Selective prodrug activation: Using the cancer cell environment to activate a protein switch.

B6 In general, current chemotherapeutic agents used for cancer treatment suffer from a lack of selectivity for cancer cells over normal cells. The need for more targeted therapies lead to the development of alternative methods, such as GDEPT. In GDEPT, a “suicide” gene that is capable of converting a prodrug into a toxin is delivered to cancer cells. This is followed by the systematic administration of that prodrug, which is converted into a toxin within the cancer cell. However, the selectivity for cancer cells over normal cells still remains an issue for GDEPT. Normal cells are often transfected with the suicide gene, making it necessary to develop vehicles for specific delivery of the suicide gene to cancer cells. We propose to circumvent the need for specific delivery vehicles by creating a “cancer-sensing” therapeutic protein whose prodrug converting activity is dependent upon the cancer cell environment. This therapeutic protein can be delivered, non-specifically, to cells within and surrounding a tumor. The cancer cells producing this therapeutic protein will convert a systematically administered prodrug into a toxin within the cancer cell, thereby killing it, while the therapeutic protein will be inactive in normal cells.
 Our lab has invented a novel protein engineering strategy by combining the domains of two independent proteins into a single hybrid protein. As a proof of principle, we have created maltose-activated beta-lactamases by combining maltose binding protein with TEM1 beta-lactamase. These “switch” proteins’ beta-lactamase enzyme activity is increased 500-fold in the presence of maltose, but they are inactive in the absence of maltose. We envision that a therapeutic protein switch could be created in the same fashion as the maltose-activated proteins, by combining a prodrug activating enzyme and a protein that binds a cancer-specific marker. We are focusing on hypoxia inducible factor-1 alpha (HIF-1a) as the cancer-specific marker. HIF-1a is completely absent in normal tissue due to rapid degradation, but is overexpressed within many types of cancer. In this manner, the therapeutic protein will be inactive in normal tissue and will be activated by HIF-1a to convert the prodrug into a toxin in cancer cells. We are using the HIF-1a binding domain from the p300 protein and a cytosine deaminase to make our therapeutic switch. Cytosine deaminase (CD) is an enzyme that is able to convert the relatively harmless 5-fluorocytosine (5-FC) prodrug into the toxic 5-fluorouracil (5-FU). Cancer cells that overexpress HIF-1a will activate the protein switch to convert 5-FC into 5-FU, thereby killing the cancer cells. This therapeutic protein could have the potential to be used in a treatment method that is analogous to GDEPT, but will have superior “killing” efficiency with reduced side-effects because the selectivity of our method arises at the molecular level.