Expression of CXCR6 and its ligand CXCL16 in the lung in health and disease.

BACKGROUND: Chemokine receptors (CR) play an important role in T cell migration, but their contribution to lung trafficking is unclear. OBJECTIVE: We hypothesized that if a particular CR was involved in T cell homing its expression would be enriched on lung T cells compared with peripheral blood T cells (PBT). METHODS: We have measured the CR expression on BAL T cells from patients with sarcoid, other interstitial lung diseases (ILD), asthma and healthy volunteers. RESULTS: Of 14 CR studied in sarcoid, CXCR6 expression was the most markedly increased in the lung compared with the blood, a finding that was also seen in ILD patients. A striking although lesser increase was also seen in asthmatics and healthy controls. Analysis of expression of the CXCR6 ligand, CXCL16, by immunohistochemistry suggested that alveolar macrophages (AM) were the major source of CXCL16 in the lung. AM expressed mRNA for CXCL16 and released nanogram quantities after adhesion to plastic as shown by RT-PCR, Western blotting and ELISA. Bronchoalveolar lavage (BAL) fluid from all subjects contained large amounts of CXCL16. The full-length CXCL16 was the predominant isoform in AM lysates, supernatants and BAL. CONCLUSION: This data suggests that CXCR6 and CXCL16 may play a role in T cell recruitment to the lung.

Idiopathic chronic cough: association with organ specific autoimmune disease and bronchoalveolar lymphocytosis.

BACKGROUND: We have recently reported a strong association between organ specific autoimmune disease and idiopathic chronic cough and have suggested that cough may be caused by airway inflammation secondary to aberrant homing of activated lymphocytes to the lung. An immunopathological study was undertaken to test the hypothesis that idiopathic chronic cough is associated with lymphocytic airway inflammation. METHODS: Bronchoscopy, bronchial biopsies, bronchoalveolar lavage (BAL), and peripheral blood and BAL flow cytometry were performed in 19 patients with idiopathic chronic cough, 14 with explained chronic cough, and 11 normal subjects. RESULTS: Organ specific autoimmune disease or positive autoantibodies were present in eight of the 19 patients with idiopathic cough, in one of the 14 patients with explained cough, and in one of the 11 normal subjects. Median BAL fluid differential lymphyocyte counts were significantly higher in patients with idiopathic cough (10.0%) than in normal subjects (6.3%, 95% confidence interval of difference 1.5 to 11.9, p = 0.01) or patients with explained cough (5.2%, 95% CI of difference 2.0 to 10.4, p = 0.001). There were no differences in bronchial biopsy T lymphocyte counts between the groups. The mean (SE) proportion of CD3+ peripheral blood mononuclear cells expressing CD4 was significantly higher in normal subjects than in patients with idiopathic cough (69 (3)% v 58 (3)%, mean difference 11%, 95% CI of difference 2 to 20, p<0.02) but not than those with explained chronic cough (63 (2)%). There were no differences in BAL T lymphocyte phenotype between groups. CONCLUSION: BAL fluid lymphocytosis occurs in some patients with idiopathic chronic cough. The association of idiopathic chronic cough with organ specific autoimmune disease raises the possibility that this might be caused by lymphocyte homing from the primary site of autoimmune inflammation or the result of an autoimmune process in the lung.

Interleukin-4 and -13 expression is co-localized to mast cells within the airway smooth muscle in asthma.

