INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: basophils; asthma; human

Basophils express high levels of the high-affinity IgE receptor, FcRI, and are effector cells in systemic allergic inflammation because of their ability, upon stimulation, to release histamine and other preformed mediators from granules and to synthesize additional mediators and cytokines. Basophils have also been implicated as effector cells in late-phase allergic reactions in the skin (1) after nasal allergen challenge (2) and in the nasal mucosa of atopics (5). Their numbers are reportedly increased in both the upper and lower airways (6, 7) during the late-phase response of allergic patients. In addition, increased numbers of basophils have been reported in the airways of patients dying from asthma compared with lung tissues of individuals dying from other causes (8), in bronchoalveolar lavage fluid (BALF) from asthmatics after ragweed challenge (9), and in both BALF and peripheral blood from patients with asthma in comparison with nonasthmatics (10). Mast cells release both tryptase and histamine whereas basophils release only histamine. Therefore, basophils were further implicated in the pathogenesis of asthma because there was a lack of correlation of tryptase with histamine in BALF from allergic asthmatics (15). These previous studies used metachromatic dyes, anti-IgE antibodies, and other indirect criteria for determination of basophil numbers. However, basophils lose their ability to stain with metachromatic dyes after exposure to aqueous media (16), and other cells besides mast cells and basophils can bind IgE (17), making an accurate assessment of basophil numbers difficult.

Previous studies have identified a basophil-specific antigen recognized by the monoclonal antibody (mAb), 2D7 (16). This antibody (Ab) does not react with mast cells, eosinophils, neutrophils, or other leukocytes. In this study, we used immunohistochemistry with 2D7 to quantify and compare postmortem lung basophil numbers in individuals who died from asthma with subjects dying from nonasthmatic causes.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Materials

The mouse anti-human basophil mAb, 2D7 (IgG1) (16), was a gift of Dr. Lawrence Schwartz (Virginia Commonwealth University, Richmond, VA). Protease Type I, 10% neutral buffered formalin (NBF), peroxidase-conjugated streptavidin, nonspecific mouse IgG1 (MOPC31-C), 30% H₂O₂, normal goat serum, and 3-amino-9-ethylcarbazole (AEC) were obtained from Sigma, St. Louis, MO. CD45 (murine IgG1, clones 2B11 and PD7/26), a pan-leukocyte antibody, was obtained from Dako, Carpinteria, CA. Biotin-conjugated, goat anti-mouse IgG was obtained from Jackson Labs, West Grove, PA.

Tissue Specimens

Paraffin tissue blocks were screened using a hematoxylin-eosin stain to identify the presence of suitable airways for analysis. Lung tissue sections were divided into three groups based on pathology records from the New Mexico Office of the Medical Investigator (OMI). The first group of sections was from individuals whose autopsy reports identified asthma as the primary or major contributing cause of death (fatal asthma [FA]; n = 5). The second group of sections was from individuals whose autopsy reports identified a history of asthma and had clear evidence of asthma histologically, but died of causes other than asthma (nonfatal asthma [NFA]; n = 6). The third group of sections was from individuals whose autopsy reports identified no history of asthma nor any histologic evidence of asthma (controls [C]; n = 4). All material was obtained with approval of the OMI Research Review Committee.

Immunohistochemistry

Immunohistochemistry was performed as described previously (16). Briefly, NBF-fixed, protease-digested tissue sections or cytospins were incubated with 2D7 mAb (10 µg/ml), CD45 (1 µg/ml), or MOPC31-C (10 g/ml) overnight at 4° C in a humid chamber. After washing, sections were incubated with biotin-conjugated goat anti-mouse IgG followed by peroxidase-conjugated streptavidin, each for 1 h. Immunoreactivity was detected using AEC. Cytocentrifuge preparations of peripheral blood basophils were used as a positive control for the immunohistochemical protocol.

Quantitation

All slides were randomly coded and analyzed by a single blinded observer without knowledge of the subjects. Light and phase microscopy as well as photography were performed with a Zeiss Axioskop 2 microscope fitted with a MC 80 DX microscope digital camera and a Zeiss photoreticule eyepiece (for tissue area calculation). The number of cells staining positively with 2D7 or CD45, shown as total number of cells/mm² in separate fields of the section, was determined by counting cells: (1) around large and small airways, including the basal lamina to the luminal edge of the smooth muscle (designated AW); and (2) in alveolar ducts and alveoli (designated AL). To adjust for differences in air space within each of the two areas, computer-assisted measurements were performed where only cells within the relevant tissue were counted and air space was excluded.

