Increased Numbers of Mast Cells in Bronchial Mucosa after the Late-Phase Asthmatic Response to Allergen 1- 3
EMANUELE CRIMI, MAURIZIO CHIARAMONDIA, MANLIO MILANESE, GIOVANNI A. ROSSI, and VITO BRUSASCO Introduction
Allergen inhalation causes airway inflammation in asthmatic patients, particularly in those who respond with a biphasic bronchospastic reaction (I, 2). The cellular characteristics of this inflammatory response have been described in studies using bronchoalveolar lavage after either inhalation or local instillation of allergen (3-5). These studies have clearly shown that eosinophils and lymphocytes are involved in the inflammatory process associated with the development of late-phase response and bronchial hyperresponsiveness (6). Mast cells were found to be activated soon after the exposure to allergen (7), and they decreased in number 6 h after allergen challenge in bronchoalveolar lavage fluid from patients with but not from patients without a late-phase response (8). Although these data seem to indicate that an enhanced mast cell degranulation may contribute to the development of the latephase response, it is also possible that they reflect a reduced dislodgement of these cells from the bronchial epithelium to airway lumen (8). By using bronchial biopsy, Beasley and coworkers (9) showed that mast cellstended to increase in the bronchial mucosa 18 h after allergen challenge, but a clear link with the development of the late-phase response was not established. In the present study we investigated if a relationship exists between the occurrence ofthe late-phase bronchospastic response to allergen and the number of mast cells or other inflammatory cell numbers in the bronchial mucosa. For this purpose we analyzed the bronchial biopsy specimens obtained in patients with and without late-phase response at different times after allergen inhalation challenge. Methods Patients The study was performed on 27 (26 male and one female) outpatients, 16 to 38 yr of age. 1282
SUMMARY Weexamined the characteristics of allergen-Induced Inflammation of the bronchial muco88 In esthmatlc patients. Studies were carried out 4 h (eight patients) and 24 h (nine patients) after allergen Inhalation challenge; 10 patients were not challenged and served as control subjects. We found that In the control group the ratio of degranulatlng to granulatad mast cells was higher In patients with than In patlants without Iste-phase response. In patients studlad 4 h sfter allergen challenge the total number of mast calls was not significantly different from that In controlsubJectsj the ratio of degranulatlng to granUlated mast calls was Increased similarly In patients with and without late-phase response. Among patients studied 24 h after allergen challenge, those who had developed the late-phase response had an Increased (p < 0.05) number of mast cells as compared with patients who had not developed the late-phase response, the number of mast calls was significantly correlated with the severity of the late-phase response (r = 0.80; P < 0.001). The numbers of eoslnophlls and mononuclear cells and the morphologic abnormalities of bronchial structure (altered ratio of cylindrical to globet cells, thickening of the basement membrane, and edema and anglectssls of lamlns propria) were similar In the different groups of patients. We conclude that the Inflammatory events leading to the development of the late-phase asthmatic response to allergen represent a stimulus for an Increase In the number of mast cells In the bronchial mUC088. AM REV RESPIR DIS 1991; 144:1282-1286
They had a history of perennial asthma, with or without rhinitis, lasting for 2 yr or longer. Allergic sensitization to house dust mite was demonstrated in all patients by skin prick test or radioallergosorbent test or both. To be included in the study, patients were required to be asymptomatic, not to have suffered from respiratory infections during the previous 4 wk, and to have a FEV l greater than 70070 of predicted (10). None of the patients used antiasthma treatment other than ~2-stimu lants on an as-necessary basis. Treatments were discontinued at least 12 h before study. All patients wereinformed on the nature and the scope of the study, and they gave written consent. Patients weredivided into three groups. Ten patients wereassigned to a control group and were not challenged with the allergen. The remaining 17 patients underwent bronchial biopsy 4 h (eight patients, the 4-h group) or 24 h (nine patients, the 24-h group) after allergen inhalation challenge. Patients of each group were further divided in two subgroups depending on whether they did (LAR +) or did not (LAR-) develop a late-phase asthmatic response either on the study day (24-h group), or on a preceding occasion (control group and 4-h group). Bronchial Challenges Forced expiratory maneuvers were recorded by a wedge spirometer (Vitalograph Ltd., Buckingham, UK) and the highest FEV 1 from
three technically acceptable maneuvers was retained for analysis. Aerosols were delivered by an ampul-dosimeter device(MEFAR, Brescia, Italy) according to a method previously described (ll). Control measurements of FEV 1 were determined after inhalation of saline. Allergen bronchial challenge was performed by using scalar solutions of Dermatophagoides pteronyssinus reconstituted from predosed (arbitrary units, AU), dried allergen (Pharmacia, Uppsala, Sweden). Fifteen minutes were allowed between allergen inhalations and FEV 1 measurements. The allergen bronchial challenge was started from a dose of 4 AU, with twofold increments until FEV 1 fell below 80% of control or up to a maximum dose of 500 AU. The twofold increments of allergen dose were obtained by
(Received in original form January 23, 1991 and in revised form Apri125, 1991) 1 From the Istituto di Medicina della Sport, Cattedra di Fisiopatologia Respiratoria, Universita di Genova, and Servizio di Anatomia Patologica Ospedale San Martino, XIII USL, and Divisione di Pneumologia, Istituto G. Gaslini, Genova, Italy. 2 Supported in part by Grants 87.00560.56 and SP2 FATMA from the Consiglio Nazionale delle Ricerche, Rome, Italy. J Correspondence and requests for reprints should be addressed to Emanuele Crimi, M.D., Istituto di Medicina dello Sport, Ospedale San Martino, Viale Benedetto XV, 10. 16132 Genova, Italy.
1283
MAST CELLS AND THE LATE ASTHMATIC REACTION
inhaling doubling allergen concentrations during 70 quiet breaths corresponding to an effective inhaled solution of 0.5 ml. FEV, was measured postchallenge 30 and 60 min, and then hourly for 8 h to detect the occurrence of the late-phase asthmatic response defined as a decrease of 15% or more in the FEV, during the 3- to 8-h period after initial earlyphase reaction. For methacholine challenge 1-, 10-,and 50-mg/ml solutions ofmethacholine chloride in saline were prepared. Double increasing doses of methacholine were obtained changing number of breaths and methacholine concentration. Challenge was started from a dose of 10 jig, with twofold increments until the FEV, measured I min after inhalation was decreased below 80010 of the control value. The maximal dose at which methacholine challenge was stopped if patients did not respond was 5,000 jig. The dose causing a 15% fall of FEV, (PD 15 ) was calculated by interpolation of the dose-response curves. A single dose of inhaled salbutamol (200 jig) was given immediately after methacholine challenge to relieve bronchospasm. Methacholine responsiveness was determined on the day before allergen challenge and again 1 h before bronchoscopy.
