Effect of IFN-gamma on the killing of S. aureus in human whole blood. Assessment of bacterial viability by CFU determination and by a new method using alamarBlue.

Given the increasing incidence of methicillin resistant Staphylococcus aureus (MRSA) and the recent emergence of MRSA with a reduced susceptibility to vancomycin, alternative approaches to the treatment of infection are of increasing relevance. The purpose of these studies was to evaluate the effect of IFN-gamma on the ability of white blood cells to kill S. aureus and to develop a simpler, higher throughput bacterial killing assay. Using a methicillin sensitive clinical isolate of S. aureus, a clinical isolate of MRSA, and a commercially available strain of MRSA, studies were conducted using a killing assay in which the bacteria were added directly into whole blood. The viability of the bacteria in samples harvested at various time points was then evaluated both by the classic CFU assay and by a new assay using alamarBlue. In the latter method, serially diluted samples and a standard curve containing known concentrations of bacteria were placed on 96-well plates, and alamarBlue was added. Fluorescence readings were taken, and the viability of the bacteria in the samples was calculated using the standard curve. The results of these studies demonstrated that the CFU and alamarBlue methods yielded equivalent detection of bacteria diluted in buffer. For samples incubated in whole blood, however, the alamarBlue method tended to yield lower viabilities than the CFU method due to the emergence of a slower growing subpopulation of S. aureus upon incubation in the blood matrix. A significant increase in bacterial killing was observed upon pretreatment of whole blood for 24 h with 5 or 25 ng/ml IFN-gamma. This increase in killing was detected equivalently by the CFU and alamarBlue methods. In summary, these studies describe a method that allows for the higher throughput analysis of the effects of immunomodulators on bacterial killing.

A neutralizing monoclonal antibody specific for the dimer interface region of IL-8.

We have generated two mAbs, 6G4.2.5 and A5.12.14, that are similarly capable of neutralizing the biologic activity of wild-type IL-8. To characterize these antibodies further, their reactivity against a series of engineered IL-8 monomer and dimer variants was examined using a neutrophil degranulation assay. While 6G4.2.5 was found to block effectively the biologic activity of all variants regardless of their dimerization status, the results for A5.12.14 differed dramatically. A5.12.14 fully inhibited the agonist activity of one of the monomer variants, partially blocked the activity of another, and had no effect on the activity of two other variants. These results suggested that the binding epitope of A5.12.14 was being affected by the particular amino acid substitutions introduced into the dimer interface region of the variants to disfavor dimerization. If A5.12.14 indeed binds to the dimer interface region of IL-8, it could be predicted that this mAb would be unable to inhibit the activity of dimeric IL-8. This was confirmed in studies which showed that A5.12.14 had no demonstrable effect on the activity of a constitutively dimeric IL-8 variant. These studies represent the first example of a mAb specific for the dimerization status of IL-8.

IL-8 single-chain homodimers and heterodimers: interactions with chemokine receptors CXCR1, CXCR2, and DARC.

Covalent single-chain dimers of the chemokine interleukin-8 (IL-8) have been designed to mimic the dimeric form of IL-8 in solution and facilitate the production of heterodimer variants of IL-8. Physical studies indicated that use of a simple peptide linker to join two subunits, while allowing receptor binding and activation, led to self-association of the tethered dimers. However, addition of a single disulfide crosslink between the tethered subunits prevented this multimer from forming, yielding a species of dimer molecular weight. Crosslinked single-chain dimers bind to both IL-8 neutrophil receptors CXCR1 and CXCR2 as well as to DARC, as does a double disulfide-linked dimer with no peptide linker. In addition, neutrophil response to these dimers as measured by chemotaxis or beta-glucuronidase release is similar to that elicited by wild-type IL-8, providing evidence that the dissociation of the dimeric species is not required for these biologically relevant activities. Finally, through construction of single-chain heterodimer mutants, we show that only the first subunit's ELR motif is the single-chain variants.

Exchanging interleukin-8 and melanoma growth-stimulating activity receptor binding specificities.

