. (Elkhart, Ind.). Sialidase (neuraminidase, from Salmonella), phorbol myristate acetate (PMA), n-formyl-methionyl-leucyl-phenylalanine (fMLP), rabbit anti-human lactoferrin, peroxidase-conjugate avidin, bovine serum albumin (BSA), and o-phenylenediamine were purchased from Sigma Chemical Co. (St. Louis, Mo.). PAF was purchased from Avanti Polar Lipids, Inc. (Birmingham, Ala.). Recombinant human IL-1{alpha} was kindly provided by Peter Lomedico (Dept. of Molecular Genetics, Hoffmann-La Roche, Nutley, N.J.). Additional preparations of recombinant human IL-1{alpha} and polyclonal rabbit anti-human IL-1{alpha} were obtained from Genzyme (Boston, Mass.). 111Indine oxine was purchased from the Radiopharmacy Service at the University of Utah. We used 12-well tissue culture plates and EIA microtiter plates from Costar (Cambridge, Mass.). Anti-human lactoferrin immununoglobulin G (IgG) was from Sigma and peroxidase-conjugated anti-human lactoferrin was from Cappel (Cochranville, Pa.). Sheep anti-human elastase and peroxidase-conjugated sheep anti-human elastase were purchased from The Binding Site Ltd. (Birmingham, England). Human neutrophil elastase and human lactoferrin standards were from CalBiochem (San Diego, Calif.). Hydrogen peroxide was purchased from J. T. Baker, Inc. (Phillipsburg, N.J.). Leukotrienes B4 and C4 were gifts from Dr. Joshua Rokach (Merck-Frosst, Montreal, Canada) and WEB 2086 was provided by Peggy Ganong of Boehringer-Ingelheim (Ridgefield, Conn.). Recombinant human IL-8 (72 and 77 amino acid forms) was a gift from Jeoff Baker, Genentech (South San Francisco, Calif.) and polyclonal goat anti-IL-8 (unconjugated and biotinylated), murine monoclonal anti-IL-8, and recombinant human IL-8 (72 amino acid form) were gifts from Ivan Lindley and Miroslav Ceska of The Sandoz Research Institute (Vienna, Austria). Additional preparations of IL-8 were purchased from Genzyme. Recombinant human TNF-{alpha} was provided by Genentech.

Monoclonal antibodies
Dreg-56 (IgG1, anti-L-selectin) (44) was a gift from T. K. Kishimoto of Boehringer Ingelheim; LAM 1.4 (IgG1, anti-L-selectin) (45) was provided by Tom Tedder (Boston, Mass.). H18/7 (IgG2, anti-E-selectin) (5) in ascites was provided by Michael Bevilacqua (San Diego, Calif.). Antibodies 60.1 (IgG1, anti-{alpha}m) and 60.3 (IgG2a, anti-ß2) were gifts from Patrick Beatty (University of Utah) and John Haran (University of Washington). IB4 (IgG2a, anti-ß2) was provided by Samuel Wright (Rockefeller University, New York, N.Y.). Primary references for these monoclonal antibodies (mAb's), which identify ß2 (CD11/CD18) integrins, and the conditions for their use in our experiments can be found in refs 36 and 46. We used the following isotype-matched control antibodies: W6/32 (IgG2a, anti-HLA class I; purified or in ascites, as appropriate), a gift from John Bohnsack (University of Utah, Salt Lake City, Utah); T10 (IgG1, anti-{alpha}IIb/ß3), provided by Rodger McEver (Oklahoma City, Okla.).

Cells
Endothelial cells were cultured in gelatinized tissue culture wells as described (47). Only primary monolayers that were tightly confluent were used.

Human blood was obtained from volunteers after informed consent. PMNs were isolated and, in the experiments indicated, were labeled with 111In, as described (48, 49). Unlabeled and radiolabeled PMNs were suspended at a final concentration of 5.5 x 106 cells/ml in HBSS containing 5 mg/ml human serum albumin (HBSS/A).

PMNs from a subject deficient in expression of ß2 integrins were collected after informed consent. Characterization of the phenotype of PMNs from this patient has been described previously (3, 36). Tom St. John (ICOS, Inc., Seattle, Wash.) generously provided mouse L cells transfected with a cDNA for E-selectin; surface expression of E-selectin was documented by specific binding of the anti-E-selectin antibody, BBA II (British Batik). Wild-type L cells were used as controls.

Assays of PMN degranulation
We developed sandwich-type enzyme-linked immunosorbent assays (ELISAs) to measure PMN secretion of primary and secondary granular constituents (see below for details). Lactoferrin was assayed as a marker for secondary granules and elastase was assayed to monitor primary granule secretion (17, 18). Because translocation of secondary granules is a more sensitive measure of degranulation than is release of primary granule contents (see opening paragraphs and refs 17–22), we routinely measured lactoferrin secretion and also mea~sured elastase release in parallel in particular experiments. In some experiments, PMNs were incubated in the presence of endothelial cells, and the samples for measurement of markers of degranulaton were collected as described in the legend to Fig. 1. In a second protocol, samples for assay of paracrine-degranulating factors released by stimulated endothelial cells (see below for details) were collected as described in the legend to Fig. 6, added to PMNs in the absence of endothelial cells, and lactoferrin and/or elastase secretion was measured. To determine total granular content of lactoferrin and/or elastase, an aliquot of PMNs from the original suspension was lysed with 1 M NH4OH, buffered, diluted, and assayed with the samples in each experiment.



View larger version (21K):
[in this window]
[in a new window]
  		Figure 1. PMNs incubated with IL-1- but not thrombin-treated endothelial cells secrete lactoferrin and elastase. Primary HUVEC monolayers were incubated for 4 h in culture medium alone (`control') or with IL-1 (50 U/ml); in a second group of experiments, endothelial cells were incubated for 10 min with control buffer or thrombin (2 U/ml). The monolayers were then washed with HBSS (37°C). PMNs (0.5 ml, 5.5x106/ml) were added to the endothelial cells and incubated at 37°C for 2 h with control and IL-1-treated endothelial cells or for 20 min with thrombin-treated and control endothelial cells. The solution overlying the monolayers was then collected, centrifuged to remove PMNs that had not remained adherent, and these samples were assayed by ELISA for lactoferrin (upper panel) and elastase (lower panel) (Methods). Each point indicates the mean of duplicate determinations in one experiment. The difference between the values for degranulation by PMNs incubated with IL-1-stimulated endothelial cells compared to control endothelial cells was significant (P<0.01).



View larger version (22K):
[in this window]
[in a new window]
  		Figure 6. PMNs suspended in supernatants from IL-1-stimulated endothelial cells secrete lactoferrin and elastase. Endothelial cell monolayers were incubated in culture medium alone or with 50 U/ml IL-1 for 4 h at 37°C and then washed. HBSS/A (0.55 ml, 37°C) was added, incubated for 2 h (37°C, 5% CO2), and collected. Pelleted PMNs (0.5 ml, 5.5x106/ml) were resuspended in these supernatants and incubated on gelatin matrices for the times shown (37°C, 5% CO2). The secretion of lactoferrin (upper panel) and elastase (lower panel) was assayed by ELISA as described in Methods. The values represent the mean and range of duplicate points in one experiment. Time-dependent triggering of degranulation by supernatants from IL-1-stimulated endothelial cells was reproduced in three additional experiments. Control EC (open bars); IL-1-treated EC (filled bars).


