Selection of fibronectin scaffolds using nucleic acid-protein fusions

Abstract

Disclosed herein are proteins that include an immunoglobulin fold and that can be used as scaffolds. Also disclosed herein are nucleic acids encoding such proteins and the use of such proteins in diagnostic methods and in methods for evolving novel compound-binding species and their ligands.

1 . A molecule comprising i) a protein comprising a fibronectin type III (Fn3) domain, wherein the Fn3 domain: (a) has at least one loop with a modified amino acid sequence relative to the sequence of the corresponding loop of a human Fn3 domain; and (b) binds to a target compound that is not bound by the corresponding human Fn3 domain; and ii) a nucleic acid, wherein the protein is bonded through a DNA-puromycin linker to the nucleic acid, and wherein the protein is encoded by said nucleic acid.  
     
     
         2 . A molecule of  claim 1 , wherein the Fn3 domain binds to said target compound with a K D  of 10 nM or less.  
     
     
         3 . A molecule of  claim 1 , wherein the Fn3 domain is a tenth type Fn3 domain ( 10 Fn3).  
     
     
         4 . A molecule of  claim 3 , wherein the loop is selected from the group of the BC loop, the DE loop and the FG loop  
     
     
         5 . A molecule of  claim 4 , wherein the  10 Fn3 domain has at least two loops with a modified amino acid sequence relative to the sequence of the corresponding loop of a human  10 Fn3 domain.  
     
     
         6 . A molecule of  claim 4 , wherein the  10 Fn3 domain has at least three loops with a modified amino acid sequence relative to the sequence of the corresponding loop of a human  10 Fn3 domain.  
     
     
         7 . The molecule of  claim 4 , wherein at least one of the modified loops in the  10 Fn3 domain is extended in length relative to the corresponding loop of a human  10 Fn3 domain.  
     
     
         8 . The molecule of  claim 7 , wherein the DE loop in the  10 Fn3 domain is extended by 10-13 amino acid residues relative to the corresponding loop of a human  10 Fn3 domain.  
     
     
         9 . A molecule of  claim 3 , wherein the integrin binding motif, RGD, of the  10 Fn3 domain is replaced by an amino acid sequence as follows: basic amino acid-neutral amino acid-acidic amino acid.  
     
     
         10 . The molecule of  claim 1 , wherein the at least one loop is randomized relative to the sequence of the corresponding loop of a human Fn3 domain.  
     
     
         11 . The molecule of  claim 1 , wherein the nucleic acid is ribonucleic acid.  
     
     
         12 . The molecule of  claim 1  selected by the method comprising the steps of a) producing a population of candidate RNA molecules, each comprising a candidate fibronectin type III (Fn3) domain sequence which differs from human Fn3 domain coding sequence, said RNA molecules each comprising a translation initiation sequence and a start codon operably linked to said candidate Fn3 domain coding sequence and each being operably linked to a DNA-puromycin linker at the 3′ end; b) in vitro translating said candidate Fn3 domain coding sequences to produce a population of candidate RNA-Fn3 fusions; c) contacting said population of candidate RNA-Fn3 fusions with the target compound; and d) selecting an RNA-Fn3 fusion, the protein portion of which has a binding affinity or specificity for said target compound that is altered relative to the binding affinity or specificity of said human Fn3 for said target molecule.  
     
     
         13 . A fibronectin type III (Fn3) domain scaffold-based protein that binds to a compound, selected by the method comprising the steps of: a) producing a population of candidate RNA molecules, each comprising a candidate (Fn3) domain scaffold-based protein sequence which differs from human Fn3 domain coding sequence, said RNA molecules each comprising a translation initiation sequence and a start codon operably linked to said candidate protein coding sequence and each being operably linked to a DNA-puromycin linker at the 3′ end; b) in vitro translating said candidate protein coding sequences to produce a population of candidate RNA-protein fusions; c) contacting said population of candidate RNA-protein fusions with said compound; and d) selecting an RNA-protein fusion, the protein portion of which has a binding affinity or specificity for said compound that is altered relative to the binding affinity or specificity of said human Fn3 for said compound.  
     
     
         14 . The protein of  claim 13 , wherein the Fn3 domain contains no free sulfhydryl moieties and no disulfide bonds.  
     
     
         15 . The protein of  claim 13 , wherein the Fn3 domain binds to said compound with a K D  of 10 nM or less.  
     
     
         16 . The protein of  claim 13 , wherein the Fn3 domain is a tenth domain ( 10 Fn3).  
     
     
         17 . The protein of  claim 16 , wherein the  10 Fn3 domain: (a) has at least one loop with a modified amino acid sequence relative to the sequence of the corresponding loop of a human  10 Fn3 domain, wherein the loop is selected from the group of the BC loop, the DE loop and the FG loop.  
     
     
         18 . The protein of  claim 17 , wherein the  10 Fn3 domain has at least two loops with a modified amino acid sequence relative to the sequence of the corresponding loop of a human  10 Fn3 domain.  
     
     
         19 . The protein of  claim 17 , wherein the  10 Fn3 domain has at least three loops with a modified amino acid sequence relative to the sequence of the corresponding loop of a human  10 Fn3 domain.  
     
     
         20 . The protein of  claim 17 , wherein at least one of the modified loops in the  10 Fn3 domain is extended in length relative to the corresponding loop of a human  10 Fn3 domain.  
     
     
         21 . The protein of  claim 20 , wherein the DE loop is extended by 10-13 amino acid residues relative to the corresponding loop of a human  10 Fn3 domain.  
     
     
         22 . The protein of  claim 17 , wherein the integrin binding motif, RGD, of the  10 Fn3 domain is replaced by an amino acid sequence as follows: basic amino acid-neutral amino acid-acidic amino acid.  
     
     
         23 . The protein of  claim 17 , wherein the  10 Fn3 domain has an amino acid sequence that is at least 70% identical to the sequence of a human  10 Fn3 domain.  
     
     
         24 . The protein of  claim 17 , wherein the at least one loop is randomized relative to the sequence of the corresponding loop of a human  10 Fn3 domain.  
     
     
         25 . The protein of  claim 19 , wherein the at least three loops are randomized relative to the sequence of the corresponding loop of a human  10 Fn3 domain.  
     
     
         26 . The protein of  claim 24 , wherein the selection method further comprises: (e) repeating (c) and (d) using a further randomized  10 Fn3 domain.