Statistical Analysis

Mean experimental values between the three groups were analyzed using analysis of variance (ANOVA). In addition, an unpaired t test with Welch's correction was used to test the significance of cell number differences found in AL or AW from each group. Differences were accepted as statistically significant at p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Clinical Characteristics and Pathologic Features

Patient data are shown in Table 1. The average age in the FA group was 31 (range 14 to 44 yr), in the NFA group 50 (range 38 to 80 yr), and for the C group 24 (range 18 to 54 yr). No information was available about steroid use for the asthmatic patients.

CD45-positive cells

Immunohistochemistry with CD45 Abs revealed a wide scatter of inflammatory cells infiltrating the lungs in all three groups of subjects. Inflammatory cell infiltrates surrounded large and small airways and in the alveolar walls in FA. Compared with alveolar walls and sacs (AL), the area immediately surrounding the airway lumen (AW) had a greater density of CD45-positive cells, but the difference was not significant (p = 0.052) (Figures 1A and 1B). CD45-positive cells were scattered throughout the lung sections (Figures 2A and 2B). Values ranged from 14,904 to 16,789 cells/mm² (x = 13,103; SEM ± 1,331) around AW and 9,040 to 12,088 cells/mm² (x = 9,249; SEM ± 926) in AL. In NFA subjects, CD45-positive cells ranged from 7,614 to 11,179 cells/mm² (x = 9,917; SEM ± 1,275) in AW and 6,695 to 9,800 cells/mm² (x = 8,603; SEM ± 1,416) in AL (Figures 1C and 1D). In control subjects CD45-positive cells ranged from 5,498 to 10,477 cells/mm² (x = 8,310; SEM ± 1,051) in AW and 6,155 to 8,467 cells/mm² (x = 7,704; SEM ± 530) in AL (Figures 1E and 1F). Although there were differences in numbers of CD45-positive cells among the three groups the only significant differences were between both FA and NFA and controls in AW CD45-positive cells. Lung sections from FA and NFA had significantly more (p = 0.03; R² = 0.48) CD45-positive cells in this region compared with similar regions in control sections.

View larger version (30K): [in this window] [in a new window]   Figure 1.   Quantitation of CD45-positive leukocytes. The numbers of inflammatory cells were counted around large and small airways (AW) and in the septae of alveolar ducts and alveoli (AL) in lung sections of FA (A and B), NFA (C and D), and control subjects (E and F ). The total numbers of CD45-positive cells/mm2 are given for each specimen in A, C, and E whereas the mean (± SEM) values are given in B, D, and F.

View larger version (175K): [in this window] [in a new window]   Figure 2.   Immunostaining for CD45 and 2D7. Lung sections from FA cases were incubated with anti-CD45 mAb (A, B) or mAb 2D7 (C, D) and immunoreactivity detected using peroxidase-conjugated secondary Abs and AEC. Substitution of CD45 or 2D7 Abs with an irrelevant mouse IgG1 Ab resulted in no staining (data not shown). Original magnification: ×300.

Basophils

Formalin-fixed tissue sections require digestion with 0.1% protease for 20 min at room temperature to permit labeling by 2D7 (18). The staining exhibited both discrete cellular localization and diffuse extracellular staining consistent with the 2D7 antigen being a releasable granule constituent upon cell activation (16).

Basophil numbers in AW and AL were quantified by counting the number of positively stained, nucleated cells and by measuring the tissue area stained. The concentration of 2D7+ basophils was higher in the lung specimens from the FA subjects than the lung specimens from NFA subjects and control subjects. In FA subjects, the basophil infiltrate ranged from 41 to 119 cells/mm² (x = 82; SEM ± 14) around AW and 56 to 108 cells/mm² (x = 70; SEM ± 10) in AL (Figures 3A and 3B). Intense staining with 2D7 occurred along both large and small airways as well as in alveolar septi (Figures 2C and 2D). Substituting the 2D7 antibody with an irrelevant mouse IgG1 resulted in no tissue labeling (data not shown). Basophils were found in the submucosa as well as the lumen of the bronchioles and around the vascular walls. In the lungs of NFA 0 to 14 cells/mm² (x = 7.6; SEM ± 5.9) in AW and 1 to 16 cells/mm² (x = 7.2; SEM ± 2.6) in AL (Figures 3C and 3D) were detected. In control patients, 0 to 13 cells/mm² (x = 7.3; SEM ± 2.8) around AW and 0 to 9 cells/mm² (x = 4; SEM ± 2.0) in AL (Figures 3E and 3F) were detected. No significant differences in the number of basophils were seen when comparing tissue areas within each group. However, when comparing total numbers of basophils observed in FA cases versus NFA or controls cases, the numbers of basophils were significantly greater in both AW (p < 0.0001; R² = 0.81) and AL (p < 0.0001; R² = 0.86) from the lungs of FA cases. No significant differences in basophil numbers were seen in lungs of NFA cases compared with control subjects.

