Structural basis of stability of human immunodeficiency virus type 1 (HIV-1) capsid

Abstract

Human immunodeficiency virus type 1 (HIV-1) is the etiologic agent of acquired immunodeficiency syndrome (AIDS). Since its discovery in early 1980, many advances have been made in the prevention and management of HIV/AIDS. One of the most important ones has been antiretroviral therapies (ART). Currently, there are more than 30 drugs approved for the treatment of HIV-1 infection. Used as combinations, they efficiently suppress viral loads and reduce AIDS-related deaths. Despite the advances, there is no cure and infection can eventually lead to fatal results. Hence, there is a need for new antivirals with novel mechanisms of action, favorable resistance and low toxicity profiles, which will offer more therapeutic options in the clinic. The HIV-1 capsid protein (CA) has been increasingly viewed as an attractive therapeutic target as it plays a critical role in multiple steps of the virus life cycle, including uncoating, reverse transcription, nuclear entry, integration site selection, and assembly. Moreover, it is also necessary for shielding the deoxyribonucleic acid (DNA) product of reverse transcription from immune surveillance of the target cell. HIV-1 CA interacts with host factors including Cyclophilin A (CypA), cleavage and polyadenylation specific factor 6 (CPSF6), and nucleoporin Nup153. It is synthesized as a central part of the Gag polyprotein, which is cleaved during maturation to give matrix (MA), CA, nucleocapsid (NC), and several other peptides. Of [about]5,000 Gag molecules in the immature HIV-1 virion, only [about]1,500 CA molecules assemble into the mature capsid core that encloses the viral ribonucleic acid (RNA) and enzymes. The core comprises [about]250 hexamers and 12 pentamers that allow the formation of the closed shape. Notably, all these diverse functions and interactions are regulated by the structure of a single, remarkably flexible protein, CA, which has been the main focus of this thesis. Over the past 25 years, several X-ray, nuclear magnetic resonance (NMR) and cryo-electron microscopy (cryo-EM) structures of HIV-1 CA fragments or engineered variants have been solved. However, none of these provide the complete set of molecular details of the critical CA-CA contacts that govern capsid stability, which is at the heart of its biological functions. In this study, we aimed to solve crystal structures of HIV-1 CA, including the elusive structure of the native full-length HIV-1 CA, in the space group P6, which allows the building of the hexameric biological unit. Our findings describe novel interactions between CA monomers related by 6-fold symmetry within a hexamer (intra-hexamer) and by 3-fold and 2-fold symmetry between neighboring hexamers (inter-hexamer). These structures help elucidate how CA builds a hexagonal lattice, the foundation of the mature capsid. Moreover, they demonstrate that the intra- and inter-hexamer interfaces are malleable and can change through an adaptable hydration layer. Disruption of this layer by crystal dehydration treatment alters inter-hexamer interfaces and condenses CA packing. The structures reveal a remarkable plasticity which explains the polymorphism observed in virions. They also establish our experimental system to be the most relevant for the study of CA interactions in a native context. We have used this system to obtain crystal structures of CA in complex with either a CA-binding antiviral (PF74) or a host factor peptide (CPSF6 or Nup153). Previous structures of CA fragments or engineered CA hexamer in complex with PF74, CPSF6 or Nup153 have provided valuable information on CA-host factor interactions, but they lack information on inter-hexamer interfaces. Hence, despite extensive work the details of CA interactions with host factors and pharmacological ligands that regulate the HIV life cycle as well as the structural basis of CA stability that determines virus infectivity are not well understood. Our findings reveal novel information about the changes at the hexamer-hexamer interfaces, thus providing structural insights into how those ligands can affect uncoating and assembly. An enormous research effort has been invested over the years to determine the phenotypes of natural and artificial mutations in HIV proteins, including CA. Those studies have been significant for defining various CA functions and mapping them to the N-terminal (CANTD) and C-terminal (CACTD) domains. Additionally, several mutations have been identified to be useful in studies aimed at dissecting how CA performs its essential nonstructural functions. However, there is currently no structural information on any biologically relevant CA mutants (not counting the cross-linked CA constructs). Such information will reveal mechanisms of stabilization of the capsid core and may be applicable to other pathogenic viruses. Thus, using our experimental setup, we have studied three groups of CA mutants. Initially, we focused on mutations in CA altering core stability and impairing viral infectivity that suggests that core of optimal stability and its proper uncoating are critical for productive infection. P38A and E45A CA mutations have been reported to destabilize or hyperstabilize the capsid, resulting in non-infectious virus. Compensatory mutations T216I and R132T partially rescue infection from the defects associated with P38A and E45A, respectively. We performed crystallographic analysis of CA proteins bearing P38A, P38A/T216I, E45A, and E45A/R132T mutations, evaluated their assembly competence in vitro, and estimated core stability and uncoating. Mutant phenotypes are not a consequence of the main structural rearrangements in CA. Structural analysis suggests mutual electrostatic repulsion between pairs of glutamic (E45) and aspartic (D51) acids forced into proximity by the CA structure provides an environmentally-sensitive switch which can control the state of assembly and disassembly of the capsid. The mutant structures highlight additional rearrangements that may affect host factor recognition and trafficking of small-molecules across the capsid shell. Furthermore, early studies determined that mutations introduced into the loop between CA helices H6 and H7 (residues 122-125, Pro-Pro-Ile-Pro, or PPIP motif) to be lethal or caused decrease in infectivity; however, they were not studied in detail. To address the role of the PPIP motif in the mature capsid, we have crystallized CA proteins bearing P122A, I124A, T58A/I124A, T58S/T107I/P122A, V11I/T58A/P122A, and V11I/T58A/I124A mutations. The structures revealed subtle structural rearrangements not only at the sites of mutations but also at the inter- and intra-hexamer interfaces, as well as at the host factor binding sites. This implies that mutations may alter the stability of the mature capsid core and/or host factor binding and recognition. Hence, the H6-H7 loop of the HIV-1 CA is a new structural element essential for inter-hexamer contacts in the immature Gag and mature capsid lattices. Finally, statistical analysis of the intermolecular interactions between two monomers either within a hexamer or pentamer in the published capsid models revealed the hydrogen bond (H-bond) between the glutamic acid at position 28 and lysine in position 30 (E28[about]K30) being more abandoned at the intra-pentamer, rather than intra-hexamer interfaces. To investigate if it may contribute to counterbalancing the electrostatic destabilization observed in the pentamers, we tested purified mutant CA proteins, harboring R18A, E28A and R18A/E28A mutations, for the cylinder formation in vitro and solved their crystal structures. The results indirectly support the presence of E28~K30 H-bond predominantly in the pentamers. This interaction may regulate pentamer stability and may be essential for the proper CA assembly, stability, and uncoating of the HIV-1 capsid core. Collectively, the structures highlight that CA plasticity is a key factor for its stability and how challenging it is to fully understand the effect of even a single mutation on this highly flexible protein. Our studies unravel the structural basis of core stability, which affects multiple steps in the virus life cycle. Moreover, they provide unique information on how CA structure controls interactions with host factors and small molecule CA-targeting antivirals.