Two-Metal Ion-Dependent Enzymes as Potential Antiviral Targets in Human Herpesviruses

The majority of drug discovery efforts against herpesviruses have focused on nucleoside analogs that target viral DNA polymerases, agents that are associated with dose-limiting toxicity and/or a narrow spectrum of activity. We are pursuing a strategy based on targeting two-metal ion-dependent (TMID) viral enzymes. This family of enzymes consists of structurally related proteins that share common active sites containing conserved carboxylates predicted to coordinate divalent cations essential for catalysis. Compounds that target TMID enzymes, such as HIV integrase and influenza endoribonuclease, have been successfully developed for clinical use. HIV integrase inhibitors have been reported to inhibit replication of herpes simplex virus (HSV) and other herpesviruses; however, the molecular targets of their antiviral activities have not been identified. We employed a candidate-based approach utilizing several two-metal-directed chemotypes and the potential viral TMID enzymatic targets in an effort to correlate target-based activity with antiviral potency. The panel of compounds tested included integrase inhibitors, the anti-influenza agent baloxavir, three natural products previously shown to exhibit anti-HSV activity, and two 8-hydroxyquinolines (8-HQs), AK-157 and AK-166, from our in-house program. The integrase inhibitors exhibited weak overall anti-HSV-1 activity, while the 8-HQs were shown to inhibit both HSV-1 and cytomegalovirus (CMV). Target-based analysis demonstrated that none of the antiviral compounds acted by inhibiting ICP8, contradicting previous reports. On the other hand, baloxavir inhibited the proofreading exonuclease of HSV polymerase, while AK-157 and AK-166 inhibited the alkaline exonuclease UL12. In addition, AK-157 also inhibited the catalytic activity of the HSV polymerase, which provides an opportunity to potentially develop dual-targeting agents against herpesviruses.