On the humanization of VHHs: Prospective case studies, experimental and computational characterization of structural determinants for functionality.

The humanization of camelid-derived variable domain heavy chain antibodies (VHHs) poses challenges including immunogenicity, stability, and potential reduction of affinity. Critical to this process are complementarity-determining regions (CDRs), Vernier and Hallmark residues, shaping the three-dimensional fold and influencing VHH structure and function. Additionally, the presence of non-canonical disulfide bonds further contributes to conformational stability and antigen binding. In this study, we systematically humanized two camelid-derived VHHs targeting the natural cytotoxicity receptor NKp30. Key structural positions in Vernier and Hallmark regions were exchanged with residues from the most similar human germline sequences. The resulting variants were characterized for binding affinities, yield, and purity. Structural binding modes were elucidated through crystal structure determination and AlphaFold2 predictions, providing insights into differences in binding affinity. Comparative structural and molecular dynamics characterizations of selected variants were performed to rationalize their functional properties and elucidate the role of specific sequence motifs in antigen binding. Furthermore, systematic analyses of next-generation sequencing (NGS) and Protein Data Bank (PDB) data was conducted, shedding light on the functional significance of Hallmark motifs and non-canonical disulfide bonds in VHHs in general. Overall, this study provides valuable insights into the structural determinants governing the functional properties of VHHs, offering a roadmap for their rational design, humanization, and optimization for therapeutic applications.

Human antibody polyreactivity is governed primarily by the heavy-chain complementarity-determining regions.

Although antibody variable regions mediate antigen-specific binding, they can also mediate non-specific interactions with non-cognate antigens, impacting diverse immunological processes and the efficacy, safety, and half-life of antibody therapeutics. To understand the molecular basis of antibody non-specificity, we sorted two dissimilar human naïve antibody libraries against multiple reagents to enrich for variants with different levels of polyreactivity. Sequence analysis of >300,000 paired antibody variable regions revealed that the heavy chain primarily mediates human antibody polyreactivity, and this is due to the high positive charge, high hydrophobicity, and combinations thereof in the corresponding complementarity-determining regions, which can be predicted using a machine learning model developed in this work. Notably, a subset of the most important features governing antibody non-specific interactions, namely those that contain tyrosine, also govern specific antigen recognition. Our findings are broadly relevant for understanding fundamental aspects of antibody molecular recognition and the applied aspects of antibody-drug design.

A Simple Analysis of the Second (Extra) Disulfide Bridge of VHHs.

Camelids produce a special type of antibody, known as VHHs, that has lost the VL domain, providing a more optimised VH domain. This particular highly stable antibody domain has interesting properties for biotechnological development. Ordinarily, those molecules possess only one disulphide bridge, but surprisingly, at least a quarter of these VHHs have a second one. Curiously, this does not seem to be essential for the stability and the function of this domain. In an attempt to characterise precisely the role and impact of this disulphide bridge at the molecular level, several in silico mutants of a VHH were created to disrupt this second disulphide bridge and those systems were submitted to molecular dynamics simulation. The loss of the second disulphide bridge leads to an increase in the flexibility of CDR1 and CDR3 and an unexpected rigidification of CDR2. Local conformational analysis shows local differences in the observed protein conformations. However, in fact, there is no exploration of new conformations but a change in the equilibrium between the different observed conformations. This explains why the interaction of VHHs is not really affected by the presence or absence of this second bridge, but their rigidity is slightly reduced. Therefore, the additional disulphide bridge does not seem to be an essential part of VHHs function.

Functional Divergence in the Affinity and Stability of Non-Canonical Cysteines and Non-Canonical Disulfide Bonds: Insights from a VHH and VNAR Study.

Single-domain antibodies, including variable domains of the heavy chains of heavy chain-only antibodies (VHHs) from camelids and variable domains of immunoglobulin new antigen receptors (VNARs) from cartilaginous fish, show the therapeutic potential of targeting antigens in a cytosol reducing environment. A large proportion of single-domain antibodies contain non-canonical cysteines and corresponding non-canonical disulfide bonds situated on the protein surface, rendering them vulnerable to environmental factors. Research on non-canonical disulfide bonds has been limited, with a focus solely on VHHs and utilizing only cysteine mutations rather than the reducing agent treatment. In this study, we examined an anti-lysozyme VNAR and an anti-BC2-tag VHH, including their non-canonical disulfide bond reduced counterparts and non-canonical cysteine mutants. Both the affinity and stability of the VNARs and VHHs decreased in the non-canonical cysteine mutants, whereas the reduced-state samples exhibited decreased thermal stability, with their affinity remaining almost unchanged regardless of the presence of reducing agents. Molecular dynamics simulations suggested that the decrease in affinity of the mutants resulted from increased flexibility of the CDRs, the disappearance of non-canonical cysteine-antigen interactions, and the perturbation of other antigen-interacting residues caused by mutations. These findings highlight the significance of non-canonical cysteines for the affinity of single-domain antibodies and demonstrate that the mutation of non-canonical cysteines is not equivalent to the disruption of non-canonical disulfide bonds with a reducing agent when assessing the function of non-canonical disulfide bonds.

