Biophysical Characterization of Nucleophosmin Interactions with Human Immunodeficiency Virus Rev and Herpes Simplex Virus US11

Nucleophosmin (NPM1, also known as B23, numatrin or NO38) is a pentameric RNA-binding protein with RNA and protein chaperon functions. NPM1 has increasingly emerged as a potential cellular factor that directly associates with viral proteins; however, the significance of these interactions in each case is still not clear. In this study, we have investigated the physical interaction of NPM1 with both human immunodeficiency virus type 1 (HIV-1) Rev and Herpes Simplex virus type 1 (HSV-1) US11, two functionally homologous proteins. Both viral proteins show, in mechanistically different modes, high affinity for a binding site on the N-terminal oligomerization domain of NPM1. Rev, additionally, exhibits low-affinity for the central histone-binding domain of NPM1. We also showed that the proapoptotic cyclic peptide CIGB-300 specifically binds to NPM1 oligomerization domain and blocks its association with Rev and US11. Moreover, HIV-1 virus production was significantly reduced in the cells treated with CIGB-300. Results of this study suggest that targeting NPM1 may represent a useful approach for antiviral intervention.


Introduction
Nucleophosmin (NPM1, also known as B23, numatrin, NO38) is a multifunctional phosphoprotein, predominantly localized in the nucleoli, which participates extensively in RNA regulatory mechanisms including transcription, ribosome assembly and biogenesis, mRNA stability,

Antibodies and fluorescent probes
Mouse monoclonal anti-NPM1 (ab10530) recognizing the C-terminal 68-amino acids and rabbit monoclonal anti-NPM1 (ab52644) recognizing the N-terminal 122-amino acids were from Abcam (Cambridge, United Kingdom), Rabbit monoclonal anti-myc from Cell Signaling Technology, Inc. (Boston, USA), Alexa fluor 488 mouse anti-rabbit IgG and Alexa fluor 633, and goat anti-mouse IgG from Molecular Probes (Oregon, USA), and normal monoclonal Rabbit IgG (sc-2027) was from Santa Cruz Biotechnology, Texas, USA.

Immunoblotting
Proteins were heated in Laemmli sample buffer and subjected to SDS-PAGE. The proteins were transferred to nitrocellulose membranes (Hybond C, GE Healthcare) using Mini Trans-Blot cell (100 volt for 1 h) (BIO-RAD, USA), and immunoblotted using monoclonal primary antibody to mouse NPM1 antibody (Abcam), rabbit NPM1 antibody (Abcam), and rabbit myc antibody (Cell Signaling) for 1 h. After three washing steps, membranes were incubated with polyclonal horseradish peroxidase-coupled secondary antibodies for 1 h and signals were visualized by the ECL detection system (GE Healthcare) and images were collected using the Che-moCam Imager ECL (INTAS science imaging, Germany).
Immunoprecipitation COS-7 cells were transiently transfected with cDNA encoding Myc-tagged US11. After 48 h, an equal number of the cells were lysed in a buffer, containing 30 mM Tris/HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 2.5 mM Na-pyrophosphate, 1 mM β-glycerophosphate, 1 mM sodium vanadate, and one EDTA-free protease inhibitor cocktail tablet (Roche, Mannheim, Germany). Lysates were centrifuged at 12,000×g for 2 min. The supernatant was precleared with protein G agarose (Roche, Mannheim, Germany) and divided to three parts for IgG control, beads control and IP, and then incubated with an anti-myc antibody (Cell Signaling) overnight at 4°C. Afterwards, protein G-Agarose beads were added to the lysate for 1 h before recovering the beads by centrifugation at 500×g for 5 min at 4°C. The beads were washed 4-times in the lysis buffer, and resuspended in Laemmli sample buffer. Precipitates and total cell lysate were subjected to SDS-PAGE, and Western blotting as described above.

Analytical size exclusion chromatography (aSEC)
The complex formation of NPM1 OD and US11 FL was analyzed using a superdex 200 10/30 column (GE Healthcare, Uppsala, Sweden) and a buffer, containing 30 mM Tris-HCL (pH 7.5), 150 mM NaCl, 5 mM MgCl 2 , and 3 mM dithiothreitol. The flow rate was sustained at 0.5 ml/min. Fractions were collected at a volume of 0.5 ml and then peak fractions were visualized by 12.5% SDS-PAGE gel and staining using coomassie brilliant blue (CBB).