Bronchial Biopsy Patients were premedicated intramuscularly with atropine (0.5 mg) and diazepam (10mg) 5 min before the procedure. After local anesthesia of the nostrils with lidocaine solution 2% followed by injection of epinephrine solution 0.1/1,000 (l ml each side), the fiberoptic bronchoscope (Olympus BF, type PIO; Lorenzatto, Torino, Italy) was passed through the nose. After local anesthesia of pharynx and airways, the tip of the instrument was wedged into a subsegmental branch of the right middle lobe, and bronchoalveolar lavage was performed as described elsewhere (6). The results of bronchoalveolar lavage have been the object of a separate report (12). Biopsies were taken from the main carina and from the carina of the right upper lobe. Biopsies were gently extracted from the forceps with a hypodermic needle and immediately fixed in 10070 buffered formalin solution at room temperature. Samples were then embedded in paraffin, cut at 5 jim with a rotative microtome, and stained with toluidine blue, for the demonstration of mast cells, and with hematoxylin-eosin, for the demonstration of other parenchymal and inflammatory cells. Bronchial biopsies were analyzed by two independent observers who were unaware of the results of bronchial challenges. Light microscopy was performed with a Zeiss Axiomat microscope (Oberkochen, Germany). Bronchial biopsies not correctly oriented were withdrawn. The following morphologic characteristics were evaluated: abnormalities in the ratio of cylindrical to goblet cells, thickening or other abnormalities of basement membrane, and edema and angiectasis of the lamina propria. These morphologic abnormalities were scored from zero to 3, with zero meaning absence of abnormali-
TABLE 1 PATIENT CHARACTERISTICS'
Age (yr) Subgroup Control group LARLAR+ 4-h group LARLAR+ 24-h group LARLAR+
Number
Mean
Range
FEV,t (% pred)
5 5
22 24
16-25 19-27
4 4
22 19
4 5
21 24
MCh PO,.
Maximal FEV, Fallt (%)
(p.g. geometric mean)
Early
late
98 ± 1 90 ± 3
1,038 97
19 ± 3 28 ± 2
2 ± 1 31 ± 3
19-25 18-19
95 ± 5 89 ± 5
234 89
17 ± 2 27 ± 4
5 ± 2 39 ± 3
17-26 17-38
106 ± 4 96 ± 7
760 44
20 ± 4 26 ± 3
3 ± 1 30 ± 5
• The early and late FEV, falls after allergen inhalation challenge in the control group and the late FEV, fall in the 4-h group are those obs,'rved on a prestudy day. t Values are mean ± SE.
ties. Cells lying within 200 jim from the epithelial basement membrane werecounted and identified by means of an eyepiece graticule at x 500 over five fields. Endothelial cells, pericytes, and Schwann cells were not included in this count. Mast cellswere also observed at x 1,000 to x 2,000 to see if they were degranulating or not. Inhaled salbutamol (200 J.l.g) was given to patients who experienced respiratory symptoms after bronchoscopy. Patients left the hospital 1 h after completion of the procedure with a contact telephone number, which they could call at any time.
Statistical Analysis Two-factoranalysis of variance (ANOVA) was used to compare differences in baseline data among groups and between subgroups. In order to control the per-experiment error rate, Kruskal-WallisANOVAwas used to assess the significance of differences in the number of each cell type among groups. When this preliminary test was significant at 5% level, mul-
tiple comparisons between groups were carried out by the Mann-Whitney-Wilcoxon test. The paired t test, correlation coefficients, and chi-square test were used when appropriate; p values less than 0.05 were considered to be statistically significant. Results
Baseline Data The age of the patients and the baseline lung function were similar in the control group and in the two challenged groups, (table 1). Methacholine PD 15 was not different among groups (p > 0.05), but it was significantly lower in the LAR + . subgroups as compared with the LARsubgroups (p < 0.05). The early maximal FEV 1 fall was also not different among groups (p > 0.05), but it was significantly larger in the LAR + subgroups as compared with the LAR- subgroups (p < 0.005).
Mast cella/rum" 25
*
20 Fig. 2. Relationship between the number of mast cells observed in bronchial biopsy 24 h after allergen challenge and the maximal fall of FEll, observed between 3 and 8 h after the end of challenge. Note that both lAR+ and lARpatients are shown in this figure. r • 0.80; p < 0.001.
15
1
1
10
5
0'-----"---'---------------
Control
4h
24 h
1284
CRIMI, CHIARAMONDIA. MILANESE, ROSSI, AND BRUSASCO
60 Mast cella/mrn" Fig. 1. Numbers of mast cells in subgroupsof patientswith (LAR+ = closed columns) or without (LAR- ~ open columns) late-phase asthmatic reaction to allergen inhalation challenge. Horizontal lines are medians. columns indicate the lower and upper quartiles. Asterisk indicates sign ificantly different (p < 0.05)from the LAR- subgroup of the 24-hgroup and from the LAR+ subgroup of the 4-h group.
40
20
o o
LJ..L.=----'*'--- -'--- - - - - - ' - - - - ' - - - - - ' - - -
10
20 30 Late FEV, fall (0/0)
50
40
Bronchial Biopsy Studies The numbers of mast cells observed in bronchial biopsy specimens are shown in figure 1.Within the 24-h group the number of mast cells was significantly (p < 0.05) higher in the LAR + subgroup than in the LAR- subgroup. In addition, in the LAR + subgroup of the 24-h group the number of mast cells was significantly (p < 0.01)higher than in the LAR + subgroup of the 4-h group and tended to be higher (P = 0.08) than in the LAR + sub-
group of the control group. Furthermore, in the 24-h group the number of mast cells was significantly correlated (p < 0.001)with the maximal late fall of FEV I observed after allergen inhalation challenge (figure 2). When the proportion of degranulating mast cells to total mast cells was evaluated (table 2), it appeared that in the 4-h group this was significantly (p < 0.001) higher than in the other groups, irrespective of the type of asthmatic re-
TABLE 2
Group Control group 4-h group 24-h group p values. 2 x 3 tables
LAR-
LAR+
01184 32/56 0/48 < 0.001
24/200 40/64 16/512 < 0.001
(2
p Values tables)
x 2
< 0.001 NS NS
• The number of mast cells lor eech subgroup was obtained by adding the number of mast cells of all individual patients belonging to that subgroup.
