Hantavirus Gn^H Nanoparticle Immunogen Elicits a Cross-Neutralizing Antibody Response in Mice

Abstract

Hantaviruses are zoonotic pathogens that are spread by rodents and can cause severe and fatal disease in humans. “New World” hantaviruses (endemic to North and South America) include the human pathogenic Andes orthohantavirus (ANDV), Choclo orthohantavirus (CHOV), and Sin Nombre orthohantavirus (SNV). Human infections can lead to hantavirus cardiopulmonary syndrome (HCPS), which is associated with ~40% mortality. Currently, there are no FDA-approved hantavirus vaccines or treatments, but neutralizing antibodies targeting the glycoproteins Gn and Gc have been shown to be protective in animals. Here, we develop nanoparticle immunogens bearing the Gn head domain (Gn^H) from ANDV, CHOV, or SNV. Initial immunization studies with the ANDV-Gn^H monomer indicated that this antigen elicited a reactive but non-neutralizing antibody response in mice. To bolster the immune response, we developed a strategy to link Gn^Hs to mi3 nanoparticles using the SpyCatcher/SpyTag bioconjugation technology. We found that ANDV-Gn^H-mi3 nanoparticles elicited a cross-reactive antibody response that neutralized pseudotyped viruses containing ANDV and CHOV glycoproteins but not SNV. In contrast, CHOV-Gn^H-mi3 nanoparticles elicited only a homotypic neutralizing response. Finally, the reactivity of sera from mice immunized with a cocktail of ANDV-Gn^H-mi3 and SNV-Gn^H-mi3 nanoparticles was similar to sera from mice immunized with ANDV-Gn^H-mi3, only indicating that the SNV-Gn^H-mi3 antibody response was not even homotypically neutralizing. These results suggest that there are differences in immunodominance that may contribute to the breadth of the hantavirus-targeting neutralizing response elicited by Gn^H-based immunogens. Nonetheless, the cross-neutralizing sera obtained by ANDV-Gn^H-mi3 immunization suggest that developing broad immunogens may be possible with appropriate engineering.

Graphical Abstract

Hantaviruses are globally distributed zoonotic viruses that are primarily transmitted by rodents; they can be classified as ‘Old World’ or ‘New World’ based on their geographical distribution. Old World hantaviruses can cause hemorrhagic fever with renal syndrome (HFRS), which has a fatality rate of ~1–10%. Infections by New World hantaviruses such as Andes virus (ANDV), Choclo virus (CHOV), and Sin Nombre virus (SNV) can lead to hantavirus cardiopulmonary syndrome (HCPS), which has case-fatality rates of up to ~40% (although somewhat lower for CHOV).^1–3 Despite multiple outbreaks caused by New World hantaviruses in the Americas and the significant risk of human-to-human transmission of ANDV,⁴ there are currently no approved vaccines or therapeutics available. The sequence diversity of hantaviruses associated with human diseases poses a significant challenge for the development of a single vaccine or therapeutic with broad activity. Thus, novel strategies to design vaccines and therapeutics that provide cross-protection against multiple hantaviruses are a high priority.

The hantavirus virion surface is decorated by a lattice comprising tetramers of the glycoprotein complex.⁵ The genome is composed of three negative-sense single-stranded RNA segments: small (S), medium (M), and large (L). The M segment encodes the two subunits, Gn and Gc, in the endoplasmic reticulum.⁶ Gn and Gc form a heterodimer, which then assembles into a tetrameric Gn/Gc glycoprotein complex (Figure 1A). The Gn subunit is responsible for viral attachment to host cells, specifically via its head domain (Gn^H), and the Gc subunit mediates viral-host membrane fusion.^7,8 The Gn subunit is highly surface-exposed and regulates Gc’s prefusion conformation. Further, Gn is extensively decorated with N-linked glycans that are predicted to be important for noncovalent Gn–Gc interactions and overall stability of the Gn/Gc heterotetramer and likely contribute to immune evasions.⁹ Amino acid sequence analysis of divergent hantaviruses reveals that Gn is less conserved among hantaviruses than is Gc. However, the protein fold of the Gn head domain is similar even among divergent sequences.¹⁰ Further, entry of New World hantaviruses has been reported to be dependent on the host receptor protocadherin-1 (PCDH1).^11,12

Figure 1.

Characterization of monomeric ANDV-Gn^H. (A) Model of hantavirus virion showing the Gn/Gc dimer and surface representation of Gn^H. (B) Reactivity of ANDV-Gn^H to rabbit sera immunized with ANDV/PUUV. (C) Schematic of the immunization timeline. (D) Reactivity of D56 mice sera immunized with monomeric ANDV-Gn^H to VSV-ANDV, VSV-CHOV, and VSV-SNV. (E) Neutralization profiles of mice sera immunized with ANDV-Gn^H to VSV-ANDV, VSV-CHOV, and VSV-SNV. (F) Neutralization curves of naïve sera (negative control). Binding and neutralization curves were generated by aggregating data from two immunization trials, with each sample tested twice in duplicate or triplicate. For the histogram in panel (D), the 1/EC₅₀ for each experiment was plotted, and the statistical comparison among VSV-ANDV and VSV-CHOV or VSV-SNV was performed using a t test.

