The puzzling role of TRAIL in endothelial cell biology

In their study published online in Arteriosclerosis, Thrombosis, and Vascular Biology, O’Brien et al demonstrate that activated protein C (APC) decreases tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) expression in endothelial cells.1 Although the molecular mechanism by which APC decreases the expression of TRAIL mRNA and protein via an Egr-1/Erk-1/2 pathway appears convincing, much less clear is whether the downregulation of TRAIL reflects a physiological relevant decrease of surface TRAIL or TRAIL release by endothelial cells. In fact, we have never detected surface TRAIL by flow cytometry in human umbilical vein endothelial cells (HUVECs), cultured with or without TNF-alpha. On the other hand, surface TRAIL was easily detectable in other cell model systems, such as peripheral blood-derived adherent macrophages. Moreover, we were also unable to document in ELISA detectable release of soluble TRAIL by HUVECs treated or not with TNF-alpha. Because in their study O Brien et al used an ELISA approach to detect the TRAIL protein in endothelial cell lysates, the reported modulation of TRAIL by TNF-alpha and APC likely involves an intracellular pool of TRAIL protein. Although this represents an interesting observation, the mechanisms involved in membrane expression/release of TRAIL in endothelial cells have not been defined yet, making unclear the physiopathological implications of this study. However, what raises some concern in the study of O’Brien et al is their interpretation on the putative role of TRAIL in endothelial cell biology in the context of the available scientific literature. In fact, although O’Brien and colleagues postulate a detrimental effect of TRAIL in vascular biology, in our opinion this hypothesis is not adequately supported. In spite of the observation that human endothelial cells obtained from different vascular districts express detectable levels of all transmembrane TRAIL receptors (TRAIL-R1–R4),2–4 most of the published studies have reported lack of toxicity in vitro when endothelial cells were exposed to recombinant TRAIL. Notwithstanding, different authors, including us, have shown that particular culture conditions, such as serum and extracellular matrix deprivation coupled to pharmacological inhibition of the PI-3 kinase/Akt prosurvival pathway were required to efficiently sensitize endothelial cells to TRAIL-mediated cytotoxicity. How these culture conditions reflect a physiopathological condition remains to be determined. Contrary to the hypothesis of a detrimental effect of TRAIL in vascular biology, in a series of in vitro studies we have documented that soluble recombinant TRAIL exhibits protective and antiinflammatory effects on endothelial cells. Moreover, we have recently reported that soluble TRAIL shows antiatherogenic effects in an in vivo model of Apo-E knockout mouse. Thus, the overall scenario emerging from these published studies is that the role of TRAIL in vascular biology deserves further investigation before any conclusion on the usefulness of drug decreasing its expression/release by endothelial cells can be drawn.