IL-1β Scavenging by the Type II IL-1 Decoy Receptor in Human Neutrophils

Reagents

All chemicals were supplied by Merck (Darmstadt, Germany) unless otherwise stated. NaCl (0.9%) and distilled water (Bieffe, Bergamo, Italy), Ficoll (Seromed-Biochem, Berlin, Germany), Percoll (Pharmacia, Uppsala, Sweden), RPMI 1640 (Life Technologies, Glasgow, Scotland), FCS (HyClone Laboratories, Logan, UT), glycine (Bio-Rad, Richmond, CA), IL-1β (Domp Research Centre, L’Aquila, Italy), ¹²⁵I-IL-1β (NEN, Bad Homberg, Germany), BSA (Sigma-Aldrich, St. Louis, MO), cyclohexamide (Sigma-Aldrich), M4 and M22 neutralizing Abs against type I and type II IL-1R, respectively (kind gift from Dr. J. Sims, Amgen, Seattle, WA), and sodium azide (Sigma-Aldrich).

Medium used routinely was RPMI 1640-10% heat-inactivated FCS supplemented with 2 mM l-glutamine (Life Technologies), 20 mM HEPES (Merck), 50 U/ml penicillin, and 50 mg streptomycin (Bristol-Myers Squibb, Sermoneta, Italy). For routine culture of HaCAT811 keratinocytes, DMEM-10% heat inactivated FCS was supplemented with Geneticin G418 (Life Technologies) at a final concentration of 500 μg/ml.

Cells

Human PMNs were separated from peripheral blood of healthy donors using a two-step buoyant density centrifugation as previously described (25). Briefly, buffy coats were centrifuged over a Ficoll gradient. PMNs collected from the pellet were then underlayed and centrifuged with 62% Percoll-RPMI 1640-10% FCS. PMNs of >95% purity, determined by morphology, were resuspended in binding buffer (RPMI 1640-1% BSA, pH 7.4).

The human keratinocyte cell line HaCAT811 was kindly supplied by D. Boraschi (Domp Research Centre). The human keratinocyte cell line HaCAT, which does not express wild-type IL-1RII, was stably transfected with IL-1RII cDNA (26). This cell line was maintained in DMEM supplemented with 2 mM l-glutamine, HIFCS, and 500 μg/ml geneticin (used originally to select out transfected cells). Cells were grown to confluency before passage or experimental use.

IL-1 binding and internalization assay

Briefly, 4 × 10⁶ PMN or 1 × 10⁶ HaCATs were incubated with 1 nM ¹²⁵I-IL-1β in the presence or absence of a 200-fold excess of IL-1β in binding buffer at 4°C for 2 h in a final volume of 50 μl. Labeled cells were washed twice, resuspended in binding buffer, and incubated at 37°C. Specific binding was calculated by subtracting nonspecific binding, determined in the presence of excess unlabeled ligand, from total binding.

The distribution of surface-bound and intracellular ¹²⁵I-IL-1β was determined by washing the cells for 5 min on ice with 50 mM glycine-0.8% NaCl/HCl (pH 3) (285 Osmol) in a volume of 75 μl per 4 × 10⁶ PMNs or 75 μl/1 × 10⁶ HaCAT811s. To separate bound from free cytokine, cells were centrifuged over sucrose gradient (20% sucrose, 1% BSA) and the activity of the pellets and the supernatants was determined by a gamma counter.

Determination of degraded and intact ¹²⁵I-IL-1β

Degradation status of the IL-1 released into the cell supernatant was determined by trichloroacetic acid (TCA) solubility. Media removed from the cells during the incubation at 37°C were precipitated in a final concentration of 10% TCA on ice for 15 min. These were then centrifuged at 14,000 rpm for 10 min and the soluble (completely degraded IL-1) and precipitate-associated (intact or partially intact IL-1) ¹²⁵I-IL-1β activity was determined.

