CI-949: A new, potential antiallergy compound inhibits antigen-induced allergic reactions in guinea-pigs in vitro and in vivo

Pulmonary Pharmacology (1990) 3 203-208 (01990 Longman Group UK Ltd

0952-0600/90/0003-0203/$10 .00

PULMONARY PHARMACOLOGY

CI-949: A New, Potential Antiallergy Compound Inhibits Antigen-Induced Allergic Reactions in Guinea-Pigs In Vitro and In Vivo R .L . Adolphson, D .O . Thueson, M .P . Finkel, J .C. Chestnut, M .C . Conroy Immunopharmacology Section, Department of Experimental Therapy, Parke-Davis Pharmaceutical Research Division, Warner-Lambert Company, Ann Arbor, MI 48105, USA SUMMARY. CI-949 (5-methoxy-3-(1-methylethoxy)-1-phenyl-N-lH-tetrazol-5-yl-lH-indole-2-carboxamide) effectively inhibited the release of histamine and the synthesis or release of immunoreactive sulfidopeptide leukotrienes C 4 -D 4 and thromboxane B 2 from antigen-challenged lung fragments of actively sensitized guineapigs. The IC 50 s were 26 .7 ± 2.8 pM for histamine, 2 .7 ± 2.4 pM for leukotriene, and 3 .0 ± 1 .8 pM for thromboxane . Drugs including ketotifen, cromolyn and nedocromil did not inhibit histamine release or did so only at high concentrations, and only cromolyn inhibited the synthesis or release of the 2 eicosanoids . A dose of 50 mg/kg i.p. of CI-949 protected conscious, aerosol-allergen challenged guinea-pigs for at least 1 h and 100 mg/kg i.p. or per os protected for at least 2 h. These results, the comparisons to standard antiallergic drugs, and other data from experiments with human lung fragments and isolated peripheral leukocytes' 2,1 .4 suggest that CI-949 should be evaluated for clinical activity against allergic and inflammatory conditions in which histamine, sulfidopeptide leukotrienes, and/or thromboxane mediate symptoms . MATERIALS AND METHODS

INTRODUCTION Recent data have convinced many investigators that the key mediators of the asthmatic response are the metabolic products of arachidonic acid generated by lipoxygenase and/or cyclooxygenase (i .e . sulfidopeptide leukotrienes C 4, D4 and E4 , prostaglandins, thromboxane A2 and others) as well as histamine . 4-8 Several authors have suggested that asthma symptoms may result from the combined actions and interactions of at least three allergic mediators . 4 .5 .9-13 Therefore, the most effective antiallergy agents may have to reduce the production of all of these mediators in order to produce significant clinical effects . Many antiallergic drugs available today do not completely fulfill the above criteria . 14,11 The approach adopted in this laboratory for the development of an antiallergy drug required that the compound inhibit the production or release of multiple mediators from antigen-stimulated lung tissue . This report presents the results of studies to quantify the activity of a new antiallergic compound, CI-949 (5methoxy-3-(1-methylethoxy)-1-phenyl-N-1 H-tetrazol5-yl-1 H-indole-2-carboxamide), shown in Figure 1 . The activity of CI-949 is also compared with other antiallergic agents currently in use: ketotifen, sodium cromolyn and a new drug, nedocromil sodium .

Mediator release from guinea pig chopped lung Preparation of lung tissue Chopped lung from actively sensitized guinea-pigs was prepared according to a modification of previously described methods . 16,17 Guinea-pigs were sensitized with 0 .1 mg of egg albumin (OA) (ICN Pharmaceuticals, Inc ., Cleveland, OH) dissolved in saline and injected i .p. After 4-6 weeks the animals were killed and the lungs were removed and washed in HEPES Ca ++ _Mg ++ buffer (mM : KCl, 5 ; NaCl, 125 ; MgC1 2 , 1 ; CaC1 2 , 1 ; and HEPES, 2-[4-(2-hydroxyethyl)piperazine-1-yl]-ethanesulfonic acid, 25), pH 7 .4, at room temperature ; care was taken to exclude extraneous blood from the parenchymal mass . The lungs were cleared of visible airways and connective tissue and then chopped into approximately 3 mm 2 pieces with a Mcllwain Chopper (Mickle Laboratories, Gonshall, Surrey, UK) . The lung fragments were set aside for 1 h before use so that mediators released as a result of tissue damage from chopping could be removed by a final wash . Assay of mediators Tissue fragments were washed twice with HEPES Ca"-Mg" buffer at room temperature immediately before use . Aliquots of about 100 mg of fragments were incubated for 10 min in tubes containing HEPES Ca++-Mg++ buffer at 37°C . Either drug or vehicle was added and the tubes were incubated for another 10 min, after which the lung fragments were

