Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Epitope-targeting platform for broadly protective influenza vaccines

  • David F. Zeigler,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing

    Current address: Lonza Pharma and Biotech, Bend, Oregon, United States of America

    Affiliation TRIA Bioscience Corp., Seattle, Washington, United States of America

  • Emily Gage,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Writing – review & editing

    Current address: Lonza Pharma and Biotech, Bend, Oregon, United States of America

    Affiliation TRIA Bioscience Corp., Seattle, Washington, United States of America

  • Christopher H. Clegg

    Roles Conceptualization, Formal analysis, Funding acquisition, Investigation, Project administration, Supervision, Writing – original draft, Writing – review & editing

    Affiliation TRIA Bioscience Corp., Seattle, Washington, United States of America

Epitope-targeting platform for broadly protective influenza vaccines

  • David F. Zeigler, 
  • Emily Gage, 
  • Christopher H. Clegg


Seasonal influenza vaccines are often ineffective because they elicit strain-specific antibody responses to mutation-prone sites on the hemagglutinin (HA) head. Vaccines that provide long-lasting immunity to conserved epitopes are needed. Recently, we reported a nanoparticle-based vaccine platform produced by solid-phase peptide synthesis (SPPS) for targeting linear and helical protein-based epitopes. Here, we illustrate its potential for building broadly protective influenza vaccines. Targeting known epitopes in the HA stem, neuraminidase (NA) active site, and M2 ectodomain (M2e) conferred 50–75% survival against 5LD50 influenza B and H1N1 challenge; combining stem and M2e antigens increased survival to 90%. Additionally, protein sequence and structural information were employed in tandem to identify alternative epitopes that stimulate greater protection; we report three novel HA and NA sites that are highly conserved in type B viruses. One new target in the HA stem stimulated 100% survival, highlighting the value of this simple epitope discovery strategy. A candidate influenza B vaccine targeting two adjacent HA stem sites led to >104-fold reduction in pulmonary viral load. These studies describe a compelling platform for building vaccines that target conserved influenza epitopes.


The rapid mutation rate of influenza viruses fuels seasonal epidemics that cause >0.25 million deaths annually and facilitates occasional pandemic outbreaks that can lead to >20 million fatalities [15]. Human infections are caused by type A (IAV) and B (IBV) viruses. Strains are classified based on antigenic variation in hemagglutinin (HA) and neuraminidase (NA), the two major surface proteins which respectively enable viral fusion and budding. While vaccination is the best prophylactic, there is still tremendous need for improvement. The fundamental problem with current vaccines is that they elicit antibodies to mutable regions of the HA head that have limited homology between strains [6, 7]. Thus, antigenic mismatch between vaccine and circulating strains severely limits effectiveness. Moreover, selective pressure on unstable epitopes favors escape mutants with substitutions that abrogate antibody binding, thereby undermining long-term protection. This continual antigenic drift forces vaccines to be updated annually based on prediction of the strains that will dominate the upcoming year [8]. Influenza vaccines that can stimulate long-term broadly protective humoral immunity are needed.

Foundational studies have identified sequences on the three surface proteins (HA, NA, and M2, a proton channel critical for viral replication) that are highly conserved across strains, making them potential targets for broadly protective vaccines. These sites do not naturally stimulate appreciable antibody responses and much effort has been devoted to overcoming this problem [9]. One of the most common targets is the HA stem, which is substantially more conserved than the head subunit [10]. Many antigens have been designed to focus antibody responses on this locale. For instance, hyper-glycosylated, computationally optimized (COBRA), and mosaic HA vaccines are in preclinical development [1115]. A phase I trial testing headless stem-ferritin nanoparticles is underway (, NCT03814720) [16]. Interim phase I results for chimeric HA-based vaccines showed suboptimal memory responses to the stem [17]. Epitope-targeting platforms–e.g., conjugate, virus-like particle (VLP), and peptide vaccines–have also been tested in clinical trials [6, 7, 1820]. These antigens present conserved epitopes outside their native context, which avoids interference from the HA head and other mutation-prone domains [21]. Epitope-targeting platforms have largely concentrated on M2e, the exposed ectodomain of M2 that is conserved separately within IAV and IBV [20, 2224]. These vaccines elicit M2e-specific antibodies in humans, but at inefficacious levels [25]. Furthermore, it is unclear whether targeting M2e alone can stimulate sufficient protection [6, 2631]. Designing antigens that elicit strong focused responses to conserved influenza epitopes remains a major challenge.

We are developing a vaccine platform based on peptide nanoparticles that induce functional antibodies to small molecules and protein-based epitopes. This technology is based on peptide monomers (~70 amino acid) made using solid phase synthesis that consist of three functional domains: an amphipathic helix that drives nanoparticle self-assembly, two universal CD4 T cell epitopes that mediate high-affinity and long-lived antibody responses, and a targeted B cell epitope at one or more predetermined sites [3235]. To enhance immunogenicity, the nanoparticles are paired with GLA-SE, an adjuvant consisting of a toll-like receptor-4 agonist in a stable emulsion [36]. This adjuvant promotes TH1-mediated antibody class switching and antibody-dependent cellular cytotoxicity (ADCC), which are requirements for protection mediated by non-neutralizing antibodies to sites such as M2e [20, 37]. Previously, we used the platform to target Helix A, a conserved site on the HA stem that is bound by broadly neutralizing antibodies [35]. The vaccine partially protected mice from a lethal H1N1 challenge, confirming the antiviral potential of the platform. This proof-of-concept also illustrated the platform’s unique ability to present helical epitopes in their native conformation, which is difficult for most epitope-targeting platforms [38, 39]. Herein, we demonstrate how this platform can be used to construct pan-subtype influenza vaccines.

Materials and methods


This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, the US Public Health Service (PHS), and the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC). Protocol #2019–17 was approved by the Institutional Animal Care and Use Committees (IACUC) of the Infectious Disease Research Institute which operates under a currently approved Assurance #A4337-01, which is in accordance with PHS Policy for Humane Care and Use of Laboratory Animals.


Peptides were synthesized by Bio-Synthesis Inc. (Lewiston, TX). All peptides contained N-terminal acetyl units and chloride counterions. The peptide monomer used in these experiments contains 4 IKKIEKR heptad repeats fused to TCEs selected from Measles virus F2 protein (LSEIKGVIVHRLEGV) and Hepatitis B surface antigens (FFLLTRILTIPQSLD) [40, 41]. These peptides were made using standard SPPS chemistry with Fmoc protecting groups. The M2e 1xC-terminus antigen was made by synthesizing the M2e sequence (SLLTEVETPT) onto the C-terminal of the coiled-coil domain with a Gly linker. Peptides with two epitope copies on the self-assembly domain were synthesized in the following manner: 1), the target B cell epitope sequence was synthesized separately and purified with its reactive functionalities protected; 2), the self-assembling peptide monomer was synthesized with labile protecting groups on the desired Lys sidechains (heptad f positions) located in the first and fourth heptad repeats, 3), while still on the SPPS solid support, these Lys ζ-amine groups were deprotected and bonded to the C-terminal carboxylic acid of the B cell epitope sequence using standard SPPS amide formation chemistry, and 4), all remaining protecting groups were removed to yield the final peptide. This strategy was used to build antigens targeting M2eIAV, M2eIBV, NA222, NA238, HA127, and HA1231. The Helix AH1 and Helix AIBV sequences were synthesized onto the N-terminus of the carrier peptide during SPPS.