Interleukin-8 (IL-8), a CXC chemokine, is known to bring about chemotaxis and activation of neutrophils through high affinity binding to at least two distinct receptors, receptor-A and receptor-B. The IL-8 homolog melanoma growth stimulating activity (MGSA) is also active toward neutrophils. In contrast to IL-8, MGSA binds receptor-B with high affinity and binds receptor-A with approximately 400-fold lower affinity. Using the structure of IL-8 (Clore et al.(1990) Biochemistry, 29, 1689-1696; Baldwin et al. (1991) Proc. Natl. Acad. Sci. U.S.A. 88, 502-506) and the NMR-determined structure of MGSA (Fairbrother et al. (1994) J. Mol. Biol. 242, 252-270), we designed variants of both IL-8 and MGSA to investigate the basis of specificity for binding of these chemokines to the IL-8 receptors. The most outstanding structural difference between IL-8 and MGSA lies in the loop preceding the first beta-strand. When the corresponding (shorter) loop from MGSA was swapped into IL-8, both receptor-A and receptor-B binding affinities were significantly (>300-fold) reduced. However, with additional mutations that affect packing interactions, an IL-8 variant specific for receptor-B binding was produced. Conversely, when the same loop from IL-8 was swapped into MGSA, receptor-B binding was maintained with only a approximately 30-fold reduction in receptor-A affinity. Again, mutations affecting packing of the loop yielded a MGSA variant with high affinity for both receptors, like IL-8. Finally, we show, through point mutations in a monomeric IL-8 framework, that individual side chain substitutions can affect receptor specificity.

Immunological priming attenuates the in vivo pathophysiological response to lipopolysaccharide. Comparison of cytokine production, tissue injury, and lethality in complete Freund's adjuvant-primed mice and in unprimed mice.

Previous reports have indicated that immunological priming of animals will result in increased cytokine production and enhanced susceptibility to the toxicity of cytokines. We primed mice with complete Freund's adjuvant and challenged 2 weeks later with 1 mg/mouse of lipopolysaccharide. Primed mice produced less tumor necrosis factor than naive mice (35 +/- 8 vs 108 +/- 20 ng/ml) and also less interleukin-6 (182 +/- 37 vs 6.39 +/- 155 ng/ml). Leukopenia developed only in the naive mice. Although neutropenia and lymphocytosis developed in both groups, the alterations manifested themselves more quickly in primed mice. Primed mice had substantially greater pulmonary neutrophil sequestration determined both enzymatically and histologically but no lung damage. However, primed mice had significantly less small bowel damage than naive mice. Mortality was substantially reduced in primed mice compared with unprimed mice. These results demonstrate that immunological priming in vivo decreases cytokine production in response to lipopolysaccharide challenge, decreases organ injury, and reduces mortality.

Regulation of interleukin 8 gene expression by oxidant stress.

Interleukin 8 (IL-8) is a recently described cytokine that functions as a potent neutrophil chemoattractant and activator. We sought to examine the link between the generation of reactive oxygen intermediates (ROI) and the regulation of IL-8 gene expression to specifically test the hypothesis that ROI would induce production of IL-8 mRNA and protein. In lipopolysaccharide-stimulated human whole blood, the OH radical scavenger dimethyl sulfoxide (Me2SO) dramatically inhibited (approximately 90%) IL-8 production, but had minimal effects on the production of tumor necrosis factor, interleukin 1 beta (IL-1), and IL-6. To determine whether NADPH-oxidase-generated free radicals were critical in the regulation of IL-8, studies were performed using blood from patients with chronic granulomatous disease. In both normal individuals and patients with chronic granulomatous disease, production of IL-8 could be initiated with lipopolysaccharide, phytohemagglutinin, or aggregated immune complexes, and this production could be inhibited by Me2SO (1% v/v). To examine if oxidant stress represents a ubiquitous mechanism for the induction of IL-8, experiments were performed in cultured cell lines. In the human hepatoma cell line Hep-G2, Me2SO dose-dependently inhibited tumor necrosis factor-stimulated IL-8 production, with a 74 +/- 1% reduction observed at a Me2SO concentration of 1%. Direct exposure to ROI demonstrated that H2O2 stimulated IL-8 production in a dose-dependent manner in Hep-G2 cells, A549 pulmonary type II epithelial cells, and human skin fibroblasts; this induction could be prevented by addition of catalase. The production of IL-8 appeared to be specific to an oxidant stress since exposure of the cells to heat shock or chemical stress did not induce expression of IL-8. These studies demonstrate that oxidant stress is an important regulator of IL-8 gene expression and support the hypothesis that low levels of ROI may serve to initiate IL-8 production which then serves to recruit neutrophils to sites of inflammation.

