A team from the Crucell Vaccine Institute and The Scripps Research Institute has identified a trio of neutralizing human mAbs that protect mice from challenge with influenza viruses, including one mAb that protects against both A and B strains.1 Crucell N.V., first plans to develop the antibodies to treat influenza-infected individuals and in the long run could use the antibody against A and B strains as a guide for developing a universal influenza vaccine.

Despite the success of prophylactic vaccines, seasonal influenza epidemics-either by the A or B strain-still cause morbidity and mortality every year. Influenza A viruses cause seasonal flu as well as influenza pandemics and are associated with more severe clinical disease. Influenza B viruses are the cause of seasonal epidemics every two to four years.2

The seasonal trivalent inactivated influenza vaccine attempts to induce protection by including two influenza A subtypes and one influenza B lineage. However, seasonal influenza vaccine mismatches occurred in about 50% of the seasons from 1998-2010.3

In an effort to take guesswork out of the equation, the Crucell-Scripps team decided to focus on finding a broadly neutralizing antibody against both influenza A and B viruses. A mAb-based immunotherapy that delivers antibodies right to the immune system could be used as a treatment for people already severely infected with influenza, as a prophylactic in cases of pandemics or in people who do not respond well to vaccination.

The researchers used a soluble recombinant version of the influenza A virus hemagglutinin (HA) proteins from various subgroups (HA9, HA7, HA5, HA3 and HA1) and influenza B HA from two lineages (Yamagata and Victoria) to screen a combinatorial display library of human B cells from volunteers vaccinated with the seasonal influenza vaccine.

The HAs were selected to cover the two lineages and main phylogenetic branches of influenza B, as well as the major subtypes of influenza A. Hits were further screened for their ability to bind viruses of both influenza B lineages and influenza A subtypes.

Crucell identified mAbs CR8033, CR9114 and CR8071 that bound to the two influenza B lineages. CR9114 also bound influenza A virus HA1 and HA3 subtypes.

Because antibody effector functions-the antibody's ability to recruit immune cells to fight infection-have been suggested to contribute to the protective efficacy of broadly neutralizing antibodies, the researchers next evaluated the three antibodies in mice challenged with influenza.

In mice given a lethal dose of Yamagata or Victoria influenza B, pretreatment with CR8033, CR8071 or CR9114 produced dose-dependent decreases in weight loss and increased survival compared with vehicle pretreatment. CR9114 also protected mice from lethal doses of H1N1 or H3N2 influenza A, making CR9114 the first mAb to provide protection against both influenza A and B.

Notably, the three antibodies did not compete with each other for HA binding, suggesting they targeted different epitopes. Thus, the researchers turned to the crystal structures of the antibodies bound to virus to identify the actual epitopes and to understand how the antibodies achieved broad neutralization.

CR8033 bound an epitope in the HA head within the receptor-binding pocket of sialic acid and surrounding antigenic site. CR8071 bound to an epitope at the base of the HA head, distant from the receptor-binding site. CR9114 bound to an epitope in the HA stem, which explained its broadly protective nature.

mAbs directed to the globular head region of HA, which is highly variable and mediates attachment of influenza virus to the cellular receptors of the host cell (see "Antibodies can target the head or stalk of the hemagglutinin protein to provide different mechanisms of action," 1.I), are typically strain specific and do not confer broad protective immunity. In contrast, mAbs directed to the conserved HA stem prevent membrane fusion and can broadly neutralize different influenza subtypes (see Figure 1.II).

For example, Celltrion Inc.'s CT120 and Humabs BioMed S.A.'s FI6 both are antibodies that target the HA stem. The antibodies are in preclinical development for influenza A. FI6 is licensed to an undisclosed pharma.

The crystal structure also showed that CR9114's epitope was similar to the epitope of FI6. However, the structure of FI6 itself is dissimilar to that of CR9114. The researchers suggested that this explains why FI6 can neutralize influenza A, whereas CR9114 can neutralize both A and B.

To flesh out the antibodies' mechanisms of action, the researchers studied influenza-infected cell lines pretreated with the antibodies. The stem-binding CR9114-but not the head-binding CR8033 and CR8071-blocked the HA pH-induced conformational changes associated with membrane fusion (see Figure 1.II)

Curiously, CR8033 and CR8071 did not block viral entry and genome replication, which is how HA head-binding antibodies typically prevent virus propagation (see Figure 1.I). Instead, the antibodies interfered with the release of progeny virus from inside the cell into the supernatant, blocking the subsequent infection of neighboring cells. Surface electron microscopy images of the cell culture system showed dense aggregation of virus on the surface of the infected cells.

