Influenza virus hemagglutinin and neuraminidase, but not the matrix protein, are required for assembly and budding of plasmid-derived virus-like particles

Abstract

For influenza virus, we developed an efficient, noncytotoxic, plasmid-based virus-like particle (VLP) system to reflect authentic virus particles. This system was characterized biochemically by analysis of VLP protein composition, morphologically by electron microscopy, and functionally with a VLP infectivity assay. The VLP system was used to address the identity of the minimal set of viral proteins required for budding. Combinations of viral proteins were expressed in cells, and the polypeptide composition of the particles released into the culture media was analyzed. Contrary to previous findings in which matrix (M1) protein was considered to be the driving force of budding because M1 was found to be released copiously into the culture medium when M1 was expressed by using the vaccinia virus T7 RNA polymerase-driven overexpression system, in our noncytotoxic VLP system M1 was not released efficiently into the culture medium. Additionally, hemagglutinin (HA), when treated with exogenous neuraminidase (NA) or coexpressed with viral NA, could be released from cells independently of M1. Incorporation of M1 into VLPs required HA expression, although when M1 was omitted from VLPs, particles with morphologies similar to those of wild-type VLPs or viruses were observed. Furthermore, when HA and NA cytoplasmic tail mutants were included in the VLPs, M1 failed to be efficiently incorporated into VLPs, consistent with a model in which the glycoproteins control virus budding by sorting to lipid raft microdomains and recruiting the internal viral core components. VLP formation also occurred independently of the function of Vps4 in the multivesicular body pathway, as dominant-negative Vps4 proteins failed to inhibit influenza VLP budding.

FIG. 1.

Protein expression levels and budding efficiencies of virus and VLPs. VLPs (A) or virions (B) were prepared by transfecting or infecting 293T cells as described in Materials and Methods. At 48 h p.t. or 20 h p.i., the culture media and cells were harvested. The cell lysate (cell, lane 1), material pelleted through a 30% sucrose cushion (30% sucr., lane 2), and fractions collected following flotation of VLPs or virions through a sucrose density gradient (lanes 3 to 8) were analyzed by SDS-PAGE followed by immunoblotting to detect viral proteins. NP appears as a doublet, which probably reflects the known proteolytic cleavage of NP in infected cells (68). (C) The budding efficiencies of VLPs were compared to those of virus by metabolically labeling transfected or infected cells with ³⁵S-Promix and harvesting the culture media after 20 h. The amount of radiolabeled M1 released into the culture media and pelleted through a 30% sucrose cushion was quantified by using Image Gauge software. Relative M1 release was determined by dividing the amount of M1 present in the 30% sucrose pellet (P) by the total amount of M1 present in the pellet and cell lysate (C). Only the full-length M1 band, and not the clipped form present in the infected cells (unpublished observation), was used for quantification. Error bars represent standard deviations from the averages for three experiments.

FIG. 2.

Glycoprotein release from 293T cells. HA, NA, and M2 proteins were expressed in 293T cells in combinations as indicated. The culture medium was harvested at 48 h p.t. and pelleted through a 30% sucrose cushion, and cells were prepared as described in Materials and Methods. Samples were analyzed by SDS-PAGE followed by immunoblotting to detect viral proteins. Where indicated, exogenous bacterial NA (exo NA) was added to the replacement media following transfection.

FIG. 3.

M1 protein release from 293T cells driven by glycoprotein expression. HA, NA, M1, M2, polymerase proteins (Pol; PB1, PB2, and PA), NP, and NS2 (NEP) were expressed in 293T cells in combinations as indicated. The culture medium was harvested at 48 h p.t. and pelleted through a 30% sucrose cushion, and cells were prepared as described in Materials and Methods. Samples were analyzed by SDS-PAGE followed by immunoblotting to detect viral proteins. Where indicated, exogenous bacterial NA (exo NA) was added to the replacement media following transfection. M1 protein in the 30% sucrose pellet was quantified by using the Odyssey infrared imaging system and normalized to the amount of M1 protein found in the cell lysate. The amount of M1 released from cells expressing all VLP proteins (lane 10) was set to 1.0.

FIG. 4.

Release of HA and M1 proteins expressed by vTF7-3 infection/transfection in CV1 and HeLa T4 cells. HA and M1 were expressed in CV1 (A) or HeLa T4 (B, upper) cells by vTF7-3-expressed, T7-RNA polymerase-driven expression, according to the protocol of Gómez-Puertas et al. (18). At 60 h p.t., the culture medium was harvested and pelleted through a 30% sucrose cushion and cells were prepared as described in Materials and Methods. Where indicated, exogenous bacterial NA (exo NA) was added to the replacement medium at 20 h p.t. HeLa T4 cells (B, lower) were also transfected with HA, NA, and M1 protein expression plasmids, and the culture media and cells were analyzed as described above. Samples were analyzed by SDS-PAGE followed by immunoblotting to detect viral proteins.

