Computational design of protein antigens that interact with the CDR H3 loop of HIV broadly neutralizing antibody 2F5

Engineered immunogens to elicit antibodies against conserved coronavirus epitopes

Immune responses to SARS-CoV-2 primarily target the receptor binding domain of the spike protein, which continually mutates to escape acquired immunity. Other regions in the spike S2 subunit, such as the stem helix and the segment encompassing residues 815-823 adjacent to the fusion peptide, are highly conserved across sarbecoviruses and are recognized by broadly reactive antibodies, providing hope that vaccines targeting these epitopes could offer protection against both current and emergent viruses. Here we employ computational modeling to design scaffolded immunogens that display the spike 815-823 peptide and the stem helix epitopes without the distracting and immunodominant receptor binding domain. These engineered proteins bind with high affinity and specificity to the mature and germline versions of previously identified broadly protective human antibodies. Epitope scaffolds interact with both sera and isolated monoclonal antibodies with broadly reactivity from individuals with pre-existing SARS-CoV-2 immunity. When used as immunogens, epitope scaffolds elicit sera with broad betacoronavirus reactivity and protect as "boosts" against live virus challenge in mice, illustrating their potential as components of a future pancoronavirus vaccine.

Computational design of vaccine immunogens

Computational protein engineering has enabled the rational design of customized proteins, which has propelled both sequence-based and structure-based immunogen engineering and delivery. By discerning antigenic determinants of viral pathogens, computational methods have been implemented to successfully engineer representative viral strains able to elicit broadly neutralizing responses or present antigenic sites of viruses for focused immune responses. Combined with improvements in customizable nanoparticle design, immunogens are multivalently displayed to enhance immune responses. These rationally designed immunogens offer unique and powerful approaches to engineer vaccines for pathogens, which have eluded traditional approaches.

Analysis of Binding Modes of Antigen-Antibody Complexes by Molecular Mechanics Calculation

Antibodies are one of the most important protein molecules in biopharmaceutics. Due to the recent advance in technology for producing monoclonal antibodies, many structural data are available on the antigen-antibody complexes. To characterize the molecular interaction in antigen-antibody recognition, we computationally analyzed 500 complex structures by molecular mechanics calculations. The presence of Ser and Tyr is markedly large in the complementarity-determining regions (CDRs). Although Ser is abundant in CDRs, its contribution to the binding score is not large. Instead, Tyr, Asp, Glu, and Arg significantly contribute to the molecular interaction from the viewpoint of the binding score. The decomposition of the binding score suggests that the hydrophilic interaction is predominant in all CDRs compared with the hydrophobic one. The contribution of the heavy chain is larger than that of the light chain. In particular, H2 and H3 largely contribute to the binding interaction. Tyr is a main contributing residue both in H2 and H3. The positively charged residue Arg also significantly contributes to the binding score in H3, while the contribution of Lys is small. The appearance of Ser is remarkable in H2, and Asp is abundant in H3. The non-charged polar residues, Thr, Asn, and Gln, appear much in H2, compared to appearing in H3. The negatively charged residues Asp and Glu significantly contribute to the binding score in H3. The contributions of Phe and Trp are not large in spite that the aromatic residues are capable of making the π-π or CH-π interaction. Gly is commonly abundant both in H2 and H3. The average distance of the shortest direct hydrogen bond between the antigen and antibody is longer than that of the hydrogen bonds observed in the complexes between compounds and their target proteins. Therefore, the antigen-antibody interface is not so tight as the compound-target protein interface. The calculation of shape complementarity is consistent with the result of the hydrogen bonds in that the fitness of the antigen-antibody contact is not so high as that of the compound-target protein contact. There exist many water molecules at the antigen-antibody interface. These findings suggest that Tyr, Asp, Glu, and Arg are rich in H3 and work as major contributors for the interaction with the antigen. Ser, Thr, Asn, and Gln are rich in H2 and support the interaction with enhancing molecular fitness. Gly is helpful in increasing flexibility and geometrical diversity. Because the antigen-antibody binding is fundamentally hydrophilic-driven, the non-polar residues are unfavorable for mediating the contact even for the aromatic residues such as Phe and Trp.

Computational tools for modern vaccine development

Vaccines play an essential role in controlling the rates of fatality and morbidity. Vaccines not only arrest the beginning of different diseases but also assign a gateway for its elimination and reduce toxicity. This review gives an overview of the possible uses of computational tools for vaccine design. Moreover, we have described the initiatives of utilizing the diverse computational resources by exploring the immunological databases for developing epitope-based vaccines, peptide-based drugs, and other resources of immunotherapeutics. Finally, the applications of multi-graft and multivalent scaffolding, codon optimization and antibodyomics tools in identifying and designing in silico vaccine candidates are described.

