Analytical Method for Experimental Validation of Computer-Designed Antibody

Accelerating antibody discovery and optimization with high-throughput experimentation and machine learning

The integration of high-throughput experimentation and machine learning is transforming data-driven antibody engineering, revolutionizing the discovery and optimization of antibody therapeutics. These approaches employ extensive datasets comprising antibody sequences, structures, and functional properties to train predictive models that enable rational design. This review highlights the significant advancements in data acquisition and feature extraction, emphasizing the necessity of capturing both sequence and structural information. We illustrate how machine learning models, including protein language models, are used not only to enhance affinity but also to optimize other crucial therapeutic properties, such as specificity, stability, viscosity, and manufacturability. Furthermore, we provide practical examples and case studies to demonstrate how the synergy between experimental and computational approaches accelerates antibody engineering. Finally, this review discusses the remaining challenges in fully realizing the potential of artificial intelligence (AI)-powered antibody discovery pipelines to expedite therapeutic development.

FASTIA: A rapid and accessible platform for protein variant interaction analysis demonstrated with a single-domain antibody

Antibodies are critical tools in medicine and research, and their affinity for their target antigens is a key determinant of their efficacy. Traditional antibody affinity maturation and interaction analyses are often hampered by time-consuming steps such as cloning, expression, purification, and interaction assays. To address this, we have developed FASTIA (Fast Affinity Screening Technology for Interaction Analysis), a novel platform that integrates rapid gene fragment preparation, cell-free protein synthesis, and bio-layer interferometry with non-regenerative analysis. Using this approach, we can analyze the intermolecular interactions of over 20 variants over 2 days, requiring only the parent protein expression plasmid and basic equipment. We have demonstrated the ability of FASTIA to discriminate between single-domain antibody variants with different binding affinities using the anti-HEL VHH antibody D2-L29, and mapped the results to the crystal structure to identify key interaction sites. FASTIA provides results comparable to those obtained using traditional methods. Our system bypasses the need for genetic engineering facilities and can be easily adopted by laboratories, accelerating the protein engineering and optimization processes. In addition, FASTIA is applicable to other protein-protein interactions, making it a versatile tool for studying molecular recognition. FASTIA facilitates efficient affinity maturation, protein engineering, and analysis of protein-protein interactions. This provides a rapid and accessible route for improving antibodies and a broader understanding of protein interactions.

BthTX-I, a phospholipase A2-like toxin, is inhibited by the plant cinnamic acid derivative: chlorogenic acid

Snakebite is a significant health concern in tropical and subtropical regions, particularly in Africa, Asia, and Latin America, resulting in more than 2.7 million envenomations and an estimated one hundred thousand fatalities annually. The Bothrops genus is responsible for the majority of snakebite envenomings in Latin America and Caribbean countries. Accidents involving snakes from this genus are characterized by local symptoms that often lead to permanent sequelae and death. However, specific antivenoms exhibit limited effectiveness in inhibiting local tissue damage. Phospholipase A2-like (PLA2-like) toxins emerge as significant contributors to local myotoxicity in accidents involving Bothrops species. As a result, they represent a crucial target for prospective treatments. Some natural and synthetic compounds have shown the ability to reduce or abolish the myotoxic effects of PLA2-like proteins. In this study, we employed a combination approach involving myographic, morphological, biophysical and bioinformatic techniques to investigate the interaction between chlorogenic acid (CGA) and BthTX-I, a PLA2-like toxin. CGA provided a protection of 71.8% on muscle damage in a pre-incubation treatment. Microscale thermophoresis and circular dichroism experiments revealed that CGA interacted with the BthTX-I while preserving its secondary structure. CGA exhibited an affinity to the toxin that ranks among the highest observed for a natural compound. Bioinformatics simulations indicated that CGA inhibitor binds to the toxin's hydrophobic channel in a manner similar to other phenolic compounds previously investigated. These findings suggest that CGA interferes with the allosteric transition of the non-activated toxin, and the stability of the dimeric assembly of its activated state.