Discovery of potent allosteric antibodies inhibiting EGFR.

In this work, we report the discovery of potent anti-epidermal growth factor receptor (EGFR) allosteric heavy-chain antibodies by combining camelid immunization and fluorescence-activated cell sorting (FACS). After immunization and yeast surface display library construction, allosteric clones were obtained by introducing the labeled EGF Fc fusion protein as an additional criterion for FACS. This sorting method enabled the identification of 11 heavy-chain antibodies that did not compete with the orthosteric ligand EGF for the binding to EGFR. These antibodies bind to a triple-negative breast cancer cell line expressing EGFR with affinities in the picomolar to nanomolar range. Those camelid-derived antibodies also exhibit interesting properties by modulating EGFR affinity for EGF. Moreover, they are also able to inhibit EGF-induced downstream signaling pathways. In particular, we identified one clone that is more potent than the approved blocking antibody cetuximab in inhibiting both PI3K/AKT and MAPK/ERK pathways. Our results suggest that allosteric antibodies may be potential new modalities for therapeutics.

Development of novel humanized VHH synthetic libraries based on physicochemical analyses.

Due to the high affinity and specificity of antibodies toward antigens, various antibody-based applications have been developed. Recently, variable antigen-binding domains of heavy-chain antibodies (VHH) have become an attractive alternative to conventional fragment antibodies due to their unique molecular characteristics. As an antibody-generating strategy, synthetic VHH libraries (including humanized VHH libraries) have been developed using distinct strategies to constrain the diversity of amino acid sequences. In this study, we designed and constructed several novel synthetic humanized VHH libraries based on biophysical analyses conducted using the complementarity determining region-grafting method and comprehensive sequence analyses of VHHs deposited in the protein data bank. We obtained VHHs from the libraries, and hit clones exhibited considerable thermal stability. We also found that VHHs from distinct libraries tended to have different epitopes. Based on our results, we propose a strategy for generating humanized VHHs with distinct epitopes toward various antigens by utilizing our library combinations.

Antibody-SGM, a Score-Based Generative Model for Antibody Heavy-Chain Design.

Traditional computational methods for antibody design involved random mutagenesis followed by energy function assessment for candidate selection. Recently, diffusion models have garnered considerable attention as cutting-edge generative models, lauded for their remarkable performance. However, these methods often focus solely on the backbone or sequence, resulting in the incomplete depiction of the overall structure and necessitating additional techniques to predict the missing component. This study presents Antibody-SGM, an innovative joint structure-sequence diffusion model that addresses the limitations of existing protein backbone generation models. Unlike previous models, Antibody-SGM successfully integrates sequence-specific attributes and functional properties into the generation process. Our methodology generates full-atom native-like antibody heavy chains by refining the generation to create valid pairs of sequences and structures, starting with random sequences and structural properties. The versatility of our method is demonstrated through various applications, including the design of full-atom antibodies, antigen-specific CDR design, antibody heavy chains optimization, validation with Alphafold3, and the identification of crucial antibody sequences and structural features. Antibody-SGM also optimizes protein function through active inpainting learning, allowing simultaneous sequence and structure optimization. These improvements demonstrate the promise of our strategy for protein engineering and significantly increase the power of protein design models.

The impact of exchanging the light and heavy chains on the structures of bovine ultralong antibodies.

The third complementary-determining regions of the heavy-chain (CDR3H) variable regions (VH) of some cattle antibodies are highly extended, consisting of 48 or more residues. These `ultralong' CDR3Hs form ß-ribbon stalks that protrude from the surface of the antibody with a disulfide cross-linked knob region at their apex that dominates antigen interactions over the other CDR loops. The structure of the Fab fragment of a naturally paired bovine ultralong antibody (D08), identified by single B-cell sequencing, has been determined to 1.6 Šresolution. By swapping the D08 native light chain with that of an unrelated antigen-unknown ultralong antibody, it is shown that interactions between the CDR3s of the variable domains potentially affect the fine positioning of the ultralong CDR3H; however, comparison with other crystallographic structures shows that crystalline packing is also a major contributor. It is concluded that, on balance, the exact positioning of ultralong CDR3H loops is most likely to be due to the constraints of crystal packing.