SIV-Specific Antibodies are Elicited by a Recombinant Fowlpox Virus Co-expressing SIV Gag and envT

Given the failures of past HIV-1 vaccine clinical trials, potential HIV-1 vaccine candidates should be rigorously screened in preclinical models including simian immunodeficiency virus (SIV) primate models and small animal models. In this study, we tested the immunogenicity of a recombinant fowlpox virus (rFPV) expressing the SIV gag and SIV envT (rFPV_sg-se) proteins in BALB/c mice, to establish a foundation for further development. rFPV_sg-se was constructed through homologous recombination techniques and purified through plaque screening assays using enhanced green fluorescent protein as the reporter gene. The integration, transcription, and translation of the SIV genes were measured by PCR (genomic DNA), RT-PCR (RNA), Western-blot, respectively. The levels of SIV-specific antibodies were assessed by ELISA following a single immunization (n = 18/group) or a prime-boost strategy (n = 24/group) with rFPV_sg-se and compared to FPV and PBS controls. Residual virus was measured in distant organs following immunization using PCR. SIV-specific IgG titers against gag and gp120 were detected following single vaccination and the prime-boost. As expected the titers were higher following the prime-boost approach. The levels of Gag- and gp120-specific antibodies were significantly higher than controls (p < 0.01) 14 days after the booster immunization. Residual rFPV_Sg-Se was detected in the muscle at the site of injection, but not in distant organs, from day 1-7 post immunization. In summary, rFPV_sg-se induced high levels of SIV-specific antibodies suggesting it may be a viable candidate for further development.

Creation of artificial protein-protein interactions using α-helices as interfaces

Designing novel protein-protein interactions (PPIs) with high affinity is a challenging task. Directed evolution, a combination of randomization of the gene for the protein of interest and selection using a display technique, is one of the most powerful tools for producing a protein binder. However, the selected proteins often bind to the target protein at an undesired surface. More problematically, some selected proteins bind to their targets even though they are unfolded. Current state-of-the-art computational design methods have successfully created novel protein binders. These computational methods have optimized the non-covalent interactions at interfaces and thus produced artificial protein complexes. However, to date there are only a limited number of successful examples of computationally designed de novo PPIs. De novo design of coiled-coil proteins has been extensively performed and, therefore, a large amount of knowledge of the sequence-structure relationship of coiled-coil proteins has been accumulated. Taking advantage of this knowledge, de novo design of inter-helical interactions has been used to produce artificial PPIs. Here, we review recent progress in the in silico design and rational design of de novo PPIs and the use of α-helices as interfaces.

In silico methods for design of biological therapeutics

It has been twenty years since the first rationally designed small molecule drug was introduced into the market. Since then, we have progressed from designing small molecules to designing biotherapeutics. This class of therapeutics includes designed proteins, peptides and nucleic acids that could more effectively combat drug resistance and even act in cases where the disease is caused because of a molecular deficiency. Computational methods are crucial in this design exercise and this review discusses the various elements of designing biotherapeutic proteins and peptides. Many of the techniques discussed here, such as the deterministic and stochastic design methods, are generally used in protein design. We have devoted special attention to the design of antibodies and vaccines. In addition to the methods for designing these molecules, we have included a comprehensive list of all biotherapeutics approved for clinical use. Also included is an overview of methods that predict the binding affinity, cell penetration ability, half-life, solubility, immunogenicity and toxicity of the designed therapeutics. Biotherapeutics are only going to grow in clinical importance and are set to herald a new generation of disease management and cure.

De novo design of antibody complementarity determining regions binding a FLAG tetra-peptide

Computational antibody engineering efforts to date have focused on improving binding affinities or biophysical characteristics. De novo design of antibodies binding specific epitopes could greatly accelerate discovery of therapeutics as compared to conventional immunization or synthetic library selection strategies. Here, we employed de novo complementarity determining region (CDR) design to engineer targeted antibody-antigen interactions using previously described in silico methods. CDRs predicted to bind the minimal FLAG peptide (Asp-Tyr-Lys-Asp) were grafted onto a single-chain variable fragment (scFv) acceptor framework. Fifty scFvs comprised of designed heavy and light or just heavy chain CDRs were synthesized and screened for peptide binding by phage ELISA. Roughly half of the designs resulted in detectable scFv expression. Four antibodies, designed entirely in silico, bound the minimal FLAG sequence with high specificity and sensitivity. When reformatted as soluble antigen-binding fragments (Fab), these clones expressed well, were predominantly monomeric and retained peptide specificity. In both formats, the antibodies bind the peptide only when present at the amino-terminus of a carrier protein and even conservative peptide amino acid substitutions resulted in a complete loss of binding. These results support in silico CDR design of antibody specificity as an emerging antibody engineering strategy.

