Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The allergic inflammatory cascade is characterized by both an early and late phase, followed by the resolution of inflammation. The early response is initiated when mast cells release preformed mediators from their granules such as histamine, tryptase, and chymase (1). In addition, activated mast cells generate prostaglandin D₂ and leukotriene C₄, as well as other mediators known to cause bronchoconstriction, mucous secretion, and edema formation. The role of the mast cell in potentiating the late allergic response has been in part attributed to mast cell-dependent secretion of proinflammatory cytokines and chemokines (1).

Allergic inflammation tends to be self-limited. In this regard, there is evidence that mast cells produce several mediators that are considered antiinflammatory, and through such mediators may contribute to the resolution of inflammation. Heparin is one such mediator that is found in mast-cell granules. It binds to antithrombin, inhibits clot formation and kinin generation (2), and may help neutralize the detrimental effects of basic proteins released by eosinophils (3). Additionally, both interleukin (IL)-10 and IL-13 are released by mast cells and exhibit antiinflammatory actions (4, 5). IL-10 is active in downregulating the T helper 1 response as well as suppressing the activity of macrophages (6). IL-13 reduces the production of IL-1, IL-6, tumor necrosis factor (TNF)-, and other cytokines by activated macrophages (7).

In a search for other antiinflammatory molecules that might be generated and released by mast cells, we chose to examine human mast cells (huMCs) for the production of IL-1 receptor antagonist (IL-1ra). This is in part because the IL-1 family of cytokines is known to be involved in allergic inflammation. IL-1 is thus present in samples taken from the sites of ragweed-challenged skin chambers (8, 9), nasal secretions (10), and asthmatic bronchial epithelium (11).

As will be shown, both CD34⁺-derived cultured huMCs and human lung mast cells produce IL-1ra which is released after aggregation of high-affinity Fc receptor for immunoglobulin (Ig) E (FcRI). Further, the 17-kD form of IL-1ra is secreted by huMCs and can be measured in bronchoalveolar lavage (BAL) fluid (BALF) after antigen challenge.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Cultures

Human peripheral blood CD34⁺ progenitor cells were obtained and processed, after informed consent, as described elsewhere (12), and placed in huMC culture medium consisting of StemPro-34 SFM (Life Technologies, Grand Island, NY) supplemented with 2 mM L-glutamine, 100 µg/ml streptomycin, and 100 IU/ml penicillin, supplemented with 100 ng/ml recombinant human (rh) stem-cell factor, 100 ng/ml rhIL-6, and 30 ng/ml rhIL-3 (first week only) (PeproTech, Rocky Hill, NJ) (12). Half of the culture media was replaced every week. Mast-cell percentages were assessed by Kimura staining (15). More than 95% of the cells were identified as mast cells 8 to 10 wk after the initiation of the culture (12). To remove contaminating monocytes/macrophages, cultured cells were incubated in a culture dish (35 × 10 mm) overnight and nonadherent cells were harvested. The final purity of huMCs was greater than 99%.

A549 cells, a type II-like human lung epithelial cell line, were purchased from the American Type Culture Collection (ATCC, Rockville, MD) and grown in 75-cm² tissue-culture flasks. Cultures were maintained in epithelial cell medium consisting of Kaighn's modified Ham's F12 medium (ATCC) supplemented with 10% heat-inactivated fetal calf serum (FCS) at 37°C with 5% CO₂. A549 cells were passaged at 80 to 100% confluence.

HMC-1 cells were maintained in Iscove's modified Dulbecco's medium (Biofluids, Rockville, MD) containing 1.2 mM -thioglycerol, 4 mM L-glutamine, 100 µg/ml streptomycin, 100 IU/ml penicillin, and 10% heat-inactivated FCS. The cells were seeded at 2 × 10⁴ cells/ml density and recultured every 4 d.

