Broadly neutralizing antibodies have been proposed as templates for HIV-1 vaccine design, but it has been unclear how similar vaccine-elicited antibodies are to their naturally elicited templates. To provide insight, here we compare the recognition of naturally elicited and vaccine-elicited antibodies targeting the HIV-1 fusion peptide, which comprises envelope (Env) residues 512–526, with the most common sequence being AVGIGAVFLGFLGAA. Naturally elicited antibodies bound peptides with substitutions to negatively charged amino acids at residue positions 517–520 substantially better than the most common sequence, despite these substitutions rarely appearing in HIV-1; by contrast, vaccine-elicited antibodies were less tolerant of sequence variation, with no substitution of residues 512–516 showing increased binding. Molecular dynamics analysis and cryo-EM structural analysis of the naturally elicited ACS202 antibody in complex with the HIV-1 Env trimer with an alanine 517 to glutamine substitution suggested enhanced binding to result from electrostatic interactions with positively charged antibody residues. Overall, vaccine-elicited antibodies appeared to be more fully optimized to bind the most common fusion peptide sequence, perhaps reflecting the immunization with fusion peptide of the vaccine-elicited antibodies.
Chimeric immunotoxins that combine antigen recognition domains of antibodies and cytotoxic RNases have attracted much attention in recent years as potential targeted agents for cancer immunotherapy. In an attempt to obtain a structurally minimized immunofusion for folding/stability studies, we constructed the chimeric protein VL–barnase. The chimera comprises a small cytotoxic enzyme barnase, ribonuclease from Bacillus amyloliquefaciens, fused to the C‐terminus of the light chain variable domain (VL) of the anti‐human ferritin monoclonal antibody F11. While the individual VL domain was expressed in Escherichia coli as insoluble protein packed into inclusion bodies, its fusion to barnase resulted in a significant (∼70%) fraction of soluble protein, with only a minor insoluble fraction (∼30%) packed into inclusion bodies. The in vivo solubilizing effect of barnase was also observed in vitro and suggests a chaperone‐like role that barnase exerted with regard to the N‐terminal VL domain. Cytoplasmic VL–barnase was analyzed for structural and functional properties. The dimeric state of the chimeric protein was demonstrated by size‐exclusion chromatography, thus indicating that fusion to barnase did not abrogate the intrinsic dimerization propensity of the VL domain. Ferritin‐binding affinity and specificity in terms of constants of association with isoferritins were identical for the isolated VL domain and its barnase fusion, and RNase activity remained unchanged after the fusion. Intrinsic fluorescence spectra showed a fully compact tertiary structure of the fusion protein. However, significantly altered pH stability of the fusion protein versus individual VL and barnase was shown by the pH‐induced changes in both intrinsic fluorescence and binding of ANS. Together, the results indicate that VL–barnase retained the antigen‐binding affinity, specificity and RNase activity pertinent to the two individual constituents, and that their fusion into a single‐chain chimeric protein resulted in an altered tertiary fold and pH stability.
Article Figures and data Abstract eLife digest Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Multivalent presentation of viral glycoproteins can substantially increase the elicitation of antigen-specific antibodies. To enable a new generation of anti-viral vaccines, we designed self-assembling protein nanoparticles with geometries tailored to present the ectodomains of influenza, HIV, and RSV viral glycoprotein trimers. We first de novo designed trimers tailored for antigen fusion, featuring N-terminal helices positioned to match the C termini of the viral glycoproteins. Trimers that experimentally adopted their designed configurations were incorporated as components of tetrahedral, octahedral, and icosahedral nanoparticles, which were characterized by cryo-electron microscopy and assessed for their ability to present viral glycoproteins. Electron microscopy and antibody binding experiments demonstrated that the designed nanoparticles presented antigenically intact prefusion HIV-1 Env, influenza hemagglutinin, and RSV F trimers in the predicted geometries. This work demonstrates that