(454f) Efficient Incorporation of Matrix Protein M2 into Influenza Virus-like Particles (VLPs) for Improved Vaccine Manufacturing and Efficacy

Influenza virus remains a heavy economic burden and poses pandemic threats to public health worldwide, causing 250,000-500,000 deaths annually and 3-5 million cases of severe illness. Over the last 15 years in the United States, reported efficacy of flu shots has been as low as 19%. The low vaccine efficacy is primarily due to key limitations of the traditional egg-based manufacturing platform, a slow process (6-9 months) during which mutations in the circulating virus (antigenic drift) or egg-adapted vaccine virus can result in antigen mismatch. This antigen mismatch is particularly problematic for influenza strains that mutate at a faster rate, such as the H3N2 circulated in the past two flu seasons, causing higher number of hospitalizations and deaths compared to previous years. To address these issues, alternative manufacturing platforms with shorter lead times (2-3 months) such as the baculovirus-insect system have been utilized for producing recombinant protein-based vaccines (Flublok®) as well as virus-like particles (VLPs). Compared to recombinant protein-based vaccines, VLPs are more immunogenic due to greater antigen display in a three dimensional structure that resembles the influenza virus but lacks viral RNA. Furthermore, the baculovirus-insect system allows for facile engineering of the antigen composition of the VLPs, which is critical for developing broadly protective flu vaccines and ultimately eliminating the current practice of updating flu shots annually necessitated by antigen mismatch.

Towards the goal of rapid manufacturing of broadly protective influenza vaccines, VLPs presenting antigens that are conserved among multiple influenza strains such as the matrix protein M2 have been attempted. However, the expression of full-length M2 showed cytopathic effects on insect cells hindering their ability to produce M2-containing VLPs. In this work, we quantified the cytopathic effects of expressing M2 in insect cells using various metrics and revealed that M2 significantly impedes the ability of insect cells to replicate baculovirus but does not directly cause cell death. Because baculovirus is the gene carrier for expressing influenza viral proteins including M2, HA and M1 in insect cells, the reduced baculovirus replication leads to reduced expression of all three influenza proteins, which in turn reduces the quantity and quality (i.e., antigen density) of the resulting VLPs. Based on the function of M2, we hypothesized that blocking its ion channel activity could address this issue. Indeed, the use of an M2 inhibitor restored the baculovirus replication when expressing M2 in insect cells, resulting in > 5-fold improvement in the expression of all three influenza viral proteins M2, HA and M1. More importantly, it further improved both the quantity and the quality of the VLPs produced, as evidenced by the increased number and antigen (HA) density of the VLPs. For future work, we will evaluate the vaccination efficacy of the M2-incorporated VLPs in mice. Given the importance of HA density and the ability of M2 ectodomain in eliciting effective, broadly protective and long lasting immunity against influenza, this work provides a novel approach for manufacturing and engineering effective and broadly protective influenza vaccines.