Purification of Human Lung Mast Cells

Normal human lung that was surgically resected was obtained from the National Disease Research Interchange (Philadelphia, PA). Lung mast cells were dispersed from chopped lung specimens by an enzymatic procedure, and were purified by magnetic bead affinity selection using the anti-Kit monoclonal antibody (mAb) YB5.B8 (BD Pharmingen, San Diego, CA) as described (16). Mast-cell percentages and numbers were assessed by counting using a Neubauer hemocytometer after metachromatic staining with the Kimura stain (15). The final purity of the human lung mast cells was greater than 98%.

Cell Activation

For high-affinity IgE receptor-dependent activation, 1 × 10⁶ huMCs were incubated with anti-4-hydroxy-3-nitrophenylacetyl (NP)-IgE (1 µg/ml) (Serotec, Raleigh, NC) for 16 h. Cells were then centrifuged at 500 × g for 10 min, the supernatant was removed, and cells were resuspended in human mast-cell culture medium. A total of 1 × 10⁶ cells in 1 ml of huMC culture medium were activated by the addition of 100 ng/ml NP-bovine serum albumin (BSA) (Biosearch Technology, Inc., Novoto, CA) for 0, 2, 4, 8, and 20 h. Cells were then centrifuged at 500 × g for 10 min, and the supernatant was removed and stored at 80°C. The cell pellet was resuspended in huMC culture medium, lysed with vigorous pipetting, and stored at 80°C.

Isolation of RNA and Ribonuclease Protection Assay

Total cellular RNA was isolated from 1 × 10⁶ huMCs with an RNeasy Mini Kit (Qiagen, Valencia, CA), according to the manufacturer's specifications. The purity of RNA was assessed on the basis of the A260/A280 ratio. A custom multiprobe template was assembled (BD Pharmingen) for the comparative analysis of the messenger RNAs (mRNAs) of IL-1, IL-1, IL-1ra, IL-1 receptor 1 (IL-1RI), and IL-1 receptor 2 (IL-1RII). Total RNA (1 µg/ sample) was applied and a ribonuclease protection assay (RPA) was performed following the manufacturer's recommendations (BD Pharmingen). Yeast transfer RNA (tRNA) (1 µg) was used as a negative control, and human control RNA (1 µg) (BD Pharmingen) was used as a positive control. The quantification of each cytokine was performed by measuring the relative expression of each cytokine with respect to the ribosomal protein L32 after each background was subtracted with the aid of ImageQuant 5.0 (Molecular Dynamics, Sunnyvale, CA).

Flow Cytometry

Fluorescence-activated cell sorter (FACS) analysis was performed as described elsewhere (17). Mast cells were fixed in phosphate-buffered saline (PBS) plus 0.5% paraformaldehyde for 15 min at room temperature, followed by two washing steps in PBS. Cells were then permeabilized in 0.5% saponin for 10 min at room temperature and washed in PBS. Intracellular staining with phycoerythrin-conjugated antihuman IL-1ra mAb (Becton Dickinson, San Jose, CA) diluted in PBS/0.1% BSA and 1% milk plus 0.5% saponin was performed for 30 min at 4°C. After washing, cell analysis was performed using a FACScalibur (Becton Dickinson) and CellQuest software (Becton Dickinson).

