Customizing a tri-domain TRAIL variant to achieve active tumor homing and endogenous albumin-controlled release of the molecular machine in vivo.

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is an attractive antitumor drug candidate for precision cancer therapy due to its superior selective cytotoxicity in a variety of tumor cells. However, the clinical application of TRAIL in cancer therapy has been limited by its poor tumorhoming capacities and short half-life. Herein, we designed a tridomain TRAIL variant, Z-ABD-TRAIL, by sequentially fusing the platelet-derived growth factor receptor beta (PDGFR beta)-specific affibody Z(pDGFR beta) and an albumin-binding domain (ABD) to the N-terminus of TRAIL. The fusion protein Z-ABD-TRAIL was produced as a soluble protein with high yield in Escherichia coli (E. coli). The Z(PDGFR beta) domain provided Z-ABD-TRAIL with P-DGFR beta binding properties and thus promoted its tumor homing via the engagement of P-DGFR beta-expressing pericytes on tumor microvessels. ABD-mediated binding of Z-ABD-TRAIL to albumin in the blood endowed TRAIL with long-lasting (>72 h for Z-ABD-TRAIL vs <0.5 h for TRAIL) abilities to kill tumor cells. Although the in vitro cytotoxicity of Z-ABD-TRAIL in tumor cells was similar to that of the parent TRAIL, the in vivo tumor uptake, apoptosis-inducing ability, and antitumor effect of Z-ABD-TRAIL were much greater than those of TRAIL, indicating that Z(pDGFR beta)-mediated tumor homing and ABD-introduced albumin binding significantly improved the pharmacodynamics of TRAIL. In addition, repeated injection of high-dose Z-ABD-TRAIL showed no obvious acute toxicity in mice. These results demonstrate that the newly designed tridomain Z-ABD-TRAIL is a promising agent for precision cancer therapy.