Broadly neutralizing antibodies target

Broadly neutralizing antibodies that target hemagglutinin epitopes on influenza virus have the potential to provide near universal protection against influenza virus infection1. However, viral mutants that largely evade neutralizing antibodies have been reported2,3. The identification of broadly neutralizing antibody classes that can neutralize viral escape mutants is essential for the design of a universal influenza virus vaccine. Here, we report a distinct class of broadly neutralizing antibodies that target a discrete membrane anchor epitope proximal to the hemagglutinin stalk domain.

Antibodies targeting anchor epitopes largely neutralize H1 viruses and can cross-react with H2 and H5 viruses which pose a pandemic threat. Antibodies that target this anchor epitope use a very restricted repertoire, which encodes two public binding motifs that make extensive contacts with conserved residues in the fusion peptide.

Additionally, B cells targeting the anchor epitope are common in the human memory B cell repertoire and have been recalled in humans by a chimeric hemagglutinin vaccine with oil-in-water adjuvant4,5, which is a potential universal vaccine against the influenza virus. To maximize protection against seasonal and pandemic influenza viruses, vaccines should aim to stimulate this previously untapped source of largely neutralizing antibodies that are prevalent in the pool of human memory B cells.

Antibodies against the major surface glycoprotein, hemagglutinin (HA), are essential to provide protection against influenza virus infection6,7. However, most HA-binding antibodies target variable epitopes of the HA leader domain, which provide limited protection against antigenically similar influenza virus strains3. Vaccine formulations that preferentially induce antibodies against conserved epitopes of the HA head and stalk domains could provide broad and potent protection against a wide range of influenza viruses.

Several leading candidate universal influenza viruses are engineered to induce antibodies specifically directed against the stem domain, but the spectrum of distinct epitopes on the stem targeted by the human B cell repertoire remains to be determined. By analyzing the specificities of B cells targeting the H1 stalk through the generation of monoclonal antibodies (mAbs), our study reveals a novel class of broadly neutralizing antibodies (bnAbs) against an underappreciated epitope where HA anchors in the membrane viral. Next-generation vaccine platforms should take advantage of this discovery to elicit antibodies against the conserved anchor epitope.

To investigate the specificities of HA-specific antibodies, we generated 358 mAbs from plasmablasts and HA+ memory B cells (MBCs) isolated from volunteers vaccinated against or naturally infected with seasonal influenza viruses or participating in a clinical trial of phase I of a chimeric HA vaccine (cHA)4,5. Of all mAbs tested, nearly half targeted the HA stem domain, of which 21% targeted the well-characterized central stem (CS) epitope (Extended Data Fig. 1a).

Notably, stalk-binding mAbs have been disproportionately isolated from infected pandemic 2009 monovalent H1N1 (pH1N1) inactivated influenza vaccine and cHA vaccine cohorts (extended data Fig. 1b), as these have previously been shown to exposure routes induced antibody responses to the HA5,8,9 stem. To investigate the epitopes targeted by the remaining 79% of stem-binding mAbs, we performed negative-staining electron microscopy with three non-CR9114 competitor stem-binding mAbs (ref. 10): 047-09 4F04, 241 IgA 2F04 and 222. -1C06. All three mAbs bound to an epitope near the HA stem anchor and oriented at an upward angle towards the epitope (Fig. 1a,b, Extended Data Fig. 1c)

suggesting that this epitope might be partially occluded by the lipid membrane and can only be exposed for antibody binding when the HA trimers flex on the membrane. FISW84, a recently identified stem-binding mAb11, targets an epitope that overlaps with the three identified anchor-binding mAbs (extended data Fig. 1d), suggesting that the anchor epitope is a stem epitope common. In addition, a proximal epitope was previously identified on group 2 viruses that is targeted by the mAbs CR8020 (ref. 12) and CR8043 (ref. 13). Despite some overlap, the epitope ci