Correspondence to: Richard L. Adolphson, PhD, Immunopharmacology Section, Department of Experimental Therapy, Parke-Davis Pharmaceutical Research Division, Warner-Lambert Company, Ann Arbor, MI 48105, USA . 203

20 4

Pulmonary Pharmacology

CN 0

Fig . 1-The structure of CI-949 (5-methoxy-3-(1-methylethoxy)-lphenyl-N-1 H-tetrazol-5-yl-1 H-indole-2-carboxamide) .

incubated for 30 min with 300 .tg/ml of OA . Portions of each of the supernatants were gassed with nitrogen and reserved at - 70°C until a radioimmunoassay was performed for leukotriene C 4 (Amersham, Arlington Heights, IL) and for the principal metabolic product of thromboxane A 2, thromboxane B 2 (New England Nuclear-Dupont, Boston, MA) . The antibody supplied with the LTC4 radioimmunoassay kit cross-reacted with LTD 4 and LTE 4 (64% for both) . Thus, the measured leukotrienes in these studies are referred to as immunoreactive leukotriene (ir-LT) . Only slight cross-reactivity was reported for the antibody to thromboxane B 2 (2% for prostaglandin D 2); the measured product is referred to as immunoreactive thromboxane B 2 (ir-TX) . Another portion of each supernatant was reserved for the immediate fluorometric assay of histamine . 18 In order to estimate total tissue histamine, the remaining tissue pellets and supernatants were acidified at 80°C for 10 min with 3 .6% perchloric acid which released residual tissue histamine . Differences in tissue sample size were corrected by normalizing the measured amount of each mediator in a sample according to the total histamine in that tissue sample . Spontaneous release was subtracted from antigeninduced control and drug-treated samples, and per cent inhibition of mediator release were calculated at each concentration of drug . The IC S0 and the standard error of the IC 50 for each treatment group were calculated using a least squares linear regression model fitting responses against the log of the dose . 11,20 The histamine and eicosanoid results are the mean and standard error of triplicate samples for each treatment consisting of (n) experiments . In these experiments a differentiation between inhibition of synthesis and inhibition of release of newly formed mediator cannot be made, hence the inhibition reported here for leukotriene and thromboxane is that of appearance in the lung fragment supernatants . Anaphylactic collapse in conscious, antigen-challenged guinea-pigs A modification" of the method of Herxheimer 21 was used to sensitize and subsequently measure bronchospasm in awake guinea-pigs . Guinea-pigs were sensi-

