Molecular anatomy of the subcellular localization and nuclear import mechanism of herpes simplex virus 1 UL6.

As an indispensable structure protein, the herpes simplex virus 1 (HSV-1) UL6 has been described to exert numerous roles in viral proliferation. However, its exact subcellular localization and subcellular transport mechanism is not well known. In the present study, by utilizing confocal fluorescent microscopy, UL6 was shown to mainly locate in the nucleus in enhanced yellow fluorescent protein or Flag tag fused expression plasmid-transfected cells or HSV-1-infected cells, whereas its predicted nuclear localization signal was nonfunctional. In addition, by exploiting dominant negative mutant and inhibitor of different nuclear import receptors, as well as co-immunoprecipitation and RNA interference assays, UL6 was established to interact with importin α1, importin α7 and transportin-1 to mediate its nuclear translocation under the help of Ran-mediated GTP hydrolysis. Accordingly, these results will advance the knowledge of UL6-mediated biological significances in HSV-1 infection cycle.


INTRODUCTION
Herpes simplex virus 1 (HSV-1), a large nuclear duplicating DNA virus, is an epidemic human microbe that can provoke a lytic infection in the mucosal epithelial cells but a life-long latent infection in neurons. As one of the fundamental structure proteins of HSV-1, UL6 has obtained remarkable concern by virtue of its association with numerous viral propagation processes, including establishing the portal for DNA entry into the HSV capsid, cleavage, processing and packaging of replicated viral DNA, assembling of a minor constituent of virions and capsids, and locating on the external surface of the viral capsid [1][2][3][4][5][6].
Besides, recent studies also showed that the tryptophan residues or putative leucine zipper of UL6 is crucial for its association with scaffold proteins, UL15 and UL28 proteins, as well as the incorporation of the portal into capsids [7][8][9][10]. However, the definite function of UL6 is still poorly understood.

Subcellular localization of UL6 in the plasmid transfected and virus infected cells
Protein is the executor of life activity, which need to be transported into certain cell compartments for its execution of specific biological function. UL6 was previously demonstrated to localize in the nucleus in chemical fixed cells [1,4,11,12]. To further detect the subcellular distribution of UL6 in plasmid transfected live cells, enhanced yellow fluorescent protein (EYFP)tagged UL6 and confocal fluorescence microscopy were adapted. Subsequently, plasmid encoding UL6 fused to the C-terminus of EYFP was constructed and transfected into COS-7 cells to test the subcellular localization of UL6, without the presence of other HSV-1 constituents. Although EYFP-UL6 could show cytoplasmic or pan-cellular localization, it largely exhibited nuclear localization ( Figure 1A and Table 1). On the contrary, the fluorescence of vector control EYFP was homogeneously dispersed throughout the cytoplasm and the nucleus in cells transfected with pEYFP-C1 ( Figure 1B and Table 1).
Since EYFP is a relatively considerable tag (~27 kDa), it may alter the nuclear localization of UL6. To avoid this hypothesis, plasmid encoding Flag-tagged UL6 (pCMV-Flag-UL6) was constructed and immunofluorescence assay (IFA) was performed to examine the subcellular localization of the UL6. As shown in Figure 1C and Table 1, Flag-tagged UL6 also localized in the nucleus following formaldehyde-based fixation method.
It is well known that viral protein may show distinct subcellular localization fashions in plasmid transfected and virus infected cells. Therefore, the subcellular localization of UL6 was investigated in HSV-1 infected cells. For this sake, Vero cells were infected with HSV-1 and then IFA was carried out. As a result, UL6 also displayed dominantly nuclear localization when cells were infected at an MOI of 1 at 8 h post-infection ( Figure 1D and Table 1). Accordingly, the above data showed that UL6 localized in the nucleus regardless in live cells or chemical fixed cells, as well as in plasmid transfected cells or HSV-1 infected cells. UL6 is shown to exert certain roles that are generally associated with the nucleus, such as constituting the portal for the access of DNA into the HSV capsid, installing of a minor constituent of virions and capsids, and cleavage, disposal and encasement of duplicated viral DNA [1][2][3][4][5][6][7][8][9]25]. Thus, it is no wonder that UL6 presents primarily nuclear localization.