The images closely resembled what occurs in cells treated with the neuraminidase (NEU1; SIAL1) inhibitor Relenza zanamivir. Biota Holdings Ltd. and GlaxoSmithKline plc market Relenza to treat and prevent influenza.

Results were published in Science.

"The fact that the Crucell-Scripps team found these three cross-neutralizing antibodies in B cells from influenza-vaccinated individuals provides an important proof of concept that humans can generate broadly protective responses," said Gary Nabel, director of the Vaccine Research Center at the NIH.

Entering the pipeline

Robert Friesen, lead principal investigator on the manuscript and VP of antibody discovery at Crucell, told SciBX, "We plan to transition our findings as therapeutic options to J&J's pipeline and to make this a smooth transition-these mAbs can be produced to sufficient quantity to take them into toxicology and safety studies."

"For these antibodies to be therapeutically actionable, the team is going to have to deal with manufacturing questions-how much antibody is needed to provide protection? Can they make enough of the antibody? Is the antibody stable enough?" said Peter Palese, chair of microbiology at the Mount Sinai School of Medicine.

Antonio Lanzavecchia, director of the Institute for Research in Biomedicine and cofounder of Humabs, also questioned the therapeutic potential of the antibodies. "Their study indicates that influenza A virus can escape from the broadest neutralizing antibody, CR9114, a fact that raises concern on the potential therapeutic use of this antibody," he said.

Both Lanzavecchia and Palese also wanted to see additional studies on the mechanism of protection for CR8033 and CR8071.

"We are very interested in further determining how CR8033 and CR8071 neutralize influenza B virus," agreed Friesen. "Because the structural work showed that they bound to or near the HA head, we predicted they would prevent binding of the virus to the cell or conformational change of HA. Unexpectedly, our results suggest that they instead contribute to protection by preventing egress of the virus from an infected cell."

The universe and everything

Although the Crucell-Scripps team identified the HA-stem epitope to which CR9114 binds, the key will be finding the specific antigen.

"Simply knowing what epitope the antibody binds to does not guarantee that a universal vaccine based on this knowledge can happen," said Palese.

"How likely it is that you can generate these broadly neutralizing antibodies through vaccination remains to be seen," Nabel pointed out. "It is worth trying, but the antibodies themselves could still provide protection even if vaccination does not succeed," he added.

Friesen agreed. "In terms of going after a universal vaccine, the concept is so far removed from our current work that it is actually a spin-off project within Crucell led by an entirely different Crucell research team. Our first priority is to get the broadly neutralizing influenza antibodies into the clinic."

Crucell has filed for a patent covering the broadly neutralizing antibody work. The IP is not available for licensing.

Baas, T. SciBX 5(33); doi:10.1038/scibx.2012.859
Published online Aug. 23, 2012

REFERENCES

1.   Dreyfus, C. et al. Science; published online Aug. 9, 2012; doi:10.1126/science.1222908
Contact: Jaap Goudsmit, Crucell Vaccine Institute, Leiden,
the Netherlands
e-mail: jaap.goudsmit@crucell.com

2.   Thompson, W.W. et al. JAMA 292, 1333-1340 (2004)

3.   Ambrose, C.S. & Levin, M.J. Hum. Vaccin. Immunother. 8, 81-88 (2012)

COMPANIES AND INSTITUTIONS MENTIONED

      Biota Holdings Ltd. (ASX:BTA), Notting Hill, Victoria, Australia

      Celltrion Inc. (KOSDAQ:068270), Incheon, South Korea

      Crucell N.V. (Pink:CRXLY), Leiden, the Netherlands

      Crucell Vaccine Institute, Leiden, the Netherlands

      GlaxoSmithKline plc (LSE:GSK; NYSE:GSK), London, U.K.

      Humabs BioMed S.A., Bellizona, Switzerland

      Institute for Research in Biomedicine, Bellizona, Switzerland

      Johnson & Johnson (NYSE:JNJ), New Brunswick, N.J.

      Mount Sinai School of Medicine, New York, N.Y.

      National Institutes of Health, Bethesda, Md.

      The Scripps Research Institute, La Jolla, Calif.