Methods for Study of Chemokine Receptors in the Tissues

Three different assays were used to study the distribution of binding sites for IL-8 in human skin and several animal tissues. An in situ binding assay was designed in which the binding of radiolabeled IL-8 to small intact tissue pieces was studied, and a histological autoradiographic technique was used to detect the bound chemokine in the subsequently prepared tissue sections. A modified assay was also performed in which the binding of unlabeled IL-8 to intact tissue pieces was visualized using monoclonal anti-IL-8 antibody. In addition, we performed a "classical" autoradiographic study in which radiolabeled IL-8 was injected subcutaneously and visualized in sections prepared from the injected sites by autoradiography. We reflect on the potentials and limitations of studying the chemokine binding in situ, compare the results, and discuss the relative advantages and disadvantages of each of the techniques used.

Some aspects of IL-8 pathophysiology. III: Chemokine interaction with endothelial cells.

Chemokines have been convincingly implicated in driving leukocyte emigration in different inflammatory reactions. However, the cellular and molecular mechanisms of chemokine involvement in leukocyte emigration are not clear. We and others suggested that leukocyte adhesion to the endothelium and transmigration are induced by chemokines immobilized on the endothelial cell surface. This would require the presence of specific chemokine binding sites in this microanatomical location. Using an in situ binding assay we demonstrated the presence of binding sites for interleukin-8 (IL-8) and RANTES, but not monocyte inflammatory protein-1 alpha on the endothelium of postcapillary venules and small veins in human skin. In contrast, venules and veins in various anatomical locations showed dramatically differing IL-8 binding patterns. The subcellular distribution of IL-8 in the venular endothelial cells following its in vivo and ex vivo injections was studied by use of electron microscopy. Our results suggest that IL-8 was internalized by the endothelial cells, transported transcellularly via plasmalemmal vesicles, and released onto the luminal surface where it appeared located preferentially on tips of membrane protrusions. We were unable to study the endothelial IL-8 binding or transport in vitro because all the in vitro propagated endothelial cell lines and primary endothelial cells tested lacked IL-8 binding sites. This includes human umbilical vein endothelial cells (HUVECs), which also did not bind IL-8 in situ. However, HUVECs provided a satisfactory in vitro system to study the secretion of IL-8 by the endothelial cells. Two possible alternative pathways were described: secretion directly from the Golgi apparatus or following storage in Weibel-Palade bodies.

Binding of RANTES, MCP-1, MCP-3, and MIP-1alpha to cells in human skin.

Based on their ability to induce leukocyte chemotaxis and adhesion to endothelial cells (ECs), chemokines have been implicated in driving inflammatory leukocyte emigration. Recently, it was suggested that chemokines can accomplish their pro-emigratory role more effectively while being bound to the luminal surface of the ECs. Previously, such binding was demonstrated in situ in human skin for the prototype alpha-chemokine interleukin (IL)-8. Here we used an in situ binding assay to investigate the binding characteristics of several beta-chemokines in intact human skin. RANTES, MCP-1, and MCP-3 bound, similar to IL-8, in a specific saturable manner to the ECs of venules and small veins but not arteries or capillaries. RANTES inhibited MCP-1 and MCP-3 binding and vice versa, indicating that the EC binding sites are shared among these beta-chemokines; moreover, IL-8 and RANTES cross-competed for each other's binding, suggesting that the same chemokine binding sites are used by members of alpha- and beta-chemokine subfamilies. Conversely, MIP-1alpha did not bind to the ECs and did not compete for binding of RANTES. Analogous to IL-8, all of the tested beta-chemokines bound to the resident dermal cells. As a novel aspect of chemokine interaction with cells in normal skin, we observed specific, saturable binding of RANTES, MCP-1, and MCP-3 but not MIP-1alpha or IL-8 to the ECs of dermal afferent lymphatic vessels. RANTES, MCP-1, and MCP-3 also cross-competed for each other's binding to lymphatics, suggesting a common binding site with a novel chemokine binding profile. We suggest that the chemokine binding to the ECs of lymphatics may be involved in the process of leukocyte entry into the afferent lymphatic vessels.

Transcytosis and surface presentation of IL-8 by venular endothelial cells.

Chemokines have been convincingly implicated in actuating inflammatory leukocyte emigration. To affect the circulating leukocytes, tissue-derived chemokines have to traverse the endothelial cells (ECs). This was thought to be accomplished by chemokine diffusion through the intercellular gaps. On the contrary, we show by electron microscopy that the prototype chemokine IL-8 is internalized by venular ECs abluminally and transcytosed to the luminal surface. Here, it is presented to the adherent leukocytes on the EC membrane, predominantly in association with the EC projections. The intact C terminus of IL-8, the molecule's "immobilization" domain, is required for the EC binding, transcytosis, and consequently, the in vivo proemigratory activity of IL-8, indicating that the described subcellular interactions of IL-8 with the ECs are functionally relevant.