HIV-1 Vpr is a viral regulatory protein that acts at many points along the viral life cycle. Vpr intracellular localization and dwell time in different compartments depend on the intrinsic molecular interactions exerted between distinctive Vpr domains and other viral or cellular partner proteins (for a review, see ). These interactions depend not only on the stage of the viral life cycle but also on the expression of intracellular binding partners, which are ultimately differentially regulated in selected cellular phenotypes. Indeed, HIV-1 Vpr has been shown to be essential to productively infect cells of the monocyte-macrophage lineage [8–10], whereas its presence is dispensable for replication within lymphocytes . This could suggest the presence of different host cellular factors differentially expressed in the two cell types, which Vpr relies upon, or of negative regulatory factors in lymphocytes (absent in monocytes) that eliminate the requirement for Vpr with respect to productive replication.
In the studies described herein, different cellular microenvironments were examined for their potential impact on the pattern of Vpr localization observed under selected experimental conditions. To this end, six different cell lines representative of important HIV-1 targets within the PB, BM, and CNS were explored. We determined that the location of the ZsGreen-emitting protein at either the N- or C-terminus did not affect Vpr localization in HEK 293T cells, unlike a previously reported study performed with the EGFP . Indeed, as reported by several independent studies [18, 44, 57–60], Vpr is found predominantly at the nuclear rim and to a lesser extent in speckles present both inside the nucleus and within the cytoplasm. In addition, localized herniations were detected at the nuclear envelope in HEK 293T cells, as has been previously reported by other investigators . The herniated areas may facilitate shuttling or leakage of Vpr molecules between the nuclear and cytoplasmic compartments. To model a more relevant in vivo scenario, studies are underway to determine whether the presence of the other HIV-1 viral proteins could influence Vpr localization by cotransfection studies. The results have clearly indicated that none of the other viral proteins, added either singly or in combination, altered the intracellular location of Vpr. Studies then proceeded to examine Vpr localization of Vpr within the two human cell lines representative of the main cellular targets for HIV-1 within the PB. In this respect, our studies clearly demonstrated that the CD4+ T-lymphocytic Jurkat cells exhibited primarily a cytoplasmic localization consisting of punctate cytoplasmic foci. These results partly contrast with published results employing a Jurkat cell line , although with a different clone than the one utilized in the present study. In this regard, Bolton and coworkers demonstrated primarily nuclear localization of Vpr, with small amounts of Vpr localized to the cytoplasm, although other cells within the same population showed a more perinuclear localization pattern. However, the difference between the two studies may relate to the difference in the posttransfection time at which cells were analyzed (24 h after transfection herein as compared with 48 h in the previous study ), although in an additional study, a Vpr peptide (residues 52-96) added extracellularly to lymphocytes and lymphoblasts was distributed exclusively within the cytoplasm . In contrast to T cells, within a promonocytic U-937 cell line, Vpr appears to accumulate exclusively within the nucleus, as also shown in one study employing transfected monocyte-derived macrophages . Two additional independent studies have also found a nuclear pattern in monocyte-derived macrophages either transduced by a Vpr-containing adenovirus vector  or treated with an extracellular synthetic Vpr .
Hematopoietic differentiation occurs within the BM, after which time cells leave the BM and eventually enter the peripheral circulation. As one approach to model this developmental process, Vpr localization studies were performed in a human CD34+ BM progenitor cell line to determine whether this was similar to the pattern observed in either differentiated cell type (lymphocytes or monocytes). TF-1 cells showed perinuclear localization, with accumulation involving punctate foci both within the nucleus and likely at the endoplasmic reticulum, a pattern of accumulation more similar to that observed in 293T cells than either the lymphocytic or monocytic cell line. These observations suggest that cellular differentiation and/or activation results in changes in the pattern of Vpr accumulation. These changes may be closely linked to differentiation- and/or state-of-activation-dependent alterations in the presence or modification of intracellular factors regulated during the differentiation process. Indeed, the Vpr localization pattern in TF-1 was clearly different from that found within representative cell lines of CD4+ T-lymphocytic or monocytic origin.
