Protein engineering of targeted cancer therapies

Abstract

Protective antigen (PA). the pore-forming component of anthrax toxin, has emerged as a platform for the development of cancer therapies because of its versatility and robust ability to translocate proteins into a cell's cytosol. More recently, development of new techniques for modifying PA is enabling it to be retargeted to receptors of interest via fusion with existing protein binders. There is a vast library of potential binders for PA based on natural or novel protein scaffolds generated by the field of protein engineering. This has allowed new approaches for tumor cells to be targeted for cytosolic delivery of toxins as a therapeutic strategy. In our work, we sought to leverage the anti-tumor properties of an antibody, Elv3, to retarget PA to epidermal growth factor receptor (EGFR).

This PA construct was shown to be capable of translocating a recently discovered protease, Ras and RapI Specific Protease (RRSP), which cleaves and inactivates the signal effector, Ras, found in the cytosol. We demonstrated that when Ras is inhibited in this manner, downstream growth signaling through pERK is ablated and health of a pancreatic cancer cell line (AsPC-l) is affected. Our results suggest that this retargeted PA, rnPA-Elv3, efficiently translocates cytotoxic material into EGFR-positive tumor cells and thus presents a possible avenue for development of a potent therapeutic. Using the same approach, we also took another previously engineered antibody, sm3e and expressed it as a fusion to PA to confer specificity to carcinoembryonic antigen (CEA). Though CEA is not typically internalized, we demonstrated that this retargeted PA (mPA-sm3e) retained the property of endocytosis and translocation and was able to deliver toxins to inhibit proliferation of AsPC-1 tumor cells.

Finally, the retargeted PA variants, mPA-Elv3 and mPA-sm3e, were further characterized for tumor growth inhibition using mouse models. Nude mice were treated with the engineered PA variants against EGFR and CEA to test delivery of toxins into the cells of subcutaneous tumors. Initial results were promising, and future work should be aimed at additional studies confirming this work in mouse models. Our work has demonstrated that protein engineering can be used to retarget PA against tumor cells with positive results. We believe that the modularity and versatility of this retargeting strategy holds great potential in the design of anti-cancer therapeutics.