Antibodies: Computer-Aided Prediction of Structure and Design of Function

Revolutionizing oncology: the role of Artificial Intelligence (AI) as an antibody design, and optimization tools

Antibodies play a crucial role in defending the human body against diseases, including life-threatening conditions like cancer. They mediate immune responses against foreign antigens and, in some cases, self-antigens. Over time, antibody-based technologies have evolved from monoclonal antibodies (mAbs) to chimeric antigen receptor T cells (CAR-T cells), significantly impacting biotechnology, diagnostics, and therapeutics. Although these advancements have enhanced therapeutic interventions, the integration of artificial intelligence (AI) is revolutionizing antibody design and optimization. This review explores recent AI advancements, including large language models (LLMs), diffusion models, and generative AI-based applications, which have transformed antibody discovery by accelerating de novo generation, enhancing immune response precision, and optimizing therapeutic efficacy. Through advanced data analysis, AI enables the prediction and design of antibody sequences, 3D structures, complementarity-determining regions (CDRs), paratopes, epitopes, and antigen-antibody interactions. These AI-powered innovations address longstanding challenges in antibody development, significantly improving speed, specificity, and accuracy in therapeutic design. By integrating computational advancements with biomedical applications, AI is driving next-generation cancer therapies, transforming precision medicine, and enhancing patient outcomes.

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 β-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 β-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 β-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.

Guiding protein design choices by per-residue energy breakdown analysis with an interactive web application

Recent developments in machine learning have greatly facilitated the design of proteins with improved properties. However, accurately assessing the contributions of an individual or multiple amino acid mutations to overall protein stability to select the most promising mutants remains a challenge. Knowing the specific types of amino acid interactions that improve energetic stability is crucial for finding favorable combinations of mutations and deciding which mutants to test experimentally. In this work, we present an interactive workflow for assessing the energetic contributions of single and multi-mutant designs of proteins. The energy breakdown guided protein design (ENDURE) workflow includes several key algorithms, including per-residue energy analysis and the sum of interaction energies calculations, which are performed using the Rosetta energy function, as well as a residue depth analysis, which enables tracking the energetic contributions of mutations occurring in different spatial layers of the protein structure. ENDURE is available as a web application that integrates easy-to-read summary reports and interactive visualizations of the automated energy calculations and helps users selecting protein mutants for further experimental characterization. We demonstrate the effectiveness of the tool in identifying the mutations in a designed polyethylene terephthalate (PET)-degrading enzyme that add up to an improved thermodynamic stability. We expect that ENDURE can be a valuable resource for researchers and practitioners working in the field of protein design and optimization. ENDURE is freely available for academic use at: http://endure.kuenzelab.org.

Bioprocess Development and Characterization of a 13C-Labeled Hybrid Bispecific Antibody Produced in Escherichia coli

Monoclonal antibodies (mAbs) are highly efficacious therapeutics; however, due to their large, dynamic nature, structural perturbations and regional modifications are often difficult to study. Moreover, the homodimeric, symmetrical nature of mAbs makes it difficult to elucidate which heavy chain (HC)-light chain (LC) pairs are responsible for any structural changes, stability concerns, and/or site-specific modifications. Isotopic labeling is an attractive means for selectively incorporating atoms with known mass differences to enable identification/monitoring using techniques such as mass spectrometry (MS) and nuclear magnetic resonance (NMR). However, the isotopic incorporation of atoms into proteins is typically incomplete. Here we present a strategy for incorporating ¹³C-labeling of half antibodies using an Escherichia coli fermentation system. Unlike previous attempts to generate isotopically labeled mAbs, we provide an industry-relevant, high cell density process that yielded >99% ¹³C-incorporation using ¹³C-glucose and ¹³C-celtone. The isotopic incorporation was performed on a half antibody designed with knob-into-hole technology to enable assembly with its native (naturally abundant) counterpart to generate a hybrid bispecific (BsAb) molecule. This work is intended to provide a framework for producing full-length antibodies, of which half are isotopically labeled, in order to study the individual HC-LC pairs.