BACKGROUND: Airway smooth muscle infiltration by mast cells is a feature of asthma and not eosinophilic bronchitis. In asthma, Th2 cytokines have been implicated as playing a critical role in the development of airway inflammation and hyper-responsiveness. Whether inflammatory cells within the airway smooth muscle release these cytokines is unknown. METHODS: We have undertaken a comparative immunohistochemical study in bronchial biopsies from 14 subjects with asthma, 10 with eosinophilic bronchitis and eight normal controls recruited from two centres. RESULTS: The median number of IL-4+ cells/mm2 smooth muscle was significantly higher in subjects with asthma than eosinophilic bronchitis and normal controls for both the anti-IL-4 mAb 3H4 (2.4, 0, 0, respectively; P=0.001) and anti-IL-4 mAb 4D9 (1.6, 0, 0, respectively; P=0.02). There were no group differences in the number of IL-5+ cells (P=0.31). In six subjects with asthma, IL-13 expression by cells within the airway smooth muscle was studied. The median (range) of IL-13+cells was 2 (0.9-2.7). Ninety-four percent of the cells expressing IL-4 (3H4), 92% of those expressing IL-4 (4D9) and 100% expressing IL-13 in the airway smooth muscle were mast cells. Fifty-five percent of the mast cells within the airway smooth muscle co-localized to IL-4 (3H4), 29% to IL-4 (4D9) and 17% to IL-13. CONCLUSIONS: In asthma, IL-4+ and IL-13+ cells were present within the airway smooth muscle and were expressed predominantly by mast cells, suggesting that IL-4 and IL-13 may play an important role in mast cell-airway smooth muscle interactions.

Comparison of airway immunopathology of eosinophilic bronchitis and asthma.

BACKGROUND: Eosinophilic bronchitis is a condition characterised by a corticosteroid responsive cough, sputum eosinophilia, and normal tests of variable airflow obstruction and airway responsiveness. We performed a detailed comparative immunopathological study to test the hypothesis that the different airway function in patients with eosinophilic bronchitis and asthma reflects differences in the nature of the lower airway inflammatory response. METHODS: Exhaled nitric oxide was measured and induced sputum, bronchoscopy, bronchial wash (BW), bronchoalveolar lavage (BAL), and bronchial biopsy were performed in 16 subjects with eosinophilic bronchitis, 15 with asthma, and 14 normal controls. RESULTS: Both eosinophilic bronchitis and asthma were characterised by an induced sputum, BW and BAL eosinophilia, an increased number of epithelial and subepithelial eosinophils, and increased reticular basement membrane thickness. The median concentration of exhaled nitric oxide was higher in those with eosinophilic bronchitis (12 ppb) or asthma (8.5 ppb) than normal controls (2 ppb) (95% CI of the difference 5 to 16, p<0.0001 and 2 to 11.3, p=0.004, respectively). There were no group differences in epithelial integrity or the number of subepithelial T lymphocytes, mast cells or macrophages. CONCLUSION: With the exception of our previously reported association of smooth muscle mast cell infiltration with asthma, the immunopathology of eosinophilic bronchitis and asthma are similar which suggests that eosinophilic airway inflammation, increased exhaled nitric oxide, and increased basement membrane thickening are regulated independently of airway hyperresponsiveness.

Characterisation of adhesion receptors mediating lymphocyte adhesion to bronchial endothelium provides evidence for a distinct lung homing pathway.

BACKGROUND: The lung is an important tertiary lymphoid organ and many lung diseases are associated with disordered lung immunity. Unlike the gut (alpha4beta7 binding to MAdCAM-1) and skin (CLA+ve T cells binding to E-selectin) where the adhesion receptors involved in organ specific homing of T cells have been identified, the molecular pathways controlling lymphocyte migration to the lung are unclear. Using a modified version of the Stamper-Woodruff assay we have investigated the receptors mediating adhesion of peripheral blood lymphocytes to airway endothelium. METHODS: Longitudinal frozen sections of bronchus (8 micro m) obtained from lung resection specimens were incubated with T cell enriched peripheral blood mononuclear cells for 30 minutes under shaking conditions in the presence of a fluorescently labelled polyclonal anti-von Willebrand antibody to identify blood vessels. After fixation the percentage of blood vessels supporting adhesion was measured. Blocking monoclonal antibodies were used to determine the role of individual adhesion receptors in lymphocyte binding. RESULTS: Specific cation dependent binding of lymphocytes to bronchial endothelium was observed which was significantly inhibited by antibodies against P-selectin, PSGL-1, L-selectin, LFA-1, ICAM-1 and ICAM-2 but not E-selectin, VLA-4, VCAM-1 or Mac-1. This was consistent with the pattern of endothelial expression of these receptors with strong expression of P-selectin and ICAM-1, but negligible expression of E-selectin on bronchial endothelium. CONCLUSION: This study suggests an important role for PSGL-1/P-selectin in T cell migration into the bronchi and provides further evidence for a pattern of recirculation for respiratory tract homing T cells distinct from the gut and skin.