Dynamic light scattering

DLS spectroscopy was performed using a Zetasizer Nano (Malvern Instruments, UK) with a 4 mW He–Ne laser (633 nm) and a fixed detection angle (173°). To avoid interference from the adjuvant emulsion, peptides were formulated without GLA-SE in PBS or MOPS (100 mM, 50 mM NaCl, pH 7.5) at the concentration used for immunizations. Solutions were filtered through a 0.2 μm nylon membrane and loaded into a plastic microcuvette. Measurements were carried out in general purpose model with the following parameters: material setting was protein (refractive index = 1.440), dispersant setting was water (viscosity = 0.8872 cP, refractive index = 1.330), 10 cycles averaged per measurement, and 30 second temperature equilibration at 25°C.

B cell epitope discovery

Protein sequences of influenza B viruses were created using human isolate sequences from the NIAID Influenza Research Database and Global Initiative on Sharing All Influenza Data [42, 43]. Sequences were sorted to exclude duplicate sequences. The final data set contained 3182 HA and 3331 NA sequences. Epitopes were aligned with this library using MUSCLE to identify contiguous regions of homology between strains [44]. Homologous regions were identified on published X-ray diffraction structures of representative HA (B/Yamanashi/166/1998) and NA (B/Brisbane/60/2008) proteins. The numbering of NA238, HA127, and HA1231 are relative to the position of the start codon Met residue in the respective proteins.

Sequence homology

Homologous and nonhomologous substitutions were tallied for each residue in the putative B cell epitopes. Amino acid identity and frequency at each position was calculated. The relative prevalence of each substitution was used to generate a visual representation of conservation.

Protein modeling

Files depicting X-ray crystallography structures of representative HA [42] and NA [43] proteins were downloaded from the RCSB Protein Data Bank (PDB). These PDB files were opened with the Visual Molecular Dynamics (VMD) viewer [45]. Proteins were depicted with the ColorID coloring method and Surf drawing method. Glycosylation was depicted using bonds.


Mice (Charles River Laboratories) were housed and handled by highly trained researchers under specific pathogen-free conditions with easy access to food and water within the Infectious Disease Research Institute vivarium (Seattle, WA). Since the vaccine’s CD4 T cell epitopes bind promiscuously to a broad repertoire of MHCII molecules, our experiments employed outbred female CD-1 mice (6–8 wks) to more accurately model immune responses in a genetically diverse population like humans. Peptides were dissolved phosphate-buffered saline (PBS) or MOPS (100 mM, 50 mM NaCl, pH 7.5) buffers and filtered through a 0.2 μm nylon membrane to create immunization stocks. Final concentrations of these stocks were determined by amino acid analysis (AAA Service Laboratory, Damascus, OR). The peptides were combined on the day of immunization with GLA-SE adjuvant containing 5 μg of the synthetic TLR4 agonist, GLA, formulated in a final 2% oil-in-water stable emulsion. The adjuvant was provided by Immune Design Corp (Seattle, WA). Mice, which were inoculated under isoflurane anesthesia, received 10 μg of each indicated peptide diluted in 100 μL total volume, 50 μL of which was injected in each hind limb using a prime-boost regimen (d0 and d21). Serum was collected on d35 and used to measure antibody responses. Influenza challenge experiments were performed by infecting mice intranasally with 5LD50 dose of A/California/07/2009 or B/Florida/04/2006 in 50 μL PBS. Mice were monitored daily for 14 days to measure overall health, body weight changes and survival rates. According to the humane endpoint guideline, mice losing 25% of their body weight relative to the baseline weight were euthanized immediately by carbon dioxide overdose followed by cervical dislocation (euthanized mice, n = 262; found dead mice, n = 26). Mice were monitored for weight loss and other signs of virus induced morbidity daily and sacrificed if weight loss exceeded 25% of initial body weight. Monovalent challenge data represents three combined experiments with ntotal = 18 (M2eIAV), 25 (NA222), 22 (Helix AH1), 21 (M2eIBV), 16 (NA328), 21 (Helix AIBV), 20 (HA127), and 20 (HA1231). The bivalent studies were single experiments (n = 10 mice/group for M2e + Helix A bivalents and n = 11/group for HA127 formulations). Each mouse received 10 μg of the indicated peptides. HA127 monovalent and M2eIBV + Helix AIBV bivalent groups (n = 3/group) were included as pulmonary controls for the second bivalent study. For the pulmonary analysis, whole lungs of a subset of mice (n = 3/group, chosen randomly from groups receiving HA127 formulations) were flash frozen on day 4 post infection for viral titer determination. Briefly, frozen lungs were homogenized using the gentleMACS Dissociator M tubes in 1mL sterile PBS and viral titers determined by 50% tissue culture infectious dose in Madin-Darby canine kidney (MDCK) cells.

Antibody assays

Serum samples were serially diluted 5-fold from 1/20 in blocking buffer (3% BSA in PBST) and IgG endpoint titers were assayed by ELISA using previously-reported methodologies [32]. Endpoint titers were calculated using GraphPad Prism (GraphPad Software, San Diego, CA). Antibodies to M2eIAV, M2eIBV, NA220, NA238, HA127, and HA1231 were detected using cysteine-terminated synthetic peptides conjugated to BSA through maleimide crosslinking chemistry. Helix A titers, as well as cross-reactivity of HA127 and HA1231 antisera, were measured using recombinant HA from A/California/07/2009 or B/Malaysia/2506/2004 (Protein Sciences). NA238 antisera was screened against an available recombinant NA (A/Thailand/1(KAN-1)/2004) with a sequence nearly identical (PRPNDGT) to the NA238 epitope.

Plaque reduction neutralization titer (PRNT)

Serum samples from immunized mice were inactivated by incubation at 56 °C for 30 min. Inactivated serum samples were serially diluted two-fold in DMEM medium without FBS in a 96-well beginning with a 1:2 dilution in a total volume of 100 μL. Following serum dilution, 100 μL of diluted B/Florida/4/2006 virus (50 pfu) was added to all serum samples with TPCK-trypsin (1 μg/mL). Virus: serum mixtures were incubated at 37 °C for 60 min. Following incubation, virus−serum mixtures were incubated with MDCK cell monolayers (200 μL/well) in 6-well plates at 33°C for 60 min with rocking to distribute the medium every 15 min. Wells were overlaid with 1% agarose-MEM and incubated for 3 days at 33°C in a CO2 incubator. Following this incubation, plaques were fixed with 4% paraformaldehyde (PFA) and stained with crystal violet prior enumeration. Negative (media only) and naive murine serum samples were also assessed. Neutralizing antibody titers are presented as the highest total serum dilution capable of reducing the number of plaques by 50% compared to a virus only control (PRNT50).