Challenges in Detection of Serum Oncoprotein: Relevance to Breast Cancer Diagnostics

Breast cancer is a highly prevalent malignancy that shows improved outcomes with earlier diagnosis. Current screening and monitoring methods have improved survival rates, but the limitations of these approaches have led to the investigation of biomarker evaluation to improve early diagnosis and treatment monitoring. The enzyme-linked immunosorbent assay (ELISA) is a specific and robust technique ideally suited for the quantification of protein biomarkers from blood or its constituents. The continued clinical relevancy of this assay format will require overcoming specific technical challenges, including the ultra-sensitive detection of trace biomarkers and the circumventing of potential assay interference due to the expanding use of monoclonal antibody (mAb) therapeutics. Approaches to increasing the sensitivity of ELISA have been numerous and include employing more sensitive substrates, combining ELISA with the polymerase chain reaction (PCR), and incorporating nanoparticles as shuttles for detection antibodies and enzymes. These modifications have resulted in substantial boosts in the ability to detect extremely low levels of protein biomarkers, with some systems reliably detecting antigen at sub-femtomolar concentrations. Extensive utilization of mAb therapies in oncology has presented an additional contemporary challenge for ELISA, particularly when both therapeutic and assay antibodies target the same protein antigen. Resolution of issues such as epitope overlap and steric hindrance requires a rational approach to the design of diagnostic antibodies that takes advantage of modern antibody generation pipelines, epitope binning techniques and computational methods to strategically target biomarker epitopes. This review discusses technical strategies in ELISA implemented to date and their feasibility to address current constraints on sensitivity and problems with interference in the clinical setting. The impact of these recent advancements will depend upon their transformation from research laboratory protocols into facile, reliable detection systems that can ideally be replicated in point-of-care devices to maximize utilization and transform both the diagnostic and therapeutic monitoring landscape.

Anion solvation enhanced by positive supercharging mutations preserves thermal stability of an antibody in a wide pH range

Proteins function through interactions with other molecules. In protein engineering, scientists often engineer proteins by mutating their amino acid sequences on the protein surface to improve various physicochemical properties. "Supercharging" is a method to design proteins by mutating surface residues with charged amino acids. Previous studies demonstrated that supercharging mutations conferred better thermal resistance, solubility, and cell penetration to proteins. Likewise, antibodies recognize antigens through the antigen-binding site on the surface. The genetic and structural diversity of antibodies leads to high specificity and affinity toward antigens, enabling antibodies to be versatile tools in various applications. When assessing therapeutic antibodies, surface charge is an important factor to consider because the isoelectric point plays a role in protein clearance inside the body. In this study, we explored how supercharging mutations affect physicochemical properties of antibodies. Starting from a crystal structure of an antibody with the net charge of -4, we computationally designed a supercharged variant possessing the net charge of +10. The positive-supercharged antibody exhibited marginal improvement in thermal stability, but the secondary structure and the binding affinity to the antigen (net charge of +8) were preserved. We also used physicochemical measurements and molecular dynamics simulations to analyze the effects of supercharging mutations in sodium phosphate buffer with different pH and ion concentrations, which revealed preferential solvation of phosphate ions to the supercharged surface relative to the wild-type surface. These results suggest that supercharging would be a useful approach to preserving thermal stability of antibodies in a wide range of pH, which may enable further diversification of antibody repertoires beyond natural evolution.