Antibody response to hantavirus infection is a topic of active investigation. A robust immunoglobulin M (IgM) response is induced during the early phase of infection, followed by an increase in immunoglobulin G (IgG) antibodies.^13,14 Furthermore, the level of hantavirus-specific neutralizing antibodies (NAbs) is correlated with disease outcome.^15,16 NAbs have been detected in convalescent patient serum years after SNV and ANDV infections, suggesting that elicitation of a robust nAb response can provide long-lasting protection.^16–18 Isolation of human monoclonal antibodies against Old and New World hantaviruses indicates that the highly conserved Gc subunit is targeted by broadly reactive antibodies with relatively poor neutralizing activity.¹⁹ By contrast, most Nabs specific for the Gn head, although generally more potent, tend to exhibit virus-specific activity, consistent with its greater sequence divergence relative to that of Gc. Broadly neutralizing antibodies (bNAbs) that have been identified to date are rare and target conserved epitopes at the interface of the Gn/Gc heterodimer.^19–22

Although no vaccine candidate has been identified to provide broad protection, antibody studies suggest that some regions of Gn/Gc may be exploited to develop immunogens that elicit similar bNAbs in vivo.^23–25 Thus, our goal was to explore Gn-based immunogens that can elicit a broadly neutralizing immune response against three human pathogenic New World hantavirus species: ANDV, CHOV, and SNV. To this end, we engineered Gn^H-mi3 nanoparticles. Antigen display on self-assembling protein-based nanoparticles has emerged as an approach to induce more potent immune responses compared with soluble monomeric antigens. By organizing antigens in a multivalent format, nanoparticle immunogens promote lymph node trafficking and B-cell receptor (BCR) cross-linking.^26,27 We found that displaying divergent Gn^Hs on mi3 nanoparticles led to drastic differences in immune responses when they were administered to mice. The display of ANDV-Gn^H on mi3 nanoparticles resulted in an immunogen that generated cross-reactive and cross-neutralizing antibody responses toward CHOV Gn/Gc. However, similar mi3 nanoparticle immunogens containing SNV-Gn^H induced antibody responses with limited cross-neutralization activity against CHOV Gn/Gc-mediated infection. Furthermore, CHOV-Gn^H on mi3 nanoparticles elicited cross-reactive antibody responses toward vesicular stomatitis virus bearing the ANDV glycoproteins (VSV-ANDV) and VSV-SNV but only neutralized homotypic VSV-CHOV. Our work is the first to report on an immunogen that can elicit Gn^H-specific cross-neutralizing antibody responses and provides tools to study the evolution of B-cell populations and IgG repertoires relevant to hantavirus infection.

RESULTS AND DISCUSSION

Monomeric ANDV-Gn^H Elicits a Non-Neutralizing Ab Response.

While we have previously reported cross-neutralizing and protective mAbs targeting Gc or the Gn/Gc interface, these larger complexes are difficult to express and present in relevant conformations in the absence of the viral particle.¹⁹ It is currently unknown whether shared epitopes that can be targeted by cross-neutralizing antibodies exist on Gn^H, which is a much simpler and easier antigen to produce. To explore this, we focused on Gn^H-based immunogens. To investigate whether Gn^H could elicit neutralizing antibody responses, we began with studies on monomeric ANDV-Gn^H. The ANDV-Gn^H protein was expressed and purified from ExpiHEK cells (Figure S1) and reactivity with sera from rabbits previously immunized with the Gn/Gc of ANDV and PUUV examined by ELISA (Figure 1B). The rabbit sera exhibited strong reactivity to the ANDV-Gn^H protein, confirming its antigenic conformation. We next immunized mice with three 8.25 μg doses of ANDV-Gn^H (Figure 1C). We utilized AddaVax adjuvant, which is a squalene-in-water emulsion that is similar to the clinically approved adjuvant MF59. We found that the resulting sera from day 56 (D56) were highly reactive toward recombinant vesicular stomatitis viruses (VSVs) displaying the glycoprotein of ANDV (VSV-ANDV), with 1/EC₅₀ (i.e., the dilution of sera at which 50% binding activity was observed) of 1:40,780, but displayed little cross-reactivity to VSV-CHOV or VSV-SNV (1/EC₅₀ values of 1:62 and 1:272, respectively; Figure 1D). We utilized VSVs bearing the hantavirus Gn/Gc to facilitate studies under routine (BSL2) containment; such VSVs have previously been demonstrated to be faithful surrogates of pathogenic hantavirus entry.^2,19 Although it was reactive to VSV-ANDV, the sera from immunized mice did not neutralize VSV-ANDV or any of the other VSVs tested (VSV-CHOV or VSV-SNV) (Figure 1E,F). The lack of neutralizing activity may be due to the fact that small, monomeric protein antigens stimulate weak immune responses because they are cleared rapidly from the bloodstream and inefficiently cross-link B-cell receptors, a critical event during affinity maturation.²⁸

Generation and Biochemical Characterization of ANDV-Gn^H-mi3 Nanoparticles.

