Universal vaccines: A holy grail for the flu?


Seasonal flu infects millions worldwide each year and is a major public health concern. This is aggravated by the perpetual threat of pandemic influenza strains that periodically bring health systems and economies to a standstill. Existing vaccines for the influenza virus suffer from low efficacy and protracted development cycles. So-called ‘universal vaccines’ have gained much attention in recent years as viable alternatives. In this blog, we review some of the main targets of universal vaccines and the key challenges to their successful implementation.

 The influenza viruses

The influenza viruses are single-stranded RNA viruses belonging to the Orthomyxoviridae family and are typically spherical or filamentous in shape. Transmitted by the coughs and sneezes of infected individuals, influenza causes the familiar symptoms of flu, including fever, cough, sore throat, aching and fatigue. Flu is most prevalent in the winter months, from October to February, generally referred to as flu season [1]. While we still aren’t entirely sure why influenza is so active during this period, it is thought that lower temperatures and changes in human behavior help to spread the virus [2].

A simplified diagram of the influenza virus. Haemagglutinin recognizes cell-surface receptors and mediates cell entry; Neuraminidase cleaves neuraminic (sialic) acid residues to promote the release of viral progeny; Matrix 1 mediates assembly of the membrane complex and confers structural integrity. Matrix 2 is a selective proton channel that regulates viral pH.

Unlike most viruses, influenza contains eight separate strands of RNA that collectively encode its 11 proteins. The most well-studied influenza proteins are Haemagglutinin (HA) and Neuraminidase (NA), which stud the envelope’s outer surface and mediate invasion and release from host cells, respectively. Matrix 1 (M1) forms a protein coat inside the viral envelope and plays an important role in virus assembly and in maintaining structural integrity. Matrix 2 (M2) is a helical proton channel that regulates influenza’s internal pH as it passes through different host membranes.

The Orthomyxoviridae family classifies four major types of influenza virus (A–D) based on their NA and M1 proteins. Of most clinical relevance are the co-circulating influenza A and B viruses that cause seasonal and pandemic flu, and are the subsequent targets of vaccines.

Drift and shift

The characterizing feature of influenza viruses that have made them such successful pathogens is their ability to rapidly mutate and evade the adaptive immune response. While this is a popular strategy for viruses, influenza achieves this to great effect through the combination of two phenomena known as antigenic drift and shift:

All influenza strains undergo antigenic drift – a continuous process in which the development of adaptive immunity by hosts creates selection pressures for mutations in influenza’s HA proteins [3]. As HA contains various structurally tolerant regions, its epitopes can change without hindering influenza’s ability to gain entry to cells, all while escaping host recognition.

Conversely, antigenic shift is characterized by abrupt changes to HA and NA through genome reassortment with other influenza subtypes, which occurs when two different influenza viruses co-infect the same cell. While all influenza viruses undergo antigenic drift, only the influenza A viruses are able to undergo antigenic shift due to their extensive animal reservoirs. In total, this gives the influenza A viruses the choice of 18 HA subtypes (H1–18) and 11 NA subtypes (N1–11), making for an astounding 198 potential combinations [4].

Haemagglutinin trimer. The head is labelled orange, with green hypervariable epitopes; The stalk is labelled blue (RCSB: 3znm).

By shifting the epitopes of their surface antigens so drastically, influenza A strains are able to catch our immune systems completely off-guard and rapidly proliferate through human populations. Indeed, the most devastating influenza epidemics in human history, including the 1918 Spanish Flu, the 2004 Bird Flu and the 2009 Swine Flu, are all the result of shifted influenza A strains [5].

Vaccinating moving targets

Every year, influenza viruses infect hundreds of millions of people and cause over half a million deaths [6]. While mortality rates for the flu are generally low, infection is a serious threat to the immunocompromised or those at the extremities of age, who account for the majority of mortalities [7]. Furthermore, flu outbreaks frequently overwhelm healthcare systems, causing further indirect morbidity and mortality, and constitute a significant economic burden [8].

It goes without saying that preventing influenza transmission is a major public health priority, and the most effective way to do so is through the development of influenza vaccines. The leading strategy for developing seasonal influenza vaccines relies on the annual development of inactivated influenza virus vaccines and live-attenuated influenza vaccines, according to the public health community’s best guess of what strains will predominate in the upcoming flu season. Vaccine strains of the virus are then produced using millions of chicken eggs as mini influenza incubators, and the vaccines are then administered to the general public.

However, this process is highly complex and time-consuming, typically taking 6–8 months to produce sufficient quantities. As a result, most vaccines are semi-obsolete by the time that they are made publicly available, due to the effects of antigenic drift [9]. Furthermore, errors in strain prediction can result in vaccines that significantly differ from the strain circulating for that season [10].

