Potential Proteins Interactions with Bombyx mori Nucleopolyhedrovirus Revealed by Co-Immunoprecipitation

Simple Summary The Bombyx mori nucleopolyhedrovirus (BmNPV) is a typical model baculovirus, representing one of the major pathogens of the silkworm. We reconstituted the virus in vitro and used it as a bait for immunoprecipitation experiments on cells and silkworm bodies, obtaining a database of proteins potentially interacting with BmNPV. This study provides a protein data reference for the screening of BmNPV receptors and the study of proteins that play a key role in the replication of BmNPV. It is also important to deal with the prevention of BmNPV in silkworms at the molecular level. Abstract Virus–host interactions are critical for virus replication, virulence, and pathogenicity. The Bombyx mori nucleopolyhedrovirus (BmNPV) is a typical model baculovirus, representing one of the most common and harmful pathogens in sericulture. Herein, we used co-immunoprecipitation to identify candidate proteins with potential interactions with BmNPV. First, a recombinant BV virus particle rBmBV-egfp-p64-3×flag-gp64sp was constructed using a MultiBac baculovirus multigene expression system. Co-immunoprecipitation experiments were then performed with the recombinant BV virus infected with BmN cells and Dazao silkworms. LC-MS/MS analysis revealed a total of 845 and 1368 candidate proteins were obtained from BmN cells and silkworm samples, respectively. Bioinformatics analysis (Gene Ontology, KEGG Pathway) was conducted for selection of proteins with significant enrichment for further confirmation of the effects on BmNPV replication. Overall, the results showed that SEC61 and PIC promoted the replication of BmNPV, while FABP1 inhibited the replication of BmNPV. In summary, this study reveals the potential proteins involved in BmNPV invasion and proliferation in the host and provides a platform for identifying the potential receptor proteins of BmNPV.


Introduction
Baculovirus is a class of double-stranded circular DNA viruses with a capsid structure and a genome size of 80-180 kb, for whom the main hosts are arthropods [1,2]. Based on the morphology of the inclusion bodies, baculoviruses are divided into two main groups, the nucleopolyhedrovirus (NPV) and the granulovirus (GV) [3,4]. The Bombyx mori nucleopolyhedrovirus (BmNPV) is a typical model baculovirus, representing one of the most common and harmful pathogens in sericulture [5][6][7][8]. When BmNPV infects a host, it produces two types of viral particles: the early budded virions (BV) [9], which spread mainly between cells, and the later occlusion-derived virions (ODV), which spread mainly between hosts [10][11][12]. ODV virus particles are packaged in polyhedra of highly symmetric covalent cross-linked lattices [13,14]. BmNPV infects B. mori larvae mainly through the oral Table 1. Primers used in this study.

Co-Immunoprecipitation
BmN cells were cultured in 75 cm 2 cell culture flask and were grown to 70-80%, the recombinant baculovirus rBmBV-egfp-p64-3×flag-gp64sp was added to the flask and incubated for 72 h at 28 • C. Cell samples were then collected using 2 mL PBS. Immunoprecipitation experiments were performed with a Pierce TM Crosslink Magnetic IP/Co-IP kit (Thermo Fisher Scientific, Waltham, MA, USA) ( Figure 1a). The recombinant baculovirus rBmBV-egfp-p64-3×flag-gp64sp was injected into Dazao silkworm, and silkworm morbidity was determined by the characteristics of sil worm disease and green fluorescence of silkworm body. The silkworm epithelium is mo (1: Proteins of the lysed cells; 3: Protein not bound to magnetic beads; 5: Protein eluted from the magnetic beads, and 2,4,6 is control).
The recombinant baculovirus rBmBV-egfp-p64-3×flag-gp64sp was injected into a Dazao silkworm, and silkworm morbidity was determined by the characteristics of silkworm disease and green fluorescence of silkworm body. The silkworm epithelium is more accessible than other silkworm tissues and less likely to be mixed with proteins from other tissues. Moreover, to ensure the expression of "3×flag-gp64sp", we chose silkworm epithelium as the experimental sample. Then, the affected silkworm epithelium was collected and ground in liquid nitrogen, and silkworm epithelium samples were collected with 2 mL PBS. Normal silkworms were used as control. Immunoprecipitation experiments were performed with a Pierce TM Crosslink Magnetic IP/Co-IP kit (Thermo Fisher Scientific) (Figure 2a). Proteins of the lysed cells; 3: Protein not bound to magnetic beads; 5: Protein eluted from the magnetic beads, and 2,4,6 is control).
The recombinant baculovirus rBmBV-egfp-p64-3×flag-gp64sp was injected into a Dazao silkworm, and silkworm morbidity was determined by the characteristics of silkworm disease and green fluorescence of silkworm body. The silkworm epithelium is more accessible than other silkworm tissues and less likely to be mixed with proteins from other tissues. Moreover, to ensure the expression of "3×flag-gp64sp", we chose silkworm epithelium as the experimental sample. Then, the affected silkworm epithelium was collected and ground in liquid nitrogen, and silkworm epithelium samples were collected with 2 mL PBS. Normal silkworms were used as control. Immunoprecipitation experiments were performed with a Pierce TM Crosslink Magnetic IP/Co-IP kit (Thermo Fisher Scientific) (Figure 2a). (1: Proteins of the lysed silkworm of the epithelium; 3: Protein not bound to magnetic beads; 5: Protein eluted from the magnetic beads, and 2,4,6 is control).

