HIV Capsid and Integration Targeting

Abstract

Integration of retroviral reverse transcripts into the chromosomes of the cells that they infect is required for efficient viral gene expression and the inheritance of viral genomes to daughter cells. Before integration can occur, retroviral reverse transcription complexes (RTCs) must access the nuclear environment where the chromosomes reside. Retroviral integration is non-random, with different types of virus-host interactions impacting where in the host chromatin integration takes place. Lentiviruses such as HIV efficiently infect interphase cells because their RTCs have evolved to usurp cellular nuclear import transport mechanisms, and research over the past decade has revealed specific interactions between the HIV capsid protein and nucleoporin (Nup) proteins such as Nup358 and Nup153. The interaction of HIV capsid with cleavage and polyadenylation specificity factor 6 (CPSF6), which is a component of the cellular cleavage and polyadenylation complex, helps to dictate nuclear import as well as post-nuclear RTC invasion. In the absence of the capsid-CPSF6 interaction, RTCs are precluded from reaching nuclear speckles and gene-rich regions of chromatin known as speckle-associated domains, and instead mis-target lamina-associated domains out at the nuclear periphery. Highlighting this area of research, small molecules that inhibit capsid-host interactions important for integration site targeting are highly potent antiviral compounds.

Keywords: CPSF6; HIV; antiviral inhibitor; capsid; integration; integration targeting; nuclear import.

Figure 1

Intracellular HIV-1 trafficking to active genes for integration (not drawn to scale). The RTC, modeled after protein data bank (PDB) code 3J3Y (see Figure 2D), is shown interacting with a microtubule for retrograde transport via CA-binding protein BICD2. After reaching the microtubule organizing complex (MTOC), antegrade transport via kinesin-FEZ1 may shuttle the RTC to the NPC for nuclear transport. Nup proteins that have been shown to interact with CA are highlighted to the right, color coded to demarcate their relative positions within the NPC and noted subcomplexes. During nuclear entry, CPSF6 aids the release of the RTC from Nup153/the NPC nuclear basket [17]. Following nuclear entry, the preintegration complex (PIC) traffics to nuclear speckles (NSs) for integration into speckle-associated domains (SPADs) in a manner that is dependent on the CA-CPSF6 interaction [12]. Lacking this interaction, PICs mislocalize to the nuclear periphery and uncharacteristically target lamina-associated domains (LADs) for integration [15,18]. The PIC is depicted partially uncoated. Transcriptionally active Zone 1 and Zone 2 regions of chromatin are highlighted by green color [13].

Figure 2

HIV-1 CA, capsomere, and capsid shell structures. (A) The structure of the CA monomer from protein data bank (PDB) entrant 4XFY [29] is color-coded to highlight the N-terminal domain (NTD) (residues 1–146; light blue), interdomain linker (residues 147–149; magenta), and C-terminal domain (CTD) (residues 150–231; dark blue). Labels demarcate secondary structural elements. The host factor cyclophilin A (CypA) engages CA via the CypA-binding loop [30]. (B) The hexameric capsomere from PDB entrant 4U0D [31]. The upper “top” view represents what would be seen from the outer surface of the core shell. Rotating this view 90° into the plane of the page yields the lower “side” view. Coloring is maintained from panel A. (C) The pentameric capsomere (PDB code 3P05) [32]. Upper and lower images are analogous to those in panel B. Orange is used to highlight NTDs of pentameric capsomeres; other coloring is same as panel A. (D) Atomic model of assembled core shell (PDB code 3J3Y). The model is composed of 186 hexameric capsomeres and 12 pentamers [33]. Coloring is same as in panels A–C.

Figure 3

HIV-1 CA and capsomere interaction structures. (A) Two common host factor binding regions of CA, which are approximated by dashed circles and labeled Region 1 and Region 2, are superimposed on the structure of the Nup358 cyclophilin-homology domain (CHD) bound to the CA NTD (PDB code 4LQW) [80]. Region 2 is a pocket encompassing alpha helices α3, α4, and α7. Residues Asn57 and Asn74 located on α3 and α4, respectively, are shown as sticks. The orientation and labelling of the CA NTD is preserved from Figure 2A. (B) Structure of HIV-1 hexamer capsomere bound by CPSF6 residues 313–327 (green; PDB code 4U0B) [31]. Backbone atoms of CPSF6 residues Phe321 and Leu326, which are shown as sticks with italicized labels, interact with CA residues Asn57 and Asn74, respectively (dashed lines). The CTD visible in the lower left (dark blue) is from a neighboring CA molecule. (C) Interaction of Nup153 residues 1407-1423 (green) with HIV-1 CA hexamer (PDB code 4U0D) [31]. The backbone atoms of Phe1417 interact with CA residue Asn57, akin to CPSF6 residue Phe321 (compare with panel B). CA residue Asn74, by contrast, is distal from the Nup153 peptide. (D) Interaction of the antiretroviral inhibitor GS-6207 in the Region 2 binding pocket (PDB code 6VKV) [81]. The drug crystallized in two different binding orientations, both of which are shown. Interactions with CA residues Asn57 and Asn74 are highlighted. Stick colorings in panels A–D: red, oxygen; blue, nitrogen; yellow, sulfur; light grey, fluorine.

Figure 4

Domain organizations of CPSF6, Nup358, and Nup153. (A) Diagram of mCPSF6-358 restriction factor relative to human CPSF6 isoforms 2 and 1; the smaller isoform 1 is the form predominantly expressed in cells. Numbers demarcate domain boundaries. RRM, RNA recognition motif; PRD, proline-rich domain; RSLD, Arg/Ser-like domain. The region of the PRD that confers binding to CA is marked by a black line in human CPSF6 isoform 1. The original mCPSF6-358 construct harbored a heterologous 18-mer sequence (pink) at its C-terminus that did not impact anti-HIV restriction activity [94]. (B) Nup358 and Nup153 domain organizations. I–IV, Ran-binding domains I–IV; ZnF, zinc finger domains; E3, E3 ubiquitin ligase domain; CHD, cyclophilin-homology domain; FG, Phe-Gly peptide repeat domain.