Potent priming by inactivated whole influenza virus particle vaccines is linked to viral RNA uptake into antigen presenting cells

Abstract

Current detergent or ether-disrupted split vaccines (SVs) for influenza do not always induce adequate immune responses, especially in young children. This contrasts with the whole virus particle vaccines (WPVs) originally used against influenza that were immunogenic in both adults and children but were replaced by SV in the 1970s due to concerns with reactogenicity. In this study, we re-evaluated the immunogenicity of WPV and SV, prepared from the same batch of purified influenza virus, in cynomolgus macaques and confirmed that WPV is superior to SV in priming potency. In addition, we compared the ability of WPV and SV to induce innate immune responses, including the maturation of dendritic cells (DCs) in vitro. WPV stimulated greater production of inflammatory cytokines and type-I interferon in immune cells from mice and macaques compared to SV. Since these innate responses are likely triggered by the activation of pattern recognition receptors (PRRs) by viral RNA, the quantity and quality of viral RNA in each vaccine were assessed. Although the quantity of viral RNA was similar in the two vaccines, the amount of viral RNA of a length that can be recognized by PRRs was over 100-fold greater in WPV than in SV. More importantly, 1000-fold more viral RNA was delivered to DCs by WPV than by SV when exposed to preparations containing the same amount of HA protein. Furthermore, WPV induced up-regulation of the DC maturation marker CD86 on murine DCs, while SV did not. The present results suggest that the activation of antigen-presenting DCs, by PRR-recognizable viral RNA contained in WPV is responsible for the effective priming potency of WPV observed in naïve mice and macaques. WPV is thus recommended as an alternative option for seasonal influenza vaccines, especially for children.

Introduction

Influenza is a highly contagious respiratory illness that is responsible for significant morbidity and mortality. Over 1 billion cases are reported each year, 3–5 million of which are severe and result in 290,000–650,000 deaths annually worldwide [1], [2], [3]. Adults 65 years and older are at particular risk of severe disease, hospitalization, and mortality due to seasonal influenza [4]. Young children are also at risk due to higher attack rates [5], [6]. Children are key virus spreaders in a community because they are more susceptible to influenza virus infection than adults and frequently make contact with friends and family members, resulting in shedding larger amounts of virus in the community over a long period of time [7], [8]. In fact, the best predictor for influenza occurring in a household is the presence of children [9], [10]. Thus, children should be prioritized for prevention of influenza through vaccination.

Annual vaccination is the most effective way to prevent influenza and to reduce an individual’s risk of severe influenza-related disease. The first inactivated seasonal influenza vaccine was developed as a whole virus particle vaccine (WPV) in the 1940s and was used for 30 years [11]. However, the administration of the WPV vaccine to children occasionally caused side effects such as fever, pain, and fatigue [12], [13], which led many people to mistakenly assume that the vaccine might cause flu-like symptoms. One of the reasons for this reactogenicity was thought to be due to impurities, such as egg-derived contaminants, in the vaccine [13]. Therefore, the manufacturing process for the newly developed seasonal influenza vaccines had an additional purification step to remove the impurities using zonal centrifugation. Moreover, most current influenza vaccines are disrupted using ether and/or detergents to decrease febrile reactions, hence the name “split” vaccine (SV). Several clinical trials demonstrated that SV was indeed less pyrogenic than WPV but was also less immunogenic, especially in young children [14], [15], [16]. Despite some disagreements, SV was approved in the United States as a standard influenza vaccine and replaced WPV in 1968 [11], and in Japan in 1972. A report of a clinical trial by Gross et al. [15], [16] highlighted three interesting findings: (1) in young children WPV causes fever more frequently than does SV, (2) young children previously vaccinated with influenza virus vaccine are unlikely to experience fever subsequent to immunization with a related antigen and (3) SV induces less vaccine strain-specific antibodies than does WPV in immunologically unprimed young children. The fact that WPV but not SV induces immunity in “naïve” children with a febrile response implies that the fever caused by WPV accompanies immune priming responses.

To re-evaluate WPV, we established the All Japan Influenza Vaccine Study Group and requested that member vaccine manufacturers in Japan prepare WPV and SV from the same purified virus batches from which egg-derived contaminants had been removed. Our previous study utilizing such matched sets of WPV and SV preparations demonstrated the superiority of WPV over SV in terms of the induction of neutralizing antibodies and CD8⁺ T cell responses in naïve mice as well as protection against severe disease following homologous virus challenge [17]. These results not only encouraged us to advance this project to preclinical study, but also raised a new question as to why WPV, rather than SV, elicits an effective priming response. To address this question, we compared priming and boosting effects of WPV and SV in cynomolgus macaques. The present study provides data supporting the strong priming effect of WPV compared to SV and provides evidence for the mechanistic basis that underpins this phenomenon.