key: cord-301180-ndiwmnv0
authors: Lin, Min-Hsuan; Sivakumaran, Haran; Apolloni, Ann; Wei, Ting; Jans, David A.; Harrich, David
title: Nullbasic, a Potent Anti-HIV Tat Mutant, Induces CRM1-Dependent Disruption of HIV Rev Trafficking
date: 2012-12-10
journal: PLoS One
DOI: 10.1371/journal.pone.0051466
sha: 
doc_id: 301180
cord_uid: ndiwmnv0

Nullbasic, a mutant of the HIV-1 Tat protein, has anti-HIV-1 activity through mechanisms that include inhibition of Rev function and redistribution of the HIV-1 Rev protein from the nucleolus to the nucleoplasm and cytoplasm. Here we investigate the mechanism of this effect for the first time, establishing that redistribution of Rev by Nullbasic is not due to direct interaction between the two proteins. Rather, Nullbasic affects subcellular localization of cellular proteins that regulate Rev trafficking. In particular, Nullbasic induced redistribution of exportin 1 (CRM1), nucleophosmin (B23) and nucleolin (C23) from the nucleolus to the nucleus when Rev was coexpressed, but never in its absence. Inhibition of the Rev:CRM1 interaction by leptomycin B or a non-interacting RevM10 mutant completely blocked redistribution of Rev by Nullbasic. Finally, Nullbasic did not inhibit importin β- or transportin 1-mediated nuclear import, suggesting that cytoplasmic accumulation of Rev was due to increased export by CRM1. Overall, our data support the conclusion that CRM1-dependent subcellular redistribution of Rev from the nucleolus by Nullbasic is not through general perturbation of either nuclear import or export. Rather, Nullbasic appears to interact with and disrupt specific components of a Rev trafficking complex required for its nucleocytoplasmic shuttling and, in particular, its nucleolar accumulation.

Both the Human immunodeficiency virus type-1 (HIV-1) Tat and Rev proteins are encoded by two exons arranged in alternative reading frames on fully spliced viral mRNA [1] . Tat and Rev are similar in size; Tat is typically 101 amino acids long and Rev is typically 116 amino acids long, and both have RNA binding domains composed of arginine and, in the case of Tat, lysine residues which bind to different HIV-1 RNA stem loop structures. Tat binds to an RNA structure in the 59 untranslated region (UTR) of all viral transcripts called the Trans-Activation Response element (TAR), while Rev binds to an intronic region retained by incompletely spliced transcripts called the Rev Response Element (RRE). The RNA binding domains of both proteins also function as a nuclear/nucleolar localization signal (NLS/NoLS), although recent evidence implies that Tat may passively enter the nucleus by diffusing through nuclear pores [2] . Both proteins are localized primarily in the nucleus; Tat is observed throughout the nucleoplasm with nucleolar accumulation, whereas the nucleocytoplasmic shuttling Rev concentrates in the nucleolus in addition to localizing to the nucleoplasm and, to a lesser extent, to the cytoplasm.

Trafficking of Rev in cells has been studied extensively ( Fig. 1 ) [3, 4] . In the nucleolus, Rev promotes the nuclear export of various HIV-1 mRNAs by directly binding to singly-spliced and unspliced viral transcripts via the RRE contained therein (Fig. 1, step 1 ).

Exportin 1 (also called CRM1 and XPO1) binds to Rev through a nuclear export signal (NES; HIV-1 NL4-3 Rev amino acids 73 to 84, LQLPPLERLTLD) [5, 6, 7] , which leads to colocalization of Rev and CRM1 in the nucleolus and subsequent export of the Rev:mRNA complex from the nucleus to the cytoplasm (Fig. 1 , step 2). Many other cellular proteins can contribute to Rev nuclear export, including hRIP/Rab, eIF5A, DDX3, DDX1, RNA helicase A, and PIMT that act through Rev, and Matrin 3 and Sam68 that bind to viral mRNA [3, 8, 9, 10, 11, 12] . The Rev:mRNA complex disassembles in the cytoplasm (Fig. 1, step 3) allowing Rev to recycle back to the nucleus using the transportin 1 or importin b nuclear import pathways (Fig. 1, step 4) [3] . Once Rev enters the nucleus, nucleophosmin (B23) facilitates transport of Rev to the nucleolus (Fig. 1 , step 5) [13] . B23 is reported to be necessary for the nucleolar localization of both Rev and Tat through interaction with their respective basic domains [13, 14, 15, 16, 17] .

