Antibody Humanization via Computational Framework Grafting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional antibody humanization methods often result in substantial decreases in stability, expression levels, and antigen binding affinity, requiring iterative mutational fine-tuning to recover the original antibody's properties.
Innovation Solution
A computational method, CUMAb, that grafts animal CDRs onto thousands of human frameworks using Rosetta atomistic simulations to rank designs by energy and structural integrity, expanding humanization from a few dozen frameworks to over 20,000.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional humanization methods graft CDRs onto human frameworks, then humanness of the antibody is improved, but stability, expression levels, and antigen binding affinity deteriorate
Solution Approach 1:
The patent applies parameter changes by systematically varying framework residues at positions surrounding the CDRs, particularly in the VH and VL frameworks. Instead of using fixed human framework sequences, the method explores multiple sequence variants with different amino acid compositions at key positions, thereby optimizing both humanness and structural stability simultaneously
Solution Approach 2:
The patent employs preliminary action through in silico modeling and computational prediction to identify favorable framework residue combinations before experimental implementation. The method uses structural modeling to predict which framework variants will maintain CDR conformation and binding affinity, allowing selection of optimal candidates prior to wet lab validation
2Reliability
If iterative backmutation is performed to restore parental antibody properties, then stability and affinity are improved, but development time and complexity increase
Solution Approach 1:
The patent performs preliminary computational optimization of framework residues before experimental testing, using in silico models to predict stable configurations. This preliminary design phase identifies favorable mutations that maintain binding affinity while achieving humanization, eliminating the need for multiple iterative rounds of experimental backmutation and validation
Solution Approach 2:
The patent uses computational copying of successful framework residue patterns from known stable antibodies. By identifying and replicating favorable amino acid compositions and structural features from validated antibody frameworks, the method accelerates the design process without requiring extensive iterative experimentation
3Ease of manufacture
If CDRs are grafted onto human frameworks with high sequence identity, then ease of framework selection is improved, but structural integrity and CDR compatibility deteriorate
Solution Approach 1:
The patent applies parameter changes by evaluating framework sequences based on multiple criteria beyond simple sequence identity, including amino acid composition at specific positions, predicted secondary structure elements, and spatial arrangement of framework residues. This multi-parameter evaluation identifies frameworks that maintain structural integrity while achieving adequate humanization
Data Source
AI summary
Provided herein is a method for obtaining a humanized antibody based on a non-human antibody having an affinity to an antigen of interest, that starts from an experimental or model structure, grafts the non-human CDRs on a library of human frameworks and uses restrained/constrained atomistic simulations to relax and rank the designs by energy.