Baculovirus GP64-Mediated Entry into Mammalian Cells

Abstract

The baculovirus Autographa californica multiple nucleopolyhedrovirus (AcMNPV) serves as an efficient viral vector, not only for abundant gene expression in insect cells, but also for gene delivery into mammalian cells. Lentivirus vectors pseudotyped with the baculovirus envelope glycoprotein GP64 have been shown to acquire more potent gene transduction than those with vesicular stomatitis virus (VSV) envelope glycoprotein G. However, there are conflicting hypotheses about the molecular mechanisms of the entry of AcMNPV. Moreover, the mechanisms of the entry of pseudotyped viruses bearing GP64 into mammalian cells are not well characterized. Determination of the entry mechanisms of AcMNPV and the pseudotyped viruses bearing GP64 is important for future development of viral vectors that can deliver genes into mammalian cells with greater efficiency and specificity. In this study, we generated three pseudotyped VSVs, NPVpv, VSVpv, and MLVpv, bearing envelope proteins of AcMNPV, VSV, and murine leukemia virus, respectively. Depletion of membrane cholesterol by treatment with methyl-β-cyclodextrin, which removes cholesterol from cellular membranes, inhibited GP64-mediated internalization in a dose-dependent manner but did not inhibit attachment to the cell surface. Treatment of cells with inhibitors or the expression of dominant-negative mutants for dynamin- and clathrin-mediated endocytosis abrogated the internalization of AcMNPV and NPVpv into mammalian cells, whereas inhibition of caveolin-mediated endocytosis did not. Furthermore, inhibition of macropinocytosis reduced GP64-mediated internalization. These results suggest that cholesterol in the plasma membrane, dynamin- and clathrin-dependent endocytosis, and macropinocytosis play crucial roles in the entry of viruses bearing baculovirus GP64 into mammalian cells.

INTRODUCTION

Viruses represent highly evolved natural vectors for the transfer of foreign genetic information into cells (74). This attribute has enabled the engineering of recombinant viral vectors based on retrovirus, lentivirus, adenovirus, adeno-associated virus, herpesvirus, and baculovirus for the delivery of the foreign genes into target tissues. While substantial progress has been made, further vector refinement is required to overcome cytotoxicity, the induction of host immune responses, and the expression of viral genes before such vectors can be approved for clinical use for any individual disorder.

Baculovirus is an insect virus possessing a large double-stranded circular DNA genome packaged into a rod-shaped capsid. Among the numerous baculoviruses, Autographa californica multiple nucleopolyhedrovirus (AcMNPV) is the species most frequently used for baculovirus studies. Although AcMNPV has long been used as an efficient gene expression vector in insect cells (41, 45), recombinant baculoviruses were shown to be capable of entering into various mammalian cells without any replication and of expressing foreign genes under the control of mammalian promoters (11, 17, 65, 70, 71). Therefore, baculovirus is now recognized as a useful viral vector, not only for abundant gene expression in insect cells, but also for gene delivery into mammalian cells. In addition to achieving efficient gene delivery, AcMNPV has also been shown to stimulate host antiviral immune responses in mammalian cell lines (1–3, 24, 33, 63, 73, 77) and to confer protection from lethal virus infection (3, 24) and progressive tumor metastasis in mice (33). Furthermore, baculovirus has been utilized as both a ubiquitous and a specific gene transfer vector in the form of a recombinant virus bearing foreign proteins on the viral surface in addition to GP64, which is the major envelope glycoprotein of AcMNPV (70, 71), and of a pseudotyped virus displaying ligands of interest alone without GP64 (32), respectively.

The mechanism of entry of AcMNPV has been studied mainly in insect cells by using the extracellular budded viruses. AcMNPV was shown to become internalized by receptor-mediated endocytosis (26, 49, 58), and GP64 is initially involved in virus attachment on the cell surface. In a recent study, heparan sulfate was shown to be essential for gene transduction by baculovirus into mammalian cells (78). After receptor-mediated internalization of AcMNPV (26), GP64 undergoes conformational change into a fusion-competent state at low pH, and the nucleocapsid is released into the cytoplasm after cell fusion (14, 43, 63, 73). Furthermore, GP64-null viruses exhibited approximately 98% reduction in viral budding (7), indicating that GP64 is involved, not only in viral entry, but also in the egress of viral particles from infected cells.

Although the mechanisms of entry of AcMNPV into not only insect cells, but also mammalian cells, have not been well characterized, previous reports suggested that AcMNPV enters cells through a clathrin-dependent (40, 44) or -independent (36) endocytic pathway. Moreover, it was shown recently that AcMNPV enters cells through direct fusion with the plasma membrane at low pH (7, 19). We reported previously that the entry of AcMNPV into mammalian cells was inhibited by treatment with purified lipids and was reduced in mutant hamster cell lines deficient in phospholipid synthesis (71). Furthermore, we have shown that GP64 interacts with CD55/decay-accelerating factor (DAF) in a lipid raft and confers resistance to serum inactivation (30). These data suggest that GP64's interaction with phospholipids on the cell surface plays an important role in the internalization of AcMNPV into mammalian cells. The lipid raft is a cholesterol-rich microdomain on the cell surface. It is characterized by detergent insolubility; light density; enrichment of cholesterol, glycosphingolipids, and glycosylphosphatidylinositol-anchored protein; and participation in cell surface receptor-mediated signal transduction, protein sorting, and membrane transport (9, 66). The lipid raft has been shown to play an important role in the entry of vaccinia virus (15) and Ebola virus (4), and in not only the entry but also the budding of influenza virus (12, 60) and human immunodeficiency virus (HIV) (55).