Mechanistic Insights into How the Single Point Mutation Change the Autoantibody Repertoire.

A recent study showed that just one point mutation F33 to Y in the complementarity-determining region 1 of heavy chain (H-CDR1) could lead to the auto-antibody losing its DNA binding ability. However, the potential molecular mechanisms have not been well elucidated. In this study, we investigated how the antibody lost the DNA binding ability caused by mutation F33 to Y in the H-CDR1. We found that the electrostatic force was not the primary driving force for the interaction between anti-DNA antibodies and the antigen single strand DNA (ssDNA), and that the H-CDR2 largely contributed to the binding of antigen ssDNA, even larger than H-CDR1. The H-F33Y mutation could increase the hydrogen-bond interaction but impair the pi-pi stacking interaction between the antibody and ssDNA. We further found that F33H, W98H and Y95L in the wiletype antibody could form the stable pi-pi stacking interaction with the nucleotide bases of ssDNA. However, the Y33 in mutant could not form the parallel sandwich pi-pi stacking interaction with the ssDNA. To further confirm the importance of pi-pi stacking, the wildtype antibody and the mutants (F33YH, F33AH, W98AH and Y95AL) were experimentally expressed in CHO cells and purified, and the results from ELISA clearly showed that all the mutants lost the ssDNA binding ability. Taken together, our findings may not only deepen the understanding of the underlying interaction mechanism between autoantibody and antigen, but also broad implications in the field of antibody engineer.

A single-domain antibody library based on a stability-engineered human VH3 scaffold.

Using conventional immunoglobulin G (IgG) molecules as therapeutic agents presents several well-known disadvantages owing to their large size and structural complexity, negatively impacting development and production efficiency. Single-domain antibodies (sdAbs) are the smallest functional antibody format (~ 15 kDa) and represent a viable alternative to IgG in many applications. However, unlike natural single-domain antibodies, such as camelid VHH, the variable domains of conventional antibodies show poor physicochemical properties when expressed as sdAbs. This report identified stable sdAb variants of human VH3-23 from a framework region 2-randomized human VH library by phage display selection under thermal challenge. Synthetic complementarity determining region diversity was introduced to one of the selected variants with high thermal stability, expression level, and monomeric content to construct a human VH sdAb library. The library was validated by panning against a panel of antigens, and target-specific binders were identified and characterized for their affinity and biophysical properties. The results of this study suggest that a synthetic sdAb library based on a stability-engineered human VH scaffold could be a facile source of high-quality sdAb for many practical applications.

Structural, Biophysical, and Computational Studies of a Murine Light Chain Dimer.

Antibodies are widely used in medicinal and scientific research due to their ability to bind to a specific antigen. Most often, antibodies are composed of heavy and light chain domains. Under physiological conditions, light chains are produced in excess, as compared to the heavy chain. It is now known that light chains are not silent partners of the heavy chain and can modulate the immune response independently. In this work, the first crystal structure of a light chain dimer originating from mice is described. It represents the light chain dimer of 6A8, a monoclonal antibody specific to the allergen Der f 1. Building on the unexpected occurrence of this kind of dimer, we have demonstrated that this light chain is stable in solution alone. Moreover, enzyme-linked immunosorbent assays (ELISA) have revealed that, when the light chain is not partnered to its corresponding heavy chain, it interacts non-specifically with a wide range of proteins. Computational studies were used to provide insight on the role of the 6A8 heavy chain domain in the specific binding to Der f 1. Overall, this work demonstrates and supports the ongoing notion that light chains can function by themselves and are not silent partners of heavy chains.

Anti-Idiotypic VHHs and VHH-CAR-T Cells to Tackle Multiple Myeloma: Different Applications Call for Different Antigen-Binding Moieties.