A case of cough, lymphocytic bronchoalveolitis and coeliac disease with improvement following a gluten free diet.

Chronic cough is a common reason for presentation to a respiratory clinic. In up to 20% of cases the cause remains unclear after investigations. We report one such case where there was bronchoscopic evidence of lymphocytic airway inflammation in association with newly diagnosed coeliac disease. All features improved markedly on a gluten free diet, suggesting a causal relationship between coeliac disease, cough, and lymphocytic bronchoalveolitis.

Expression of chemokine receptors by lung T cells from normal and asthmatic subjects.

The lung is an important tertiary lymphoid organ with constant trafficking of T cells through the lung in both health and disease. T cell migration is controlled by a combination of adhesion receptors and chemokines expressed on vascular endothelium and in the tissue, often in an organ-specific manner. This leads to selective accumulation of different T cell subsets, a process called lymphocyte homing. There is evidence for a distinct lung-homing pathway, but no specific lung-homing receptors have been described. We analyzed the chemokine receptor profile of lung T cells to determine the extent to which lung T cells shared homing pathways with other organs such as the gut and skin. In addition, we compared expression of receptors in normal and asthmatic individuals to determine whether different pathways were used in health and disease. We observed that lung T cells expressed a profile of chemokine and adhesion receptors distinct from that of gut- and skin-homing T cells although no chemokine receptor specific for the lung was found. In particular, lung T cells expressed CCR5 and CXCR3, but not CCR9 or cutaneous lymphocyte Ag, and only low levels of CCR4 and alpha(4)beta(7). No differences were observed between lung T cells from normal vs asthmatic subjects. This study provides added support for the concept of a lung-homing pathway separate from other mucosal organs such as the gut and suggests that the chemokine pathways that control T cell migration in normal homeostasis and Th2-type inflammatory responses are similar.

Interleukin-13 induces PSGL-1/P-selectin-dependent adhesion of eosinophils, but not neutrophils, to human umbilical vein endothelial cells under flow.

Selective eosinophil accumulation is a hallmark of diseases such as asthma. In a model of chronic eosinophilic inflammation, we have previously shown that the tethering step in eosinophil adhesion is mediated by PSGL-1 binding to P-selectin. The Th2-associated cytokine IL-13 is of potential importance in allergic disease. We have therefore investigated whether IL-13 can mediate eosinophil binding to human umbilical vein endothelial cells (HUVEC) through P-selectin. IL-13 caused dose- and time-dependent increases of P-selectin expression, as assessed by flow and laser scanning cytometry. A similar degree of expression was observed with IL-4. There was no effect on E-selectin or ICAM-1 expression. Tumor necrosis factor-alpha induced the expression of VCAM-1, E-selectin, and ICAM-1 but had no effect on P-selectin expression. IL-13 increased the production of mRNA for surface and soluble variants of P-selectin. Under flow conditions, eosinophils, but not neutrophils, showed enhanced binding to IL-13 and to IL-4-stimulated HUVEC compared to medium-cultured cells. Eosinophil adhesion was completely inhibited by a blocking monoclonal antibody against PSGL-1 and P-selectin. Anti-VLA-4 and anti-VCAM-1 antibodies inhibited binding to a lesser extent. Thus, at physiologic levels of expression induced by Th2 cytokines, P-selectin/PSGL-1 supported eosinophil but not neutrophil adhesion. This mechanism is likely to be a key event leading to the selective accumulation of eosinophils in allergic inflammation.

Interleukin-5 enhances eosinophil adhesion to bronchial epithelial cells.