We reasoned that presentation of ANDV-Gn^H in a multivalent nanoparticle could induce stronger antibody responses with potential cross-functionality toward other New World hantaviruses. Protein nanoparticles generally have a longer in vivo half-life compared with monomeric antigens. This persistence, along with enhanced multivalent interactions with BCRs, promotes effective trafficking to lymph nodes and facilitates antibody responses.^29–31 We used the protein-based nanoparticle (NP) platform SpyCatcher003-mi3, which spontaneously assembles into a 60 subunit nanocage and allows facile conjugation to antigens containing a SpyTag (Figure 2A). Several other viral glycoproteins have been conjugated to the SpyCatcher003-mi3 and used successfully as vaccine candidates in rodents and nonhuman primates.^32–35 The mi3 nanoparticle can display antigens of different sizes and symmetries, with 3- or 4-fold symmetry axes on the nanoparticle. Given that the hantavirus glycoprotein assembles in a tetramer with C4 symmetry that requires Gc–Gc interactions, we reasoned that the integration of Gn^H into the mi3 nanoparticle could mimic elements of the tetramer in the absence of Gc. Furthermore, the modular nature of the SpyCatcher/SpyTag conjugation system allows for the incorporation of different Gn^Hs in a “plug-and-play” format.

Figure 2.

Biochemical characterization of ANDV-Gn^H-mi3. (A) Conjugation of SpyTagged Gn^H with SpyCatcher003 to produce ANDV-Gn^H-mi3. (B) Schematic of the SpyCatcher003-mi3 construct and purification gel. (C) SEC chromatogram and gel of SpyCatcher003-mi3 NPs. An aggregate peak and the homogeneous nanoparticle are labeled. (D) EM image of the SpyCatcher003-mi3 NP. (E) SEC purification of ANDV-Gn^H-mi3 and the SDS-PAGE gel showing individual components and the conjugated ANDV-Gn^H-mi3 nanoparticle. (F) EM image of ANDV-Gn^H-mi3. (G) Reactivity of rabbit sera (immunized with ANDV/PUUV) to ANDV-Gn^H-mi3.

SpyCatcher003-mi3 was expressed in Escherichia coli (E. coli), as previously described (Figure 2B), and purified initially by ammonium sulfate precipitation and then by SEC (Figure 2C).³² We confirmed the morphology of the SpyCatcher003-mi3 nanoparticles by electron microscopy (EM) (Figure 2D). Next, SpyTagged ANDV-Gn^H and SpyCatcher003-mi3 were conjugated to form ANDV-Gn^H-mi3 nanoparticles, which were purified by SEC to remove any unconjugated monomeric ANDV-Gn^H (Figure 2E). The morphology of ANDV-Gn^H-mi3 particles was uniform with a diameter of ~80 nm (Figure 2F). Serum from Gn/Gc-immunized rabbits was utilized to confirm the structural integrity of ANDV-Gn^H displayed on mi3 (Figure 2G).

Immunogenicity of ANDV-Gn^H-mi3 Nanoparticles.

We next immunized mice with three 12 μg doses of ANDV-Gn^H-mi3 (Figure 3A, the nanoparticle dose was weight adjusted to be an equivalent molar dose to the Gn^H monomer studies above) and characterized the binding reactivity of the resulting sera to VSV-ANDV, VSV-CHOV, and VSV-SNV by ELISA (Figure 3B). The sera exhibited high reactivity toward VSV-ANDV, with a 1/EC₅₀ value of 1:78,300, while showing comparatively lower reactivity to both VSV-CHOV and VSV-SNV, with 1/EC₅₀ values of 1:166 and 1:676, respectively (Figure 3B). Next, we assessed the neutralization potency of the sera against VSV-ANDV, VSV-CHOV, and VSV-SNV (Figure 3C). In contrast to sera from mice immunized with the ANDV-Gn^H monomer, the sera of mice immunized with ANDV-Gn^H-mi3 were able to neutralize VSV-ANDV with a 1/IC₅₀ of 1:6830. More noteworthy, the ANDV-Gn^H-mi3 sera had cross-neutralizing activity against VSV-CHOV, albeit with a 1/IC₅₀ value of 1:61 (~100-fold lower) but not against VSV-SNV (Figure 3C). We speculate that the differences in cross-neutralizing activity induced by ANDV-Gn^H-mi3 toward VSV-CHOV and VSV-SNV may be due to the different sequence conservation levels between ANDV-Gn^H and CHOV-Gn^H versus ANDV-Gn^H and SNV-Gn^H, which is 71% versus 63%, respectively (Figure 4A and Figure S2).