NA-Star assay

Serum samples from NA1 and NA2 immunized and naive mice were assayed for NA enzymatic inhibition using the NA-Star influenza neuraminidase inhibitor resistance detection kit (Applied Biosystems). To measure sera-mediated inhibition, immunized and naive sera was serially diluted two-fold in NA-Star assay buffer in white, flat-bottom, 96-well cell culture plates. Virus (B/Florida/04/06 or B/Malaysia/2506/04) was diluted to the determined 3EC50 (half-maximum effective concentration) and 25 μL was added to each well. The plates were incubated for 30 min at 37 °C. Data points were expressed as percent inhibition of maximal NA enzymatic activity, which was determined by the activity of virus without the addition of sera. ELISA signals were fit with an inhibition regression algorithm and IC50 values determined using GraphPad Prism.


Maximizing immunogenicity to an M2e-targeting antigen

We have previously reported that peptides synthesized with one B cell epitope located at either the N- or C-terminus induced equivalent antibody responses [35]. To test whether immunogenicity could be improved by increasing B cell epitope multiplicity [32, 34, 46, 47], mice received a prime boost immunization (S1 Fig) with peptides containing either a c-terminal M2eIAV epitope or two M2eIAV epitopes located on the self-assembly domain (S1 Table). As indicated in S2 Fig, antibody titers were >103-fold higher in animals receiving peptides containing two M2eIAV epitopes. This result provides further evidence that increasing epitope valency enhances B cell receptor engagement and resultant immune responses.

Monovalent and bivalent vaccines targeting known influenza epitopes

We next investigated whether the 2x-self-assembly domain template could generate protective responses to other linear influenza epitopes, including M2eIAV and M2eIBV [20, 4851] and a sequence lining the NA active site, NA222 (S2 Table), that is nearly 100% conserved across all influenza subtypes [52]. Also included in this study were two stem-targeting peptides (the previously reported Helix AH1 and a new pan-IBV antigen, Helix AIBV) [5355]. The Helix A monomer designs contain one B cell epitope copy at the N-terminus. This design utilizes the natural helicity of the peptide’s self-assembly domain to constrain and present the epitope sequence as a helix [35]. Dynamic light scattering (DLS) verified each peptide formed nanoparticles (20–40 nm mean hydrodynamic diameters) in aqueous buffer (S3 Fig). We have previously verified that peptides lacking the self-assembly domain fail to reach these size distributions (data not shown). Antibody titers induced by each epitope were comparable (Fig 1A and 1B), although Helix AIBV yielded more variable titers to recombinant HA than Helix AH1. Mice were then challenged with H1N1/A/California/07/2009 (Fig 1C and 1D) or B/Florida/04/2006 (Fig 1E and 1F). Respectively, M2e, Helix A and NA222 vaccines conferred approximately 75%, 70% and 50% survival regardless of the challenge strain, although weight loss trends across experiments were indistinct. This consistency exemplifies the versatile “plug-and-play” nature of the platform and substantiates its potential for building antiviral vaccines.

Fig 1. Peptides targeting conserved IAV and IBV epitopes stimulate robust antibody responses and confer protection against lethal challenge.

CD-1 mice (n = 8) were immunized in a prime-boost regimen with the indicated (A) IAV or (B) IBV peptides plus GLA-SE (or GLA-SE only as a control) and d35 sera was assayed for titers by ELISA. A one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test was used for statistical analysis (*P<0.05, n.s. not significant). On d42, mice were challenged with (C,D) A/California/07/2009 or (E,F) B/Florida/04/2006 and then monitored for survival and weight loss plotted as mean ± S.E.M. Monovalent challenge data was compiled from three experiments (n = 5-8/experiment). Survival curves were compared by log-rank Mantel-Cox test. Data from each weight loss time point were compared by one-way ANOVA followed by Dunnett’s multiple comparisons test. Color coded asterisks without brackets denote significance between control and indicated test group. Brackets indicate comparison between test groups. For weight loss, significance over the control is shown until maximum difference and comparison between test groups was maximum on the designated day (*P<0.05, **P<0.01, ***P<0.001, ****P<0.0001).

The partial protection conferred by these vaccines signaled that combining peptide antigens may further improve efficacy. To test this concept, the two best antigens (M2e and Helix A) were mixed to create pan-H1 and -IBV formulations. Peptide mixtures exhibited ~30 nm diameters (S4 Fig) by DLS, suggesting co-formulation does not interfere with assembly or cause aggregation. Bivalent formulations stimulated antibodies to each epitope (Fig 2A and 2B), boosted survival to 90% (Fig 2C and 2E) and statistically decreased weight loss (Fig 2D and 2F) over control mice 1–2 days earlier than their composite monovalent vaccines. Dose-ranging studies comparing 20 μg monovalent formulations to the bivalent made from 10 μg of each peptide have confirmed that the improved protection conferred by bivalent formulations is not due to peptide dose (data not shown). These data suggest that targeting two influenza epitopes simultaneously has the potential to improve protection.

Fig 2. M2e + Helix A coformulations stimulate epitope-specific antibody responses to both targets and improve antiviral protection.

CD-1 mice (n = 11) were immunized with (A) M2eIAV + Helix AH1 or (B) M2eIBV + Helix AIBV formulations (10 μg/peptide) in GLA-SE and d35 sera was assayed for antibody titers by ELISA. Statistical differences were calculated with an unpaired two-tailed t-test (*P<0.05). On d42, mice were challenged with (C,D) A/California/07/2009 or (E,F) B/Florida/04/2006 and then monitored for survival and weight loss plotted as mean ± S.E.M. Survival curves were compared by log-rank Mantel-Cox test. Data from each weight loss time point were compared by one-way ANOVA followed by unpaired two-tailed t-test. Asterisks denote significance between control and indicated test group. For weight loss, significance over the control is shown until maximum difference (**P<0.01, ***P<0.001, ****P<0.0001).

