In contrast to oncoretroviruses that replicate only in dividing cells and require nuclear envelope (NE) disassembly during mitosis to integrate their genetic material into the host cell genome, HIV-1 and other lentiviruses have the ability to productively infect non-dividing cells, such as terminally-differentiated macrophages . In the case of HIV-1, these cell populations represent important targets during the initial steps of infection and largely contribute to the establishment of viral reservoirs . The ability of HIV-1 to infect non-dividing cells relies on mechanisms allowing active transport of the so-called "preintegration complex" (PIC), the nucleoprotein complex containing the viral DNA, from the cytoplasm to the nuclear compartment through the intact NE. While nuclear import of the PIC is essential for virus replication in non-dividing cells, it was also proposed that uncoating of the viral capsid after virus entry might rather be the rate-limiting step in the ability of HIV-1 to infect such non-dividing cells . The molecular details underlying this process are still unknown, but a certain body of evidence suggests that the PIC may be transported along the microtubule network to accumulate at the nuclear periphery before anchoring to the NE (for review, see Ref. ).
Although the composition of the HIV-1 PIC changes during its travel to the nucleus, three viral proteins, namely the matrix protein (MA), integrase (IN) and the auxiliary viral protein R (Vpr), remain tightly associated with the viral DNA and have thus been proposed as potential mediators of the nuclear import of the PIC. The central DNA flap structure generated upon completion of the reverse transcription process has been involved in this active process. While the exact contribution of these distinct viral determinants in the nuclear import of the PIC is still controversial (for review, see Ref. ), HIV-1 Vpr specifically facilitates virus replication in non-dividing cells and differentiated macrophages [5-8]. In addition, it was recently reported that some tRNA species incorporated into virus particles may also promote nuclear import of the viral DNA .
HIV-1 Vpr is a highly conserved 96-amino acid (a.a.) basic protein (14 kDa). The analysis of the soluble full length Vpr polypeptide by nuclear magnetic resonance (NMR) allowed the three-dimensional (3D) structure determination of the protein. Vpr consists of an hydrophobic central core domain, with three α-helices (H1 a.a. 17–33, H2 a.a. 38–50 and H3 a.a. 55–77), that are connected by loops and surrounded by two flexible N- and C-terminal domains negatively and positively charged, respectively . By contrast with other HIV-1 auxiliary proteins, Vpr is specifically incorporated at a high copy number in virus particles [11-15], and is consequently present in the cytoplasm of newly infected cells, indicating that it certainly plays specific roles in the early post-entry steps of viral replication . In addition to its role in the nuclear import of the viral PIC, Vpr displays several other activities, including an effect on the fidelity of the reverse-transcription process, an arrest of the cell cycle at the G2/M transition, an induction of apoptosis and the transactivation of the HIV-1 LTR as well as host cell genes (for review, see Ref. . Although the exact contribution of these activities along the virus life cycle is still debated, Vpr-induced G2-arrest has been proposed to provide a favorable cellular environment for optimal transcription of HIV-1 , while the modulation of the virus mutation rate seems required for efficient spreading of HIV-1 in primary macrophages .
When expressed either in dividing or non-dividing cells, HIV-1 Vpr displays evident karyophilic properties and is clearly concentrated at the NE at steady state [20-23]. This latter observation was correlated with its binding to several components of the nuclear pore complex (NPC) which selectively regulates the trafficking of macromolecules or complexes between the nucleus and cytoplasm [24-26]. The NPC is a large supramolecular structure embedded into the NE and composed of around 30 unique proteins termed nucleoporins (Nups) . About half of these Nups contain Phe-Gly repeats (FG-repeats) that contribute directly to the active nucleo-cytoplasmic transport. While initial studies supported the idea that Vpr could bind the FG-rich regions of several Nups, including the human Nup54 and Nup58 , the rodent Pom121  and the yeast Nsp1p , a more recent study described a direct interaction between Vpr and the human CG1 nucleoporin . This interaction does not require the FG-rich region of hCG1 but rather a region without consensus motif found in the N-terminal domain of the protein. Using an in vitro nuclear import assay, it has been demonstrated that hCG1 contributed in the accumulation of Vpr to the NE .
Only a few reports have tried so far to evaluate the virological impact related to the property of HIV-1 Vpr to localize at the NE [25,29]. In the present study, we have explored the role of Vpr accumulation at the NE for the Vpr functions, including G2-arrest and pro-apoptotic activities, and for virus replication in primary macrophages. Single-point Vpr mutants, including two new independent mutants that specifically failed to interact with hCG1, were characterized. Like other mutants with substitutions within the first α-helix of Vpr, they failed to localize at the NE and were impaired for G2-arrest and cell death induction, indicating a functional link between these activities and the Vpr accumulation at the NE. Finally, the replication of the hCG1-binding deficient Vpr mutant viruses was impaired in monocyte-derived macrophages (MDMs) from some but not all donors, suggesting that Vpr targeting to the nuclear pore complex is not absolutely required, but can improve HIV-1 replication in macrophages.