Influenza virus-like particles are currently evaluated in clinical trials as vaccine candidates for influenza viruses. Most commonly they are produced in baculovirus- or mammalian- expression systems. Here we used different vaccination schemes in order to systematically compare virus-like particle preparations generated in the two systems. Our work shows significant differences in immunogenicity between the two, and indicates superior and broader immune responses induced by the baculovirus-derived constructs. We demonstrate that these differences critically influence protection and survival in a mouse model of influenza virus infection. Finally, we show that the enhanced immunogenicity of the baculovirus-derived virus-like particles is caused by contamination with residual baculovirus which activates the innate immune response at the site of inoculation.
Citation: Margine I, Martinez-Gil L, Chou Y-y, Krammer F (2012) Residual Baculovirus in Insect Cell-Derived Influenza Virus-Like Particle Preparations Enhances Immunogenicity. PLoS ONE 7(12): e51559. https://doi.org/10.1371/journal.pone.0051559
Editor: Paul Zhou, Pasteur Institute of Shanghai, Chinese Academy of Science, China
Received: August 20, 2012; Accepted: November 5, 2012; Published: December 7, 2012
Copyright: © 2012 Margine et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was partially supported by PATH and CEIRS (Centers for Excellence for Influenza Research and Surveillance, HHSN26620070010C) grants to Dr. Peter Palese. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Vaccination remains the most effective counter measure against infection with influenza viruses , . The seasonal influenza vaccines are effective when well matched with the circulating strain . However, the occurrence of the swine-origin H1N1 virus in 2009 highlighted the need for fast and easy production strategies in the event of a pandemic. Influenza virus-like particles (VLPs) are evaluated as candidate vaccines, since they can be produced in a timely manner in response to a newly occurring influenza virus strain and they can efficiently present the hemagglutinin (HA) protein. Antibodies directed against this immunodominant influenza virus protein block entry and thereby inhibit viral replication.
Influenza VLPs have been produced in various expression systems including plants, mammalian cells, and most prominently, the baculovirus expression system –. VLPs expressed in the latter system are currently being tested in clinical trials in Mexico, and were recently reported to have yielded encouraging results in a phase II trial in 4563 healthy adults , . General advantages of the baculovirus expression over mammalian cell culture-systems include higher yields, lower media costs and higher cellular growth rates. However, it is also associated with several disadvantages, such as non-mammalian-like protein glycosylation and the presence of high titers of contaminating baculovirus particles in the expression supernatants, especially when Sf9 cells are used which support the growth of baculovirus very well . These contaminations are also found in VLP preparations used for vaccination since influenza VLPs and baculovirus virions have very similar densities and cannot be separated efficiently by density gradient ultracentrifugation , –. Mammalian cell-derived influenza VLPs are also being developed as vaccine candidates, and have proved successful in pre-clinical evaluations , , . In the present study we compare influenza A VLPs produced in a mammalian and a baculovirus expression system, in terms of immunogenicity and protective ability in a mouse influenza virus infection model. Since generation of influenza VLPs has been described by co-expression of HA with the influenza matrix 1 (M1) protein , as well as with the retroviral Gag protein , we also evaluate the effect of the budding partner on immunogenicity. We find dramatic differences between the immune responses triggered by the different preparations in terms of anti-HA antibody titers, hemagglutination inhibition (HI) titers, antibody isotype profiles, cytokine induction by the antigen formulation and, finally, survival upon challenge with an influenza A virus. Interestingly, the differences in all these aspects correlate with the presence or absence of residual baculovirus in the preparations.
Materials and Methods
Cells, Plasmids and Viruses
Sf9 insect cells (ATCC# CRL-1711) were grown in TNM-FH media (Gemini Bioproducts) supplemented with 0.1% (w/v) Pluronic F-68 (Sigma), 100 U/mL penicillin and streptomycin (Gibco) and 5% (v/v) fetal calf serum (FCS). BTI-TN-5B1-4 (High Five) cells  were grown in Hyclone SFX media (Fisher Scientific). 293T and MDCK cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and were maintained either in Dulbecco’s Modified Eagle Medium (DMEM) or in Minimum Essential Medium (MEM) (Gibco, Invitrogen) supplemented with 10% FCS (HyClone; Thermo Scientific) and penicillin and streptomycin (Gibco, Invitrogen). Mouse adapted influenza virus strain A/Puerto Rico/8/1934 (PR8) was grown in 10 day old embryonated chicken eggs for 2 days at 37°C and titered by plaque assay on MDCK cells. Expression plasmids encoding for M1, eGFP-Gag (Gag), PR8 HA and influenza B/Yamagata/16/1988 neuraminidase were constructed as described elsewhere –.