TABLE 3 TOTAL CELLS. MONONUCLEAR CELLS, EOSINOPHILS. AND MORPHOLOGIC ABNORMALITIES'
Subgroups Control group LARLAR+ 4-h group LARLAR+ 24-h group LARLAR+
Total Cells per mm2 (x 10')
Mononuclear Cells per mm2 (x 10')
10 (IH1) 15 (6-17)
8 (8-9) 8 (6-10)
6.5 (5.5-9.5) 12 (10.5-12.5)
5 (4--8) 7 (6.5-8.5)
8.5 (6-12) 13 (10-15)
5 (5-6.5) 8 (7-8)
Eosinophils per mm2
3 (0-6) 8 (3-25.5)
6 (3-9) 20.5 (13-28) 5 (3-18.5) 19 (8-48)
Airway Responsiveness In the LAR + subgroup of the 4-h group, but not in the LAR- subgroup, methacholine POlS (geometric mean) was decreased 3 h after allergen challenge, from 89 to 32 IJ.g (p < 0.005). Similarly, in the LAR + subgroup, but not in the LARsubgroup, of the 24-h group, methacholine POlS was decreased23 h after allergen challenge, from 44 to 24 IJ.g (p = 0.05). Discussion
PROPORTION OF DEGRANULATING TO TOTAL MAST CELLS' DegranulatingiGranulated (per mmZj
sponse; by contrast, within the control group, the proportion of degranulating mast cells was significantly (p < 0.001) higher in the LAR + subgroup than in the LAR- subgroup. Other data of bronchial biopsy specimens are shown in table 3. The number of total cells was significantly higher (P < 0.05) in the LAR + subgroups than in the LAR- subgroups, without differences among groups (p > 0.05). The number of eosinophils tended to be higher in the LAR + subgroup than in the LAR - subgroup of the 4 h group (p = 0.06). The number of mononuclear cells and the morphologic abnormalities score were similar in all groups and subgroups (p > 0.05, all comparisons).
Morphologic Abnormalities Score
3 (1-3) 4 (3-7) 2 (1.5-2.5) 3 (3-3.5) 3 (1.5-4) 3
(3-4)
• The number of total cells was significantly different (p < 0.05) betwesn subgroups but not betwesn groups (ANOVA). Values are median (lower and upper quartiles).
The results of the present study confirm that mast cell degranulation is associated with the bronchial response to inhaled allergen. In addition, wehave shown that the events determining the occurrence and the intensity of the late-phase asthmatic response are also a stimulus for mast cell recruitment in the bronchial mucosa. Before discussing our results we must comment on methodology. Both control patients and patients who were studied 4 h after allergen challenge were divided in subgroups depending on whether they did or did not develop the late-phase response to allergen challenge on a prestudy day. This was done because on study days patients received medications before and after bronchoscopy; this might have prevented or blunted the latephase response. Wethink the assumption that these patients would have developed a late-phase response also on study days is justified for two reasons. First, the dose of allergen able to provoke the earlyphase response was similar on the study day and on the prestudy day. Second, patients known to have late-phase response, but not single responders, showed an increased airway responsiveness 3 h after allergen challenge. An increase of airway responsiveness is associated,as also shown
MAST CELLS AND THE LATE ASTHMATIC REACTION
in our patients studied 24 h after allergen challenge, with the late-phase response and often precedes its development (13, 14). A limitation that must be kept in mind in interpreting the results of the present study is that we used formalin as standard fixation method; this method does not allow the identification of either basophils or formalin-sensitive mast cells. Therefore, no conclusion can be drawn about the role of these cells in the airway exposed to allergen. There isgood agreement that mast cells have a role in initiating the bronchial inflammatory reaction to allergen exposure (15-19). A relationship has been established between the degree of sensitivity to methacholine, the number of mast cells, and the level of histamine in bronchoalveolar lavage (17). Whether these cells playa role in perpetuating the inflammatory processes that sustain the asthmatic disease remains, however, still unknown. Most studies aimed at investigating the inflammatory events that follow the inhalation of allergen were carried out by bronchoalveolar lavage. In patients with dual response, but not in patients with single early response, Diaz and coworkers (8) showed a significant decrease of the number of mast cells in bronchoalveolar lavage performed 6 h after allergen inhalation as compared with that at baseline. These investigators postulated two hypotheses to explain their findings. The first hypothesis is that inflammatory cell-derived factors associated with the development of the late-phase response may induce, in a nonspecific way, mast cell degranulation. The second hypothesis is that luminal mast cells activated during the development of the late-phase response are more adherent to bronchial epithelium and, therefore, more difficult to dislodge by the lavage procedure. Indeed, the use of bronchoalveolar lavage for studying mast cells has some inherent limitations. First, the number of mast cells that can be recovered by BAL in asthmatic patients is usually small and most times below 1070 the total cells; second, degranulated mast cells are unrecognizable; third, mast cells tend to be trapped among epithelial cells. Bronchial biopsy seems to be a more suitable method to study mast cells in asthmatic patients. By using this technique, Beasley and coworkers (9) found mast cells with various stages of degranulation in the bronchial lamina propria of asthmatics but not in that of healthy subjects. Consistent with these findings
1285
we have shown that the percentage of degranulating mast cells observed 4 h after allergen challenge was higher than in control subjects. Moreover, both the total number of mast cells and the proportion of degranulating mast cellswas similar in patients with and in patients without late-phase response studied 4 h after allergen challenge. These findings suggest that mast cell degranulation plays a role in the early-phase response, whether or not the late-phase response will occur. The increase of bronchial responsiveness that precedes the late-phase response seems, therefore, not to be related to an enhanced mast cell degranulation. Nevertheless, the finding that in the control group patients with late-phase response had a larger proportion of degranulating mast cells than did patients without late-phase responses, supports the suggestion that baseline hyperresponsiveness is sustained by ongoing mast cellsdegranulation (17). An important finding of the present study is that 24 h after allergen challenge the number of mast cellswas significantly higher in patients who had developed the late response than in patients who had not. In addition, unlike the mast cells observed 4 h after allergen challenge, the majority of these mast cells were granulated. These findings indicate that the occurrence of a late-phase response is associated with some stimuli, leading to an increase of the number of mast cells in the bronchial mucosa. These events may include the migration of mature mast cellsinto the mucosal epithelium, the rapid maturation of immature cells, and the differentiation of other cellsalready present. Several cytokines (IL-4, IL-3, OMCSF, IL-5) can promote growth, differentiation, and phenotypical switches in mast cells (20). In patients without latephase response 24 h after challenge the number of mast cells remained low, whereas granulation was complete, suggesting regranulation of the existing mast cells. In a canine experimental model (21) mast cells have been shown to increase rapidly, i.e., within 1 h, in the peripheral airways after exposure to specific stimuli and in the central airwaysafter exposure to both specific and nonspecific stimuli. The fact that both specificand nonspecific stimuli caused mast cells to increase in the tracheal epithelium independent of the type of stimulus and of the dose inhaled suggested that this response might be the result of epithelial irritation or damage. In our study, mast cells were