BACKGROUND: Eosinophil-bronchial epithelial cell interactions are thought to be central to the pathogenesis of asthma, both in terms of the epithelium as a source of pro-inflammatory mediators and as a target for eosinophil-mediated damage. We have therefore investigated adhesion interactions between these two cell types. OBJECTIVES: To determine the role of eosinophil and epithelial activation on eosinophil adhesion to bronchial epithelium and to characterize the adhesion receptors mediating eosinophil adhesion. METHODS: Eosinophils were purified from human peripheral blood by immunomagnetic selection and adhesion to confluent cultures of the airway epithelial cell lines A549 and BEAS-2B was studied. RESULTS: Stimulation of A549 cells with TNFalpha, IFNgamma or a combination of 50 ng/mL of TNFalpha, IFNgamma and IL-1 (cytomix) did not effect eosinophil binding despite an increase in ICAM-1 expression. Similarly stimulation of eosinophils with PAF or IL-5 had no effect on eosinophil binding to medium- or cytokine-treated A549 cells. In contrast stimulation of BEAS-2B cells with cytomix caused a significant increase in eosinophil adhesion. This was associated with an increase in expression of ICAM-1 and induced expression of VCAM-1. Treatment of eosinophils with Mn2+ and IL-5 but not eotaxin, RANTES or PAF also significantly enhanced eosinophil adhesion to medium-treated BEAS-2B cells. Using blocking mAbs we were able to demonstrate that the increased adhesion resulting from stimulation of eosinophils or BEAS-2B cells was in both cases mediated by a combination of CD18 and alpha4 integrins. CONCLUSIONS: This study demonstrates a selective role for IL-5 in mediating integrin-dependent eosinophil adhesion to airway epithelium and once again emphasizes the importance of this cytokine in controlling eosinophil activation in diseases such as asthma.

Characterization of the integrin and activation steps mediating human eosinophil and neutrophil adhesion to chronically inflamed airway endothelium.

We have used the Stamper-Woodruff frozen-section assay (FSA) to characterize the integrin and activation steps involved in adhesion of peripheral blood eosinophils and neutrophils to nasal polyp endothelium (NPE). Eosinophil and neutrophil adhesion was significantly inhibited by monoclonal antibodies (mAbs) against CD18 (beta2) and CD11a-c. Eosinophil adhesion was also inhibited to a lesser extent by mAbs against CD29 (beta1), CD49d (alpha4), and vascular cell adhesion molecule-1. The involvement of integrins raised the possibility of an activation step being involved in the adhesion process. Although stimulation of the cells with granulocyte macrophage colony-stimulating factor (GM-CSF) before the assay failed to modulate adhesion, binding was inhibited by up to 50% by treatment of the leukocytes with azide. In addition, neutrophil adhesion was completely abrogated by pertussis toxin (PT) and inhibited by about 50% by the platelet-activating factor antagonist WEB 2086 and antibodies against interleukin (IL)-8 and the two IL-8 receptors IL8RA and IL8RB (C-X-CR1 and -CR2). In contrast, eosinophil adhesion was unaffected by PT, WEB 2086, or anti-IL8R mAbs. mAbs against CCR-3, IL-3, IL-5, and GM-CSF also had no effect. This study demonstrates that eosinophil and neutrophil adhesion to NPE in the FSA conforms to the multistep paradigm for leukocyte adhesion and can be used to model the molecular basis for adhesion to endothelium in the context of chronic inflammatory disease. Using this assay, we have observed significant differences in integrin usage between eosinophils and neutrophils and a striking difference in the mechanism of integrin activation. These differences could explain, in part, the preferential accumulation of eosinophils in diseases such as asthma.

Objective quantitative analysis of eosinophils and bronchial epithelial cells in induced sputum by laser scanning cytometry.

BACKGROUND: Sputum induction is an important non-invasive technique for measuring airway inflammation in asthma. Cell numbers are often too low for flow cytometric analysis. Laser scanning cytometry (LSC) is a novel technique that allows objective multicolour fluorescence analysis of cells on a microscope slide. METHODS: LSC was used to determine sputum eosinophil and bronchial epithelial cell counts. We first confirmed that we could measure eosinophil counts accurately in peripheral blood using alpha-major basic protein (MBP) immunofluorescent staining. Sputum induction was performed according to standard protocols. Sputum samples from eight normal controls and 12 asthmatic patients were analysed by LSC and manual counting by two independent observers. Octospot cytospins were fixed and stained with mouse-alpha-human-MBP monoclonal antibody or mouse-alpha-human-cytokeratin antibody and goat-alpha-mouse Oregon Green conjugated second antibody. RESULTS: Sputum induction provided a mean (SE) of 0.99 (0.2) x 10(6) cells per donor. More than 3000 cells on three cytospins per slide were analysed per cell type. The intraclass correlation coefficient (R) and standard deviation (SD) of differences in eosinophils determined by manual counting and LSC were 0.9 and 2.1, respectively, and for bronchial epithelial cell counts they were 0.7 and 2.0. Selective detection of labelled cells was confirmed visually after relocation. CONCLUSION: Eosinophils and bronchial epithelial cells can be accurately and reproducibly counted in an objective manner. LSC is therefore a potentially powerful new method for immunophenotyping leucocytes and epithelial cells objectively in induced sputum in patients with asthma.