Recombinant Baculovirus Generation
Coding sequences for HA and M1 of PR8 and of eGFP-Gag were amplified by polymerase chain reaction (PCR) and cloned into a modified version of the baculovirus shuttle vector pFastBacDual (Invitrogen) using NheI and KpnI (M1, eGFP-Gag) or BamHI and NotI (HA). Primer sequences are available upon request. Sequences were confirmed using Sanger sequencing. Recombinant vectors and empty vector (as control - referred to as ‘wild type’ herewith) were transformed into DH10Bac bacteria (Invitrogen) according to the manufacturer's recommendations; colonies with the desired phenotype were picked, propagated and bacmids were prepped using a Hipure Plasmid Filter Midiprep kit (Invitrogen). Bacmids were then transfected into Sf9 cells using Cellfectin II (Invitrogen) according to the manufacturer's recommendations. The rescued recombinant baculovirus was propagated and titered on Sf9 cells.
VLP Generation, Purification and Quantification
Insect cell-derived VLPs were expressed as previously described . Briefly, BTI-TN5B1-4 cells were co-infected at a multiplicity of infection of 10, either with baculoviruses that express HA and M1 or HA and eGFP-Gag, or with eGFP-Gag only (control VLPs). Infected cells were cultured in shaker flasks at 28°C at 80 rpm in a rotation shaker. Cultures were harvested 72 hours post-infection and supernatants were cleared by low-speed centrifugation at 2000 g at room temperature for 20 min. For the production of mammalian cell-derived VLPs, transient polyethylenimine (PEI) mediated transfections of the expression plasmids expressing either HA, M1 and influenza B neuraminidase, HA, eGFP-Gag and influenza B neuraminidase, or eGFP-Gag alone, were performed on 60–70% confluent 293T human embryo kidney cell layers. 1.1×107 293T cells were seeded in a 175 cm2 Roux flask and 24 hours later were transfected with PEI (Sigma-Aldrich) combined with the desired plasmid DNAs (45 µg total DNA per flask), at a 10∶1 ratio, as previously described . Cleared supernatants from insect or 293T cultures were laid over 30% sucrose cushions and spun at 136,000 g, for 90 minutes at 4°C, and the pellet containing the VLPs was resuspended in phosphate buffered saline (PBS, pH 7.4). To determine the amount of HA protein in the different VLP preparations, we performed a Western blot followed by densitometric analysis (Image J, National Institutes of Health), using a purified recombinant PR8 HA  of known concentration as a standard. Western blot analysis was performed as previously described , by using the monoclonal antibody PY102  to detect the PR8 HA (1∶1000 dilution) and a horseradish peroxidase-conjugated anti-mouse IgG secondary antibody (1∶5000 dilution). VLPs were further characterized as described before ,  (Supplemental Material S1).
This study was carried out in full compliance with the guidelines of the Mount Sinai School of Medicine Institutional Animal Care and Use Committee (IACUC). The protocol was approved by the Mount Sinai School of Medicine IACUC (animal protocol # LA12-00028). Procedures (vaccination, virus inoculation, etc.) were performed under ketamine/xylazine anesthesia (0.15 mg ketamine and 0.03 mg xylazine per mouse), and all efforts were made to minimize suffering. For the immunogenicity studies, groups of female, six- to eight-week old BALB/c mice were primed either intraperitoneally (i.p.) or intramuscularly (i.m.) with unadjuvanted mammalian cell- or baculovirus-derived eGFP-Gag or M1 driven VLPs harboring 500 ng of HA. Three weeks post prime, the animals primed via i.p. immunization were boosted i.p. with 500 ng of HA in the form of homologous VLPs. Animals that received an i.m. prime were either boosted i.m. or intranasally (i.n.) with equal amounts (500 ng of HA) of homologous VLPs under anesthesia. Group size was n = 3 for all immunogenicity studies except for the i.m./i.p. administered mammalian HA and M1 VLPs were the group size was n = 1. The control mice received similar amounts of either mammalian- or baculovirus-derived VLPs containing only eGFP-Gag, in an analogous manner (i.p. - i.p.). On day 42, mice were anesthetized and blood was collected by terminal cardiac puncture, followed by immediate sacrificing. Serum was separated from blood cells by performing a brief centrifugation and was stored at −20°C until further use.
In order to see differences in protection studies between the two immunogens we used a constant challenge dose (100 LD50) and titered down the vaccine dose. At a vaccine dose of 50 ng HA some morbidity in animals vaccinated with the baculovirus produced VLPs was observed and this amount of immunogen was then used for the comparative challenge as follows: Six- to eight-week old female BALB/c mice (n = 5) were vaccinated (i.m.) with mammalian- or baculovirus- derived, M1 driven VLPs containing 50 ng of HA, or were mock vaccinated with PBS. Three weeks post-vaccination, the mice were intranasally infected, while under anesthesia, with 100 mouse lethal dose 50 (mLD50) of PR8 virus. Weight was monitored daily for a period of 14 days, and mice losing more than 25% of their initial body weight were sacrificed and scored as dead.