In the present study, we investigated the mechanisms of GP64-mediated viral entry into mammalian cells by using vesicular stomatitis virus (VSV) pseudotyped particles bearing various viral envelope proteins, including GP64 of AcMNPV and a recombinant AcMNPV possessing a luciferase gene under the control of a mammalian promoter. Depletion of membrane cholesterol by methyl-β-cyclodextrin (MβCD) impaired GP64-mediated entry in a dose-dependent manner but did not affect attachment to the cell surface. Inhibition of the dynamin-, clathrin-, and macropinocytosis-mediated pathways by using inhibitors and dominant-negative mutants impaired GP64-mediated entry, whereas the inhibition of caveolin-mediated endocytosis had no such effect. These results suggest that cell surface cholesterol, dynamin- and clathrin-mediated endocytoses, and macropinocytosis participate in the internalization of viral particles bearing GP64 into mammalian cells.

MATERIALS AND METHODS

Cells.

Sf9 cells derived from Spodoptera frugiperda were grown in SF900-II medium (Invitrogen, Carlsbad, CA) supplemented with or without 10% fetal bovine serum (FBS) (Sigma, St. Louis, MO) at 27°C. Huh7 and 293T cell lines were cultured in Dulbecco's modified Eagle's medium (DMEM) (Sigma) supplemented with 10% FBS.

Antibodies.

Anti-GP64 monoclonal antibodies (AcV1 and AcV5) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Rabbit anti-VP39 polyclonal antibody was prepared by intradermal injection of the synthetic peptides of amino acid residues 37 to 49 (SPDAYHDDGWFIC) of AcMNPV VP39 into a Japanese white rabbit purchased from Kitayama Laboratories (Nagano, Japan). Antibodies to caveolin-1, calreticulin, β-actin, and the FLAG tag (M2) were purchased from Sigma. Antibodies to the hemagglutinin (HA) tag and green fluorescent protein (GFP) were purchased from Covance (Richmond, CA) and Clontech, respectively.

Plasmids.

The cDNA encoding AcMNPV GP64 was generated by PCR, cloned into pCAGGS/MCS-PM (56), and designated pCAGGP64. pCAGVSVG, a plasmid encoding VSV-G protein under the control of the CAG promoter, was constructed as described previously (69). HA-tagged dynamin 2 (Dyn2), a dominant-negative Dyn2 (K44A) in which lysine⁴⁴ is replaced with alanine, FLAG-tagged epidermal growth factor (EGF) receptor substrate 15 (Eps15), a dominant-negative Eps15 (Δ95-295), Glu/Glu (EE)-tagged Rab34, and a dominant-negative Rab34 (T66N) in which threonine⁶⁶ is replaced with asparagine were cloned into pcDNA3.1. Caveolin-1 (CAV1) and a dominant-negative mutant (CAV1^DGV) were cloned into pAcGFP-C1 (Clontech, Mountain View, CA) (61). pSilencer-CAV1, carrying a short hairpin RNA (shRNA) targeted to CAV1 under the control of the U6 promoter, was constructed by cloning of the oligonucleotide pair 5′-GATCCGATTGACTTTGAAGATGTGTTCAAGAGACACATCTTCAAAGTCAATCTTTTTTGGAAA-3′ and 5′-AGCTTTTCCAAAAAAGATTGACTTTGAAGATGTGTCTCTTGAACACATCTTCAAAGTCAATCG-3′ between the cloning sites of pSilencer 2.1-U6 hygro (Ambion, Austin, TX). The pSilencer negative-control plasmid (Ambion) has no homology to any human gene. The plasmids used in this study were confirmed by sequencing with ABI Prism genetic analyzer 3130 (Applied Biosystems, Tokyo, Japan).

Production of pseudotyped VSVs and recombinant AcMNPV.

Pseudotyped VSVs and recombinant baculoviruses were generated as described previously (30). To generate pseudotyped VSVs bearing the GP64 protein of AcMNPV (NPVpv), the G protein of VSV (VSVpv), and the GP70 protein of murine leukemia virus (MLVpv), 293T cells were transfected with pCAGGP64, pCAGVSVG, and pFBASALF (provided by T. Miyazawa, Kyoto University), respectively; they were then infected with VSVΔG/Luc-*G, in which the G gene was replaced with the luciferase gene and was pseudotyped with the G protein (69), at a multiplicity of infection (MOI) of 0.1. The virus was adsorbed for 2 h at 37°C and extensively washed four times with serum-free DMEM. After 24 h of incubation at 37°C with DMEM-10% FBS, the culture supernatants were centrifuged to remove cell debris and stored at −80°C. The resulting VSV pseudotyped viruses transiently display heterologous or homologous envelope proteins. AcCAluc was generated by insertion of a luciferase gene under the control of the CAG promoter (40). Schematic representations of the recombinant baculovirus and the pseudotyped VSVs used in this study are shown below (see Fig. 2A). Recombinant baculovirus AcGP64TC, possessing a tetracysteine (TC) tag (Cys-Cys-Pro-Gly-Cys-Cys) at the N terminus of GP64, was generated by using a transfer vector, pAc-SURF2-TC, in which the TC tag sequence was cloned in frame under the signal sequence of GP64 of pBACsurf-1 (Merck KGaA, Darmstadt, Germany) by using the KpnI site (see Fig. 7A). AcGP64TC was then inoculated into Sf9 cells at an MOI of 0.1 and cultured in the presence of FlAsH-EDT2 (Invitrogen) (23). Culture supernatants were collected at 4 days postinfection, purified by sucrose gradient ultracentrifugation, and stored at −80°C.

Fig 2.