tized 13 to 16 days before testing by a subcutaneous injection of 10 mg of OA dissolved in saline . The animals, 6 per treatment group, were exposed to a I aerosol of OA in a 48 x 32 x 23 cm plexiglass box using a Monaghan, model 675 ultrasonic nebulizer . The nebulizer output was adjusted to obtain a mean control collapse time of 1 .7 to 2 .1 min . Drugs were administered i .p ., 20 min before the OA challenge . During a 5-min exposure to the OA aerosol, those animals whose collapse times were statistically different from the mean collapse time of control animals were considered `protected', while those animals which had not collapsed by 5 min were considered `completely protected' and the experiment was terminated. This challenge period was chosen since, in earlier experiments it was determined that animals not collapsing by 5 min rarely collapsed at all . The use of times greater than 5 min, hence, would have biased the data relative to animals with collapse times between 3 and 5 min . To minimize the skewing effect of a poorly sensitized or tolerant animal, the mean values reported were calculated excluding from each group the animal having the longest collapse time (i .e . the mean was calculated for five animals) . Compounds CI-949 (L-arginine and sodium salts), 23 sodium cromolyn, nedocromil sodium and mepyramine maleate were synthesized at Parke-Davis Pharmaceutical Research Division (Ann Arbor, MI) ; ketotifen was from Sandoz Pharmaceuticals, East Hanover, NJ ; aminophylline (theophylline ethylenediamine) was from Sigma Chemical Co ., St Louis, MO . Guinea pigs Male, Hartley Strain guinea-pigs (Charles River Breeding Laboratories, Wilmington, MA), weighing 175-250 g were used in these experiments . Preparation of drugs Drugs for in vitro use were dissolved in water and the solutions were adjusted to physiologic ionic strength with concentrated HEPES buffer . For oral or i .p . administration the drugs were suspended in a mixture of 0.9% saline (w/v), 0 .9% benzyl alcohol (v/v), 0 .5% carboxymethyl cellulose (w/v), and 0 .8% Tween 80 (v/v) . All compound doses and concentrations are expressed as the weight of the active moiety . Statistical analysis The differences between the responses of untreated and treated animals in both the in vivo models were compared using an unpaired, 1-tailed Student's t-test . RESULTS Inhibition of the release/synthesis of mediators from guinea-pig chopped lung fragments Guinea-pig lung tissue incubated in buffer alone for the test period (50 min) spontaneously released low

Antiallergic Effects of CI-949 205 Table 1 . Mediator production release from guinea-pig lung fragments Mediator

(n)

Production or release of mediators (Mean ± SEM) Spontaneous Stimulated

Histamine Leukotriene Thromboxane

10 8 5

2.4±0.6 5 .0+1 .1 2.1±0.5

10.5±2.1 12 .4+5 .2 2 .7±0 .8

Lung fragments were incubated for 30 min either with 300 µg/ml of OA (stimulated release) or with HEPES Ca"-Mg + ' buffer (spontaneous release) Histamine release is expressed as the percent of total tissue histamine in the supernatant . Immunoreactive leukotriene C 4 (pg) and immunoreactive thromboxane B 2 (ng) are per mg of wet lung fragments

levels of histamine, it-LT, and it-TX into the tissue supernatants (Table 1) . The amount of each mediator released by lung fragments from sensitized animals after a 30 min challenge with 300 µg/ml of OA is also shown . CI-949 inhibited the release of histamine, it-LT, and it-TX in a concentration-dependent manner over the entire concentration range of drug used : 0 .075 gM

100- Histamine

U)

to 75 µM (Fig. 2) . The IC 50 s for CI-949 are compared with those of standard compounds in Table 2 . Cromolyn, nedocromil and ketotifen (Table 2) did not effectively reduce histamine release over the same concentration range . Nedocromil and ketotifen failed as well to consistently inhibit the accumulation of the eicosanoids (less than 50% inhibition at 75 gMequal to 20 to 30 µg/ml) . In contrast, CI-949 almost completely inhibited mediator release/production at this concentration (Fig . 2) . Ketotifen failed to inhibit histamine release at all of the concentrations tested, and even augmented antigen-induced histamine release at 75 gM, in agreement with results of other investigators .' 5 Anaphylactic collapse in conscious, antigen-challenged guinea-pigs Actively sensitized guinea-pigs were given i .p . doses of 30, 50, or 100 mg/kg of CI-949 between 20-120 min before aerosol challenge with antigen (Table 3) . The animals were protected from collapse for at least I h after 50 and 100 mg/kg, and 100 mg/kg afforded complete protection up to 2h . An oral dose of 100 mg/kg at 2 h, but not at 4 h before challenge also inhibited collapse (Table 3) . A dose of 100 mg/kg at 4 h and again at 2 h before challenge was more effective than a single dose at 2 h . Companion experiments were performed to compare CI-949 with other antiallergy drugs dosed 20 min before antigen challenge (Table 4) . Theophylline (100 mg/kg i.p .) and ketotifen (10 mg/kg i.p .) effectively extended collapse times, while cromolyn and nedocromil (100 mg/kg each i.p .) were inactive .

