INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Bronchus-associated lymphoid tissue (BALT) has been observed in the bronchial wall of several mammalian species. It consists of aggregates of lymphoid and nonlymphoid cells with a specialized lymphoepithelium and is more frequently located at the bifurcations of the tracheobronchial tree. Differences in structure and function have been observed between the different species (1). In rabbits and rats, BALT is constitutively present and is believed to have a central role in the initiation of the adaptive immune response through the uptake and presentation of antigen to the immune system (2). In humans, even though almost universally demonstrated in the lungs of children up to 5 yr of age (3), BALT is not considered to be constitutively present or to be a major component of the pulmonary immune system in adults (4). BALT can be induced in immunogenically stimulated lungs (5), as it has been reported in 8% to 82% of individuals with chronic or recurrent lower respiratory tract infections, recurrent pneumonia, and cigarette smokers (6). However, BALT foci have been observed in 14% (2 of 14) of nonsmokers, suggesting that they may be an integral constituent of the healthy human lung (9).

In order to shed more light on the controversy around the presence of BALT in human airways, we examined endobronchial biopsies from a study population of elite cross-country skiers and normal healthy control subjects. We observed lymphoid aggregates in the subepithelial regions of the bronchial mucosa and in the submucosa of both skiers and control subjects.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

Forty-four nonsmoking, elite, competitive cross-country skiers from central Norway and Sweden and 12 healthy, nonsmoking medical students from Trondheim were recruited to this study (Table 1). In skiers, the mean age at the start of competitive skiing was 9.9 yr (range, 6 to 15 yr), the mean (± SD) annual training was 426 ± 95 h, and mean competitive skiing experience was 7.6 ± 2.3 yr.

Asthma-like symptoms, defined as the presence of at least two of the following symptomswheeze, abnormal breathlessness, or chest tightness present either on exertion, at rest, or on exposure to irritantswere reported by 26 (59%) skiers in a self-completed questionnaire and were absent in control subjects. The use of 2 agonist bronchodilator therapy within the previous year was reported by 11 (25%) skiers. Three of these skiers used topical steroids within this period.

Lung function was measured with the Microlab 3300 Mk2 spirometer (Micro Medical Ltd, Gillingham, Kent, UK). Bronchial responsiveness to methacholine was measured by a controlled tidal volume technique using an automatic inhalation synchronized dosimeter jet nebulizer (Spira Elektro 2; Respiratory Care Centre, Hameenlinna, Finland) as previously described (10). Bronchial hyperresponsiveness, defined as provocative dose of methacholine causing a 20% fall in FEV₁ (PD₂₀ FEV₁) of  1,800 µg (9.1 µmol), was detected in 35 (80%) skiers and absent in control subjects.

Written informed consent was obtained from all subjects, together with written parental consent for those younger than 18 yr of age. The study was approved by the ethics committee of the Medical Faculty of the Norwegian University of Science and Technology, Trondheim, Norway.

Bronchoscopy

Subjects were premedicated with nebulized salbutamol (1 mg/ml, 2.5 ml; Glaxo Wellcome, Greenford, UK) and ipratropium bromide (0.25 mg/ml, 1 ml; Boehringer Ingelheim, Ingelheim, Germany), and intravenous glycopyrronium (0.2 mg/ml, 1.5 to 2.5 ml; Wyeth-Ayerst International Inc., Philadelphia, PA), midazolam (1 mg/ml, 1 to 2 ml; F. Hoffmann-La Roche & Co., Basel, Switzerland), and alfentanil (0.5 mg/ml, 0.5 ml; Janssen Pharmaceutica BV, Beerse, Belgium). Bronchoscopy was performed with the Olympus BF-XT 20 or BF-IT 30 fiberbronchoscope (Olympus Optical Co., Tokyo, Japan) under local anesthesia with xylocaine hydrochloride. Olympus FB 35C or FB 36C forceps were used to obtain a total of 3 to 4 bronchial biopsy specimens from the carinae of the second- and third-generation bronchi. Supplementary oxygen (2 L/min) was administered via a nasal catheter and oxygenation status was monitored by pulse oximetry during the procedure.

Processing of Bronchial Biopsies

Bronchial biopsy specimens from 42 skiers and control subjects were snap-frozen in liquid nitrogen and stored at 70° C before processing for immunohistochemical studies. Biopsy specimens from two skiers with asthma-like symptoms and bronchial hyperresponsiveness to methacholine were fixed in 4% buffered formaldehyde solution for routine histopathologic studies.