We recently described a mutant of the two-exon HIV-1 Tat protein, termed Nullbasic, that exhibits antiviral properties by inhibiting multiple steps of the HIV-1 replication cycle [18] . Nullbasic was created by replacing the entire arginine-rich RNA binding domain of wild type Tat with glycine and alanine residues. Like similarly mutated one-exon Tat mutants, Nullbasic exhibits transdominant negative effects on Tat-dependent HIV-1 gene expression [18] . However, unlike previously reported mutants [19, 20, 21] , Nullbasic also effectively suppresses the steady state levels of unspliced and singly-spliced viral mRNA, an activity caused by the inhibition of HIV-1 Rev activity [18] . The inhibition by Nullbasic was attributed, in part, to a subcellular redistribution of Rev from the nucleolus to the nucleoplasm and cytoplasm, but precisely how Nullbasic mediates this effect is unknown.

In this study, we investigated the effect of Nullbasic on the nucleocytoplasmic transport machinery responsible for Rev trafficking. Not surprisingly, Rev recruited and colocalized with several cellular proteins in the nucleolus, including CRM1, B23 and nucleolin (C23). However, coexpression of Nullbasic with Rev resulted in the unexpected redistribution of CRM1 and other nucleolar proteins from the nucleolus to the nucleoplasm in a Revdependent manner, whereas Nullbasic did not affect the distributions of these nucleolar proteins when expressed alone. Our experiments support the possibility that Nullbasic interferes with an unknown component of the Rev nucleocytoplasmic transport complex required for the nucleolar organization and function of Rev.

HEK293T and HeLa cells were cultured in RPMI 1640 medium supplemented with 10% (v/v) newborn bovine serum (Invitrogen) and 1% (v/v) penicillin-streptomycin. All cells were typically incubated at 37uC in a humidified 5% CO 2 atmosphere. When 50% -80% confluent, cells were transfected with desired plasmids using FuGENE 6 transfection reagent (Roche Applied Science) according to the manufacturer's instructions. At 24 h post-transfection, cells were harvested for further analysis. In certain experiments where Rev nuclear export was blocked, Leptomycin B (Sigma-Aldrich) was added to growth medium at a final concentration of 20 nM and incubated for 2 h before cell fixation.

The two-exon Tat (101 amino acids) expression plasmid with FLAG epitope (Tat-FLAG) was a gift from Monsef Benkirane, Institute de Génétique Humaine, France. GFP 2 -cNLS and GFP 2 -M9core plasmids were gifts from Ralph Kehlenbach, Georg- August-University of Göttingen, Germany [22] . The GFP 2 expression plasmid was obtained from Addgene, plasmid #20738. The construction of the Nullbasic-FLAG and Myc-Rev plasmids have been previously described [18] . The pGCH infectious molecular clone is a HIV proviral plasmid which expresses authentic HIV-1 RNA using the CMV immediate early promoter. The plasmid consists of a pGEM7z (-) (Promega) backbone containing the CMV promoter fused so that the transcription initiation site uses the correct HIV-1+1 position. With respect to the HIV-1 transcription start site, HIV-1 sequences +1 to +933 from the HIV-1 pNL4-3 proviral plasmid (obtained from the ADIS Research and Reference Reagent Program, plasmid #114) were inserted into the pGEM7z (-) via MluI and SphI restriction enzymes sites and +934 to +9251 from the HIV-1 pNL4-3 were inserted into the pGEM7z (-) via SphI and XbaI restriction enzymes sites. The HIV-eGFP expression plasmid was engineered by inserting an eGFP sequence from pcDNA3.1-eGFP into the HIV-1 nef gene in pGCH proviral expression vector through BamHI and XbaI restriction enzyme sites. The expression vector encoding Nullbasic-mCherry was generated by polymerase chain reaction (PCR) amplification of mCherry from pmCherry-LacRep (obtained from Addgene, plasmid #18985), which was then inserted into the Nullbasic-FLAG plasmid via BsrGI and XhoI restriction enzyme sites. The eluates were then separated by sodium dodecylsulfatepolyacrylamide gel electrophoresis (SDS-PAGE) and electroblotted onto a polyvinylidene fluoride (PVDF) membrane (Pall) using a semi-dry transfer system (Bio-Rad Laboratories). Tat-FLAG and Nullbasic-FLAG proteins were detected with a rabbit anti-FLAG polyclonal antibody (Cell Signalling Technology). Myc-Rev was detected with a rabbit anti-Myc polyclonal antibody (Cell Signalling Technology). Endogenous CDK9 was detected with a rabbit anti-CDK9 monoclonal antibody (Cell Signalling Technology). Primary antibodies were detected with houseradish peroxidase (HRP)-conjugated goat anti-rabbit antibody (Invitrogen).

