Peptide amphiphile micelles as a universal influenza vaccine delivery vehicle
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Thesis
Subject
Abstract
[EMBARGOED UNTIL 12/01/2025] Despite a plethora of influenza vaccines and treatment options, there are millions of cases of influenza each year in the United States alone. Although increasing vaccination rates could potentially alleviate some of this public health burden, a significant problem with existing influenza vaccines is the fact that, because vaccine production must begin well before an increase in cases is observed in a given region, there is a significant reliance on predictions of what will be the prevalent influenza viruses for a given season. Unfortunately, if predictions are inaccurate, vaccine efficacy suffers.1 Avoiding this dependence on inconsistent forecasts is the driving force for research into a universal influenza vaccine. The premise of the concept of a universal vaccine is that epitopes are selected based on their conservation across a wide range of influenza strains. Because the epitopes are conserved, they should offer more broad-spectrum protection and reduce the need for effective predictions. Epitopes in a vaccine can be chosen based on a number of factors including degree of conservation as well as the magnitude and form of immunogenicity. An influenza virus expresses three proteins on its capsid surface, predominantly hemagglutinin (HA) and neuraminidase (NA) with lesser quantity of matrix protein 2 (M2). Traditionally, HA and NA have been the proteins of focus because they contain the most immunogenic epitopes of the influenza virus. However, epitopes in these proteins are under significant evolutionary pressure, so their amino acid sequences are constantly changing. In fact, there are 18 known subtypes of HA and 11 of NA (with additional variants within these subtypes).2 Nonetheless, there are many epitopes that are reasonably well conserved within the three surface proteins of the influenza virus that could prove to be viable candidates for a universal influenza vaccine. For example, the HA stalk, which is less exposed and slightly less immunogenic than the HA head, tends to be more conserved. Another such region is the N-terminal ectodomain of M2 (M2e), which is known to be highly conserved.3 Subunit vaccines, which contain only select parts of the target pathogen, are a promising avenue for taking advantage of conserved epitopes. Because subunit vaccines deliver only the foreign material required to elicit a desired immune response, it can be possible to deliver a higher dose while avoiding side effects caused by unproductive viral components. However, the delivery of disassembled viral components can pose problems because antigens (especially conserved epitopes) can be poorly immunogenic and relatively unstable on their own.4 Nanoparticles offer a way to address this issue and are currently being heavily researched for many drug and vaccine applications.5-7 The use of nanoparticles for immunization involves anchoring the antigenic components to the particle surface or entrapping them within the particle structure. This approach offers several advantages over delivering neat antigen including increasing local antigen concentration and protecting the antigenic components from degradation. There are a wide variety of nanoparticles currently being studied for vaccine development, each with their own advantages and limitations, which will be discussed more in depth in the first chapter.6, 8-11 One particularly promising nanoparticle vaccine delivery technology is micelles, specifically those composed of peptides or lipidated peptides, termed peptide amphiphiles (PAs). Peptide amphiphile micelles (PAMs) can have several advantages in addition to those already mentioned for nanoparticle vaccines -- namely the enhancement of antigen-cell interactions, trafficking, and immunogenicity.11- 13 Additionally, PAMs form spontaneously in aqueous conditions, making their preparation simple and relatively inexpensive. In this work, PAMs were developed using antigens from the well- conserved ectodomain of M2 (M22-16 and M21-24) and their potential as a universal influenza vaccine was evaluated. In the second chapter, I discuss work I completed with M22-16 peptides and PAMs, including the discovery that the M22-16 antigen without any lipidation or modification also self-assembles into nanoparticles. This allowed us to investigate the impact lipidation has on micelle physical properties and immunogenicity decoupled from the effect of self-assembly. This was a unique opportunity because in other studies done by our research group and others, peptide antigens alone have not self-assembled. Excitingly, there were distinct differences in physical properties between the PAMs and the unmodified M22-16 peptides that formed peptidyl micelles (PMs). The differences in the murine immune response against PAMs versus PMs was less pronounced, as the IgG titers and BMDC activation were quite similar between both micelle types. However, the IgM and initial IgG responses were significantly stronger in PAM-vaccinated mice. In the third chapter, I further explored a phenomenon discovered in the second chapter, namely that PAMs elicited a lower-than-expected IgG antibody titer compared to unmodified M2 antigen. The M21-24 antigen was used in this research chapter to see if the same effect was observed when the antigen was expanded from the previously used M22-16 antigen. Indeed, the trends between vaccine groups were quite similar whether the M22-16 or M21-24 antigen was used. Interestingly, we found that PAMs elicited off-target antibody production against the non-native parts of the PA, thus demonstrating the importance of careful consideration when chemically modifying peptide antigens. Despite this, antibodies generated in response to PAMs were still able to recognize the M2 protein just as well as the antibodies generated in PM-vaccinated mice. In the fourth chapter, I discuss approaches that were taken to mitigate off-target antibody production, namely the use of different linkers between the non-native flanking regions and the antigen. While some of the linkers induced changes in micelle shape and peptide secondary structure, all of the linker-containing formulations still induced the production of off-target antibodies. However, one of the formulations (Pam2CS-M22-16-PEG2-(KE)4), which was lipidated with a Pam2CS- moiety on the N-terminus in an attempt to mimic the adjuvanting effects of Pam2CSK4, did elicit antibody titers similar to mice vaccinated with a combination of PAMs and Pam2CSK4. This, along with BMDC activation results, confirmed that the Pam2CS- moiety retained its ability to act as an adjuvant, as well as stimulate antibody production against the M22-16 antigen, which could have applications in situations where co-delivery of an adjuvant and antigen is required. In the fifth chapter, I delineate the remaining work needing to be completed to publish my three manuscripts and future directions for this project. Developing a non-micellized M22-16 peptide control will be essential for differentiating between the effects of micellization and lipidation on immunogenicity. Another important additional work to increase the legitimacy of this platform should include evaluating the type and magnitude of the immune response to intranasal vaccination as this would likely generate a more favorable response in line with the natural immune response to influenza -- i.e., IgA and IgG2a production. Finally, it would be valuable to evaluate the functionality and protectivity of the immune response by antibody functionality assays, survival after influenza challenge, and other similar approaches.
Table of Contents
DOI
PubMed ID
Degree
Ph. D.