Identification and Characterization of the Nucleolar Localization Signal of Autographa californica Multiple Nucleopolyhedrovirus LEF5

Many viruses, including human and plant viruses, target nucleolar functions as part of their infection strategy. However, nucleolar localization for baculovirus proteins has not yet been characterized. In this study, two nucleolar proteins, SfNS and SfFBL, were identified in Sf9 cells. Our results showed that Autographa californica multiple nucleopolyhedrovirus (AcMNPV) infection resulted in redistribution of the nucleoli of infected cells. We demonstrated that AcMNPV late expression factor 5 (LEF5) could localize to the nucleolus and contains a nucleolar localization signal (NoLS), which is important for nucleolar localization of AcMNPV LEF5 and for production of viral progeny and yield of occlusion bodies.

ABSTRACT

Autographa californica multiple nucleopolyhedrovirus (AcMNPV) late expression factor 5 (LEF5) is highly conserved in all sequenced baculovirus genomes and plays an important role in production of infectious viral progeny. In this study, nucleolar localization of AcMNPV LEF5 was characterized. Through transcriptome analysis, we identified two putative nucleolar proteins, Spodoptera frugiperda nucleostemin (SfNS) and fibrillarin (SfFBL), from Sf9 cells. Immunofluorescence analysis demonstrated that SfNS and SfFBL were localized to the nucleolus. AcMNPV infection resulted in reorganization of the nucleoli of infected cells. Colocalization of LEF5 and SfNS showed that AcMNPV LEF5 was localized to the nucleolus in Sf9 cells. Bioinformatic analysis revealed that basic amino acids of LEF5 are enriched at residues 184 to 213 and may contain a nucleolar localization signal (NoLS). Green fluorescent protein (GFP) fused to NoLS of AcMNPV LEF5 localized to the nucleoli of transfected cells. Multiple-point mutation analysis demonstrated that amino acid residues 197 to 204 are important for nucleolar localization of LEF5. To identify whether the NoLS in AcMNPV LEF5 is important for production of viral progeny, a lef5-null AcMNPV bacmid was constructed; several NoLS-mutated LEF5 proteins were reinserted into the lef5-null AcMNPV bacmid with a GFP reporter. The constructs containing point mutations at residues 185 to 189 or 197 to 204 in AcMNPV LEF5 resulted in reduction in production of infectious viral progeny and occlusion body yield in bacmid-transfected cells. Together, these data suggested that AcMNPV LEF5 contains an NoLS, which is important for nucleolar localization of LEF5, progeny production, and occlusion body production.

IMPORTANCE Many viruses, including human and plant viruses, target nucleolar functions as part of their infection strategy. However, nucleolar localization for baculovirus proteins has not yet been characterized. In this study, two nucleolar proteins, SfNS and SfFBL, were identified in Sf9 cells. Our results showed that Autographa californica multiple nucleopolyhedrovirus (AcMNPV) infection resulted in redistribution of the nucleoli of infected cells. We demonstrated that AcMNPV late expression factor 5 (LEF5) could localize to the nucleolus and contains a nucleolar localization signal (NoLS), which is important for nucleolar localization of AcMNPV LEF5 and for production of viral progeny and yield of occlusion bodies.

INTRODUCTION

The family Baculoviridae represents a diverse group of insect-specific DNA viruses. Members of the Baculoviridae are characterized by having large, rod-shaped enveloped viruses with circular double-stranded DNA genomes which vary in size from approximately 80 to 180 kb and contain 90 to 180 open reading frames (ORFs). Baculoviruses are divided into four genera, Alphabaculovirus (lepidopteran baculoviruses), Betabaculovirus (granuloviruses), Deltabaculovirus (dipteran baculoviruses), and Gammabaculovirus (hymenopteran baculoviruses) (1). A hallmark of baculovirus infection is a biphasic replication cycle with production of two morphologically different virion phenotypes: budded virus (BV) early in infection and occlusion-derived virus (ODV) later in infection (2). BV is responsible for cell-to-cell transmission within a host, while ODVs spread virus infection from insect to insect.

The nucleolus is a non-membrane-bound subnuclear organelle responsible for mainly ribosome subunit biogenesis (3–5). The subnuclear compartment is composed of at least three major regions, the fibrillar center (FC), the dense fibrillar component (DFC), and the granular component (GC). The nucleolus consists of many different components, including nucleolin, fibrillarin (FBL), nucleostemin (NS), and B23. The most abundant and well-understood proteins in the nucleolus are nucleolin, fibrillarin, and B23. Nucleolin accounts for approximately 10% of the total nucleolar protein content. This protein is highly phosphorylated and methylated. Many functions, including the first cleavage step of rRNA, have been ascribed to this protein (6). Fibrillarin is highly conserved in sequence, structure, and function in eukaryotes, and fibrillarin is involved in many posttranscriptional processes, including pre-rRNA processing, pre-rRNA methylation, and ribosome assembly (7). B23 is widely distributed in different species and contains two isoforms. The major isoform, B23.1, is located in the nucleolus and the minor isoform, B23.2, is situated in the cytoplasm. B23 has many functions in ribosome assembly, binding to nucleolar proteins, and nucleocytoplasmic shuttling (8, 9). NS was discovered as a GTP-binding protein in the nucleoli of neural stem cells and plays important roles in stem cell proliferation, cell cycle maintenance, ribosome biogenesis, and embryogenesis (10).

Many viruses, including human and plant viruses, target nucleolar functions as part of their infection strategy. Viral infection could result in changes to nucleolar morphology and the proteome in infected cells (11). Many studies have shown that nucleolar localization plays important roles for human viruses. For example, the pseudorabies virus (PRV) early protein UL54 contains a nucleolar localization signal (NoLS), and the mutation or deletion of the nucleolar localization sequence resulted in severe defects in viral gene expression and DNA synthesis (12). Schmallenberg virus (SBV), a trisegmented negative-strand RNA virus, encoded a nonstructural major virulence factor, which is targeted to mainly the nucleolus through an NoLS, and B23 undergoes a nucleolus-to-nucleoplasm redistribution in SBV-infected cells (13). Assembly-activating protein (AAP) of adeno-associated virus serotype 2 (AAV2) is a nucleolus-localizing protein that plays a critical role in transporting viral capsid VP3 protein to the nucleoli of infected cells for assembly. The AAP NoLS mutations not only resulted in aberrant intracellular localization but also attenuated AAP2 protein expression to various degrees, and both of these abnormalities have a significant negative impact on capsid production (14). Nucleolar localization is also important in plant viruses. Potato mop-top virus (PMTV) triple gene block1 (TGB1) movement protein is required for viral systemic infection. Importin-α-mediated nucleolar targeting of PMTV TGB1 is a crucial step in establishing the efficient systemic infection of the entire plant (15). Citrus tristeza virus (CTV) encodes a single multifunctional protein (p23), which accumulated preferentially in the nucleolus and functions in suppression of RNA silencing. Deletions or substitutions in p23 NoLS have important effects on suppressing RNA silencing in Nicotiana benthamiana and in eliciting a pathogenic reaction (16). Nuclear inclusion protein a (NIa) of potato virus A (PVA) has multiple functions and accumulates in the nuclei of virus-infected cells. The regions in the viral genome-linked protein (VPg) domain were found to constitute both a nuclear localization signal (NLS) sequence and NoLS sequences. The mutations in the VPg NoLS region reduced multiplication and virulence of PVA (17).

The lef5 gene is present in all baculovirus genomes sequenced to date (18–20). The lef5 gene is thought to be involved directly in baculovirus late gene transcription (21). LEF5 protein from Spodoptera exigua multiple nucleopolyhedrovirus (SeMNPV) can substitute for the Autographa californica multiple nucleopolyhedrovirus (AcMNPV) LEF5 protein in transient late expression assays (22). lef5 knockout in AcMNPV resulted in abolishment of progeny virus production and caused a drastic decrease in the expression of the late genes. In vitro experiments indicated that LEF5 functions as an initiation factor for late and very late gene transcription (23). Expression of the lef5 gene from SeMNPV contributes to baculovirus stability in cell culture (24). Colocalization of enhanced green fluorescent protein (EGFP)-LEF5 and cMyc-LEF5 demonstrated that LEF5 could localize to the nucleus (25). However, the specific intranuclear localization, such as in the nucleolus, has not yet been determined.