Identification of the nuclear localization signal of UL6
Nuclear localization signal (NLS), predominantly possessed of basic residues, is vital for the nuclear accumulation of specific protein [26]. Bioinformatics analysis using PSORT II predicted that UL6 contains a potential NLS in the basic residue rich region, namely PILRKRQ at aa171-177 (pat7). However, the potential nuclear export signal of UL6 was not predicted. In order to identify the functional NLS, UL6 was firstly divided into two segments (amino acids (aa) 1-296 and aa297-676) and fused to the C-terminus of EYFP to construct aa1-296-EYFP and aa297-676-EYFP ( Figure 2A). Then, these two plasmids were analyzed in COS-7 cells. As shown in Figure 2B and Table 2, the fluorescence of aa1-296-EYFP showed cytoplasmic localization, whereas aa297-676-EYFP showed pan-cellular distribution, suggesting these two regions may not contain functional NLS. To further explore the functional NLS, plasmids encoding EYFP fused to two diverse segments aa1-177 and aa171-296, which encompass the predicted NLS aa171-177, were constructed ( Figure 2A) and assessed in COS-7 cells. As shown in Figure 2B and Table 2, both of the fluorescence of aa1-177-EYFP and aa171-296-EYFP were similar to that of aa1-296-EYFP, indicating the predicted NLS was non-functional, and the functional NLS of UL6 may be generated by spatial conformation.

Characterization of the nuclear import mechanism of UL6
To date, Ran GTPase is reported to be indispensable for the nuclear transport process of most nuclear target protein [27]. To probe the nuclear import mechanism of UL6, the dominant negative (DN) mutant of RanGTP, with deficiency in GTP hydrolysis (Ran-Q69L) [28], was utilized to inspect whether Ran participates in the nuclear translocation of UL6. Plasmids expressing Ran-Q69L-mCherry and FLAG-UL6 were co-transfected into COS-7 cells, then their subcellular distributions were analysed by IFA. As a result, co-transfection of Ran-Q69L significantly abolished the nuclear accumulation of UL6 ( Figure 3A and Table 3). Considering the evolutionary conserved nuclear pore complex (NPC) only endorses the dispersion of small proteins with approximate molecular masses of 40~60 kDa [29,30], and FLAG-UL6 has a molecular mass of about 76 kDa, it cannot be proposed to export the AGING nucleus by simple dispersion. Consequently, UL6 is a Ran-associated protein and is transported into the nucleus from the cytoplasm through a canonical nuclear transport pathway mediated by GTP hydrolysis.

AGING
co-transfection of Bimax2 and M9M could efficiently diminished the nuclear import of UL6, whereas DN kα1 or DN kβ1 did not obviously lessened the nuclear trafficking of UL6. As negative control, UL6 was not relocalized by mCherry when COS-7 cells were cotransfected with pCMV-Flag-UL6 and mCherry vector ( Figure 3C and Table 3). These data revealed that the nuclear transport of UL6 was mediated by transportin-1, and may be one of the cellular transporters of importin α1, α3, α6 and α7, but not importin α5 or importin β1.

