Introduction

The very recent reports (January, March 2025) of foot-and-mouth disease (FMD) episodes in Germany1 (last outbreak, 1988), Hungary and Slovakia2 (last outbreaks, 1970s), are a salutary reminder that even highly-surveilled regions of the world are vulnerable to the reemergence of this worrying disease. A highly contagious viral infection affecting cloven-hoofed animals like cattle, sheep, and pigs, FMD remains a persistent concern for the livestock industry3,4. Infection quickly disseminates among susceptible animals, often culminating in widespread epidemics causing illness, discomfort, and serious productivity losses. Classified by the World Organization for Animal Health (WOAH) as a reportable disease, FMD prompts trade restrictions by FMD-free countries on animals and products from endemic areas, with severe economic and social repercussions5,6. In regions where FMD is prevalent, such as Asia and Africa, vaccination remains the primary control method, mainly by inactivated virus vaccines7,8. Although such classic vaccines have demonstrated efficacy in disease control and prevention9, limitations such as the risk of live virus escape during production10, plus the broad antigenic variability of FMDV, with seven different serotypes11, each subdivided into genetically and geographically distinct topotypes, complicate control efforts12,13,14 and call for alternative approaches.

In this regard, subunit vaccines, by definition free of any infectious agent, are currently viewed as smarter, safer alternatives to virus-based formulations. Within the subunit class, vaccines based on the multiple antigenic peptide concept pioneered by Tam (1988)15 are an increasingly attractive choice, with advantages like unquestioned safety, antigenic versatility, easy production and handling, and especially, marker features allowing distinction between infected and vaccinated animals (DIVA concept)16,17. In such peptide-based formulations, selected antigenic motifs can be purposefully integrated into higher-order assemblies eliciting both B- and T-cell immune responses18,19, conferring protection, adaptability, and multivalency20,21.

Previously, we developed a vaccine prototype named B₂T, incorporating two copies of a conserved B-cell epitope from VP1 (residues 140–158) and one copy of a T-cell epitope from nonstructural protein 3A (residues 21–35), both derived from type-O FMDV. These epitopes were tethered onto a well-defined dendrimeric scaffold using chemoselective ligation methods22. A single dose of B2T, at either 2 mg or as little as 0.5 mg, sufficed to induce a rapid and robust immune response. By 15 days post-immunization (dpi), animals had developed high levels of FMDV-neutralizing antibodies and IFN-γ-secreting T cells. Moreover, 80% of the B2T-vaccinated animals were fully protected, exhibiting no FMD signs following viral challenge at 25 dpi23,24.

Importantly, this dendrimeric platform has also been successfully adapted to harbor epitopes from other FMDV serotypes (e.g., C-S8) tailored for pig and cattle, as well as from other pathogens such as classical swine fever virus, demonstrating its versatility and potential for broader vaccine applications25,26,27. Also, given the relevance of swine major histocompatibility complex (SLA) polymorphism in immune response and vaccine design, SLA class II typing was performed in B2T-immunized animals, revealing associations between specific low-resolution haplotypes (Lr-Hp) and the magnitude of T cell responses, and to a lesser extent, neutralizing antibody levels28.

Building upon these results, we have designed a new construct, B₂T-TB₂, generated by fusing two B₂T units via click chemistries (thiol-ene addition and copper(I)-catalyzed azide-alkyne cycloaddition, CuAAC), resulting in a dendrimer containing four B-cell and two T-cell epitope copies29. The aim was to enhance immunogenicity through increased valency and epitope density while retaining the molecular definition and safety of the original platform. Importantly, the increased potency allows for protection at reduced antigen doses, an essential feature for emergency use and scalable field application. Moreover, the modular design of B₂T-TB₂ enables rapid adaptation to emerging FMDV variants, as epitopes can be interchanged or updated without the lengthy production cycles required for conventional inactivated vaccines. This flexibility, combined with its potent and durable immunogenicity persisting for up to four months after immunization, positions B₂T-TB₂ as a promising next-generation FMD vaccine platform30.

Moving on from these findings, we now evaluate for the first time the efficacy of B2T-TB2 in a challenge model in pigs, one of the two main FMDV hosts, as a crucial step toward validating the platform’s potential for real-world application. Our in-depth evaluation shows full clinical protection, broad intraserotype cross-neutralization, and early induction of neutralizing antibodies and IFN-γ-secreting T-cells following a single low-dose of B₂T-TB₂, all of them key features supporting a possible role in emergency FMD vaccination strategies. In addition, SLA typing has shown all analyzed pigs to have homogeneous haplotypes, among the most common in European farmed pigs.

Results

Design and synthesis of the multiepitopic B2T-TB2 dendrimeric vaccine platform using click chemistry

In a quest for chemically well-defined peptide vaccines that display several immunorelevant epitopes in multivalent fashion, we have designed the dendrimeric B2T-TB2 platform. An outline of the synthetic route to this construct, where several peptide moieties are assembled by chemoselective conjugation techniques, is shown in Fig. 1.

Fig. 1: Click chemistry synthesis of B2T-TB2.
figure 1

General scheme of the chemoselective click approach used for the synthesis of a B2T-TB2 multiepitopic peptide platform.