Identification and validation of conserved influenza B antibody epitopes

These encouraging results suggested that bivalent vaccine efficacy might be improved by substituting more protective antigens in the formulation. To identify potential new epitopes, homologous amino acid stretches within IBV HA and NA were located on published X-ray diffraction structures [56, 57]. Suitable antibody targets were ≥6 amino acids in length, surface-exposed, had a linear or looped conformation, and did not possess a glycosylation motif (Asn-X-Ser/Thr/Cys). Three sites (HA127, HA1231, NA328) were identified that showed strong sequence homology across >3000 IBV strains (S5 Fig). HA127 (Fig 3A and 3C) is situated along a raised ridge on the HA stem and lies end-to-end with the Helix A epitope, which is rotated toward a recessed hydrophobic pocket. HA127 abuts conserved glycosylated Asn residues (N25, N301, N330). HA1231 (Fig 3A and 3B) is a loop flush with the HA head. It is adjacent to several potential glycosylation sites that vary by strain (e.g. N59, N145, N163). Each residue in these HA epitopes is >99% conserved across type B viruses. The NA328 epitope (Fig 3D and 3E) lies near but is more surface exposed than NA222. To gauge their antiviral activity, peptides targeting these 3 putative epitopes were vetted in vivo. Two copies of each sequence were grafted onto identical locations within the peptide monomer (S3 Table) and prior to immunization, their ability to form nanoparticles was confirmed (S6 Fig), as was their sequence conservation with the virus challenge strain (B/Florida). Importantly, these peptides induced epitope-specific antibodies that bound recombinant protein (S7 Fig) and protected against virus (Fig 3F and 3G). As indicated, HA127 conferred 100% survival, while HA1231 and NA328 groups exhibited 80% and 65% survival, respectively. Average weight loss in the HA127 and HA1231 groups remained less than 10%, exhibiting statistically better protection than the NA328 vaccine. These results demonstrate a general ability to identify and target novel antibody epitopes using protein sequence and structural data.

Fig 3. Identification of 3 pan-IBV antibody epitopes that mediate robust antiviral protection.

X-ray diffraction images of an (A) HA trimer and (D) NA tetramer. Magnified views of the (B) HA1231, (C) HA127, and (E) NA222/NA328 epitopes are shown, with amino acid residues labeled. Protein subunits are colored grey and purple. Known and putative antibody epitopes are depicted in red and yellow, respectively, and the NA active site is shown in pink. Glycosylated Asn residues are depicted with green highlighting. CD-1 mice were immunized with peptides containing the indicated epitopes plus GLA-SE (or GLA-SE only) and antibody responses were confirmed using d35 sera. On day 42, mice were challenged with 5LD50 B/Florida/04/2006 and monitored for (F) survival and (G) weight loss plotted as mean ± S.E.M. Challenge data is compiled from three experiments (n = 5-8/experiment). Survival curves were compared by log-rank Mantel-Cox test. Data from each weight loss time point were compared by one-way ANOVA followed by Dunnett’s multiple comparisons test. Color coded asterisks without brackets denote significance between control and indicated test group; brackets indicate comparison between test groups. Significance between test and control weights are shown for each time point until the group’s maximum statistical significance. Weight loss comparisons using brackets represent the most significant difference between indicated test groups, which occurred on the designated day (*P<0.05, **P<0.01, ***P<0.001, and ****P<0.0001).

To further characterize these epitopes, the mechanism of antibody protection was determined. Neutralizing capacity was measured by plaque reduction neutralization titers (PRNT50). M2eIBV antisera served as a non-neutralizing negative control. HA1231 was the only antigen that led to PRNT50 values above the limit of detection in all mice (S8A Fig). The NA neutralizing ability of NA328 antisera was also assayed (S8B Fig). NA222 antisera served as a positive control since this epitope lies in the active site. Each target produced neutralizing antibodies with similar potency against strains representing both Yamagata (B/Florida) and Victoria (B/Malaysia) lineages. These data confirm that the epitope discovery method has antiviral utility and help explain the protection conferred by these vaccines.

Construction of maximally protective influenza B vaccines

Given its strong protection, we tested whether HA127 could be paired with the next-best IBV antigens (HA1231, M2eIBV, Helix AIBV) to enhance protection relative to the M2eIBV + Helix AIBV vaccine. Bivalent mixtures exhibited 20–40 nm nanoparticles and stimulated antibody titers to each component (S9 Fig and Fig 4A). Following challenge, the HA127 + Helix AIBV and HA127 + M2eIBV combinations conferred complete survival (Fig 4B). The HA127 + Helix AIBV group showed the least weight loss (Fig 4C) relative to the adjuvant only control and statistical improvement over the other HA127 formulations; in this group, only one mouse lost >5% body weight and all mice showed ≥100% original body weight by day 12. Pulmonary viral loads (Fig 4D) further differentiated efficacy, using the HA127 and M2eIBV + Helix AIBV formulations as positive control references. The HA127 + Helix AIBV combination led to a >104-fold reduction compared to the adjuvant only group and was the only combination to perform statistically better than HA127 and M2eIBV + Helix AIBV benchmarks. Although it is not clear how broadly applicable this multivalent strategy is for building influenza vaccines, this experiment provides further evidence that targeting two HA epitopes simultaneously can significantly improve antiviral protection.

Fig 4. Bivalent HA127-based influenza B vaccines confer strong protection against influenza B challenge.

CD-1 mice (n = 11) were immunized with HA127 + M2eIAV, HA127 + Helix AIBV, and HA127 + HA1231 formulations (10 μg/peptide) in GLA-SE. (A), Day 35 antisera were assayed for titers to each target separately by ELISA. One-way ANOVA followed by Tukey’s multiple comparisons test was used for statistical analysis (**P<0.01). On day 42, mice were challenged with 5LD50 B/Florida/04/2006 and were monitored for (B) survival, (C) weight loss plotted as mean ± S.E.M, and (D) viral load. Survival curves were compared by log-rank Mantel-Cox test. Data from each weight loss time point were compared by one-way ANOVA followed by Dunnett’s multiple comparisons test. Color coded asterisks without brackets denote significance between control and indicated test group. Brackets indicate comparison between test groups. For weight loss, significance over the control is shown until maximum difference and comparison between test groups was maximum on the designated day (*P<0.05, ***P<0.001, ****P<0.0001). For viral load, lungs were assayed four days after infection for mean tissue culture infectious dose (TCID50), with the limit of detection depicted with dashed line. One-way ANOVA followed by Tukey’s multiple comparisons test was used for statistical analysis (*P<0.05).


More broadly protective influenza vaccines would greatly reduce the global health burden caused by seasonal and pandemic outbreaks. Here, we describe a platform for building these vaccines using a ~70 amino acid peptide containing a self-assembling domain, the targeted conserved B cell epitope, and two universal CD4 T cell epitopes that bind a broad repertoire of MHCII molecules [34, 35]. Advantages of this technology include the elimination of non-relevant immunogenic sequences common to conjugate vaccines and VLPs that induce competing antibody responses including carrier suppression [32, 34, 58, 59]. Its manufacture by solid phase peptide synthesis prevents the need for chemical conjugation reactions that require extensive downstream purification [47, 60, 61], and unlike many peptide-based vaccines, it assembles into nanoparticles that facilitate humoral immunity [35, 62].

