|
 |
Contents |
|---|
TOP
Contents
ASTHMA AND AIRWAY BIOLOGY
ALLERGIC RHINITIS
REFERENCES
|
|---|
Asthma and Airway Biology
Genetics (16)
Epidemiology (3)
Airway Inflammation
Animal Models (5)
Induced Sputum (5)
Bronchoalveolar Specimens (4)
Blood (3)
Exhaled Nitric Oxide (6)
Other Exhaled Markers (3)
Ex Vivo Studies (2)
Airway Hyperreactivity
Animal Models: Antigen Challenge (7)
Animal Models: Other Challenges and Mediators (9)
Ex Vivo Studies (5)
Early and Late Asthmatic Responses (5)
Antigen and Methacholine Challenge (3)
Exercise-Induced Asthma (6)
Drugs (1)
Other Pathophysiological Mechanisms in Asthma
Tachykinins, Neural Activity (8)
Deep Inspiration Bronchoconstriction (3)
Blood Flow (2)
Immunology (1)
Airway Narrowing (4)
Remodeling (1)
Classic Mechanics (2)
Treatment
-Agonists (1)
Inhaled Glucocorticoids (7)
Glucocorticoids (1)
Muscarinic Antagonists (2)
Leukotriene Inhibitors (1)
Combination Regimens (5)
Specific Clinical Scenarios
Nocturnal Asthma (3)
Acute Severe and Fatal Asthma (4)
Dyspnea (2)
Quality of Life (2)
Age (1)
Differential Diagnosis: Eosinophilic Bronchitis (1)
Occupational Asthma
Diisocyanates (2)
High Molecular Weight Agents (1)
Farmers (1)
Prevalence and Severity (2)
Natural History (2)
Allergic Rhinitis
Risk Factors (2)
Quality of Life (1)
Inflammation and Hyperreactivity (6)
Treatment (4)
References
| |
ASTHMA AND AIRWAY BIOLOGY |
|---|
|
TOP
Contents
ASTHMA AND AIRWAY BIOLOGY
ALLERGIC RHINITIS
REFERENCES
|
|---|
Genetics
Because several reports have implicated the chromosomal region 5q31-q33 in harboring the gene for the 2-receptor, Ulbrecht and coworkers (1) undertook an association study focusing on bronchial hyperresponsiveness as a qualitative trait.
From a population sample of 1,150 people, all 152 hyperreactive probands (index cases) were extracted and compared with
295 control subjects who had normal bronchial reactivity. All
individuals were genotyped for three single nucleotide polymorphisms of the 2-receptor gene, resulting in variants at the
amino acid positions 16, 27, and 164. No individual polymorphism (an allele with a frequency of at least 1%) was associated
with bronchial hyperreactivity, but the Gly16/Gln27/Thr164
haplotype was underrepresented in the case group, indicating
that this haplotype protects against hyperreactivity.
The FCER2 gene encoding the low-affinity receptor for IgE has been mapped to chromosome 19p13 and is implicated as a candidate for IgE-mediated allergic disorders. Laitinen and coworkers (2) studied three phenotypic markers (asthma, total serum IgE, and IgE specific for aeroallergens) in 124 families from Finland and Catalonia. Eight polymorphic markers in a region around the FCER2 gene were analyzed. The allele and haplotype association study revealed a suggestive haplotype association for a high IgE response. (A haplotype consists of closely linked alleles on a single chromosome that are usually inherited as a group and determine a particular phenotype.) In families with two or more affected members, a single haplotype was highly enriched but sequence polymorphisms were not detected. The authors conclude that chromosome 19p13 might harbor a genetic determinant of IgE-related traits.
Because variants of the FCER1B gene are associated with
increased risk of atopy and asthma, Hizawa and coworkers (3)
screened 24 patients with asthma for new mutations. A common
109C/T polymorphism at the promoter region of FCER1B
was identified. Genotyping of this promoter polymorphism in
226 asthmatic and 226 control subjects revealed similar distribution. Homozygosity for the 109T allele was associated
with increased IgE in the asthmatics. The authors conclude
that the 109C/T polymorphism of the FCER1B promoter region is one of the genetic factors affecting serum IgE in the Japanese population.
Because total IgE is related to active smoking in the general population, Oryszczyn and coworkers (4) studied 122 asthmatic probands, 430 first-degree relatives, and 190 control subjects to see whether asthma influenced the relationship. Levels of total IgE in first-degree relatives were intermediate between cases and control subjects, and current smokers had higher levels than never smokers. IgE was increased in female first-degree relatives in relation to passive smoking. The authors conclude that active smoking was related to higher total IgE independently of gender, or personal or family history of asthma, and that active smoking increased IgE even in asthmatic subjects.
Genes on chromosomes 5q and 11q exhibit the most consistent association with the asthma-related phenotypes of bronchial hyperresponsiveness and atopy. Few studies have focused on genes protecting against oxidative stress, which plays a critical role in airway inflammation. Fryer and coworkers (5) undertook an association study to determine whether allelic variation at the glutathione S-transferase GSTP1 locus influences expression of bronchial hyperreactivity and atopy. (Locus is the site or location of a gene in a chromosome.) The enzyme encoded by GSTP1 uses a variety of lipid and DNA products of oxidative stress, and polymorphic variants of this gene are associated with altered catalytic function of this enzyme. The frequency of GSTP1 Val105/Val105 was significantly lower in asthmatic than in control subjects. The presence of this genotype conferred one-sixth the risk of asthma than did GSTP1 Ile105/Ile105. Val105 homozygotes had one-tenth the risk of atopy defined by IgE level. The authors conclude that GSTP1 polymorphism is strongly associated with asthma, and provides an alternative explanation for linkage of chromosome 11q13 with bronchial hyperreactivity and atopy.
Because it is not known whether serum IgE and airway hyperresponsiveness arise from distinct or shared genetic determinants, Palmer and coworkers (6) investigated the genetic and environmental components of the variance of these two traits in 232 Australian nuclear families. Total serum IgE had a narrow-sense heritability (h2N of 47%); specific serum IgE had an h2N of 34%; and airway hyperreactivity had an h2N of 30%. Extended modeling revealed that the genetic determinants of total and specific IgE had 70% overlap, whereas determinants of IgE and airway hyperresponsiveness had less than 30% sharing. The authors conclude that the pathophysiologic traits associated with asthma have multiple genetic determinants and that airway hyperresponsiveness is genetically distinct from atopy.
Because a marker locus on chromosome 5q23-31 is linked
to asthma in Hutterites, Summerhill and coworkers (7) assessed whether the 2-adrenergic receptor gene is the 5q-linked asthma locus in this population. Genotyping 361 individuals for polymorphisms in the receptor gene showed that
neither Arg16Gly nor Gln27Glu polymorphisms were linked
to bronchial hyperresponsiveness or serum IgE. Individuals
homozygous for the Arg16 allele had lower FEV1 and forced
vital capacity (FVC). The authors conclude that the observed
linkage to asthma in the 5q23-31 region is not related to variation in the 2-adrenergic receptor gene.
Atopy has been linked to the chain of the high-affinity
IgE receptor on chromosome 11q13, and polymorphisms of
that gene (I181L, V183L, and E237G) are reported to be associated with asthma and atopy. Thomas and coworkers (8) investigated linkage to the IgE receptor in a sample of 131 families recruited at random and 109 families found via an
asthmatic proband. Four microsatellite markers were genotyped, including the IgE receptor. The frequencies of I181L,
V183L, and E237G polymorphisms were determined. No evidence for linkage to the high-affinity IgE receptor was found.
The authors conclude that their study fails to strengthen the
evidence for a candidate gene on chromosome 11q13.
In 381 twin pairs (183 monozygous, 198 dizygous), Clarke and coworkers (9) measured asthma (diagnosed by a physician), atopy (skin tests), and bronchial hyperresponsiveness. Associations between monozygous pairs were greater than between dizygous pairs as shown by their respective odds ratios: asthma, 26 versus 2; atopy, 15 versus 3; and bronchial hyperreactivity, 14 versus 4. The associations between each pair of traits within an individual were slightly greater than the association between one trait in a twin and the other trait in the cotwin (cross-trait cross-pair concordance). The authors conclude that the strong cross-sectional associations between these three traits are due to overlap between the genetic factors involved in each of the traits.
The role of Clara cell secretory protein in immunomodulation together with its chromosomal location suggests that the governing gene may be involved in the inheritance of asthma. To address this issue, Laing and coworkers (10) examined the relationship of the A38G polymorphism to gene expression of the Clara cell secretory protein in 100 children. Levels of Clara cell secretory protein were 20% lower in asthmatic children than in healthy controls, and 40% lower in 38A/38A subjects than in 38G/38G subjects; the predisposition, however, was reduced on correcting for plasma levels of the Clara cell secretory protein. The authors conclude that the 38A sequence was associated with less Clara cell secretory protein, and individuals with lower levels of this protein were more likely to have asthma.