For pandemic vaccines, it’s a similar story. Upon recognition of a major outbreak, a mad scramble ensues as multiple manufacturers race to produce an effective vaccine. The drawbacks parallel those of seasonal vaccines and were evidenced by the devastating 2009 H1N1 swine flu epidemic, in which delayed vaccine administration was estimated to have caused hundreds of thousands of additional deaths [11,12].

Universal vaccines

It goes without saying that antigenic plasticity, the perpetual scramble to develop new vaccines, and byzantine production processes create a headache for public health authorities and vaccine manufactures, alike. As such, the status quo has culminated in the search for a more sustainable and effective alternative.

Existing vaccine design strategies are based the understanding that HA’s globular head is the most potent target for neutralizing antibody responses. The flipside is that HA’s head continuously mutates, which allows influenza strains to quickly adapt and commits any new vaccine to rapid obsolescence. As such, there has been a major thrust in recent years to develop immunogens that protect against both existing and future strains.

One way to address this conundrum is by targeting influenza’s Achille’s heel: it’s highly conserved protein regions. By stimulating immune responses against more conserved epitopes, it may be possible to induce cross-protective antibody responses against a range of subtypes to provide lasting ‘heterotypic’ protection. In essence, this is the concept for a so-called ‘universal vaccine’. The development strategies for universal vaccines are highly diverse, but can largely be divided by their target antigens:

The stalk

One of the most popular strategies for a universal vaccine is based on targeting HA’s stalk domain: a thin helical region that interacts with the viral envelope and shows high sequence conservation due to its fundamental role in membrane fusion (see previous figure) [13].

The catch is that stalk regions are somewhat shy, as they are overshadowed by the immunodominant head, which gains much of the immune response’s attention [14]. Fortunately, researchers have devised various means of getting around this issue, from which two major strategies have emerged.

The first, relies on total removal of the head region to leave a truncated stalk protein. The advantage of such a strategy is that the totally exposed stalk is relatively immunogenic. However, anti-stalk antibodies are very particular and must bind to highly specific conformational structures to exhibit neutralizing ability [13]. And given that removal of the head can misshape the remaining protein, the major obstacle in the development of these vaccines has been the consistent production of correctly folded stalk.

An alternative strategy employs chimeric haemagglutinins (cHAs), in which HA head domains from distant influenza subtypes fused to conserved stalk regions [15]. The thinking is that, by administrating multiple vaccine doses formulated with the same stalk but different heads, the immune response to conserved stalk epitopes can be considerably strengthened. This has been corroborated in mouse models where initial vaccination appears to ‘prime’ the humoral response, which is then boosted by successive vaccinations [14]. What’s more, recent studies have shown that in addition to inducing cell-mediated cytotoxicity and complement cascades, stalk-reactive antibodies are able to lock HA trimers in a conformation that prevents viral entry, essentially trapping them until they are cleared by immune cells [16].

Chimeric haemagglutinin proteins that share the same stalk domain, but include head domains from various human and zoonotic subtypes (RCSB: 1ti8).

A variation on this theme are the so-called mosaic haemagglutinins (mHA), in which only immunodominant epitopes of the HA head are replaced, as opposed to the entire domain. This strategy has the added benefit of eliciting head-specific antibody responses that aren’t immunodominant [14].

Matrix 2e

Matrix 2 (M2) is a highly conserved proton channel that functions to maintain influenza’s internal pH as it passes through the host’s cell and organelle membranes. M2 is sparsely found on the envelope of influenza virions, but can be found at high copy numbers in the membranes of infected cells [17].

Matrix two proton channels in closed and open conformations, respectively (RCSB: 6pvr, 6pvt).

As most of M2’s residues are buried within the lipid bilayer, choice of epitopes are limited. Of these, the 22-amino acid ectodomain, M2e, has been evaluated as a promising vaccine target due of its sequence conservation and relative exposure to the virion’s environment. However, M2e is a poor natural immunogen as it is still shielded by the bulkier HA and NA surface proteins, which results in negligible M2e antibody titers during natural infection [18]. Therefore, vaccine constructs have been used to improve epitope binding and have successfully elicited potent humoral and cell-mediated responses.

A various constructs other for M2e have been explored, including virus-like particles (VLPs) [19], bacteriophages [20], tetrameric M2e proteins [21] and flagellin-fused M2e [17], all of which have shown success against a broad range of influenza viruses.

Due to lack of exposure to the environment, antibodies against M2e aren’t neutralizing. Instead, they contribute to T-cell responses, which are directed against infected cells to limit viral replication. Therefore, while M2e vaccines would not be able to prevent initial infection, they are able to protect from severe forms of infection, as evidenced by studies of passive immunity [22].