(6)

ao 100- Leukotriene

DISCUSSION

a) 80 0

0

60-

0

40-

(3)

(s)

20E C

0 (4)

Q 100Thromboxane

(3)

so-

60 40 200

0.1

1 .0 10 CI-949 (AM)

100

Fig . 2-C oncentration-effect curves of CI-949 in OA-challenged, actively sensitized guinea-pig lung fragments . Each point on the curves represents the mean ±SEM of N experiments .

It is now well accepted that allergic mediators exert individual effects which combine to generate asthma symptoms . Allergic mediators may also interact to produce the inflammatory characteristics of the disease . Such interactions have been described for bronchial hyperreactivity and increased releasability of mediators from inflammatory cells (for review see 24), as well as for the generation of bronchospasm .4,5,11,25 Interactions of lipoxygenase-derived metabolic products with other mediators for the production of asthma symptoms have been postulated, 26 and cyclo-oxygenase inhibitors, especially in combination with other inhibitors and antagonists, have been shown to reduce immediate allergen-induced bronchospasm . s There are reports which emphasize the importance of histamine in the development of the asthmatic response, 4,11 and show that newer antihistamines may, indeed, reduce asthma symptoms .' Finally, Schellenberg and co-workers 4 showed that complete inhibition of allergen-induced

2 06

Pulmonary Pharmacology

Table 2. Inhibition of mediator release from guinea-pig lung by CI-949 and other antiallergic drugs gM IC s , (mean±SEM) Immunoreactive (n) Thromboxane B 2

Drug

Histamine

(n)

Immunoreactive Leukotriene C 4

CI-949 Ketotifen Sodium cromolyn Nedocromil sodium

26.7±2 .8 AU NA

(10) (6) (5)

2 .7+2 .4 NA 9 .7+4 .4

(8) (4) (5)

3 .0+1 .8 NA 4 .3+5 .1

(5) (3) (3)

NA

(4)

NA

(4)

NA

(5)

(n)

A drug was judged `NA', not active, if it did not produce inhibition >_50% at 75 gM, the highest dose used Ketotifen produced augmented histamine release, `AU', at 75 µM, and did not inhibit release at lower concentrations Table 3. Effect of dose, route, and time of pretreatment with CI949 on antigen-induced collapse in conscious guinea-pigs

Table 4 . Effect of antiallergy drugs on the collapse time of antigen-challenged guinea-pigs

Pretreatment time (min)

Drug

Dose (mg/kg, i .p .)

Collapse time (min±SEM)

(n)

CI-949 Theophylline Ketotifen Sodium cromolyn Nedocrmil sodium

100 100 10 100 100

4 .9±0 .1* 4 .7±0 .3* 5 .0±0* 2 .2±0 .2 2 .1±0 .1

(5) (10) (5) (5) (5)

Control 20 20 60 75 90 30 45 90 120 120 240 120 & 240

Collapse times 2.2±0 .2 Intraperitoneal, 30 mg/kg 3 .4±0 .4 Intraperitoneal, 50 mg/kg 3 .4+0 .4 3 .7+0 .8 2 .2+0 .1 2 .1+0 .1 Intraperitoneal, 100 mg/kg 5 .0±0 .0 5 .0+0 .0 4 .4+0 .5 5 .0+0 .0 Oral, 100 mg/kg 3 .7+0 .4 1 .7+0 .6 4 .8+0 .6

(n) (5) (15)* (15)* (5)* (5)* (5) (5)* (5)* (5)* (5) (10)* (5) (15)*

'Experiments terminated after 5 min *P<0 .05, compared to controls

contractions in isolated human bronchus occurred only after pretreatment with a combination of diphenhydramine, indomethacin, and piriprost (a lipoxygenase enzyme inhibitor) . Thus, essential roles in the allergic response for histamine and the mediators derived from the oxidation of arachidonic acid have been established . In addition, cell-cell interactions during mediator release have also been described, 27,28,29 as have the effects of released mediators to modulate the subsequent release of other mediators." Since there is considerable information showing that CI-949 inhibits the activation of several inflammatory cell types in man, 1'2' 3 the present experiments were directed toward studying the combined response of all mediator releasing cells within the lung fragments . CI-949 effectively inhibited the appearance of all three mediators, histamine, sulfidopeptide leukotrienes and thromboxane B 2 , after antigen challenge of guinea-pig lung . In the assay system used for these experiments, at concentrations below 75 AM, other antiallergic drugs tested did not inhibit the release of all three mediators . Histamine release was inhibited only weakly or required drug concentrations much