Full size image

The main components of B2T-TB2 were peptide building blocks embodying B- and T-cell epitopes [VP1(140–158), 3A(21–35), light green and light blue respectively, Fig. 1] of FMDV isolate O/UK/11/2001. They were readily produced by Fmoc solid phase peptide synthesis (SPPS), with site-specific extra residues bearing functional groups enabling chemoselective Lego-like buildup of B2T-TB2 by two distinct click chemistries. Thus, the T-cell moiety (pale blue) incorporated a Lys residue providing a branching point (dark blue), with both Nα and Nε amino groups derivatized as maleimide (orange). For its part, the B-cell epitope (light green) had an extra C-terminal Cys residue (green) for subsequent thiol−ene addition. Likewise, azide- or alkyne-functionalized residues (purple and red, respectively), enabling click CuAAC, were integrated into the T-cell peptide sequence as required. Finally, Lys-Lys dipeptide motifs were placed at both ends of the T-epitope sequence (wavy lines within light blue segment) to serve as a cleavage site for cathepsin D, a protease involved in in vivo antigen processing for MHC class II presentation31.

With the above constituents in hand, Michael type thiol−ene click chemistry was applied to make the first two selectively functionalized units of B2T, itself an effective vaccine prototype22,23,24. Next, the two B2T variants were fused tail-to-tail by click CuAAC to give the B2T-TB2 dimeric end product (Fig. 1), where tetra- and bivalent display of B- and T-cell epitopes, is respectively achieved. Further details, including solutions to some often-ignored pitfalls found on assembling constructs of such complexity and size, can be found in (29).

Single-dose B2T-TB2 vaccination confers complete clinical protection against FMDV challenge in swine

The B2T-TB2 vaccine described above has been reported to elicit robust B- and T-cell-specific responses in swine, the natural FMDV host, at reduced doses30. Animals immunized with B2T-TB2 show a consistent trend to antibody titers equal or higher than those elicited with the B2T prototype, which is known to provide full, long-lasting protection against viral challenge23,24. Here we compare the protective responses to FMDV challenge afforded by a low dose (0.5 mg) of either B2T-TB2 or its monomer version B2T, including clinical signs and other immune parameters associated to protection.

All immunological and clinical data presented in this study were obtained from the same vaccination and challenge experiment, which included both vaccinated and control animals. The results of protection experiments are summarized in Table 1. Domestic pigs, in two different groups, were immunized once at day 0 with B2T or B2T-TB2, or non-immunized (PBS+adjuvant), and subsequently challenged with FMDV O-UK/11/2001 at 25 dpi. Animals were examined daily for clinical signs (see “Methods”) and judged protected if lesions were observed only at the infection site and/or at a single site32.

Table 1 Evidence for protection in animals immunized with B2T-TB2 and B2T vaccine candidates
Full size table

As expected, non-immunized (PBS-inoculated) pigs #11 and #12 showed typical clinical FMD signs at 4–5 dpc, with vesicular lesions on the snout and all four feet. In contrast, peptide-immunized pigs (#1 to #10) showed no clinical FMD signs after challenge with O-UKG/11/2001 (100% protection). Pig #8 died spontaneously at 25 dpi, prior to challenge, for causes unrelated to the experiment, hence no protection and viremia data are reported for this animal. In general, a correlation between low body temperature and protection was observed. Thus, animals protected by B2T or B2T-TB2 immunization did not develop fever (<39 °C), except for B2T-immunized pig #9, which reached 39.8 °C (at 7 dpc only). Furthermore, RT-qPCR assay after challenge showed the presence of FMDV RNA only in serum and nasal samples of non-immunized controls #11 and #12. Among immunized pigs, viral RNA was only detected at very low level (Ct 39.54) in a nasal swab from pig #9 at 6 dpc, the same animal testing negative at 11 dpc. In all cases, FMDV detection in challenged pigs correlated with severity of lesions. In conclusion, immunization with a single 0.5 mg dose of either B2T or B2T-TB2 afforded consistent protection against FMDV challenge that prevented virus shedding.

B2T-TB2 induces significantly higher and earlier neutralizing antibody responses than B2T at equal dose

In the same experiment, the humoral response was analyzed at different time points in the animals that were immunized and subsequently challenged with the virus. Antibody titers serve as key indicators of vaccine effectiveness, given their reliable correlation with protection rates33. In the present work, swine FMDV-specific antibodies in sera from 15 and 25 dpi, plus 10 dpc, were tested by ELISA. As shown in Fig. 2A, all pigs given a 0.5 mg single dose developed consistent, comparable IgG titers already at 15 dpi [3.5 ± 0.5 and 3.5 ± 0.4 for B2T-TB2 and B2T, respectively] and showed concomitant titer increases at 25 dpi [4.1 ± 0.4 and 4.2 ± 0.3 for B2T-TB2 and B2T, respectively]. After challenge, anti-FMDV antibodies remained stable in both vaccinated groups up to 36 dpi (11 dpc), with no clear-cut differences among the two constructs.

Fig. 2: Antibody response elicited by dendrimer peptides.
figure 2

Two groups of 5 animals each were immunized on day 0 with B2T-TB2 (orange circles) or B2T (green circles) and challenged at 25 dpi (down-pointing arrows). A ELISA antibody response in sera from 15, 25, and 36 dpi. B Neutralizing antibody responses at 15, 25, and 36 dpi, determined as detailed in “Methods”. Dotted line indicates the detection limit. Numbers next to each circle indicate pig number. Horizontal lines indicate the geometric mean for each animal group. No spontaneous reactivity was observed in titers at day 0. Statistically significant differences are indicated by asterisks (*) for p < 0.05.

