Applying valency-based immuno-selection to generate broadly cross-reactive antibodies against influenza hemagglutinins

Hinke, Daniëla Maria; Anderson, Ane Marie; Katta, Kirankumar; Laursen, Marlene Fyrstenberg; Tesfaye, Demo Yemane; Werninghaus, Ina Charlotta; Angeletti, Davide; Grødeland, Gunnveig; Bogen, Bjarne; Braathen, Ranveig

doi:10.1038/s41467-024-44889-w

Download PDF

Article
Open access
Published: 12 February 2024

Applying valency-based immuno-selection to generate broadly cross-reactive antibodies against influenza hemagglutinins

Nature Communications volume 15, Article number: 850 (2024) Cite this article

1878 Accesses
1 Citations
20 Altmetric
Metrics details

Subjects

Abstract

Conserved epitopes shared between virus subtypes are often subdominant, making it difficult to induce broadly reactive antibodies by immunization. Here, we generate a plasmid DNA mix vaccine that encodes protein heterodimers with sixteen different influenza A virus hemagglutinins (HA) representing all HA subtypes except H1 (group 1) and H7 (group 2). Each single heterodimer expresses two different HA subtypes and is targeted to MHC class II on antigen presenting cells (APC). Female mice immunized with the plasmid mix produce antibodies not only against the 16 HA subtypes, but also against non-included H1 and H7. We demonstrate that individual antibody molecules cross-react between different HAs. Furthermore, the mix vaccine induces T cell responses to conserved HA epitopes. Immunized mice are partially protected against H1 viruses. The results show that application of valency-based immuno-selection to diversified antigens can be used to direct antibody responses towards conserved (subdominant) epitopes on viral antigens.

Multivalent next generation influenza virus vaccines protect against seasonal and pre-pandemic viruses

Article Open access 16 January 2024

Broadly neutralizing antibodies target a haemagglutinin anchor epitope

Article Open access 23 December 2021

A chimeric thermostable M2e and H3 stalk-based universal influenza A virus vaccine

Article Open access 29 June 2022

Introduction

Seasonal influenza epidemics caused by influenza A viruses (IAVs) and influenza B viruses (IBVs) are responsible for approximately 500,000 deaths worldwide every year^1,2. Moreover, IAVs have caused four pandemics during the last century alone³. The main target for neutralizing antibodies against IAVs is viral surface glycoprotein hemagglutinin (HA)⁴. There are 18 different HA subtypes^5,6,7,8. Based on similarities in amino acid sequences, these 18 subtypes are divided into group 1 (H1, H2, H5, H6, H8, H9, H11, H12, H13, H16, H17 and H18), and group 2 (H3, H4, H7, H10, H14 and H15). At present, H1 and H3 viruses circulate in the human population and regularly cause seasonal epidemics. In addition, a number of zoonotic IAV (H5, H6, H7, H9 and H10) have caused isolated but severe disease outbreaks in humans⁹.

The adaptive immune system often responds to antigen in a hierarchical manner; highly immunogenic epitopes are immunodominant, while less immunogenic epitopes are subdominant¹⁰. Current influenza vaccines predominantly elicit antibodies towards immunodominant epitopes in the membrane-distal head domain of HA. Unfortunately, frequent mutations in the HA head allow viral escape from pre-existing immune responses^11,12,13, necessitating annual reformulations of seasonal influenza vaccines¹⁴. It is therefore important to develop vaccine formats that focus B cell responses towards conserved, subdominant HA epitopes, thus enhancing the generation of broadly cross-reactive antibodies.

Human monoclonal antibodies with broad reactivity to many HAs have been isolated¹⁵, suggesting that it should be possible to induce broadly reactive antibodies to subdominant HA epitopes by means of innovative vaccination strategies. The main specificities of broadly neutralizing antibodies are the receptor binding site in the HA head domain^16,17,18, and epitopes in the fusion peptide in the conserved stem domain^19,20,21.

We have recently developed a new vaccine format based on an ACID/BASE-modified Fos-Jun zipper where bivalency versus monovalency of vaccine antigens can be controlled²². Using this strategy, we have shown that antigen bivalency enhances the ability of APC-targeted vaccines to induce potent antibody responses²³. As an explanation, we hypothesized that antigen bivalency increased B cell stimulation by cross-linking B cell receptors (BCRs) in APC-B cell synapses. According to this hypothesis, if two variants of an antigen are expressed together in a heterodimer, B cells with specificity for conserved epitopes shared by the two variants should be preferentially stimulated²³. To test this idea, we here developed a vaccine strategy where single vaccine protein heterodimers should always express two different HA subtypes (e.g. HA_x, and HA_y), each in a monovalent manner. In such molecules, only conserved epitopes shared between HA_x and HA_y should be bivalent while strain-specific HA epitopes should be monovalent. This vaccination strategy should preferentially stimulate B cells (antibodies) with specificity for conserved epitopes, while monovalent strain-specific epitopes should be less immunogenic. To test if such a valency-based immuno-selection strategy could enhance induction of broadly reactive antibody responses against ordinarily subdominant HA epitopes, we have here immunized mice with mixtures of plasmids that together encoded up to 18 different HA subtypes, each HA being monovalently expressed in a large number of heterodimeric vaccine molecules.

Results

APC-targeted heterodimeric proteins with different HAs delivered as DNA plasmids

We selected HA variants from 18 IAV subtypes based on their commercial availability as proteins and their similarity to strains that presently circulate or have circulated in the past (Table 1). A phylogenetic tree (Fig. 1a) of selected HA subtypes was generated based on their amino acid sequences (Supplementary Fig. 1). We inserted genes for the 18 HA subtypes into A and B plasmids that together encode heterodimeric vaccine proteins due to a centrally located modified Fos-Jun zipper heterodimerization motif called A/B (ACID/BASE) (Fig. 1b)^22,24. The different HA subtypes were expressed C-terminally of the dimerization unit. To bivalently target proteins to APC, A and B chains both expressed an N-terminal MHCII (I-E^d)-specific scFv. In non-targeted controls a scFv specific for the hapten 4-hydroxy-3-iodo-5-nitrophenylacetic acid (NIP) replaced the MHCII-specific scFv.

Table 1 Overview of IAV HA subtype virus strains used in the plasmid mix vaccines, ELISA coats and challenge of mice

Full size table

**Fig. 1: Expression of 18 different HA subtypes in heterodimeric vaccine proteins.**

We first tested if all 18 HA subtypes were properly expressed in heterodimeric proteins. HEK293E cells were co-transfected with plasmid pairs that together encoded MHCII-targeted heterodimers with identical HA on their two antigenic arms. Heterodimers could be detected in supernatant by Western blot with a mAb specific for the assembled A/B heterodimerization motif (Fig. 1c), and by ELISA using a combination of mAbs against A/B and a conserved HA stem epitope (Fig. 1d). We verified the secretion of each of the 18 HA subtypes on either A or B chains by co-transfection of A plasmids expressing any of H1-18 together with B plasmids expressing H1, and vice versa. We detected similar levels of heterodimeric proteins with all MHCII-targeted plasmid combinations (Fig. 1e). The results were confirmed for non-targeted scFv^αNIP plasmid combinations (Supplementary Fig. 2). Expression of scFv^αMHCII and scFv^αNIP, on either A or B chains, did not influence the results (Fig. 1e, Supplementary Fig. 2). These findings suggest that A and B chains were assembled and secreted as correctly folded heterodimeric vaccine proteins with any of the 18 HA subtypes as antigen.

MHCII-targeted heterodimeric DNA vaccines that bivalently express two identical HA (H1-18) induce subtype-specific antibody responses

BALB/c mice were vaccinated once with A/B plasmid pairs that encode MHCII-targeted heterodimers with identical HA on their two arms (i.e. bivalent expression). Plasmids were injected intradermally (i.d.) followed by electroporation (EP) to increase cellular uptake of DNA. Six weeks after vaccination, serum IgG reactivity was clearly highest against the HA subtype expressed in the vaccine (Fig. 2a). Some minor cross-reactivity was observed towards phylogenetically related HA subtypes. While most of the 18 HA subtypes induced high antibody titers, H2 and H5 were less immunogenic. The observed IgG titers correlated to the expression levels of bivalent HA vaccines (Supplementary Fig. 3). These results suggest that all 18 bivalently expressed HA subtypes are immunogenic in heterodimeric vaccine molecules.

**Fig. 2: MHCII-targeted heterodimeric DNA vaccines that bivalently express two identical HA induce subtype-specific antibody responses.**

Ten weeks after a single immunization with the various group 1 HA, mice were challenged with influenza H1N1 PR8 virus (Fig. 2b–d and Supplementary Fig. 4a–c). Vaccines expressing homologous MHCII-targeted H1/H1 from PR8 completely protected mice against PR8 virus, consistent with previous results²⁵. Vaccines expressing other HA subtypes within group 1 generally failed to protect against PR8. The minor protection induced by H11, H12, and H13 could be due to T cell cross-reactivity since the antibody responses elicited by these HA subtypes poorly cross-reacted to H1 (Fig. 2a). Mice vaccinated with group 2 HA subtypes were challenged ten weeks later with H7N1 virus (A/turkey/Italy/3889/1999) (Fig. 2e and Supplementary Fig. 4d). The MHCII-targeted vaccine expressing H7/H7 (A/chicken/Italy/13474/1999 H7N1) protected against this heterologous virus strain, while vaccines with other group 2 HA subtypes did not. Overall, to obtain full vaccine-induced protection with these antigenically bivalent vaccines, the HA subtype had to match the subtype of challenging virus.

Different HA subtypes can be combined in one heterodimeric mix vaccine

We have previously shown that antigen bivalency in dimeric vaccines increases humoral responses²³. Heterodimers that carry two different HA subtypes are bivalent for conserved epitopes while monovalent for unique epitopes, as illustrated in Fig. 3a. We therefore hypothesized that forced monovalency for HA subtypes in dimeric vaccine molecules could selectively favor activation of B cells reactive towards shared conserved regions over those specific for unique epitopes on HA subtypes.