Cholesterol plays an important role in the internalization of baculovirus and the pseudotyped VSV bearing GP64. (A) Schematic representations of the recombinant baculovirus and the pseudotyped VSVs used in this study. Recombinant baculovirus AcCAluc has a luciferase gene under the control of the CAG promoter. The pseudotyped VSVs, NPVpv, VSVpv, and MLVpv, have a luciferase gene in place of the G gene and transiently display heterologous or homologous envelope proteins. (B) Huh7 cells were treated with various concentrations of MβCD for 30 min, followed by infection with AcCAluc (MOI, 5), NPVpv (MOI, 0.5), or VSVpv (MOI, 0.5), and the infectivity of each virus was determined in relative luciferase units (RLU) at 24 h postinfection. (C) Huh7 cells treated with 5 mM MβCD were replenished with various concentrations of water-soluble cholesterol for 30 min, followed by infection with the viruses; the infectivity of each virus was determined in RLU at 24 h postinfection. (D) Huh7 cells treated with 5 mM MβCD were incubated with AcCAluc at an MOI of 50 for 30 min. GP64 was stained with anti-GP64 antibody (AcV5) and Alexa Fluor 488-conjugated goat anti-mouse immunoglobulin G antibody after fixation with 4% paraformaldehyde, followed by permeabilization with 0.1% Triton X-100. Nuclei were stained with DAPI (4′,6-diamidino-2-phenylindole). (E) Huh7 cells were treated with various concentrations of lovastatin for 24 h, followed by inoculation with the viruses, and RLU and cell viability were determined at 24 h postinfection. The results are the averages of three independent assays, with the error bars representing SD. The bar and line graphs indicate RLU and cell viability, respectively. *, P < 0.05.

Fig 7.

Live images of internalization of the recombinant baculovirus into mammalian cells. (A) AcGP64TC incorporates the TC tag sequence (Cys-Cys-Pro-Gly-Cys-Cys) under the signal peptide sequence (SP) in the N terminus of the mature GP64. The flanking amino acid sequences are -Ala-Asp-Leu-Gln- and -Pro-Arg-Ala-Glu-. Sf9 cells were infected with AcGP64TC at an MOI of 0.1 and cultured in the presence of FlAsH-EDT2 for 4 days, and the labeled viral particles were incubated with Huh7 cells at 4°C for 30 min, together with Texas Red-labeled dextran as a marker of macropinocytosis. The temperature was then raised to 37°C for examination of the internalization of the particles in the living cells by confocal microscopy. (B) Time-lapse images of the internalization of FlAsH-labeled AcGP64TC (green), together with dextran (red), into Huh7 cells. Internalization of the particles into cells was viewed at 37°C by confocal microscopy.

Reagents.

Chlorpromazine hydrochloride (CPZ), dynasore, lovastatin, MβCD, water-soluble cholesterol, and 5-N-ethyl-N-isopropyl amiloride (EIPA) were obtained from Sigma. Cells were preincubated with the drugs for 30 min (except for lovastatin) or 24 h (lovastatin) and infected with AcCAluc or pseudotyped VSV. In the MβCD treatment and cholesterol replenishment experiments, cells were cultured in DMEM with 10% lipid-free serum (BioWest, Nuaille, France). In the cholesterol replenishment experiments, cells were washed and incubated with DMEM containing water-soluble cholesterol for 30 min after treatment with Texas Red-labeled dextran, a marker of macropinocytosis (Invitrogen).

Determination of cellular cholesterol.

MβCD-treated cells were washed with phosphate-buffered saline (PBS), and cellular cholesterol was determined as follows. Huh7 cells (10⁶ cells/100 μl of PBS) were lysed by three cycles of freeze-thawing followed by sonication in a water bath sonicator (three bursts of 20 s each at 4°C). Cholesterol was extracted from the cell lysates by adding chloroform (200 μl) and methanol (100 μl) to the sonicated lysates (100 μl), and the bottom layer was collected and evaporated. The residual cholesterol was dissolved in ethanol and assayed using the Amplex Red cholesterol kit (Invitrogen).

Quantitative PCR.

Entry and binding of baculovirus to cells were determined by quantitative PCR (qPCR). AcCAluc was incubated with cells for 30 min at 4°C or 37°C and then washed three times with PBS with or without 0.1% trypsin. The amount of baculovirus genome in the total cellular DNA (20 ng) was determined with Platinum SYBR green qPCR SuperMix (Invitrogen) using GP64-specific primers (5′-AGCTGATGTACGAAAACGAT-3′ and 5′-TCGTGCAGCATATTGTTTAG-3′). Viral genomes extracted from Huh7 cells incubated with AcCAluc, those from cells washed three times with 0.1% trypsin-PBS after virus incubation, and those from mock-infected cells were used as a positive control (PC) and negative control 1 (NC1) and NC2, respectively.

Immunofluorescent staining.

Cells cultured to be 50% confluent on a chambered slide were treated with chemicals or transfected with the plasmids and then inoculated with the viruses after incubation at 37°C for 24 h. The cells were fixed at 4°C for 30 min with 4% paraformaldehyde in PBS, permeabilized with 0.2% Triton X-100 in PBS for 5 min, and blocked with 1% bovine serum albumin (BSA) in PBS for 1 h at room temperature. The cells were reacted with appropriate primary antibodies diluted with 1% BSA in PBS for 1 h, followed by secondary antibodies; washed three times with PBS; and observed with a FluoView FV1000 laser scanning confocal microscope (Olympus, Tokyo, Japan).

Time-lapse microscopy.

Huh7 cells were seeded at 1 × 10³ cells in a μ-Dish (iBidi, Munich, Germany), grown overnight to 50% confluence, and incubated with FlAsH-labeled AcGP64TC at an MOI of 50 and with Texas Red-labeled dextran (2 mg/ml) at 4°C for 30 min. Internalization of the particles into living cells was observed at 37°C by laser scanning confocal microscopy. Digital images were analyzed with Image-Pro software (Media Cybernetics, Bethesda, MD).