found to be increased 24 but not 4 h after allergen challenge; furthermore, the stimuli causing mast cell migration seem to be specific and dose-dependent since the mast cell increase was related to the occurrence and the severityof the preceding late asthmatic response. These differences between our study and that of Turner and coworkers (21) may be due to a difference in bronchial allergic response between humans and dogs. Previous studies in patients with rhinitis have shown that the mast cell number in nasal mucosa correlates with the clinical symptoms of allergy (22), and the patients with greater sensitivity to pollen have more intraepithelial mast cells during the pollen season (23). Our findings suggest that, like prolonged natural exposure, also the experimental exposure to the sensitizing allergen may cause mast cells to increase in the exposed tissues provided that the allergenic stimulus is sufficiently strong to trigger the inflammatory process associated with the latephase response. We found no significant differences in the number of other inflammatory cells in bronchial biopsy by comparing patients with and patients without the latephase response. However, the number of eosinophils in patients with a late-phase response tended to be higher than in patients without a late-phase response 4 h after allergen challenge. Even though statistical significance was not achieved, this finding is in agreement with the increase of eosinophils in bronchoalveolar lavage shown in a separate report (12). The possible interaction between the kinetics of mast cells and eosinophils deserves further investigation. In conclusion, we have shown that the number of mast cells is increased in the bronchial mucosa 24 h after the experimental inhalation of allergen only in patients who developed a late-phase response. The close correlation between the severity of the late-phase response and the number of mast cells in the bronchial mucosa suggests that during the latephase bronchial response to allergen a release of factors that cause the subsequent mast cellincrease occurs. Whether a similar increase of mast cells in bronchial mucosa is also induced by natural exposure to allergen remains to be elucidated. References I. Abraham WM, Perruchoud AP, Sielczak MW, YergerLD, Stevenson J8. Airwayinflammation during antigen-induced late bronchial obstruction. Prog Respir Res 1985; 19:48-55.
1286
ST. Cellular events in bronchi in mild asthma and 2. O'Byrne PM, Dolovich J, Hargreave FE. Late asthmatic response. Am Rev Respir Dis 1987; after bronchial provocation. Am Rev Respir Dis 1989; 139:806-17. 136:740-51. 10. Quanjer PH. Standardized lung function test3. De Monchy JGR, Kauffman HF, Venge P, et al. Bronchoalveolar eosinophilia during allergen- ing. Report of working party on standardization induced late-phase asthmatic reactions. Am Rev of lung function tests: European Community for Coal and Steel. Bull Eur Physiopathol Respir 1983; Respir Dis 1985; 13l:373-6. 19(5uppl:5:1-94). 4. Metzger WJ, Richerson HB, Worden K, Mon. Crimi E, Brusasco V, Losurdo E, Crimi P. nick M, Hunninghake GW. Bronchoalveolar lavage of allergic asthmatic patients following allergen Predictive accuracy of late asthmatic reaction to Dermatophagoides pteronyssinus. J Allergy Clio bronchoprovocation. Chest 1986; 89:477-83. Immunol 1986; 78:908-1003. 5. Metzger WJ, Zavala D, Richerson HB, et al. 12. Rossi GA, Crimi E, Lantero S, et al. The lateLocal allergen challenge and bronchoalveolar laphase asthmatic reaction to inhaled allergen is asvage of allergic asthmatic lungs. Am Rev Respir sociated with an early recruitment of eosinophils Dis 1987; 135:433-40. in the airways. Am RevRespirDis 1991; 144:379-83. 6. Brusasco V, Crimi E, Gianiorio P, Lantero S, 13. Thorpe JE, Steinberg D, Bernstein IL, MurRossi GA. Allergen induced increase in airway relas CG. Bronchial reactivity increases soon after sponsiveness and inflammation in mild asthma. J Appl Physiol 1990; 69:2209-14. the immediate response in dual responding asth7. Wenzel SE, Fowler AA, III, Schwartz LB. Acmatic subjects. Chest 1987; 91:21-5. 14. Durham SR, Craddock CF, Cookson WO, tivation of pulmonary mast cellsby bronchoalveolar Benson MK. Increases in airway responsiveness to allergen challenge. Am Rev Respir Dis 1988; histamine precede allergen-induced late asthmatic 137:1002-8. 8. Diaz P, Gonzalez MC, GalleguiIlos FR, et a/. . responses. J AllergyClin Immunol1988; 82:764-70. IS. Flint KC, Leung PKB, Hudspith BN, et al. Leukocytes and mediators in bronchoalveolar laBronchoalveolar mast cells in extrinsic asthma: a vage during allergen-induced late-phase asthmatic reactions. Am Rev Respir Dis 1989; 139:1383-9. mechanism for the initiation of antigen specific bronchoconstriction. Br Med J [C1in Res] 1985; 9. Beasley R, Roche WR, Roberts JA, Holgate
CRIMI, CHIARAMONDIA, MILANESE, ROSSI, AND BRUSASCO
291:923-6. 16. Kay AB. Provoked asthma and mast cells.Am Rev Respir Dis 1987; 135:1200-3. 17. Casale TB, Wood D, Richerson HB, Zehr B, Zavala D, Hunninghake GW. Direct evidence of a role for mast cells in the pathogenesis of antigeninduced bronchoconstriction. J Clin Invest 1987; 80:1507-11. 18. Kaliner M. Asthma and mast cell activation. J Allergy Clin Immunol 1989; 83:510-20. 19. Friedman MM, KalinerMA. Human mast cells and asthma. Am Rev Respir Dis 1987;135:n57-64. 20. Denburg JA, Dolovich J, Harnish D. Basophil mast cell and eosinophil growth and differentiation factors in human allergic disease. Clin Exp Allergy 1989; 19:249-54. 21. Turner CR, Kolbe J, Spannhake EW. Rapid increase in mast cell numbers in canine central and peripheral airways. J Appl Physiol1988; 65:445-51. 22.• Otsuka H, Denburg J, Dolovich J, et al. Heterogeneity of metachromatic mast cells in the human nose: significance of mucosal mast cells. J Allergy Clin Immunol 1985; 76:695-702. 23. Enerback L, Pipkorn U, Granerus G. Intraepithelial migration of nasal mucosal mast cells in hay fever. Int Arch Allergy Appl Immunol1986; 80:44-51.