HeLa cells were grown on glass coverslips and transfected with appropriate expression vectors. After 24 h, the cells were fixed in 3% (w/v) paraformaldehyde at room temperature for 10 min and quenched with 50 mM NH 4 Cl for 5 min. Cells were then permeabilized with 0.1% (v/v) Triton X-100 for 15 min and blocked in 10% (v/v) normal goat serum (Millipore) for 15 min. Cellular endogenous proteins were detected with rabbit anti-CRM1 polyclonal (Santa Cruz Biotechnology), rabbit antifibrillarin monoclonal (Cell Signalling Technology), mouse antinucleophosmin monoclonal (Invitrogen) and mouse anti-nucleolin monoclonal (Santa Cruz Biotechnology) antibodies, and Myc-Rev was probed with either mouse anti-Myc monoclonal antibody (Millipore) or rabbit anti-Myc polyclonal antibody (Cell Signalling Technology). Rev expressed from proviral HIV-1 was probed with a mouse anti-HIV-1 Rev monoclonal antibody (Santa Cruz Biotechnology). Primary antibodies were detected with FITCconjugated goat anti-mouse or anti-rabbit antibodies (Invitrogen) and Alexa Fluor 647-conjugated goat anti-rabbit (Invitrogen), Cy5-conjugated goat anti-mouse antibodies (Invitrogen) or Cy3congugated goat anti-rabbit antibodies (Invitrogen) (to detect the rabbit anti-FLAG polyclonal antibody). Nuclei were stained with 1 mM 49,6-diamidino-2-phenylindole (DAPI, Invitrogen). Finally, coverslips were mounted onto slides with ProLong Gold antifade reagent (Invitrogen). Fluorescent images were captured using a Leica TCS SP2 confocal scanning microscope (Leica Microsystems) and DeltaVision deconvolution microscope (Applied Precision) with 636 objective lenses and standard lasers and filters for FITC, mCherry, Cy5, Cy3 and DAPI fluorescence.

We previously observed that Nullbasic induces considerable redistribution of Rev within the cell, from the nucleolus to the nucleoplasm and cytoplasm, leading to significant down regulation of steady state levels of unspliced or singly-spliced viral mRNA [18] . To investigate whether Nullbasic is able to alter Rev's subcellular localization in the context of HIV-1, we visualized the subcellular distribution of Rev derived from HIV-1 provirus in the presence or absence of Nullbasic. To overcome transcriptional inhibition caused by Nullbasic [18] , we utilized a CMV-driven HIV-1 proviral plasmid, pGCH, which expresses viral factors independent of Tat. The plasmid also has the eGFP gene inserted into the nef reading frame (here called HIV-eGFP) to enable visualization of HIV expression in cells. HIV-eGFP was transfected into HeLa cells with or without FLAG-tagged Nullbasic fused to the mCherry fluorescent reporter protein (Nullbasic-mCherry). We noted strong nuclear accumulation of Nullbasic-mCherry compared to a previous study [18] , whereas one exon Tat proteins with basic domain mutation were also shown to predominantly localized in the cytoplasm [20, 23, 24] . Further comparison of Nullbasic localization in HeLa cells showed that Nullbasic fused to FLAG (Fig. S1 , rows 3 and 5) or mCherry ( Fig. 2A , row 2) had very similar patterns of distribution in the nucleus and cytoplasm. The localization of Rev was determined by indirect immunofluorescence microscopy using an anti-Rev antibody 24 h post-transfection. As expected, Myc epitope-tagged Rev (Myc-Rev) expressed alone was prominent in the nucleolus and the nucleoplasm but to a lesser extent ( Fig. 2A, row 1) . When Rev was expressed by HIV-eGFP, it was distributed throughout the nucleus (Fig. 2B, row 1 ). This nuclear distribution of Rev is consistent with its role in nucleocytoplasmic trafficking of viral mRNA [3, 25] . Coexpression of HIV-eGFP and Nullbasic-mCherry resulted in the distinct redistribution of Rev to the nucleoplasm and cytoplasm (Fig. 2B, row2) , comparable to the subcellular redistribution of Myc-Rev by Nullbasic-mCherry ( Fig. 2A, row2) . Although Nullbasic was shown to inhibit Revmediated viral mRNA transport leading to limited expression of viral proteins from unspliced or singly spliced viral mRNA [18] , these data indicate that Nullbasic-induced redistribution of Rev is independent of other HIV-1 factors under these experimental conditions.