**Fig. 3: The mix plasmid vaccines are translated into vaccine proteins in vitro and induce HA subtype-specific antibody responses upon immunization of mice.**

To test our hypothesis, we expressed nine HA subtypes with A plasmids, and the residual nine subtypes with B plasmids, equally splitting both group 1 and 2 HAs (Fig. 3b). Theoretically, 81 HA monovalent heterodimers should be formed. To test if plasmid mixtures were translated into anticipated vaccine proteins, HEK293E cells were transiently co-transfected with a mixture of the nine A plus the nine B plasmids. For both MHCII-targeted and NIP-targeted mixes, we detected heterodimers in sandwich ELISAs using either H1- and H3-specific mAbs or a broadly reactive anti-HA stem mAb in combination with the anti-A/B mAb, suggesting formation of heterodimeric vaccine proteins (Fig. 3c). When capturing with an anti-H1 mAb, dimers were detected with anti-stem and anti-H3 mAbs but not with the same anti-H1 mAb as used for capture, demonstrating formation of H1/H3 but not H1/H1 heterodimers. This result indicates a lack of heterodimers expressing two identical HA. Non-targeted heterodimers were secreted more efficiently than MHCII-targeted versions, consistent with previous observations^25,26.

A heterodimeric HA mix vaccine induces antibody responses against HAs included in the mix

To test the immunogenicity of the mix vaccines, we prepared mixes of 16 HA subtypes as described in Fig. 3a, but leaving out H1 (group 1) and H7 (group 2), resulting in 64 possible heterodimers. We hypothesized that a high vaccine dose might favor induction of strain-specific rather than cross-reactive antibody responses. We therefore performed an initial titration of mix vaccines which showed that vaccination with 1 µg/plasmid induced only low amounts of strain-specific antibody responses for all HA subtypes after one immunization (Supplementary Fig. 5). We therefore selected this dose for further experiments. Mice were immunized i.d. three times with either an MHCII-targeted or a non-targeted version of the DNA mix vaccine. Five weeks after the first immunization, IgG responses were observed against each of the tested HA variants included in the plasmid mix (Fig. 3d). However, the antibody titers were generally lower than those obtained after a single vaccination with plasmids encoding MHCII-targeted heterodimers with bivalent HA expression (Fig. 2a). This could be due to a lack of bivalency, lower amounts of a given HA in the mix, or both. After two boosts, antibody levels increased and reached titers of approximately 10⁴ (Fig. 3d). No differences between the MHCII-targeted and non-targeted mixes were observed. This result is consistent with the observation that the APC-targeting effect observed with bivalent antigen expression^{25,26,27,28,29} is less evident when antigen is monovalent²³. Similar results were obtained after intramuscular (i.m.)/EP immunizations (Supplementary Fig. 6). Note that the HA-reactive antibodies detected in these experiments could be either cross-reactive or subtype-specific.

Vaccination with DNA plasmid and protein mixes induces antibodies against HA subtypes not included in the mix

To test whether the mix vaccine induces cross-reactive antibodies, mice were vaccinated i.d. with DNA mix vaccines that contained either 16 HAs (ΔH1ΔH7) or all 18 subtypes. IgG titers against H1 and H7 were low after a single immunization but increased after two boosts (Fig. 4a). Interestingly, the inclusion of H1 and H7 in the mix did not improve antibody responses against H1 and H7. This is to be expected since a strong anti-H1 response requires bivalency²³. MHCII-targeting did not significantly enhance the levels of H1- and H7-reactive IgG.

**Fig. 4: Vaccination with an HA plasmid mix induces antibodies against HA subtypes not included in the mix.**

We obtained similar results after i.m. immunization (Fig. 4b). Here, MHCII-targeting slightly enhanced antibody responses against H1 but not H7 (Fig. 4b–d). The cross-reactive antibodies were of both IgG1 and IgG2a subclasses (Fig. 4c). Plasmid mix vaccines also induced antibodies reactive to HA of another H1 strain, A/California/07/2009 (Cal07) (Supplementary Fig. 7). The H1-reactive antibodies bound inactivated H1N1 PR8 virus in ELISA, indicating that the epitopes bound by cross-reactive antibodies are expressed on viral particles (Fig. 4d). These results were essentially confirmed in experiments using separate mixes for group 1 and group 2 HA (Supplementary Fig. 8). Serum antibodies induced by the mix showed no detectable virus neutralization activity against PR8 H1N1 virus (Fig. 4e and Supplementary Fig. 9a), nor against Cal07 H1N1, H3N2 (A/HongKong/1/1968) or H7N1 (A/turkey/Italy/3889/1999) virus (Supplementary Fig. 9a). To explore the ability of vaccine-induced antibodies to engage in Fc-mediated antibody-dependent cellular cytotoxicity (ADCC), we tested whether induced H1-reactive antibodies could activate mouse NK cells. Mice immunized with the MHCII-targeted (ΔH1ΔH7) mix vaccine induced H1-reactive serum antibodies that enhanced levels of IFN-γ⁺ and CD107a⁺ NK cells compared to the NaCl and H7/H7 control-vaccinated mice (Fig. 4f and Supplementary Fig. 9b, c). Taken together, immunizations with plasmid mixes encoding most HA subtypes, monovalently expressed in heterodimeric vaccine proteins, induce cross-reactive antibodies against non-included HA.

The induction of cross-reactive antibodies using MHCII-targeted plasmid mixes was extended to use of a mix of purified heterodimeric proteins (Supplementary Fig. 10a, b). The heterodimeric mix of 16 HA in combination with AddaVax adjuvant induced IgG antibodies against non-included H1 (Fig. 4g). The H1-reactive antibodies were of both IgG1 and IgG2a subtypes (Supplementary Fig. 10c). In the absence of adjuvant, the MHCII-targeted mix proteins induced some anti-H1 IgG1, but not IgG2a. This result supports the idea that immunization with a plasmid mix vaccine induces transfected cells to secrete a protein mix vaccine that elicits antibody responses.

Mix plasmid vaccines induce antibodies that cross-react between closely related HA

To directly analyze HA cross-reactivity of antibodies induced by the ΔH1ΔH7 16 HA plasmid mix, we performed inhibition ELISAs. Sera of vaccinated mice were pre-incubated with various recombinant HAs before adding the samples to wells coated with H1 or H7 (Fig. 5a, b). In such inhibition assays, a reduced signal in the presence of a competitor indicates that single antibody molecules bind both the coat HA and the competing HA. Binding to H1 was inhibited more effectively by group 1 HAs as compared to group 2 HAs. Moreover, the level of inhibition was most pronounced for group 1 HAs with the highest sequence homology to H1 (H2, H5, H6) (Fig. 5a, Supplementary Fig. 11). In contrast, in bivalent H1/H1-vaccinated mice, serum antibodies binding to H1 coat could only be inhibited by soluble H1 and no other HAs. (Fig. 5a). The inhibition pattern was not influenced by the inclusion of H1 and H7, nor by APC-targeting versus non-targeting.

**Fig. 5: Vaccination with HA plasmid mixes induces serum antibodies that cross-react between HA subtypes.**

Using H7 as a coat antigen, serum antibodies from mix-vaccinated mice were only inhibited from binding to H7 coat by H15 (~78% homology), but not by H3, H4 or H14 (each ~50% homology) (Fig. 5b, Supplementary Fig. 11). Antibodies induced by a bivalent H7/H7 vaccine had a different specificity since binding to H7 coat was only inhibited by H7 and not H15. Again, the inclusion of H1 and H7 in the mix, or APC-targeting versus non-targeting, did not influence induction of cross-reactive antibodies binding H7.

We extended the studies to HAs included in the mixes. Results with two group 1 HAs, H2 and H5, were similar to that described above for H1 (Fig. 5c, d) since HAs of closest homology inhibited the best, and APC-targeting or not did not influence the results. Using group 2 H3 as coat, we only observed effective inhibition with H3 and only weak inhibition with the other HAs (Fig. 5e).

To confirm the results of the inhibition ELISA, we used H1^PR8 Sepharose beads to purify H1-reactive antibodies from pooled sera of vaccinated mice. Purified H1-reactive antibodies from mix-vaccinated mice [scFv^aMHCII group 1 + 2(∆H1, ∆H7)] bound H1 and a range of other HA subtypes in ELISA (Fig. 5f, Supplementary Fig. 12). The binding reactivity of these antibodies mimicked the cross-reactivity observed with the inhibition ELISA. Antibodies purified from mice vaccinated with bivalent H1/H1 showed minor cross-reactivity to closely related H2 and H5, but had no significant reactivity to the other HA subtypes tested. This result confirms that serum antibodies elicited by mix vaccination show broad reactivity against various HAs.

Collectively, these results show that the plasmid mix vaccine strategy induces antibodies that cross-react with the most similar HAs, suggesting that cross-reactive epitopes often may be limited to discrete groups of related HAs. Cross-reactivity between group 1 and 2 HAs appears lower than cross-reactivity within the groups, most likely due to the low sequence homology between HA of the two groups.

Plasmid mix vaccines induce antibodies reacting to head and stems of HAs not included in the vaccine

The membrane proximal stem of HA is conserved and subdominant while the distal globular head is variable and immunodominant (Supplementary Fig. 1)^4,30. We measured antibodies against the H1 and H7 globular heads and stem domains in ELISAs. We observed that all mice immunized with the group 1 + 2 (ΔH1ΔH7) 16 mix vaccine, whether MHCII-targeted or not, generated IgG that bound the H1 stem (Fig. 6a). IgG responses against the H7 stem were detected only in some mice (Fig. 6b), and titers were lower. Antibodies against globular heads of H1 and H7 were found in many but not all mice. Inclusion of H1 and H7 in the mix vaccine, containing all 18 HA subtypes, gave similar results.