   EXPERIMENTAL PROTOCOLS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 EXPERIMENTAL PROTOCOLS
RESULTS
 DISCUSSION
 REFERENCES
 
`Rapid' activation of endothelial cells to induce expression of tethering and signaling molecules
Endothelial cells were incubated at 37°C in 5% CO2 in control buffer (HBSS/A) alone, in buffer containing thrombin (2 U/ml) for 10 min, or in buffer containing LTC4 (1 µM) for 30 min; these conditions were previously shown to induce maximal neutrophil adhesion (48, 49). The incubation medium was removed and the endothelial cells were washed once with 0.5 ml HBSS (37°C). Neutrophils (0.5 ml, 5.5x106/ml), isolated as detailed above, were layered on the endothelial cells and incubated for various times, as indicated, at 37°C, 5% CO2. PMA (10 ng/ml) or fMLP (0.1 or 1 µM) was added directly to PMNs overlying the control unactivated endothelial cells and incubated in parallel as a positive control, demonstrating that the PMNs could secrete granular contents under these conditions. After the incubation times indicated, the buffer and nonadherent PMNs were removed and centrifuged to pellet the cells and separate them from the buffer (450xg for 5 min at 4°C). These cell-free buffer samples were stored at 4°C until assayed by ELISA. The endothelial cell monolayers were washed immediately with 0.5 ml HBSS, and PMN adhesion was assayed by counting adherent PMNs using phase contrast microscopy or, in some experiments, with 111In-labeled PMNs as described below.

Activation of endothelial cells with inflammatory cytokines to induce expression of tethering and signaling molecules
Endothelial cell monolayers were incubated at 37°C in 5% CO2 for 4 h in complete culture medium without agonist or in culture medium containing IL-1{alpha} or TNF-{alpha} (2). In some experiments, a polyclonal antibody to human IL-1{alpha} (1:200) or a control antibody was added before IL-1. The monolayers were washed with HBSS (37°C), then 0.5 ml of PMN suspension was layered on the endothelial cells and incubated for various times at 37°C in 5% CO2. At the end of the incubation period, the buffer and nonadherent PMNs were aspirated, the PMNs were removed from this aspirated supernatant by centrifugation, and the supernatants were saved for assay of PMN granule constituents by ELISA as in experiments with `rapidly activated' endothelial cells (see description above). Controls for PMN degranulation and assays of PMN adhesion were also done in parallel (see above).

Release of paracrine factors that signal PMN degranulation by activated endothelial cells
To assay for paracrine factors that are released into solution, we measured degranulation of PMNs when they were sus~pended in supernatants from control or cytokine-stimulated endothelial cells. Endothelial cells were incubated first in control culture medium without agonist or with IL-1{alpha} or TNF-{alpha} for 4 h at 37°C. The medium was removed and the monolayers were washed with 0.5 ml HBSS (37°C). HBSS/A (0.55 ml) was then added for various times as indicated; in most experiments this was a period of 2 h. These conditioned `supernatants' were collected, centrifuged, and stored at 4°C until assayed.

To assay the supernatants for degranulating factors, 0.5 ml of 0.2% gelatin in HBSS was first added to 16 mm tissue culture plates and incubated at 37°C for 1 h. The solution was then aspirated and the plates were washed with 0.5 ml HBSS (37°C). PMNs (0.5 ml, 5.5x106/ml) were pelleted (450xg, 5 min, room temperature), resuspended in 0.5 ml of endothelial cell supernatant (thawed immediately before assay) or control buffer, and incubated on the gelatin matrices for various periods of time (37°C, 5% CO2). Each assay also included wells in which PMNs were treated with an agonist known to induce granular secretion (fMLP or PMA) as a positive control. In some experiments, an antibody to human IL-1 (1:200), IL-8 (see text for a description of conditions), or a control antibody was incubated with the endothelial cell supernatants before addition of PMNs. After the 1 h incubation, the supernatants were collected from the gelatin matrices, centrifuged (450xg, 5 min, room temperature) to remove any PMNs that had not tightly adhered to the gelatin, and immediately assayed for the concentration of granular markers by ELISA (see below) or rapidly frozen for later assay.

This assay utilizes adhesion of PMNs to the gelatin matrix mediated by {alpha}m/ß2 integrin (36, 46) to enhance translocation of primary and secondary granules (50), obviating the need for treatment with cytochalasin B (17, 18).

Assay of PMN adhesion to endothelial cells
The binding assays for 111In-labeled PMNs have been reported previously (2, 48). Briefly, control or activated endothelial cell monolayers were washed and 111In-labeled PMNs were added and incubated for various times. In experiments with thrombin- or LTC4-stimulated endothelial cells, this incubation period was 5 min because maximal binding occurs during this interval (2, 48, 49). In experiments with IL-1{alpha}- or TNF-{alpha}-activated endothelial cells, the incubation period was 5 or 10 min (2). Nonadherent PMNs were removed and the monolayers were washed twice with HBSS. The adherent radiolabeled PMNs and endothelial cells were solubilized and the fraction of adherent PMN (percent of total radiolabeled PMN added) was quantitated as described (48). Side-by-side comparisons of PMN adhesion assessed by phase contrast microscopy and by radiolabeled PMN counting yield equivalent data (ref 48 and unpublished observations).

ELISAs for lactoferrin, elastase, and IL-8
Antibody to either human lactoferrin or human elastase was dissolved in coating buffer (pH 9.6, 10 µg antibody/ml) and immobilized on 96-well EIA microtiter plates by incubating at 4°C for 12 h or at 25°C for 1 h. The plates were then washed six times with 0.05% Tween-20 in phosphate-buffered saline (PBS). Immediately before use, samples from stimulated PMNs (see above) were diluted with 1% BSA in PBS to appropriate concentrations for the assay. The samples were added to the plates and incubated 90 min at 37°C. The plates were washed six times with 1% BSA, and solutions of horseradish peroxidase-labeled antibody raised against human elastase or lactoferrin were added and incubated for 1 h at 37°C. The rabbit anti-human lactoferrin antibody was diluted 1:400 in PBS containing 1% BSA and the ovine anti-human elastase was diluted 1:1500 in the same buffer. Both antibody solutions were prepared fresh on the day of use. The plates were again washed six times, and peroxidase substrate (o-phenylenediamine, H2O2) was added and incubated for 15–20 min to develop color. The reaction was stopped with 8N H2SO4 and the optical density of the mixture in each well at 492 nm was determined with a microplate reader (Molecular Devices, Menlo Park, Calif.). Serial dilutions of human neutrophil elastase or human lactoferrin were added to parallel wells on the ELISA plates in each assay as standards.

An ELISA was also used to measure IL-8 secreted by cytokine-activated endothelial cells. Polyclonal goat anti-IL-8 (51) was immobilized and recombinant IL-8 standards or supernatants to be assayed were incubated on the plates using a protocol similar to that described above for ELISAs for lactoferrin and elastase. Biotin-conjugated polyclonal goat anti-IL-8 in 1% BSA/PBS was then incubated in the wells for 1 h at 37°C; the plates were washed six times. Peroxidase-conjugated avidin was then incubated in the wells for 1 h at 37°C. The plates were again washed six times and developed as in the assays for lactoferrin and elastase.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 EXPERIMENTAL PROTOCOLS
 RESULTS
DISCUSSION
 REFERENCES
 
Endothelial cells stimulated with inflammatory cytokines express degranulating factors and adhesion molecules for PMNs
We examined the adhesive and secretory responses of PMNs when they were incubated with endothelial cells stimulated with inflammatory cytokines. Pretreatment of endothelial cells with IL-1 resulted in enhanced PMN binding in a time- and concentration-dependent fashion similar to that previously reported by others (5; not shown). Incubation of PMNs with IL-1 did not directly alter their adhesiveness. In contrast to an earlier observation (35), we found that PMNs degranulate when incubated with IL-1-stimulated endothelial cells ( Fig. 1). Lactoferrin, a marker of specific granules, and elastase, a primary granule constituent, were both released.