Verification of the nuclear import mechanism of UL6
To finally validate the nuclear import mechanism of UL6, short hairpin RNA (shRNA) expression plasmids were constructed to knock down the expression of importin α1, importin α7 and transportin-1. Compared to the shRNA control vector (shRandom), shImportin-α1, shImportin-α7 and shTransportin-1 could effectively knock down the expression of importin α1, importin α7 and transportin-1, respectively ( Figure 5A), suggesting the related shRNA expression plasmids were successfully AGING constructed. Then, one or two or three plasmids combination of shImportin-α1, shImportin-α7 and shTransportin-1 were co-transfected with pFLAG-UL6 into COS-7 cells and IFA was carried out to analyze whether these shRNA expression plasmids can influence the nuclear import of UL6. As results, the nuclear translocation of UL6 was not obviously affected when one of importin α1, importin α7 and transportin-1, or two of importin α1/importin α7 and importin α1/transportin-1, were knocked down. However, the nuclear trafficking of UL6 was significantly inhibited when importin α7/transportin-1 or importin α1/importin α7/transportin-1 were simultaneously knocked down ( Figure 5 and Table 4), confirming UL6 could be imported into the nucleus via various transport pathways, which was primarily mediated by importin α7 and transportin-1.
As we known, HSV-1 encodes more than 80 structural proteins, some of which need to be transported into the nucleus for their functions execution, such as promoting viral proliferation, restraining host transcription and expression, inhibiting host innate immunity, etc. The nuclear accumulation of these proteins is mediated by one or more different nuclear import receptors, of course including importin α1, importin α7 and transportin-1. In addition, some host proteins also need to be transported into the nucleus by different nuclear import receptors, to perform their corresponding functions. Therefore, it is bound to affect the nuclear accumulation of many proteins of HSV-1 (and host) when the DN mutants of importin α 1, importin α 7 and transportin-1 are transfected into cells or these nuclear import receptors are knocked down by shRNA expression plasmid. Consequently, it is difficult for us to determine whether the reduction of DNA replication, nucleocapsid assembly and virions production of HSV-1 is the direct outcome of the inhibition of UL6 nuclear translocation.  Table 4. AGING Table 4. Verification of the nuclear import mechanism of HSV-1 UL6. shVector, shRandom, one or two or three plasmids of shImportin-α1, shImportin-α7 and shTransportin-1 were co-transfected with FLAG-UL6 into COS-7 cells. 24 h post-transfection, cells were examined for the subcellular localization of UL6 by IFA using confocal fluorescence microscopy.
In conclusion, we had proved that UL6 was a genuine nuclear localization protein. Although the predicted NLS of UL6 was nonfunctional, it was identified to be transported into the nucleus through Ran-, transportin-1-, importin α1and importin α7-dependent nuclear import mechanism, which was largely mediated by the later two nuclear import receptor. These results dissected the molecular determinant for the nuclear transport of UL6, and will shine light for the further study of its biological roles during HSV-1 infection.

Plasmid transfection and fluorescence analysis
Plasmid transfection and fluorescence analysis were carried out, as described in our previous studies [13, 15, 18, 22-24, 41, 45]. Briefly, COS-7 cells were transiently transfected with the indicated plasmid DNA mixed with Thermo Scientific TurboFect Transfection Reagent in line with the manufacturer's instructions. 24 h post-transfection, DAPI staining, which is widely applied in our previous studies of related fluorescent experiments [12-15, 17, 18, 20, 21], was employed to investigate whether the target protein locates in the nucleus or in the cytoplasm. Then, cells were analyze by live cells fluorescence microscopy or IFA, using a laser scanning confocal microscopy (Leica SP8). The image shown represents a great proportion of the cells with homogeneous subcellular distribution. EYFP fusion proteins were shown in pseudocolor green, FITC-labeled proteins and mCherry fusion proteins were shown in their original colors green and red, respectively, and the merged image was presented in yellow signal. All scale bars indicate 10 um, and images were processed using Adobe Photoshop.

Virus infection and IFA
Vero cells infected with HSV-1 (MOI=1) for 8 h were fixed with 4% paraformaldehyde, permeabilized with 0.5% Triton X-100, and stained with the anti-UL6 polyclonal antibody (pAb) [46]. Then, cells were incubated with fluorescein isothiocyanate (FITC)conjugated goat anti-rabbit immunoglobulin G (Zymed Laboratories) and stained with DAPI. Cells were finally detected with a laser scanning confocal microscopy. All scale bars indicate 10 um, and images were processed with Adobe Photoshop.

Co-IP and immunoblotting
Co-IP and immunoblotting (IB) assays were manipulated as described previously [13, 15, 18, 22-24, 41, 47, 48]. Summarily, HEK293T cells were cotransfected with FLAG-or EYFP-tagged expression plasmids for 24 h. Cells were then collected and lysed on ice with 1 mL of lysis buffer. The lysate was subsequently incubated with anti-Flag monoclonal antibody (mAb, Sigma) or nonspecific mouse control antibody (IgG) and a 1:1 slurry of Protein A/G PLUS-Agarose (Santa Cruz Biotechnology) for at least 4 h or overnight at 4 o C. Then, lysis buffer was used to wash beads for three times. Finally, cell lysates and the Co-IPed proteins, were subjected to IB analysis with anti-Flag mAb and anti-YFP pAb (Santa Cruz Biotechnology). All Co-IP were duplicated at least two times, and analogous data were obtained.