Enzyme Linked Immunosorbent Assay
96-well ELISA plates (Immulon 4 HBX) were coated with 0.2 µg per well of PR8 virus or with 0.1 µg per well of purified recombinant HA protein in carbonate coating buffer (100 mM, pH 9.6). PR8 virus used for this purpose was prepared from allantoic ﬂuid by concentration through a 30% sucrose cushion. The recombinant HA proteins of the A/Puerto Rico/8/1934, A/California/04/2009, A/Viet Nam/1203/2004 isolates, as well as the cH6/1 protein, were purified in the baculovirus system, as previously described . Plates were blocked for one hour at room temperature with 3% milk in PBS. Afterwards, threefold serial dilutions of mouse sera were incubated on the coated plates, for two hours at room temperature. After extensive washing with PBS containing 0.1% Tween-20 (v/v), the bound antibody was detected with an alkaline phosphatase-linked anti-mouse IgG antibody (Invitrogen) (1∶5000) for one hour at RT, followed by another extensive washing step and detection with p-nitrophenyl phosphate (PNPP) substrate (Sigma). Optical density measurements were taken at 405 nm. For detection of IgA antibodies in nasal washes, a similar assay was employed, except that we used an alkaline phosphatase-linked anti-mouse IgA antibody (Southern Biotech) diluted 1∶500 and all incubations were performed at 37°C for three hours. Analysis of the isotype distribution in sera was determined by performing an ELISA assay, using recombinant PR8 HA as a substrate and an isotyping kit (Invitrogen), which contains a collection of secondary antibodies specific for each subtype and an AP-conjugated tertiary antibody that allows for detection of binding.
Hemagglutination Inhibition Assay
PR8 virus was diluted to eight chicken erythrocyte hemagglutination units (3 wells), and 25 µl of this virus dilution was pre-incubated with 25 µL of two-fold dilutions of inactivated antisera . After incubating for one hour on ice, 50 µl of a 0.5% chicken erythrocyte suspension was added to the mixture which was then incubated on ice for an additional hour. Virus mixed with PBS was used as a negative control, while PBS with no virus and no sera served as a background control.
Detection of Interferon β and Mx1 Following Vaccination
For measurement of cytokine induction following vaccination, groups of six- to eight-week old BALB/c mice (n = 3) were immunized either i.n. or i.m. (calf muscle) with mammalian- or baculovirus-derived, M1 driven VLPs containing 500 ng of HA. Additionally, control mice received either PBS or a dose of purified wild-type baculovirus containing similar titers of infectious particles as the ones present in the administered dose of baculovirus-derived VLP preparation (4.5×106 PFU/dose). Six hours post-inoculation, the calf muscle or the lungs were harvested and homogenized in 1 mL of Trizol-LS (Invitrogen) using a FastPrep-24 instrument (MP). The homogenized samples were passed through a QIAshredder column (Qiagen) and the RNA was extracted according to the manufacturer's protocol. Residual DNA in the samples was removed using the DNA-free kit (Ambion). To ensure for optimal RNA quality, the samples were subjected to treatment with the RNeasy kit (Qiagen). Complementary DNA was synthesized using oligoT primers and SuperScript III Reverse Transcriptase (Invitrogen) according to the manufacturer's protocol. Real-time PCR was performed using SYBR Green mix (Roche) and a LightCycler 480 Instrument (Roche). The sequences of the primers used for amplification are available upon request.
Baculovirus-derived VLPs Induce a Stronger and Broader Immune Response Compared to Mammalian Cell-derived VLPs
To compare mammalian- and baculovirus-derived VLPs in terms of immunogenicity, we vaccinated BALB/c mice using a prime and boost regimen with VLPs containing 500 ng of PR8 HA. We compared VLPs that were made by co-expression of HA with either eGFP-Gag or M1, and administered them i.p./i.p., i.m./i.m. or i.m./i.n. We first wanted to confirm if, and to what extent, animals from the differently vaccinated groups seroconverted. Because of possible non-specific binding of sera from baculovirus-vaccinated animals to baculovirus-expressed proteins used for ELISA assays, we compared levels of reactivity in sera to both recombinant PR8 HA and purified PR8 virus.
Irrespective of the vaccine protocol, we saw a significant difference in reactivity between sera of mice vaccinated with mammalian cell- or baculovirus-derived VLPs although all animals sero-converted to HA. Sera from animals vaccinated with preparations obtained in the latter system, whether M1 and eGFP-Gag driven, had significantly higher reactivity against both substrates. The highest titers were raised by the i.p./i.p. immunization scheme compared to the i.m./i.m. and i.m./i.n. regimens. No significant difference was appreciated in terms of immunogenicity induced by M1 versus eGFP-Gag driven VLP preparations. It is of note that sera from mice vaccinated with eGFP-Gag only VLPs (mammalian- and baculovirus-derived) had very low unspecific reactivity against the purified PR8 HA preparation in comparison to whole virus substrate (Figure 1A and B). Therefore, we decided to use purified HAs for all subsequent analysis but included sera from control animals vaccinated with HA-negative eGFP-Gag VLPs in all assays we performed.