Figure 3.

Characterizing the immunogenicity of ANDV-Gn^H-mi3. (A) Schematic of the ANDV-Gn^H-mi3 immunization timeline. (B) Binding reactivity curves of mice immunized with ANDV-Gn^H-mi3 (D56 sera samples) to VSV-ANDV, VSV-CHOV, and VSV-SNV and histogram. (C) Neutralization curves of mice immunized with ANDV-Gn^H-mi3 (D56 sera samples) to VSV-ANDV, VSV-CHOV, and VSV-SNV and histogram. Binding and neutralization curves were generated by aggregating data from two immunization trials. For the ELISA histogram in panel (B), the 1/EC₅₀ for each experiment was plotted, and the statistical comparison among VSV-ANDV and VSV-CHOV or VSV-SNV was performed using a t test. For the neutralization histogram in panel (C), the data fitting provided a reliable IC₅₀ value for only two of the four experiments against VSV-CHOV (both plotted), and thus statistical comparisons were not performed. NN = non-neutralizing.

Figure 4.

Characterizing the immunogenicity of CHOV-Gn^H-mi3. (A) Surface conservation map of ANDV-, CHOV-, and SNV-Gn^H. (B) Schematic of the CHOV-Gn^H-mi3 conjugation. (C) EM image of CHOV-Gn^H-mi3. (D) Binding reactivity curves of mice immunized with ANDV-Gn^H-mi3 (D56 sera samples) to VSV-ANDV, VSV-CHOV, and VSV-SNV and the histogram quantifying different groups. (E) Neutralization curves of mice immunized with CHOV-Gn^H-mi3 (D56 sera samples) to VSV-ANDV, VSV-CHOV, and VSV-SNV and the histogram quantifying different groups. Binding and neutralization curves were generated by aggregating data from two immunization trials. For the ELISA histogram in panel (D), the 1/EC₅₀ for each experiment was plotted, and the statistical comparison among VSV-ANDV and VSV-CHOV or VSV-SNV was performed using a t test. For the neutralization histogram in panel (E), the 1/IC₅₀ was similarly plotted for each experiment. NN = non-neutralizing.

Biochemical Properties and Immunogenicity of CHOV-Gn^H-mi3.

Next, we wondered whether more distantly related New World hantavirus Gn^H immunogens would elicit stronger cross-reactive and cross-neutralizing antibody responses. We developed CHOV-Gn^H-mi3 (Figure 4B) by conjugating CHOV-Gn^H and SpyCatcher003-mi3, as described above, followed by SEC purification (Figure 2E). We visually confirmed the morphology of CHOV-Gn^H-mi3 by EM, which indicates a uniform assembly of NPs with a size of ~80 nm (Figure 4C). Furthermore, the serum from rabbits previously immunized with the Gn/Gc of ANDV and PUUV is highly reactive to CHOV-Gn^H-mi3 (Figure 2G), indicating that the antigenic integrity of CHOV-Gn^H displayed on mi3 is preserved. We immunized mice with three 12 μg doses of CHOV-Gn^H-mi3 and determined the binding reactivity of the mouse sera to VSV-ANDV, VSV-CHOV, and VSV-SNV by ELISA (Figure 4D). The sera from mice immunized with CHOV-Gn^H-mi3 were equally reactive toward VSV-ANDV and VSV-SNV, with 1/EC₅₀ values of 1:3360 and 1:3420, respectively, and had ~10-fold lower binding to VSV-CHOV, with a 1/EC₅₀ of 1:360 (Figure 4D). Nonetheless, the sera had a higher cross-reactivity than sera from ANDV-Gn^H-mi3-immunized mice. Next, we characterized the neutralization potency of CHOV-Gn^H-mi3 sera and found that it was able to neutralize VSV-CHOV with a 1/IC₅₀ of 1:1970 but could not neutralize VSV-ANDV or VSV-SNV. The unidirectional cross-neutralizing responses of sera induced by ANDV-Gn^H-mi3, which was able to neutralize VSV-CHOV, in comparison to CHOV-Gn^H-mi3 sera, which are unable to neutralize VSV-ANDV, may be due to differences in immunodominant type-specific epitopes present on CHOV-Gn^H.

Immunogenicity of an ANDV-Gn^H-mi3 + SNV-Gn^H-mi3 Cocktail.

