Clinical and Experimental Allergy. 1992, Volume 22, pages 984-990

Anti-inflammatory efl^ects of nedocromil sodium: inhibition of alveolar macrophage function L. BORISH, J. WILLIAMS. S. JOHNSON. J. J. MASCALI, R. MILLER* and L. J. ROSENWASSER National Jewish Center for Immunology and Respiratory Medicine, University of Colorado Health Sciences Center Denver, Colorado. U.S.A. and*Fisons Corporation, Rochester, Ne\v York. U.S.A. Summary

The IgE-dependent activation of mononuclear phagocytic cells through their capacity to express low affinity IgH receptors (FceRII) has been proposed as a mechanism for the development of airways inflammation in allergic asthma. This Fcf:RII expression leads to the IgE-dependent production ofthe potent pro-inflammatory cytokines IL-I^ and TNF--y. Expression by monocytes of FcfiRII is regulated by several cytokines including interleukin-4, y- and :z-interferons. and granulocyte-macrophage and macrophage colony stimulating factors. An anti-inflammatory eflfect of nedocromil on monocytes has been proposed as a possible mechanism for its anti-asthma activity. We therefore investigated the capacity of nedocromil to modulate mononuclear phagocyte FccRII expression and cytokine production. We used an anti-FceRII antibody and flow cytometric analysis to assess the capacity of nedocromil to modulate cytokine-induced FcfiRII expression in normals and asthmatics. Monocytes, THP-1 monocyte leukaemia cells, and alveolar macrophages were exposed to varying concentrations of these cytokines for 48 hrat 37'C with or without the additional presence of nedocromil (1-10 /iM) and the per cent of monocytes expressing FceRIT was determined. No changes in Fcj^:RII expression were observed. Subsequently, we investigated the capacity of nedocromil to affect the capacity of IgE plus anti-IgE comple.xes, allergen, and LPS (16 hr/37 C) to stimulate IL-1^ and IL-6 production. No changes were observed when nedocromil was applied concomitant with the stimulus. However, pre-treatment (30 min) with nedocromil was associated with a 59 5 ± 5-6'^'ii inhibition of TL-6 production stimulated by allergen and 34-5 ± 5-1'' i. by anti-IgB. In conclusion, nedocromil does not modulate mononuclear phagocytic cell Fc^RII expression but does suppress IgEdependent cytokine production. This may represent an important anti-inflammatory action of nedocromil in the treatment of reactive obstructive airways disease. Clinical and Experimental Allergy, Vol, 22, pp. 9K4-990. Submitted 24 October 19Q1; revised 7 April 1992; accepted 13 April 1992. Introduction

Activation of alveolar macrophages (AM) has been proposed as a pathophysiological mechanism for the development of airways inflammation in asthma [1,2]. Such activation of AM may occur in an allergen-specific and IgE-dependent fashion through the capacity of CorrespundL-nci.', Dr L Borish. National Jewish Center for Immunology and Respiratory Medicine, University ofColorado Health Sciences Center, t400 Jackson Strccl, Denver, CO 80206. U,S.A, Supported by NIH grant At24S48 and a grant from the Fisons Corporation,

984

mononuciear phagocytes to express low affinity IgE receptors (FetRlI or CD23) [3-7]. Monocytes obtained from non-allergic individuals do not express CD23 [4,6,8.9]. However, asthma is associated with up-regulation of CD23 and the presence of surface bound igF on AM [4,9]. This stimulation of Fc/:RT1 may occur secondary to the allergen-induced release of several cytokines. including interleukin 4. y- or x-interferons, and both granulocyte-macrophage- and rnacrophage-colony stimulating factors [9-12], When activated, AM release numerous pro-inflammatory mediators including iysosomai enzymes, respiratory