The degranulation response was time dependent, varying with the period of preincubation of endothelial cells with IL-1 and with the time of incubation of PMNs with the monolayers after a fixed pretreatment of endothelial cells with IL-1 ( Fig. 2). Release of granular markers from PMNs incubated with IL-1-stimulated endothelial cells was uniformly greater than that from PMNs incubated with control endothelial cells at each time point in three experiments. This was consistent whether the levels of released granular marker were expressed as the absolute concentration in the sample or as a percent of the total granular content ( Fig. 2and legend, and data not shown), although the magnitude of granular secretion by neutrophils incubated with IL-1-stimulated endothelial monolayers varied from experiment to experiment ( Figs. 1 and 2and data not shown). The magnitude of degranulation induced by IL-1-stimulated endothelial cells was similar to, or greater than, that triggered by addition of exogenous fMLP (not shown) or IL-8 to neutrophils incubated in the absence of endothelial monolayers (see below). In contrast, degranulation by PMNs incubated with IL-1-stimulated endothelial cells was less than that triggered by addition of the potent agonist PMA to PMNs incubated with control endothelial cells (at 0.5 h of incubation, there was 0.36% lactoferrin release by PMNs incubated with control endothelial cells, 0.93% release by PMNs incubated with IL-1-treated endothelial cells, and 12.3% in response to PMA; at 1 h of incubation, 0.42% lactoferrin release by PMNs on control endothelial cells, 1.58% release by PMNs on IL-1-treated endothelial monolayers, and 30.4% release in response to PMA. A second experiment gave a similar pattern).



View larger version (19K):
[in this window]
[in a new window]
  		Figure 2. Degranulation of PMNs adherent to EC stimulated with IL-1 is time dependent. Upper panel: Monolayers of human umbilical vein endothelial cells were incubated with IL-1 (50 U/ml) or control buffer as described in Methods and Fig. 1. The horizontal axis indicates the time of incubation with IL-1 before PMNs were added. Neutrophils were then layered over the endothelial cells, incubated for 2 h, and lactoferrin secretion was assayed as described in Fig. 1. Open circles indicate the lactoferrin concentration in supernatants from PMNs adherent to control endothelial cells; this value (1.4% of total PMN lactoferrin) did not vary at any time point in the experiment. Filled circles indicate values from incubations of PMNs with endothelial cells stimulated with IL-1. When expressed as a percent of total PMN lactoferrin, these values were as follows: 1 h, 3.2%; 2 h, 3.2%; 4 h, 4.2%. Lower panel: endothelial cell monolayers were incubated in culture medium alone or with IL-1 (50 U/ml) for 4 h. Neutrophils were then layered over the endothelial cells, incubated, and the release of lactoferrin was assayed as described in Fig. 1. The horizontal axis indicates the time of incubation after PMNs were added. When expressed as a percent of total PMN lactoferrin, these values were as follows: PMNs incubated with control endothelial cells (open circles), 0.4% at 0.5 h, 0.5% at 1 h, and 0.9% at 2 h; PMNs incubated with IL-1-stimulated endothelial cells (filled circles), 1.2% at 0.5 h, 2.0% at 1 h, 3.4% at 2 h. The values shown are the means and ranges of duplicate points in one experiment. Two additional experiments demonstrated similar time courses although the magnitude of lactoferrin secretion at the intervals varied from experiment to experiment.

IL-1 induces additional IL-1 synthesis by endothelial cells (reviewed in refs 52, 53), and IL-1 has been reported to directly induce PMN degranulation (54). However, experiments with a neutralizing monoclonal antibody against IL-1 indicated that neither en~dogenous synthesis of IL-1 by the endothelial cells nor carryover of recombinant IL-1 mediated the degranulation response. In these studies, a blocking monoclonal antibody against IL-1 did not inhibit the degranulating activity in conditioned supernatants from IL-1-treated endothelial cells but potently inhibited the ability of recombinant IL-1 to induce endothelial cells to generate degranulating activity, documenting its blocking efficacy (not shown). Also, recombinant IL-1 (10 or 50 U/ml) did not directly stimulate degranulation greater than that of PMNs in control buffer when they were incubated on gelatinized wells for 0.5–2 h (lactoferrin release by IL-1- or buffer-treated PMNs was 0.5% or less of total PMN lactoferrin at each time point, n=2). In contrast, recombinant IL-1 from the same preparation induced endothelial cells to generate degranulating activity, demonstrating that it was biologically active.

Treatment of endothelial cells with TNF-{alpha} (1–1000 U/ml), followed by washing, also induced PMN adhesion and granular secretion in a concentration-dependent fashion, as did treatment of endothelial monolayers with IL-1. In contrast, thrombin-stimulated endothelial cells, which express a different pattern of surface molecules compared to monolayers stimulated by IL-1 or TNF-{alpha} (see opening paragraphs and ref 15), did not trigger degranulation by PMNs ( Fig. 1). Similarly, endothelial cells stimulated with LTC4, which, like thrombin, induces rapid expression of P-selectin and PAF (2, 4, 49), did not cause granular secretion (not shown). Together, the results indicate that degranulation by PMNs at the surfaces of stimulated endothelial cells is regulated differentially depending on the endothelial agonist.

Secretion of primary and specific granular markers by PMNs can be dissociated from adhesion
Juxtacrine activation of neutrophils, which is one mechanism of adhesion-dependent signaling, occurs at the surfaces of stimulated endothelial cells (reviewed in ref 1). Furthermore, neutrophil integrins deliver outside-in signals that can modify degranulation responses (50). We therefore explored the roles of adhesion molecules in PMN degranulation at the surfaces of stimulated endothelial cells.

E-selectin
E-selectin, which tethers leukocytes to endothelial cells stimulated with IL-1, TNF-{alpha}, or LPS, is reported to mediate juxtacrine activation of PMNs (7, 8) and could signal PMN degranulation. However, a blocking monoclonal antibody against E-selectin consistently reduced PMN adhesion to IL-1-stimulated endothelial cells by a mean of 30%, as previously reported (5), but did not inhibit leukocyte degranulation ( Fig. 3). In a second strategy, we modified the ligands for E-selectin on the PMN surface by treating them with sialidase (neuraminidase) (55, 56). Although this treatment reduced adhesion of PMNs to IL-1-stimulated endothelial cells by a mean of 35%, it did not inhibit degranulation ( Fig. 3). Because inhibition of adhesion was incomplete in these experiments, we next examined the degranulation responses of PMNs adherent to murine L cells transfected with a cDNA for E-selectin. Although there was enhanced adhesion of PMNs to transfected cells compared to wild-type L cells, there was no secretion of granular contents (not shown). These experiments indicated that the degranulation response can be dissociated from adhesion mediated by E-selectin.