Pull-down assay
GST, GST-fused NPM1 and HSV-1 US11 variants as well as HIV-1 Rev were expressed in E. coli and purified using standard protocols [30,31]. In order to obtain prey proteins the GSTtag was cleaved off with purified tobacco etch virus (tev) protease and removed by reverse GSH affinity purification. Pull-down experiments were performed by adding 50 μg purified proteins, e.g. HIV-1 Rev and HSV US11 variants, or COS-7 cell lysate transfected with pcDNA-mycUS11 FL to 25 μg of GST-fused NPM1 proteins, immobilized on 100 μl glutathione-conjugated Sepharose 4B beads (Macherey-Nagel, Duren, Germany). The mixture was incubated at 4°C for 1 h in a buffer containing 30 mM Tris/HCl, pH 7.5, 150 mM NaCl, 5 mM MgCl 2 , and 3 mM Dithiothreitol. In cases of RNase treatments, 70 U RNase A (Qiagen, Hilden, Germany) were added to the same buffer in order to determine an RNA dependent interaction between the NPM1 variants and HIV-1 Rev. After four washing steps with the same buffer, proteins retained on the beads were heat-denatured (7 min at 90°C) and analyzed by SDS-PAGE followed by coomassie brilliant blue (CBB) staining or by Western blotting. Mixed samples prior to pull-down (PD) analysis were used as input controls.

Isothermal titration calorimetry (ITC)
All proteins were prepared in ITC buffer, containing 30 mM Tris-HCl, pH 7.5, 150 mM NaCl, 5 mM MgCl 2 , and 1 mM Tris (2-carboxyethyl) phosphine (TCEP) on a size exclusion chromatography (SEC) column (Superdex 200, 16/60, GE Healthcare, Uppsala, Sweden). ITC measurements were performed at 25°C using a VP-ITC system (Microcal, Northampton, MA, USA) as previously reported [32]. The final data analysis was carried out using Origin software (Microcal). The experimental data were evaluated using Origin 7.0 software (Microcal) to determine the binding parameters including association constant (K a ), number of binding sites (n), and enthalpy (ΔH). Control measurements were carried out by titrating buffer to the protein.

Analytical ultracentrifugation (AUC)
Sedimentation velocity centrifugation experiments at 50,000 rpm and 20°C were carried out in a Beckman Optima XL-A (Beckman-Coulter, Brea, CA, USA), equipped with absorption optics, and a four-hole rotor. Samples (volume 400 μL) were filled into standard aluminum double sector cells with quartz glass windows. Measurements were performed in absorbance mode at detection wavelengths 230 nm. Radial scans were recorded with 30 μm radial resolution at~1.5 min intervals. The software package SEDFIT v 14.1 (www.analyticalultracentrifugation.com) was used for data evaluation. After editing time-invariant, noise was calculated and subtracted. In SEDFIT continuous sedimentation coefficient distributions c(s) were determined with 0.05 S resolution and F-ratio = 0.95. Suitable s-value ranges between 0 and 20 S and f/f 0 between 1 and 4 were chosen. Buffer density and viscosity had been calculated with SEDNTERP v 20111201 beta (bitcwiki.sr.unh.edu) [33]. The partial specific volume of NPM1 OD fragment, NPM1 FL and US11 FL were calculated according to the method of Cohn and Edsall [34] as implemented in SEDNTERP. NPM1 OD was analyzed at 0.25 concentrations in 30 mM Tris-HCl, pH 7.5, 150 mM NaCl, and TCEP (1 mM). After equilibrium was reached, concentration profiles were recorded with 10 μm radial resolution and averaging of seven single registrations per radial value. Equilibria had been established at 14,000, 16,000, 25,000, 42,000 and 50,000 rpm. Data evaluation was performed using SEDPHAT.

Multi angle light scattering (MALS)
MALS experiments were performed as described [35]. Briefly, light scattering measurement of purified NPM1 OD alone or combined with US11 FL was performed on a MALS instrument (miniDAWN™ TREOS). For exact protein mass calculation, UV absorptions at 280 nm (Agilent Infinity 1260) and refractive index (RI) signals (OptilabRex, Wyatt Technology) were collected. Raw data was analyzed and processed using ASTRA software (Wyatt Technology) to calculate molecular mass averages and polydispersity indexes of analyzed protein samples.

