Methods for characterizing and engineering protein-protein interactions
Inventors
YOUNGER, David • Colby, David • Lopez, Randolph • Wittekind, Michael
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11726097-B2
Publication Date
2023-08-15
Expiration Date
Abstract
Characterization of the binding dynamics at the interface between any two proteins that specifically interact plays a role in myriad biomedical applications. The methods disclosed herein provide for the high-throughput characterization of the specific interaction at the interface between two protein binding partners and the identification of functionally significant mutations of one or both protein binding partners. For example, the methods disclosed herein may be useful for epitope and paratope mapping of an antibody-antigen pair, which is useful for the discovery and development of novel therapies, vaccines, diagnostics, among other biomedical applications.
Core Innovation
The invention provides a method for identifying compensatory mutations between two mutant protein binding partners using first and second protein binding partner libraries expressed on haploid yeast. The first library includes a first wild-type polypeptide and a plurality of mutant polypeptides, and the second library includes a second wild-type polypeptide and a plurality of mutant polypeptides. The method culturing haploid yeast cells such that diploid yeast cells are produced if the first and second protein binding partners interact.
The method measures an observed affinity value between each protein binding partner of the first library and each protein binding partner of the second library. Compensatory mutations are identified based on whether the observed affinity value for a mutant-mutant pair is substantially different from a respective expected affinity value. The expected affinity value for a given pair is calculated from observed affinity values for corresponding wild-type-mutant relationships, using the observed affinity value between the first wild-type polypeptide and a mutant polypeptide, and the observed affinity value between a mutant polypeptide and the second wild-type polypeptide.
In the described examples, the approach is applied to protein interaction characterization and engineering workflows that measure binding affinities between antibody/antigen, receptor/ligand, and other protein binding partners. Library-on-library screening is used to detect correct cognate pairs and to map mutational intolerance and interface-relevant residues, while identifying compensatory mutation pairs whose observed affinity differs from an expected value derived from wild-type-mutant effects. The document links affinity or intensity readouts to yeast mating frequency via synthetic agglutination and describes scalable assay capacity and epitope/paratope mapping workflows.
Claims Coverage
The independent claim is directed to a method for identifying compensatory mutations between two mutant protein binding partners using haploid yeast libraries and a calculated expected affinity based on corresponding wild-type-mutant observations. The inventive features are expressed in the independent claim and further refined by dependent claims specifying measurement by synthetic agglutination, quantitative directions/thresholds for the observed-versus-expected difference, selection of binding partner types, and an orthogonal binding relationship constraint.
Compensatory mutation identification via observed-versus-expected affinity calculation
Identifying compensatory mutations between one or more pairs of mutant polypeptides, wherein an observed affinity value for a mutant-mutant pair is substantially different than a respective expected affinity value between the respective pair, and wherein the expected affinity value for a given pair is calculated based on observed affinity values for corresponding wild-type-mutant pairs.
Interaction-dependent diploid yeast production from first and second haploid yeast libraries
Providing a first library of first protein binding partners on a surface of a first plurality of haploid yeast cells and providing a second library of second protein binding partners on a surface of a second plurality of haploid yeast cells, and culturing the first plurality and second plurality such that diploid yeast cells are produced if the first and second protein binding partners interact.
Library-by-library measurement of observed affinity values
Measuring an observed affinity value between each protein binding partner of the first library and each protein binding partner of the second library.
Synthetic agglutination-based observed affinity measurement
Measuring observed affinity values between protein binding partners from two libraries by performing synthetic agglutination using two pluralities of haploid yeast cells.
Observed affinity substantially lower than expected by a quantitative threshold
Identifying pairs where the observed affinity value is lower than the expected affinity value by about 50% or less.
Observed affinity substantially higher than expected
Identifying pairs where the observed affinity value is substantially higher than the corresponding expected affinity value.
Antibody-format first protein binding partners
The first library of first protein binding partners is selected as antibodies, scFvs, Fabs, or VHH species.
Orthogonal binding relationship condition for mutant-mutant pairing
Mutating polypeptides so that the mutated polypeptides form an orthogonal binding pair, where each mutant binds the other mutant while not binding the corresponding wild-type polypeptide.
Across the independent claim and its refinements, the claimed approach identifies compensatory mutation pairs by comparing a measured observed affinity for mutant-mutant combinations to an expected affinity calculated from corresponding wild-type-mutant observed affinities, with optional refinement to synthetic agglutination measurement, quantitative directions/thresholds of the observed-versus-expected difference, narrowed protein-binding-partner types, and an orthogonal binding relationship criterion.
Stated Advantages
Identifies compensatory mutations between two mutant protein binding partners by using a calculated expected affinity based on corresponding wild-type-mutant observations.
Detects compensatory mutation pairs where observed affinity between mutant-mutant combinations is substantially different from expected values computed from observed wild-type-mutant effects.
Supports measurement of affinity or intensity readouts linked to yeast mating frequency via synthetic agglutination.
Provides a scalable assay capacity and supports epitope/paratope mapping workflows.
Enables identification of correct antibody-antigen cognate pairs with low cross-reactivity.
Maps mutational intolerance and interface-relevant residues at PD-1/pembrolizumab.
Documented Applications
Characterizing and engineering protein-protein interactions, including antibody-antigen binding and receptor-ligand binding.
Epitope and paratope mapping workflows using library-on-library screening.
Applications related to CAR-T cell therapy are referenced in the document context.
Application to PD-1/pembrolizumab interaction characterization, including mapping interface-relevant residues and identifying compensatory mutation pairs.
Interested in licensing this patent?