Full size image

Neutralizing antibody titers were determined by virus neutralization tests (VNT) performed in 96-well plates following the standard microneutralization procedure recommended by the WOAH. In contrast to ELISA results, statistically significant differences in VNT titers were found between the groups immunized with B2T and B2T-TB2 at 15 dpi (1.2 ± 0.2 vs. 1.8 ± 0.5, respectively) and 25 dpi (1.8 ± 0.3 vs. 2.3 ± 0.5) (Fig. 2B). Post-challenge titers, on the other hand, increased at 36 dpi in both groups, with higher values in pigs given the B2T-TB2 construct. As expected, no neutralizing antibodies were detected in the two control animals before challenge (not shown). It is worth noting that immunization with the same amount of B-cell epitope [VP1(140–158)] peptide led to significantly heightened levels of FMDV-neutralizing antibodies in the B2T-TB2 homodimer vs. the B2T monomer, using the same 0.5 mg dose in all cases. Although it is difficult to establish a VNT cut-off that corresponds to protection, previous studies indicated that animals with VNTs ≥1.65 have more probabilities to be protected34. In this study, all B2T-TB2-immunized pigs elicited VNTs above this threshold at 25 dpi, most of them (4 out of 5) even at 15 dpi. Such early production of neutralizing antibodies in significant levels by B2T-TB2 suggests potential differences in the T-cell help evoked by the homodimeric vs. the monomeric B2T version. On the other hand, physical features of antigens, rather than inherent immunogenic properties of their epitopes, are known to account for the activation of non-T-helper signals, e.g., TLRs on B cells and APCs, essential for triggering the humoral response35,36. Thus, one cannot exclude the possibility that the differing VNTs of monomeric vs. dimeric peptides are due to distinct physical properties. In any event, the results confirm that a reduced 0.5 mg dose of B2T-TB2 homodimer induces higher levels of neutralizing antibodies faster than the previous B2T vaccine.

B2T-TB2 elicits significantly higher intraserotype cross-neutralizing antibody responses against FMDV Type O isolates than B2T

Immunization against a FMDV serotype does not warrant heterotypic/interserotype protection. However, recent data show that inter- or intraserotype protection can be induced by sequential challenge with multiple FMDV serotypes or topotypes37,38. In this study, we tested B2T and B2T-TB2 suitability as broad-spectrum vaccines against FMDV type O, prevalent in Asia, Middle East, and Africa. To this end, we determined cross-reactivity towards several (n = 6) FMDV isolates, representing two of the most predominant type O topotypes, in VN assays of antisera at 25 dpi from pigs immunized with each of the constructs. Results showed positive and highly specific cross-neutralizing responses for both B2T and B2T-TB2, and no cross–reactivity against FMDV serotype C, indicating that both vaccines mimic the type O neutralizing antigenic site at the VP1 GH loop in an efficient manner (Fig. 3). Furthermore, as observed in the VN assays against homologous virus, VNTs for other O isolates were also significantly higher in animals immunized with the B2T-TB2 homodimer than with the B2T monomer. This underscores the robustness of the B2T-TB2 vaccine, which elicits higher levels of neutralizing antibodies with an equal amount of B-epitope peptide than the B2T version. Since both B2T and B2T-TB2 embody O/UK/11/2001 virus-specific B- and T-cell epitopes, the cross-neutralizations observed suggest that, despite sequence variability among type O isolates at the VP1(140–158) antigenic site on the FMDV capsid39,40 (Fig. 3), the respective amino acid changes are conservative enough in structural terms (e.g., charge and/or hydrophobicity) to have only minor impact on B-cell epitope recognition, in contrast with the lack of cross-neutralization with serotype C-S8c1, whose cognate VP1(136–154) antigenic site differs widely from those of O isolates (Fig. 3).

Fig. 3: Cross-neutralizing responses and antigenic differences in FMDV.
figure 3

Top panel shows cross-neutralizing activity against O- and C-type FMDV isolates in sera at 25 dpi. VNTs represent the geometric mean for each animal group immunized with the indicated peptide. Dotted line indicates detection limit. No reactivity observed in VNTs at day 0. Statistically significant differences are indicated by asterisks (*) for p < 0.05, (**) for p < 0.005 and (***) for p < 0.0005. Bottom panel depicts differences (highlighted blue) in B-cell antigenic site residues [VP1 (140–158) for O-type; VP1 (136–154) for C-type C-S8c1] among tested FMDV isolates and the standard O/UK/11/2001 used for challenge.

Full size image

The near-equivalence among O isolates, however, is not wide-ranging, as shown by the fact that both B2T and B2T-TB2 antisera were less cross-reactive with isolate O/IRN/8/2005, of the same ME-SA topotype but with significantly lower VNTs. It would therefore appear that changes at single positions within the B-cell epitope, slightly altering antigen recognition and immune response, are tolerated only to some extent41. In particular, it remains to be understood how antigenic recognition of the Val → Thr replacement at VP1 position 141 of FMDV O/IRN/8/2005 (Fig. 3) bears upon the substantially altered neutralizing response. Insights into this point may be helpful towards developing even more broadly protective vaccines42.