In the present study, we identified two nucleolar proteins in Sf9 cells, Spodoptera frugiperda nucleostemin (SfNS) and fibrillarin (SfFBL), and showed that AcMNPV infection resulted in redistribution of the nucleoli of infected cells. Here we report that AcMNPV LEF5 accumulates in the nucleolus. We identified an NoLS between amino acids (aa) 184 and 213 of AcMNPV LEF5, which are required for LEF5 nucleolar localization. Multiple point substitutions in the region of LEF5 NoLS abrogated nucleolar localization. We also constructed a lef5 knockout AcMNPV bacmid, and different repair bacmids were generated by transposing the lef5 wild-type (WT) gene or ones containing the point mutations. Our results demonstrated that NoLS of AcMNPV LEF5 is important for nucleolar localization, budded-virus yield, and occlusion body (OB) production.

RESULTS

Identification of nucleolar proteins in Sf9 cells.

The nucleolus is composed of three distinct subregions termed FC, DFC, and GC in eukaryotic cells. Human NS (or GNL3) is enriched in the granular component, while fibrillarin is located in the DFC and considered a typical marker for the subnuclear compartment (26). To identify NS and FBL of Sf9 cells, cells were infected with wild-type AcMNPV (E2 strain) at a multiplicity of infection (MOI) of 10. RNAs were isolated from infected Sf9 cells or from control (mock-infected) cells at 12 h postinfection (hpi). In total we collected 345,567,364 raw reads. After removing rRNA and low-quality reads, we obtained 290,846,932 high-quality reads that were subsequently used to assemble an Sf9 transcriptome, which contains 116,261 transcripts with an N₅₀ value of 2,045 nucleotides (nt) and a mean length of 850 nt, the unigenes with lengths longer than 200 bp were annotated. From the annotated unigenes, we found an NS homolog (SfNS) and an FBL homolog (SfFBL) from Sf9 cells. SfNS (GenBank accession number GHKU01024144.1) showed 34% amino acid identity to the human NS (Fig. 1A), while there was 79% amino acid identity between SfFBL (GenBank accession number GHKU01017105.1) and human FBL (Fig. 1B). We then PCR amplified Sfns and Sffbl genes from cDNA of Sf9 cells. Amino acid sequence analysis showed that Sfns contains a 1,743-bp ORF encoding a 580-amino-acid protein with a predicted molecular mass of 65.29 kDa, and the Sffbl contains a 963-bp ORF encoding a 320-amino-acid protein with a predicted molecular mass of 33.01 kDa. The transcripts of ns and fbl were upregulated early in infection and downregulated later in infection in AcMNPV-infected Trichoplusia ni cells (27). We searched for relevant domains of SfNS and SfFBL using the Conserved Domain Database (CDD) (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi); the results revealed that the SfNS protein consists of a basic domain at the N terminus, coiled-coil domain, GTP-binding motif, putative RNA-binding domain, and acidic domain near the C terminus, while SfFBL contains three major structural domains arranged from an N-terminal glycine- and arginine-rich (GAR) domain, central domain with presumed RNA-binding capacity, and a C-terminal α-helical domain, which are typical characteristics for NS and FBL, respectively.

FIG 1.

Identification and nucleolar localization of NS and FBL in Spodoptera frugiperda cells. (A) Schematic diagrams of NS proteins from Homo sapiens and S. frugiperda. B, basic domain; C, coiled-coil structure; GTP, GTP-binding motifs; R, RNA-binding motif; A, acidic domain. Black horizontal lines above the schematic indicate nucleolar localization signal. (B) Domain structure of FBL from H. sapiens and S. frugiperda. GAR, glycine- and arginine-rich domain; RBD, RNA-binding domain; α, α-helical domain. The black horizontal line above the schematic indicates nucleolar localization signal. (C) Subcellular distributions of NS and FBL proteins. Sf9 cells were transfected with plasmid pBlue-NS:GFP, expressing green NS:GFP. At 24 hpt, cells were treated with mouse monoclonal anti-fibrillarin antibody and then incubated with Alexa Fluor 594-conjugated goat anti-mouse antibody. Nuclei were stained with Hoechst 33342. Photographs were taken by using light microscopy (DIC), GFP fluorescence microscopy (NS:GFP), and red FBL fluorescence or blue Hoechst staining (Hoechst). Green, red, and blue fluorescence images were merged in the panel labeled “N+F+H,” and light, green, red, and blue fluorescence images were merged in the column labeled “D+N+F+H.” C, cytoplasm; N, nucleus; No, nucleolus; DFC, dense fibrillar component of the nucleolus.

To identify whether both putative proteins have a nucleolar localization, a transient expression vector was constructed. Sfns fused to gfp was cloned into pBluescript, generating plasmid pBlue-NS:GFP, where Sfns was expressed under the control of the AcMNPV ie-1 promoter and simian virus 40 (SV40) poly(A) signal. Sf9 cells were transfected with pBlue-NS:GFP. The native protein of Sf9 cells, SfFBL, the homolog of DFC marker protein FBL, was monitored by immunofluorescence using mouse monoclonal anti-FBL antibody conjugated with the red Alexa Fluor 594 goat anti-mouse antibody, and SfNS was observed by green GFP fluorescence. By 24 h posttransfection (hpt), red immunofluorescence of SfFBL was observed in numerous irregularly shaped bodies inside the nucleus, while green fluorescence of NS:GFP was found to embrace the FBL-positive DFC structures (Fig. 1C), which matches the morphology of subnucleolar compartments in other species, such as Drosophila melanogaster and Homo sapiens (26, 28). These results indicated that SfNS and SfFBL were localized to the nucleolus and that they are nucleolar proteins of Sf9 cells.

AcMNPV infection resulted in redistribution of nucleolar proteins.

Viral transcription, DNA replication, viral genome packaging, and progeny capsid assembly occur in the nuclei of infected cells, suggesting that baculovirus infection may be involved in a subnuclear organelle. To determine if AcMNPV infection affects nucleolar organization, we used nucleolar proteins, SfNS and SfFBL, to track changes in the nucleoli of infected cells. Sfns fused to gfp was inserted into the polyhedrin locus of the AcMNPV bacmid, creating Bac-NS:GFP. The polyhedrin (polh) gene was also introduced into the same locus by transposition. Sfns fused to gfp was expressed under the control of the AcMNPV ie-1 promoter and the SV40 poly(A) site. All constructs were confirmed by PCR analysis.

Sf9 cells were infected with Bac-NS:GFP at an MOI of 5. Intracellular localization of SfNS was observed by green GFP fluorescence, and SfFBL of Sf9 cells was identified by red immunofluorescence. Colocalization of NS:GFP and FBL was detected by confocal fluorescence microscopy at 4 hpi (Fig. 2A), and both proteins displayed their typical nucleolar pattern even up to 12 hpi (Fig. 2A). However, the appearance of the NS and FBL localizations changed beginning at 18 hpi. Green NS:GFP fluorescence was enriched in the approximate middle of the nuclei, while colocalization of green NS:GFP fluorescence and red FBL immunofluorescence was observed in the nuclear periphery of infected cells even up to 48 hpi (Fig. 2A). These results indicated that AcMNPV infection did not affect nucleolus morphology integrity at the early stage of infection but resulted in nucleolar reorganization in infected cells late in infection.

FIG 2.

Effect of AcMNPV infection on nucleolar integrity. (A) Nucleolar integrity analysis during AcMNPV infection. Sf9 cells were infected by Bac-NS:GFP at an MOI of 5 and observed by confocal microscopy at the indicated hours postinfection. Photographs were taken by using light microscopy (DIC), GFP fluorescence microscopy (NS:GFP), red immunofluorescence (FBL), or blue nuclear staining (Hoechst). Green, red, and blue fluorescence images were merged in the column labeled “N+F+H,” and light, green, red, and blue fluorescence images were merged in the column labeled “D+N+F+H.” (B) Steady-state protein levels of FBL during AcMNPV infection. Sf9 cells were infected by WT AcMNPV at an MOI of 5. At the indicated times postinfection, Sf9 cells were harvested for FBL detection using anti-fibrillarin antibody in the presence (+) or absence (−) of AcMNPV infection. β-Actin was used as a control for protein loading.

