INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Local eosinophilia and activation of eosinophils are characteristic features of allergic airway (1). Recent investigations have provided some understanding of the development and selective recruitment of eosinophils from the bone marrow into local tissue sites. Granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin-5 (IL-5) play a crucial role in hematopoiesis, eosinophil maturation and activation, and prolonged eosinophil survival (4). Elevated GM-CSF and IL-5 mRNA and protein expression, as well as increased numbers of eosinophils, have been detected in nasal mucosa and/or secretions of patients with allergic rhinitis (5). In this disease virtually all mucosal eosinophils may exhibit distinct signs of degranulation (3). Presence of GM-CSF and IL-5 in nasal biopsies and secretions indicates that these cytokines may be important in allergic eosinophilic airway disease in vivo.

Topical allergen provocation, similar to the reaction to natural allergen exposure (10, 11), increased levels of mucosal mRNA as well as protein levels of GM-CSF and IL-5 (5, 12). The nasal cytokines are produced in picogram quantities and, therefore, may not be readily measured in nasal lavage fluids that typically involve application of many milliliters of fluid on the mucosal surface. However, small filter paper strips applied on the nasal mucosa have successfully recovered measurable levels of nasal cytokines after allergen challenge (7, 13, 14). This technique samples both surface and tissue fluids from airway mucosa (15), which may be advantageous in studies of cytokines that to a significant extent would be released beneath the epithelial barrier.

In agreement with observations by Alam and coworkers (13) and Sim and coworkers (7), we recently demonstrated that "filter paper levels" of GM-CSF increase markedly during the nonsymptomatic allergen-challenge-induced late phase of nasal allergic response (14). We also demonstrated that the capacity of the nasal mucosa of subjects suffering from seasonal allergic rhinitis to produce GM-CSF was similar during and outside the allergic season. Furthermore, we noted that this sampling process was noninjurious, in the sense that no increase in exudative inflammatory indices such as ₂-macroglobulin (16) occurred in baseline samples of nasal secretions (C. Svensson and coworkers, unpublished data).

Topical corticosteroids have proved to be a highly efficacious treatment of allergic rhinitis (17). Maintenance treatment with topical steroids exerts a range of antiinflammatory nasal effects. These include inhibition of pleiotropic proeosinophilic cytokines GM-CSF and IL-5 (11, 21). However, little is known about acute effects of nasal steroids. Particularly, there have been no data on effects in the nose that would correspond to inhibition of the asthmatic late phase response produced by single dose pretreatment with inhaled steroids (22).

In the present study we explored acute effects of a single dose of topical budesonide (256 µg; 128 µg in each nostril) on nasal mucosal output of cytokines after allergen challenge in patients with allergic rhinitis. Nasal mucosal fluids were collected at various time points after allergen challenge using filter paper strips for noninjurious sampling of mucosal surface and tissue fluids. We selected GM-CSF as our prime variable since we (14) and others (7, 13) have previously demonstrated robust increases in nasal levels of this cytokine in allergen challenge experiments. In a previous study Sim and coworkers (7) reported inconsistent effects of allergen challenge on nasal output of IL-5, which is considered to be a particularly important cytokine in allergic eosinophilic airway disease (23). Hence, in addition, we further explored the effects on IL-5 of allergen challenge and examined if any increase in IL-5 would be sensitive to the present steroid pretreatment.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Nine nonsmoking subjects (six men/three women; 20-36 yr of age) with a history of seasonal allergic rhinitis due to birch (n = 2) or timothy (n = 7) pollen allergy and positive skin prick test were recruited from the ENT clinic, University Hospital, Lund, Sweden. They were asymptomatic out of the pollen season and had not used any antiallergic medication for at least 2 mo before the study started. Protocol and consent forms were reviewed and approved by the ethics committee, and all subjects gave their written, informed consent to participate.

