Development of Bronchus-Associated Lymphoid Tissue in Chronic Hypersensitivity Pneumonitis* Takafumi Suda, MD, PhD; Kingo Chida, MD, PhD; Hiroshi Hayakawa, MD, PhD; Shiro Imokawa, MD, PhD; Masatoshi Iwata, MD, PhD; Hirotoshi Nakamura, MD, PhD; and Atsuhiko Sato, MD, FCCP
Study objectives: Bronchus-associated lymphoid tissue (BALT) is well defined in animals. In humans, however, BALT has been reported to be inducible under pathologic conditions, such as chronic respiratory infection, although it is not present in healthy adults. Thus, induced BALT is considered to be involved in the mucosal immunity of the human lung under these conditions. However, there have been few studies to investigate BALT development in hypersensitivity pneumonitis. The aim of this study was to examine the presence of BALT in hypersensitivity pneumonitis, especially in its chronic form. Methods: The subjects included five patients with chronic hypersensitivity pneumonitis (CHP) diagnosed from clinical and histologic findings. We investigated histologically the development of BALT in these patients. Further, the cellular distribution of BALT was also examined by immunohistochemistry. Results: BALT was present in three of five patients with CHP. Immunohistochemical examination revealed the follicular area of BALT to be composed mainly of B cells, while the parafollicular area comprised predominantly T cells. Centroblasts located in the germinal center of BALT expressed Ki-67 antigen, a marker of cell proliferation, suggesting that these cells were actively proliferating after antigenic stimulation. Cells expressing bcl-2, which is present primarily on memory B cells, were confined to the follicular area, devoid of any germinal centers. S-100positive, CD1a-negative interdigitating dendritic cells were observed in the dome area of BALT. Conclusions: These observations suggest that chronic antigenic stimulation and/or inflammation in CHP may cause BALT development, which, in turn, is likely to play an important role in the mucosal immune response of this disease. (CHEST 1999; 115:357–363) Key words: BALT; bronchus-associated lymphoid tissues; hypersensitivity pneumonitis Abbreviations: BALT 5 bronchus-associated lymphoid tissue; CHP 5 chronic hypersensitivity pneumonitis; DPB 5 diffuse panbronchiolitis; GALT 5 gut-associated lymphoid tissue
lymphoid tissue (BALT) was B ronchus-associated first described in animals by Bienenstock, who 1–3
characterized it as lymphoid tissue in the lung, analogous to gut-associated lymphoid tissue (GALT). BALT is considered to play a crucial role in the mucosal immunity of the lung in these animals as the site of antigen-uptake and induction of antigenspecific immune response.4,5 In humans, it has been reported that no BALT is present in the normal *From the 2nd Division of Internal Medicine, Hamamatsu University School of Medicine (Drs. Suda, Chida, Hayakawa, Imokawa, Iwata, Nakamura), Hamamatsu, and Kyoto Preventive Medical Center (Dr. Sato), Kyoto, Japan. Manuscript received June 1, 1998; revision accepted August 3, 1998. Correspondence to: Takafumi Suda, MD, PhD, 3600 Handa-cho, Hamamatsu, Shizuoka, 431-31 Japan
lung.6 However, several studies have demonstrated that BALT can develop under certain pathologic conditions, including chronic respiratory infection, immunodeficiency, and autoimmune disease.7–10 We previously reported the presence of BALT in diffuse panbronchiolitis (DPB), which is a chronic inflammatory disease of the airways characterized by peribronchiolitis with mononuclear cells, and bronchiolar disease associated with rheumatoid arthritis.9,10 In these diseases, it was speculated that chronic microbial stimulation, persistent inflammation, or systemic immunologic disorders may be responsible for the development of BALT. Chronic hypersensitivity pneumonitis (CHP) is an immunologically mediated disorder caused by continuous inhalation of specific environmental organic CHEST / 115 / 2 / FEBRUARY, 1999
357
antigens.11,12 Patients with CHP manifest chronic inflammation predominantly around small airways caused by an antigen-specific immune reaction. As seen in DPB, it is possible that continuous antigenic stimulation and persistent airway inflammation in CHP may lead to the development of BALT. However, no data have been available on BALT in CHP. In the present study, we investigated histological BALT development in patients with CHP. Further, the cellular distribution of BALT in CHP was also examined by immunohistochemistry. BALT was observed in three of five patients with CHP, suggesting that induced BALT may be involved in the mucosal immunity of this disease by functioning as its inductive site. Materials and Methods Patients The study population consisted of five patients with CHP who underwent an open lung