Developing Targeting Techniques for the Advancement of In Vivo Gene Therapies

Abstract

Hematopoietic Stem Cell (HSC) gene therapy is a promising route to curing patients with a variety of hematologic diseases and disorders. HSC gene therapy is currently performed ex vivo and requires rare cleanroom infrastructure, which is a significant obstacle to patients in resource impoverished areas. Application of gene therapy agents directly in the patient (in vivo) would overcome this bottleneck. We have previously identified the CD34+CD90+ subset to be exclusively responsible for short- and long-term engraftment. However, purification and enrichment of this subset is laborious and expensive. HSC-specific delivery agents for the direct modification of rare HSCs are currently lacking. Here, we developed novel targeted viral vectors to specifically transduce CD90-expressing HSCs. Anti-CD90 single chain variable fragments (scFvs) were engineered onto measles- and VSVG-pseudotyped lentiviral vectors that were knocked out for native targeting. We further developed a custom hydrodynamic titration methodology to assess the loading of surface-engineered capsids, measure antigen recognition of the scFv, and predict the performance on cells. Engineered vectors formed with minimal impairment in the functional titer maintained their ability to fuse with the target cells and showed highly specific recognition of CD90 on cells ex vivo. Most importantly, targeted vectors selectively transduced human HSCs with secondary colony-forming potential. However, agents recognizing only a single marker on target cells are often insufficient due to epitope sharing in between on- and off-target tissues. The Baker Lab previously designed a system of protein switches known as Co-LOCKR, which enable the specific recognition of two antigens simultaneously. Co-LOCKR targeted CAR T cells highly specifically killed tumor cells in vitro with virtually no off-target effects. Therefore, we further designed and evaluated Co-LOCKR-targeted viral vectors for ex vivo as well as in vivo applicability and comprehensively evaluated the on-target specificity in a murine tumor model. Together this work lays a foundation and provides a robust toolset for the exploration of in vivo gene therapies.