Here, we measured the protection induced by conserved epitopes on all three influenza surface proteins. Vaccines targeting M2e induced high antibody titers and the two different M2eIAV and M2eIBV sequences mediated equivalent survival (~75%) in their respective challenge experiments. This internal consistency implies that protection was dependent upon its lower copy number relative to HA and NA (~1:60:5 M2:HA:NA) [63] and/or epitope accessibility [64, 65], but independent of challenge strain or epitope sequence. With respect to the two Helix A epitope sequences, Helix AH1 induced a high uniform antibody titer, whereas Helix AIBV antibody levels were more variable which may be due to subtle differences in conformation of the IBV epitope or the glycosylation at the nearby Asn residue (N330; see Fig 3), which is not present in H1 HAs [66]. Despite these difference in titers, both epitopes induced the same level of protection that closely approximated the M2e epitopes. The NA222 epitope also induced a robust antibody titer, but protection was more limited (~50%) than the M2e and Helix A epitopes following IAV and IBV challenge. Again, this may be due to antibody accessibility since NA222 is buried in the NA active site (Fig 3E). Finally, we tested whether protection rates could be improved with bivalent formulations using the M2e and Helix A epitopes, and in both IAV and IBV challenge experiments, overall survival improved to 90% and body weights were better maintained than the monovalent vaccines. Thus, establishing the ability to co-formulate and target 2 epitopes to improve overall survival.

Having validated the technology using known epitopes, we searched for new highly conserved sequences that were readily exposed on the surface IBV hemagglutinin and neuraminidase, which has considerably less sequence variability than IAV [67]. Three sequences, HA127, NA328, HA231, were selected that are >99% identical across 3,000 independently isolated IBV strains and their induced survival following viral challenge was, respectively, 100%, 80%, and 66%. To the best of our knowledge, this is the first reported characterization of these epitopes, although residues in NA328 may reside within the epitope of a recently described anti-H5N1 monoclonal antibody [68]. Interestingly, while these three antisera recognized their peptide-conjugate ELISA reagents equivalently, the antibodies induced by the linear epitope (HA27) demonstrated better native protein binding and antiviral protection than the looped/curved epitopes (HA231 and NA328), thus suggesting a bias in the ability to induce antibodies to linear versus constrained epitopes. In this same experiment HA231 and NA328 induced similar recombinant protein antibody titers, although HA231 stimulated better protection, which could be due to the relative abundance of these two proteins on the virus. Preliminary mechanism of action studies indicated that antibodies directed against HA231 neutralized virus infectivity, which may be related to its location near the receptor binding site, and anti-NA222 and NA328 antibodies inhibited neuraminidase activity. Presumably, ADCC is also an important mechanism of protection given its role in mediating anti-M2e and anti-HA stem antibody activity [4850]. Futures studies will confirm this and test whether the novel HA stem binding antibodies directed against HA27 and Helix A can also inhibit endosomal fusion [6, 7, 10, 26].

The HA stem is the primary target for building broadly protective influenza vaccines, which is also supported by these studies. The HA127 + Helix AIBV IBV vaccine stimulated 100% survival, negligible weight loss and a 104-fold decline in viral titer relative to controls, and outperformed HA127 formulations with either HA1231 or M2eIBV, two antigens that showed better monovalent activity than Helix AIBV. Targeting these highly conserved IBV epitopes may be superior to existing antigen designs that include stem regions with lower homology or are obscured by glycosylation, the HA head or viral envelope [18, 26, 45]. Future experiments will test whether this improved antibody protection involves coordinated Fc receptor engagement and/or neutralizing activity.

Our method for eliciting antibodies to highly conserved sequences represents a new paradigm for building improved influenza vaccines. Given the putative role that these conserved subdominant epitopes play in maintaining viral function, they should be much less susceptible to mutation. To this point, mAbs specific to conserved epitopes in the HA2 stem [13, 69, 70], NA [71, 72], and M2e [73, 74] are very effective in preventing viral escape. However, the development of vaccines using this approach will require an escape mutant analysis and a need to show protection against multiple strains of virus. This is especially true for epitopes that lie in mutation-prone regions, such as HA1231. Establishing protection against strains bearing different glycosylation patterns would also corroborate their utility. Additionally, antisera should be screened against host cells or tissues to test for autoreactivity, as reported for a class of stem-specific B cells [6, 26, 75]. The improved efficacy with bivalent formulations establishes the framework for multi-epitope influenza vaccines, which is also supported by studies showing improved vaccine efficacy following antibody induction to multiple proteins [6, 7, 2831]. It is also akin to combination monoclonal therapies, where targeting disparate sites on cytomegalovirus, rabies, HIV, Zika, and Chikungunya viruses enhanced antiviral activity and prevented viral escape synergistically [7679]. Confirmation of the safety and efficacy of this vaccine platform for IBV will support its use for targeting highly conserved epitopes in IAV and other viruses.

Supporting information

S1 Table. Peptide vaccine designs used in these studies.


S2 Table. Targeted IAV and IBV epitopes in M2e, NA, and HA proteins and their corresponding peptide vaccine design.


S3 Table. Novel epitopes in IBV HA and NA (see Fig 3) and their corresponding peptide vaccine design.


S2 Fig. Increasing epitope valency improves antibody responses.

(A) Amino acid sequences of M2e antigens. The M2eIAV epitope (italics) was synthesized onto the C-terminus of the peptide monomer (1xC-terminus) or grafted onto two lysine sidechains within the self-assembly domain using isopeptide bonds (2xself-assembly domain). CD4 T cell epitopes from Measles and Hepatitis B are shown in bold. (B) Immunogenicity of peptides. CD-1 mice (n = 5) received a prime-boost immunization with GLA-SE (or GLA-SE only) and d35 titers were assayed by ELISA. A one-way analysis of variance (ANOVA) followed by Tukey’s multiple comparisons test was used for statistical analysis (***P<0.001).


S3 Fig. Peptides targeting conserved influenza A and B epitopes assemble into nanoparticles.

Dynamic light scattering was used to verify nanoparticle size of (A) M2eIAV, (B) NA222, (C) Helix AH1, (D) M2eIBV, and (E) Helix AIBV.


S4 Fig. M2e + Helix A peptide mixtures form nanoparticles.

Dynamic light scattering was used to verify nanoparticle size of (A) M2eIAV + Helix AH1 and (B) M2eIBV + Helix AIBV formulations.


S5 Fig. Evolutionary sequence profiles of new IBV antibody targets.

The amino acid sequence of each epitope is depicted, with the residue letter height proportional to its mutational frequency in aligned HA or NA sequences. Amino acids are colored according to chemical properties: green (hydrophilic), black (hydrophobic), red (acidic), and blue (basic).


S6 Fig. Peptides targeting putative influenza B epitopes assemble into nanoparticles.

Dynamic light scattering was used to verify nanoparticle size of (A) HA127, (B) HA1231, and (C) NA328.


S7 Fig. Peptides targeting putative IBV epitopes stimulate epitope-specific antibodies that bind recombinant protein.

CD-1 mice (n = 8) were immunized the indicated peptide plus GLA-SE (or GLA-SE only). Antisera (d35) from each group was screened for titers to (A) BSA-epitope conjugates or (B) recombinant HA/NA. One-way ANOVA followed by Tukey’s multiple comparisons test was used for statistical analysis of titers (***P<0.001, ****P<0.0001, n.s. not significant).


S8 Fig. Neutralization capacity varies by influenza B target.

