Introduction

Abstract Introduction Materials and Methods Results Discussion References

Asthma is characterized by eosinophilic inflammation of the airway that is thought to contribute to bronchial hyperresponsiveness (1, 2). Eosinophils, in common with other leukocytes, develop from CD34⁺ pluripotent progenitor cells. Interleukin (IL)-5 causes the selective expansion of eosinophils from bone-marrow precursors in model systems (3, 4). IL-5 responsiveness appears early during in vitro growth of eosinophils from CD34⁺ cells (5), and is presumably determined by receptor expression. IL-5 receptors comprise a cytokine-specific  subunit (IL-5R) that associates with a  chain shared with IL-3 and granulocyte macrophage colony-stimulating factor (GM-CSF) to form a high-affinity complex (6). Recent evidence suggests that eosinophil precursors may be recognized as CD34⁺/IL-5R⁺ cells (7).

Eidelman and colleagues demonstrated an increase in numbers of major basic protein-positive (MBP⁺) eosinophils after in vitro antigen challenge of bronchial explants removed from allergen-sensitized rats (8). In addition, increased eosinophil colony-forming activity was increased in the peripheral blood of atopic asthmatics after allergen challenge or controlled exacerbations of asthma, and circulating CD34⁺ cells were increased in atopic subjects (9, 10). Together, these observations led us to the hypothesis that, in addition to recruitment of mature eosinophils that differentiated in the bone marrow, local expansion of eosinophils from CD34⁺/IL-5R⁺ cells may contribute to eosinophilic airway inflammation in asthma. To determine whether immature progenitors that may have the potential to develop into eosinophils are present in the bronchial mucosa in asthma, we examined CD34⁺ cell numbers and expression of IL-5R messenger RNA (mRNA) by CD34⁺ cells in bronchial biopsies of atopic asthmatic, atopic nonasthmatic, and nonatopic control subjects.

	Materials and Methods

Abstract Introduction Materials and Methods Results Discussion References

Subjects

Nine atopic asthmatic and 10 atopic nonasthmatic patients, and 9 normal volunteers were examined. Patients were recruited from the Royal Brompton Hospital (London, UK), and volunteers responded to advertisements. Asthma was defined as a history of variable breathlessness and/or wheeze, documented 20% reversibility in peak flow or forced expiratory volume at 1 s (FEV₁), and airway hyperresponsiveness (a histamine PC₂₀  6 mg/ml). Atopy was defined by positive skin-prick tests (> 3 mm weal diameter, compared with control, at 15 min), or positive radioallergosorbent test (> 0.7 IU/ml) (Phadebas; Pharmacia, Uppsala, Sweden) to one or more of a panel of common aeroallergens. Pregnant women and smokers were excluded. None of the subjects was currently taking oral or inhaled corticosteroids. The study was approved by the Ethics Committee of the Royal Brompton Hospital. Patient details are given in Table 1.

Bronchoscopy and Processing of Bronchial Biopsies

Fiberoptic bronchoscopy was performed as previously described (11). An IT30 bronchoscope (Olympus Corp., Tokyo, Japan) was passed to the right or left lower lobe and biopsies were taken using FB-20C forceps (Olympus). Biopsies were immediately fixed in 4% paraformaldehyde (Sigma Chemicals, Poole, UK) for 1 h, followed by two changes of 15% sucrose (Sigma) for 1 h each. Biopsies were then mounted in OCT embedding medium (Miles, Inc., Elkhart, IN) and snap-frozen to 80°C.

Immunocytochemistry

Immunocytochemistry was performed using alkaline phosphatase antialkaline phosphatase (APAAP) as described previously (7), with some modifications. The monoclonal anti-CD34 (QBEND 10) was from Becton-Dickinson (San Jose, CA); anti-MBP was from Sanbio (Uden, The Netherlands); rabbit antimouse immunoglobulin (Ig), APAAP, and control IgG₁ were purchased from Dako (High Wycombe, UK). Optimal concentrations of all antibodies used were determined in pilot experiments. Omission or substitution of the primary antibody with an irrelevant antibody of the same species was used as a negative control.

In Situ Hybridization

The IL-5R complementary DNA (cDNA) (6) was used as a template for synthesis of a 92-base pair ³⁵S-labeled riboprobe specific for the membrane-associated isoform, as previously described (7). All reagents were from Sigma Chemicals, unless otherwise indicated. Permeabilization, prehybridization, and hybridization protocols were as described previously (7, 11). Incubation in N-ethyl maleimide, iodoacetamide, and triethanolamine reduced nonspecific binding of the ³⁵S-labeled probes. Negative controls employed hybridization with the sense probe and pretreatment of slides with RNAse A (Promega, Southampton, UK) before hybridization with the antisense probe.