Others have shown that wild type Tat does not interact with Rev within cells [26, 27, 28] . To formally exclude the possibility that Nullbasic may interact with Rev, co-immunoprecipitation reactions were performed using HEK293T cells transfected with plasmids encoding FLAG epitope-tagged Nullbasic (Nullbasic-FLAG) and Myc-Rev. Immunoprecipitation of wild type Tat-FLAG with anti-FLAG beads did not co-immunoprecipitate Myc-Rev ( Fig. 2C ) but did, as expected, co-immunoprecipitate endogenous CDK9, a member of the pTEF-b protein complex that interacts with the Tat activation domain [29] . Similarly, Nullbasic-FLAG did not show evidence of co-immunoprecipitation of Myc-Rev but did with CDK9 (Fig. 2C ). The latter interaction is most likely similar to the interaction between wild type Tat and CDK9, through the activation domain preserved in Nullbasic-FLAG. These experiments imply that a Nullbasic:Rev interaction is not likely to be responsible for the observed mislocalization of Rev.

In order to investigate the effects of Nullbasic on Rev nucleocytoplasmic transport, we performed the following analyses on Myc-Rev expressed by our Myc-Rev expression plasmid instead of HIV-expressed Rev, since the subcellular localization of Myc-Rev and proviral Rev is similarly altered by Nullbasic.

CRM1 can directly bind to the NES sequence in Rev to facilitate its export from the nucleus [3] . We therefore investigated the effect of Nullbasic-mCherry on the association between Myc-Rev and endogenous CRM1 in transfected HeLa cells. In control cells transfected with the parental pcDNA3.1 plasmid, CRM1 was observed widely throughout the nucleus and along the nuclear membrane (Fig. 3, row 1) . As previously observed [28, 30, 31] , Myc-Rev expression induced CRM1 relocalization and prominent colocalization of Rev and CRM1 in nucleoli (Fig. 3, row 2) . This colocalization could be reversed by treating cells with leptomycin B (LMB), a specific CRM1 inhibitor that directly binds to and inactivates CRM1 [32] , thereby preventing its interaction with Rev (Fig. 3, row 3) . Expression of Nullbasic alone had no apparent effect on CRM1 distribution (Fig. 3, row 4) . Colocalization between Rev and CRM1 could also be partially disrupted by the expression of Nullbasic-mCherry, in the presence of which Myc-Rev, but not CRM1, redistributed to the cytoplasm (Fig. 3, row 5) . Importantly, treatment of Myc-Rev and Nullbasic-mCherry expressing cells with LMB resulted in the complete restoration of Myc-Rev nucleolar accumulation (Fig. 3, row 6) , suggesting that Nullbasic requires functional CRM1 to mislocalize Rev most likely by an indirect interaction.

To test this hypothesis further, we utilized the Rev mutant RevM10 which contains a non-functional NES and hence is unable to be recognized by CRM1 [33] . As reported for RevM10 [17, 34] , Myc-RevM10 (Fig. 4, row 1 ) was unable to interact and colocalize with CRM1 in the nucleolus. Interestingly, coexpression of Nullbasic-mCherry had no disruptive effect on the nucleolar localization of Myc-RevM10 since, unlike wild type Myc-Rev, no redistribution to the cytoplasm was observed (Fig. 4, row 2) . We thus conclude that an intact Rev:CRM1 interaction is required for Nullbasic to relocalize Rev to the cytoplasm.