Pooled sera (A) or sera from individual mice (B) vaccinated by different immunization schemes with either mammalian cell- or baculovirus-derived VLPs, were tested for binding to purified PR8 virus or recombinant PR8 HA, respectively.
To measure levels of HA head domain-directed antibody titers, we performed an HI assay with PR8 virus and sera obtained from the vaccinated animals and found similar trends as those seen in ELISA. The i.p./i.p. administration was able to induce the highest titers of hemagglutination inhibiting antibodies, followed by the other two vaccination regimens. A higher geometrical mean titer was observed for baculovirus-derived VLPs in general, although this finding is not significant for all cases. We also observed a higher variation in terms of HI activity in sera from mice vaccinated with mammalian cell-derived VLPs, compared to their baculovirus VLP vaccinated counterparts. No significant differences were observed between eGFP-Gag and M1 driven VLPs. All sera from control mice that received eGFP-Gag only VLPs tested negative (Table 1).
Because of the increasing interest in the field to develop vaccine immunogens that provide broad protection to a variety of influenza virus subtypes, we next wanted to test reactivity to other group 1 HAs, as well as to the PR8 stalk domain. Antibodies with specificities against the highly conserved stalk domain of the HA glycoprotein have been described to have broadly reactive and cross-neutralizing characteristics –. For these purposes, we tested serum reactivity against HA proteins coming from the Cal09 and VN04 viral isolates, and a chimeric H6/1 HA , respectively. The highest antibody titers against the Cal09 substrate were found in sera from groups that received baculovirus-derived VLPs containing M1 and HA, followed by eGFP-Gag plus HA VLPs (Figure 2). Interestingly, vaccination with mammalian cell produced VLPs did not induce reactivity above background. Similar results were obtained when testing binding activity against VN04 or cH6/1 HA proteins. Our findings suggest that HA-M1 VLPs are better at inducing antibodies against the stalk domain of the influenza virus HA.
Pooled sera from animals vaccinated with either mammalian cell- or baculovirus-derived VLPs were tested in ELISA for the ability to bind recombinant chimeric H6/1 (cH6/1 HA), H1 (A/California/04/2009 HA) or H5 (A/Viet Nam/1203/2004 HA) hemagglutinin. Data shown are from serum samples diluted 1∶300. Baselines represent background levels detected in this assay, when testing sera from animals vaccinated with HA negative, eGFP-Gag containing, mammalian cell- (red) or baculovirus-derived (black) VLPs.
Baculovirus-derived VLPs Trigger Preferable Antibody Isotype Profiles
Next we analyzed the serum antibody isotype distribution, as a measure of the level of affinity maturation triggered by the different vaccination regimens, and also as a potential estimation of the neutralizing efficiency of the induced antibody responses. For this purpose, we compared relative levels of IgM, IgA and IgG subtypes in sera of mice immunized by different protocols with M1 containing baculovirus- or mammalian cell-derived VLPs using an ELISA. Expectedly, we detected an overall increased level of all serotypes in the mice vaccinated with insect cell-derived preparations (Figure 3A). Interestingly, we detected similar levels of IgM antibodies in the differently vaccinated groups that received baculovirus-derived VLPs, indicating similar levels of initial seroconversion, while the mammalian cell produced VLPs only induced significant levels of this class of antibodies when administered i.p./i.p. However, all groups underwent somatic maturation of serum antibodies, since differentiated IgG-types were present in all groups of mice, albeit at much lower levels in sera of mice who had received mammalian cell-derived VLP vaccines by i.m./i.m. or i.m./i.n. administration. When looking at the IgG subtype distribution, we saw similar levels of IgG1 type antibodies irrespective of the expression system in which the VLPs were produced. However, the baculovirus-derived VLPs produced significantly higher levels of IgG2a and IgG2b antibodies in all of the vaccination protocols assessed. These subclasses of IgGs are known to be the most pro-inflammatory IgG subtypes in mice and have been described to have a higher biological activity than IgG1 or IgG3 subtypes .
(A) Serum antibody isotype profile induced by different VLPs and immunization regimens. Serum was collected 3 weeks after the second immunization and analyzed for relative distribution of immunoglobulin subclasses by ELISA assay, using a recombinant PR8 HA protein as substrate and different secondary antibodies recognizing the different Ig isotypes. Signal was detected using a tertiary antibody coupled to an alkaline phosphatase (AP) enzyme. (B) Mucosal IgA antibody levels induced by the different vaccination regimens as analyzed by testing binding activity in nasal washes obtained from animals three weeks after the second immunization. Samples were pooled and binding was detected using an AP-coupled secondary antibody specific for the IgA subtype. Baselines represent background levels detected in this assay, when testing sera from animals vaccinated with HA negative, eGFP-Gag containing, mammalian cell- (orange) or baculovirus-derived (yellow) VLPs.