Given the differences in cross-neutralizing activity induced by different Gn^H-mi3s and in an effort to increase the cross-neutralizing activity to VSV-SNV, we generated SNV-Gn^H-mi3 nanoparticles (Figure 5A) to combine with ANDV-Gn^H-mi3. We conjugated SNV-Gn^H and SpyCatcher003-mi3, as described above, and again visually confirmed the morphology of SNV-Gn^H-mi3 by EM (Figure 5B). The Gn/Gc rabbit serum was highly reactive to SNV-Gn^H-mi3 (Figure 2G), again indicating that the antigenic integrity of SNV-Gn^H displayed on mi3 is preserved. We next immunized mice with three doses of a cocktail composed of ANDV-Gn^H-mi3 (6 μg) and SNV-Gn^H-mi3 (6 μg). Overall, the cocktail sera displayed similar binding and neutralization profiles as the sera from mice immunized with ANDV-Gn^H-mi3 only but at half the nanoparticle dosage (Figure 5C). Equally high reactivity toward VSV-ANDV and VSV-SNV (1/EC₅₀ of 1:33,700 and 1:14,900, respectively) and slightly reduced binding to VSV-CHOV, with a 1/EC₅₀ of 1:513 was observed (Figure 5C). Of note, the cocktail group displayed higher reactivity to VSV-SNV than sera from the ANDV-Gn^H-mi3 group. The cause for the consistently lower binding reactivity to VSV-CHOV induced by all Gn^H-mi3s is unclear but may be due to lower amounts of Gn/Gc displayed on those particles.^36,37 Next, we characterized the neutralization potency of the cocktail sera against different VSVs (Figure 5D). Even though the sera from mice immunized with the cocktail displayed equal broad reactivity toward tested VSVs, it neutralized VSV-ANDV with a 1/IC₅₀ of 1:1280 and cross-neutralized VSV-CHOV approximately 10-fold less potently (1/IC₅₀ of 1:114, Figure 5D). The neutralization profiles of the cocktail sera were similar to the sera from the mice immunized with ANDV-Gn^H-mi3 alone in that it could neutralize VSV-ANDV and VSV-CHOV but not VSV-SNV, suggesting that the SNV nanoparticles do not contribute to a neutralizing Ab immune response in mice.

Figure 5.

Characterizing the immunogenicity of an ANDV-Gn^H-mi3 + SNV-Gn^H-mi3 cocktail. (A) Schematic of the SNV-Gn^H-mi3 conjugation. (B) EM image of SNV-Gn^H-mi3. (C) Binding reactivity curves of mice immunized with a cocktail of ANDV-Gn^H-mi3 + SNV-Gn^H-mi3 (D56 sera samples) to VSV-ANDV, VSV-CHOV, and VSV-SNV and the histogram quantifying different groups. (D) Neutralization curves of mice immunized with the cocktail (D56 sera samples) to VSV-ANDV, VSV-CHOV, and VSV-SNV and the histogram quantifying different groups. Binding and neutralization curves were generated by aggregating data from two immunization trials. For the ELISA histogram in panel (C), the 1/EC₅₀ for each experiment was plotted, and the statistical comparison among VSV-ANDV and VSV-CHOV or VSV-SNV was performed using a t test. For the neutralization histogram in panel (D), the 1/IC₅₀ was similarly plotted for each experiment, and comparisons among VSV-ANDV and VSV-CHOV were similarly performed. NN = non-neutralizing.

CONCLUSIONS

Here, we investigated the possibility of eliciting cross-reactive and cross-neutralizing immune responses toward hantaviruses based on the Gn^H domain only. Our results support the hypothesis that the immunogenicity of hantavirus Gn^H immunogens could be improved by integrating them into multivalent mi3 nanoparticles. As a proof-of-concept, we focused on ANDV, CHOV, and SNV, the main causative agents of hantavirus disease in the Americas. Our designed ANDV-Gn^H-mi3 immunogens elicited both cross-reactive and cross-neutralizing antibody responses toward both VSV-ANDV and VSV-CHOV. However, the antibody response was not reciprocally cross-neutralizing when mice were immunized with CHOV-Gn^H-mi3. Furthermore, SNV-Gn^H-mi3 did not elicit even a homotypic neutralizing response. Given that all three viruses utilize the PCDH1 receptor, one possibility is that a cross-neutralizing antibody response may be directed to the conserved receptor-binding site (RBS). However, it is unclear why the serum neutralizing activity did not extend to SNV, which utilizes the same receptor.¹¹ Additionally, the differences observed in hetero- or homotypic neutralizing responses may result from the immunodominance of epitopes that alter the degree of RBS-directed antibody that is induced. The improved neutralizing response observed with ANDV-Gn^H relative to mi3-presented ANDV-Gn^H is likely due to improved BCR cross-linking from polyavid antigen presentation, preferential trafficking of nanoparticles to lymph nodes, improved serum half-life of the nanoparticle relative to the Gn_H monomer, or other adjuvanting behavior induced by the nanoparticles. This work further highlights the importance of antigen display on nanoparticles to induce cross-neutralizing immune responses and the benefits of “plug-and-play” versatility for the presentation of different antigens.