(A) Plaque reduction neutralization titers. CD-1 mice (n = 5) were immunized (d0, d21) with the indicated peptide plus GLA-SE (or GLA-SE only). Day 35 antisera was assayed for neutralizing activity in a PRNT assay. One-way ANOVA followed by Dunnett’s multiple comparisons test was used for statistical analysis between control and indicated test group (*P = 0.0156, **P = 0.0044, n.s. not significant). Limit of detection depicted with dashed line. (B) NA neutralizing ability. CD-1 mice (n = 3) were immunized as above. Day 35 antisera was assayed for its ability to prevent cleavage of an NA substrate (see Materials and methods). One-way ANOVA followed by Tukey’s multiple comparisons test was used for statistical analysis. Color coded asterisks without brackets denote significance between control and indicated test group; brackets indicate comparison between test groups (*P<0.05, **P<0.01, n.s. not significant).


S9 Fig. Bivalent HA127-based formulations exhibit normal nanoparticle sizes.

Dynamic light scattering was used to verify nanoparticle size of (A) HA127 + M2eIAV, (B) HA127 + Helix AH1, and (C) HA127 + HA1231 formulations.



We thank Dr. Leo Poon (University of Hong Kong) for alignment and homology analysis of influenza B HA and NA sequences.


  1. 1. Paget J, Spreeuwenberg P, Charu V, Taylor RJ, Iuliano AD, Bresee J, et al. Global mortality associated with seasonal influenza epidemics: New burden estimates and predictors from the GLaMOR Project. J Glob Health. 2019;9: 020421. pmid:31673337
  2. 2. Sullivan SG, Cowling BJ. Reconciling estimates of the global influenza burden. Lancet Respir Med. 2018;7: 8–9. pmid:30553846
  3. 3. Taubenberger JK, Morens DM. 1918 influenza: the mother of all pandemics. Emerg Infect Dis. 2006;12: 15–22. pmid:16494711
  4. 4. Viboud C, Simonsen L, Fuentes R, Flores J, Miller MA, Chowell G. Global mortality impact of the 1957–1959 influenza pandemic. J Infect Dis. 2016;213: 738–745. pmid:26908781
  5. 5. Krammer F, Smith GJD, Fouchier RAM, Peiris M, Kedzierska K, Doherty PC, et al. Influenza. Nat Rev Dis Primers. 2018; 4: 3. pmid:29955068
  6. 6. Sautto GA, Kirchenbaum GA, Ross TM. Towards a universal influenza vaccine: different approaches for one goal. Virol J. 2018;15: 17. pmid:29370862
  7. 7. Estrada LD, Schultz-Cherry S. Development of a universal influenza vaccine. J Immunol. 2019;202: 392–398. pmid:30617121
  8. 8. Boni MF. Vaccination and antigenic drift in influenza. Vaccine. 2008;28, C8–14. pmid:18773534
  9. 9. Angeletti D, Kosik I, Santos JJS, Yewdell WT, Boudreau CM, Mallajosyula VVA, et al. Outflanking immunodominance to target subdominant broadly neutralizing epitopes. Proc Natl Acad Sci USA. 2019;116: 13474–13479. pmid:31213541
  10. 10. Coughlan L, Palese P. Overcoming barriers in the path to a universal influenza vaccine. Cell Host Microbe. 2018;24: 18–24. pmid:30001520
  11. 11. Eggink D, Goff PH, Palese P. Guiding the immune response against influenza virus hemagglutinin toward the conserved stalk domain by hyperglycosylation of the globular head domain. J Virol. 2014;88: 699–704. pmid:24155380
  12. 12. Sun W, Kirkpatrick E, Ermler M, Nachbagauer R, Broecker F, Krammer F, et al. Development of influenza B universal vaccine candidates using the “mosaic” hemagglutinin approach. J Virol. 2019;93: e00333–19. pmid:30944178
  13. 13. Broecker F, Liu STH, Suntronwong N, Sun W, Bailey MJ, Nachbagauer R, et al. A mosaic hemagglutinin-based influenza virus vaccine candidate protects mice from challenge with divergent H3N2 strains. npj Vaccines. 2019;4: 31. pmid:31341648
  14. 14. Carter DM, Darby CA, Lefoley BC, Crevar CJ, Alefantis T, Oomen R, et al. Design and characterization of a computationally optimized broadly reactive hemagglutinin vaccine for H1N1 influenza viruses. J Virol. 2016;90: 4720–4734. pmid:26912624
  15. 15. Allen JD, Ray S, Ross TM. Split inactivated COBRA vaccine elicits protective antibodies against H1N1 and H3N2 viruses. PLoS One. 2018;13: e0204824.
  16. 16. Mathew NR, Angeletti D. Recombinant influenza vaccines: saviors to overcome immunodominance. Front Immunol. 2020;10: 2997. pmid:31998299
  17. 17. Bernstein DI, Guptill J, Naficy A, Nachbagauer R, Berlanda-Scorza F, Feser J. Immunogenicity of chimeric haemagglutinin-based, universal influenza virus vaccine candidates: interim results of a randomised, placebo-controlled, phase 1 clinical trial. Lancet Infect Dis. 2020;20: 80–91. pmid:31630990
  18. 18. Kumar A, Meldgaard TS, Bertholet S. New platforms for the development of a universal influenza vaccine. Front Immunol. 2018;9: 600.
  19. 19. Atsmon J, Kate-Ilovitz E, Shaikevich D, Singer Y, Volokhov I, Haim KY, et al. Safety and immunogenicity of Multimeric-001 –a novel universal influenza vaccine. J Clin Immunol. 2012;32: 595–603. pmid:22318394
  20. 20. Mezhenskaya D, Isakova-Sivak I, Rudenko L. M2e-based universal influenza vaccines: a historical overview and new approaches to development. J Biomed Sci. 2019;26: 76. pmid:31629405
  21. 21. Wang TT, Tan GS, Hai R, Pica N, Ngai L, Ekiert DC, et al. Vaccination with a synthetic peptide from the influenza virus hemagglutinin provides protection against distinct viral subtypes. Proc Natl Acad Sci USA. 2010;107: 18979–18984. pmid:20956293
  22. 22. Lu W, Li H, Chen Y-H. N-terminus of M2 protein could infudce antibodies with inhibitory activity against influenza virus replication. FEMS Immunol Med Microbiol. 2003; 35:141–146. pmid:12628550
  23. 23. Song J-M, Wang B-Z, Park K-M, Rooijen NV, Quan F-S, Kim M-C, et al. Influenza virus-like particles containing M2 induce broadly cross protective immunity. PLoS One. 2011;6:e14538. pmid:21267073