Immunofluorescence microscopy

Medium was removed from HaCAT811 cells (seeded at 2 × 10⁵ cells/ml in 10-cm-diameter petri dishes and grown to subconfluency), replaced with medium containing 1 nM IL-1β, and cells were incubated for 10 min. Control cells incubated in medium alone were done in parallel. Cells were fixed for 20 min at room temperature with 4% paraformaldehyde/0.25% glutaraldehyde in 125 mM phosphate buffer (pH 7.4). Cells were scraped and collected by centrifugation. Pellets were washed with the phosphate buffer, infiltrated with concentrated sucrose, and frozen in a 3:1 (v/v) mixture of propane and cyclopentane cooled with liquid nitrogen. Sections (∼1-μm thick) were prepared using a cryosection apparatus (RMC; Boeckeler Instruments, Tucson, AZ), placed on glass slides, treated briefly with 0.1 M glycine in PBS (pH 7.4) followed by 0.3% Triton X-100, 15% filtered goat serum, 0.45 M NaCl, and 10 mM phosphate buffer (pH 7.4). Sections were incubated overnight at 4°C with mAb early endosome Ag 1 (EEA-1; BD Transduction Laboratories, Lexington, KY) and anti-IL-1RII M22 (see above), washed, and incubated for 1 h at room temperature with the appropriate fluorescein (FITC) and rhodamine (tetramethylrhodamine isothiocyanate) conjugate secondary Abs (Jackson ImmunoResearch, West Grove, PA). Following a final wash, sections were mounted with glycerol. Confocal microscopy was carried out using a Radiance 2100 microscope (Bio-Rad) equipped with a kripon/argon laser. Noise reduction was achieved by kalman filtering during acquisition.

For the endocytosis assay, subconfluent cells grown on coverslips were incubated for 1 h on ice with FITC-wheat germ agglutinin (WGA conjugates; Molecular Probes, Eugene, OR) in PBS containing 0.1 mM Ca²⁺, 1 mM Mg²⁺, and 1% BSA. The cells were washed and incubated at 37°C in the absence and presence of 1 nM IL-1β. Cells were washed and fixed in 4% paraformaldehyde. The coverslips were washed again and the above protocol of immunofluorescence was carried out using anti-human IL-1RII Ab 8.5 (27) (a kind gift from G. Peri, Istituto Mario Negri, Milan, Italy) and rhodamine (tetramethylrhodamine isothiocyanate)-conjugated secondary Ab.

IL-1β internalization and scavenging in human neutrophils

To directly measure the ability of freshly isolated human PMNs to sequester active IL-1 from the extracellular environment, we monitored the disappearance over time of radiolabeled IL-1β from medium removed from the cells. Radioactivity within the medium was found to decrease rapidly within the first 10- to 20-min incubation with the PMNs at 37°C indicating scavenging of radioactive IL-1 by the cells (Fig. 1⇓A). We calculated that upon incubation with PMNs, 2 × 10⁶ IL-1β molecules/cell were sequestered over a period of 30 min. This was followed by a plateau indicating a state of equilibrium in which further uptake of IL-1 is balanced by release of intact and degraded IL-1 into the medium.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