Flotation assay.

The flotation assay was described previously (30). Briefly, 2 million Huh7 cells were infected with AcCAluc or incubated with GP64ΔTM, lacking a transmembrane region, which was expressed in insect cells and purified as described previously (2) at 4°C. The cells were washed three times with ice-cold PBS, harvested with a rubber policeman, suspended in 500 μl of ice-cold TNE lysis buffer (25 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, 0.1% Triton X-100), and incubated on ice for 30 min. The cell lysates were mixed with OptiPrep (Sigma) to 1 ml of a final concentration of 35%. This mixture was overlaid with 1 ml of 30%, 25%, and 10% OptiPrep, after which 0.5 ml of the lysis buffer was centrifuged at 40,000 rpm at 4°C for 16 h in an SW55Ti rotor (Beckman Coulter, Fullerton, CA). Each fraction was collected at 0.5 ml from the top of the centrifuge tube at 4°C. The fractions were subjected to immunoblotting.

Transfection and immunoblotting.

The plasmids were transfected into cells using the FuGene 6 Transfection Reagent (Roche, Mannheim, Germany) according to the manufacturer's instructions. The protein samples were subjected to 12.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The proteins were transferred to an Immobilon-P Transfer Membrane (Millipore, Tokyo, Japan) and were reacted with the appropriate antibodies. The immune complexes were visualized with Super Signal West Femto substrate (Pierce, Rockford, IL) and detected by an LAS-3000 image analyzer system (Fujifilm, Tokyo, Japan).

Statistical analysis.

The results were expressed as means ± standard deviations. The significance of differences in the means was determined by Student's t test.

RESULTS

Cholesterol participates in the internalization of baculovirus into mammalian cells through direct interaction with GP64.

A lipid raft is a microdomain on the cellular membrane and is sensitive to treatment with MβCD, which is a lipophilic cyclic oligosaccharide that is known to remove cholesterol from cellular membranes and to disrupt lipid raft formation (29). We have previously shown that GP64 expressed in mammalian cells was localized on the lipid rafts and that pseudotyped viruses bearing GP64 incorporated human decay-accelerating factor (DAF) during the budding process in mammalian cells (30). DAF is a membrane protein that regulates the complement system on the cell surface. To examine the direct interaction between GP64 and the lipid rafts during the entry process, a flotation assay was carried out by mixing purified GP64ΔTM protein, which lacks a transmembrane region, with Huh7 cells. The GP64ΔTM was detected in fractions 3 to 6, which overlapped partially with those containing caveolin-1, a marker of lipid rafts (Fig. 1A), suggesting that the GP64ΔTM directly interacts with lipid rafts on the cell surface. We also performed this assay using AcCAluc, and GP64 was observed in the lipid raft fraction (data not shown). Next, to examine the role of cholesterol in the binding and internalization of baculovirus and pseudotyped viruses, Huh7 cells were treated with various concentrations of MβCD and then inoculated with AcCAluc bearing a luciferase gene under the CAG promoter. Treatment with MβCD, a chemical that depletes cholesterol (29), reduced the cellular cholesterol in Huh7 cells in a dose-dependent manner (Fig. 1B). Treatment with MβCD up to 2 mM exhibited no effect on virus binding, but 5 mM MβCD treatment showed 30% reduction. However, 5 mM MβCD treatment induced 20% reduction in living cells, as shown in Fig. 1B. In contrast, cholesterol reduction by MβCD treatment inhibited the internalization of AcCAluc into Huh7 cells in a dose-dependent manner (Fig. 1D). MβCD treatment clearly inhibited the internalization of AcCAluc in a dose-dependent manner, whereas 5 mM MβCD treatment exhibited a maximum 30% reduction in viral attachment and no dose dependency, suggesting that cholesterol participates mainly in the internalization of AcMNPV and slightly in its binding to mammalian cells.

Fig 1.

Cholesterol participates in the internalization of baculovirus into mammalian cells through direct interaction with GP64. (A) Huh7 cells were incubated with the purified GP64ΔTM at 4°C for 1 h, lysed with ice-cold lysis buffer, and subjected to a flotation assay. Aliquots of each fraction were assayed by Western blot analysis, which detected GP64ΔTM, caveolin-1, a lipid raft marker, and calreticulin, an endoplasmic reticulum marker. (B) The amount of cholesterol and the viability of Huh7 cells treated with various concentrations of MβCD for 30 min. The open diamonds and closed circles indicate the cholesterol content and cell viability, respectively. (C) Binding of AcCAluc to cholesterol-depleted Huh7 cells. Huh7 cells treated with various concentrations of MβCD were inoculated with AcCAluc at an MOI of 5, incubated at 4°C for 30 min, and washed three times with PBS, after which quantitative PCR determined the baculovirus genome in the total cellular DNA extracted from the cells (20 ng). PC represents viral genomes extracted from untreated Huh7 cells incubated with AcCAluc, NC1 represents viral genomes extracted from untreated cells washed three times with 0.1% trypsin-PBS after incubation with virus, and NC2 represents viral genomes extracted from mock-infected cells. ND, not detected. (D) Internalization of AcCAluc in MβCD-treated Huh7 cells. Huh7 cells treated with various concentrations of MβCD were inoculated with AcCAluc at an MOI of 5, incubated at 37°C for 30 min, and washed three times with 0.1% trypsin-PBS to remove the viral particles attached to the cell surface, after which quantitative PCR determined the intracellular viral genomes in the total cellular DNA extracted from the cells (20 ng). The results are the averages of three independent assays, with the error bars representing standard deviations (SD). *, P < 0.05.

Cholesterol plays important roles in the internalization of baculovirus and the pseudotyped VSV bearing GP64.