Snap-frozen biopsies were embedded in Tissue Tek OCT medium (Miles Inc., Elkart, IN). Cryosections of 5 µm thickness were cut on a Leitz 1720 Digital Cryostat (Ernst Leitz GmbH, Wetzlar, Germany). Slide preparations were immunostained with BerMAC3 (dilution 1:25) monoclonal antibody (mAb) to identify macrophages and with anti-CD3 (1:1,000) mAb to identify T lymphocytes. Only slide preparations that contained cryosections from biopsies with bronchial epithelium, basement membrane, lamina propria, and submucosa were evaluated. Additional slide preparations were prepared from remaining biopsy material for immunostaining with anti-CD8 (1:1,000) mAb for T-cytotoxic lymphocytes, anti-CD4 (1:100) mAb for T-helper lymphocytes, and anti-CD20 (1:10 to 1:25) mAb for B lymphocytes. Specific antibody binding was visualized by the alkaline phosphatase antialkaline phosphatase (APAAP) method. The number of slide prep-arations in the two groups that were evaluated for each cell phenotype is shown in Table 2.

Formaldehyde-fixed biopsies were embedded in paraffin and 4-µm-thick sections were prepared. In these sections, staining with the routine hematoxylin-eosin-saffran method and immunostaining for T lymphocytes with anti-CD3 (1:200) mAb and B lymphocytes with anti-CD20 (1:900) mAb were performed. Specific antibody binding was visualized by the labeled streptavidin-biotin-peroxidase method.

All mAbs were supplied by Dako AS, Glostrup, Denmark.

Lymphoid Aggregate

A lymphoid aggregate was defined as a follicle-like cluster of more than 50 cells, which was closely related to the bronchial epithelium or to the bronchial gland or ducts in the submucosa.

Statistical Analysis

The association of lymphoid aggregates in skiers with biopsy level, self-reported allergy, medication, asthma-like symptoms, and bronchial hyperresponsiveness was assessed by the chi-square test (two-tailed p, Fisher exact test when appropriate). The independent samples t test was used to assess for significant differences in age and lung function in the two groups. A value of p  0.05 was considered to be statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Lymphoid aggregates were observed in biopsies from 28 (64%) skiers and three (25%) control subjects. The frequency of these aggregates was 2.5-fold greater (p = 0.02) in skiers than in control subjects (Figure 1). Aggregates were seen in skiers of all ages, and the distribution is shown in Figure 2. Of the skiers with lymphoid aggregates, 21 were males, and three (11%) skiers had no history of asthma-like symptoms, respiratory allergy, or use of 2 agonists, and were not hyperresponsive to methacholine.

View larger version (34K): [in this window] [in a new window]   Figure 1.   Lymphoid aggregates in control and cross-country skiers. Frequency is significantly greater in skiers than in control subjects (p = 0.02, chi-square test).

View larger version (39K): [in this window] [in a new window]   Figure 2.   Distribution of lymphoid aggregates by age of the skiers. Bar chart on left axis shows the number of skiers with and without aggregates, while graph on right axis shows the percentage of skiers with aggregates at each age.

Cryosection Slide Preparations

A total of 182 slide preparations from 42 skiers and 23 slide preparations from 12 control subjects were examined by light microscopy (Table 2). Lymphoid aggregates were observed in a total of 69 (38%) slide preparations from skiers and five (22%) slide preparations from control subjects, altogether from 26 skiers and three control subjects.

Morphological appearances. The aggregate was usually a discrete, easily recognized structure on light microscopy. It was located in the lamina propria, commonly closely below the basement membrane and thus in close association to the bronchial epithelium in 20 skiers. In one of these skiers, the aggregate was sandwiched between the ducts of a submucosal gland and the subepithelial basement membrane, while in four skiers two or more aggregates were visible within the same cryosection. In another six skiers and in two control subjects, the aggregates were seen in the submucosa.

Germinal centers and high endothelial venules could not be identified with certainty within the aggregates. In a few preparations, an area of the epithelium overlying the aggregate appeared to be flattened with an apparent reduction of the superficial layer of epithelial cells.

Immunohistochemistry. T lymphocytes were present within all aggregates identified in slide preparations immunostained with anti-CD3 mAb. Other lymphocyte phenotypes were not always identified in additional slide preparations from biopsies of skiers with aggregates. The frequency of T-helper, T-cytotoxic, and B-cell phenotypes in aggregates identified in slide preparations that were immunostained for these cell phenotypes is shown in Table 2. The lymphocytes were usually seen as clusters within the aggregate.

Lymphocytes were seen in the overlying epithelium in 25 of 55 (45%) slide preparations from skiers with lymphoid aggregates. These cells were more often of the T-cell phenotype than of the B-cell phenotype. Intraepithelial lymphocytes were not observed in lymphoid aggregates in control subjects.

Macrophages were present in 12 of 14 (86%) lymphoid aggregates from skiers and in 2 of 2 lymphoid aggregates from control subjects. They were usually solitary cells diffusely distributed within the aggregate.

Level of biopsy. In the skiers, 82 (45%) slide preparations were from biopsies of the carina of the left upper and lower lobe bronchi, and 67 (37%) slide preparations were from biopsies of the segmental carinae of the left upper lobe. In the control group, 10 (43%) slide preparations were from biopsies of the carina of the left upper and lower lobe bronchi, and seven (30%) slide preparations were from biopsies of the segmental carinae of the right or left upper lobes. The remaining slide preparations in the skiers (18%) and control subjects (27%) were from biopsies either of the carina of the lower and middle lobe bronchi or of the segmental carinae of the right lower lobe.