B2T-TB2 induces more consistent FMDV-specific IFN-γ-T-cell responses than B2T

Specific T-cell responses elicited by B2T-TB2 and B2T at 15 and 25 dpi were examined by ELISPOT analysis of IFNγ-producing peripheral blood mononuclear cells (PBMCs). High frequencies of IFN-γ-producing cells were found in both groups in response to in vitro recall with the homologous peptide at 15 dpi [1498.3 ± 124.6 and 1266.3 ± 466.7 for B2T-TB2 and B2T, respectively] (Fig. 4A). These responses significantly increased at 25 dpi [2557.3 ± 149.8 and 2085.3 ± 500.8 for B2T-TB2 and B2T, respectively] (Fig. 4B).

Fig. 4: Specific T-cell responses to B2T-TB2 and related peptides.
figure 4

Specific T-cell responses measured by an ex vivo IFN-γ ELISPOT at 15 (A) and 25 dpi (B). IFN-γ release by PBMCs stimulated in vitro with either B2T-TB2, B2T, or with single T or B peptide epitopes. The frequency of FMDV-specific IFN-γ secreting cells was determined as detailed in Methods. The box-and-whisker plot shows each individual value as a dot, from the minimum up to the maximum value. The box extends from 25 to 75th percentiles where the dotted line represents the median. The numbers next to each symbol indicate the pig number, as in Table 1. Statistically significant differences are indicated by asterisks (*) for p < 0.05.

Full size image

Specifically, all pigs immunized with B2T-TB2 showed homogeneous frequencies of IFN-expressing PBMCs, which on average were slightly higher than those obtained with the B2T construct, differences being statistically significant from 15 dpi. All IFN-γ responses were specific, as no peptide-driven IFN-γ-producing cells were detected in non-immunized pigs (not shown). In addition, PBMC stimulation with the T3A (21–35) peptide in both dendrimer constructs paralleled B2T results at 15 (1183.0 ± 449.9 vs. 1266.3 ± 466.7) and 25 dpi (2034.4 ± 612.7 vs. 2085.3 ± 500.8), supporting the recognition of such a motif as a T-cell epitope. Interestingly, for B2T-TB2, IFN-γ-release by PBMC stimulated with the T3A peptide occurred to a lesser extent at 15 pi (931.9 ± 382.1 vs. 1498.3 ± 124.6) and 25 dpi (1572.8 ± 585.7 vs. 2557.3 ± 149.8); possibly because the sequence becomes less accessible to processing and presentation within the structurally more complex construct, resulting in immune lymphocytes that do not recognize T3A in vitro as efficiently as those from B2T pigs. It is also noteworthy that data dispersion in the B2T-TB2 group at both 15 and 25 dpi is much lower than that for B2T pigs; this clustering again underscores the superior robustness of the cellular response induced by B2T-TB2.

Both B2T and B2T-TB2 vaccines induce multifunctional CD4⁺ T-cells expressing IFN-γ and TNF-α

Studies on multifunctional T-cells have highlighted their crucial role in protective immune responses43. Although IFN-γ is the key effector cytokine produced by Th1 cells, other pro-inflammatory cytokines also play important roles in Th1-mediated immunity, contributing to a comprehensive and effective T-cell response. To evaluate whether CD4+ T-cells activated by B2T or B2T-TB2 peptides induce effector cytokines beyond IFN-γ, intracellular cytokine staining of PBMCs at 25 dpi was performed to detect both IFN-γ and TNF-α expression (Fig. 5). The analysis revealed significant induction of IFN-γ- and TNF-α-producing CD4+T-cells in response to stimulation in both B2T (0.43 ± 0.15% for IFN-γ; 0.854 ± 0.35% for TNF-α) and B2T-TB2 (0.32 ± 0.09% for IFN-γ; 0.6 ± 0.23% for TNF-α) groups, with no statistically significant differences. Furthermore, a subset of IFN-γ-secreting CD4+ T-cells also produced TNF-α, reflecting a multifunctional phenotype in both B2T (0.26 ± 0.09%) and B2T-TB2 (0.16 ± 0.06%) groups, again with no significant differences. Nevertheless, a considerable fraction of CD4+ T-cells produced TNF-α alone. Individual variability was evident: the highest and lowest proportions of CD4+ T-cells producing both cytokines were observed in pig #6 (0.64% for IFN-γ and 1.92% for TNF-α) from the B2T group and pig #4 (0.24% for IFN-γ and 0.18% for TNF-α) from the B2T-TB2 group, respectively. In any case, these differences alone cannot be directly associated with the degree of protection.

Fig. 5: CD4⁺ T-cell cytokine responses.
figure 5

IFN-γ- and TNF-α-producing CD4+ T-cell subsets from immunized pigs upon stimulation with homologous peptide, analyzed by intracellular cytokine staining (ICS) at 25 dpi. A Surface CD4 staining and expression was analyzed (representative plot of pig #1 is shown). B CD4+ T-cells were gated and analyzed for production of a single cytokine (IFN-γ or TNF-α). PBS-inoculated animals (#11 and #12) were stimulated with PHA, thus serving as positive controls. The numbers shown within the plots indicate cell percentages relative to the CD4+ T-cell population. C Percentages of CD4+ T-cell populations from each pig, expressing a single cytokine (IFN-γ or TNF-α) or both.

Full size image

B2T-TB2-induced immune responses are consistent across diverse SLA haplotypes in pigs

The high polymorphism of the swine major histocompatibility complex (SLA) plays a crucial role in shaping immune responses to pathogens and vaccines. Despite selective breeding and inbreeding associated with domestication, pigs retain considerable SLA diversity, which is particularly relevant for the design of effective subunit vaccines. These vaccines must include T-cell epitopes that are broadly recognized and presented by SLA alleles commonly found in natural populations, to minimize the risk of poor responders.

