Refers to article:
Key message from this paper is that the swine flu vaccine they used could induce anti-influenza (flu) haemagglutinin (HA) antibodies that are not only unhelpful but actively harmful.
In this study, pigs were vaccinated with a vaccine called WIV-H1N2, which is a whole UV-irradiation inactivated, adjuvanted H1N2 (human-like virus). They were then challenged with a different flu virus, the pandemic H1N1 flu strain, H1N1pdm09. The vaccine and challenge viruses shared HA stem/stalk epitopes but not HA head epitopes or NA epitopes
Authors showed the vaccine
- Induced anti-WIV-H1N2 HA stem/stalk antibodies.
- These antibodies cross-reacted with H1N1pdm09 HA stem/stalk.
- Such anti-HA stem/stalk antibodies enhanced viral fusion and increased flu virus entry into cells in vitro.
- OTOH, vaccine didn’t appear to have induced antibodies capable of neutralizing the challenge virus.
- The vaccine-induced anti-HA antibodies could bind strongly to HA derived from the challenge virus.
- The vaccine-induced anti-HA antibodies competed with several mAbs (monoclonal antibodies) that were stem-specific and neutralizing, implying that in vivo, such antibodies could potentially outcompete them thereby enhancing rather than eliminating the flu infection.
- Virus fusion promoting activity correlated with enhanced lung pathology in vaccinated pigs challenged with a flu virus different from the one in the vaccine (heterologous).
By binding where they did in the manner they did, authors speculate these antibodies facilitated a conformational change in HA that enhanced its fusion to host cells (see figure below from 1). As a result, rather than preventing/blocking/inhibiting flu infection, such virus-bound antibodies ended up enhancing disease, specifically more frequent and severe pneumonia in vaccinated pigs.
This phenomenon of enhanced disease post-vaccination was already described in pigs. Called VAERD (vaccine-associated enhanced respiratory disease), pigs vaccinated with other flu antigen vaccines or with mismatched inactivated viruses developed worse disease when challenged with H1 viruses (2, 3).
Relevance of anti-HA stem/stalk antibodies
HA is the most abundantly expressed glycoprotein on the surface of the flu virus envelope. It has two domains, a variable globular head and a relatively conserved stalk (see figure below from 4).
The globular HA head binds to sialic acid residues on cell surface receptors. Such binding drives a conformational change in the HA protein that facilitates its fusion to the host cell membrane (see figure below from 5).
Studies show the HA head domain is immunodominant.
- A frequently mis-understood word, in the case of T cells, an immunodominant antigen means a peptide that dominates by its ability to withstand cellular proteolytic degradation processes, and prevails over other antigenic candidates in antigen processing and presentation processes inside antigen-presenting cells (APCs) by effectively out-competing them for ability to bind to MHC molecules. Upshot is an immunodominant peptide would be more likely to get presented to a T cell by an APC and would thus become the focus of T cell responses compared to sub-dominant and non-dominant peptides.
- In the case of B cells, an immunodominant antigen is one that dominates over others by its ability to bind a specific B cell receptor (BCR).
In terms of attracting immune response, the head domain of the HA protein is immunodominant such that much of the anti-flu immune response becomes directed against it. In particular, anti-HA head domain antibodies tend to neutralize only a particular flu strain. OTOH, anti-HA stalk antibodies are broadly neutralizing and can bind to a wide range of flu strains (6). Anti-HA head domain antibodies tend to bind close to the globular head’s sialic acid-binding site. This prevents flu virus from attaching to the cell while anti-HA stem/stalk antibodies can restrict the conformational changes necessary for flu virus to fuse with host cell membrane.
Since HA stem/stalk’s relatively conserved across flu subtypes, vaccines that can drive robust anti-HA stem/stalk antibodies hold greater promise for a universal flu vaccine. The figures below showing the range of HA subtypes, their distribution across different species (from 7), the HA subtypes associated with 20th and 21st century flu pandemics (from 7), and the HA phylogenetic tree (from 8) help explain the attraction an universal flu vaccine holds.
The original 2011 observation of HA-stalk specific antibodies being broadly neutralizing (9, 10) thus spurred a stampede of studies to create vaccines that could drive such immune responses.
Implications of this data are several-fold
- Inactivated versus Live virus drive entirely different immune responses: Inactivated flu vaccines show limited cross-protective immunity in pig farms against flu virus strains other than the one used in the vaccine, i.e., against heterologous flu vaccine strains (3). OTOH, live H1N1 and H2N2 viruses show better cross-protective immunity (11).