View larger version (23K):
[in this window]
[in a new window]
  		Figure 3. Granular secretion by PMNs incubated with IL-1-stimulated endothelial cells can be dissociated from adhesion mediated by E-selectin. Upper panel: Endothelial monolayers were incubated in culture medium alone (control, open bars) or with IL-1 (50 U/ml, filled bars) for 4 h. The anti-E-selectin mAb H18/7 (5) or an isotype-matched control antibody, W6/32 (20 µg antibody/ml in HBSS/A), was then added to the endothelial cells for 10 min (37°C). Neutrophils were added to achieve a final volume of 0.5 ml and final concentrations of 5.5 x 106 PMN/ml and 10 µg antibody/ml. The PMNs were incubated with the endothelial cells for 2 h and lactoferrin secretion was assayed as described in Fig. 1. Adherence was monitored in parallel by phase contrast microscopy. The control mAb W6/32 did not significantly affect lactoferrin secretion (not shown) or adherence, whereas H18/7 reduced adherence to IL-1-treated endothelial cells by 30%. In three additional experiments, mAb H18/7 did not reduce secretion of lactoferrin by PMNs incubated with IL-1-stimulated endothelial cells to a level below that of cells incubated in the presence of the control antibody and/or in the absence of antibody. Lower panel: PMNs were suspended in buffer (0.15 M NaCl, 0.4 mM CaCl2, pH 5.5) with or without 1 U/ml sialidase and incubated for 1 h (37°C). After this, they were washed three times with HBSS/A and resuspended in HBSS/A to a final concentration of 5.5 x 106/ml. They were then incubated with control or IL-1-treated endothelial cells and assayed for lactoferrin release, as described in Fig. 1. The values indicate the mean and range of duplicate points. Adhesion was monitored by phase contrast microscopy; sialidase treatment reduced PMN adherence to IL-1-treated endothelial cells by 30% compared to control neutrophils. In five additional experiments, treatment of PMNs with sialidase decreased PMN adhesion to IL-1-stimulated endothelial cells (mean 35% decrease) but had no consistent effect on secretion of lactoferrin. Control (open bars); IL-1 (filled bars).

PMN adhesion molecules
ß2 Integrins on the neutrophil plasma membrane mediate a component of the adhesive interaction with cytokine-stimulated endothelial cells (7, 8, 10, 34, 57–59), a feature we confirmed by using PMNs from a subject with leukocyte adhesion deficiency type I and by pretreating PMNs with the blocking anti-ß2 antibody mAb, 60.3 (not shown). In contrast to its effect on adhesion (mean 25% inhibition in two experiments), mAb 60.3 did not inhibit secretion of lactoferrin or elastase from PMNs incubated with cytokine-treated endothelial monolayers ( Fig. 4 and data not shown). A second anti-ß2 antibody, mAb IB4, inhibited PMN adhesion to endothelial cells stimulated with IL-1 but did not inhibit degranulation (mean 30% decrease in adhesion vs. mean 0% inhibition of degranulation in four experiments). The conditions for the experiments with monoclonal antibodies 60.3 and IB4 were based on previous studies in which these antibodies completely inhibited ß2-in~tegrin-dependent adhesion of stimulated PMNs to human endothelial cells (36, 46). However, because we used complete monoclonal antibodies that may conceivably have had spurious effects by interacting with Fc receptors on the PMNs, we conducted additional experiments using a different strategy. In this alternative approach, incubations of the leukocytes with cytokine-stimulated endothelial cells were done on an orbital platform to induce shear (44, 45), which inhibits ß2 integrin-dependent adhesion (60, 61). Again, adhesion was inhibited (mean 30% inhibition) but secretion of lactoferrin and elastase was not (mean 0% inhibition, n=2). Though inhibition of adhesion to IL-1 stimulated endothelial cells was only partial when using these independent approaches, as expected from previous reports (see above), these results together argued that degranulation does not absolutely require engagement of ß2 integrins under the conditions of our experiments, although the leukocyte integrins clearly modify neutrophil secretory responses (50, 62–64).



View larger version (13K):
[in this window]
[in a new window]
  		Figure 4. Degranulation of PMNs incubated with IL-1-stimulated endothelial cell is not blocked by antibodies against ß2 integrins. Endothelial cell (EC) monolayers were incubated in culture medium alone (control) or with IL-1 (50 U/ml) for 4 h. PMNs were preincubated with buffer alone, with the anti-CD18 mAb, 60.3, (20 µg/ml), or with an equal concentration of isotype-matched control antibody W6/32 for 15 min. The PMNs were layered onto the endothelial cells and incubated for 2 h. The values indicate the mean and ranges of duplicate points in which PMNs treated with mAb 60.3 were compared to buffer-treated PMNs. Secretion of lactoferrin from PMNs treated with W6/32 (not shown) was not different from PMNs treated with buffer. Control EC (open bars); IL-1-treated EC (filled bars).

A second component of PMN adhesion to IL-1-stimulated endothelial cells involves binding of L-selectin on the leukocyte surface to one or more inducible counterligand on the endothelial cells (45), and engagement of L-selectin can trigger activation responses of PMNs (65). We examined this mechanism using mAb LAM 1.4 and DREG 56, which inhibit L-selectin-mediated adhesion (44, 45, 61). Using anti-L-selectin antibodies alone or in combination with shear, we found that inhibiting engagement of L-selectin did not reduce the degranulation response of neutrophils incubated with IL-1-stimulated endothelial cells (not shown). Combining DREG-56 with the anti-E-selectin mAb H18/7 or with the anti-ß2 mAb 60.3 did not inhibit degranulation, although high-level inhibition of adhesion was achieved with these combinations (not shown).

These experiments demonstrated that molecular adhesive interactions known to operate when PMNs interact with cytokine-stimulated endothelial cells are not required for secretion of components of the primary and secondary granules. An additional observation demonstrated that adhesion and degranulation of PMNs could be dissociated in this system: the concentration–response relationships for the two events were different. Adhesion was induced when endothelial cells were treated with IL-1 at concentrations between 1 and 10 U/ml and was maximal at approximately 5 U/ml, whereas the maximal degranulation response occurred at a ~10-fold higher concentration and was absent or low at 1–10 U/ml IL-1 ( Fig. 5 and data not shown).



View larger version (18K):
[in this window]
[in a new window]
  		Figure 5. The concentration–response relationships for adhesion and degranulation are different when PMNs interact with IL-1-stimulated endothelial cells. Upper panel: Endothelial monolayers were incubated for 4 h with control buffer or with IL-1 in the concentrations shown. The monolayers were washed, 111In-labeled PMNs were layered over them, and adhesion was determined after a 10 min incubation. Values represent the mean and range of two experiments. Lower panel: Endothelial monolayers were incubated with or without IL-1 as described above. PMNs were layered onto the endothelial cells, incubated for 2 h, and lactoferrin secretion was assayed as described in Fig. 1. Values represent the mean and range of duplicate points in a single experiment. Two additional experiments demonstrated a similar concentration–response relationship for IL-1-induced expression of degranulating activity by endothelial cells.

Endothelial cells stimulated with IL-1 release paracrine degranulating factors
The studies outlined above suggested that endothelial cells generate one or more paracrine factors that are released into solution and do not require adhesion of neutrophils to the endothelial surface in order to signal degranulation. In experiments to confirm this, we found that PMNs suspended in supernatants from endothelial cells stimulated with IL-1 secreted lactoferrin and elastase ( Fig. 6). Triggering of neutrophil degranulation by supernatants from IL-1-stimulated endothelial cells was demonstrated in more than two dozen additional experiments (see Figs. 8, 9; data not shown).

