(P32) Rational design and characterization of novel HIV-1 capsid inhibitors


Rafael Ceña [1], Anders Sönnerborg [1,2,3], Ujjwal Neogi [1], Kamal Singh [1,3]


1. Division of Clinical Microbiology, ANA Futura Laboratory, Department of Laboratory Medicine, Karolinska Institute, Stockholm, Sweden. 2. Division of Infectious Diseases, Department of Medicine Huddinge, Karolinska Institute, Stockholm, Sweden. 3. Department of Molecular Microbiology and Immunology, and Bond Life Sciences Center, University of Missouri, Columbia, MO, United States.


Background: HIV-1 capsid interacts with many host factors plays key roles at multiple steps of viral replication. At the initial steps, controlled uncoating (disassembly) of capsid ensures efficient reverse transcription of the RNA genome into the double-stranded DNA and facilitates nuclear import of pre-integration complex. At the later steps, capsid protomers assemble into a fullerene-like core and form a mature and infectious virus. Thus, the inhibition of capsid assembly and/or disassembly are potential therapeutic strategies, and many small molecule capsid inhibitors have been reported. Despite the fact that none of these molecules have been approved for clinical use, compound PF-3450074 (PF74) has been extensively studied experimentally. Poor metabolic properties of PF74 have restricted its clinical use but triggered the search for more efficient compounds. Recently reported GS-CA compounds are the best-in-class capsid inhibitors. Location of GS-CA1 and PF74 resistance mutations, and our recent structural studies suggested that GS-CA compounds and PF74 share the binding pocket, and extended interactions of GS-CA compounds with adjacent monomer are most likely contribute to their greater potency. Hence, we used rational approach and designed novel capsid inhibitors. In our two-pronged strategy, we developed peptides, and small molecule inhibitors using the crystal structures of capsid bound to host factors, PF74, BI-2 as well as the modelled structure of capsid/GS-CA complexes.

Methods: Microscale thermophoresis (MST) and molecular docking was used to determine capsid binding affinities with peptides and small molecules. Hexameric capsid (stabilized by inter-monomer disulfide bonds) was used in MST assays, whereas crystal structure of native form of capsid was used in molecular docking studies. Thermal shift assays and transmission electron microscopy was used for determining the impact of peptides and compounds on the assembly and disassembly of the native form of capsid.

Results: The MST data showed that the capsid hexamers bind linear and cyclic peptides that consist structural moieties of CPSF6 (cleavage and polyadenylation specific factor 6), NUP153 (nucleoporin 153) and PF74 with nanomolar affinities (Kd ranging between 32 nM to 100 nM). Structural studies showed that the peptides docked at GS-CA-binding site with comparable docking score and potential energy. Transmission electron microscopy (TEM) data showed that the cyclic peptide interfered with disassembly of the capsid tubes, whereas linear peptide inhibited assembly of the capsid monomers to form tubes. These results were confirmed in thermal shift assays. Designed small molecule inhibitors were studied in the MST assays for their capsid binding affinities. These molecules also showed high affinities for capsid hexamers (Kd = 34 nM to 227 nM). The structural data showed that substituted-phenyl rings of designed compounds superposed on the phenyl rings of PF74 and modified ring structures (in place of indole ring of PF74) have extensive interactions with adjacent capsid monomer.

Conclusions: Our results suggest that designed peptides consisting of chemical moieties from host factors and small molecules that have interactions with both subunits have strong potential for further characterization in virological experiments, which will be presented during the conference.