Nucleoprotein and Matrix 1

While the vaccines discussed so far have focused on surface antigens, the effects of antigenic drift and shift are only skin deep. There are various other highly conserved antigens found within influenza’s envelope that have been evaluated as universal vaccine targets. Nucleoprotein and Matrix 1 are found within influenza virions and infected host cells but are not readily accessible by antibodies, and therefore don’t make good targets for eliciting a humoral response. However, like M2, Nucleoprotein and M1 can confer protective immunity through potent T-cell responses to reduce the severity of influenza infection [23]. Oxford-based Vaccitech (UK) have developed a Modified Vaccinia Ankara virus (MVA) and Adenovirus platforms for the expression of NP and M1, which have so far shown promising results in clinical trials [24,25].

There are various other vaccine strategies beyond the scope of this article. These include synthetic fusion peptides (one of which is the first universal vaccine to make it to phase III trials [26]), so-called COBRA Haemagglutinins [27] and live single cycle replication virus vaccines [28].

Challenges ahead

Overall, these novel approaches are moving closer us to the promise of a broadly reactive influenza vaccine. However, there are still multiple hurdles to achieving a truly universal vaccine that deserve some consideration.

As of yet, the breadth of cross-protection conferred by existing vaccine candidates is limited to small groups of subtypes that fail to stimulate high truly heterotypic protection. Therefore, a sizable gap still remains between current candidates and ultimate goal of a universal influenza vaccine. Likewise, there are also discrepancies between coverage of protection. For example, stalk vaccines have so far struggled to provide complete protection against the flu [29]. Alternatively, M1, M2 and NP fail to elicit robust humoral responses and can only modulate disease severity. As a result, it is very likely vaccines will need to be formulated as a combination of antigens or given as successive boosters. With this in mind, Guo et al. recently developed a chimeric M2e-HA fusion protein that has successfully shown to prevent lethal challenge in mice and induce high antibody titers [30].

Another important consideration for vaccine development is the duration of protection, as a universal vaccine that is only able to protect patients from infection for a few months is of little use. Immune responses to certain influenza proteins fail to provide lasting immunity, requiring vaccine developers to evaluate various adjuvants to improve the potency and duration of the immune response.

A safety concern of universal influenza vaccines is original antigenic sin, in which our immune systems can preferentially stimulate humoral responses based on previous pathogen or vaccine exposures when a marginally different version of a virus is encountered [31]. This has the unpleasant effect of making influenza infections more severe as mis-matched antibodies cannot properly neutralize the invader, effectively paralyzing the humoral response, or even enhancing infection. Therefore, vaccine candidates will require extensive testing in a diverse range of human populations that differ in their leukocyte haplotypes and immunoglobulin germlines [32].

A universal vaccine is the holy grail of influenza. But like the holy grail, it has proven difficult to find. Yet, in spite of the challenges ahead, these novel vaccines have so far shown vast potential when compared with existing technologies. A single vaccine formulation could provide protection against a broad range of seasonal and pandemic influenza A and B subtypes, boosting immunity to levels that can minimize symptoms and drastically reduce the rates of severe infection and mortality. Universal vaccines would also abolish the need for annual reformulation and re-administration and will maybe make the prospect of getting your yearly flu jab seem more worthwhile.

The Native Antigen Company

The Native Antigen Company offers an extensive panel of influenza reagents and has recently launched a custom development service to rapidly create HA and NA proteins from specific influenza strains to support manufacturers in the development of vaccines and diagnostics, facilitating studies of how vaccines and diagnostics perform on with newly identified strains.