Western Blotting
Collected cell samples and silkworm samples were processed by lysate, and after SDS-PAGE, the proteins were transferred into PVDF membranes (Roche Diagnostics, Basel, Switzerland) and incubated with the primary antibody Flag (Beyotime, Suzhou, China), and the secondary antibody, horseradish peroxidase-conjugated goat anti-mouse IgG (Beyotime). The results of Western blotting were analyzed by a ECL Western blotting Detection System (Bio-Rad, Hercules, CA, USA).

Transmission Electron Microscopy
The supernatant of BmN cells infected with the recombinant baculovirus rBmBV-egfp-p64-3×flag-gp64sp was collected and centrifuged at 4 • C for 60 min at 100,000× g. The supernatant was discarded, and the precipitate was resuspended with PBS solution and stored at 4 • C. Hydrophilic treatment of copper mesh was conducted. Then, 10 µL of sample was pipetted into the copper mesh and stained with 2% uranium acetate for 45 s. After drying for 2 h, the samples were observed using a transmission electron microscope (Talos F200s, TEM, Thermo Fisher Scientific).

LC-MS/MS Analysis
The sample was loaded onto a preloaded column (3 µm, 120 Å) at a flow rate of 4 µL/min, and were further rinsed and desalted for 5 min. After completion of desalting, the sample was separated by an analytical column with a gradient of mobile phase B from 8 to 38% in 60 min. The ion source spray voltage was set as 2.4 kV, while the atomization air pressure was 12 PSI. In addition, the air curtain air pressure was 35 PSI, while the heating temperature was 150 • C. The sweep time for a single spectrum was 250 ms, and a maximum of 40 secondary profiles with charges of 2 + to 4 + and counts greater than 260 cps acquired at a time with a cumulative time of 60 ms. Each cycle lasted 2.5 s. The raw data from the mass spectrometry acquisition were processed and analyzed by PEAKS Studio 8.5 software (Thermo Fisher Scientific) with Uniprot (https://www.uniprot.org/, accessed on 5 June 2021) under the B. mori species protein database.
Pathway enrichment of the candidate proteins identified by LC-MS/MS wasperformed for identifying candidate enriched metabolic pathways with the KEGG database (https: //www.kegg.jp/, accessed on 20 September 2021). Statistical analysis was performed for pairs of different types of proteins.

RNAi
With the primers listed in Table 1, T7 RIBO MAX TM expressing the RNAi system kit (Promega, Madison, WI, USA) was used to synthesize the dsRNA (double-strand RNA), and the product was stored at −80 • C. The dsRNA was mixed with the transfection reagent (Promega) and added to BmN cells for incubation at 28 • C. After 24 h.p.i, the recombinant baculovirus BmBV-egfp-p64-3×flag-gp64sp was added. Samples from two timepoints (12 and 24 h) were collected, respectively. Normal BmN cells were used as control.