An increase in exhaled nitric oxide is a critical component of the asthmatic phenotype, but levels of nitric oxide are very variable among patients identified by a standard case definition of asthma. Because neurologic disorders are known to be associated with allelic variations of trinucleotide repeat sequences, Wechsler and coworkers (11) assessed whether exhaled nitric oxide is associated with various alleles at a trinucleotide repeat locus in a prominent asthma candidate gene (neuronal nitric oxide synthase, or NOS1). Individuals having at least 12 AAT repeats at a locus in intron 20 of the nitric oxide synthase 1 gene had both lower and less variable levels of exhaled nitric oxide. The addition of this genotype to the case definition of asthma allows a uniform cohort of patients to be identified that otherwise look phenotypically similar. The authors conclude that the data provide a link between a genotype in a candidate gene and an important component of the asthmatic phenotype.
The variation in the response to certain pharmacological agents has been shown to be genetic in nature, but equivalent studies have not been conducted to explain the marked variation in response to bronchodilator agents. Niu and coworkers (12) investigated the degree of familial clustering for the response to inhaled albuterol in 1,161 index families with asthma in China. Each family unit consisted of a mother, a father, a first offspring, and a second offspring. Bronchodilator response showed correlations for: father-first offspring pairs, mother-first offspring pairs, mother-second offspring pairs, and first offspring-second offspring pairs. The correlations (0.09-0.16) suggest a modest degree of familial clustering. The largest odds ratio (3.1) was in a second offspring whose mother and first offspring had a bronchodilator response above the median. The authors conclude that the significant familial aggregation indicates that genetic factors influence bronchodilator responsiveness.
Because regular use of -agonists may produce adverse effects in some patients, Israel and coworkers (13) studied 190 patients with asthma to determine whether polymorphism of the
2-adrenergic receptor might affect the response to regular versus as-needed use of albuterol. Among patients homozygous
for arginine at 2-AR-16 (Arg/Arg), regular albuterol users had
a lower morning peak flow (31 liters per minute) than the as-needed users. No significant differences in outcome between
regular and as-needed use were associated with polymorphisms
at position 27 of the 2-adrenergic receptor or in patients homozygous for glycine at 2-AR-16. The authors conclude that
polymorphisms of the 2-adrenergic receptor may influence airway responses to regular inhaled -agonist treatment.
Studying cells of human airway smooth muscle, Moore and
coworkers (14) examined the influence of two polymorphic
forms of the 2-adrenergic receptor, the Gly16 and Glu27 alleles, on acute and long-term desensitization of the receptor.
When cells were pretreated with isoproterenol, they became less
responsive to subsequent treatment with isoproterenol. When
the data were stratified post hoc by genotype, cells containing at
least one Glu27 allele showed greater acute and long-term desensitization than cells without that allele. Cells containing the
Gly16Glu27 haplotype showed less acute and long-term desensitization than did cells without that haplotype, suggesting that the
influence of Glu27 is not through its association with Gly16. The
Glu27 allele was in strong linkage disequilibrium with the Arg19
allele, and cells containing any Arg19 allele showed greater
acute and long-term desensitization. The authors conclude that
the presence of the Glu27 allele is associated with greater acute
and long-term desensitization of human airway smooth muscle.
The results of gene mapping studies are critically dependent on a standardized definition of the asthma phenotype, and Panhuysen and coworkers developed an algorithm for this purpose in a Dutch population. Celedon and coworkers (15) used a modified version of this algorithm to classify members of 2,756 Chinese families. Among 4,097 Chinese parents, 10% were classified as definite asthma, 7% probable asthma, 15% unclassifiable airway obstruction, 15% chronic obstructive pulmonary disease (COPD), and 39% unaffected. Among 6,424 offspring, 17% were classified as definite asthma, 13% probable asthma, 31% unclassifiable airway obstruction, 4% COPD, and 36% unaffected. The use of this algorithm in a population with a high prevalence of smoking would exclude subjects with asthma who smoke or who have severe airway obstruction from linkage analysis, and hinder the exploration of any potential interaction between genetic factors and smoking.
Dizier and coworkers (16) conducted a genome-wide search in 107 French families having at least two siblings with asthma as part of the EGEA (Epidemiological Study on the Genetics and Environment of Asthma, Bronchial Hyperresponsiveness and Atopy) study. Analysis had two stages: the total sample of 107 families was divided into two independent subsets, and the potential linkages detected in the first subset of 46 families were tested for replication in the second subset of 61 families. A total of 254 markers were typed in the first set and 70% in the second set. Linkage was investigated for asthma and four asthma-related phenotypes: bronchial responsiveness, skin tests, total IgE, and eosinophil count. Three regions were detected: 11p13 for IgE, 12q24 for eosinophils, and 17q12-21 for asthma and skin tests. All regions reported by published genome scans in different populations were examined, and seven were found: 11p13 for IgE, 12q24 for eosinophils, 17q12-21 for asthma and skin tests, 1p31 for asthma, 11q13 for IgE, 13q31 for eosinophils, and 9q13 for bronchial hyperresponsiveness. The authors conclude that they identified three regions involved in asthma-like phenotypes.
A series of review articles focus on the question, "What determines asthma phenotypes?" (17).
Epidemiology
To determine asthma mortality in Hispanic populations in the United States, Homa and coworkers (22) used national vital statistics data for 1990 through 1995. Age-adjusted annual asthma mortality rate, on a per million basis, was Puerto Ricans 40.9, Cuban-Americans 15.8, and Mexican-Americans 9.2; for comparison, non-Hispanic whites 14.7, and non-Hispanic blacks 38.1. The northeast accounted for 81% of asthma deaths among Puerto Ricans. The authors conclude that Hispanics do not represent a uniform group in terms of health outcome.
To estimate the sex-specific incidence of asthma from birth to 44 years, de Marco and coworkers (23) analyzed data on 18,659 participants in the European Community Respiratory Health Survey. During childhood, girls had a lower risk of developing asthma than boys (relative risk [RR]: 0.56 to 0.74). Around puberty, risk was almost equal. After puberty, risk in women was always higher than in men (RR: 1.38 to 5.91). A case-control analysis showed that the greater susceptibility in women was partly explained by their smaller airway caliber. Smoking did not increase risk. The authors conclude that the incidence of asthma shows a sex reversal around puberty.
To determine the relationship between attacks of asthma and the menstrual cycle, Zimmerman and coworkers (24) studied 228 women presenting to an emergency department. Exclusions included pregnancy, hormonal therapy, postmenopausal state, and hysterectomy. Only 13% reported reproductive factors as an asthma trigger. Visits in the preovulatory phase (days 5-11) were more frequent (33%) than in the periovulatory phase (days 12-18; 26%), postovulatory phase (days 19-25; 20%), and perimenstrual phase (days 26-4; 21%). Asthma severity was not related to the cycle. The authors conclude that acute asthma is more frequent during the preovulatory phase of the menstrual cycle.
Airway Inflammation
Animal models.
Inoue and coworkers (25) investigated the
contribution of endogenous nitric oxide to airway inflammation and hyperresponsiveness, and the potential involvement
of interleukin-8. In transformed cells of the bronchial epithelium of humans, nitric oxide donors increased production of
interleukin-8; production was also increased by tumor necrosis
factor-
plus interleukin-1 plus interferon-
, and production
was attenuated by a combination of nitric oxide synthase inhibitors (aminoguanidine and N4-nitro-L-arginine methyl ester
[L-NAME]). Exposure of guinea pigs to ozone induced airway
hyperresponsiveness and increased neutrophils in bronchoalveolar fluid; nitric oxide synthase inhibitors prevented these
changes 5 hours after the exposure, and decreased interleukin-8 mRNA expression in epithelial cells. The authors conclude
that endogenous nitric oxide may play a role in airway inflammation and hyperresponsiveness caused by ozone, presumably
through upregulation of interleukin-8.
0.64). The levels of human interleukin-5 were six times higher in allergic mice than
in mice that had not been reconstituted, and the level was correlated with the degree of airway hyperresponsiveness. The
authors conclude that airway hyperresponsiveness occurs after allergen exposure in this mouse model.
Induced sputum. To develop reference standards for cell counts in induced sputum, Belda and coworkers (30) studied 118 healthy subjects. Adequate samples were obtained in 96 subjects. The mean total cell count was 4.1 × 106 cells per g, and proportions of eosinophils were 0.4%, neutrophils 37.5%, macrophages 58.8%, lymphocytes 1%, and bronchial epithelial cells 1.6%. Female gender and atopy were associated with increased eosinophils.