In previous studies, we SLA-typed 63 pigs immunized with B2T dendrimeric vaccines at both SLA class I (SLA-1, SLA-3, SLA-2) and class II (DRB1, DQB1, DQA) loci. We then analyzed correlations between specific low-resolution SLA haplotypes (Lr-Hp) and the magnitude of neutralizing antibody and T-cell responses. Our findings revealed a strong association between SLA class II haplotypes and T-cell responses, with some haplotypes, such as Lr-0.24, also linked to reduced antibody production. Building on these findings28, we aimed to expand our understanding by performing the same SLA typing analysis on the animals included in the present study.

Table 2 represents SLA-I and SLA-II allele-groups and Lr Haplotypes from immunized animals. Thirteen allele-groups at 3 SLA class I loci (5 for SLA-1, 4 for SLA-3, and 4 for SLA-2) were identified in the cohort of pigs, comprising 5 haplotypes, while 8 SLA class II allele-groups were found at the 3 loci analyzed (3 for DRB1, 2 for DQB1, and 3 for DQA), giving rise to 4 haplotypes. Lr haplotypes Lr-1.0 (45%), followed by Lr-37.0 (25%) and Lr-22.0 (20%) for SLA class I and Lr-0.15b (75%) for class II, were the most frequently observed (Table 2). Accordingly, a higher frequency of certain allele groups was observed in each SLA class I loci [SLA-1*01XX (45%), SLA-3*01XX (50%) and SLA-2*01XX (45%)] and SLA class II loci [DRB1*04XX (80%), DQB1*02XX (90%) and DQA*02XX (85%)]. All these haplotypes and allele groups are among the most abundant in European farmed pigs, as expected44. Although significant correlations cannot be established due to the limited sample size, it can be stated that there is considerable homogeneity in the humoral and cellular responses, which resulted in 100% protection, regardless of the haplotype. We observed only one example of potential correlation in one animal immunized with B2T-TB2 (#4), which showed significantly lower neutralizing antibody titers and had a SLA II Lr haplotype, Lr-0.24, previously associated with a low humoral response28.

Table 2 SLA-I and SLA-II allele-groups and Low-resolution haplotypes (Lr-Hp) in the B2T-TB2 and B2T-immunized pigs
Full size table

Discussion

Vaccination against FMDV is considered the most efficient method of disease prevention. However, despite extensive efforts over decades to control FMD, success has been limited, largely because bespoke vaccines, tailored for prevalent circulating viruses, are not easily accessible in sufficient amounts and with proper standards of quality, effectiveness and safety. In this scenario, the present study aimed to evaluate a recent alternative, namely a synthetic multivalent modular platform of small-protein size (~120 residues, ~15 kDa), integrating up to six immunorelevant (15–20 residue) peptide epitopes, as a broadly protective vaccine candidate against this most troublesome of animal diseases. In this paper, we have shown that a single low dose (0.5 mg) of B2T-TB2 vaccine is successful in preventing FMDV infection in pigs, inducing innate and adaptive responses that outperform earlier candidates such as our B2T prototype45.

Vaccination with B2T-TB2 specifically achieves two crucial goals for FMD control: expeditious protection from clinical disease, and remarkable reduction of virus shedding by robust neutralizing antibody and virus-specific T-cell responses. All peptide-immunized pigs in this work were protected against FMD and did not develop clinical signs after challenge with FMDV O-UK/11/2001 (100% protection). Interestingly, the same amount of B-cell VP1(140–158) epitope in the B2T-TB2 homodimer induced significantly higher titers of FMDV-neutralizing antibodies than the monomer B2T version, with values that were, at least, similar to those observed in conventionally vaccinated pigs46. Since high levels of neutralizing antibodies are deemed a reliable benchmark of FMD protection47, our data highlight dendrimer peptides, particularly B2T-TB2, as successful vaccine candidates conferring effective immunity at low doses. Furthermore, B2T-TB2 antisera afford broad cross-neutralization among O serotypes, with statistically significant differences relative to B2T. This would suggest that FMD vaccination using specific and/or universal peptide epitopes in the dendrimer fashion of B2T-TB2 is a viable approach to achieve interserotype cross-protection. An additional advantage of our dendrimer approach is its rapid adaptability to emerging variants, achieved by incorporating alternative B-cell epitopes to broaden immune coverage as needed—unlike conventional inactivated vaccines, which require months to reformulate.

Analysis of specific in vitro T-cell responses recalled by B2T-TB2 and B2T peptides at 15 and 25 dpi, with notably reduced data dispersion at both timepoints, suggests again a more robust response by B2T-TB2 than by B2T. The strong T-cell responses observed with our dendrimer vaccines likely result from the focused inclusion of a single, immunodominant T-cell epitope, in contrast to natural infection or inactivated vaccines, which often induce weaker cellular immunity due to viral immune evasion or the presence of immunosuppressive viral proteins. On the other hand, multifunctional CD4+ T lymphocyte populations did not show significant differences between B2T-TB2 and B2T-vaccinated pigs, consistent with the similar protection levels observed.