Nedocrnmil modulates mononuclear phagocytic cell function

burst products, and eicosanoids [13 14]. Among the most potent inflammatory monocyte mediators are a series of newly generated peptides termed cytokines [15-21]. Cytokines produce inflammation through the chemotaxis and activation of granulocytes and inducing the expression of adhesion molecules on endotheiial cells. We have recently described the IgE-dependent production of interleukin ifH (IL-I/?) and tumour necrosis factor a (TNF-x) by monocytes [8]. Nedocromil has recently been introduced as an antiasthma medication with an anti-inflammatory efl'ect [22,23]. Thus, nedocromil modulates monocytic arachidonate metabolism, the respiratory burst and degranulation [24 26], We therefore investigated the capacity of nedocron:iil to modulate FcsRII expression and to mitigate IgE-dependent cytokine production. These studies were performed both to provide new understanding of the basis of activity of this drug and to lend credence to our hypothesis that these are important parameters of the pathophysiology of asthma. Methods Suhjects Asthmatics were obtained from the National Jewish Center's allergy and pulmonary clinics and their asthma was documented by the presence of a consistent medieal history and documentation of either an FEVi of < 80"o predicted on at least one occasion with > 1 5'',, improvement after administration of bronchodilator or via a positive methaeholine challenge. Asthma was controlled on j6-adrenergics alone and subjects were excluded from the study if they had received inhaled or systemic corticosteroids, theophylline, or cromolyn within 6 weeks. Normals were laboratory and clinic employees and had no history consistent with asthma or other atopic diseases. Subjects were excluded from the study if they had recent viral infections, Monocytes Preliminary studies were performed on monocytes and THP-1 cells. Cells of the monocyte leukaemia cell line THP-1 were obtained from American Type Culture Collection (Rockville, Maryland, U.S.A.). Cells were cultured in complete medium consisting of RPM I I MO medium supplemented with 1 niM HFPES buflcr (pH 7 20), 100 U/ml penicillin G, 100 //g ml streptomycin, 2 mM L'glutamine, and 10"/,. NuSerum (Collaborative Research, Bedford, Massachusetts, U.S.A.). Monocytes were obtained from hcparini/ed peripheral blood and isolated as described [27] via the sequential application of ficoll-hypaque centrifugation and plastic adherence. This

985

preparation contained >90% monocytes via Giemsa staining and fewer than 2% contaminating B lymphocytes via surface immunoglobulin staining and flow cytometry. Alveolar maerophages Alveolar macrophages were obtained via bronchoalveolar lavage. Subjects were admitted to the clinical study unit at the National Jewish Center, Subsegments of either the right middle lobe or iingula were lavaged with four 60 ml aliquots. as previously described [28]. The return from each aliquot was pooled. Alveolar macrophages (AM) were purified from BAL cell pellets via plastic adherence. The AM preparation was >95'Vn pure and contained ^ 4 - 5 % lymphocytes and < 1% other cell types (mostly eosinophils). FccRII induction Mononuclear phagocytic cells were allowed to interact with various cytokines. The choice and concentrations of cytokines used and incubation period were previously determined to give optimal stimulation of FCJ:RIT expression [9]. The cytokines were interleukin 4 (IL-4; 400 U; Genzyme, Boston, Massachusetts, U.S.A.). granulocyte-macrophage colony stimulating factor (GM-CSF; 100 U; Genzyme), Tnterferon-:r (IFN-x; 1000 U; Collaborative Research) and IFN-y (1000 U. Genzyme), Cocultures were performed in complete medium, in a 37 C/ 5"/n CO: atmosphere with or without the additional presence of nedocromil (0 1 10 fiM). Nedocromil was applied concomitant with or 30 min prior to the various cytokines. Surface expression ot" Fcj;RII was quantified via flow cytometric analysis 48 hr after co-culture with the various cytokines.