To provide normal reference values for sputum cells, Spanevello and coworkers (31) studied 114 nonatopic nonsmoking volunteers with normal lung function. An adequate sputum sample was produced by 96 subjects (84%). The cell counts were: macrophages 69%, neutrophils 27%, eosinophils 0.6%, lymphocytes 1%, and epithelial cells 1.5%. Only the macrophages and neutrophils conformed to a normal distribution. The authors conclude that the data serve as reference values for future work. To determine the relationship between severity of asthma and airway inflammation, Louis and coworkers (32) studied induced sputum in 74 asthmatic and 22 control subjects with disease severity classified according to the Global Initiative for Asthma (GINA) guidelines. Asthma severity was related to sputum eosinophilia, eosinophil cationic protein, and, less strongly, to sputum neutrophilia and myeloperoxidase. Tryptase, a marker of mast cell activation, was raised in patients with mild-to-moderate asthma. Despite treatment with high doses of corticosteroids, eosinophilic inflammation was prominent. The authors conclude that the severity of persistent asthma is related to airway inflammation that continues despite treatment with corticosteroids. In macrophages, eosinophils, and neutrophils of induced sputum, Taha and coworkers (33) studied expression of two isoforms of prostaglandin H synthase (also termed cyclooxygenase). Patients with asthma and COPD had increased levels of prostaglandin H synthase-1 (the constitutively expressed isoform) and prostaglandin H synthase-2 (the isoform induced by inflammatory stimuli). Levels were higher for asthma than for COPD, and evidence was also found in the airway epithelium and inflammatory infiltrates. The authors conclude that prostaglandin H synthase-2 may contribute to inflammation in patients with asthma. Gauvreau and coworkers (34) characterized the kinetics of basophil and mast cell recruitment to the airways of atopic asthmatic subjects after allergen inhalation. Of 19 subjects, 14 developed both early- and late-phase responses, and 5 developed only early responses. At 7 and 24 hours after allergen challenge, both groups had increased numbers of sputum eosinophils and basophils. The dual, but not early, responders also had an increased number of activated eosinophils and mast cells. Dual responders had higher levels of allergen-induced basophils than early responders, and sputum basophils were correlated with airway hyperresponsiveness at 24 hours (r = 0.66). The authors conclude that late responders have higher levels of allergen-induced basophils than early responders, which may contribute to airway hyperresponsiveness.Bronchoalveolar specimens. Although interleukin-16 is one of the earliest chemoattractants found after antigen challenge, little is known about its activity beyond 6 hours. To address this issue, Krug and coworkers (35) performed segmental bronchial challenge in 13 patients with mild asthma. Levels of interleukin-16 in bronchoalveolar fluid were increased 24 hours after the challenge. T-cell chemoattractant activity was increased, and it was correlated with anti-interleukin 16 antibodies (r = 0.90) and with the level of Fas ligand expression on CD4+ cells (r = 0.80). The authors conclude that increased chemotactic activity for T cells in patients with asthma 24 hours after a challenge is mainly attributable to interleukin-16.
In patients with atopic asthma, Lamkhioued and coworkers (36) found an increase in monocyte chemoattractant protein-4, a CC chemokine, in bronchoalveolar fluid, and the levels were correlated with eosinophils and eotaxin, a chemokine that causes selective recruitment of eosinophils to sites of inflammation. In bronchial epithelium and submucosa, monocyte chemoattractant protein-4 mRNA and protein was significantly upregulated, and its expression could be induced in these cells by tumor necrosis factor- and interleukin-1. Interferon-gamma acted synergistically in inducing expression
of the chemokine, whereas expression was decreased by dexamethasone. The authors conclude that monocyte chemoattractant protein-4 is upregulated in asthmatic airways and it
plays a role in the recruitment of eosinophils.
Boulet and coworkers (37) studied two groups of patients
with asthma: 16 diagnosed within the preceding 2 years and 16 carrying the diagnosis for at least 13 years. Baseline FEV1 was
similar in the two groups, but PC20 on a methacholine challenge (the provocative dose of methacholine causing a 20%
decrease in FEV1) was lower in the long-standing group than
in the recently diagnosed group (0.44 and 3.37 mg per ml).
Bronchial cell counts for CD3+, CD4+, CD8+, CD25+, EG1+,
CD45ro+, and AA1+ cells were similar in both groups, as was
type 1 and type 3 collagen deposition. Inhaled fluticasone
proprionate (1,000 µg daily) for 8 weeks improved PC20 and
decreased EG1+, EG2+, and AA1+ counts equally in both
groups, but decreased CD45ro+ only in the long-standing group.
The authors conclude that the degree of airway inflammation
and subepithelial fibrosis, and response to inhaled glucocorticoids are similar in patients with recent or long-standing asthma,
although airway responsiveness in patients with long-standing
asthma cannot be reverted to normal.
To determine whether the cellular patterns of airway inflammation differ in atopic and nonatopic asthma, Amin and
coworkers (38) did bronchial biopsies in 13 patients with atopic
asthma, 9 patients with nonatopic asthma, and 7 healthy controls. Eosinophil and mast cell counts were increased in both
asthmatic groups, and higher in patients with atopy. Neutrophils were increased only in nonatopic asthmatic patients. The
number of T-lymphocytes (CD3-, CD4-, CD8-, CD25-positive
cells) was higher in atopic than nonatopic asthmatic patients.
Interleukin-4 and -5 positive cells were found in atopic asthmatic, and interleukin-8 positive cells in nonatopic asthmatic
patients. Atopic asthmatic subjects had greater epithelial damage, and a thicker tenascin and laminin layer. In the atopic
group, epithelial integrity was correlated with the eosinophil
count and the number of CD25-positive cells. The authors conclude that atopic and nonatopic patients with asthma differ in
bronchial mucosal membrane and airway inflammation.
Blood. The view that atopy results from immune deviation towards activation of type 2 (Th2) helper T cells has arisen from studies mainly performed on purified cells. Because this approach may give an incomplete picture, Magnan and coworkers (39) reassessed the Th1/Th2 paradigm in whole blood from 69 subjects. Atopy was characterized by increased production of interleukin-4, which was correlated with total IgE levels, and by impairment of T cells to produce interferon-gamma, which was correlated with the number of positive skin tests. Asthma was also characterized by increased production of interleukin-4 if atopy was present, and, unexpectedly, by overproduction of interferon-gamma, which was related to increased capacity of CD8+ T cells. The number of CD8+ T cells producing interferon-gamma was related to asthma severity, bronchial hyperresponsiveness, blood eosinophilia, and interleukin-12 production. The authors conclude that CD8+ T cells producing interferon-gamma are found in the blood of patients with asthma.
To examine the influence of leukotrienes and leukotriene inhibitors on eosinophil survival, Lee and coworkers (40) obtained eosinophils and lymphocytes from the peripheral blood of patients with asthma. Eosinophil survival was promoted by leukotriene-B4, LTC4, and LTD4, granulocyte-macrophage colony-stimulating factor, and fibronectin. The increased survival of eosinophils induced by granulocyte-macrophage colony-stimulating factor was reversed by blockade of cysteinyl leukotriene receptors, and inhibition of 5-lipoxygenase and 5-lipoxygenase activating protein; these same compounds also increased basal rates of eosinophil apoptosis (programmed cell death). The authors conclude that the autocrine cysteinyl leukotriene pathway supports eosinophil survival.1-Acid glycoprotein is an acute-phase protein containing
five N-linked glycons, which differ in their degree of branching (i.e., relative proportions of di-, tri-, and tetraantennary
glycons). To determine the degree of branching (which affects
the protein's immunomodulatory properties), Van Den Heuvel (41) measured plasma concentrations in three groups of
patients. Atopic patients with asthma and atopic patients without asthma had normal levels of 1-acid glycoprotein, whereas
patients with interstitial lung disease had increased levels.
Only the patients with atopic asthma had altered branching,
and the increased branching correlated with asthma symptoms, FEV1, response to histamine challenge, and eosinophils.
The authors conclude that the changes in branching of 1-acid
glycoprotein in asthma indicate an inflammatory reaction and
differ markedly from the decreased branching of an acute-phase response.
Exhaled nitric oxide. In patients with acute asthma, Hunt and coworkers (42) measured the pH of exhaled vapor condensates. The pH was more than 2 log orders lower in the patients than in the controls. The values were highly reproducible and identical to samples from the lower airway. Glucocorticoid therapy caused the values to return to normal. Airway acidity appears to be relevant to asthma because it causes conversion of the endogenous airway compound, nitrite, to nitric oxide in quantities sufficient to account for the concentrations of nitric oxide in exhaled air, and it accelerates necrosis of human eosinophils. The authors conclude that regulation of airway pH plays a previously unsuspected role in asthma pathophysiology. An editorial commentary by Marshall and Stamler accompanies this article (43).