B23 is a nucleolar chaperone and shuttling protein that, in HeLa cells, is observed in both the nucleolus and nucleoplasm [35] , being associated with nucleolar ribonucleoprotein structures and involved in ribosome biogenesis. B23 is reported to assist Rev and Tat transport into the nucleolus through interactions with their respective basic domains [13, 14, 15, 16, 17, 36] . Mutant Tat proteins lacking a basic domain, like Nullbasic, have been reported not to interact with B23 in vitro [14] . As previously observed, Myc-Rev colocalized with B23 in nucleolar structures (Fig. 5, row 1) . While expression of Nullbasic-mCherry alone had no apparent effect on B23 nuclear distribution (Fig. 5, row 2) , coexpression of Myc-Rev with Nullbasic-mCherry resulted in the strong redistribution of both Myc-Rev and B23 throughout the nucleus, and the observation of Myc-Rev in the cytoplasm (Fig. 5, row 3) . Since B23 is known to strongly bind to Rev [13] , it is likely that the redistribution of B23 by Nullbasic is Rev-dependent.

The data thus far suggest that Nullbasic can alter the localizations of nucleolar proteins in a Rev-dependent manner.

To determine if Nullbasic can generally disrupt nucleolar composition, we monitored the localization of C23 and fibrillarin, two nucleolar proteins that have not been reported to interact with Rev. Both C23 and fibrillarin colocalized with Myc-Rev within the nucleoli of transfected cells (Fig. 6 , rows 1 and 4, respectively). While expression of Nullbasic-mCherry alone had no apparent effect on the nucleolar localization of either protein (Fig. 6 , rows 2 and 5), coexpression of Nullbasic-mCherry with Myc-Rev induced relocalization of C23 but not fibrillarin from the nucleolus to the nucleoplasm, in a manner concomitant with the redistribution of Myc-Rev (Fig. 6, rows 3 and 6 , respectively). We therefore conclude that Nullbasic does not cause a generalized mislocalization of nucleolar proteins, but effects the relocalization of only certain, specific proteins (including CRM1, B23 and C23) in a manner dependent on the presence of Rev.

Five different importin b family members, including importin b, transportin 1, importin 5, importin 7 and importin 9, have been reported to recognize the Rev NLS and facilitate the transport of Rev into the nucleus, with different pathways being used in a cell line-dependent manner [22, 37, 38, 39] . In HeLa cells, for example, transportin 1 is the dominant Rev nuclear import pathway [22, 40] , while the importin b pathway is favored in HEK293T cells [40] . To eliminate the possibility that Nullbasic-mediated blockade of a nuclear import pathway leads to the observed cytoplasmic accumulation of Rev, two NLS-containing reporters, GFP 2 -cNLS and GFP 2 -M9core, which mimic cargo for nuclear import by importin b and transportin 1 respectively, were expressed individually or along with Nullbasic-mCherry in HeLa cells. As shown in figure 7 , there was no change in the nuclear accumulation of either reporter in the presence of Nullbasic-mCherry (Fig. 7, rows 2 and 4) compared to in its absence (Fig. 7,  rows 1 and 3) . Nullbasic-mCherry did not affect localization of GFP 2 lacking a nuclear import signal (Fig. 7, rows 5 and 6) . Similar experiments indicated that coexpression of Myc-Rev with Nullbasic-mCherry likewise had no effect on the nuclear localizations of the respective reporters (data not shown). While we cannot exclude interference of other pathways, we conclude that the accumulation of Rev in the cytoplasm induced by Nullbasic is not caused by a general inhibition of the importin b and transportin 1 nuclear import pathways.