Generation of the paracrine degranulating activity by endothelial cells depended on the concentration of IL-1, with the same concentration relationship for granular secretion shown in Fig. 5. Degranulating activity was found in the supernatants from both primary and passed endothelial monolayers stimulated with IL-1 and was time dependent, with a measurable response within 1 h and an increase in release over the next 3 h. A 1 h `pulse' of endothelial cells with 50 U/ml IL-1 induced release of degranulating activity for the next 24 h. TNF-{alpha} also induced release of degranulating activity into conditioned supernatants (not shown).

Analysis of the degranulating activity released by IL-1-stimulated endothelial cells
PAF, a known signaling molecule (1, 15), is reported to mediate activation of PMNs at the surfaces of IL-1- or TNF-{alpha}-stimulated endothelial cells (8, 12) and could potentially account for the degranulating activity. To explore this possibility, we performed an experiment in which the PAF receptor on PMNs was desensitized by pretreatment with exogenous PAF (3 nM) (48) before their incubation with IL-1-stimulated endothelial cells and found that this did not block degranulation (lactoferrin release 90% of that by PMNs pretreated with buffer in parallel). In a second experiment, pretreatment of PMNs with a PAF receptor antagonist (WEB 2086, 4 µM) reduced, but did not block, degranulation by PMNs adherent to IL-1-stimulated endothelial cells (lactoferrin release by antagonist-pretreated or control PMNs was 2.5% and 3.8%, respectively, compared to 0.8% for control PMNs incubated with unstimulated endothelial cells). The reduction of lactoferrin release could possibly have been due to PAF synthesis by the PMNs in response to signals from the IL-1-stimulated endothelial cells rather than by the endothelial cells themselves (52). Therefore, we measured PAF synthesis in IL-1-stimulated endothelial cells and found little or no production under the conditions of our experiments: in four incubations in which endothelial cells were stimulated with IL-1 (50 U/ml) for 4 or 6 h, incorporation of radiolabeled acetate into PAF (47, 49) was a mean 1.8-fold greater than in monolayers treated with control buffer (range: 0- to 3-fold), whereas there was a mean 49-fold increase in radiolabeled PAF (range: 21- to 80-fold) in endothelial cells stimulated in parallel with thrombin. These findings, in addition to the fact that PAF is not released from stimulated endothelial cells (2) whereas the degranulating factor is ( Fig. 6), indicated that PAF does not account for the degranulating activity. This was further supported by experiments in which inhibitors of transcription and translation blocked generation of the degranulating activity (see below), suggesting that the factor is a protein or polypeptide, whereas PAF is a lipid. In addition, the degranulating activity released by IL-1-stimulated endothelial cells partitioned into the aqueous phase of a Bligh/Dyer extraction and was removed by passage of conditioned supernatants through 3000 mol wt exclusion filters, findings again inconsistent with PAF (47, 49).

IL-8 is synthesized and released by endothelial cells stimulated with IL-1 or TNF-{alpha} (6, 9, 12) and induces secretion of granular contents from PMNs treated with cytochalasin B (66, 67). We found that recombinant IL-8 induces PMN degranulation under the conditions of our assay ( Fig. 7, upper panel), indicating that cytochalasin pretreatment is not required and that IL-8 may be an agonist for degranulation under physiologic conditions. Both the dominant 77 amino acid form secreted by endothelial cells and the 72 a.a. form, which is released in lesser concentrations by endothelium (67), were active and had similar concentration–response relationships. Furthermore, IL-8 was present in the conditioned supernatants from endothelial cells stimulated with IL-1 in concentrations of 1–2 nM ( Fig. 7, lower panel). However, when we compared these concentrations to the concentration–response relationship for degranulation induced by recombinant IL-8 ( Fig. 7), analysis indicated that IL-8 could not account for all of the degranulating activity secreted by stimulated endothelial cells: concentrations of IL-8 severalfold higher than those present ( Fig. 7, lower panel) would be required to induce the degree of granular secretion stimulated by supernatants from IL-1-treated endothelial cells ( Fig. 6). This was supported in several additional experiments by direct comparison of the total degranulating activity released by IL-1-stimulated endothelial cells, the concentration of IL-8 measured in the same sample, and the magnitude of degranulation induced by the same concentration of recombinant IL-8 (not shown).



View larger version (21K):
[in this window]
[in a new window]
  		Figure 7. Upper panel: IL-8 induces PMN degranulation. Neutrophils (0.5 ml, 5.5x106/ml) were incubated with recombinant IL-8 (72 amino acid form) at various concentrations or with control buffer for 2 h on gelatinized matrices. Lactoferrin secretion was measured by ELISA. Values are presented as mean and range of duplicate points in one experiment. Lower panel: Interleukin-8 is present in supernatants from endothelial cells stimulated with IL-1. Endothelial monolayers were incubated in culture medium alone (open bars) or with 50 U/ml IL-1 for 4 or 25.5 h, washed, then 0.5 ml HBSS/A was added and incubated for 2 h (37°C, 5% CO2). These supernatants were removed and assayed for IL-8 by ELISA. The values represent the means of duplicate points in assays of four supernatants from endothelial cells treated with IL-1 for 4 h (filled bar) and the parallel control supernatants, and one supernatant from endothelial cells stimulated with IL-1 for 25.5 h (hatched bar) and the parallel control. The error bars indicate standard deviations for the 4 h periods and the range of duplicate determinations for the 25.5 h points.

These experiments suggested that additional factors were present that enhanced the neutrophil response: either one or more separate degranulating factors, or a factor that does not have degranulating activity itself but potentiates the effect of low concentrations of IL-8. In the latter case, blocking or removing IL-8 would be expected to completely eliminate the degranulating activity. We examined this issue with immunoneutralization experiments. Supernatants from endothelial cells treated with IL-1 were preincubated with a polyclonal antibody against IL-8; neutrophils were resuspended in these supernatants and assayed for lactoferrin secretion. Degranulation was only partially reduced by anti-IL-8 ( Fig. 8). Using the polyclonal anti-IL-8 antibody, we found the same result with a second supernatant from a different IL-1-stimulated endothelial monolayer and with four different supernatants treated with a monoclonal anti-IL-8. In each case, the antibody completely neutralized recombinant IL-8 (100 nM, a concentration 50- to 100-fold higher than that present in supernatants from stimulated endothelial cells; Fig. 7, lower panel) but incompletely inhibited degranulating activity in the endothelial supernatants. The range of inhibition was 0–50% in six experiments that used neutralizing antibodies, the greatest degree of inhibition being that shown in Fig 8. The inability of the antibodies against IL-8 to completely block the degranulating activity indicates that an additional signaling molecule that induces PMN granular secretion is present.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 8. An antibody against IL-8 incompletely neutralizes the degranulating activity released by stimulated endothelial cells. Supernatants from IL-1-stimulated endothelial cells were collected and assayed for degranulating activity as described in Fig. 6; activity in the supernatant from stimulated endothelial cells (filled bar) induced fivefold more lactoferrin secretion from PMNs than did control buffer (not shown). Volumes of the supernatant from stimulated endothelial cells were then incubated alone (no antibody: left filled bar) or with polyclonal anti-IL-8 (hatched bar) and assayed for degranulating activity. In parallel, recombinant IL-8 was incubated in buffer alone (right filled bar) or with anti-IL-8 (right hatched bar; value near zero). In additional parallel incubations, a control polyclonal antibody (against human von Willebrand factor) did not reduce degranulating activity in the supernatant from IL-1-treated endothelial cells or the activity of recombinant IL-8 (not shown).