biopsy. The diagnosis was made from a compatible exposure history to an antigen or environmental factor, the presence of long-standing clinical and radiologic findings, and a lung biopsy showing characteristic histologic features. There were two men and three women, with a mean age of 50 years. One subject was a current smoker at the time of diagnosis; the other four were nonsmokers. All had chronic dyspnea and cough. The duration of symptoms ranged from 12 to 60 months, with a mean duration of 37 months. In every case, improvement in clinical findings was observed with the avoidance of the causative antigen or environment, although the degree of improvement varied among the patients. No patient kept a bird as a pet in their home. Exposure leading to hypersensitivity pneumonitis was determined in all cases. We performed a challenge test exposing patients to either causative antigens or environments after hospitalization. Over the next 24 h after the challenge test, the patient’s symptoms and signs were observed, and measurements were made of body temperature, leukocyte count, C-reactive protein, and pulmonary function. The challenge test was interpreted according to the criteria of Yoshida et al.13 One patient was a confectioner exposed to fine wheat flour. Provocation with emulsified wheat flour solution in phosphate-buffered saline solution for 10 min using a ultrasonic nebulizer gave a positive reaction. Another was a farmer who planted muskmelons in a greenhouse. The other three developed their symptoms in their homes. In these four patients, an environmental challenge was made by having the patients stay in the greenhouse or home for 3 to 12 h until a positive response was exhibited. Our institutional committee approved the challenge test and every patient tested gave informed consent.
Histologic Studies Lung biopsy specimens were obtained from at least two lobes in every case. They were fixed in 10% formaldehyde and embedded in paraffin. Then 4-mm-thick sections were cut and stained with hematoxylin-eosin or Elastica van Gieson. The histologic sections of three or four lung tissues in each case were reviewed by two observers. In this study, lymphoid follicle-like structures in the bronchiolar wall, which were covered by a specialized lymphoepithelium, were defined as BALT, as origi358
nally described by Bienenstock.14 The lymphoepithelium overlying the follicles is characterized by nonciliated, cuboidal epithelial cells and intraepithelial lymphocytes.1,14
Immunohistochemistry To study the cellular distribution of BALT, immunohistochemistry was performed using a modified streptavidin-biotin-peroxidase complex method with a Histofine SAB-PO kit (Nichirei; Tokyo, Japan). The following antibodies were used: mouse monoclonal antibody O10 (anti-CD1a; Immunotech; Marseille, France), 144B (anti-CD8; Dako Japan; Kyoto, Japan), L26 (anti-CD20; Dako Japan), 124 (anti-bcl-2; Dako Japan), MIB1 (anti-Ki-67; Dako Japan), rabbit polyclonal antibody anti-CD3 (Dako Japan), and anti-S-100a (Dako Japan). Briefly, 4-mm sections were deparaffined in xylene and rehydrated in ethanol. For staining against CD4, CD8, CD20, bcl-2, and Ki-67, nonspecific protein staining was blocked with the rabbit serum. The sections were autoclaved at 121°C for 20 min in a stainless steel pot filled with 10 mM citrate buffer (pH 6.0). The slides then were treated with 3% hydrogen peroxidase in methanol for 20 min at room temperature, to eliminate endogenous peroxidase, and incubated with the primary antibody at 4°C overnight followed by biotinylated anti-mouse immunoglobulin antibody for 20 min at room temperature. The slides then were incubated with streptavidin-biotin-peroxidase complex for 15 min at room temperature. They were developed with 3,39-diaminobenzidine tetrahydrochloride and counterstained with methyl green. For staining against CD3 and S-100, biotinylated anti-rabbit immunoglobulin antibody was used as the secondary antibody. The coexpression of different antigens on the same cells was determined by staining the sequential sections.
BAL BAL was performed with a fiberoptic bronchoscope (Olympus Corp.; Tokyo, Japan) in a segmental or subsegmental bronchus of the middle lobe with 3 3 50 mL of sterile 0.9% saline solution. BAL fluid was centrifuged at 800g for 10 min to obtain the cellular components. Total cell count was determined using a hemocytometer and a differential cell count was taken on Giemsa-stained cytocentrifuged preparations. To characterize the phenotype of the T cells in the BAL fluid, flow cytometric analysis was performed in a flow cytometer (EPICS Profile; Coulter Electronics; Hialeath, France) using mouse antibodies OKT3 (anti-CD3; Coulter Electronics), OKT4 (anti-CD4; Coulter Electronics), and OKT8 (anti-CD8; Coulter Electronics).
