Since the identification of the avian influenza A (H7N9) infection in human in March 2013, the viral threat has prompted a global effort to develop vaccine candidates as humans have little immunity to this reassorted virus [18–20]. This leap of the virus from birds to humans was attributed to multiple reassortment events, and continuing reassortment events may lead to emergence of more infective strains. Disease management at both the human and animal levels are thus of major importance. In this study, A/Taiwan/S02076/2013(H7N9), a strain isolated in Taiwan [13], was used for the development of a vaccine candidate through the preparation of VLPs. Co-infection of recombinant baculoviruses in insect cells yielded influenza A/H7N9 VLPs that coexpressed HA, NA, and M1 proteins, which have molecular weights consistent with previously published studies [10, 11]. The VLPs are monodisperse and contain approximately 800 HA proteins per particles. These VLPs were found to induce proper humoral and cellular immunity against the H7N9 virus in animal models. In particular, the VLPs were found to be superior in eliciting viral-antigen-specific humoral and cellular immune responses as compared to a free protein formulation in an avian model, which has not been investigated previously.
VLPs are a versatile system that presents a compelling alternative to conventional vaccine formulations as they possess the morphological semblance to natural viral particles. In the present study we applied nanoparticle tracking analysis and transmission electron microscopy to validate the structure and antigenicity of the VLPs. These analyses showed virus-like features consistent with other VLP formulations reported in previous studies [6, 10, 11, 21]. It is worth noting that although an Sf9 insect cell system was adopted for the production of the H7N9 VLPs in this study, the VLP platform is highly versatile and may be produced using other culture systems. For instance, a previous study by Chang et al. employed an alternate High Five insect cell culture system for VLP production [22, 23], demonstrating high production yield upon High Five system optimization. In the present study, we showed that varying the MOI of the different recombinant baculoviruses impacted the VLP production, offering information toward further optimization of the Sf9 system for VLP generation.
We first validated the potency of the VLPs in a mouse model. Given that neutralizing antibodies against the globular head of hemagglutinin protein are the primary mediators of most vaccine-induced protection against influenza, a hemagglutination inhibition assay was used to examine the VLPs as a vaccine candidate against H7N9. An HI antibody titer of 1:40 is the accepted correlate of protection for human HA split inactivated vaccines [24]. Immunization with the VLPs resulted in a mouse serum HI titer of 1:160 against the H7N9 virions. Although future animal studies with viral challenges are warranted, the present result validates the immunopotentiation effect of our VLPs, which is on par with those in previous reports.
Despite previous studies that examined different formulations of H7N9 VLPs in mouse models [9–11, 25, 26], no examination of H7N9 vaccination in avian models has been reported to the best of our knowledge. Given that the management of the avian influenza virus would benefit from both human and poultry vaccinations, we investigated the VLPs’ vaccination effect in SPF chickens. As compared to a free protein formulation, the VLP vaccination yielded increased HI serum titers, anti-NA titers, and anti-M1 titers. It is also worth noting that upon examination of cellular immune responses, we observed that the VLP immunization resulted in elevated splenic IFN-? and IL-4 upon subsequent viral exposure. The cellular immune responses elicited by VLPs in SPF chickens are consistent with previous studies examining VLP immunizations in other animal models [21].
In the present study, we purified our VLPs and made sure that no detectable baculovirus titer was present in any of our batches. The purification is critical as residual baculoviral proteins were previously shown to trigger innate immune responses through TLR9 and other pattern recognition receptors [27, 28]. Towards future clinical translation, the elimination of residual baculovirus contamination is of high importance.
Even though the free protein formulation, which is a mixture consisting of free HA, NA, and M1 proteins at a 1:1:1 ratio, also elicited humoral and cellular responses in our study, the responses were either weaker or more variable as compared to those elicited by the VLPs. The finding fortifies the notion that the H7N9 VLPs may serve as a compelling vaccine candidate in poultry farming, which typically employs inactivated or subunit vaccines for disease management [29–31].