Quantitative Real-Time PCR (qPCR)
The total RNAs of BmN cells were extracted with the Kit RNA fast 2000 (Fastagen, Shanghai, China) and then reverse transcribed into cDNA using an RT reagent kit with gDNA Eraser (TaKaRa, Shiga, Japan). qPCR was further performed with the primers listed in Table 1, and Ribosomal protein RP49 was used as an internal gene control. The qPCR was carried out in a 20 µL reaction mixture that contained 0.8 µL of cDNA, 0.5 mM of each specific primer, and 10 µL of 2xiTaq TM Universal SYBR Green Supermix (Bio-Rad), and each test was conducted in triplicate. The reaction conditions were set as 95 • C for 5 min, followed by 40 cycles at 95 • C for 15 s and 60 • C for 35 s.

Co-Immunoprecipitation
BmN cells were infected with the recombinant baculovirus, and the baculovirus infestation was determined by identifying the green fluorescence by inverted fluorescence microscopy ( Figure 1b). A 50 mL volume of BmN cells infected with the recombinant baculovirus was then enriched with 2 mL of PBS at 72 h.p.i. Collection of samples at different stages of co-immunoprecipitation was performed for qualitative analysis. Westernblotting results showed that the lysed cells had contained the recombinant baculovirus proteins and the magnetic bead-eluted samples contained protein complexes that interacted with the recombinant baculovirus proteins (Figure 1c).
The recombinant baculovirus was then injected into Dazao silkworm, and the affected silkworms were identified by disease symptoms and green fluorescence targeting the silkworm body (Figure 2b). The epithelium of infected silkworms was collected for liquid nitrogen grinding. As with cell samples, qualitative analysis of each stage of coimmunoprecipitation was performed. The Western blotting results showed that the epithelium of affected silkworm had contained the recombinant baculovirus proteins and the magnetic bead-eluted samples had contained the protein complexes that interacted with the recombinant baculovirus proteins (Figure 2c). The recombinant baculovirus was passed for three generations to obtain virulencestabilized progeny baculovirus. The green fluorescence of infected BmN cells gradually stabilized at 48-72 h (Figure 4a). The recombinant baculovirus capsid protein GP64 was detected by SDS-PAGE and Western-Blot, and the results showed successful expression of GP64 protein (Figure 4b,c). The recombinant baculovirus structure was examined by transmission electron microscopy, and the results showed that the baculovirus was structurally intact (Figure 4d).

Co-Immunoprecipitation
BmN cells were infected with the recombinant baculovirus, and the baculovirus infestation was determined by identifying the green fluorescence by inverted fluorescence microscopy ( Figure 1b). A 50 mL volume of BmN cells infected with the recombinant baculovirus was then enriched with 2 mL of PBS at 72 h.p.i. Collection of samples at different stages of co-immunoprecipitation was performed for qualitative analysis. Western-blotting results showed that the lysed cells had contained the recombinant baculovirus proteins and the magnetic bead-eluted samples contained protein complexes that interacted with the recombinant baculovirus proteins (Figure 1c).
The recombinant baculovirus was then injected into Dazao silkworm, and the affected silkworms were identified by disease symptoms and green fluorescence targeting the silkworm body (Figure 2b). The epithelium of infected silkworms was collected for liquid nitrogen grinding. As with cell samples, qualitative analysis of each stage of co-

LC-MS/MS Analysis
The magnetic bead eluate was analyzed by LC-MS/MS. A total of 5156 matched peptides were identified in the cells, with a false discovery rate of 1.8% (Figure 5a). Peptidespectrum matches were mainly concentrated at 0-5 ppm (Figure 5b). Lastly, 845 proteins were matched in the Uniprot protein database for B. mori species (Table S1).
A total of 10,790 matched peptides were identified in the silkworm samples, and the false discovery rate was 0.9% (Figure 5c). The peptide-spectrum matches were also mainly concentrated in 0-5 ppm (Figure 5d). A total of 1368 proteins were matched B. mori species in the Uniprot protein database (Table S2).