P- and L-selectin mediate binding of T cells to chronically inflamed human airway endothelium.

The inflammatory process that underlies allergic diseases such as asthma is characterized by tissue infiltration of eosinophils and T cells. We have used the Stamper-Woodruff frozen-section assay to characterize the receptors involved in adhesion of human peripheral blood T cells to nasal polyp endothelium (NPE) as a model of T cell migration in allergic disease. T cells bound specifically to NPE in a temperature-, cell concentration- and shear stress-dependent fashion. Adhesion was inhibited by approximately 70% by antibodies against P-selectin and its counter-receptor P-selectin glycoprotein-1 (PSGL-1). In addition, a blocking monoclonal antibody (mAb) against L-selectin caused significant although lesser inhibition. Cells adhering to NPE were primarily of the CD45RO+ memory subset. Although only a minority subset of peripheral blood T cells expressed functional PSGL-1, as determined by binding of a P-selectin Fc chimera, the majority of the P-selectin chimera-binding cells were found to be CD45RO+. This is consistent with the observation that memory T cells bind to NPE via P-selectin. Using blocking mAb we also investigated which integrins and their counter-structures were involved in T cell binding. A combination of anti-beta1 and beta2 mAb was able to inhibit adhesion by almost 50%. An antibody against intercellular adhesion molecule (ICAM)-2 gave an inhibition similar to that by anti-CD18 mAb, suggesting ICAM-2 was the major counter-receptor involved for the beta2 integrin component. This study suggests that P-selectin, and to a lesser extent L-selectin, may be acting as specific homing receptors for the airway mucosa in the context of chronic allergic disease.

Selectins and their counter receptors: a bitter sweet attraction.

Selectins are adhesion receptors expressed by leucocytes, platelets, and endothelial cells. They mediate the initial binding of leucocytes to vascular endothelium in the post-capillary venules. This is an essential first step in leucocyte migration into tissue. The selectin family of adhesion receptors consists of three C-type lectins (E, P, and L selectin). Their ligands (counter structures) are sialylated and fucosylated carbohydrate molecules which, in most cases, decorate mucin-like glycoprotein membrane receptors. Studies using blocking monoclonal antibodies have shown that inhibition of selectin function can ameliorate a range of inflammatory processes, offering the possibility that antagonists of selectin function may be useful in the treatment of inflammatory lung diseases such as asthma.

Adhesion interactions involved in eosinophil migration through vascular endothelium.

Considerable progress has been made in our understanding of the molecular mechanisms involved in eosinophil migration into sites of allergic inflammation. A number of differences between eosinophils and neutrophils have been observed in their pattern of adhesion interactions. These include expression of alpha 4 beta 1, alpha 4 beta 7, and alpha 6 beta 1 by eosinophils but not neutrophils and structural differences between eosinophil and neutrophil PSGL-1 associated with increased eosinophil binding to P-selectin. On the basis of current evidence the various receptors and mediators involved are summarized in FIGURE 1. Once in the tissues eosinophils may persist for several days or weeks, surviving under the influence of locally generated cytokines and this persistence may also partly explain selective tissue accumulation of eosinophils. Understanding the molecular mechanisms involved in leucocyte migration offers the possibility of selective and effective antagonists to treat allergic disease by preventing cell migration. Results in a number of animal models already offer the hope that this approach may be successful. The development of drugs that can be tested in the clinic are awaited with considerable interest.

Integrin alpha 4 beta 7 mediates human eosinophil interaction with MAdCAM-1, VCAM-1 and fibronectin.