View larger version (23K): [in this window] [in a new window]   Figure 3.   Quantitation of 2D7-positive basophils. The numbers of basophils were counted around AW and in AL in lung sections of FA (A and B), NFA (C and D), and control subjects (E and F ). The total numbers of 2D7-positive cells/mm2 are given for each specimen in A, C, and E whereas the mean (± SEM) values are given in B, D, and F.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have used a basophil-specific antibody to determine basophil numbers in postmortem lung tissue from subjects who died from asthma and from subjects who died from other causes, with and without asthma. The antigen recognized by 2D7 has not yet been characterized, but its specificity for basophils was established in previous studies (16, 18). Our study demonstrates an increased number of basophils in the lungs of subjects with FA.

Previously, Koshino and coworkers also identified increased basophil numbers in the airways of subjects with FA examined postmortem (8). These studies distinguished basophils from mast cells and other cells on the basis of their positive staining with anti-IgE antibodies and their negative staining with anti-tryptase antibodies and so ran the risk of including other IgE-receptor-positive cell types such as dendritic cells and Langerhans cells (17). Nonetheless, the densities of basophils in the lungs of asthmatics reported here (x = 76 basophils/mm²) and previously (x = 82 basophils/mm²), are very similar.

There was also a significant increase in total inflammatory cells, as identified by CD45 positivity, in the FA and NFA versus the control lung. Other studies have found a similar abundant leukocyte infiltrate in the lungs of fatal asthmatics (19- 24). In particular, an increased number of eosinophils has been reported as an important component of the peripheral inflammation in FA. Eosinophils are absent from the airways of normal healthy control subjects but, like basophils, their numbers are significantly higher in bronchoscopies from asthmatics (25) as well as in the lungs of patients dying from asthma (22, 27, 28). Interestingly, we have previously reported that both basophils and eosinophils can be detected using Abs to eosinophil peroxidase (EPO) and eosinophil cationic protein (EG1 and EG2) which are used typically for eosinophil quantitation (29). It is thus possible that the mAbs used in previous studies to identify eosinophils may also have detected basophils and that the similar adhesion receptor profiles of these two cell types may cause the recruitment of both basophils and eosinophils to the lungs in FA.

In contrast to fatal episodes of asthma, an increase in basophil numbers in stable clinical asthma has not been well documented. In BALF, Kirby and coworkers did not find an increase in basophil numbers in asthmatics verses nonasthmatics (30). In other studies, only small increases in basophil numbers have been reported in BALF (9, 31) and bronchial biopsies (32) from airways during the late inflammatory response after antigen challenge. Consistent with these results, we did not detect substantial numbers of basophils in lungs from subjects with asthma who died of other causes, even though their total leukocyte numbers were substantially increased over control. Therefore, basophils may play a more important role in the pathogenesis of FA than in nonfatal, chronic asthmatic inflammation.

Increased basophils in the lungs could contribute to the pathogenesis of FA in a variety of ways. Besides mast cells, basophils are the only cell type which store and release large amounts of histamine. When inhaled by asthmatics, histamine induces bronchoconstriction, increases bronchial blood flow and increased microvascular permeability to macromolecules, promotes the increased secretion of mucus by isolated human airways, and is a chemoattractant for eosinophils (33). Basophils are also a major source of interleukin-4 (IL-4) and IL-13 (34), two potent proinflammatory cytokines involved in the pathogenesis of asthma. Finally, basophils store and release large amounts of leukotriene C₄ and B₄ (35), which are the most powerful bronchoconstrictor agents tested in vitro and in vivo (33). Thus, basophils can release noxious, granule-derived mediators implicated in asthma severity; the release of these agents in the lung may have severe and sometimes fatal consequences.