EMANUELE CRIMI, MAURIZIO CHIARAMONDIA, MANLIO MILANESE, GIOVANNI A. ROSSI, and VITO BRUSASCO Introduction
Allergen inhalation causes airway inflammation in asthmatic patients, particularly in those who respond with a biphasic bronchospastic reaction (I, 2). The cellular characteristics of this inflammatory response have been described in studies using bronchoalveolar lavage after either inhalation or local instillation of allergen (3-5). These studies have clearly shown that eosinophils and lymphocytes are involved in the inflammatory process associated with the development of late-phase response and bronchial hyperresponsiveness (6). Mast cells were found to be activated soon after the exposure to allergen (7), and they decreased in number 6 h after allergen challenge in bronchoalveolar lavage fluid from patients with but not from patients without a late-phase response (8). Although these data seem to indicate that an enhanced mast cell degranulation may contribute to the development of the latephase response, it is also possible that they reflect a reduced dislodgement of these cells from the bronchial epithelium to airway lumen (8). By using bronchial biopsy, Beasley and coworkers (9) showed that mast cellstended to increase in the bronchial mucosa 18 h after allergen challenge, but a clear link with the development of the late-phase response was not established. In the present study we investigated if a relationship exists between the occurrence ofthe late-phase bronchospastic response to allergen and the number of mast cells or other inflammatory cell numbers in the bronchial mucosa. For this purpose we analyzed the bronchial biopsy specimens obtained in patients with and without late-phase response at different times after allergen inhalation challenge. Methods Patients The study was performed on 27 (26 male and one female) outpatients, 16 to 38 yr of age. 1282
SUMMARY Weexamined the characteristics of allergen-Induced Inflammation of the bronchial muco88 In esthmatlc patients. Studies were carried out 4 h (eight patients) and 24 h (nine patients) after allergen Inhalation challenge; 10 patients were not challenged and served as control subjects. We found that In the control group the ratio of degranulatlng to granulatad mast cells was higher In patients with than In patlants without Iste-phase response. In patients studlad 4 h sfter allergen challenge the total number of mast calls was not significantly different from that In controlsubJectsj the ratio of degranulatlng to granUlated mast calls was Increased similarly In patients with and without late-phase response. Among patients studied 24 h after allergen challenge, those who had developed the late-phase response had an Increased (p < 0.05) number of mast cells as compared with patients who had not developed the late-phase response, the number of mast calls was significantly correlated with the severity of the late-phase response (r = 0.80; P < 0.001). The numbers of eoslnophlls and mononuclear cells and the morphologic abnormalities of bronchial structure (altered ratio of cylindrical to globet cells, thickening of the basement membrane, and edema and anglectssls of lamlns propria) were similar In the different groups of patients. We conclude that the Inflammatory events leading to the development of the late-phase asthmatic response to allergen represent a stimulus for an Increase In the number of mast cells In the bronchial mUC088. AM REV RESPIR DIS 1991; 144:1282-1286
They had a history of perennial asthma, with or without rhinitis, lasting for 2 yr or longer. Allergic sensitization to house dust mite was demonstrated in all patients by skin prick test or radioallergosorbent test or both. To be included in the study, patients were required to be asymptomatic, not to have suffered from respiratory infections during the previous 4 wk, and to have a FEV l greater than 70070 of predicted (10). None of the patients used antiasthma treatment other than ~2-stimu lants on an as-necessary basis. Treatments were discontinued at least 12 h before study. All patients wereinformed on the nature and the scope of the study, and they gave written consent. Patients weredivided into three groups. Ten patients wereassigned to a control group and were not challenged with the allergen. The remaining 17 patients underwent bronchial biopsy 4 h (eight patients, the 4-h group) or 24 h (nine patients, the 24-h group) after allergen inhalation challenge. Patients of each group were further divided in two subgroups depending on whether they did (LAR +) or did not (LAR-) develop a late-phase asthmatic response either on the study day (24-h group), or on a preceding occasion (control group and 4-h group). Bronchial Challenges Forced expiratory maneuvers were recorded by a wedge spirometer (Vitalograph Ltd., Buckingham, UK) and the highest FEV 1 from
three technically acceptable maneuvers was retained for analysis. Aerosols were delivered by an ampul-dosimeter device(MEFAR, Brescia, Italy) according to a method previously described (ll). Control measurements of FEV 1 were determined after inhalation of saline. Allergen bronchial challenge was performed by using scalar solutions of Dermatophagoides pteronyssinus reconstituted from predosed (arbitrary units, AU), dried allergen (Pharmacia, Uppsala, Sweden). Fifteen minutes were allowed between allergen inhalations and FEV 1 measurements. The allergen bronchial challenge was started from a dose of 4 AU, with twofold increments until FEV 1 fell below 80% of control or up to a maximum dose of 500 AU. The twofold increments of allergen dose were obtained by
(Received in original form January 23, 1991 and in revised form Apri125, 1991) 1 From the Istituto di Medicina della Sport, Cattedra di Fisiopatologia Respiratoria, Universita di Genova, and Servizio di Anatomia Patologica Ospedale San Martino, XIII USL, and Divisione di Pneumologia, Istituto G. Gaslini, Genova, Italy. 2 Supported in part by Grants 87.00560.56 and SP2 FATMA from the Consiglio Nazionale delle Ricerche, Rome, Italy. J Correspondence and requests for reprints should be addressed to Emanuele Crimi, M.D., Istituto di Medicina dello Sport, Ospedale San Martino, Viale Benedetto XV, 10. 16132 Genova, Italy.
1283
MAST CELLS AND THE LATE ASTHMATIC REACTION
inhaling doubling allergen concentrations during 70 quiet breaths corresponding to an effective inhaled solution of 0.5 ml. FEV, was measured postchallenge 30 and 60 min, and then hourly for 8 h to detect the occurrence of the late-phase asthmatic response defined as a decrease of 15% or more in the FEV, during the 3- to 8-h period after initial earlyphase reaction. For methacholine challenge 1-, 10-,and 50-mg/ml solutions ofmethacholine chloride in saline were prepared. Double increasing doses of methacholine were obtained changing number of breaths and methacholine concentration. Challenge was started from a dose of 10 jig, with twofold increments until the FEV, measured I min after inhalation was decreased below 80010 of the control value. The maximal dose at which methacholine challenge was stopped if patients did not respond was 5,000 jig. The dose causing a 15% fall of FEV, (PD 15 ) was calculated by interpolation of the dose-response curves. A single dose of inhaled salbutamol (200 jig) was given immediately after methacholine challenge to relieve bronchospasm. Methacholine responsiveness was determined on the day before allergen challenge and again 1 h before bronchoscopy.