Western Blotting

The mast-cell supernatants and cell pellets were concentrated using Centricon filters (Millipore Corp., Bedford, MA). Gel electrophoresis supplies were obtained from Invitrogen (San Diego, CA). A total of 10 to 20 µl of the prepared samples were loaded onto 4 to 12% NuPage Tris-Bis gels, within NuPage (2-N-morpholino)ethanesulfonic acid-sodium dodecyl sulfate running buffer according to the manufacturer's protocol. The proteins were then transferred onto a Hybond-P: PVDF Membrane (Amersham Pharmacia Biotech, Piscataway, NJ). The membranes were washed twice with Tris-buffered saline (TBS) containing 0.05% Tween 20, then blocked for 1 h in TBS containing 3% fat-free milk (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and 0.05% Tween 20. The blocking buffer was removed and replaced with polyclonal goat anti-IL-1ra (R&D Systems, Minneapolis, MN) suspended in 4% BSA, and the membranes were incubated in this solution at room temperature for 1 h. The membranes were then rinsed four times for 5 min each time in TBS containing 0.05% Tween 20, and incubated in TBS containing 4% BSA, 0.05% Tween 20, and antigoat IgG horseradish peroxidase (1:10,000) (Vector, Burlingame, CA) at room temperature for 30 min. The membranes were washed once in TBS containing 0.05% Tween 20 for 15 min, followed by three washings for 5 min each time and then twice for 2 min each time in TBS without Tween 20. The bands were visualized using enhanced chemiluminescence plus a Western blot chemiluminescence kit (Amersham Pharmacia Biotech) according to the manufacturer's protocols.

Enzyme-Linked Immunosorbent Assay

IL-1ra in mast-cell supernatants and cell lysates were quantitated by an enzyme-linked immunosorbent assay (ELISA) kit for IL-1ra (sensitivity limit, 14 pg/ml) (R&D Systems). IL-8 in A549 cell supernatants was quantitated by an ELISA kit for IL-8 (sensitivity limit, 10 pg/ml) (R&D Systems).

Bioassay

A549 cells were plated in 96-well plates (Becton Dickinson) at a cell density of 2 × 10⁴ cells per well in epithelial cell medium and incubated for 2 d, and the culture supernate was removed and epithelial cell media were added. Plates were then incubated for 16 h with rhIL-1 (50 pg/ml) (R&D Systems) alone or in combination with either rhIL-1ra (R&D Systems) or activated huMC lysates. Before the addition of IL-1 and IL-1ra to the A549 cells, the cytokines were mixed and preincubated in epithelial cell media at 37°C, 5% CO₂, for 15 min. After 16 h, cell-free supernatants were collected and assayed for IL-8 by ELISA.

Immunohistochemistry

Human lung that was surgically resected was obtained from the National Disease Research Interchange. Lung specimens were fixed in ice-cooled 40% acetone before being embedded in paraffin. Sequential sections, 2 µm thick, were cut and placed on silanated slides. After deparaffination with xylene, the slides were rehydrated and then washed in TBS with 0.05% Tween 20 (twice for 5 min each time). The following antibodies were applied for 30 min at previously titrated optimal dilutions: goat antihuman IL-1ra (R&D Systems), AA1 to mast-cell tryptase (Dako Corp., Carpinteria, CA), or goat IgG or mouse IgG₁ control (Dako Corp.). Slides were then washed in TBS with 0.05% Tween 20 (twice for 5 min each time), labeled with a biotinylated second stage for 15 min, washed, and detected using the streptavidin-biotin peroxidase system (Dako Corp.). After further washing, the Fuchsin substrate system (Dako Corp.) was applied as the chromogen, giving a red reaction product, and the sections were counterstained with Mayers hematoxylin.

Bronchoscopy and Segmental Antigen Challenge

The protocol to obtain BALF (#95-1-0043) was approved by the NIAID IRB at the National Institutes of Health, Bethesda, MD. All subjects signed informed consent. Four subjects with a clinical history of allergen-induced asthma and 10 nonatopic control subjects without asthma took part in the study. All subjects with asthma had a methacholine (MCh) concentration causing a 20% drop (PC₂₀) in forced expiratory volume in 1 s (FEV₁) of less than 10 mg/ml, a skin test positive to one or more common allergens (oak, box elder, ragweed, elm, rye grass, Bermuda grass, and cat dander), and an FEV₁ greater than 55%. Subjects had no evidence of an upper respiratory infection in the 3 wk before being studied; had not used cromolyn sodium or inhaled, oral, or parenteral steroids in the 4 wk before the lavage; and had not taken theophylline or antihistamines within 1 wk of bronchoscopy. Subjects had received no immunotherapy in the last 2 yr, and had not smoked in the past 10 yr. The nonatopic, healthy control subjects had a lifelong absence of any symptoms indicative of allergic disease and had normal bronchial reactivity to MCh (PC₂₀ FEV₁ greater than 25 mg/ml).

