Research Paper Mediators of Inflammation 3, 155,-160 (1994)

In the present study we examined whether endothelin-1 stimulation of human monocytes causes release of chemotactic factors. It was found that monocytes released neutrophil- and monocyte-chemotactic activity in a dose- and time-dependent manner in response to ET-1. ET-1 did not show any chemotactic activity by itself. NCA was detected in monocyte supernatants in response to ET-1 (0.01–100 nM) after 1, 4, 8 and 24 h stimulation. MCA was detected only after 24 h stimulation with ET-1 (0.1–100 nM). Preincubation of the monocyte cultures with the lipoxygenase inhibitors nordihydroguaiaretic acid (10−4 M) or diethylcarbamazine (10−9 M) completely abolished the appearance of NCA and MCA. NCA was neutralized by > 75% using a polyclonal antibody against human interleuktn-8. The ET-1 induced release of IL-8 was confirmed by IL-8 ELISA. A monoclonal antibody against human monocyte chemotactic protein-1 neutralized MCA by > 80%. It is concluded that ET-1 stimulation of monocytes in vitro causes release of neutrophil- and monocyte-chemotactic activity identified as IL-8 and MCP-I respectively. An intact lipoxygenase pathway is crucial for this effect of ET-1 to occur.


Introduction
Endothelin-1 is a 21-amino acid peptide, initially isolated from endothelial cells and described as the most potent biological vasoconstrictor known. 1," It has recently been shown that ET-1 also might activate monocytes and macrophages, causing release of tumour necrosis factor-(x (TNF00, interleukin-l (IL-11) and interleukin-6 (IL-6).
When activated, monocytes release several chemotactic factors for monocytes and neutrophils. Some chemotactic factors such as TNF0t and leukotriene B (LTB4) show chemotactic activity for both monocytes and neutrophils, 4,5 while interleukin-8 (IL-8) has chemotactic activity for neutrophils and lymphocytes, but not for monocytes6. Monocyte chemotactic protein-1 (MCP-1) has specific chemotactic and activating activity for monocytes.7 To further examine the pro-inflammatory effects of ET-1, human monocytes in vitro were stimulated with ET-1 for different time intervals, and the supernatants tested for monocyte-and neutrophil chemotactic activity. To identify the factors responsible for the chemotactic activity, neutralization experiments with antibodies against known chemotactic factors were performed. For comparison, supernatants from ET-1 stimulated human endothelial cells were also tested for chemotactic activity.