Monocytes or alveolar macrophages lU*'') wore stained with a commercially available biotin-conjugated monoclonal mouse anti-human Fc/:RII antibody (1:100 dilution; The Binding Site. San Diego. California, U,S,A,). Antibodies were diluted in Hank s bullered saline supplemented with bovine serum albumin (0-1 "A,; Sigma Chemicals, St Louis, Missouri, U.S.A.) and sodium azide (OOT'u; Sigma) (HBSS/BSA/a/ide), As B lymphocytes may constitutively express FCJ:RII [29 30], monocytes were additionally stained for the mononuclear phagocyte marker CDI4. Monocytes were allowed to simultaneousl\ interact with a commercially available fluorescein-conjugated monoclonal mouse anti-human CDI4 antibody {LeuM3; Becton Dickinson, Mtn View, Califor-

9S6

L. Borish

nia. L'.S.A.). An isotype matched fluorescein-conjugated control mouse antibody was used to determine nonspecitic binding (Becton-Dickinson). Alveolar macroph.iges were stained only with the anti-FceRIT antibody. Rinding was allowed to occur for 30 min at 4 C with continuous agitation. After washing twice in the HBSS BSA/azide, monocytes (and THP-1 cells) were allowed to interact with streptavidin-phycoerythrin (avidin-PE; Becton-Dickinson). FccRII on AM were labelled with streptavidin-allophycocyanin (avidin-APC; Becton Dickinson), APC was chosen as a fluor for the alveolar macrophages as under the conditions used to detect APC fluorescence, the AM demonstrate minimal auto-fluorescence [31]. For both of these experiments, binding ofthe relevant avidin conjugate in the absence of primary antibody was performed to determine background fluorescence. After 30 min at 4 C, cells were again washed twice in HBSS/BSA/azide and finally resuspended in fixative (2-5':n paraformaldehyde; Sigma). Dual immunofluorescence was performed on monocytes on a Coulter Epics Profile fluorescence-activated cell sorter and data were collected only for cells which stained with the LeuM3. Alveolar macrophages were analysed on a Coulter EPTCS multiparameter system-751 flow cytometer with the flow cytometer carefully gated to collect data only on alveolar macrophages. Data were analysed both as the per cent of cells expressing FesRII and the mean fluorescence channel, after subtracting our binding with the control antibodies.

1 ng endotoxin via the limulus lysate assay. Polymyxin b (10 ^ig ml; Sigma) was added to the non-LPS samples to stoichiometrically bind any remaining contaminating endotoxin. IL-1/5 production was assayed by ELISA. Enzyme linked imnnmosorbent assays

iELISA)

Monocytes were stimulated as described above for 16 hr in a 37 = C/5% CO: atmosphere and supematants were collected. ELTSAs were performed for IL-l^ (Cistron Biotechnology. Pine Brooks, New Jersey, U.S.A.) and IL6 (Amgen Biologicals. Thousand Oaks, California, U.S.A.) according to the manufacturer's instructions. Briefly, flat bottomed microtitre plates were coated with the appropriate anti-cytokine antibodies. Cytokine standards and samples (1 :40 and 1:100 dilution) were applied in duplicate and allowed to bind overnight at 4 C. After washing, the enzyme linked anti-cytokine antibodies were applied. Finally, substrate was introduced and colour change determined on a Dynatech MR580 ELISA reader (Chantilly. Virginia, U.S.A.). Statistics Data are presented as the mean + s.e.m. ofthe replicate experiments. Statistical significance was calculated using Student's /-test to compare the paired means. Significance and regression analyses were performed on a Macintosh Ilex computer with the Statl'iew software program. Results