antigen-displaying protein nanoparticles can be designed from scratch, and provides a systematic way to investigate the influence of antigen presentation geometry on the immune response to vaccination. eLife digest Vaccines train the immune system to recognize a specific virus or bacterium so that the body can be better prepared against these harmful agents. To do so, many vaccines contain viral molecules called glycoproteins, which are specific to each type of virus. Glycoproteins that sit at the surface of the virus can act as 'keys' that recognize and unlock the cells of certain organisms, leading to viral infection. To ensure a stronger immune response, glycoproteins in vaccines are often arranged on a protein scaffold which can mimic the shape of the virus of interest and trigger a strong immune response. Many scaffolds, however, are currently made from natural proteins which cannot always display viral glycoproteins. Here, Ueda, Antanasijevic et al. developed a method that allows for the design of artificial proteins which can serve as scaffold for viral glycoproteins. This approach was tested using three viruses: influenza, HIV, and RSV – a virus responsible for bronchiolitis. The experiments showed that in each case, the relevant viral glycoproteins could attach themselves to the scaffold. These structures could then assemble themselves into vaccine particles with predicted geometrical shapes, which mimicked the virus and maximized the response from the immune system. Designing artificial scaffold for viral glycoproteins gives greater control over vaccine design, allowing scientists to manipulate the shape of vaccine particles and test the impact on the immune response. Ultimately, the approach developed by Ueda, Antanasijevic et al. could lead to vaccines that are more efficient and protective, including against viruses for which there is currently no suitable scaffold. Introduction Multivalent antigen presentation, in which antigens are presented to the immune system in a repetitive array, has been demonstrated to increase the potency of humoral immune responses (Bennett et al., 2015; Snapper, 2018). This has been attributed to increased cross-linking of antigen-specific B cell receptors at the cell surface and modulation of immunogen trafficking to and within lymph nodes (Irvine et al., 2013; Tokatlian et al., 2019). An ongoing challenge has been to develop multimerization scaffolds capable of presenting complex oligomeric or engineered antigens (Sanders and Moore, 2017; Jardine et al., 2013; McLellan et al., 2013a), as these can be difficult to stably incorporate into non-protein-based nanomaterials (e.g. liposomes, polymers, transition metals and their oxides). Epitope accessibility, proper folding of the antigen, and stability are also important considerations in any strategy for antigen presentation. Several reports have utilized non-viral, naturally occurring protein scaffolds, such as self-assembling ferritin (Kanekiyo et al., 2013; Sliepen et al., 2015; Darricarrère et al., 2018), lumazine synthase (Sanders and Moore, 2017; Abbott et al., 2018), or encapsulin (Kanekiyo et al., 2015) nanoparticles, to present a variety of complex oligomeric or engineered antigens. These studies have illustrated the advantages of using self-assembling proteins as scaffolds for antigen presentation (López-Sagaseta et al., 2016; Kanekiyo et al., 2019), including enhanced immunogenicity and seamless integration of antigen and scaffold through genetic fusion. More recently, computationally designed one- and two-component protein nanoparticles (Hsia et al., 2016; King et al., 2014; Bale et al., 2016) have been used to present complex oligomeric antigens, conferring additional advantages such as high stability, robust assembly, ease of production and purification, and increased potency upon immunization (Marcandalli et al., 2019; Brouwer et al., 2019). The ability to predictively explore new structural space makes designed proteins (Parmeggiani et al., 2015; Brunette et al., 2015) attractive scaffolds for multivalent antigen presentation. In our previous work with computationally designed nanoparticle immunogens (Marcandalli et al., 2019; Brouwer et al., 2019), the nanoparticles were generated from naturally occurring oligomeric proteins without initial consideration of geometric compatibility for antigen presentation. A more comprehensive solution would be to de novo design nanoparticles which present complex antigens of