To determine the usefulness of noninvasive markers of airway inflammation for monitoring the loss of asthma control, Jatakanon and coworkers (44) induced mild exacerbations of asthma in 15 patients by decreasing the dosage of inhaled glucocorticoids to one-quarter of baseline. Over the subsequent 8 weeks, seven patients developed mild exacerbations and eight did not. Sputum eosinophil count was the only measurement at baseline that discriminated between the groups. Decreases in morning FEV1 and peak expiratory flow were correlated with increases in sputum eosinophils and exhaled nitric oxide. The authors conclude that sputum eosinophils are useful in predicting loss of asthma control. To determine whether increased concentrations of nitric oxide in exhaled gas of asthmatics result from increased production or increased effectiveness of clearance from the airway wall to the lumen, Silkoff and coworkers (45) developed a model of nitric oxide diffusion in the airways. The airways diffusing capacity for nitric oxide was four-fold higher in 25 patients with asthma than in 10 healthy controls, but the concentration of nitric oxide in the airway wall was not different. Inhaled beclomethasone produced a decrease in nitric oxide in the airway wall in the patients, but no change in the nitric oxide diffusing capacity. The authors conclude that the increased airways diffusing capacity for nitric oxide in asthmatic patients may reflect upregulation of nonadrenergic, noncholinergic, nitric-oxide producing nerves in the airways in compensation for a decrease in the sensitivity of airway smooth muscle to the relaxant effects of endogenous nitric oxide. The 36% decrease in exhaled nitric oxide concentration that occurs in asthmatic patients performing repeated FVC maneuvers has been attributed to a decrease in neurally derived nitric oxide consequent to the deep breaths. To address this issue, Deykin and coworkers (46) studied 10 healthy subjects. FVC maneuvers repeated every 15 minutes over 1 hour produced a 25% decrease in mixed expired nitric oxide. Repeated measurements of airway resistance in a body plethysmograph had no effect. The authors conclude that the increase in exhaled nitric oxide after an FVC maneuver is related to volume history, rather a mechanism involved in the pathobiology of asthma. Symptoms of asthma often decrease or disappear in adolescence, but many subjects later relapse. van Den Toorn and coworkers (47) asked, "Do subjects in remission still have inflamed airways?" Exhaled nitric oxide was higher in 21 adolescents in remission (with no symptoms or medications for at least 1 year) than in 18 healthy controls (19 and 1 ppb), but not different from 21 patients with current asthma (22 ppb). Adolescents in remission had greater bronchial hyperresponsiveness to both methacholine and adenosine-5'-monophosphate (AMP) than the controls, but less than in patients with current asthma. Exhaled nitric oxide was correlated with responsiveness to AMP but not with responsiveness to methacholine. The authors conclude that airway inflammation persists during clinical remission of asthma. To determine whether the increase in exhaled nitric oxide in patients with asthma is mediated through an increase in eosinophils induced by leukotrienes, Deykin and coworkers (48) did bronchoprovocation challenges with methacholine and leukotriene E4 in 16 subjects with atopic asthma. Sputum eosinophils increased by 4.3% after inhaling leukotriene E4 and fell by 1.4% after methacholine. Exhaled nitric oxide did not change with either challenge. The authors conclude that neither leukotriene E4 nor recruitment of eosinophils is sufficient to acutely increase exhaled nitric oxide in patients with asthma.Other exhaled markers. A reaction between nitric oxide and superoxide anions results in the formation of peroxynitrite (a highly reactive oxidant species), which, in turn, reacts with tyrosine residues to form nitrotyrosine (a stable product). Hanazawa and coworkers (49) found that 15 patients with mild asthma had higher concentrations of nitrotyrosine in exhaled breath condensates than had 15 healthy controls (15.3 and 6.3 µg/ml). The levels were lower in 12 patients with moderate asthma (5.0 µg/ml) and 12 patients with severe asthma (3.3 µg/ml). In patients with mild asthma, nitrotyrosine in breath condensate was correlated with exhaled nitric oxide (r = 0.65). The authors conclude that nitrotyrosine in exhaled breath condensates may be a marker of oxidative stress in the airways of patients with asthma.
Because oxidative stress plays a role in asthma, Paredi and coworkers (50) measured ethane, a product of lipid peroxidation that results from oxidative stress, in 26 patients with asthma. Ethane levels were higher in 12 patients not receiving glucocorticoids (2.1 ppb) than in 14 patients being treated with steroids (0.8 ppb) or 14 healthy controls (0.9 ppb). In patients not receiving steroids, ethane was correlated with nitric oxide (r = 0.55) and with the ratio of residual volume to total lung capacity (r = 0.60). The authors conclude that exhaled ethane is elevated in patients with asthma and is decreased by glucocorticoid therapy. Heme oxygenase is an antioxidant enzyme that catalyzes the oxidative degradation of heme to produce carbon monoxide and biliverdin. The enzyme has two isoforms: heme oxygenase-2 is constitutively expressed, and heme oxygenase-1 is inducible by agents that lead to oxidant stress. Lim and coworkers (51) investigated the expression of the enzymes in bronchial biopsies from 10 subjects with asthma and 10 healthy controls. Both enzymes were widely distributed in the submucosa, especially in the airway epithelium and submucosal macrophages; extent and intensity were the same in patients and controls. After a month of inhaled budesonide, expression of either enzyme did not change, although airway eosinophils and bronchial hyperresponsiveness decreased; levels of exhaled carbon monoxide were not changed by treatment. The authors conclude that both heme oxygenase isoforms are extensively and equally distributed in asthmatic and healthy subjects, and are not modulated by inhaled glucocorticoid therapy.Ex vivo studies.
H2 synthase exists in two isoforms: cyclooxygenase (COX)-1, which is constitutively expressed in most cells,
and cyclooxygenase (COX)-2, which is induced by inflammatory
stimuli. In cells of human airway smooth muscle, Bonazzi and coworkers (52) studied the effect of prostaglandin E2 on the expression of cyclooxygenase-2 induced by interleukin-1. Treatment
with exogenous prostaglandin E2 enhanced the expression of cyclooxygenase-2 induced by interleukin-1. A nonselective cyclooxygenase inhibitor decreased the production of prostaglandin E2
and reduced the expression of cyclooxygenase-2 induced by interleukin-1 (supporting a role for endogenous cyclooxygenase metabolites in modulating the expression of cyclooxygenase-2). Administration of blocking agents linked the effect of prostaglandin E2 to the transcriptional level of cyclooxygenase-2 mRNA. The authors conclude that the data support the involvement of prostaglandin E2 in a self-amplifying loop leading to increased biosynthesis of prostaglandin E2 during inflammatory events in the airways.
Airway Hyperreactivity
Animal models: antigen challenge. To assess the effect of prolonged allergen exposure on the structure and function of airways, Palmans and coworkers (89) studied varying exposures to ovalbumin in sensitized brown Norway rats. Exposure for 2 weeks produced goblet-cell hyperplasia, increased bromodeoxyuridine-positive cells in airway epithelium, increased fibronectin deposition, and thickening of the wall of the inner airway. These changes coincided with increased airway responsiveness to aerosolized carbachol. Exposure for 12 weeks produced increased fibronectin and collagen deposition in the submucosa; neither area of the airway wall nor responsiveness to carbachol differed from controls. The authors conclude that the findings are similar to the airway remodeling in patients with asthma and that changes in the extracellular matrix can enhance or protect against airway hyperreactivity.
CD4+ T cells are thought to play a major role in initiating and perpetuating airway inflammation associated with the type 2 (Th2) helper T lymphocyte phenotype, but it is not known whether activation of resident specific CD4+ T cells is sufficient to induce this phenotype. To address this issue, Knott and coworkers (90) used Balb/c DO11.10 mice that are transgenic for the T-cell receptor specific for the immunodominant epitope of ovalbumin. (A transgenic animal bears a foreign gene [termed a transgene] that is usually spliced to a tissue-specific or cell-specific promoter. The transgene is inserted into a fertilized egg in vitro, and thus becomes integrated into the animal's germ line. An epitope is the structure on an antigen that is recognized by an antigen receptor.) An aerosol of ovalbumin produced marked neutrophilia and lesser eosinophilia in bronchoalveolar fluid of sensitized DO11.10 mice, but not in wild-type mice. In vivo depletion of CD4+, but not CD8+, T cells abrogated the response. Cytokines in the lavage fluid were indicative of a type 1 (Th1) helper T cell-like immune response. Neutralizing interferon-gamma caused an increase in the eosinophilia, suggesting that production of interferon-gamma was limiting the development of a Th2 response. Multiple exposures to ovalbumin aerosol failed to induce airway hyperresponsiveness in the DO11.10 mice, unlike in the wild-type mice. The authors conclude that in vivo stimulation of resident CD4+ T cells with antigen caused inflammation with characteristics of both a Th1 and Th2 response but was not sufficient to induce airway hyperresponsiveness. To examine the effect of inhibiting several critical molecules of the tyrosine kinase-signaling cascade, Tsang and Fred Wong (91) did in vitro studies of antigen challenge of guinea-pig airways. Inhibition of transmembrane protein tyrosine kinase caused a decrease in the bronchial contraction that results from anaphylaxis, and it facilitated relaxation. Selective inhibition of mitogen-activated protein kinase failed to suppress the contraction but it facilitated relaxation. In chopped lung, the inhibitors prevented the release of histamine and peptidoleukotrienes in response to allergen. The authors conclude that inhibition of tyrosine kinase signaling cascade can markedly attenuate contraction of the airways during anaphylaxis. Very late antigen-4, also known as41 integrin, is expressed on eosinophils and lymphocytes and binds to its ligand,
vascular-cell adhesion molecule 1. In allergic sheep, Abraham
and coworkers (92) showed that BIO-1211, an inhibitor of 4,
blocked the early- and late-asthmatic response and airway hyperresponsiveness induced by antigen. The inhibitor reduced
eosinophils in bronchoalveolar tissue. The time to recover
from airway hyperreactivity induced by antigen was decreased
from 9 to 3 days. The authors conclude that an inhibitor of the
integrin 41 effectively blocks antigen-induced airway responses.