The swine major histocompatibility complex, SLA, exhibits a high polymorphism, enabling presentation of diverse antigenic peptides and influencing immune responses to disease and vaccines. Despite selective breeding, pigs retain substantial SLA diversity. Given the role of SLA in regulating immune responses, SLA-typed pigs are valuable for vaccine development. In previous studies, we typed a total of 63 pigs immunized with B2T to explore possible correlations between specific low-resolution SLA haplotypes (Lr-Hp) and neutralizing antibody and T-cell responses elicited by these peptides. In particular, the Lr-0.24 haplotype was found to be associated with reduced antibody responses, suggesting that both cellular and humoral immunity is shaped by SLA genotype. These findings provided the rationale for extending SLA typing to the animals in the present study, to assess whether similar response patterns could be identified in this independent cohort28. Typing of the animal groups in the present study shows greater homogeneity, particularly with respect to class II haplotypes, albeit group size does not allow establishing solid correlations. Even so, we observed a potential case of correlation in an animal immunized with B2T-TB2, which exhibited significantly lower neutralizing antibody titers and carried the SLA class II Lr-0.24 haplotype, previously associated with reduced humoral responses. In any event, the levels of both humoral and cellular responses are sufficient for the 100% protection achieved, regardless of haplotype. One may thus conclude that our peptide-based vaccines appear to be effective against the most abundant haplotypes in European farmed pigs. Targeting common haplotypes may reduce diversity over time, leading to susceptibility to other diseases and a lack of vaccine efficacy. Hence, a vaccine that works across a wide range of haplotypes potentially could be a safer strategy44,48.

In conclusion, the B2T-TB2 vaccine, even if more structurally complex than the earlier B2T version29, has multivalency and modularity features that provide distinct advantages: (i) the branched structure enhances immune response for antibody production and full protection49,50; (ii) it performs similar to B2T but with a lower-dose single shot that maximizes coverage and reduces costs; (iii) the modular assembly adds an extra level of customization; (iv) the flexible choice of immunorelevant motifs allows to target a broad epidemiological spectrum, and (v) the versatile design and ready production are well suited to outbreak situations requiring emergency vaccines. Even so, expanding and fine-tuning the protective capability of the B2T-TB2 vaccine will require steady surveillance to detect the emergence of new variants51.

Methods

General peptide synthesis

Linear peptides were assembled in a Prelude synthesizer (Protein Technologies, Inc., Tucson, AZ) running Fmoc SPPS protocols at 0.1 mmol scale on Rink-amide ChemMatrix resin (PCAS Bio-Matrix, Inc., Saint-Jean-sur-Richelieu, Canada) using Fmoc-protected amino acids (Iris Biotech., Marktredwitz, Germany). Side chain functionalities were protected with tert-butyl (Asp, Glu, Ser, Thr, Tyr), tert-butyloxycarbonyl (Lys, Trp), NG-2,2,4,6,7-pentamethyldihydrobenzo-furan-5-sulfonyl (Arg), and trityl (Asn, Gln, His) groups. Eight-fold excess of Fmoc-amino acid and HBTU, in the presence of double molar amount of DIEA, were used for the coupling steps, with DMF as solvent. All peptides were fully deprotected and cleaved from the resin with TFA/H2O/triisopropylsilane (95:2.5:2.5 v/v, 90 min, r.t.), precipitated by addition of chilled diethyl ether, taken up in aqueous AcOH (10% v/v), and lyophilized. Reverse-phase HPLC purification gave homogeneous materials of satisfactory purity (>95%) with the expected mass by ESI-MS in an LCMS-2010 EV mass spectrometer (Shimadzu, Kyoto, Japan) controlled by LabSolutions LCMS software and/or by MALDI-TOF MS in a ABSciex 4800 Plus spectrometer (Applied Biosystems, Foster City, CA), using α-cyano-4-hydroxycinnamic acid as matrix.

Dendrimer peptide synthesis and characterization

Peptides embodying the B- and T-cell epitopes of FMDV O/UK/11/2001 in various branched arrangements are shown in Fig. 1. All were made from precursors prepared by SPPS and purified prior to conjugation as specified above. In the B2T synthesis, branching was achieved by N-terminal elongation of the T epitope sequence with two Lys units plus an additional Lys used as a branching point. For conjugation, the Lys-elongated T epitope sequence was functionalized with two maleimide groups that reacted with two copies of an N-acetylated B epitope peptide with a C-terminal Cys that allowed thiol−ene (click) addition. For the synthesis of B2T-TB2, two B2T units were prepared with a non-native azide- or alkyne-bearing residue at the C-terminus of the T epitope sequence. After cleavage and purification of the azide- and alkyne- modified B2T versions, their tail-to-tail fusion was achieved by CuAAC. Additional details are available in refs. 23,29 and in Supplementary Information (SI), including peptide sequences and analytical data such as HPLC retention times and mass spectra. The final products were purified to near homogeneity (>95%) by HPLC on a C18 column (Luna, 4.6 mm × 50 mm, 3 mm; Phenomenex, Torrance, CA) eluted with a 20–60% linear gradient of solvent B (0.036% TFA in MeCN) into A (0.045% TFA in H2O) over 15 min, and satisfactorily characterized by LC/MS.

