In the mature viral particle, two plus sense copies of the HIV-1 RNA genome and viral enzymatic proteins are enclosed in a conical core, which is composed of a lattice of CA hexamers and pentamers. After the HIV-1 Envelope fuses with the target-cell membrane, the viral core is released into the cytoplasm. Although previous models suggested that the core immediately disassembled after membrane fusion, the persisting integrity of the core in the cytoplasm and the presence of CA at the NPC indicate that CA is not an immediate castaway during early steps of replication and is instead functionally associated with the PIC facilitating critical steps in infection. HIV-1 infects dividing and nondividing cells equally well due to efficient utilization of NPCs . NPCs are transport channels that span the nuclear envelope and regulate bidirectional transport of macromolecules between the nucleus and cytoplasm. With an estimated molecular mass of more than 100 MDa, NPCs are composed of multiple copies of 30 different proteins called nucleoporins (Nups); including transmembrane Nups (Poms), structural Nups, and FG (phenyalanine-glycine) Nups. FG Nups are essential components of the nuclear diffusion barrier, and provide docking sites for transport receptors. NPCs allow the passive diffusion of molecules with a diameter of up to 9 nm, and active translocation of large cargoes with a diameter of up to 39 nm. Structural analysis suggests that the NPC can dilate up to 50 nm in diameter. How the PIC traverses the NPC has been a subject of great inquiry. Early experiments with HIV-1/MLV chimeric viruses revealed that replacement of HIV-1 CA with MLV CA impairs the ability of HIV-1 to infect nondividing cells. These findings were supported by studies showing specific mutations in CA also prevented HIV-1 infection of nondividing cells. Studies with host factors that regulate nuclear transport solidified a role for CA in this process. Genome-wide screens for HIV-1 host factors exhibited significant overlap in identifying karyopherins as well as nuclear pore protein components. In particular, transportin-3 (TNPO3 or TRN-SR2), Nup358 (also known as RNABP2), and Nup153 emerged as potent regulators of HIV-1 infection. Nup358, the largest FG Nup, is the main component of NPC filaments that extend towards the cytoplasmic side of the pore, and plays an essential role in regulating cargo trafficking by forming a physical meshwork of FG repeats. Another FG Nup, Nup153, is anchored in the nuclear side of the NPC and its FG enriched filaments extend into the nucleoplasm. Depletion of Nup358 or Nup153 impairs HIV-1 infection and reduces 2-LTR circle formation but does not affect reverse transcription. Parallel studies from our group on CPSF6 had revealed that mutation forms of the protein blocked HIV-1 nuclear entry. Notably, we selected for a virus resistant to this nuclear entry block and obtained N74D HIV-1, again implicating a role of CA in regulating HIV-1 nuclear entry. However, unlike previous HIV-1 CA mutants that were impaired for nuclear entry, the N74D mutant virus efficiently infected nondividing cells. We exploited this property to test if N74D HIV-1 had different nuclear pore requirements relative to WT HIV-1. Indeed we soon discovered that TNPO3, Nup358, and Nup153 depletion impaired WT HIV-1 but not N74D HIV-1 infection. WT HIV-1 and N74D HIV-1 do overlap in utilization of some Nups, and N74D HIV-1 appears to be more dependent on Nup85 and Nup155 relative to WT HIV-1. With this finding, we and others more carefully investigated the relationship between CA and NPC components. Nup358 and Nup153 are thought to have distinct interactions with CA. One group has shown that Nup358 interacts with CA through a CypA-homology domain in C-terminal domain of Nup358, while other studies have found that three FG repeats in the N-terminal domain are sufficient to support HIV-1 infection. In contrast, Nup153 has been shown to bind to a conserved pocket in CA, also targeted by CPSF6 and the antiviral compounds PF-74 and BI-2. Interestingly, both Nup153 and CPSF6 possess 'FG' dipeptides that are necessary for binding CA. Notably, HIV-1 with CA mutations that prevent CypA-interaction also appeared to exhibit reduced sensitivity to Nup153 depletion. CypA is a highly abundant cellular protein and binds to the HIV-1 CA residues glycine 89 and proline 90 on the proline-rich loop between helices 4 and 5. Disruption of this interaction by CA mutation (such as G89V or P90A), by cyclosporine A (CsA, a competitive inhibitor of CypA), by CypA knockdown, or by homozygous deletion of the CypA gene, impairs HIV-1 replication in most human cells. Although CypA is incorporated into HIV-1 particles, CypA-CA interaction in the target cell, rather than in the producer cell, is necessary to enhance viral replication. CypA can also have deleterious effects on HIV-1 replication. Passage of HIV-1 in a CD4+ HeLa cells in the presence of CsA selected two mutants (A92E and G94D) in the CypA-binding loop of CA, although neither mutation affected the affinity of CA for CypA. Viruses bearing either A92E or G94D, show attenuated HIV-1 infectivity in some human cell lines, such as HeLa and H9 cells; however, reducing the CypA-CA interaction by CsA treatment or by introducing an additional mutation at proline 90, rescued HIV-1 infectivity in these cells. In contrast, the A92E and G94D mutants are able to replicate in the presence or absence of CsA in other cell lines, such as Jurkat cells. How these CA mutants behave very differently following CsA treatment in different target cells is still a puzzle. One possible explanation for these phenotypes is that differential effects of CypA mutants on HIV-1 replication in different cell types might be due to variations in CypA expression levels. In support of this idea, some studies have shown that there is correlation between different CypA expression levels and CsA effects in different cell lines. Another possibility to explain the CsA-dependency of these mutants in some cell types is a CypA-dependent restriction factor activity. In owl monkey cells, TRIMCyp blocked HIV-1 infection, but restriction was released by CA mutants that disrupt the interaction with CypA, and by CsA treatment. Since A92E HIV-1 infection was also rescued by CA mutants or CsA, it was hypothesized that a CypA-dependent restriction factor was inhibiting A92E replication in these cells. Use of heterokaryons has shown that CsA-dependent infection by A92E and G94D mutants is due to a dominant cellular restriction factor, but the restriction is not by retrovirus restriction factor TRIM5alpha. Because A92E HIV-1 and G94D HIV-1 are impaired at the level of nuclear entry in nonpersmissive cell types, we sought to understand whether CypA interaction with the WT HIV-1 PIC also had the potential to restrict nuclear transport - but HIV-1 had adapted in use of the NPC to overcome the impediment. To investigate the role of nuclear pore subcomplexes in HIV-1 infection, we systematically depleted all thirty-two human nuclear pore proteins in HeLa cells with siRNA, and then infected with VSV-G pseudotyped reporter viruses. In this effort, we have identified HIV-1 dependency on nucleoporins that are regulated by CA interactions with CypA. Strikingly, CypA dictates the nuclear import route utilized by HIV-1, favoring an FG-receptor mediated pathway. We hypothesize that CypA regulates access the N74 pocket of HIV-1 CA and facilitates subsequent interaction with FG-nucleoporins.