To determine the effect of prostaglandin E2, a product of
the cyclooxygenase pathway, on cytokine secretion, Peebles
and coworkers (93) used a murine model of allergic sensitization and airway hyperresponsiveness. Mice treated with indomethacin, a nonselective cyclooxygenase inhibitor, had
greater airway hyperresponsiveness, increased interstitial eosinophilia, higher levels of interleukin-5 and -13, and increased
lung mRNA expression of the CC chemokine, monocyte chemotactic protein (MCP)-1; levels of interleukin-4 and serum
IgE were not different. The authors conclude that cyclooxygenase inhibition during allergen sensitization increases hyperresponsiveness of the airways, and causes interstitial eosinophilia and production of interleukin-5 and -13.
In Sprague-Dawley rats, depletion of CD8+ T cells causes
an increase in the late airway response to antigen challenge.
Allakhverdi and coworkers (94) used this model to investigate
the pattern of chemokine and cytokine production. In CD8+
depleted rats, a challenge with ovalbumin produced a 3-fold increase in the late response, accompanied by increased eosinophilic infiltration of the airway, increased expression of eotaxin, and decreased expression of interferon-gamma mRNA;
expression of RANTES, monocyte chemoattractant protein-1,
and interleukin-4 and -5 was not affected. The authors conclude that CD8+ T cells in sensitized rats can suppress the late
airway response, possibly through decreasing expression of eotaxin and increasing expression of interferon-gamma mRNA.
Because the influx of eosinophils into the lungs is dependent on expression of very late activating antigen-4 (VLA-4),
Kanehiro and coworkers (95) investigated the effect of administering an antibody to this antigen in BALB/c mice sensitized
and challenged with ovalbumin. When given 2 hours before
challenge, the antibody prevented the expected increase in
lung resistance. Injecting the antibody from 2 hours before the
challenge until 42 hours after the challenge prevented the increase in resistance, and inhibited the concentrations of interleukin-5 and leukotrienes in bronchoalveolar fluid. The antibody prevented changes in dynamic compliance and goblet
cell metaplasia only if given 2 hours before the challenge. The
authors conclude that the timing of administering an inhibitor
of very late activating antigen-4 can selectively affect pathologic processes occurring in central and peripheral airways after allergen challenge.
Animal models: other challenges and mediators. To determine the effect of pulmonary vascular engorgement on the response to inhaled methacholine, Uhlig and coworkers (96) inflated a balloon in the left atrium of six piglets. Congestion alone increased airway resistance by 15% and tissue resistance by 8%. A small dose of methacholine alone increased airway resistance by 11% and tissue resistance by 42%. The combination of congestion and methacholine increased airway resistance by 68% and tissue resistance by 38%. Morphometry of transverse sections showed that thickness of the inner airway wall was similar in the three groups. Thickness of the outer airway wall was greater in the two congested groups than in the methacholine alone group. The amount of shortening of the airway smooth muscle in the piglets experiencing both congestion and methacholine challenge was greater than in the other two groups. The authors conclude that pulmonary vascular engorgement increases the airway, but not lung tissue, response to inhaled methacholine.
To better understand the importance of air pollution in the initiation of asthma, Hamada and coworkers (97) studied neonatal mice that were also exposed to residual oily fly ash, a surrogate for ambient air particles. Repeated exposure of the neonatal mice to allergen alone or pollutant alone had no effect on airway hyperresponsiveness and did not cause antibody production. Adult mice exposed to allergen or allergen plus pollutant did develop airway hyperreactivity, but the combination did not have an additive effect. Neonatal mice exposed to both allergen and pollutant developed greater airway hyperreactivity than on exposure to allergen alone, and reexposure to allergen produced IgE and IgG specific to ovalbumin. The authors conclude that exposure to pollutant aerosols can disrupt normal resistance to sensitization to inhaled allergens, and thereby promote the development of airway hyperreactivity. To assess the relationship between the transcription factor, nuclear factor-
B, and acute airway obstruction, Bureau and
coworkers (98) studied bronchial cells from horses experiencing heaves (acute airway obstruction that naturally occurs in
horses when exposed to moldy hay). Bronchial cells from
healthy horses had small amounts of the transcription factor,
and all horses developed high levels during heaves. Three
weeks later, the level of nuclear factor-B in horses affected
by heaves was highly correlated with the degree of residual
lung dysfunction. The active complexes of nuclear factor-B
were mainly p65 homodimers, rather than the classic heterodimer composed of the p65 and p50 subunits. The activity
of these homodimers was paralleled by expression of intercellular adhesion molecule-1. The authors conclude that the kinetics of nuclear factor-B is strongly related to the course of
acute airway obstruction in horses.
To determine relationships between airway responsiveness
and in vitro measurements of smooth-muscle contractility, Duguet and coworkers (99) studied strains of mice with known interstrain differences in airway hyperresponsiveness. The rank
order of hyperresponsiveness to methacholine among the strains
was the same order as for velocity of shortening in explanted airways. Shortening velocity correlated with the achieved degree
of airway narrowing. In contrast, airway hyperresponsiveness
occurring in vivo did not correlate with generation of isometric
tension in the trachealis, morphometric analysis of airway
smooth muscle, or tracheal myosin content. The authors conclude that the dynamics of airway smooth muscle are important in understanding airway responsiveness.
In guinea pigs, Folkerts and coworkers (100) investigated
the role of bradykinin, a metabolite of the kallikrein-kinin
system, on airway hyperresponsiveness arising from infection
with parainfluenza-3 virus. Hoe 140, a 2-antagonist that also
inhibits bradykinin, prevented the airway hyperresponsiveness that usually occurs 4 days after viral infection. Inhalation
of bradykinin produced comparable airway hyperresponsiveness in captopril-treated animals. Viral infection produced an
influx of cells in bronchoalveolar fluid that could not be prevented by pretreating with Hoe 140. The authors conclude
that bradykinin is involved in events causing airway hyperresponsiveness after viral infection without affecting the influx
of cells into bronchoalveolar fluid.
The protease-activated receptor-2 (PAR-2) belongs to a
family of G-protein-coupled receptors that are activated by
proteolysis. Ricciardolo and coworkers (101) examined the location of this receptor in guinea pigs and the effect of its activation. The receptor was located in airway epithelial and smooth
muscle cells. Trypsin, which cleaves the receptor and exposes
an N-terminal tethered ligand (SLIGRL), caused bronchoconstriction that was inhibited by antagonists of tachykinin-NK1
and -NK2 and was potentiated by inhibiting nitric oxide synthase. Trypsin and the N-terminal tethered ligand caused contraction of isolated intrapulmonary bronchi and relaxation of
the trachea and main bronchi; inhibitors of cyclooxygenase and
nitric oxide synthase abolished the relaxation. Protease-activated receptors-1, -3, and -4 were not involved in the bronchomotor action of trypsin. The authors conclude that protease-activated receptor-2 is localized to the guinea-pig airway and is
involved in multiple bronchomotor activities.
The role of eotaxin, a chemokine that causes selective recruitment of eosinophils to sites of inflammation, in airway hyperresponsiveness is not clear. To address this issue, Fukuyama and coworkers (102) instilled eotaxin into the trachea of
guinea pigs. Eosinophils accumulated in the airways but not in
the alveoli, and airway hyperresponsiveness was not altered
even when eotaxin was combined with subthreshold doses of
interleukin-5 or leukotriene D4. Eotaxin combined with platelet-activating factor enhanced airway hyperresponsiveness, and produced a larger increase in eosinophils and peroxidase
activity in bronchoalveolar fluid than seen with eotaxin alone.
The authors conclude that eotaxin alone causes eosinophil accumulation in airways but not airway hyperresponsiveness,
and that additional factors, such as platelet-activating factor,
are needed to activate eosinophils for the development of airway hyperresponsiveness.