A, Disappearance of ¹²⁵I-IL-1 from medium removed from the cells. PMNs (5 × 10⁵ cells/50 μl) were incubated with ¹²⁵I-IL-1β for the indicated periods of time. Medium was then removed from the cells and cpm were determined by a gamma counter. Total radioactivity added was measured by adding 3 × 10⁻¹² mol of ¹²⁵I-IL-1β to 50 μl of medium in the absence of cells. Results are expressed as number of molecules × 10¹¹ of ¹²⁵I-IL-1β within the medium. B, Internalization of IL-1β in PMNs. Human PMNs (4 × 10⁶ cells/50 μl) were incubated with 1 nM ¹²⁵I-IL-1β in the presence or absence of a 200-fold excess of cold IL-1β in binding buffer at 4°C for 2 h. Labeled cells were washed, resuspended in binding buffer, and incubated at 37°C for the indicated time periods. Cells were transferred to 4°C, the medium was removed, and the cell pellet was acid washed to remove surface-bound ¹²⁵I-IL-1β. The distribution of surface-bound and intracellular ¹²⁵I-IL-1β was determined by separating acid wash removable and cell pellet-associated counts by sucrose gradient. Medium removed from the cells was analyzed by TCA precipitation to determine the degradation status of the IL-1β released. The radioactive counts of the cell pellets (internalized IL-1), the acid wash (surface-associated IL-1), TCA-precipitatable (intact or partially degraded IL-1), and TCA-soluble (degraded IL-1) fractions from the supernatants were determined by a gamma counter. Data are presented as a percentage of specific bound ¹²⁵I-IL-1β on the PMN surface after 2 h of binding and are representative of five experiments.

To further characterize the fate of IL-1β bound to IL-1 surface receptors, PMNs were incubated with saturating amounts of ¹²⁵I-IL-1 at 4°C in the absence and presence of a 200-fold excess of unlabeled cytokine. Unbound cytokine was then removed and the IL-1β-bound cells were incubated at 37°C in fresh medium to allow receptor internalization. At various time points, aliquots were removed, medium was collected, and cells were treated with acidic glycine buffer. This acid wash removes surface bound IL-1 without affecting the internalized cytokine. Preliminary experiments carried out at 4°C showed that the acid wash was 85–95% effective in removal of cell surface-associated ligand (results not shown). The levels of radioactivity contained in the supernatant, acid wash, and cell pellet, representing released, surface-bound, and internalized ligand, respectively, were determined for each time point. Fig. 1⇑B illustrates the rate of internalization of ¹²⁵I-IL-1β into human PMNs. The IL-1-IL-1R complex was rapidly internalized with 50–60% of radioactivity associated with the pellet after 1 h. The percentage of receptor within the cell remained relatively unchanged up to 4 h of incubation. Over the same 60-min period, cell surface-bound cytokine decreased to 25% and remained constant thereafter. This was accompanied by a concomitant accumulation within the medium of both TCA-precipitatable and nonprecipitatable radioactive fractions. The nonprecipitatable fraction represents fully degraded IL-1, whereas the precipitatable fraction represents both intact and partially degraded molecules which are indistinguishable from each other using the TCA method. Indeed SDS-PAGE analysis of the medium (results not shown) revealed bands representing partially degraded IL-1 in addition to intact and fully degraded IL-1. Thus, much of the IL-1 released back into the medium following internalization has been rendered inactive within the cell.

Internalization of ¹²⁵I-IL-1β is mediated by the IL-1 type II receptor

Previous studies have demonstrated that human PMNs express >90% IL-1RII (28). This would suggest that the IL-1 internalization observed in these cells is primarily mediated by the IL-1RII. To determine the exact contributions of the type I and the type II receptors to internalization, binding and internalization experiments were carried out in the presence of blocking mAbs against the IL-1RI or IL-1RII. Anti-type I Ab (M4) had no effect on binding and internalization, whereas the anti-type II Ab (M22) inhibited both binding and internalization by up to 98% (Fig. 2⇓). These data indicate that the IL-1 type I receptor, and thus signaling by IL-1, is not required to mediate internalization of the type II receptor.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Internalization of ¹²⁵I-IL-1β by PMNs is mediated by the IL-1 type II receptor. Binding and internalization were carried out as described previously in the absence and presence of blocking Abs to IL-1RI (M4), IL-1RII (M22), or an irrelevant IgG2b control at 10 μg/ml. Following 1-h internalization, cells were removed and subjected to an acid wash and the percentage of internalized IL-1 was determined. Data are expressed as the percentage inhibition relative to untreated cells and is representative of three independent experiments. Statistical analysis was carried out using the student t test with ∗∗∗ representing a p < 0.01.