Evaluable biopsies were from second-generation carinae in 19 skiers and from third-generation carinae in 23 skiers. Lymphoid aggregates were observed in 11 (58%) and 15 (65%) of these skiers (p = 0.6). In the control subjects, aggregates were observed in 2 of 6 evaluable biopsies from the second-generation carinae and in 1 of 6 evaluable biopsies from the third-generation carinae.

Paraffin Slide Preparations

Lymphoid aggregates were observed in slide preparations from both skiers on staining with hematoxylin-eosin-saffran and immunohistochemical methods. The morphology and the immunohistochemistry of the lymphoid aggregate from one of these skiers are shown in Figure 3.

View larger version (134K): [in this window] [in a new window]   View larger version (136K): [in this window] [in a new window]   View larger version (141K): [in this window] [in a new window]   Figure 3.   Serial 4-µm-thick sections of lymphoid aggregate in endobronchial biopsy (fixed in buffered formalin and embedded in paraffin) from cross-country skier with asthma-like symptoms, bronchial hyperresponsiveness to methacholine, and without antiasthmatic medication. Top panel: aggregate as seen on hematoxylin-eosin-saffran staining. Middle panel: immunostaining with anti-CD3 monoclonal antibody (mAb) for T lymphocytes. Note clustering of T lymphocytes in peripheral area of aggregate and lymphocyte infiltration of overlying epithelium. Bottom panel: immunostaining with anti-CD20 mAb for B lymphocytes. Note clustering of B lymphocytes in central area of aggregate. Original magnification: ×200 for all panels.

Lymphoid Aggregates and Subject Characteristics in Skiers

The lymphoid aggregates were seen more frequently in skiers using 2 agonists (91% versus 55%, p = 0.04) and with bronchial hyperresponsiveness to methacholine (71% versus 33%, p = 0.053) (Table 3). They were not associated with a history of respiratory allergy (p = 0.69) or with asthma-like symptoms (p = 0.35). Aggregates were present in all three skiers who had used topical steroids within the last year.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

To our knowledge, this is the first reported observation of lymphoid aggregates in endobronchial biopsies from human airways. Previous observations of lymphoid aggregates in human airways were made in studies of lung tissue obtained either at surgery or at post mortem.

The aggregates in our study have similarities as well as differences to BALT as described in animal and human lung tissue (1, 2, 11, 12). The similarities are location at the airway bifurcations and close relationship to the bronchial epithelium and submucous glands, infiltration of the overlying epithelium with lymphocytes, and the presence of macrophages and B- and T-lymphocyte zones within the aggregates. The localization of the B- and T-lymphocytes zones within the aggregates in our study is similar to that reported in the BALT in young children (13). The differences are the absence of features such as germinal centers and high endothelial venules and of the other criteria of a lymphoepithelium, and their small size. However, these differences do not preclude the possibility that our aggregates may be a manifestation of BALT. It has been observed both in the pig (1) as well as in humans that the histologic features of BALT may be determined by the degree of antigenic stimulation. BALT foci in patients over 5 yr of age with chronic respiratory tract and parenchymal infection had all the features of BALT as seen in rabbits and rats (8). In contrast, Richmond and coworkers reported that cardinal features such as germinal centers and nonciliated M cells in the lymphoepithelium were not present in at least 75% of BALT foci in smokers and healthy nonsmokers (9). Furthermore, in that study the size of the aggregates was not dissimilar from that in our study, and submucosal location of aggregates was also observed.

We are unable to explain the unexpected high frequency of these aggregates in apparently healthy cross-country skiers and control subjects and can only speculate about their possible function. With the exception of one study (9), lymphoid aggregates have not been reported in other studies of the presence of BALT in healthy adult humans. One explanation for this discrepancy may be the large number of slide preparations that were examined in our study. These preparations were made from serial cryosections of biopsy material and were subjected to strict quality control. Another possible explanation may be that the antigenic load on the lower airways may have been greater in our skiers than in other healthy subjects in those studies. During training and competition, nasal filtration of antigens in the inspired air is ineffective, as respiration of the high minute ventilation volumes is predominantly oronasal. Interestingly, similar lymphoid aggregates have been reported in the conjunctiva of contact-lens wearers (14). The aggregates may thus be part of the immunological defense mechanisms. The presence of aggregates in biopsies from the three skiers on topical corticosteroids may suggest that anti-inflammatory treatment does not suppress their expression. Finally, lymphoid aggregates may be a constitutive feature of the bronchial wall in a proportion of healthy adults, as they were present in 25% of healthy control subjects and in three clinically healthy skiers.

In summary, we have observed a high frequency of lymphoid aggregates in endobronchial biopsies from a population of young adult cross-country ski athletes and healthy young adults. The frequency of these aggregates is significantly greater in the skiers than in control subjects. While these aggregates share some resemblance with what is usually defined as BALT, their exact nature and function await further clarification.