This study aimed to elucidate a possible mechanism by which the two-exon anti-viral acting Tat mutant, Nullbasic, perturbs Rev subcellular localization through the active mislocalization of a set of cellular proteins that control Rev trafficking. Nucleocytoplasmic shuttling of Rev is critical for HIV-1 infectivity since viral protein synthesis, and subsequent virion assembly, require the coordinated transport of incompletely spliced viral mRNAs from the nucleus. Nullbasic's action in downregulating Rev's mRNA transport function by mechanisms that include the substantial mislocalization of Rev from nucleolus to nucleoplasm and cytoplasm (see Fig. 2 ) [18] is therefore of great importance. Immunoprecipitation of either Nullbasic-FLAG or Tat-FLAG from lysates of cells coexpressing Myc-Rev revealed no co-immunoprecipitation of Myc-Rev (Fig. 2C) , supporting the idea that Nullbasic does not interact with Rev. Furthermore, fluorescence resonance energy transfer analyses between a Tat-GFP fusion protein and a Rev-BFP fusion failed to detect any interaction in living cells [26, 27] , a result which we reason will likewise apply to Nullbasic and Rev. It is thus reasonable to conclude that Nullbasic interferes with Rev nucleolar localization by an indirect mechanism.

The fact that Nullbasic strongly disrupts the nucleolar localization of Rev in an indirect manner suggests that Nullbasic may interfere with a cellular factor important for Rev nucleolar accumulation. The nucleolus is a complex structure where several thousands of different cellular proteins have been identified [41] . Among these, CRM1 is a Rev binding protein that regulates Rev trafficking throughout the nucleus. CRM1 is a cellular receptor that recognizes and binds to the leucine-rich NES domain of Rev, an interaction which facilitates the egress of Rev/viral mRNA complexes through nuclear pores [3, 4, 42] and is readily visualized in cells as a colocalization in the nucleolus (Fig. 3, row 2 ). Our evidence strongly indicates that CRM1 is a mediator through which Nullbasic induces Rev redistribution, since disrupting the interaction between CRM1 and Rev mitigates the effect of Nullbasic. This was evident through the use of LMB (Fig. 3) , a specific inhibitor of CRM1, and the RevM10 mutant, whose altered NES is no longer recognized or bound by CRM1 (Fig. 4) . In both cases, Rev retained a nucleolar localization despite high levels of Nullbasic expression. As sole expression of Nullbasic does not appear to directly affect CRM1 localization (Fig. 3 , row 1 compared to row 4), the precise mechanism by which CRM1 mediates Nullbasic-induced redistribution remains to be elucidated.

B23, C23 and fibrillarin are among the most abundant proteins in the nucleolus [43] . B23 is a nucleolar shuttling factor that recognizes and binds to the NoLS of Rev [13] , Tat [44] and C23 [45] . While primarily located in the nucleolus, B23 can display a nucleocytoplasmic distribution as a consequence of its shuttling function [46] . Nullbasic-mediated redistribution of Rev in this study was accompanied by the mislocalizations of B23 (Fig. 5) and C23 (Fig. 6 ), but not fibrillarin (Fig. 6) . It is likely that the mislocalization of B23 in the presence of redistributed Rev is a consequence of their strong interaction [13] rather than through an interaction between Nullbasic and B23 because, while Tat has been reported to interact with B23 through its basic domain [44] , the same domain in not present in Nullbasic [18] . Due to the lack of evidence of a direct interaction between C23 and Rev, the mislocalization of C23 may be a result of its interaction with mislocalized B23 [45] . The observed mislocalizations, however, were largely confined to the nucleoplasm even when Rev was substantially redistributed into the cytoplasm.

It is important to note that Nullbasic did not affect the localizations of CRM1, B23 and C23 in the absence of Rev (Figs. 3, 4, 5 and 6 ). That Nullbasic did not affect the nucleolar distribution of fibrillarin (Fig. 6) , either in the presence or absence of Rev, indicates that Nullbasic is not inducing a generalized deformation or reconfiguration of cellular nucleoli. Further, the retention of the importin b and transportin 1 nuclear import reporters in the nucleus in the presence of Nullbasic demonstrates that Nullbasic does not effect a general perturbation of these trafficking pathways (Fig. 7) . It is evident, therefore, that Nullbasic induces the redistribution of CRM1, B23 and C23 only in the presence of Rev.