Differential expression of tumor necrosis factor and interleukin-6 by peritoneal macrophages in vivo and in culture.

To investigate the differences in cytokine regulation in vitro as compared to in vivo, we examined the synthesis of tumor necrosis factor-alpha (TNF-alpha) and interleukin-6 (IL-6) by peritoneal macrophages in response to lipopolysaccharide (LPS). Mice (CBA/J) were primed with an intraperitoneal injection of complete Freund's adjuvant and after 2 weeks, peritoneal cells were harvested for culture or mice were injected intraperitoneally with LPS for in vivo studies. In ascites fluid, TNF-alpha peaked 1 hour after LPS and returned to baseline levels by 4 hours. In contrast, TNF-alpha in the media reached maximum at 7 hours. Expression of TNF-alpha messenger (m)RNA in vivo was rapid but transient, as levels peaked at 15 minutes and returned to baseline 1 hour after LPS. In contrast, TNF-alpha mRNA in vitro became maximal at 1 hour, but remained elevated to 5 hours after LPS. In vivo, IL-6 in ascites fluid peaked at 2 hours, whereas in vitro, IL-6 continued increasing to 24 hours. In vivo, IL-6 mRNA reached maximum at 30 minutes, but fell below baseline by 1.5 hours after LPS. In contrast, IL-6 mRNA in vitro was sustained at maximal expression between 5 to 9 hours after LPS. These results demonstrate that both TNF-alpha and IL-6 synthesis is more rapid in vivo than in vitro. The rapid kinetics of cytokine expression in vivo must considered when designing strategies to inhibit cytokine action in vivo.

Oxygen radical scavengers selectively inhibit interleukin 8 production in human whole blood.

The hydroxyl radical (OH.) scavenger dimethyl sulfoxide (DMSO) was found to dose-dependently inhibit interleukin 8 (IL-8) production in LPS-stimulated human whole blood. At a concentration of 1% (vol/vol), DMSO blocked IL-8 release by approximately 90% in the presence of 1 microgram/ml LPS at a 24-h time point, but did not affect cell viability or reduce the production of tumor necrosis factor (TNF), interleukin 6, or interleukin-1 beta (IL-1 beta). DMSO was found to directly inhibit IL-8 expression at the level of transcription. Furthermore, this effect was not LPS-specific, in that IL-8 production was reduced by DMSO to a similar extent upon stimulation of blood with phytohemagglutinin, aggregated immune complexes, TNF, or IL-1 beta. Other oxygen radical scavengers that have been shown to inhibit OH.-dependent reactions (dimethyl thiourea, thiourea, mannitol, and ethanol) also inhibited IL-8 production. Conversely, addition of H2O2 caused a dose-dependent stimulation of IL-8 release. These results provide evidence that reactive oxygen metabolites play an important role in the regulation of IL-8 production and suggest that reduction of IL-8 release may contribute to the beneficial effects of antioxidants in experimental models of inflammation and ischemia/reperfusion injury.

Comparison of methods for incorporating a radioiodinated residualizing cholesteryl ester analog into low density lipoprotein.

Two different methods were evaluated for incorporating [125I]cholesteryl iopanoate ([125I]CI), a non-hydrolyzable cholesteryl ester analog, into LDL. The first procedure was an organic solvent delipidation-reconstitution procedure (R[125I-CI]LDL) while the second involved incubation of detergent (Tween-20)-solubilized [125I]CI with whole plasma (D[125I-CI]LDL). R[125I-CI]LDL behaved similar to native LDL in vitro, but was markedly different in vivo, apparently due to a heterogeneity in particle size. D[125I-CI]LDL, however, was metabolized normally both in vitro and in vivo. These results, combined with the residualizing nature of [125I]CI, demonstrate that D[125I-CI]LDL is appropriate for tracing LDL uptake in vivo.

Profile of cytokines in synovial fluid specimens from patients with arthritis. Interleukin 8 (IL-8) and IL-6 correlate with inflammatory arthritides.