Study Design

A randomized, placebo-controlled, double-blind, crossover study was performed to examine the effects of single-dose treatment with intranasal budesonide or placebo on the recovery of IL-5 and GM-CSF in nasal secretions after allergen challenge. The study was performed out of the Swedish pollen season (1998). Once entered into the study, the subjects were randomly assigned to one of the two groups (active drug or placebo). They were started on two sprays 64 µg per nostril: 256 µg total dose of budesonide (Rhinocort Aqua Nasal Spray; AstraZeneca R&D, Lund, Sweden) or placebo 30 min prior to allergen challenge. The placebo drug was identical in taste and appearance to the active drug. Allergen challenges were performed with birch or timothy pollen extracts; 10,000 SQ/U per nostril of Aquagen (ALK, Denmark) was administered with a mechanical pump spray. Both recovery of nasal secretions and symptom scoring were done before challenge and at different time intervals up to 9 h after allergen challenge. A washout period of 4 wk followed before the subjects were crossed over to the other treatment and another allergen challenge.

Nasal Symptoms

Nasal symptoms were recorded before and after two nasal lavages with saline, and 30 min, 1, 3, 5, 7, and 9 h after allergen challenge. Nasal itching, secretion, and blockage were scored from 0 to 3 (0 = no, 1 = mild, 2 = moderate, 3 = severe symptoms). Furthermore, the number of sneezes was counted and transformed into a score (0 = 0, 1 = 1-4, 2 = 5-9, 3 = 10 or more sneezes). A total symptom score was then calculated by addition of the four scores.

Recovery of Nasal Mucosal Fluids

Nasal fluids were collected at various time points by using filter paper strips, a modified "matrix method" described by Alam and coworkers (13). Briefly, two strips (size 7 × 30 mm/strip) of filter paper (Whatman No. 40; Whatman Paper Ltd, Maidstone, UK) were placed, one on the nasal septum and one on the inferior turbinates (two strips in each nostril), for 10 min. The filter strips were weighed before and after the application for estimation of nasal secretion amount. Nasal fluids were collected before (for "baseline" cytokine level estimations), after two nasal lavages with saline, and 20 min after active drug or placebo treatment (for "preallergen challenge" cytokine level estimations), and 30 min, 1, 3, 5, 7, and 9 h after allergen challenge. After removal and weighing, the filter paper strips (four from each time point and subject) were air dried at room temperature and stored at 70° C until assayed.

GM-CSF and IL-5 in Nasal Secretions

Cytokines from filter strips from all subjects were eluted simultaneously the same day. Four filter paper strips (from right and left nostrils from a single subject and time point) were cut into approximately 5 mm pieces. They were placed into a volume of 500 µl wash buffer (0.1 M Tris buffer, pH 7.4, 0.01% sodium azide, 0.002% Tween 20, and 0.3% human serum albumin), and gently shaken overnight at 4° C. After centrifugation at 400 × g for 15 min, the eluate was collected and stored at 70° C until cytokine analyses. Enzyme-linked immunosorbent assay (ELISA) of GM-CSF and IL-5 was performed using specific ELISA kits (Quantikine immunoassays [R&D Systems, UK]; detection limit 7.8 pg/ml and 4 pg/ml, respectively). Samples from a single subject were analyzed simultaneously. Samples and cytokine standards were assayed in duplicate according to the manufacturer's instructions.

Statistics

Differences between each time point were analyzed using Friedman's test. If any statistical significance emerged, Wilcoxon's sign rank test was used for further analyses. Wilcoxons' signed rank test was also used for comparisons between paired data. Results of the study were presented as mean and SEM. Correlations between IL-5 and GM-CSF were tested by calculation of the Spearman correlation coefficient r. p Values of < 0.05 were considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nasal Symptoms

Nasal provocation caused immediate symptoms peaking 30 min after allergen challenge (Figure 1; p = 0.007 compared with baseline). Although the total symptom score consistently continued to be somewhat higher (1 h, p = 0.007; 3 h, p = 0.02; 5 h, p = 0.02; 7 h, p = 0.02; 9 h, p = 0.07) after allergen challenge in comparison with the baseline symptoms, there was no clear late phase symptomatic response. Also, budesonide had no effect on nasal symptoms compared with placebo pretreatment (p > 0.05) (Figure 1).

