Human endothelial cells cultured on microporous filters used for leukocyte transmigration studies form monolayers on both sides of the filter.

A growing number of studies on the mechanism of leukocyte transendothelial migration use endothelial cells grown on microporous filters as an in vitro model of endothelium. Ultrastructural examination of such a model system previously demonstrated that human pulmonary artery endothelial cells (HPAEC) formed confluent monolayers on both sides of the 3-microm-pore filter (Mackarel et al., 1999). To determine whether this was a characteristic specific to pulmonary artery endothelial cells, the growth characteristics of a human pulmonary microvascular endothelial cell type (HMVEC-L) and the widely used human umbilical vein endothelial cells (HUVEC) on 3-microm microporous filters were examined by transmission electron microscopy (TEM). Similar to HPAEC, HMVEC-L and HUVEC were also found to grow on both sides of the filter. All three endothelial cell types were capable of migrating through the 3 microm pores of the filter to form a monolayer on the filter underside. The endothelial cells on the underside were orientated in an inverted position with the luminal surface facing away from the filter. Such 'bilayer' formation was observed at a range of seeding densities and in different culture media. Despite the presence of a bilayer of endothelial cells, TEM demonstrated that neutrophils migrated successfully across the cell-filter-cell system. Previous transmigration reports in which an in vitro model similar to ours was used have often assumed only one layer of endothelial cells. The observations reported here indicate that while endothelial cells on microporous filters are useful models for examining leukocyte-endothelial interactions, they are not appropriate for studies examining endothelial cell 'sidedness.'

Migration of neutrophils across human pulmonary endothelial cells is not blocked by matrix metalloproteinase or serine protease inhibitors.

It has long been speculated that neutrophils deploy proteases to digest subendothelial matrix as they migrate from the bloodstream. Direct evidence for the involvement of proteases in neutrophil transendothelial migration is, however, lacking. To address this issue we used transmission electron microscopy to verify the presence of continuous basal lamina beneath pulmonary endothelial cells grown on microporous filters, and then examined the effects of protease inhibitors on neutrophil migration through the endothelial cells and their associated subcellular matrix. Inhibitors of the two major matrix-degrading protease groups present in neutrophils, the matrix metalloproteinases (MMPs) and serine proteases, were assessed for their ability to modulate neutrophil transendothelial migration in response to the chemoattractant n-formylmethionyl leucylphenylalanine (FMLP). Neither the naturally occurring MMP inhibitor, tissue inhibitor of metalloproteinase-1, nor the hydroxamic acid-based inhibitors GM-6001, BB-3103, or Ro 31-9790 had any significant effect on FMLP-stimulated neutrophil migration across endothelial cells and associated basal lamina, with >/= 80% of neutrophils migrating through the system, even in the presence of inhibitors, at concentrations that totally inhibited all the gelatinase B (MMP-9) released upon stimulation with FMLP. Similarly, with serine protease inhibitors no significant inhibition of neutrophil migration was observed with a naturally occurring inhibitor, secretory leukocyte protease inhibitor, or a low molecular-weight synthetic inhibitor, Pefabloc SC. These results indicate that neither MMP nor serine protease digestion of sub-endothelial matrix is required for successful neutrophil transendothelial migration.

Interleukin-8 and leukotriene-B(4), but not formylmethionyl leucylphenylalanine, stimulate CD18-independent migration of neutrophils across human pulmonary endothelial cells in vitro.

Although neutrophil migration from the systemic circulation involves the beta2- (or CD18) integrin family, the existence of an alternative, CD18-independent route of neutrophil extravasation to tissues has been demonstrated in animal models. The molecular interactions involved in this alternative migratory route have not yet been characterized. The objective of this study was to assess the CD18-dependency of neutrophil migration across human endothelial cells from an organ known to support CD18-independent migration, the lung, with a view to establishing an in vitro model to facilitate study of CD18-independent migration. Neutrophil migration across human pulmonary artery endothelial cells (HPAECs) in response to three different chemoattractants, formylmethionyl leucylphenyl-alanine (FMLP), interleukin (IL)-8, and leukotriene (LT) B(4), was examined. Results demonstrated that a function-blocking antibody to CD18 decreased FMLP-stimulated migration by 71.7 +/- 4.4% (P < 0.001). In contrast, migration in response to LTB(4) was decreased by only 20.5 +/- 10.2% (P < 0.01), and no significant decrease was observed with migration to IL-8. Neutrophils that migrated to FMLP had 1.7-fold more surface CD11b/CD18 compared with nonmigrated neutrophils (P < 0.01), whereas this integrin complex was not significantly upregulated on neutrophils that had migrated to IL-8 or LTB(4). Further investigation of this migratory route indicated that it did not involve the beta1 integrins (CD29) or the endothelial selectins, E- or P-selectin, nor did it require the activity of either metalloproteinases or neutrophil elastase. These results indicate that neutrophil migration across HPAECs in vitro to IL-8 and LTB(4) is predominantly CD18-independent and provides a much-needed in vitro system for examination of the neutrophil-endothelial interactions involved in this alternative migratory route.