Although B-cells are known to be crucial for antibody production, little is known about the specific evolution of B-cell populations following hantavirus infection or vaccination with Gn^H immunogens.¹ The isolation and characterization of monoclonal antibodies from mice immunized with monomeric and mi3 nanoparticle-displayed ANDV-Gn^H may shed light on the epitopes and modes of antibody binding that afford homotypic and cross-reactive, cross-neutralizing antibody responses. This understanding should inform future efforts to design broadly protective vaccines and therapeutics against hantaviruses. Additionally, further protein engineering approaches can focus on designing glycan variants of Gn^H given their role in assembly to potentially elicit stronger antibody responses directed toward glycan sites that can disrupt heterotetramers on viral particles.³⁸ Lastly, even though our hantavirus Gn^H-mi3 NPs elicited limited cross-neutralizing antibody responses, our study is the first to report any type of cross-neutralizing activity toward different New World hantaviruses that are based solely on the Gn^H domain and provides tools for dissecting protective immunological responses against hantaviruses.

METHODS

Ethics Statement.

Immunization studies were carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocols were approved by the Institutional Animal Care and Use Committee at the Albert Einstein College of Medicine (Animal Welfare Assurance D16–00200).

Expression and Purification of Hantavirus Antigens and Nanoparticles.

To generate the Gn^H of ANDV, CHOV, and SNV gene fragments with BssHII and NheI overhangs, an N-terminal SpyTag and C-terminal His tag (HHHHHHH) were codon optimized for mammalian expression and cloned into the pMAZ vector using Gibson assembly. Once the sequence was verified, the manufacturer’s (Thermo Fisher Scientific) standard protocol for transfecting 100 mL of Expi293F cells was followed. Following transfection, cells were incubated for 4–6 days at 37 °C and 8% CO₂ while being shaken at 120 rpm. Then, cultures were harvested by centrifugation at 4000 rpm and 4 °C for 20 min, after which the supernatant was collected and filtered using a 0.22 μm filter (Millipore). The supernatant was then incubated for 2 h with 1 mL of packed Ni-NTA beads pre-equilibrated with binding buffer (50 mM Tris-HCl, 250 mM NaCl, 5 mM imidazole, pH 8 at 4 °C). After incubation, the supernatant and Ni-NTA mixture was flowed through an Econo-Pac column (Biorad) and washed with binding buffer, Wash 1 buffer (50 mM Tris-HCl, 250 mM NaCl, 10 mM imidazole, pH 8 at 4 °C), and Wash 2 buffer (50 mM Tris-HCl, 250 mM NaCl, 30 mM imidazole, pH 8 at 4 °C). Then, it was eluted with elution buffer (50 mM Tris-HCl, 250 mM NaCl, 250 mM imidazole, pH 8 at 4 °C) and dialyzed overnight at 4 °C with storage buffer (20 mM Tris-HCl, 150 mM NaCl, pH 8 at 4 °C). Proteins were quantified by UV-spectrometry measurement at 280 nm and then either flash frozen in liquid nitrogen and stored at −80 °C or immediately used for conjugation with nanoparticles.

The expression and purification of SpyCatcher003-mi3 were as described previously.³² Briefly, the SpyCatcher003-mi3 gene fragment with XbaI and XhoI overhangs was codon optimized for bacterial expression and cloned into the pet28a vector using Gibson assembly. Once the sequence was verified, the plasmid was transformed into E. coli BL21(DE3) RIPL cells (Agilent) by chemical transformation, incubated for 1 h at 37 °C to allow for recovery, and then plated on Luria broth (LB)-agar plates supplemented with 50 μg/mL of kanamycin and incubated overnight (O/N) at 37 °C. A single colony was picked and cultured in LB media supplemented with 50 μg/mL of kanamycin and incubated O/N at 37 °C while shaking at 200 rpm. The preculture was diluted 1:100 into 1 L of LB media supplemented with 50 μg/mL kanamycin and 0.8% (w/v) glucose. The culture was incubated at 37 °C, shaking at 200 rpm to grow until OD₆₀₀ = 0.6. The expression of SpyCatcher003-mi3 was induced with 0.2 μM isopropyl β-d-1-thiogalactopyranoside (IPTG). Then, the culture was incubated at 22 °C while being shaken at 200 rpm for O/N hours. The next day, the bacterial cells were harvested, and the bacterial pellet was resuspended in 40 mL of lysis buffer (25 mM Tris-HCl, 300 mM NaCl, pH 8.5, supplemented with 0.1 mg/mL lysozyme, 1 mg/mL cOmplete mini EDTA-free protease inhibitor (Merck), and 1 mM phenylmethanesulfonyl fluoride (PMSF)) and stored at −80 °C.