  24. 24. Stepanova LA, Kotlyarov RY, Kovaleva AA, Potapchuk MV, Korotkov AV, Sergeeva MV, et al. Protection against multiple influenza A virus strains induced by candidate recombinant vaccine based on hetereologous M2e peptides linked to flagellin. PLoS One. 2015;10:e0119520. pmid:25799221
  25. 25. Kolpe A, Schepens B, Fiers W, Saelens X. M2-based influenza vaccines: recent advances and clinical potential. Expert Rev Vaccines. 2017;16: 123–136. pmid:27653543
  26. 26. Jang YH, Seong BL. The quest for a truly universal influenza vaccine. Front Cell Infect Microbiol. 2019;9: 344. pmid:31649895
  27. 27. Sun W, Zheng A, Miller R, Krammer F, Palese P. An inactivated virus approach to targeting the conserved hemagglutinin stalk and M2e domains. Vaccines. 2019;7: 117. pmid:31540436
  28. 28. Deng L, Chang TZ, Wang Y, Li S, Wang S, Matsuyama S, et al. Heterosubtypic influenza protection elicited by double-layered polypeptide nanoparticles in mice. Proc Natl Acad Sci USA. 2018;115: E7758–E7767. pmid:30065113
  29. 29. Tsybalova LM, Stepanova LA, Shuklina MA, Mardanova ES, Kotlyarov RY, Potapchuk MV, et al. Combination of M2e peptide with stalk HA epitopes of influenza A virus enhances protective properties of recombinant vaccine. PLoS One. 2018;13: e0201429. pmid:30138320
  30. 30. Wang Y, Deng L, Gonzalez GX, Luthra L, Dong C, Ma Y, et al. Double-layered M2e-NA protein nanoparticle immunization induces broad cross-protection against different influenza viruses in mice. Adv Healthc Mater. 2020;9: e1901176. pmid:31840437
  31. 31. Schotsaert M, Ysenbaert T, Smet A, Schepens B, Vanderschaeghe D, Stegalkina S. Long-lasting cross-protection against influenza A by neuraminidase and M2e-based immunization strategies. Sci Reports. 2016;6: 24402. pmid:27072615
  32. 32. Miller KD, Roque RP, Clegg CH. Novel anti-nicotine vaccine using a trimeric coiled-coil hapten carrier. PLoS One. 2014;9: e114366. pmid:25494044
  33. 33. Zeigler DF, Roque RP, Clegg CH. Construction of an enantiopure bivalent nicotine vaccine using synthetic peptides. PLoS One. 2017;12: e0178835. pmid:28570609
  34. 34. Zeigler DF, Roque RP, Clegg CH. Optimization of a multivalent peptide vaccine for nicotine addiction. Vaccine. 2017;37: 1584–1590.
  35. 35. Zeigler DF, Gage E, Roque RP, Clegg CH. Epitope targeting with self-assembled peptide vaccines. npj Vaccines. 2019;4: 30. pmid:31341647
  36. 36. Clegg CH, Roque R, Van Hoeven N, Perrone L, Baldwin SL, Rininger JA, et al. Adjuvant solution for pandemic influenza vaccine production. Proc Natl Acad Sci USA. 2012;109: 17585–17590. pmid:23045649
  37. 37. Cauwelaert ND, Desbien AL, Hudson TE, Pine SO, Reed SG, Coler RN, et al. The TLR4 agonist vaccine adjuvant, GLA-SE, requires canonical and atypical mechanisms of action for TH1 induction. PLoS One. 2016;11: e0146372. pmid:26731269
  38. 38. Li W, Joshi MD, Singhania S, Ramsey KH, Murthy AK. Peptide vaccine: progress and challenges. Vaccines 2014;2: 515–536. pmid:26344743
  39. 39. Oscherwitz J. The promise and challenge of epitope-focused vaccines. Hum Vaccines Immunother. 2016;12: 2113–2116. pmid:27058686
  40. 40. Falugi F, Petracca R, Mariani M, Luzzi E, Mancianti S, Carinci V, et al. Rationally designed strings of promiscuous CD4 (+) T cell epitopes provide help to Haemophilus influenzae type b oligosaccharide: a model for new conjugate vaccines. Eur J Immunol 2001;31: 3816–24. pmid:11745403
  41. 41. Greenstein JL, Schad VC, Goodwin WH, Brauer AB, Bollinger BK, Chin RD, et al. A universal T cell epitope-containing peptide from hepatitis B surface antigen can enhance antibody specific for HIV gp120. J Immunol. 1992;148: 3970–3977. pmid:1376346
  42. 42. Zhang Y, Aevermann BD, Anderson TK, Burke DF, Dauphin G, Gu Z, et al. Influenza Research Database: an integrated bioinformatics resources for influenza virus research. Nucleic Acids Res. 2017;45: D466–D474. pmid:27679478
  43. 43. Bogner P, Capua I, Cox NJ, Lipman DJ, et al. A global initiative on sharing avian flu data. Nature. 2006;442: 981.
  44. 44. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32: 1792–1797. pmid:15034147
  45. 45. Yusuf M, Konc J, Bing CS, Konc JT, Khairudin NBA, Janezic D, et al. Structurally conserved binding sites of hemagglutinin as targets for influenza drug and vaccine development. J Chem Inf Model. 2013;53: 2423–2436. pmid:23980878
  46. 46. Li Q, Rodriguez LG, Farnsworth DF, Gildersleeve JC. Effects of hapten density on the induced antibody repertoire. Chembiochem. 2010;11: 1686–1691. pmid:20602400
  47. 47. McCluskie MJ, Thorn J, Mehelic PR, Kolhe P, Bhattacharya K, Finneman JI, et al. Molecule attributes of conjugate antigen influence function of antibodies induced by anti-nicotine vaccine in mice and non-human primates. Int Immunopharmacol. 2015;25: 518–527. pmid:25737198
  48. 48. Saelens X. The role of matrix protein 2 ectodomain in the development of universal influenza vaccines. J Infect Dis. 2019;219: 568–574.
  49. 49. Zebedee SL, Lamb RA. Influenza A virus M2 protein: monoclonal antibody restriction of virus growth and detection of M2 in virions. J Virol. 1988;62: 2762–2772. pmid:2455818
  50. 50. Bakkouri KE, Descamps F, De Filette M, Smet A, Festjens E, Birkett A, et al. Universal vaccine based on ectodomain of matrix protein 2 of influenza A: Fc receptors and alveolar macrophages mediate protection. J. Immunol. 2011;186:1022–1031. pmid:21169548
  51. 51. Paterson RG, Takeda M, Ohigashi Y, Pinto LH, Lamb RA. Influenza B virus BM2 protein is an oligomeric integral membrane protein expressed at the cell surface. Virol. 2003;306:7–17. pmid:12620792
  52. 52. Gravel C, Li C, Wang J, Hashem AM, Jaentschke B, Xu K, et al. Qualitative and quantitative analyses of virtually all subtypes of influenza A and B viral neuraminidases using antibodies targeting the universally conserved sequences. Vaccine. 2010;28: 5774–5784. pmid:20621113