interest. For homo-oligomeric class I viral fusion proteins, a large group that includes many important vaccine antigens (Harrison, 2015), a close geometric match between the C termini of the antigen and the N termini of a designed nanoparticle component would enable multivalent presentation without structural distortion near the glycoprotein base, and potentially allow for better retention of antigenic epitopes relevant to protection. More generally, precise control of antigen presentation geometry through de novo nanoparticle design would enable systematic investigation of the structural determinants of immunogenicity. Results De novo design of protein nanoparticles tailored for multivalent antigen presentation We sought to develop a general computational method for de novo designing protein nanoparticles with geometries tailored to present antigens of interest, focusing specifically on the prefusion conformations of the trimeric viral glycoproteins HIV-1 Env (BG505 SOSIP) (Wang et al., 2017; Sanders et al., 2013), influenza hemagglutinin (H1 HA) (Kadam et al., 2017), and respiratory syncytial virus (RSV) F (DS-Cav1) (McLellan et al., 2013a). To make the antigen-tailored nanoparticle design problem computationally tractable, we employed a two-step design approach (Figure 1). In the first step, we de novo designed antigen-tailored trimers, featuring N termini geometrically matched to the C termini of the viral glycoproteins. In the second step, we generated tetrahedral, octahedral, and icosahedral two-component nanoparticles by designing secondary interfaces between a designed trimer (fusion component) and a de novo homo-oligomer (assembly component) (Fallas et al., 2017). This design approach yielded nanoparticles tailored to present 4, 8, or 20 copies of the viral glycoproteins in defined geometries (Figure 1d). Sequences for all designed trimers and homo-oligomers, two-component nanoparticles, and antigen-fused components in this study can be found in Supplementary file 1A, B, and C, respectively. Details on each step of the design approach are described in the following sections. Figure 1 with 1 supplement see all Download asset Open asset De novo design of protein nanoparticles tailored for multivalent antigen presentation. (a) Computational docking of monomeric repeat proteins into C3-symmetric trimers using the RPX method. (b) Selection of trimers for design based on close geometric match between their N termini (blue spheres) and C termini (red spheres) of a viral antigen (green, BG505 SOSIP shown for illustration). (c) Design of two-component nanoparticles incorporating a fusion component (cyan) and assembly component (gray). (d) Nanoparticle assembled with antigen-fused trimeric component yields multivalent antigen-displaying nanoparticle. Computational design of trimers tailored for fusion to specific viral glycoproteins We chose to design our antigen-tailored trimers from monomeric repeat proteins composed of rigidly packed 20- to 50-residue tandem repeat units (Parmeggiani et al., 2015; Brunette et al., 2015; Urvoas et al., 2010; Kajander et al., 2007; Main et al., 2003), as their high stability and tunable length (through variation of repeat number) are desirable properties for the design of protein-based nanomaterials. These structurally diverse alpha-helical repeat proteins featured three to six repeat modules and total lengths between 119 and 279 residues. They were docked into C3-symmetric trimers using our RPX docking method, which identifies configurations likely to accommodate favorable side chain packing at the de novo designed interface (Fallas et al., 2017). To identify trimeric configurations with N termini compatible for fusion to the C termini of the three viral glycoproteins, docks with an RPX score above 5.0 were screened using the sic_axle protocol (Marcandalli et al., 2019). Geometrically compatible docks (non-clashing termini separation distances of 15 Å or less) were subjected to full Rosetta C3-symmetric interface design and filtering (see Materials and Methods), and twenty-three designs were selected for experimental characterization (Figure 1—figure supplement 1). Structural characterization of designed trimers Synthetic genes encoding each of the designed trimers were expressed in E. coli and purified from lysates by Ni2+ immobilized metal affinity chromatography (Ni2+ IMAC) followed by size-exclusion chromatography (SEC). Twenty-two designs were found to express in the soluble fraction, and