All normal volunteers underwent bronchoscopy with BAL, and patients with asthma underwent segmental allergen challenge with lavage of the challenged site 24 h later. After topical anesthesia with lidocaine, a bronchoscope (Olympus BFP20; Olympus Corp., London, UK) was introduced transnasally and additional 1% lidocaine was administered. The bronchoscope was advanced into the right mainstem bronchus and then placed into the right middle lobe bronchus. Five 60-ml saline bronchial lavages were performed. For healthy control subjects, the bronchoscope was withdrawn and the subject monitored in the intensive care unit. For subjects with asthma, the bronchoscope was then advanced into the right lower lobe. Bronchial antigen challenge was performed as described (18), with the initial concentration of allergen based on the skin-test reactivity of each patient. If, after 2 min, bronchial edema was not observed, a 10-fold greater concentration of allergen was instilled, up to a maximum of three total allergen concentrations. After the procedure, the bronchoscope was removed and the patient monitored in the intensive care unit for 1 to 4 h. In the challenged asthmatic subjects, bronchoscopy was repeated 24 h after antigen challenge. The average recovery of BALF was 43% from normal control subjects and 39 and 34% from asthmatic subjects before and after challenge, respectively.

Processing of BALF

BALF was passed through two layers of sterile gauze to remove debris and then centrifuged at 500 × g for 10 min at 4°C. The cells were separated from the supernatant and the supernatant was kept at 80°C. Supernatants were assessed for IL-1ra by ELISA.

Statistical Analysis

All data are expressed as means ± standard error of the mean (SEM). Statistical significance of differences was performed using the two-tailed unpaired Student's t test. Differences were considered significant when the probability (P) was < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

IL-1ra mRNA in huMCs

To determine whether huMCs synthesize IL-1ra, we first examined huMCs cultured from CD34⁺ cells for the presence of IL-1ra mRNA before and after IgE-mediated activation. Figure 1A presents the time course of the IL-1 family of mRNAs produced by huMCs after FcRI aggregation. At rest, these mast cells demonstrate mRNA to IL-1ra and IL-1RI. There was minimal mRNA for IL-1, IL-1, and IL-1RII. After activation through FcRI, expression of mRNAs for IL-1, IL-1RI, and IL-1ra was upregulated. IL-1 was elevated by 2 h, whereas IL-1ra and IL-1RI increased compared with control by 4 h (Figures 1B-1D). mRNAs for IL-1, IL-1ra, and IL-RI returned to the level of control by 20 h. Thus, huMCs expressed mRNA for IL-1ra that was upregulated after activation of huMCs through FcRI.

View larger version (35K): [in this window] [in a new window]   Figure 1.   Time course of mRNA expression of IL-1 cytokines and receptors produced by cultured huMCs after aggregation of FcRI. Cells were exposed to anti-NP-IgE overnight, followed by the addition of antigen (+) to paired cell cultures. Total RNA was isolated at the indicated times and the RPA was performed. A representative result is shown (A). To establish the identity of each protected fragment, the known size and migration distance of the unprotected probe was used to prepare a standard curve. Human control RNA (BD Pharmingen) was used as a positive control and yeast tRNA was used as a negative control. Semiquantification of IL-1ra (B), IL-1RI (C), and IL-1 (D) expression in huMCs before (open bars) or after ( filled bars) aggregation of FcRI was performed by measuring the band density of the relative expression of each mRNA with respect to L32 mRNA after each background was subtracted. IL-1RII and IL-1 mRNAs were below detection levels. For IL-1ra and IL-1 the results were repeated in five separate experiments through 8 h using cells cultured from different donors. These data are presented as means ± SEM; *P < 0.05, **P < 0.005, ***P < 0.001 compared with 0 time. The 4-h error bar in activated cells in D was too small to diagram.