CIGB-300 synthesis
The CIGB-300 peptide was synthesized at room temperature by manual solid-phase peptide synthesis using a Rink Amide resin (0.59 mmol/g loading). Briefly, the resin (200 μmole scale) was pre-swollen by suspending in 3 mL of NMP for 10 min and the N-terminal Fmoc-protecting group cleaved by treating the resin with 3 mL of a stock solution of 20% piperidine (v/v) in N-methyl-2-pyrrolidone (NMP) (

Fluorescence polarization
Fluoresceinated CIGB-300 (also referred to as FITC-labelled CIGB-300) was synthesized as described above. Increasing amounts of different variants of NPM1, GST-Rev, GST-US11 and GST as a negative control were titrated into FITC-labeled CIGB-300 (0.1 μM) in a buffer containing 30 mM Tris/HCl (pH 7.5), 150 mM NaCl, 5 mM MgCl 2 , 1 mM tris-(2-carboxyethyl) phosphine and a total volume of 200 μl at 25°C using a Fluoromax 4 fluorimeter. Displacement assay was performed by titrating increasing amount of Rev and US11 to the complex of NPM1 and FITC-labelled CIGB-300. The concentration dependent binding curve was fitted using a quadratic ligand binding equation.
Virus production assay

Structural bioinformatics
Model of the complex between NPM1 and CIGB-300 was created in two steps. Tat part of the peptide was first docked to the structure of NPM1 (PDB ID: 4N8M) [36] with the help of Haddock web portal (http://haddocking.org/). Acidic residues on three subunits were defined as active residues for docking while the setup of the Easy interface was used. Docked pose with best score that enables building of cyclic part of the peptide was then used in the second step. Model of the cyclic peptide was first generated and then placed with program CHARMm [37] in different orientations and positions on the surface of NPM1 in a way that enabled its interaction with the Tat portion of the peptide construct. After linking, the geometry of whole complex was optimized by energy minimization applying 500 steps of steepest descent method. Complex with lowest minimized energy was used as a final mode.

HIV-1 Rev directly binds to two distinct regions of NPM1
Previous reports have shown that NPM1 is co-localized and co-immunoprecipitated with HIV-1 Rev in cells [4,38]. To investigate a direct interaction between NPM1 and Rev, pulldown experiments under cell-free conditions were performed using Rev FL and NPM1 variants as GST-fusion proteins. As indicated in Fig 2A (upper panel), Rev FL interacts with NPM1 FL , NPM1 OD , NPM1 HBD and NPM1 HRBD , but not with the NPM1 RBD , suggesting that two different regions of NPM1, namely OD and HBD, have tight physical interaction with the HIV-1 Rev. To show whether this interaction is RNA-dependent, the pull-down experiments were performed under the same conditions in the presence of RNase A. As shown in Fig 2A (lower panel), RNase treatment had no effect on HIV-1 Rev association with NPM1. These results clearly indicate that HIV-1 Rev specifically binds to NPM1, and the binding is not RNAdependent.
Next, we purified all proteins in high quantities (Fig 1B), and after cleaving the tag, isothermal titration calorimetry (ITC) experiments were conducted in order to examine the stoichiometry of binding and to determine the binding affinity of Rev FL for the NPM1 variants. Consistent with the data obtained by pull-down assay, Rev FL revealed variable affinity for the NPM1 variants with calculated dissociation constants (K d ) between 18 and 0.013 μM for 1:1 stoichiometry (Fig 2B and S1 Fig; Table 2). No interaction was detected between Rev FL and NPM RBD (Fig 2C) suggesting that a low micromolar affinity for the interaction between Rev and NPM1 HRBD actually stems from the central histone binding domain of NPM1 (NPM1 HBD ). The obtained dissociation constant (K d ) for the Rev FL and NPM1 HBD interaction was 5.8 μM indicating a stronger affinity for Rev FL as compared to that of NPM HRBD , which could be due to a binding site that partially masked by the C-terminal RBD.

