Combinatorial In Silico Drug‐Lead Optimization for Targeted Inhibitors of P‐Glycoprotein

P‐glycoprotein (P‐gp) is an ATP‐powered efflux pump of the plasma membrane that is responsible for removing toxins from cells. While this is beneficial in normal cells, P‐gp overexpression is often responsible for the failure of chemotherapies in cancer and HIV treatments. Because P‐gp transports drugs that encompass a wide variety of structures, multidrug resistances (MDR) to many unrelated therapeutics is often a consequence of P‐gp overexpression. Attempts to identify inhibitors of P‐gp that could be used to reverse MDR have often failed because they are also pumped out of the cell. We recently identified three inhibitors that target the nucleotide binding domains of P‐gp and avoid the drug binding domains. These compounds reverse MDR in cancer cell lines. A big problem in optimizing any drug‐lead is that the number of possible chemical variations of the lead always out‐numbers those that can be synthesized and tested. We present here a novel computational approach that generates many thousands of compounds related to a drug‐lead structure, analyzes them for interaction with the target protein, and rates them for improvements in target binding potential. We have applied this approach to inhibitors of P‐gp that reverse multidrug resistances in cancer and report on the application of these variants to high performance computational drug docking experiments with Pgp. These methods aimed at expediting the optimization of drug leads should be generalizable to many different systems. Supported by NIH NIGMS [Grant 1R15‐GM094771‐01A1] to PDV and JGW, and SMU Undergraduate Research Assistantship, SMU Engaged Learning, and Hamilton Undergraduate Research Scholarships to KO.