Bioinformatic Analysis of Gene Ontology and KEGG
Gene ontology analysis was performed on the 845 candidate proteins obtained from cells. Regarding enrichment in the biological process in cells, the largest proportion was found to be enriched in the metabolic processes of organic nitrogen compounds, followed by the protein metabolic processes. Cellular component analysis found that the largest proportion was enriched in proteins of cell components, followed by proteins of intracellular components, while molecular function analysis indicated that the largest proportion was enriched in proteins involved in the binding function of small molecules, followed by nucleotide-binding proteins (Figure 6a). The KEGG pathway enrichment analysis reported the top ten signaling pathways with significant levels (p < 0.05), and the results showed that the 845 candidate proteins were mainly involved in the signaling pathways of cellular metabolism, followed by signaling pathways of ribosome metabolism (Figure 6b). By gathering the results from GO and KEGG analysis in the candidate proteins obtained from cells, the results indicated that 1027 proteins were identified in the biological process, 210 of which were with significantly enriched; cell component enriched 242 proteins, 92 of which were significant; molecular function enriched 556 proteins, 117 of which were significant; Kegg pathway enriched 79 proteins, 25 of which were significant (Figure 6c) (Table S3). magnetic bead-eluted samples had contained the protein complexes that interacted with the recombinant baculovirus proteins (Figure 2c).

LC-MS/MS Analysis
The magnetic bead eluate was analyzed by LC-MS/MS. A total of 5156 matched peptides were identified in the cells, with a false discovery rate of 1.8% (Figure 5a). Peptidespectrum matches were mainly concentrated at 0-5 ppm (Figure 5b). Lastly, 845 proteins were matched in the Uniprot protein database for B. mori species (Table S1).
A total of 10,790 matched peptides were identified in the silkworm samples, and the false discovery rate was 0.9% (Figure 5c). The peptide-spectrum matches were also mainly concentrated in 0-5 ppm (Figure 5d). A total of 1368 proteins were matched B. mori species in the Uniprot protein database (Table S2).

Bioinformatic Analysis of Gene Ontology and KEGG
Gene ontology analysis was performed on the 845 candidate proteins obtained from cells. Regarding enrichment in the biological process in cells, the largest proportion was found to be enriched in the metabolic processes of organic nitrogen compounds, followed by the protein metabolic processes. Cellular component analysis found that the largest proportion was enriched in proteins of cell components, followed by proteins of intracellular components, while molecular function analysis indicated that the largest proportion was enriched in proteins involved in the binding function of small molecules, followed by nucleotide-binding proteins (Figure 6a). The KEGG pathway enrichment analysis reported the top ten signaling pathways with significant levels (p < 0.05), and the results showed that the 845 candidate proteins were mainly involved in the signaling pathways of cellular metabolism, followed by signaling pathways of ribosome metabolism ( Figure  6b). By gathering the results from GO and KEGG analysis in the candidate proteins obtained from cells, the results indicated that 1027 proteins were identified in the biological  (Table S3). The results of GO and KEGG analysis in the silkworm samples were similar to those of BmN cells, however, there were differences in the number of proteins (Figure 7a,b). The biological process enriched 1147 proteins, 311 of which were significant; cell component enriched 276 proteins, 117 of which were significant; molecular function enriched 682 pro- The results of GO and KEGG analysis in the silkworm samples were similar to those of BmN cells, however, there were differences in the number of proteins (Figure 7a,b). The biological process enriched 1147 proteins, 311 of which were significant; cell component enriched 276 proteins, 117 of which were significant; molecular function enriched 682 proteins, 145 of which were significant; Kegg pathway enriched 98 proteins, 45 of which were significant (Figure 7c) (Table S4).

dsRNA Interferes with the Expression of FABP1, PIC, and SEC61
Combined with bioinformatics analysis, we selected some significant proteins in the Kegg pathway to investigate their effects on BmNPV replication. These included cellular retinoic acid binding protein FABP1 (ID: Q2QEH2), phosphate carrier protein PIC (ID: Q1HPL2) and transport protein SEC61 (ID: Q19AA9). The dsRNA fragments of FABP1, PCI, and SEC61 were synthesized with the primers shown in Table 1. The results of qPCR showed that down-regulation of FABP1 mRNA level resulted in up-regulation of the relative expression of baculovirus nucleocapsid protein VP39 at 12 h.p.i and 24 h.p.i ( Figure  8a), while down-regulation of PIC mRNA level resulted in down-regulation of the relative expression of VP39 at 12 h.p.i and 24 h.p.i (Figure 8b). In addition, down-regulation of SEC61 mRNA level resulted in down-regulation of the relative expression of VP39 at 12 h.p.i and 24 h.p.i (Figure 8c). These results demonstrate that PCI, SEC61 protein together might promote BmNPV replication in the host, while FABP1 protein likely inhibits BmNPV replication in the host.

