Viruses have evolved elaborate mechanisms to evade or inactivate the complex system of sensors and signaling molecules that make up the host innate immune response. Here we show that human coronavirus (HCoV) NL63 and severe acute respiratory syndrome (SARS) CoV papain-like proteases (PLP) antagonize innate immune signaling mediated by STING (stimulator of interferon genes, also known as MITA/ERIS/MYPS). STING resides in the endoplasmic reticulum and upon activation, forms dimers which assemble with MAVS, TBK-1 and IKKε, leading to IRF-3 activation and subsequent induction of interferon (IFN). We found that expression of the membrane anchored PLP domain from human HCoV-NL63 (PLP2-TM) or SARS-CoV (PLpro-TM) inhibits STING-mediated activation of IRF-3 nuclear translocation and induction of IRF-3 dependent promoters. Both catalytically active and inactive forms of CoV PLPs co-immunoprecipitated with STING, and viral replicase proteins co-localize with STING in HCoV-NL63-infected cells. Ectopic expression of catalytically active PLP2-TM blocks STING dimer formation and negatively regulates assembly of STING-MAVS-TBK1/IKKε complexes required for activation of IRF-3. STING dimerization was also substantially reduced in cells infected with SARS-CoV. Furthermore, the level of ubiquitinated forms of STING, RIG-I, TBK1 and IRF-3 are reduced in cells expressing wild type or catalytic mutants of PLP2-TM, likely contributing to disruption of signaling required for IFN induction. These results describe a new mechanism used by CoVs in which CoV PLPs negatively regulate antiviral defenses by disrupting the STING-mediated IFN induction.

The design, synthesis, X-ray crystal structure, molecular modeling, and biological evaluation of a series of new generation SARS-CoV PLpro inhibitors are described. A new lead compound 3 (6577871) was identified via high-throughput screening of a diverse chemical library. Subsequently, we carried out lead optimization and structure-activity studies to provide a series of improved inhibitors that show potent PLpro inhibition and antiviral activity against SARS-CoV infected Vero E6 cells. Interestingly, the (S)-Me inhibitor 15h (enzyme IC50 = 0.56 μM; antiviral EC50 = 9.1 μM) and the corresponding (R)-Me 15g (IC50 = 0.32 μM; antiviral EC50 = 9.1 μM) are the most potent compounds in this series, with nearly equivalent enzymatic inhibition and antiviral activity. A protein-ligand X-ray structure of 15g-bound SARS-CoV PLpro and a corresponding model of 15h docked to PLpro provide intriguing molecular insight into the ligand-binding site interactions.

We describe here the design, synthesis, molecular modeling, and biological evaluation of a series of small molecule, nonpeptide inhibitors of SARS-CoV PLpro. Our initial lead compound was identified via high-throughput screening of a diverse chemical library. We subsequently carried out structure-activity relationship studies, and optimized the lead structure to potent inhibitors that have shown antiviral activity against SARS-CoV infected Vero E6 cells. Based upon the X-ray crystal structure of one of the potent inhibitors 24-bound to SARS-CoV PLpro, a drug-design template was created. Our structure-based modification led to the design of a more potent inhibitor, 2 (enzyme IC50 = 0.46 μM; antiviral EC50 = 12.5 μM). Interestingly, its methylamine derivative 49 displayed good enzyme inhibitory potency (IC50 = 1.3 μM) and most potent SARS antiviral activity (EC50 = 2.5 μM) in the series. We have carried out computational docking studies and generated a predictive 3D-QSAR model for SARS-CoV PLpro inhibitors.