Viruses

A FMDV stock derived from isolate O/UK/11/2001 (The Pirbright Institute, UK) by two amplifications in IBRS-2 swine kidney cells was used for swine challenge. The virus was virulent in vivo, as demonstrated by its ability to produce generalized clinical signs in naive pigs23,24. The VP1 nucleotide sequence (see SI) of the FMDV used to challenge pigs in our study differs from the O-UKG/35/2001 reference (GenBank: AJ539141.1) in only an A/C change at nucleotide 589, leading to a Ser/Arg replacement at VP1 position 197, a residue not reported to affect FMDV pathogenicity52. Stocks of O/SKR/1/2002, O1Manisa/TUR/69, O1BFS 1860/UK/67, O/Iran/8/2005 (The Pirbright Institute), and O1Campos (OPS-PanAftosa, Washington, DC) were likewise amplified in IBRS-2 cells. Type C C-S8c1 FMDV53 was amplified in BHK-21 cells.

Animals and experimental design

The immune responses were assessed in domestic Landrace X Large White pigs (Agropardal SL, Cuenca, Spain), free of antibodies against FMDV. Animal handling and experimental procedures were conducted at CISA in accordance with protocols approved by CISA-INIA-CSIC Committee on Ethics of Animal Experiments and Biosafety, as well as by the National Committee on Ethics and Animal Welfare (PROEX 088.1/21). Two-month-old pigs (numbering as in Table 1) were randomly assigned to 2 groups of 5 animals each and immunized with a single intramuscular injection at day 0 with 2 mL of Montanide ISA 50V2 emulsion (Seppic, Colombes, France) containing 0.5 mg of either B2T-TB2 (pigs #1 to #5) or B2T (pigs #8 to #10). Two additional pigs (#11 and #12), housed in separate units of a high-containment facility and inoculated with a mock PBS and adjuvant emulsion, were kept as infection controls. Vaccinated and control pigs were challenged at 25 dpi with 1.3 × 104 plaque forming units of FMDV O/UK/11/2001, by inoculation at two sites of both main claws of the left-hindfoot pad (0.1 mL/site). This viral dose and inoculation route produced in naive pigs severe FMD clinical signs23,45. Animals were monitored for clinical signs of disease for 11 days. At 0, 15, 25, 29, 31, and 36 dpi, blood samples were collected to obtain sera and/or PBMC. Nasopharyngeal swabs and sera for viremia analysis were taken at 25, 29, 31, and 36 dpi. Pigs were classified as protected if, at most, lesions were found at the injection sites and one additional site on the body23,25,32. Pigs were anesthesized and terminated at the end of the study (earlier for animals with lesions on all 4 feet). To induce gentle loss of consciousness prior to cessation of cardiac and/or respiratory function, and to minimize pain and distress to the animal, an anesthetic solution of zoletil (100 mg/mL, 5 mg/kg), plus atropine (1 mg/mL, 0.05 mg/kg) plus detomidine (10 mg/mL, 0.035 mg/kg) was administered intramuscularly. The anesthesized pigs were then euthanised by intravenous administration of 20–25 mL of Dolethal (sodium pentobarbital) solution (200 mg/mL, 133 mg/kg).

Virus detection after challenge

Nucleic acid extraction was performed manually from 200 μL serum and from nasal swabs, using IndiSpin Pathogen kit (Indical Biosciences, Leipzig, Germany) according to manufacturer instructions. Presence of FMDV RNA was determined by Real-time RT-PCR, using Luna Universal Probe One-Step RT-qPCR Kit. Reactions (in 20 μL) were set up with 5 μL of RNA and 10 μM of serotype-specific primer/probe sets targeting the conserved VP1-coding region of FMDV type O genome54. Amplification was done in a Roche Lightcycler 96 (Thermo Fisher, Waltham, MA) running the following protocol: RT step, 55 °C for 10 min; denaturation step, 95 °C for 1 min; 95 °C for 10 s, 58 °C for 1 min. RT-PCR values were judged positive if the cycle threshold (CT) was below 40.

Virus neutralization (VN) test

The quantitative VN microtest for FMDV neutralizing antibodies was performed in flat-bottomed tissue-culture grade microtitre plates, following recommended procedures from WOAH Terrestrial Manual 2022 (Chapter 3.1.8). Serial two-fold dilutions of each serum sample were incubated with 100 infection units—50% tissue culture infective doses (TCID50)—of FMDV O/UKG/01, for 1 h at 37 °C. Then, a suspension of IBRS-2 cells in DMEM was added and plates were incubated for 72 h. Monolayers were controlled for the development of cytopathic effect, fixed, and stained. End-point titers in a log10 scale were calculated as the reciprocal of the highest serum dilution that neutralized 100 TCID50 of FMDV in 50% of the wells55. For cross neutralization assays, incubation of 25 dpi sera with a panel of FMDVs representing different O topotypes was performed in parallel to that of the homologous isolate O UK/11/2001, with type-C isolate C-S8c1 as a negative control.

Detection of specific anti-FMDV antibodies by ELISA

Total anti-FMDV antibodies were determined by ELISA. Briefly, Nunc 96-well plates (Thermo Fisher) were coated with B peptide (1 µg/well) in PBS overnight at 4 °C. Duplicate three-fold dilution series of each serum sample were prepared in 50 µL, starting at 1/100. Pre-immune sera from peptide-immunized and from non-immunized animals were used as negative controls. Specific antibodies were detected with HRP-conjugated protein A (Thermo Fisher), diluted 1/4000. Color development was obtained after addition of 100 µL/well of TMB (Sigma Aldrich, Madrid, Spain) and stopped by an equal volume of 1 M H2SO4. Plates were read at 450 nm and titers in a log10 scale were expressed as the reciprocal of the last dilution giving the absorbance recorded in the control wells (serum at day 0) plus 2 SD.