Bronchial Biopsy Patients were premedicated intramuscularly with atropine (0.5 mg) and diazepam (10mg) 5 min before the procedure. After local anesthesia of the nostrils with lidocaine solution 2% followed by injection of epinephrine solution 0.1/1,000 (l ml each side), the fiberoptic bronchoscope (Olympus BF, type PIO; Lorenzatto, Torino, Italy) was passed through the nose. After local anesthesia of pharynx and airways, the tip of the instrument was wedged into a subsegmental branch of the right middle lobe, and bronchoalveolar lavage was performed as described elsewhere (6). The results of bronchoalveolar lavage have been the object of a separate report (12). Biopsies were taken from the main carina and from the carina of the right upper lobe. Biopsies were gently extracted from the forceps with a hypodermic needle and immediately fixed in 10070 buffered formalin solution at room temperature. Samples were then embedded in paraffin, cut at 5 jim with a rotative microtome, and stained with toluidine blue, for the demonstration of mast cells, and with hematoxylin-eosin, for the demonstration of other parenchymal and inflammatory cells. Bronchial biopsies were analyzed by two independent observers who were unaware of the results of bronchial challenges. Light microscopy was performed with a Zeiss Axiomat microscope (Oberkochen, Germany). Bronchial biopsies not correctly oriented were withdrawn. The following morphologic characteristics were evaluated: abnormalities in the ratio of cylindrical to goblet cells, thickening or other abnormalities of basement membrane, and edema and angiectasis of the lamina propria. These morphologic abnormalities were scored from zero to 3, with zero meaning absence of abnormali-
TABLE 1 PATIENT CHARACTERISTICS'
Age (yr) Subgroup Control group LARLAR+ 4-h group LARLAR+ 24-h group LARLAR+
Number
Mean
Range
FEV,t (% pred)
5 5
22 24
16-25 19-27
4 4
22 19
4 5
21 24
MCh PO,.
Maximal FEV, Fallt (%)
(p.g. geometric mean)
Early
late
98 ± 1 90 ± 3
1,038 97
19 ± 3 28 ± 2
2 ± 1 31 ± 3
19-25 18-19
95 ± 5 89 ± 5
234 89
17 ± 2 27 ± 4
5 ± 2 39 ± 3
17-26 17-38
106 ± 4 96 ± 7
760 44
20 ± 4 26 ± 3
3 ± 1 30 ± 5
• The early and late FEV, falls after allergen inhalation challenge in the control group and the late FEV, fall in the 4-h group are those obs,'rved on a prestudy day. t Values are mean ± SE.
ties. Cells lying within 200 jim from the epithelial basement membrane werecounted and identified by means of an eyepiece graticule at x 500 over five fields. Endothelial cells, pericytes, and Schwann cells were not included in this count. Mast cellswere also observed at x 1,000 to x 2,000 to see if they were degranulating or not. Inhaled salbutamol (200 J.l.g) was given to patients who experienced respiratory symptoms after bronchoscopy. Patients left the hospital 1 h after completion of the procedure with a contact telephone number, which they could call at any time.
Statistical Analysis Two-factoranalysis of variance (ANOVA) was used to compare differences in baseline data among groups and between subgroups. In order to control the per-experiment error rate, Kruskal-WallisANOVAwas used to assess the significance of differences in the number of each cell type among groups. When this preliminary test was significant at 5% level, mul-
tiple comparisons between groups were carried out by the Mann-Whitney-Wilcoxon test. The paired t test, correlation coefficients, and chi-square test were used when appropriate; p values less than 0.05 were considered to be statistically significant. Results
Baseline Data The age of the patients and the baseline lung function were similar in the control group and in the two challenged groups, (table 1). Methacholine PD 15 was not different among groups (p > 0.05), but it was significantly lower in the LAR + . subgroups as compared with the LARsubgroups (p < 0.05). The early maximal FEV 1 fall was also not different among groups (p > 0.05), but it was significantly larger in the LAR + subgroups as compared with the LAR- subgroups (p < 0.005).
Mast cella/rum" 25
*
20 Fig. 2. Relationship between the number of mast cells observed in bronchial biopsy 24 h after allergen challenge and the maximal fall of FEll, observed between 3 and 8 h after the end of challenge. Note that both lAR+ and lARpatients are shown in this figure. r • 0.80; p < 0.001.
15
1
1
10
5
0'-----"---'---------------
Control
4h
24 h
1284
CRIMI, CHIARAMONDIA. MILANESE, ROSSI, AND BRUSASCO
60 Mast cella/mrn" Fig. 1. Numbers of mast cells in subgroupsof patientswith (LAR+ = closed columns) or without (LAR- ~ open columns) late-phase asthmatic reaction to allergen inhalation challenge. Horizontal lines are medians. columns indicate the lower and upper quartiles. Asterisk indicates sign ificantly different (p < 0.05)from the LAR- subgroup of the 24-hgroup and from the LAR+ subgroup of the 4-h group.
40
20
o o
LJ..L.=----'*'--- -'--- - - - - - ' - - - - ' - - - - - ' - - -
10
20 30 Late FEV, fall (0/0)
50
40
Bronchial Biopsy Studies The numbers of mast cells observed in bronchial biopsy specimens are shown in figure 1.Within the 24-h group the number of mast cells was significantly (p < 0.05) higher in the LAR + subgroup than in the LAR- subgroup. In addition, in the LAR + subgroup of the 24-h group the number of mast cells was significantly (p < 0.01)higher than in the LAR + subgroup of the 4-h group and tended to be higher (P = 0.08) than in the LAR + sub-
group of the control group. Furthermore, in the 24-h group the number of mast cells was significantly correlated (p < 0.001)with the maximal late fall of FEV I observed after allergen inhalation challenge (figure 2). When the proportion of degranulating mast cells to total mast cells was evaluated (table 2), it appeared that in the 4-h group this was significantly (p < 0.001) higher than in the other groups, irrespective of the type of asthmatic re-
TABLE 2
Group Control group 4-h group 24-h group p values. 2 x 3 tables
LAR-
LAR+
01184 32/56 0/48 < 0.001
24/200 40/64 16/512 < 0.001
(2
p Values tables)
x 2
< 0.001 NS NS
• The number of mast cells lor eech subgroup was obtained by adding the number of mast cells of all individual patients belonging to that subgroup.