Next, we assessed the effect of the vaccination route on the generation of mucosal IgA. Baculovirus-derived VLPs have been previously reported to induce an IgA antibody type response in lung extracts of vaccinated mice, when administered i.n. . We were particularly interested in the levels of IgA antibodies in the nasal washes of the animals, since these are known to play an important role in minimizing infections with influenza virus by neutralizing infectious particles in the respiratory tract . Interestingly, we were indeed able to detect HA reactive IgA in samples from animals that received baculovirus-produced VLPs (both M1- and eGFP-Gag-derived) i.m./i.n., but not in sera from animals similarly immunized with mammalian cell-derived VLPs (Figure 3B).
Residual Baculovirus in VLP Preparations Stimulates Innate Immune Responses
Considering the significant differences in quantity and quality of the immune responses induced by vaccinations with baculovirus- and mammalian cell-derived VLP preparations, we further investigated the possible causative factors. Baculoviruses do not replicate in mammalian cells , but they have been reported to enter these cells and trigger an innate immune response, most probably through activation of TLR9 , –. We hypothesized that residual baculovirus in the insect cell-derived VLPs may activate the innate immune response, which may in turn be responsible for the observed enhanced immunogenicity of these preparations. When examining the baculovirus content in terms of plaque forming units (PFU) found in our VLP preparations, we detected 4.5×106 PFU of baculovirus per vaccine dose, which is in the range of previous reports , , . To investigate the effect this residual baculovirus may have on cytokine and interferon stimulated gene (ISG) induction, we compared mammalian cell-derived VLPs, baculovirus-derived VLPs and purified wild-type baculovirus. We inoculated mice either i.n. or i.m. with preparations of mammalian cell-derived VLPs, baculovirus-derived VLPs (both M1 triggered), a corresponding dose of purified baculovirus (corresponding PFU as in the baculovirus-derived VLP dose) and a PBS control. Six hours post-inoculation we harvested lungs or muscle and performed quantitative PCR analysis to measure mRNA levels of interferonβ and Mx1 genes. Interestingly, the baculo-derived VLPs and baculovirus, but not the mammalian VLPs were able to induce IFNβ and Mx1 (Figure 4). This finding strongly suggests that the baculovirus contaminations in the insect cell preparations can induce innate immunity, and thereby have a potential immuno-enhancing effect for the antigen preparation , . The lower levels of IFNβ induction determined for the purified baculovirus, when compared to the baculovirus-derived VLPs, may be explained by the wild type purified baculovirus having a lower PFU-to-defective interference (DI) particle ratio than baculovirus expressing recombinant proteins. Therefore, the number of baculovirus particles might be higher in the VLP vaccine than in the purified preparation, despite similar PFUs .
Mice were immunized (intramuscularly or intranasally) with one dose of vaccine or the corresponding dose of wild type baculovirus (in PFU/dose) and 6 hours post-vaccination lung or muscle tissue was harvested and the level of induction of gene specific mRNA for IFN β, and Mx1 was analyzed as fold induction over tissue samples from animals mock-immunized with PBS.
Enhanced Protection by Baculovirus-derived VLPs in Comparison to Mammalian-derived VLPs
Finally, we assessed the protective effect of the enhanced immunogenicity shown by the baculovirus-derived VLPs in the context of an in vivo infection. In order to detect differences in protection, we chose a low single immunization dose (50 ng of HA in HA plus M1 VLPs) that was found to be the lowest protective dose (from mortality) for baculovirus-derived VLPs in a challenge experiment with 100 mLD50 of PR8 virus (data not shown). Mice were immunized once, intramuscularly, with 50 ng of HA in form of M1 driven, mammalian- or baculovirus-derived VLPs. We chose this vaccination route since i.n., but not i.m. immunization with wild-type baculovirus protects from viral challenge through stimulation of innate immunity . Three weeks post immunization, mice were challenged intranasally with 100 mLD50 of PR8 virus. The mice that received this low dose of mammalian cell produced VLPs lost weight with similar kinetics as the mock-vaccinated group, and succumbed to disease caused by the virus infection by day 7 (Figure 5). Animals that received baculovirus produced VLPs however, only lost approximately 15% of their body weight, looked otherwise healthy (scores not shown), regained their initial body weight quickly and survived the viral infection. We therefore conclude that the differences in serum antibody levels translate to enhanced in vivo protection in a mouse influenza virus infection model.
Mice were vaccinated once with mammalian cell-, or baculovirus- derived VLPs containing 50 ng of HA, and were intranasally infected with 100 mLD50 of PR8 virus three weeks post vaccination. (A) Body weight loss was monitored daily for 14 days and (B) animals that lost more than 25% of their initial body weight were scored dead (Kaplan-Meier curve).
In this study we compared influenza A VLPs produced in a mammalian and a baculovirus expression system, in terms of immunity and protection induced in a mouse model of influenza virus infection. We detected a significantly higher immunogenicity induced by the baculovirus-derived VLPs in terms of homologous- and heterologous- HA specific IgG and IgA antibody titer, and a similar trend in HI titers although mice sero-converted to both immunogens. Furthermore, the baculovirus-derived VLP preparations significantly biased the isotype distribution of the antibody response towards the more active IgG2 immunoglobulin subtype. Additionally we observed that M1, but not eGFP-Gag driven particles are able to induce a small but significant antibody response directed to the stalk domain of the PR8 HA. These antibodies have recently been described to have broadly neutralizing abilities, and vaccination strategies that can efficiently induce them are considered to be promising for the development of a universal influenza virus vaccine .