We have investigated the contribution of integrin alpha 4 beta 7 to human peripheral blood eosinophil adhesive interactions. Immunofluorescence and flow cytometry demonstrated constitutive expression of alpha 4 beta 7 by eosinophils. Expression of alpha 4 beta 7 or alpha 4 beta 7 was not enhanced by eosinophil activation with platelet-activating factor (PAF). Expression of alpha 4 beta 7 was confirmed by immuno-precipitation of 125I-labeled lysates analysed by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS PAGE). Approximately 20% of unstimulated eosinophils were adherent to L1-2 cells transfected with vascular cell adhesion molecule-1 (VCAM-1) cDNA, while very few resting eosinophils adhered to mouse mucosal adressin cell adhesion molecule-1 (MAdCAM-1) transfectants. Binding of unstimulated eosinophils to VCAM-1 transfectant was inhibited by HPI 2 (an antibody that blocks both alpha 4 beta 1 and alpha 4 beta 7 functions), but not Act-1, and alpha 4 beta 1 monoclonal antibody (mAb). PAF stimulation resulted in increased binding of eosinophils to MAdCAM-1 transfectants, which was inhibited by both HPI 2 and Act-1. In contrast, PAF did not enhance binding to VCAM 1 transfectants, although binding of PAE-stimulated eosinophils to VCAM-1 could be partially inhibited by Act-1. Stimulation of eosinophils with the beta 7-activating mAb TS2 16 resulted in enhanced binding of eosinophils to both VCAM-1 and MAdCAM-1 transfectants. The increased binding was largely alpha 4 beta 7-dependent. Unstimulated eosinophils bound to soluble recombinant human (rh) VCAM-1 and fibronectin (Fn), coated on 96-well plates in dose-dependent manner. Binding was inhibited by HPI-2 and 4b4, an anti-beta 1 mAb, but not by Act-1. TS2 16 treatment increased adherent cell numbers and this enhanced binding was inhibited by Act-1. We have therefore confirmed that alpha 4 beta 7 is functionally active on unstimulated eosinophils. In contrast, PAF-induced enhancement of eosinophils binding to VCAM-1 or MAdCAM-1 was alpha 4 beta 7-dependent. In addition treatment with TS2 16 resulted in a alpha 4 beta 7-dependent enhancement of eosinophil binding to VCAM-1, MAdCAM-1 and Fn. We therefore hypothesize that alpha 4 beta 7 may have an important role in eosinophil localization in diseases such as asthma and inflammatory bowel disease.

Functional and structural characterization of the eosinophil P-selectin ligand.

Our recent studies have indicated an important role for P-selectin in eosinophil adhesion. We have therefore compared eosinophil and neutrophil binding with nasal polyp endothelium as well as purified P-selectin. We have also compared the structure and expression of the eosinophil and neutrophil P-selectin ligands. Using the frozen section assay, eosinophils bound to 2-fold more blood vessels within the nasal polyp tissue than neutrophils. Up to 10-fold more eosinophils than neutrophils bound per unit length of endothelium. Neutrophil and eosinophil binding was inhibited by a mAb against P-selectin and a P-selectin chimera which binds to the P-selectin ligand. Eosinophils bound with approximately 2-fold greater avidity to purified P-selectin under flow conditions. Using SDS-PAGE we characterized the eosinophil P-selectin ligand as a sialylated, homodimeric glycoprotein consistent with the known structure of PSGL-1. However, expression of PSGL-1 by eosinophils was significantly greater than on neutrophils. The eosinophil ligand had a calculated molecular mass by SDS-PAGE of approximately 10 kDa greater than the neutrophil ligand, which was not due to differences in N-glycosylation. Eosinophils expressed the 15-decapeptide repeat form of PSGL-1 compared with neutrophils that have the 16-decapeptide repeat form. The increased binding of eosinophils, compared with neutrophils, to P-selectin in both an ex-vivo and in vitro assay suggests that P-selectin may have a role directing the specific migration of eosinophils in diseases such as asthma. The increased avidity may be due to increased expression of PSGL-1 by eosinophils, differences in the peptide backbone, or post-translational modifications.

