*3.1.2 Consensus based approach: computationally optimized broadly reactive antigens (COBRAs)*

Furthermore, in order to overcome the extreme variability of influenza HA, in particular at the head region, Giles and Ross, [44] described the generation of computationally optimized broadly reactive antigens (COBRAs) for the influenza HA. The COBRA-based approach is a classic reverse vaccinology approach based on multiple layering of consensus HA protein sequences, followed by the generation of a final consensus sequence capable of recapitulating, in a unique protein, amino acid changes undergone by influenza virus from the past years to the present [45]. In this approach, a phylogenetic tree is inferred using hemagglutinin (HA) amino

acid sequences. Primary and secondary consensus sequences are constructed, and the secondary consensus sequences are subsequently aligned to provide the resultant consensus, known as COBRA. In multiple clinical investigations, a firm called Sanofi-Pasteur used this method with a mechanism called Elicite HAI+ antibodies [46–49].

More specifically, vaccination of mice with H1N1-based COBRA candidates resulted in broad HAI activity against a panel of 17 H1N1 virus strains. Furthermore, when inoculated mice were challenged, there was little or no detectable viral replication, as found in animals immunized with a matching approved vaccine [46]. Similarly, previous studies describing the design and generation of H5N1-based COBRA found that mice, ferrets, and nonhuman primates (Cynomolgus macaques) vaccinated with COBRA clade 2 HA H5N1 virus-like particles (VLPs) had higher HAI antibody titers recognizing different isolates representing divergent subclades [44, 49].

Aside from the COBRA-based strategy, there are several potential vaccines targeting the HA head. Song et al. [50] demonstrated the production of a fusion protein comprised of the globular HA head domains (HA1–2, spanning amino acids 62–284) from H7N9 and the Salmonella typhimurium flagellin (fliC) produced in *Escherichia coli* (*E. coli*). The authors chose fliC as a potent Toll-like receptor-5 (TLR5) ligand in order to induce an innate immune response with subsequent induction of cytokine production and dendritic cell activation, ultimately leading to higher titers of antigen-specific IgG recognizing different isolates representing divergent subclades [44, 49].

#### *3.1.3 Vaccines targeting internal viral proteins*

Internal influenza virus proteins are often highly conserved, making them viable targets for a universal vaccination. Although these proteins are rarely detected on virions or cell surfaces, making them inaccessible to antibodies, they are abundant in infected cells, where they are also processed and presented to T cells through major histocompatibility complex molecules. T cells have therefore been proven to play a significant role in influenza virus immunity. In this approach, NP and M1 have been widely studied as possible targets for universal T cell–based vaccine. Virus-based and DNA vaccination approaches have been shown in animal models to induce protective immune responses, and they are now being studied in a variety of clinical trials [41]. Over two consecutive influenza seasons, Evans et al. [51] conducted a phase 2b, randomized, placebo-controlled, double-blind trial of a recombinant viral-vectored vaccine (modified vaccinia Ankara expressing virus nucleoprotein and matrix protein 1; MVA-NP + M1), which has been shown to induce both CD4 and CD8 T cells, at eight outpatient clinical trial sites in Australia. They wanted to see if generating extra responses to conserved CD4 and CD8 T-cell antigens improves routine influenza vaccination. Based on their findings, they concluded that MVA-NP + M1 was well tolerated, with no vaccine-related major side effects. When administered within 28 days of normal QIV immunization, a vaccine intended to stimulate modest T-cell responses to cross-reactive internal proteins of influenza A did not result in an increase in incidence.

Finally, another notable vaccine technique is the epitope-based Multimeric-001 (M-001) candidate vaccine, which is now being tested in clinical trials. This vaccine, initially published by Ben-Yedidia et al. and later produced by BiondVax Pharmaceuticals Ltd., is made up of B- and T-cell epitopes taken from influenza A and B strains, containing nine conserved epitopes from the HA (including the globular

*Influenza Viruses: Targetting Conserved Viral Ha-Stem, Matrix and Nucleo-Proteins to Disarm… DOI: http://dx.doi.org/10.5772/intechopen.104770*

head), NP, and M1 proteins [38, 52, 53]. To compensate for M-001 peptides' poor immunogenicity and expensive cost, the epitopes are concatenated in triplicate into a single recombinant protein generated in E. coli. M-001 has been evaluated in both preclinical and clinical research, and it has been shown to protect mice against infection with various influenza strains while also being safe and generating both B- and T-cell specific immune responses [38, 53]. However, M-001 alone does not elicit HAI antibodies, which can only be generated when M-001 is followed by a boosting with seasonal or pandemic strain specific vaccinations [54].