CAR-T cell therapy is at the forefront of next-generation multiple myeloma (MM) management, with two B-cell maturation antigen (BCMA)-targeted products recently approved. However, these products are incapable of breaking the infamous pattern of patient relapse. Two contributing factors are the use of BCMA as a target molecule and the artificial scFv format that is responsible for antigen recognition. Tackling both points of improvement in the present study, we used previously characterized VHHs that specifically target the idiotype of murine 5T33 MM cells. This idiotype represents one of the most promising yet challenging MM target antigens, as it is highly cancer- but also patient-specific. These VHHs were incorporated into VHH-based CAR modules, the format of which has advantages compared to scFv-based CARs. This allowed a side-by-side comparison of the influence of the targeting domain on T cell activation. Surprisingly, VHHs previously selected as lead compounds for targeted MM radiotherapy are not the best (CAR-) T cell activators. Moreover, the majority of the evaluated VHHs are incapable of inducing any T cell activation. As such, we highlight the importance of specific VHH selection, depending on its intended use, and thereby raise an important shortcoming of current common CAR development approaches.

Anti-idiotype isolation of a broad and potent influenza A virus-neutralizing human antibody.

The VH6-1 class of antibodies includes some of the broadest and most potent antibodies that neutralize influenza A virus. Here, we elicit and isolate anti-idiotype antibodies against germline versions of VH6-1 antibodies, use these to sort human leukocytes, and isolate a new VH6-1-class member, antibody L5A7, which potently neutralized diverse group 1 and group 2 influenza A strains. While its heavy chain derived from the canonical IGHV6-1 heavy chain gene used by the class, L5A7 utilized a light chain gene, IGKV1-9, which had not been previously observed in other VH6-1-class antibodies. The cryo-EM structure of L5A7 in complex with Indonesia 2005 hemagglutinin revealed a nearly identical binding mode to other VH6-1-class members. The structure of L5A7 bound to the isolating anti-idiotype antibody, 28H6E11, revealed a shared surface for binding anti-idiotype and hemagglutinin that included two critical L5A7 regions: an FG motif in the third heavy chain-complementary determining region (CDR H3) and the CDR L1 loop. Surprisingly, the chemistries of L5A7 interactions with hemagglutinin and with anti-idiotype were substantially different. Overall, we demonstrate anti-idiotype-based isolation of a broad and potent influenza A virus-neutralizing antibody, revealing that anti-idiotypic selection of antibodies can involve features other than chemical mimicry of the target antigen.

De Novo Synthesis and Structural Elucidation of CDR-H3 Loop Mimics.

The binding affinity of antibodies to specific antigens stems from a remarkably broad repertoire of hypervariable loops known as complementarity-determining regions (CDRs). While recognizing the pivotal role of the heavy-chain 3 CDRs (CDR-H3s) in maximizing antibody-antigen affinity and specificity, the key structural determinants responsible for their adaptability to diverse loop sequences, lengths, and noncanonical structures are hitherto unknown. To address this question, we achieved a de novo synthesis of bulged CDR-H3 mimics excised from their full antibody context. CD and NMR data revealed that these stable standalone ß-hairpin scaffolds are well-folded and retain many of the native bulge CDR-H3 features in water. In particular, the tryptophan residue, highly conserved across CDR-H3 sequences, was found to extend the kinked base of these ß-bulges through a combination of stabilizing intramolecular hydrogen bond and CH/π interaction. The structural ensemble consistent with our NMR observations exposed the dynamic nature of residues at the base of the loop, suggesting that ß-bulges act as molecular hinges connecting the rigid stem to the more flexible loops of CDR-H3s. We anticipate that this deeper structural understanding of CDR-H3s will lay the foundation to inform the design of antibody drugs broadly and engineer novel CDR-H3 peptide scaffolds as therapeutics.

An important consideration when expressing mAbs in Escherichiacoli.

Monoclonal antibodies (mAbs) are a driving force in the biopharmaceutical industry. Therapeutic mAbs are usually produced in mammalian cells, but there has been a push towards the use of alternative production hosts, such as Escherichia coli. When the genes encoding for a mAb heavy and light chains are codon-optimized for E. coli expression, a truncated form of the heavy chain can form along with the full-length product. In this work, the role of codon optimization in the formation of a truncated product was investigated. This study used the amino acid sequences of several therapeutic mAbs and multiple optimization algorithms. It was found that several algorithms incorporate sequences that lead to a truncated product. Approaches to avoid this truncated form are discussed.

Characterization of a novel format scFv×VHH single-chain biparatopic antibody against metal binding protein MtsA.