Simultaneous In Situ Hybridization and Immunohistochemistry

To examine expression of IL-5R mRNA by CD34⁺ cells, sections were first stained with anti-CD34 monoclonal antibodies and developed with fast red, as above, then processed for in situ hybridization using ³⁵S-labeled riboprobe for IL-5R membrane-associated isoform (7).

Quantification of Immunohistochemistry and In Situ Hybridization

Slides were counted in duplicate blind to the patients' clinical status by means of an eyepiece graticule, as previously described (7, 11). Results are expressed as the numbers of cells per millimeter of basement membrane. The coefficient of variability of the duplicate counts obtained from all slides was less than 5%.

Statistical Analysis

Between patient groups, comparisons were performed using the nonparametric Mann-Whitney U test with Bonferoni's correction for comparisons between the three groups (Minitab, Inc., State College, PA). Correlations were sought using Spearman's rank correlation coefficient. P values < 0.05 were considered significant.

	Results

Abstract Introduction Materials and Methods Results Discussion References

Eosinophil numbers were increased in the bronchial mucosa of asthmatic subjects when compared with atopic control subjects (median 7.0 MBP⁺ cells/mm basement membrane, range 5.5-21.0 for asthmatic subjects; and 1.1, range 0-2.5, for atopic control subjects; P < 0.001) and when compared with nonatopic control subjects (median 0.88, range 0-2.5, P < 0.005).

In situ hybridization showed increased numbers of cells expressing IL-5R mRNA in the airways of asthmatic subjects (median 7.0/mm basement membrane, range 4.5- 16.0) when compared with atopic nonasthmatic subjects (2.1, range 0-3.0, P < 0.01) and nonatopic control subjects (0.88, range 0-2, P < 0.001). The mean total length of basement membrane counted was 1.6 mm, with a range of 0.5 to 1.8 mm.

Immunohistochemistry showed increased numbers of CD34⁺ cells in bronchial biopsies from both atopic asthmatic and atopic nonasthmatic subjects when compared with nonatopic nonasthmatic control subjects (Figure 1a). There was no significant difference in CD34⁺ cell numbers in biopsies between asthmatic and atopic nonasthmatic subjects. CD34 immunostaining was seen in a pattern suggesting endothelial cell staining, but also in scattered cells within the subepithelial cell layer.

View larger version (17K): [in this window] [in a new window]   Figure 1.   Numbers of CD34+ cells enumerated by immunohistochemistry (a) and CD34+/IL-5R mRNA+ cells enumerated by simultaneous immunohistochemistry and in situ hybridization (b) of bronchial mucosal biopsies from atopic asthmatic (n = 9), atopic nonasthmatic (n = 10), and nonatopic control subjects (n = 9). Slides were counted in duplicate, and cell counts in the submucosa are expressed per millimeter of basement membrane.

Simultaneous immunohistochemistry for CD34 and in situ hybridization for IL-5R mRNA showed colocalization of signal in cells scattered within the subepithelial layer (Figure 2). CD34⁺/IL-5R mRNA⁺ cell numbers were increased in bronchial biopsies from asthmatic subjects when compared with atopic nonasthmatic and nonatopic control subjects (Figure 1b). There were also increased numbers of CD34⁺/IL-5R mRNA⁺ cells in biopsies from atopic control subjects when compared with those from nonatopic control subjects.

View larger version (95K): [in this window] [in a new window]   Figure 2.   Immunohistochemical localization of CD34+ cells in the bronchial mucosa of an asthmatic subject. Bronchial biopsies were stained with monoclonal antibody against CD34 (QBEND 10), and staining was developed with APAAP and fast red. The basement membrane is arrowed (bm). (b) In situ hybridization for IL-5R chain mRNA in a bronchial biopsy from an asthmatic subject. Positive cells are overlaid by dense deposits of silver grains. The basement membrane is arrowed (bm).

Among the asthmatic subjects there was a significant correlation between eosinophil numbers and numbers of IL-5R mRNA⁺ cells in bronchial biopsies (r_s = 0.90, P < 0.001). There were also correlations between FEV₁ and CD34⁺/IL-5R mRNA⁺ cell numbers for asthmatic subjects (r_s = 0.72, P < 0.02). Among atopic asthmatic and atopic nonasthmatic subjects there was a correlation between CD34⁺/IL-5R mRNA⁺ number and eosinophil number in bronchial biopsies (r_s = 0.70, P < 0.01).

	Discussion

Abstract Introduction Materials and Methods Results Discussion References

In this study we show increased numbers of CD34⁺ cells in the bronchial mucosa of both atopic asthmatic and atopic nonasthmatic subjects. However, CD34⁺/IL-5R mRNA⁺ cell numbers were increased only in the airways of asthmatic subjects, and numbers of these cells were related to asthma severity as judged by airway caliber. This suggests that airway progenitor cell numbers are increased in atopy, but that increased numbers of CD34⁺ cells with the potential to respond to IL-5 are a feature of asthma. These findings raise the possibility that eosinophils may differentiate in situ, at the site of allergic bronchial inflammation as well as from the bone marrow.