Release of IL-1ra Protein by huMCs

There has been no report that IL-1ra protein is produced by huMCs. We thus next determined whether resting cultured huMCs both contain IL-1ra protein and release IL-1ra protein after activation through IgE-dependent mechanisms. Using flow cytometry, we found that resting huMCs constitutively expressed intracellular IL-1ra (Figure 2A). After stimulating huMCs through FcRI, IL-1ra protein was detectable in the supernatant by ELISA (Figure 2B). Western blot analysis was then performed to identify the specific isoform of IL-1ra that is present in huMCs and in the huMC supernatant. As can be seen in Figure 2C, the 17-kD isoform was identified in both the huMC lysate and supernate. In contrast, monocytes constitutively make the 17-kD form but secrete a glycosylated 22-kD form (19). Lysate of the HMC-1 mast-cell line also demonstrates IL-1ras which appear to be of the 16- and 18-kD isoforms (19). Thus, huMCs contained and secreted the 17-kD isoform of IL-1ra.

View larger version (19K): [in this window] [in a new window]   Figure 2.   IL-1ra protein in huMCs and in supernates from huMCs after aggregation of FcRI. (A) After staining of permeabilized cells with a phycoerythrin-conjugated anti-IL-1ra mAb (thick line) or an isotype control mouse IgG1 (thin line), FACS analysis was performed to detect the intracellular expression of IL-1ra. (B) huMCs were incubated with anti-NP-IgE overnight at 37°C and washed, and NP-BSA was added to aggregate FcRI. Plates were then centrifuged and the supernates removed. IL-1ra was measured by ELISA in supernates of stimulated (filled bar) and unstimulated (open bar) cells at 20 h (n = 4; with cells for each experiment cultured from separate donors). (C) Western blot analysis of IL-1ra in huMC supernatants and lysates. ST, rhIL-1ra standard; Lys, concentrated lysate; Supp, concentrated supernate; Lys HMC-1, concentrated HMC-1 lysate. These samples were subjected to electrophoresis on 4 to 12% Bis-Tris gel under reducing conditions. After electrophoresis, proteins were blotted and incubated with goat antihuman IL-1ra.

Kinetics of IL-1ra Release

We next performed a time course of IL-1ra release from huMCs after FcRI-dependent activation. As can be seen, release of IL-1ra occurred between 8 and 20 h, with little increase in IL-1ra in supernatants thereafter (Figure 3). There also appeared to be a decrease in intracellular stores of IL-1ra at later time points. The total IL-1ra protein in the pellet and supernatant at 20 h was 13,570 pg, whereas the total IL-1ra obtained at time 0 was 7,671 pg. This increase in total IL-1ra protein is consistent with the observation of elevated mRNA for IL-1ra through 8 h (Figure 1). These results demonstrate that huMCs released the 17-kD form of IL-1ra over a period of 8 to 20 h after activation.

View larger version (24K): [in this window] [in a new window]   Figure 3.   Time course of IL-1ra release from cultured huMCs after FcRI aggregation. huMCs were incubated with anti-NP-IgE overnight at 37°C and washed, and NP-BSA was added to aggregate FcRI (time 0). Plates were then centrifuged at the indicated time points, the supernates removed, and cells lysed. IL-1ra was measured by ELISA in both huMC lysates (open bars) and supernates (filled bars) of mast cells (*P < 0.05, and **P < 0.005 compared with time 0). Data presented are the results of three experiments performed on huMCs cultured from a single donor.