In various assays, the M1-derived VLPs were found to be more immunogenic when compared to eGFP-Gag driven preparations. It is possible that by coupling HA with a natural budding partner, M1-derived VLPs may be more immunogenic due to preferential antigen uptake, receptor binding by the VLPs, fusion with the endosomal membrane, or antigen presentation. This was suggested by previous studies that detected broader reactivity in baculovirus-derived VLP vaccines containing M1 –. Furthermore, we showed that this enhanced immunogenicity translates well into protection from lethal influenza virus challenge in a single-dose immunization model. We hypothesize that residual baculovirus particles in the VLP preparations are responsible for their enhanced immunogenicity, and we show that baculovirus-derived VLPs, as well as wild type baculovirus, induce high levels of interferon at the site of inoculation. This induction may trigger the recruitment of innate immune cells, as well as activation and maturation of dendritic cells that produce inflammatory cytokines and present antigen to B cells. Collectively, these processes could be responsible for the observed effect. Along these lines, a recent report interestingly showed that MyD88−/− mice have a diminished response to vaccination with baculovirus-derived VLPs, as compared to wild-type mice . Actually, the pattern demonstrated in MyD88−/− mice resembles the response we observed with mammalian cell-derived VLPs and supports our hypothesis, since baculovirus induced signaling occurs mainly through activation of TLR9, and MyD88 downstream , , .
Our results therefore have practical implications. Since baculovirus is hard to separate from influenza VLPs due to the similar density and enveloped nature they share, it is always present in “purified” VLP preparations , –. We hypothesize that the enhanced immunogenicity seen with baculovirus-expressed influenza VLPs in animal models is dependent on the presence of these viral contaminations. As a consequence, pre-clinical results with non-inactivated preparations may be hard to translate into clinical trials where these preparations have to be inactivated because of safety concerns . Inactivation of baculoviruses with different chemical (binary ethylenimine, Triton X100) or physical (UV irradiation) reagents has been shown to abolish the interferon-inducing and immuno-enhancing ability of baculovirus impurities . The presence of active baculovirus particles in vaccine preparations might lead to enhanced reactogenicity including local inflammation, soreness and fever. Additionally, there is a general concern by regulatory agencies about live, contaminating viruses or virus genomes in human vaccines.
In summary, we demonstrate the higher immunogenicity of baculovirus-derived VLPs compared to mammalian-cell based systems in terms of eliciting of differentiated IgG and also mucosal IgA responses. These differences can be directly translated into protection in an in vivo model of influenza virus infection. The enhanced immunogenicity of the insect cell-derived influenza VLPs can be explained by residual, contaminating baculovirus that induces a local antiviral response.
We thank Natalie Pica (Mount Sinai School of Medicine) for help with manuscript preparation and animal studies and Christopher W. Seibert (Mount Sinai School of Medicine) for help with nasal washes. We also thank Dieter Palmberger and Reingard Grabherr (Vienna Institute of BioTechnology (VIBT)) for the VIBT sub-clone of BTI-TN-5B1-4 insect cells . Finally, we would like to thank Dr. Peter Palese for valuable discussions and help with manuscript preparation.
Conceived and designed the experiments: IM FK. Performed the experiments: IM LM YC FK. Analyzed the data: IM FK. Wrote the paper: IM FK.
- 1. Tosh PK, Jacobson RM, Poland GA (2010) Influenza vaccines: from surveillance through production to protection. Mayo Clin Proc 85: 257–273.
- 2. Nichol KL (2008) Efficacy and effectiveness of influenza vaccination. Vaccine 26 Suppl 4D17–22.
- 3. Palese P (2006) Making better influenza virus vaccines? Emerg Infect Dis 12: 61–65.
- 4. D'Aoust MA, Couture MM, Charland N, Trépanier S, Landry N, et al. (2010) The production of hemagglutinin-based virus-like particles in plants: a rapid, efficient and safe response to pandemic influenza. Plant Biotechnol J 8: 607–619.
- 5. Krammer F, Grabherr R (2010) Alternative influenza vaccines made by insect cells. Trends Mol Med 16: 313–320.
- 6. Szécsi J, Boson B, Johnsson P, Dupeyrot-Lacas P, Matrosovich M, et al. (2006) Induction of neutralising antibodies by virus-like particles harbouring surface proteins from highly pathogenic H5N1 and H7N1 influenza viruses. Virol J 3: 70.
- 7. Giles BM, Bissel SJ, Dealmeida DR, Wiley CA, Ross TM (2012) Antibody breadth and protective efficacy are increased by vaccination with computationally optimized hemagglutinin but not with polyvalent hemagglutinin-based H5N1 virus-like particle vaccines. Clin Vaccine Immunol 19: 128–139.