How HIV-1 entry mechanism and broadly neutralizing antibodies guide structure-based vaccine design

PURPOSE OF REVIEW

An HIV-1 vaccine that elicits broadly neutralizing antibodies (bNAbs) remains to be developed. Here, we review how knowledge of bNAbs and HIV-1 entry mechanism is guiding the structure-based design of vaccine immunogens and immunization regimens.

RECENT FINDINGS

Isolation of bNAbs from HIV-1-infected donors has led to an unprecedented understanding of the sites of vulnerability that these antibodies target on the HIV-1 envelope (Env) as well as of the immunological pathways that these antibody lineages follow to develop broad and potent neutralization. Sites of vulnerability, however, reside in the context of diverse Env conformations required for HIV-1 entry, including a prefusion-closed state, a single-CD4-bound intermediate, a three-CD4-bound intermediate, a prehairpin intermediate and postfusion states, and it is not always clear which structural state optimally presents a particular site of vulnerability in the vaccine context. Furthermore, detailed knowledge of immunological pathways has led to debate among vaccine developers as to how much of the natural antibody-developmental pathway immunogens should mimic, ranging from only the recognized epitope to multiple antigens from the antibody-virus coevolution process.

SUMMARY

A plethora of information on bNAbs is guiding HIV-1-vaccine development. We highlight consideration of the appropriate structural context from the HIV-1-entry mechanism and extraordinary progress with replicating template B-cell ontogenies.

Protein Engineering by Combined Computational and In Vitro Evolution Approaches

Two alternative strategies are commonly used to study protein-protein interactions (PPIs) and to engineer protein-based inhibitors. In one approach, binders are selected experimentally from combinatorial libraries of protein mutants that are displayed on a cell surface. In the other approach, computational modeling is used to explore an astronomically large number of protein sequences to select a small number of sequences for experimental testing. While both approaches have some limitations, their combination produces superior results in various protein engineering applications. Such applications include the design of novel binders and inhibitors, the enhancement of affinity and specificity, and the mapping of binding epitopes. The combination of these approaches also aids in the understanding of the specificity profiles of various PPIs.

Broadly Neutralizing Antibody-Guided Carbohydrate-Based HIV Vaccine Design: Challenges and Opportunities

The HIV envelope (Env) is heavily glycosylated, facilitating the spread and survival of HIV in many ways. Some potent broadly neutralizing antibodies (bnAbs) such as 2G12, PG9, PG16, and PGTs can recognize the conserved glycan residues on Env. The bnAbs, which often emerge after many years of chronic infection, provide insight into the vulnerability of HIV and can therefore guide the design of vaccines. Many carbohydrate-conjugated vaccines have been designed to induce bnAb-like antibodies, but none have yet been successful. The low antigenicity of these vaccines is one possible explanation. New strategies have been applied to obtain high-affinity antigens of glycan-dependent and other bnAbs. However, when used as immunogens in vivo, high-affinity antigens are still insufficient in eliciting bnAb-like antibodies. bnAbs generally possess some unusual features and may therefore be suppressed by the host immune system. In view of this situation, some immunization regimens based on the affinity maturation of antibodies have been tested. Herein we summarize recent studies into the design of carbohydrate-based HIV vaccines and some valuable experiences gained in work with other bnAb-based HIV vaccines.

A Framework to Simplify Combined Sampling Strategies in Rosetta

A core task in computational structural biology is the search of conformational space for low energy configurations of a biological macromolecule. Because conformational space has a very high dimensionality, the most successful search methods integrate some form of prior knowledge into a general sampling algorithm to reduce the effective dimensionality. However, integrating multiple types of constraints can be challenging. To streamline the incorporation of diverse constraints, we developed the Broker: an extension of the Rosetta macromolecular modeling suite that can express a wide range of protocols using constraints by combining small, independent modules, each of which implements a different set of constraints. We demonstrate expressiveness of the Broker through several code vignettes. The framework enables rapid protocol development in both biomolecular design and structural modeling tasks and thus is an important step towards exposing the rich functionality of Rosetta's core libraries to a growing community of users addressing a diverse set of tasks in computational biology.

Emerging Vaccine Technologies

Vaccination has proven to be an invaluable means of preventing infectious diseases by reducing both incidence of disease and mortality. However, vaccines have not been effectively developed for many diseases including HIV-1, hepatitis C virus (HCV), tuberculosis and malaria, among others. The emergence of new technologies with a growing understanding of host-pathogen interactions and immunity may lead to efficacious vaccines against pathogens, previously thought impossible.