The synovial fluid aspirated from patients with symptomatic arthritis was analyzed for the presence of tumor necrosis factor (TNF), interleukin 6 (IL-6) and interleukin 8 (IL-8). All three cytokines were found in both inflammatory and non-inflammatory arthritides: IL-8 levels ranged from less than 20 to 38,990 pg/ml, IL-6 from less than 10 to 72,300 pg/ml and TNF from less than 4 to 61 pg/ml. No inhibitors of cytokine activity were found. IL-8 and IL-6 were present in significantly higher levels in patients with inflammatory arthritis compared to patients with osteoarthritis, and there was significant correlation between the IL-6 and IL-8 levels. These findings document the presence of multiple cytokines in the synovial fluid specimens of patients with arthritis, and demonstrate that higher cytokine levels accompany inflammatory arthritis.

Local production of interleukin-8 is associated with nosocomial pneumonia.

One hundred five (70%) of 151 patients hospitalized in the intensive care unit and undergoing mechanical ventilation had bronchial secretions that tested positive for interleukin-8 within 36 hours of admission. Arterial blood, mixed venous blood, and urine collected simultaneously all tested negative, except for 11 patients admitted with intra-abdominal septic foci. The presence of interleukin-8 in the pulmonary air space early in the course of hospitalization was significantly associated with patients with multiple injuries, the need for greater ventilatory support, the occurrence of pulmonary dysfunction, and a 66% incidence of nosocomial bacterial pneumonia. We conclude that the early local production of interleukin-8 in the lungs is an early marker of pulmonary injury and may be involved in the pathogenesis of nosocomial bacterial pneumonia.

Interleukin-1 receptor antagonist protein inhibits interleukin-8 expression in lipopolysaccharide-stimulated human whole blood.

Interleukin-8 (IL-8) is a neutrophil and lymphocyte chemoattractant and activator that may play an important role in mediating events at sites of inflammation. IL-8 is produced by many cell types in response to a variety of inducers, including interleukin-1 (IL-1). Studies were conducted to address the question of whether an inhibitor of IL-1 action, IL-1 receptor antagonist protein (IRAP), would suppress IL-8 production. Lipopolysaccharide (LPS)-stimulated human whole blood was used as an ex vivo model of local cytokine production. Preliminary time course studies showed that plasma IL-1 beta levels were fully induced by 6 hours (approximately 15 ng/ml) and persisted at this level over 24 hours. IL-8 production, in contrast, reached a plateau between 6 to 12 hours (5 ng/ml) and then increased rapidly to 17 ng/ml at 24 hours. Addition of IRAP was found to dose-dependently inhibit IL-8 expression at the levels of both protein and mRNA. A 50% reduction in IL-8 protein release occurred at an IRAP dose of 8 micrograms/ml (5 x 10(-7) mol/l) at 24 hours. A 2 hour delay in the addition of IRAP relative to LPS still permitted optimal reduction in IL-8 release, whereas delays of 4-8 hours reduced or eliminated the inhibitory effect. IRAP was found to have no effect on the LPS-stimulated production of IL-1 alpha or IL-1 beta. In addition, experiments performed with isolated peripheral blood cells demonstrated that, whereas monocytes were the major producers of IL-8, IRAP was equally effective in reducing IL-8 production in neutrophil and mononuclear cell suspensions. These studies further document the role of IL-1 in inducing the production of IL-8 and indicate that the ability of IRAP to suppress IL-8 expression may be an important mechanism of the anti-inflammatory properties of this molecule.

Detection of plasma tumor necrosis factor, interleukins 6, and 8 during the Jarisch-Herxheimer Reaction of relapsing fever.

The Jarisch-Herxheimer Reaction (J-HR) is a clinical syndrome occurring soon after the first adequate dose of an antimicrobial drug to treat infectious diseases such as Lyme disease, syphilis, and relapsing fever. Previous attempts to identify factors mediating this reaction, that may cause death, have been unsuccessful. We conducted a prospective trial in Addis Ababa, Ethiopia on 17 patients treated with penicillin for proven louse-borne relapsing fever due to Borrelia recurrentis to evaluate the association of symptoms with plasma levels of tumor necrosis factor (TNF), interleukins 6, and 8 (IL-6 and -8). 14 of the 17 (82%) patients experienced a typical J-HR consisting of rigors, a rise in body temperature (1.06 +/- 0.2 degrees C) peaking at 2 h, leukopenia (7.4 +/- 0.6 x 10(-3) cells/mm3) at 4 h, a slight decrease, and then rise of mean arterial blood pressure. Spirochetes were cleared from blood in 5 +/- 1 h after penicillin. There were no fatalities, but constitutional symptoms were severe during J-HR. Plasma TNF, IL-6, and -8 were raised in several patients on admission, but a seven-, six-, and fourfold elevation of these plasma cytokine concentrations over admission levels was detected, respectively, occurring in transient form coincidental with observed pathophysiological changes of J-HR. Elevated plasma cytokine levels were not detected in the three patients who did not suffer J-HR. We conclude that the severe pathophysiological changes characterizing the J-HR occurring on penicillin treatment of louse-borne relapsing fever are closely associated with transient elevation of plasma TNF, IL-6, and -8 concentrations.