nine formed the intended trimeric oligomerization state as assessed by SEC in tandem with multi-angle light scattering (SEC-MALS; examples in Figure 2 top panel, second row; SEC-MALS chromatograms for the remaining designs are in Figure 2—figure supplement 2 and data in Figure 2—figure supplement 1—source data 1; SEC chromatograms for remaining designs with off-target retention volumes are in Figure 2—figure supplement 2). Four of the designs that were trimeric and expressed in high yield, 1na0C3_2, 3ltjC3_1v2, 3ltjC3_11, and HR04C3_5v2, were selected for solution small angle X-ray scattering (SAXS) experiments. The proteins exhibited scattering profiles very similar to those computed from the corresponding design models, suggesting similar supramolecular configuration (Figure 2 top panel, third row; metrics in Table 1 and Figure 2—source data 1). These four trimers were derived from three distinct designed helical repeat proteins from TPR, HEAT, or de novo topological families (1na0, 3ltj, and HR04, see Materials and Methods) (Brunette et al., 2015; Urvoas et al., 2010; Main et al., 2003). Crystals were obtained for the two designs 1na0C3_2 and 3ltjC3_1v2. Structures were determined at resolutions of 2.6 and 2.3 Å, revealing a backbone root mean square deviation (r.m.s.d.) between the design model and structure of 1.4 and 0.8 Å, respectively (Figure 2—figure supplement 3, and Figure 2—figure supplement 3—source data 1, crystallization conditions, structure metrics, and structure-to-model comparisons are described in Materials and Methods). The structures confirmed in both cases that the designed proteins adopt the intended trimeric configurations, and that most of the atomic details at the de novo designed interfaces are recapitulated. Figure 2 with 7 supplements see all Download asset Open asset Biophysical characterization of antigen-tailored trimers and nanoparticles. Top rows, design models. Middle rows, SEC chromatograms and calculated molecular weights from SEC-MALS. Bottom rows, comparisons between experimental SAXS data and scattering profiles calculated from design models. (a) 1na0C3_2. (b) 3ltjC3_1v2. (c) 3ltjC3_11. (d) HR04C3_5v2. (e) T33_dn2. (f) T33_dn10. (g) O43_dn18. (h) I53_dn5. Figure 2—source data 1 Biophysical properties of designed trimers and two-component nanoparticles. Experimentally-measured data (exp) is compared to predicted design data (model). Molecular weights (MW) were obtained using the ASTRA software. Rg and Dmax calculations performed in Scatter3 SAXS analysis software with the determined qmax values. X values computed from the FoXS online SAXS web server between the designed model and the experimental scattering data. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data1-v1.docx Download elife-57659-fig2-data1-v1.docx Figure 2—source data 2 1na0C3_2 SEC-MALS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data2-v1.txt Download elife-57659-fig2-data2-v1.txt Figure 2—source data 3 3ltjC3_1v2 SEC-MALS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data3-v1.txt Download elife-57659-fig2-data3-v1.txt Figure 2—source data 4 3ltjC3_11 SEC-MALS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data4-v1.txt Download elife-57659-fig2-data4-v1.txt Figure 2—source data 5 HR04C3_5v2 SEC-MALS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data5-v1.txt Download elife-57659-fig2-data5-v1.txt Figure 2—source data 6 1na0C3_2 SAXS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data6-v1.txt Download elife-57659-fig2-data6-v1.txt Figure 2—source data 7 3ltjC3_1v2 SAXS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data7-v1.txt Download elife-57659-fig2-data7-v1.txt Figure 2—source data 8 3ltjC3_11 SAXS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data8-v1.txt Download elife-57659-fig2-data8-v1.txt Figure 2—source data 9 HR04_5v2 SAXS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data9-v1.txt Download elife-57659-fig2-data9-v1.txt Figure 2—source data 10 T33_dn2 SEC-MALS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data10-v1.txt Download elife-57659-fig2-data10-v1.txt Figure 2—source data 11 T33_dn10 SEC-MALS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data11-v1.txt Download elife-57659-fig2-data11-v1.txt Figure 2—source data 12 O43_dn18 SEC-MALS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data12-v1.txt Download elife-57659-fig2-data12-v1.txt Figure 2—source data 13 I53_dn5 SEC-MALS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data13-v1.txt