We then examined the requirements for transcription and translation in synthesis of the degranulating factor. Actinomycin D completely blocked the release of degranulating activity from endothelial cells stimulated with IL-1 ( Fig. 9), indicating that transcription is required. The translational inhibitors cycloheximide and emetine also blocked the release of degranulating activity (not shown), indicating that protein synthesis by the stimulated endothelial cell is required. PMN adhesion, which depends on transcription and de novo synthesis of E-selectin under these conditions (5), was also inhibited by actinomycin D, cycloheximide, and emetine in parallel (not shown).



View larger version (13K):
[in this window]
[in a new window]
  		Figure 9. Transcription is required for release of the endothelial degranulating activity. Monolayers of endothelial cells were treated with a transcription inhibitor, actinomycin D (5 µg/ml), or with buffer for 30 min (37°C, 5% CO2); IL-1 (50 U/ml) or control buffer was then added to the appropriate wells. The endothelial cells were incubated for 4 h (37°C, 5% CO2) and washed with HBSS (37°C). HBSS/A (0.55 ml) was then added and the endothelial cells were incubated for 2 h (37°C, 5% CO2). The supernatants were collected and assayed for degranulating activity as described in Methods and shown Fig. 6. Values represent the mean and range of duplicate points. In two additional experiments, actinomycin D blocked release of the degranulating activity; in one of these, emetine and cycloheximide, which are inhibitors of protein synthesis, also blocked release. Control EC (open bar); IL-1-treated EC (filled bar).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 EXPERIMENTAL PROTOCOLS
 RESULTS
 DISCUSSION
REFERENCES
 
When neutrophils are activated, receptors, adhesion molecules, and enzymes are translocated to the cell surface from intracellular granules and some of these factors are secreted. The process of degranulation influences PMN functional responses including adhesion to surfaces, aggregation, motility, and microbial killing. In subjects with granule deficiencies or impaired degranulation mechanisms, these events are altered and host defense is compromised (17, 18). In patients with sepsis and other pathologic syndromes, in contrast, exaggerated granular secretion may cause vascular or tissue injury. Although little is known regarding the influence of endothelial cells on granular secretion (17, 18, 21), interaction of PMNs with the endothelium may be an early and critical modifier of this response. We examined degranulation by PMNs when they interact with human endothelial cells stimulated by inflammatory cytokines and other agonists, and found that endothelial cells generate factors that induce degranulation and defined contributions of specific adhesion and signaling molecules.

In a previous report (68), resting unstimulated endothelial cells inhibited granular secretion when compared to acellular surfaces, possibly because of an undefined antiadhesive property. This suggested that resting endothelium may negatively modify granular secretion. We have also found evidence for an endothelial cell-associated inhibitor of degranulation (M. K. Topham et al., unpublished experiments). However, we subsequently found that endothelial cells stimulated with thrombin or leukotriene C4 prime PMNs for enhanced granular secretion when the leukocytes are activated by fMLP or leukotriene B4 (4). Priming resulted from juxtacrine signaling of PMNs by endothelial cell-associated PAF, in coordination with tethering by P-selectin. Vercellotti et al. (69) also reported that PAF expressed by stimulated endothelial cells primes PMNs for enhanced granular secretion. These experiments demonstrated that degranulation responses of PMNs are different at the surfaces of stimulated vs. quiescent endothelial cells. They also indicated that interaction of PMNs with appropriately stimulated endothelial cells may condition them for enhanced secretory responses during or after transmigration into the extravascular space. We found no evidence that resting endothelial cells or endothelial cells stimulated with thrombin directly induce secretion of primary or specific granular contents (4). The latter finding is confirmed here ( Fig. 1).

In contrast to resting endothelial cells or those stimulated with thrombin, endothelial monolayers stimulated with IL-1 or TNF generate factors that induce PMN degranulation ( Fig. 1). PMNs translocate specific granule markers to their plasma membranes as they emigrate across endothelial monolayers stimulated with IL-1 (12, 13), which is consistent with this finding. In an earlier study, however, there was no detectable granular secretion when PMNs were incubated with IL-1-stimulated endothelial cells (35). The discrepancy between this report and our experiments is likely explained by the fact that the endothelial cells were stimulated with lower concentrations of cytokine (1 U/ml IL-1 or 5 U/ml TNF-{alpha}) and for shorter periods of time (1 h) than was optimal to induce the degranulating activity in our studies (Figs. 2, 5). Also, B12 binding protein, which may underestimate granular secretion (70), was used as the marker for specific granule release (35). Together, the experiments we report here and earlier observations (12, 13, 35) indicate that PMNs interacting with endothelial cells that have been stimulated with low concentrations of IL-1 or TNF-{alpha} for relatively short periods translocate specific granule markers to the surface, but do not secrete granule contents into solution (or do so only at low levels). Interaction of PMNs with endothelial cells stimulated with higher concentrations of cytokine or for longer durations induces release of lactoferrin and elastase (Figs. 1, 2, 5). Thus, the nature and concentration of the endothelial agonist, the duration of stimulation of the endothelial cells, and the time of interaction of PMNs with stimulated endothelial cells ( Fig. 2) each influence degranulation.

We dissected known adhesion mechanisms that tether PMNs to IL-1-stimulated endothelium (including E-selectin, leukocyte ß2 integrins, and L-selectin) and found that adhesion and degranulation could be dissociated (Figs. 3–5; text). This was unexpected because adhesion-dependent signaling provides an important regulatory mechanism for PMN activation (reviewed in ref 1). Our results do not exclude the possibilities that adhesion molecules modify degranulation responses of PMNs at the endothelial surface (7, 8, 62–65, 71), that the paracrine degranulating factors released by IL-1-stimulated endothelial cells may also act in a juxtacrine fashion under some conditions (reviewed in refs 1, 16), or that translocation of neutrophil secretory organelles, which are more easily mobilized than are primary and secondary granules (17, 18, 21), occurs in response to juxtacrine signals at the surfaces of cytokine-stimulated endothelial cells (ref 13 and our unpublished experiments). However, our experiments clearly show that paracrine degranulating factors can be generated and released by endothelium stimulated by inflammatory cytokines (Figs. 6–9). Release of degranulating factors that act in solution, together with induction of this activity at higher concentrations of IL-1 than those required to induce endothelial cell-dependent adhesion ( Fig. 5) and transmigration (12, 13, 35; our unpublished observations), suggest that this process may have a role in pathologic conditions (see below).

We analyzed the activity released by IL-1-stimulated endothelial cells in order to identify the relevant degranulating factors. PAF could potentially account for the activity (8, 12, 72), but experiments using several strategies made this possibility unlikely (see Results). Furthermore, PAF is an extremely weak agonist for release of primary and specific granule contents (ref 72 and our unpublished experiments), although it can prime PMNs and synergize with other agonists (4). In contrast to the results with PAF, we found that IL-8, another endothelial signaling molecule with multiple effects on PMN functions (6, 9, 12, 15, 16, 66, 67), directly induced PMN degranulation ( Fig. 7) and in some experiments accounted for part of the degranulating activity in conditioned supernatants from IL-1-stimulated endothelial monolayers ( Fig. 8). However, neutralization of IL-8 did not block all of the degranulating activity in supernatants collected from endothelial cells stimulated with IL-1. This indicates that one or more degranulating factors in addition to IL-8 is released by endothelial cells stimulated with IL-1. A caveat to this interpretation is that experiments with antibodies can at times lead to inaccurate interpretations because of cryptic activation of the PMNs by immune complexes or direct interaction with Fc receptors (73). However, in a companion manuscript that describes experiments that extend findings in this study, we provide additional unambiguous evidence that human endothelial cells and endothelial cell lines produce degranulating factors different from IL-8 (74).