Anti-tumor necrosis factor antibody therapy fails to prevent lethality after cecal ligation and puncture or endotoxemia.

Cytokines have been studied intensively to delineate their role in the altered pathophysiology observed in septic shock. We studied the role of TNF in the lethality of two well characterized models of septic shock by inhibiting TNF's activity with a specific antibody. In the first model, sepsis was induced by cecal ligation and puncture (CLP), and in the second model sepsis was induced by either an i.p. or i.v. injection of LPS. After CLP, plasma endotoxin was detectable within 4 h and reached a peak at 8 h (136 +/- 109 ng/ml). TNF bioactivity peaked at 12 h (528 +/- 267 pg/ml) at a significantly higher level than sham-operated control mice (64 +/- 31 pg/ml). After i.p. LPS, TNF peaked much more quickly (90 min) compared with CLP and at a significantly higher level (107,900 +/- 25,000 pg/ml). Another cytokine studied in septic shock, IL-6, peaked at 12 h after CLP at 1011 +/- 431 pg/ml, and at 90 min after lethal LPS at 16,300 +/- 3,700 pg/ml. Mice were treated with an anti-TNF antibody that has been shown previously to inhibit in vivo TNF activity. Antibody treatment of mice subjected to CLP significantly reduced TNF bioactivity but did not reduce mortality or pulmonary neutrophilic infiltration. In the i.v. LPS model, anti-TNF antibody treatment concomitant with LPS injection reduced plasma TNF activity from 80,000 +/- 20,000 pg/ml to undetectable levels. However, anti-TNF treatment immediately before either i.v. or i.p. LPS did not reduce mortality. Additionally, when the antibody was administered 4 h before the lethal i.v. LPS, there was no reduction in lethality. These data show that in two separate models of septic shock blockade of TNF biologic activity will not prevent lethality.

Biphasic production of IL-8 in lipopolysaccharide (LPS)-stimulated human whole blood. Separation of LPS- and cytokine-stimulated components using anti-tumor necrosis factor and anti-IL-1 antibodies.

TNF, IL-1, and IL-6 are integral components of the cytokine cascade released in the response to inflammatory stimuli such as LPS. IL-8 is produced both in response to LPS as well as TNF and IL-1. The early, local production of TNF and IL-1 may therefore contribute to the subsequent expression of IL-8. This hypothesis was tested using LPS-stimulated human whole blood as an ex vivo model of local cytokine production. The production of TNF, IL-1 alpha, IL-1 beta, IL-6, and IL-8 was found to be responsive to a wide range of LPS concentrations (0.1 ng/ml-10 micrograms/ml). These cytokines were first detected between 1 to 4 h post-LPS stimulation, and reached plateau levels after 6 to 12 h. IL-8, however, also displayed a secondary wave of production, with the levels again increasing between 12 to 24 h. The IL-8 present in the plasma after LPS stimulation was biologically active, as assessed by neutrophil chemotaxis. In further studies, addition of anti-TNF and anti-IL-1 neutralizing antibodies, alone and in combination, to LPS-stimulated blood resulted in nearly complete ablation of the secondary phase of IL-8 synthesis at both the levels of protein and mRNA, while leaving the first, LPS-mediated phase of IL-8 synthesis unaffected. This model of cytokine production in human whole blood may reflect the sequence of events in a localized environment of inflammation where both a primary stimulus and the induced early cytokine mediators may serve to elicit multiple, temporally distinct phases of IL-8 production.

Synthesis and evaluation of radioiodinated cholesteryl ethers as lipoprotein probes.