**Fig. 6: Vaccination with HA plasmid mixes induces antibodies against both the stem and head domains of HAs not included in the mix.**

These results were confirmed in a second experiment employing a large number of ΔH1ΔH7 mix-immunized mice (Supplementary Fig. 13a, b). Both H1- and H7-reactive IgG in serum from group 1 + 2 (ΔH1ΔH7)-vaccinated mice were skewed towards stem reactivity, as compared to control mice immunized with H1/H1 and H7/H7 bivalent vaccines (Fig. 6c). However, there was considerable variability in antibody specificity for H1 and H7 stem and head domains between individual mix-vaccinated mice (Fig. 6d, e), suggesting an oligoclonality of cross-reactive antibodies and a stochastic influence on specificity patterns.

Plasmid mix vaccines induce antibodies skewed away from immunodominant antigenic sites

We further analyzed the specificity of serum from vaccinated mice to the five main immunodominant antigenic sites (Sa, Sb, Ca1, Ca2 and Cb) in the head domain of H1^4,31,32. We coated ELISA wells with recombinant HA proteins derived from various mutated PR8 viruses in which four out of five (Δ4) antigenic sites were altered, leaving only the indicated immunodominant antigenic site intact (e.g. Sa in Sa∆4)³³. A sixth mutant, S12, contained modifications in all five immunodominant sites³⁴. The reactivity of serum from ΔH1ΔH7 mix-vaccinated mice to PR8 HA was not hampered by mutations that abrogated the immunodominant antigenic sites and showed strong reactivity to S12, indicating targeting of non-canonical antigenic sites. In contrast, sera from mice vaccinated with H1/H1 bivalent vaccines had reduced reactivity to most of the mutants, as would be expected (Fig. 6f and Supplementary Fig. 14). This indicates that the specificity pattern of antibodies elicited by mix vaccines is shifted away from the immunodominant antigenic sites. The inclusion of H1 (and H7) in the mix vaccine resulted in a reactivity pattern to the H1^PR8 mutants that is more similar to that induced by the H1/H1 bivalent vaccine.

Plasmid mix vaccine induces HA-specific T-cell responses

To test if the mix vaccines could induce T cell responses, we immunized mice for various number of times with different plasmid mixes and analyzed T cell responses in draining lymph nodes (dLN) and spleen employing three sources of H1 peptides: IYSTVASSL, a K^d-restricted CD8 epitope³⁵, conserved between the group 1 HA subtypes (Supplementary Fig. 1); an overlapping PR8 peptide pool; and an overlapping Cal07 peptide pool (Fig. 7, Supplementary Fig. 15a–c). Based on previous experiments with APC-targeted DNA vaccines²⁵, T-cell responses were detected by the production of IFN-γ. The results show that: (i) the group 1 + 2 (∆H1∆H7) mix, lacking H1, induced IFN-γ T cell responses towards all three sources of H1 peptides; (ii) weak T cell responses were detected in dLN already 5 days after a single mix immunization and increased by day 11 post-vaccination (Fig. 7a, Supplementary Fig. 15a); (iii) three mix immunizations elicited increased responses, almost equal to those elicited by a positive control vaccine (MHCII-targeted H1/H1) (Fig. 7b); (iv) MHCII-targeting of mix vaccine enhanced responses only early (day 5) after a single vaccination (Fig. 7a); (v) The 18 HA mix vaccine, where PR8 H1 was included, induced higher T cell responses against H1 peptides than did the ΔH1ΔH7 mix (Fig. 7b, Supplementary Fig. 15a, b); (vi) H1-specific responses were mainly seen for CD8⁺ T cells (Supplementary Fig. 15b). Taken together, mix vaccines elicit T cells cross-reactive to non-included HA (H1). Inclusion of H1 in the mix, MHCII-targeting, and multiple immunizations, enhanced responses.

**Fig. 7: The HA mix vaccine induces IFN-γ secreting T cells.**

Cross-reactive antibodies mediate protection against challenge with an HA heterologous influenza virus

Next, we explored if the cross-reactive antibodies and T cells induced by the HA mix vaccines could protect against challenge with a heterologous virus that expresses an HA subtype not included in the mix. Mice were immunized thrice with a group 1 + 2 (ΔH1ΔH7) mix vaccine. Four weeks after the final vaccination, mice were challenged with a lethal dose (2.5xLD₅₀) of H1 PR8 virus. Both MHCII-targeted and non-targeted mix vaccines induced a partial protection, with most mice losing a considerable amount of weight. However, both vaccines protected about 50% of the mice from reaching the endpoint of 20% weight loss (Fig. 8a). The positive control, MHCII-targeted H1/H1(PR8), induced full protection, as expected²⁵. The mix vaccines reduced viral titers in the lungs of PR8-challenged mice (Supplementary Fig. 16a). When extending the challenge experiments to other H1 and H7 viruses we obtained divergent results. The plasmid mix vaccine did not offer significant protection to H7N1 virus (A/turkey/Italy/3889/1999) (Supplementary Fig. 16b) which, in our hands, is more pathogenic than PR8 virus. In contrast, mix-vaccinated mice were fully protected against challenge with 5 x LD₅₀ Cal07 H1N1 virus two weeks after a third immunization (Fig. 8b). When the vaccine was generated as protein, two immunizations with the MHCII-targeted ΔH1ΔH7 mix protein fully protected mice against challenge with 5xLD₅₀ PR8 virus, but only when delivered with AddaVax adjuvant (Fig. 8c).

**Fig. 8: The HA mix vaccine induces partial protection against heterologous influenza virus.**

The partial protection against the PR8 heterologous challenge could be due to H1 cross-reactive antibodies (Figs. 4–6) or T cells (Fig. 7) induced by the group 1 + 2 (ΔH1ΔH7) mix vaccination. To test the contribution of antibodies and T cells, mice vaccinated twice with the MHCII targeted group 1 + 2 (ΔH1ΔH7) mix were depleted of T cells by injection of anti-CD4 and anti-CD8 mAbs (or isotype-matched control mAbs) starting two days prior to a challenge with 5 x LD₅₀ of PR8 virus (Fig. 8d, Supplementary Fig. 16c). T cell-depleted mice were as well protected as non-depleted mice. However, only 10–20% of mice survived the challenge in either group, perhaps due to the high dose of influenza virus used for challenge. Nevertheless, the protection conferred by mix-vaccination, either in the absence or presence of T cell depletion, was significant compared to negative controls (NaCl and MHCII-targeted H7/H7 immunized mice). These results support a role for cross-reactive antibodies as mediators of the protection against virus with an HA subtype that was not included in the mix vaccine.

Discussion

Current influenza vaccines induce predominantly strain-specific antibodies. These are protective as long as the infecting influenza strain matches the vaccine strains. An important goal in influenza vaccine research is therefore to develop universal vaccines that induce broadly reactive antibodies that protect against many, if not all, influenza strains.

We recently observed that MHCII-specific heterodimeric vaccine proteins generated APC-B cell synapses²³. Moreover, antigen bivalency, meaning that heterodimeric proteins express two identical antigens on both antigenic arms, was necessary to induce potent antibody responses, most likely due to BCR crosslinking. We here hypothesized that in a mixture of many different heterodimeric vaccine proteins, each of which always expresses two different but related antigens on its two arms, only conserved epitopes shared between the related but different antigens would be bivalent in a single heterodimeric vaccine protein. Consequently, B cells with a BCR that recognize these conserved bivalent epitopes should preferentially be stimulated. To test this hypothesis of valency-based immuno-selection, we generated a mixture of DNA plasmids that encoded heterodimeric vaccine proteins expressing 16 different IAV HA subtypes, each heterodimer carrying two different HAs. The plasmid mix induced cross-reactive antibodies against H1 and H7, which were not included in the mix. The cross-reactive antibodies partially and fully protected against two different hetero-subtypic H1 virus challenges in the absence of CD4⁺ and CD8⁺ T cells.

Analysis of sera from mix-immunized mice suggested that a fraction of induced antibodies was cross-reactive between H1 (or H7) and certain HA subtypes expressed in the mixed vaccine. These cross-reactive antibodies apparently bound shared conserved sites that ordinarily are subdominant. In inhibition ELISA, cross-reactive antibodies were only inhibited by HAs that were most homologous to non-included H1 (or H7) suggesting a restricted sharing of conserved epitopes among all HA subtypes. Supporting this notion, the mix vaccines elicited antibodies that cross-reacted between related HAs that had been included in the mix, again indicating that the vaccination strategy skews antibody responses towards conserved epitopes shared between similar HAs.

The assay used here demonstrated cross-reactivity at the polyclonal antibody level. Generating mAb from mix-vaccinated mice should enable structural insights into the localization of conserved epitopes recognized by broadly cross-reactive antibodies. Our results suggest that relevant epitopes are localized to both the head and the stem of HA. There was, however, considerable heterogeneity between individual mice, indicating that cross-reactive B cell responses may be oligoclonal.

Sera from mix-vaccinated mice did not have any meaningful neutralizing activity, but they were able to activate NK cells. Broadly reactive antibodies against HA often protect through FcγR-dependent mechanisms, such as antibody-dependent cellular cytotoxicity^36,37,38. Hence, it might be important that mixed plasmid vaccines skewed cross-reactive antibody responses towards the IgG2a subclass over IgG1. Important in this respect, the specificity of targeted vaccines for surface molecules on APC influences antibody isotype. For example, chemokines MIP-1α²⁶ and Xcl1³⁹ predominantly induce FcγR-activating IgG2a antibodies. The MHCII-targeting used herein induced both IgG1 and IgG2a cross-reactive antibodies. Thus, substituting this targeting unit with MIP-1α or Xcl1 should further increase the IgG2a profile of cross-reactive antibodies and improve the ability of serum antibodies to engage in Fc-mediated ADCC.

Other approaches that combine different HA variants into one vaccine have been used to induce cross-reactive antibodies against HA. We previously showed that simultaneous immunization with six group 1 HA subtypes expressed in MHCII-targeted vaccine proteins induced antibodies against H1 not included in the mix⁴⁰. However, that study used DNA plasmids encoding protein chains that could form homodimers. Therefore, a fraction of secreted dimers are likely to bivalently express the same HA⁴⁰, making interpretation more complicated compared to the present use of exclusively heterodimeric molecules.