huMC Lysates Inhibit IL-8 Production by Human Epithelial Cells

Having determined that IL-1ra protein is released by activated huMCs, the ability of huMC lysates to inhibit the action of IL-1 on human epithelial cells was investigated. Lysate was obtained from huMC cultures stimulated by FcRI aggregation for 20 h. A 17-kD form of rhIL-1ra was used as a control to inhibit the production of IL-8. IL-1 at 50 ng/ml was used to stimulate IL-8 production of A549 epithelial cells, on the basis of previous study (20). At this concentration there was 30 ng/ml of IL-8 production, compared with 0.5 ng/ml of IL-8 from nonstimulated epithelial cells. When the 17-kD form of IL-1ra was preincubated with IL-1, this rhIL-1ra reduced the production of IL-8 in a dose-dependent manner (Figure 4A). When mast-cell lysates were substituted for standard IL-1ra, a dose-dependent inhibition of IL-8 production was documented (Figure 4B), consistent with the presence of IL-1ra within the huMC lysates.

View larger version (33K): [in this window] [in a new window]   Figure 4.   Inhibition of IL-1-induced IL-8 production by A549 epithelial cells by IL-1ra or mast-cell lysates. A549 cells were incubated with IL-1 at a concentration of 50 ng/ml for 16 h in the presence or absence of either rhIL-1ra (A) or huMC lysate (B). Cell-free supernates were collected and assayed for IL-8 by ELISA. The increase was downregulated by (A) increasing concentrations of standard IL-1ra, and (B) increasing concentrations of huMC lysate.

IL-1ra in Human Lung Mast Cells

To extend the observations on IL-1ra in cultured huMCs to human lung mast cells, we examined human lung tissues for the presence of IL-1ra within lung mast cells by immunohistochemistry and documented the release of IL-1ra from isolated human lung mast cells after FcRI aggregation. IL-1ra was detected in cells that also stained positive for tryptase in two sequential 2-µm sections of a lung tissue (Figure 5A). In Figure 5B, the production of IL-1ra from isolated human lung mast cells after FcRI aggregation was measured by ELISA. A total of 10 ng of IL-1ra was produced by 10⁶ human lung mast cells, similar to the amount of IL-1ra produced by alveolar macrophages stimulated by lipopolysaccharide, phorbol myristate acetate, and adherence (21). Thus, human lung mast cells produced and secreted IL-1ra.

View larger version (104K): [in this window] [in a new window]   View larger version (9K): [in this window] [in a new window]   Figure 5.   IL-1ra in human lung mast cells. (A) Sequential 2-µm sections of a surgically resected lung immunostained for tryptase (i) and IL-1ra (ii), demonstrating colocalization of IL-1ra to tryptase+ mast cells (filled arrows). Open arrows indicate IL-1ra+ tryptase cells. (B) IL-1ra production from human lung mast cells (106 cells/ml) stimulated with FcRI aggregation. Purified human lung- derived mast cells were incubated with anti-NP-IgE overnight at 37°C and washed, and NP-BSA was added to aggregate FcRI. Plates were then centrifuged and the supernates removed. IL-1ra was measured by ELISA in supernates of stimulated (filled bar) and unstimulated (open bar) cells at 20 h (n = 2).

IL-1ra in Allergic Inflammation within the Human Lung

We next wished to confirm the presence of IL-1ra in human allergic inflammation by examining BALF from the lungs of patients undergoing segmental antigen challenge. Figures 6A and 6B show the effect of antigen challenge on the cell number of eosinophils and on the total number of cells, both of which were significantly increased after challenge (P < 0.05 or P < 0.005). This BALF was next examined for IL-1ra by ELISA (Figure 6C) and values were compared with levels found in normal lung. BALF from normal volunteers contained 0.35 ± 0.12 ng/ml IL-1ra (n = 10 subjects). The before-challenge BALF samples from allergic subjects contained 0.19 ± 0.04 ng/ml IL-1ra (n = 4 subjects). There was no statistical difference between these two levels. The after-challenge lavage samples from allergic subjects had 1.6 ± 0.57 ng/ml IL-1ra (n = 4 subjects), which is statistically increased when compared with IL-1ra levels in both normal BALF and before-challenge levels in atopic subjects (P < 0.05). Mast-cell tryptase was also measured in the BALF (Figure 6C). Although not statistically significant, there was a trend toward increasing levels of -tryptase in the after-challenge BALF. Figure 6D shows that the 17-kD isoform is prominent in BALF; in addition, there is a less intense band at 22 kD. The 17-kD isoform is consistent with the isoform released from huMCs after FcRI aggregation. The 22-kD isoform of IL-1ra is observed with lipopolysaccharide stimulation of alveolar macrophages (21). Thus, human lung mast cells produced IL-1ra, and the predominant form of IL-1ra found in lung after antigen challenge was 17 kD.