dsRNA Interferes with the Expression of FABP1, PIC, and SEC61
Combined with bioinformatics analysis, we selected some significant proteins in the Kegg pathway to investigate their effects on BmNPV replication. These included cellular retinoic acid binding protein FABP1 (ID: Q2QEH2), phosphate carrier protein PIC (ID: Q1HPL2) and transport protein SEC61 (ID: Q19AA9). The dsRNA fragments of FABP1, PCI, and SEC61 were synthesized with the primers shown in Table 1. The results of qPCR showed that down-regulation of FABP1 mRNA level resulted in up-regulation of the relative expression of baculovirus nucleocapsid protein VP39 at 12 h.p.i and 24 h.p.i (Figure 8a), while down-regulation of PIC mRNA level resulted in down-regulation of the relative expression of VP39 at 12 h.p.i and 24 h.p.i (Figure 8b). In addition, down-regulation of SEC61 mRNA level resulted in down-regulation of the relative expression of VP39 at 12 h.p.i and 24 h.p.i (Figure 8c). These results demonstrate that PCI, SEC61 protein together might promote BmNPV replication in the host, while FABP1 protein likely inhibits BmNPV replication in the host. Insects 2022, 13, x FOR PEER REVIEW 11 of 15

Discussion
Viruses are obligate parasites. Upon their invasion of the host and subsequent replication, they are met by interactions with host proteins [30,31]. In this study, we reconstructed the budding viral particle BV of BmNPV using the multi-bac baculovirus expression system and conducted co-immunoprecipitation experiments with the BV as bait to investigate the interaction between viral proteins and host proteins during the invasion and replication of BmNPV. LC-MS/MS analysis revealed a total of 845 and 1368 candidate proteins were obtained in cells and silkworm samples, respectively. Among them, 476 candidate proteins were detected in both cells and silkworm body samples.
There are several approaches for identifying the interaction between viruses and hosts. Feng et al. obtained the protein SINAL10, which interacts with the baculovirus capsid protein GP64, by using a yeast two-hybrid and baculovirus library screening method [25], while Dong et al. obtained the BmREEPa protein by proteomic differential analysis