Guinea-pigs were pretreated with drug 20 min before antigen challenge . Experiments were terminated after 5 min *P<0.05 compared to controls

greater than described for CI-949 . Some authors report a similar lack of effect for cromolyn and nedocromil to inhibit histamine release at micromolar concentrations, 31 while others have described potent inhibition of histamine release, but only from human mast cells obtained by bronchial lavage . 31-33 Studies with human lung in this laboratory have also revealed variable inhibition of leukotriene and thromboxane release by cromolyn and nedocromil .' These studies further emphasize the lack of effect of these drugs to inhibit histamine release. CI-949 has no antihistaminic activity in vitro at clinically relevant concentrations (personal communication), and only poorly inhibits broken cell enzyme preparations of cyclo-oxygenase (IC 50 > 200 AM vs IC 50 of 0 .6 AM for indomethacin) and lipoxygenase (32% inhibition vs 95% inhibition at 32 AM for phenidone) [personal communication] . It is clear, therefore, that a majority of the activity of CI-949 is other than specific enzyme inhibition or antagonism of histamine . Other possible actions of CI-949 which may explain these data include the inhibition of an early cell activation process . Such inhibition would, in turn, depress mediator release and/or production, and could account for the coincident reduction of all three mediators in lung preparations treated with CI-949 . Indeed, there is evidence using basophils stimulated with agents which release histamine by acting at different stages of the excitation-response pathway' showing that CI-949 inhibits a very early stage of the excitation-response pathway . CI-949 was effective in vivo in a guinea-pig model of allergic hypersensitivity in which histamine 22 is the

Antiallergic Effects of CI-949 207

predominant mediator . In unpublished studies CI-949 was also effective in a guinea-pig model where the effects of leukotrienes reportedly predominate, 16,34 in contrast, cromolyn inhibited the appearance of leukotriene and thromboxane in supernatants from the guinea-pig lung fragments, yet was not active in either model in vivo (this report and unpublished results) . Since histamine release from lung fragments was not inhibited by cromolyn, the lack of activity in the histamine-dependent collapse model is understandable . However, antihistamine- and cromolyn-treated animals also developed severe bronchospasm after antigen challenge ." It is not clear whether cromolyn's lack of protection in this model may be due to the interval between its administration and challenge or the type of antibody(ies) induced by this sensitization procedure. Both of these parameters have been shown to be important for the activity of cromolyn in animal models . 3 s .36 It is clear, in any case, that the profile of CI-949 is quite different from that of cromolyn, ketotifen and nedocromil . In the dog, antigen-induced bronchospasm is known to be largely caused by cyclo-oxygenasederived mediators . 37 The fact that CI-949 inhibits such bronchospasm 38 is further evidence that the drug inhibits the release or production of mediators from several distinct metabolic origins in inflammatory cells . Based on the range of in vivo and in vitro models in which CI-949 is effective, including human granulocytes' , " and lung tissue, this compound should be effective in clinical studies . Moreover, the unique profile of activity of CI-949 suggests that it may provide a real therapeutic advantage over present asthma therapy . Acknowledgements The authors wish to thank Dr Paul Unangst and Mr Norm Colbry of Parke-Davis Pharmaceutical Research Division of Warner-Lambert Company for synthesizing CI-949 and nedocromil sodium . References 1 . Conroy M C, Kennedy J A, Chestnut J C, Wright C D, Adolphson R L, Thueson D O. Inhibition of histamine, leukotriene C 4/D 4, and thromoxane B 2 release from human basophils and human chopped lung. Submitted for publication . 2 . Thueson D 0, Kennedy J A, Conroy M C . Inhibition of human basophil histamine release by the new antiallergy compound, CI-949, following challenge with various stimuli . J Allergy Clin Immunol 1988 ; 81 : 233. 3 . Wright C D, Hoffman M D, Thueson D 0, Conroy M C. Inhibition of human neutrophil activation by the allergic mediator meditor release inhibitor CI-949 : Mechanism of inhibitory activity. J Cell Biol 1988; 107 : 58a . 4. Schellenberg R R, Foster A, Duff M J . Anti-IgE induced contraction of human bronchus in vitro . J Allergy Clin Immun 1985 ; 75 : 126 . 5 . Phillips G D, Holgate S T . Interaction of inhaled