We and others have previously shown evidence for increased expression of IL-5, together with IL-3 and GM-CSF in the airways of atopic asthmatic subjects (11). IL-5 expression correlated with measures of asthma severity such as FEV₁, bronchial responsiveness, and symptom scores, and probably acts to prolong survival of eosinophils in the airway and prime for their activation and degranulation (14). In addition, animal studies suggest that IL-5 can act distantly both to mobilize a bone marrow pool of eosinophils and to increase eosinophil differentiation from bone marrow precursors (17). However, recent experiments involving in vitro antigen challenge of isolated rat airway explants raised the possibility that eosinophil differentiation might also occur in the tissue (8). We have also shown increased MBP⁺ cells in human nasal mucosal explants after in vitro allergen challenge (18). Our current data show that CD34⁺/IL-5R⁺ cells are present in the airway in asthma. Further work is required to determine whether these cells can differentiate into eosinophils, but their presence supports the concept that local tissue differentiation and expansion of eosinophils may occur in human asthma.

IL-5R chain can be expressed in several isoforms by differential gene splicing to generate different mRNAs: one membrane-associated isoform and two soluble isoforms which, at least in vitro, have the potential to antagonize the actions of IL-5 (19). We used a riboprobe selective for the membrane-associated isoform, but did not show that IL-5R is indeed expressed at the surface of the CD34⁺ cells studied. However, in previous studies of CD34⁺/IL-5R⁺ cells from bone marrow (7), and in our own work on cord blood progenitors (unpublished data), mRNA expression and surface expression were closely related.

In addition to eosinophils, IL-5R is also expressed by basophils (20), and mixed eosinophil/basophil colonies in response to IL-5 have been described (21), so that CD34⁺/ IL-5R⁺ cells may contribute to local basophil expansion in the airway, as well as eosinophils.

Much current interest is focused on targeting mechanisms of eosinophil recruitment into the airway in asthma. Of particular importance is the role of eotaxin and other chemokines acting on CCR3 (22), which is restricted in expression to eosinophils, basophils, and some T-helper- 2 cells (23). These chemokines also act in recruitment and mobilization of bone marrrow eosinophils (24). Our current data raise the possibility that control of CD34⁺ cell influx and local differentiation and expansion of CD34⁺/ IL-5R⁺ cells may also be required to reduce eosinophilic inflammation in the airway in asthma. We did not detect any increase in CD34⁺/IL-5R⁺ cells in peripheral blood in asthmatic compared with control subjects (data not shown). However, Sehmi and coworkers showed increased numbers of CD34⁺/IL-5R⁺ cells in bone marrow of atopic asthmatic subjects after allergen inhalation challenge (7). Whether CD34⁺/IL-5R⁺ are recruited from the bone marrow or whether CD34⁺ cells acquire IL-5R in the airway remains to be established. To date, SDF-1 is the only chemokine characterized as acting in the mobilization of CD34⁺ progenitor cells (25). Whether this or other chemokines act in recruitment of CD34⁺ cells to the airway in atopic subjects is unknown.

We show increased CD34⁺ cell numbers in both asthmatic and nonasthmatic atopic subjects when compared with nonatopic control subjects. Similar findings are reported for bone marrow and peripheral blood (10). In contrast, CD34⁺/IL-5R mRNA⁺ cell numbers were more prominent in atopic asthmatic than nonasthmatic atopic subjects. Whether this results from local proliferation of CD34⁺/IL-5R mRNA⁺ cells, selective survival of CD34⁺ cells expressing IL-5R, or factors driving CD34⁺ cells to express IL-5R in asthma remains to be determined. In general, lineage commitment of CD34⁺ progenitors is thought to arise from selective expansion of subsets arising in a stochastic or random manner (26). Whether this is true of circulating CD34⁺ cells or whether selected subpopulations are released from the bone marrow also remains to be determined. It is of note that there was a small but significant increase in CD34⁺/IL-5R mRNA⁺ cell numbers in the bronchial mucosa of nonasthmatic atopic subjects compared with nonatopic subjects, but no increase in mature eosinophils. Airway IL-5 expression may be required for full maturation, expansion, and survival of the eosinophil phenotype.

In conclusion, we show here that there are increased numbers of CD34⁺/IL-5R⁺ cells in the airway in asthma. This raises the possibility of local differentiation of eosinophils from progenitors. These findings have implications for other inflammatory conditions, and an understanding of the processes regulating local differentiation of progenitor cells may have therapeutic potential in asthma and other chronic inflammatory diseases.