Modeling Immunity with Rosetta: Methods for Antibody and Antigen Design

Structure-based antibody and antigen design has advanced greatly in recent years, due not only to the increasing availability of experimentally determined structures but also to improved computational methods for both prediction and design. Constant improvements in performance within the Rosetta software suite for biomolecular modeling have given rise to a greater breadth of structure prediction, including docking and design application cases for antibody and antigen modeling. Here, we present an overview of current protocols for antibody and antigen modeling using Rosetta and exemplify those by detailed tutorials originally developed for a Rosetta workshop at Vanderbilt University. These tutorials cover antibody structure prediction, docking, and design and antigen design strategies, including the addition of glycans in Rosetta. We expect that these materials will allow novice users to apply Rosetta in their own projects for modeling antibodies and antigens.

Third generation antibody discovery methods: in silico rational design

Owing to their outstanding performances in molecular recognition, antibodies are extensively used in research and applications in molecular biology, biotechnology and medicine. Recent advances in experimental and computational methods are making it possible to complement well-established in vivo (first generation) and in vitro (second generation) methods of antibody discovery with novel in silico (third generation) approaches. Here we describe the principles of computational antibody design and review the state of the art in this field. We then present Modular, a method that implements the rational design of antibodies in a modular manner, and describe the opportunities offered by this approach.

In silicoevaluation of the interactions among two selected single chain variable fragments (scFvs) and ESAT-6 antigen ofMycobacterium tuberculosis

Nowadays antibody engineering is an important approach in the design and manufacture of therapeutic and diagnostic antibodies. The study of interactions between antibodies and antigens is the critical step in the design of antibodies with desirable properties. Computational docking is a useful tool for structural characterization of bimolecular interactions. Docking is the process of predicting bound conformations and binding enthalpy of antibody–antigen complexes. In this study, the three-dimensional structures of two ribosome displayed-selected scFv antibodies were constructed by Kotai Antibody Builder. By using ClusPro 2.0 web server, the ESAT-6 antigen (a tuberculosis-specific antigen) structure was docked to both scFv models to obtain the structures of the binding complexes and molecular dynamics (MD) simulations were performed using GROMACS 4.5.3 package. By analyzing of the ESAT-scFv complexes, important amino acids involved in antigen–antibody interactions were identified which were Asn164 in VL3, Ser164 in VL7 and Asn55 in VH7. All three amino acids belonged to the CDRs. In conclusion, results achieved from this bioinformatics study can help in the design and development of novel antibodies with improved affinities for tuberculosis diagnosis.

Structural constraints-based evaluation of immunogenic avirulent toxins from Clostridium botulinum C2 and C3 toxins as subunit vaccines

Clostridium botulinum (group-III) is an anaerobic bacterium producing C2 and C3 toxins in addition to botulinum neurotoxins in avian and mammalian cells. C2 and C3 toxins are members of bacterial ADP-ribosyltransferase superfamily, which modify the eukaryotic cell surface proteins by ADP-ribosylation reaction. Herein, the mutant proteins with lack of catalytic and pore forming function derived from C2 (C2I and C2II) and C3 toxins were computationally evaluated to understand their structure-function integrity. We have chosen many structural constraints including local structural environment, folding process, backbone conformation, conformational dynamic sub-space, NAD-binding specificity and antigenic determinants for screening of suitable avirulent toxins. A total of 20 avirulent mutants were identified out of 23 mutants, which were experimentally produced by site-directed mutagenesis. No changes in secondary structural elements in particular to α-helices and β-sheets and also in fold rate of all-β classes. Structural stability was maintained by reordered hydrophobic and hydrogen bonding patterns. Molecular dynamic studies suggested that coupled mutations may restrain the binding affinity to NAD(+) or protein substrate upon structural destabilization. Avirulent toxins of this study have stable energetic backbone conformation with a common blue print of folding process. Molecular docking studies revealed that avirulent mutants formed more favorable hydrogen bonding with the side-chain of amino acids near to conserved NAD-binding core, despite of restraining NAD-binding specificity. Thus, structural constraints in the avirulent toxins would determine their immunogenic nature for the prioritization of protein-based subunit vaccine/immunogens to avian and veterinary animals infected with C. botulinum.