Considering all of the available data, we propose that Rev induces the formation of a nuclear or nucleolar complex to reconcile why Nullbasic redistributes only certain nucleolar proteins in a Rev-dependent manner. This Rev containing complex would be enriched for CRM1 and colocalized with B23 and C23 in order to facilitate the proper processing and nuclear export of HIV-1 unspliced and singly-spliced transcripts [27, 47, 48] . Nullbasic may destabilize the complex that includes Rev by binding to an unidentified factor that is important for its formation, resulting in the redistribution of Rev and concomitant redistributions of B23 and C23 throughout the nucleoplasm. The liberated Rev:CRM1 complexes would freely traffic out of the nucleus, resulting in the observed redistribution of Rev into the cytoplasm, and which would be counteracted by LMB treatment or the RevM10 mutation.

Similar nucleolar hijacking and modification have been described for other viruses. The replication of certain DNA and RNA viruses (such as adenovirus, herpes simplex virus, hepatitis B virus, HIV-1, hepatitis C virus, Japanese encephalitis virus, Hendra virus, Nipah virus, SARS coronavirus and dengue virus) require the trafficking of viral proteins through the nucleolus [42] . In some instances, for example during herpes simplex virus and adenovirus infections, this results in the dramatic mislocalization of cells were immunostained with anti-Myc (green) and either anti-C23 (magenta, upper panels) or anti-fibrillarin (magenta, lower panels) antibodies before the subcellular localizations of Myc-Rev, endogenous C23 or endogenous fibrillarin, and NB-mCherry (red) were analyzed by fluorescence microscopy. Nuclei were stained with DAPI. The overlay panels show the merge of the Myc-Rev panel with either the C23 or fibrillarin panels (rows 1, 3, 4 and 6) and the NB-mCherry panel with either the C23 or fibrillarin panels (rows 2 and 4). All figures are representative of five fields each from four independent experiments. The white bar in last panel is equal to 10 mm. doi:10.1371/journal.pone.0051466.g006 C23 and B23 [42, 49] . Given such evidence from other viruses, it is entirely consistent that Rev may similarly manipulate nucleoli factors in HIV-1 infected cells.

Identification of Nullbasic interacting factors that are required for Rev nucleolar accumulation may provide insights into the design of potent yet specific antiretroviral inhibitors. Overall, our data provide evidence that Nullbasic appears to interact with and disrupt specific components of a Rev trafficking complex important for Rev's nucleocytoplasmic trafficking and nucleolar accumulation, resulting in CRM1-mediated Rev redistribution. Mutant HIV-1 proteins such as Nullbasic thus not only provide insights into the functions of their wild type counterparts, but may also reveal novel drug targets in previously unrecognized host/ pathogen interactions.

CRM1-dependent subcellular redistribution of Rev from the nucleolus by Nullbasic is not through general perturbation of either nuclear import or export. Rather, our data suggest that Nullbasic appears to interact with and disrupt specific components of a Rev trafficking complex required for Rev nucleocytoplasmic shuttling and, in particular, nucleolar accumulation. Figure 7 . Nullbasic does not affect the nuclear import of importin b and transportin 1 cargoes. HeLa cells were transfected to express the importin b-dependent nuclear import reporter GFP 2 -cNLS, or the transportin 1-dependent nuclear import reporter GFP 2 -M9core, either alone (rows 1 and 3, respectively) or with Nullbasic (NB)-mCherry (rows 2 and 4, respectively). The subcellular localizations of the reporter proteins (green) and NB-mCherry (red) were analyzed in fixed cells by fluorescence microscopy. Nuclei were stained with DAPI. The overlay panels show the merge of the GFP reporter and DAPI panels. Plasmid expressing GFP 2 was also transfected into cells either alone or with NB-mCherry as a vector control (rows 5 and 6, respectively). Figures are representative of five fields each from four independent experiments. The white bar in last panel is equal to 10 mm. doi:10.1371/journal.pone.0051466.g007 CRM1-Dependent Inhibition of Rev Trafficking

HIV-1 replication

In vivo study of HIV-1 Tat arginine-rich motif unveils its transport properties

Cellular proteins and HIV-1 Rev function

Virus-host interactions: role of HIV proteins Vif, Tat, and Rev

CRM1 is an export receptor for leucine-rich nuclear export signals

CRM1 is responsible for intracellular transport mediated by the nuclear export signal

The importin-beta family member Crm1p bridges the interaction between Rev and the nuclear pore complex during nuclear export

Characterization of the HIV-1 RNA associated proteome identifies Matrin 3 as a nuclear cofactor of Rev function