We also performed a Western immunoblot analysis to assess the effect of nucleolus redistribution on nucleolar protein levels during AcMNPV infection. Sf9 cells were infected with WT AcMNPV at an MOI of 5, and infected cells were harvested at different times. Nucleolar protein SfFBL was detected by using anti-fibrillarin antibody. The results showed that FBL was stable during AcMNPV infection, and the steady-state protein level was equivalent to that of mock-infected cells at all considered time points (Fig. 2B). These data suggested that although AcMNPV infection altered nucleolus integrity, nucleolar protein stability was not affected.

Subcellular localization of AcMNPV LEF5.

Previous results showed that AcMNPV LEF5 localizes to the nucleus (25), and it is thought to be involved directly in baculovirus late gene transcription (21). To better define the intranuclear localization of AcMNPV LEF5 in the nucleolus during virus infection, a bacmid expressing LEF5 fused to a hemagglutinin (HA) tag was constructed. We first generated a lef5-null mutant bacmid, Bac-LEF5^KO (Fig. 3A). Bac-LEF5^KO (lacking lef5) was then used as the backbone to reinsert (repair) the lef5 gene under the control of the native lef5 promoter, and the poly(A) site was cloned into the polyhedrin locus of the bacmid, generating Bac-NS:mCherry-LEF5^HA (Fig. 3B). The polyhedrin and ns:mCherry genes were also introduced by transposition into the same locus. The constructs were confirmed by PCR analysis. Sf9 cells were infected with Bac-NS:mCherry-LEF5^HA at an MOI of 5. Confocal fluorescence microscopy showed that the green immunofluorescence of AcMNPV LEF5 fused to HA was first visible at 18 hpi but only in the nucleus; at the same time, some nucleoli could be seen as NS:mCherry puncta (Fig. 3C). Green immunofluorescence of AcMNPV LEF5 could be observed mainly in the nucleus, with colocalization of red fluorescence SfNS at 24 and 48 hpi (Fig. 3C). SfFBL was also used for colocalization analysis of AcMNPV LEF5 by immunofluorescence microscopy with antibody to FBL; similar results were obtained for subcellular localization of AcMNPV LEF5 (Fig. 3D).

FIG 3.

Construction of lef5 knockout AcMNPV bacmids, schematic of the WT and mutant lef5 repair viruses, and AcMNPV LEF5 localization analysis. (A) Strategy for construction of a lef5 knockout bacmid containing a deletion of the AcMNPV lef5 gene by recombination in E. coli. An internal 315-bp portion of a lef5 ORF was deleted and replaced with the chloramphenicol acetyltransferase (CAT) gene. (B) Schematic of the indicated constructs showing the polyhedrin (polh) and ns:mCherry/gfp genes inserted into the polyhedrin locus by Tn7-mediated transposition. (C) Colocalization analysis of LEF5 and SfNS during AcMNPV infection. Sf9 cells were infected with Bac-NS:mCherry-LEF5^HA at an MOI of 5. At the indicated hours postinfection, cells were treated with mouse monoclonal anti-HA antibody and then incubated with Alexa Fluor 488-conjugated goat anti-mouse antibody for the detection of HA-tagged LEF5. (D) Colocalization analysis of LEF5 and SfFBL during AcMNPV infection. Sf9 cells were infected with Bac-LEF5^HA at an MOI of 5. At the indicated hours postinfection, LEF5 and SfFBL were detected by immunofluorescence microscopy using rabbit polyclonal anti-HA antibody and mouse monoclonal anti-fibrillarin antibody, respectively. Photographs were taken by using light microscopy (DIC), GFP immunofluorescence microscopy (LEF5:HA), red fluorescence microscopy (NS:mCherry or FBL), or blue nuclear staining (Hoechst). Green, red, and blue fluorescence images were merged in the column labeled “L+N+H” (or “L+F+H”), and light, green, red, and blue fluorescence images were merged in the column labeled “D+L+N+H” (or “D+L+F+H”).

Virus infection could result in drastic morphological and structural changes of nucleoli (29), which may affect localization of LEF5. To investigate the subcellular distribution of LEF5 in the absence of infection, AcMNPV LEF5 fused to HA was cloned into pBlue, creating pBlue-HA:LEF5, which was transfected into Sf9 cells in the absence of virus infection. Using anti-HA antibody conjugated with Alexa Fluor 488 goat anti mouse antibody and confocal fluorescence microscopy, green immunofluorescence was observed for pBlue-HA:LEF5 at 24 hpt, indicating that this fused protein was expressed. Colocalization of HA:LEF5 and NS:mCherry showed that LEF5 had significant nucleolar enrichment in a few nucleolar clear foci (Fig. 4). Furthermore, when only pBlue-HA:LEF5 was transfected into Sf9 cells but in the presence of WT AcMNPV infection, green immunofluorescence of LEF5 fused to HA was detected mainly in the nucleus, with nuclear colocalization of red fluorescence SfNS (Fig. 4). Thus, LEF5 showed nucleolar localization in independent pBlue-HA:LEF5-transfected cells but more diffuse nuclear localization for pBlue-HA:LEF5 in the presence of WT AcMNPV infection at 24 hpi. Together, these results demonstrated that LEF5 could translocate to the nucleolus in plasmid-transfected cells, but these data also indicated that AcMNPV infection resulted in nucleolar dispersal in cells late in infection.

FIG 4.

Colocalization of LEF5 and NS. Plasmids pBlue-HA:LEF5 and pBlue-NS:mCherry were cotransfected into Sf9 cells. The plasmids were transfected in the absence or presence of WT AcMNPV infection. By 24 hpt, cells were treated with mouse monoclonal anti-HA antibody and then incubated with Alexa Fluor 488-conjugated goat anti-mouse antibody for the detection of HA-tagged LEF5. Photographs were taken by using light microscopy (DIC), GFP immunofluorescence microscopy (LEF5:HA), red fluorescence microscopy (NS:mCherry), or blue nuclear staining (Hoechst). Green, red, and blue fluorescence images were merged in the column labeled “L+N+H,” and light, green, red, and blue fluorescence images were merged in the column labeled “D+L+N+H.”

Bioinformatic analysis of AcMNPV and other baculovirus LEF5 NoLSs.

LEF5 could localize to the nucleolus on its own, indicating that LEF5 may contain an NoLS or another nucleolar targeting signal. To identify a possible sequence for nucleolar localization, we searched for the presence of a possible NoLS or relevant domains in LEF5 using an online resource (NoD from http://www.compbio.dundee.ac.uk). The results showed a putative NoLS between amino acids 184 and 213 (LKKKEKHTSTGCTRKKKIKHRQILNDKVIY) of AcMNPV LEF5 (Fig. 5A). This region shares features characteristic of NoLSs, including a high proportion of basic amino acids and proximity to the C terminus of AcMNPV LEF5.

FIG 5.

Nucleolar localization analysis using pBlue transient-expression plasmids. (A) Sequence analysis of AcMNPV LEF5. Secondary structures, α-helix and β-sheet, are predicted by PSIPRED (http://bioinf.cs.ucl.ac.uk/psipred). Domain compositions of LEF5 are recognized by SMART (http://smart.embl-heidelberg.de/). Baculo_LEF5 domain, Baculoviridae late expression factor 5 domain, amino acids 29 to 187 (Pfam entry PF04838); Baculo_LEF5_C domain, Baculoviridae late expression factor 5 C-terminal domain, amino acids 221 to 262 (Pfam entry PF11792). (B) GFP localization analysis of LEF5 putative NoLSs. Sf9 cells were cotransfected with pBlue-NS:mCherry and one of plasmids pBlue-GFP:LEF5^184-195, pBlue-GFP:LEF5^196-213, and pBlue-GFP:LEF5^184-213. Plasmid pBlue-GFP was used as a control. Photographs were taken at 24 hpt by using light microscopy (DIC), GFP fluorescence microscopy (GFP), red fluorescence microscopy (NS:mCherry), or blue nuclear staining (Hoechst). Green, red, and blue fluorescence images were merged in the column labeled “G+N+H,” and light, green, red, and blue fluorescence images were merged in the column labeled “D+G+N+H.” White arrowheads indicate location of NS:mCherry-positive nucleoli.