Download elife-57659-fig2-data13-v1.txt Figure 2—source data 14 T33_dn2 SAXS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data14-v1.txt Download elife-57659-fig2-data14-v1.txt Figure 2—source data 15 T33_dn10 SAXS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data15-v1.txt Download elife-57659-fig2-data15-v1.txt Figure 2—source data 16 O43_dn18 SAXS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data16-v1.txt Download elife-57659-fig2-data16-v1.txt Figure 2—source data 17 I53_dn5 SAXS. https://cdn.elifesciences.org/articles/57659/elife-57659-fig2-data17-v1.txt Download elife-57659-fig2-data17-v1.txt Table 1 Summary of the experimental characterization for designed trimers and two-component nanoparticles. 1na0C3_2 and 3ltjC3_1v2 structures determined by X-ray crystallography and T33_dn10, O43_dn18, and I53_dn5 structures determined by cryo-EM. DesignTargeted AntigensExperimental Molecular Weight (kDa)Target Molecular Weight (kDa)SAXS X valueResolution, backbone r.m.s.d. structure (Å, Å)1na0C3_2HA, SOSIP, DS-Cav148451.42.6, 1.43ltjC3_1v2SOSIP, DS-Cav156631.12.3, 0.83ltjC3_11SOSIP, DS-Cav150661.6--HR04C3_5v2SOSIP71691.5--T33_dn2HA, SOSIP, DS-Cav13973454.8--T33_dn5HA, SOSIP, DS-Cav14224221.7--T33_dn10HA, SOSIP, DS-Cav15465562.33.9, 0.65O43_dn18HA,SOSIP, DS-Cav18108762.94.5, 0.98I53_dn5HA, SOSIP, DS-Cav1200019601.25.3, 1.30 Table 1—source data 1 Summary of the experimental characterization for designed trimers and two-component nanoparticles. https://cdn.elifesciences.org/articles/57659/elife-57659-table1-data1-v1.docx Download elife-57659-table1-data1-v1.docx Computational design of two-component nanoparticles incorporating designed trimers As secondary assembly components were required to design our antigen-tailored nanoparticles, validated trimers were docked pairwise with de novo designed symmetric homo-oligomers (Fallas et al., 2017) to generate tetrahedral, octahedral, and icosahedral nanoparticle configurations using the TCdock program (King et al., 2014; Bale et al., 2016). To increase the probability of generating icosahedra which confer the highest valency among the targeted symmetries, three naturally occurring homopentamers were also included in the docking calculations (PDB IDs 2JFB, 2OBX, and 2B98). Analogously to the designed trimers, nanoparticle docks were scored and ranked using the RPX method (Fallas et al., 2017) to identify configurations likely to accommodate favorable side chain packing at a secondary de novo designed interface. High-ranking and non-redundant nanoparticle configurations featuring outward-facing N termini for antigen presentation were selected for Rosetta interface design (King et al., 2014; Bale et al., 2016). Fifty-three nanoparticle designs across all three targeted symmetries that exhibited the best interface metrics were selected for experimental characterization (see Materials and Methods). The nomenclature for the eleven tetrahedra, twenty-one octahedra, and twenty-one icosahedra indicate the symmetry of the nanoparticle (T, O, or I), the oligomeric state of the first component (A) and second component (B) used in each design, the letters "dn" reflecting the de novo nature of the input oligomers, and the rank by RPX score from the docking stage (e.g., "I53_dn5" indicates an icosahedral nanoparticle constructed from a pentameric and trimeric component, ranked 5th in RPX-scoring for the two input oligomers). Synthetic genes encoding each of the two-component nanoparticles were obtained with one of the components fused to a His6-tag, and the designs were purified using Ni2+ IMAC (see Materials and methods). Pairs of proteins at the expected molecular weights were found to co-elute by SDS-PAGE for twenty-four of the designs, consistent with spontaneous assembly of the nanoparticles followed by pulldown His6-tagged component (featured co-eluting designs are presented in Figure 2—figure supplement 4). SEC chromatograms revealed that nineteen designs did not form assemblies of the expected size or that the resulting assemblies were heterogeneous (Figure 2—figure supplement 5). Five designs comprising a panel of unique geometric configurations, T33_dn2, T33_dn5, T33_dn10, O43_dn18, and I53_dn5, ran as monodisperse particles of the predicted molecular mass by SEC-MALS and were further investigated by SAXS. The experimental solution scattering curves closely matched the scattering curves computed from the design models (Schneidman-Duhovny et al., 2010) for all