View larger version (25K): [in this window] [in a new window]   Figure 1.   Total nasal symptom scores after placebo or budesonide single dose pretreatment and after allergen challenge. Values presented are mean and SEM at each scoring period for all subjects (n = 9).

GM-CSF and IL-5 in Nasal Fluids

After placebo pretreatment, GM-CSF levels were increased in the nasal mucosal fluids (Figure 2). This increase was already significant in the 3 h sample (p  0.02, compared with the prechallenge levels) and elevated GM-CSF levels were maintained at 9 h. Pretreatment with a single dose of budesonide significantly depressed the allergen challenge-induced increase in GM-CSF (p < 0.05, compared with GM-CSF levels obtained after placebo pretreatment) (Figure 2).

View larger version (25K): [in this window] [in a new window]   Figure 2.   Serial levels of GM-CSF in nasal mucosal fluids of allergic subjects after placebo or budesonide single dose pretreatment and after allergen challenge. Values presented are mean and SEM at each sampling period for all subjects (n = 9). *p < 0.05.

IL-5 levels were similarly increased (p  0.02, compared with the prechallenge levels) (Figure 3) and similarly suppressed by the budesonide pretreatment even at 9 h (p < 0.05, compared with IL-5 levels obtained after placebo pretreatment) (Figure 3). There was a significant positive correlation between the IL-5 and GM-CSF levels after placebo (r = 0.791; p < 0.001) or budesonide (r = 0.961; p < 0.001) pretreatment.

View larger version (21K): [in this window] [in a new window]   Figure 3.   Serial levels of IL-5 in nasal mucosal fluids of allergic subjects after placebo or budesonide single dose pretreatment and after allergen challenge. Values presented are mean and SEM at each sampling period for all subjects (n = 9). *p < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study demonstrates that off-season allergen challenge of individuals with seasonal allergic rhinitis evokes a nonsymptomatic nasal late phase response involving markedly elevated mucosal fluid levels of IL-5 and GM-CSF. Furthermore, these cytokines are inhibited by budesonide given as a single clinical dose just prior to the challenge. These data are of interest in part due to the potential importance of cytokines such as IL-5 in the pathogenesis of allergic airway disease, and in part due to the current paucity of experimental clinical models that readily unravels therapeutic potency of airway steroids.

In the present study, markedly increased levels of GM-CSF occurred as a late phase reaction in response to allergen challenge in allergic rhinitis. This observation agrees with previous reports of Alam and coworkers (13), Sim and coworkers (7), and Linden and coworkers (14). Furthermore, we demonstrated a significant coappearance of nasal mucosal IL-5. These latter data are at variance with the study of Sim and coworkers (7) who reported inconsistent effects of allergen challenge on nasal IL-5 that occasionally appeared to increase and exhibit "maximum levels" by 30 min after challenge in their study. This discrepancy may be due to differences in the experimental methodology, such as use of IL-5 ELISAs with different sensitivity. Sim and coworkers (7) included treatment with topical decongestants, which were not used in the present study. However, it appears unlikely that a vasoconstrictor should selectively affect IL-5 in this kind of study.

Contrary to the late phase response in GM-CSF and IL-5 production, single nasal challenge with pollen allergens in patients with seasonal allergic rhinitis caused immediate symptoms but no clear late phase symptomatic response. Whether dual responses occur in human nasal airways is controversial. It appears that selected individuals may experience a recurrence of symptoms after allergen challenge (24), but this feature seems highly variable (28). Differing from the present design, previous studies that do report the occurrence of nasal late phase responses involve administration of local nasal decongestants prior to the allergen challenge (25). Because the vasoconstructive effects of these drugs may decline markedly a few hours after allergen challenge (29, 30), a pharmacological artifact may cause a degree of mucosal swelling contributed to any recorded late phase nasal obstruction. Indeed, nasal blockage is the dominating symptom in the previous reports on nasal late phase response (25).