TABLE 3 TOTAL CELLS. MONONUCLEAR CELLS, EOSINOPHILS. AND MORPHOLOGIC ABNORMALITIES'
Subgroups Control group LARLAR+ 4-h group LARLAR+ 24-h group LARLAR+
Total Cells per mm2 (x 10')
Mononuclear Cells per mm2 (x 10')
10 (IH1) 15 (6-17)
8 (8-9) 8 (6-10)
6.5 (5.5-9.5) 12 (10.5-12.5)
5 (4--8) 7 (6.5-8.5)
8.5 (6-12) 13 (10-15)
5 (5-6.5) 8 (7-8)
Eosinophils per mm2
3 (0-6) 8 (3-25.5)
6 (3-9) 20.5 (13-28) 5 (3-18.5) 19 (8-48)
Airway Responsiveness In the LAR + subgroup of the 4-h group, but not in the LAR- subgroup, methacholine POlS (geometric mean) was decreased 3 h after allergen challenge, from 89 to 32 IJ.g (p < 0.005). Similarly, in the LAR + subgroup, but not in the LARsubgroup, of the 24-h group, methacholine POlS was decreased23 h after allergen challenge, from 44 to 24 IJ.g (p = 0.05). Discussion
PROPORTION OF DEGRANULATING TO TOTAL MAST CELLS' DegranulatingiGranulated (per mmZj
sponse; by contrast, within the control group, the proportion of degranulating mast cells was significantly (p < 0.001) higher in the LAR + subgroup than in the LAR- subgroup. Other data of bronchial biopsy specimens are shown in table 3. The number of total cells was significantly higher (P < 0.05) in the LAR + subgroups than in the LAR- subgroups, without differences among groups (p > 0.05). The number of eosinophils tended to be higher in the LAR + subgroup than in the LAR - subgroup of the 4 h group (p = 0.06). The number of mononuclear cells and the morphologic abnormalities score were similar in all groups and subgroups (p > 0.05, all comparisons).
Morphologic Abnormalities Score
3 (1-3) 4 (3-7) 2 (1.5-2.5) 3 (3-3.5) 3 (1.5-4) 3
(3-4)
• The number of total cells was significantly different (p < 0.05) betwesn subgroups but not betwesn groups (ANOVA). Values are median (lower and upper quartiles).
The results of the present study confirm that mast cell degranulation is associated with the bronchial response to inhaled allergen. In addition, wehave shown that the events determining the occurrence and the intensity of the late-phase asthmatic response are also a stimulus for mast cell recruitment in the bronchial mucosa. Before discussing our results we must comment on methodology. Both control patients and patients who were studied 4 h after allergen challenge were divided in subgroups depending on whether they did or did not develop the late-phase response to allergen challenge on a prestudy day. This was done because on study days patients received medications before and after bronchoscopy; this might have prevented or blunted the latephase response. Wethink the assumption that these patients would have developed a late-phase response also on study days is justified for two reasons. First, the dose of allergen able to provoke the earlyphase response was similar on the study day and on the prestudy day. Second, patients known to have late-phase response, but not single responders, showed an increased airway responsiveness 3 h after allergen challenge. An increase of airway responsiveness is associated,as also shown
MAST CELLS AND THE LATE ASTHMATIC REACTION
in our patients studied 24 h after allergen challenge, with the late-phase response and often precedes its development (13, 14). A limitation that must be kept in mind in interpreting the results of the present study is that we used formalin as standard fixation method; this method does not allow the identification of either basophils or formalin-sensitive mast cells. Therefore, no conclusion can be drawn about the role of these cells in the airway exposed to allergen. There isgood agreement that mast cells have a role in initiating the bronchial inflammatory reaction to allergen exposure (15-19). A relationship has been established between the degree of sensitivity to methacholine, the number of mast cells, and the level of histamine in bronchoalveolar lavage (17). Whether these cells playa role in perpetuating the inflammatory processes that sustain the asthmatic disease remains, however, still unknown. Most studies aimed at investigating the inflammatory events that follow the inhalation of allergen were carried out by bronchoalveolar lavage. In patients with dual response, but not in patients with single early response, Diaz and coworkers (8) showed a significant decrease of the number of mast cells in bronchoalveolar lavage performed 6 h after allergen inhalation as compared with that at baseline. These investigators postulated two hypotheses to explain their findings. The first hypothesis is that inflammatory cell-derived factors associated with the development of the late-phase response may induce, in a nonspecific way, mast cell degranulation. The second hypothesis is that luminal mast cells activated during the development of the late-phase response are more adherent to bronchial epithelium and, therefore, more difficult to dislodge by the lavage procedure. Indeed, the use of bronchoalveolar lavage for studying mast cells has some inherent limitations. First, the number of mast cells that can be recovered by BAL in asthmatic patients is usually small and most times below 1070 the total cells; second, degranulated mast cells are unrecognizable; third, mast cells tend to be trapped among epithelial cells. Bronchial biopsy seems to be a more suitable method to study mast cells in asthmatic patients. By using this technique, Beasley and coworkers (9) found mast cells with various stages of degranulation in the bronchial lamina propria of asthmatics but not in that of healthy subjects. Consistent with these findings
1285
we have shown that the percentage of degranulating mast cells observed 4 h after allergen challenge was higher than in control subjects. Moreover, both the total number of mast cells and the proportion of degranulating mast cellswas similar in patients with and in patients without late-phase response studied 4 h after allergen challenge. These findings suggest that mast cell degranulation plays a role in the early-phase response, whether or not the late-phase response will occur. The increase of bronchial responsiveness that precedes the late-phase response seems, therefore, not to be related to an enhanced mast cell degranulation. Nevertheless, the finding that in the control group patients with late-phase response had a larger proportion of degranulating mast cells than did patients without late-phase responses, supports the suggestion that baseline hyperresponsiveness is sustained by ongoing mast cellsdegranulation (17). An important finding of the present study is that 24 h after allergen challenge the number of mast cellswas significantly higher in patients who had developed the late response than in patients who had not. In addition, unlike the mast cells observed 4 h after allergen challenge, the majority of these mast cells were granulated. These findings indicate that the occurrence of a late-phase response is associated with some stimuli, leading to an increase of the number of mast cells in the bronchial mucosa. These events may include the migration of mature mast cellsinto the mucosal epithelium, the rapid maturation of immature cells, and the differentiation of other cellsalready present. Several cytokines (IL-4, IL-3, OMCSF, IL-5) can promote growth, differentiation, and phenotypical switches in mast cells (20). In patients without latephase response 24 h after challenge the number of mast cells remained low, whereas granulation was complete, suggesting regranulation of the existing mast cells. In a canine experimental model (21) mast cells have been shown to increase rapidly, i.e., within 1 h, in the peripheral airways after exposure to specific stimuli and in the central airwaysafter exposure to both specific and nonspecific stimuli. The fact that both specificand nonspecific stimuli caused mast cells to increase in the tracheal epithelium independent of the type of stimulus and of the dose inhaled suggested that this response might be the result of epithelial irritation or damage. In our study, mast cells were
