Hemagglutinin stem immunogens may hold the key for broad protection against influenza A viruses

By Neha MathurReviewed by Aimee MolineuxNov 15 2022

In a recent article published in Immunity, researchers proposed that vaccines targeting the immunosubdominant yet conserved hemagglutinin (HA) stem could trigger broadly neutralizing antibodies (bnAbs) against influenza A viruses.

Background

Influenza viruses are highly contagious respiratory pathogens, better prevented than combatted, just as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Thus, vaccines remain the most effective countermeasure against influenza viruses, reduce disease severity and slow down their transmission.

Current influenza vaccines trigger immune responses against the hypervariable HA head, thus, requiring frequent vaccination. Also, influenza vaccines are needed because it is impossible to predict an influenza pandemic wave with the available research tools. To this end, the ideal approach would be to develop and stockpile a broadly-reactive vaccine against group 1 and 2 influenza A viruses, and deploy it early during an influenza pandemic, negating the need for strain-matched vaccines typically devised and updated annually.

About the study

In the present study, researchers demonstrated how a cocktail of the group 1 and 2 headless HA stem-nano-particles could help elicit within-group and broadly-reactive neutralizing bnAbs and protective immune responses against a wide range of influenza A viruses in mice, ferrets, and non-human primates (NHPs).

The H10 stem occupies a unique position between group 2 HA stems, i.e., H3 and H7 phylogenetically, and possesses slightly higher sequence and solvent-exposed residue identities than H3 and H7. So the researchers hypothesized that the H10-based group 2 stem immunogen(s) might have a better antigenic surface to elicit broadly cross-reactive antibody responses against multiple influenza viruses.

In the pursuit, they structurally stabilized the H10 HA stem in native-like trimers and displayed them on self-assembling ferritin nanoparticles called H10ssF using a previously established method. Next, they purified self-assembled H10ssF from the culture supernatant of transfected mammalian cells, which gave a distinct peak of ~1.2 MDa on size exclusion chromatography (SEC).

Single-particle cryoelectron microscopy (cryo-EM) reconstruction of the H10ssF particle showed structural elements of the ferritin core and the spikes of HA-stem trimers at a 4.8 A ̊ resolution structure. Further, they measured the thermostability of H10ssF by differential scanning fluorometry (DSF). Furthermore, the team evaluated the antigenic integrity of H10ssF.

The researchers used biolayer interferometry (BLI) to measure the binding kinetics of the antigen-binding fragment of an antibody (Fab) of MEDI8852, CT149, and CR8020 to H10ssF. They measured calcium (Ca2+) flux by flow cytometry (FC) using recombinant Ramos B cells to evaluate the ability of H10ssF to activate B cells expressing BCRs of the unmutated common ancestors (UCA) of human bnAbs. The team immunized mice with H3ssF, H7ssF, or H10ssF using a sigma adjuvant system [SAS]). Likewise, they immunized groups of mice, ferrets, and cynomolgus macaques with a cocktail of H1ssF and H10ssF (termed herein H1+10ssF) with a squalene oil-in-water emulsion adjuvant (AddaVax).

Results

Virus-based microneutralization assays revealed heterosubtypic virus-neutralizing antibody responses in mice and ferrets, and their sera neutralized all five pseudotyped viruses representing group 1 and group 2 human influenza A viruses. However, NHP immune sera did not show detectable neutralization against the H3N2 virus.

Preclinical studies evaluating vaccine immunogenicity in small animal models do not recapitulate complex immune responses observed in humans. For instance, VH1-69 gene usage largely constrained bnAb responses to group 1 HA stem, which do not exist in these animal species. Even NHPs do not possess the human VH1-69 gene, the most prevalent gene utilized for group 1 HA stem-directed bnAbs. Instead, they use IGHV3S5 and IGHV3S25 VH genes to target the same stem epitope.

Even though NHPs generate human-like bnAb responses to either group 1 or group 2 HA stem following infection or vaccination, no NHP bnAbs broadly neutralized both group 1 and group 2 influenza viruses reported to date.

NHPs immunized with H1+10ssF induced a diverse repertoire of cross-reactive HA-specific B cells, including some that cross-reacted with both group 1 and group 2 HAs. So the team single-cell sorted H1+H10+ B cells from peripheral blood mononuclear cells (PBMCs) in week 12 from animal M08145 and sequenced their Ig genes. Most importantly, the researchers deciphered a mode of HA recognition via the DH gene-encoded motif, common in NHP and human bnAbs. The human DH3-3 gene encoded the canonical CDR H3 motif of the human VH6-1+DH3-3 class of bnAbs. In addition, 789-203-3C12 targeted the canonical HA stem supersite, with HA1 and HA2 subunits providing 15% and 85% of a total buried surface area (BSA) of 988 A ̊2, respectively. Then, the researchers identified the NHP bnAb 789-203-3C12, a cross-group bnAb, resembling this class of bnAbs, a public human bnAb clonotype, thus, offering a blueprint for broadly protective influenza vaccines for humans.

Conclusions

The IGHD3-41 gene appears conserved in macaques. However, it was surprising that this gene, out of many DH genes with a particular reading frame, facilitated the recognition of the influenza HA stem between humans and macaques. However, many crucial questions remain unanswered regarding this phenomenon. For example, it is unclear whether or not the VH gene poses an immunogenetic bottleneck in macaques like the IGHV6-1 gene does for the VH6-1+DH3-3 public clonotype. Furthermore, studies have yet to elucidate why bnAb 789-203-3C12 does not neutralize H3N2 viruses and whether this is a common limitation among all NHP cross-group bnAbs. More concerted antibody discovery efforts would be needed to address these questions and leverage the NHP model for studying the germline gene-endowed immunological processes in response to the HA stem.

A recent study showed that the subtype of influenza virus that the individual was first exposed to in their early childhood (H1N1, H2N2, or H3N2) determines one's "immunological imprint". It likely introduces birth year-dependent bias in immune repertoire and influence group preference for vaccine-induced immunity, making it difficult to achieve universal influenza immunity in humans. Yet, vaccine candidates in active clinical development have shown promise in phase I human clinical trials by eliciting HA stem-directed antibody responses in adults with pre-existing influenza immunity. Even the study reported group 1 HA stem- and group 2 HA stem nanoparticles have entered phase 1 clinical trials evaluating their safety and immunogenicity in humans. Shortly, there will be a wealth of information regarding HA stem-based vaccination strategies showing ways to elicit and reshape the immune repertoire in humans in the face of pre-existing influenza immunity. Alternatively, establishing a stem-directed broadly cross-reactive immunological imprint through vaccination in infants is feasible, although the same seems improbable in adults.

The building block of HA stem nanoparticles was encoded by a single gene (human DH3-3 gene) and its assembly was precise and efficient, making it a candidate for genetic vaccine delivery approaches such as mRNA and recombinant viral vectors. These vaccine modalities have helped successfully combat the ongoing SARS-CoV-2 pandemic and are anticipated to play a major role in future vaccines.

Overall, this study provided proof of concept that these vaccine modalities with innovative new adjuvants would help overcome challenges, such as pre-existing cellular immunity and durability. This approach would also offer better manufacturing and dose-sparing options alongside eliciting broad cross-group protective immunity against multiple influenza viruses.