five designs (Figure 2, bottom panel and Figure 2—figure supplement 6; metrics in Table 1 and Figure 2—source data 1, bottom five designs). Due to its high valency and production yield, we selected the I53_dn5 nanoparticle to investigate the capacity of its two components to be separately produced and assembled in vitro. The two components of I53_dn5 were re-cloned, expressed, and separately purified (pentameric "I53_dn5A" with His6-tag and trimeric "I53_dn5B"). Nanoparticle assembly appeared to be complete within minutes after equimolar mixing (Figure 2—figure supplement 7). This capability is noteworthy as it enables production of each component independently, even from different host systems, which provides more flexibility in nanoparticle manufacturing. In vitro assembly also confers more control over nanoparticle assembly and composition, for example by assembling with a mixture of components fused to different antigens (Boyoglu-Barnum et al., 2020). Structural characterization of designed two-component nanoparticles The five SAXS-validated nanoparticles were structurally characterized using negative stain electron microscopy (NS-EM) (Lee and Gui, 2016; Ozorowski et al., 2018). 2,000–5000 particles were manually picked from the electron micrographs acquired for each designed nanoparticle and classified in 2D using the Iterative MSA/MRA algorithm (see Materials and Methods). 3D classification and refinement steps were performed in Relion/3.0 (Zivanov et al., 2018). Analysis of the NS-EM data confirmed high sample homogeneity for all five nanoparticle designs as evident from the micrographs and 2D class-averages (Figure 3). While some free nanoparticle components were detected in the T33_dn5 sample, suggesting a certain propensity towards disassembly, analysis of the reconstructed 3D maps revealed that all five nanoparticles assemble as predicted by the design models, at least to the resolution limits of NS-EM. Figure 3 Download asset Open asset NS-EM analysis of antigen-tailored nanoparticles. From left to right: designed trimers incorporated in each designed nanoparticle, nanoparticle design models fit into NS-EM density (views shown down each component axis of symmetry), designed nanoparticle 2D class-averages, raw electron micrographs of designed nanoparticles. (a) T33_dn2. (b) T33_dn5. (c) T33_dn10. (d) O43_dn18. (e) I53_dn5. In order to obtain higher-resolution information, three designs, T33_dn10, O43_dn18, and I53_dn5, representing one nanoparticle from each targeted symmetry (T, O, I), were subjected to cryo-electron microscopy (cryo-EM). Cryo-EM data acquisition was performed as described in the Materials and Methods section and data acquisition statistics are displayed in Figure 4—source data 1. The data processing workflow is presented in Figure 4—figure supplement 1. Appropriate symmetry (T, O, and I for T33_dn10, O43_dn18, and I53_dn5, respectively) was applied during 3D classification and refinement and maps were post-processed in Relion/3.0 (Zivanov et al., 2018). The final resolutions of the reconstructed maps for the T33_dn10, O43_dn18, and I53_dn5 nanoparticles were 3.9, 4.5, and 5.3 Å, respectively. Some structural heterogeneity was observed in the cryo-EM data, particularly in the case of I53_dn5. In 2D classification results we generated particle projection averages that range from spherical to ellipsoid shape (Figure 4—figure supplement 1c), indicating some degree of flexibility. There is less evidence of flexibility in T33_dn10 and O43_dn18, in agreement with the higher final map resolution for these nanoparticles. Nanoparticle design models were relaxed into the corresponding EM maps by applying multiple rounds of Rosetta relaxed refinement (Wang et al., 2016) and manual refinement in Coot (Emsley and Crispin, 2018) to generate the final structures. Refined model statistics are shown in Figure 4—source data 2. Reconstructed cryo-EM maps for T33_dn10, O43_dn18, and I53_dn5 and refined models are superimposed in Figure 4. Overall, the refined structures show excellent agreement with the corresponding Rosetta design models. Backbone r.m.s.d. values estimated for the asymmetric unit (consisting of a single subunit of component A and component B) were 0.65, 0.98, and 1.3 Å for T33_dn10, O43_dn18, and I53_dn5, respectively (Table 1). Figure 4 with 1 supplement see all Download asset Open asset Cryo-EM analysis of antigen-tailored nanoparticles. From left to right: cryo-EM maps with refined nanoparticle