  1. Centers for Disease Control and Prevention. Flu Season cdc.gov/flu/season/index.html
  2. Schultz-Cherry S. Roles of humidity and temperature in shaping influenza seasonality. Virol. 88 (14) 7692–7695 (2014).
  3. Henslet SE, Das SR, Bailey AL et al. Hemagglutinin receptor binding avidity drives influenza A virus antigenic drift. Science 326(5953), 734–736 (2009).
  4. Centers for Disease Control and Prevention. Types of influenza viruses. cdc.gov/flu/about/viruses/types.htm
  5. Zhang N, Zheng B, Lu L et al. Advancements in the development of subunit influenza vaccines. Infect. 17(2), 123–124 (2015).
  6. World Health Organization. Fact Sheets: Influenza (seasonal). www.who.int/news-room/fact-sheets/detail/influenza-(seasonal)
  7. Wong S, Webby RJ. Traditional and new influenza vaccines. Microbiol. Rev. 26(3), 476–492 (2013).
  8. Mao L, Yang Y, Qiu Y, Yang Y. Annual economic impacts of seasonal influenza on US counties: Spatial heterogeneity and patterns. J. Health Geog. 11, 16 (2012).
  9. Jong JC. Rimmelzwaan GF, Fouchier RAM, Osterhaus ADME. Influenza virus: a master of metamorphosis J. Infect. 40(3), 218–228 (2000)
  10. Grand Global Challenges. Ending the pandemic threat: A grand challenge for universal influenza vaccine development. https://gcgh.grandchallenges.org/challenge/ending-pandemic-threat-grand-challenge-universal-influenza-vaccine-development
  11. Dawood FS, Iuliano AD, Reed C et al. Estimated global mortality associated with the first 12 months of 2009 pandemic influenza A H1N1 virus circulation: a modelling stud Lancet Infect. Dis. 12(9), 687–695 (2012).
  12. Fineberg HV. Pandemic preparedness and response — Lessons from the H1N1 influenza of 2009. Engl. J. Med. 370, 1335–1342 (2014).
  13. Valkenburg SA, Cowling BJ. Turning influenza vaccinology on its head to reveal the stalk. Lancet Infect. Dis. 20(1), 5–7 (2020).
  14. Broecker F, Liu STH, Suntronwong N et al. A mosaic hemagglutinin-based influenza virus vaccine candidate protects mice from challenge with divergent H3N2 strains. Npj Vaccines. 4, 31 (2019).
  15. Krammer F, Palese P. Advances in the development of influenza virus vaccines. Rev. Drug Disc. 1, 167–182 (2015).
  16. Chen Y, Kan LY, Huang M, Henry C, Wilson PC. Hemagglutinin stalk-reactive antibodies interfere with influenza virus neuraminidase activity by steric hindrance. Virol. 93(4), e01526–18 (2019).
  17. Pielak RM, Chou JJ. Influenza M2 proton channels. Bophys. Biomemb. 1808(2), 522–529 (2011).
  18. Schotsaert M, Ysenbaert T, Neyt K et al. Natural and long-lasting cellular immune responses against influenza in the M2e-immune host. Mucosal Imm. 6, 276–287 (2013).
  19. Kim M, Song J, Eunju O et al. Virus-like particles containing multiple M2 extracellular domains confer improved cross-protection against various subtypes of influenza virus. 21(2), 485–492 (2013).
  20. Deng L, Ibañez LI, Van den Bossche V et al. Protection against influenza A virus challenge with M2e-displaying filamentous Escherichia coli phages. PLoS ONE. doi:1371/journal.pone.0126650 (2015).
  21. Eliasson DG, Omokanye A, Schön K et al. M2e-tetramer-specific memory CD4 T cells are broadly protective against influenza infection. Mucosal Imm. 11, 273–289 (2018).
  22. Ramos EL. Mitcham JL, Koller TD et al. Efficacy and safety of treatment with an anti-M2e monoclonal antibody in experimental human influenza. Infect. Dis. 211(7), 1038–1044 (2015).
  23. Aleen JD, Ray S, Ross TM. Split inactivated COBRA vaccine elicits protective antibodies against H1N1 and H3N2 influenza viruses. PLoS ONE. doi:1371/journal.pone.0204284 (2018).
  24. Berthoud TK, Hamill M, Lillie PJ et al. Potent CD8+ T-cell immunogenicity in humans of a novel heterosubtypic influenza A vaccine. MVA−NP+M1. Infect. Dis. 52(1), 1–7 (2011).
  25. Arunkumar GA, McMahon M, Pavot V et al. Vaccination with viral vectors expressing NP, M1 and chimeric hemagglutinin induces broad protection against influenza virus challenge in mice. 37(37), 5567–5577 (2019).
  26. Van Doorn E, Liu H, Ben-Yedidia T et al. Evaluating the immunogenicity and safety of a BiondVax-developed universal influenza vaccine (Multimeric-001) either as a standalone vaccine or as a primer to H5N1 influenza vaccine. Medicine 96(11), 6339 (2017).
  27. Aleen JD, Ray S, Ross TM. Split inactivated COBRA vaccine elicits protective antibodies against H1N1 and H3N2 influenza viruses. PLoS ONE. doi:1371/journal.pone.0204284 (2018).
  28. Sarawar S, Hatta Y, Watanabe S et al. M2SR, a novel live single replication influenza virus vaccine, provides effective heterosubtypic protection in mice. Vaccine 34(42) 5090–5098 (2016).
  29. Eisenstein M. Towards a universal flu vaccine. nature.com/articles/d41586-019-02751-w
  30. Guo Y, He L, Song N et al. Highly conserved M2e and hemagglutinin epitope-based recombinant proteins induce protection against influenza virus infection. Infect. 19(12), 641–647 (2017).
  31. Khurana S, Loving CL, Manischewitz J et al. Vaccine-induced anti-HA2 antibodies promote virus fusion and enhance influenza virus respiratory disease. Science 5(200), 200ra114 (2013).
  32. Nachmagauer R, Krammer F. Universal influenza virus vaccines and therapeutic antibodies. Clin Microbiol. Infect. 23(4), 222–228 (2017).

You might also like:


Leave A Comment