Because heparin attenuates exercise-induced asthma, Suzuki
and Freed (103) determined whether it would inhibit hyperventilation-induced bronchoconstriction in a canine model of exercise-induced asthma. Aerosolized heparin had no effect on
baseline peripheral airway resistance, measured with a wedged
bronchoscope. Heparin achieved a 50-60% decrease in bronchoconstriction caused by hyperventilation. Despite causing acute
infiltration of macrophages and eosinophils into bronchoalveolar fluid, pretreatment with heparin attenuated or abolished the
release of leukotrienes, prostaglandins, and thromboxane consequent to hyperventilation. The authors conclude that inhaled
heparin inhibits the production and release of eicosanoid mediators caused by hyperventilation with dry air and attenuates hyperventilation-induced bronchoconstriction.
Endotoxin is thought to contribute to pulmonary hyperresponsiveness in asthma and the acute respiratory distress syndrome. In an isolated mouse lung, Held and Uhlig (104) studied
the effect of endotoxin on airway and vascular hyperreactivity.
Infusion of lipopolysaccharide induced hyperreactivity of the
airways to methacholine and of the pulmonary vasculature to
platelet-activating factor. Both forms of hyperreactivity were
completely prevented by blocking the thromboxane/endoperoxide receptor. Blocking cyclooxygenase-2 abolished vascular
hyperreactivity, but it had only a marginal effect on airway hyperreactivity. Pretreatment with an oxygen-radical scavenger,
N-acetylcysteine, partly protected against the increase in airway
hyperreactivity. The authors conclude that endotoxin induces
airway and vascular hyperreactivity by activating the thromboxane/endoperoxide receptor, that vascular hyperactivity depends
on cyclooxygenase-2 activity, and that airway hyperactivity largely depends on other mechanisms.
Ex vivo studies. Because in vitro contractility of human airways depends on sensitization status, Ammit and coworkers (105) asked whether increased contractility of sensitized airways was related to differences in content of isoforms of myosin light-chain kinase in smooth muscle. Immunoblotting of bundles of smooth muscle from five subjects with and five subjects without sensitized airways showed that the major isoform of myosin light-chain kinase was of the smooth muscle type. Sensitized airways contained about three times more myosin light-chain kinase; the content of myosin heavy chain did not differ. The authors conclude that an increase in myosin light-chain kinase in smooth muscle may be responsible for the increased contractile reactivity in sensitized tissue.
Airway smooth muscle may participate in the airway inflammatory response by expressing various cytokines, but the intracellular signal regulating cytokine expression in airway smooth muscle has not been determined. Maruoka and coworkers (106) examined the role of p38 mitogen-activated protein (MAP) kinase and extracellular signal-regulated kinase (Erk) in the production of RANTES, a member of the C-C chemokine family, by airway smooth muscle cells that had been stimulated by platelet-activating factor and by tumor necrosis factor-. Platelet-activating factor induced the threonine and tyrosine phosphorylation of p38 mitogen-activated
protein kinase and extracellular signal-regulated kinase. SB
203580 almost completely inhibited p38 mitogen-activated
protein kinase activity, and PD98059 almost completely inhibited extracellular signal-regulated kinase activity. Both inhibitors acted additively to inhibit the production of RANTES induced by platelet-activating factor. The authors conclude that
both p38 mitogen-activated protein kinase and extracellular
signal-regulated kinase are involved in the production of
RANTES by airway smooth muscle cells that are stimulated by platelet-activating factor.
To determine whether smooth muscle in human bronchi
contains ryanodine calcium-release channels and the associated
calcium-induced calcium release (CICR) mechanism, Hyvelin
and coworkers (107) obtained bronchial muscle samples during
thoracotomy. Ryanodine and caffeine induced transient increases in cytoplasmic calcium concentration. Higher doses of
ryanodine inhibited the caffeine-induced response, as did inhibitors of the calcium-induced calcium release mechanism. Additional studies revealed the existence of mRNAs encoding only
the type 3 ryanodine calcium-release channel. The authors conclude that functional ryanodine calcium-release channels are
present in human bronchial smooth muscle.
To examine the relationship between allergic inflammation
and changes in the production of extracellular matrix proteins
by human smooth muscle, Johnson and coworkers (108) used
an in vitro model of hyperresponsiveness (passive sensitization). Sera from asthmatic patients caused smooth muscle cells
to produce increased amounts of fibronectin, perlecan, laminin 1, and chondroitin sulfate. Beclomethasone did not reverse the increase in extracellular matrix protein. The authors
conclude that an interaction between the allergic process and
airway smooth muscle may alter components of the airway
wall in asthma, and that glucocorticoids do not prevent the fibrosis induced by resident cells within the airways.
In patients with asthma, desquamation of airway epithelial
cells is often attributed to airway inflammation. Because this finding might be an artifact, Ordonez and coworkers (109) did bronchial biopsies in 14 subjects with mild-to-moderate
asthma and 12 healthy subjects. The percentage of basement
membrane denuded of epithelium was 11.4% in asthmatic
subjects and 14.8% in healthy subjects. The percentage of
basement membrane covered by a single layer of basal cells
was 46% in asthmatic subjects and 55% in healthy subjects.
Neither measurement was correlated with FEV1 or PC20 on
methacholine provocation. The authors conclude that denudation of the bronchial epithelium in endobronchial biopsies is
an artifact of tissue sampling.
Early and late asthmatic responses. To determine the effect of allergen inhalation on eicosanoid mediators, Macfarlane and coworkers (110) sampled airway secretions in 14 patients with a confirmed late asthmatic response. At baseline, the percentage of eosinophils was correlated with the level of cysteinyl leukotrienes (r = 0.84), but not with prostaglandin D2 or prostaglandin E2. Allergen challenge produced a 3.5-fold increase in cysteinyl leukotrienes, which were correlated with the eosinophilia (r = 0.55), but no change in prostanoid mediators. The authors conclude that cysteinyl leukotrienes, but not prostanoids, are involved in the late asthmatic response.
To determine whether the late asthmatic response could be predicted from baseline characteristics, Avila and coworkers (111) analyzed data in 60 subjects undergoing an allergen challenge. A fall in FEV1 of at least 15% between 3 and 7 hours occurred in 57% of the subjects. Classification tree analysis revealed that a threshold of 0.25 mg per ml for the PC20 on a methacholine challenge (dose producing a 20% decrease in FEV1) was the best predictor of a late response: 87% of subjects below this dose developed a late response compared with 38% of subjects above that dose. Baseline FEV1 or sputum eosinophils did not predict the late response. The authors conclude that a methacholine challenge helps in identifying subjects likely to develop a late response to an allergen challenge. Platelet-activating factor, an important lipid mediator in the pathogenesis of asthma, is degraded by platelet-activating factor-acetylhydrolase, resulting in inactive lyso-platelet activating factor. In 14 patients with asthma, Henig and coworkers (112) did a double-blind crossover study of the effect of recombinant platelet-activating factor-acetylhydrolase on the early and late response to allergen challenge. The early and late responses, sputum eosinophils, eosinophilic cationic protein, and tryptase were not influenced. The authors conclude that recombinant platelet-activating factor-acetylhydrolase was not effective in patients with dual asthmatic responses to allergen. Prostaglandin E2 inhibits the early and late response to allergen inhalation. To determine the mechanism of its bronchoprotective action, Hartert and coworkers (113) did a double-blind crossover study in 10 atopic asthmatic patients. Nebulized prostaglandin E2 was well tolerated; it caused a decrease in the levels of prostaglandin D2 in bronchoalveolar fluid that result with allergen challenge. The authors conclude that prostaglandin E2 blocks the early asthma response by decreasing prostaglandin D2. Because cyclosporin A is known to inhibit the late asthmatic reaction and the associated increase in blood eosinophils, Khan and coworkers (114) studied 24 atopic subjects to determine whether the effect was mediated by cytokines and chemokines. Cyclosporin A inhibited allergen-induced increases in interleukin-5 and granulocyte-macrophage colony-stimulating factor in bronchoalveolar cells. The authors conclude that inhibition of the late asthmatic response by cyclosporin A may be related to inhibitory effects on eosinophil-associated cytokines and chemokines.Antigen and methacholine challenge. Persistent asthma is associated with deposition of extracellular matrix proteins, which may be mediated by release of matrix metalloproteinases and tissue inhibitor of metalloproteinase-1. To advance understanding, Kelly and coworkers (115) measured metalloproteases in bronchoalveolar fluid after segmental bronchoprovocation in 17 allergic subjects. Concentrations of matrix metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 were increased in the airway, but not in the serum, at 48 hours. Matrix metalloproteinase-9 was the predominant form, and it was correlated with airway neutrophils and less so with alveolar macrophages. The authors conclude that antigen challenge leads to the generation of matrix metalloproteinases, which may lead to remodeling in asthma.