To determine whether NoLS is conserved, we also analyzed other baculovirus LEF5 proteins. The results showed that the LEF5 proteins of all group I and II alphabaculoviruses contain at least one NoLS, and these NoLSs are distributed in the LEF5 center and/or C terminus (see Table S1 in the supplemental material). Nine betabaculoviruses contain a single NoLS in LEF5, yet the LEF5 proteins of another 14 betabaculoviruses did not contain easily identifiable NoLSs (Table S1). LEF5 from the deltabaculovirus did not show any evidence of an NoLS. Among the gammabaculoviruses, only Neodiprion sertifer NPV contains two NoLSs in LEF5, while the NoLSs were not identified in the LEF5 proteins from another two gammabaculoviruses (Table S1). These data demonstrated that the LEF5 proteins of many baculoviruses in all four genera contain NoLS and that NoLS is conserved in LEF5 proteins of at least group I and II alphabaculoviruses, indicating the NoLS is important for the function of LEF5.

Functional identification of AcMNPV LEF5 NoLS.

The prediction of an NoLS sequence in the AcMNPV LEF5 does not prove that it functions as an NoLS; it may simply be a nonfunctional sequence that matches such a consensus. Thus, it was important to demonstrate that the putative NoLS sequence is functional as an NoLS. To confirm if the AcMNPV LEF5 30-amino-acid sequence mediated nucleolar localization, we constructed a GFP-fused peptide encompassing LEF5 NoLS at aa 184 to 213, creating the expression plasmid pBlue-GFP:LEF5^184-213 (LKKKEKHTSTGCTRKKKIKHRQILNDKVIY). To identify the requirement of the basic amino acid regions of putative NoLS for nucleolar localization, we also constructed GFP tag peptides representing the basic amino acid regions at aa 184 to 195 (LKKKEKHTSTGC) and aa 196 to 213 (TRKKKIKHRQILNDKVIY), resulting in expression plasmids pBlue-GFP:LEF5^184-195 and pBlue-GFP:LEF5^196-213 (Fig. 5B). The expression plasmids described above were transfected into Sf9 cells. pBlue-GFP was used as a GFP control. By 24 hpt, an obvious nucleolar fluorescent signal was observed for pBlue-GFP:LEF5^184-213 (Fig. 5B). However, a more diffuse signal was observed in both the cytoplasm and nucleus in Sf9 cells for pBlue-GFP, pBlue-GFP:LEF5^184-195, and pBlue-GFP:LEF5^196-213 (Fig. 5B). Compared to GFP alone or GFP tag peptides at the basic amino acid regions, the GFP-tagged peptide from aa 184 to 213 colocalized with NS:mCherry red fluorescence. These results suggested that residues 184 to 213 allowed for efficient nucleolar localization of GFP, while shorter peptides including only basic amino acid regions at residues 184 to 195 or 196 to 213 did not allow for nucleolar localization of GFP-tagged fusions.

To investigate whether basic amino acids in the putative NoLS motif of LEF5 are important for nucleolar localization, we generated multiple point mutations within the basic stretches. We replaced the cluster of the basic amino acids Lys (K) and Arg (R) in the motif with alanines (A) in specific regions, resulting in pBlue-HA:LEF5^Mut185-189, pBlue-HA:LEF5^Mut197-200, and pBlue-HA:LEF5^Mut202-204 (Fig. 6A). These resultant plasmids were transfected into Sf9 cells, and pBlue-HA:LEF5 was used as a positive control for nucleolar localization. Using confocal fluorescence microscopy, green fluorescence was observed in the nucleus, with strong accumulation in the nucleolus for pBlue-HA:LEF5 and pBlue-HA:LEF5^Mut185-189 by 24 hpt (Fig. 6B and C), indicating that the proteins were expressed and could localize to the nuclei and nucleoli of transfected cells. However, since the green fluorescence foci were not as intense, nucleolar association appeared to be reduced for pBlue-HA:LEF5^Mut197-200 and pBlue-HA:LEF5^Mut202-204 (Fig. 6B and C), suggesting that these mutations in NoLS of AcMNPV LEF5 proteins resulted in a decrease in nucleolar localization. Plasmid pBlue-HA:LEF5^Mut197-204 was generated by mutation of positively charged amino acids from LEF5 at residues for both 197 to 200 and 202 to 204 at once. When Sf9 cells were transfected with pBlue-HA:LEF5^Mut197-204, green immunofluorescence was found to localize exclusively to the nucleus, with little or no nucleolar localization, by 24 hpt (Fig. 6B and C). These results demonstrated that AcMNPV LEF5 residues 197 to 204 play an important role in nucleolar but not nuclear targeting.

FIG 6.

Mutation analysis of AcMNPV LEF5 NoLS. (A) Mutational sites of NoLS basic amino acids in LEF5 tagged with HA. (B) Nucleolar localization analysis of mutated LEF5. Sf9 cells were transfected with the indicated plasmids. Subcellular localization of wild-type or mutated LEF5 was monitored through Alexa Fluor 488-labeled (green) HA tags. Green (LEF5:HA) and blue (Hoechst) fluorescence images were merged in the column labeled “L+H,” and light (DIC), green, and blue fluorescence images were merged in the column labeled “D+L+H.” Images of cells that are representative of the entire population are shown. (C) Graphs showing the mean values ± the SDs from the experiment whose results are shown in panel B. The results are presented as percentage of cells with LEF5 localizing in nucleolus and nucleus in comparison with the total number of fluorescent cells counted under confocal microscopy. At least 200 cells from each transfection were scored. Bars in the panels represent SDs determined from three independent replicates. Statistical analysis was performed using Student’s t test.

Effects of NoLS mutations on infectious virus yield and occlusion body production.

To test if the NoLS sequence of AcMNPV LEF5 is important for BV and ODV production, all amino acid residues of each motif of the NoLS were replaced by alanine. The wild-type and alanine-substituted LEF5 proteins were then reinserted into Bac-LEF5^KO, resulting in Bac-GFP-LEF5^REP, Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204. polyhedrin and gfp were also cloned into the same loci of the bacmids (Fig. 3B). Bac-GFP-LEF5^REP was used as a positive control and Bac-GFP-LEF5^KO, the AcMNPV lef5 knockout bacmid with the reporter egfp, was a negative control.

Sf9 cells were transfected with the bacmid DNAs described above, and viral replication and spread were monitored by GFP fluorescence. There was no difference in the relative number of GFP-positive cells for all transfected monolayers at 24 hpt, indicating equivalent transfection efficiencies (Fig. 7A). GFP fluorescence was observed to spread to neighboring cells at 48 hpt and almost all the cells for Bac-GFP-LEF5^REP by 72 hpt. There was an increased spread of fluorescence to adjacent cells at 72 hpt for Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204 but lower levels for Bac-GFP-LEF5^Mut185-189 (Fig. 7A). By 120 hpt, GFP fluorescence spread completely from the initially transfected cells to adjacent cells for Bac-GFP-LEF5^REP, showing viral spread identical to that of the lef5 repair bacmid (25). There was some compromise in viral spread for Bac-GFP-LEF5^Mut185-189 at 120 hpt. In contrast, the number of GFP-positive cells did not increase for Bac-GFP-LEF5^KO up to 120 hpt or even at 144 hpt, indicating that there was no spread of the virus from the initially transfected cells (Fig. 7A). Occlusion bodies (OBs) were detected microscopically in 0%, 42.8%, 49.2%, and 4.7% of the Sf9 cells transfected with the Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204 at 120 hpt, respectively. However, OBs were observed in 79.8% of Bac-GFP-LEF5^REP-transfected cells (Fig. 7D). Viral spread and polyhedron formation for the LEF5:HA fusions showed the same results as LEF5 expressed independently, and LEF5 proteins were expressed in bacmid-transfected cells (Fig. 7B and C). These results suggested that the mutations of basic amino acids in LEF5 NoLS motifs affected virus spread and OB production.