  53. 53. Wu Y, Cho MS, Shore D, Song M, Choi JA, Jiang T, et al. A potent broad-spectrum protective human monoclonal antibody crosslinking two haemagglutinin monomers of influenza A virus. Nat Commun. 2015;6: 7708. pmid:26196962
  54. 54. Dreyfus C, Laursen NS, Kwaks T, Zuijdgeest D, Khayat R, Ekiert DC, et al. Highly conserved protective epitopes on influenza B viruses. Science. 2012;337: 1343–1348. pmid:22878502
  55. 55. Ekiert DC, Bhabha G, Elsliger M-A, Friesen RHE, Jongeneelen M, Throsby M, et al. Antibody recognition of a highly conserved influenza virus epitope: implications for universal prevention and therapy. Science. 2009;324: 246–251. pmid:19251591
  56. 56. Ni F, Kondrashkina E, Wang Q. Structural basis for the divergent evolution of influenza B virus hemagglutinin. Virology. 2013;446: 112–122. pmid:24074573
  57. 57. Escuret V, Collins PJ, Casalegno J-S, Vachieri SG, Cattle N, Ferraris O, et al. A novel I221L substitution in neuraminidase confers high-level resistance to oseltamivir in influenza B viruses. J Infect Dis. 2014;210: 1260–1269. pmid:24795482
  58. 58. McCluskie MJ, Evans DM, Zhang N, Benoit M, McElhiney SP, Unnithan M, et al. The effect of preexisting anti-carrier immunity on subsequent responses to CRM197 or Qb-VLP conjugate vaccines. Immunopharmacol Immunotoxicol. 2016;38: 184–196. pmid:27121368
  59. 59. Jegerlehner A, Wiesel M, Dietmeier K, Zabel F, Gatto D, Saudan P, et al. Carrier induced epitopic suppression of antibody responses induced by virus-like particles is a dynamic phenomenon caused by carrier-specific antibodies. Vaccine. 2010;28: 5503–5512. pmid:20307591
  60. 60. Frasch CE. Preparation of bacterial polysaccharide-protein conjugates: analytical and manufacturing challenges. Vaccine. 2009;27: 6468–6470. pmid:19555714
  61. 61. Herbert JA, Kay EJ, Faustini SE, Richter A, Abouelhadid S, Cuccui J, et al. Production and efficacy of a low-cost recombinant pneumococcal protein polysaccharide conjugate vaccine. Vaccine. 2018;36: 3809–3819. pmid:29778517
  62. 62. Malonis RJ, Lai JR, Vergnolle O. Peptide-based vaccines: current progress and future challenges. Chem Rev. 2020;120: 3210–3229. pmid:31804810
  63. 63. Hutchinson EC, Charles PD, Hester SS, Thomas B, Trudgian D, Martinez-Alonso M, et al. Conserved and host-specific features of influenza virion architecture. Nat Commun. 2014;5: 4816. pmid:25226414
  64. 64. Cho KJ, Schepens B, Moonens K, Deng L, Fiers W, Remaut H, et al. Crystal structure of the conserved amino terminus of the extracellular domain of matrix protein 2 of influenza A virus gripped by an antibody. J Virol. 2016;90: 611–615. pmid:26468526
  65. 65. Schepens B, De Vlieger D, Saelens X. Vaccine options for influenza: thinking small. Curr Opin Immunol. 2018;53: 22–29. pmid:29631195
  66. 66. An Y, Parsons LM, Jankowska E, Melnyk D, Joshi M, Cipollo JF. N-glycosylation of seasonal influenza vaccine hemagglutinins: implication for potency testing and immune processing. J Virol. 2019;93: e01693–18. pmid:30355697
  67. 67. Nobusawa E, Sato K. Comparison of the mutation rates of human influenza A and B viruses. J Virol. 2006;80: 3675–3678. pmid:16537638
  68. 68. Ramamurthy M, Sankar S, Abraham AM, Nandagopal B, Sridharan G. B cell epitopes in the intrinsically disordered regions of neuraminidase and hemagglutinin proteins of H5N1 and H9N2 avian influenza viruses for peptide‐based vaccine development. J Cell Biochem. 2019;120: 17534–17544. pmid:31111560
  69. 69. Corti D, Voss J, Gamblin SJ, Codoni G, Macagno A, et al. A Neutralizing Antibody Selected from Plasma Cells that Binds to Group 1 and Group 2 Influenza A Hemagglutinins. Science. 2011,333:850–856. pmid:21798894
  70. 70. Wu Y, Cho M, Shore D, Song M, Choi J, et al. A Potent Broad-spectrum Protective Human Monoclonal Antibody Crosslinking Two Hemagglutinin Monomers of Influenza A Virus. Nature Comm. 2015,6:7708. pmid:26196962
  71. 71. Doyle TM, Hashem AM, Li C, Van Domselaar G, Larocque L, Wang J, et al. Universal anti-neuraminidase antibody inhibiting all influenza A subtypes. Antiviral Research. 2013, 100:567–574. pmid:24091204
  72. 72. Doyle TM, Li C, Bucher DJ, Hashem AM, Van Domselaar G, Wang J, et al. A monoclonal antibody targeting a highly conserved epitope in influenza B neuraminidase provides protection against drug resistant strains. Biochem Biophys Res Comm. 2013,441: 226–229. pmid:24140051
  73. 73. Beerli RR, Bauer M, Schmitz N, Buser RB, Gwerder M, Muntwiler S, et al. Prophylactic and therapeutic activity of fully human monoclonal antibodies directed against influenza A M2 protein. Virol J. 2009, 6:224. pmid:20025741
  74. 74. Treanor JJ, Tierney EL, Zebedee SL, Lamb RA, Murphy BR. Passively transferred monoclonal antibody to the M2 protein inhibits influenza A virus replication in mice. J Virol 1990, 64:1375–1377. pmid:2304147
  75. 75. Bajic G, van der Poel CE, Kuraoka M, Schmidt AG, Carroll MC, Kelsoe G, et al. Autoreactivity profiles of influenza hemagglutinin broadly neutralizing antibodies. Sci Rep. 2019;9: 3492. pmid:30837606
  76. 76. Salazar G, Zhang N, Fu T-M, An Z. Antibody therapies for the prevention and treatment of viral infections. npj Vaccines. 2017;2: 19. pmid:29263875
  77. 77. Mendoza P, Gruell H, Nogueira L, Pai JA, Butler AL, Millard K, et al. Combination therapy with anti-HIV-1 antibodies maintains viral suppression. Nature. 2018;561: 479–484. pmid:30258136
  78. 78. Keeffe JR, Van Rompay KKA, Olsen PC, Wang Q, Gazumyan A, Azzoparddi SA, et al. A combination of two human monoclonal antibodies prevents Zika virus escape mutations in non-human primates. Cell Rep. 2018;25: 1385–1394. pmid:30403995
  79. 79. Pal P, Dowd KA, Brien JD, Edeling MA, Gorlatov S, Johnson S, et al. Development of a highly protective combination monoclonal antibody therapy against Chikungunya virus. PLoS Pathog. 2013;9: e1003312. pmid:23637602