In patients being screened for bronchial hyperresponsiveness, Fowler and coworkers (116) compared methacholine and adenosine monophosphate challenges. Of 185 patients, 76% responded to methacholine and 56% responded to adenosine monophosphate. For those unresponsive to adenosine monophosphate, 57% responded to methacholine. For those unresponsive to methacholine, 11% responded to adenosine monophosphate. Patients who responded only to methacholine had a lower FEV1 and were receiving a higher dose of inhaled glucocorticoids. Subjects with the glycine allele at codon 16 had increased bronchial hyperresponsiveness to methacholine, but not to adenosine monophosphate. The authors conclude that a methacholine challenge is a more sensitive screening tool than a challenge with adenosine monophosphate. Guidelines for methacholine and exercise challenge testing are presented in an ATS statement (117).Exercise-induced asthma. Because of the suggestion that intense physical training may cause airway hyperresponsiveness, Langdeau and coworkers (118) compared the prevalence of physician-diagnosed asthma and airway hyperresponsiveness in 100 athletes involved in high-level competitions with 50 sedentary subjects. The prevalence of airway hyperresponsiveness in the controls was 28% versus 49% in the athletes; the prevalence was especially high in athletes breathing humid air (76%) or cold air (52%) during training. Parasympathetic tone, detected by a measure of heart rate variability, was increased in the athletes and weakly correlated with the PC20 on methacholine challenge. The authors conclude that airway hyperresponsiveness is increased in athletes, possibly because of the content of air inhaled during training.
Because asthma-like symptoms are common in cross-country skiers, Karjalainen and coworkers (119) characterized the morphologic changes in the bronchial mucosa in 40 competitive skiers without a diagnosis of asthma, 12 subjects with mild asthma, and 12 healthy controls. Skiers had a 43-fold increase in T-lymphocytes, 26-fold increase in macrophages, and 26-fold increase in eosinophils over the controls; respective values in the asthmatic subjects were 70-fold, 63-fold, and 8-fold. Skiers had twice the number of neutrophils as asthmatic subjects, whereas mast cell counts were normal. Expression of tenascin, an extracellular matrix protein in the basement membrane, was increased 8-fold in the skiers and 11-fold in the asthmatic subjects, and did not correlate with the inflammatory cell counts. The authors conclude that repeated exposure to poorly conditioned air may induce inflammation and remodeling in the airways of competitive skiers. Inhalation of mannitol provides an osmotic stimulus that identifies subjects who develop exercise-induced asthma. To improve understanding of its mechanism, Brannan and coworkers (120) did a randomized double-blind trial of necrodomil, which was given before progressively increasing doses of mannitol. The dose of mannitol producing a 15% decrease in FEV1 was increased 2.6-fold after necrodomil, and FEV1 never decreased by 15% in half of the subjects. The authors conclude that necrodomil significantly inhibits the airway response to inhaled mannitol. Exercise-induced bronchoconstriction is associated with increased airway osmolarity, resulting from water loss, and alterations in airway temperature, resulting from heat loss. To define the mechanisms involved, Hashimoto and coworkers (121) developed a model exposing human bronchial epithelial cells to hyperosmolarity, and cooling and rewarming. In this model, the production of RANTES, a member of the C-C chemokine family, and interleukin-8 was inhibited by beclo- methasone diproprionate and budesonide, but these agents failed to prevent the release of p38 mitogen-activated protein (MAP) kinase and c-Jun NH2-terminal kinase (JNK). The authors conclude that interleukin-8 and RANTES are produced in a model of hyperosmolarity and cooling and rewarming, and their production is inhibited by inhaled glucocorticoids. In 10 subjects with mild asthma, Gauvreau and coworkers (122) asked, "Is exercise-induced bronchoconstriction associated with airway inflammation and hyperresponsiveness?" Exercise produced a 21% fall in FEV1, which was comparable to the fall achieved with inhaled methacholine (30%) and inhaled allergen (29%). Allergen inhalation produced hyperresponsiveness and increased sputum eosinophils, whereas exercise had no effect on inflammatory cells or airway hyperresponsiveness. The authors conclude that exercise-induced bronchoconstriction does not cause airway inflammation or hyperresponsiveness. Because hyperemia of the bronchial circulation could contribute to the airway narrowing associated with hyperpnea- induced bronchospasm, Kaminsky and Lynn (123) measured pulmonary blood volume in this setting. Hyperpnea produced a 26% decrease in FEV1 in 13 subjects with asthma, but neither diffusing capacity nor pulmonary capillary blood volume changed. The authors conclude that hyperpnea-induced bronchospasm is not accompanied by change in pulmonary capillary blood volume.Drugs. To determine whether angiotensin II receptors are involved in bronchial hyperresponsiveness, Myou and coworkers (124) did a double blind crossover study of losartan, a type 1 receptor antagonist, in 8 patients with asthma. Losartan produced an increase in PC35-PEF40 (the concentration of methacholine producing a 35% fall in standardized partial expiratory flow at 40% of FVC). The authors conclude that angiotensin II type 1 receptors are involved in bronchial hyperresponsiveness in asthmatic patients.
A series of review articles focus on the question, "What makes the airways contract abnormally in asthma?" (125).Other Pathophysiological Mechanisms in Asthma
Tachykinins, neural activity. Because inhaling cigarette smoke causes transient bronchoconstriction through acting on pulmonary C-fiber afferent endings, Wu and coworkers (130) asked whether the effect would be enhanced in the presence of airway hyperresponsiveness. In guinea pigs sensitized with aerosolized ovalbumin, inhalation of cigarette smoke produced a 6.3% greater increase in lung resistance than in nonsensitized controls. Smoke also produced greater increases in substance P-like immunoreactivity and calcitonin gene-related peptide-like immunoreactivity in bronchoalveolar fluid. Pretreatment with a selective neurokinin-2 receptor antagonist inhibited more than 85% of the enhanced bronchoconstrictive response to cigarette smoke. The authors conclude that airway hyperresponsiveness increases the bronchoconstrictive response to inhaled cigarette smoke, and that activation of lung C-fibers by endogenous tachykinins plays a primary role in this response.
In rats, Ho and coworkers (131) investigated the action of prostaglandin E2 on vagal C-fiber afferents. Infusion of prostaglandin E2 did not alter baseline activity of vagal C-fibers but it markedly enhanced the effect of capsaicin, a potent and selective stimulant of these fibers. Prostaglandin E2 also increased the response of C-fibers to lactic acid, adenosine, and lung inflation, but it did not alter the responses of either slowly adapting or rapidly adapting pulmonary receptors to inflation. The authors speculate that release of prostaglandin E2 may contribute to the dyspnea and airway irritation caused by airway inflammation. Sensory innervation of the airway originates from afferent neurons whose somata reside in the superior (jugular) and inferior (nodose) vagal ganglia. In guinea pigs immunized with ovalbumin, Moore and coworkers (132) investigated whether airway inflammation caused by inhaling antigen would unmask neurokinin-2 tachykinin responses in vagal afferent somata. Within 24 hours of allergen inhalation, exogenously applied tachykinins acting on neurokinin-2 receptors depolarized 80% of nodose neurons. Vagotomy attenuated the antigen-induced unmasking of tachykinin responses. The authors conclude that the vagus is involved in the transduction of a signal essential for unmasking of tachykinin responses. Because toluene diisocyanate stimulates the release of substance P from sensory neurons in the airways, Hunter and coworkers (133) determined whether it also affected the levels of substance P and preprotachykinin mRNA in the sensory neurons that innervate the nasal epithelium. Fisher-344 rats were exposed to a vapor of toluene diisocyanate for 2 hours. The level of substance P in nerve fibers was increased at 12, 24, and 48 hours after exposure. The proportion of cell bodies expressing high levels of preprotachykinin mRNA was increased at 24 and 48 hours. The authors conclude that inhalation of toluene diisocyanate increases release of substance P and the levels of neuropeptide within nerve fibers of the nasal epithelium. To determine the effect of nerve growth factor on vagal neurons containing substance P, Hunter and coworkers (134) used immunochemistry and retrograde tracing techniques. In the trachea of untreated guinea pigs, more than 99% of neurons containing substance P were located in the jugular ganglia. These neurons were small in diameter (23 µm) and negative for neurofilament immunoreactivity. In contrast, neurons in the nodose ganglia were large (40 µm), did not contain substance P, and were positive for neurofilament immunoreactivity. Injections of nerve growth factor- into the tracheal wall
caused 10% of the nodose neurons to become positive for substance P. The authors conclude that nerve growth factor increases expression of substance P in airway neurons, but also
changes the neuronal phenotype such that large nodose neurons provide a component of the tachykinergic innervation.