FIG. 5.

Budding of VLPs with protein omissions. VLPs were prepared by transfecting 293T cells as described in Materials and Methods. Where indicated, plasmids encoding individual proteins were omitted from the transfection, the plasmids encoding the three polymerase proteins (PB1, PB2, and PA) were omitted (Pol−), or exogenous bacterial NA (exo NA) was added to the replacement media following transfection. The culture medium was harvested at 48 h p.t. and pelleted through a 30% sucrose cushion, and cells were prepared as described in Materials and Methods. Samples were analyzed by SDS-PAGE followed by immunoblotting to detect viral proteins. M1 protein in the 30% sucrose pellet was quantified by using the Odyssey infrared imaging system and normalized to the amount of M1 protein found in the cell lysate. The amount of M1 released from cells expressing all VLP proteins (WT, lane 1) was set to 1.0. Error bars represent standard deviations from the averages for three experiments.

FIG. 6.

Morphologies of VLPs as shown by EM. Virions or VLPs containing all VLP proteins (WT VLP) or lacking M1 protein (M1− VLP) were prepared by infecting or transfecting 293T cells as described in Materials and Methods. The culture medium was harvested at 20 h p.i. or 48 h p.t. and pelleted through a 30% sucrose cushion. The pellet was resuspended in NTE and prepared for EM. Virions and VLPs were immunostained with a monoclonal anti-HA antibody followed by IgG conjugated to 15-nm gold and then negative stained. Bar, 100 nm.

FIG. 7.

Release of VLPs with mutant glycoproteins or VSV G protein. VLPs were prepared by transfecting 293T cells as described in Materials and Methods. Where indicated, plasmids encoding wt HA or wt NA were replaced by plasmids encoding HAt−, NAt−, or HA-SSS mutant proteins (A), or plasmids encoding HA and NA protein were omitted and replaced by increasing amounts (0.5 μg, 1.0 μg, and 2.0 μg) of plasmid encoding VSV G protein (B). The culture medium was harvested at 48 h p.t. and pelleted through a 30% sucrose cushion, and cells were prepared as described in Materials and Methods. Samples were resolved by SDS-PAGE followed by immunoblotting to detect viral proteins. M1 protein in the 30% sucrose pellet was quantified by using the Odyssey infrared imaging system and normalized to the amount of M1 protein found in the cell lysate. The amount of M1 released from cells expressing wt VLP proteins (WT, lane 6) was set to 1.0.

FIG. 8.

Infectivities of VLPs with protein omissions. VLPs were prepared by transfecting 293T cells as described in Materials and Methods. Where indicated, plasmids encoding polymerase proteins (PB1, PB2, and PA), NP, NS2 (NEP), or M1 were omitted. The culture medium containing VLPs was harvested at 48 h p.t. and treated with trypsin. VLPs that were not treated with trypsin were used as a negative control. VLPs were transferred to target 293T cells expressing PB1, PB2, PA, and NP. Twenty-four hours later, the target cells were fixed and GFP mean fluorescence intensity (MFI) was quantified by flow cytometry. The MFI of target cells overlaid with complete VLPs (WT) was set to 100%. Error bars represent standard deviations from the averages for three experiments.

FIG. 9.

Effect of dominant-negative Vps4 proteins on influenza VLP, HIV-1, PIV-5 VLP, and VSV budding. (A) HIV-1 budding in the presence of Vps4A-KQ or Vps4B-KQ proteins was examined as described previously (17, 62). (B and C) Influenza VLPs were prepared by transfecting 293T cells as described in Materials and Methods. Where indicated, plasmids encoding Vps4A-KQ (B) or Vps4B-KQ (B) or increasing amounts (0.5 μg, 1.0 μg, and 2.0 μg) of plasmid encoding Vps4A-EQ (C) were included in the transfection. The culture medium containing VLPs was harvested at 48 h p.t. and pelleted through a 30% sucrose cushion, and cells were prepared as described in Materials and Methods. Samples were resolved by SDS-PAGE followed by immunoblotting to detect viral proteins and Vps4 proteins. M1 protein in the 30% sucrose pellet was quantified by using the Odyssey infrared imaging system and normalized to the amount of M1 protein found in the cell lysate. The amount of M1 released from cells transfected with VLP plasmids and empty vector (lane 5) was set to 1.0. (D) PIV-5 VLP budding and VSV budding in the presence of increasing amounts (0.5 μg and 1.0 μg) of plasmid encoding Vps4A-EQ were performed essentially as described previously (25, 55). The relative amounts of PIV-5 and VSV M proteins released into the media and detected in the 30% sucrose pellet were quantified as for panel C. (E) Infectivities of influenza VLPs prepared as for panel C were analyzed as described in the legend to Fig. 8 and Materials and Methods.