Novel proteins with targeted binding
AssigneeAvidia Research Institute
InventorsJoost Kolkman, Willem Stemmer, Per-Ola Freskgard
Life of this patent
Estimating the term…
Abstract
Methods for identifying discrete monomer domains and immuno-domains with a desired property are provided. Methods for generating multimers from two or more selected discrete monomer domains are also provided, along with methods for identifying multimers possessing a desired property. Presentation systems are also provided which present the discrete monomer and/or immuno-domains, selected monomer and/or immuno-domains, multimers and/or selected multimers to allow their selection. Compositions, libraries and cells that express one or more library member, along with kits and integrated systems, are also included in the present invention.
Claims97
A method for identifying a monomer domain that binds to a target molecule, the method comprising,
- providing a library of monomer domains, wherein the monomer domains each bind an ion;
- screening the library of monomer domains for affinity to a first target molecule;
- and
- identifying at least one monomer domain that binds to at least one target molecule.
A method for identifying a multimer that binds to at least one target molecule, the method comprising:
- providing a library of multimers, wherein each multimer comprises at least two monomer domains and each monomer domain exhibits a binding specificity for a target molecule;
- and
- screening the library of multimers for target molecule-binding multimers.
A method for identifying a multimer that binds to a target molecule, the method comprising,
- providing a library of monomer domains and/or immuno domains;
- screening the library of monomer domains and/or immuno domain for affinity to a first target molecule;
- and
- identifying at least one monomer domain and/or immuno domain that binds to at least one target molecule;
- linking the identified monomer domain and/or immuno domain to a library of monomer domains and/or immuno domains to form a library of multimers, each multimer comprising at least two monomer domains, immuno domains or combinations thereof;
- screening the library of multimers for the ability to bind to the first target molecule;
- and
- identifying a multimer that binds to the first target molecule.
A library of multimers, wherein
- each multimer comprises at least two monomer domains connected by a linker;
- and
- each monomer domain binds an ion.
A polypeptide comprising at least two monomer domains separated by a heterologous linker, wherein each monomer domain specifically binds to a target molecule and each monomer domain binds an ion.
The polypeptide of claim 74, wherein the domains form a secondary structure by the formation of disulfide bonds.
A method for identifying a human chimeric monomer domain that binds to a target molecule, said method comprising:
- providing a sequence alignment of at least two naturally occurring human monomer domains from the same family of monomer domains;
- identifying amino acid residues in corresponding positions in the human monomer domain sequences that differ between the human monomer domains;
- generating a library of human chimeric monomer domains, wherein each human chimeric monomer domain sequence consists of amino acid residues that correspond in type and position to residues from two or more naturally occurring human monomer domains from the same family of monomer domains;
- screening the library of human chimeric monomer domains for binding to a target molecule;
- and
- identifying a human chimeric monomer domain that binds to a target molecule.
The method of claim 83, wherein the polypeptide is produced by recombinant gene expression.
A non-naturally-occurring polypeptide comprising an LDL receptor class A domain monomer, wherein the monomer comprises the following sequence:
- (SEQ ID NO:219)
- CaX3-15CbX3-15CcX6-7Cd(D,N)X4CeX4-6DEX2-8Cf
- wherein C is cysteine, Xn-m represents between n and m number of independently selected amino acids, and (D,N) indicates that the position can be either D or N;
- and
- wherein Ca-Cc, Cb-Ce and Cd-Cf form disulfide bonds.
A non-naturally-occurring polypeptide comprising an EGF domain monomer, wherein the EGF domain monomer comprises the following sequence:
- (SEQ ID NO:232)
- CaX3-14CbX3-7CcX4-16CdX1-2CeX8-23Cf
- wherein C is cysteine, Xn-m represents between n and m number of independently selected amino acids;
- and
- wherein Ca-Cc, Cb-Ce and Cd-Cf form disulfide bonds.
A non-naturally-occurring polypeptide comprising an amino acid sequence in which:
- at least 10% of the amino acids in the sequence are cysteine;
- and
- the amino acid sequence is at least 50 amino acids long;
- and/or
- at least 25% of the amino acids are non-naturally-occurring amino acids.
Dependencies are inferred from the claim text.