Because inhaled furosemide attenuates cough, bronchoconstriction, and dyspnea, Sudo and coworkers (135) examined its effects on tracheobronchial receptors of spontaneously breathing rats. Receptors were classified on the basis of
single or pauci unit recordings from the vagus nerve. The
slope of airway pressure versus spike frequency of slowly
adapting receptors was increased from 9 to 15 Hz per cm H2O
by inhalation of furosemide. The activity of rapidly adapting
receptors was attenuated by furosemide. The authors conclude that inhaled furosemide sensitizes slowly adapting receptors and desensitizes rapidly adapting receptors.
Because of the link between the neurogenic and immune
systems, Maghni and coworkers (136) assessed the effect of selectively inhibiting neurokinin-1 and -2 receptors on the airway inflammatory response in atopic rats sensitized to ovalbumin. Neither antagonist prevented the early airway response
to allergen challenge, whereas both antagonists blocked the
late response. The neurokinin-2, but not neurokinin-1, antagonist decreased bronchoalveolar eosinophilia, and both helper
T cell Th1 (interferon-gamma) and Th2 (interleukin-4 and -5)
cytokine expression in bronchoalveolar cells. The authors conclude that neurokinins are linked to the regulation of cytokine expression in cells without discrimination as to their phenotype, and that the dichotomy between neurokinin receptors
has therapeutic implications.
To determine whether trypsin, which activates protease-activated receptors 2 and 4, induces release of neurokinins from the sensory C-fibers that innervate airways of guinea pigs, Carr and coworkers (137) studied the isolated bronchus. Trypsin evoked contractions in proportion to its concentration, and the contractions were markedly attenuated in the presence of neurokinin receptor antagonists. Capsaicin, which depletes neurokinins in sensory nerves, also decreased trypsin's ability to produce contractions. Trypsin did not evoke action potentials in C-fiber afferents whose receptive fields were located in the trachea or main bronchi. The
authors conclude that trypsin enables the local release of sensory
neurokinins from afferent C-fibers and that this release occurs independently of the sensory functions of these nerves.
Deep inspiration bronchoconstriction. In healthy subjects, a deep inspiration produces bronchodilation and it also produces bronchoprotection. To determine which effect is stronger, Scichilone and coworkers (138) performed methacholine challenges at two doses in 10 healthy subjects. A bronchodilation index was derived from spirometry after methacholine, and a bronchoprotection index was derived from spirometry before methacholine. With mild obstruction (decrease in FEV1 of 10- 20%), the bronchodilator and bronchoprotector effects of a deep inspiration were equal. With greater obstruction (decrease in FEV1 of 20-40%), the bronchoprotector effect was twice as strong. The authors conclude that the bronchoprotector effect of lung inflation is stronger than the bronchodilator effect.
When airway smooth muscle is contracted, a deep inhalation causes bronchodilation in healthy subjects and causes bronchoconstriction in asthmatic subjects. Reasoning that the difference relates to the dynamic behavior of airway smooth muscle and lung parenchyma, Brown and Mitzner (139) studied this issue in anesthetized dogs using high-resolution computed tomography. When the airways were contracted with methacholine, rapid lung inflation caused the airways to dilate at only one-quarter the rate of the parenchyma. The authors conclude that the slower dynamic response of the contracted airway probably involves intrinsic properties of the smooth-muscle contractile process. Healthy subjects develop a decrease in airway resistance with deep inspiration, whereas patients with asthma develop an increase in resistance or no change. Because usual pulmonary function tests do not measure airway dimensions with sufficient precision to resolve the different behaviors, Mitzner and Brown (140) measured the dimensions with direct imaging. Normal tidal stresses allowed airway smooth muscle to respond normally to deep inspiration. Removing tidal stresses after methacholine was sufficient to change the normal bronchodilator response to a deep inspiration into an abnormal contraction response. The authors conclude that alterations in sensitivity of airway smooth muscle to normal tidal stresses may be involved in the pathogenesis of asthma.Blood flow. In 19 patients with mild asthma who had never used glucocorticoids, Brieva and coworkers (141) found that blood flow in their airway mucosa was 26% higher than in 12 healthy controls. Inhaled albuterol increased mucosal blood flow by 27% in the controls, but had no effect in the patients. A 2-week course of inhaled fluticasone (440 µg daily) decreased mucosal blood flow by 11% in the patients, and restored its responsiveness to albuterol. The authors conclude that blood flow in the mucosa of asthmatic airways has a subnormal response to albuterol, which may be caused by airway inflammation.
Because the vasoconstrictive action of corticosteroids on the skin (skin blanching) is used to assess relative potency, Kumar and coworkers (142) examined whether inhaled fluticasone causes vasoconstriction in the airway mucosa. At baseline, blood flow in the airway mucosa was 25% higher in 10 asthmatic than in 10 healthy subjects. Fluticasone decreased mucosal blood flow by 37% in the patients and by 21% in the controls; the change was transient and values approached normal after 90 minutes. The authors conclude that measurement of airway mucosal blood flow may provide a more relevant test of the potency of inhaled corticosteroids than skin blanching.Immunology. Because autoimmunity may explain the link between asthma and chronic inflammation of the airways mediated by T cells, Arnold and coworkers (143) measured the expression of perforin, a cytotoxic molecule implicated in autoimmunity. The percentage of lymphocytes expressing perforin was higher in 20 patients with asthma than in 18 controls. The subpopulations of lymphocytes with increased expression of perforin were CD3+, CD4+, CD5+, and CD56+ in 13 patients with extrinsic asthma, and CD4+ and CD56+ in 7 patients with intrinsic asthma. The authors conclude that allergic and intrinsic asthma is associated with increased expression of perforin in T-lymphocyte subsets.
Airway narrowing. In 7 patients with asthma, Kaminsky and coworkers (144) determined the mechanisms responsible for increased resistance in the peripheral airways. The stop flow method during bronchoscopy was used to measure the decay of segment pressure over time. The absence of a sudden drop in pressure and its smooth decay suggest that airway resistance was negligible. The authors conclude that peripheral airway resistance in patients with asthma is predominantly located in the collateral airways rather than in the more proximal airways.
To assess the presence and characteristics of airway wall thickening in patients with asthma, Niimi and coworkers (145) did helical computed tomography of the apical bronchus of the right upper lobe in 81 patients with asthma and 22 healthy controls. Airway wall thickness was increased in 13 patients with mild asthma, 39 patients with moderate asthma, and 22 patients with severe asthma, but not in 7 patients with intermittent asthma. Wall area alone, and corrected for body surface area, was weakly correlated with duration and severity of asthma, FEV1, and FEV1/FVC. Airway luminal area was not related to severity of disease. Intraobserver and interobserver reproducibility was good. The authors conclude that thickening of the airway wall in patients with asthma is not limited to those with severe disease, and the degree of thickening relates to the duration and severity of the disease. To determine the relationship between spirometric measures of airway obstruction and the area of the airway lumen, Brown and coworkers (146) performed high-resolution computed tomography and partial spirometry in five healthy subjects challenged with increasing doses of methacholine who were prohibited from taking a deep breath. Increasing doses of methacholine decreased progressively the airway area, reaching down to 80% of baseline. Change in airway area was correlated with partial FEV1 (r = 0.46). The authors conclude that spirometric change consequent to methacholine is associated with decreased area of the conducting airways. King and coworkers (147) developed an automated algorithm for analyzing images of the lumen and wall thickness of airways seen on high-resolution computed tomography. For tubes of varying size, embedded in Styrofoam and scanned at different angles, the algorithm provided accurate estimates of the area of airway lumen; the area of the airway wall was overestimated in proportion to airway size. In two excised and inflated pig lungs, the mean error for area of the airway lumen was 0.52 mm2 and for area of the airway wall it was 0.17 mm2. The authors conclude that their automated algorithm provides accurate measurements of airway dimensions on high-resolution computed tomography.Remodeling. To determine whether structural remodeling alters the mechanical properties of the airways, Brackel and coworkers (148) used esophageal balloons and probes at five locations between the right lower lobe and midtrachea in 10 patients with asthma and 14 controls. Plots of area versus transmural pressure revealed a larger cross-sectional area in men than in women. Specific airway compliance was decreased in downstream locations. It was also decreased in the patients, although the difference from controls became less marked towards the trachea. The authors conclude that stiffening of the airways suggests the occurrence of remodeling in asthma.
Classic mechanics. In 18 asthmatic patients (mean age 59 years) with irreversible lung function despite aggressive treatment, Gelb and Zamel (149) investigated the mechanism of airflow limitation. All patients had markedly abnormal flow- volume curves and hyperinflation, normal computed tomography, normal transdiaphragmatic strength, and normal or elevated diffusing capacity. All but three patients had marked loss of static elastic recoil pressure of the lung. Loss of lung elastic recoil explained the flow limitation on maximal expiration in only four elderly patients, and then only at low lung volumes. The loss of elastic recoil in the other 11 patients was estimated to account for 35-55% of the decrease in maximal expiratory flow. The authors conclude that loss of elastic recoil is unexpected