Methods
Isolation and cultivation of monocytes Highly purified monocytes were obtained using the method described previously*. Briefly, unseparated mononuclear cells (PBMC) were isolated from human A blood buffy coat (The Blood Bank, University Hospital of Tromso, Norway) by density centrifugation with Lymphoprep (Nycomed Pharma AS, Oslo, Norway). The cells were washed three times with HBSS (Gibco, Glasgow, UK) and resuspended in medium consisting of RPMI-1640 with 100 IU/ml penicillin, 100 l.tg/ml streptomycin and supplemented with 25% A human serum (The Bloodbank, University Hospital of TromsO). The peripheral blood mononuclear cells (PBMC) were then seeded in 24 Isolation of human neutrophils: Neutrophils for use in chemotaxis assay, were isolated from heparinized blood from healthy adult donors by density centrifugation with Polyprep (Nycomed Pharma AS, Oslo, Norway), as described. Contaminating erythrocytes were lysed with 0.2% NaCl for 60 s. The cells were kept on ice, washed and resuspended at 1 x 106/ml in RPMI-1640 supplemented with 2% human serum and used immediately in experiments. The purity of the neutrophil preparations was > 95% as assessed morphologically with Giemsa staining, and the viability was > 98% tested by Trypan blue exclusion.
Stimulation of monocyte cultures with ET-I: The monocyte cultures were stimulated with various concentrations of ET-1 ranging between 0.005 nM and 100 nM for 1, 4, 8 and 24 h. As control, RPMI-1640 instead of ET-1 was used in each experiment. The supernatants were harvested at the indicated time points and stored at-20C before being assayed for neutrophil-and monocyte chemotactic activity.
To exclude possible contamination with endotoxin, boiling of ET-1 for 30 min, which denaturates the peptide but leaves endotoxin intact, was used in stimulation experiments. In some experiments ET-1 was preincubated with ET-1 antiserum prior to addition to the cell cultures. In addition, supernatants and ET-1 were tested by Endospecy (Seikagaku Kogyo, Tokyo, Japan). Chemotaxis assay: The chemotaxis assay was performed in a 48-well chemotaxis chamber (Neuroprobe Inc., Cabin John, MD, USA) as described previously. 11 The bottom wells of the chamber were filled with 25 l-tl of supernatant in duplicate or triplicate. A polyvinyl pyrrolidone (PVP) polycarbonate filter (Neuroprobe Inc., Cabin John, MD, USA) with a pore size of 5 l.tm for monocyte chemotaxis, or a 5 m PVP free polycarbonate filter 12 for neutrophil chemotaxis assay was placed over the bottom wells. The upper wells were filled with 50 l.tl of cell suspension. The chamber was then incubated in humidified air in 5% CO2 at 37C for either 40 min for neutrophil chemotaxis assay or 3 h for monocyte chemotaxis assay. After incubation the chamber was disassembled, and the filter was fixed in 2.5% glutaraldehyde (Merck, Darmstadt, Germany), 0.1 M sucrose, and 0.1 M cacodylate buffer, pH 7.3, for 30 min at room temperature. The filter was then stained with 1% Giemsa (Sigma) for 30 min, and mounted on a glass slide. Cells completely migrated through the filter were counted in 6-12 high power fields (HPF, x 100). Chemotactic response was expressed as the mean number of cells per one HPF in duplicate or triplicate wells. RPMI-1640 was used to determine background migration. Formyl-methionyl-leucylphenylalanine (FMLP, Sigma) at a concentration of 10 -7 M, was used as positive control.
Neutralization studies: In the neutralization studies, supernatants from monocytes stimulated with ET-1 for different time intervals were incubated with a polyclonal goat anti-human IL-8 antibody (British Biotechnology, UK); or a monoclonal anti-human TNF0t antibody (Boehringer Mannheim, Germany); or a monoclonal anti-human MCP-1/MCAF antibody (PeproTech Inc., NJ, USA) for 1 h at 37C, before being tested for neutrophil and monocyte chemotactic activity. Controls were performed using non-immune serum. Anti-human IL-8 antibody, 1 mg/ml, was used at an initial dilution of 1:75 in the experiments. Anti-human TNF0t antibody, 200 t.tg/ml, was used at an initial dilution of 1:50. Anti-human MCP-1/MCAF antibody, 1 mg/ml, was used at an initial dilution of 1:50 in the experiments. The decrease in chemotaxis was expressed as the percentage decrease in migration after addition of the neutralizing antibody ([before antibody-after antibody] x 100 / [before antibody]). IL-8 ELISA: Supernatants from monocytes stimulated with ET-1 for different time intervals, were analysed for IL-8 content by interleukin-8 ELISA (Quantikine kits, British Biotechnology). A NovaPath TM Mini Reader (BioRad, Richmond, CA, USA) was used in reading of the microplates at OD45o.
Effect of lipoxygenase inhibitors on release of NCA and MCA The lipoxygenase inhibitors nordihydroguaiaretic acid (NDGA, 50 l.tM-400 lttM, Sigma) or diethylcarbamazine (DEC, 0.62 nM-10 nM, Sigma) was added to the monocyte cultures 5 min before ET-1. The cell cultures were incubated at 37C for different time intervals as described above, and the supernatants were collected and examined for NCA and MCA. The effect on NCA was examined after 1, 4, 8 and 24 h incubation, while the effect on MCA was examined after 24 h incubation. NDGA or DEC did not affect monocyte viability as tested by Trypan blue exclusion. Statistics: The difference between groups was tested for significance using a Student's t-test, or one-way ANOVA supplied with Scheffe's test. 13 In all cases, a

Results
Release of neutrophil (NCA) and monocyte chemotactic activity (MCA) from ET-1 stimulated monocytes Monocytes released NCA in response to ET-1 in a biphasic manner as shown in Fig. 1A. The maximum response was measured at 0.5 nM ET-1. This was comparable to the NCA seen in response to FMLP 10 -7 M. ET-1 concentrations lower than 0.01 nM gave no significant increase in NCA. Background migration, in response to RPMI-1640 was 6.2 +_ 3.2 neutrophils/HPF. The time course study (Fig. 1B) showed that the release of NCA was significant by 1 h after addition of ET-1, with a maximal response measured after 4 h, and a further increase after 24 h stimulation. Monocytes released MCA in response to ET-1 in a dose dependent manner as shown in Fig. 2A Immunoadsorption of monocyte and neutrophil chemotactic activity: Incubation with an antibody against IL-8 (dilution 1/100) reduced NCA in the supernatants by > 75% (Fig. 3). The migration after addition of antibody was not significantly different from controls incubated with medium only; 22_+ 3 cells/HPF. Higher antibody concentrations did not increase the inhibition of NCA, while lower antibody concentrations showed a significant reduction in inhibitory effect (data not shown). Release of IL-8 from ET-1 stimulated monocytes was further confirmed using IL-8 ELISA as shown in Table 1.
Incubation with a monoclonal antibody against MCP-1 (dilution 1/200) reduced MCA by > 80% (Fig.  4). The migration after addition of antibody was not significantly different from controls incubated with Endothelin-1 stimulates monocytes to release chemotactic activity Effect of lipoxygenase inhibitors on release of monocyteand neutrophil chemotactic activity: ET-1 has been shown previously to activate the arachidonic acid cascade system in macrophages. 14 Therefore we wanted to determine the possible contribution of lipoxygenase pathway for the ET-1 induced release of NCA and MCA. As shown in Table 2 NCA and MCA were significantly reduced when the monocytes were incubated with one of the lipoxygenase inhibitors, NDGA or DEC, before ET-1 was added.
The reduction was dependent on the dose of lipoxygenase inhibitor used, and was independent of incubation time (data not shown). NDGA and DEC alone did not show any NCA or MCA.