Immunologic Studies Serum samples obtained from patients were examined by the Ouchterlony gel double immunodiffusion method for detecting precipitating antibodies to various antigens: Aspergillus fumigatus, Cephalosporium acremonium, Cryptostroma corticale, Micropolyspora faeni, Pullularia pullulans, Sitophilus granarius, Thermoactinomyces vulgaris, Trichoderma viride, pigeon droppings and serum (Hollister-Stier Laboratories; Ontario, Canada), Alternaria kikuchiana, Candida albicans, Cladosporium cladosporoides, Penicillium lutem (Torii Corp; Tokyo, Japan), Trichosporon cutaneum (which was a gift of Dr. M. Ando, Kummamoto University; Kummamoto, Japan), Aspergillus niger, and Sphaerotheca fuliginea prepared by Dr. K. Nishimura (Research Center for Pathogenic Fungi and Microbial Toxicoses, Chiba University, Chiba, Japan) as described by Ando et al.15 Clinical Investigations
ND 1 Home 51/M 5
12
Home 63/F 4
31
Home 46/F 3
30
Wheat flour 52 46/M 2
*VC 5 vital capacity; BALF 5 BAL fluid; TCC 5 total cell count; Lym 5 lymphocytes; AM 5 alveolar macrophages; Eos 5 eosinophils; Neut 5 neutrophils. †Ratio of CD41 to CD81 lymphocytes. ‡1 5 positive response; ND 5 not done.
225
62.2
36.8
76.8
60.2
1.2
43.0 56.0
0
1.0
5.1
ND 1 305
71.4
28.2
76.4
75.5
1.7
22.2 73.6
2.7
1.5
19.2
ND 1 277
76.0
47.0
47.0
77.0
1.1
5.2 92.7
0
2.1
0.39
1 ND ND 0 0 55.2 44.8 2.0 76.0 59.0 40.8 94.6 279
ND 1 1.2 0.8 0 46.4 51.4 5.2
Exposure to Inhalation Causative of Causative CD4/CD8† Environment Antigen Neut, %
BALF Findings
Eos, % FEV1, %
84.6 68.4 31.6 48.5 490
Micronodular densities Reticulonodular, linear densities Reticulonodular densities Reticulonodular densities Reticulonodular densities Greenhouse 60 43/F 1
Histologic characteristics are summarized in Table 3. In every case, the distribution of the lesions basically reflected peribronchiolar predominance. All patients had bronchiolitis and alveolitis infiltrated with small round cells, and intraluminal granulation tissue. Bronchiolitis obliterans-organizing pneumonia was present in two cases. Two cases had nonnecrotizing granulomas. Interstitial fibrosis was observed in three cases. In cases 3, 4, and 5, prominent hyperplastic lymphoid follicles in the bronchioles were found (Table 3; Fig 1, top and middle). The epithelium overlying the surface of these follicles consisted of low cuboidal, nonciliated cells and had intraepithelial infiltrating lymphocytes, the characteristics of lymphoepithelium of BALT (Fig 1, bottom). Elastica van Gieson staining showed disruption of the elastic fibers running through the lamina propria in the lymphoid follicles (data not shown). Taken together, these findings were consisted with BALT. The BALT in CHP was composed of four distinct re-
Chest Radiograph Findings
Histologic Findings
Pulmonary Functions
Clinical characteristics of five patients with CHP are summarized in Table 1. All cases, except for case 2, had hypoxemia. Pulmonary function tests showed restrictive impairment in all cases. In the BAL fluid, an increase in the percentage of lymphocytes was observed in every case except case 3. However, the CD4 to CD8 ratio of lymphocytes in their BAL fluid varied. Cases 1 and 3 through 5 developed symptoms associated with the greenhouse and home, respectively, and their environmental challenge tests gave a positive response. Precipitating antibodies detected in the patients’ serum and fungi isolated from their environments are listed in Table 2. No patient had precipitating antibody to T cutaneum, which causes summer-type hypersensitivity pneumonitis associated with infected houses in Japan. Case 1 had precipitating antibodies to S fuliginea isolated from her greenhouse. In case 4, A niger was isolated from the home together with serum precipitins to it. Thus, it is possible that these fungi were the causative antigen in each case; this was not confirmed, however, since inhalation tests with these fungi could not be performed.