PBMC isolation and IFN-γ detection by ELISPOT

Porcine PBMCs were isolated by density gradient centrifugation using Sigma Histopaque-1077 (Merck, Burlington, MA) and Leucosep tubes (Greiner Bio-One, Kremsmunster, Austria) as previously described56. Cell counting and viability tests were done by Trypan blue staining. In general, fresh cells were used, and those remaining were cryopreserved in 90% FBS and 10% DMSO in liquid nitrogen. For the IFN-γ ELISPOT, 2.5 × 105 PBMC were placed in triplicate wells of Immobilon-P 96-well plates (Merck) coated as reported23 before and in vitro stimulated with 50 µg/mL of the dendrimer used for pig immunization or with T [3A (21–35)]57 or B [VP1 (140–158)]58 peptides. For positive control, PBMCs were incubated with 1X of stimulation cocktail (500X), PMA and ionomycin (Merck), using cells incubated without antigen as a negative control. After 48 h at 37 °C—5% CO2, plates were washed and incubated with 2 µg/mL of biotinylated anti-pig IFN-γ antibody (clone P2C11, BD Biosciences, Franklin Lakes, NJ) followed by HRP-streptavidin (BD Biosciences). Antibody was visualized with 3-amino-9-ethyl carbazole (BD Biosciences). The frequency of peptide-specific T cells in the responding population was expressed as the mean number of spot-forming cells/106 PBMC, with background values (number of spots in negative control wells) subtracted from the respective counts of stimulated cells.

Intracellular cytokine staining (ICS)

To detect the phenotype of cytokine-producing cells, intracellular staining of purified PBMC at 25 dpi, was performed using 5 × 106 defrosted PBMC from each pig that were in vitro stimulated or not for 18 h at 37 °C with their respective peptide at a final concentration of 25 µg/mL. Brefeldin A (Sigma Aldrich) was added at a final concentration of 5 µg/mL 10 h before the end of the stimulation time. Purified PBMC from PBS-inoculated pigs, stimulated with PHA-M at a final concentration of 25 µg/mL were used as positive controls, while medium-incubated cultures served as negatives. Cells were washed in PBS-0.05% EDTA-5% FBS and surface-stained by incubation with mouse anti-pig cell-surface monoclonal primary antibody, anti-CD4 (74-12-4, IgG2b)59 for 30 min at 4 °C in PBS-0.05% EDTA-5% swine serum. After three washes in PBS-0.05% EDTA-5% FBS, cells were incubated with Alexa 488-labeled goat anti-mouse secondary antibody (Thermo Fisher) for 30 min at 4 °C, washed three times, fixed and permeabilized for 20 min at 4 °C with Cytofix/Cytoperm buffer (BD Biosciences). Cells were then washed in PERM-WASH buffer (BD Biosciences), free binding sites were blocked with whole mouse IgG, and incubation with PE-conjugated anti-pig IFN-γ (P2G10, BD Biosciences) or APC-conjugated anti-human TNF-α (Mab11, BD Biosciences) for 30 min at 4 °C was carried out to detect intracellular IFN-γ or TNF-α, respectively. Cells were next washed three times in PERM-WASH, fixed with 2% paraformaldehyde and further washed prior to FACS analysis in a CantoA cytometer (BD Biosciences). Data was processed using FACSDiva software (BD Biosciences) or FlowJo software (www.flowjo.com/, Tree Star, Ashland, OR), and transferred to Microsoft Excel for further calculations and preparation of graphs with GraphPad Prism Software 5.0 (Dotmatics, Boston, MA).

SLA Typing via PCR-SSP

SLA polymorphisms were analyzed as previously described28. Briefly, genomic DNA was isolated from PBMC of immunized pigs, following the protocol outlined in the DNeasy Blood and Tissue Kit (Qiagen, Düsseldorf, Germany). SLA class I (SLA-I) and class II (SLA-II) Lr-Hp were determined using a PCR-based sequence-specific primer (PCR-SSP) assay to define the animals’ MHC background at the allele-group level. This typing utilized a complete primer set targeting allele groups within three SLA class I loci (SLA-1, SLA-2, SLA-3) and three SLA class II loci (DRB1, DQB1, DQA)44.

Ethical approval for animal experimentation and sample size estimation

The experimental procedures were conducted in accordance with the guidelines of Directive 2010/63/EU for animal experiments and were approved by the National Committee on Ethics and Animal Welfare under the reference number PROEX 088.1/21. In line with the 3Rs (Reduce, Refine, Replace), a series of policies designed to improve the welfare of animals used in scientific procedures, we determined the minimum sample size for this exploratory study on animals. Based on a resource equation approach60 for a one-way ANOVA group comparison with a minimum error degree of freedom of 10, the minimum number of animals required was n = 5. In line with this design, we and others have published articles evaluating protection against FMDV vaccines using a similar number of animals45,56,61,62,63.

Statistical analyses

Differences in anti-FMDV titers and number of IFN-γ producing cells among peptide-immunized groups were analysed by one-way ANOVA, followed by Tukey’s post-hoc comparisons tests. Values are cited in the text as means ± SD. All p values are two sided, and p < 0.05 were considered significant. Statistical analyses were conducted using GraphPad Prism 5.0.