Optimized Neural Stem Cell Delivery of the TRAIL Protein for Targeted Glioblastoma Therapy

Abstract

Effective delivery of therapeutics to the brain remains elusive, contributing to poor outcomes in diseases like glioblastoma, where median survival is under 15 months. Here, we explore a novel therapeutic strategy using human neural stem cells derived by direct reprogramming of peripheral blood erythroblasts (PBiNSCs). Pilot studies show that PBiNSCs have an intrinsic ability to home to glioblastoma cells in vivo and in vitro, supporting their use as delivery vehicles for targeted treatment of aggressive brain tumors. We genetically engineered PBiNSCs to express the membrane-bound variant of TNF-related apoptosis-inducing ligand (TRAIL), a potent pro-apoptotic protein with stronger cytotoxicity than its soluble form. Lentiviral transduction resulted in successful exogenous expression of TRAIL but showed some toxicity to the PBiNSCs themselves. To overcome this, we used CRISPR-Cas9 to knock out the endogenous DR5 receptor, generating a TRAIL-resistant PBiNSC population (DR5KO-PBiNSC). Following ectopic expression of TRAIL in this new population (TRAIL-DR5KO-PBiNSCs), we detect TRAIL predominantly in the media in exosomes. TRAIL-DR5KO-PBiNSCs exhibited improved viability and enhanced cytotoxicity against a primary glioblastoma cell line when tested either in co-culture or using conditioned media from these modified cells. We further tested several sensitizing agents to enhance TRAIL-mediated apoptosis. AZD5582, a small-molecule Smac mimetic and inhibitor of IAPs, demonstrated sensitizing effects at low concentrations and killing of >90% in one primary glioblastoma cell line. Our findings establish a genetic engineering strategy for TRAIL delivery in neural stem cells for targeted glioblastoma therapy.