design models fit into electron density, view of designed nanoparticle interface region fit into cryo-EM density with indicated resolution (res.), designed nanoparticle 2D class-averages, raw cryo-EM micrographs of designed nanoparticles. (a) T33_dn10. (b) O43_dn18. (c) I53_dn5. Figure 4—source data 1 Cryo-EM data acquisition metrics for designed nanoparticles T33_dn10, O43_dn18, and I53_dn5. https://cdn.elifesciences.org/articles/57659/elife-57659-fig4-data1-v1.docx Download elife-57659-fig4-data1-v1.docx Figure 4—source data 2 Cryo-EM model building and refinement statistics for designed nanoparticles T33_dn10, O43_dn18, and I53_dn5. https://cdn.elifesciences.org/articles/57659/elife-57659-fig4-data2-v1.docx Download elife-57659-fig4-data2-v1.docx Characterization of viral glycoprotein-displaying nanoparticles To explore the capability of the designed nanoparticles to present viral glycoproteins, we produced their trimeric fusion components genetically linked to a stabilized version of the BG505 SOSIP trimer. Synthetic genes for BG505 SOSIP fused to the N termini of T33_dn2A, T33_dn10A, and I53_dn5B (BG505 SOSIP–T33_dn2A, BG505 SOSIP–T33_dn10A, and BG505 SOSIP–I53_dn5B) were transfected into HEK293F cells. The secreted fusion proteins were then purified using a combination of immuno-affinity chromatography and SEC. The corresponding assembly component for each nanoparticle was produced recombinantly in E. coli, and in vitro assembly reactions were performed as equimolar mixtures of the two components overnight. Assembled nanoparticles were purified by SEC and analyzed by NS-EM to assess particle assembly and homogeneity. ~ 1000 particles were manually picked and used to perform 2D classification and 3D classification/refinement in Relion (Zivanov et al., 2018). Models for the BG505 SOSIP-displaying nanoparticles fit into their reconstructed 3D maps are displayed in Figure 5 (left). BG505 SOSIP trimers are clearly discernible in 2D class-averages and reconstructed 3D maps. However, the trimers appear less well-resolved than the corresponding nanoparticle core in the three reconstructions, likely due to the short flexible linkers between the BG505 SOSIP trimer and the fusion component. The self-assembling cores of the antigen-fused T33_dn2, T33_dn10, and I53_dn5 nanoparticles were very similar to the NS-EM maps of the unmodified nanoparticles (at least to the resolution limits of NS-EM), demonstrating that fusion of the BG505 SOSIP trimer did not induce any major structural changes to the underlying nanoparticle scaffolds. Free components were detected in raw EM micrographs of BG505 SOSIP–I53_dn5, indicating some degree of disassembly. This finding is supported by stability data reported in a parallel study, where BG505 SOSIP–I53_dn5 demonstrated sensitivity to various physical and chemical stressors (Antanasijevic et al., 2020). Figure 5 with 2 supplements see all Download asset Open asset NS-EM analysis of BG505 SOSIP-displaying nanoparticles. From left to right: BG505 SOSIP-displaying nanoparticle models fit into NS-EM density, 2D class-averages, raw NS-EM micrographs of assembled BG505 SOSIP-displaying nanoparticles. (a) BG505 SOSIP–T33_dn2. (b) BG505 SOSIP–T33_dn10. (c) BG505 SOSIP–I53_dn5. To further characterize the capability of the designed nanoparticles to present viral glycoproteins, we characterized the structures and antigenic profiles of I53_dn5 fused to the prefusion influenza HA and RSV F glycoproteins (HA–I53_dn5 and DS-Cav1–I53_dn5). Constructs were generated with each glycoprotein genetically linked to the N terminus of the I53_dn5B trimeric fusion component, and the proteins were secreted from HEK293F cells and purified by Ni2+ IMAC. The fusion proteins were mixed with equimolar pentameric I53_dn5A for HA–I53_dn5 or I53_dn5A.1 (a stabilized and redox-insensitive variant of I53_dn5A lacking cysteines, see Materials and Methods) for DS-Cav1–I53_dn5, and the assembly reactions purified by SEC. For both assemblies, the majority of the material migrated in the peak expected for assembled nanoparticles, and NS-EM analysis showed formation of I53_dn5 nanoparticles with spikes emanating from the surface (Figure 5—figure supplement 1 and 2). In both cases, there was considerable variation in the spike geometry, again suggesting some flexibility between the glycoproteins and the underlying scaffold. The GG linker connecting DS-Cav1 to I53_dn5 likely accounts for the observed flexibility and suboptimal