Discussion
Viruses are obligate parasites. Upon their invasion of the host and subsequent replication, they are met by interactions with host proteins [30,31]. In this study, we reconstructed the budding viral particle BV of BmNPV using the multi-bac baculovirus expression system and conducted co-immunoprecipitation experiments with the BV as bait to investigate the interaction between viral proteins and host proteins during the invasion and replication of BmNPV. LC-MS/MS analysis revealed a total of 845 and 1368 candidate proteins were obtained in cells and silkworm samples, respectively. Among them, 476 candidate proteins were detected in both cells and silkworm body samples.
There are several approaches for identifying the interaction between viruses and hosts. Feng et al. obtained the protein SINAL10, which interacts with the baculovirus capsid protein GP64, by using a yeast two-hybrid and baculovirus library screening method [25], while Dong et al. obtained the BmREEPa protein by proteomic differential analysis of two cell lines, BmN-SWU1 and BmN-SWU2, which interacts with GP64 protein through the BmPTCHD protein [27]. In the present study, BmREEPa protein was also obtained by screening in cells and silkworm samples. In terms of the time of viral infection, Xue et al., confirmed by transcriptomic analysis that BmNPV reached a maximum interaction of viral proteins with host proteins at 24 h.p.i [32]. In this study, we collected samples for co-immunoprecipitation experiments after the baculovirus infection stabilized at 72 h. Therefore, further optimization of the sample collection time would allow increasing amount of protein.
Viruses can interact directly with host proteins or indirectly through other proteins. The receptor protein for BmNPV has not been identified. The capsid protein GP64 of BmNPV plays an important role in its invasion of the host. The current research shows that the proteins that may interact with the GP64 protein are as follows. (1) Phospholipids, TANI et al., treated human hepatoma cells HepG2 with phospholipase C and then coincubated with recombinant baculovirus Ac64-CAluc and AcVSVG-CAluc, respectively. The expression of fluorescein was used to determine whether the virus entered the host cells. The results showed a dose-dependent expression level of the luciferase gene in the Ac64-CAluc experimental group, while no significant changes were observed in the AcVSVG-CAluc experimental group [33]. Chernomordik et al., treated host insect cells with phospholipase A, which affected the composition of their membrane phospholipids. In addition, the formation of syncytia or lack thereof was used to determine whether the host cells were infected or not [34]. These results suggest that phospholipids on the cell surface play an important role in both baculovirus infection of insect cells and transduction of mammalian cells. (2) Cholesterol, it is essential for animal life activities, accounting for about 30% of animal cell membranes, and its main function is to maintain membrane stability and regulate membrane fluidity. The cholesterol recognition domain (CRAC) is required to mediate viral endocytosis via lattice proteins. Luz-Madrigal et al., found that the GP64 gene contains a cholesterol recognition domain and that baculovirus infection was significantly reduced when host cell cholesterol was inhibited with MβCD compared to the control group [35]. Hao et al., found that two Cholesterol Recognition Amino Acid Consensus Motifs of GP64 with uncleaved Signal Peptide are required for B. mori Nucleopolyhedrovirus infection [36]. (3) Acetyl heparin sulfate, it is a linear polysaccharide present in all animal tissues, which usually occurs as a proteoglycan and binds various protein ligands. Acetyl heparin sulfate has been shown to be a receptor for a variety of viruses, e.g., acetyl heparin sulfate plays an important role in novel coronavirus SARS-CoV-2 infection, especially when the virus is bound to ACE2 [37]. Makkonen et al., showed that a heparin recognition region exists at amino acid sites 271-292 of the baculovirus capsid protein GP64, and in an acidic environment, GP64 binds heparin, especially at pH 6.2, where heparin-binding is highest [38].
Protein-protein interaction (PPI) analysis of FABP1, PIC and SEC61 using STRING (https://string-db.org/, accessed on 15 March 2022). PPI analysis revealed that FABP1 interacts with BGIBMGA013486-TA, which belongs to the ubiquitin-binding enzyme family E2 (Figure 9a), and E3 ubiquitin protein-linked enzymes have been shown to promote the proliferation of BmNPV by facilitating the fusion of its capsid protein GP64 with the host cell membrane. E2 and E3 can interact with each other to transfer ubiquitin to target cells. Therefore, it is speculated that FABP1 may have a competitive relationship with E3, which in turn inhibits the replication of BmNPV. SEC61 plays an important role as a transporter protein with ribosomal proteins and is mainly involved in protein translocation across membranes (Figure 9b). It may be involved in protein translocation in BmNPV replication. PIC belongs to the mitochondrial carrier (TC 2.A.29) family, which is mainly involved in mitochondrial activities, including ATP synthesis coupled proton transport, and may provide energy for the replication of BmNPV (Figure 9c). Insects 2022, 13, x FOR PEER REVIEW 13 of 15

Conclusions
In summary, the present study obtained a large amount of protein data related to BmNPV invasion and replication. This provides a protein data reference for the screening of BmNPV receptors and the study of proteins that play a key role in the replication of BmNPV. Moreover, it provides a basis for subsequent studies on FABP1, SEC61, and PIC proteins.
Supplementary Materials: The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Table S1: protein data of BmN; Table S2: protein data of silkworms; Table  S3: Bioinformatics Analysis of the BmN; Table S4: Bioinformatics Analysis of the silkworms.

Conclusions
In summary, the present study obtained a large amount of protein data related to BmNPV invasion and replication. This provides a protein data reference for the screening of BmNPV receptors and the study of proteins that play a key role in the replication of BmNPV. Moreover, it provides a basis for subsequent studies on FABP1, SEC61, and PIC proteins.