Effects of different agents upon the internalization of IL-1RII

To examine the underlying mechanism of IL-1R-mediated internalization of ligand, we tested the effects of a range of known inhibitors of receptor endocytosis on IL-1 internalization in human PMNs. Phenylarsine oxide, a general inhibitor of receptor-mediated endocytosis, inhibited internalization in the PMNs by 70% (Fig. 3⇓). The metabolic blocker sodium azide inhibited IL-1 uptake by 84%, indicating the involvement of ATP synthesis in the internalization process. Monodansylcadaverine, a potent inhibitor of transglutaminases, is thought to block clathrin-mediated endocytosis at the receptor-invagination step. Pretreatment of the PMNs with this agent caused 45% inhibition of internalization, indicating the possible involvement of clathrin-mediated mechanisms in IL-1 internalization in these cells. This is corroborated by results of immunofluorescent microscopic analysis (see below), which show colocalization of IL-RII with the early endosome compartments known to be downstream during clathrin-mediated endocytosis. Lysosomotropic agents (e.g., ammonium chloride, chloroquine, methylamine hydrochloride, and, an inhibitor of actin assembly, dihydrocytochalasin B) all failed to inhibit receptor-mediated IL-1 internalization at 37°C (Fig. 3⇓).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Effects of different agents upon the internalization of IL-1RII. Briefly, 4 × 10⁶ PMNs were incubated for 30 min at 37°C in binding buffer in the presence of inhibitors at the following concentrations: phenylarsine oxide (50 μM), monodansylcadaverine (100 μM), sodium azide (0.1%), ammonium chloride (20 mM), chloroquine (100 μM), methylamine hydrochloride (10 mM), and dihydrocytochalasin B (10 μM). These concentrations were maintained throughout the course of the experiment. A control without addition of any agent was carried out in parallel. Saturation binding followed by internalization was carried out as described previously. Results are pellet-associated radioactivity in the presence of inhibitors expressed as percentage of control.

Recycling of the IL-1R after internalization

PMNs were first incubated with saturating amounts of unlabeled IL-1β at 37°C to allow internalization of IL-1RII. Then unbound ligand was removed and the cells were further incubated at 37°C for various periods of time to evaluate the reappearance of the receptor at the cell surface. At each time point, cells were washed and subjected to binding with ¹²⁵I-IL-1β at 4°C. As shown in Fig. 4⇓, IL-1 binding gradually increased with time, peaking at 15 min and decreasing again possibly due to another cycle of receptor internalization. The IL-1 binding capacity of PMNs re-expressing IL-RII was found to be approximately 60–70% of that of the same cell population before the internalization process. Since reappearance of the protein could be due to either de novo synthesized molecules and/or re-expression of the internalized receptors, we carried out this experiment in the absence and presence of the protein inhibitor cycloheximide. The re-expression was unaffected by the presence of this agent, suggesting that the receptors reappearing at the surface are preformed and de novo protein synthesis is not required. Since the IL-1RII does not signal, functional properties of this molecule other than binding could not be measured.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Recycling of the IL-1RII after internalization. Human PMNs (4 × 10⁶ cells) were incubated for 2 h at 4°C in binding buffer containing 10 nM unlabeled IL-1β in the absence (○) and presence (•) of 10 μg/ml cyclohexamide and then transferred to 37°C for 1 h. Cells were acid washed, resuspended in binding buffer, and further incubated at 37°C for the indicated times. Cells were then subjected to binding with ¹²⁵I-IL-1β at 4°C in the absence and presence of 200-fold excess of IL-1β. The pellet-associated activity was then determined by a gamma counter. Cyclohexamide concentration was kept constant throughout the course of the experiment. Data are representative of three independent experiments.