Kanekiyo et al. showed that heterotypic nanoparticles with eight different H1 head constructs induced broadly reactive antibodies⁴¹. The authors suggested that having an avidity advantage improves the activation of cross-reactive B cells, and subverts the activation of strain-specific B cells unable to bind shared epitopes⁴¹. Other groups have shown that intranasal delivery of a mixture of four different virus-like particles, each carrying a different HA subtype, induced significant protection against a variety of heterosubtypic IAV strains. However, this protection was likely not due to cross-reactive antibodies⁴². Sequential immunizations with chimeric HAs that combine the same stem domain with different irrelevant head domains induce cross-reactive antibody responses against the subdominant HA stem domain⁴³. The computationally optimized broadly reactive antigen (COBRA) methodology, using multiple rounds of layered consensus building to generate influenza vaccine HA immunogens, induced broad hemagglutination inhibition activity as a head-based strategy⁴⁴.

The use of “mix vaccines” has also been studied for several other infectious diseases, including malaria^45,46,47, HIV⁴⁸ and infection with SARS-like viruses⁴⁹. In silico work suggests that combining multiple malaria antigens in one vaccine favors affinity maturation towards cross-reactive antibodies⁴⁵. Immunization with four different variants of AMA1 malaria antigen in rabbits enhanced cross-reactive responses as compared to immunization with one single AMA1⁴⁶. In silico models with HIV gp120 mix vaccines suggest that the generation of an optimal mix vaccine requires careful consideration of the number of antigens, their concentration, and the mutational distance between different antigens⁴⁸. Single mosaic nanoparticles expressing the receptor binding site of eight SARS-CoV-2 related viruses induced cross-reactive antibodies targeting more conserved sites in mice and non-human primates⁴⁹.

The vaccination strategy used herein may be optimized and expanded. Many more HAs may be introduced into an expanded plasmid mix. Alternatively, mixes with only variants of H1, H3 or IBV HA could be generated to induce broad protection against strains that currently cause seasonal influenza epidemics. Besides influenza, the mixed strategy could be valuable for generating vaccines against other viral diseases, such as the highly polymorphic gp120 of HIV viruses or the spike of SARS-CoV-2.

The mixed vaccine could be delivered both as plasmids and proteins, with similar results. This is to be expected since cells transfected with many (or all) mixed plasmids are likely to secrete heterodimeric molecules assembled stochastically in the ER^22,23. Most likely, the heterodimeric mix vaccine could also be delivered as modified mRNA. Supporting this possibility, a recent publication from Arevalo et al. describes a multivalent mRNA vaccine encoding 18 IAV and 2 IBV HA subtypes. Vaccinated mice developed antibodies against all the included HA subtypes. However, it is unclear to what extent antibodies were cross-reactive or not⁵⁰.

Based on the present results using DNA delivery, and the apparently high immunogenicity of mRNA vaccines⁵⁰, one may ask whether the valency-based immuno-selection mechanism afforded by our mix APC-targeting heterodimers would be maintained upon mRNA rather than DNA delivery. If so, the high immunogenicity of mRNA vaccines could nevertheless be outweighed by several advantages of plasmid delivery such as (i) ease of engineering and production, (ii) independence of a cold chain, and (iii) needle-free jet delivery⁵¹.

A vaccine that reliably induces protection against multiple influenza strains may eliminate the need for annual updates of seasonal influenza vaccines. An important question is when such a vaccine should be administered since previous exposures to influenza viruses could thwart its effectiveness due to imprinting (original antigenic sin)^52,53,54,55. Perhaps a valency-based mix vaccine should be the first immunization to influenza, given very early in childhood, before any imprinting has occurred. Arguing against a necessity of early mix vaccination, Arevalo et al. showed that prior infection of mice with an H1N1 virus did not hamper the induction of antibodies elicited by their mRNA mix vaccine⁵⁰.

Methods

Mice, cell lines, and viruses

Female BALB/cAnNRj/c mice (Janvier Labs, Le Genest Saint Isle, France or Taconic biosciences inc, NY, USA), 6-8 weeks old, were used for the experiments. All experiments were reviewed and approved by the Norwegian Animal Research Authority and were carried out in accordance with the recommendations from the Guide for the Care and Use of Laboratory Animals of the Norwegian National Institute of Health. HEK293E cells were from ATCC (RRID:CVCL0045, cat# CRL-1573, Manassas, USA). Madin-Darby canine kidney (MDCK) cells were likewise purchased from ATCC (RRID:CVCL0422, cat# CCL-34). Cells lines were cultured in complete medium: RPMI 1640 (R2405, Invitrogen) with 10% heat-inactivated FCS (F7524, Sigma), 24 mg/l gentamicin (Gensumycin®, VNR 453130, Sanofi-Aventis Norge AS), 50 µM monothioglycerol (M6145, Sigma), 1 mM sodium pyruvate (11360-039, Gibco), and 0.1 mM non-essential amino acids (11140-035, Gibco). The growth medium for MDCK cells contained the same supplements, but in DMEM (41965062, Life Technologies). Mouse-adapted (VR-95) and MDCK-adapted (VR-1469) viruses from strain A/Puerto Rico/8/1934 (H1N1, PR8) were purchased from ATCC. Virus strain A/California/07/2009 (H1N1, Cal07) was kindly donated by Dr. A. Germundsson (The National Veterinary Institute, Oslo, Norway), the LD₅₀ dose was established by the Reed and Muench method and titrations of virus in mice²⁵. Virus from strain A/turkey/Italy/3889/1999 (H7N1) was kindly donated by Prof. Rebecca Cox (UiB, Bergen, Norway), and was mouse adapted in our lab before use.

Generation of phylogenetic tree and multiple sequence alignment of HA proteins

The phylogenetic tree was generated using a Clustal Omega multiple alignment of the amino acid sequences of the HA proteins used in this paper. The tree was made with the neighbor-joining method without distance correction (URL: https://www.ebi.ac.uk/Tools/msa/clustalo/⁵⁶). The radial tree was visualized using an online version of TreeDyn 198.3 (URL: http://www.phylogeny.fr/⁵⁷). The graphic amino acid alignment of the HA proteins used in this paper was made using blastp suite (URL: https://blast.ncbi.nlm.nih.gov/Blast.cgi?PROGRAM=blastp&PAGE_TYPE=BlastSearch&LINK_LOC=blasthome⁵⁸).

Cloning of heterodimeric vaccine vectors

DNA vaccines were encoded in the pLNOH2 vector under a CMV promotor and the signal peptide sequence from the V_H gene of B1-8 mAb⁵⁹. The targeting units, anti-MHCII (I-E^d, scFv^αMHCII from mouse 14-4-4 S mAb) and anti-NIP (scFv^αNIP from mouse B1-8 mAb) were subcloned from previously constructed plasmids²⁷. The heterodimerization unit was based on a modified Fos-Jun based leucine zipper motif consisting of ACID (QLEKELQALEKENAQLEKELQALEKELAQ) and BASE (QLKKKLQALKKKNAQLKKKLQALKKKLAQ) sequences on separate protein chains. Sequences for the A/B dimerization motif were kindly donated by Dr. Elisabeth Mellins²⁴. Within the previously constructed vaccine plasmids, two A or two B motifs were spaced by a linker (GGGSGGLTKFGGSTTAPS), and placed C-terminal of the hinge exon 1 from human IgG3²². The HA subtype strains used for H1-18 of IAV are listed in Table 1, and the amino acid sequences of each HA subtype strain are shown in Supplementary Fig. 1. H1 from PR8 and A/California/07/2009 (Cal07)²⁵, H5, H6, H8, H9, H11, H13⁴⁰ and H7⁶⁰ were subcloned from existing plasmids with restriction enzyme SfiI²⁵. The remainder of the HA genes were amplified from mammalian codon optimized cDNA (Sino Biological, category numbers indicated in Supplementary Table I) by PCR with specifically designed primers (Sigma):

H2₁₆5′: ACCGGCCTCGGTGGCCTGGACCAGATTTGTATTGGCTACC,

H2₅₃₉3′: ACCGGCCCTGCAGGCCTCAGAGGCTCAGGGATCCAG,

H3₁₇5′: GGCCTCGGTGGCCTGCAAGACCTTC,

H3₅₃₈3′: ACCGGCCCTGCAGGCCTCATCAAATAGCAAAAG,

H4₁₇5′: ACCGGCCTCGGTGGCCTGCAGAACTACACAGGCAACC_,

H4₅₃₇3′: ACCGGCCCTGCAGGCCTCAGGAGATGCTGAAGGAGAAC,

H10₁₈/H15₁₉5′: ACCGGCCTCGGTGGCCTGGACAAGATTTGTCTGGGACACC,

H10₅₃₄/H15₅₄₃3′: ACCGGCCCTGCAGGCCTCAGGAGGCTCCAAAGGAGAAC,

H12₁₈5′: ACCGGCCTCGGTGGCCTGGACAAGATTTGTATTGGCTACCAG,

H12₂₃₉3′: ACCGGCCCTGCAGGCCTCACAGGGAGGATGCCACAGA,

H14₁₈5′: ACCGGCCTCGGTGGCCTGCAGATTACCAATGGCACCAC,

H14₅₄₁3′: ACCGGCCCTGCAGGCCTCAACTCATACTGAAGGAAATCCACAG,

H16₂₀5: ACCGGCCTCGGTGGCCTGGACAAGATTTGTATTGGCTACCTG,

H16₅₄₁3′: ACCGGCCCTGCAGGCCTCACAGCACAATGCTGGAGGC,

H17₁₉5′: ACCGGCCTCGGTGGCCTGGACAGGATTTGTATTGGCTACCAG,

H17₅₄₁3′: ACCGGCCCTGCAGGCCTCACAGCACCACAGAGGAGGC,

H18₁₅5′: ACCGGCCTCGGTGGCCTGGACCAGATTTGCATCGGC, and

H18₅₃₈3′: ACCGGCCCTGCAGGCCTCACAGCACGACACTTGAGGC (SfiI sites indicated in italic, stop codon in bold). Primers were designed to exclude the intrinsic HA signal peptide and most of the transmembrane region from the HA gene. The primer names show the amino acid numbers included for the individual HA subtypes.