FIG 7.

GFP fluorescence and virus production of bacmids with mutations in NoLS regions of AcMNPV LEF5. (A) Green GFP fluorescence images of independent LEF5. Monolayers of cells were transfected with the indicated constructs at 24, 72, and 120 hpt. (B) Green GFP fluorescence images of LEF5 fused to HA. Monolayers of cells were transfected with the indicated constructs at 24, 72, and 120 hpt. (C) Analysis of intracellular LEF5 in cells transfected with the indicated constructs. Sf9 cells were harvested at 36 hpt, and lysates at equivalent total protein levels were subjected to immunoblot analysis using anti-HA antibody with β-actin as a control for protein loading. (D) OB production of cells transfected with the indicated constructs at 120 hpt. OB production was quantified by counting the cells containing one or more OBs. The percentage of cells containing OBs was estimated by counting the number of cells containing OBs and the total number of cells in a measured area of an ocular on Nikon inverted microscope. (E) Virus growth curves. Cells were transfected with 2 μg of bacmid DNAs, and the resulting supernatants were used to determine the yield of infectious BV. (F) Kinetics of viral DNA accumulation of the indicated constructs in Sf9 cells. Total DNA was harvested from transfected cells at 24 hpt, and viral DNA copy numbers were determined by qPCR. (G) Quantitative analysis of viral gene transcription. Sf9 cells were infected with the indicated constructs at an MOI of 1 and harvested at 24 hpi. The transcript levels were normalized to that of the host 18S rRNA transcripts and are shown as the percentages of the corresponding genes in cells infected with Bac-GFP-LEF5^REP. Bars represent SDs determined from three independent replicates. Statistical analysis was performed using Student’s t test.

To determine the effect of mutations of basic amino acids in NoLS on BV production, viral growth curves were used to monitor infectious virus production. Cells were transfected with 2 μg of the bacmid DNAs described above. The viral growth curves showed a steady increase in virus production for Bac-GFP-LEF5^REP following transfection, while virus yields for Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204 were lower than for Bac-GFP-LEF5^REP at 120 hpt (Fig. 7E). There was no increase in any detectable levels up to 120 hpt for Bac-GFP-LEF5^KO. The final virus titers at 120 hpt for the four mutants, Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204, were 5.65 logs, 8.00 logs, 8.09 logs, and 7.12 logs, compared to 8.33 logs for Bac-GFP-LEF5^REP, respectively, while no virus was detected for Bac-GFP-LEF5^KO (Fig. 7E). There were 478.6-fold, 2.1-fold, 1.7-fold, and 16.2-fold reductions in virus yield for Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204, respectively. These results suggested that the mutations in LEF5 NoLS motif resulted in reduced levels of infectious virus production.

Quantitative PCR was also performed to monitor the accumulation of intracellular total viral DNA in transfected cells. The levels of viral DNA accumulation were determined over the first 24 hpt in cells transfected with the bacmid DNAs described above. The intracellular DNA accumulation increased 52.1-fold, 39.7-fold, 43.9-fold, 32.1-fold, and 31.0-fold for Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, Bac-GFP-LEF5^Mut197-204, and Bac-GFP-LEF5^KO over 24 h, respectively, while the levels of intracellular viral DNA increased about 48.6-fold for Bac-GFP-LEF5^REP over the same time course (Fig. 7F). These results are not significantly different from each other and indicated that LEF5 containing an NoLS is not required for efficient viral DNA synthesis.

Effects of NoLS mutations on expression of late and very late genes.

LEF5 functions as an initiation factor for late and very late gene transcription (23), which suggested that NoLS mutations may affect expression of late or very late genes. Thus, it is important to identify transcription levels of late and very late genes for the viruses containing point mutations of LEF5. Cells were infected with the repair viruses containing point mutations of LEF5, and quantitative reverse transcription-PCR (qRT-PCR) was used to analyze the transcription kinetics of late and very late genes in the repair viruses described above. Five representative viral genes, vp39, 38k, p6.9, polh, and p10, were selected for determination of transcription levels. vp39, 38k, and p6.9 are late genes, while polh and p10 are very late genes. The transcription levels of vp39, 38k, p6.9, polh, and p10 for Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204 showed obvious reduction compared to those for Bac-GFP-LEF5^REP (Fig. 7G). By 24 hpi, the transcription levels of vp39 for Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204 were reduced to 4%, 31%, 48%, and 15% of that for Bac-GFP-LEF5^REP, respectively (Fig. 7G). The transcripts of 38k for Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204 were decreased to 16%, 62%, 57%, and 36% of that for Bac-GFP-LEF5^REP, respectively (Fig. 7G). The transcription levels of p6.9 for Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204 were reduced to 3%, 38%, 49%, and 14% of that for Bac-GFP-LEF5^REP, respectively (Fig. 7G). The transcripts of polh for Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204 were decreased to 7%, 41%, 53%, and 21% of that for Bac-GFP-LEF5^REP, respectively (Fig. 7G). The transcription levels of p10 for Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204 were reduced to 17%, 38%, 39%, and 29% of that for Bac-GFP-LEF5^REP, respectively (Fig. 7G). These results revealed that the mutations of NoLS resulted in significant decrease of transcriptions of late and very late genes.

DISCUSSION

Subcellular localization of protein is important for understanding its biological role. In this study, we characterized nucleolar localization of AcMNPV LEF5 and identified its NoLS. In preliminary studies we tried to map the NoLS sequence by truncating LEF5; unfortunately, truncated LEF5 was degraded in transfected Sf9 cells (data not shown). Then we used human NoLS predictor to search for possible NoLS sequences in AcMNPV LEF5. Our results showed that the basic-amino-acid-rich sequence from residues 184 to 213 might represent a potential NoLS. Fluorescence microscopy analysis demonstrated that this predicted peptide sequence fused to GFP could localize to the nucleolus in Sf9 cells (Fig. 5). Further multiple point substitution in the putative NoLS resulted in attenuated or abrogated accumulation of AcMNPV LEF5 in the nucleolus (Fig. 6). These results demonstrated that LEF5 possesses an NoLS at amino acids 184 to 213.

Many studies showed that NoLS and NLS sequences reside in separate or joint locations (12, 14, 30, 31). Our nucleolar localization study revealed that LEF5 fused to HA could localize to the nucleolus (Fig. 4), while mutations of amino acids 197-RKKK-200 and 202-KHR-204 did not result in localization into the nucleolus (Fig. 6B), and thus that amino acids 197 to 204 are essential for nucleolar localization. However, these NoLS-mutated LEF5 proteins fused to HA were exclusively localized in the nucleus. This confirms that the only function of the LEF5 NoLS is in nucleolar localization, but the NoLS does not function in transporting LEF5 to the nucleus. However, the mutations of NoLS in AcMNPV LEF5 did not alter its nuclear localization (Fig. 6B), suggesting that LEF5 contains an NLS(s). Since we could not find a classic NLS sequence in AcMNPV LEF5, we speculate that LEF5 has a nonclassical NLS, which may be necessary for its nuclear translocation. Alternatively, perhaps LEF5 uses a chaperone for that purpose, much like AcMNPV P143 is transported to the nucleus via LEF3 in infected cells (32).

