No abstract

1 Expression of cell adhesion molecules (CAM) on the lung microvascular endothelium is believed to play a key role in the recruitment of leukocytes in pulmonary in¯ammation. Moreover, regulation of CAM expression may be an important mechanism through which this in¯ammation may be controlled. Experimental evidence has suggested that combined phosphodiesterase (PDE) 3 and 4 inhibitors increase cyclic AMP levels within cells greater than inhibition of either isoenzyme alone. In the present study we assessed the eect of combinations of rolipram (PDE4 inhibitor), ORG 9935 (PDE3 inhibitor) and salbutamol (b-agonist) on CAM expression and neutrophil or eosinophil adhesion to human lung microvascular endothelial cells (HLMVEC). 2 Tumour necrosis factor-a (TNF-a)-induced intercellular adhesion molecule (ICAM)-1, vascular cell adhesion molecule (VCAM)-1 and E-selectin expression were measured on HLMVEC monolayers at 6 h by a speci®c ELISA technique in the presence of dierent combinations of medium, rolipram, ORG 9935 and salbutamol. 3 Rolipram in combination with salbutamol, but neither agent alone, inhibited TNF-a-induced Eselectin expression, whilst ICAM-1 and VCAM-1 expression were not aected. ORG 9935 had no signi®cant eect on CAM expression alone. However, in combination with rolipram a syngergistic inhibition of VCAM-1 and E-selectin, but not ICAM-1, expression was observed. No further inhibition was seen in the additional presence of salbutamol. 4 Neutrophil adhesion to TNF-a-stimulated (6 h) HLMVEC was mainly E-selectin dependent in this model, as ENA 2 an anti-E-selectin monoclonal antibody (mAb) abrogated neutrophil adhesion. Eosinophil adhesion was E-selectin-, ICAM-1-and VCAM-1-dependent, as assessed by the inhibitory activity of ENA 2 and the ability of a mAb to the ICAM-1 ligand, CD18, and a mAb to the VCAM-1 ligand, VLA 4 , to attenuate adhesion. 5 Rolipram in the presence of salbutamol or ORG 9935 signi®cantly inhibited neutrophil adherence to TNF-a-stimulated HLMVEC. Eosinophil adherence to monolayers was inhibited only when HLMVEC were activated in the presence of rolipram and ORG 9935. 6 Collectively, the ®ndings presented in this manuscript suggest that inhibition of PDE4 with appropriate activation of adenylate cyclase is sucient to inhibit induction of E-selectin expression on HLMVEC to a level that has functional consequences for neutrophil adhesion. In contrast, combined inhibition of PDE3 and 4 isoenzymes is necessary to inhibit VCAM-1 and to have inhibitory eects on eosinophil adhesion to activated HLMVEC. Upregulation of ICAM-1 expression on HLMVEC does not appear to be modulated by PDE3 and 4 inhibition. These data may have implications for the use of selective PDE4 inhibitors in lung in¯ammation.