Ligation of CD69 induces apoptosis and cell death in human eosinophils cultured with granulocyte-macrophage colony-stimulating factor.

Peripheral blood (PB) eosinophils rapidly undergo apoptosis and cell death in vitro unless cultured in the presence of cytokines such as granulocyte-macrophage colony-stimulating factor (GM-CSF) in which their survival is prolonged for up to 10 days. CD69 is a type II membrane antigen expressed by cytokine-activated, but not freshly isolated, PB human eosinophils. We have examined the effect of ligation of CD69 by specific monoclonal antibody (MoAb) on the viability of human eosinophils cultured with recombinant human (rh)GM-CSF. Eosinophils were purified by immunomagnetic selection and cultured with GM-CSF (10(-10) mol/L). Eighteen hours after the start of culture, a panel of CD69 MoAb or controls (anti-CR3 or isotype-matched control MoAb) were added. Viability was assessed by trypan blue exclusion and apoptosis by morphologic assessment, DNA laddering, and flow cytometric analysis of eosinophil red autofluorescence. Up to 50% of the eosinophils had undergone apoptosis 48 hours after addition of anti-CD69 MoAb compared with less than 10% apoptosis for CR3 or the isotype matched control. The majority of apoptotic eosinophils excluded trypan blue at 48 hours post CD69 ligation. More apoptotic eosinophils were observed at later time-points and this was associated with loss of viability. At 120 hours post-addition of the anti-CD69 MoAb MLR3, 24% +/- 10.6% eosinophils were viable compared with 84% +/- 3.4% for the CR3 control (P < .001). A F(ab)2 fragment of CD69 MoAb P8, also induced apoptosis in GM-CSF cultured eosinophils. A more rapid induction of eosinophil apoptosis was obtained with CD69 MoAb immobilized via their Fc portions on protein-A coated plastic 96 well plates. Ligation of CD69 or CR3 resulted in the release of comparable quantities of eosinophil peroxidase at 48 hours post-ligation. These levels of EPO were consistent with the viability of these cells at 48 hours as assessed by exclusion of trypan blue. Finally, a neutralizing MoAb to TGF beta 1 had no effect on CD69-dependent apoptosis induction nor were there detectable quantities of TGF beta 1 in supernatants from GM-CSF--cultured eosinophils ligated with CD69 or control MoAb. These results suggest that eosinophils cultured with GM-CSF can be induced to undergo apoptosis as a result of cell signalling mediated by perturbation of CD69. This may represent an important physiologic mechanism for eosinophil removal in vivo.

Human eosinophils preferentially survive on tissue fibronectin compared with plasma fibronectin.

BACKGROUND: Eosinophil-derived inflammatory mediators including cytokines are considered to be important in the pathogenesis of allergic inflammation. Fibronectin (Fn) has been shown to be a physiological trigger of autocrine cytokine production by human eosinophils. Fn is encoded by a single gene, but alternate splicing of the primary RNA transcript results in polypeptide diversity in a cell type-specific fashion. Thus, tissue Fn contains approximately 50% more of the CS-1 cell binding region recognized by the integrin alpha 4 beta 1 compared with plasma Fn. OBJECTIVE: Since eosinophils are predominantly tissue-dwelling cells we compared the effect of tissue and plasma Fn on eosinophil survival in culture. METHODS: The viability and cytokine generation of eosinophils (> 99.9% pure) cultured for up to 4 days in 96 well plates coated with tissue Fn, plasma Fn or BSA was compared. RESULTS: There was a significant difference in the ability of tissue Fn to support eosinophil survival compared with plasma Fn (P < 0.01). Optimal survival with tissue Fn was seen at 25 micrograms/well (70% +/- 2.0% viability at 3 days vs 7% +/- 2.2% viability on BSA). Significant (P < 0.001) cell viability on tissue Fn was observed for up to 4 days in culture (54% +/- 6.0%) compared with BSA coated wells. Addition of autologous mononuclear cells (final concentration 0.5%, 1% or 2%) resulted in plasma Fn-dependent eosinophil survival comparable to that of 99.9% pure eosinophils adherent to tissue Fn. Tissue Fn-dependent survival was significantly inhibited by anti-interleukin-3, anti-granulocyte macrophage colony stimulating factor (GM-CSF) and anti-IL-5 monoclonal antibodies. Picogram quantities of these three cytokines were detected in supernatants from eosinophils cultured for 3 days on tissue Fn using specific enzyme-linked immunosorbent assays (ELISAs). Eosinophil survival on tissue Fn was significantly inhibited by anti-beta 1 and alpha 4 beta 1 monoclonal antibody (MoAb) and also by a MoAb specific for the CS-1 motif in the IIICS region of Fn. CONCLUSION: These observations show preferential survival of eosinophils cultured on tissue Fn as a result of alpha 4 beta 1-dependent interaction with the CS-1 region of tissue Fn triggering autocrine cytokine synthesis and release, thereby promoting their survival and persistence within the tissues.