To examine the role of cholesterol in GP64-mediated entry, Huh7 cells were treated with various concentrations of MβCD, and the viruses were then inoculated. MβCD treatment suppressed luciferase expression in cells inoculated with AcCAluc and NPVpv, but not with VSVpv, in a dose-dependent manner (Fig. 2B). Gene transduction in Huh7 cells upon infection with AcCAluc and NPVpv was reduced by approximately 50% and 90% by treatment with 2 mM and 5 mM MβCD, respectively. To confirm cholesterol's effect on the entry of the viruses, Huh7 cells treated with 5 mM MβCD were incubated with DMEM containing various concentrations of water-soluble cholesterol for 30 min and inoculated with the viruses. Internalization of viral DNA was recovered with AcCAluc, and NPVpv in the cholesterol-depleted cells was recovered by the replenishment of cholesterol in a dose-dependent manner (Fig. 2C). Next, we examined the effect of cholesterol on the entry of baculovirus by immunofluorescence microscopy (Fig. 2D). Although MβCD treatment dramatically reduced the detected level of GP64 in the cytoplasm, replenishment with cholesterol returned the level to comparability with that of the untreated cells. Finally, to confirm the effect of cholesterol on the internalization of baculovirus, Huh7 cells were treated with lovastatin, an inhibitor of cholesterol biosynthesis. Lovastatin treatment reduced the luciferase expression in cells inoculated with AcCAluc and NPVpv, but not in those inoculated with VSVpv, in a dose-dependent manner (Fig. 2E). Collectively, these results suggest that cholesterol plays an important role in the internalization of AcMNPV and the pseudotyped viruses bearing GP64 into mammalian cells.

Baculovirus and pseudotyped VSV bearing GP64 internalize into mammalian cells through an endosome.

There are conflicting reports on the entry of AcMNPV suggesting that baculovirus internalizes into cells through a clathrin-dependent endocytic pathway (40, 44) or a clathrin-independent and phagocytosis-like endocytic pathway (36). To determine whether AcMNPV and the pseudotyped viruses internalize into cells through endosomes, Huh7 cells were treated with ammonium chloride (NH₄Cl), an inhibitor of endosome acidification. Luciferase expression in cells inoculated with AcCAluc, NPVpv, and VSVpv was reduced by treatment with NH₄Cl in a dose-dependent manner, in contrast to cells infected with MLVpv, which enters cells through a direct fusion between viral and plasma membranes at neutral pH (Fig. 3A). Next, to determine the subcellular localization of baculovirus, infected Huh7 cells expressing GFP-fused RhoB, an endosome marker, were stained with anti-GP64 antibody (AcV5). The colocalization of GP64 and RhoB was detected by confocal microscopy (Fig. 3B). These results support the notion that AcMNPV and the pseudotyped VSV bearing GP64 internalize into mammalian cells through endocytosis.

Fig 3.

Baculovirus is internalized into mammalian cells through an endosome. (A) Huh7 cells were treated with various concentrations of NH₄Cl for 30 min, followed by infection with AcCAluc (MOI, 5), NPVpv (MOI, 0.5), VSVpv (MOI, 0.5), or MLVpv (MOI, 0.5), and RLU and cell viability were determined at 24 h postinfection. The results are the averages of three independent assays, with the error bars representing SD. The bar and line graphs indicate RLU and cell viability, respectively. *, P < 0.05. (B) Huh7 cells expressing GFP-fused RhoB were infected with AcCAluc at an MOI of 50 for 45 min. GP64 was stained with anti-GP64 antibody (AcV5) and Alexa Fluor 594-conjugated goat anti-mouse immunoglobulin G antibody after fixation with 4% paraformaldehyde, followed by permeabilization with 0.1% Triton X-100. Nuclei were stained with DAPI. The boxed region in the merged image is magnified in the inset.

Dynamin participates in the internalization of baculovirus and the pseudotyped VSV bearing GP64.

We further examined the roles of the molecules essential for endocytosis, including dynamin, clathrin, and caveolin, in the GP64-mediated internalization of viral particles. The role of dynamin, a large GTPase promoting fission of the endocytic membrane and a pivotal factor in endocytosis, was determined by using a dominant-negative mutant of dynamin (K44A) and dynasore, an inhibitor of dynamin. Expression of an HA-tagged dynamin mutant (K44A) inhibited the gene transduction of AcCAluc, NPVpv, and VSVpv, but not that of MLVpv, compared to the empty vector (EV) or the wild-type (WT) HA-tagged dynamin (Dyn2) (Fig. 4A). Huh7 cells that expressed either HA-tagged wild-type or K44A mutant dynamin or that had been treated with various concentrations of dynasore were inoculated with AcCAluc and examined by immunofluorescence microscopy after immunostaining. Expression of the K44A mutant, but not of wild-type dynamin, inhibited the internalization of baculovirus (Fig. 4B), and treatment with dynasore also reduced the internalization of the viruses in a dose-dependent manner (Fig. 4C and D), indicating that dynamin participates in the internalization of baculovirus and the pseudotyped VSV bearing GP64.

Fig 4.

