We analyzed a novel bifunctional fusion protein, CD40ed-CD95Led, consisting amino-terminally of the extracellular domain of CD40 and carboxy-terminally of the extracellular domain of CD95L. On cells lacking CD40L, this fusion protein is poorly active with respect to CD95 activation [median effective dose (ED50)>1 microg/ml], but it stimulates CD95 signaling with high efficiency upon binding to membrane-expressed CD40L (ED50<1 ng/ml). Thus, cell surface immobilization mediated by the CD40 part of the molecule unmasks the high-latent, CD95-stimulating capacity of the otherwise poorly active CD95L fusion protein. Moreover, interaction of the CD40 part of CD40ed-CD95Led with CD40L prevents the activation of cellular CD40. The CD40ed-CD95Led fusion protein therefore simultaneously blocks antiapoptotic CD40 activation and induces CD95-mediated apoptosis. Indeed, T47D cells displaying an antiapoptotic autocrine CD40-CD40L signaling loop were significantly more sensitive toward CD40ed-CD95Led than toward soluble CD95L artificially activated by crosslinking. Fusion proteins of RANK and CD95L (RANKed-CD95Led) and CD40 and tumor necrosis factor-related apoptosis inducing ligand (TRAIL) (CD40ed-TRAILed), with domain architectures similar to CD40ed-Cd95Led, displayed RANKL-dependent CD95 and CD40L-dependent TRAILR2 activation, respectively, indicating the principle feasibility of this fusion protein design.