Transient transfection of HEK293E cells

HEK293E cells (5 × 10⁵) in 5 ml complete medium per well were seeded in 6 well culture plates (3506, Corning), and incubated at 37 °C with 5% CO₂. The following day, the complete medium was replaced by FreeStyle™ 293 serum free medium (12338002, Life Technologies) with 24 mg/l gentamicin before cells were transfected with 2.5 µg DNA of A and B plasmids (5 µg in total) with 10 µg Polyethylenimine (23966, Polysciences). To produce HA mix proteins with 18 different plasmids, 5 µg DNA/plasmid and 200 µl polyethylenimine were used for transient transfection of HEK293E cells in T175 flasks. Supernatants were collected after 3-4 days, and centrifuged 5 min at 350 xg. For Western blot analysis and HA mix heterodimer ELISA, supernatants were concentrated 10 times using Vivaspin 20 spin columns with 10 kDa cut off (VS2002, Sartorius).

Protein purification

For the production of larger amounts of MHCII-targeted ΔH1ΔH7 mix and MHCII-targeted H1/H1 vaccine proteins, HEK293E cells were transfected with a total of 250 µg DNA in BD Falcon® 5-layer multiflasks (353114, Corning) in FreeStyle^TM 293 serum free medium. Proteins were purified using an affinity column with anti-A/B mAb (clone 2H11, produced from hybridoma kindly provided by E.L. Reinherz⁶¹) coupled to CNBr activated Sepharose 4 Fast Flow (17-0981-01, GE healthcare).

ELISA to detect vaccine proteins

High binding 96 well plates (3590, Corning) were coated with 1 µg/ml anti-A/B mAb, or anti-IAV H1N1 (A/Puerto Rico/8/1934) HA (clone H36-4-52, produced from hybridoma provided by Siegfried Weiss⁶² mAb in PBS, incubated at 4 °C overnight, and blocked with 200 µl blocking buffer (1% w/v BSA in PBS with 0.02% w/v Na-Azide) per well. Supernatants from transfected HEK293E cells were added (50 µl/well) and serially diluted 3-fold in triplicates with ELISA buffer (0.1% w/v BSA and 0.2% Tween 20 in PBS/Na-Azide). Plates were incubated for 2 hours at room temperature. Vaccine proteins were detected for 1 hour with 1 µg/ml biotinylated anti-H1 mAb, a cross-reactive rabbit mAb specific for IAV HA stem (1:3000, clone #2, 86001-RM01, Sino Biological), or 1 µg/ml rabbit-anti-AIV H3N2 (A/Brisbane/10/2007) HA mAb (clone# 104, 11056-R104, Sino Biological). Biotinylated mAb was detected with streptavidin-Alkaline Phosphatase (ALP) conjugate (1:3000, RPN1234, Southern Biotech) and rabbit mAbs with polyclonal (p) anti-rabbit IgG Ab-ALP conjugate (1:3000, A3687, Sigma Aldrich). Signals were developed with phosphatase substrate (1 mg/ml, P4744, Sigma) dissolved in diethanolamine substrate buffer. OD_405nm was measured using EnVision 2104 Multilabel Reader with EnVision Manager 1.12 software (PerkinElmer).

Western blot analysis

Concentrated supernatants were normalized on a dot blot with 3-fold serial dilutions of supernatant in PBS and detected with 0.33 µg/ml anti-A/B mAb. Approximately similar amounts of vaccine proteins were run on a Bolt™ 4-12% Bis-Tris Plus gel (NW04122BOX, Invitrogen). Proteins were transferred to a PVDF membrane using the iBlot2 gel transfer system with a PVDF transfer stack (IB24002, Invitrogen). Membranes were blocked using 2% blocking reagent (RPN2125, GE Healthcare) in 0.1% tween 20 in PBS. Vaccine proteins were detected with 0.33 µg/ml biotinylated anti-A/B mAb followed by streptavidin-HRP conjugate (1:3000, 7100-05, Southern Biotech). Blots were captured using G:box Chemi XX6 device (SYNGENE).

DNA vaccination

Mice were anaesthetized by an interperitoneal injection with 0.06 ml/10 gr body weight of a cocktail of Tiletamin/Zolazepam (250 mg/ml, Zoletil Forte, Vibrac), xylazine (Rompun, 20 mg/ml, Bayer Animal Health), and fentanyl (50 µg/ml, Actavis). Plasmids were purified using Endofree Megaprep plasmid purification kits (12381, Qiagen), and recovered in 0.9% NaCl solution (VNR 451773, B. Braun). Bivalent scFv^αMHCII-H1-18/H1-18 vaccines contained 0.05 µg/µl − 0.5 µg/µl of both A and B DNA plasmids (0.1–1 µg/µl DNA in total). Mix vaccines contained 0.02 µg/µl in 50 µl (i.d. vaccination) or 0.01 µg/µl in 100 µl (i.m. vaccination) of each included plasmid vector (for a final dose of 1 µg/plasmid in total per mouse). For i.d vaccinations, 25 µl DNA was injected in the lower back near the base of the tail on both flanks, immediately followed by EP with DermaVax (BTX Harvard Apparatus), delivered in two pulses of 450 V/cm × 2.5 µs plus eight pulses of 110 V/cm × 8.1 ms. For i.m. vaccination, 50 µl DNA was injected in the quadriceps femoris muscle of each leg, followed by EP with Elgen 1000 Needle Electroporator (Inovio Biomedical), delivered as 5 × 60 ms at 50 V/400 mA with 200 ms delay. In many experiments, mice were immunized three times, at week 0, 5 and 9 unless otherwise specified. Blood samples were collected from the saphenous vein, spun down twice, and sera frozen at −20 °C. Blood was collected at weeks 1, 3, on the day of immunization and two weeks after the second and third immunization.

Protein vaccination

Proteins (50 µg/ml in PBS) were combined in a 1:1 ratio with adjuvant AddaVax (vac-adx-10, Invivogen) or PBS for non-adjuvanted controls. Hind legs of mice were shaved, and 50 µl vaccine protein was injected in the quadriceps femoris muscle on each side, resulting in a final vaccine dose of 2.5 µg protein per mouse. Mice were immunized with protein twice, at week 0 and 7, and blood was collected at 1, 2, 3, 6, 9 and 13 weeks after the first immunization.

ELISA for HA-reactive antibodies in serum

The serum ELISAs were run as described above for protein ELISAs except for the following: Recombinant HA or HA1 subunit protein (0.5 µg/ml, Sino Biologicals, category numbers indicated in Supplementary Table I), or inactivated mouse adapted H1N1 PR8 virus (1:1600, 10100782, Charles River Laboratories) in PBS were used as coat. Serum was serially diluted 3-fold in ELISA buffer starting at 1:50. HA-reactive antibodies were detected with anti-mouse IgG pAb-ALP conjugate (1:5000, A2429, Sigma Aldrich). IgG subtypes were detected with biotinylated 1 µg/ml anti-mouse IgG1^[a] (1:500, clone10.9, 553500, BD Biosciences) or anti-mouse IgG2a^[a] (1:500, clone 8.3, 553502, Biosciences) followed by detection with streptavidin-HRP (1:5000, Southern Biotech) and 100 µl TMB solution (CL07-1000ML, Merck Millipore). Signal development was stopped with 100 µl 1 M sulfuric acid, and OD_450nm was measured. The Ab titer was defined as the final dilution that resulted in OD_405nm or OD_450nm values > (2 x mean) of NaCl-vaccinated mice. Samples with a titer <50 were given an endpoint titer of 1.

Some ELISAs were performed with recombinant HA protein from PR8 viruses that were mutated to abolish four out of five of the major H1 antigenic sites (Δ4 mutants)³³, and a mutant that was altered all five major antigenic sites, abrogating binding of most tested mAb towards these sites (S12)³⁴. The mutant proteins were produced and purified as described previously³³. The recombinant HA proteins were used as ELISA coat at a concentration of 1 µg/ml. Serum IgG titers towards the mutated HA was analyzed as described above.

ELISA for detection of HA stem-specific antibodies

HEK293E cells were transfected with plasmids encoding dimeric fusion proteins that expressed a scFv against the hapten 4-ethoxymethylene-2-phenyl-2-oxazoline-5-one (phOx)⁶³ linked to the human Ig-hinge regions h1 and h3, and CH3 from human IgG3 as dimerization unit²⁷, and to the HA stem derived from either H1 or H7 as antigenic unit⁶⁴ to form scFv^αphOx-CH3-HA stem proteins. phOx (E0753-1G, Sigma) was coupled to BSA. The resulting phOx-BSA (1 µg/ml) was used as coat, and supernatant with scFv^αphOx-C_H3-HA stem proteins was added and incubated for 2 hours. The scFv^αphOx -C_H3-HA stem supernatant was exchanged with fresh supernatant twice to obtain a saturated coating with protein dimers. The presence of scFv^αphOx -C_H3-HA stem in the ELISA was confirmed through binding of mAb towards the human IgG3-based dimerization motif (anti-human IgG-Biotin, 1 µg/ml, clone HP-6017, B3773, Sigma) and rabbit anti-HA stem mAb (1:3000) detected with anti-rabbit IgG Ab-ALP. Mice sera were serially diluted 3-fold in ELISA buffer, starting at 3:50. HA stem-specific antibodies were detected with anti-mouse IgG-ALP conjugate.