Dynamin participates in the internalization of baculovirus and the pseudotyped VSV bearing GP64. (A) Huh7 cells transfected with EV or plasmids encoding the wild type or a dominant-negative mutant (K44A) of Dyn2 were infected with AcCAluc (MOI, 5), NPVpv (MOI, 0.5), VSVpv (MOI, 0.5), or MLVpv (MOI, 0.5). The RLU and cell viability were determined at 24 h postinfection. The results are the averages of three independent assays, with the error bars representing SD. The bar and line graphs indicate RLU and cell viability, respectively. *, P < 0.05. (B) Huh7 cells were transfected with plasmids encoding either HA-Dyn2 or HA-K44A and were infected with AcCAluc at an MOI of 50 at 24 h posttransfection. HA-Dyn2 and HA-K44A were stained with rabbit antibodies against HA- and Alexa Fluor 594-conjugated goat anti-rabbit immunoglobulin G antibody. GP64 was stained with anti-GP64 antibody (AcV5) and Alexa Fluor 488-conjugated goat anti-mouse immunoglobulin G antibody. Nuclei were stained with DAPI. (C) Huh7 cells were treated with various concentrations of dynasore for 30 min, followed by infection with AcCAluc at an MOI of 50. GP64 was stained with anti-GP64 antibody (AcV5) and Alexa Fluor 488-conjugated goat anti-mouse immunoglobulin G antibody after fixation with 4% paraformaldehyde, followed by permeabilization with 0.1% Triton X-100. Nuclei were stained with DAPI. (D) Huh7 cells treated with various concentrations of dynasore for 30 min were incubated with AcCAluc at an MOI of 50 at 37°C for 30 min, after which cell viability was determined. The cells were washed three times with 0.1% trypsin-PBS to remove the viral particles attached to the cell surface, and intracellular viral genomes in the total cellular DNA extracted from the cells (20 ng) were determined by quantitative PCR. The results are the averages of three independent assays, with the error bars representing SD. The bar and line graphs indicate RLU and cell viability, respectively. *, P < 0.05.

Clathrin-dependent and caveola-independent endocytoses participate in GP64-mediated viral entry into mammalian cells.

Both clathrin- and caveola-dependent endocytoses utilize dynamin. Therefore, we next examined the involvement of these pathways in GP64-mediated entry. The dominant-negative mutant of caveolin-1, CAV1^DGV, was shown to inhibit caveola-mediated endocytosis (61). Expression of CAV1^DGV (Fig. 5A) and the knockdown of caveolin-1 by small interfering RNA (siRNA) targeted to caveolin-1 (Fig. 5B) exhibited no significant effect on gene transduction in Huh7 cells upon infection with AcCAluc and the pseudotyped VSVs, suggesting that caveola-mediated endocytosis is not required for GP64-mediated viral entry. Next, we examined the involvement of clathrin in GP64-mediated entry. Treatment with CPZ, an inhibitor of clathrin-mediated endocytosis, inhibited the gene transduction of AcCAluc, NPVpv, and VSVpv, but not that of MLVpv, in a dose-dependent manner (Fig. 5C). In immunofluorescence analysis, CPZ treatment abrogated the internalization of GP64 into Huh7 cells (Fig. 5D). Eps15, which localizes in the clathrin-coated pit, is a key factor in clathrin-mediated endocytosis (5, 72). Expression of a dominant-negative mutant of Eps15 (Δ95-295) produced a 40% reduction in gene transduction of AcCAluc, NPVpv, and VSVpv but no reduction in that of MLVpv (Fig. 5E). These results suggested that baculovirus and the pseudotyped VSV bearing GP64 internalize into mammalian cells through clathrin-dependent and caveola-independent endocytoses.

Fig 5.

Clathrin-dependent and caveola-independent endocytosis is involved in GP64-mediated viral entry into mammalian cells. (A) (Left) Huh7 cells transfected with plasmids encoding either GFP, GFP-fused CAV1, or GFP-fused CAV1^DGV were infected with AcCAluc (MOI, 5), NPVpv (MOI, 0.5), VSVpv (MOI, 0.5), or MLVpv (MOI, 0.5), and RLU and cell viability were determined at 24 h postinfection. (Right) Expression of the GFP fusion CAV1 proteins was determined by immunoblotting. (B) (Left) The knockdown cells were infected with AcCAluc (MOI, 5), NPVpv (MOI, 0.5), VSVpv (MOI, 0.5), or MLVpv (MOI, 0.5). The RLU were determined at 24 h postinfection. (Right) Huh7 cells transfected with plasmids encoding shRNA targeted to CAV1 (shCAV1) or nonspecific targets (shCTRL) were cultivated for 2 days in the presence of hygromycin and lysed with buffer containing 1% Triton X-100. Expression of CAV1 and actin was detected by immunoblotting. (C) Huh7 cells were treated with various concentrations of CPZ for 30 min, followed by infection with AcCAluc (MOI, 5), NPVpv (MOI, 0.5), VSVpv (MOI, 0.5), or MLVpv (MOI, 0.5), and the RLU and cell viability were determined at 24 h postinfection. (D) Huh7 cells treated with 5 μg/ml of CPZ were incubated with AcCAluc at an MOI of 50 for 30 min. GP64 was stained with anti-GP64 antibody (AcV5) and Alexa Fluor 488-conjugated goat anti-mouse immunoglobulin G antibody after fixation with 4% paraformaldehyde, followed by permeabilization with 0.1% Triton X-100. Nuclei were stained with DAPI. (E) Huh7 cells transfected with EV or plasmids encoding the wild type or a dominant-negative mutant (Δ95-295) of Eps15 were infected with AcCAluc (MOI, 5), NPVpv (MOI, 0.5), VSVpv (MOI, 0.5), or MLVpv (MOI, 0.5). The RLU and cell viability were determined at 24 h postinfection. The results are the averages of three independent assays, with the error bars representing SD. The bar and line graphs indicate RLU and cell viability, respectively. *, P < 0.05.

Macropinocytosis participates in GP64-mediated viral entry into mammalian cells.