View larger version (27K): [in this window] [in a new window]   Figure 6.   IL-1ra levels and isoforms in BALF from segmental antigen challenge. Segmental antigen challenge was performed as described in MATERIALS AND METHODS, with (A) before- (Pre) and after- (Post) antigen challenge eosin- ophil number (** indicates a significant difference between eosinophil number in BALF after antigen challenge when compared with before challenge; P < 0.005; (B) before- and after-challenge total cell number (* indicates a significant difference between total cell number in BALF after antigen challenge when compared with before challenge; P < 0.05); and (C) IL-1ra ( filled bars) and tryptase (open bars) found in cell-free supernates from BALF. Normal subjects (Normal) were 10 patients who were nonatopic and nonasthmatic. There was a significant difference (*P < 0.05) between IL-1ra levels in BALF after antigen challenge when compared with before-challenge and normal levels. (D) Western blotting of BALF shows the 17-kD form as prominent in BALF with a minimal band at 22 kD. ST, rhIL-1ra 17-kD standard; BAL 1 and 2 are representative samples of after-challenge BALF.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The results of this study support the conclusions that huMCs have the ability to produce and secrete IL-1ra (Figures 1-3 and 5), and that the IL-1ra is bioactive (Figure 4). Mast cells also appear to contribute to the IL-1ra measured in BALF after segmental antigen challenge (Figure 6). This suggests that IgE-dependent activation of huMCs may contribute to the local control of IL-1-dependent inflammation through the release of IL-1ra.

Using the RPA, expression of mRNAs for IL-1, IL-1RI, and IL-1ra were shown to increase after activation of cultured huMCs through FcRI aggregation (Figure 1). As observed in alveolar macrophages (21), upregulation of IL-1 is followed by IL-1ra upregulation (Figure 1). In agreement with the time course of IL-1ra mRNA, the release of IL-1ra was essentially completed by 20 h (Figure 3). This time course is similar to that seen in the secretion of IL-1ra from monocytes (22) and alveolar macrophages (21). The decrease in intracellular stores of IL-1ra at later time points may be due to increased intracellular turnover, or to a decrease in steady-state IL-1ra production. The bioactivity of the huMC IL-1ra was confirmed using a bioassay (Figure 4).

Human lung mast cells were shown to stain for IL-1ra by immunohistochemistry (Figure 5A), and to release ~ 10 ng IL-1ra/10⁶ cells when FcRI was aggregated (Figure 5B). This amount of IL-1ra is similar to that secreted by alveolar macrophages when stimulated by adherence, phorbol myristate acetate, and lipopolysaccharide (21).

To examine the possible contribution of mast cells to IL-1ra in pulmonary inflammation, we first measured IL-1ra in BALF after segmental antigen challenge. Figures 6A and 6B show the degree of inflammation that resulted after challenge, and Figure 6C shows that IL-1ra is elevated in BALF after antigen challenge. Figure 6D shows that Western blotting of BALF indicates that some of the IL-1ra present after segmental antigen challenge is of the 17-kD isoform. The identification of the 17-kD isoform of IL-1ra in both the huMC lysate and supernate (Figure 2C) as well as in BALF (Figure 6D) suggests that lung mast cells contribute to IL-1ra levels in lung after antigen challenge. In this study, because we focused on the antiinflammatory aspect of mast cell function in allergic inflammation, we did not examine allergen challenge in the control subjects. In the lung, it has been reported that alveolar macrophages are a major source of IL-1ra and that they secrete a glycosylated 22-kD isoform of IL-1ra (21). Further, it has been reported that IL-1ra released in vitro from monocytes and neutrophils is of the 22-kD isoform (23). Human epithelial cells produce an intracellular form of IL-1ra that is not secreted (11).