- 8. Wu CY, Yeh YC, Yang YC, Chou C, Liu MT, et al. (2010) Mammalian expression of virus-like particles for advanced mimicry of authentic influenza virus. PLoS One 5: e9784.
- 9. López-Macías C (2012) Virus-like particle (VLP)-based vaccines for pandemic influenza: Performance of a VLP vaccine during the 2009 influenza pandemic. Hum Vaccin Immunother 8.
- 10. López-Macías C, Ferat-Osorio E, Tenorio-Calvo A, Isibasi A, Talavera J, et al. (2011) Safety and immunogenicity of a virus-like particle pandemic influenza A (H1N1) 2009 vaccine in a blinded, randomized, placebo-controlled trial of adults in Mexico. Vaccine 29: 7826–7834.
- 11. Krammer F, Schinko T, Palmberger D, Tauer C, Messner P, et al. (2010) Trichoplusia ni cells (High Five) are highly efficient for the production of influenza A virus-like particles: a comparison of two insect cell lines as production platforms for influenza vaccines. Mol Biotechnol 45: 226–234.
- 12. Haynes J, Dokken L, Wiley J, Cawthon A, Bigger J, et al. (2009) Influenza-pseudotyped Gag virus-like particle vaccines provide broad protection against highly pathogenic avian influenza challenge. Vaccine 27: 530–541.
- 13. Pushko P, Tumpey T, Bu F, Knell J, Robinson R, et al. (2005) Influenza virus-like particles comprised of the HA, NA, and M1 proteins of H9N2 influenza virus induce protective immune responses in BALB/c mice. Vaccine 23: 5751–5759.
- 14. Song JM, Choi CW, Kwon SO, Compans RW, Kang SM, et al. (2011) Proteomic characterization of influenza H5N1 virus-like particles and their protective immunogenicity. J Proteome Res 10: 3450–3459.
- 15. Easterbrook JD, Schwartzman LM, Gao J, Kash JC, Morens DM, et al. (2012) Protection against a lethal H5N1 influenza challenge by intranasal immunization with virus-like particles containing 2009 pandemic H1N1 neuraminidase in mice. Virology 432: 39–44.
- 16. Latham T, Galarza J (2001) Formation of wild-type and chimeric influenza virus-like particles following simultaneous expression of only four structural proteins. J Virol 75: 6154–6165.
- 17. Steel J, Lowen AC, Wang TT, Yondola M, Gao Q, et al.. (2010) Influenza virus vaccine based on the conserved hemagglutinin stalk domain. MBio 1.
- 18. Hermida-Matsumoto L, Resh MD (2000) Localization of human immunodeficiency virus type 1 Gag and Env at the plasma membrane by confocal imaging. J Virol 74: 8670–8679.
- 19. Hai R, Martínez-Sobrido L, Fraser KA, Ayllon J, García-Sastre A, et al. (2008) Influenza B virus NS1-truncated mutants: live-attenuated vaccine approach. J Virol 82: 10580–10590.
- 20. Krammer F, Nakowitsch S, Messner P, Palmberger D, Ferko B, et al. (2010) Swine-origin pandemic H1N1 influenza virus-like particles produced in insect cells induce hemagglutination inhibiting antibodies in BALB/c mice. Biotechnol J 5: 17–23.
- 21. Boussif O, Lezoualc'h F, Zanta MA, Mergny MD, Scherman D, et al. (1995) A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proceedings of the National Academy of Sciences of the United States of America 92: 7297–7301.
- 22. Krammer F, Margine I, Tan GS, Pica N, Krause JC, et al.. (2012) A carboxy-terminal trimerization domain stabilizes conformational epitopes on the stalk domain of soluble recombinant hemagglutinin substrates. Plos One.
- 23. Varga ZT, Ramos I, Hai R, Schmolke M, Garcia-Sastre A, et al. (2011) The influenza virus protein PB1-F2 inhibits the induction of type I interferon at the level of the MAVS adaptor protein. PLoS pathogens 7: e1002067.
- 24. Dinca L, Neuwirth S, Schulman J, Bona C (1993) Induction of antihemagglutinin antibodies by polyclonal antiidiotype antibodies. Viral Immunol 6: 75–84.
- 25. Pica N, Hai R, Krammer F, Wang TT, Maamary J, et al. (2012) Hemagglutinin stalk antibodies elicited by the 2009 pandemic influenza virus as a mechanism for the extinction of seasonal H1N1 viruses. Proc Natl Acad Sci U S A 109: 2573–2578.
- 26. Lowen AC, Steel J, Mubareka S, Carnero E, Garcia-Sastre A, et al. (2009) Blocking interhost transmission of influenza virus by vaccination in the guinea pig model. Journal of virology 83: 2803–2818.