As expected, the acute pretreatment with a single dose of budesonide had no effect on the nasal symptoms observed during immediate nasal response to allergen challenge. This finding is in a good agreement with the work of Andersson and coworkers (31) who demonstrated lack of effect on nasal symptoms after 2 h pretreatment with a single dose of budesonide in a model of allergen-induced allergic rhinitis. Importantly, however, we also demonstrated that a single dose of budesonide within 3 h after its topical application on the human allergic mucosa efficiently inhibited the allergen-challenge-induced mucosal output of GM-CSF and IL-5. Thus, our study presents new evidence that budesonide exerts antiinflammatory airway effects with a very rapid onset of action. The present results also demonstrate that a single-dose pretreatment with topical budesonide is sufficient for inhibition of nasal late phase cytokine generation (3-9 h after allergen challenge), assessed as inhibition of GM-CSF and IL-5 output. Although other studies have provided information on cytokine inhibition by topical corticosteroids in allergic rhinitis, these reports did not address effects of a single-dose steroid treatment. For example, Sim and coworkers (21), Weido and coworkers (32), and Masuyama and coworkers (11) studied nasal output of GM-CSF and/or IL-5 after pretreatment with beclomethasone dipropionate or fluticasone propionate. In these previous studies allergen challenge was administered only after prolonged steroid treatment, that is, 1 wk (21, 32) or 6 wk (11) of topical steroid treatment. Data generated in studies of asthmatic bronchi support the notion that effects obtained a few hours after topical airway application of steroids are therapeutically relevant. Thus, single-dose treatment inhibits allergen-challenge-induced late phase exudation and obstruction in allergic and experimental asthma (16, 33); similarly, in chronic asthma budesonide significantly increases peak expiratory flow rate during the first 8 h after drug inhalation (34). Indeed, the present human airway data suggest that inhibition of cytokine generation may be importantly involved in the clinical efficacy recorded only a few hours after a single dose of airway steroid.

Theoretically, the present reduction of cytokine recovery from the nasal mucosa may have resulted from steroid-induced inhibition of cytokine production, decreased cytokine secretion, and/or reduced recruitment of cells. Although this effect develops within the first few hours after steroid application, it is compatible with a genomic action of these drugs involving inhibition of the expression of genes that encode for cytokines (35, 36). There are several candidates for recruited and resident cellular sources of GM-CSF and IL-5; Th2- and Tc2-like lymphocytes, eosinophils, mast cells, and basophils are regarded as principal producers. Moreover, nasal epithelial cells may also produce GM-CSF (37) as well as IL-5 (38). Previous reports involving patients with allergic rhinitis indicate that GM-CSF mRNA may be colocalized to macrophages, T lymphocytes, and eosinophils (39), whereas IL-5 mRNA is expressed predominantly by T cells (11). These two previous reports (11, 39) also indicate that topical nasal steroids inhibit production of GM-CSF and IL-5 by T lymphocytes and, possibly, by eosinophils. The present data, demonstrating excellent correlation between GM-CSF and IL-5, their levels, their time courses of appearance, and their inhibition by budesonide, suggest that these two cytokines are produced jointly by steroid-sensitive nasal mucosal cells. Further studies are warranted to identify the type of the cells involved in nasal late phase production of GM-CSF and IL-5.

We conclude that allergen-challenge-induced nonsymptomatic nasal late phase responses include increased nasal output of IL-5 and GM-CSF. Also, a single dose of a topical nasal steroid inhibits this cytokine response, in this case demonstrating that budesonide exerts an antiinflammatory nasal effect within 3 h after its topical application. We suggest that the present experimental model can be used to explore both cytokine involvement in allergic airway inflammation and potency as well as onset of action of airway glucocorticoid compounds.