LTC4 with histamine and PGD2 on airway caliber in asthma . J Appl Physiol 1989; 66: 304-312 . 6. Fuller R W, Dixon C M S, Dollery C T, Barnes P J. Prostaglandin D2 potentiates airway responsiveness to histamine and methacholine . Am Rev Respir Dis 1986; 133 :252-254. 7 . Rafferty P, Beasley R, Holgate S T . The contribution of histamine to immediate bronchoconstriction provoked by inhaled allergen and adenosine 5' monophosphate in atopic asthma. Am Rev Respir Dis 1987 ;136 :369-373 . 8 . Curzen N, Rafferty P, Holgate S T . Effects of a cyclooxygenase inhibitor, flurbiprofen, and an H, histamine receptor antagonist, terfenadine, alone and in combination on allergen induced immediate bronchoconstriction in man . Thorax 1987; 42 : 946-952 . 9 . Barnes N C, Piper P J, Costello J F . Actions of inhaled leukotrienes and their interactions with other allergic mediators . Prostaglandins 1984; 28 : 629-631 . 10 . Brunelli G, Daffonchio L, Hernandez A, Sala A, Omini C, Berti F . PGD 2 induces pulmonary hyperresponsiveness to acetylcholine in vivo in the guinea-pig . J Anton Pharmacol 1987 ; 7 : 281-285 . 11 . Engineer D M, Morris H R, Piper P J, Sirois P . The release of prostaglandins and thromboxanes from guinea-pig lung by slow reacting substance, of anaphylaxis, and its inhibition . Br J Pharmacol 1978 ; 64: 211-218 . 12. Berti F, Folco G C, Nicosia S, Omini C, Pasargiklian R . The role of histamine H l - and H2 -receptors in the generation of thromboxane A2 in perfused guinea-pig lungs . Br J Pharmacol 1979 ; 65: 629-633 . 13 . Soter N A, Lewis R A, Corey E J, Austen K F . Local effects of synthetic leukotrienes (LTC4, LTD, LTE,, and LTB4 ) in human skin . J Invest Dermatol 1983 ; 80 : 115-119 . 14 . Salari H, Borgeat P, Fournier M, Herbert J, Pelletier G . Studies on the release of leukotrienes and histamine by human lung parenchymal and bronchial fragments upon immunologic and nonimmunologic stimulation : Effects of nordihydroguaiaretic acid, aspirin, and sodium cromoglycate . J Exp Med 1985 ; 162 :1904-1915 . 15 . Katayama S, Tsunoda H, Sakuma Y, Kai H, Tanaka I, Katayama K. Effect of azelastine on the release and action of leukotriene C 4 and D4 . Int Archs Allergy appl Immunol 1987; 83 : 284-289 . 16 . Barrett-Bee K J, Green L R . The relationship between histamine release and lung c-AMP levels during anaphylaxis in the guinea-pig . Prostaglandins 1975 ; 10 : 589-595 . 17 . Robichaud L J, Stewart S F, Adolphson R L . CI-922 - a novel, potent antiallergic compound-I : inhibition of mediator release in vitro . Int J Immunopharmac 1987; 9 : 41-49 . 18 . Siraganian R P. An automated continuous-flow system for the extraction and fluorometric analysis of histamine . Analyt Biochem 1974 ; 57 : 383-894 . 19 . Finney D L. Statistical Methods in Biological Assay . 3rd ed . Charles Griffen & Co. Ltd, London 1978 . 20. Snedecor G G, Cochran W G. Statistical Methods . 6th ed . Iowa State University Press, Ames 1967 . 21 . Adolphson R L, Finkel M P, Robichaud L J . CI-922 - a novel, potentially antiallergic compound-II : activity in animal models of allergy . Int J Immunopharmac 1987 ; 9 : 51-60 . 22. Herxheimer H . The protective action of antihistaminic and sympathomimetic aerosols in anaphylactic microshock of the guinea-pig . Br J Pharmac 1953; 8: 461-465 . 23 . Unangst P C, Connor D T, Stabler S R, Weikert R J, Carethers M E, Kennedy J A, Thueson D 0, Chestnut J C, Adolphson R L, Conroy M C . Novel indole carbamoyltetrazoles as potential antiallergy agents . J Med Chem 1989; 32 : 1360-1366 . 24 . Burka J F

