At the turn of the twentieth century, the immunologist Paul Ehrlich put forward the idea of 'magic bullets', which could be used in the fight against human diseases. Seventy years later, this concept received support, thanks to technology that enabled the production of antibodies designed to target almost any antigen — molecular structures produced by cells and viruses. In 2016, five of the eight top-grossing prescription drugs were monoclonal antibodies (those that bind to only one site on an antigen), which target viruses, bacteria, toxins and proteins (go.nature.com/2yidu1o). Writing in Science Translational Medicine, Shen et al.1 describe a monoclonal antibody that, in animals, potently cross-protects against all influenza B viruses tested. Such an antibody could protect humans against influenza B and could inform the design of broadly protective flu vaccines.

Seasonal outbreaks of influenza A and B strains are estimated to be responsible for the death of up to 50,000 people in the United States each year (see go.nature.com/2ao6jua). This disease burden is likely to remain steady in the near future, because current commercial flu vaccines are suboptimal. The occasional emergence of pandemic strains of the influenza A virus adds an extra element of unpredictability. Typically, seasonal and pandemic influenza A viruses pose the greatest threat to the population — but influenza B can also be associated with high mortality, especially in infants and children.

Flu viruses dock to host cells through a glycoprotein called haemagglutinin, which binds to host-cell receptors. Antibodies that bind to haemagglutinin to prevent this docking can therefore protect animals and humans infected with flu viruses. Theoretically, treatment with such antibodies could both prevent and treat flu by providing people with immediate, short-term protection (passive immune protection).

Currently, circulating influenza B viruses belong to either the Victoria or the Yamagata lineage — the two split in the 1980s from an earlier precursor (see go.nature.com/2hivvpf). Antibodies that would protect against variants of both lineages are highly desirable. Three groups2,3,4 have previously reported human monoclonal antibodies that bind to specific sites on B virus haemagglutinins to provide protection against multiple strains of B virus in mice. Shen et al. have now built on these findings.

The authors immunized mice with live influenza B viruses from both lineages to trigger antibody production by the animals' immune systems. They isolated hundreds of monoclonal antibodies produced from eight immunization regimens and analysed the ability of each antibody to block the attachment of influenza B viruses to cells (determined by measuring the level of haemagglutination — the clumping of red blood cells and viral particles bound together through haemagglutinin). The researchers selected the best antibody, dubbed 12G6, and combined the antigen-binding region of this antibody with the portion of a human IgG1 antibody, called the Fc region, that can bind to receptors on host immune cells to activate the human innate immune system. The group used the resulting chimaeric antibody, which they named C12G6, for all further studies.

C12G6 protected mice and ferrets against influenza B viruses when administered preventively.

Shen et al. demonstrated that C12G6 is a broadly protective antibody. C12G6 exhibited haemagglutination-inhibition activity against all 18 influenza B viruses that the researchers tested and showed higher potency than the previous cross-protecting antibodies2,3,4. Importantly, C12G6 protected mice and ferrets against influenza B viruses when administered preventively, as well as providing excellent therapeutic effects when administered after viral infection.

Typically, an antibody that acts to inhibit the attachment of specific flu viruses to the host cell binds to a highly variable region on haemagglutinin's 'head' region. C12G6 is different, in that it binds to an evolutionarily conserved docking point on the head near the receptor-binding site. The group showed that binding of C12G6 to this site interferes with the attachment of all B virus strains (not only a specific variant) to host cells, providing an explanation for the antibody's broad activity (Fig. 1a).

Figure 1: Antibody protection against influenza B viruses.
figure 1

Shen et al.1 have created an antibody, dubbed C12G6, that can protect mice and ferrets against multiple strains of influenza B virus. a, C12G6 binds to an evolutionarily conserved region of the head of the viral glycoprotein haemagglutinin, thereby blocking haemagglutinin binding to receptors on host cells and so preventing virus replication. b, Unusually, C12G6 also seems to exert antiviral effects through its Fc region. This region binds to Fc receptors on immune cells, activating the cells to destroy the invading virus, and interacts with extracellular immune proteins called complement factors, again leading to virus breakdown.

Full size image

Next, Shen et al. showed that C12G6 can inhibit influenza B virus replication by not only neutralizing the virus, but also activating the antiviral activities of the host's immune system through Fc-mediated mechanisms (Fig. 1b). In particular, C12G6 shows complement-dependent cytotoxicity, in which binding of the antibody's Fc region by immune-system proteins called complement factors triggers an extracellular signalling cascade that leads to the degradation of the virus. C12G6 also stimulates antibody-dependent cell-mediated cytotoxicity (ADCC), in which Fc receptors on immune cells are activated by antibody binding, triggering a direct immune response from the cell that again leads to virus destruction and elimination.

The ability of C12G6 to both block haemagglutination of host cells by a virus and elicit ADCC is highly unusual. It has previously been reported5,6 that haemagglutination-inhibiting antibodies block optimal activation of immune cells and so do not induce ADCC — more work is needed to explain this discrepancy.

What is next to be discovered? First, it has not yet been proved that C12G6 is protective in humans. Second, even if it is effective, passive immunization against flu (especially influenza B) may be of limited value. Because of the short duration of protection and high cost of antibody therapeutics, this approach might not be as cost effective as seasonal vaccines. However, it could be useful for certain groups of people who are at risk of exposure, including health-care workers, emergency personnel and the military, because of the ability of broadly protective antibodies to be effective against multiple influenza variants.

Finally, it will be valuable to find out whether the head of a B virus haemagglutinin can be developed for use as a vaccine. To be of use, the haemagglutinin head region used would need to induce the human immune system to produce high titres of antibodies that have properties similar to those of C12G6, and provide long-lasting and broad immune protection. If successful, such a vaccine could be transformative.

Footnote 1