Table 1—Clinical Characteristics of Patients with Chronic Hypersensitivity Pneumonitis*
Clinical Features
TCC, 3105/ mL Lym, AM, BALF % %
Results
Serum IgA Level, Pao2, Paco2, mg/mL mm Hg mm Hg % VC
Cultures of indoor samples from the home or greenhouse were performed by open plate culture on Sabouraud’s agar media. All isolated fungi were identified by Dr. K. Nishimura.
Duration of Causative Case Symptoms, Antigen, No. Age/Sex mo. Environment
Challenge Tests‡
Isolation of Fungi
CHEST / 115 / 2 / FEBRUARY, 1999
359
Table 2—Precipitating Antibodies and Isolated Fungi* Case No.
Fungi Isolated from Causative Environment
Serum Precipitating Antibodies
1
S fuliginea
2 3 4
Flour
S fuliginea Cladosporium sp Penicillium sp Fusarium sp Alternaria sp ND ND A niger Cladosporium sp Penicillium sp Acremonium sp Curvularia sp Fusarium sp Pithomyces sp A nidulans A flavus Penicillium sp Acremonium sp Mucor sp Arthroderma sp
2 A niger C acremonium
2
5
*ND 5 not done.
gions: the lymphoepithelium, the dome area, the follicular area, and the parafollicular area (Figure 1, middle). There was no significant difference in the clinical findings, including serum IgA levels, between the patients with BALT and those without BALT. Immunohistochemistry Immunohistochemistry revealed the cellular distribution of the BALT in CHP. The follicular area consisted primarily of CD20-positive B cells (Fig 2, top left). CD3-positive T cells were found mainly in the parafollicular area, although these cells were also present in the dome area (Fig 2, top right). Most CD3-positive T cells had no reactivity for anti-CD8, suggesting that CD4-positive T cells predominated over CD8-positive T cells there (Fig 2, middle left). Most Ki-67-positive cells were found in the ger-
Figure 1. Histologic features of BALT in CHP. Top, A: BALT is located in a bronchial wall (arrow head) (hematoxylin-eosin stain, original 325). Middle, B: BALT manifests four distinct lesions: a lymphoepithelium (LE), a dome area (DA), a follicular area (FA), and a parafollicular area (PFA) (hematoxylin-eosin stain, 366). Bottom, C: higher magnification of the lymphoepithelium. LE consists of cuboidal, nonciliated cells, and intraepithelial lymphocytes (arrow heads) (hematoxylin-eosin stain, original 3200).
Table 3—Histologic Findings* Case Bronchio- AlveoNo. litis litis 1 2 3 4 5
1 1 1 1 1
1 1 1 1 1
Organizing NonnecroTissues/ tizing Interstitial BOOP Granuloma Fibrosis BALT 1/2 1/2 1/1 1/1 1/2
2 2 2 1 1
1 2 1 1 2
2 2 1 1 1
*BOOP 5 bronchiolitis obliterans-organizing pneumonia; 1 5 present; 2 5 absent. 360
minal center of the follicular area (Fig 2, middle right). These cells had an expanded cytoplasm and large nuclei, characteristics of centroblasts. In contrast, bcl-2-positive cells were restricted to where Ki-67-positive cells were absent in the follicular area (Fig 2, bottom left). A few Ki-67-positive cells were also observed within the lymphoepithelium. Although S-100-positive cells were found in both Clinical Investigations
Figure 2. Cellular distribution of BALT in CHP. Top left, A: anti-CD20; top right, B: anti-CD3; middle left, C: anti-CD8; middle right, D: anti-Ki-67; bottom left, E: anti-bcl-2 (methyl green, original 366). The follicular area of BALT is composed primarily of CD20-positive B cells, whereas the parafollicular area comprises predominantly CD-3-positive, CD-8-negative T cells. Centroblast located in the germinal center expresses Ki-67 antigen, while bcl-2-positive cells are confined mainly to part of the follicular area, devoid of the germinal center.