208 Pulmonary Pharmacology bronchoconstrictor mediators. Can J Physiol Pharmacol 1987; 65 : 442-447. 25. Holgate S T, Emanuel M B, Howarth P H . Astemizole and other H 1 -antihistaminic drug treatment of asthma . J Allergy Clin Immun 1985 ; 76 : 375-380 . 26 . Peters S P, Kagey-Sobotka A, MacGlashan D W, Siegel M 1, Lichtenstein L M . The modulation of human basophil histamine release by products of the 5-lipoxygenase pathway . J Immun 1982 ; 129 : 797-803 . 27. Jakschik B A, Wei Y F, Rengers T A. Modulation of eicosanoid production due to mast cell-macrophage interaction . Adv Prostaglandin Thromboxane Leukotriene Res 1986 ; 16 : 141-152 . 28. Salari H, Chan-Yeung . Mast cell mediators stimulate synthesis of arachidonic acid metabolites in macrophages . J Immunol 1989; 142: 2821-2827 . 29. Kloprogge E, de Leeuw A, de Monchy J G R, Kauffman H F . Cellular communication in leukotriene C 4 production between eosinophils and neutrophils. Int Arch Allergy App! Immunol 1989 ; 90 : 20-23 . 30. Kay A B. Mediators and inflammatory cells in allergic disease . Ann Allergy 1987 ; 59: 35-42 . 31 . Leung K B P, Flint K C, Brostoff J, Hudspith B N, Johnson N McI, Pearce F L . A comparison of nedocromil sodium and sodium cromoglycate on human lung mast cells obtained by bronchoalveolar lavage and by dispersion of lung fragments . Eur J Respir Dis 1986 ; 69 : 223-226. 32 . Flint K C, Leung K B P, Pearce F L, Hudspith B N, Brostoff J, Johnson N Mcl . Human mast cells recovered from bronchioalveolar lavage : their morphology, histamine release, and the effects of sodium cromoglycate . Clin Sci 1985 ; 68 : 427-432.

33 . Moqbel R, Cromwell 0, Walsh G M, Wardlaw A J, Kurlak L, Kay A B . Effects of nedocromil sodium (Tilade®) on the activation of human eosinophils and neutrophils and the release of histamine from mast cells . Allergy 1988; 43 : 268-276 . 34 . Anderson W H, O'Donnell M, Simko B A, Welton A F. An in vivo model for measuring antigen-induced SRS-A-mediated bronchoconstriction and plasma SRS-A levels in the guinea pig . Br J Pharmacol 1983 ; 78 : 67-74 . 35 . Thomson D S, Evans D P . Inhibition of immediate hypersensitivity reactions by disodium cromoglycate : Requirements for activity in two laboratory models . Clin Exp Immunol 1973 ; 13 : 537-544 . 36 . Taylor W A, Roitt I M . Effect of disodium cromoglycate on various types of anaphylactic reaction in the guinea pig . Int Archs Allergy 1973 ; 45 : 795-807 . 37 . Kleeberger S R, Kolbe J, Adkinson N F, Peters S P, Spannhake E W . Central role of cyclooxygenase in the response to canine peripheral airways to antigen . J appl Physiol 1986 ; 61 : 1309-1315 . 38 . Finkel M P, Mallard S P, Thueson D 0, Conroy M C. The in vivo anti-allergic activity of CI-949, 5methoxy-3-(1-methylethoxy)-l-phenyl-N-1 H-tetrazol-5yl-lH-indole-2-carboxamide, arginine salt . J Allergy Clin Immunol 1988 ; 81 : 313 .

Date received : 23 July 1989 Date revised : 15 February 1990 Date accepted: 10 April 1990