Biparatopic antibodies (bpAbs) are engineered antibodies that bind to multiple different epitopes within the same antigens. bpAbs comprise diverse formats, including fragment-based formats, and choosing the appropriate molecular format for a desired function against a target molecule is a challenging task. Moreover, optimizing the design of constructs requires selecting appropriate antibody modalities and adjusting linker length for individual bpAbs. Therefore, it is crucial to understand the characteristics of bpAbs at the molecular level. In this study, we first obtained single-chain variable fragments and camelid heavy-chain variable domains targeting distinct epitopes of the metal binding protein MtsA and then developed a novel format single-chain bpAb connecting these fragment antibodies with various linkers. The physicochemical properties, binding activities, complex formation states with antigen, and functions of the bpAb were analyzed using multiple approaches. Notably, we found that the assembly state of the complexes was controlled by a linker and that longer linkers tended to form more compact complexes. These observations provide detailed molecular information that should be considered in the design of bpAbs.

Engineering a Low-Immunogenic Mirror-Image VHH against Vascular Endothelial Growth Factor.

Immunogenicity is a major caveat of protein therapeutics. In particular, the long-term administration of protein therapeutic agents leads to the generation of antidrug antibodies (ADAs), which reduce drug efficacy while eliciting adverse events. One promising solution to this issue is the use of mirror-image proteins consisting of d-amino acids, which are resistant to proteolytic degradation in immune cells. We have recently reported the chemical synthesis of the enantiomeric form of the variable domain of the antibody heavy chain (d-VHH). However, identifying mirror-image antibodies capable of binding to natural ligands remains challenging. In this study, we developed a novel screening platform to identify a d-VHH specific for vascular endothelial growth factor A (VEGF-A). We performed mirror-image screening of two newly constructed synthetic VHH libraries displayed on T7 phage and identified VHH sequences that effectively bound to the mirror-image VEGF-A target (d-VEGF-A). We subsequently synthesized a d-VHH candidate that preferentially bound the native VEGF-A (l-VEGF-A) with submicromolar affinity. Furthermore, immunization studies in mice demonstrated that this d-VHH elicited no ADAs, unlike its corresponding l-VHH. Our findings highlight the utility of this novel d-VHH screening platform in the development of protein therapeutics exhibiting both reduced immunogenicity and improved efficacy.

For antibody sequence generative modeling, mixture models may be all you need.

MOTIVATION: Antibody therapeutic candidates must exhibit not only tight binding to their target but also good developability properties, especially low risk of immunogenicity. RESULTS: In this work, we fit a simple generative model, SAM, to sixty million human heavy and seventy million human light chains. We show that the probability of a sequence calculated by the model distinguishes human sequences from other species with the same or better accuracy on a variety of benchmark datasets containing >400 million sequences than any other model in the literature, outperforming large language models (LLMs) by large margins. SAM can humanize sequences, generate new sequences, and score sequences for humanness. It is both fast and fully interpretable. Our results highlight the importance of using simple models as baselines for protein engineering tasks. We additionally introduce a new tool for numbering antibody sequences which is orders of magnitude faster than existing tools in the literature. AVAILABILITY AND IMPLEMENTATION: All tools developed in this study are available at https://github.com/Wang-lab-UCSD/AntPack.

Applications and challenges in designing VHH-based bispecific antibodies: leveraging machine learning solutions.

The development of bispecific antibodies that bind at least two different targets relies on bringing together multiple binding domains with different binding properties and biophysical characteristics to produce a drug-like therapeutic. These building blocks play an important role in the overall quality of the molecule and can influence many important aspects from potency and specificity to stability and half-life. Single-domain antibodies, particularly camelid-derived variable heavy domain of heavy chain (VHH) antibodies, are becoming an increasingly popular choice for bispecific construction due to their single-domain modularity, favorable biophysical properties, and potential to work in multiple antibody formats. Here, we review the use of VHH domains as building blocks in the construction of multispecific antibodies and the challenges in creating optimized molecules. In addition to exploring traditional approaches to VHH development, we review the integration of machine learning techniques at various stages of the process. Specifically, the utilization of machine learning for structural prediction, lead identification, lead optimization, and humanization of VHH antibodies.

Recombinant Production and Characterization of VHHs/Nanobodies Targeting Tau to Block Fibrillar Assembly.

Tau protein was extensively studied using nuclear magnetic resonance spectroscopy, providing a powerful way to determine interaction sites between Tau and partner proteins. Here we used this analytical tool to describe the epitopes of Tau-specific VHHs (variable domain of the heavy chain of the heavy chain-only antibodies, aka nanobodies) selected from a synthetic library. An in vitro Tau aggregation assay was subsequently used as a functional screen to check VHH efficacy as aggregation inhibitors. We have observed a correlation between the targeted epitope and the aggregation-inhibition capacity of a series of Tau-specific VHHs.