HSV-1 US11 associates with NPM1 in cells
The fact that Rev physically binds to NPM1 and US11 alone can fulfill Rex and Rev's function in transactivating envelope glycoprotein gene expression [14], led us to examine a potential US11-NPM1 interaction. We first analyzed the intracellular distribution of endogenous NPM1 and overexpressed myc-US11 in HeLa cells using confocal imaging. Fig 3A shows a nucleolar co-localization of NPM1 and US11 where the overall pattern of these proteins is different. In contrast to a predominant nucleolar localization of NPM1, US11 was found in the cytoplasm and also accumulated, to certain extent, in the nucleoli. To confirm the association of US11 with NPM1, COS-7 cells overexpressing myc-US11 were lysed and endogenous NPM1 was immunoprecipitated. Fig 3B shows that NPM1 co-precipitated with myc-US11 indicating that US11 forms a complex with NPM1. We, next, used purified GST-NPM1 FL and pulled down myc-US11, transiently overexpressed in COS-7 cells. As shown in Fig 3C, the myc-US11 FL clearly bound to NPM FL , but not to the GST control, indicating that there may be a direct interaction between US11 and NPM1.

US11 associates with NPM1 OD in its oligomeric state
To clarify whether the interaction observed above is a direct interaction, we used purified, RNase A treated NPM1 and US11 variants from E. coli. Fig 4A shows that NPM1 FL and NPM1 OD but not NPM1 HRBD and NPM1 RBD , directly interact with US11 FL . We repeated the experiments to map the NPM1 binding region of US11 by using purified, GST-fused, N-terminal and C-terminal fragments of US11. As shown in Fig 4A, both US11 Cterm and US11 Nterm bound, with the same pattern as US11 FL bound to NPM1 FL and NPM1 OD . However, binding affinities of isolated N-or C-terminal domains of US11 towards NPM1 seemed markedly  reduced compared to the full-length protein. In the light of above mentioned, we conclude that NPM1 and US11 physically interact with each other via NPM1 OD and largely US11 Cterm . Next, ITC measurement was also performed to determine the binding affinity between NPM1 and US11 by titrating NPM1 FL (1.2 mM) to US11 FL solution (60 μM); both proteins were treated with RNase A. As shown in Fig 4B, the association of NPM1 FL with US11 FL is endothermic (positive peaks). As a control experiment, buffer was titrated to 60 μM US11 FL under the same experimental condition with no calorimetric changes (Fig 4C). Based on ITC analysis we estimated an apparent K d value of 4 μM. The NPM1 OD interaction with US11 FL was also analyzed by aSEC combined with MALS, after treating the proteins with RNase A. Fig  4D (lower panel) shows a co-elution of the RNase-treated NPM1 OD and US11 FL proteins from the Superdex 200 (10/300) column indicating that these proteins form a complex. MALS Pull-down experiments were conducted with purified proteins in the presence of RNase A by using GST-fused US11 FL , US11 Nterm , US11 Cterm , and GST as a negative control. For the detection of NPM1 variants two different antibodies were used, ab52644 recognized an N-terminal epitope containing in NPM1 FL and NPM1 OD , and ab10530 recognized a C-terminal epitope containing in NPM1 HRBD and NPM1 RBD . The same pattern of interaction was obtained for the N-terminal and the C-terminal parts of US11, although the interaction between NPM1 FL and NPM1 OD with US11 Nterm was much weaker than with US11 Cterm . The exposure time was 1 min for all the blots. (B-C) US11 binds NPM1 with a binding constant in the low micromolar range. To measure the binding parameter for the NPM1-US11 interaction, 1.2 mM NPM1 FL (B) and buffer (C) were titrated to 60 μM US11 FL . Both NPM1 and US11 were treated with RNase A. Conditions were the same as described in analysis revealed that NPM1 OD oligomerized to a pentameric state and formed a 1:1 complex with the monomeric US11 FL (Fig 4D upper panel). To further investigate the oligomerization states of US11 and NPM1, AUC experiments were performed. Results obtained were consistent with the MALS data, and revealed that NPM1 FL and NPM1 OD are pentameric and globular while US11 FL was monomeric and adopts an elongated structure (Table 3 and S2 Fig). Together, the data clearly demonstrates that US11 selectively binds to the N-terminal oligomerization domain of NPM1 in an RNA-independent manner.