the dome area of the BALT and the bronchiolar epithelium, the cells in the dome area did not express CD1a (Fig 3, top and middle). The S-100positive, CD1a-negative cells in the dome area had irregular dendritic morphology (Fig 3, bottom). Discussion BALT has been reported to play a crucial role in the development of a local immune response to inhaled antigens in animals, such as rabbits and rats.1,2 However, there are significant differences between species in the degree of organization of BALT. In human lungs, no organized BALT is found under normal circumstances.6 Thus, in contrast to GALT, most of the studies on BALT have been performed in animals. A few reports focused on BALT in humans.7,8,16,17 We previously reported the development of BALT in patients with DPB and
bronchiolar disease associated with rheumatoid arthritis,9,10 indicating that BALT is inducible under certain pathologic conditions. Further, in the present study, we demonstrated the development of BALT in patients with CHP. All patients described herein presented with chronic respiratory symptoms related to their exposure to causative antigens or environments. The exposure causing CHP in each case was determined on the basis of a positive response in a challenge test. On histologic examination, two cases had a complete histologic triad of hypersensitivity pneumonitis, described by Coleman and Colby,18 including cellular bronchiolitis, interstitial infiltrates, and nonnecrotizing granulomas. Although no granuloma was found in the other three subjects, their histologic findings were compatible with hypersensitivity pneumonitis. The lymphoid follicles seen in our patients had CHEST / 115 / 2 / FEBRUARY, 1999
361
Figure 3. Distribution of dendritic cells in BALT and bronchiole of CHP. Top, A: CD1a-positive cells are located in the bronchiolar epithelium (arrow heads) (methyl green, original 350). Middle, B: S-100-positive cells are seen in both the dome area of BALT and the bronchiolar epithelium (arrow heads) (methyl green, original 350). Bottom, C: higher magnification of S-100positive cells in the dome area. S-100-positive cells in the dome area of BALT have irregular dendritic morphology (arrow heads) (methyl green, original 3160).
lymphoepithelium consisting of nonciliated, cuboidal epithelial cells and intraepithelial lymphocytes. In addition, they were composed of four distinct regions: lymphoepithelium, dome area, follicular area, and parafollicular area. These findings indicated that the lymphoid follicles in CHP had the characteristics of BALT.19 BALT is considered a major site of 362
induction and amplification of the local immune response in the lungs of animals, because inhaled antigens can be taken up through its lymphoepithelium and presented to the lymphocytes in BALT.4,5 Thus, it is suggested that the BALT observed in our patients could act as an inductive site, resulting in amplification of the mucosal immune response in this disease, once it develops. In our previous report, BALT was observed in 12 of 17 patients with DPB.9 In DPB, we found that patients with BALT had a significantly higher level of serum IgA than did those without BALT. However, there was no difference in the clinical findings, including serum IgA levels, between patients with and without BALT in CHP. Immunohistochemical examination showed the cellular distribution of the BALT in CHP to be similar to that in DPB.9 The follicular area was composed mainly of B cells, while the parafollicular area comprised predominantly T cells. The centroblasts located in the germinal center expressed Ki-67 antigen, which is expressed on cells during all active stages of the cell cycle, suggesting that they were proliferating after antigenic stimulation in BALT. Bcl-2-positive cells were confined to part of the follicular area. Since bcl-2 antigen is expressed mainly on memory B cells, the cells in the follicular area primarily were memory B cells. In the dome area, we found S-100-positive, CD1a-negative cells having a dendritic shape. These cells were considered to be phenotypically and morphologically identical to the interdigitating dendritic cells in other lymphoid tissues.20,21 Thus, it is likely that inhaled antigens taken up through the lymphoepithelium are translocated to the dome area, where, in turn, the interdigitating dendritic cells in the dome area process and present these antigens to the local T cells. Little is known about the mechanism involved in the development of BALT. In DPB, persistent infection around the bronchioles and continuous microbial stimulation appear to be responsible for BALT development. In animals, chronic antigenic stimulation was reported to induce the full expression of organized BALT.1,2,22 On the other hand, systemic immunologic disorders are considered also to induce BALT formation as seen in rheumatoid arthritis.10 In CHP, it is possible that persistent respiratory inflammation elicited by antigen-specific immunologic reaction, in addition to repeated direct stimulation by a causative antigen, may lead to BALT development. However, further study is required to determine the precise mechanism and causative agents involved in BALT development, such as the cytokinetics over the period of BALT development. The present study indicates BALT development in CHP. BALT is considered to be induced by chronic antigenic stimulation and/or inflammation in CHP, Clinical Investigations
which, in turn, may play an important role in the mucosal immunity of this disease by acting as its inductive site. ACKNOWLEDGMENT: We thank Dr. K. Nishimura, Research Center for Pathologic Fungi and Microbial Toxicoses, Chiba University, for the assistance in identifying the fungi.