IL-1ra mRNA has been reported in HMC-1 cells in a survey of cytokines produced by these cells (24). Protein levels were not measured; but when HMC-1 IL-1ra protein was examined, its molecular weight appeared to differ from the molecular weight of IL-1ra found in normal huMC (Figure 2C). Release of IL-1ra after IgE-mediated activation could not be studied in HMC-1 cells because they lack the FcRI receptor.

The ratio of IL-1 to IL-1ra may influence the course of inflammation. In normal individuals there is a 100-fold excess of IL-1ra to IL-1. When exogenous IL-1ra is administered in disease states such as endotoxic shock, sepsis, graft-versus-host disease, and rheumatoid arthritis, the ratio of IL-1ra to IL-1 must be considerably greater (~ 100,000-fold) to limit disease severity (25). In status asthmaticus the reported ratio is approximately 1,000-fold (26). Additional evidence for a role for the IL-1/IL-1ra ratio in allergic disease is suggested by studies that show elevated levels of IL-1 in nasal secretions throughout the ragweed pollen season. The same nasal fluid shows no IL-1ra in Weeks 1 to 3, indicating a possible imbalance that contributes to the progression of allergic inflammation (10). In a follow-up study to the findings that IL-1ra and IL-1 are present in airway bronchial epithelial cells, it was reported that treatment with inhaled steroids reduced IL-1 that was present but did not affect the amount of IL-1ra (27).

Several studies have demonstrated that treatment with IL-1ra may be useful in allergic disease. In the mouse model of allergic conjunctivitis, topical IL-1ra suppressed allergic eye inflammation by downregulating the recruitment of eosinophils and other inflammatory cells responsible for ocular inflammation (28). In a guinea-pig model of asthma, intravenous administration of IL-1ra before antigen exposure reduced the generation of a late-phase reaction as measured by pulmonary resistance and by the downregulation of the recruitment of eosinophils (29). A second study, also using a guinea-pig model, showed the benefit of aerosolized IL-1ra immediately before antigen challenge, with protection against bronchial hyperreactivity and pulmonary eosinophil accumulation (30). Thus, IL-1ra secreted by human lung mast cells after FcRI aggregation may be instrumental in limiting the initial inflammation that is induced by allergic mechanisms.

In addition to the role of IL-1 in allergic inflammation, IL-1 is a cytokine that stimulates production of other proinflammatory and profibrotic cytokines. IL-1 and TNF- have been implicated in the pathogenesis of fibrotic lung disease. The risk of fibrosing alveolitis increases with specific IL-1ra polymorphisms (31). There is similar evidence of a protective role for exogenous IL-1ra in a number of models of acute lung injury. IL-1ra partially reverses changes of pulmonary fibrosis and, to a lesser extent, BALF cellularity in bleomycin-induced fibrosis in mice (32).

In summary, we have shown that human cultured and lung-derived mast cells release IL-1ra. This cytokine, along with other antiinflammatory cytokines secreted by mast cells, such as IL-10 and IL-13, may be instrumental in limiting the inflammation that is induced by allergic mechanisms. In addition, mast cells have been associated with other disease states, such as fibrotic lung disease. The fact that mast cells secrete IL-1ra could, in some instances, provide a protective effect. Whether manipulation of endogenous production of IL-1ra by mast cells could be used as a strategy to treat IL-1-enhanced pulmonary inflammation remains to be determined.