Baculovirus is 40 to 50 nm in diameter and 200 to 400 nm in length (64), while the clathrin-coated pit is about 100 to 150 nm in diameter. Therefore, baculovirus particles are likely a little too large for uptake by clathrin-mediated endocytosis. To determine the role of macropinocytosis in the entry of baculovirus, Huh7 cells were treated with various concentrations of EIPA, an inhibitor of Na⁺/H⁺ ion exchange and filopodium formation, after which they were inoculated with the viruses. Treatment with EIPA inhibited the gene transduction of AcCAluc most efficiently, followed by that of NPVpv, in a dose-dependent manner, whereas it had no effect on that of VSVpv and MLVpv (Fig. 6A). Rab34 has been shown to be involved in the formation of macropinosomes (67), and the expression of a dominant-negative mutant of Rab34 (T66N) in Huh7 cells inhibited gene transduction of AcCAluc and NPVpv, but not that of VSVpv or MLVpv (Fig. 6B). In the immunofluorescence assay, treatment with EIPA reduced the internalization of both baculovirus and dextran in a dose-dependent manner (Fig. 6C and data not shown). These results indicate that macropinocytosis also participates in the internalization of baculovirus and that GP64 per se is capable of inducing macropinocytosis.

Fig 6.

Macropinocytosis participates in GP64-mediated viral entry into mammalian cells. (A) Huh7 cells were treated with various concentrations of EIPA for 30 min, followed by infection with AcCAluc (MOI, 5), NPVpv (MOI, 0.5), VSVpv (MOI, 0.5), or MLVpv (MOI, 0.5), and RLU and cell viability were determined at 24 h postinfection. The results are the averages of three independent assays, with the error bars representing SD. *, P < 0.05. (B) Huh7 cells transfected with EV or plasmids encoding the wild type or a dominant-negative mutant (T66N) of Rab34 were infected with AcCAluc (MOI, 5), NPVpv (MOI, 0.5), VSVpv (MOI, 0.5), or MLVpv (MOI, 0.5). The RLU and cell viability were determined at 24 h postinfection. The bar and line graphs indicate RLU and cell viability, respectively. *, P < 0.05. (C) Huh7 cells were treated with 40 μm of EIPA for 30 min, followed by incubation with either AcCAluc at an MOI of 50 or Texas Red-labeled dextran (2 mg/ml). GP64 was stained with anti-GP64 antibody (AcV5) and Alexa Fluor 488-conjugated goat anti-mouse IgG antibody after fixation with 4% paraformaldehyde, followed by permeabilization with 0.1% Triton X-100. Nuclei were stained with DAPI.

Live imaging of the internalization of baculovirus into mammalian cells.

To examine the internalization of baculovirus by live-cell imaging, we generated a recombinant baculovirus possessing tetracysteine-tagged GP64 (AcGP64TC) (Fig. 7A). The labeled tetracysteine tag allows imaging of the location of the protein. The labeled AcGP64TC and Texas Red-labeled dextran were incubated with Huh7 cells. The labeled GP64 was incorporated into cells together with dextran (Fig. 7B) over time. These results support the notion that macropinocytosis also participates in the internalization of baculovirus into mammalian cells.

DISCUSSION

Baculovirus has a large capacity to incorporate foreign genes and exhibits a broad spectrum for internalization into mammalian cells but low cytotoxicity due to a lack of replication; thus, baculovirus is thought to be a promising viral-vector candidate for future human gene therapy (16, 34, 73). Baculovirus GP64 has been shown to participate in attachment to both insect and mammalian cells, as well as in the low-pH-triggered membrane fusion following endocytosis (7, 8, 14, 16, 19, 28, 36, 38, 49, 50, 75). Although the mechanisms of entry of AcMNPV have not been well characterized, previous studies suggested that AcNMPV enters cells through a clathrin-dependent (40, 44) or -independent (36) endocytic pathway. In this study, we utilized a recombinant baculovirus possessing a luciferase gene under the control of a mammalian promoter and pseudotyped viruses bearing envelope glycoproteins of AcMNPV, VSV, and MLV to facilitate more reliable and quantitative analyses of GP64-mediated entry into mammalian cells. The resultant data suggest that cholesterol participates in the internalization of baculovirus but not in its binding to mammalian cells and that baculovirus is internalized into cells through clathrin-dependent but caveola-independent endocytosis or macropinocytosis.

We reported previously that the entry of AcMNPV into mammalian cells was inhibited by treatment with either phosphatidylethanolamine or phosphatidylinositol and was reduced in mutant hamster cell lines deficient in phospholipid biosynthesis (70). We also previously showed that incorporation of human DAF into viral particles through interaction with GP64 in the lipid raft confers resistance to the viral particles bearing GP64 against serum inactivation (30). The lipid raft plays crucial roles in signal transduction, protein sorting, and membrane transport (9, 66). Clustering of the lipid rafts was shown to be essential for the signaling cascade (51, 66). In this study, we have shown that the depletion and inhibition of the biosynthesis of cholesterol by treatment with MβCD and lovastatin, respectively, impaired GP64-dependent internalization but not the binding of the viruses. Phosphatidylserine (PS), which is localized in the inner plasma membrane, is known to be exposed on the envelopes of viral particles of vaccinia virus and HIV (10, 46). Morizono et al. reported that the serum-soluble protein Gas6 binds to both PS on viral particles and TAM receptor tyrosine kinase Axl on target cells and mediates virus binding by bridging the virus to target cells (52). Because Gas6-dependent entry is not limited by a specific interaction between viral envelope proteins and cell surface receptors, this alternative pathway may broaden the host range and enhance the infectivity of various envelope viruses. Infection with pseudotyped Sindbis virus bearing baculovirus GP64 to human microvascular endothelial cells was shown to be enhanced by treatment with Gas6 (51). However, our treatment with Gas6 had no effect on gene transduction by baculovirus and NPVpv into both Huh7 and A549 cells expressing low and high levels of Axl, respectively (data not shown). Differences of pseudotyped virus, VSV based or Sindbis based, might raise this discrepancy. Further studies are needed to clarify the involvement of cholesterol in the lipid raft in the internalization of baculovirus into mammalian cells.