To purify SpyCatcher003-mi3, the bacterial suspension was incubated at 22 °C for 30 min to thaw and then sonicated on ice 4 times for 30 s intervals. Then, the lysate was clarified by centrifugation at 35,000g, 4 °C, for 45 min. The supernatant was filtered through both 0.45 and 0.22 μm syringe filters (STARLAB). The 170 mg of ammonium sulfate was added per mL of lysate to precipitate SpyCatcher003-mi3 particles, and the lysate was incubated at 4 °C for 1 h while mixing at 100 rpm with a magnetic stirrer. Precipitated SpyCatcher003-mi3 particles were pelleted by centrifugation at 30,000g, 4 °C, for 30 min. The supernatant was discarded, and the pellet was resuspended with 8 mL of dialysis buffer (25 mM Tris-HCl, 150 mM NaCl, pH 8.5). Residual ammonium sulfate was removed by dialysis with a 500-fold excess of the dialysis buffer. The next day, the dialyzed SpyCatcher003-mi3 material was centrifuged at 17,000g, 4 °C, for 30 min to remove any insoluble material. Then, the supernatant was filtered through a 0.22 μm syringe filter to remove any aggregates and loaded into a HiPrep Sephacryl S-400 HR 16–600 size exclusion chromatography (SEC) column (GE Healthcare), equilibrated with dialysis buffer, using a KTA Pure 25 system (GE Healthcare), to isolate pure SpyCatcher003-mi3 nanoparticles. The elution fractions containing SpyCatcher003-mi3 nanoparticles were identified by SDS-PAGE, pooled and concentrated using a Vivaspin 20 100 kDa MW cutoff spin concentrator, and stored at −80 °C.

Assembly of Hantavirus Nanoparticles.

To generate hantavirus nanoparticles, ANDV-, CHOV-, and SNV-Gn^Hs were conjugated to SpyCatcher003-mi3 nanoparticles. A 2:1 molar excess of Gn^H to the SpyCatcher003-mi3 nanoparticle mixture was prepared and incubated O/N while gently mixing on a rocker at 4 °C. Then, the reaction was filtered to remove any aggregates and loaded into a HiPrep Sephacryl S-400 HR 16–600 SEC column (as described above) to separate conjugated Gn^H-SpyCatcher003-mi3 nanoparticles from excess monomeric Gn^H. The elution fractions containing conjugated Gn^H-SpyCatcher003-mi3 (referred to as Gn^H-mi3) nanoparticles were identified by SDS-PAGE, pooled and concentrated to 1 mg/mL using a 30 kDa MWCO Millipore Amicon Ultra Centrifugal Filter, then filtered and flash frozen in liquid nitrogen, and stored at −80 °C.

Negative Stain Electron Microscopy.

A carbon only grid, 400 mesh, was cleaned using a Tergeo-EM Plasma Cleaner (PIE Scientific, United States). Purified Gn^H-mi3 nanoparticles were adsorbed onto grids for 10 min and then washed with distilled water. Then, samples were negatively stained with 1% uranyl acetate and viewed on a Tecnai 20 transmission electron microscope (Thermo Fisher) at 120 kV.

ELISA of Hantavirus Gn^H Conjugated to SpyCatcher003-mi3 Nanoparticles.

To validate the structural integrity of the hantavirus Gn^H displayed on Gn^H-mi3 nanoparticles, we performed enzyme-linked immunosorbent assays (ELISAs) with sera from rabbits previously immunized with the Gn/Gc of ANDV and Puumala orthohantavirus (PUUV). Briefly, half-area binding plates (Corning) were coated with 0.2 ug of protein per well of Gn^H or Gn^H-mi3 in phosphate-buffered saline (PBS) at pH 8.0 and O/N at 4 °C. The next day, wells were blocked with 1% (w/v) bovine serum albumin (BSA) in PBS for 2 h at room temperature (RT) while shaking. The rabbit sera were serially diluted by a factor of 10 in PB-T (PBS supplemented with 0.5% (w/v) BSA and 0.05% (v/v) Tween 20) and incubated with immobilized Gn^H or Gn^H-mi3 for 1 h at RT. After incubation, the wells were washed 5× with PBS-T, and then goat-antimouse monoclonal antibody conjugated to horseradish peroxidase (HRP) in PB-T (1:5000 dilution) was added and incubated for 1 h while shaking at RT. Then, plates were washed 5× with PBS-T, developed with 3,3′,5,5′-tetramethylbenzidine (TMB), and quenched with 0.5 M sulfuric acid. The absorbance signal at 450 nm served as a readout for binding reactivity and was measured using a Synergy 4 plate reader (Biotek). The data were fit to a four-parameter logistic regression equation using GraphPad Prism to obtain binding curves and EC₅₀ values.