Displacement of the NPM1-CIGB-300 complex by Rev and US11
Synthetic peptide CIGB-300 (also called p15-Tat; Fig 5A) has been described as a proapoptotic and anti-cancer peptide, which directly targets and antagonizes NPM1 function in cancer cells [39,40]. Fluorescence polarization analysis revealed that a FITC-labelled CIGB-300 tightly associates with NPM1 FL and NPM1 OD but not with NPM1 HRBD and NPM1 RBD (Fig 5B). Calculated K d values for the FITC-labelled CIGB-300 interaction with NPM1 FL and NPM1 OD were 1.4 and 6.6 μM, respectively.
We used the NPM1 FL -FITC-labelled CIGB-300 complex to further investigate NPM1 interactions with Rev and US11. The idea here was that titrating Rev or US11 to the complex may result in displacement of NPM1 FL from the FITC-labelled CIGB-300. Fig 5C shows that increasing concentrations of US11, but not Rev, significantly displaced NPM1 FL from the FITC-labelled CIGB-300 complex. This result was surprising for two reasons: First, Rev binds NPM1 in a higher nanomolar range (Table 2) and should be able to compete with CIGB-300 provided that both bind to the same surface of the NPM1 protein. Interestingly, Rev revealed a 30-fold lower affinity for NPM1 FL as compared to NPM1 OD (Table 2), which may explain why Rev did not displace NPM1 FL from FITC-labelled CIGB-300. Second, US11, which evidently exhibits an approximately 10-fold lower binding affinity for NPM1 FL as compared to Rev, is able to displace NPM1 FL from its complex with the synthetic FITC-labelled CIGB-300 ( Fig  5C). To address this issue we repeated the displacement experiments under the same conditions as before but used the FITC-labelled CIGB-300 complex with NPM1 OD instead of NPM1 FL . Data obtained revealed that both Rev and US11 efficiently displace FITC-labelled CIGB-300 by binding to NPM1 OD (Fig 5D), indicating that Rev, US11 and FITC-labelled CIGB-300 have overlapping binding sites on NPM1 OD .
To obtain a first structural assessment of NPM1 OD site targeted by CIGB-300 we conducted a multistage protein-ligand docking approach. Assuming that basic part of CIGB-300 determines the binding, its Tat tail was docked in the first step. In the second step, the cyclic part was placed on the surface of NPM1 OD and linked to the peptide fulfilling geometry and energy criteria. Whole peptide contacted three out of five monomeric units of the pentameric NPM1 OD , but in a way that enables five copies of CIGB-300 to be generated without sterical clashes (Fig 5E). It is important to note that a stoichiometry of 1:1 emerged spontaneously, as the criteria that five peptides should bind to NPM1 OD pentamer was not applied while generating of the model. The feature that CIGB-300 wraps around at least several monomeric units (Fig 5E, middle panel) points to a stabilization effect of bound peptides and is consistent with the model of NPM1 in complex with R-rich proteins, such as p19 ARF , ARF6, Rev and the ribosomal protein L5 [36].

HIV-1 production is influenced in CIGB-300 treated cells
In order to investigate the possible role of NPM1-Rev interaction for HIV-1 replication, HOS-CD4.CXCR4 cells were incubated with CIGB-300 for 30 min or left untreated. After removing the peptide, cells were infected with HIV-1 (clone NL4.3, MOI 1). Culture supernatants were collected 48 and 72 h post infection and were quantified by titration on the HIV-1 reporter cells TZM-bl. In cells treated with CIGB-300, the virus production was reduced by 63% and 70% after 48 h and 72 h post infection, respectively (Fig 6). Thus, CIGB-300 may interfere with an NPM1-Rev interaction in cells and affect Rev-dependent gene expression and subsequently HIV infection.