Internalization of IL-1RII in different cellular contexts and localization in early endosomes

The type II decoy receptor is the main IL-1-binding molecule expressed in monocytes (29); therefore, the study was extended to include this cell type. Levels of internalization in freshly isolated human monocytes were low, with a maximum of 10–25% of radioactivity detectable within the cell pellet after a 2-h incubation at 37°C (results not shown). No internalization was detectable in the monocyte cell lines THP-1 or MonoMac (results not shown). In addition, we tested other cell types of immune origin such as macrophages and dendritic cells (mature and immature) in which little or no internalization was detectable. In contrast, when we carried out the same experiment with the human keratinocyte cell line HaCAT811, significant levels of internalization were observed. This line was stably transfected to overexpress the gene encoding IL-1RII, ensuring high levels of decoy receptor surface expression. Similar to results in the PMNs, this cell type also rapidly internalized the ligand receptor complex, peaking at 42% internalization after 1.5 h incubation (Fig. 5⇓A).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Internalization of IL-1RII in the IL-1RII-transfected human keratinocyte cell line HaCAT811. A, HaCAT811 cells (1 × 10⁶/50 μl) were incubated with 1 nM ¹²⁵I-IL-1β in the presence or absence of a 200-fold excess of cold IL-1β in binding buffer (DMEM-0.1% BSA, pH 7.4) at 4°C for 4 h. Labeled cells were washed, resuspended in binding buffer, and incubated at 37°C for the indicated time periods. Cells were transferred to 4°C, the medium was removed, and the cell pellet was acid washed to remove surface-bound ¹²⁵I-IL-1β. Cell-associated and acid-removable counts were separated by sucrose gradient. Medium removed from the cells was analyzed by TCA precipitation to determine the degradation status of the IL-1β released. The radioactive counts of the cell pellets (internalized IL-1), the acid wash (surface-associated IL-1), TCA-precipitatable (intact or partially degraded IL-1), and TCA-soluble (degraded IL-1) fractions from the supernatants were determined by a gamma counter. Data are presented as a percentage of specific bound ¹²⁵I-IL-1β on the PMN surface after 2 h of binding and are representative of five independent experiments. B, Colocalization with WGA: Subconfluent HaCAT811 cells were labeled with FITC-WGA (a marker for membrane transport) at 0°C and incubated at 37°C in the absence (−IL-1β) and presence (+IL-1β) of 1 nM IL-1β before fixation and staining with anti-human IL-1RII Ab. In the presence of IL-1β the confocal image shows colocalization (yellow) of IL-1RII (red) with the internalized labeled glycoproteins (green), which is not apparent in untreated cells. Colocalization with EEA-1: HaCAT811 cells incubated at 37°C in the absence (−IL-1β) and presence (+IL-1β) of 1 nM IL-1β were fixed, collected, and permeabilized. Cells were incubated with anti-EEA-1 (a marker for early endosomes) Ab and anti-IL-1RII Ab. The presence of the yellow staining in cells treated with IL-1β shows colocalisation of IL-RII with EEA-1 which indicates the presence of IL-1RII within the early endosomes of the cell.

The process of IL-1R internalization was also examined immunohistochemically. The HaCAT811 cell line was used in preference to the PMNs to ensure sufficiently high, consistent levels of IL-1RII expressed at the cell surface, albeit this also resulted in highly elevated levels of receptor detectable within the cytoplasm in transit to cell membrane (Fig. 5⇑B). Following a 10-min incubation with IL-1β, IL-1RII receptors were found to colocalize with internalized WGA-labeled surface glycoproteins, indicating the involvement of receptor-mediated endocytosis during internalization of IL-1β in these cells (30, 31). At later time points (results not shown), similar colocalization was also observed occurring in some cells even in the absence of IL-1β. Furthermore, IL-1RII colocalized extensively with EEA-1 (an established marker of early endosomes (32)) in IL-1β-treated cells when compared to controls. Early endosomes are the first cellular compartments into which endocytosed molecules pass during the process of endocytosis (33). Therefore association with these compartments further corroborates the finding that the IL-1-IL-1RII complex is readily internalized by endocytosis into HaCAT811s upon exposure to IL-1.