It is now clear that, when stimulated by specific agonists, human endothelial cells synthesize multiple protein or polypeptide signaling factors that trigger specific activation responses of neutrophils (see Results; refs 75–77). It is not yet clear why mediators with overlapping patterns of neutrophil activation, such as those that induce degranulation, are generated by stimulated endothelium. One possibility is that this amplifies and/or prolongs signaling of the leukocytes for a particular functional response (75). If so, generation of degranulating factors by stimulated endothelial cells, as we have described here, may be involved in physiologic defensive inflammation (77), in which a heightened or sustained neutrophil secretory response (17, 18, 21, 32) is required to deal with a large bacterial inoculum, or with particularly long-lived or virulent pathogens. However, if such degranulating factors are induced by pathologic agonists or are produced in an unregulated fashion, they may contribute to vascular injury (40–43). This possibility is supported by additional studies in which we examined the generation of degranulating activity by endothelial cells stimulated by LPS, a potent agonist for vascular injury, and further characterized the degranulating factors that are synthesized (74).


   ACKNOWLEDGMENTS
 
The authors thank Tada-atsu Imaizumi and Bart Tarbet for helpful discussions, Susan Cowley and Donelle Benson for excellent technical assistance, Jason Beales for preparation of the figures, and Leona Montoya and Michelle Bills for skillful preparation of the manuscript. We greatly appreciate the assistance of the staff of the Labor and Delivery Service at the LDS hospital, who collected umbilical samples. We also appreciate the generous gifts of biologic reagents and monoclonal antibodies from the following people: Jeoff Baker, Patrick Beatty, Michael Bevilacqua, Miroslav Ceska, Peggy Ganong, John Harlan, T. K. Kishimoto, Ivan Lindley, Peter Lomedico, Rod McEver, Joshua Rokach, Boris Schleiffenbaum, Tom St. John, and Tom Tedder. We thank John Bohnsack and Anne Shigeoka for help in studies using PMNs from a subject with leukocyte adhesion deficiency. This work was supported by the Nora Eccles Harrison Foundation and the Richard A. and Nora Eccles Harrison Fund for Cardiovascular Research, grants from the National Institutes of Health (RO1 HL44525, RO1 HL34127, T32 HL076), an Institutional Basic Sciences Research Support Grant, and a Special Center of Research in ARDS (P50 HL50153). G.A.Z. and S.M.P. were Established Investigators of the American Heart Association at the time some of this work was done.


   FOOTNOTES
 
1 Correspondence: University of Utah, CVRTI, 95 S. 2000 E. Back, Salt Lake City UT 84112–5000, USA.

2 Abbreviations: PMN, neutrophil; PAF, platelet-activating factor; IL, interleukin; TNF, tumor necrosis factor; HBSS/A, Hank's balanced salt solution containing 5 mg/ml human serum albumin; PMA, phorbol myristate acetate; fMLP, n-formyl-methionyl-leucyl-phenylalanine; BSA, bovine serum albumin; EC, endothelial cells; ELISA, enzyme-linked immunosorbent assay; PBS, phosphate-buffered saline; Ig, immunoglobulin; mAb, monoclonal antibody; LPS, lipopolysaccharide.