Mononuclear phagocytic cell stimulation

Effect of nedocromil on Fcr.RII induction

IgE was obtained from the U266 human IgE myeloma line (courtesy of Dr D. Vercelli). IgE was affinity purified as described [8] on a Hydropore EP column (Rainin Instrument Co, Woburn, Massachusetts, U.S.A.) to which had been conjugated 10 mg of goat anti-human IgE antibodies (Kirkegaard & Perry Laboratories Inc., Gaithersburg. Maryland, U.S.A.), Mononuclear phagocytic cells (10") were either left in a resting state or stimulated with either the combination of IgE (10"^ M) and the goat anti-human IgE(l: 100dilution)or with lipopolysaccharide (LPS; OT ^g/ml; Sigma) in a total volume of 1 ml. Alveolar macrophages (5x10'^) were either left in a resting state or stimulated with anti-human IgE. allergen to which the subject had demonstrated specific reactivity, or LPS, AM were not opsonized with TgE as when these ceils are obtained from asthmatics they have been shown to constitutively express surface IgE [9]. Experiments on both monocytes and AM were performed with or without the additional presence of nedocromil (OT-10 /JM) which was applied concomitant with or 30 min prior to the stimuli. Both the IgE and the algE contained less than

Cells ofthe human monocytic leukaemia cell line THP-1 were left in a resting state or stimulated with IL-4 (400 U/ ml), GM-CSF (100 U), IFN-a (1000 U) and IFN-y (1000 U) with or without the additional presence of nedocromi! (OT-10 ^iM). After 48 hr. the per cent of monocytes expressing FceRII was determined by flow cytometry. Preliminary experiments demonstrated that 01-10 /iM of nedoeromil did not modulate FctRII expression indtieed by IL-4 (data not shown), Theeflectsof 10/JM nedocromil on the various cytokines are shown in Fig. 1 (data represent the mean + s.e.m. of eight replicate experiments). As previously demonstrated, all four eytokines induced Fci^RII. However, no differences were observed in the additional presence of nedocromil. This was true regardless of whether or not the mononuclear phagocytic cells were pretreaLed (30 min) or concomitantly treated with the nedocromil (data not shown). Similar experiments were performed with peripheral blood monocytes and alveolar macrophages. Results on monocytes derived from normal subjects demostrated no elTect of nedocromil on induction of Fc/:;RII (data not shown).

Nedocromi! mmlulates monomiclear phagocyiic cell junction

30 f-

987

Alveolar macrophages were obtained by bronchoalveolar lavage from asthmatic volunteers. FcsRIl were expressed on 72-7+ 5'6 percent of alveolar macrophages (mean + s.e.m. of eight subjects) and this per cent was not modulated by either cytokine or nedocromil exposure (data not shown). However, exposure to the cytokines IL4, I F N - ; . IFN-a. or G M - C S F was associated with an increase in the number of receptors expressed per cell, as determined by shifts in mean fluorescence (Table 1). Concomitant exposure to nedocromil did not modulate the capacity of these cytokines to increase the expression

1=1

ofFceRII. Resting

XL-4

IFN-a Stimulus

GM-CSF

Fig. 1. Cells of the monocyte leukaemia line THP-1 were incubated (48 hr at 37 C) with the cytokines with or without the additional presence of nedocromil (10 /JM). Cells were subsequently allowed lo sequentially inleract with biotinylated mouse anti-human Fc^RII monoclonal antibody and avidinphycoerythrin. Fluorescence was evaluated via flow cytometry as the per cent of cells expressing FcrRII. Data represent the mean values |+s,e.m,) of eight experiments, • , control, D. nedocromil (10

Table 1. Effect of nedocromil on cytokine induction of FccRlI on alveolar macrophages

FcfiRII expression* Cytokine

Control

Resting IL-4 IFN-y IFN-a GM-CSF

2-5+0'9 3-9 + 0-91 3-2 + 0-74} 4-2+1-1J ?r-.+2-?j

Nedocromilt 3 0 + 1 Oil 4'5±0'9^ 3-3 +0-61 3-3 + 0-8"! 5-9 + 2-2«^

* Alveolar macrophages were obtained from asthmatic subjects and stimulated (48 hr) with the cytokines. Data represent mean+ s.e.m. fluorescence (arbitrary units) and are the mean of nine experiments, t Nedocromil at 10 /IM, X P