David Liu’s group reprogrammed a botulinum toxin protease to trigger cancer cell death
David Liu’s group reprogrammed a botulinum toxin protease to trigger cancer cell death
On August 3, a team from the Broad Institute and Harvard published a study of the BoNT/X protease, the part of botulinum toxin that cleaves proteins. Using phage-assisted continuous evolution, the researchers trained it to cleave procaspase-1 or gasdermin D, proteins involved in inflammatory cell death. The protein crossed cell membranes and triggered cell death in cultured cells. In mice with pancreatic tumors, the X(PC) variant slowed tumor growth.
Botulinum toxin is known for its ability to relax muscles. Its protease enters cells and cleaves specific proteins. In a study from David Liu’s group, the researchers redirected this ability toward proteins that trigger pyroptosis, an inflammatory form of cell death.
During pyroptosis, caspase-1 cleaves gasdermin D. The released gasdermin D fragment forms pores in the cell membrane, causing the cell to break apart and release signals to the immune system. The X(GD) variant cleaved gasdermin D directly and caused this form of death in several cancer cell lines. The X(PC) variant activated the caspase-1 precursor and triggered lytic death together with apoptosis, a form of programmed cell self-destruction, in leukemia cells.
The authors produced these variants using PACE, or phage-assisted continuous evolution. In this system, phage replication depends on whether the associated protease cleaves a specified target. Natural BoNT/X cleaves the protein VAMP1, whereas X(PC) acquired activity against procaspase-1. The botulinum toxin transport domain allowed both variants to enter cancer cells.
Here, the protein performs two tasks at once: it recognizes a new intracellular target and crosses the cell membrane. The authors tested the uptake of purified protein in cell-based experiments. In a mouse model of drug-resistant pancreatic cancer, the tumor cells received the X(PC) construct in advance. Activating the construct then slowed tumor growth.
Controlling cell death requires treating the choice of molecular target and the route into the cell as parts of the same engineering problem. This study demonstrates both parts in cultured cells and in a mouse tumor model.