Many enveloped viruses have been shown to internalize into cells through an endocytosis pathway. Clathrin-mediated endocytosis is involved in the internalization of viruses and receptors through a clathrin-coated pit (100 to 150 nm in diameter). The pit buds into the cytosol and delivers a virus first to the early endosomes and then to the late endosomes and lysosomes (25). Many molecules participate in this process, such as Dyn2, Eps15, and accessory protein 2 (13). A number of viruses, including VSV, influenza virus, and hepatitis C virus, are known to use this pathway for entry (6, 54, 68, 76). Conflicting data have been published concerning the involvement of clathrin-dependent endocytosis in the internalization of baculovirus into mammalian cells (36, 40, 44, 63). To obtain more reliable data, in this study, we examined GP64's role in the entry into mammalian cells by using the pseudotyped VSV bearing GP64, together with the pseudotypes possessing envelope glycoproteins of VSV and MLV, as viral controls for entry through the endocytic pathway and direct fusion, respectively. VSV is known to internalize into target cells through clathrin-mediated endocytosis, and the G glycoprotein of VSV exhibits characteristics similar to those of baculovirus GP64 (22). The data obtained in this study support the notion that AcMNPV is internalized into mammalian cells through clathrin-dependent endocytosis (36, 40, 44, 63).

Caveolin-mediated endocytosis, a different form of clathrin-mediated endocytosis, is known to be used by simian virus 40 (57). Caveolae are the caveolin-coated and flask-shaped membrane invaginations associated with lipid rafts (35, 37). Caveolin-1 is typically found in the lipid raft fraction of plasma membranes and is essential for caveola formation through interaction with cholesterol. For a molecule to be internalized through caveolin-mediated endocytosis, it must be no larger than 50 to 80 nm, the diameter of a flask-shaped caveola. Both clathrin- and caveolin-mediated endocytoses require dynamin, a large GTPase promoting the fission of the endocytic membrane (27). The expression of a dominant-negative mutant of caveolin-1 and the suppression of caveolin-1 did not affect the internalization of baculovirus; this is consistent with previous reports suggesting that baculovirus is internalized into mammalian cells through the lipid raft microdomain in the caveola-independent pathway (36, 40, 44).

Macropinocytosis is associated with membrane raffles driven by actin polymerization, such as filopodia and lamelipodia, and has been studied mainly in antigen presentation and the uptake of extracellular fluid and solids, as well as particles, such as bacteria, apoptotic cell fragments, and viruses (21, 31). Viruses, such as Ebola virus, vaccinia virus, coxsackievirus, herpes simplex virus (HSV), and HIV, are reported to use macropinocytosis to enter target cells (18, 42, 47, 48, 53, 62). Macropinocytosis is a transient, growth-factor-induced, actin-dependent endocytic process that leads to the internalization of fluid and membrane in large vacuoles. Like the factors involved in other endocytic mechanisms, structural changes in the plasma membrane during macropinocytosis are tightly regulated by many factors, including the Na⁺/H⁺ exchanger, Rho GTPase family, and protein kinase C (PKC) (48). In this study, macropinocytosis appeared to participate in the internalization of baculovirus into mammalian cells, based on data obtained by treatment with EIPA, an inhibitor of the Na⁺/H⁺ exchanger; by the expression of the dominant-negative mutant of Rab34, which is involved in macropinosome closure; and by the time-lapse microscopic observation of the tetracystein-tagged AcMNPV, in contrast to the previous observation (36). This discrepancy might be attributable to the differences in cell types and assay systems. Furthermore, the expression of a dominant-negative mutant of Rab34 suppressed gene transduction in cells infected, not only with AcCAluc, but also with NPVpv (Fig. 6B), suggesting that the interaction of GP64 with the cellular receptor(s) triggers macropinocytosis, as seen in infection with Ebola virus (53, 62).

The data in this study suggest that baculovirus interacts with the molecules on the lipid raft via GP64 and then is internalized through clathrin-mediated endocytosis and macropinocytosis. This pathway is reminiscent of the entry of EGF receptor, which localizes in the caveola-containing lipid rafts (59), exits from the lipid rafts upon interaction with EGF, and is internalized into cells through clathrin-mediated endocytosis (39). A hypothetical model of baculovirus entry into mammalian cells based on data in the present and previous studies is shown in Fig. 8. In this model, baculovirus binds to a not-yet-identified cellular receptor(s) present in the lipid rafts, this association induces cellular remodeling through signal transduction, and baculovirus is internalized into cells through clathrin-mediated endocytosis and macropinocytosis.

Fig 8.

Putative model of internalization of baculovirus into mammalian cells. Baculovirus binds to a not-yet-identified cellular receptor(s) present in the lipid raft. This association induces cellular remodeling through signal transduction. In clathrin-mediated endocytosis, baculovirus is internalized into the clathrin-coated pit. In macropinocytosis, filopodia formed by actin dynamics wrapped the baculovirus into a macropinosome. The viral genome is released from the endosome or macropinosome through membrane fusion induced by low pH.

Baculovirus is known to use heparan sulfate on the cell surface as an attachment factor (20). In a recent study, Wu and Wang identified the sequence in GP64 responsible for binding to heparin, and this binding is essential for baculovirus internalization into mammalian cells (78). Although there are many studies on baculovirus entry, no baculovirus entry receptors have been identified yet. Since baculovirus is capable of internalizing into various cells originating from not only insects, but also mammals, baculovirus would utilize ubiquitously distributed or multiple molecules to enter the cells. The data presented in this study should provide clues for the future development of baculovirus vectors suitable for efficient and specific gene delivery into target cells and tissues.