Designing Multi-Epitope Supramolecular Nanofiber Vaccines for Autoimmunity and Infectious Disease

Thumbnail Image




Shores, Lucas Scott


Collier, Joel H

Journal Title

Journal ISSN

Volume Title

Repository Usage Stats



While promoting a highly targeted immune response, linear peptide epitopes generally have reduced immunogenicity and a lack of conformational specificity which has hampered the development of peptide-based vaccines. In contrast, supramolecular assembly is capable of considerably enhancing immunogenicity and promoting long-term immune responses to peptide epitopes. Immune responses generated with different vaccine platforms but targeted to the same peptide epitope, however, can vary greatly in magnitude and phenotype depending on the physicochemical cues, adjuvant, and density with which that immune epitope is displayed, influencing vaccine outcomes. Greater understanding of how design criteria such as adjuvant choice, epitope density, and multi-epitope immunization adjust the magnitude and phenotype of immune responses—in addition to how these changes in magnitude and phenotype affect therapeutic outcomes—would greatly benefit the development of peptide-based vaccines for treatment and prevention of disease. In this effort, the Collier lab has previously established that the density of T-cell epitope content within molecular self-assemblies can alter the cellular phenotype of antigen-specific T-cells in mice. Further modifications to nanofiber physical properties such as size and surface charge have also been recently shown to modulate the immunogenicity of peptide nanofiber constructs. It remains unclear, however, the extent to which combinations of B- and T-cell epitope content within nanofibers affects the antibody and T-cell responses. Further, the magnitude of antibody and T-cell responses can be magnified by the presence of adjuvants, but with adjuvant specific alterations to antibody subclass and T-cell phenotype. This work capitalizes on these advances in supramolecular nanofiber vaccines and extends them further in three separate contexts. Using peptide epitopes from inflammatory cytokines (Chapter 3), HIV protein mimics (Chapter 4), and influenza A virus (Chapter 5), we demonstrate how titrating epitope content controls adaptive immune responses and how these responses are influenced by the presence of different adjuvants. While previous work with adjuvants has identified general principles of phenotype, typically characterized as Th1 vs Th2 responses, this work seeks to understand and expand context-specific effects, e.g. optimized phenotypes for anti-cytokines immunizations compared to infectious disease. In Chapter 3, epitopes were predicted, tested, and optimized in the Q11 system to determine an immunogenic IL-17A epitope for the treatment of psoriasis. The epitopes were identified through a literature search and Kolaskar-Tongaonkar antigenicity prediction. High-scoring epitopes were tested in mice for immunogenicity and one was selected based on antibody titer. The formulation space combining the selected epitope with different combinations of T-cell epitope was also characterized, and combinations of B- and T-cell epitope content were determined to be predictive of antibody titer in this system. Responses were then enhanced with either a Th1 or a Th2 biasing adjuvant and tested for efficacy in a mouse model of psoriasis. Reduced disease symptoms were correlated with specific antibody subclasses in a first demonstration of the importance of antibody subclass for active immunotherapy. To further understand the quantitative importance of B cell signaling induced by supramolecular nanofibers, we developed a system of peptide-based activation of human B cells in vitro. In Chapter 4, we describe the first use, to our knowledge, of a peptide epitope for activation of the VRC01 Ramos B cell line. This cell line was described in 2017 and contains a B-cell Receptor (BCR) that is identical to the anti-HIV broadly-neutralizing monoclonal antibody VRC01. Other groups had previously published epitope mimicking peptide epitopes, termed mimotopes, which bind the VRC01 monoclonal antibody despite little to no sequence similarity to the native protein. We tested the highest performing mimotopes with the Q11 system and found an epitope-density-dependent effect for one of the mimotopes that was independent of nanofiber concentration. We then tested this epitope at a variety of epitope densities, with or without adjuvant. This study provides evidence that epitope density on supramolecular nanofibers has the potential to directly influence B-cell signaling, and how adjuvants can modulate this effect. Finally, we applied a Design-of-Experiments (DOE) approach to a multi-epitope system and combined it with the novel adjuvant KEYA20Q11. As attempts to make a universal influenza vaccine have become more refined, the specificity of peptide-based vaccines has had increased interest for their ability to avoid immunodominant epitopes in the native protein. Combining T-cell and B-cell responses has further shown promise in the field, but there remains a significant deficit in methods to enhance the T-cell response to a given epitope. In Chapter 5, we first optimize peptide nanofiber formulations in the context of adjuvant-free formulations. The addition of adjuvants does not alter the optimal response profile for antibody and T cell responses in this context, and we first use traditional adjuvants to enhance these responses before challenging with influenza virus. Unfortunately, this did not result in protection from viral challenge and a novel adjuvant based on the therapeutic polypeptide Glatiramer Acetate was used. This adjuvant provides universal T-cell help to enhance B-cell responses and increases the magnitude of co-assembled CD4+ and CD8+ responses. The protective efficacy of the peptide nanofiber formulation was enhanced by a combination of this novel adjuvant with the more traditional adjuvant of cyclic-di-AMP. While formulations with and without KEYA20Q11 both provided equal protection from influenza-induced weight loss, the enhanced T cell responses from the KEYA20Q11 group correlated with reduced viral titers. This work demonstrates how quantitative optimization in conjunction with adjuvant choice can lead to the enhancement of universal B- and T-cell responses for protection from influenza. Taken as a whole, this work represents a significant contribution to the understanding of how peptide epitope content and phenotype of adaptive immune responses influences their efficacy in the contexts of autoimmune and infectious disease. While not all materials-based vaccine platforms may follow the precise rules laid out through the use of the supramolecular peptide systems, the quantitative approach and analysis of immune cell phenotype should be broadly applicable. This work relies heavily on responses from in vivo models or from human cell responses which broadens the translatability of the approach. The completion of this work should facilitate the next generation of supramolecular nanofiber vaccines for autoimmunity and infectious disease.




Biomedical engineering, Immunology, Anti-cytokine Vaccine, Biomaterials, Immune Engineering, Influenza Vaccine, Nanofiber Vaccines



Shores, Lucas Scott (2021). Designing Multi-Epitope Supramolecular Nanofiber Vaccines for Autoimmunity and Infectious Disease. Dissertation, Duke University. Retrieved from


Dukes student scholarship is made available to the public using a Creative Commons Attribution / Non-commercial / No derivative (CC-BY-NC-ND) license.