definition of the glycoprotein trimer in two-dimensional class averages (Figure 5—figure supplement 1, bottom right). There was no engineered linker between the glycoprotein and fusion component in the case of HA–I53_dn5, and more clearly defined spike density was observed in the class averages (Figure 5—figure supplement 2, bottom right). To determine if the presented glycoproteins were properly folded, we examined their reactivity with conformation-specific monoclonal antibodies (mAbs). The DS-Cav1–I53_dn5 nanoparticle was found by an enzyme-linked immunosorbent assay (ELISA) to bind the RSV F-specific mAbs D25 (McLellan et al., 2013b), Motavizumab (Cingoz, 2009), and AM14 (Gilman et al., 2015) similarly to soluble DS-Cav1 trimer with foldon (McLellan et al., 2013a), indicating that the F protein is presented in the desired prefusion conformation on the nanoparticle (Figure 4—figure supplement 1, top). Biolayer interferometry binding experiments with anti-HA head - and stem-specific mAbs (Krause et al., 2011; Ekiert et al., 2009) analogously showed that both the HA–I53_dn5 nanoparticle and the HA–I53_dn5B trimer presented the head and stem regions with wild-type-like antigenicity (Figure 5—figure supplement 2, top). Tuning BG50
Abstract Since the outbreak of the COVID-19 pandemic, widespread infections have allowed SARS-CoV-2 to evolve in human, leading to the emergence of multiple circulating variants. Some of these variants show increased resistance to vaccines, convalescent plasma, or monoclonal antibodies. In particular, mutations in the SARS-CoV-2 spike have drawn attention. To facilitate the isolation of neutralizing antibodies and the monitoring the vaccine effectiveness against these variants, we designed and produced biotin-labeled molecular probes of variant SARS-CoV-2 spikes and their subdomains, using a structure-based construct design that incorporated an N-terminal purification tag, a specific amino acid sequence for protease cleavage, the variant spike-based region of interest, and a C-terminal sequence targeted by biotin ligase. These probes could be produced by a single step using in-process biotinylation and purification. We characterized the physical properties and antigenicity of these probes, comprising the N-terminal domain (NTD), the receptor-binding domain (RBD), the RBD and subdomain 1 (RBD-SD1), and the prefusion-stabilized spike ectodomain (S2P) with sequences from SARS-CoV-2 variants of concern or of interest, including variants Alpha, Beta, Gamma, Epsilon, Iota, Kappa, Delta, Lambda, Mu, and Omicron. We functionally validated probes by using yeast expressing a panel of nine SARS-CoV-2 spike-binding antibodies and confirmed sorting capabilities of variant probes using yeast displaying libraries of plasma antibodies from COVID-19 convalescent donors. We deposited these constructs to Addgene to enable their dissemination. Overall, this study describes a matrix of SARS-CoV-2 variant molecular probes that allow for assessment of immune responses, identification of serum antibody specificity, and isolation and characterization of neutralizing antibodies.
Both SIV and SHIV are powerful tools for evaluating antibody-mediated prevention and treatment of HIV-1. However, owing to a lack of rhesus-derived SIV broadly neutralizing antibodies (bnAbs), testing of bnAbs for HIV-1 prevention or treatment has thus far been performed exclusively in the SHIV NHP model using bnAbs from HIV-1-infected individuals. Here we describe the isolation and characterization of multiple rhesus-derived SIV bnAbs capable of neutralizing most isolates of SIV. Eight antibodies belonging to two clonal families, ITS102 and ITS103, which target unique epitopes in the CD4 binding site (CD4bs) region, were found to be broadly neutralizing and together neutralized all SIV strains tested. A rare feature of these bnAbs and two additional antibody families, ITS92 and ITS101, which mediate strain-specific neutralizing activity against SIV from sooty mangabeys (SIVsm), was their ability to achieve near complete (i.e. 100%) neutralization of moderately and highly neutralization-resistant SIV. Overall, these newly identified SIV bnAbs highlight the potential for evaluating HIV-1 prophylactic and therapeutic interventions using fully simian, rhesus-derived bnAbs in the SIV NHP model, thereby circumventing issues related to rapid antibody clearance of human-derived antibodies, Fc mismatch and limited genetic diversity of SHIV compared to SIV.