- 27. Wang TT, Tan GS, Hai R, Pica N, Petersen E, et al. (2010) Broadly protective monoclonal antibodies against H3 influenza viruses following sequential immunization with different hemagglutinins. PLoS Pathog 6: e1000796.
- 28. Ekiert DC, Bhabha G, Elsliger MA, Friesen RH, Jongeneelen M, et al. (2009) Antibody recognition of a highly conserved influenza virus epitope. Science 324: 246–251.
- 29. Sui J, Hwang WC, Perez S, Wei G, Aird D, et al. (2009) Structural and functional bases for broad-spectrum neutralization of avian and human influenza A viruses. Nat Struct Mol Biol 16: 265–273.
- 30. Kashyap AK, Steel J, Oner AF, Dillon MA, Swale RE, et al. (2008) Combinatorial antibody libraries from survivors of the Turkish H5N1 avian influenza outbreak reveal virus neutralization strategies. Proc Natl Acad Sci U S A 105: 5986–5991.
- 31. Tan GS, Krammer F, Eggink D, Kongchanagul A, Moran TM, et al. (2012) A pan-h1 anti-hemagglutinin monoclonal antibody with potent broad-spectrum efficacy in vivo. J Virol 86: 6179–6188.
- 32. Krammer F, Pica N, Hai R, Tan GS, Palese P (2012) Hemagglutinin stalk reactive antibodies are boosted following sequential infection with seasonal and pandemic H1N1 influenza virus in mice. J Virol.
- 33. Nimmerjahn F, Ravetch JV (2006) Fcgamma receptors: old friends and new family members. Immunity 24: 19–28.
- 34. Quan F, Huang C, Compans R, Kang S (2007) Virus-like particle vaccine induces protective immunity against homologous and heterologous strains of influenza virus. J Virol 81: 3514–3524.
- 35. Renegar KB, Small PA Jr, Boykins LG, Wright PF (2004) Role of IgA versus IgG in the control of influenza viral infection in the murine respiratory tract. Journal of immunology 173: 1978–1986.
- 36. Guarino L (2001) Baculoviruses. eLS: John Wiley & Sons, Ltd.
- 37. Abe T, Matsuura Y (2010) Host innate immune responses induced by baculovirus in mammals. Curr Gene Ther 10: 226–231.
- 38. Hervas-Stubbs S, Rueda P, Lopez L, Leclerc C (2007) Insect baculoviruses strongly potentiate adaptive immune responses by inducing type I IFN. J Immunol 178: 2361–2369.
- 39. Abe T, Hemmi H, Miyamoto H, Moriishi K, Tamura S, et al. (2005) Involvement of the Toll-like receptor 9 signaling pathway in the induction of innate immunity by baculovirus. J Virol 79: 2847–2858.
- 40. Hu Y (2008) Baculoviral vectors for gene delivery: a review. Curr Gene Ther 8: 54–65.
- 41. Toporovski R, Morrow MP, Weiner DB (2010) Interferons as potential adjuvants in prophylactic vaccines. Expert Opin Biol Ther 10: 1489–1500.
- 42. Wickham TJ, Davis T, Granados RR, Hammer DA, Shuler ML, et al. (1991) Baculovirus defective interfering particles are responsible for variations in recombinant protein production as a function of multiplicity of infection. Biotechnol Lett July v. 13 (7): 483–488.
- 43. Abe T, Takahashi H, Hamazaki H, Miyano-Kurosaki N, Matsuura Y, et al. (2003) Baculovirus induces an innate immune response and confers protection from lethal influenza virus infection in mice. J Immunol 171: 1133–1139.
- 44. Wang TT, Palese P (2011) Biochemistry. Catching a moving target. Science 333: 834–835.
- 45. Khurana S, Wu J, Verma N, Verma S, Raghunandan R, et al.. (2011) H5N1 virus-like particle vaccine elicits cross-reactive neutralizing antibodies in humans that preferentially bind to oligomeric form of influenza hemagglutinin. J Virol.
- 46. Tao P, Luo M, Zhu D, Qu S, Yang Z, et al. (2009) Virus-like particle vaccine comprised of the HA, NA, and M1 proteins of an avian isolated H5N1 influenza virus induces protective immunity against homologous and heterologous strains in mice. Viral Immunol 22: 273–281.
- 47. Bright R, Carter D, Daniluk S, Toapanta F, Ahmad A, et al. (2007) Influenza virus-like particles elicit broader immune responses than whole virion inactivated influenza virus or recombinant hemagglutinin. Vaccine 25: 3871–3878.
- 48. Kang SM, Yoo DG, Kim MC, Song JM, Park MK, et al. (2011) MyD88 plays an essential role in inducing B cells capable of differentiating into antibody-secreting cells after vaccination. J Virol 85: 11391–11400.
- 49. Abe T, Kaname Y, Wen X, Tani H, Moriishi K, et al. (2009) Baculovirus induces type I interferon production through toll-like receptor-dependent and -independent pathways in a cell-type-specific manner. J Virol 83: 7629–7640.