Inhibition of HIV replication by dominant negative mutants of Sam68, a functional homolog of HIV-1 Rev

Sam68 functions in nuclear export and translation of HIV-1 RNA

Rev-ing up post-transcriptional HIV-1 RNA expression

Matrin 3 is a co-factor for HIV-1 Rev in regulating post-transcriptional viral gene expression

Specific complex of human immunodeficiency virus type 1 rev and nucleolar B23 proteins: dissociation by the Rev response element

Spatial association of HIV-1 tat protein and the nucleolar transport protein B23 in stably transfected Jurkat T-cells

The post-transcriptional regulator Rev of HIV: implications for its interaction with the nucleolar protein B23

The cytotoxicity of human immunodeficiency virus type 1 Rev: implications for its interaction with the nucleolar protein B23

Analysis of intracellular trafficking and interactions of cytoplasmic HIV-1 Rev mutants in living cells

Potent inhibition of HIV-1 replication by a Tat mutant

Trans-dominant Tat mutants with alterations in the basic domain inhibit HIV-1 gene expression

A transdominant tat mutant that inhibits tat-induced gene expression from the human immunodeficiency virus long terminal repeat

Inhibition of human immunodeficiency virus type 1 replication is enhanced by a combination of transdominant Tat and Rev proteins

The nuclear pore component Nup358 promotes transportin-dependent nuclear import

Mutational analysis of the conserved basic domain of human immunodeficiency virus tat protein

Inhibition of human immunodeficiency virus type 1 and type 2 Tat function by transdominant Tat protein localized to both the nucleus and cytoplasm

Immunodeficiency virus rev trans-activator modulates the expression of the viral regulatory genes

Automatic and quantitative measurement of protein-protein colocalization in live cells

In vivo HIV-1 Rev multimerization in the nucleolus and cytoplasm identified by fluorescence resonance energy transfer

Kinetic and molecular analysis of nuclear export factor CRM1 association with its cargo in vivo

Interactions between Tat and TAR and human immunodeficiency virus replication are facilitated by human cyclin T1 but not cyclins T2a or T2b

A synthetic HIV-1 Rev inhibitor interfering with the CRM1-mediated nuclear export

Nucleoporins nup98 and nup214 participate in nuclear export of human immunodeficiency virus type 1 Rev

Leptomycin B inactivates CRM1/exportin 1 by covalent modification at a cysteine residue in the central conserved region

Functional dissection of the HIV-1 Rev trans-activator-derivation of a trans-dominant repressor of Rev function

Inhibition of human immunodeficiency virus Rev and human T-cell leukemia virus Rex function, but not Mason-Pfizer monkey virus constitutive transport element activity, by a mutant human nucleoporin targeted to Crm1

NPM1/B23: A Multifunctional Chaperone in Ribosome Biogenesis and Chromatin Remodeling

Intracellular trafficking and interactions of the HIV-1 Tat protein

The arginine-rich domains present in human immunodeficiency virus type 1 Tat and Rev function as direct importin betadependent nuclear localization signals

Interactions between HIV Rev and nuclear import and export factors: the Rev nuclear localisation signal mediates specific binding to human importin-beta

Multiple importins function as nuclear transport receptors for the Rev protein of human immunodeficiency virus type 1

Intermolecular masking of the HIV-1 Rev NLS by the cellular protein HIC: novel insights into the regulation of Rev nuclear import

NOPdb: Nucleolar Proteome Database-2008 update

Involvement of the nucleolus in replication of human viruses

Involvement of the plant nucleolus in virus and viroid infections: parallels with animal pathosystems

Protein B23 is an important human factor for the nucleolar localization of the human immunodeficiency virus protein Tat

C23 interacts with B23, a putative nucleolar-localization-signal-binding protein

Major nucleolar proteins shuttle between nucleus and cytoplasm

The roles of nucleolar structure and function in the subcellular location of the HIV-1 Rev protein

Recruitment of the Crm1 nuclear export factor is sufficient to induce cytoplasmic expression of incompletely spliced human immunodeficiency virus mRNAs

Adenovirus protein V induces redistribution of nucleolin and B23 from nucleolus to cytoplasm

We thank Dongsheng Li and Kylie Warren for their advice and comments.