Investigation of vascular endothelial growth factor effects on pulmonary endothelial monolayer permeability and neutrophil transmigration.

This study sought to determine whether vascular endothelial growth factor (VEGF)-induced permeabilisation of pulmonary endothelium to macromolecules could be related to a permissive role for neutrophil-derived VEGF in neutrophil transmigration. Treatment of human pulmonary artery endothelial cell (HPAEC) monolayers with 1, 10 or 100 ng/ml VEGF for 15 min or 1, 10 ng/ml for 90 min significantly increased endothelial permeability to trypan blue-labelled albumin (TB-BSA). These increases were correlated with changes in the cellular distribution of F-actin, as visualised by rhodamine-phalloidin staining: increased stress fibre formation, cellular elongation and formation of intercellular gaps after 15 min; at 90 min, there was also evidence of microspike formation and extension of spindle processes from the cell surface. Treatment of human neutrophil suspensions with 200 nM phorbol myristyl acetate (PMA), n-formyl-methionyl leucylphenylalanine (fMLP, 10 nM), interleukin-8 (IL-8, 10 nM) (but not with leukotriene B(4) (LTB(4)) 100 nM), for 30 min caused significant extracellular release of neutrophil VEGF stores. A permissive role for neutrophil-derived VEGF in facilitating migration across HPAEC monolayers was assessed in experiments using a functional blocking antihuman VEGF antibody. In the presence of this antibody (10 microg/ml), neutrophil migration in response to fMLP (10 nM), IL-8 (10 nM) or LTB(4) (100 nM) was not significantly different to that in the absence of antibody. We conclude that neutrophil-derived VEGF does not play a functional role in facilitating neutrophil migration across pulmonary vascular endothelium, despite its ability to induce cytoskeletal changes and enhance endothelial macromolecular permeability.

Matrix metalloproteinases and tissue inhibitor of metalloproteinase-1 in sarcoidosis and IPF.

The purpose of this study was to examine the role of interstitial collagenases, members of the family of matrix metalloproteinases, in the development of pulmonary fibrosis. The activity, levels and molecular forms of collagenases (matrix metalloproteinases (MMP)-1, -8 and -13), gelatinase B (MM P-9) and its main endogenous inhibitor, tissue inhibitor of metalloproteinase-1 (TIMP-1) were assessed in bronchoalveolar lavage fluid (BALF) from patients with idiopathic pulmonary fibrosis (IPF) and sarcoidosis patients with varying degrees of pulmonary parenchymal involvement. Collagenase activity was elevated in IPF and group 3 sarcoidosis patients. A positive correlation between BALF collagenase activity and MMP-8 levels was also observed. Western immunoblotting revealed the presence of two isoforms of MMP-8 in patient samples; an 80 kD form representing latent enzyme from polymorphonuclear neutrophils and a 55 kD form representing the fibroblast-type proform. MMP-9 levels were also elevated in both IPF and group 3 sarcoidosis patients, while TIMP-1 levels remained normal, indicating a shift in the balance between the enzyme and inhibitor, favouring MMP-9. Matrix metalloproteinase-8 is the major contributor to the bronchoalveolar lavage fluid collagenase activity in the airways of patients with idiopathic pulmonary fibrosis and sarcoidosis and may initiate collagen destruction and remodelling leading to the development of pulmonary fibrosis.

CD18 dependency of transendothelial neutrophil migration differs during acute pulmonary inflammation.

Neutrophil extravasation during inflammation can occur either by a mechanism that requires the neutrophil integrin complex, CD18, or by an alternative CD18-independent route. Which of the two pathways is used has been shown to depend on the site and nature of the inflammatory insult. More recent evidence suggests that selection may also depend on whether inflammation is chronic or acute, but why this is the case remains unknown. Using an in vitro model that supports both migratory mechanisms, we examined the CD18 dependency of migration of neutrophils isolated from patients with either chronic or acute pulmonary infection. Chronic neutrophils were found to behave like normal neutrophils by migrating to IL-8 and leukotriene B(4) using the CD18-independent pathway, but to the bacterial product, FMLP, using the CD18-dependent route. In contrast, migration of acute neutrophils to all of these stimuli was CD18 dependent. Normal neutrophils could be manipulated to resemble acute neutrophils by exposing them to FMLP before migration, which resulted in a "switch" from the CD18-independent to -dependent mechanism during migration to IL-8 or leukotriene B(4). Although treatment of normal neutrophils with FMLP caused selective down-regulation of the IL-8 receptor, CXCR2, and acute neutrophils were found to have less CXCR2 than normal, a functional relationship between decreased CXCR2 and selection of CD18-dependent migration was not demonstrated. Results indicate that selection of the CD18-dependent or -independent migration mechanism can be controlled by the neutrophil and suggest that the altered CD18 requirements of acute neutrophils may be due to priming in the circulation during acute infection.