Microneutralization assay

The viral 50% TCID of MDCK-adapted H1N1 PR8 influenza virus was determined using the Reed-Muench method⁶⁵. Serum from vaccinated mice was mixed in 1:3 ratio with receptor destroying enzyme from cholera filtrate (C8772, Sigma) incubated at 37 °C for 20 hours, and heat-inactivated at 56 °C for 60 min. Starting at 1:10 dilution, 2-fold serial dilutions of treated serum was incubated with equal volumes of 100xTCID₅₀ PR8 virus in virus diluent (DMEM supplemented with 48 mg/l gentamicin, 100 μM monothioglycerol, 2 mM sodium pyruvate, 0.2 mM non-essential amino acids, 1% bovine albumin fraction V (10735078001, Roche), 0.02 M HEPES (83264-100ML-F, Sigma), and 1 µg/ml TPCK-Trypsin (T1426-100MG, Sigma)) for 2 hours on 96-wells plates at 37 °C at 5% CO₂. As controls, four wells on each plate contained only virus and four wells contained only virus diluent. MDCK cells (2 × 10²) were added to each well and incubated for 18 hours at 37 °C and 5% CO_2. Cells were washed with PBS, fixed with cold 80% acetone for 10 min, and viral nucleoprotein was detected in ELISA using biotinylated anti-nucleoprotein mAb (1 µg/ml, clone HB65, produced from hybridoma, H16-L10-4R5, ATCC) followed by streptavidin-ALP (1:3000). Plates were developed and read as described above.

Analysis of antibody-mediated NK cell activation

To determine ADCC, we adapted a protocol from Jegaskanda et al.³⁶ Briefly, spleens were harvested from naïve BALB/cAnNRj/c mice and dissociated in gentleMACS™ C Tubes (130-096-334, Miltenyi Biotech). Splenocytes were treated with Tris-buffered ammonium chloride for 5 min on ice. Cells were filtered through a 70 nm Nylon strainer (732-2758, VWR). NK cells were isolated using an NK cell negative selection kit (130-115-818, Miltenyi Biotech) according to the manufacturer’s protocol. NK cell purity was determined by Flow cytometry. Purified NK cells were pre-activated overnight with IL15 (10 ng/ml, 447-ML-010, R&D systems) in a complete medium at 37 °C and 5% CO₂. High-binding 96 well plates were coated with 0.25 µg/ml recombinant H1^PR8 protein in PBS, at 4 °C overnight. Coated plates were washed five times with PBS and blocked for 2 hours in 200 µl 1% w/v BSA in PBS at room temperature. Then, wells were incubated with heat-inactivated serum from vaccinated mice at 4 °C overnight. The next day, plates were washed five times with PBS, and 0.5 × 10⁶ pre-activated NK cells resuspended in 100 µl complete cell medium supplemented with 10 ng/ml IL15, anti-mouse CD107a (LAMP-1)-PE antibody (2 µg/ml, clone 1D4B, 121612, Biolegend) and GolgiPlug™ protein transport inhibitor (1:1000, 555029, BD Biosciences) were added to the wells. Cells were incubated for 5 hours at 37 °C and 5% CO₂, after which NK cells were washed, stained with Ghost Dye™ Violet 510 (1:400, SKU 1-0870-T500, TONBO) and blocked with 20% rat serum, followed by staining with anti-mouse CD3e-FITC (5 µg/ml, clone 145-2C11, 35-0031-U500, TONBO) and NKp46-eFluor 450 (2 µg/ml, clone 29A1.4, 48-3351-82, ThermoFisher Scientific). Stained cells were fixed in 3% formaldehyde, permeabilized with Perm/Was buffer (554723, BD Biosciences) and stained intracellularly with anti-mouse IFN-γ-APC (5 µg/ml, clone XMG1.2, 20-7311-U100, TONBO). Cells were analyzed using an Attune NxT flow cytometer (Thermo Fisher Scientific) and FlowJo software version 10.8.1 (BD Biosciences).

Inhibition analysis in ELISA

Sera were diluted to approximately 0.5 OD_405nm, as determined in preceding serum ELISA titrations for coat HA-reactive antibodies, and incubated with various recombinant HA (5 µg/ml, H1, H2, H3, H4, H5, H6, H7, H9, H15, or H16) in binding buffer (PBS with 1% BSA and 0.2% tween 20) for 1 hour at room temperature. Serum-HA mixtures were transferred in triplicates to plates coated with recombinant H1, H7, H2, H3 or H5 (0.5 µg/ml) and incubated at room temperature for 1 h, before total IgG was detected as described above. OD_405nm from NaCl mice sera (diluted 1:50) was subtracted as background before calculating the relative inhibition (%) of OD_405nm signal of competed serum compared to serum that was pre-incubated with binding buffer alone (non-competed serum).

Purification of H1-reactive antibodies from mouse serum

Recombinant H1 protein from PR8 (11684-V08H, Sino Biological) was coupled to CNBr-activated Sepharose 4 fast flow (1.5 mg protein:1 ml sepharose, 17098101, Cytiva). Antibodies were purified from 300 µl pooled mouse serum using the coupled sepharose in Pierce™ spin columns (69705, Thermo Fisher). After extensive washing of beads, antibodies were eluted with 1 column volume 0.1 M glycine HCl pH 2.7. The pH of the eluate was neutralized by adding 2 column volumes of 0.1 M glycine HCl pH 8.0. The glycine HCl solution was exchanged for PBS using three rounds of centrifugation with centrifugal filters with 10 kDa cut-off (UFC501, Merck), ending with a final volume of 100 µl.

T cell ELISPOT

IFN-γ secreting T cells were analyzed using a Mouse IFN-γ ELISpot^PLUS kit (3321-4APT-2, Mabtech) according to the manufacturer’s protocol. Spleens or draining lymph nodes were harvested from vaccinated mice and dissociated in gentleMACS™ C Tubes. Splenocytes were treated with Tris-buffered ammonium chloride for 5 min on ice. Cells were filtered through a 70 nm Nylon strainer before counting, and 5 × 10⁵ viable splenocytes or 2.5 × 10⁵ draining lymph node cells per well in a complete medium were seeded in the pre-coated ELISPOT plates. Cells were re-stimulated for 20 hours at 37 °C at 5% CO₂ with overlapping HA peptides from H1N1 viruses PR8 (PM-INFA-HAPR) or Cal07 (PM-INFA-HACal, 15-mers with 11 aa overlap, 0.1 µg/peptide/ml, JPT Peptide Technologies GmbH), or with 2 µg/ml H1 peptide (aa 532–540, IYSTVASSL, K^d, GenScript). Spots were counted automatically using an ImmunoSpot® CORE cell counter (Cellular Technology Limited). For each sample, spot counts from wells stimulated with 2 µg/ml of an irrelevant peptide (aa 91-107, ALWFRNHFVFGGGTKVT, of the λ2 chain of the M315 mouse myeloma protein, GenScript) were subtracted as background.

Viral challenge with influenza virus

BALB/c mice were anaesthetized as described under DNA vaccination. H1N1 PR8 virus, H1N1 Cal07 virus or A/turkey/Italy/3889/1999 (H7N1) virus stocks were diluted to either 2.5xLD₅₀ or 5xLD₅₀ in PBS as described per experiment, and administered in 10 µl to each nostril. Weight loss was monitored, and 20% weight loss was used as humane endpoint as approved by the Norwegian Animal Research Authority. Mice that reached this endpoint were euthanized and excluded from the weight curves from that point on.

T cell depletion

BALB/c mice were treated with intraperitoneal injections with 100 µl PBS with 100 µg of both purified anti-CD4 (clone GK1.5, produced from hybridoma TIB207, ATCC) and anti-CD8 (clone 53-6.72, produced from hybridoma TIB105, ATCC), or with isotype matched controls (clone SFR8-B6 produced from HB-152, and clone Y13-238 produced from CRL-1741, both from ATCC), starting two days before challenge. Injections were repeated every second day until the end of the experiment. Mice were challenged with 5xLD₅₀ H1N1 PR8 virus, as described above.

Statistical analysis

Results were graphed and analyzed for significance using GraphPad Prism 8 software. Antibody titers and data from T cell ELISPOT and flow cytometry analysis were compared using Kruskal-Wallis with Dunn’s multiple comparisons test. Protein expression by transfected HEK239E cells and weight loss at day 4-6 after viral challenges were compared between groups using one-way ANOVA with Dunnet’s multiple comparisons test. Complete weight loss curves after infection were compared using two-way ANOVA. Survival curves were analyzed for significance using the Mantel–Cox test. In inhibition ELISA, correlations between % homology of competing HA with coat HA and % inhibition was analyzed with the Pearson correlation coefficient. P values < 0.05 were considered as statistically significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Data and material not included in the manuscript supplementary material will be made available to qualified academic researchers upon request to the corresponding author. The data used to generate the main results shown in the main figures and Supplementary Information are available as Source data. Source data are provided with this paper.