The CNS, although anatomically compartmentalized, is surveyed by cells involved in immune surveillance and is also breached by potentially neurotoxic substances and/or cells infected by any of a number of infectious agents, including HIV, particularly under circumstances of pathogenic insult. Within the CNS, HIV-1 preferentially targets microglia, trafficking monocyte-derived macrophages and brain microvascular endothelial cells, and, to a lesser extent, astrocytes, which likely represent an important CNS reservoir simply based on their numbers. Owing to the lack of a reliable microglial cell line, we have investigated two different types of astroglioma cell lines (U-373 MG and U-87 MG) and found that Vpr displayed a similar localization pattern in both. A predominant accumulation around the nuclear envelope was observed, although Vpr also localized within the cytoplasm and to a lesser extent in the nucleus and around the nucleoli. This localization pattern was most similar to that observed in one previous study performed in U-87 MG cells .
The differential localization patterns we observed in the aforementioned cell lines indicate that the cellular phenotype affects the intracellular distribution of Vpr and point to a role that Vpr might play in each cell type. In this regard, a nuclear localization pattern in monocyte-derived macrophages may reflect a role Vpr could have as a transcription factor. This possibility is substantiated by studies that confirmed a lack (or low level) of infectivity for Vpr-deficient molecular clones in cells of the monocytic lineage. Conversely, a cytoplasmic accumulation in CD4+ T lymphocytes could denote a more active participation in later phases of the viral cycle, such as virion incorporation or as a secreted protein. In the former case, the presence of Vpr within the cytoplasm could reflect an increased interaction with p6 proteins as more viral transcription takes place. Indeed, within CD4+ T cells, Vpr is known to synergize with Nef in activating nuclear factor of activated T cells (NFAT)-dependent and cyclic AMP response element (CRE)-directed transcription . This leads to increased viral genome production owing to the presence of NFAT and CRE-responsive elements within the HIV-1 promoter. In the latter case, because Vpr has been shown to be released from Jurkat cells  and extracellular Vpr has been detected within the blood of HIV-1-infected patients [66, 67], this cell type could be responsible for secretion of Vpr within the PB compartment, given the Vpr cytoplasmic distribution in this T-cell line. However, it remains possible that Vpr release is just the consequence of Vpr's apoptogenic nature rather than the result of an active secretion process (for a review, see Ferrucci et al., 2011, manuscript in press). A further explanation of the different intracellular localization of Vpr between Jurkat and U-937 cells could lie in the differential expression of Vpr binding partners between the two cell lines. For instance, Vpr is known to interact with UNG, a uracil DNA glycosylase repair enzyme , whose gene gives rise to UNG1 and UNG2 by alternative splicing, found cytoplasmically and nuclearly, respectively . A possible phenotype-specific differential expression of either form in the two cell lines, such that UNG1 were found predominantly in Jurkat (or CD4+ T lymphocytes) and UNG2 primarily in U-937 (or monocytic cells), could then explain why Vpr localizes within the cytoplasm or the nucleus, respectively. Other important intracellular host factors Vpr has been shown to interact with are the Sp1 and CCAAT enhancer binding proteins (C/EBP) , which are differentially regulated in several different cell lines. For instance, the overabundance of C/EBP in U-937 monocytic cells compared with Jurkat lymphocytes (Y. Liu and B. Wigdahl, unpublished observations) could explain a more nuclear localization of Vpr in the former cell line. Conversely, a more perinuclear localization in the other cell lines examined could be the cause of decreased affinity with the aforementioned proteins, rather a stronger binding for factors such as importin-α for which Vpr was shown to interact [4, 18, 21].
Within the CNS, infected infiltrating CD4+ T lymphocytes and trafficking monocytes play an important role in HIV-1 pathogenesis by continually seeding virus into the CNS. Because Vpr has been found in a soluble form within the cerebrospinal fluid of HIV-1-infected patients , these two cell types could participate in the release process both within the PB and CNS compartments. In this regard, after either lymphocytes or monocytes infected with HIV-1 and expressing Vpr have crossed the blood-brain barrier it is possible that intra-CNS cytokines or other signaling molecules induce changes in the levels and activities of intracellular factors. In turn, these may play a role in altering the intracellular accumulation of Vpr, which ultimately leads to functional changes in these infected cell populations in ways that may alter the course of HIV-1-associated neurologic disease. In addition, given the extensive amount of Vpr in the cytoplasm of astrocytes and because of their number within the CNS, these cells also contribute to Vpr release either through a regulated secretory process or through necrosis or apoptosis. Among other CNS-resident cells, brain perivascular and parenchymal microglial cells could also contribute to seeding viral infection within the CNS. Only one study has been published that showed a predominant nuclear localization of Vpr in these two cell types, although the study was performed in Vpr-transgenic mice .