The nucleolus is a subnuclear structure responsible for ribosome-subunit biogenesis. Many viruses, including DNA viruses, retroviruses, and RNA viruses, take over the nucleolus and recruit nucleolar protein to favor viral replication and production. Herpes simplex virus 1 (HSV-1) infection led to redistribution of nucleolin, B23, and fibrillarin, and viral DNA replication requires a high level of nucleolin expression (33). In Schmallenberg virus (SBV)-infected cells, B23 undergoes a nucleolus-to-nucleoplasm redistribution, which is important for SBV-induced inhibition of cellular transcription (13). Adenovirus infection results in the redistribution of fibrillarin and B23 from the nucleolus to the cytoplasm (34). In this study, our results showed that AcMNPV infection triggered nucleolar disruption at a late stage of infection and redistribution of SfNS and SfFBL to the nucleoplasm in infected cells (Fig. 2A). Our results demonstrated that nucleolar morphogenesis during viral infection in insects is similar to that in human viral infections. AcMNPV LEF5 functions as a nucleolar protein, but its localization changed along with the distribution of nucleolar proteins during viral infection (Fig. 3C), which may contribute to more efficient virus production.

While previous studies revealed that LEF5 was transported into the nucleus (25), our results showed that LEF5 could localize to the nucleolus and contains an NoLS, which is important for viral replication, efficient budded-virus yield, and occlusion body production, but the NoLS is not involved in nuclear localization of AcMNPV LEF5. The normal nucleolus structure could be observed only at early stages of infection when Sf9 cells were infected by AcMNPV, while the nucleolus started to disorganize and nucleolar antigens were redistributed at 18 hpi. Later, LEF5 was distributed throughout the nucleoplasm (Fig. 3C), resulting in only nuclear localization of LEF5.

Until now, protein nucleolar localization has not been well understood. Many nucleolar proteins contain basic-residue-rich regions that function as NoLSs, although a consensus sequence of an NoLS motif has not been well defined. The nucleolus is a non-membrane-containing compartment. Unlike protein nuclear localization, which requires nuclear transport machinery to target the nucleus, nucleolar accumulation of proteins was considered to be mediated by a retention mechanism by interaction with distinct ribosomal proteins or RNA in its nucleolar targeting (35). The human protein C23 (nucleolin) uses a bipartite NLS to enter the nucleus and accumulates within the nucleolus via binding to B23 by its RNA-binding and glycine/arginine-rich domains (9, 30). Residues 41 to 72 from the nucleocapsid protein of porcine reproductive and respiratory syndrome virus (PRRSV) are involved in targeting the nucleolus (31). It is not yet known if the LEF5 NoLS we identified can bind to ribosomal proteins or RNA in the nucleolus in Sf9 cells. Further analysis of AcMNPV LEF5 NoLS interaction with nucleolar proteins or RNA is important to our understanding of the mechanism of nucleolar localization in baculovirus-infected cells.

In summary, we identified two nucleolar proteins, SfNS and SfFBL, in Sf9 cells and showed that AcMNPV infection resulted in redistribution of nucleolar proteins in infected cells. Furthermore, AcMNPV LEF5 localizes to the nucleolus, and amino acids 184 to 213 of LEF5 are required for this nucleolar localization. Our results also demonstrated that the NoLS in LEF5 is important for progeny viral yield and occlusion body production.

MATERIALS AND METHODS

Cells and viruses.

Spodoptera frugiperda clonal isolate 9 (Sf9) cells were cultured at 27°C in Grace’s insect medium supplemented with 10% fetal bovine serum (FBS; Invitrogen, USA), penicillin (25 U/ml), and streptomycin (25 μg/ml). Viral stocks of the parental E2 strain of AcMNPV and the recombinant viruses generated in the present study were all amplified in Sf9 cells.

Construction of transient-expression plasmids.

An HA-tag fusion expression cassette plasmid, pBlue-ie1-HA-SV40, was constructed by inserting the AcMNPV ie-1 promoter and HA tag-fused PCR fragment using primers 1 and 2 and simian virus 40 poly(A) signal, amplified from pFastBACHta (Invitrogen), using primers 3 and 4 into the SacI/XbaI and HindIII/XhoI sites of pBluescript II SK(+), respectively (Table 1). Coding sequences of the lef5 gene were amplified using primers 20 and 21 and subcloned into the XbaI/PstI sites of pBlue-ie1-HA-SV40 to make pBlue-HA:LEF5.

TABLE 1.

Primers used in this study

Primer no. and name	Primer sequence (5ʹ–3ʹ)	Position in genome
(1) Ie1PupSacI	AAGAGCTCGCGAATGCAGCTGATCACG	SacI + AcMNPV (126881–126899)
(2) Ie1PdnHAXbaI	GCTCTAGAAGCGTAGTCTGGAACATCGTAGGGATACATAGTCACTTGGTTGTTCACGATC	XbaI + HA tag + AcMNPV (127176–127197)
(3) SV40FHindIII	CCCAAGCTTGGGGATCATAATCAGCCATACCAC	HindIII + simian virus (2755–2774)
(4) SV40RXhoI	CCGCTCGAGCGGGATCCAGACATGATAAGATACATTG	XhoI + simian virus (2532–2553)
(5) Ie1PdnNotI	AAGCGGCCGCAGTCACTTGGTTGTTCACGATC	NotI + AcMNPV (127176–127197)
(6) EgfpFEcoRI	CCGGAATTCATGGTGAGCAAGGGCGAGGA	EcoRI + EGFP (1–20)
(7) EgfpREcoRV	CGCGATATCCTTGTACAGCTCGTCCATG	EcoRV + EGFP (699–717)
(8) mCherryFEcoRI	CCGGAATTCATGGTGAGCAAGGGCGAGGAG	EcoRI + mCherry (1–21)
(9) mCherryREcoRV	CGCGATATCCTTGTACAGCTCGTCCATGC	EcoRV + mCherry (689–708)
(10) lef5KO-1	CATCTGTTTCACTCGTTGCACGCGTACGTGCCCAGCGTCAGTGATTTGGTTACCCATATGAATATCCTCCTTAG	AcMNPV (86131–86180) + pKD3 (1025–1042)
(11) lef5KO-2	TATCATTCAATATTTGCCTGTGTTTGATTTTCTTTTTGCGTGTACACCCAGGTACCGTGTAGGCTGGAGCTGCT	AcMNPV (86496–86545) + pKD3 (31–48)
(12) lef5cko-1	CGTTCAACTTTTCGTCTACCG	AcMNPV (86108–86128)
(13) lef5cko-2	AAACTAAGCCCGCTAAGCTC	AcMNPV (86587–86606)
(14) lef5rep-1	CGCGAGCTCTCGGTTATGAGAGTGCTGTCC	SacI + AcMNPV (85541–85561)
(15) lef5rep-2	CCGCTCGAGTGGTTTATCGTCGCCGTCGC	XhoI + AcMNPV (86859–86878)
(16) lef5repHA-FL	ATGTATCCCTACGATGTTCCAGACTACGCTATGTCGTTTGATGATGGCGTC	HA tag + AcMNPV (85918–85938)
(17) lef5repHA-FS	ATGTCGTTTGATGATGGCGTC	AcMNPV (85918–85938)
(18) lef5repHA-RL	AGCGTAGTCTGGAACATCGTAGGGATACATGTTCTCGTTTTAAGCGAGTACG	HA tag + AcMNPV (85896–85917)
(19) lef5repHA-RS	GTTCTCGTTTTAAGCGAGTACG	AcMNPV (85896–85917)
(20) lef5-upX	GCTCTAGAATGTCGTTTGATGATGGCGTC	XbaI + AcMNPV (85918–85938)
(21) lef5-dnP	AAACTGCAGACAACCAGACATTCCACACAGC	PstI + AcMNPV (86691–86712)
(22) lef5mut1-FL	TTGGCGGCGGCGGAAGCGCATACCAGCACTGGGTGTACACG	AcMNPV (86467–86507); mutation position (185-KKKEK-189 to AAAEA)
(23) lef5mut1-FS	ACCAGCACTGGGTGTACACG	AcMNPV (86488–86507)
(24) lef5mut1-RL	ATGCGCTTCCGCCGCCGCCAACAACAAACTTTTGTAATTTAGGG	AcMNPV (86444–86487); mutation position (185-KKKEK-189 to AAAEA)
(25) lef5mut1-RS	CAACAAACTTTTGTAATTTAGGG	AcMNPV (86444–86466)
(26) lef5mut2-FL	ACAGCGGCGGCGGCGATCAAACACAGGCAAATATTGAATG	AcMNPV (86503–86542); mutation position (197-RKKK-200 to AAAA)
(27) lef5mut2-FS	AAACACAGGCAAATATTGAATG	AcMNPV (86521–86542)
(28) lef5mut2-RL	GATCGCCGCCGCCGCTGTACACCCAGTGCTGGTATGTT	AcMNPV (86483–86520); mutation position (197-RKKK-200 to AAAA)
(29) lef5mut2-RS	ACACCCAGTGCTGGTATGTT	AcMNPV (86483–86502)
(30) lef5mut3-FL	ATCGCGGCGGCGCAAATATTGAATGATAAAGTTATTTATTTAC	AcMNPV (86518–86560); mutation position (202-KHR-204 to AAA)
(31) lef5mut3-FS	CAAATATTGAATGATAAAGTTATTTATTTAC	AcMNPV (86530–86560)
(32) lef5mut3-RL	CGCCGCCGCGATTTTCTTTTTGCGTGTACACCC	AcMNPV (86497–86529); mutation position (202-KHR-204 to AAA)
(33) lef5mut3-RS	TTTCTTTTTGCGTGTACACCC	AcMNPV (86497–86517)
(34) lef5mut4-FL	ACAGCGGCGGCGGCGATCGCGGCGGCGCAAATATTGAATGATAAAGTTATTTATTTAC	AcMNPV (86503–86560); mutation position (197-RKKKIKHR-204 to AAAAIAAA)
(35) lef5mut4-RL	CGCCGCCGCGATCGCCGCCGCCGCTGTACACCCAGTGCTGGTATGTTTTTC	AcMNPV (86479–86529); mutation position (197-RKKKIKHR-204 to AAAAIAAA)
(36) lef5mut4-RS	ACACCCAGTGCTGGTATGTTTTTC	AcMNPV (86479–86502)
(37) GNL3-upB	CGGGATCCATGGCCAAGTTTAAGTTGAAGAAAC	BamHI + CDS of Sfgnl3 (1–25)
(38) GNL3-upP	AAACTGCAGTAGGGAAAAGTCTTCTTTAAAGTCG	PstI + CDS of Sfgnl3 (1716–1740)
(39) GNL3-upS	ACGCGTCGACCGCGAATGCAGCTGATCACG	SalI + AcMNPV (126880–126899)
(40) GNL3-dnS	ACGCGTCGACATCCAGACATGATAAGATACATTGATG	SalI + simian virus (2535–2561)
(41) qPCR-1	CGTAGTGGTAGTAATCGCCGC	AcMNPV (65972–65992)
(42) qPCR-2	AGTCGAGTCGCGTCGCTTT	AcMNPV (66054–66072)
(43) lef5cki-1	TTGCAGACAGAATATTTGGCA	AcMNPV (86203–86223)
(44) lef5cki-2	CAACAAACTTTTGTAATTTAGGGA	AcMNPV (86443–86466)
(45) qvp39-F	TTGCGCAACGACTTTATACC	AcMNPV (75628–75647)
(46) qvp39-R	TAGACGGCTATTCCTCCACC	AcMNPV (75535–75554)
(47) q38k-F	GCCATACGACCACAAGACT	AcMNPV (85447–85465)
(48) q38k-R	CATAACCGAAGAGGAGCAA	AcMNPV (85531–85549)
(49) qp6.9-F	GGCGACCTGTCGATGAA	AcMNPV (86738–86754)
(50) qp6.9-R	CGCAGAAGCTCGGGTTA	AcMNPV (86803–86819)
(51) qpolh-F	TTAGGTGCCGTTATCAAGA	AcMNPV (4589–4607)
(52) qpolh-R	GCCACTAGGTAGTTGTCT	AcMNPV (4669–4686)
(53) qp10-F	CATATTGACCGGCGACATTG	AcMNPV (119009–119028)
(54) qp10-R	TTACTTGGAACTGCGTTTACC	AcMNPV (119103–119123)