Discussion
In the present study it is shown that ET-1 functions as a signal molecule for monocytes causing release of chemotactic factors for neutrophils and monocytes. The dose-response pattern for NCA with a maximal response at an ET-1 concentration of 10-10 M is similar to the dose-response pattern previously observed for release of TNFa, IL-lb and IL-6 from monocytes and macrophages. Viability tests, however, excluded the possibility that higher concentrations of ET-1 had a toxic effect on the cells. The striking difference in dose-response and the time course for NCA and MCA indicated that different chemotactic factors for neutrophils and monocytes were produced in response to ET-1. This was confirmed in immunoadsorption studies showing that specific antibodies against IL-8 neutralized the neutrophil chemotactic activity, while monoclonal antibodies against MCP-1 neutralized the monocyte chemotactic activity. Although these antibodies neutralized MCA and NCA by > 80% and > 75% respectively, we cannot exclude the possibility that other chemotactic factors are released after stimulation of monocytes with ET-1.
A significant increase in IL-8 secretion was demonstrated at all time-points measured. This time course, with rapidly induced and long-lasting release of IL-8, is in accordance with previous observations on IL-8 release from activated mononuclear phagocytes. 6,15a6 Although monocytes are determined to be the predominant producing cells for IL-8, vascular endothelial cells have also been shown to produce IL-8 when stimulated with lipopolysaccharide or TNF and IL-1. In the present study no NCA is released from endothelial cells in response to ET-1, indicating a disparate regulation of IL-8 release from monocytes and endothelial cells.
The monocyte chemotactic activity, which was neutralized with monoclonal antibodies against MCP-1, was observed in the supernatants only after 24 h stimulation with ET-1. This is in accordance with previous time course studies on MCP-1 release from activated monocytes, showing a maximal response in MCP-1 production after 24 h. 1,17 MCP-1 was recently shown to be responsible for the majority of the chemotactic activity released from vascular endothelium. TM However, the present results show that ET-1 causes no release of monocyte chemotactic activity from endothelial cells.
Preincubation of the monocytes with the lipoxygenase inhibitors NDGA and DEC, inhibited the release of both monocyte and neutrophil chemotactic activity. ET-1 has previously been reported not to release LTB from monocytes, 19 while ET-1 stimulation of alveolar macrophages causes release of arachidonic acid. 14 Evidence is increasing that the lipoxygenase pathway and lipoxygenase products such as LWB might play a role of feedback regulators or intracellular messengers for production of cytokines, including IL-8. , 21 Thus the present results indicate that an intact lipoxygenase pathway is crucial for the ET-1 induced release of IL-8 and MCP-1.
ET-1 has previously been shown to cause a leukocyte dependent increase in microvascular permeability in isolated rat lungs. = The mechanism for the ET-1 induced lung injury might be explained by a synergistic action of TNFx, IL-I[ and IL-8 from monocytes and tissue macrophages, causing neutrophil recruitment and increased microvascular permeability. 7,1 Patients with atherosclerosis have increased circulating levels of ET-1, which exhibit a positive correlation with the extension of disease. 23 MCP-1 and IL-8 have also been detected in early vascular lesions in atherosclerosis, predominantly localized to tissue macrophages. 2<2 On this background we suggest that released ET-1 at predilection sites for atherosclerosis, '3,26 might stimulate mono-cytes and/ or tissue macrophages in the vascular wall to release MCP-1 and IL-8, and thus contribute to recruitment of inflammatory cells into the vessel wall causing a chronic inflammatory condition and finally atherosclerosis. 18,23,24 To conclude, ET-1 stimulation of human monocytes in vitro caused release of monocyte and neutrophil chemotactic activity provided that an intact lipoxygenase pathway was present. The neutrophil-and monocyte chemotactic activity were identified as IL-8 and MCP-1, respectively.