Two novel cholesteryl ether derivatives were synthesized and radioiodinated: (1) [125I]cholesteryl m-iodobenzyl ether (125I-CIBE) and (2) [125I]cholesteryl 12-(m-iodophenyl)dodecyl ether (125I-CIDE). These radioiodinated ethers were incorporated into low-density lipoprotein (LDL) by incubating the compounds (solubilized in saline with Tween-20) with isolated LDL or with whole plasma. Such LDL preparations were taken up by cultured fibroblasts in a receptor-dependent manner similar to that of radioiodinated LDL. Upon injection into guinea pigs, 125I-CIBE-labeled guinea pig LDL cleared from the plasma similarly to radioiodinated guinea pig LDL. The primary sites of 125I-CIBE uptake were the adrenal and the liver, and the compound was stable to both hydrolysis and deiodination over 24 h. In summary, 125I-CIBE and 125I-CIDE, like previously described tritiated cholesteryl ethers, appear to be potentially useful tracers of cholesteryl ester uptake. Moreover, these radioiodinated probes have the advantage of being more easily quantitated in tissue samples as well as being detectable by noninvasive scintigraphic imaging.

IL-8 in septic shock, endotoxemia, and after IL-1 administration.

Much effort has been directed toward elucidating the host response to sepsis and inflammation, resulting in the definition of a cascade of endogenous mediators that direct metabolic and immunological responses. Here we report that IL-8, a novel cytokine produced by a variety of cells in vitro in response to stimulation with bacterial LPS and the proinflammatory cytokines, appears in the circulation of primates in vivo during septic shock, sublethal endotoxemia, and after the administration of IL-1 alpha. The magnitude of the IL-8 response correlates with the severity of the insult, and levels of IL-8 peak relatively late, after those of TNF-alpha and IL-1 beta, and simultaneously with those of IL-6. IL-8 has been primarily defined as a selective activator and chemoattractant of neutrophils, and we demonstrate that after LPS or IL-1 alpha infusion, circulating neutrophil numbers rapidly recover from an initial neutropenia while IL-8 concentrations are maximal, supporting the hypothesis that IL-8 influences circulating leukocyte populations in vivo. We conclude that IL-8 is another participant in the cytokine cascade elicited by sepsis and inflammation and, as such, may play a significant role in host defense and disease.

Sandwich ELISA for detection of picogram quantities of interleukin-8.

A sensitive sandwich enzyme-linked immunosorbent assay (ELISA) was developed for interleukin-8 (IL-8), a neutrophil chemoattractant and activator. A polyclonal antibody to recombinant human IL-8 was raised in rabbits, and the IgG was isolated from the antisera using a protein A column. Native and biotinylated forms of this antibody served as the capture antibody and developing antibody for the ELISA, respectively, and avidin-conjugated horse radish peroxidase provided the means for enzymatic color development. The lower limit of sensitivity for the assay was found to be 84 +/- 20 pg/ml IL-8 (mean +/- SD for 10 determinations). An inter-assay variability of 15-29% and an intra-assay variability of 12% were observed. The assay was able to detect IL-8 when the samples were prepared in either normal saline, RPMI, or human plasma. The development of this rapid, sensitive assay should provide a means to more fully evaluate the role of this cytokine in diverse disease states.

Kinetics of TNF, IL-6, and IL-8 gene expression in LPS-stimulated human whole blood.

While the production of tumor necrosis factor (TNF) and interleukin-6 (IL-6) in septic shock and other inflammatory states is well established, the role of interleukin-8 (IL-8), a recently described neutrophil chemoattractant and activator, has yet to be fully elucidated. Using lipopolysaccharide (LPS)-stimulated human whole blood as an ex vivo model of sepsis, the kinetics of messenger RNA (mRNA) up-regulation and protein release of these cytokines were examined. Two waves of cytokine gene activation were documented. TNF and IL-6 were induced in the first wave with mRNA levels peaking between 2-4 hours and then rapidly declining. TNF and IL-6 protein peaked at 4-6 hours and then stabilized. IL-8 mRNA and protein were induced in the first wave, reached a plateau between 6-12 hours, and rose again in a second wave which continued to escalate until the end of the 24 hour study. These data demonstrate the complex patterns of cytokine gene expression and suggest that production of early mediators may augment continued expression of IL-8 to recruit and retain neutrophils at a site of inflammation.