The SacI/NotI ie-1 promoter PCR fragment from AcMNPV ie-1 (using primers 1 and 5), the BamHI/PstI nucleostemin ORF from the Spodoptera frugiperda cDNA library (using primers 37 and 38), and the HindIII/XhoI fragment for simian virus 40 poly(A) signal from pFastBACHta (Invitrogen) (using primers 3 and 4) were inserted into pBluescript II SK(+), generating pBlue-NS. The ORFs of GFP and mCherry were inserted into the EcoRI/EcoRV sites of pBlue-NS to generate plasmids pBlue-NS:GFP (using primers 6 and 7) and pBlue-NS:mCherry (using primers 8 and 9), respectively.

Generation of the lef5 knockout from the AcMNPV bacmid.

The construction of the lef5 knockout in a bacmid, bMON14272, containing the AcMNPV genome was accomplished via a PCR-based protocol involving the λ Red homologous recombination system in Escherichia coli (BW2113), as previously described. Briefly, a chloramphenicol resistance cassette (CAT) was amplified using primers 10 and 11 from the template plasmid pKD3. These two primers contained a 50-bp sequence homologous to the 5′ and 3′ regions of the AcMNPV lef5 gene, respectively. The PCR fragment was gel purified and electroporated into E. coli BW25113, which harbored the AcMNPV bacmid (bMON14272) and temperature-sensitive plasmid pKD46, which expressed λ Red recombinase with the addition of l-arabinose. The electroporated cells were allowed to recover in LB broth devoid of antibiotics at 37°C for 5 h, plated onto LB plates containing kanamycin (50 μg/ml) and chloramphenicol (34 μg/ml), and incubated at 37°C overnight. Plasmid pKD46 was cured by overnight incubation at a nonpermissive temperature of 43°C.

Colonies isolated from chloramphenicol and kanamycin plates were further screened by PCR using primers 12 and 13 to confirm the replacement of the CAT with lef5 locus in the AcMNPV bacmid. Primers 43 and 44, specific to the lef5 gene, were also used to verify the absence of the lef5 gene from the AcMNPV bacmid.

Construction of lef5 repair and subcellullar localization plasmid.

A 1.3-kb lef5 fragment containing the entire LEF5 ORF and its native promoter and poly(A) signal was PCR amplified from the parental AcMNPV bacmid by using primers 14 and 15 and cloned into pGEM-T Easy vector to make pGEM-LEF5. An HA tag (YPYDVPDYA) was inserted into the N terminus (after the start codon) of LEF5 by using primers 16 to 19, creating pGEM-HA:LEF5. Plasmids pGEM-LEF5 and pGEM-HA:LEF5 were digested with SacI/XhoI and subcloned into pFACT-GFP and pFACT to create pFACT-GFP-LEF5 and pFACT-HA:LEF5, respectively. An NS- and mCherry-coexpressing cassette was amplified from pBlue-NS:mCherry with primers 39 and 40 and inserted into the SalI site of pFACT-HA:LEF5 to create pFACT-NS:mCherry-HA:LEF5.

Multiple alanine point substitutions in AcMNPV LEF5.

To explore the role of basic amino acid clusters in nucleolar localization of AcMNPV LEF5, plasmids pBlue-HA:LEF5 and pGEM-LEF5 were used as the template for site-directed mutagenesis. Each reaction included four primers.

Codons for basic amino acid clusters in putative LEF5 NoLS of pBlue-HA:LEF5 were substituted by alanine by using primers 22 to 25 for the mutation of 185-KKKEK-189 (underlines indicate alanine replacement positions), primers 26 to 29 for the mutation of 197-RKKK-200, primers 30 to 33 for the mutation of 202-KHR-204, and primers 31 and 34 to 36 for the mutation of 197-RKKKIKHR-204, creating plasmids pBlue-HA:LEF5^Mut185-189, pBlue-HA:LEF5^Mut197-200, pBlue-HA:LEF5^Mut202-204, and pBlue-HA:LEF5^Mut197-204, respectively.