Discussion
Since its discovery 34 years ago, intensive research has been performed on NPM1. NPM1 is ubiquitously expressed and significantly upregulated in response to cellular stress signals [18,19,41,42] leading to the alteration of nucleolar structures and its re-localization to other cellular compartments. As a global effector, it has been implicated in maintenance of genomic stability, transcriptional gene regulation, ribosome biogenesis, centrosome duplication, DNA repair, control of cellular senescence, protection against radiation-induced apoptosis, tumor suppression, and has been increasingly emerging as a potential cellular factor for viral infection (see Table 1). Most of these functions have hitherto remained obscure and unexplained.
To shed light on the association of NPM1 with viral proteins, we have investigated its physical interaction with HIV-1 protein Rev and HSV-1 protein US11. Based on our results Rev exhibits affinity towards two NPM1 binding sites: on the pentameric, N-terminal oligomerization domain (NPM1 OD ) and on the central histone-binding domain (NPM1 HBD ), while HSV-US11 has only one binding site on NPM1 OD . We suggest that the different NPM1 domains interact in a mechanistically different mode with the Rev and US11 proteins. Rev association with NPM1 is the result of presumably an RNA-independent bimodal binding mechanism, according to our data, of (i) a low-affinity binding to the histone-binding domain of NPM1 (K d = 5.8 μM) and (ii) a very high-affinity binding to oligomerization domain of NPM1 (K d = 0.013 μM), leading to an overall K d value of 0.4 μM for the full-length NPM1 ( Table 2). In the case of the NPM1-US11 interaction, we observed a strong binding of US11 to NPM1 OD , which is most probably achieved via its C-terminal RBD (US11 Cterm ; See Figs 1A and 4A). While the data regarding US11 reports its unprecedented direct interaction with NPM1, our measurements with Rev confirm previously obtained observations. It has been shown that two different transcripts of NPM1, B23.1 and B23.2, prevent the aggregation of Rev via their proposed chaperone activity [43]. B23.1, which was also used in this study, is identical to B23.2 but has a 35-amino acid longer C-terminus. As the prevention of Rev aggregation by both constructs was nearly identical, this C-terminus was excluded from the interaction with Rev [43], which is in agreement with our results from PD and ITC experiments (Fig 2 and S1  Fig; Table 2). Our finding of a 1:1 ratio (n 0.84) between NPM1 and Rev obtained by ITC (Table 2) is also consistent with earlier studies that have suggested a stoichiometric interaction between NPM1 and Rev, and a maximal stimulation of the import of Rev into the nucleus by NPM1 at a 1:1 molar ratio [4,43]. This stoichiometric ratio suggests that NPM1 FL exhibits one binding site for one HIV-1 Rev molecule. Since Rev has the tendency to aggregate also under normal physiological conditions [44], it is very likely that NPM1, by acting as a molecular chaperone, increases Rev's solubility and mobility during the import into and throughout the nucleus.
US11 is an abundant HSV-1 protein, which is expressed late during infection [45]. It has been reported that US11 functionally substitutes Rev and Rex proteins by stimulating expression of glycoproteins required for retroviral envelope synthesis [14]. US11 interaction with cellular proteins may, therefore, be required during HSV-1 infection. However, so far, only a few proteins including 2'-5'-oligoadenylate synthetase [46], cellular kinesin light-chain-related protein PAT1 [45], human ubiquitous kinesin heavy chain [24], protein kinase R (PKR) [47], protein activator of the interferon-induced protein kinase (PACT) [48], and nucleolin [23] have been reported. NPM1 and nucleolin are among the most abundant nucleolar proteins [5] with high functional but not structural similarities. They are usually found in the granular components and dense fibrillar components of nucleoli, have the same distribution as US11 [49], and are re-localized during HSV-1 infection [7,50]. With NPM1, we have identified in this study a new nucleolar protein partner for US11 and characterized the subdomains responsible for their interactions. US11 has two domains ( Fig 1A): An N-terminal domain called effector domain (ED) and a C-terminal RNA-binding domain (RBD). C-terminal domain consisting of 20-24 XPR (X, any amino acid; P, proline; R, arginine) repeats has a polyproline type II helix organization and is usually engaged in interactions with other proteins [15]. US11 ED is necessary for transactivation of gene expression, transport, and mRNA translation [15]. Therefore, we designed two deletion variants of US11 (N-and C-terminus) to determine the part involved in the interaction with NPM1. In contrast to nucleolin, which has been reported to interact with the C-terminus of US11 [23], our data clearly shows that both domains are apparently required for the interaction with NPM1. The