12
13
References 1 Bienenstock J, Johnston N, Perey DY. Bronchial lymphoid tissue: I. Morphologic characteristics. Lab Invest 1973; 28: 686 – 692. 2 Bienenstock J, Johnston N, Perey DY. Bronchial lymphoid tissue: II. Functional characteristics. Lab Invest 1973; 28: 693– 698 3 Bienenstock J. Gut and bronchus associated lymphoid tissue: an overview. Adv Exp Med Biol 1982; 149:471– 477 4 Racz P, Racz KT, Myrvik QN, et al. Functional architecture of bronchial lymphoid tissue and lymphoepithelium in pulmonary cell-mediated reactions in the rabbit. J Reticuloendothel Soc 1977; 22:59 – 83 5 Watanabe N, Kato H, Mogi G. Induction of antigen-specific IgA-forming cells in the upper respiratory mucosa. Ann Otol Rhinol Laryngol 1989; 98:523–529 6 Pabst R, Gehrke I. Is the bronchus-associated lymphoid tissue (BALT) an integral structure of the lung in normal mammals, including humans? Am J Respir Cell Mol Biol 1990; 3:131–135 7 Meuwissen HJ, Hussain M. Bronchus-associated lymphoid tissue in human lung: correlation of hyperplasia with chronic pulmonary disease. Clin Immunol Immunopathol 1982; 23: 548 –561 8 Rogers J, Langston C, Guerra IC. Pulmonary follicular lymphoid hyperplasia in a child with HTLV-III-related immunodeficiency. Pediatr Pulmonol 1986; 2:175–178 9 Sato A, Chida K, Iwata M, et al. Study of bronchus-associated lymphoid tissue in patients with diffuse panbronchiolitis. Am Rev Respir Dis 1992; 146:473– 478 10 Sato A, Hayakawa H, Uchiyama H, et al. Cellular distribution of bronchus-associated lymphoid tissue in rheumatoid arthritis. Am J Respir Crit Care Med 1996; 154:1903–1907 11 Hapke EJ, Seal RME, Thomas GO, et al. Farmer’s lung: a
14 15
16
17
18 19 20
21
22
clinical, radiographic, functional, and serologic correlation of acute and chronic stages. Thorax 1968; 23:451– 468 Buschman DL, Gamsu G, Waldron JJ, et al. Chronic hypersensitivity pneumonitis: use of CT in diagnosis. Am J Roentgenol AJR 1992; 159:957–960 Yoshida K, Ando M, Sakata T, et al. Environmental mycological studies on the causative agent of summer-type hypersensitivity pneumonitis. J Allergy Clin Immunol 1988; 81:475– 483 Bienenstock J. Bronchus-associated lymphoid tissue. Int Arch Allergy Appl Immunol 1985; 1:62– 69 Ando M, Sakata T, Yoshida K, et al. Serotype-related antigen of Trichosporon cutaneum in the induction of summer-type hypersensitivity pneumonitis: correlation between serotype of inhalation challenge-positive antigen and that of the isolates from patients’ homes. J Allergy Clin Immunol 1990; 85: 36 – 44 Richmond I, Pritchard GE, Ashcroft T, et al. Bronchus associated lymphoid tissue (BALT) in human lung: its distribution in smokers and nonsmokers. Thorax 1993; 48:1130 – 1134 Tschernig T, Kleemann WJ, Pabst R. Bronchus-associated lymphoid tissue (BALT) in the lungs of children who had died from sudden infant death syndrome and other causes. Thorax 1995; 50:658 – 660 Coleman A, Colby TV. Histologic diagnosis of extrinsic allergic alveolitis. Am J Surg Pathol 1988; 12:514 –518 Sminia T, van dB, Gamelkoorn GJ, et al. Structure and function of bronchus-associated lymphoid tissue (BALT). Crit Rev Immunol 1989; 9:119 –150 Wood GS, Tuner RR, Shiurba RA, et al. Human dendritic cells and macrophages: in situ immunophenotypic definition of subsets that exhibit specific morphologic microenvironmental characteristics. Am J Pathol 1985; 119:73– 82 Krenacs L, Tiszalvicz L, Krenacs T, et al. Immunohistochemical detection of CD1a antigen in formalin-fixed and paraffinembedded tissue sections with monoclonal antibody 010. J Pathol 1993; 171:99 –104 Weisz CP, Grimes SJ, Lamm ME. Gut-associated lymphoid tissue as source of an IgA immune response in respiratory tissues after oral immunization and intrabronchial challenge. Cell Immunol 1987; 106:132–138
CHEST / 115 / 2 / FEBRUARY, 1999
363