References

Iuliano, A. D. et al. Estimates of global seasonal influenza-associated respiratory mortality: a modelling study. Lancet 391, 1285–1300 (2018).
Article PubMed Google Scholar
WHO. Influenza (Seasonal) Fact Sheet 2023. Last update: January 12, 2023. Accessed 6 August 2023. https://www.who.int/news-room/fact-sheets/detail/influenza -(seasonal).
Neumann, G. & Kawaoka, Y. Predicting the next influenza pandemics. J. Infect. Dis. 219, S14–S20 (2019).
Article CAS PubMed Google Scholar
Altman, M. O., Bennink, J. R., Yewdell, J. W. & Herrin, B. R. Lamprey VLRB response to influenza virus supports universal rules of immunogenicity and antigenicity. eLife 4, e07467 (2015).
Article PubMed PubMed Central Google Scholar
Webster, R. G., Bean, W. J., Gorman, O. T., Chambers, T. M. & Kawaoka, Y. Evolution and ecology of influenza A viruses. Microbiol. Rev. 56, 157–179 (1992).
Article Google Scholar
Röhm, C., Zhou, N., Süss, J., Mackenzie, J. & Webster, R. G. Characterization of a novel influenza hemagglutinin, H15: criteria for determination of Influenza A subtypes. Virology 217, 508–516 (1996).
Article PubMed Google Scholar
Fouchier, R. A. M. et al. Characterization of a novel Influenza A virus hemagglutinin subtype (H16) obtained from black-headed gulls. J. Virol. 79, 2814 (2005).
Article CAS PubMed PubMed Central Google Scholar
Wu, Y., Wu, Y., Tefsen, B., Shi, Y. & Gao, G. F. Bat-derived influenza-like viruses H17N10 and H18N11. Trends Microbiol. 22, 183–191 (2014).
Article CAS PubMed PubMed Central Google Scholar
Freidl, G. S. et al. Influenza at the animal–human interface: a review of the literature for virological evidence of human infection with swine or avian influenza viruses other than A(H5N1). Eur. Surveill. 19, 20793 (2014).
Article Google Scholar
Angeletti, D. & Yewdell, J. W. Understanding and manipulating viral immunity: antibody immunodominance enters center stage. Trends Immunol. 39, 549–561 (2018).
Article CAS PubMed Google Scholar
Kirkpatrick, E., Qiu, X., Wilson, P. C., Bahl, J. & Krammer, F. The influenza virus hemagglutinin head evolves faster than the stalk domain. Sci. Rep. 8, 10432–10432 (2018).
Article ADS PubMed PubMed Central Google Scholar
Heaton, N. S., Sachs, D., Chen, C. J., Hai, R. & Palese, P. Genome-wide mutagenesis of influenza virus reveals unique plasticity of the hemagglutinin and NS1 proteins. Proc. Natl Acad. Sci. USA 110, 20248–20253 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Doud, M. B., Lee, J. M. & Bloom, J. D. How single mutations affect viral escape from broad and narrow antibodies to H1 influenza hemagglutinin. Nat. Commun. 9, 1386 (2018).
Article ADS PubMed PubMed Central Google Scholar
Gerdil, C. The annual production cycle for influenza vaccine. Vaccine 21, 1776–1779 (2003).
Article PubMed Google Scholar
Corti, D. et al. Tackling influenza with broadly neutralizing antibodies. Curr. Opin. Virol. 24, 60–69 (2017).
Article CAS PubMed PubMed Central Google Scholar
Whittle, J. R. R. et al. Broadly neutralizing human antibody that recognizes the receptor-binding pocket of influenza virus hemagglutinin. Proc. Natl Acad. Sci. USA 108, 14216–14221 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Lee, P. S. et al. Heterosubtypic antibody recognition of the influenza virus hemagglutinin receptor binding site enhanced by avidity. Proc. Natl Acad. Sci. USA 109, 17040–17045 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Ekiert, D. C. et al. Cross-neutralization of influenza A viruses mediated by a single antibody loop. Nature 489, 526–532 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Ekiert, D. C. et al. Antibody recognition of a highly conserved influenza virus epitope. Science 324, 246–251 (2009).
Article ADS CAS PubMed PubMed Central Google Scholar
Corti, D. et al. A neutralizing antibody selected from plasma cells that binds to group 1 and group 2 influenza A hemagglutinins. Science 333, 850–856 (2011).
Article ADS CAS PubMed Google Scholar
Guthmiller, J. J. et al. Broadly neutralizing antibodies target a haemagglutinin anchor epitope. Nature 602, 314–320 (2022).
Article ADS CAS PubMed Google Scholar
Braathen, R. et al. A DNA vaccine that encodes an antigen-presenting cell-specific heterodimeric protein protects against cancer and influenza. Mol. Ther. Methods Clin. Dev. 17, 378–392 (2020).
Article CAS PubMed PubMed Central Google Scholar
Hinke, D. M. et al. Antigen bivalency of antigen-presenting cell-targeted vaccines increases B cell responses. Cell Rep. 39, 110901 (2022).
Article CAS PubMed Google Scholar
Busch, R., Pashine, A., Garcia, K. C. & Mellins, E. D. Stabilization of soluble, low-affinity HLA-DM/HLA-DR1 complexes by leucine zippers. J. Immunol. Methods 263, 111–121 (2002).
Article CAS PubMed Google Scholar
Grødeland, G., Mjaaland, S., Roux, K. H., Fredriksen, A. B. & Bogen, B. DNA vaccine that targets hemagglutinin to MHC class II molecules rapidly induces antibody-mediated protection against influenza. J. Immunol. 191, 3221–3231 (2013).
Article PubMed PubMed Central Google Scholar
Grødeland, G., Mjaaland, S., Tunheim, G., Fredriksen, A. B. & Bogen, B. The specificity of targeted vaccines for APC surface molecules influences the immune response phenotype. PLoS ONE 8, e80008 (2013).
Article ADS PubMed PubMed Central Google Scholar
Fredriksen, A. B., Sandlie, I. & Bogen, B. DNA vaccines increase immunogenicity of idiotypic tumor antigen by targeting novel fusion proteins to antigen-presenting cells. Mol. Ther. 13, 776–785 (2006).
Article CAS PubMed Google Scholar
Andersen, T. K. et al. Enhanced germinal center reaction by targeting vaccine antigen to major histocompatibility complex class II molecules. NPJ Vaccines 4, 9 (2019).
Article PubMed PubMed Central Google Scholar
Braathen, R. et al. The magnitude and IgG subclass of antibodies elicited by targeted DNA vaccines are influenced by specificity for APC surface molecules. ImmunoHorizons 2, 38–53 (2018).
Article CAS PubMed Google Scholar
Wilson, I. A., Skehel, J. J. & Wiley, D. C. Structure of the haemagglutinin membrane glycoprotein of influenza virus at 3 Å resolution. Nature 289, 366–373 (1981).
Article ADS CAS PubMed Google Scholar
Gerhard, W., Yewdell, J., Frankel, M. E. & Webster, R. Antigenic structure of influenza virus haemagglutinin defined by hybridoma antibodies. Nature 290, 713–717 (1981).
Article ADS CAS PubMed Google Scholar
Caton, A. J., Brownlee, G. G., Yewdell, J. W. & Gerhard, W. The antigenic structure of the influenza virus A/PR/8/34 hemagglutinin (H1 subtype). Cell 32, 417–427 (1982).
Article Google Scholar
Angeletti, D. et al. Defining B cell immunodominance to viruses. Nat. Immunol. 18, 456–463 (2017).
Article CAS PubMed PubMed Central Google Scholar
Das, SumanR. et al. Defining Influenza A Virus Hemagglutinin Antigenic Drift by Sequential Monoclonal Antibody Selection. Cell Host Microbe 13, 314–323 (2013).
Article CAS PubMed PubMed Central Google Scholar
Tamura, M., Kuwano, K., Kurane, I. & Ennis, F. A. Definition of amino acid residues on the epitope responsible for recognition by influenza A virus H1-specific, H2-specific, and H1- and H2-cross-reactive murine cytotoxic T-lymphocyte clones. J. virol. 72, 9404–9406 (1998).
Article CAS PubMed PubMed Central Google Scholar
Jegaskanda, S. et al. Cross-reactive influenza-specific antibody-dependent cellular cytotoxicity antibodies in the absence of neutralizing antibodies. J. Immunol. 190, 1837–1848 (2013).
Article CAS PubMed Google Scholar
DiLillo, D. J., Palese, P., Wilson, P. C. & Ravetch, J. V. Broadly neutralizing anti-influenza antibodies require Fc receptor engagement for in vivo protection. J. Clin. Invest. 126, 605–610 (2016).
Article PubMed PubMed Central Google Scholar
DiLillo, D. J., Tan, G. S., Palese, P. & Ravetch, J. V. Broadly neutralizing hemagglutinin stalk–specific antibodies require FcγR interactions for protection against influenza virus in vivo. Nat. Med. 20, 143–151 (2014).
Article CAS PubMed PubMed Central Google Scholar
Fossum, E. et al. Vaccine molecules targeting Xcr1 on cross-presenting DCs induce protective CD8+ T-cell responses against influenza virus. Eur. J. Immunol. 45, 624–635 (2015).
Article CAS PubMed Google Scholar
Anderson, A. M., Baranowska-Hustad, M., Braathen, R., Grødeland, G. & Bogen, B. Simultaneous targeting of multiple hemagglutinins to APCs for induction of broad immunity against influenza. J. Immunol. 200, 2051–2066 (2018).
Article Google Scholar
Kanekiyo, M. et al. Mosaic nanoparticle display of diverse influenza virus hemagglutinins elicits broad B cell responses. Nat. Immunol. 20, 362–372 (2019).
Article CAS PubMed PubMed Central Google Scholar
Schwartzman, L. M. et al. An intranasal virus-like particle vaccine broadly protects mice from multiple subtypes of Influenza A virus. MBio 6, e01044 (2015).
Article CAS PubMed PubMed Central Google Scholar
Krammer, F., Pica, N., Hai, R., Margine, I. & Palese, P. Chimeric hemagglutinin influenza virus vaccine constructs elicit broadly protective stalk-specific antibodies. J. Virol. 87, 6542–6550 (2013).
Article CAS PubMed PubMed Central Google Scholar
Carter, Donald et al. Design and characterization of a computationally optimized broadly reactive hemagglutinin vaccine for H1N1 influenza viruses. J. Virol. 90, 4720–4734 (2016).
Article CAS PubMed PubMed Central Google Scholar
Chaudhury, S., Reifman, J. & Wallqvist, A. Simulation of B cell affinity maturation explains enhanced antibody cross-reactivity induced by the polyvalent malaria vaccine AMA1. J. Immunol. 193, 2073 (2014).
Article CAS PubMed PubMed Central Google Scholar
Dutta, S. et al. Overcoming antigenic diversity by enhancing the immunogenicity of conserved epitopes on the malaria vaccine candidate apical membrane Antigen-1. PLOS Pathog. 9, e1003840 (2013).
Article PubMed PubMed Central Google Scholar
Turner, L., Theander, T. G. & Lavstsen, T. Immunization with recombinant Plasmodium falciparum Erythrocyte membrane protein 1 CIDRα1 domains induces domain subtype inhibitory antibodies. Infect. Immun. 86, e00435–00418 (2018).
Shaffer, J. S., Moore, P. L., Kardar, M. & Chakraborty, A. K. Optimal immunization cocktails can promote induction of broadly neutralizing Abs against highly mutable pathogens. Proc. Natl Acad. Sci. USA 113, E7039–E7048 (2016).
Article ADS CAS PubMed PubMed Central Google Scholar
Cohen, A. A. et al. Mosaic RBD nanoparticles protect against challenge by diverse sarbecoviruses in animal models. Science 377, eabq0839 (2022).
Article CAS PubMed Google Scholar
Arevalo, C. P. et al. A multivalent nucleoside-modified mRNA vaccine against all known influenza virus subtypes. Science 378, 899–904 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Hillemanns, P. et al. A therapeutic antigen-presenting cell-targeting DNA vaccine VB10.16 in HPV16-positive high-grade cervical intraepithelial neoplasia: results from a phase I/IIa trial. Clin. Cancer Res. 28, 4885–4892 (2022).
Article CAS PubMed Google Scholar
Davenport, F. M., Hennessy, A. V. & Francis, T. Epidemiologic and immunologic significance of age distribution of antibody to antigenic variants of influenza virus. J. Exp. Med. 98, 641–656 (1953).
Article CAS PubMed PubMed Central Google Scholar
Fonville, J. M. et al. Antibody landscapes after influenza virus infection or vaccination. Science 346, 996–1000 (2014).
Article ADS CAS PubMed PubMed Central Google Scholar
Miller, M. S. et al. Neutralizing antibodies against previously encountered influenza virus strains increase over time: a longitudinal analysis. Sci. Transl. Med. 5, 198ra107 (2013).
Article PubMed PubMed Central Google Scholar
Gostic, K. M., Ambrose, M., Worobey, M. & Lloyd-Smith, J. O. Potent protection against H5N1 and H7N9 influenza via childhood hemagglutinin imprinting. Science 354, 722–726 (2016).
Article ADS CAS PubMed PubMed Central Google Scholar
Chojnacki, S., Cowley, A., Lee, J., Foix, A. & Lopez, R. Programmatic access to bioinformatics tools from EMBL-EBI update: 2017. Nucleic Acids Res. 45, W550–W553 (2017).
Article CAS PubMed PubMed Central Google Scholar
Dereeper, A. et al. Phylogeny.fr: robust phylogenetic analysis for the non-specialist. Nucleic Acids Res. 36, W465–W469 (2008).
Article CAS PubMed PubMed Central Google Scholar
Altschul, S. F. et al. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402 (1997).
Article CAS PubMed PubMed Central Google Scholar
Norderhaug, L., Olafsen, T., Michaelsen, T. E. & Sandlie, I. Versatile vectors for transient and stable expression of recombinant antibody molecules in mammalian cells. J. Immunol. Methods 204, 77–87 (1997).
Article CAS PubMed Google Scholar
Andersen, T. K., Zhou, F., Cox, R., Bogen, B. & Grødeland, G. A DNA vaccine that targets hemagglutinin to antigen presenting cells protects mice against H7 influenza. J. Virol. 91, e01340–01317 (2017).
Article CAS PubMed PubMed Central Google Scholar
Chang, H. C. et al. A general method for facilitating heterodimeric pairing between two proteins: application to expression of alpha and beta T-cell receptor extracellular segments. Proc. Natl Acad. Sci. USA 91, 11408–11412 (1994).
Article ADS CAS PubMed PubMed Central Google Scholar
Staudt, L. & Gerhard, W. Generation of antibody diversity in the immune response of BALB/c mice to influenza virus hemagglutinin. I. Significant variation in repertoire expression between individual mice. J. Exp. Med. 157, 687–704 (1983).
Article CAS PubMed PubMed Central Google Scholar
Reiersen, H. et al. Covalent antibody display-an in vitro antibody-DNA library selection system. Nucleic Acids Res. 33, e10–e10 (2005).
Article PubMed PubMed Central Google Scholar
Grødeland, G. et al. Induction of cross-reactive and protective antibody responses after DNA vaccination with MHCII-targeted stem domain from influenza hemagglutinin. Front. Immunol. 11, 431 (2020).
Article PubMed PubMed Central Google Scholar
Reed, L. J. & Muench, H. A simple method of estimating fifty per cent endpoints. Am. J. Hyg. 27, 493–497 (1938).
Google Scholar

