Controversies in Experimental Dermatology

The emigration of leukocytes from peripheral vasculature into the dermis is dependent on, and regulated by, a sequence of adhesion molecule-mediated events. Such events involve specific families of adhesion molecules which are expressed on endothelial cells and leukocytes as cognate receptor/ligand pairs. Initial contact between a leukocyte rolling over endothelial cells is for the most part up-regulated by constitutive L-selectin expression or expression of α4β1 integrin (1). If the endothelium has been stimulated by cytokines such as tumour necrosis factor (TNF)-α, interleukin (IL)-1, IL-4, or lipopolysaccharide then leukocytes are rapidly activated and bind initially with L-selectin. Transient expression of P-selectin, released from storage in Weibel-Palade bodies (2), occurs rapidly – within 30 min – after stimulation by either histamine, IL-4, or IL-13 and induces early recruitment of leukocytes. The ligand for P-selectin is P-selectin glycoprotein ligand-1 (PSGL-1) expressed on myeloid, lymphoid, and dendritic cells (3) – P-selectin/PSGL-1 binding is crucial for early neutrophil rolling. Only endothelial cells synthesize and express E-selectin; it is not constitutively expressed. Up-regulation of E-selectin is stimulated by TNF-α, IL-1, lipopolysaccharide, or substance-P (4,5); peak expression occurs after 4 h. In all inflammatory dermatoses dermal endothelial cells express E-selectin (6), indeed E-selectin is most probably expressed preferentially in skin relative to other organs (7). The main T-cell ligand for E-selectin is cutaneous lymphocyte associated antigen (CLA). It appears that those T cells which migrate preferentially to skin, as opposed to other organs, express CLA; this observation is corroborated by the finding that T cells in inflammatory skin disease express CLA, whereas T cells in non-cutaneous inflammation are mainly CLA negative (8). Cell-adhesion molecules are ligands for integrins; one of their functions is to bind leukocytes to endothelial cells thereby facilitating leukocyte extravasation. Intercellular adhesion molecule (ICAM)-1 belongs to the immunoglobulin superfamily and is constitutively but weakly expressed on normal dermal endothelium, i.e. in non-diseased skin (9). The expression of ICAM-1 is significantly increased following stimulation by IL-1, TNF-α, or complement fraction C5a (10). The main leukocyte ligand for ICAM-1 is leukocyte function-associated antigen (LFA)-1 (CD11a/CD18) or Mac-1 (CD11b/CD18) (11). Vascular cell adhesion molecule (VCAM)-1 also belongs to the immunoglobulin superfamily and is expressed on activated endothelial cells, some macrophages, dendritic cells, and fibroblasts. Its main ligands are integrins, either α4/β4 [very late antigen-4 (VLA-4)] or α4/β7 on lymphocytes, monocytes, and eosinophils. VCAM-1 is important in modulating leukocyte adhesion to endothelial cells and thereby migration into the skin. These discrete pathways, both E-selectin and immunoglobulin mediated, present attractive targets for therapy of inflammatory skin disease. Targeting can be either selective, i.e. aimed at disrupting specific ligand-receptor pairs, e.g. E-selectin/CLA; or non-selective aimed at inhibiting those cytokines which are responsible for up-regulating or inducing adhesion molecules, e.g. TNF-α. This article is written from the viewpoint of targeting specific adhesion molecule ligand-receptor pairs as opposed to those upstream cytokines which induce adhesion molecule expression. The main targets for adhesion molecule-directed therapy are thus: E-selectin/CLA; VCAM-1/VLA-4; and ICAM-1/LFA-1. Despite knowledge of the existence of these three potential targets for at least a decade surprisingly little research has been directed towards this approach to therapy of skin disease. Naturally these ligand/receptor targets are not unique to inflammatory skin disease and have been investigated in other inflammatory disorders, notably inflammatory bowel disease and stroke (12–14). Overall, the results are disappointing. The most attractive target is E-selectin. We performed a study to assess the safety and efficacy of a humanized antibody to E-selectin (CDP850, Celltech, Slough, UK) in the treatment of chronic plaque psoriasis (15). Subjects were randomly assigned to a single intravenous dose of either CDP850 or placebo. At the end of the study (8 weeks post infusion), there was no significant difference in clinical severity of psoriasis between the two treatment groups. In all patients who received CDP850, there were significant detectable blood levels of CDP850. Furthermore, staining of lesional skin with an antibody, ENA-2, which binds the same epitope on E-selectin as CDP850, was reduced – implying binding of CDlP850 to E-selectin on dermal endothelium. There was a concomitant rise in the number of epidermal Langerhans cells in CDP850 but not placebo-treated subjects, indicative of inhibition of migration of Langerhans cells from epidermis to afferent lymphatics. The conclusions from this study were that anti-E selectin although well tolerated and safe is not an effective therapy for clearing chronic plaque psoriasis. ICAM-1/LFA-1 binding is key not only for leukocyte endothelial cell binding but also for binding leukocytes to keratinocytes. Indirect evidence of the importance of the latter interaction to maintenance of psoriasis plaques comes from the observation that in patients with psoriasis treated with cyclosporin, frequency of intraepidermal leukocytes within plaques declines following reduction in keratinocyte expression of ICAM-1 which in turn is secondary to cyclosporin-induced inhibition of T-cell production of interferon-γ– a key stimulant of ICAM-1 expression (16). Efalizumab – a humanized monoclonal antibody to the CD11a component of LFA-1 has recently been approved in the USA for treatment of severe psoriasis. This follows extensive trialling which showed that 12-week therapy with once-weekly subcutaneous injections of efalizumab produced significant reduction in clinical severity of psoriasis in approximately 25% of subjects (17, 18). This clinical improvement was accompanied by an increase in the number of circulating lymphocytes – indicative of inhibition of leukocyte extravasation probably by blockade of ICAM-1/LFA-1 binding. This conclusion should not, however, be taken as proof of concept, i.e. blocking leukocyte extravasation produces improvement in inflammatory skin disease. This is because ICAM-1/LFA-1 are costimulatory molecules important for binding T cells to antigen-presenting cells thereby amplifying T-cell activation. Efalizumab blocks this process thereby inhibiting T-cell activation. What does this mean? My view is that blockade of endothelial/leukocyte adhesion is not best placed as a strategy to clear inflammatory skin diseases – the non-efficacy of anti-E-selectin and modest efficacy of efalizumab are testimony to this. However, I believe that we should change our paradigm for assessment of efficacy these biologicals to one of long-term prevention. Those biologicals which block leukocyte extravasation may be best placed when used long-term thereby maintaining remission or clearance of disease once achieved by other fast-acting, remission-inducing therapies such as cyclosporin or infliximab.