- Role of Vaccine Adjuvant in type of immune response the vaccine elicited: The vaccine used in this study, called WIV-H1N2, was adjuvanted with what the authors call a commercial oil-in-water emulsion. Though the authors don’t discuss this adjuvant at all, it’s quite likely that this adjuvant played a major role in the class of innate, T and B cell responses induced by this vaccine.
- Thus, it’s possible that anti-HA stalk/stem antibody enhancement of flu disease severity may be adjuvant-dependent or species-specific or both.
- The species-specific aspect of the conundrum emerges from flu vaccine studies in ferrets. Ferrets, which like pigs develop flu infections with features similar to those in humans, did not develop disease-enhancing antibodies when vaccinated with HA-stalk based H1N2 vaccines (12).
- Conundrum is the ferret vaccines were non-adjuvanted, ferrets were vaccinated intransally while the pigs in this study were vaccinated intramuscularly. So too many critical differences render a side-by-side comparison practically impossible.
- Nevertheless, role of adjuvant and importance of species remain crucial issues that could explain the difference in outcome of flu H1N2 vaccine followed by heterologous flu virus challenge between pigs and ferrets.
- Understanding how anti-HA stem/stalk antibodies neutralize versus enhance flu infection is a more important priority than new vaccine design. Though directed against HA stem/stalk, antibodies in this study weren’t neutralizing because they bind to a linear portion of the HA stem/stalk whereas neutralizing anti-HA stem/stalk antibodies tend to be conformational, i.e., directed to a three-dimensional epitope and as such can’t be detected by measuring binding to linear peptides derived from HA.
- Need to understand relative contribution of all types of anti-flu antibodies in flu virus clearance. After all, promise of universal flu vaccine notwithstanding, focusing only on anti-stem/stalk antibodies at the expense of antibodies with other specificities may well be throwing the baby out with the bath water.
- Relative neglect of T cell responses in flu vaccine design is indefensible. After all, B cells generate anti-flu IgG antibodies based on the kind of T cell help they receive. Yet the most basic questions about this process remain unanswered. For e.g., what’s the nature of T cell help that drives anti-HA stem/stalk antibodies versus anti-HA head versus anti-NA and so on.
- B cell population differences between anti-HA stem/stalk antibody secreting versus anti-HA head secreting versus anti-NA secreting B cells are unexplored.
- In particular, propensity for memory B cell formation capacity between B cells specific for different flu virus antigens is not well understood.
In short, tremendous technical prowess is being expended to understand the chemistry of antibody binding to flu HA without enough attention paid to the underlying biology, particularly the processes necessary to activate and maintain T and B cells in a manner that favors a flu-neutralizing antibody response.
1. Crowe, James E. “Universal flu vaccines: primum non nocere.” Science translational medicine 5.200 (2013): 200fs34-200fs34.
2. Heinen, Paul P., et al. “Vaccination of pigs with a DNA construct expressing an influenza virus M2–nucleoprotein fusion protein exacerbates disease after challenge with influenza A virus.” Journal of General Virology 83.8 (2002): 1851-1859.
3. Vincent, Amy L., et al. “Failure of protection and enhanced pneumonia with a US H1N2 swine influenza virus in pigs vaccinated with an inactivated classical swine H1N1 vaccine.” Veterinary microbiology 126.4 (2008): 310-323.
4. Lofano, Giuseppe, et al. “B cells and functional antibody responses to combat influenza.” Frontiers in immunology 6 (2015).
5. Krammer, Florian, and Peter Palese. “Advances in the development of influenza virus vaccines.” Nature Reviews Drug Discovery 14.3 (2015): 167-182.
6. Corti, Davide, et al. “A neutralizing antibody selected from plasma cells that binds to group 1 and group 2 influenza A hemagglutinins.” Science 333.6044 (2011): 850-856.
7. Webster, Robert G., and Elena A. Govorkova. “Continuing challenges in influenza.” Annals of the New York Academy of Sciences 1323.1 (2014): 115-139.
8. Wheatley, Adam Kenneth, and Stephen John Kent. “Prospects for antibody-based universal influenza vaccines in the context of widespread pre-existing immunity.” Expert review of vaccines 14.9 (2015): 1227-1239.
9. Ekiert, Damian C., et al. “A highly conserved neutralizing epitope on group 2 influenza A viruses.” Science 333.6044 (2011): 843-850.
10. Corti, Davide, et al. “A neutralizing antibody selected from plasma cells that binds to group 1 and group 2 influenza A hemagglutinins.” Science 333.6044 (2011): 850-856.
11. Vincent, Amy L., et al. “Swine influenza viruses: a North American perspective.” Advances in virus research 72 (2008): 127-154.
12. Krammer, Florian, et al. “Assessment of influenza virus hemagglutinin stalk-based immunity in ferrets.” Journal of virology 88.6 (2014): 3432-3442.