Immunogenicity of Hantavirus Immunogens.

Groups of five 6- to 8-week-old female BALB/c mice were immunized by intraperitoneal injection with 0.2 mL of immunogens using a 1 mL insulin syringe. Two independent groups of 5 mice were immunized (for a total of 10 mice per treatment group). Hantavirus immunogens were adjuvanted with AddaVax, following the manufacturer’s protocol. The vaccination schedule included a prime, followed by two booster injections. Blood was collected from the animals by submandibular bleeds, 1 week to 1 day prior to prime immunization and 2 weeks following each boost injection. The vascular bundle at the back of the jaw was punctured with a 5 mm lancet to collect up to 0.2 mL of blood in a serum separation tube. The samples were allowed to coagulate at RT for up to 1 h and then centrifuged at 13,000g for 2 min to separate the serum. Serum was aliquoted and flash frozen in liquid nitrogen for storage at −20 °C.

Reactivity of Mouse Sera Immunized with Hantavirus Immunogens.

First, vesicular stomatitis virus (VSV) stocks displaying ANDV, CHOV, and SNV glycoproteins were normalized. To do so, the VSV stocks were labeled with functional component spacer diacyl lipid (FSL)-biotin (Sigma-Aldrich) following manufacturer instructions. Briefly, 3 μL of VSV stock was mixed with 10 μL of 1× NT buffer (10 mM Tris-HCl, 135 mM NaCl at pH 7.5) and 10 μL of FSL-biotin (10 μg/mL) and incubated for 1 h at 37 °C. Then, the VSV-FSL mixture was mixed with 120 μL of PBS and serially diluted 1:2 in deep-well 96-well plates (CELLTREAT). Next, half-area binding plates (Corning) were coated with 25 μL of the serially diluted VSV-FSL and incubated for 1 h at 37 °C. After incubation, coating solution was decanted, and plates were washed with PBS. Then, plates were blocked with 3% BSA in PBS and incubated for 1 h at 37 °C. Plates were washed with PBS, and then streptavidin-HRP (diluted 1:5000 in PBS) was added and incubated for 1 h at 37 °C. Then, plates were washed with PBS, developed with TMB, and quenched with 0.5 M sulfuric acid. The absorbance signal at 450 nm served as a readout for binding reactivity and was measured using a Synergy 4 plate reader (Biotek). The data were fit to a four-parameter logistic regression equation using GraphPad Prism to obtain binding curves and EC₅₀ values. The EC₅₀ values were used to determine the number of VSV particles in each hantavirus VSV stock.

Once the concentration of VSVs was determined, half-area binding plates (Corning) were coated with 25 μL of normalized VSV diluted in PBS and incubated for 1 h at 37 °C. Then, plates were washed with PBS, blocked with 5% (w/v) nonfat dry milk in PBS, and incubated for 1 h at 37 °C. Plates were washed with PBS, and then mice sera were added. The mice sera from the different groups were serially diluted 1:3 (after a 1:50 dilution) in 3% (w/v) BSA in PBS, added to plates, and incubated for 1 h at 37 °C. Then, plates were washed with PBS, and goat-antimouse-HRP (diluted 1:5000 in PBS) was added and incubated for 1 h at 37 °C. After incubation, plates were washed with PBS, developed with TMB, and quenched with 0.5 M sulfuric acid. The absorbance signal at 450 nm served as a readout for binding reactivity and was measured using a Synergy 4 plate reader (Biotek). The data were fit to a four-parameter logistic regression equation using GraphPad Prism to obtain binding curves and EC₅₀ values. Sera from each independent group of five mice were pooled for binding ELISA analysis, which was performed twice independently on each sample in duplicate.

Neutralization Profiles of Mouse Sera.

For neutralization experiments, pretitrated pseudotyped rVSV particles displaying either the ANDV, CHOV, or SNV Gn/Gc glycoproteins were incubated with serially diluted mice serum (diluted in corresponding media) and at RT for 1 h. Then, the mixture was added to a confluent monolayer of Vero cells to allow for any infection and incubated at 37 °C for 14–16 h. Postinfection, infectivity and neutralization were determined using a Cytation 5 cell imaging multimode plate reader (BioTek) for automated counting of eGFP+ cells. The data were fit to a four-parameter logistic regression equation using GraphPad Prism to obtain neutralization curves and IC₅₀ values. Sera from each independent group of five mice were pooled for neutralization analysis, which was performed twice independently on each sample in triplicate.

Supplementary Material

Supporting Information

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsinfecdis.5c00415.

SDS-PAGE of ANDV-Gn^H; sequence alignment of ADNV-/CHOV-/SNV-Gn^H (PDF)