found to be increased 24 but not 4 h after allergen challenge; furthermore, the stimuli causing mast cell migration seem to be specific and dose-dependent since the mast cell increase was related to the occurrence and the severityof the preceding late asthmatic response. These differences between our study and that of Turner and coworkers (21) may be due to a difference in bronchial allergic response between humans and dogs. Previous studies in patients with rhinitis have shown that the mast cell number in nasal mucosa correlates with the clinical symptoms of allergy (22), and the patients with greater sensitivity to pollen have more intraepithelial mast cells during the pollen season (23). Our findings suggest that, like prolonged natural exposure, also the experimental exposure to the sensitizing allergen may cause mast cells to increase in the exposed tissues provided that the allergenic stimulus is sufficiently strong to trigger the inflammatory process associated with the latephase response. We found no significant differences in the number of other inflammatory cells in bronchial biopsy by comparing patients with and patients without the latephase response. However, the number of eosinophils in patients with a late-phase response tended to be higher than in patients without a late-phase response 4 h after allergen challenge. Even though statistical significance was not achieved, this finding is in agreement with the increase of eosinophils in bronchoalveolar lavage shown in a separate report (12). The possible interaction between the kinetics of mast cells and eosinophils deserves further investigation. In conclusion, we have shown that the number of mast cells is increased in the bronchial mucosa 24 h after the experimental inhalation of allergen only in patients who developed a late-phase response. The close correlation between the severity of the late-phase response and the number of mast cells in the bronchial mucosa suggests that during the latephase bronchial response to allergen a release of factors that cause the subsequent mast cellincrease occurs. Whether a similar increase of mast cells in bronchial mucosa is also induced by natural exposure to allergen remains to be elucidated. References I. Abraham WM, Perruchoud AP, Sielczak MW, YergerLD, Stevenson J8. Airwayinflammation during antigen-induced late bronchial obstruction. Prog Respir Res 1985; 19:48-55.
1286
ST. Cellular events in bronchi in mild asthma and 2. O'Byrne PM, Dolovich J, Hargreave FE. Late asthmatic response. Am Rev Respir Dis 1987; after bronchial provocation. Am Rev Respir Dis 1989; 139:806-17. 136:740-51. 10. Quanjer PH. Standardized lung function test3. De Monchy JGR, Kauffman HF, Venge P, et al. Bronchoalveolar eosinophilia during allergen- ing. Report of working party on standardization induced late-phase asthmatic reactions. Am Rev of lung function tests: European Community for Coal and Steel. Bull Eur Physiopathol Respir 1983; Respir Dis 1985; 13l:373-6. 19(5uppl:5:1-94). 4. Metzger WJ, Richerson HB, Worden K, Mon. Crimi E, Brusasco V, Losurdo E, Crimi P. nick M, Hunninghake GW. Bronchoalveolar lavage of allergic asthmatic patients following allergen Predictive accuracy of late asthmatic reaction to Dermatophagoides pteronyssinus. J Allergy Clio bronchoprovocation. Chest 1986; 89:477-83. Immunol 1986; 78:908-1003. 5. Metzger WJ, Zavala D, Richerson HB, et al. 12. Rossi GA, Crimi E, Lantero S, et al. The lateLocal allergen challenge and bronchoalveolar laphase asthmatic reaction to inhaled allergen is asvage of allergic asthmatic lungs. Am Rev Respir sociated with an early recruitment of eosinophils Dis 1987; 135:433-40. in the airways. Am RevRespirDis 1991; 144:379-83. 6. Brusasco V, Crimi E, Gianiorio P, Lantero S, 13. Thorpe JE, Steinberg D, Bernstein IL, MurRossi GA. Allergen induced increase in airway relas CG. Bronchial reactivity increases soon after sponsiveness and inflammation in mild asthma. J Appl Physiol 1990; 69:2209-14. the immediate response in dual responding asth7. Wenzel SE, Fowler AA, III, Schwartz LB. Acmatic subjects. Chest 1987; 91:21-5. 14. Durham SR, Craddock CF, Cookson WO, tivation of pulmonary mast cellsby bronchoalveolar Benson MK. Increases in airway responsiveness to allergen challenge. Am Rev Respir Dis 1988; histamine precede allergen-induced late asthmatic 137:1002-8. 8. Diaz P, Gonzalez MC, GalleguiIlos FR, et a/. . responses. J AllergyClin Immunol1988; 82:764-70. IS. Flint KC, Leung PKB, Hudspith BN, et al. Leukocytes and mediators in bronchoalveolar laBronchoalveolar mast cells in extrinsic asthma: a vage during allergen-induced late-phase asthmatic reactions. Am Rev Respir Dis 1989; 139:1383-9. mechanism for the initiation of antigen specific bronchoconstriction. Br Med J [C1in Res] 1985; 9. Beasley R, Roche WR, Roberts JA, Holgate
CRIMI, CHIARAMONDIA, MILANESE, ROSSI, AND BRUSASCO
291:923-6. 16. Kay AB. Provoked asthma and mast cells.Am Rev Respir Dis 1987; 135:1200-3. 17. Casale TB, Wood D, Richerson HB, Zehr B, Zavala D, Hunninghake GW. Direct evidence of a role for mast cells in the pathogenesis of antigeninduced bronchoconstriction. J Clin Invest 1987; 80:1507-11. 18. Kaliner M. Asthma and mast cell activation. J Allergy Clin Immunol 1989; 83:510-20. 19. Friedman MM, KalinerMA. Human mast cells and asthma. Am Rev Respir Dis 1987;135:n57-64. 20. Denburg JA, Dolovich J, Harnish D. Basophil mast cell and eosinophil growth and differentiation factors in human allergic disease. Clin Exp Allergy 1989; 19:249-54. 21. Turner CR, Kolbe J, Spannhake EW. Rapid increase in mast cell numbers in canine central and peripheral airways. J Appl Physiol1988; 65:445-51. 22.• Otsuka H, Denburg J, Dolovich J, et al. Heterogeneity of metachromatic mast cells in the human nose: significance of mucosal mast cells. J Allergy Clin Immunol 1985; 76:695-702. 23. Enerback L, Pipkorn U, Granerus G. Intraepithelial migration of nasal mucosal mast cells in hay fever. Int Arch Allergy Appl Immunol1986; 80:44-51.