The same strategy as described above was used for mutagenesis in pGEM-LEF5 to generate plasmids pGEM-LEF5^Mut185-189, pGEM-LEF5^Mut197-200, pGEM-LEF5^Mut202-204, and pGEM-LEF5^Mut197-204. These plasmids were digested with SacI/XhoI and subcloned into the pFACT-GFP vector, constructing pFACT-GFP-LEF5^Mut185-189, pFACT-GFP-LEF5^Mut197-200, pFACT-GFP-LEF5^Mut202-204, and pFACT-GFP-LEF5^Mut197-204, respectively. The resulting plasmids were used to generate the corresponding bacmids.

Bacmid constructions.

The plasmid pFACT-GFP was used to transpose fragments into the polyhedrin locus of lef5 knockout AcMNPV bacmid as described above according to the Bac-to-Bac expression manual (Invitrogen) for the construction of Bac-GFP-LEF5^KO, lef5 knockout bacmid expressing polyhedrin and GFP. The lef5 repair and subcellullar localization bacmids, Bac-GFP-LEF5^REP, Bac-LEF5^HA, and Bac-NS:mCherry-LEF5^HA, were constructed by transferring cloned fragments into the lef5 knockout bacmid using plasmids pFACT-GFP-LEF5, pFACT-HA:LEF5, and pFACT-NS:mCherry-HA:LEF5, respectively.

For mutated LEF5, the Tn7 cassettes from plasmids pFACT-GFP-LEF5^Mut185-189, pFACT-GFP-LEF5^Mut197-200, pFACT-GFP-LEF5^Mut202-204, and pFACT-GFP-LEF5HA^Mut197-204 were transferred to the lef5 knockout bacmid, creating Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204.

To monitor the status of the nucleolus during AcMNPV infection, an NS- and GFP-coexpressing cassette was amplified from pBlue-NS:GFP with primers 39 and 40 and inserted into the SalI site of pFACT to create pFACT-NS:GFP, which was transferred to the AcMNPV bacmid to generate bacmid Bac-NS:GFP.

Transfection.

Bacmid DNAs were purified from 500-ml LB cultures using a HiPure plasmid midiprep kit (Invitrogen), and concentration was quantified using a NanoDrop 2000c spectrophotometer (Thermo Scientific). Sf9 cells were seeded in six-well plates with an approximate density of 1 × 10⁶ per well. Cells were transfected with 2 μg of bacmid DNA by using Cellfectin (Invitrogen) according to the instructions provided by the manufacturer. Cells were incubated with transfection mixture for 5 h, after which time the transfection mixture was discarded and supplemented with fresh Grace’s medium.

Viral growth curves.

Bacmid DNAs were transfected into 1 × 10⁶ Sf9 cells per well in triplicate using Cellfectin. Cell monolayers were incubated for 5 h with transfection mixture and rinsed twice with Grace’s medium, which was replaced with 2 ml of fresh Grace’s medium supplemented with 10% FBS. Virus supernatants were collected from extracellular medium with an interval of 24 h. Viral yields were determined in Sf9 cells by a 50% tissue culture infective dose (TCID₅₀) endpoint dilution assay, and virus titers were determined according to the Reed-Muench method (36). Signs of infection were determined by GFP fluorescence. For all transfections, time zero (0 hpt) was considered to be when the monolayer was replenished with fresh medium after the transfection solution had been aspirated from the wells. Each single-step growth curve experiment was replicated three times.

Quantitative real-time PCR.

Quantitative real-time PCR (qPCR) was performed to determine increases in the level of viral DNA copies with primer pair 41 and 42 for the amplification of a 100-bp region within the gp41 ORF, which contains four DpnI restriction sites. Treatment with DpnI digests the input bacmids but preserves replicated DNAs. In brief, Sf9 cells (1 × 10⁶) were transfected with 1 μg of bacmid DNA and harvested at 0 and 24 hpt. Total intracellular DNA was extracted using MiniBEST universal genomic DNA extraction kit (TaKaRa) and resuspended in 100 μl of sterile water. Equal amounts of total DNA from each time point were digested with 20 U of DpnI at 37°C overnight. A 2.5-μl aliquot of the digested DNA was mixed with TB Green Premix Ex Taq II (TaKaRa), and the resulting mixture was subjected to qPCR analysis in a CFX96 real-time system (Bio-Rad). The gp41 ORF was PCR amplified and cloned into the pMD19-T vector. The number of plasmid copies was calculated based on the DNA concentration. Standard curves were generated using serial dilutions of the plasmid.

Immunoblotting.

Sf9 cells were infected with wild-type AcMNPV at an MOI of 5 and collected by centrifugation at various times postinfection. Proteins from cells lysed in 1% sodium dodecyl sulfate (SDS) and β-mercaptoethanol (β-ME) were subjected to SDS-polyacrylamide gel electrophoresis (PAGE) and transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore). For immunoblotting, membranes were incubated with antibodies against fibrillarin (1:2,000; Abcam) and β-actin (1:2,500; Abbkine) as a loading control. Signal development was conducted by using horseradish peroxidase (HRP)-conjugated goat anti-mouse secondary antibody (Abmart) at a dilution of 1:5,000 at room temperature for 2 h, and bands were detected with a SuperSignal West Pico Trail kit (Thermo). Films were scanned at 300 dots per inch (dpi) and prepared using CS 5 Adobe Photoshop and Illustrator.

Quantitative reverse transcription-PCR analysis of viral transcripts.

Sf9 cells were infected with Bac-GFP-LEF5^REP, Bac-GFP-LEF5^Mut185-189, Bac-GFP-LEF5^Mut197-200, Bac-GFP-LEF5^Mut202-204, and Bac-GFP-LEF5^Mut197-204 at an MOI of 1 and harvested at 24 hpi. Total RNA was isolated using the RNeasy minikit (Qiagen) according to the manufacturer’s protocol. One microgram of each purified RNA was used for cDNA synthesis using a PrimeScript RT reagent kit with gDNA Eraser (TaKaRa). The RNA extracted from uninfected cells was used to determine the background level of expression of each gene. qPCR was performed with TB Green Premix Ex Taq II (TaKaRa) by using primers 45 and 46 for vp39, primers 47 and 48 for 38k, primers 49 and 50 for p6.9, primers 51 and 52 for polh, and primers 53 and 54 for p10. The qPCR data of viral genes were normalized to host 18S rRNA using primers 5′-TACCGATTGAATGATTTAGTGAGG-3′ and 5′-TACGGAAACCTTGTTACGACTTT-3′ (37), and the relative expression of genes was calculated using the threshold cycle (2^−ΔΔCT) method (38).

Confocal microscopy.

Sf9 cells (5 × 10⁵) were seeded onto 35-mm glass bottom dishes (In Vitro Scientific) to attach for 5 h. Monolayers were washed twice in phosphate-buffered saline (PBS) and overlaid with 2 ml of PBS before imaging. For immunofluorescence analysis, transfected or infected cells were fixed in 4% paraformaldehyde for 15 min and washed three times with PBS, followed by a 15-min permeabilization in PBS containing 0.3% Triton X-100 and a 1-h blocking in PBS with 2% bovine serum albumin (BSA). Fixed and BSA-blocked cells were incubated with a mouse monoclonal anti-HA antibody (1:300; Abmart) or anti-fibrillarin antibody (1:200; Abcam) overnight at 4°C, washed three times for 10 min each time in 2% BSA blocking buffer, and incubated with an Alexa Fluor 488-conjugated goat anti-mouse IgG (1:300) (Invitrogen) for 1 h and nuclear staining dye Hoechst 33342 (10 μg/ml) for 15 min. Cells were washed with PBS and imaged on a Zeiss LSM 700 confocal microscope with ZEN 2011 software (Zeiss) for fluorescent imaging.

Statistical analysis.

Statistical analysis was performed using Student’s t test with GraphPad Prism software. Significance is indicated in figures as follows: *, P < 0.05; **, P < 0.01; ***, P < 0.001; and ns, not significant.

Accession number(s).

The transcriptome shotgun assembly project in this study has been deposited at DDBJ/EMBL/GenBank under the accession number GHKU00000000. The version described in this paper is the first version, GHKU01000000.