Received for publication September 25, 1997. Accepted for publication January 19, 1998.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 EXPERIMENTAL PROTOCOLS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Zimmerman, G. A., McIntyre, T. M., and Prescott, S. M. (1996) Adhesion and signaling in vascular cell–cell interactions. J. Clin. Invest. 98, 1699–1702[Medline]
  2. Zimmerman, G. A., McIntyre, T. M., Mehra, M., and Prescott, S. M. (1990) Endothelial cell-associated platelet-activating factor: a novel mechanism for signaling intercellular adhesion. J. Cell. Biol. 110, 529–540[Abstract/Free Full Text]
  3. Lorant, D. E., Patel, K. D., McIntyre, T. M., McEver, R. P., Prescott, S. M., and Zimmerman, G. A. (1991) Coexpression of GMP-140 and PAF by endothelium stimulated by histamine or thrombin: a juxtacrine system for adhesion and activation of neutrophils. J. Cell. Biol. 115, 223–234[Abstract/Free Full Text]
  4. Lorant, D. E., Topham, M. K., Whatley, R. E., McEver, R. P., McIntyre, T. M., Prescott, S. M., and Zimmerman, G. A. (1993) Inflammatory roles of P-selectin. J. Clin. Invest. 92, 559–570
  5. Bevilacqua, M. P., Stengelin, S., Gimbrone, M. A., Jr., and Seed, B. (1989) Endothelial leukocyte adhesion molecule 1: an inducible receptor for neutrophils related to complement regulatory proteins and lectins. Science 243, 1160–1165[Abstract/Free Full Text]
  6. Strieter, R. M., Kunkel, S. L., Showell, H. J., Remick, D. G., Phan, S. H., Ward, P. A., and Marks, R. M. (1989) Endothelial cell gene expression of a neutrophil chemotactic factor by TNF-{alpha}, LPS and IL-1ß. Science 243, 1467–1469[Abstract/Free Full Text]
  7. Lo, S. K., Suining, L., Ramos, R. A., Lobb, R., Rosa, M., Chi-Rosso, G., and Wright, S. D. (1991) Endothelial-leukocyte adhesion molecule 1 stimulates the adhesive activity of leukocyte integrin CR3 (CD11b/CD18, Mac-1, {alpha}mß2) on human neutrophils. J. Exp. Med. 173, 1493–1500[Abstract/Free Full Text]
  8. Kuijpers, T. W., Hakkert, B. C., Hoogerwerf, M., Leeuwenberg, J. F. M., and Roos, D. (1991) Role of endothelial leukocyte adhesion molecule-1 and platelet-activating factor in neutrophil adherence to IL-1-prestimulated endothelial cells. Endothelial leukocyte adhesion molecule-1 mediated CD18 activation. J. Immunol. 147, 1369–1376[Abstract]
  9. Huber, A. R., Kunkel, S. L., Todd, R. F., and Weiss, S. J. (1991) Regulation of transendothelial migration by endogenous interleukin-8. Science 254, 99–102[Abstract/Free Full Text]
  10. Luscinskas, F. W., Cybulsky, M. I., Kiely, J.-M., Peckins, C. S., Davis, V. M., and Gimbrone, M. A., Jr. (1991) Cytokine-activated human endothelial monolayers support enhanced neutrophil transmigration via a mechanism involving both endothelial-leukocyte adhesion molecule-1 and intercellular adhesion molecule-1. J. Immunol. 146, 1617–1625[Abstract]
  11. Hakkert, B. C., Kuijpers, T. W., Leeuwenberg, J. F. M., van Mourik, J. A., and Roos, D. (1991) Neutrophil and monocyte adherence to and migration across monolayers of cytokine-activated endothelial cells: the contribution of CD18, ELAM-1, and VLA-4. Blood 78, 2721–2726[Abstract/Free Full Text]
  12. Kuijpers, T. W., Hakkert, B. C., Hart, M. H. L., and Roos, D. (1992) Neutrophil migration across monolayers of cytokine-prestimulated endothelial cells: a role for platelet-activating factor and IL-8. J. Cell. Biol. 117, 564–572
  13. Kuijpers, T. W., Hoogerwerf, M., and Roos, D. (1992) Neutrophil migration across monolayers of resting or cytokine-activated endothelial cells. Role of intracellular calcium changes and fusion of specific granules with the plasma membrane. J. Immunol. 148, 72–77[Abstract]
  14. Furie, M. B., Burns, M. J., Tancino, M. C. A., Benjamin, C. D., and Lobb, R. R. (1992) E-selectin (endothelial-leukocyte adhesion molecule-1) is not required for the migration of neutrophils across IL-1-stimulated endothelium in vitro. J. Immunol. 148, 2395–2404[Abstract]
  15. Zimmerman, G. A., Prescott, S. M., and McIntyre, T. M. (1992) Endothelial cell interactions with granulocytes: tethering and signaling molecules. Immunol. Today 13, 93–100[Medline]
  16. Ebnet, K., Kaldjian, E. P., Anderson, A. O., Shaw, S. (1996) Orchestrated information transfer underlying leukocyte endothelial interactions. Annu. Rev. Immunol. 14, 155–177[Medline]
  17. Henson, P. M., Henson, J. E., Fittschen, C., Bratton, D. L., Riches, D. W. H. (1992) Degranulation and secretion by phagocytic cells. In Inflammation. Basic Principles and Clinical Correlates (Gallin J. I., Goldstein, I. M., and Snyderman, R., eds) 2nd Ed, 511–540, Raven Press, New York
  18. Gallin, J. I. (1985) Neutrophil specific granule deficiency. Annu. Rev. Med. 36, 263–274[Medline]
  19. Dewald, B., Bretz, U., and Baggiolini, M. (1982) Release of gelatinase from a novel secretory compartment of human neutrophils. J. Clin. Invest. 70, 518–525
  20. English, D., and Graves, V. (1992) Simultaneous mobilization of Mac-1 (CD11b/CD18) and formyl peptide chemoattractant receptors in human neutrophils. Blood 80, 776–787[Abstract/Free Full Text]
  21. Borregaard, N., Lollike, K., Kjeldsen, L., Sengelov, H., Bastholm, L., Nielsen, M. H., and Bainton, D. F. (1993) Human neutrophil granules and secretory vesicles. Eur. J. Hematol. 51, 187–198[Medline]
  22. Sengelov, H., Kjeldsen, L., Diamond, M. S., Springer, T. A., and Borregaard, N. (1993) Subcellular localization and dynamics of Mac-1 ({alpha}mß2) in human neutrophils. J. Clin. Invest. 92, 1467–1476
  23. Gallin, J. I., Wright, D. G., and Schiffmann, E. (1978) Role of secretory events in modulating human neutrophil chemotaxis. J. Clin. Invest. 62, 1364–1374
  24. Todd, R. F., III, Arnaout, M. A., Rosin, R. E, Crowley, C. A., Peters, W. A., and Babior, B. M. (1984) Subcellular localization of the large subunit of Mo1 (Mo1a; formerly gp 110), a surface glycoprotein associated with neutrophil adhesion. J. Clin. Invest. 74, 1280–1290
  25. Singer, I. I., Scott, S., Kawka, D. W., and Kazazis, D. M. (1989) Adesomes: specific granules containing receptors for laminin, C3bi/fibrinogen, fibronectin, and vitronectin in human polymorphonuclear leukocytes and monocytes. J. Cell Biol. 109, 3169–3182[Abstract/Free Full Text]
  26. Gallin, J. I. (1980) Degranulating stimuli decrease the negative surface charge and increase the adhesiveness of human neutrophils. J. Clin. Invest. 65, 298–306
  27. Oseas, R., Yang, H.-H., Baehner, R. L., and Boxer, L. A. (1981) Lactoferrin: a promoter of polymorphonuclear leukocyte adhesiveness. Blood 57, 939–945[Abstract/Free Full Text]
  28. Boxer, L. A., Haak, R. A., Yang, H.-H., Wolach, J. B., Whitcomb, J. A., Butterick, C. J., and Baehner, R. L. (1982) Membrane-bound lactoferrin alters the surface properties of polymorphonuclear leukocytes. J. Clin. Invest. 70, 1049–1057
  29. Brock, J. (1995) Lactoferrin: a multifunctional immunoregulatory protein. Immunol. Today 16, 417–419[Medline]
  30. Hoffstein, S. T., Friedman, R. S., and Weissmann, G. (1982) Degranulation, membrane addition, and shape change during chemotactic factor-induced aggregation of human neutrophils. J. Cell Biol. 95, 234–241[Abstract/Free Full Text]
  31. Kobayashi, T., and Robinson, J. M. (1991) A novel intracellular compartment with unusual secretory properties in human neutrophils. J. Cell Biol. 113, 743–756[Abstract/Free Full Text]
  32. Brumell, J. H., Volchuk, A., Sengelov, H., Borregaard, N., Cieutat, A.-M., Bainton, D. F., Grinstein, S., and Klip, A. (1995) Subcellular distribution of docking/fusion proteins in neutrophils, secretory cells with multiple exocytic compartments. J. Immunol. 155, 5750–5759[Abstract]
  33. Bretscher, M. S. (1996) Moving membrane up to the front of migrating cells. Cell 85, 465–467[Medline]
  34. Smith, C. W., Rothlein, R., Hughes, B. J., Mariscalco, M. M., Rudloff, H. E., Schmalsteig, F. C., and Anderson, D. C. (1988) Recognition of an endothelial determinant for CD18-mediated human neutrophil adherence and transendothelial migration. J. Clin. Invest. 82, 1746–1756
  35. Moser, R., Schleiffenbaum, B., Groscurth, P., and Fehr, J. (1989) Interleukin 1 and tumor necrosis factor stimulate human vascular endothelial cells to promote transendothelial neutrophil passage. J. Clin. Invest. 83, 444–455
  36. Bohnsack, J. F., Akiyama, S. K., Damsky, C. H., Knape, W. A., and Zimmerman, G. A. (1990) Human neutrophil adherence to laminin in vitro: evidence for a distinct neutrophil integrin receptor for laminin. J. Exp. Med. 171, 1221–1237[Abstract/Free Full Text]
  37. Hughes, B. J., Hollers, J. C., Crockett-Torabi, E., and Smith, C. W. (1992) Recruitment of CD11b/CD18 to the neutrophil surface and adherence-dependent cell locomotion. J. Clin. Invest. 90, 1687–1696
  38. Huber, A. R., and Weiss, S. J. (1989) Disruption of the subendothelial basement membrane during neutrophil diapedesis in an in vitro construct of a blood vessel wall. J. Clin. Invest. 83, 1122–1136
  39. Wright, D. G., and Gallin, J. I. (1979) Secretory responses of human neutrophils: exocytosis of specific (secondary) granules by human neutrophils during adherence in vitro and during exudation in vivo. J. Immunol. 123, 285–294[Abstract/Free Full Text]
  40. Harlan, J. M., Killen, P. D., Harker, L. A., Striker, G. E., and Wright, D. G. (1981) Neutrophil-mediated endothelial injury in vitro. Mechanism of cell detachment. J. Clin. Invest. 68, 1394–1403
  41. Harlan, J. M., Schwartz, B. R., Reidy, M. A., Schwartz, S. M., Ochs, H. D., and Harker,