Tumour necrosis factor‐α (TNFα) increases the expression of the adhesion molecule intercellular adhesion molecule‐1 (ICAM‐1) on cultured endothelial and epithelial cells and modulation of this may be important in controlling inflammation. Activation of tyrosine kinase(s) is known to be involved in the signal transduction pathways of many cytokines. In this study we have investigated the effects of the tyrosine kinase inhibitors, ST638, tyrphostin AG 1288 and genistein, on TNFα‐induced ICAM‐1 expression in human alveolar epithelial (A549) and vascular endothelial (EAhy926) cell lines and also normal human lung microvascular endothelial cells (HLMVEC). ICAM‐1 expression on cultured cells was determined by a sensitive enzyme‐linked immunosorbant assay (ELISA). Endothelial or epithelial monolayers were exposed to increasing doses of TNF‐α (0.01–10 ng ml−1), in the presence or absence of either ST638 (3–100 μm), AG 1288 (3–100 μm) or genistein (100 μm) and ICAM‐1 expression was measured at 4 and 24 h. Control experiments examined the effect of ST638 on phorbol 12‐myristate 13‐acetate (PMA, 20 ng ml−1, 4 h)‐stimulated ICAM‐1 and compared it to that of a specific protein kinase C inhibitor, Ro31‐8220 (10 μm). Also, functional consequences of changes in ICAM‐1 expression were assessed by measuring adhesion of 111In‐labelled human neutrophils to EAhy926 endothelial and A549 epithelial monolayers treated with TNFα, in the presence or absence of ST638. ST638 caused a concentration‐dependent reduction in TNFα‐ (0.1–10 ng ml−1‐induced ICAM‐1 on EAhy926 endothelial (at 4 h) and A549 epithelial monolayers (at 4 and 24 h). In contrast, ST638 caused a concentration‐dependent increase in TNFα‐ (0.1–10 ng ml−1‐induced ICAM‐1 on EAhy926 endothelial cells at 24 h. Similar effects were seen with AG 1288 or genistein. ST638 (100 μm) significantly (P < 0.01) inhibited ICAM‐1 expression on HLMVEC endothelial cells induced by 0.01 ng ml−1 TNFα at 4 or 24 h or 0.1 ng ml−1 at 4 h, but increased ICAM‐1 expression induced by 0.1 ng ml−1 TNFα at 24 h. ST638 did not significantly change the expression of PMA‐stimulated ICAM‐1 on either A549 epithelial, EAhy926 or HLMVEC endothelial cells. However, PMA‐induced ICAM‐1 expression was inhibited by Ro31‐8220. Also, treatment of epithelial or endothelial monolayers with TNFα and ST638 altered adhesion of human neutrophils to A549 epithelial or EAhy926 endothelial cells in a manner that corresponded to the alteration in ICAM‐1 expression. These results show that tyrosine kinase inhibitors alter TNFα‐induced ICAM‐1 expression, but that the cell type, concentration of TNFα and time of exposure to this cytokine determine whether expression is decreased or increased by the inhibitor. An increased understanding of the signal transduction pathway(s) involved in TNFα‐induced ICAM‐1 expression on lung epithelial and vascular endothelial cells may be of potential therapeutic value in the treatment of inflammatory disease.

Eosinophil adhesion to airway epithelium is believed to facilitate eosinophil accumulation and retention in asthmatic airways. Monoclonal antibodies (mAb) against intercellular adhesion molecule-1 (ICAM-1) and its CD18 leukocyte integrin ligands have been shown to inhibit airway eosinophilia in animal models of asthma, although the role of this pathway in eosinophil-epithelial adhesion is not fully understood. To investigate the role in vitro of CD18 and ICAM-1, we measured adhesion of fluorescently labeled human eosinophils to normal human bronchial epithelial cell (NHBEC) monolayers pretreated for 24 h with culture medium (low constitutive ICAM-1) or tumor necrosis factor-alpha (TNF-alpha; 1 ng/ml) and interferon-gamma (IFN-gamma) (10 ng/ml; increased ICAM-1). Stimulation of eosinophils with C5a (10(-7) M) increased adhesion measured at 30 min to unactivated NHBEC from 11.4 +/- 0.7 to 15.5 +/- 0.4% (n = 4), and this increase was CD18/ICAM-1-independent, whereas phorbolmyristate acetate (PMA) (10(-8) M)-induced adhesion (20.7 +/- 1.7%) was abolished by anti-CD18 and reduced by anti-ICAM-1. In contrast, C5a- and PMA-induced adhesion to TNF-alpha/IFN-gamma-activated NHBEC (increased from 11.1 +/- 1.3% to 21.9 +/- 1.0% and 27.6 +/- 1.9%, respectively) was CD18- and ICAM-1-dependent. Eotaxin, but not regulated on activation normal T cells expressed and secreted, macrophage inflammatory protein-1, formyl methionyl leucyl phenylalanine, leukotriene B4 or platelet-activating factor, also induced CD18/ICAM-1-dependent adhesion to activated NHBEC. In the absence of added chemoattractants, eosinophil adhesion to NHBEC increased with time and, at 120 min, was significantly greater (P < 0.01) to activated NHBEC (37.3 +/- 2.4%, n = 5) than to unactivated monolayers (24.3 +/- 1.9%); mAb against CD18 or ICAM-1 abolished increased, but not basal, adhesion. These results suggest that CD18/ICAM-1 mediated eosinophil adhesion to activated NHBEC but that adhesion to resting NHBEC was largely independent of this pathway.

RAGE is an inflammation-perpetuating receptor with a diverse range of ligands. Evidence supporting a role of the RAGE axis in the pathogenesis of systemic inflammation, ALI and myocardial dysfunction is compelling with numerous animal experiments showing the beneficial effects of inhibiting the RAGE axis. Despite a number of unanswered questions that need to be further addressed, the potential for inhibiting RAGE-mediated inflammation in humans undoubtedly exists.