Mechanisms of eosinophil and basophil migration.

Considerable progress has been made in our understanding of the molecular mechanisms involved in eosinophil and basophil migration into sites of allergic inflammation. It is clearly a staged process, each stage offering a level of control over the cell specificity and degree of migration. On the basis of current evidence, the various receptors and mediators involved are summarized in Table 4. Once in the tissues, eosinophils may persist for several days or weeks, surviving under the influence of locally generated cytokines, and this persistence may also partly explain the selective tissue accumulation of eosinophils and basophils. Understanding the molecular mechanisms involved in leukocyte migration may lead to the discovery of selective and effective antagonists to treat allergic disease by preventing cell migration. Results in a number of animal models already suggest that this approach may be successful. The development of drugs that can be tested in the clinic is awaited with much interest.

Eosinophil adhesion to nasal polyp endothelium is P-selectin-dependent.

Tissue eosinophilia is a characteristic feature of a number of inflammatory diseases including asthma and nasal polyposis. Eosinophil migration into tissues is controlled in part by interactions between eosinophil adhesion receptors and counter-structures on the vascular endothelium. To determine the receptors used by eosinophils to adhere to vascular endothelium in allergic inflammation we have adapted the Stamper-Woodruff frozen section assay (FSA) to study eosinophil adhesion to nasal polyp endothelium. Immunohistology indicated that intercellular adhesion molecule 1 (ICAM-1), E-selectin and P-selectin were well expressed by nasal polyp endothelium, whereas expression of vascular cell adhesion molecule 1 (VCAM-1) was weak or absent. Unstimulated human peripheral blood eosinophils adhered specifically to nasal polyp endothelium. Adherence was temperature and divalent cation-dependent and saturable at cell densities > 5 x 10(6) cells/ml. Eosinophil adhesion was almost completely inhibited by a monoclonal antibody (mAb) against P-selectin and by a chimeric molecule consisting of the Fc portion of human IgG and the lectin binding domain of P-selectin, which binds to the P-selectin ligand on leucocytes. Anti-Mac-1 mAb partially inhibited eosinophil adhesion whereas mAb against E-selectin, L-selectin, ICAM-1, VCAM-1, very late activation antigen 4, and lymphocyte function-associated antigen 1 had no effect. P-selectin is stored in intracellular granules within the endothelial cell and in vitro is only transiently expressed. To determine if P-selectin was expressed on the membrane of the nasal polyp endothelium we compared P-selectin expression in normal skin and nasal polyps after acetone fixation, which permeabilizes cells, and paraformaldehyde, which only allows staining of membrane expressed receptors. In the skin, good expression was seen with acetone fixation but no expression was seen after paraformaldehyde treatment, whereas in nasal polyps, similar expression was observed with both fixatives. In addition immunofluorescence with confocal microscopy demonstrated lumenal staining of nasal polyp endothelium indicating that P-selectin was located on the surface of endothelial cells while in skin only an intracellular granular distribution was apparent. Lastly, whereas eosinophils bound consistently to nasal polyp endothelium, no binding was observed to blood vessels in normal skin further supporting the idea that eosinophils were binding to membrane expressed and not intracellular P-selectin. The importance of P-selectin in eosinophil adhesion to nasal polyp endothelium suggests that P-selectin antagonists may be effective at inhibiting eosinophil accumulation at sites of allergic inflammation.