C-terminal domain of US11, which is involved in the nucleolar localization of US11, binds to NPM1 stronger than the N-terminal domain (Fig 4A). Since C-terminus of US11 is rich in arginine, these results support the idea that arginine-rich motif (R-rich) mediates the interactions with NPM1 [36]. Synthetic peptide CIGB-300 used in our investigation also falls into this category as it is the conjugate of R-rich peptide Tat, and the cyclic peptide (hence is called p15-Tat; Fig 5A). This peptide, which has been described as a proapoptotic peptide with antiproliferative activity in vitro and antitumoral activity in vivo [51], has been reported to directly bind to NPM1 [39,40]. We observed in this study that only NPM1 OD , but not the other domains of NPM1, associates with fCIGB-300. Interestingly, the K d value for the fCIGB-300 interaction with NPM1 FL , derived from our polarization measurements (Fig 5B), was indicative of almost 5-fold higher affinity than that of fCIGB-300-NPM1 OD interaction. This higher affinity can be explained by an avidity effect that originates from core N-terminal domain and the dynamic flexible tails, similarly to the model proposed for nucleoplasmin interaction with histones [52]. NPM1 OD is followed by the two highly acidic regions with disordered structure and a C-terminal RBD that folds as a threehelix bundle [53]. The biological significance of the acidic regions (A1-A3; Fig 1A) has not been established. The A1 region in NPM1 OD has been recently shown to play a crucial role in the interaction with R-rich motifs of NPM1 binding proteins, such as p19ARF, ARF6, the ribosomal protein L5, and HIV1 Rev [36]. A model of the complex between NPM1 OD and CIGB-300 provided insights into different sites for the association of the CIGB-300 peptide, especially the R-rich motif of the CPP Tat contacting negative charges of the A1 region of NPM1 OD (Figs 1A and 5E). Additionally, our displacement experiment with Rev indicates that CIGB-300 shares the same binding site on NPM1 and may act as an inhibitor of NPM1-Rev interaction. Most likely for the same reason, we observed a reduced expression of viral production in HIV-1 infected cells treated with the CIGB-300 peptide (Fig 6).
Furthermore, our displacement data shows that the NPM1-US11 interaction was also modulated by CIGB-300 (Fig 5C and 5D). Thus, it is tempting to speculate that US11 and Rev, two functionally homologous viral proteins, share a similar binding site on NPM1 as suggested in this study for CIGB-300. An amino acid sequence analysis revealed clear differences in the Rrich motifs between Rev ( 38 RRNRRRRWRARAR 48 ) and US11, which consists of 21`XPRŕ epeat motifs in US11 Cterm . R-rich motifs act as NLS by binding to the nuclear import receptors in nuclear translocation of viral proteins [10,12,54,55]. On the other hand, nucleolar shuttling and accumulation of Rev requires interaction with NPM1 [4,12]. US11 is similarly shuttling between the nucleus and the cytoplasm in transiently transfected cells and HSV-1-infected cells [20,56]. Mutagenesis and modeling studies of the C-terminus of US11, containing XPR repeats, have shown that this region is critical for both nucleolar accumulation of US11 and its nucleocytoplasmic export [15,57]. As mentioned above, CIGB-300 has the cell penetrating peptide Tat with R-rich motif, which corresponds to the presumed nuclear localization signal (NLS). Tat moves across the nuclear envelope and consequently drives CIGB-300 to the nucleus. Thus, we hypothesize that, (i) R-rich motifs of viral proteins serve as NPM1 binding sites that facilitate their nuclear transport analogous to NLS-importin system, and (ii) NPM1 most likely acts as an auxiliary factor for R-rich motif-containing viral proteins, such as HIV-1 Rev and HSV-1 US11, and achieves their transport into different nuclear compartments and subnuclear domains, leading to nuclear egress of infectious viral particles. Thus, NPM1 seems to represent a key protein in viral infections that is hijacked by invading pathogens to facilitate infection. As a consequence, NPM1 may represent a novel promising target for antiviral therapeutic intervention. For presentation, curves had been normalized to maximum peak height. Results revealed that NPM1 FL and US11 FL are pentameric and monomeric, respectively. (B) The left panel contains data obtained from the sedimentation velocity analysis of NPM1 OD , which shows the population of pentamer, and the right panel are data obtained from sedimentation equilibrium analysis of 0.25 μM NPM1 OD at 14000 (purple), 16000 (blue), 25000 (cyan), 42000 (green) and 50000 rpm (yellow) at 20°C. Experimentally determined concentration profiles were fitted globally with a single species model resulting in a molecular mass of 65180 ±640 Da corresponding to a pentamer of NPM1 OD . The experimental data together with the fitted concentration profiles are shown on the top, and at the bottom, residuals from the fit are documented. (TIF)