Download references

Acknowledgements

The authors are grateful to Hilde Omholt, Marianne Øygaarden, Elisabeth Vikse, and Peter Hofgaard for their expert technical assistance. This work was funded by grants from the K.G. Jebsen Center for Influenza Vaccine Research, University of Oslo (to B.B.), a grant from The Research Council of Norway, project 300049 (to B.B.), South-Eastern Norway Regional Health Authority, Norway, projects 2019127, and 2020046 and innovation grants in 2020 and 2021 (to R.B.), and the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No 801133 (to M.F.L.).

Author information

These authors jointly supervised this work: Bjarne Bogen, Ranveig Braathen.

Authors and Affiliations

K.G. Jebsen Centre for Influenza Vaccine Research, University of Oslo, Oslo, Norway
Daniëla Maria Hinke, Ane Marie Anderson, Gunnveig Grødeland, Bjarne Bogen & Ranveig Braathen
Institute of Immunology (IMM), University of Oslo and Oslo University Hospital, Oslo, Norway
Daniëla Maria Hinke, Ane Marie Anderson, Kirankumar Katta, Marlene Fyrstenberg Laursen, Demo Yemane Tesfaye, Ina Charlotta Werninghaus, Gunnveig Grødeland, Bjarne Bogen & Ranveig Braathen
Department of Microbiology and Immunology, Institute of Biomedicine, University of Gothenburg, Gothenburg, Sweden
Davide Angeletti

Authors

Daniëla Maria Hinke
View author publications
You can also search for this author in PubMed Google Scholar
Ane Marie Anderson
View author publications
You can also search for this author in PubMed Google Scholar
Kirankumar Katta
View author publications
You can also search for this author in PubMed Google Scholar
Marlene Fyrstenberg Laursen
View author publications
You can also search for this author in PubMed Google Scholar
Demo Yemane Tesfaye
View author publications
You can also search for this author in PubMed Google Scholar
Ina Charlotta Werninghaus
View author publications
You can also search for this author in PubMed Google Scholar
Davide Angeletti
View author publications
You can also search for this author in PubMed Google Scholar
Gunnveig Grødeland
View author publications
You can also search for this author in PubMed Google Scholar
Bjarne Bogen
View author publications
You can also search for this author in PubMed Google Scholar
Ranveig Braathen
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

R.B. and B.B. developed the hypothesis of valency-based immuno-selection. D.M.H., A.M.A., G.G., R.B. and B.B. designed the study and experiments. D.M.H., A.M.A., K.K., M.F.L., D.Y.T., I.C.W. and G.G. performed experiments and analyzed the results. D.A. provided reagents and analyzed the results. D.M.H., R.B. and B.B. wrote the manuscript. The manuscript was reviewed by all authors before submission.

Corresponding authors

Correspondence to Bjarne Bogen or Ranveig Braathen.

Ethics declarations

Competing interests

The authors declare the following competing interest: the research funding did not have any role in the conceptualization, design, data collection, analysis, decision to publish, or preparation of the manuscript. B.B. holds shares in Nykode A.S. The TTO office of the University of Oslo and Oslo University Hospital, Inven2, has filed a patent application, WO 2019048928 National phase, “Vaccine molecules”, related to the valency-based immune-selection strategy described in the manuscript. R.B. and B.B. are inventors of this patent. The remaining authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Reporting Summary

Source data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Hinke, D.M., Anderson, A.M., Katta, K. et al. Applying valency-based immuno-selection to generate broadly cross-reactive antibodies against influenza hemagglutinins. Nat Commun 15, 850 (2024). https://doi.org/10.1038/s41467-024-44889-w

Download citation

Received: 26 October 2022
Accepted: 09 January 2024
Published: 12 February 2024
DOI: https://doi.org/10.1038/s41467-024-44889-w

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Subjects

Abstract

Similar content being viewed by others

Multivalent next generation influenza virus vaccines protect against seasonal and pre-pandemic viruses

Broadly neutralizing antibodies target a haemagglutinin anchor epitope

A chimeric thermostable M2e and H3 stalk-based universal influenza A virus vaccine

Introduction

Results

APC-targeted heterodimeric proteins with different HAs delivered as DNA plasmids

MHCII-targeted heterodimeric DNA vaccines that bivalently express two identical HA (H1-18) induce subtype-specific antibody responses

Different HA subtypes can be combined in one heterodimeric mix vaccine

A heterodimeric HA mix vaccine induces antibody responses against HAs included in the mix

Vaccination with DNA plasmid and protein mixes induces antibodies against HA subtypes not included in the mix

Mix plasmid vaccines induce antibodies that cross-react between closely related HA

Plasmid mix vaccines induce antibodies reacting to head and stems of HAs not included in the vaccine

Plasmid mix vaccines induce antibodies skewed away from immunodominant antigenic sites

Plasmid mix vaccine induces HA-specific T-cell responses

Cross-reactive antibodies mediate protection against challenge with an HA heterologous influenza virus

Discussion

Methods

Mice, cell lines, and viruses

Generation of phylogenetic tree and multiple sequence alignment of HA proteins

Cloning of heterodimeric vaccine vectors

Transient transfection of HEK293E cells

Protein purification

ELISA to detect vaccine proteins

Western blot analysis

DNA vaccination

Protein vaccination

ELISA for HA-reactive antibodies in serum

ELISA for detection of HA stem-specific antibodies

Microneutralization assay

Analysis of antibody-mediated NK cell activation

Inhibition analysis in ELISA

Purification of H1-reactive antibodies from mouse serum

T cell ELISPOT

Viral challenge with influenza virus

T cell depletion

Statistical analysis

Reporting summary

Data availability

References

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding authors

Ethics declarations

Competing interests

Peer review

Peer review information

Additional information

Supplementary information

Supplementary Information

Reporting Summary

Source data

Source data

Rights and permissions

About this article

Cite this article

Share this article

Comments

Search

Quick links